The Factors Affecting the Marketing

Channel Choices of Rice Farmers

Evidence from the Philippines

Hazel V. Antonio

A thesis submitted in partial fulfilment of the requirements for the degree of PhD in Development
Studies, The University of Auckland, 2021.
ii
Abstract
Rice farmers in developing countries remain poverty-stricken despite increasing agricultural
productivity. Some argue that this is due to their low market engagement, yet there is scant
information on how farmers choose their buyers. This study aims to help fill this knowledge
gap by providing information on the factors affecting their marketing channel choices using
300 surveys and 45 interviews with farmers from the top 42 rice-producing provinces of the
Philippines. Considering the non-separability of the producer, consumer, and worker decisions,
the contributing factors were found to range from personal, social, cultural, and external
environment factors. Using random-effects logistic regression, it was found that farmers who
are younger, non-employed, landowners, live closer to trader-millers, or who are in towns with
a higher income class, and provinces with lower production are more likely to choose direct
marketing channels such as trader-millers, cooperatives, wholesalers, retailers, and consumers.
However, farmers with higher education, lower off-farm income, less training, bigger product
volume, and without indigenous group origin and farm traditions sell to indirect buyers such
as agents, assemblers, and buying stations. The qualitative data suggest that a major part of
their decisions can be attributed to personality traits such as motivation, risk-aversion, and
labour disutility; social ties such as kinship, friendship, or suki relationship; cultural values
such as debt of gratitude, ethos of shame, and loyalty; and the constraints in the external
environment presented by an unfavourable task environment and adverse general environment
factors. The results of the movement analysis also demonstrate that the majority of farmers
stick with the same buyer because of credit and its cultural implications, risk-aversion, and
personal alliance system. These results underscore the importance of creating projects and
policies such as those that will provide farmers with diverse income sources, allow fast and
accessible loans, ease postproduction processes, and link buyers to farmers. More importantly,
these results underscore that to help farmers improve their marketing income effectively, it is
crucial to consider the various non-economic factors – such as personality, social capital, and
cultural values – when creating policy interventions for the Philippine rice industry.

iii
Acknowledgements
It is said that THESIS means “True Happiness Ended Since It Started”. Following that thought,
submitting this thesis means happiness, which I owe to the people and organisations who
helped me make it through. Thus, I would like to express my deepest gratitude to the following:

My supervisors, Professor Andreas Neef and Dr Steffen Lippert, for patiently helping me
through this journey from day one. Without their expertise and superb guidance, this thesis
would not be as it is;

NZAid Scholarship – for funding my studies in the University of Auckland and allowing me
to be in New Zealand and develop my knowledge with the aid of outstanding professors, great
facilities, and an excellent environment;

The Philippine Rice Research Institute (PhilRice) and its management – for believing in me
and allowing me to pursue a PhD while on study leave, for consenting to the use of the RBFHS
database that facilitated my thesis sampling and data gathering, and for the assistance during
my data collection;

The management of DA Regional Offices, especially the RIOs, for the great help during my
data collection ‒ especially for providing transportation and research assistants;

My 300 farmer-respondents – for allotting time for the survey and qualitative interview and for
patiently answering my questions with all honesty and to the best of your ability;

All my Filipino and international friends in Auckland, especially Arisarawan ‘Ming’, Gay,
Hasith, Mai, Michael, the Tangos, the Tionkos, and the Wong Family with Ryan – you all
served as my family and made my stay in New Zealand a fantastic experience so I will forever
cherish you in my heart. Thank you for loving me – for driving me, feeding me, and for
supporting me in all my endeavours;

My parents, Dr. Leopoldo C. Antonio and Anita V. Antonio, this thesis and the accompanying
degree is for you – thank you for all the support, guidance, and most of all, the eternal love;

My siblings, Dr. Marie Ann Antonio and Engr. Florence Antonio, for the love, moral support,
guidance, and prayers; my two beautiful and lovable nieces, Aja and Gemmalet, and my smart

iv
and handsome nephew, Tot, for making me smile even through tough times; Papa Ricky and
Mommy Lucy for cooking delicious and nutritious meals and taking care of me and Franzelle
while I was doing my final revisions; and to all my other relatives for always wishing me
success;

Our baby Franzelle for being my inspiration and for literally kicking me from time to time to
work; my husband, Mark Francis Beltran – for the unconditional love, thoughtfulness, moral
support, delicious meals, for trying to instil discipline in me to accomplish my tasks, and for
fasting and praying for me ‒ I love you and our baby so much;

Above all, to God, for putting me in this university and this country, for surrounding me with
kind and loving people, and for always blessing me abundantly – nothing is impossible with
You, and I would not have done this without Your grace and love.

To ALL of you ‒ thank you, maraming salamat, and kia mihi!

v
Table of Contents
Abstract..................................................................................................................................... iii
Acknowledgements .................................................................................................................. iv
Table of Contents ..................................................................................................................... vi
List of Figures............................................................................................................................ x
List of Tables ............................................................................................................................ xi
Chapter 1. Introduction ............................................................................................................. 1
1.1 Rice in the Philippines ..................................................................................................... 3
1.1.1 The Rice Plant .......................................................................................................... 3
1.1.2 Rice Farming in the Philippines ............................................................................... 4
1.1.3 The Filipino Rice Farmers ........................................................................................ 5
1.2 Rice Marketing in the Philippines ................................................................................... 6
1.2.1 The Rice Value Chain and Farmers’ Marketing Practices ....................................... 7
1.2.2 The Marketing Issues ............................................................................................... 9
1.2.3 Marketing Support Initiatives ................................................................................. 12
1.3 The Research ................................................................................................................. 15
1.3.1 Research Problem ................................................................................................... 15
1.3.2 Research Objectives ............................................................................................... 17
1.3.3 Significance of the Study........................................................................................ 17
1.3.4 Thesis Structure ...................................................................................................... 20
Chapter 2. Literature Review ................................................................................................. 22
2.1 Farmers' Market Participation...................................................................................... 22
2.2 The Factors Affecting Market Participation ................................................................ 26
2.2.1 Personal Factors................................................................................................... 26
2.2.2 Economic Factors ................................................................................................ 30
2.2.3 Institutional Factors ............................................................................................. 35
2.2.4 External Environment Factors ............................................................................. 36
2.3 The Factors Affecting Channels of Sale ...................................................................... 38
2.4 The Research Gap ........................................................................................................ 47
Chapter 3. Research Framework ............................................................................................. 52
3.1 The Agricultural Household Model ............................................................................. 52

vi
 3.2 Social Capital Theory .................................................................................................. 57
3.3 Cultural Capital Theory ............................................................................................... 59
3.4 The External Environment ........................................................................................... 60
3.5 Conceptual Framework ................................................................................................ 62
Chapter 4. Methodology .......................................................................................................... 65
4.1 Quantitative Methodology ........................................................................................... 65
4.1.1 Areas of Research and Sampling ......................................................................... 65
4.1.2 Data Gathering..................................................................................................... 66
4.1.3 The Variables....................................................................................................... 69
4.1.4 Data Preparation .................................................................................................. 84
4.1.5 Data Analysis....................................................................................................... 88
4.1.5.1 Descriptive Statistics ........................................................................................ 88
4.1.5.2 Econometric Model .......................................................................................... 89
4.1.5.3 Movement Analysis .......................................................................................... 91
4.2 Qualitative Methodology ............................................................................................. 92
4.2.1 Study Population and Sampling .......................................................................... 93
4.2.2 Data Collection Methods ..................................................................................... 93
4.2.3 Data Preparation .................................................................................................. 95
4.2.4 Data Analysis....................................................................................................... 96
4.2.5 Research Positionality ......................................................................................... 98
4.3 Limitations of the Study .............................................................................................. 99
4.4 Research Process Summary ....................................................................................... 101
Chapter 5. Farmers’ Typology and their Marketing Practices .............................................. 103
5.1 Farmers’ Typology .................................................................................................... 103
5.1.1 Personal Characteristics..................................................................................... 104
5.1.2 Economic Characteristics .................................................................................. 105
5.1.3 Social Characteristics ........................................................................................ 107
5.1.4 Cultural Characteristics ..................................................................................... 108
5.1.5 External Environment Characteristics ............................................................... 108
5.2 Marketing Practices ................................................................................................... 109
5.2.1 Form of the Product Sold .................................................................................. 109
5.2.1.1 Fresh Paddy Rice ............................................................................................ 110

vii
 5.2.1.2 Dried Paddy Rice ............................................................................................ 114
5.2.1.3 Milled Rice ..................................................................................................... 116
5.2.2 Marketing Preference between Direct and Indirect Buyers and the Typology of
Farmers based on their Choices ..................................................................................... 117
5.2.3 Specific Marketing Channel Choices ................................................................ 121
5.3 Chapter Synthesis ...................................................................................................... 130
Chapter 6. The Different Factors of Marketing Channel Choices......................................... 132
6.1 Personal Factors ......................................................................................................... 132
6.2 Economic Factors ...................................................................................................... 141
6.3 Social Factors............................................................................................................. 148
6.4 Cultural Factors ......................................................................................................... 151
6.5 External Environment Factors ................................................................................... 154
6.5.1 The Task Environment Factors ......................................................................... 155
6.5.2 The General Environment Factors..................................................................... 159
6.6 Chapter Synthesis ...................................................................................................... 169
Chapter 7. Channel Movement and Its Factors ..................................................................... 172
7.1 The Factors Influencing the Extension of the Market Value Chain .......................... 175
7.2 The Factors Influencing the Cut in the Market Value Chain..................................... 179
7.3 The Factors of Same Level Movement ...................................................................... 182
7.4 The Factors of Invariable Marketing Channels ......................................................... 184
7.4.1 The Suki ............................................................................................................. 184
7.4.2 The Factors Keeping the Suki Relationship....................................................... 187
7.5 Chapter Synthesis ...................................................................................................... 193
Chapter 8. Discussion ............................................................................................................ 195
8.1 The Marketing Practices ............................................................................................ 195
8.2 The Factors of Marketing Channel Choices .............................................................. 197
8.3 The Factors of Movement .......................................................................................... 213
8.4 The Typology of the Farmers based on Price ............................................................ 220
8.4.1 The Price Prioritiser ........................................................................................... 221
8.4.2 The Unit Price Prioritiser................................................................................... 222
8.4.3 The Non-Price Prioritiser .................................................................................. 223
Chapter 9. Conclusion ........................................................................................................... 229

viii
 9.1 Summary and Conclusion .......................................................................................... 229
9.2 Practical and Policy Implications .............................................................................. 233
9.3 Further Research ........................................................................................................ 240
Appendices ............................................................................................................................ 244
Appendix 1. Pearson correlation of variables with possible multicollinearity. ................. 244
Appendix 2. The Variance Inflation Factors (VIF) of the independent variables. ............ 245
Appendix 3. The full STATA results of the random-effects logistic regression............... 246
Appendix 4. The average marginal effects of the independent variables. ......................... 247
Appendix 5. Frequency tables of discreet and dummy variables. ..................................... 248
Appendix 6. Selected correlations between certain variables. .......................................... 254
Appendix 7. Number of farmers with suki, classified based on harvest frequency. ......... 256
Appendix 8. Movement of farmers with and without suki within marketing levels. ........ 256
References ............................................................................................................................. 258

ix
List of Figures
Figure 1. Marketing channels of paddy/milled rice in Nueva Ecija, Philippines. ..................... 8

Figure 2. Conceptual framework of the factors affecting the marketing channel choices. ..... 63

Figure 3. Summary of the research process and its timeline. ................................................ 102

Figure 4. Summary of the factors affecting the rice farmer’s marketing channel choices. ... 171

Figure 5. The different levels of marketing channels used in this study. .............................. 172

Figure 6. Changes in the marketing channel levels of farmers within three cropping seasons.
....................................................................................................................................... 174

Figure 7. A simple representation of the farmers who are PP, UP, and NP. ......................... 220

x
List of Tables
Table 1. A summary of the literature on marketing channel choices in the grains industry. .. 46

Table 2. A summary of the variables in the binomial random effects logistic regression
model. .............................................................................................................................. 82

Table 3. Comparison of the prices offered by direct and indirect marketing channels. .......... 85

Table 4. Descriptive Statistics of the continuous variables of the whole population. ........... 103

Table 5. Comparing the means of the continuous independent variables of farmers who sold
to direct and indirect buyers. ......................................................................................... 118

Table 6. Independent sample test showing the significant difference in the characteristics of
farmers who sold to direct and indirect buyers. ............................................................ 119

Table 7. The main marketing channel choices of farmers in the Philippines. ....................... 121

Table 8. The coefficients and average marginal effects of the choices between direct and
indirect marketing channels, using 900 observations from 300 farmer-respondents. ... 132

Table 9. Distribution of the marketing channel choices of farmers with credit. ................... 144

Table 10. The total movement of farmers between marketing levels within seasons 1 to 3. 175

Table 11. The movement of farmers between marketing levels within the two dry seasons.175

Table 12. The movement of farmers between Seasons 1 & 2, 2 & 3, and 1 & 3. ................. 176

Table 13. Distribution of the farmers based on the suki system. ........................................... 185

Table 14. The frequency that farmers have sold to their suki................................................ 185

Table 15. The underlying relationships of the farmers to their suki. ..................................... 186

Table 16. Similarities of the farmers with their suki. ............................................................ 188

Table 17. Reasons why farmers keep their suki relationships. .............................................. 188

xi
Introduction

Chapter 1. Introduction
There are 1.4 billion people in the world who are extremely poor or living on less than USD1.25

a day (Lakner, Negre, & Prydz, 2014; Mottaleb, Mohanty, & Nelson, 2014). Around 40% or

560 million live in major rice-producing areas (Mottaleb et al., 2014). In particular,

approximately 200 million of them are farm households in developing countries with less than

one hectare of land cultivated with their staple food such as corn, wheat, sorghum, and rice

(Rapsomanikis, 2015; Rios, Masters, & Shively, 2008). Most of them are subsistence

households, who rely solely on their own labour and consume their harvest to manage risk

caused by market imperfections, weather shocks, missing credit or insurance market, price

volatility, and transaction costs (De Janvry, Fafchamps, & Sadoulet, 1991; Singh, Squire, &

Strauss, 1986). They are mostly detached from the market, thereby limiting their opportunities

to improve their income (Mottaleb et al., 2014).

In the 1980s, policies to promote market liberalisation were initiated in Latin America and Asia

to improve farmers' income and thus reduce poverty (Boughton et al., 2007; Reardon &

Timmer, 2007). This resulted in the many initiatives that aim to transition farmers from

subsistence farming to commercial farming, which eventually headed to the market led-

paradigm of agricultural development (Reardon & Timmer, 2007). Since then, market structure

and market access have been seen as crucial in improving the livelihood strategy of highly

agriculture-dependent farm households (Gelaw, Speelman, & Van Huylenbroeck, 2016;

Mattos & Zinn, 2016; Vollmer, Hermann, & Mußhoff, 2017; Wickramasinghe & Weinberger,

2013). Consequently, agencies, including the Food and Agriculture Organization (FAO) of the

United Nations, initiated various projects aimed at improving farmers’ incomes in developing

countries through, for example, investments in marketing infrastructure, alternative marketing

1
Introduction

channels, safety nets, and risk reduction (Acharya, 2003; FAO, 2012). However, despite the

many initiatives to improve marketing income through better market participation, recent

studies provide evidence that efforts have not been very successful (Benfica & Tschirley, 2012;

Ojiako et al., 2017).

Particularly in the Philippines, reports show that farmers' marketing practices are still not

resulting in optimal income. In fact, in 2014, only 59.75% of the harvested paddy rice was

marketed, with prices comprising only 44% of the consumer prices (Philippine Statistics

Authority, 2015; PhilRice, 2011). This low share in consumer prices was a result of selling

produce at the farm gate through indirect marketing channels such as agents or brokers

(Philippine Statistics Authority, 2015; PhilRice, 2017). The farmers’ marketing income has

also gone much lower in 2019 with the recent enactment of free rice trade in the country, which

drastically lowered fresh paddy rice prices to as low as PHP 7 (USD 0.14) per kilogram (Rivas,

2019). This enactment presents the need not just to enhance farmers’ production but also to

improve their marketing income. Thus, it is crucial to understand the factors affecting their

marketing channel choices and their movement between channels of sale – which are the focus

of this research.

Before presenting the goal of this research in more detail, the following section offers the

background of rice farming and its importance in the Philippines. Following this section is the

context of rice marketing – discussing the rice value chain, the rice marketing issues, and the

programs created to help improve rice farmers' marketing income. Finally, the last section

presents the problem, objectives, and significance of this research.

2
Introduction

1.1 Rice in the Philippines

This section aims to enhance understanding of rice farming in the Philippines. The first part

describes the rice plant's physiology, followed by statistics to show the level and efficacy of

rice farming in the Philippines. The last part discusses the importance of rice in the lives of

Filipinos.

1.1.1 The Rice Plant

The rice plant, or Oryza, is a type of grass that serves as the staple food of more than half of

the world’s population, making it the most important and widely cultivated crop in the world.

It is believed that the Oryza genus originated about 14 million years ago in what is presently

known as Southeast Asia and the Philippines. It has two main types, Oryza glaberrima or

African rice, which is mostly grown in West Africa, and Asian rice or Oryza sativa, which is

commonly grown worldwide (CGIAR, 2017).

Unlike most grass types, rice is a cultivated plant species, bred and developed by humans for

consumption. About 40,000 different varieties of Asian rice are planted by farmers across all

continents, except Antarctica. While the rice environment is either described as upland or

lowland, irrigated or rainfed, rice cultivation is widespread because it can grow in different

environments ‒ even in areas where other crops would be unproductive (CGIAR, 2017). In

fact, in the Philippines, new varieties can now also grow in adverse environments such as saline

water, dry land, or water-submerged areas (PhilRice, 2017).

It takes about three to six months to cultivate a rice plant and harvest its grains. Specifically, in

the Philippines, where most farmers grow white rice varieties, it takes about 106-144 days

(PhilRice, 2003). The harvested rice grains consist of the endosperm, the germ or embryo, the

bran, and the hull from inner to outer parts. When processed, these grains can be turned into

3
Introduction

either polished or unpolished rice, depending on what parts are removed during processing.

Either way, the grains are threshed from the plant and dried before they can be milled and

stored.

During the first passing in the mill, called dehulling, the hull or the hard outer covering that

comprises 20% of the overall grain weight is removed. This process results in unpolished

(brown) rice – rice containing the nutritious bran that makes 5-8% of the paddy grain weight.

Despite its nutrients, however, the bran is commonly removed through several more polishing

stages that result in white or polished rice. White rice is more prevalent than brown rice in the

Philippines because of the peculiar chewy and nutty taste, short shelf life, and meticulous

cooking preparation required of brown or unpolished rice. Thus, most rice grains in the

Philippine market are white or polished, comprising mainly of the starchy endosperm (Antonio

& Acpal, 2016).

1.1.2 Rice Farming in the Philippines

The Philippines is the eighth-largest rice producer in the world, with rice having a share of

about 19% of the gross agricultural output (Philippine Statistics Authority, 2017). Moreover,

rice contributed 19% of the PHP 660B (USD 13.2B) Gross Value Added (GVA) coming from

agriculture and fisheries in 2010 (Philippine Statistics Authority, 2010), while reports from

1993 to 2003 demonstrate that paddy rice consistently provided about two to three per cent of

the Philippine Gross Domestic Product (GDP) (Cororaton, 2014).

The historical average of the land planted with rice in the Philippines was 4.55 million hectares

from 2007 to 2016. The majority of these rice farms were in Luzon – Nueva Ecija (6.9%),

Isabela (6.1%), and Pangasinan (5.7%) (Philippine Statistics Authority, 2017). In 2017, the

total area planted with rice had further increased to 4.81 million hectares; thus, accounting for

4
Introduction

as much as 66.9% of the 7.19 million hectares of farm holdings in the country (Philippine

Statistics Authority, 2018b). This was in stark contrast to the pre-colonial period when rice was

only planted in the upland and some water-logged areas, thus limiting the supply and elite

branding (Aguilar, 2005).

Given topography issues and other farming challenges, the average yield of a rice farm in the

Philippines from 2007 to 2016 was only 3.8 metric tons per hectare (Philippine Statistics

Authority, 2017). Specifically, during the crop year 2013-2014, Philippine rice productivity

was only 3.84 tons per hectare (t/ha), while Vietnam achieved 6.33 t/ha, China 6.10 t/ha,

Indonesia 5.42 t/ha, and Thailand recorded 5.31 t/ha (Bordey, Moya, Beltran, & Dawe, 2016).

In 2016, the national average was 3.87 metric tons, with the top three average yields being 5.37

t/ha in Nueva Ecija, 4.70 t/ha in Ilocos Norte, and 4.60 t/ha in Biliran.

The country's average total rice production was 17.3 million metric tons from 2007 to 2016,

with Nueva Ecija, the top rice-producing province, contributing around a decile to the country’s

total harvest (Philippine Statistics Authority, 2017). In 2017, the volume of paddy rice

produced was high at 19.28 million metric tons (Philippine Statistics Authority, 2018b). Yet,

it was still insufficient to feed the Philippine population of more than 100 million, with an

annual per capita consumption of 114-119 kg (CGIAR, 2017; OECD, 2017).

1.1.3 The Filipino Rice Farmers

About 2.4 million farmers are involved in rice production in the Philippines (Mataia et al.,

2018). While they provide the country's staple food, they were reported to be the poorest sector

in 2015, with 34.3% living below the Philippine poverty threshold (Philippine Statistics

Authority, 2016; The World Bank, 2018). Similarly, according to World Bank reports, a

significant part of the decile with the lowest income in the Philippines from 2006 to 2015 were

5
Introduction

the rice farmers, with an increasing poverty rate. Specifically, in 2015, they comprised 10.1%

of the bottom 10% of the income distribution, with a 31.1% poverty rate. In 2016, this worsened

further, with farmers comprising 16.1% of the lowest decile, with a 39.3% poverty rate (The

World Bank, 2018). According to this data, the average income of a rice farmer per year is just

around USD 845, which is equivalent to USD 169 per person for a family of five (PhilRice,

2011). During the same year that the data was collected, the Philippines' per capita poverty

threshold was USD 271. Thus, this shows that what farmers earn from rice farming is only 62%

sufficient to meet their daily needs.

While not all farmers are solely dependent on rice, most of them (93%) perceive rice as their

primary source of income; only 4% and 3% consider other agricultural products and non-

agriculture as their primary sources of livelihood, respectively. In addition, they are only

farming 1.41 hectares on average, with 60% of them cultivating only one hectare or less.

Moreover, half of them are not landowners but are just tenants or workers (Philippine Statistics

Authority, 2015; PhilRice, 2017). Furthermore, around 45% of Filipino farmers usually borrow

their capital. Specifically, 80% of them borrow from informal lenders despite very high interest

rates, while only 20% borrow from formal institutions like banks despite their much lower

credit rates (PhilRice, 2017). Finally, 72% of rice farmers limit their role to producers – selling

their produce fresh at the farm gate, thereby positioning themselves at the start of a rather long

market value chain. This will be further described in the following section.

1.2 Rice Marketing in the Philippines

Rice marketing in the Philippines can be very long and complicated since rice can come in

different forms – from fresh to milled and anything in between – which entails various

processes, commands different prices, and involves different sets of market players (CGIAR,

2017; Mataia et al., 2018). To understand this better and determine the Filipino rice farmers'

6
Introduction

position, this section describes the rice marketing process in the country – from the rice value

chain, the marketing practices of farmers, the different marketing challenges they face, and the

various marketing support provided to them.

1.2.1 The Rice Value Chain and Farmers’ Marketing Practices

About 84% of the farmers have a marketable surplus, leaving only 16% as subsistence farmers

(PhilRice, 2017). Thus, after cultivating the rice plant for three to four months, most farmers

sell part, if not all, of their product somewhere along the market value chain.

After harvesting, among the logistic components are drying, transporting, milling, classifying,

packaging, storing, and distribution (Alvero, 2008; Mataia et al., 2018). Inside the rice value

chain are the farmers, input suppliers, and the different marketing channels. The marketing

channel could be indirect or direct buyers. The indirect buyers are those who buy paddy rice

and sell them again in the same form, which includes agents, barangay assemblers, municipal

assemblers, provincial assemblers, regional assemblers, interregional assemblers, trader-

millers, and buying stations. The direct buyers can be buyers who buy processed paddy or

milled rice, such as consumers, retailers, cooperatives, trader-millers, and the government

regulatory arm called the National Food Authority (Philippine Statistics Authority, 2015).

There are 112 players in the rice market value chain in the Philippines, where paddy and milled

rice may pass through before they reach the consumers. Specifically, there are 78 identified

marketing players in Luzon, 88 in the Visayas, and 68 in Mindanao. However, the traditional

supply chain is composed of farmers, paddy traders, millers, wholesalers, retailers, and

consumers (Mataia et al., 2018, 2020). Being the producer, the farmer can sell their products

to direct or indirect buyers along any part of the chain. However, the farmer’s role usually ends

at threshing or removing the grains from the plant as they tend to sell their harvest fresh at the

7
Introduction

farm gate immediately. In fact, in 2011, the PhilRice’s Rice-Based Farm Household Survey

(RBFHS) recorded that, of all the rice farm households that had a marketable surplus, 72% sold

immediately at the farm gate, and 82% of them sold their rice in its fresh form (PhilRice, 2017).

Exceedingly few farmers dry and mill their harvest to sell brown or white rice in the market.

Most of them dry and mill only a small amount for personal consumption through the village

miller using the rice bran, usually used for animal feed, as their milling fee (Alvero, 2008).

Farmers are, thus, usually at the start of an extended market value chain since most of them

sell to indirect buyers at the farm gate. They reportedly sell to agents or assemblers (Fig. 1)

because they are on-the-spot buyers and a source of immediate "sure money", among others

(PhilRice, 2017).

Ó PSA, 2015

Figure 1. Marketing channels of paddy/milled rice in Nueva Ecija, Philippines.

8
Introduction

Specifically, many farmers sell to agents who buy fresh paddy rice at the farm gate and then

put a cap on the price at around PHP 0.5-1.0/kg (USD 0.01-0.02/kg) (Alvero, 2008). These

men then sell the same product and get an income by merely serving as an intermediary or

negotiating the transaction between the farmers and the assembler, i.e., municipal level. This

municipal assembler then sells to a provincial assembler, who then passes the product to an

interregional assembler, who then markets to a miller, who mills the rice and sells it to

wholesalers who finally retail the rice to consumers (Fig. 1). Farmers also usually sell to a

barangay assembler, who then sells to a municipal assembler and thus, the market value chain

is equally long as selling to an agent (Philippine Statistics Authority, 2015; PhilRice, 2017). In

contrast to these, very few farmers market to a direct buyer, such as a cooperative, which

directly process the paddy rice into milled rice and sell it to different consumer channels,

including small retail markets, big retail markets, and food service institutions (Fig. 1) (PSA,

2015).

1.2.2 The Marketing Issues

Rice farmers have been at the centre of Philippine government initiatives since 1987. Despite

that, they still receive a low share in consumer price – even with the Philippines' high gross

marketing margin that gives it one of the highest prices for milled rice in Asia (Bordey et al.,

2016; McKay, Cazzuffi, & Perge, 2015; PhilRice, 2017). Notably, in 2016, the share of Filipino

rice farmers in consumer price was only 27.16%, while that of Indonesia’s was already around

52% (Jegho, 2016; Philippine Statistics Authority, 2018a). Reports show that during the same

year, there was a 115% gap between the farm gate price and wholesale price of rice in the

Philippines, making rice farmers the lowest share receiver in the final cost of the product among

all grains, and 14th among the 26 commodities evaluated in the country. Peanut farmers

9
Introduction

received the highest share at 79% because they process their products and sell them directly to

consumers or wholesalers (Guzman, Madamba, Banzon, & Zapata Jr, 2011).

One of the reasons for the low share in consumer prices is the several marketing channels that

rice must go through in the Philippines before reaching the consumers – with the traditional

supply chain comprising around six players with brokers in both rice buying and distribution.

This makes the Philippines' marketing supply chain more lengthy because of more players and

interactions compared with other Asian countries whose players have integrated and diversified

functions to shorten the chain (Mataia et al., 2018). Based on a study by PSA (2015), because

farmers sell fresh paddy rice to indirect marketing channels at the farm gate, they receive a

gross margin of only PHP 9-14 (USD 0.18-0.28) per kilogram. Thus, they get only about 22.5

to 30 per cent of the retail rice price that is PHP 40/kg (USD 0.77/kg), leaving them with barely

0.15 net profit cost ratio given production expenses that can be as high as PHP 11.49 (USD

0.23) a kilogram (PhilRice, 2015). In comparison, a provincial assembler who buys paddy rice

from the farmer gets a net income of as much as PHP 4.28/kg (USD 0.09/kg), with a 0.26 net

profit cost ratio. Similarly, a village assembler who buys paddy rice to process into milled rice

to sell to consumers earns a net income of as much as PHP 8.99/kg (USD 0.18/kg), giving a

net profit cost ratio of 0.3 (Philippine Statistics Authority, 2015). Thus, the farmer who

cultivated the rice for four months is earning much less than a trader who spent at most a month

to re-sell the product, or an assembler who spent less than half of the farmers’ efforts.

Another major issue that farmers are facing is the full implementation of trade liberalisation.

The Philippines decided to maintain special treatment for rice since it joined the World Trade

Organisation (WTO) in 1995, extending it several times until June 2017, to protect millions of

local rice farmers from the surge of cheaper imported rice that they are still unable to compete

with (USTR, 2015; WTO, 1999). In February 2019, however, Philippine President Rodrigo

10
Introduction

Duterte signed the Republic Act No. 11203 titled “Act liberalising the importation, exportation,

and the trading of rice, lifting for the purpose the quantitative restriction on rice, and for other

purposes” (Corpuz, 2019). While this is expected to help the poor, especially the landless and

the urban consumers, it is also anticipated to lower the price of paddy rice further, given the

profit-maximising goals of traders. Specifically, it was projected to further decrease farmers’

income by 29%, likely positioning more of them below the poverty line (Balisacan & Ravago,

2003; Javier, 2017). Unfortunately, the law's repercussion was worse than expected – the

consumers are yet to enjoy the promised cheaper milled rice at PHP 27 (USD 0.54). Still, the

farmers are already selling at a loss ‒ for as low as PHP 7 (USD 0.14) per kilogram ‒ which

has resulted in negative income for many, given the average production cost of PHP 12 (USD

0.24) per kilogram (Montemayor, 2020; Rivas, 2019).

EFSIM similarly reported that distorted and inefficient markets, specifically those lacking

information in other markets, are hindering farmers from getting the maximum income from

better commercialisation practices (EFSIM, 2009). Specifically, there is limited knowledge on

paddy grades and standards set by government offices such as the NFA (Mataia et al., 2020)

and even a lack of definitive product quality standards (EFSIM, 2009). Without adequate

market information on pricing, farmers cannot negotiate prices or plan for longer-term

activities and are thus, left at the mercy of traders who dictate the farm gate prices, especially

during peak harvest season.

Another market issue is the weak linkages that deprive agro-processors of access to stable

markets (EFSIM, 2009). More importantly, there is a lack of stability in the rice market and a

non-transparent assessment of paddy rice due to the limited access to moisture meters (Mataia

et al., 2020). Farmers likewise lack entrepreneurial skills that hinder them from effectively

11
Introduction

using market investment information to plan for longer-term production and marketing and

take advantage of economic opportunities (EFSIM, 2009).

Others cited limited access to rice mills due to bad road conditions as another impediment in

maximising marketing income, as well as underutilised rice mills (Alvero, 2008; Mataia et al.,

2020; Muthayya et al., 2014; PhilRice, 2017). Similarly, poor transport facilities and high

transport costs cause market distortion and inefficiency (EFSIM, 2009; Mataia et al., 2020). In

addition, there is limited access to drying and storage facilities that result in low quality and

low prices for fresh paddy rice and high handling costs owing to the manual labour involved

from procurement to marketing (EFSIM, 2009; Mataia et al., 2020). There is also the issue of

high marketing costs resulting from different constraints such as lower economies of scale and

high packaging and maintenance costs (Mataia et al., 2020). Adding to these factors is the poor

access to low-cost credit and crop insurance. Finally, the high unit margin in the rice price is

exacerbated by the profit-seeking behaviour of the buyers within the chain (Mataia et al., 2020).

1.2.3 Marketing Support Initiatives

To help farmers get through the marketing challenges mentioned above, the government has

been easing the way towards market participation in the past decades. While spending on

research and development, basic transport infrastructure, and institutional development

between 1990 and 2000 was negligible, it dramatically increased during the last decade. Recent

projects address different agricultural challenges, including inefficient supply chains by

providing farm-to-market roads, research and development, and extension services (OECD,

2017).

The Department of Agriculture (DA) and the National Food Authority (NFA) also provide

marketing assistance through training and establishing market linkages between farmer

12
Introduction

cooperatives and buyers through forums and market encounters. Market information was

similarly made available through the Agricultural Marketing and Information System. It is an

online service that aims to help farmers make rational and viable marketing decisions through

information on what, where, when, how, and to whom they can sell their rice (DA, 2017;

Pabuayon, 2006). To address high procurement price and low quality of milled rice, drying and

storage facilities have been established in selected major rice-producing provinces while the

Cabotage Law was implemented to increase the competition among shipping companies to

lower the shipping costs of domestic and even international goods (Mataia et al., 2020).

Finally, the government has a price support system carried out by the NFA, which regulates

rice buying and selling price by buying from farmers and their organisations at a reasonable

price and selling cheap milled rice to distributors nationwide. Notably, farmers can sell clean,

partially dried paddy rice to the NFA at PHP 17/kg (USD 0.34/kg) pick-up-price, resulting in

an additional 28.5% over the farm gate price of PHP 13.23/kg (USD 0.26/kg) (NFA, 2011;

PhilRice, 2015). They can even get an additional 5.3%, resulting in a 33.8% higher selling price

than the farm gate price of private buyers. This is by drying the paddy rice until it reaches 14%

moisture content, selling as part of a group, and delivering the paddy rice to the nearest NFA

warehouse or buying station strategically located in all provinces. Being a regulatory

instrument that aims to guarantee a fair return on production investment and ensure a buyer for

the farmers’ produce, any farmer, whether an individual or an organisation, can sell to NFA.

The only requirement is to have a farmer’s passbook, which requires an information sheet, a

certification from the community leader or agriculturist office of the location and area of the

farmland, and a picture that can be taken at the NFA office free of charge (NFA, 2011, 2017).

To partly address the unstable price of fresh paddy rice, some local government units and

farmers’ associations and cooperatives have been buying fresh paddy rice to stabilise its prices,

13
Introduction

especially given the influx of imported milled rice brought about by the rice free trade (Mataia

et al., 2020).

However, since the rice tariff law enactment, most of these marketing projects/supports have

been discontinued, and efforts were once again focused on increasing farmers’ productivity.

Specifically, the government has significantly abated policy support. While the NFA was

originally tasked with regulating rice buying and selling price, its role now is limited to buffer-

stock management (Corpuz, 2019; DA, NEDA, & DBM, 2019). Thus, with the influx of

cheaper imported rice because of trade liberalisation, paddy rice prices fell from around PHP17

(0.34 USD) in 2018 to as low as PHP7 (0.14 USD) in 2019, thereby making farmers suffer

drastic losses (Montemayor, 2020; Rivas, 2019). Moreover, instead of marketing projects, the

government has reverted to focusing on increasing yield, even though empirical evidence

shows that poverty can prevail among farmers despite agricultural productivity if marketing is

inefficient (Acharya, 2003). In fact, to address the very low rice farm gate prices brought about

by the free rice trade, the government has annually allocated PHP 10B (200M USD) of the

tariffs to be collected from imported rice to put towards increasing domestic farm productivity.

This is through the project “Rice Competitiveness Enhancement Fund (RCEF)”, which

distributes free seeds, farm production equipment and postproduction facilities, and extension

support to farmers (Corpuz, 2019; DA et al., 2019; Mataia et al., 2020). Loans with lower

interest rates are also part of the support package available to farmers, but these are rarely put

in place due to repayment issues. These provisions are to be given through farmer cooperatives.

However, most farmers in the Philippines are not organised into groups ‒ especially those in

far-flung areas and those in provinces with low rice production volumes ‒ so they hardly benefit

from these projects. These RCEF projects, along with other productivity-related initiatives, are

not planned nor implemented through the local government. Instead, they are implemented

14
Introduction

through specific government agencies, so they are not holistic. To date, farmers are still

complaining about the low farm gate prices of rice, but the government has no current project

nor upcoming initiatives to address it.

1.3 The Research

Given the importance of rice in the Filipino diet and farmers' livelihoods, it is crucial to make

rice farming sustainable. With the rice buying price that is as low as PHP7 (0.14 USD) per

kilogram resulting from the newly enacted free rice trade in the country, the need to improve

farmers’ marketing incomes has become even more vital. Therefore, it is essential to find out

how rice farmers can be helped in marketing their products at optimal prices. One way to do

that is by understanding their marketing channel choices since asymmetric information is a big

problem in developing countries like the Philippines. This issue results in many types of traders

or marketing channels that give volatile prices and enforce different buying or paying set-ups

that usually disadvantage the farmers (Kherallah, Delgado, Gare-Madhin, Minot, & Johnson,

2000). This is especially true in the case of rice, which can be sold in different forms that also

call for various buyers.

1.3.1 Research Problem

In addressing the rice farmers’ poverty, it is important to understand why they choose specific

buyers, i.e., indirect buyers, which lengthens the market value chain, thereby lessening their

share in consumer prices (Crawford, 1997). For instance, why do farmers sell at the farm gate

or to commission agents, which results in other market players receiving much of the gain from

the different marketing projects and policies that are supposed to help them (Barrett et al., 2005;

Gelaw et al., 2016; Javier, 2017; Soe, 2015). Similarly, what are their reasons for not selling to

their cooperatives when they are direct rice processors and can get additional income through

15
Introduction

dividends? Another question is, why they do not aim to sell their products wholesale to large

companies despite the willingness of the latter to source products from them (Gani & Adeoti,

2011; Mataia et al., 2018; Ojiako et al., 2017; PhilRice, 2017; Reardon, Barrett, Berdegué, &

Swinnen, 2009)? Lastly, it is crucial to understand why they do not sell directly to individual

consumers when such direct marketing can give the highest profit margin, as it eliminates the

middlemen along with their fees (Nxumalo, Oduniyi, Antwi, & Tekana, 2019).

This research recognises that marketing channels involve many complex factors, which can

make farmers choose less direct buyers. Nevertheless, despite the crucial role of marketing

channel choices in improving farmers' incomes, there is an inadequate number of studies on

the many factors affecting them (Nxumalo et al., 2019; Soe, 2015; Soe, Moritaka, & Fukuda,

2015) ‒ especially in the Philippines.

To fill this knowledge gap and consequently assist rice farmers in improving their marketing

income, this study will address this problem by answering the main research question: What

factors affect the marketing channel choices of the rice farmers in the Philippines?

The following are sub-questions created to answer the main research question in more detail:

1. What is the typology of the rice farmers and their marketing practices?

2. What are the different variables, and how are they affecting the choice of direct versus

indirect marketing channels based on quantitative data?

3. What are the different factors, and how are they affecting farmers' marketing channel

choices based on the qualitative data?

4. How do farmers move between channels of sale, and what factors lead to such

movements?

16
Introduction

1.3.2 Research Objectives

This study seeks to identify, describe, and explain the different factors affecting farmers'

marketing channel choices with the aim of helping improve farmers' lives through better

marketing income.

Specifically, this study has the following objectives:

1. To describe the rice farmers' typology in terms of their marketing practices and choices.

2. To enumerate the different variables and explain how they affect marketing channel

choices using quantitative and qualitative data.

3. To identify farmers' movement between channels of sale and discuss the factors that

extend, shorten, or keep them in their value chain position.

1.3.3 Significance of the Study

To date, there is still limited information on the factors affecting the marketing channel choices

of rice farmers, particularly in the Philippines, when it is vital for producers because it

significantly impacts their profitability (Soe et al., 2015). There are limited studies that explain

why individual rice farmers choose to sell at the farm gate rather than directly to milled rice

buyers or agents rather than to millers when the channels of sale prove to have huge

implications on farmer welfare (Fafchamps & Hill, 2005). The factors affecting their

movement between channels of sale is remarkably scant despite several studies expressing the

importance of deepening the knowledge and understanding of marketing practices to solve

chronic poverty (Gani & Adeoti, 2011; Hulme & McKay, 2007; Jackson, 2010; Pecotich,

Rahtz, & Shultz, 2010). While there has also been a call for identifying the operations and

decisions of small-scale farmers as early as the 1980s to understand them fully and thus, make

intervention strategies successful (Beckford, 2002; Gladwin & Murtaugh, 1980), studies within

17
Introduction

this space are still lacking. Moreover, much previous research highlights the potential benefits

of understanding farmers’ marketing decisions from employing a combination of techniques

involving both qualitative and quantitative approaches (Aklilu, Almekinders, Udo, & Van der

Zijpp, 2007; Musser, Patrick, & Eckman, 1996). However, to date, to the best of my

knowledge, there is still no research on the marketing channels of farmers, particularly of rice

– the staple food for more than half of the world’s population (FAO, 2015; Khanal & Maharjan,

2015) – which uses such a mix of methods.

In response to the need for such a study and to help fill the knowledge gap, this research

examines rice farmers' marketing channel choices in the Philippines, using both quantitative

and qualitative methods. Therefore, this research contributes to advancing the body of

knowledge in development studies, particularly in agricultural economics and rural

development, by providing quantitative and qualitative data on farmers’ marketing channel

choices and their factors. Specifically, this study investigates the different factors affecting the

choice of marketing channels – holistically looking at all the elements, including personality,

social, cultural, and external environment factors. While there are numerous studies on farmer’s

market participation, these are typically confined to identifying the sociodemographic and

socioeconomic factors that affect market participation, rarely including channels of sale (Azam,

Imai, & Gaiha, 2012; Benfica, Boughton, Mouzinho, & Uaiene, 2014; Khanal & Maharjan,

2014, 2015; Van Huong, Tran Huu, & Lebailly, 2016). Similarly, they give a mechanistic view

of the farmers by predicting marketing decisions solely based on these factors. Farmers,

however, are normally more complicated than that, considering the environment in which they

operate. Hence, while sociodemographic or socioeconomic factors may help explain it, they

are only one part of the bigger picture. This study also uses longitudinal data analysis to

18
Introduction

determine the influence of different factors on the choice of marketing channels without the

unintended effect of fixed factors that are unobservable.

Furthermore, this research offers information on farmers' movements between channels of sale

over time and their reasons behind such actions using the panel data. This information could

help explain why farmers add or cut market players, move from one buyer to another, or

patronise just one buyer over the years. Based on the literature found, this area of study is not

well explored yet ‒ despite its potential to help understand change – particularly in marketing

behaviour or choices. Thus, this study contributes not only to agricultural economics and rural

development but also to other areas, such as behavioural economics.

Moreover, this research is of paramount importance in improving farmers’ lives and making

rice production sustainable. An increase in rice production has consistently been among global

targets since more than half of the world population eat rice as their staple food (FAO, 2015).

Most rice is produced in developing countries such as India, Indonesia, Bangladesh, Vietnam,

Thailand, Myanmar, and the Philippines, where farmers are reported to be poor (Rapsomanikis,

2015; “World Rice Production 2017,” 2017). Thus, apart from increasing rice production, an

increase in farmer income has also been among the primary targets of the national governments

of these countries to keep rice production sustainable (Gabriel & Hundie, 2006; Soe et al.,

2015).

Most importantly, this research makes practical contributions as it provides information that is

important in the formulation of effective and efficient private and public policies, advisory

support, and measures to improve farmers’ marketing practices. This study's results will be

very timely in helping farmers in the Philippines minimise the ill effects of trade liberalisation

and help save them from being exploited by players who hold market power and better

information. Understanding the factors affecting the marketing channel choices is crucial in

19
Introduction

creating programs and policies that could effectively help farmers sell their products to direct

buyers ‒ particularly to consumers. Farmers have been shown to earn more if they can better

see, realise, and understand their practices and tendencies (Gani & Adeoti, 2011; Sartwelle et

al., 2000; Vollmer et al., 2017). Hence, this study's findings have the potential to help increase

farmer marketing income by showing them better strategies, i.e., providing suggestions on how

to cut the long rice market value chain and thus increase their share in consumer prices.

Particularly, this study will help achieve two of the Philippine rice strategic plan goals for 2022:

one, to improve rice trade through efficient postproduction, better product quality, and reliable

supply and distribution system and two, create a science-based and supportive rice policy

environment (Philippine Statistics Authority, 2017).

Moreover, because the study results are location- and crop-specific, it will significantly help

create custom-fit and appropriate market-related projects and policies in the Philippines. It will

provide information on what needs to be addressed to improve institutional contexts and

promote more profitable channels of sale. Directing farmers towards better marketing practices

may prove beneficial as it can optimise resource allocation and income that could help them

out of poverty (Mottaleb et al., 2014). Overall, these will help improve the lives of farmers in

the Philippines and other developing countries as well. Therefore, this research can contribute

to the Sustainable Development Goals, particularly to Goal 1, which is eradicating poverty;

Goal 2, ending hunger; and Goal 12, responsible consumption and production.

1.3.4 Thesis Structure

This thesis contains nine chapters. Following the first chapter, Chapter 2 reviews the related

literature, including the factors affecting the market participation of farmers in the grains

industry, given the lack of research on marketing channel choices. Chapter 3 presents the

framework that explains the theories and resulting conceptual framework that guided this

20
Introduction

research. Chapter 4 is divided into the quantitative and qualitative methodology. The

quantitative section explains the methods used to derive farmers' typology, the factors affecting

their marketing choices, and their movement between channels of sale. The second part

describes how the qualitative interviews were designed to gather more information on the

factors affecting the farmers’ marketing decisions and the reasons behind their movement

between channels of sale.

The results of this research, both from the quantitative and qualitative inquiry, comprise

Chapters 5 to 7. Chapter 5 presents farmers' typology and their marketing practices; Chapter 6

enumerates the different factors affecting the rice farmers' marketing channel choices based on

mixed methods. Chapter 7 presents the factors that result in the movement or non-movement

within different marketing channel levels. Chapter 8 provides a discussion of the results

mentioned above, synthesising the various factors that farmers say matter to them and what

actually matters to them when choosing a marketing channel, cutting or adding a market player

in the chain, or staying with a particular buyer. Finally, Chapter 9 offers conclusions and

presents policy implications following the results of this research.

21
Literature Review

Chapter 2. Literature Review

Many studies have attempted to understand market participation and the factors affecting it.

However, detailed research on the factors affecting farmers' marketing channel choices ‒

especially in the grains industry ‒ is minimal despite being a crucial aspect of marketing. Given

this scarcity, most of the studies reviewed in the following sections are on the market

participation of grain farmers and its factors, assuming that they are the same factors affecting

the choice of marketing channels. The few studies on the marketing channel choices of farmers

in the grains industry are presented and discussed towards the end of the chapter.

2.1 Farmers' Market Participation

There are several studies aimed at understanding the market participation of farmers in

developing countries (Gani & Adeoti, 2011; Hulme & McKay, 2007; Jackson, 2010; Pecotich,

Rahtz, & Shultz, 2010). These, however, mostly document the rate and level of farmers’ market

participation in Africa, with just a few in Asia. One of the studies conducted in Asia is that of

Cazzuffi and McKay (2012), where they analysed rice farmers' market participation and their

channels of sale in Vietnam. Their study used panel data for 2006, 2008, and 2010 to investigate

rice farmers' marketing practices. They record that the number of farmers selling their rice has

increased from 51% to 56% and that their output sold has also increased from 46% to 54%

from 2006 to 2010.

Mottaleb et al. (2014) similarly explore rice farmers' market participation in Bangladesh, one

of the world's significant rice-producing and rice-consuming countries. They used the data of

8290 farmers from Bangladesh’s Household Income and Expenditure Survey collected in 2000,

2005, and 2010, and a community survey to understand the farmers’ market participation

22
Literature Review

decisions. The results show that 71% of the farmers sold rice during the period sampled, which

was a little higher than that recorded by Cazzuffi and McKay (2012).

The rest of the studies are mainly in the context of Africa. These studies include one conducted

in Mozambique wherein Benfica and Tschirley (2012) used panel data from 2008 to 2011 to

look at the dynamics of smallholder farmers' market participation. Among its 1186

respondents, 87.6% were farming maize, 40.9% were producing cassava, 24.3% were

cultivating sorghum, 22.4% were growing pigeon pea, 12.3% were planting common beans,

and 7.1% were producing groundnut. Rice farmers, however, accounted for only 6.8% of the

sampled population. The results suggest that rice and sorghum farmers’ market participation

are meagre compared with that of other crops. It was only 10% for sorghum and 22.6% for

rice, making them the products with the first and second lowest market participation,

respectively.

Looking specifically into rice, Benfica and Tschirley (2012) showed that only 2.9% and 2.8%

of rice farmers in Mozambique sold rice to the market in 2008 and 2011, respectively. Of the

6.8% farmers cultivating rice, 77.4% never sold to the market from 2008 to 2011, around 9.9%

sold only in 2008, 6.7% sold only in 2011, and only about 6% sold consistently from 2008 to

2011. These numbers are very low compared with the other farmers' market participation in

the study, who planted different crops, including maize farmers who had 28.3% and 32.5%

market participation in 2008 and 2011, respectively. According to Benfica and Tschirley

(2012), maize had the highest proportion of households selling because of the aggressive

buying by a poultry producer from 2008. Overall, the farmers' market participation in

Mozambique is small compared with the studies in Vietnam and Bangladesh. Nevertheless,

Mottaleb et al. (2014) concur with their results – that rice is the least marketed commodity

among all major food crops such as wheat and corn. However, the enormous disparity is hard

23
Literature Review

to explain because Benfica and Tschirley (2012) did not discuss the probable reasons for the

low market participation or the very low volume of the rice marketed reflected in their findings.

It was also hard to deduce because their study did not mention the average area planted with

rice. Their graph only indicates the average sales per household, which is about 100 kg, and

the share of sales in production, which is about 50%. Inferring from this, the rice farmers in

their study only produce about 200 kg of rice, which is relatively small for a farmer who grows

rice for a living. Comparing this with Philippine farmers, statistics show that Filipino rice

farmers sell 53% of the total yield, retain 19% for home consumption, and use the rest as input

payment and seeds (PhilRice, 2011). With an average yield of 3.81 t/ha and a land area of 1.16

ha, given some processing losses, this would leave them with approximately 2200 kg for

marketing. This is around 22 times more than the rice sold in the research of Benfica and

Tschirley (2012). Therefore, the rice farmers' low market participation in the said study,

particularly as compared with the results of Cazzuffi and McKay (2012), could just be because

of the differences in respondents. Specifically, while the farmer-respondents of Benfica and

Tschirley (2012) were farming rice mostly for their own consumption, the farmers who

provided the data for Cazzuffi and McKay (2012) are cultivating rice for a living.

Gani and Adeoti (2011) studied the market participation rate in 24 villages in the Northern part

of Taraba State in Nigeria. It featured farmers cultivating various crops, including rice, maize,

sorghum, millet, cassava, sweet potato, cocoyam, tomato, melon, banana, mangoes, oranges,

beans, groundnut, garden eggs, and other vegetables. Of the 2456 farmers in the whole state,

120 farmers were selected through multi-stage stratified random and purposive random

sampling. Specifically, six local government areas were chosen first, followed by the random

selection of two wards in each area, from which two villages were then randomly selected.

Overall, Gani and Adeoti (2011) chose 120 respondents from these 24 villages proportional to

their population size. Results from the indexing and Tobit and logistic regressions showed very

24
Literature Review

high market participation. In particular, 110 or 92% participated in the market at various levels.

Specifically, 8.33% of the farmers were at level 1 or selling 450 kg and below, 12.5% were at

level 2 or selling 451 to 1000 kg, 28.83% were at level 3 or selling between 1001 and 1450 kg,

while 50% were at level 4 or selling above 1450 kg.

While these results are detailed, the study's farmer-respondents were not specified or

dichotomised into specific products. Thus, it is hard to compare these results with other crop-

specific results. Some of the products involved are also highly perishable, such as fruits and

vegetables. This difference could result in higher market participation since the high

perishability of goods leaves farmers no choice but to sell their products even at very low prices

to prevent postharvest losses (Adepoju, Owoeye, & Adeoye, 2015). This might also result in

the marketing of grains or other non-perishable products since farmers have already incurred

the fixed transaction cost for the perishable goods.

Maize, the most studied staple grain, was the focus of Olwande et al. (2015) on a five-set panel

data on farmers' agricultural marketing in Kenya, which suggested that only a minority of the

farmers sold from year to year. It was recorded at 40% and 36.6% in 2000 and 2010,

respectively, demonstrating a decreasing trend. Moreover, the study showed only 13.1%

consistently selling to the market, 57.9% inconsistently selling, while 29% did not sell at all

throughout the study period. Like Gani and Adeoti (2011), Omiti et al. (2009) also investigated

the intensity of smallholder farmers' participation in Kenya. To determine the difference in the

market participation between rural and peri-urban farmers, Omiti et al. (2009) purposively

selected eight villages from each of the two districts. They gathered data through rapid rural

appraisal with 15-40 farmers in each of the 16 villages. The first data-gathering stage was

through a structured checklist questionnaire to capture information from the focused group

discussions conducted. After one month, they conducted another household survey with the

25
Literature Review

same villages but with different households, purposively selecting farmers who sell milk,

vegetables, or maize. In the end, the study only had 76 maize farmers selling 65.64% of their

maize, which is relatively high compared with the other studies because the respondents were

limited to those marketing their produce.

2.2 The Factors Affecting Market Participation

Similar to the rate of market participation, there are several studies on the factors affecting the

market participation of different farmers. However, very few studies focus on grains despite

the stark differences compared to fruits and vegetables. Furthermore, a large share of the

studies on grains was conducted in Africa, mostly exploring the effects of demographic and

economic factors on market participation and its extent or sales volume. These studies are

discussed in the following sections since some of their results were used as the basis for using

some independent variables in this study.

2.2.1 Personal Factors

Personal factors typically refer to the farmers' demographics, such as age, sex, marital status,

educational achievement, and family size. These are usually included in the quantitative models

of the different studies exploring the factors affecting market participation, most of which

proved to be significant determinants. Similarly, personal factors include personality traits,

emotions, and risk tolerance (Chitra & Sreedevi, 2011; Olga & Monowar, 2015; Young,

Gudjonsson, Carter, Terry, & Morris, 2012). While personality traits and emotions are rarely

included in market participation models, a few studies show the effect of risk aversion on

farmers' market participation.

First, age was found to be a strong determinant. Heltberg and Tarp (2002) provided evidence

for this using the 1996-97 Living Standard Measurement Survey data to study the likelihood

26
Literature Review

and extent of maize farmers' market participation in Mozambique. Using a Heckman or

switching regression model to analyse their data, the results suggested that the head of the

household’s age strongly affects market participation decisions. Boughton et al. (2007) and

Alene et al. (2008) also looked into this with their research that evaluated maize farmers’

market participation. Their results showed that households with older household heads are less

likely to participate in the market or that market participation declines with age. By contrast,

Olwande et al. (2015) explored the effect of the number of prime-aged men in the household

on market participation. The results demonstrated that the more prime-aged men there are in

the household, the higher the household's chances to participate in the output market. This

means that while older farmers are less likely to participate in the market, having other prime-

aged men in the household could alter that scenario.

Moreover, empirical evidence suggests that gender influences market participation and the

volume of output sold by the household. Studying maize farmers in Mozambique using cross-

sectional data from 4900 farmers, Boughton et al. (2007) showed that male-headed homes are

more likely to participate in the market. The studies of Mottaleb et al. (2014) in Bangladesh,

Mmbando et al. (2015) in Tanzania, and Woldeyohanes et al. (2017) in Ethiopia found similar

results. By contrast, using Heckman’s sample selection model to analyse 802 maize producers

from eight districts in Western Kenya, Alene et al. (2008) found otherwise – that female-headed

households have a higher probability of market participation by 10% compared with male-

headed ones. Furthermore, Boughton et al. (2007) found that female-headed families in

Mozambique were likely to sell a lower output volume than male-headed households. This

contrasts with the results of Mottaleb et al. (2014), which demonstrated that female-headed

homes proportionately sold more rice than male-headed families, explaining that once women

farmers are engaged in the market, they sell more than males do. Also, contrary to these, Owach

27
Literature Review

et al. (2017) found that the sex of the household head has no significant effect on the product

quantity sold in the market by the household.

These differences in findings could be a result of several factors. Firstly, the participants'

selection and research areas were different; some were random, while others were purposively

selected. Secondly, the aims of the studies were dissimilar. While Alene et al. (2008) and

Mmbando et al. (2015) centred on the effect of transaction costs on market participation,

Boughton et al. (2007) focused on the impact of assets. Moreover, Woldeyohanes et al. (2017)

concentrated on the effect of off-farm income, and Owach et al. (2017) highlighted the factors

simultaneously affecting storage, consumption, and sales, under the assumption that the

farmers’ decisions are made under price uncertainty. Thirdly, their analyses differed from each

other, ranging from selection model, seemingly unrelated regressions with risk neutrality tests,

correlated random effects with a control function approach, to a two-stage Heckman selection

model. Lastly, the countries of research – Mozambique, Tanzania, Ethiopia, Kenya,

Bangladesh – could have differences in culture, especially within ethnic groups, that could

affect women's access to markets and their role in marketing. These make the data and results

hard to compare and dispute.

Education was another factor found to be a significant determinant of market participation.

Mottaleb et al. (2014) showed that the number of schooling years affects market participation

decisions and the proportion of market sales. They explained that higher education leads to

higher literacy and numeracy that help farmers obtain better information and allow them to

plan and manage their resources well, resulting in higher productivity. In addition, better

education can help the farm household acquire other income sources that could aid them in

getting and using better inputs to increase farm yield. This was similar to the findings of

Mmbando et al. (2015) and Olwande et al. (2015), which showed that education is a significant

28
Literature Review

factor in market participation. Owach et al. (2017), however, provide evidence that education

only affects the number of products sold to the market. Interestingly, Alene et al. (2008) did

not find any significant relationship between education and market participation or marketed

volume. These differences in findings could again result from sampling disparities, dissimilar

study focus, and differences in estimation models.

Moreover, many studies suggest that household size significantly affects both market

participation and the amount of marketed surplus. Sharma and Gupta (1970) showed that

family size substantially impacts the marketed surplus, explaining that families of larger sizes

consume more and thus, sell a lower volume of products than smaller families. This is similar

to the results in Mottaleb et al. (2014) and Woldeyohanes et al. (2017), which demonstrated

that farm households with more family members are less likely to participate in the market;

and if they do, they tend to sell a lesser amount. They explained that small grain farmers in

developing countries are primarily producing staples for their consumption. Thus, it only

follows that the bigger the household's family size, the less marketed surplus there would be,

and the less volume there would be to sell in the output market.

Furthermore, some research has documented the effect of risk on market participation.

According to Gabriel and Hundie (2006), the risk of postharvest grain losses is one of the most

critical factors that explain why farmers sell their harvest instantly, even at their lowest levels.

Their study on grain farmers in Ethiopia found that 93% of the farmers are worried about the

imminent risk of grain loss due to storage problems for a prolonged period. Thus, 51% of the

total marketable output for the year was sold in the first quarter. However, some studies suggest

that uncertainty in future food prices results in food storage and lower market participation

since farm households want insurance against high future costs. Park (2006) demonstrated this

in his study on the grain management of farmers under risk. Using data from 288 households

29
Literature Review

in 24 villages in Shaanxi, China, where grain production and storage were typical, he chose

parameters that would reflect the farmers' economic environment to simulate their decisions.

Results showed that subsistence farmers usually do not sell their grains in the market. Park

(2006) explained that farmers prefer to store their marketable surplus in a good year because it

is more profitable than buying grains during a bad year. Given the negative correlation of price

and yield ‒ especially in areas where markets are isolated ‒ the prices of grains are low during

the peak harvest months and high during the lean harvest period. Thus, it is less appealing for

farmers to sell during the harvest season and then buy back when their stocks run out because

the expected return to production is low. They would rather store the grains since the expected

future prices might be equivalent to the storage cost less the risk of selling at low prices and

buying back supplies at high prices, which would result in the non-maximisation of income.

Thus, farmers have a low propensity to sell surplus grains except for specific amounts, which

they need to cover some cash expenditure (Park, 2006).

2.2.2 Economic Factors

Several studies have provided evidence on the influence of economic factors on market

participation. For instance, price is a significant factor, whether cultivating staple crops or other

crops. However, the effect depends on whether the farm household is a net seller or buyer of

the staple good. Specifically, Singh et al. (1986) demonstrated that staple grain farmers who

are net sellers increase their sales when the staple grain price increases and vice versa. These

results are similar to Goetz (1992), that looked at the market participation and volume traded

in a Senegal grain market using a switching regression model. His works noted the positive

effect of an increase in the price of grains on market participation and its intensity. Omiti et al.

(2009), Woldeyohanes et al. (2017), and Mmbando et al. (2015) concluded the same for

smallholder farmers in Kenya, farmers in Ethiopia, and maize farmers in Tanzania,

30
Literature Review

respectively. The findings of Mottaleb et al. (2014) on smallholder rice farmers in Bangladesh

demonstrated the positive and significant effect of output prices on market participation,

showing that farmers predominantly respond to prices despite having small farms and

producing rice for their own consumption.

On the contrary, Boughton et al. (2007) discovered that while output prices have a strong

positive relationship with the volume of farmers’ sales, they do not affect the probability of

spot market participation, explaining that this is because of the covariates that condition both

variables operate independently. Similarly, Alene et al. (2008) did not find a relationship

between maize price and market participation. They noted that it is consistent with farmers'

practice wherein they sell immediately after harvest to meet cash needs without considering

prices. However, their findings suggested that the volume of maize sales increases as its price

increases with a 1.67% price elasticity. But this is once the decision to sell or participate in the

output market has been made.

Several studies likewise exhibit a strong association between market participation and asset

endowments. The research of Owach et al. (2017) on Ugandan grain farmers provided evidence

that households below the poverty line, who typically have fewer assets, significantly sold more

produce. On the contrary, most works show a positive association between assets and market

participation and sales volume. Some demonstrate that landholding or farmers with bigger farm

sizes are more likely to participate and sell a higher amount of harvest in the market. The same

is true for those with farm machinery (Boughton et al., 2007; Cazzuffi & McKay, 2012;

Heltberg & Tarp, 2002; Mmbando et al., 2015; Mottaleb et al., 2014). However, Alene et al.

(2008) did not find a significant relationship between livestock ownership and market

participation, only between livestock ownership and the volume of marketed sales.

Nevertheless, they showed that transportation, even just bicycles or motorcycles, has a strong

31
Literature Review

positive effect on market participation because it reduces transport costs in accessing market

information rather than product transport. By contrast, Boughton et al. (2007) found that

transportation positively affects the sales volume of Mozambique maize farmers because it can

reduce the cost of transportation to the point of sale. Overall, farmers with assets were seen to

participate better in the output market because they have lower liquidity constraints that give

them better access to inputs and resources, which help increase their productivity (Barrett,

2008; Boughton et al., 2007; Cadot et al., 2005; Cazzuffi & McKay, 2012; Nyoro et al., 1999;

Olwande et al., 2015).

Transaction cost, or the cost related to searching, bargaining, transporting, or monitoring, is

another factor that highly influences the market participation of farmers (Mottaleb et al., 2014)

since high transaction costs limit the gains that farmers get from participating in the market

(Larson & Deininger, 2001). Some researchers provided evidence that extremely high

transaction costs can result in missing markets because of the negative utility from market

participation, leaving farmers better off being self-sufficient (De Janvry et al., 1991; Sadoulet

& De Janvry, 1995). Goetz (1992) showed that fixed transaction costs prevent market

participation. He explained that transaction cost could be in terms of money or time, thus

making it a choice between labour and leisure. His data also suggest that poor market access

or other barriers such as language or ethnicity can influence transaction costs.

Key et al. (2000) and Mmbando et al. (2015) likewise note that apart from fixed transaction

costs (FTC), proportional transaction costs (PTC) affect the selling decisions of farmers, with

PTC being more critical. Using bivariate probity regression on the 2006-2007 wet season Rice-

Based Farm Household Survey (RBFHS) data, Cuevas and Clarete (2015) also provided

evidence that transaction costs affect market participation decisions. Mainly, fixed and

proportional transaction costs and their variables affect market participation, while only

32
Literature Review

proportional transaction costs and their variables affect the level of market participation. Based

on their results, irrigation, farm size, and number of years in farming positively affect the

chances of market participation since they increase productivity and, thus, lower transaction

costs due to economies of scale. However, a higher wage rate for farm labour increases the

transaction costs and, therefore, lowers the probability of market participation.

Moreover, Cuevas and Clarete (2015) show that variables such as years of education, the

municipality's income class, and access to informal credit affect both the decisions to

participate and the level of market participation. The years of schooling mitigate transaction

costs because of higher productivity and better ability to obtain market information. The

income class proxies for physical and market infrastructure and institutions; it engages farmers

to the market because of easier access. Informal credit, similarly, facilitates market

participation since its provision is usually coupled with marketing arrangements that reduce

search, information, and negotiation costs in marketing the products. They also found that

production shifters like irrigation affect market participation and its level since irrigation

decreases costs.

Moreover, many research studies suggest that household location affects market participation

via transaction costs. For example, Mmbando et al. (2015) observed that households closer to

the market are more likely to sell. Similarly, Omiti et al. (2009) demonstrated that a higher

distance from the farm to the point of sale is a constraint to market participation. Alene et al.

(2008) likewise found a decline in maize volume as the distance to the maize market increases.

Their results also showed that the probability of market participation is lower as the distance

to the fertiliser market increases since this does not facilitate achieving the production threshold

that would result in a higher marketable surplus. Notably, Woldeyohanes et al. (2017) provided

evidence that the distance to nearby markets does not affect the household's decision on

33
Literature Review

whether to sell to the market or not. Instead, their findings showed that if the first hurdle or the

initial decision is to participate (or not), then a significant relationship comes into play in such

a way that the higher the distance to the market, the lesser the quantity of outputs is sold to the

market.

Furthermore, Mottaleb et al. (2014) attested that the market's distance is positively correlated

with market participation. It is also positively correlated with the proportion of rice sold to the

market, which means that farmers living farther from the market sell more rice. This could

imply that the direct market transportation cost is not a deterring factor towards the market

participation of Bangladesh rice farmers. The researchers explained that this was because the

farmers living close to the market also sold other in-demand crops, which helped with their

finances. Hence, they sold a lower volume of grains. Overall, these dissimilarities may again

be attributed to the differences in the focus and econometric model of the studies.

Finally, some research showed that off-farm income is another factor affecting market

participation. However, empirical data and theoretical reasoning suggest that it can affect

market participation both positively and negatively. Oseni and Winters (2009) explained that

it might depend on the amount of income earned, activity types, and liquidity. On the one hand,

it can have a positive relationship with commercialisation or market participation if the farm

household uses the off-farm income, typically from higher wage or self-employment activities,

to help overcome working capital or liquidity constraints for farm investments (Woldehanna,

2000). This is because it could increase productivity and, thus, result in a higher marketable

surplus. Similarly, off-farm income could help serve as a safety net for the farmers and,

therefore, assist farmers in consumption smoothing given risks in production and the market

after commercialisation (Holden, Shiferaw, & Pender, 2004).

34
Literature Review

On the other hand, off-farm income could result in non-commercialisation if it leads to an

increase in demand for consumption goods (Oseni & Winters, 2009). If the farm household

sought off-farm income to consume more goods and not invest in farm inputs, then the off-

farm work and farm work are not complementary but rather a competition for labour and other

resources. This, in turn, lowers productivity and, thus, the marketed surplus (Alene et al., 2008;

Omiti et al., 2009; Woldeyohanes et al., 2017). However, the results of Woldeyohanes et al.

(2017) showed that an increase in off-farm income does not influence the decision whether to

participate in the output market or not; nor does it support the claim that off-farm income help

smallholder farmers overcome liquidity constraints. However, conditional on positive

participation, the findings suggest that off-farm income could result in a lower sales volume

because of the income effect as posited by Woldehanna (2000). More specifically, this happens

when the farm households use off-farm income to purchase a good that was previously bought

using the farm income, subsequently using their produce for personal consumption, resulting

in a lower marketable surplus.

2.2.3 Institutional Factors

Of the studies demonstrating the social factors that affect market participation, Mmbando et al.

(2015) provided evidence that membership in agricultural associations or cooperatives

increases the probability of market participation by 9%; Cazzuffi and McKay (2012) found that

union membership facilitates market participation because associations or cooperatives are

central in lowering fixed transaction costs and are a good platform for information exchange

and linkage to buyers; Alene et al. (2008) showed that farmers who belong to a group or

organisation supply 56% more volume than those who do not.

Moreover, Alene et al. (2008) and Cazzuffi and McKay (2012) demonstrated that access to

support services promoting input use and facilitating increased productivity improves farmers'

35
Literature Review

market participation. In particular, Alene et al. (2008) found that contacting extension service

personnel increases both the probability of market participation and the marketed volume of

output. This is because they help increase the farmers’ farm yield, thus, the marketable surplus.

By contrast, Owach et al. (2017) demonstrated that access to extension services does not

significantly affect the quantities sold by the household in the market.

There is also evidence that households with financial credit access are more likely to engage in

the market (Alene et al., 2008; Cadot et al., 2006; Cuevas & Clarete, 2015; Stephens & Barrett,

2011). Alene et al. (2008) noted that credit-worthy farmers have a 19% higher chance of

participating in the output market. This is primarily because working capital is needed in the

production and marketing of products. Credit availability affects the business capacity of

farmers and helps in the arbitrage of the market. Similarly, credit availability also results in

differences between transfer costs among crops (Larson & Deininger, 2001).

2.2.4 External Environment Factors

The external environment is another factor that directly or indirectly affects farmers’ market

participation decisions (Schermerhorn, 2011). It includes all the elements that the farmer has

no control over, which can affect decision-making. It can be further categorised into general

and task environments. The general environment includes the widely dispersed factors that

influence decisions indirectly and over time, such as economic, legal-political, sociocultural,

technological, international, and natural conditions. The task environment, in contrast, includes

factors that directly influence decision-making. This includes customers, competitors,

suppliers, and the labour market (Daft, 2008; Schermerhorn, 2011).

While not much research has delved into these, there is evidence that external environment

factors affect market participation. For example, in Uganda, violence, political instability, and

economic collapse resulted in farming for home consumption (Larson & Deininger, 2001).

36
Literature Review

Moreover, natural conditions affect market participation, as explored in the study of Mottaleb

et al. (2014), which looked at the effect of weather on the market participation and the volume

of rice sold by Bangladeshi rice farmers to the market. Their results showed that the monthly

average maximum temperature in the sub-district is not a significant determinant of market

participation. However, it was a positive determinant of the proportion or volume of rice that

the household sold to the market, suggesting that places with higher temperatures have higher

sales volume. This sounds paradoxical, yet the researchers explained that higher temperature

means higher solar radiation and productivity, resulting in a higher marketable surplus.

Mottaleb et al. (2014) also found that the total rainfall at the sub-district level affects market

participation and the proportion of rice sold to the market in a quadratic way. They explained

that this is because rice production has a water level threshold. After a certain amount of

rainfall, water becomes detrimental to the harvest, and thus, might negatively affect farm

productivity, and therefore, the market participation and the volume of marketed rice.

Furthermore, drought and floods were found to affect market participation and the volume of

marketed surplus. More specifically, floods were identified to negatively affect both market

participation and the intensity of market engagement. By contrast, drought was only found to

influence whether or not to participate in the output market (Mottaleb et al., 2014). Mottaleb et

al. (2014) explained that while floods and droughts are infamous for decreasing farm yield,

floods often cause more damage than droughts because Bangladesh is a low-lying delta, and

floods are harder to mitigate and control.

The effect of technology has also been documented by some researchers. For instance,

Olwande et al. (2015) concluded that farmers who use farming technologies, such as irrigation

equipment and hybrid seeds, are more likely to participate in the market. Alene et al. (2008)

also provided evidence that technology or input use is a significant factor in the volume of

37
Literature Review

marketed supply. This is because technology adoption leads to increased farm productivity,

which encourages a higher sales volume (Bardhan, 1970; Cazzuffi & McKay, 2012; Haessel,

1975). In fact, several studies suggest that for every increase in output, farmers are more likely

to retain a smaller proportion for consumption purposes and make a larger proportion available

for the market (Bardhan, 1970; Haessel, 1975; Heltberg & Tarp, 2002).

Finally, the paper by Reardon et al. (2014) noted a few environmental factors affecting the rice

value chain. Using information from China, India, Vietnam, and Bangladesh, they found that

massive and growing cities and heavy investments of local traders on transportation such as

boats result in the geographical lengthening of the rice value chain. By contrast, the regulations

on using licensed traders for food commerce and the increasing reach of the rice mills,

wholesale markets, and supermarkets have reduced the role of the village trader resulting in

shortened rice value chains. Similarly, they found that the development of the zones that result

in lower transaction costs also encourages direct sales by farmers to millers.

2.3 The Factors Affecting Channels of Sale

A marketing channel is an interdependent organisation that facilitates the availability of a

product or service for consumers (Stern, EL–Ansary, & Coughlan, 1996). They include

intermediaries involved in the process of delivering the paddy and milled rice from the farm to

the consumers (Mataia et al., 2018). In countries where information is asymmetrically held,

e.g., developing countries, farm households are faced with many types of traders or marketing

channels who give volatile prices and enforce different buying or paying set-ups (Kherallah et

al., 2000). Thus, the channel decision is one of the most crucial aspects of marketing since it

strongly affects all other commercialisation decisions. However, while there are several studies

on farmers' marketing channel choices for fruits and vegetables, those on grain marketing

38
Literature Review

channels are scarce. Since product characteristics primarily affect these choices, the significant

factors for non-grain products could be different from that of grains (Nwokoye, 1981).

One of the few studies that addresses the marketing channel choices of grain farmers is that of

Cazzuffi and McKay (2012), which along with market participation, explored the channels that

rice farmers use when marketing. Their data suggest that selling at the farm gate is more

convenient but gives limited marketing channel choices. Nevertheless, 41% to 62% of the rice

farm households sell via trader or enterprise, resulting in an 8% lower income than selling to

other families because of lower share in consumer prices and unfair adjustments in measuring

scales. Furthermore, their study provided evidence that households producing larger quantities

of grains are more likely to sell to a trader. This is because of their better storage facilities that

allow them to store the grains and wait for a trader without the risk of spoiling the product.

Their study likewise demonstrated that farmers with a smaller household and those with

children at primary school are more likely to sell to traders to ease their time constraints.

Similarly, wealthier farmers, who have less liquidity issues, also prefer to save time. Thus, they

sell to traders who can buy and pay for their product all at once rather than sell to individual

households that would require more than one transaction and a longer timeframe. However,

based on the findings, while the wealthier farmers are more likely to sell to traders to save on

time, this possibility decreases as the quantity of output to be sold increases. At a particular

volume, wealthier households prefer to look for better marketing channels that offer higher

prices since they can afford to transport their product and meet the other associated costs.

Moreover, their study showed that the probability of farm households selling to traders

increases as the quality of the transport network gets better or when the farm household

acquires a means of transportation. However, this possibility decreases the farther they are from

39
Literature Review

the road or with the presence of a daily market. In addition, they found evidence that older,

male-headed households and farmers who belong to ethnic minorities are more likely to sell to

traders.

Furthermore, Cazzuffi and McKay (2012) analysed farmers' decisions whether to travel or sell

at the farm gate using a two-stage probit with correction for sample selection since the distance

travelled was only available for 2006 and 2010, not for 2008. Results showed that farmers are

more likely to travel rather than sell at the farm gate when they are wealthier; male-headed;

with transportation; own a phone; belong to an ethnic minority; more isolated; or living in

communes without a daily market, where the overall quality of infrastructure is lower, or where

the distance from the community centre to the main road is higher. However, they found that

larger quantities were associated with a lower probability of selling at the farm gate for those

who sold product in 2010. Yet the opposite was true for consistent sellers or those who sold

both in 2006 and 2010. They explained that this might be because of the established

relationship with the potential buyer and their better reputation as a seller, which could increase

the bargaining power or the willingness of the buyer not just to give a higher price but also to

travel and pick up the grains at the farm gate.

Soe et al. (2015) explored the factors affecting paddy rice farmers' marketing channel choices

in Myanmar. They used primary data collected from 196 farmers from the five main rice-

producing areas of Myanmar and secondary data from its Ministry of Agriculture. The findings

of their study exhibited that only 21% of the farmers sold to millers. A total of 79% of the

respondents dealt at the farm gate, with 64.8% selling to brokers or agents and 14.3% selling

to collectors. Thus, the prices that the latter group received, excluding transportation costs,

were 4% to 6% lower than those selling directly to millers. This is even lower than Cazzuffi

and McKay (2012) found.

40
Literature Review

The paper by Reardon et al. (2014) studied the value chain of rice in China, India, Vietnam,

and Bangladesh using information from stacked field surveys. They reported that only 7% of

the paddy rice in Noagoan/Bangladesh is sold to village traders, 30% is sold to rice millers,

while the rest is sold to wholesale markets; in Heilongjiang/China, 29% is sold to village traders

while 63% is sold to rice millers. In both countries, no farmer was selling to the government.

By contrast, 40% of the paddy rice in RRD/Vietnam are sold to local traders, 12% to small rice

mills, and 36% to retailers. Moreover, 13% and 14% in Eastern UP/India and West UP/India,

respectively, were selling to the government, 18% in both were selling to village traders, 15%

versus 5% sold to millers, while 54% and 63% sold to wholesale markets. These results from

India were attributed to the regulation to use licensed traders for commerce in food. Finally, in

Jiangxi/China, 82% of the paddy rice is sold to village traders, 9% to the government, 2% to

rice millers, and 7% to the wholesale market, while 97% of the paddy rice in MRD/Vietnam is

sold to local brokers. None sold to rice millers nor the government. Reardon et al. (2014)

attributed the high sales to local brokers in Vietnam to their heavy investment in boats to collect

harvests from the farms reached by interlaced waterways while they related the Jiangxi, China

findings to the disappearance of small rice mills. The paper also highlighted the transformations

affecting the rice value chain, particularly noting the effect of marketing and finance revolution

such as increasing disintermediation, the disappearance of tied-output credit markets, vertical

integration/coordination, and geographical lengthening of the value chain; and changes in the

value chain structure such as land concentration in some zones, decrease in the share of rice

mills, reduction in the role of village traders, and the emergence of supermarkets.

Soe et al. (2015) used a multinomial logit model to analyse farmers' choices among different

marketing channels. They categorised the dependent variable into (1) selling to brokers or

commission men at the farm gate, (2) selling to collectors or traders at the farm gate, and (3)

selling directly to the trader-millers in the nearby town, which was used as the reference

41
Literature Review

category in their analysis. Based on the regression results, the distance to the selling place, the

volume output, storage facility, transportation, and road conditions affect the choice of farmers’

marketing channels. Specifically, those who live further from rice mills are more likely to sell

to agents and traders at the farm gate. By contrast, the volume of output sold has a negative

relationship with the channel of sale, which means that a lower marketable volume results in a

higher probability of marketing at the farm gate to intermediaries, such as brokers and traders.

The possession of a storage facility and transportation means are also negatively correlated to

selling to brokers or traders.

Similarly, farmers who urgently need cash and who live in towns with bad road conditions are

more likely to sell their harvest to brokers or commission men at the farm gate. Interestingly,

marketing information was found only to affect selling to traders at the farm gate, so farmers

with insufficient market information are more likely to sell to traders at the farm gate. Finally,

the study showed no significant relationships between the choice of marketing channels and

education, farming experience, off-farm income, and gender – factors which have been

established by other studies to significantly affect the decision whether to participate in the

output market and the amount of output to be sold to the market.

Two of the studies on marketing channel choices were conducted by Dessie et al. (2018, 2019).

One of their studies looked at the determinants of wheat marketing channel choices using

primary data gathered through semi-structured interviews with 163 smallholder farmers in

North Gondar Zone, Ethiopia (Dessie et al., 2018). Their other study was on an African cereal

called teff, and they used data from 154 randomly selected farmers from South Gondor Zone,

Ethiopia (Dessie et al., 2019). Both studies used a multivariate probit model to analyse the

various marketing channels given by their farmer-respondents.

42
Literature Review

Specifically, Dessie et al. (2018) analysed whether wheat farmers sell to assemblers,

wholesalers, retailers, consumer outlets, or a combination of them. The data disclosed that a

large proportion of the farmers’ produce was sold to retailers (40.49%), followed by assemblers

(39.2%), consumers (37.5%), and wholesalers (23.93%). The study found that age, education,

credit access, landholding size, off-farm income, and livestock number are significant factors

in choosing which marketing channel to use. Their results suggest a higher probability for

younger farmers to sell to assemblers because the older farmers do not trust their price.

Moreover, they showed that education or the farm household head's literacy status positively

influenced choosing wholesaler and retailer outlets by 50.2% and 44.2%, respectively. This is

because they have better skills and knowledge in agricultural marketing. By contrast, literacy

was found to have a negative effect on choosing consumer marketing channels, with a

decreased probability of 46%.

Access to credit also negatively influenced choosing assembler marketing channels, which

means that farmers with credit have a lower probability of selling to assemblers by 85.5%.

Similarly, the probability of selling to an assembler decreases for every unit increase in

livestock holding. Finally, the findings showed that a one-hectare increase in the farmer’s total

farm size increases the probability of selling to both assembler and retailer outlets by 43.3%

and 58.8%, respectively. Still, it decreases the probability of selling to consumer marketing

channels by 50%. The researchers explained that this is because farmers with large

landholdings can produce a higher volume of wheat that can be sold in bulk to assemblers and

retailers, which can command a reasonable price while avoiding the marketing costs associated

with the consumer market. They did not, however, explain the absence of significant results in

terms of wholesalers.

43
Literature Review

The other study by Dessie et al. (2019) found that many of the teff farmers in the South Gondar

Zone of Ethiopia sell to consumers (37.01%), followed by retailers (35.71%), wholesalers

(31.82%), and cooperatives (16.88%). Their findings demonstrated that seven predictors

significantly affect teff grain farmers' marketing channel choices. Specifically, their findings

suggested that every quintal increase in the quantity of teff produced results in a 17% and a

10.1% increase in the probability of choosing wholesaler and retailer marketing channels,

respectively. In addition, every year increase in the household head's age decreases the

probability of selling to the consumer market by 3.3%, while a larger family size increases the

probability of selling to it. Meanwhile, membership in cooperatives decreases the probability

of selling to wholesaler and retailer outlets by 72% and 65.6%, respectively, because

memberships in cooperatives give them access to production and market information.

Furthermore, every kilometre increase in the distance between the farmer's residence to the

nearest market resulted in a 0.8% decrease in the probability of selling to the retailer market

outlet but increased the probability of selling to the cooperative marketing channel by 0.9%.

The lagged price of teff (every birr/kg increase) was also found to increase the probability of

selling to retailer market outlets by 23.6%. The authors attributed this increase to the higher

price coming from the retail market outlet. Furthermore, the findings suggest that every hectare

increase in teff farm size results in a 47.9% lower probability of selling to the retail market but

an 84.6% higher probability of selling to the consumer market outlet. The researchers explained

that this implies that farmers get a higher price from consumer markets than from retailers

because of the fewer marketing players in the chain. It also shows that those with bigger

farmlands are more likely to participate in the consumer market because they can reduce

transaction costs through economies of scale. This, however, is contrary to their findings of the

wheat grain marketing channel choices, where a bigger farm size results in a lower probability

44
Literature Review

of selling to consumer marketing channels and an increased probability of selling to assembler

and retailer marketing outlets.

Yang and Guo (2011) examined the factors influencing the rice farmers of Jilin, China, on their

choices between grain brokers and food processing enterprises when selling. After analysing

data obtained from 258 farmers from 21 cities and counties using a logit regression model, their

research demonstrated similar results as Soe et al. (2015). In particular, it showed that having

facilities like grain storage and the distance from grain depots affect farmers’ selection of

brokers. It also confirmed that farmers consider the means of grain delivery when choosing the

channel of sale.

Finally, Nxumalo et al. (2019) explored the factors affecting farmers' marketing channel

choices in the North West Province in South Africa using a multinomial logit regression model.

They grouped the marketing channels into informal and formal markets and used the cases of

farmers who did not participate in the market as the base outcome. Specifically, their findings

using the unadjusted model, where no variables were kept constant, showed that farmers living

closer to their farms are more likely to sell to informal markets than not sell at all. By contrast,

an increase in farming experience and increase in distance to the market lessens the probability

of selling to formal markets versus non-market participation. However, under their adjusted

model, where some variables were kept constant, the data suggested that an increase in farm

experience, being female, and having a higher educational level reduced the probability of

selling to informal markets. Moreover, being male and having higher educational attainment

result in a lower probability of selling to formal markets. These results, however, were based

on the data of both maize and sunflower farmers.

A summary of all the previous works related to marketing channel choices is presented below

(Table 1).

45
Literature Review

Table 1. A summary of the literature on marketing channel choices in the grains industry.

Study Product Marketing Options Data source Econometric

focus model

Cazzuffi Rice (1) Traders/enterprises, Secondary data from 1411 Sequential

and McKay (2) Household/ households; panel data Probit model
(2012) individual buyers; and covering 2006, 2008, 2010

(1) Farmgate sale vs coming from 12 provinces in

(2) Travelling to a Vietnam

distant market

Dessie et Wheat (1) Assembler, Survey data from 163 Multivariate

al. (2018) (2) Wholesaler, smallholder farmers from 4 probit model
(3) Retailer, districts of the North Gondar
(4) Consumer; with the zone in Ethiopia
possibility of multiple
channels

Dessie et Teff (1) Cooperative, Survey data from 154 Multivariate

al. (2019) (2) Assembler, smallholder farmers in Dera probit model
(3) Retailer, district, South Gondar zone,
(4) Consumer; with the Ethiopia
possibility of multiple
channels

Nxumalo et Maize (1) Informal markets, Semi-structured Multinomial

al. (2019) and sun- (2) Formal market, questionnaire interview with Logit
flower (3) Non-participation 163 farmers from the North Regression
West Province in South Model
Africa

Soe et al. Rice (1) Brokers or Survey data from 196 Multinomial
(2015) commission men at the farmers (with secondary Logit Model
farm gate, (2) Collectors data) from 5 main rice-
or traders at the farm producing provinces in
gate, (3) Rice mills in Myanmar
nearby towns

46
Literature Review

Study Product Marketing Options Data source Econometric

focus model

Reardon et Rice Distribution of Stacked field surveys in (Not

al. (2014) farmers’ choice China, India, Vietnam, and specified)
between village Bangladesh
traders, rice millers,
government, retailers,
and wholesalers

Yang and Rice (1) Brokers, Survey data obtained from Logit
Guo (2011) (2) Food processing 258 farmers from 21 cities regression
enterprise and counties in Jilin, China

2.4 The Research Gap

In summary, most of the studies related to market participation are centred on whether farmers

are selling to the market or not, how much volume they are selling, and the factors related to

those decisions. Previous studies have shown that demographic factors such as age, gender,

education, household size, and risk aversion are significant market participation factors.

Nevertheless, the direction in which these factors affect it is inconsistent. Specifically,

economic factors such as product price, assets, transaction costs, distance to the market,

communication access, and off-farm income were proven to be significant. But just as the

personal factors, their effect varies among studies. Lastly, institutional factors such as

membership in organisations, access to production support services, and credit access were

also found to positively affect market participation, while external environment factors such as

political instability, economic collapse, non-adoption of farming technology, and unfavourable

weather affect market participation decisions negatively. Overall, while several studies

47
Literature Review

document grain farmers' market participation, their research areas, methodologies, and focus

vary, which often result in opposing findings and conclusions.

Furthermore, studies on the marketing channel choices suggest that personal factors such as

age, education, gender, and the number of children in the household attending primary school

are significant. Economic variables such as assets, off-farm income, liquidity, product volume,

transportation, distance to the market/buyer, and farm size were also important determinants.

Finally, factors such as access to credit and storage facilities, indigenous origins, and

membership in a cooperative similarly proved their effect on marketing channel choices. Yet,

the literature on the topic is still limited, and thus, there are not many studies to compare. More

importantly, they are hardly comparable because of the many substantial differences among

them, often resulting in seemingly inconsistent findings. First, some use nationally

representative random samples, while others involve just one province or state. Second, the

study areas are different, from Asian countries featuring rice (Cazzuffi & McKay, 2012; Soe,

2015; Soe et al., 2015) to African countries featuring other grains such as wheat and teff (Dessie

et al., 2018, 2019). Third, the studies' foci are diverse; while some focus on very specific

marketing channels (Dessie et al., 2018, 2019), others are very general (Yang & Guo, 2011),

hence the difficulty of comparing the results. Finally, the econometric model also differs from

one to another; while some use sequential probit models, others use multinomial logit, binary

logit, and multivariate probit models, which may cause variation among findings.

Overall, among the literature published on marketing channel choices, two studies are very

similar and are somewhat comparable. Still, the findings appear to be contradicting in some

respects. Specifically, the papers on the marketing outlet choices of teff and wheat grain

farmers conducted by Dessie et al. (2018, 2019) in Ethiopia – while very similar in

48
Literature Review

methodology, econometric model, and even led by the same group of researchers – still had

conflicting findings, especially in terms of the effect of farm size on marketing channel choices.

Given the limited studies on the marketing channel choices of grain farmers in Asia and given

that the results of the existing studies seem to depend strongly on the particular institutional

environment and the crop under consideration, it is fundamental to conduct this research.

Studying and, consequently, improving rice marketing channel choices would play a critical

role in warranting that farming is profitable and ensuring the food security of more than half

of the world’s population that depends on rice as a staple food (CGIAR, 2017; FAO, n.d.).

Given the many studies conducted on the marketing channels used by other crop farmers, it is

very timely to investigate the factors affecting the choices of grain farmers ‒ particularly rice.

This is especially important since grains are different from other agricultural products, and

farmers choose their marketing strategy based on the market and physical characteristics of the

product being sold (Larson & Deininger, 2001). These variances result in differences in how

markets work even within the same country or community. Therefore, the factors affecting the

marketing channel choices of vegetable or fruit farmers may not be the same as those of grain

farmers. This is especially true since, unlike fruits and vegetables, grains can be sold in different

forms such as fresh, dried, or milled, which affects its perishability and price. Marketing also

involves more processes and a longer chain compared to other products. Thus, there is a need

for a separate study on farmers whose main source of income is grain cultivation, especially

noting the form of the grains at sale, and the channels used since they determine the income

that farmers get from marketing.

In future studies, it is vital that the sampled farmers plant grain as their main crop. Otherwise,

they may not represent marketing grain farmers as they might only grow grains primarily for

home consumption, and thus, their marketing choices may not be representative (Benfica &

49
Literature Review

Tschirley, 2012). Similarly, the studies should focus on the grain farmers in developing

countries as there are many differences between them and those in developed countries. For

instance, while markets in developing countries have information systems that are informal and

network-based, those in developed or industrialised countries have markets where transactions

tend to be impersonal and formal institutions assist in ensuring quality and performance. Since

transaction costs are lower in formal and impersonal markets, then there is a lower marketing

margin. These differences are expected to affect marketing outcomes because they affect

transaction cost, the marketing margins, the farmer’s choice of whether to participate in the

market, how much product to sell, and what channels to use (Larson & Deininger, 2001). It

would not be ideal to use the findings of the studies on developed or industrialised countries

when creating programs or policies for developing countries. Hence, there is a need to conduct

more studies in developing countries, particularly Asia, because results from African countries

may not apply to Asian countries considering the dissimilarities in cultural and social

backgrounds.

It is also crucial to study factors other than demographic and economic factors to better

understand farmers' decisions. Arrow and Arrow (2012) posited as early as 1951 that values,

influenced by one’s standards of equity or relationships, rationalise and determine an

individual’s action. Yet still, most studies do not include factors other than socio-demographic

and socioeconomic. Thus, it is imperative to conduct studies that include personality, social,

cultural, and external environmental factors. Given the assumption of the non-separability of

farm households’ decisions in an incomplete market that is almost always the case for

developing countries, it is only appropriate that social and cultural factors be included in the

marketing channel choice model.

50
Literature Review

Social and cultural factors are better explained not just by numbers but with details from more

in-depth interviews. Thus, there is a need to analyse farmers’ marketing channel choices using

quantitative and qualitative methods to get in-depth information on what facilitates and deters

farmers from optimising their marketing income by choosing the best marketing channel.

Qualitative data not only complements quantitative results by explaining the deviation of

individuals from predicted behaviour based on an econometric model but also adds depth to

the relationships between variables. More importantly, there is a need for a qualitative

methodology to gain deeper insights into farmers' movements from one marketing channel

level to another, considering the crucial factors causing such actions. Ultimately, including a

qualitative aspect in the research can help explore the most important considerations for

farmers when choosing a marketing channel and determine what factors were left out in the

previous studies. The resulting information from this study will help improve future analyses

on marketing channels. Finally, it presents evidence to support or challenge one of the central

assumptions used in the many studies on market participation – that choices aim to pursue

one’s own welfare or self-interest and are geared towards profit maximisation. Overall, these

are the research gaps that this study attempts to fill using the conceptual framework described

in Chapter 3.

51
Conceptual Framework

Chapter 3. Research Framework

In order to address the need for a study that encompasses all the possible factors affecting the

marketing channel choices of grain farmers in Asia, particularly rice farmers in the Philippines,

this research builds its conceptual framework from several models and theories. However, just

as the other quantitative studies in this field of research, it follows one of the central

assumptions of neoclassical economic theory that an individual decides towards maximum

utility, and as such, makes rational choices given external restrictions or constraints. This

theory is most applicable in analysing market participants' decisions, particularly producers

(Zafirovski, 2000).

According to neoclassical economic models, the Homo oeconomicus decides based on a

sophisticated cost-benefit analysis that leads to an optimal decision in the pursuit of self-interest

and personal utility, given a situation defined by constraints (Marinescu, 2016). More precisely,

the rational choice theory assumes that agents calculate the result of all available choice

alternatives and probability of events, considering all available information and potential costs

and benefits, and select the best decision, pursuing self-interest. The choice is expected to be

consistent because the agent has clear preferences among the available choice alternatives,

which are complete and transitive (Marinescu, 2016; Sandri, 2009; Zafirovski, 2000).

How different factors can affect farmers’ marketing decisions, specifically the marketing

channel choice, is explained in the following sub-sections.

3.1 The Agricultural Household Model

The agricultural household model is a term commonly used to refer to households that are

jointly engaged in production and consumption (Löfgren & Robinson, 1999). It was developed

52
Conceptual Framework

by Singh, Squire, and Strauss (1986) but was elaborated further by De Janvry, Fafchamps, and

Sadoulet (1991) to explain subsistence farmers’ decision-making.

The model assumes that in an actual subsistence household, the producer, consumer, and labour

supply decisions are made simultaneously. This is because a household is both a producer and

consumer of goods such as food and labour, which are both supplied and demanded by itself.

Thus, it produces its own food and relies solely on its own labour. However, for semi-

commercial households or those who participate in the market by purchasing inputs and selling

outputs, decisions are not simultaneous but rather interrelated (De Janvry et al., 1991; Singh et

al., 1986).

On the one hand, when there is a perfect market for all products and factors, agricultural

households are indifferent between consuming their products or selling them and between

hiring labour for their farm or allocating their own time for their production activities. This is

because all prices are exogenous and tradable without transaction costs. Thus, decisions are

made sequentially since consumption/work decisions are not dependent on production

decisions, making the two separable (Singh et al., 1986). The household follows the recursive

process of first maximising its income as a producer and then allocating its income to maximise

its utility as a consumer (Sadoulet & De Janvry, 1995). On the other hand, when there is market

failure, such as when the opportunity cost of any product held by the household is not equal to

its market price, the decisions of the farm household become non-separable. These market

failures happen when food is perishable or at high price risk or when there is wage

discrimination. A market could also fail when there is a wide price margin between the selling

and buying price of products or factors (Sadoulet & De Janvry, 1995).

Non-separability occurs when the shadow price of the household’s production-consumption

good is endogenously determined by the interaction between their demand and supply, instead

53
Conceptual Framework

of being exogenously determined by the market (Löfgren & Robinson, 1999). There is non-

separability because production decisions then depend on consumption variables such as

preferences and endowments or income. More specifically, the household’s income constraint

from farm production, product sales, and non-farm earnings is used to maximise utility over a

bundle of consumption goods, both produced and bought from the market. The market value

of consumed products cannot exceed the market value of produced products less the market

value of the inputs purchased (De Janvry et al., 1991; Singh et al., 1986). Thus, farmers'

decisions are guided by prices that are endogenously determined by observed market prices

and other factors influencing transaction costs coming from input and output market

participation (Olwande et al., 2015).

Household decisions are usually non-separable in developing economies and marginal

production environments because markets are almost always incomplete. With market failures,

a farm household's maximum utility depends on its combined utility as a producer, consumer,

and worker. As a producer, the farm household aims to maximise net revenues with respect to

products and factors, given constraints. These are determined by market prices, fixed factors

including private assets and public goods, and technology. As a consumer, utility maximisation

is a function of the consumed bundle of goods subject to disposable income, tastes or

preferences, market prices, and household characteristics. Finally, as a worker, the maximum

utility is a function of income and home time or leisure, determined by the total time available,

worker characteristics, and income (Sadoulet & De Janvry, 1995). Households face fixed time

and monetary costs when scouting for the optimal marketing strategy since searching for

marketing options or information is sometimes costly in terms of money and time. Thus, labour

supply or labour-leisure choice also affects marketing decisions since the farmer has to decide

whether to use the time for work or leisure, which include relaxation or time with family

(Goetz, 1992).

54
Conceptual Framework

The non-separability of production and consumption/worker decisions is also a typical result

of shallow local markets, price risk and risk aversion, and transaction costs. A market is said

to be shallow when there is high negative covariation between the household supply and

effective market prices, such as when there is plentiful or low harvest. A shallow market is

characterised by drastic falls or increases in price that may result in households remaining self-

sufficient. This results in a thin market, market failure, and the non-separability of household

production and consumption/work decisions.

Risk and risk aversion could similarly increase the size of the price band and result in market

failure and the non-separability of decisions for a particular household: the greater the level of

risk and risk aversion, the wider the price band is, and the greater the probability of market

failure. This is because the household uses the certainty equivalent price for decision-making,

which is equal to the expected price discounted by a mark-up reflecting the degree of risk

aversion and level of risk. Sales prices are discounted negatively, while purchase prices are

discounted positively to address the risk associated with it. Thus, risk and risk aversion result

in the non-separability of decision-making.

Finally, the non-separability of decisions because of market failure due to transaction costs

happens when the distance to the market is far, infrastructure is poor, search and recruitment

costs are high (because of imperfect information and supervision and incentive costs on hired

labour), and marketing margins are excessive (such as when there is cartel or monopoly power)

(Sadoulet & De Janvry, 1995). The farmer produces outputs and decides on the amount to

consume and sell depending on the price of the crop and transaction costs, which can either be

household- or crop-and-location-specific. Household-specific characteristics such as age,

gender, education, language, or ethnicity might affect transaction costs through search cost and

negotiating or bargaining skills. Similarly, net sales volume, assets, and liquidity from non-

55
Conceptual Framework

farm earnings are crop- and location-specific characteristics that also affect transaction costs

in various ways (Barrett, 2008; Goetz, 1992).

The household model assumes that farm households participate in the market differently

because of the two types of transaction costs they face: fixed transaction costs, which are

invariant to the quantity of goods traded and proportional transaction costs, which depend on

the number of goods traded. Thus, the household first chooses whether or not to participate,

considering the fixed transaction costs, and then decide how much and whom to sell given the

proportional transaction costs (Key et al., 2000). Generally, when the household is close

enough to the market and can cover the transaction costs from its trading income with sufficient

profit margin for its products, then the household will be better off engaging in trade. However,

households in remote areas have high transaction costs, leaving the household without a

profitable margin. Thus, isolated or remote markets result in a low number of buyers and

sellers, resulting in thin markets and the non-separability of household decisions (De Janvry et

al., 1991).

In summary, the agricultural household model shows that with market failure caused by

transaction costs, shallow markets, and risk and risk aversion, farm households in developing

countries such as the Philippines have non-separable production, consumption, and worker

decisions. This is especially true for rice given its perishability when fresh and the price

volatility and wage discrimination during the peak harvest period. Thus, this research uses the

non-separable household model as the main framework in identifying and analysing the factors

affecting the rice farmers’ choice of marketing channels in the Philippines. The model infers

that a marketing choice is a function of the factors affecting a farmer's maximum utility as a

producer, consumer, and labour provider. Specifically, it postulates that the producer’s

decisions are affected by market prices, costs, and private assets (economic factors) and risk-

56
Conceptual Framework

aversion (personal factor). Consumer decisions are influenced by income and market prices

(economic factors) as well as demographic information, personality, attitude, perception, and

tastes and preferences (personal factors). Labour supply decisions are affected by time-use and

labour disutility (personal factors) and wage and income (economic factors). Therefore,

overall, the model supposes that a marketing channel choice is affected by both personal and

economic factors.

3.2 Social Capital Theory

Apart from personal and economic factors, social factors are also assumed to affect rice

farmers' marketing channel choices. This is based on the social capital theory, positing that

social relationships are essential inputs in facilitating and coordinating economic transactions

(Gelaw et al., 2016). It explains that people use their networks and membership in organisations

for financial gain, which depend on social obligations and connections. It also confirms that a

person could access resources through social capital subject to one’s personal or extended

networks, the strength of these connections, and the resources available to these connections

(Sobel, 2002).

Arrow (2000) posits that there are two aspects of social capital – those acquired with cost and

those obtained without cost. He argues that individuals make calculated decisions when

creating networks, such as when they join a club, or when they do favours, or as they make and

maintain relationships with an eye looking forward to future benefits. Thus, while they do not

make direct investments in a commodity, termed social capital, it still costs them something.

The other type is called fixed social capital, which does not involve a conscious calculation of

the cost or sacrifice and is like cultural capital.

Arrow (2000) likewise asserts that social capital can only be transferred to some extent to

another person, but not completely. This is exemplified by a storeowner whose social capital

57
Conceptual Framework

includes the reputation of the shop. When the shop is sold to a new owner, this ownership is

transferred without destroying the reputation of the store. Similarly, an individual does not

lessen or destroy their social capital when they transfer it through teaching another person.

Solow (1999) differentiated social capital from physical capital by explaining that the latter has

a return on investment that can be computed with the investment total and its depreciation rate,

while that of social capital does not have a formula. Unlike physical capital, Ostrom (1990)

believes that social capital appreciates with use and that its usage increases the stock of social

capital available for future use. It operates like gift-giving in that it creates an obligation to

accept and to reciprocate. Thus, social capital establishes a commitment to honour favours or

requests in the future.

Finally, social capital theory models social relationships as capital that result in better economic

outcomes depending on density and presence. It also states that individuals learn the benefits

of reciprocity, loyalty, trustworthiness, commitment, and reputation from past feedback (Gelaw

et al., 2016). High social capital results in reciprocity that encourages compromise and

bargaining, among others. A close and continuous social structure is needed to increase social

capital. These serve as the organisational structure and means to facilitate information

exchange, observe behavioural orthodoxies, and sanction nonconformities. Positive results

then result in commitment and trust (Granovetter, 1985).

In summary, social capital theory shows that social capital or social relationships affect

economic transactions. It suggests that social factors can influence farmers' decisions under the

non-separable household model, along with personal and economic factors. It is an essential

factor since the topic is about choosing a buyer, which can be significantly influenced by social

relationships, compared with market participation decisions that could be more dependent on

58
Conceptual Framework

economic factors. Therefore, social factors are incorporated in the framework of this study, and

consequently, in the econometric model.

3.3 Cultural Capital Theory

Cultural factors acquired through cultural capital are also expected to affect farmers' marketing

channel choices, just as social factors. Bourdieu (1986) introduced the notion of cultural

capital, explaining that while relationships or connections that are made purposefully are social

capital, networks without cost or sacrifice are cultural capital. As such, cultural capital includes

language, accent, manner, and familiarity with religious rituals.

Cultural capital has three types – embodied, objectified, and institutionalised. Embodied

cultural capital is acquired consciously and inherited passively from tradition and culture. It is

non-transferable or non-transmissible and takes time to develop. It is absorbed into an

individual’s character or thinking, making the person more receptive to similar cultural

influences (Bourdieu, 1986).

The objectified cultural capital, in contrast, comes in the form of property such as a work of art

or instruments. Unlike embodied cultural capital, it can be transmitted or transferred for

economic gains or the prestige of ownership. However, it cannot be shared with the mere

transfer of the item but rather through the knowledge of its significance. Thus, the transfer of

cultural capital can only be successful when the person getting it has the historical and

conceptual foundation of prior cultural capital (Bourdieu, 1990).

Lastly, institutionalised cultural capital is the formally recognised cultural capital of a person

through an institution. Examples of this form of cultural capital are academic or professional

qualifications. This type is primarily used for the labour market and is considered economic

59
Conceptual Framework

capital as an institution recognises it. It is also measurable and comparable both qualitatively

and quantitatively (Bourdieu, 1990).

Moreover, Throsby (1999) described cultural capital as the asset or stock of cultural value,

which is vital in the flow of goods and services. He categorised it into tangible and intangible

cultural capital – the tangible being assets such as cultural heritage, artworks, and artefacts and

the intangible being beliefs, traditions, values, and ideas that exist and identify a group of

people. He described that both types could be incorporated into economic analysis, although

intangible cultural capital has a different relationship to monetary value. He explained that

unlike tangible cultural capital with both economic values from its cultural content and physical

worth, intangible cultural capital, such as beliefs or dialects, has no economic price despite its

huge cultural worth. Therefore, the economic price of intangible cultural factors is augmented

by cultural worth.

Overall, this theory suggests that intangible or, more specifically, embodied cultural capital

such as the language, ethnicity, values, and beliefs of the farmers can affect economic

transactions and, thus, their marketing channel choices. It can hold true whether they are

deciding as a producer, consumer, or worker. Therefore, this study includes cultural factors in

analysing farmers’ marketing channel choices primarily because cultural factors such as

language or ethnicity are believed to affect transaction costs through search cost and

negotiating or bargaining skills (Barrett, 2008; Goetz, 1992).

3.4 The External Environment

The external environment is also crucial in understanding rationality (Simon, 2000). The

environment is one of the considerations of the paradigm of microeconomic theory because it

affects farmers’ options and could likewise affect their behaviour and decisions. Among these

environments are resources, technological, and institutional (Kreps, 1990). Moreover, an

60
Conceptual Framework

organisation is not different from the farm household in terms of decision-making. According

to neoclassical economics, both are considered to be business entities aiming to maximise

utility (Marinescu, 2016). As such, the organisation's external environment that is believed to

affect its decision-making directly or indirectly also affects the decision-making of the farm

household. According to Schermerhorn (2011), an organisation's operation ‒ particularly

decision-making ‒ is affected not only by internal factors but also by its external environment.

He explained that the factors in the external environment are essential in decision-making.

Environmental uncertainty or insufficient information regarding what exists and what

developments may occur in the external environment complicates the decision-making process

as it makes it difficult to understand and predict what events or developments will occur.

The external environment includes all the elements outside of the organisation that has the

potential to affect it. It can be further categorised into general and task environments. Daft

(2008) defined the general environment as the factors that are widely dispersed and influence

the organisation indirectly and over time, such as economic, legal-political, sociocultural,

technological, and international conditions. The international environment includes the factors

originating from other countries such as global competitors, customers, or suppliers that could

likewise shape technological, social, and economic trends. The technological environment

includes scientific and technological advancements such as the internet, computers, cell

phones, and networks. The economic environment is the general status of the economy of the

country where the organisation is located, which is measured mostly through purchasing

power, interest rates, and unemployment rates. The legal-political environment includes

government regulations or political activities that affect the operations and behaviour of the

business. Finally, the sociocultural environment involves the general population's

characteristics and structure, including its norms, values, geographic distribution and density,

age, and educational levels (Daft, 2008). Schermerhorn (2011) included natural conditions or

61
Conceptual Framework

the environment's nature and conditions as another factor. Overall, the external environment is

similar to the environments of optimising agents, which are defined by the international market,

technology, policy (analogous to legal-political), availability of public goods (related to the

economy), and institutions that shape the sociocultural environment (Sadoulet & De Janvry,

1995).

In contrast to the general environment, the task environment encompasses factors that directly

influence the organisation's performance and operations. In terms of the microeconomic

approach, this is the domestic market environment (Sadoulet & De Janvry, 1995). According

to Daft (2008), this includes customers, competitors, suppliers, and the labour market. The

customers are the ones who acquire the goods and services of the organisation; competitors are

the ones who offer the same goods and services; suppliers are the providers of the raw materials

that the organisation use; while the labour market represents the people who could supply

labour for the organisation (Daft, 2008).

Given these definitions, external environment factors are, thus, essential in this research model.

This is because public goods and technology, which are explicit factors affecting farmers'

decisions as producers, are part of the external environment. More obviously, the farmer's

decision as a worker is also expected to be significantly affected by the external environment

‒ particularly the labour market under the task environment.

3.5 Conceptual Framework

This research follows the conceptual framework below (Fig. 2) in identifying and analysing

the different factors affecting the marketing channel choices of rice farmers in the Philippines.

62
Conceptual Framework

Figure 2. Conceptual framework of the factors affecting the marketing channel choices.

The framework shows the use of the non-separable household model given the market

imperfections and failures in the Philippines. Thus, it uses all the factors affecting farmers'

decisions as producers, consumers, and workers. It includes economic factors or those related

to production, income, and wealth. It similarly covers personal characteristics such as

demographics, motivation, perception, and personality that affect taste, preferences, and risk

aversion. Moreover, it includes social factors such as kinship and friendship, as explained by

the social capital theory since the topic of interest can be influenced by the relationship between

the farmer and the buyer. It also enlists cultural factors such as beliefs, customs, values, and

practices described by the cultural capital theory.

Lastly, while these four groups of factors mainly affect the decision-making process of a

farming household, the organisation theory postulates that the external environment may

likewise influence it. Thus, outside the internal environment is the task environment, where the

63
Conceptual Framework

buyers, competitors, suppliers, and the labour market are included to reflect their direct

influence on the farmer’s decisions. Meanwhile, the most exterior area is the general

environment where economic, legal-political, sociocultural, technological, international

conditions, and natural environment are included as they indirectly affect the marketing

channel choices. These task and external environments, along with the internal environment,

are the primary considerations in designing this research methodology, which is described

further in the subsequent chapter.

64
Methodology

Chapter 4. Methodology
This chapter describes how the data were gathered and analysed following the conceptual

framework of this research. It has two main parts, the quantitative and qualitative sections, as

this research uses a mixed-methods approach – combining deductive and inductive

methodologies to obtain a comprehensive picture of the factors affecting the marketing channel

choices of the rice farmers in the Philippines. More specifically, it uses a sequential research

data gathering, starting with the quantitative approach to first find the significant factors based

on an econometric model. This is followed by the qualitative data collection phase, which

builds on the findings of the quantitative design. It aims to explain the quantitative results and

gain a deeper understanding of the research problem (Small, 2011). In the analysis, however,

the outcomes of both designs were used together and triangulated. Mixed methods triangulation

is applied to sensibly combine different insights on the factors affecting the marketing channel

choices and unite the traditions of economic and social thought (Downward & Mearman,

2007).

4.1 Quantitative Methodology

This research starts by using the quantitative methodology to identify the factors affecting the

marketing channel choices based on farmers' collective data. This section describes the

quantitative aspect of the research – from areas of study, sampling, data gathering procedure,

description of variables, data preparation, and various analyses.

4.1.1 Areas of Research and Sampling

It is vital that the research areas were chosen objectively and that the sampling procedure was

not biased to ensure sound research design. Thus, this study's data were gathered from the top

42 rice-producing provinces used in the Rice-based Farm Household Survey (RBFHS) of the

65
Methodology

Department of Agriculture’s Philippine Rice Research Institute (DA-PhilRice). The RBFHS

covers 2,000 to 3,200 farmers every five years, who are selected through multi-stage sampling,

wherein villages are first classified into irrigated or rainfed rice farming systems. At least two

villages per agro-ecosystem are then chosen in each province. Finally, farmers are selected

using the right coverage approach – determining a random start and selecting every third farm

household in the village.

The RBFHS sample was used in selecting respondents for this research. However, those who

did not participate in the output market during the 2016 wet season were removed from the list

in line with the topic and objectives of the current research. After ensuring that all farmers in

the compiled list had marketed at least a portion of their harvests, the population was

randomised using the Microsoft Excel randomiser. The first 300 farmers from the randomised

list who agreed to participate in the phone interview were selected and then interviewed at their

most convenient time. Only one farmer refused to be interviewed, while three other farmers

initially on the list were removed from the sample after finding out that they did not cultivate

their rice farms during one of the cropping seasons covered in the study.

A commonly applied rule to determine the adequate sample size for regression analysis is N ≥

50 + 8m, where N is the sample size and m is the number of independent variables (Green,

1991). Thus, the sample of 300 farmers is big enough to give statistically valid results even for

a model with as many as 31 independent variables. This study has only 24 independent

variables.

4.1.2 Data Gathering

Before describing how the data were gathered, it is noteworthy to state that the proposal was

first reviewed and approved by the University of Auckland Human Participants Ethics

Committee before proceeding with the research. This was to ensure that the investigation would

66
Methodology

not cause any adverse effects on the researcher and that all study participants were safe and

protected. The rigorous process was completed by submitting all the required documents

ensuring that the respondents’ consent forms for the survey and qualitative interviews were

well-documented, their identities and responses kept confidential, and their participation or

non-participation was not detrimental.

After acquiring ethics approval for the research, the quantitative data were gathered from

February to June 2018. Telephone interviews were used to ensure the researcher’s safety

considering the declaration of martial law in the Mindanao Island during the time and the

available time and resources. The telephone interview is a principal survey method and the

most widely-used survey modality in developed countries, given its advantages and

convenience. Compared with face-to-face interviews, which is considered the gold standard,

telephone interviews are more cost- and time-efficient, have greater reach, and offer

interviewer safety. It is also more versatile, allowing respondents to be more relaxed and to

disclose information freely since they are within their comfort zones and more anonymous

(Novick, 2008), which was confirmed in this study. This interview method also allowed the

researcher to interview all 300 farmers scattered across the country, even gaining access to

farmers in very remote areas and those in closed-site access because of martial law. One of the

disadvantages of telephone over face-to-face interviews is the inability to gather social cues

(Opdenakker, 2006). Nevertheless, Opdenakker (2006) suggested that voice and intonation can

be observed in lieu of social cues, making telephone interviews a good option.

Based on the RBFHS in 2011-2012, only 79% of the farmers have mobile phones. While these

data were collected six years before starting the data gathering of this research, it was safe to

assume that not all the farmers in the sample population have mobile phones. Thus, to ensure

that there was no selection bias, those without mobile phones were contacted through their

family members or neighbours. In some instances, farmers with mobile phones in the same

67
Methodology

Barangay were contacted and asked to visit the prospective respondent in the neighbourhood

and get a family member's mobile phone. Otherwise, a neighbour’s phone was used during the

interview. Thus, all the respondents were interviewed by telephone, whether it be their personal

phone or not.

The respondents were individually called to provide background about the research and to

determine whether they could market their farm outputs based on personal or collective

decisions since marketing channel choices are a vital aspect of the study. After getting a

favourable answer, their rights as prospective participants were explained before they were

asked about their willingness to serve as survey participants. After receiving their consent to

participate, they were asked for information during their last three harvests, covering the dry

and wet seasons of 2017 and the dry season of 2018. Each survey interview lasted for about an

hour. While some literature cites the need to keep telephone interviews short because of

possible distractions in their environment (Novick, 2008; Opdenakker, 2006), this was not an

issue in this research. Similarly, the lack of telephone coverage, which is another disadvantage

of the interview method (Carr & Worth, 2001), was addressed by asking the interviewee for

alternative contact numbers or calling them either when at home or at the farm, depending on

which had better network signal reception.

The farmer-respondents were asked during the survey interviews about their

sociodemographic, economic, social, cultural, and production and marketing data for three

seasons resulting in a strongly balanced panel data with 900 observations. The dataset of this

study did not have any missing values since farmers with questionable information and missing

data were contacted again. The collected data are described in more detail below.

68
Methodology

4.1.3 The Variables

The following data describe the information gathered from the farmers during the survey

interviews. These are the dependent and independent variables of the quantitative analysis of

this research.

Dependent variable. The farmers’ marketing channel is the primary variable of interest in this

study. The marketing channel system is defined by Kolter (2009) as a set of interdependent

entities or organisations involved in making a product available for use, while Lake (2007)

described it as an organised network of institutions that work to link producers with end

customers. Moreover, a channel is characterised as an institution through which goods and

services are marketed, which gives place and time utilities to consumers by charging a margin

(Crawford, 1997). The specific marketing channels in this study are as follows, with some of

the definitions adapted from the RBFHS 2011 questionnaire:

● Agent/broker – someone who finds/scouts buyers for the farmer’s paddy rice

● Assembler – someone who buys from farmers and assembles the paddy rice in large

volumes, procuring them at the farm gate; includes strikers who travel from one

province to another to buy rice in volume and bring the product back to their province

● Buying station – someone with a fixed location who buys paddy rice from farmers and

sells them again to another channel in paddy rice form

● Creditor-trader – an informal moneylender who only buys fresh paddy rice from

debtors, receiving them as payment-in-kind; usually individuals providing fast credit

but often with excessive interest rates

● Cooperative – a registered group of at least 15 members who equitably contribute the

required share capital and accept a fair share of risks and benefits of their undertakings

following universally accepted cooperative principles and practices

69
Methodology

● NFA (National Food Authority) – a government-owned and -controlled corporation

tasked to procure rice from farmers at a fair price and disburse them to consumers at a

low price

● Trader-miller – someone who owns, manages, and operates a rice mill and buys rice

from farmers to sell them to wholesale buyers, retailers, and consumers of milled rice.

● Wholesaler – someone who sells milled rice in large quantities, usually to institutions

or companies

● Retailer – someone who sells milled rice in small quantities, e.g., per kilogram or sack,

usually to consumers

● Consumer – the end buyer of milled rice; a person who primarily buys milled rice to

cook a meal

These marketing channels were further categorised into two groups. While the research initially

aimed to group the marketing channels into three – assemblers, direct rice traders, and milled

rice buyers – this proved to be not ideal because the number of farmers selling milled rice is

minimal. The independent variable was also grouped based on value-addition, resulting in

grouping farmers who sold fresh rice versus those who sold dried and milled rice. However,

the number of farmers selling rice with added value was too low, so there was insufficient

evidence to reject the null hypothesis. In the end, the binary dependent variables described

below were crafted instead:

● Indirect buyers (Y = 0) – people who serve as brokers or agents and those who buy the

paddy rice from the farmers to sell them again in paddy rice form to another channel;

includes agents, assemblers, buying stations, and creditor-traders.

● Direct buyers (Y = 1) – channels that are end purchasers of the product or those who

are at the far end of the market value chain; this category includes all the entities that

70
Methodology

process the paddy rice into milled rice such as cooperatives, trader-millers, and the

National Food Authority (NFA); it also consists of the buyers of milled rice such as the

wholesalers, retailers, and consumers.

Every actor in the market value chain charges a margin. Hence, a market value chain that

includes more players results in lower margins or share in consumer prices for the farmers.

Consequently, selling to direct buyers would result in a higher share in consumer prices

(Crawford, 1997). Given this, direct and indirect marketing channels were used as the

dependent variables since this differentiates the marketing channels based on profit. Overall,

selling to direct buyers should result in higher profit. Even if direct buyers procure the product

as fresh paddy, they should be able to offer higher prices since they have the capacity to process

them on their own, thereby cutting some players from the chain.

Independent variables This study includes twenty-four independent variables, including

personal, economic, social, cultural, and external environmental factors. These variables of

interest, their units, and their expected relationship with the binary dependent variable, direct

and indirect marketing channels, are discussed in more detail below. A summary is also

presented at the end of the section (Table 2).

Personal factors. This study's personal factors are the usual demographics, which include

individual statistics or characteristics of a population such as age, gender, marital status,

education, and family size. Nevertheless, only age, education, and the number of children in

the household are included in this study as they are the ones shown to significantly affect the

choice of marketing channels based on the results of previous studies. Gender, while commonly

used in an agricultural econometric model, was not included in this study since most of the

marketing channel choices are often a result of joint judgements, as shown during the survey

interviews. Other personality factors, such as knowledge, attitude, motivation, and

71
Methodology

expectations, have been included in modelling utilisation in other fields (Isacson & Bingefors,

2002; McElroy, Hendrickson, Townsend, & DeMarie, 2007). Similarly, some literature about

peasant farmers attributes their behaviour to their motives, particularly by their satisfaction of

needs (De Janvry et al., 1991). Thus, the econometric model of this research includes personal

characteristics such as personality traits or the inner characteristics and behaviour of a person

such as individualism, courage, and firmness; attitude or the person’s emotional feelings,

evaluation, and preferences; and motivation or the force that drives a person towards achieving

an aim or goal (Gajjar, 2013; Stávková, Stejskal, & Toufarová, 2008).

Age (Age). This continuous variable is predicted to negatively affect choosing direct buyers

because selling to them usually involves more work. The expected relationship is also based

on Dessie et al. (2019) study results on the teff grain market, which suggest that older farmers

are less likely to sell to consumer market outlets. Similarly, the study of Boughton et al. (2007)

and Alene et al. (2008) on maize farmers’ market participation provide evidence that

households with older household heads are less likely to participate in the market.

Education (Educ). It is the farmer’s years in school, in continuous form. It is likewise predicted

to have a negative effect on the choice of direct buyers since better education leads to other

sources of income, e.g., regular jobs, which could limit their time spent on farming (Mottaleb

et al., 2014). This was also included based on the study results of Dessie et al. (2018) on the

wheat grain market, where their findings suggest that farmers' literacy status results in a lower

probability of selling to consumers, even by as much as 46%.

Household children (HHchildren). This refers to the number of children in the household below

12 years old during each cropping season. It is predicted to have a negative effect on choosing

direct buyers based on the study findings of Cazzuffi and McKay (2012), which demonstrated

72
Methodology

that farmers with children at primary school are more likely to sell to traders to ease their time

constraints.

Inclination towards fast cash payment (WantsFastcash_D). This dummy variable pertains to

the farmer’s preference towards receiving full cash payment immediately upon harvest (1)

versus getting credit, partial, and check payments, and selling dried or milled rice (0). This is

categorised under personal factors because it is part of the farmer’s personality trait. It does not

reflect liquidity alone because not all farmers who prefer this option are in urgent need of

money. It is expected to have a negative effect on the choice of direct buyers ‒ particularly

NFA and rice mills as they are known for their meticulous process and prolonged payments.

Similarly, selling directly to consumers takes a longer amount of time given the rate of retail

sales or consignment arrangements.

Inclination towards free transportation services (WantsPickUp_D). This dummy variable

pertains to whether the farmers would choose buyers because they prefer free transportation or

trucking services (1) or not (0). However, it does not only reflect transportation costs; it is also

about the farmer’s preference for convenience since there are many instances when the farmer

has the capacity to deliver, yet they choose not to because they do not want to exert effort even

if they have nothing more profitable to do. This reflects their preference and, thus, their

personality. It is likewise predicted to have a negative effect on the choice of direct buyers

since many of the millers refuse to collect harvests from the field or impose a minimum volume

before they will collect them.

Money management (Moneymgt_D). This is another dummy variable that covers part of the

farmer’s personality. It aims to measure the farmers risk avoidance or whether they sell at the

farm gate because they want to avoid the postproduction risks involved when trading dried

paddy or milled rice. Risk avoidance means that farmers are risk-averse and tend to prioritise

73
Methodology

small risks (Alexander et al., 2020). Since business is involved when drying and milling rice,

the farmer’s preference whether to invest their extra money (1) or just spend/save it (0) needs

to be included. Money management, therefore, represents the farmer’s business perspective

which Rosenberg (1976) also included in his economic research studies focused on the

variables of adoption. Overall, it is expected to affect the choice of marketing channel in such

a way that those who choose to invest are predicted to sell to direct buyers. This is because

selling to direct buyers is like investing, given that it requires more capital and time before it

can result in higher profit.

Hierarchy of work motivation (HierarchyOfWorkMotivation_O). This is an ordinal variable

that pertains to the farmer’s reason for working, which is ordered according to Maslow’s

hierarchy of needs – beginning from basic needs (1), followed by psychological needs (2), and

then by self-fulfilment needs (3). It is predicted that a higher level of need would result in a

lower chance of selling to direct buyers. This is because when the farmer gets less focused on

basic needs, choosing channels with higher remuneration does not become a priority (McLeod,

2007; Poston, 2009).

Price motivation (PriceMotivation_D). This dummy variable refers to the farmer's main

motivation when choosing a marketing channel, whether it be the price (1) or non-price (0).

Price motivation was used instead of price because the farmers could not determine the exact

price difference that made them choose one buyer over another. Specifically, the variable

represents whether the farmer sold because of economic benefit (income or savings) rather

thanor other factors. Initially, this variable had five categories – price, friendship, filial

relationship, credit, or debt of gratitude. It was turned into a dummy variable given the

uncertainty and struggle of farmers in answering whether they sell because of credit, social

relationships, or cultural values. Thus, to minimise errors, instead of classifying the variable

into five categories, it was turned into a dummy. Particularly, the odds of selling to a direct

74
Methodology

buyer is predicted to be higher as the farmer chooses a channel because of the higher price.

While this variable may seem related to the work motivation described above, their bivariate

correlation does not prove to be significant, which means that non-price motivation does not

necessarily equate to a higher level of needs and vice versa.

Economic factors. These are defined as the set of characteristics that directly affect the farmer's

rice business and their disposable income or buying power. Thus, this includes the fixed income

sources, off-farm income, largest farm area, yield, frequency of harvest, and as well as the

distance to the trader-miller and the market. The irrigation mode was similarly included in the

model but only for control purposes and to prevent omitted variable bias.

Fixed Income (FixedIncome_D). This dummy variable refers to whether the farmer has fixed

sources of revenue, i.e., regular job (1) or not (0). As the farmer secures a regular job or fixed

income during the cropping season, he/she is expected to sell to indirect buyers. This is because

of less farm time that often hinders postproduction work such as drying and milling and lessens

their time available to find the buyer with the highest price. Thus, farmers with regular jobs or

fixed income sources most often settle for the fastest and easiest marketing channel, usually

agents or assemblers at the farm gate.

Off-farm income (OffFarmInc_W). This pertains to the whole household’s total income (PHP)

coming from non-rice sources such as sales of other agricultural products, salaries, and business

revenues. It is recorded in continuous form for each cropping season to reflect the farmer’s

disposable income. It also proxies for having their own capital since farmers with other income

sources can generally self-finance their farm production. Therefore, they do not need to borrow

capital that will subject them to sell to specific buyers. This variable was included instead of

the “capital source” because the latter variable has a high possibility of having simultaneity

issues with the dependent variable. This is owing to the fact that while the source of capital can

75
Methodology

affect the marketing channel decision, the latter can also affect the farmers access to credit

since some farmers get financed in return for contract selling. Overall, the household’s off-

farm income is predicted to positively affect choosing direct buyers because of higher

disposable income and less liquidity constraints that allow them to process the paddy rice and

sell them to direct buyers. This is apart from the opportunity to avoid selling to indirect buyers

because of credit. While there is no known study to date on the relationship between off-farm

income and marketing channels, empirical evidence suggests that liquidity constraints and

credit are among the major reasons that compel farmers to sell their harvest to indirect buyers.

Land ownership (Landownership_D). This is a dummy variable that pertains to the ownership

of the largest farm parcel (the area of interest) – whether the farmer owns the land (1) or not,

i.e., being just a tenant (0). It was recorded per season since the biggest farmland can change

from one cropping season to another, and it could be owned or rented by the farmer. This is

predicted to positively affect choosing a direct buyer since owning the farm results in a higher

disposable income that could be used for postproduction processes and eases the need to

borrow from creditors. It is also a proxy for liquidity constraint since it is an asset that could

be used to secure cash. In addition, land ownership provides more incentive for the farmer to

obtain the best price and for the buyer to offer their best price since they may be looking at a

long-term business relationship.

Largest farm area (LParea_W). This refers to the farmer’s largest farm parcel that is cultivated

with rice. It is measured in hectares with a continuous value. It is predicted to have a positive

relationship with choosing direct buyers because it can be equated to higher income that can

be used to dry, stock, or sell milled rice to direct buyers. Similarly, it proxies for the volume of

the product marketed by the farmer, given that the actual volume sold could be an endogenous

variable. This is because it is common for farmers in the Philippines to determine the volume

they would sell in consideration of the buyer. Hence, the marketing channel is not only

76
Methodology

dependent on the volume, but the volume is also dependent on the marketing channel. As a

proxy for volume sold, it is also predicted to positively affect choosing direct marketing

channels since a larger volume can command a higher price and encourage farm pick up from

direct buyers. Moreover, farmers with a more significant product volume are more likely to

sell processed paddy rice because of economies of scale when processing the product.

Yield (Yield). This refers to the largest farm parcel’s production per season, measured as tons

of rice per hectare presented in a continuous value. Just like the largest farm parcel, it is

predicted to have a positive relationship with selling to direct buyers. This means that the higher

the yield, the more likely farmers are to sell to direct buyers because of the possibility of selling

higher volumes that can command better prices.

Harvest frequency (HarvestFrequency). This is a continuous variable that refers to the number

of possible cropping cycles per year. This similarly reflects the water abundance of the farm or

its productivity. It is predicted to have a positive relationship with choosing direct marketing

channels, so those with more frequent cropping per year are expected to sell to direct buyers

because of their higher disposable income from more harvest times per year. Similarly, a more

frequent harvest could result in more recurrent marketing that could build a better relationship

between the farmers and the non-seasonal local buyers in the town, such as the trader-millers.

Distance to the nearest trader-miller (DMiller_W). This pertains to the distance from the farm

to the nearest rice trader-miller, estimated in kilometres. It aims to proxy for the transportation

cost when processing the harvest or when selling to trader-millers. It is expected to have an

inverse relationship with choosing direct marketing channels, so the longer the distance to the

nearest trader-miller, the lower the chances of marketing to them and other direct buyers. This

is based on studies by Soe et al. (2015) and Yang and Guo (2011), which found that those who

live closer to rice mills are less likely to sell to brokers or agents and traders at the farm gate.

77
Methodology

Distance to the nearest market (DMarket_W). This refers to the total distance from the farm to

the nearest market calculated in kilometres, which aims to proxy for the transaction cost when

canvassing for buying prices or selling to the market. It is also expected to have a negative

relationship with the choice of direct marketing channels since closer markets encourage direct

marketing to wholesalers, retailers, and consumers (Ferris et al., 2014).

Social factors. These are social aspects that affect economic transactions (Gelaw et al., 2016).

They mainly include social networks or social embeddedness. However, these are complicated

factors that are tricky to gather, especially when a farmer has to recall information older than a

year. More specifically, it is challenging to document changes in social relationships

throughout three seasons or 1.5 years, which is the time coverage of this research’s panel data.

Thus, the social factors in this study's econometric model were limited to the farmers’

memberships in organisations or associations and the number of training sessions they have

attended.

Membership in organisation (NumofOrgMem). This continuous variable accounts for the

number of the farmer's organisation – be it an association or a cooperative. This is an important

social factor since farmers' economic transactions are subject to one’s personal or extended

networks built primarily through these associations, the strength of these connections, and the

resources available to these connections (Sobel, 2002). It is expected to have a positive

relationship with choosing direct marketing channels since the social capital theory supposes

that people can use their networks, e.g., membership in organisations, to reap economic gains

such as lower transaction costs and pricing information. Specifically, it is assumed that

membership to more organisations can help them get a higher marketing income through better

access to direct marketing channels. A few studies documented the effect of membership in

agricultural associations or groups on market participation (Alene et al., 2008; Mmbando et al.,

2015; Cazzuffi & McKay, 2012). Their results showed a positive relationship since

78
Methodology

associations or cooperatives are central in lowering fixed transaction costs, and they provide a

good platform for information exchange and linkage to suitable buyers.

Training (Training_O). This ordinal variable refers to the number of training sessions the

farmers have attended since they started farming, coded as follows: (0) never, (1) 1-5 trainings,

(2) 6-10 trainings, (3) 11-15 trainings, (4) more than 15 trainings. The variable was counted as

a range rather than continuous because the farmers were not sure of the exact number. This

independent variable is expected to have a positive relationship with the choice of direct

marketing channels since training attendance widens the farmers’ networks and strengthens

their connections, which could theoretically help them access more direct marketing channels.

Likewise, this is a good proxy for farmers’ access to extension services, which is similarly

predicted to encourage selling to direct marketing channels. The studies of Alene et al. (2008)

and Cazzuffi and McKay (2012) demonstrated that access to support services promoting input

use and facilitating increased productivity improves farmers' market participation because of

the increase in marketable surplus. Thus, it is also predicted that reasonable access to extension

services positively affects choosing direct buyers given the higher marketable surplus.

Cultural factors. Culture is interpreted as beliefs, habits, dispositions, attitudes, tendencies,

social positions, and practices that are fundamental to society and expressed in a society’s

values and customs that are passed on from one generation to another (King, 2004; Throsby,

1999). Bourdieu (1986) explained that cultural capital – such as language, accent, manner,

ethnicity, and familiarity with religious rituals – is absorbed into an individual’s character or

thinking, making the person more receptive to similar cultural influences. While this study aims

to account for the effects of these cultural factors in rice farmers' decisions when it comes to

marketing channels, most of the farmers do not personally know their buyers to determine if

they share the same culture. Thus, this study could only include in the model indigenous group

membership and practice of farm rituals and tradition.

79
Methodology

Indigenous group membership (IndigenousGroup_D). This dummy variable refers to the

membership to any indigenous group in the Philippines, exhibited by appearance, language, or

clothing. It is one (1) if a family member belongs to an indigenous group. Otherwise, it is coded

as zero (0). It is expected to have a positive effect on choosing direct buyers, particularly

consumers. This is because language or ethnicity is believed to increase the cost of observing

market prices that can prevent farmers from looking for trading opportunities (Goetz, 1992),

thereby encouraging them to sell their surplus harvest to the nearby homes of their peers.

Farm rituals and traditions (FarmTraditions_D). This refers to the farmer’s practice of rituals

and traditions related to farming, such as planting during specific dates, placing ritual objects

in the farm, or chanting. This dummy variable is coded as one (1) when the farmer believes in

farm traditions. Otherwise, it is zero (0). Just as indigenous group membership, having farm

rituals and traditions is expected to have a positive correlation with selling to direct marketing

channels. This is owing to the embodied cultural factors, which have been imbibed in these

farmers for years. Thus, they are expected to follow the examples of their forefathers that

include traditional farming rituals and the tradition of selling milled rice.

External environment factors. While the internal environment is believed to affect the

decision-making of the farmers significantly, the external environment, such as the general and

task environments, also affect it. Daft (2008) defined the general environment factors as those

that influence the organisation indirectly and over time, including economic, legal-political,

sociocultural, technological, and international conditions with the natural environment

(Schermerhorn, 2011). The task environment, by contrast, includes customers, competitors,

suppliers, and the labour market. Given these definitions, this study has the cropping season,

town income class, and provincial rice production as factors under the external environment.

80
Methodology

Cropping season (DrySeason_D). There are two cropping seasons in the Philippines – the dry

and wet seasons. Specifically, this dummy variable represents whether the cropping was done

in the dry season (1) or wet season (0). Dry season cropping is expected to have a positive

effect on choosing direct marketing channels since sunny weather enables farmers to dry and

mill their harvests, which are better sold to direct buyers.

Town income class (TownIncClass_O). This ordinal variable is the income classification of the

city or municipality based on the data of the Philippine Statistics Authority (PSA), where a

higher income is coded using a higher number. It proxies for population density, infrastructure,

road quality, and frequency of market in the town. It is expected to have a positive relationship

with choosing direct buyers. Specifically, farmers in towns with a higher income class are more

likely to sell to direct buyers because of lower transaction costs resulting from good road

conditions and infrastructure and more markets and milled rice buyers because of the higher

population density. While there is no known study that documents the effect of town income

class on the choice of marketing channel, the study of Cuevas and Clarete (2015) shows that

town income class positively affects market participation decisions. They explained that it

proxies physical and market infrastructure and institutions that engage the farmers in the market

because of easier access.

Provincial rice production (RiceProdVol_W). This continuous variable refers to the volume

(tons) of rice produced in the province during the covered cropping season, based on the PSA.

It proxies for the number of fresh paddy rice buyers in the area and is expected to have a

negative effect on choosing direct marketing channels. This is mainly because there are more

buyers at the farm gate in provinces where the volume of rice production is higher, with some

buyers even coming from neighbouring areas. Thus, farmers may opt to sell to them (indirect

buyers) at the farm gate because of the convenience and the fast payment that they offer.

81
Methodology

Finally, a summary of all the independent variables is presented in the following table, along

with the dependent variable (Table 2):

Table 2. A summary of the variables in the binomial random effects logistic regression model.
Variable Code Description Measurement Expected
Sign

Dependent Variable
Indirect Buyers Agents, assemblers, buying Y=0
stations, and creditor-traders
Direct Buyers Cooperatives, NFA, trader- Y=1
millers, wholesalers, retailers,
and consumers

Independent Variables
Age The age of the farm household Years –
head
DMarket_W The distance of the farmer’s Kilometres –
biggest rice field to the main
town market
DMiller_W The distance of the farmer’s Kilometres –
biggest rice field to the nearest
trader-miller
DrySeason_D Raising the rice crops under the 0 – No (Wet season) +
condition of dry weather 1 – Yes
Educ The household head’s number of Number of years –
years in school
FarmTraditions_D The farmer’s belief in farm 0 – None +
traditions or cultural practices 1 – Yes
FixedIncome_D The farm household head’s 0 – No –
sources of fixed income 1 – Yes
HarvestFrequency The number of cropping per Number of cropping +
year
HHchildren The number of children below Number of persons +
12 years old in the household
HierarchyOfWork The farmer’s work motivation 1 – Basic needs –
Motivation_O based on Maslow’s Hierarchy of 2 – Psychological needs
Needs 3 – Self-fulfilment needs

82
Methodology

Variable Code Description Measurement Expected

Sign
IndigenousGroup_D The farmer’s membership to an 0 – No +
indigenous group 1 – Yes
Irrigation_D The condition in which the 0 – Natural (rain, pump, Control
crops were irrigated river, swamp, etc.) variable
1 – Irrigation system
Landownership_D Land ownership; a proxy for 0 – Tenant +
liquidity and possibility of a 1 – Landowner
long-term marketing
relationship
LParea_W The largest parcel of land Number of hectares +
cultivated with rice; proxy
variable for volume sold
MoneyMgt_D The farmer’s financial 0 – Not business-minded +
management style 1 – Business-minded
NumOrgMem The farmer’s membership to any Number of memberships +
organisation, association, or
cooperative
OffFarmInc_W The total income of the farm Pesos +
household from off-farm sources
during the semester; a proxy
variable for own capital
PriceMotivation_D The farmer’s motivation to 0 – No (Non-price) +
choose a buyer because of price 1 – Yes
RiceProdVol_W The volume of rice produced by Tons _
the province during the given
season; a proxy for the number
of buyers
TownIncClass_O The income class of the farmer’s 1 – 6th class municipality +
town; a proxy for road networks, 2 – 5th class municipality
consumer population, and 3 – 4th class municipality
frequency of markets
4 – 3rd class municipality
5 – 2nd class municipality
6 – 1st class municipality
7 – 6th class city
8 – 5th class city
9 – 4th class city
10 – 3rd class city
11 – 2nd class city
12 – 1st class city

83
Methodology

Variable Code Description Measurement Expected

Sign
Training_O The estimated number of 0–0 +
training/seminars/field 1 – 1-5
demonstration or any skill- 2 – 6-10
enhancement program that the
3 – 11-15
farmer has attended since he/she
started farming 4 – Above 15

WantsFastCash_D The inclination to choose buyers 0 – No (Credit/partial) –

who provide speedy payment for 1 – Yes
their products
WantsPickUp_D The inclination to choose buyers 0 – No (Delivered) –
who provide free transportation 1 – Yes
services
Yield The volume of paddy rice Tons/hectare +
harvested from the largest farm
area

4.1.4 Data Preparation

In order to perform a binary logistic regression using the variables discussed in the previous

section, the gathered information from the survey interviews were encoded first. They were

then cleaned and verified by checking the interviewer-accomplished survey form and, in some

cases, calling the farmers again for clarification or missing information, resulting in a dataset

with no missing values.

Afterwhich, the assumption of the dependent variable – that marketing channels assigned to

Y=1 offer higher profit to farmers than those assigned to Y=0 ‒ was tested. Results show that

the buying price of direct buyers for fresh paddy and dried rice are both higher by PhP0.15

(USD 0.003) at PhP17.8 (USD 0.356) and PhP20.07 (USD 0.401), respectively (Table 3).

Moreover, the overall average buying price of direct buyers is higher by PhP3.94 (USD 0.079)

at PhP21.78 (USD 0.436) compared with the indirect buyers’ PhP17.84 (USD 0.357). These

84
Methodology

results confirm the validity of this study’s assumption, hence the use of direct and indirect

buyers as the dependent variable in the quantitative model of this research.

Table 3. Comparison of the prices offered by direct and indirect marketing channels.

Dependent
Soldform N (900) Min (PhP) Max (PhP) Mean (PhP) Std. Deviation
Variable

Indirect Fresh 539 8.00 25.00 17.6195 2.67757

Buyers Fully dried 56 14.00 27.00 19.9107 2.27739
(Y=0)
Sub total 595 8.00 27.00 17.8351 2.72435

Direct Fresh 220 10.50 28.00 17.8037 2.48693

Buyers Fully dried 32 16.00 25.50 20.0703 2.29940
(Y=1)
Milled 53 26.30 44.00 39.3453 3.36832

Sub total 305 10.50 44.00 21.7848 8.51331

Meanwhile, the other assumptions for an excellent binary logistic regression model such as

dichotomous dependent variable, no outliers, no multicollinearity, and large sample size

(Tabachnick & Fidell, 2007) were ensured to be met through the following:

Dichotomous dependent variable There are different marketing channels for rice in the

Philippines. Farmers can sell to agents or brokers, creditors, buying stations, cooperatives,

trader-millers, wholesalers, retailers, or consumers. This, however, was categorised into binary

dependent variables based on their buying price -- direct and indirect buyers. On the one hand,

the indirect buyers are the people who serve as brokers or agents and those who buy the paddy

rice from the farmers but sell them again in paddy rice form, hence the lower buying price from

them. It includes agents, assemblers, buying stations, and non-miller creditors. On the other

hand, the direct buyer category comprises all the entities that have the capacity to dry or mill

paddy rice on their own, thus the higher buying price. This category thus consists of NFA

offices, cooperatives, trader-millers, wholesalers, retailers, and consumers.

85
Methodology

No outliers. While there are no assumptions about the distributions of the explanatory variables

for logistic regression models, it is assumed that there are no outliers in the data. Thus, the 11

independent variables of interest in this study that are continuous – age, education, children in

the household, off-farm income, largest farm area, yield, distance to the trader-miller, distance

to the market, harvest frequency, number of organisations, and the rice production volume of

the province ‒ were checked for outliers using the SPSS boxplot and histogram. Outliers were

detected among five variables given the spread of the farmers’ characteristics and the

susceptibility of the mean and variance to outliers. These variables were the off-farm income,

largest farm area, distance to the trader-miller, distance to the market, and the province's rice

production volume.

In order to address this issue, the data that were an abnormal distance from the majority were

winsorised. The winsorisation technique was named after Charles Winsor but was introduced

by W.J. Dixon. Its primary goal is to minimise outliers' effect in the data primarily by changing

the value of the outliers into something that is close to other values or by assigning a lower

weight to the outliers (Dixon, 1960; Ruppert, 2006). The winsorisation technique is used to

remove extreme data without eliminating some observations and thus, improve the robustness

or efficiency of statistical inferences (Ruppert, 2006). For this research, less than 5% of the

data points or four of the five concerned continuous variables were winsorised by changing the

values of the extreme data at the threshold. Specifically, 3.44% of the off-farm income data

were winsorised, 4.33% of the farm area data, 1.33% of the distance to the trader-miller data,

and 1% of the market distance data. Only the data on the rice produced in the province had a

higher rate of winsorisation, which was at 6.44%.

To identify the winsorised variables easier, their codes include “W” at the end. The exact

optimal winsorising cut-off points were chosen since the underlying distribution was known

and the sample was sufficiently large. The precise cut-off points resulted in a better estimation,

86
Methodology

as shown by the mean squared errors of the modified data set, which is lower than that of the

unmodified (Kokic & Bell, 1994).

No multicollinearity. Another assumption underlying logistic regression analysis is non-

multicollinearity – requiring that the independent variables are not highly correlated with each

other. A model with multicollinearity issues will yield consistent results but with inflated

standard errors. Thus, while multicollinearity does not reduce the model's overall reliability, it

leads to “less accurately” measured coefficients because it affects the standard errors of the

estimates for each independent variable (Mayers, 1990).

A correlation coefficient resulting from the bivariate correlation that is higher than 0.7 or 0.8

suggests a severe multicollinearity problem (Mayers, 1990). Similarly, a variance inflation

factor (VIF) with values exceeding ten is considered to have a multicollinearity problem

(Wooldridge, 2012). However, Allison (2012) explained that in logistic regression models,

which are usually weaker, values above 2.5 could begin to severely distort model estimation

and subsequent predictions.

Using these as guides in the bivariate correlation and VIF analyses, results show that age and

years in farming, and total farm area and largest farm size are multicollinear. Thus, the years

in farming and total farm area were removed from the model. Moreover, the analysis displays

a robust significant correlation (0.785) between the largest farm size and the volume sold, with

likewise high VIFs (Appendices 1 & 2). Thus, one was eliminated. While volume sold is

usually included in related studies, this was removed in favour of the largest farm size because

of possible endogeneity issues with the dependent variable.

Large sample size. A large sample size is highly recommended when doing binomial logistic

regression as there needs to be a sufficient number of responses for each category. This is

because the logit estimator is consistent (asymptotically “true”) but not unbiased (“true” in

87
Methodology

expectation) and therefore needs a large sample. Similarly, the more independent variables

there are, the bigger the sample size is required. While the sample size calculation for logistic

regression is a complex problem, several studies advocate ways to compute for a statistically

sound sample size. A commonly applied rule to determine the adequate sample size for

regression analysis is the formula given by Green (1991), where the sample size should be at

least equal to N ≥ 50 + 8m, where N is the sample size and m is the number of independent

variables. Following that rule, having 24 independent variables, this research should have at

least 242 respondents. In a more strict sense, Peduzzi et al. (1996) provide the following

guideline: N = 10*k/p, where k is the number of independent variables and p is the percentage

of the positive or negative cases in the population. Applying that to this research ‒ with 24

independent variables and 33.89% positive cases ‒ the required sample size is only 708. Thus,

this research data set with 900 observations is large enough to have a sound binary logistic

regression model.

4.1.5 Data Analysis

After ensuring that the data were well-prepared and the assumptions of the analyses had been

satisfactorily met, the following quantitative analyses were conducted to fulfil some of the

objectives of this research:

4.1.5.1 Descriptive Statistics

Descriptive statistics were used to satisfy the first objective of this research – describe the

typology and the farmers' marketing practices. The resulting data are summarised in tables,

graphs, and other numerical forms that are mostly found in Chapter 5 to facilitate the

population's description in terms of their different characteristics and their marketing practices

and choices. Measures of central tendency such as mean, mode, and measures of variability

such as standard deviation, minimum, maximum, and frequency were obtained using SPSS.

88
Methodology

4.1.5.2 Econometric Model

The factors affecting the rice farmers' marketing channel choices, which pertain to the second

objective of this research, were determined using the random-effects binomial logistic

regression for panel data.

Binomial logistic regression, instead of multinomial, was used to analyse the data of this study,

given the not very robust sample size. Other studies, such as that of Dessie et al. (2018), used

a multivariate probit model because the farmers in their research had multiple marketing

channel choices. In this research, however, only the primary buyer of the produce was

documented. This was to avoid the risk of getting inaccurate data from asking farmers to recall

their previous marketing channel choices and the volume they sold to each of them. Conditional

logit was not used either since the goal was to analyse the problem using the individual as the

unit and explain the dependent variable using that individual’s characteristics. Thus, logistic

regression was more appropriate than the conditional logit, which focuses on the set of

alternatives and their characteristics. In addition, logit was used instead of the linear probability

model (LPM) since the slopes of the independent variables are not assumed to be constant

(Wooldridge, 2012).

Furthermore, a random-effects model was used instead of one with fixed effects. One of the

underlying reasons is that there was a minimal change over time within the dependent variable

– the farmers' marketing channel. Even the independent variables had minimal within-person

variation between the different leg of the data sets. Given this, only 26 of the 300 farmers in

each of the three data sets, resulting in just 78 observations, would be analysed if a fixed-effects

model was used. Thus, using it for this study would result in imprecise estimates and huge

standard errors that are hard to tolerate. Considering that the variables vary more between

farmers than within farmers, and given that the fixed-effects model is most useful when there

89
Methodology

are substantial within-person changes, this study employed a random-effects model instead

(Allison, 2009; Hilbe, 2009).

Moreover, while the fixed-effects model controls for unobserved variables, it only does for

those that are the same over time, not for those that change over time. If this model has

unobserved variables, chances are, it would be a social factor, which also usually change over

time, e.g., over a cropping season. As such, using the fixed-effects method would likewise

result in biased estimates, and thus, would have no advantage over the random-effects model

(Allison, 2009). Therefore, overall, the random-effects model is best for this study because it

results in the smallest standards errors, uses all the data, considers between-person differences,

and estimates the effects of even the time-invariant variables (Allison, 2009).

The random-effects logistic regression for binary response is given by the following formula

(Hilbe, 2009): 𝑦!" = 𝛽# + 𝛽! 𝑥!" + 𝜎! + 𝜀!"

Where

𝑦 = binary response

i = individual farmer

t = harvest season of i

𝑥 = 𝑣𝑒𝑐𝑡𝑜𝑟 𝑜𝑓 𝑖𝑛𝑑𝑒𝑝𝑒𝑛𝑑𝑒𝑛𝑡 𝑣𝑎𝑟𝑖𝑎𝑏𝑙𝑒𝑠

𝛽 = 𝑣𝑒𝑐𝑡𝑜𝑟 𝑜𝑓 𝑟𝑒𝑔𝑟𝑒𝑠𝑠𝑖𝑜𝑛 𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡𝑠

𝜎! = random covariate having a known probability distribution, Gaussian

𝜀!" = error term

The following equation gives the probability or odds ratio:

𝑃$
log E I = log[𝑃$ ] − log[1 − 𝑃$ ] = 𝛽# + 𝛽! 𝑥!" + 𝜎! + 𝜀!"
1 − 𝑃$

%&' (* +*" ,"# +-" + ."# )

!
Where 𝑃$ = $ + %&' (*
! +*" ,"# +-" + ."# )

90
Methodology

Having chosen the best model to determine the factors affecting the rice farmers' marketing

channel choices, a strongly balanced panel data with 300 groups and 900 observations were

prepared. These data were analysed using the random-effects logistic regression in STATA,

with a non-adaptive integration method to get consistent results over all the permutations of

the independent variables.

4.1.5.3 Movement Analysis

Apart from describing the farmers' typology and enumerating the factors affecting their

marketing channel choices, this research also aims to discover the specific factors that result in

the switch into a different buyer or marketing channel level. It is to precisely determine the

factors facilitating their movement downwards or upwards in the market value chain. This was

analysed by following their position within the market value chain for three seasons, thereby

generating two consecutive movements and one cross-season movement. Specifically, it was

noted whether the farmers cut the market value chain by moving upward or selling to a more

direct buyer, e.g., trader-miller instead of an agent; whether they lengthened it by moving down

the value chain or selling to a less direct buyer, e.g., selling to a trader-miller instead of a

retailer; or whether they remained within the same channel level for three seasons.

In order to do this while making the data as robust as possible, the marketing channels were

grouped into four, according to their level in the market value chain. This was preferred over

just using the direct and indirect categories. Given that selling milled rice to consumers,

retailers, or wholesalers is the goal to maximise income, direct buyers of milled rice were

assigned to level 4, which is the highest level. Selling to processors, such as trader-millers,

cooperatives, and the NFA, was assigned to level 3 since they are direct buyers of paddy rice

who process them into dried or milled rice. Level 2, by contrast, are the indirect marketing

channels because they sell whatever they buy from the farmers in the same form to the

91
Methodology

processors or the trader-millers, who are in level 3. This category thus includes buying stations

and assemblers. Finally, agents or brokers were assigned to level 1 because they sell to

whomever they can make a good deal with, even to assemblers and buying stations that are

already indirect buyers. The movements are then analysed using the following:

𝑀0 = 𝑀1 − 𝑀$

Where

𝑀0 = Marketing channel movement

𝑀$ = Marketing channel used during the later season

𝑀1 = Marketing channel used during the more recent season

Both the movement from cropping seasons one to two and seasons two to three were analysed.

Likewise, the marketing channel choices of farmers between the dry seasons of 2017 and 2018

were compared. A resulting positive value means a movement upwards in the value chain,

which means that the farmers cut the market value chain by selling to more direct buyers. A

negative value means the opposite. By contrast, a zero value implies a movement within the

same level but does not necessarily mean selling to the same buyer. All these movements were

accounted for using descriptive statistics.

Overall, the results of these quantitative analyses are described in the next chapters, together

with the information gathered through the qualitative inquiry to ensure more robust discussions.

4.2 Qualitative Methodology

In order to complement and augment the findings of the quantitative data analysis described

above and get in-depth information on the marketing channel choices of the rice farmers in the

Philippines, qualitative data were gathered and analysed as follows:

92
Methodology

4.2.1 Study Population and Sampling

Given the goal to get in-depth qualitative information to complement the quantitative results

of this study, the farmers who provided the data for the qualitative interviews were purposively

selected from the 300 rice farmers surveyed for this research's quantitative aspect. They were

selected on a case by case basis, including the marketing channels they choose, their primary

reason for selecting such channels, their marketing experience, their socio-economic

characteristics, and the marketing environments in which they operate, among others. They are

widely distributed around the top rice-producing provinces of the Philippines, coming from 34

cities/municipalities within 22 provinces. Thus, while they were chosen purposively, they can

still be considered good representatives of the rice farming population given their geographical

location, economic profile, sociodemographic features, and marketing practices.

Overall, 45 farmers were selected purposively for the qualitative interviews from the pool of

the 300 surveyed farmers. Specifically, at least 15 were chosen from Luzon, 15 from the

Visayas, and 15 from Mindanao. While most qualitative researchers would argue that the point

of data saturation will depend on the particular research topic and the complexity of the local

situation, this research will adopt the sampling suggestion of Guest, Bunce, and Johnson

(2006), who concluded from their study that data saturation could be achieved after 12

interviews, and Francis et al. (2010) who suggested the 10+3 criterion. Nevertheless, to ensure

that the topic is well-covered and will reach data saturation so as not to miss important

information, 15 interviews were targeted for each of the three major islands, resulting in a total

of 45 interviews.

4.2.2 Data Collection Methods

The selected farmers were initially contacted to ask for their willingness to participate before

interview travel plans were arranged. Given the farmers’ widespread locations, the use of the

93
Methodology

phone in this particular phase was efficient since it allowed wider reach and was more time-

and cost-efficient than other communication methods (Novick, 2008). The actual interviews,

however, were mostly face-to-face, except for five respondents from Mindanao. Face-to-face

interviews are the gold standard for qualitative interviews (Novick, 2008). However, due to

unavoidable circumstances – the high risks inherent to the areas because of terrorist attacks –

telephone interviews had to be conducted for some farmers located in a few regions in

Mindanao to ensure a wide variety of data without compromising the researcher’s safety.

Besides, telephone interviews also present many advantages, including their ability to make

respondents feel more relaxed and disclose more information because of the feeling of

anonymity (Novick, 2008), which was confirmed when the information that farmers shared

during the telephone and face-to-face interviews were compared. Many farmers shared more

information during the survey phone interview. The shortcomings of telephone interviews for

social cues were likewise addressed by observing the interviewees' voices and intonation.

While these are not as rich as the social cues during face-to-face interviews, they are sufficient

to make telephone interviews a suitable alternative (Opdenakker, 2006).

The 15 pre-identified farmers in each major island were first ranked based on their cases. They

were called in that order to explain the research and the rights of a respondent. They were then

asked whether they would like to be interviewed again about their marketing practices. Those

who agreed to participate in the research were then scheduled for a visit to conduct face-to-

face interviews.

The farmers were visited through the assistance of the different offices of the Department of

Agriculture (DA) and DA-Philippine Rice Research Institute (DA-PhilRice). All interviews

were conducted in Filipino or in the local language to facilitate a smooth conversation flow and

allow the interviewees to express their answers well. Research assistants who served as

translators were also provided by DA and DA-PhilRice since some farmers prefer to speak

94
Methodology

their local dialect because they could not freely express themselves in Filipino or English.

These translators work for government offices related to agriculture and are familiar with the

local agricultural setup and farmers’ practices. However, while they could share additional

information before and after the interview, they were briefed that they should strictly translate

what the farmer was saying without interpreting or summarising anything during the interview.

The researcher conducted all interviews to ensure the quality of data. They were semi-

structured to ensure that the key questions were answered while allowing the respondents to

elaborate responses and the interviewer to probe further on whatever interesting factor the

farmer would mention. Specifically, respondents were asked to describe their usual rice

marketing practices and their marketing channel choices. They were asked about the factors

affecting their choice of marketing channels, focusing on personality, social, cultural, and

external environmental influences. Given the difficulty of getting these data, questions were

not directly asked, but data were instead inferred based on their stories and answers.

Nevertheless, the extrapolated data were confirmed through data triangulation to assess

consistency, specifically asking similar questions approximately three times but in different

instances and ways. Each qualitative interview lasted for about 1 to 2 hours and was

documented using field notes and audio recordings; these were consented to prior to the

interview schedule being set.

4.2.3 Data Preparation

All the recorded interviews were transcribed in Filipino, including those conducted in the local

dialect since these were translated during the interviews. Suh, Kagan, and Strumpf (2009) hold

that translation can be done before, during, or after analysis but recommend it is done during

analysis to ensure the authenticity of the data. Keeping the interview transcriptions in their

original language avoids the potential limitations of using a secondary language and minimises

95
Methodology

the possibility of losing some of the interviewee’s unspoken expression (Larkin, de Casterlé,

& Schotsmans, 2007; Nes, Abma, Jonsson, & Deeg, 2010). The qualitative analysis that

involves codes, categories, themes, and quotes was done in English to facilitate translation and

for publication purposes (Nes et al., 2010).

4.2.4 Data Analysis

The files were imported into the NVivo Software after the preparation, transcription, review,

and familiarisation with the data. Coding commenced following Braun and Clarke’s (2006)

analytical framework. Coding is vital in organising data systematically and in a meaningful

way and developing categories and themes out of larger textual datasets (Braun & Clarke, 2006;

Maguire & Delahunt, 2017). Overall, the codes used were not pre-set but instead developed

and modified as the coding progressed – reorganising and renaming them depending on the

data trend. The coding was also both concept-driven and data-driven since both inductive and

theoretical thematic analyses were used to examine the interview data (Clarke & Braun, 2013;

Gibbs, 2008). For instance, initially, all data concerning credit were coded under “credit”.

However, given that most of the data came from comparing credit from price or sociocultural

factors, credit was divided into either “credit over price” and vice versa, and “credit over

sociocultural” and vice versa, since this made the ranking more obvious. Eventually, codes for

credit contracts were explicitly created to refer to credit with the obligation to sell. The non-

obligatory credit was also divided into two motives – selling because of credit due to the socio-

cultural values of dishonour and debt of gratitude, which were mostly correlated with credit

with interest and without interest, respectively.

The thematic analysis was partially theoretical as the themes were theoretical and highly related

to the questions asked during the interview. The study was thereby driven by the theory and

the researcher’s analytic interest (cf. Braun & Clarke, 2006). In the same manner, the thematic

96
Methodology

analysis was inductive because the data were not limited to pre-existing coding frames or the

researcher’s critical predispositions. The research questions were open, thereby allowing it to

evolve through the coding process – permitting anything beyond the theoretical coding frame

or the themes that previous related studies have identified to be analysed as well (Braun &

Clarke, 2006). However, only the relevant ideas or answers to the main research question were

coded, following the theoretical thematic analysis; this contrasts with the usual line-by-line

coding done for inductive thematic analysis (Braun & Clarke, 2006).

Furthermore, this research used thematic analysis at the latent level. Between semantic/explicit

or latent/interpretative levels, latent analysis seemed more relevant, especially for social and

cultural factors, given that farmers gave very complicated answers and found it difficult to

pinpoint the factors of their decisions. Thus, themes were identified by looking beyond the

interviewees’ responses and examining the underlying factors for their marketing decisions,

which involved interpretive work and theorised analysis. Moreover, this research used both the

essentialist/realist and constructionist perspectives for thematic analysis as research

epistemology. While the essentialist/realist approach focuses on motivation or individual

psychologies, the constructionist approach concentrates on the structural conditions and socio-

cultural contexts that lead to the interviewees' meaning, experience, and perspectives (Braun

& Clarke, 2006; Burr, 1995). Thus, with the aim of this research in mind – exploring the factors

affecting rice farmers' decisions when choosing their marketing channels – it was essential to

use both of the epistemological approaches as guides when interpreting the data and theorising

them.

Overall, themes or patterns within the farmers’ answers were identified as a whole. These were

then subjected to second-order analysis to find recurrent factors, themes, patterns, and clusters

resulting in category development and meaning-discovery for farmers’ market participation

channels. Results were then visualised using network diagrams and maps to demonstrate the

97
Methodology

relationships between the different codes and themes and visualise the causality of choices or

decisions with these factors. All these were done using the NVivo software.

4.2.5 Research Positionality

Given the importance of conveying the researcher’s positionality relative to the research stance,

the following provides my positionality.

I am an external-insider from the research group in focus. I was born in the Philippines' rice

granary and have been with the Philippine Rice Research Institute (DA-PhilRice) for 16 years,

directly working with farmers to interview them about rice farming since I finished my

bachelor’s degree. Thus, while I am considered external from the group because I do not farm

myself, I can be regarded as an insider since my family owns rice farms, and I have more than

a decade of work experience at DA-PhilRice. My work affiliation has also given me good

access to the respondents. It has likewise provided me with good knowledge on the topic and

the situation of the farmers. Despite that, however, I have maintained a goal to co-construct

knowledge between myself and the participants by letting them share their thoughts and receive

information objectively while verifying their accuracy.

I am in favour of cutting the marketing channel chain because I believe that farmers will be

better off if they can eliminate some players and sell their produce directly to the consumers.

This is based on the empirical evidence of my previous project on the market matching of

unpolished or brown rice, wherein I linked the producers directly to hotels and restaurants and

tried to have smaller mills installed in certain towns to help the farmers sell milled rice instead

of fresh paddy rice. Nevertheless, some farmers expressed their hesitations and limitations with

regard to direct marketing. This is contrary to the profit maximisation theory and my nature to

maximise utility – either in terms of money, time, or effort. While I believe that not all people

are geared towards profit maximisation like me, I don’t know and understand the factors

98
Methodology

influencing those choices. Moreover, I want to find out if those decisions will lean towards

utility-maximising options once the constraints are minimised, if not removed.

Finally, I am aware of the enormous disparity between the farmers’ income from that of the

traders. Witnessing how poor most of the farmers are when they are the ones who worked hard

to cultivate the rice, in contrast with how rich the trader-millers are and how easy income is for

agents, I felt the need to help them improve their marketing income. However, I need to find

out the marketing issues or constraints influencing their choices, which would serve as a sound

basis to help make the project successful, hence this research. Overall, my objective is to

understand the factors affecting the marketing channel choices of the rice farmers in the

Philippines in order to help them effectively; it is not to prove whether my theories or ideas are

right. Therefore, I did not let my ideology or biases affect my gathering and interpretation of

the results; I remained as objective as possible throughout the whole study to make it a sound

and effective basis when creating projects that can inform policies.

4.3 Limitations of the Study

Despite all efforts to make this thesis as sound and rigorous as possible, it still has its

limitations. These are presented in this section, along with how they were addressed.

First, this research mainly used the process of recall to gather past data. This is primarily

because very few farmers in the Philippines practise record-keeping. Often, recall data are

subject to potential biases such as recollection issues that might result in overestimation or

underestimation (World Bank, 2006). However, for this study, recall is believed to be relatively

reliable since the latest data asked of them were three cropping seasons ago, or just more than

a year old during the interview. Moreover, overestimation and underestimation were dealt with

by allowing the respondents to answer in range, where the median was used. Distortion of

socially desirable or undesirable behaviour is also said to be one of the problems of having

99
Methodology

recall data (World Bank, 2006). This was dealt with by conducting telephone interviews to

minimise the possibility of social desirability bias since they somehow feel anonymous

(Novick, 2008). Finally, recall data are subject to how questions are formulated (World Bank,

2006). Thus, an ethnographic interviewing style was used to help the respondents reimagine

the past or go back to the harvest times of interest with better accuracy. Similarly, questions

were asked more than twice or thrice but phrased differently to triangulate information and

countercheck for inconsistencies, ensuring that they do not reinvent the past. Finally, to

minimise recall bias due to the differences in questioning, the researcher interviewed all 300

respondents, allowing for uniform treatment of the questions and comparable, if not similar,

probing styles.

Second, some important variables were not included in the quantitative model because they

were hard to quantify owing to the unreliability of the available information, both from primary

and secondary data. For instance, it was difficult to account for the number of buyers in a

locality during a particular harvest season, given that the assemblers can just come and go

without records. Moreover, the local traders are not registered, nor are the agents. The same is

true for labour rate and availability since it is an informal labour market, and people can offer

their labour from time to time at varying rates, depending on the demand. The suppliers per

season are also undocumented since they sometimes do not register their business and trade

paddy rice without a permit. Moreover, accounting for the number of farmers selling during a

cropping period is also problematic since the farmers’ master list is unreliable and outdated.

Likewise, it is common for a farmer to stop cultivating their farm for one season because they

used it as loan collateral. Meanwhile, if secondary data are unreliable, if not non-existent, using

farmers’ recall for these variables would result in more unrealistic and inaccurate data.

Third, this study was limited to naming and explaining the different factors affecting rice

farmers' marketing channel choices, but not the process of decision-making. As such, personal

100
Methodology

traits – while very complicated subjects – were confined to the farmers’ hierarchy of work

motivation, their motivation when choosing a buyer, their risk avoidance, and their inclination

to choose buyers who provide free trucking services and fast payment.

Lastly, the researcher gathered the data as a student to get the most objective answers from the

respondents. However, some farmers accidentally found out the researcher’s affiliation with

the Department of Agriculture and PhilRice, which was beyond her control. In such cases, they

were interviewed with the reassurance that it was purely research-motivated and that the

researcher was on study leave that made her independent from the institutions. Thus, any

information they share will be kept confidential and will not be used to disadvantage them in

the future.

4.4 Research Process Summary

This research involves a sequential methodology, starting with the quantitative and then the

qualitative aspect. The quantitative data were gathered through telephone interviews from 300

randomly selected farmers from the database of DA-PhilRice. These data refer to their

sociodemographic characteristics, along with their production and marketing practices during

the last three cropping seasons. The factors affecting the choice of direct versus indirect

marketing channels were analysed using a random-effects logistic regression in STATA, with

a non-adaptive integration method. By contrast, the qualitative data were collected, mostly

through face-to-face interviews, from 45 purposively selected farmers from the pool of the 300

surveyed farmers. These interviews were transcribed, coded, and analysed with the aid of the

NVIVO software.

The following diagram (Fig. 3) summarises the research process that resulted in the data

presented in the next three chapters, along with the timeline of writing this whole thesis.

101
Methodology

Figure 3. Summary of the research process and its timeline.

102
Farmers’ Typology and their Marketing Practices

Chapter 5. Farmers’ Typology and their Marketing Practices

This chapter is the first of three chapters on findings. It aims to present the typology of the

farmers to help better understand their marketing channel choices. The first section shows the

spread or the general characteristics of the farmers interviewed in this research, while the

second part presents the farmers’ different marketing channel choices and describes their

marketing practices using both quantitative and qualitative data.

5.1 Farmers’ Typology

This section includes descriptive statistics and frequency data tabulated in Table 4 and

Appendix 5 to provide the typology of the farmers interviewed for this study. It is divided and

arranged into personal, economic, social, cultural, and external environment factors.

Table 4. Descriptive Statistics of the continuous variables of the whole population.

Variables Range Min Max Mean Std. Deviation

Age 60 23 83 55.54 10.727

Educ 17 0 17 8.99 3.232

HHchildren 8 0 8 .95 1.295

OffFarmInc_W 390000 0 390000 80269.40 88916.160

LParea_W 4.36 .12 4.48 1.3850 .95531

Yield 10.80 .40 11.20 4.3205 1.74236

DMiller_W 30.90 .00 30.90 6.3842 6.63014

DMarket_W 21.85 .05 21.90 5.2721 4.43021

HarvestFrequency 1.5 1.5 3.0 2.070 .2680

NumofOrgMem 5 0 5 .97 .840

RiceProdVol_W (tons/s) 668486 31514 700000 212346.18 191169.191

103
Farmers’ Typology and their Marketing Practices

5.1.1 Personal Characteristics

The farmers interviewed for this research were aged 23 to 83 years old, with 84.23% of them

being 40 to 69, giving a mean age of 56. Men farmers account for 91% of the interviewees,

which could be explained by the hard labour involved in farming (PhilRice, 2013). Out of the

9% population of women farmers, 59% were widows, while 11% were single or unmarried.

Overall, 84% of all the farmers were married, while 9.7% were widowed, 3.7% were single,

and 2.7% were separated (Appendix 5).

The farmers have been farming rice for an average of 32 years, with one-third (33.3%) of them

having farm experience of 31-40 years. The lowest farming experience was less than one year,

while the highest was 73 years. A quarter of the farmers sampled (25.3%) finished elementary

or primary education, while 7.33% finished a bachelor’s degree. Their mean education length

was nearly nine years; one had barely any formal schooling while another had as many as 17

years. Results of the analysis also show that the farmers had an average household size of five

members, 2.22% of them are alone, and 0.33% had 18 people living in the same abode.

Furthermore, the average number of children who were 12 years or younger was one per

household, with 0.33% having as many as eight children. About half (52.2%) did not live with

any children.

For a great majority of the cases (92.22%), farmers were inclined to receive fast full cash

payments when choosing a buyer, and they preferred a marketing channel that collects the

harvest 85.89% of the time. In addition, around 72.33% of the farmers were not business-

minded, while 27.67% were interested in doing business given better liquidity. Many of them

(47.67%) regard psychological needs (belongingness and love) and esteem needs (prestige and

accomplishment), in reference to Maslow’s hierarchy of needs, as their work motivation. Yet

44.67% consider physiological (food, water, shelter) and safety needs as their driving force,

104
Farmers’ Typology and their Marketing Practices

and 7.67% take self-fulfilment (self-actualisation) as their work motivation. Finally, 46.7% of

the respondents declared that price is their primary consideration when choosing a buyer. In

comparison, the rest (53.3%) said that other factors such as social, cultural, and convenience

are more important to them than price.

5.1.2 Economic Characteristics

Rice farming alone provides for the household income of 20% of the farmers. The rest combine

it with other income sources such as other agricultural products, business, odd jobs (69%), or

fixed employment (11%). For the majority of the cases (55.22%), the farmers earned from only

PHP 1 to 50,000 (USD 0.02 to 1,000) per cropping season. Worse, in some instances, they

made nothing encompassing 4.89% of the cases, while others received negative income as low

as PHP -100,000 (-2,000 USD) 8.22% of the time because of crop failure due to typhoon, pests,

or diseases. The average off-rice-farm-income of the cases was PHP81,000 (1,620 USD) per

semester, while the majority (52.89%) received only up to PHP50,000 (1,000 USD) from their

other income sources. The most significant portion of the cases (31.56%) only had a total

household income of PHP50,000 (1,000 USD) every six months from combined farm and off-

farm sources. Given the annual poverty threshold is PHP109,000 (2,096 USD) (Philippine

Statistics Authority, 2016; The World Bank, 2018) and considering only their sales income,

the interviewed farmers are living below the poverty line in 55% of the sampled cases. This

poverty incidence is far worse than the 34.3% poverty incidence of farmers in the Philippines

reported by the Philippine Statistics Authority (PSA, 2017).

Given these low incomes, it is understandable why the data show that only 31.6% of the cases

were self-sufficient farmers or not borrowing their capital for rice production. The majority of

the time (68.44%), they use financing, relying solely on it for their farm expenses 23.22% of

the time, while they combined borrowed money with their own money in 45.22% of the cases.

105
Farmers’ Typology and their Marketing Practices

Most of the farmers borrow money or farming inputs from informal lenders, which impose

interest rates that are as high as 15% a month. Yet, because of the immediate needs and the

convenient process, farmers still prefer them over formal lending institutions such as banks.

Moreover, about 21.67% of the farmer-respondents do not have a car, while 65.33% have a

small vehicle such as a hand tractor, motorcycle, or tricycle. Only about 7.33% of them have

vehicles suited to carry small amounts of rice, such as a multi-cab (small jeepney), SUV car,

or pick-up truck, while the remaining 4.33% have a jeepney or truck suited to carry more sacks

of rice.

The farmers' biggest land parcel is 1.38 hectares on average, with as low as 0.12 and as large

as 19 hectares (before winsorisation). The majority of the farmlands (76%) were less than 2

hectares, with 34.6% being barely one hectare. Finally, only 4% of the cases farmed 5 hectares

and above. Exactly 58.99% of the cases involved rented land, either through commission or

fixed terms, while the rest of the cases involved self-owned lands. Around 60% of the cases

had farm irrigation, although some irrigation systems still need water pumps to produce

sufficient water. The rest (40%) depended on rain, stream, creek, or river for water.

The majority (86.33%) of the farmers harvest twice a year, but some harvest either once or

twice (2.67%), twice or thrice (5.33%), and thrice (5.67%) per year; thus, showing the

frequency of marketing transactions they carry out annually could also affect their social

relationship with their buyers. The farmers' average yield was 4.32 tons per hectare (t/ha),

ranging from 0.4 to 11.2 t/ha. Moreover, the total volume sold per season can range from 133

to 28,400 kilograms, with farmers keeping about 1,000 to 1,500 kg a season for family

consumption. More than half of the farmers (56.7%) sold less than 3,000 kg or 3 tons of rice

per cropping.

106
Farmers’ Typology and their Marketing Practices

Furthermore, 59.7% of the farms were within 1 to 5 km of the nearest trader-miller, with an

overall average distance of 6.4 km. The closest distance was just 1 metre, and the farthest was

31.5 km. The National Food Authority (NFA), the government agency designed to regulate the

rice buying price, is even farther – with an average distance of 24 km and as far as 102 km

from the farmers’ houses. A total of 65% of the respondents’ homes are within 5 km of the

nearest accessible market, with an average distance of 5.27 km and the farthest distance of 21.9

km.

In general, most of these data are quite different from the rice-based farm household survey

(RBFHS) in 2011-2012. This could be because the data were from 2017, and thus, are newer

by five years. Secondly, almost 10% of the interviewed farmers were not part of the 2011-2012

RBFHS as they are from the eight new provinces in the 2016-2017 survey.

5.1.3 Social Characteristics

A great majority of the farmers (70.7%) had memberships to either an association, organisation,

or cooperative. Specifically, 50.7% belonged to one association or cooperative, while 20% had

memberships in two or more. However, it is interesting to note that 29.3% of them were not

affiliated with any agriculture-related organisation despite the government's efforts to organise

farmers in groups to provide them with better facilities.

Moreover, only 9.33% of the farmers had not ever attended a training, seminar, workshop, or

technology demonstration. Many of them (39.33%) have experienced 1 to 5, 22.33% have

joined about 6 to 10, 11.33% have attended about 11 to 15, while 17.67% have even completed

more than 15. Overall, their average training session attendance was 1.89.

107
Farmers’ Typology and their Marketing Practices

5.1.4 Cultural Characteristics

Only 6.67% of the farmers were members of indigenous groups such as Maranao, Blaan, or

Lumad. While this research also wanted to determine if similar cultural values affect farmers’

marketing choices, it was challenging to get that information from all farmers because only

24% of the respondents had a close relationship with their buyers being relatives or

friends/acquaintances. About 24.67% of the farmer-respondents were still practising old farm

traditions. These farmers are considered traditional as they perform chanting, putting

lemongrass upside down in the corner of the field, following a moon calendar, choosing a

planting date based on the number stroke, and burning a sample crop harvest in the kitchen,

among others. They believe that these practices could protect their fields from pests and

diseases and give them bountiful harvests.

5.1.5 External Environment Characteristics

Most of the farmers (88%) live in municipalities, while 12% live in cities. The biggest group

of them (28.7%) live in first-class municipalities with an average annual income of at least PHP

55 M (1.1M USD). In comparison, only about 6.7% live in fifth-class municipalities with an

average yearly income ranging from PHP 15 to 25 M (0.3 to 0.5 M USD). About 6% of the

farmers live in second- and third-class cities, while 2.7% are in first-class cities.

Furthermore, 53.22% of the farmers live in provinces producing at most 150,000 tons of rice

per season, while only 21.33% live in provinces producing more than 300,000 tons of rice per

cropping. Finally, 50.67% of the farmers live in towns with good farm market roads – with

continuous concrete and asphalt roads. By contrast, 40% of them live in communes with

moderately good road conditions with partial dirt roads, while 9.33% of them are in towns with

mostly rough and dirt roads, or even without roads because of a river or canal.

108
Farmers’ Typology and their Marketing Practices

5.2 Marketing Practices

The marketing practices of the farmers are presented in this section in three parts. These are

the product form choices, the specific marketing channel choices, and other marketing-related

information. This segment includes both the frequency of the practice and its rationale, which

can help create a better picture of the farmers’ marketing choices.

5.2.1 Form of the Product Sold

Rice can be sold in any form – fresh or field wet, with a moisture content of 24% to 26%;

partially dried or dried up to approximately 18% moisture content; thoroughly/fully dried or

dried up to about 13% to 14% moisture content; and milled (Gummert & Rickman, 2011).

Farmers choose which form to sell depending on different factors, which also affects the

marketing channel choices, e.g., some indirect buyers such as agents and assemblers only buy

fresh produce. By contrast, milled rice is only sold to direct buyers such as trader-millers,

wholesalers, retailers, and consumers. Based on the survey interview results, fresh paddy rice

was sold for 83.8% of the cases, while partially and fully dried paddy rice were sold 0.6% and

9.8% of the time, respectively. On the other hand, milled rice was sold in 5.9% of the cases.

Overall, the selling price ranged from PHP 8 to 44 (USD 0.16 to 0.88) per kg, depending on

the type of marketed output. Specifically, the price range of fresh paddy rice was PHP 8 to 28

(USD 0.16 to 0.56) per kg with an average price of PHP 17.67 (USD 0.35); partially-dried

paddy rice was from PHP 16 to 22 (USD 0.32 to 0.44) per kg with an average of PHP 18.1

(USD 0.36); dried paddy rice was from PHP 14 to 27 (USD 0.28 to 0.54) per kg with an average

of PHP 19.97 (USD 0.4); milled rice was sold from PHP 26.30 to 44 (USD 0.53 to 0.88) per

kg with an average of PHP 39.34 (USD 0.79). Typically, the price of milled rice is the highest,

followed by dried paddy rice, partially dried paddy rice, and finally fresh paddy rice.

Nevertheless, there are outliers in the data, reflecting a lower selling price of dried paddy rice

109
Farmers’ Typology and their Marketing Practices

than fresh or partially-dried rice. One reason for this is some of the farmers who sold partially

dried paddy rice had premium varieties that commanded high prices ‒ even higher than other

types that are dried. Second, some farmers who were affected by unfavourable weather couldn’t

sell their paddy rice as fresh because the buyers temporarily stopped buying. Thus, to save their

product from further deteriorating, they dried it and sold it when the buyers opened again.

Despite being dried, however, the paddy rice grains were sold at a low price because they were

a bit muddy, and their colour had already darkened. Finally, the price is highly dependent on

the location of the farmers – while some areas have high buying prices because of the high

competitiveness, other places have limited buyers, even monopsonies, resulting in low buying

prices.

Moreover, 79.67% or 239 farmers sold fresh paddy rice for all three seasons, while 2.33% and

7.33% sold partially and fully dried paddy rice for three seasons, respectively. This reflects that

there are only about 10% of farmers willing to dry their harvest. By contrast, only 5%

consistently sold milled rice throughout all the seasons, while the rest of the farmers (5.67%)

were not consistent with their choice of product form because of different circumstances.

5.2.1.1 Fresh Paddy Rice

Fresh paddy rice was sold 83.8% of the time for several reasons. The most common cause is

drying, which can be because of unfavourable weather, lack of drying facilities, and labour

issues. While some farmers try to maximise income by taking the chance to dry when there is

a sunny sky, most of them do not. This is because there are areas where manual drying, which

is usually the only option available for farmers, is challenging given the erratic weather

patterns. Farmers shared their experience when they had to pay hired labour for three days

because of the prolonged drying caused by the intermittent rain. This was especially true during

the second cropping, i.e., October to November. Thus, even the most willing farmer opts to sell

110
Farmers’ Typology and their Marketing Practices

fresh paddy rice given drying constraints ‒ especially when there is a typhoon forecast, lest

their harvest will go bad.

Alongside this is the problem with drying facilities, that is, whether it be solar or mechanical

drying. Most of the farmers only use roads for drying, especially because many do not have a

backyard to spare or because the basketball courts they used before have been covered. There

are no mechanical dryers in most farmers' villages, hence the indifference to selling dried or

milled rice. While there are a few places where the government has installed small mechanical

dryers through local cooperatives or associations, these are small and insufficient to serve all

farmers in the area. This results in long queues that discourage other eager farmers from using

mechanical dryers, explaining that they cannot afford to wait until it is their turn to use them,

lest their paddy rice will discolour and command a lower price instead of a higher one. The

farmers also complained that the order of its use is not objective but rather self-serving and

based on nepotism, so they are not likely to benefit from it. Oftentimes, these also have unfair

terms of usage. For instance, in one town, the farmers said that they never get to use the dryer

or miller donated by the government because the politician-trader who volunteered his lot for

the project is charging very high drying and milling fees, and it is better to just sell fresh paddy

rice to them and likewise buy milled rice from them. Meanwhile, commercial dryers are

available in some areas, but they are usually expensive to use and discourage farmers from

drying and milling. In addition, they are likewise full during the peak harvest season.

Moreover, some farmers believe that mechanical dryers are less practical than solar drying

because apart from fuelling the dryer and paying the usage fee, they have to transport the sacks

of rice to and from the drying shed, which adds costs. Likewise, they deem it more laborious

because it requires frequent and thorough mixing lest it gets overdried and becomes chalky and

breakable. Others even claim that they are generally not efficient and thus, result in poorly

111
Farmers’ Typology and their Marketing Practices

dried rice no matter what. These are but some of the reasons why there are some areas where

the dryers are not being used at all and were just left to deteriorate.

Another significant issue of drying is the lack of labour. While many of the farmers used to

perform postproduction processes such as drying and milling themselves, mainly to save on

cost, many were not able to continue doing so because of health and physical limitations, e.g.,

hypertension and old age that prevent them from standing long hours under the sun and hauling

the sacks of rice on their own. Similarly, others do not have the time for postproduction work

because of off-farm jobs. And while hired labour is an option for some, most do not opt for it

because of its cost. Specifically, hiring labour for both hauling and drying is relatively hard and

expensive, in some areas, because of high demand during the peak harvest season. Sometimes,

farmers must also provide snacks and a drinking session as part of their social obligation, which

adds to their costs. Some farmers similarly related that the wage they pay should be higher than

the going rate. Otherwise, no one will work for them the next harvest season. This is made even

worse because they must hire labourers not just for one day but for a couple of days, given the

erratic weather in their area and the limited cement pavement or roads for drying. These

scenarios and costs make farmers presume that drying and milling will not result in significant

additional income.

Overall, many farmers believe that the price they are getting from selling fresh paddy rice is

good enough; thus, they opt to sell fresh paddy rice despite favourable drying conditions. Many

of them think that drying yields insignificant additional income, which is not worth the extra

expenses, effort, and time spent on it. Furthermore, farmers said that the price for early harvest

is good because the supply during that time is still low. They believe that if they dry and sell

rice a few days later, then they might miss the off-peak price. There is also a special price for

particular varieties, in which the fresh price is at par with the price of other dried paddy rice

112
Farmers’ Typology and their Marketing Practices

varieties. This is typically offered by travelling assemblers who only buy these varieties in

fresh form. Hence the farmers sell to them when they arrive in town; they do not bother to dry

because they are already satisfied with the price.

Finally, drying reduces the weight of paddy rice by around 16% (IRRI, n.d.). Hence, some

farmers prefer to sell their harvests fresh because of the higher weight brought about by their

freshness. According to some farmers, this is especially true with the mechanical reaper

because it finishes within hours, unlike before when they must pile the harvested grains under

the sun for a day or two, thereby losing its freshness and some weight. Based on the interviews,

this maximised fresh weight results in comparable income with selling dried paddy rice.

However, computing based on the data in one of the study locations that had this popular

opinion, selling dried paddy rice seems to be still more profitable. For instance, in Tuguegarao,

while the fresh paddy rice is sold at PHP 15 (USD 0.3) per kg, the dried paddy is around PHP

18.5 (USD 0.37) per kg. Considering weight lost from drying, which is 16.3% (IRRI, n.d.), the

farmer still has an additional profit of PHP 0.48 (USD 0.01) per kg, less PHP 10 (USD 0.2) per

sack (50 kg) for the manual drying fee. After costs, the farmer still has at P 14.22 (USD 0.284)

per sack additional income, which is an additional net income of PHP 1,080 (USD 21.6) per

hectare if multiplied by the average yield of 3.8 t or 76 bags per hectare.

Farmers also have money-related issues that result in selling at the farm gate. Specifically,

some farmers reported the need to pay off debts. They likewise emphasised the need for

immediate cash to pay for necessary expenses at home and their children’s schooling. Finally,

they explained that they do not have the required additional capital for postproduction

processes such as hauling, trucking, drying, and milling. Similarly, they do not have spare

money to use while waiting for the sales of dried paddy rice or milled rice.

113
Farmers’ Typology and their Marketing Practices

Other reasons for selling fresh paddy rice at the farm gate similarly includes storage issues.

Transportation issues are also included among the concerns given the need for trucking when

hauling the rice grains from the field to the house, then to the drying area, then back to the

home or the place of the buyer. Alternatively, if rice is to be sold milled, trucking to the rice

mill, then to the storage, then to the buyer. Given that only 4.3% of the farmers have a suitable

delivery vehicle, this is a feat that is elusive for most of them. By contrast, this is not an issue

when farmers sell at the farm gate because indirect buyers instantaneously provide

transportation. Meanwhile, some farmers have no option but to sell fresh paddy rice because

all the buyers in their area only buy fresh paddy rice from the field. These areas include some

localities in Ubay, Bohol and San Francisco, and Agusan del Sur. Finally, some farmers sell at

the farm gate simply because of labour disutility or because they want to avoid the hard work

required to sell dried or milled rice.

Given all these challenges and the need to dry the paddy rice immediately ‒ especially during

the rainy season to prevent further deductions caused by very wet or discoloured grains ‒

farmers often just rush and sell to whoever is in the field at the time of harvesting. These are

usually indirect buyers, such as agents and assemblers.

5.2.1.2 Dried Paddy Rice

During the three cropping seasons documented in this research, the farmers sold partially dried

paddy rice just 0.56% of the time and sold dried paddy rice 9.73% of the time. While most

farmers would want to sell thoroughly dried paddy rice with 14% moisture content, which is

ready for milling, they are limited by the circumstances mentioned previously, such as labour

issues, capital, and facilities. Nevertheless, these farmers appreciate the additional gain they

could have from drying; thus, they dry their products as much as their circumstances allow.

114
Farmers’ Typology and their Marketing Practices

Fully dried paddy rice, or those dried up to 14% moisture content, has a significant price

advantage. Yet many farmers are only able to do this during the dry season because of manual

drying issues. Other larger-scale or well-connected and financially endowed farmers, in

contrast, can dry during any season. While there are areas where farmers only sell fresh paddy

rice because of buyer preference or other limitations, there are likewise areas where farmers

almost always dry their harvest before selling, such as in Mindoro Occidental and Mindoro

Oriental in Luzon. Specifically, farmers in those provinces claim that almost nobody buys fresh

paddy rice in their areas, so they must dry it. Those who cannot dry using family labour, thus,

hire some people to help. Rain is not a big issue for them as they have learned to adapt to the

situation, e.g., using fishnet or tarpaulin, a heavy-duty waterproof cloth, to dry their harvest as

these materials are handy when packing up the grains in times of sudden rain and when there

is no cement pavement to use.

Overall, these farmers believe that drying will give them a higher profit than selling fresh paddy

rice. One farmer computed that the additional income is 70 cents (0.014 USD) per kilogram,

which for him, is large enough to go through the whole process because it will result in an

additional 10% net income. Some farmers shared that they dry their harvests when there is a

considerable disparity between the fresh and dried paddy rice prices. In such instances, they

endure the laborious drying process because they believe that the reward for drying is

significant. In general, farmers dry depending on the additional income – some will dry for an

additional net income of PHP 0.50 (0.01 USD), while others would only do it for an extra

income of at least PHP 1 (0.02 USD) per kilogram.

Furthermore, drying seems to be more beneficial for farmers with a capital endowment because

they can afford the related postproduction costs and wait for the best price to sell. Practically,

they dry the paddy rice to be able to stock it without affecting its quality. They then sell it

115
Farmers’ Typology and their Marketing Practices

during the off-peak season, when the price of rice is highest. One farmer from Bulacan

explained that this practice results in around PHP 8 (0.16 USD) additional income per

kilogram, which is PHP 400 (8 USD) a sack. While he doesn’t have any storage facilities, he

rents space from the rice mill where he usually sells, explaining that the rent is minimal

compared with the additional gains.

5.2.1.3 Milled Rice

Finally, a small percentage of the cases (5.9%) involved drying, milling, and selling their

products in the form of milled rice. While drying and stockpiling dried paddy rice can already

result in a good income, selling milled rice is maximises it. In fact, for other farmers, selling

dried paddy rice instead of milled rice is a big waste given the PHP 20 (0.4 USD) per kilogram

average price difference between the two. Overall, only very few farmers managed to sell

milled rice directly to consumers instead of selling fresh paddy rice to indirect buyers. It results

in an additional net profit of at least PHP 4 (USD 0.08) per kilogram, which is very close to

the estimates of Mataia et al. (2018) at PHP 3.99 per kilogram additional income.

A farmer from Bulacan divulged that apart from selling dried paddy rice at the peak price, he

likewise sells some of his harvest as milled rice. He said that he delivers it to Cavite Province,

which is about 100 km away, where he has been able to secure a wholesale buyer through a

relative. Another farmer, who owns a small rice mill, supplies milled rice in bulk to his

province's capital city. But apart from that, he sells to retail or per sack to some households

using the public utility vehicle as a mode of delivery. Many other farmers also sell milled rice

by the bag because they do not have bulk buyers, yet they would be very eager to supply by

volume. They usually do that through their small sundry “sari-sari” stores. In general, farmers

sell milled rice to direct consumers, retailers, wholesalers, and trader-millers (which also

function as wholesalers).

116
Farmers’ Typology and their Marketing Practices

Nevertheless, overall, the majority of the cases (94.1%) sell non-milled rice because many

farmers believe that selling milled rice is not ideal. First, this is because it involves a large

quantity of labour from transportation, drying, milling, and hauling. Hence, many farmers

believe that the additional income is insignificant unless even the by-product such as the bran

is used. However, not every farm household has a swine or poultry in order to use it, and selling

bran is not very beneficial because trader-millers buy it at a non-competitive price.

Second, most farmers do not have access to dryers, rice mills and transportation. This is

especially true for a province in Mindanao where the cost of processing milled rice would result

in higher expenses than buying milled rice from the market – that is because of corruption and

monopsonies in the area that aims to discourage farmers from competing in the market.

The third and fourth issues both involve marketing itself. The third issue is the limited buyers,

especially in places where almost every family has a farm. Apart from that, many farmers do

not have access or contact with bulk buyers. The fourth issue is that most farmers do not have

the permit to retail milled rice. Thus, this can cause trouble, especially if they are negatively

affecting the business of the registered sellers. One farmer relayed that he used to sell milled

rice through his sister’s store, but a competitor reported him to the authorities for selling

illegally. Since then, he had not retailed milled rice again.

Lastly, selling milled rice requires investment and good liquidity or high disposable income,

which is a known challenge for small-scale farmers. This is because it usually gives a staggered

payment from retail selling or consignment terms.

5.2.2 Marketing Preference between Direct and Indirect Buyers and the
Typology of Farmers based on their Choices

The surveyed farmers mentioned seven main marketing channels – agents, assemblers,

creditors, buying stations, trader-millers, cooperatives, and wholesalers/retailers/consumers.

117
Farmers’ Typology and their Marketing Practices

Although many of them mentioned National Food Authority (NFA), the government regulatory

institution, nobody among the respondents had it as their primary buyer, with a few farmers

selling only a small part of their harvest to them. Overall, 65.8% of the cases involved farmers

selling to indirect buyers, including creditors, buying stations, assemblers, and agents, while

only 34.22% involved selling to direct buyers such as cooperatives, trader-millers, wholesalers,

retailers, and consumers. The data show that the farmers who sold to direct buyers significantly

differed from those who sold to indirect buyers in terms of the number of children, yield,

distance to the nearest miller, number of organisations, and the rice production volume in the

province where they live at (Tables 5 & 6). Specifically, those who sold to indirect buyers had

a lower average number of children (0.89) and membership in farmer organisations (0.91). By

contrast, those who sold to direct buyers had a higher average number of children (1.09) and

membership to organisations (1.09). The yield of those who sold to indirect buyers was 4.49

t/ha on average, while those who sold to direct buyers had an average of only 3.98 t/ha. Finally,

those who sold to indirect buyers live 6.94 km away from the nearest rice mill and were located

in provinces with an average rice production of 231,985 tons per season, while those who sold

to direct buyers live closer to the nearest rice miller at a mean distance of only 5.29 km and

were located in provinces with lower rice production volume at 174,033 tons per season.

Table 5. Comparing the means of the continuous independent variables of farmers who sold to
direct and indirect buyers.

Marketing N Mean Std. Deviation Std. Error

channel Mean

Age Indirect Buyers 595 55.85 10.792 .442

Direct Buyers 305 54.94 10.590 .606

Educ Indirect Buyers 595 9.05 3.129 .128

Direct Buyers 305 8.87 3.426 .196

HHchildren Indirect Buyers 595 .89 1.228 .050

Direct Buyers 305 1.09 1.409 .081

OffFarmIncW Indirect Buyers 595 76649.75 84114.664 3448.365

118
Farmers’ Typology and their Marketing Practices

Marketing N Mean Std. Deviation Std. Error

channel Mean

Direct Buyers 305 87330.69 97364.129 5575.054

LPareaW Indirect Buyers 595 1.3655 .92866 .03807

Direct Buyers 305 1.4230 1.00577 .05759

Yield Indirect Buyers 595 4.4937 1.70419 .06986

Direct Buyers 305 3.9826 1.76906 .10130

DMarketW Indirect Buyers 595 5.3100 4.65977 .19103

Direct Buyers 305 5.1981 3.95040 .22620

DMillerW Indirect Buyers 595 6.9426 6.99156 .28663

Direct Buyers 305 5.2949 5.71691 .32735

Harvest Indirect Buyers 595 2.062 .2519 .0103

Frequency
Direct Buyers 305 2.085 .2968 .0170

NumofOrg Indirect Buyers 595 .91 .798 .033

Mem
Direct Buyers 305 1.09 .906 .052

RiceProdVolW Indirect Buyers 595 231985.36 208554.950 8549.919

(tons/s)
Direct Buyers 305 174033.68 144427.581 8269.899

Table 6. Independent sample test showing the significant difference in the characteristics of
farmers who sold to direct and indirect buyers.
Levene's
Varia- Assump- Test for
t-test for Equality of Means
bles tion Equality of
Variances
95% Confidence
Sig.
Mean Std. Error Interval of the
F Sig. t df (2-
Difference Difference Difference
tailed)
Lower Upper
Equal
.288 .591 1.200 898 .231 .906 .755 -.576 2.388
variances
Age
Variances
1.207 623.395 .228 .906 .751 -.568 2.380
not equal

Equal
5.005 .026 .783 898 .434 .178 .228 -.269 .625
variances
Educ
Variances
.760 566.461 .447 .178 .234 -.282 .639
not equal

119
Farmers’ Typology and their Marketing Practices

Equal
2.166 .141 -2.192 898 .029 -.200 .091 -.378 -.021
HHchil variances
dren Variances
-2.098 544.596 .036 -.200 .095 -.386 -.013
not equal

Equal
6.337 .012 -1.708 898 .088 -10680.939 6255.057 -22957.172 1595.294
Off Farm variances
IncW Variances
-1.629 540.611 .104 -10680.939 6555.337 -23557.993 2196.115
not equal

Equal
3.858 .050 -.854 898 .393 -.05745 .06729 -.18951 .07460
variances
LP areaW
Variances
-.832 571.870 .406 -.05745 .06904 -.19305 .07814
not equal

Equal
1.310 .253 4.204 898 .000 .51116 .12158 .27254 .74977
variances
Yield
Variances
4.154 593.304 .000 .51116 .12305 .26949 .75283
not equal

Equal
1.685 .195 .358 898 .720 .11188 .31214 -.50072 .72449
DMar variances
ketW Variances
.378 707.972 .706 .11188 .29607 -.46940 .69317
not equal

Equal
11.317 .001 3.552 898 .000 1.64770 .46393 .73720 2.55821
DMil- variances
lerW Variances
3.787 729.403 .000 1.64770 .43510 .79351 2.50190
not equal

Equal
7.302 .007 -1.222 898 .222 -.0231 .0189 -.0601 .0140
Harvest variances
Frequency Variances
-1.160 532.782 .247 -.0231 .0199 -.0621 .0160
not equal

Equal
3.244 .072 -3.044 898 .002 -.179 .059 -.295 -.064
Numof variances
Org Mem Variances
-2.923 549.453 .004 -.179 .061 -.300 -.059
not equal

Equal
Rice 56.327 .000 4.347 898 .000 57951.679 13330.607 31788.908 84114.451
variances
Prod
Variances
VolW 4.872 821.091 .000 57951.679 11895.056 34603.381 81299.977
not equal

120
Farmers’ Typology and their Marketing Practices

5.2.3 Specific Marketing Channel Choices

Exactly 93.8% of the cases involved farmers selling paddy rice to traders, while only 6.2%

involved selling milled rice to households, retailers, and wholesalers. Although slightly lower,

this is comparable with the 100% sales rate to traders of the 196 farmers in Myanmar surveyed

by Soe et al. (2015). By contrast, it is much higher than the 41% to 62% that Cazzuffi and

McKay (2012) recorded in their study of rice farmers in Vietnam and the 39.2% that Dessie et

al. (2018) found in their research of wheat farmers in Ethiopia.

Specifically, 29.8% of the cases sold rice to buying stations or traders without rice mills (Table

7). These traders just buy paddy rice and usually sell it fresh to the trader-millers. According

to some farmers' speculations, buying stations earn as much as PHP 1.50 (USD 0.03) per

kilogram from their product. Yet, they still prefer to sell to them because of convenience owing

to their fast payment and transaction. Credit obligation is also a factor because some farmers

obtain their farming inputs or equipment rentals from them, as well as cash credit. Many of

them say that they must sell to them to maintain the good credit standing and smooth

relationship, enabling them to borrow cash or inputs from them again during the next cropping

period. In addition, farmers said that buying stations are less stringent when pricing fresh paddy

rice than trader-millers, resulting in lower deductions per sack.

Table 7. The main marketing channel choices of farmers in the Philippines.

Marketing Channel Frequency Per cent

Agent 225 25.0
Assembler 94 10.4
Buying station 268 29.8
Creditor 5 .6
Cooperative 16 1.8
Rice mill 236 26.2
Wholesaler/retailer/consumer 53 5.9
Others 3 .3
Total 900 100.0

121
Farmers’ Typology and their Marketing Practices

Similarly, there are municipalities without trader-millers, leaving farmers without many

options. Finally, farmers sell to buying stations instead of the trader-millers because they

cannot deliver in bulk, which is a requirement of most rice mills. They also like the idea that

they do not have to spend resources on trucking or to deliver their products, which they think

is a problematic and time-consuming task. Many farmers likewise sell to buying stations not

because of their services but because of friendship or blood relations. Finally, some farmers

sell to them because of their accurate weighing scales. They explained that some buyers, e.g.,

assemblers, sometimes offer a slightly higher price but have a weighing scale that is inaccurate,

resulting in as low as 2 kilograms per sack and thus, a lower overall income.

Many cases (25.0%) also sold through agents (Table 7). These agents canvass for prices and

transact with the buyer on behalf of the farmers, in return for about 20 cents (0.004 USD) a

kilogram. Agents are mostly dispensable players in the market value chain, as shown by some

farmers' experiences. For instance, a farmer from Leyte (Visayas) shared her experience that

she cut the agent, went directly to the rice mill owner, and successfully earned an additional 20

cents (0.004 USD) per kilogram. Another farmer in Bulacan (Luzon) similarly recounted that

he cuts the agent to get the extra income that would otherwise go to them. However, many of

the farmers believe that they play a crucial role in their marketing. For instance, a farmer cited

an experience when he went directly to the trader-miller but was referred back to an agent

because the trader-miller only wants to transact with agents. Another farmer shared that he

tried selling directly to the trader-miller, only to find out one day later that they could not accept

his harvest because they had reached their buying capacity. He thinks that if he had used an

agent, the trader-miller would have taken it because of their better relationship.

Finally, many farmers use agents believing that they are not paying them, but rather, the trader-

millers. They think that having agents help them get them higher prices because they bargain

122
Farmers’ Typology and their Marketing Practices

for them. They likewise appreciate that agents make their marketing easier – they usually do

not have to go anywhere after harvesting; they just need to wait for their price offers. The

agents take a sample of the harvest and scout for the best buyer of fresh paddy rice, where they

can profit most. They then come back to the farmer with price quotations, leaning towards their

choice. Thus, the farmer can sell their harvest without much effort when using an agent. This

convenience is one of the most common reasons cited by the farmers for their use of agents in

marketing their rice. While selling to buying stations was the most popular choice of the

farmers in this study, selling to brokers or commission agents was dominant in the study of Soe

et al. (2015), with 64.8% of the farmers selling to them. This is very high compared to the 25%

recorded in this study, making commission men or agents just the third-most preferred

marketing channel by Filipino rice farmers. However, this disparity could be attributed to the

grouping of the marketing channels. Specifically, Soe et al. (2015) only classified their

marketing channels into three – the brokers or the commission men buying at the farm gate,

the collectors or the large-scale farmers buying at the farm gate, and the trader-millers in the

nearby town. As such, he might have categorised all the small-scale farmer-buyers into the

broker or commission men category. By contrast, this study accounted for each specific

marketing channel, thus limiting the agent classification to brokers only; hence, the low

number. Nonetheless, given the assumed categorisation, this study would result in 65.8% sales

to brokers or commission men at the farm gate, close to the 64.8% recorded by Soe et al. (2015).

Assemblers or buyers without fixed stations are the fourth most popular marketing channel

used by the farmers in this study. Most of them travel from province to province to assemble

the paddy rice harvest and supply it to low-rice-producing areas. Many of them are provided

with working capital by trader-millers, but some use their own money that gives them more

freedom to choose where to deliver their assembled rice. Farmers speculate that the share of

these assemblers from the price they receive is only PHP 0.20 (0.004 USD). Nevertheless,

123
Farmers’ Typology and their Marketing Practices

Mataia et al. (2018) showed that the assemblers’ mark-up ranges from PHP 1 to 1.5 (USD 0.02

to 0.03) per kilogram.

Based on the qualitative data, one of the reasons why farmers prefer assemblers is convenience

and availability. They are usually even more accessible than buying stations because they are

at the farm during harvesting. Thus, they can immediately pick up the harvest from the farmers’

fields and weigh the rice sacks on site. This is much faster than selling to trader-millers, which

usually requires trucking and queueing for a truck scale. It is also quicker than buying stations

where the farmers still have to notify the owners that their harvest is ready and then wait for

their availability for pick up. Consequently, they pay the fastest of all the channels. Farmers

related that assemblers pay them in cash immediately after weighing their harvest, unlike

trader-millers, who pay them after three to four days.

Farmers also choose assemblers because they stick to the price and deduction agreements,

unlike millers who tend to renegotiate towards a lower price after the product has been

delivered to them. Some farmers also appreciate the fact that some assemblers are very exact

when recording the weight; they do not round-off the weight of each sack, unlike in buying

stations. In addition, they sometimes provide inputs without interest to ensure that they can

purchase the farmers’ harvest later. Furthermore, many farmers who sell to assemblers believe

that they offer higher prices than buying stations, and sometimes even higher than the trader-

millers because this is their way to get clients from the area. One farmer said that assemblers

even buy rice with low milling quality at a much higher price than trader-millers and buying

stations because they mix it with the premium class grains.

By contrast, one reason why other farmers do not sell to assemblers is the speculation that their

weighing scales are not accurate to compensate for the higher price. This is not an issue for

those who choose assemblers, though. They address the issue by comparing the weights of their

124
Farmers’ Typology and their Marketing Practices

weighing scales with that of the assemblers to check for accuracy. A few farmers also said that

assemblers give higher deductions. Hence, they pay less, so the farmers receive a lower income

despite the higher price. Moreover, there have been issues with some assemblers due to fraud.

A farmer related that one farmer in their village lost a significant sum of money because the

assembler paid him with counterfeit money, strategically placed in the middle of real

banknotes. He recalled that out of the PHP50,000 (1000 USD), about 90% was fake. Hence,

some farmers do not sell to assemblers, especially because they come and go; they say that they

don’t know where to find them when there is a problem.

In contrast with the previous marketing channels, a very small number of cases sell to creditors

(0.6%). Creditors are neither trader-millers nor buying stations or agents because they only

“buy” or receive paddy rice from their debtors as payment for their debts. Most of them do not

accept cash for payment and instead get the farmers' harvests at a non-competitive price. They

are generally like loan sharks because apart from high credit interest rates, which can be as

high as 60% per season, they also take advantage of the farmers through their low buying

prices.

Overall, 34.2% of the cases directly sold to trader-millers, cooperatives, wholesalers, retailers,

and consumers. More specifically, 26.2% of all the cases sold to trader-millers, which is

comparable to the 20.9% documented in Myanmar by Soe et al. (2015). Based on the qualitative

data, one of the main reasons for selling to trader-millers is their best price offer for paddy rice.

They even give additional income or commission for referrals. Many large-scale farmers

likewise sell to them because they provide free trucking, and they can afford to pay for the

whole volume at once. Some farmers also mentioned that they sell to trader-millers because

they dry and mill their rice as part of a package, with some even providing free storage. One

farmer cited that they can take out one sack at a time, depending on their needs, which is very

125
Farmers’ Typology and their Marketing Practices

convenient. Another main reason that made trader-millers popular among farmers is credit,

which they usually avail at the start of the season. Some farmers similarly reported selling to

millers because of trust, given years of consistently transacting with them. Some sell because

they provide a compassionate price when the yields are low, and they do not return not-well-

dried paddy rice but rather just impose deductions. The final advantage of selling to trader-

millers is they buy fresh, dried, and milled rice.

By contrast, many farmers cannot sell to trader-millers because they cannot meet the minimum

volume requirement for free pick up, nor do they have the capacity to deliver to them. Some

farmers explained that they only provide pick up service for around 100 sacks; anything less

or more will not be included in the free hauling. The farmers also reported that selling to trader-

millers requires time – time for queueing that sometimes requires another day, and time for

waiting to get paid that can take around three to four days. Finally, many trader-millers are also

notorious for their unfair practices – they are known to farmers for lowering the price or

increasing deductions once the harvest has already been delivered, thereby leaving the farmer

without any other more manageable option than to accept the new terms.

Farmers also sparingly sell to their cooperatives, accounting for only 1.8% of the cases. These

farmers sell to their cooperatives with the conscious effort to support their business, knowing

that it will benefit them in the end. Cooperatives provide free transportation services to their

members, supply inputs, and use accurate weighing scales. On the contrary, the bureaucratic

procedure discourages farmers from selling to their cooperatives ‒ especially if they need fast

cash to pay their obligations. According to some farmers, the pay from their cooperatives can

take days or even about a week. Most of the time, the payment is also in a cheque form, so they

must go to the bank and cash it. Similarly, many farmer-cooperatives do not engage in rice

trade given their inadequate funds, facilities, and workforce. Thus, their business is mostly

126
Farmers’ Typology and their Marketing Practices

limited to credit and facility rentals. However, in rare cases, they act as an assembler to collect

bulk orders and deliver them to the National Food Authority for a higher income.

Interestingly, only a small percentage of the farmers sell milled rice to wholesalers, retailers,

and consumers, accounting for only 5.9% of the cases. According to the farmers, among the

reasons for the low sales rate to this marketing channel is limited access to drying and milling

facilities and the lack of access to buyers of milled rice ‒ especially in farming communities.

Thus, only those who are not living in pure agricultural communities are able to retail milled

rice. Others sell through retailers/wholesalers but on a consignment basis, where the farmers

are paid only once the product has been sold.

Nevertheless, despite small social networks and limited buyers in their communities, some

farmers find ways to secure large-volume buyers. They go to other provinces and negotiate

with companies who provide rice allowance or supply rice to retailers to maximise their profit.

Meanwhile, the farmers with limited harvest try to get the most out of their produce by retailing

whatever they have left to their neighbours.

The percentage of farmers in this research selling milled rice to direct buyers is lower than

previous findings. For instance, Cazzuffi and McKay (2012) noted that at least 38% of the

farmers in their research could be selling to non-traders, or in effect, other households.

Moreover, Dessie et al. (2018) found that 40.49% of wheat farmers’ produce was sold to

retailers, 37.5% to consumers, and 23.93% to wholesalers; while their study on teff showed

that 37.01% of the farmers market to consumers, 35.71% to retailers, and 31.82% to

wholesalers (Dessie et al., 2019). Although Dessie et al. (2018, 2019) accounted for the

multiple marketing channels that farmers used in one cropping season, even considering the

value of just one of them is enough to provide evidence that the rate of farmers selling to

consumers in the Philippines can still increase given more favourable circumstances.

127
Farmers’ Typology and their Marketing Practices

Finally, no farmer among the respondents considers the NFA as their primary buyer.

Nevertheless, some sell part of their harvest to them, but mainly the non-premium varieties.

One farmer from the Visayas expressly reserves the premium types for milling and sells them

through wholesale and retail, while they pool the non-premium varieties for selling to the NFA.

Ostensibly, he is one of the only three who were able to sell to the NFA amongst all the

respondents, although others said that they were able to sell to the NFA back when their price

was more competitive. This is because most of the farmers believe that marketing to the NFA

is presently not profitable. Worse, many of them do not know that the NFA exists and buys

paddy rice – thinking that their function is limited to selling cheap rice.

Among the other reasons why farmers rarely sell to the NFA are the strict product

specifications, accessibility issues, sales promptness, and price. First, farmers think it is hard

to meet the NFA's product specifications, which includes the 14% moisture content. They said

that the NFA does not accept fresh paddy rice most of the time. Hence, even if farmers want to

sell to them, they cannot do so because of the aforementioned drying issues. Some farmers

likewise complained that the NFA wants a cleaned harvest, free from hay and other impurities,

thus requiring the use of blowers which adds to the cost. Otherwise, they will receive a lower

price unless they use their social connections or bribe the classifier.

Another issue with the NFA is accessibility. On the one hand, while the NFA offices are said

to be strategically located nationwide with warehouses in almost all provinces, many farmers

said that they do not have access to them because they are either non-existent or always closed.

On the other hand, those that are open are still hard to access because farmers do not have the

resources to transport their harvest. The NFA officials, documents, and policies outline that the

NFA picks up fresh paddy rice from the farmers’ field without minimum volume requirements.

However, the farmers said that this is not true for anything less than 100 sacks. In fact, even

128
Farmers’ Typology and their Marketing Practices

the farmers who used to sell and are currently selling to the NFA said that they must deliver

the harvest to their warehouse. They often have to rent trucks if the volume is large, while some

hire motorcycles with sidecars for smaller volumes, i.e., four sacks (more or less 200

kilograms) per trip. This tedious, and most of the time, costly trucking of their harvest is

hindering farmers from selling to the NFA.

In addition, farmers also do not like the fact that they must wait so long to get the classification,

valuation, and payment for their harvest. The farmers reported that there is a queue and that the

NFA classifier is often slow. One farmer even shared his experience when the classifier took

three days to classify his harvest resulting in grain discolouration and, consequently, a lower

price. The farmer was even more enraged when he got the payment after a week. He despised

the process because he had to go to and from the NFA warehouse a couple of times that wasted

his time and efforts, spent a considerable amount of money undertaking the numerous trips and

lost some money from the trade because his harvest was paid out at a lower rate even though it

was their fault. The farmer said that after that, he never considered selling to the NFA again.

Many other farmers shared this sentiment, especially after hearing their friends and co-farmers

complain about the NFA's slow processes and payment.

Last but most importantly, the farmers said that the NFA price is outdated and no longer

competitive. They explained that the prices of other commercial traders are much better than

that of the NFA. For instance, a farmer sold his fresh produce to a trader-miller for PHP17

(0.34 USD) while NFA buys dried paddy rice at the same price. Some farmers were even able

to sell their fresh paddy rice at PHP20 (0.4 USD) to trader-millers because of their timely

harvest and premium variety. Overall, many farmers feel that even if NFA would start buying

dried paddy rice at PHP19 (0.38 USD), it would still be more beneficial for them to sell to

129
Farmers’ Typology and their Marketing Practices

trader-millers at PHP17 (0.34 USD) because of the less meticulous buying process, more

comfortable trucking conditions, and prompt payment.

5.3 Chapter Synthesis

This chapter provides the typology of the farmer-respondents of this study, along with their

common marketing practices. Overall, the farmers are 91% male and between the ages of 40

to 69 years old. They have an average household size of five, but the number of people living

in the same house could range from one to 18. There is an average of one child per household,

but 52.2% of them only have adults. Moreover, they have been farming for 32 years on average,

with a mean formal education of nine years. About 39.9% of them only had at most five

trainings, and 70.7% are members of at least one association, organisation, or cooperative. In

addition, 6.67% of the farmers are members of an indigenous group, and 24.67% are still

practising farm traditions.

Moreover, 72.33% of the farmers are not business-minded and would instead put their extra

money into savings than invest it. The work motivation of 47.67% of them is belongingness,

love, prestige, and accomplishment, while 44.67% are working just to provide their families'

basic and safety needs such as food, water, shelter, and education. In addition, for the majority

(53.33%) of the time, farmers identified non-price factors such as credit, social ties, cultural

values, and convenience as their reasons for choosing a marketing channel. In comparison,

there were only 46.7% cases wherein they primarily chose their buyers based on price.

Furthermore, 20% of the farmer-respondents are solely dependent on rice for a living, while

69% of them cultivate rice alongside other irregular sources of income, 11% have fixed

employment, and 44.3% of them have the means to transport more than 50 sacks of rice.

Farmers were earning only PHP 1 to 50,000 (USD 0.02 to 1,000) from rice per cropping for

130
Farmers’ Typology and their Marketing Practices

the majority of the cases (55.22%). About 68.44% of the cases involved the use of financing ‒

mostly from informal lenders.

On average, farmers are cultivating 1.38 hectares of irrigated land and harvest twice a year.

About 41.11% of the farmers own at least a part of the land they are farming. Most of them

live in municipalities (88%) located in provinces producing at most 150,000 tons of rice per

cropping (53.22%). Around 59.7% and 65% of the farms are within five kilometres of the

nearest trader-miller and market, respectively, with 50.67% having good farm-to-market roads.

Furthermore, the farmers’ average yield is 4.32 tons per hectare, with a 3.9 tons average volume

sold to a single marketing channel. Fresh paddy rice was sold the majority of the time (83.8%)

at an average price of PHP 17.67 (USD 0.35) per kilogram, while only 5.9% of the cases

involved selling milled rice at an average price of PHP 39.34 (USD 0.79). The rest of the cases

either sold partially-dried (0.57%) or fully-dried paddy rice (9.78%) at an average price of PHP

18.1 (USD 0.36) and PHP 19.97 (USD 0.4), respectively.

Finally, indirect buyers such as agents, assemblers, buying stations, and creditor-traders – with

buying stations being most preferred (29.8%) ‒ were chosen 65.8% of the time, followed by

the assemblers (10.4%). By contrast, 34.2% of the cases sold to direct buyers, of which 26.2%

sold to trader-millers, 1.8% sold to cooperatives, and 5.9% sold to

wholesalers/retailers/consumers. The factors affecting these choices are explained in the next

chapter.

131
The Different Factors of Marketing Channel Choices

Chapter 6. The Different Factors of Marketing Channel Choices

While the previous section describes the typology of the farmers and their specific marketing

channel choices along with their motivations and issues, this section focuses on the different

factors affecting the choice of marketing channels. It presents the combined quantitative and

qualitative inquiry results – from personal, economic, social, cultural, to environmental factors.

Overall, the quantitative model has a total of 24 variables and strongly balanced panel data

with 900 observations each since farmers with missing or inaccurate information were

contacted again. The model has a Wald Chi-Square of 63.14 and a p-value of less than 0.01.

Therefore, the output suggests that all 24 independent variables are essential in this model.

Nevertheless, based on the outcome of the random-effects logistic regression, there are only 16

highly significant factors affecting the rice farmers’ choice of direct versus indirect marketing

channels (Table 8; Appendices 3 & 4). The effect of these variables based on the econometric

model and the qualitative analysis are presented in the following sections:

6.1 Personal Factors

Based on the econometric analysis findings, the significant personal factors affecting the choice

of direct versus indirect marketing channels are age (Age) and education (Educ). Specifically,

the log odds of selling to direct buyers decrease by 0.11 for every year increase in the farmer's

age, or the probability of selling to direct buyers decreases by 0.27% for every year increase in

age (Table 8; Appendices 3 & 4).

Table 8. The coefficients and average marginal effects of the choices between direct and indirect
marketing channels, using 900 observations from 300 farmer-respondents.

Coefficients Marginal Effects

Est. S.E. P>|z| Est. S.E. P>|z|
Age -0.1052 0.0523 0.044 -0.0027 0.0013 0.039
DMarket_W -0.0017 0.0884 0.985 0.0000 0.0023 0.985

132
The Different Factors of Marketing Channel Choices

Coefficients Marginal Effects

Est. S.E. P>|z| Est. S.E. P>|z|
DMiller_W -0.4087 0.0722 0.000 -0.0106 0.0012 0.000

DrySeason_D 0.0924 0.4533 0.838 0.0024 0.0117 0.838

Educ -0.3397 0.1320 0.010 -0.0088 0.0032 0.007

FarmTraditions_D 3.8077 0.8595 0.000 983 0.0177 0.000

FixedIncome_D -3.6778 1.2410 0.003 -0.095 0.0290 0.001

Harvest-Frequency 3.7565 1.4733 0.011 0.097 0.0360 0.007

HierarchyOfWork
-1.4325 0.6525 0.028 -0.037 0.0163 0.023
Motivation_O

HHchildren -0.0343 0.2838 0.904 -0.0009 0.0073 0.904

Irrigation_D -0.9723 0.5008 0.052 -0.0251 0.0126 0.046

Landownership_D 3.4711 0.8354 0.000 0.0897 0.0179 0.000

LParea_W -0.9005 0.4312 0.037 -0.0233 0.0106 0.028

IndigenousOrigin_D 10.0101 1.8992 0.000 0.2586 0.0356 0.000

MoneyMgt_D 0.9367 0.7420 0.207 0.0242 0.0188 0.199

NumOrgMem 0.4439 0.3833 0.247 0.0115 0.0098 0.243

OffFarmInc_W 0.00001 4.25E-06 0.018 2.59E-07 1.7e-07 0.016

PriceMotivation_D 0.0713 0.5825 0.903 0.0018 0.015 0.902

-4.58E-
RiceProdVol_W 1.72E-06 0.008 -1.18E-07 4.13e-08 0.004
06

TownIncClass_O 0.512 0.2203 0.020 0.0132 0.0054 0.014

Training_O 0.6778 0.3407 0.047 0.0175 0.0084 0.038

WantsFastcash_D -6.6916 1.3329 0.000 -0.1729 0.0282 0.000

WantsPickUp_D -3.1016 0.8901 0.000 -0.0801 0.0204 0.000

Yield -0.3012 0.2239 0.178 -0.0078 0.0057 0.173

_cons 4.1221 5.3308 0.439

/Insig2u 4.7313 .226

Sigma_u 10.6511 1.419
rho .9718 .007

133
The Different Factors of Marketing Channel Choices

Based on the qualitative interviews, farmers find postproduction work, such as drying and

milling, tiring or inconvenient as they grow older. The older farmers explained that they

stopped processing their harvest and just sold them to agents or buying stations because they

were not fit enough for the hard work. This includes working long hours under the sun to mix

the paddy rice to make sure that they are evenly dried, lifting loads of rice sacks from point to

point, or standing long hours just to supervise the process. It also negatively affects the choice

of direct marketing channels because of labour disutility since selling dried or milled rice

requires manual drying and hauling ‒ difficult processes for older farmers, especially if they

intend to labour on their own and there are limited drying facilities.

Another significant predictor is the education of the farm household head. The log odds of

selling to direct versus indirect buyers decreases by 0.34 as another year of schooling is

acquired; holding all other factors constant, the probability of selling to direct buyers decreases

by 0.88% for every additional year in education (Table 8; Appendices 3 & 4). In general, this

is because farmers with more years of schooling, particularly those who finished at least a

bachelor’s degree, are more likely to have regular jobs that limit their marketing time. Thus,

they prefer to sell to the fastest and most convenient buyers that are often agents, assemblers,

and buying stations. Moreover, the farmers who stopped studying at an early age also started

farming early, as supported by the -0.392 correlation coefficient between education and

farming experience at the 0.01 significance level. Sending children to the farms when they

cannot send them to school is typical in the Philippines because of poverty. Having started

farming at a younger age, these farmers with less education have been trained earlier by their

families on postproduction work and hence, are accustomed to drying, which was common

practice before the indirect buyers thrived in the Philippines. Therefore, these farmers are more

likely to sell dried paddy rice or milled rice, which are better sold to direct buyers. By contrast,

those with higher education are motivated to sell fresh paddy rice at the farm gate to indirect

134
The Different Factors of Marketing Channel Choices

buyers because most of them started farming when indirect buyers were easily accessible and

are thus, not accustomed to drying their produce.

The qualitative interviews suggest that health is another personal factor affecting marketing

choices. Some farmers mentioned that their health issues, e.g., hypertension, kidney problems,

and disability, limit them from drying the paddy rice themselves. Given their low willingness

or inability to pay for hired labour because of its costliness, they can only sell freshly harvested

rice, which is generally bought at the farm gate by indirect buyers such as agents and

assemblers.

The results of the quantitative model also demonstrate that preference for fast payment

(WantsFastcash_D) and free transportation services (WantsPickUp_D), and work motivation

(HierarchyOfWorkMotivation_O) are significant predictors of marketing channel choices.

Specifically, the relative log odds of selling to direct buyers decrease by 6.69 and 3.1, while

the probabilities of selling to direct buyers decrease by 17.29% and 8.01%, as the farmers are

inclined to receive fast full payment and free transportation services, respectively. The

qualitative data suggest that even farmers who are not in dire need of cash prefer to be fully

paid in cash immediately to avoid inconveniences or just because of over-enthusiasm to get

their income. In fact, this personality was demonstrated in an anecdote from the qualitative data

when a farmer joined the “hybrid rice derby”, a nationwide competition for the highest hybrid

rice yielder, where the winner gets a cash price and trophy. On the one hand, the company

provides all the inputs, particularly seeds and technical support. On the other hand, the farmer

provides the land, labour, and the promise to strictly follow its production procedure to get the

variety’s optimal yield. Overall, the farmer followed all instructions except harvesting since he

threshed the grains a few days earlier than recommended, resulting in non-optimal weight. He

did this despite the company's pleading to wait for a few more days to achieve the grain’s full

135
The Different Factors of Marketing Channel Choices

maturity. The farmer was eager to turn his paddy rice into cash, even if he did not spend a

single cent on the production, thereby having no agricultural credit to pay. He didn’t even

consider the chances that he could win the contest, the extra cash from the prize money, and

the company's favour.

This anxiousness to get paid usually results in selling to indirect buyers. Based on the data, this

is because direct buyers, such as the NFA and cooperatives, typically pay in cheques, which

take more time and effort to encash. Selling to trader-millers also has a lengthy process due to

seller queues and the accounting involved ‒ especially in the large rice mills. Similarly, selling

milled rice to consumers is considered slow because it is demand-based while selling to

retailers is usually consignment. In general, the desire to be paid fast in full is a substantial

determinant for the farmers when choosing a marketing channel. One farmer even related that

he used to sell to a trader-miller who offers the highest price in the area but pays after three

days. This delay made him move to another buyer, who pays a little less but promptly.

Furthermore, results show that selling to direct buyers is less likely as the farmers are inclined

to receive free transportation services. The free transportation service is a decisive factor

affecting marketing channel choices and is almost always the deciding factor when social and

cultural factors are not in place. Based on the qualitative data, most farmers sell to indirect

buyers offering free transportation services, regardless of whether they have the best price since

they pick up the harvest without volume requirements. By contrast, direct buyers impose a

minimum volume for free trucking services. They are not very keen on picking up crops from

the field, so they just offer delivery incentives to their rice mills to compensate for the trucking

cost. Yet farmers don’t find this appealing because 95.67% do not own a vehicle suitable for

trucking, and hiring trucking services is inefficient given their small volume. They can only

make trucking cost-efficient if they pool their harvests with other farmers, which requires a lot

136
The Different Factors of Marketing Channel Choices

of effort. Apart from that, many farmers believe that the additional gain from trucking their

yields is not enough to compensate for the transaction costs ‒ particularly their effort and time.

In fact, many of them are only willing to transport their harvest on their own when the price

difference would result in a net income of more than PHP 0.5 (USD 0.01) per kilogram. This

shows that the inclination to choose indirect buyers, who provide free transportation services,

can also be related to labour disutility, as evident from their choice for convenience over price.

Hence, labour disutility is another crucial factor, which is a personality trait affecting the

farmers’ buyer choices.

In addition, the farmers’ work motivation in reference to Maslow’s Hierarchy of Needs

(HierarchyOfWorkMotivation_O) proved to be significant when choosing a marketing

channel. Results show that the log odds of selling to direct buyers decrease by 1.43, and its

probability decreases by 3.70% as the motivation of the farmer for working hard falls in a

higher category under the Maslow’s hierarchy of needs, which starts at basic needs, followed

by psychological needs, and then by self-fulfilment needs. This suggests that farmers will tend

to sell to direct buyers when their goal is to fulfil their basic needs, such as food, shelter,

clothing, and education. By contrast, farmers are less likely to sell to direct buyers when their

work motivation is for psychological needs, which include the need for love and close

relationships with family and friends. Consequently, farmers are even less likely to sell to direct

buyers when their motivation is for self-fulfilment needs, which is the need to make their lives

meaningful or rewarding, usually by helping others.

This pattern was supported by the qualitative data, particularly when Participant 1 (2019) said

that they are “not after the highest price and allow the buyer to dictate the price so they can

earn”. Similarly, this result was exhibited in the case Participant 28 (2019) who said

“everything I do is for God's glory”, so she tries to treat others with kindness and ensure that

137
The Different Factors of Marketing Channel Choices

she doesn’t step into another person in all her marketing strategies. Thus, she doesn’t mind

sharing a part of her income with agents even if she can directly sell to the trader-miller and

get the agent’s commission.

Another personal trait affecting the farmers’ marketing channel decisions unveiled during the

qualitative interviews was risk avoidance. While this was somewhat proxied under the money

management variable, where the farmers are considered risk-loving if they would instead invest

their income in business and risk-averse if they would save or spend it, the results did not prove

significant since selling to both direct and indirect buyers have risks.

For instance, many farmers do not sell to indirect buyers, especially new ones, because of the

risk that they may be scammers. This was especially true after they heard stories that an

assembler from another province ran away with the farmers’ harvest without paying. They are

also suspicious that their dealings may not be truthful, particularly when it comes to their

weighing scales. Thus, many farmers do not engage with “stranger-buyers” even if they present

an opportunity for higher income.

Similarly, some farmers do not sell to direct buyers because of the risk of not getting the

expected price. They explained that some trader-millers usually do not stick with the pre-agreed

price and that they usually change their price within a day or even hours. Thus, even if they

have a pre-agreed price, many greedy trader-millers lower the prices of the harvest when they

get the actual volume because they know that the farmers will not take the harvest back given

the costs associated with it. This is evidenced by the case of a farmer who sold his paddy rice

to a trader-miller and was promised payment within one week. When he received the payment,

the farmer was surprised to receive less than the pre-agreed amount. Apparently, the buyer

lowered the price without prior notice, saying that the sold paddy rice was of lower quality

upon further inspection. The farmer could not do anything because it would be costly to pull

138
The Different Factors of Marketing Channel Choices

the stocks back, and it was also too late. A similar incident happened with another farmer who

brought samples of their paddy rice to a trader-miller for pricing. After agreeing on a price, he

rented a truck to deliver the pooled harvest of his mother, brother, and himself. Upon delivery,

contrary to the usual practice, each sack was again classified and graded. Worse, he ruled the

harvests as of lower quality, and he was offered a lower price. The farmer related how bad that

experience was. Since then, he has stopped selling to trader-millers, especially if he must

deliver himself.

Similarly, some farmers related that selling dried paddy rice or milled rice to direct buyers also

involve risk owing to price uncertainty. One farmer explained that sometimes the price of

milled rice suddenly goes down because of government intervention or oversupply. Others

explained that risk could likewise come from storage because the stockpiled dried paddy rice

can go bad – mostly if not appropriately dried or when infested because of poor storage

conditions.

Finally, farmers also consider the risk when delivering the harvest to direct buyers. For

instance, some farmers are unwilling to do trucking by themselves despite having

transportation means because of the risks associated with travelling in their area. Thus, instead

of delivering to a trader-miller, they instead sell to the buying stations in their locality to avoid

the risks of being robbed in the savannah.

Overall, the qualitative interviews show that many of the farmers’ marketing choices were

rooted in risk avoidance. Even choosing a buyer because of loyalty is sometimes rooted in risk

avoidance rather than culture or price. For example, some farmers related that they sell to

specific buyers continuously, regardless of price, and remain loyal to them so that they will

buy their harvest even when it has turned bad, muddy, or too wet, or in times of peak harvest

when most rice mills stop buying because of full storage. Similarly, risk avoidance is one

139
The Different Factors of Marketing Channel Choices

reason a farmer would sell to a friend or a relative buyer regardless of the lower price. They

related that they want to help grow their business so that they can help them in times of need,

especially during financial hardship.

By contrast, there are farmers who are risk-tolerant and not worried about safety nets. Hence,

they sell to the highest buyer, especially if the price difference is significant. They do not hold

on to social ties nor the suki system to have a buyer during the peak season or when the harvest

quality is bad or to have a creditor in times of financial distress. These farmers explained that

they always choose the best price, just thinking of the present. If they get hit by a typhoon and

their harvest gets muddy, they will find the highest price possible for that harvest. They

acknowledge that traders might buy their rice at a lower price or even refuse it, but they believe

that is not something that should stop them from choosing the highest buyer at the moment

because those times are fewer than the good ones; hence, they will lose a lot if they live in fear.

Similarly, another farmer described that he always tries to negotiate the price with his creditor-

buyer. In times that the creditor refuses to buy it at his desired price, he then sells it to another

buyer. He explained that he is not afraid of losing that creditor because many other buyers and

creditors would be happy to lend him money since his credit track record is good. Finally,

farmers that are risk-tolerant tend to accept more risky buyers and buying terms.

Other personal traits were highlighted during the qualitative interviews, which affect marketing

channel choices; being impulsive is one of them. Some farmers are impulsive spenders, so they

do not want to come to the centre to scout and sell to the highest buyer. This trait was

exemplified by one farmer in Mindanao, who specified that the buying prices at the town centre

are higher than in their village, yet he still prefers to sell in the latter to avoid unnecessary

spending, which usually happens when he goes to the city. The farmer elaborated that the last

time he went to the centre to sell, he earned an additional PHP 500 (10 USD) but ended up

140
The Different Factors of Marketing Channel Choices

spending double because he was drawn to buy many things, which he would have otherwise

not bought if he was not in the city with his farm income.

These are but some of the personality traits affecting farmers' marketing channel choices,

which, when coupled with other factors discussed in the following sections, can significantly

impact decisions.

6.2 Economic Factors

The econometric model shows that among the significant economic factors affecting the choice

of direct versus indirect marketing channel choice are the fixed income source (Incomesource),

off-farm income (OffFarmInc_W), harvest frequency (HarvestFrequency), landownership

(Landownership_D), largest farm size (LParea_W), and distance to trader-miller (DMiller_W)

(Table 8; Appendices 3 & 4). Specifically, the log odds of selling to direct buyers decrease by

3.68 or its probability decreases by 9.5% as the farmer acquires an additional source of income

that is fixed ‒ usually a regular job. Based on the qualitative interviews, employed farmers

would rather sell to indirect buyers because selling to direct buyers is time-consuming. They

typically sell at the farm gate to indirect buyers because they do not have much time to spare

for drying and milling or to canvass buyers who provide more competitive prices because of

their regular jobs. These farmers regularly work for eight hours a day, so they cannot cultivate

the rice themselves; hence they have a hired steward or caretaker. They just visit during crucial

production processes such as planting, fertiliser application, and harvesting.

Moreover, the qualitative data show that these employed farmers do not think that the

additional income they will get from drying is worth their time. In fact, to dry their harvest,

some of them want as much as PHP 3 (USD 0.06) net additional income per kilogram that will

translate to an additional net profit of PHP 12,000 (USD 240). That amount is already

141
The Different Factors of Marketing Channel Choices

equivalent to 39.67% and 51.32% of their wet and dry seasons’ net income, respectively,

demonstrating how pricey they regard their drying labour to be.

By contrast, the value of the total household off-farm income (OffFarmInc_W) or any income

coming from other sources apart from rice is positively related to selling to direct buyers.

However, the effect is minimal with the log odds and the probability of selling to direct buyers

increasing by 1 and 2.59%, respectively, for every PHP 100,000 (USD 2,000) increase in off-

farm income. As mentioned by the farmers during the qualitative interviews, it is because their

higher disposable income enables them to pay for postproduction processes that make selling

to direct buyers more ideal. While this positive relationship appears to sightly contradict the

effect of fixed employment, note that this variable accounts for all the household income.

Hence, it is possible for the farmer not to have a fixed job while having a high total household

income that could come through other household members or a business. Based on the data,

24.2% of the surveyed population have four adults, 22.9% have three, 13.1% have five, 7.95%

have six, and 4.1% even have seven. Moreover, there are 1.9% of households with 8 to 12

adults. While not all of them may be working, this set of information suggests that many of the

household members can work. Thus, the 0.204 significant correlation at the 0.01 significance

level between the number of adults in a household and higher off-farm income.

The qualitative data support this with farmers living with other employed household members

reporting that they are not significantly constrained by money since they can get temporary

financial aids from their children who live with them. Overall, 72.9% of the farmers with at

least an off-farm income of PHP 33,000 (USD 660), which is roughly the cost of rice

production per season (Mataia et al., 2018), did not have to borrow capital, which usually

comes from indirect buyers. Thus, they were not compelled to sell to them and instead, they

had the ability to wait longer, dry, stockpile, or mill their harvests and sell them to direct buyers.

142
The Different Factors of Marketing Channel Choices

The harvest frequency (HarvestFrequency) was also found to be a significant factor.

Specifically, the log odds of choosing direct marketing channels increases by 3.76 or the

probability increases by 9.7% for every 0.5 increase in cropping frequency (Table 8;

Appendices 3 & 4). The higher frequency of harvest proxies for more harvested grains per year;

that results in higher income and gives the farmers the capacity to self-finance their

postproduction work. Thus, the positive relationship can also be explained primarily by the

farmers’ higher disposable income as affirmed by the 0.072 positive correlation coefficient

between having their own capital and higher frequency of harvesting at the 0.05 significance

level (Appendix 6). Moreover, the data show that this variable also has social implications

since the farmers mentioned that the frequency of their sales is instrumental in having a good

social relationship with the trader-miller in the area, which results in favours such as free

harvest pick-up and special prices. This is affirmed by the positive correlation (0.092) at the

0.01 significance level between the frequency of harvest and the number of times a farmer sold

to the same buyer, which is expected to increase social capital (Appendix 6).

The relative log odds of selling to a direct versus indirect buyer increase by 3.47 or the

probability by 8.97% as the farmer moves from being a land tenant to an owner

(Landownership_D) (Table 8; Appendices 3 & 4). Based on the qualitative interviews, farmer-

tenants lack the financial capacity to sell to direct buyers because they get limited farming

income given the rent they must pay, which can be as high as 50% of the net profit. By contrast,

the landowner-farmers get the farm’s full harvest so apart from having a bigger volume to sell,

they can better afford to process their rice and sell them to direct buyers for a higher profit.

This better liquidity also minimises their chance of borrowing farm production capital, as

supported by the 0.086 coefficient correlation between land ownership and having enough

production capital at the 0.01 significance level (Appendix 6).

143
The Different Factors of Marketing Channel Choices

Furthermore, land ownership gives the farmer more incentive to sell to direct buyers and build

a stable buyer-suki relationship with them as they are sure to cultivate the same land for a long

time. Similarly, direct buyers are more eager to offer them the best price, with some even giving

them favours to win their loyalty because of the possibility of long-term relationships. The

farmer-landowners related that their suki relationship started with them trying to bargain prices

by saying that they would be their suki since they own the land. Thus, the relationships they

have with the direct buyers have been built through time since they are irreplaceable, unlike

tenants. In this study, 77.5% of the farmers with their own lands have a suki relationship with

their direct buyers, such as trader-millers (44.2%) and cooperatives (6%). Therefore, this

variable also has social factor implications separate from economic consequences.

Overall, variables such as higher off-farm income, more frequent harvesting, and land

ownership have positively influenced the choice for direct buyers primarily because of higher

disposable income or better liquidity. A fixed income source and higher household income

have significant correlation coefficients of 0.120 and 0.187, respectively, with having their own

capital at the 0.01 significance level (Appendix 6). This means that farmers who have higher

disposable income do not depend on credit. Thus, they do not depend on creditors, which both

the quantitative and qualitative data demonstrate as a crucial determinant of marketing channel

choices. In fact, 22.22% of the marketing channel choices were made because of credit and

majority of the farmers (61%) with credit sold to indirect marketing channels, i.e., agents,

assemblers, buying stations, and creditor-traders (Table 9).

Table 9. Distribution of the marketing channel choices of farmers with credit.

Marketing Channel Frequency Per cent

Agent 17 8.5
Assembler 13 6.5
Buying station 87 43.5

144
The Different Factors of Marketing Channel Choices

Creditor-trader 5 2.5
Cooperative 9 4.5
Trader-miller 69 34.5
Total 200 100.0

Moreover, the findings suggest that farm size (LParea_W) negatively affects the choice of

direct marketing channels, which is contrary to the predicted effect. Specifically, the log odds

of selling to indirect buyers is higher by 0.9 or the probability increases by 2.33% for every

hectare increase of farmland. This suggests that larger volumes result in selling to indirect

buyers.

Some of the qualitative interviews explain this. First, large-scale farmers or those with a farm

size larger than two hectares (FAO, 2017) sell to indirect buyers because they can bargain for

prices at par with that of trader-millers. Yet, they can get fast payment and free transportation

services. They explained that because they have a large farm size, indirect buyers, especially

the travelling assemblers from other provinces, are very eager to buy from them because of

their harvest volume. This helps the assemblers shorten the collection period to fill a truckload;

thus, they are willing to pay them at par with the trader-miller’s price.

Second, 83.3% of the large-scale farmers do not have the means to deliver all their harvests in

one trucking, unlike small-scale farmers who can use a hand tractor, tricycle or even a jeepney

to deliver their produce to trader-millers, retailers, or consumers. Further analysis revealed a

negative 0.129 correlation coefficient between delivering the harvest to the buyer and the total

rice farm's size at a 0.01 significance level (Appendix 6). This further affirms that farmers with

a larger farm size, which usually translates to larger product volume, are not transporting their

harvest themselves and vice versa. This likewise suggests that even farmers with large volumes

do not prefer to deliver their harvest even with the advantage of economies of scale.

145
The Different Factors of Marketing Channel Choices

Third, their large volume requires a huge drying area and more workers that are limited during

the peak harvest season. In particular, the volume affects the farmer's drying capacity, whether

it be in terms of drying facilities or labour. For instance, one farmer is more willing to manually

dry his harvest when it is less than 100 sacks, which he sells to direct buyers of paddy rice, e.g.,

trader-millers, cooperatives, NFA. Another farmer explained that they could not afford to dry

a large volume of harvest given the narrow road they have available to use for drying and their

limited household workforce. Thus, they are more willing to dry, mill, and sell to consumers,

even without mechanical dryers, when the volume is around 50 sacks or less. This is also

because a larger volume would require partnerships with wholesalers or retailers, which the

farmers do not usually have.

Fourth, the volume also affects the risk-aversion of farmers. Notably, some of them are not

willing to deliver a huge harvest volume because of risk. For instance, a farmer said that he

only delivers to the trader-miller when he has just about 100 sacks. However, if it is 200 sacks

or more, he would not risk delivering it on his own even if he has his trucks because the roads

are unsafe. Thus, he would rather have it picked up and not get the maximum profit just to

secure the money and avoid the risk of losing his harvest along the way. Thus, if the trader-

miller refuses to pick up the crop within a specific timeframe, considering the perishability of

fresh paddy rice, he said he would instead break the volume into smaller portions and distribute

it to the indirect buyers in his town since none of them can handle all his harvest. He said, “it

is better to lose some money to those people because they are working on getting an income

than to bandits preying on the travellers at the prairie”.

The distance to the nearest rice trader-miller (DMiller_W), which proxies for transaction cost,

also exhibits a significant effect on the choice between selling to a direct versus indirect buyer.

The findings demonstrate that the log odds of selling to a direct buyer decrease by 0.41, or the

146
The Different Factors of Marketing Channel Choices

probability of selling decreases by 1.06% for every kilometre increase in distance to the nearest

rice trader-miller. This provides evidence that the closer the farmers’ fields are to the direct

buyers, the more likely they are to sell to them. During the qualitative interviews, when asked

why they sold to their buyers, many farmers simply answered, “it was close”. Overall, the

qualitative data suggest that farmers especially consider the buyer's distance when canvassing

and when they are doing the trucking themselves. It enables them to deliver their harvests

themselves because of the more affordable trucking. Sometimes, they just ask the public

transportation vehicles to take it to the rice mill for a person’s fare per sack. Shorter distances

mean low transportation costs, which also encourage the trader-millers to provide free harvest

pick up.

Similarly, the shorter distance to the trader-miller makes it cheaper and easier to process the

paddy rice into milled rice, which they sell to direct buyers. The higher probability of selling

to a direct buyer when a trader-miller is close-by could also be because of the variable proxies

for better infrastructure or postproduction facilities available in the area. Farmers living close

to the trader-millers mentioned that they have roads and pavements where they can dry their

harvests when the weather is favourable. This is backed by the -0.128 correlation between the

distance to the trader-miller and good road conditions, which imply that the farmers living

farther away from rice mills indeed have bad roads that are often not cemented, muddy, or

stony (Appendix 6). Most of the time, the trader-millers provide drying services that are easier

to access given the shorter distance. Furthermore, the short distance, which can be as close as

a few meters (being next door), helps create a social relationship that results in a special price

for the farmer’s harvest.

Finally, price is a factor that came predominantly from quantitative data. About 46.7% of the

farmers are believed to have sold to specific marketing channels because they offer the highest

147
The Different Factors of Marketing Channel Choices

prices. This is also the most common initial answer that farmers gave during the qualitative

interviews as influencing their buyer choices. An example of this is a farmer who has a daughter

who buys paddy rice and then sells it to a trader-miller for a brokerage fee. Without hesitation,

this farmer would just sell directly to the trader-miller to cut out his daughter from the market

value chain and save some cents. He explained that any increase in buying price would benefit

him and help in his next cropping. While recognising that his daughter needs to earn for her

own family, he said that he also needs to maximise his profit to feed his wife and remaining

children at home. Thus, he sells to whoever the highest buyer is.

Other farmers similarly choose the highest buyer versus their buyer relatives or friends. They

said that if “business is business”, they do not have to take whatever offer they give, especially

if it is not the highest in the market. Nevertheless, not all farmers prioritise price. Others sell

because of non-price factors such as social ties or cultural values, which are discussed below.

6.3 Social Factors

There are also social factors in the quantitative model that proved to be important when

choosing direct versus indirect buyers – for instance, training (Training_O). Specifically, every

five additional attendances to a training, seminar, technology demonstration, farmers’ field

school or anything similar results in 0.68 higher log odds or 1.75% higher probability of

choosing direct marketing channels (Table 8; Appendices 3 & 4). This implies that knowledge

sharing and learning venues or activities are an excellent opportunity to improve the farmers’

production practices and expand their networks. The positive relationship between attendance

to these activities and choosing direct marketing channels suggests that an increase in the

network of the farmer through training can lead to selling to buyers who offer higher prices.

While this was never directly mentioned in the qualitative interviews, training allows farmers

to expand their network by establishing a connection with like-minded farmers who are keen

148
The Different Factors of Marketing Channel Choices

on increasing their income, whether through higher productivity or better marketing practices

(PhilRice, 2016). More specifically, training extends the farmer’s social network that could

help them access more direct buyers and more affordable service providers or even share new

marketing strategies. It can even lead them to better creditors who may give them more freedom

to sell to any buyer ‒ especially because they are usually briefed about government credit

programs during those events. On the one hand, the qualitative data show that family members

are the primary influencers of farmers when choosing their marketing channels, while friends

or co-farmers are secondary. On the other hand, the quantitative data demonstrate that their co-

farmers, whom farmer-respondents could have met during training, introduced 31% of them to

their current suki. Specifically, these sukis are 56% direct buyers, with 51.5% rice trader-millers

and 4.5% retailers/consumers.

Membership in an organisation (NumOrgMem), however, was found to be an insignificant

factor. This could be because the number of memberships does not translate to the number of

networks, especially since some members do not really attend meetings or trainings; they are

inactive members and are just members on paper. This was common among the farmers

interviewed, with most confessing that they are members of an organisation, but they have

never attended even a single meeting. They just became a member because the agricultural

technician or a farmer-leader in the area included their name to create a group. This is especially

true in the Philippines, where unions, particularly cooperatives and associations, are non-

functional and only organised to get government support and projects. In fact, while there are

34,203 registered agricultural cooperatives in the Philippines, many of them are inactive and

too small to become successful (EFSIM, 2009). This – the inactivity and the namesake to get

government support – may also be the case for agricultural associations. This could be one

reason why it yielded a non-significant effect on marketing channel choices that is contrary to

149
The Different Factors of Marketing Channel Choices

the significant positive results found by Mmbando et al. (2015) and Cazzuffi and McKay

(2012) with market participation.

While the econometric model did not show many social factors, the qualitative interviews

revealed how important they are, particularly social relationships, when choosing a marketing

channel. In fact, while most farmers initially said that price is their primary factor when

choosing a marketing channel, many of them afterwards realised that social factors are critical

to them such that it almost singularly determines their choices. For instance, despite claiming

that price is the most important factor, many farmers overlook it when a relative offers to buy

their paddy rice, saying “blood is thicker than water”. They believe that they should always

support the business of a family member or relative rather than others. They explained that

selling to a relative, even at a loss, is alright because their loss is the gain of a relative who

someday could help them. The importance of the social relationship was exhibited by a farmer

who, despite their meagre farming income, still chose to sell to their relative even if there were

better options. She explained that is the tradition that her father-in-law has taught them – to be

“die-hard loyal” – because their marketing practice is not based on money but values. She

further explained that their relationship could not be equated to any price because apart from

being cousins, he is also her ritual kin through godparenthood. Thus, while they need money

themselves, she just sells to him to show support and because it gives her self-respect. She is

one of the farmers who emphasised the importance of self-respect and self-esteem during the

qualitative interviews, inferring that because they are poor, they only have their image to

protect and the respect of others to keep. And while loyalty to relatives is a social imperative,

many farmers also reported some benefits from selling to them. These were considerate pricing

and sometimes additional income, having a sure buyer in times of a bad harvest, free harvest

pick-up, and credit access that can sometimes be interest-free.

150
The Different Factors of Marketing Channel Choices

Similarly, many farmers would forego price to favour friends who buy paddy rice, believing

that these real friends will also return their favour someday. This is not just true for large-scale

farmers who can afford to lose a portion of their income, but more so for small-scale ones. In

fact, 46% of the farmers with two hectares or less are selling because of relationships, which

is lower than the 43.7% of the farmers with more than two hectares. An example of this is the

Luzon farmer who reported that he continually sells to his friend, even when the price

difference from the other buyers was as high as PHP 0.3 (USD 0.006) per kilogram, believing

that it is harder to buy a friend than to earn money. However, based on the data, many farmers

remain that loyal to their buyer-friends because of other advantages such as higher prices, better

communication, good relationship, reliability, fast and easy credit access, and sometimes, even

interest-free credit.

Meanwhile, social factors are not always the determinants of a marketing channel choice for

some farmers. They are only most important to them up to a distinct price difference. Thus,

they do not outrightly sell because of social relationships. They may or may not overlook the

price difference depending on the weight of the association at stake. These farmers weigh their

potential loss or forgone benefit against social relationships by assessing whether the additional

profit would compensate for the loss of a friend, the dismay of a family member, or the

inconveniences of breaking a suki relationship.

6.4 Cultural Factors

The econometric model's cultural factors also showed a significant effect on the choice of the

marketing channel. The results suggest that the log odds of selling to direct versus indirect

buyers increase by 10.01, or the probability increases by 25.86% when a farmer belongs to an

indigenous group (Table 8; Appendices 3 & 4). Specifically, 61.7% of the farmers who belong

to an indigenous group sell to direct buyers, mostly to trader-millers (45%). While these

151
The Different Factors of Marketing Channel Choices

farmers did not attribute any of their decisions to their indigenous roots, the data provide

evidence that there is a positive correlation (0.107) between membership in an indigenous

group and price motivation at the 0.01 significance level. It means that these farmers are selling

to the highest buyer that are usually direct buyers. According to one farmer, this is because

they grew up with the mindset that they need to work hard to give their children a good

education to live more comfortable lives.

Furthermore, other cultural capitals such as language, accent, or origins were not mentioned

during the qualitative interviews as the main reason for selling to specific buyers. Nevertheless,

the quantitative data show that cultural capital can be instrumental in the pricing since these

can eventually help create a good relationship between the two parties, as exemplified by most

farmers with suki who share the same dialect, accent, or origin with their buyers. As Throsby

(1999) described, the intangible cultural capital was not the farmers' primary consideration for

initially choosing the buyer. Nevertheless, the farmers said that cultural similarities helped

foster and strengthen the relationship that developed into sukis and allowed them to get better

prices eventually.

In addition, the practice of farm traditions also displayed a significant effect on the choice of

marketing channels. Specifically, the log odds or the probability of choosing direct marketing

channels increase by 3.81 and 9.84%, respectively, as the farmer keeps their forefathers'

traditions. Based on the qualitative data, many farmers who used to follow or are still pursuing

their predecessors' farming rituals are likewise selling to direct buyers because that was their

fathers' marketing practice. They follow many cultural traditions or rituals such as using a

particular calendar, planting from the four corners of the field, hanging a bunch of paddy rice

on the tree, and taking home the first sack of harvest and placing a cross with three small stones

inside it to get a high yield. Following these processes exhibits how loyal these farmers are to

152
The Different Factors of Marketing Channel Choices

their traditions and how much they value the techniques taught by their forefathers, believing

that these would help improve their yield. As such, they are more willing to follow their more

popular strategies to get a higher income, such as selling dried or milled rice to direct buyers.

The quantitative data concur with this – with a positive correlation (0.082) between having

farm traditions and selling more processed rice (i.e., dried paddy rice, milled rice) at the 0.05

significance level (Appendix 6).

Furthermore, the qualitative data provided evidence for more influential cultural factors. These

are the values such as debt of gratitude, the culture of shame, and loyalty. Based on the

interviews, farmers develop debt of gratitude (utang na loob) towards their past or present

creditor, regardless of whether they impose interest on the credit or not, since the creditors

often take pride in the fact that they helped them either way. Thus, in return, the farmers feel

that they must support their creditor’s business. In general, all the farmers are aware of the

credit interest separate from their creditors' generally lower buying price. Many of them realise

that the creditors are making good profits and getting rich because of them. Yet the cultural

factor of debt of gratitude, which takes root from treating the credit as a personal favour, is so

strong that many farmers still sell to them without minding their lower buying price.

Nevertheless, incorporating interest rates usually makes the debt of gratitude last only until

after the money has been repaid. There are only a few instances when the farmers truly consider

the favour as a debt of gratitude in its most valid form, which lasts a lifetime. An example of

this is the Luzon farmer, who used to sell to the highest buyer because he is a businessman.

But this changed when he became deeply indebted to a trader-creditor because he borrowed

money from him when his wife was hospitalised and again later to finance his son's schooling.

Thus, he has committed to sell to that buyer consistently regardless of any price difference –

saying that he is even planning to take that debt of gratitude to his grave.

153
The Different Factors of Marketing Channel Choices

Another cultural factor that affects the marketing channel choice is the culture of shame.

According to some farmers, they sell to their relatives, friends, and acquaintance not because

of social capital but only because of shame. They feel it is shameful not to support their

business, even if they are not helping them in return. Not selling to them makes them feel less

self-respect or self-esteem and would make it hard for them to face them again in the future.

Some of them even said that pride is all they have, and they do not want to lose it by selling

because of money. One farmer explained that he sells to buyers with whom he has close

associations because he does not want to be branded as “mukhang-pera” or money-driven.

Likewise, many farmers disclosed that they are selling to more than one buyer because they

want to face them all without embarrassment, which shows how influential the culture of shame

is for them.

The third cultural factor is loyalty, which results in the suki system. Based on the quantitative

data, farmers have been suki or selling to their buyers for as long as 28 years because of loyalty.

This relationship is strengthened by the two parties' continuous market exchange, usually

motivated by credit or a good price. Nevertheless, when fully developed, this loyalty makes

some farmers sell to their suki buyers regardless of whether they are the highest buyers or

provided them with credit. They stay loyal and do not break this economic relationship for

many reasons, including plain convenience or just to prevent discomfiture. In general, this

loyalty makes many farmers stick with their buyers for a long time despite better price options

or social relationships, unless they have been wronged or too disadvantaged.

6.5 External Environment Factors

The results of the quantitative analysis similarly provide evidence that environmental factors

such as town income class (TownIncClass_O) and rice production volume (RiceProdVol_W)

are both significant predictors of marketing channel choices (Table 8; Appendices 3 & 4).

154
The Different Factors of Marketing Channel Choices

However, apart from these, the farmers mentioned other external environment factors during

the qualitative interviews. These factors are categorised into two – the task and the general

environment, which directly and indirectly affect the marketing channel choices.

6.5.1 The Task Environment Factors

The task environment factors highlighted in this study include the buyers, competitors (co-

farmers), labour market, and suppliers. Described below are how they directly affect the

marketing channel choices of farmers.

Buyers. Daft (2008) referred to this group as customers or those who acquire the goods and

services. Given that definition, these are the rice buyers/traders because they acquire the

farmers' produce. Overall, these buyers vary widely from as high as fifty to just one based on

the data. For instance, some areas cultivate premium rice varieties, so agents and assemblers

flock there during the harvest season resulting in competitive prices. Other towns are far-flung

and isolated, e.g., on the other side of a river, resulting in a monopsony.

Similarly, the type of marketing channels available also varies per area. Other places have good

access to many huge trader-millers being at the top rice-producing province. By contrast, some

sites have no access to trader-millers, so farmers must settle with buying stations or assemblers.

Moreover, agents are indispensable in some areas, whereas they are not common in others.

Furthermore, some towns only have fresh paddy rice buyers, other towns only have dried paddy

rice buyers, while some have minimal milled rice buyers since most households have their own

rice farms.

The competition of buyers and the availability of different marketing channels largely affect

the farmers’ choices. For instance, if there is only one available buyer, then the farmers have

no choice but to sell to that buyer unless they can process and market milled rice. Similarly, if

155
The Different Factors of Marketing Channel Choices

most town residents have a supply of rice from their farm or a relative’s farm, then the farmer

has limited buyers for milled rice. The case is similar for farmers in areas with only buyers of

fresh paddy rice, as their choices are limited to indirect buyers.

The number of fresh paddy rice buyers is proxied using the volume of rice produced in a

province (RiceProdVol_W) since it is assumed that there are more buyers in the top rice-

producing areas. Overall, the results show that when the volume of rice produced in the

province gets higher by 100,000 metric tons, the log odds of selling to a direct buyer decrease

by 0.46. Conversely, the probability of selling to direct buyers decreases by 1.18% (Table 8;

Appendices 3 & 4). This negative relationship implies that an increase in the number of buyers

in an area encourages farmers to sell to indirect buyers at the farm gate. The PSA (2015) data

affirm this – with their data showing the lengthy market value chain of the farmers in the top

rice-producing provinces because of the involvement of many indirect channels. Many

travelling assemblers often visit these provinces to buy fresh paddy rice at competitive prices.

This was frequently mentioned during the interviews with farmers from Nueva Ecija, Isabela,

and Bulacan, the Philippines' top rice-producing provinces. The farmers revealed that the

buyers in the area are so numerous, especially agents and assemblers, that they do not have to

leave their farms anymore because they come to them with irresistibly convenient offers. Thus,

although they offer slightly lower prices for fresh paddy rice than trader-millers, the

convenience and their presence at the farm gate make farmers sell to them – particularly those

who are old, with frail health, with fixed employment, and with labour disutility. Furthermore,

numerous agents in these provinces likewise increase farmers' chances of having a relative or

friend that is an agent or local assembler. Hence, this also increases the probability of farmers

who highly regard social and cultural capital to sell to indirect buyers.

156
The Different Factors of Marketing Channel Choices

Competitors. This refers to the rice farmers' population selling their produce per cropping

season, which is subject to the timing of the harvest. Off-peak harvesting would mean a smaller

number of competitors, while peak harvesting means higher competition. This timing affects

farmers' access to marketing channels and the prices of the goods, as dictated by the law of

supply and demand. For instance, farmers related that they often sell to trader-millers during

the off-peak season because they are less strict when classifying during that time. By contrast,

it is difficult to sell to them during the peak harvest season because they scrutinise each sack's

quality thoroughly and usually stop buying fast, having met their storage capacity. Apart from

that, most farmers can only sell to assemblers during peak season because that is the time that

they are in town. Likewise, only a few farmers can afford to wait to sell dried paddy rice or

milled rice during the off-peak season, which results in a higher price because of the lower

competition. Lastly, the competition also affects the marketing channel choices through labour

supply availability. This is described in more detail in the following section.

Labour market. The availability of people who can supply labour for rice farms significantly

affects the marketing channel choices because it determines the final product form. Some areas

are still heavily reliant on manual labour when harvesting, drying, and hauling. Hence,

labourers are scarce and expensive during the peak harvest season. Some farmers even

mentioned that they got stood up by labourers because another farmer offered them a higher

price. Farmers similarly complained of the many other labour-hire conditions to ensure services

for the next season, e.g., snacks and drinking sessions.

On the one hand, the large-scale farmers who are determined to sell to direct buyers are able to

use the available labour force in the market. They can tap groups of labourers because they

regularly hire them and because they offer more working days considering the volume of their

harvests. On the other hand, the small-scale farmers who want to do postproduction work rely

157
The Different Factors of Marketing Channel Choices

on their household workforce. Those who cannot work themselves or those without any

additional household workforce just sell to indirect buyers at the farm gate, e.g., assemblers,

agents, and buying stations. This is to avoid issues relating to the low labour supply, including

the fees and other conditions, along with the inconvenience of finding labourers one by one.

The lack of porters and pricey services also persuade farmers to sell to those waiting in the

field. The farmers explained that this saves them from paying hauling fees because they directly

collect their harvests from the farm. Overall, the availability and affordability of the labour

force in the market is a considerable determinant of whether a farmer can do postproduction

processes, and consequently, sell to direct buyers.

Suppliers. This is defined as the providers of raw materials (Daft, 2008), which, in this case,

are the input suppliers. Nevertheless, it can similarly refer to a supplier of service, specifically

of postproduction facilities for drying and milling, transportation, and credit services. Based

on the data, many suppliers lend their inputs to farmers in return for their fresh harvest. In most

of these instances, despite the interest rates, the farmers are obliged to sell to them and do not

have a real marketing channel choice.

Furthermore, the number of cash credit providers in the area also affects the farmers' marketing

channel choices. For instance, having a limited number of creditors usually encourages farmers

to sell and be loyal to them to avail of future credit. By contrast, having numerous creditors

gives the farmers other options that enable them to bargain for better prices or break the

relationship when they are too disadvantaged. Many farmers exemplified these during the

qualitative interviews, thus showing the effect of a competitive credit market on their marketing

channel choices.

Moreover, the availability and accessibility of postproduction facilities influence the form of

the product that the farmers are selling and thus affect the marketing channels available to

158
The Different Factors of Marketing Channel Choices

them. For instance, some farmers do not have access to mechanical dryers or transportation

services. Therefore, they just usually sell fresh paddy rice at the farm gate, generally to indirect

buyers. Some farmers do not have access to trader-millers. Hence, they are missing the

opportunity to sell milled rice to direct buyers. In addition, some farmers do not have access to

storage facilities. Thus, they are missing the chance to stockpile dried paddy rice or milled rice

that are best sold during off-peak season to direct buyers.

6.5.2 The General Environment Factors

This study's findings show that general environment factors such as international, legal-

political, technological, economic, socio-cultural, and natural also affect the farmers’

marketing channel choices, albeit indirectly, as shown below.

International environment. Part of the general environment is the global rice market. At

present, this is significantly affecting the marketing income of the rice farmers because of the

Republic Act (RA) No. 11203 titled “Act liberalizing the importation, exportation, and the

trading of rice, lifting for the purpose the quantitative restriction on rice, and for other

purposes” signed by the Philippine President Rodrigo Duterte on February 14, 2019 (Corpuz,

2019). The law allows the free trade of rice in the Philippines, permitting imported rice, which

is significantly cheaper than domestic produce, to be sold in the country. The expected increase

in the inflow of cheaper imported rice because of the RA 11203 resulted in a significant drop

in paddy rice prices, by more than 50%, starting on the announcement day. It had an immediate

effect on the rice farmers, given that the harvest season starts in March. Farmers were

disappointed, and it was undeniable as they shared their stories during their interviews from

February to May. One effect was their indifference in selling their produce. Most of them were

lethargic in finding better buyers because of the sudden massive drop in the price of their

159
The Different Factors of Marketing Channel Choices

harvest, resulting in low income or even negative income for those who underwent weather

extremities such as drought.

The opening of the country’s rice market to the world has resulted in changes in the farmers'

marketing channel choices. Large-scale farmers who were selling milled rice lost some of their

projected income because of the influx of imported rice. Thus, it discouraged many of them

from selling milled rice after that. By contrast, considering their small harvest volume along

with the very low price of fresh paddy rice, small-scale farmers thought of ways to maximise

their marketing income lest they end up without any net profit. Hence, some of them processed

their harvest into dried or milled rice, while others scouted for the highest buyers in town and

ways to deliver them.

Nevertheless, in general, trade liberalisation has discouraged many farmers from using other

strategies geared towards higher marketing income, such as stockpiling dried or milled rice to

sell during the lean harvest season for the best price. A farmer from Luzon was disappointed

explicitly upon hearing that given open entry and free trade, rice can come in the country at

any time, so there will be no lean months when he can sell his dried paddy rice best. Hence, he

confessed that instead of selling dried paddy rice to millers, he might just sell fresh paddy rice

to assemblers next time because drying will not be as profitable.

Lastly, open trade has turned some direct buyers into indirect buyers. One example is the

farmer-miller in Palawan who used to buy paddy rice, mill it, and sell it to consumers. Yet

because of the rice tariff law, he just bought and sold paddy rice, thereby functioning as an

indirect buyer. He explained that while he wanted to increase his price to be more competitive

and, at the same time, help his fellow farmers, the big trader-millers in the area did not allow

him to do so. They wanted to stop him from opting out of the colluded price. Thus, they

increased their price up to the point that he would only just break even or go bankrupt, given

160
The Different Factors of Marketing Channel Choices

the cost of imported milled rice. Because he could no longer compete with the paddy prices of

his counterparts, he just functioned as an indirect buyer in the area, buying paddy rice from

farmers and selling it to the trader-millers using his small truck. Despite that, however, many

farmers still sell to him because of social capital. He revealed that these farmers do not mind a

slight price difference, mostly because he picks up the harvest from their farms. Overall, this

proves the international market's effect on the farmers’ buyer choices – it limits the number of

buyers to large businesses who are only focused on profit, thereby decreasing the chances of

farmers to sell to direct buyers who are more considerate.

Legal-political environment. Another external environment factor is the legal-political

environment, which includes corruption. One of the forms of corruption in the rice industry is

the politician’s use of government facilities such as dryers and rice mills as if they have

exclusive rights. One farmer in Mindanao related that a local government official in their area

takes all the government facilities or projects to himself. These include drying pavements,

storage facilities, and rice mills constructed inside their gated compounds. Hence they are

using these as if they are privately owned and part of their rice trading business. Thus, farmers

have no choice but to sell to them because there is neither pavement space left for drying, nor

rice mill to rent, or other traders to choose. And while farmers are selling to that corrupt official

– someone who can be considered a direct buyer because he sells milled rice – the farmers are

barely benefiting because of his monopsonistic buying price. Yet farmers sell to him everything

they harvest because of the costly rice milling fee that is around four times more than the cost

in other provinces, making processing of fresh paddy rice an inefficient option. Thus, farmers

do not only sell fresh paddy rice to him but also buy their milled rice from him.

Another aspect of corruption is the delivery of low-quality inputs and services to the farmers,

which affects their harvest and sold volume, thereby affecting marketing channel choices, as

161
The Different Factors of Marketing Channel Choices

described in Section 6.2. For instance, one farmer shared that she received seeds from the

government, which got infested by snails, resulting in substantial production losses instead of

gains. As such, she had to borrow more money from a buyer-creditor, making her even more

indebted to him. This situation left her no option but to sell everything at a price dictated by

the buyer-creditor.

Another example of corruption that discourages farmers from selling to direct buyers is related

to the NFA. Several farmers mentioned that they do not sell to the NFA because of its slow and

biased process. They specifically cited that some of them must wait long periods of time before

their harvest can be classified, while others get it quickly. Similarly, they believe that their

crops are classified unfairly, which seems to depend on their relationship with the classifier or

someone influential in the office. Thus, most farmers do not use the NFA and just sell fresh

paddy rice to agents and assemblers at the farm gate.

Lastly, the corruption within associations and cooperatives has also worsened farmers'

marketing position in some areas. Specifically, the farmers described how several of their

cooperatives and associations have closed because of funds corruption. Thus now, it does not

have the capital to buy the harvests of its farmer-members nor the capacity to assist them in

marketing their products to direct buyers.

Local policies likewise add to the legal-political environment factors. For instance, there are

municipalities or towns in Mindanao with ordinances preventing outsiders from buying paddy

rice within the town’s jurisdiction. Thus now, the farmers are complaining that they do not

have many choices, especially because the prices are all the same due to price collusion. They

recalled that they used to have buyers from other provinces who came to their municipality

during the harvest period. They said that these buyers were offering better prices and providing

fair competition with the local buyers in the area until the ordinance prevented them from

162
The Different Factors of Marketing Channel Choices

entering the town, which resulted in a significantly lower local buying price of paddy rice. The

farmers suppose that this local ordinance was not made to protect them but rather to guard the

interest of those in power who are also in the rice trading business.

Politics, especially in small villages, is another legal-political factor affecting the farmers’

buyer choices. Given some farmers' political aspirations to be village heads or councillors, they

choose to sell to the buyers in their village despite lower prices compared with the buyers in

the city. It is to build social capital and get their votes during the election period. Thus, if there

are two buyers in their village, they will divide the harvest and sell to both to please them and

not create conflict with anybody. These farmers explained that they must support their

residents' businesses to get their votes during the election time. While they do not give financial

support during the campaign period, their electoral votes are good enough for these farmer-

politicians. Thus, they sell to indirect buyers in the village and lose some money in return for

their votes that may or may not be enough to elect them.

Technological environment. Another general environmental factor is the technological

aspect. Many farmers related during the qualitative interviews that before the dawn of the

mechanical reaper or rice combine harvester, people would dry and sell to rice mills to

maximise their profit. But now that they can sell the harvest within just a few hours, they would

rather sell it immediately at the farm gate to an agent or an assembler. They claim that they

profit more in this new style since they can sell the paddy rice in its heaviest state. This is unlike

the manual threshing days when it could take a day to finish harvesting and another day before

the buyer could pick it up, lowering the harvest quality. Along the process, the grains lose some

weight as they are left to dry under the sun. A farmer from Luzon shared that with the rice

combine harvester's availability, he now schedules his harvest at dawn when it is still foggy to

make the grains slightly heavier and give him additional profit margin since they are paid by

163
The Different Factors of Marketing Channel Choices

the kilogram. Given this, he believes that it is more practical to sell fresh paddy rice at the farm

gate because the faster he can sell, the better.

The use of technological advancement to monitor typhoons has also helped farmers adjust their

marketing timing. Generally, without easy access to weather updates, farmers are left at the

mercy of indirect buyers, like buying stations, during typhoon seasons when their harvests are

soaking wet, brownish, and muddy. Buyers take advantage of this and buy at unjustifiably low

prices. Now, given easy access to this information through radio, television, and the internet,

the farmers said that they could schedule their harvest a few days before the typhoon hits their

farm. They shared that this is mostly accomplished using mechanical harvesters, which can

harvest a hectare in just about two hours. Thus, unlike before, when only a few farmers could

escape the typhoon's detrimental effects of typhoons on their farms and selling prices, most

farmers with access to mechanical harvesters are now spared.

Similarly, many farmers explained that the availability of mechanical dryers and small millers

had inspired them to process their harvest to sell to more direct buyers for higher profit margins.

Specifically, they decided to stop selling at the farm gate and instead started drying their paddy

rice using mechanical dryers for stockpiling. Unlike others who prefer solar or manual drying,

these farmers appreciate mechanical drying. They are willing to pay to finish drying within a

few hours, explaining that paying the fees per kilogram is still more cost-effective than hiring

labourers to dry it manually for a day or two. Moreover, they expressed that the costs are

acceptable considering the price at which dried paddy rice can be sold, especially during the

off-peak season. In addition, many farmers were also able to sell milled rice because of small-

scale rice mills. For instance, a farmer from Luzon was able to retail milled rice using a

travelling/mobile miller. He explained that while it may not be as fast as the huge rice mills, it

is enough for his small-scale rice retail business. He particularly appreciates its convenience –

164
The Different Factors of Marketing Channel Choices

that he doesn’t need to go anywhere to avail himself of milling services because the miller can

come to him. Thus, it helps him maximise his profit by selling milled rice without undergoing

the cost and hassle of trucking to and from the big rice mills.

This research also shows that having a mobile phone in the household could also facilitate

selling to direct buyers. Specifically, a few farmers shared that the use of mobile phones

enabled them to canvass for prices, bargain, and make a marketing agreement without even

travelling, just by mere text messages or by calling. Likewise, they are able to notify the buyer

when the harvest is ready for pick up without the hassle of going there or sending someone to

inform them. It also aids the retailing of milled rice since consumers can just text the farmers

their orders, and they can arrange the delivery immediately. Most importantly, farmers could

access buyers located beyond their villages or towns because of easier communication through

mobile phones. It primarily prevents the farmers from being subjected to the price collusion of

the traders in their areas. This is shown by a Luzon farmer's experience, who related that she

was able to escape the price collusion in their area by contacting a trader-miller in the nearby

town, who offered her a buying price that is higher by PHP 1 (USD 0.02) per kilogram. Lastly,

mobile phones can help farmers sell to direct buyers by cutting out the agents. This is especially

true because the farmers can canvass for buying prices efficiently and conveniently on their

own.

Economic environment. The economic environment is another external factor that affects the

marketing channel choices of farmers. Specifically, rice farmers' meagre income that is often

below the poverty threshold forces them to resort to credit to provide for their family needs,

such as health and education. Banks and other formal institutions offer loans with credit interest

of only around 3% to 5% per annum. Yet, farmers cannot take advantage of these loans because

they do not have anything to offer as collateral, being landless and mostly without vehicles.

165
The Different Factors of Marketing Channel Choices

The formal institutions, nevertheless, cannot waive this requirement lest they go bankrupt. This

is because farmers in the Philippines, in general, have a poor repayment track record, especially

when the loan with the banks are part of a government project. Thus, this situation leaves the

farmers no choice but to borrow from informal creditors, who usually impose 3% to as much

as 15% a month, and bind them to sell their fresh produce to them without bargaining power

over the price.

Poverty in the Philippines is one of the main reasons for criminality. These crimes are

widespread and pose significant risks to farmers in various ways. Examples of which are the

possibility of being hijacked or robbed while delivering harvests to the trader-miller, being

swindled by assemblers with fake money, and even losing the pile of crops at home. Farmers

are also afraid of snatchers in the town centre, where most direct buyers are, so they opt not to

go there to canvass for prices or sell their harvests. Farmers consequently just sell at the farm

gate to the local assemblers whom they know, not minding the foregone benefits.

Other crimes can also result in bankruptcy, as evident in the case shared by one farmer. He

related the story of his suki or favourite buyer, who is the highest buyer in the area being very

kind and considerate. This suki became insolvent because of her dishonest assistant, who stole

her money. In effect, the business stopped, so the farmer had to settle and sell to the next best

buyer.

Lastly, the Philippines, being a developing country, still has many unpaved or uncemented

roads, which farmers quoted as another reason they sell at the farm gate. This situation makes

trucking extra hard, not to mention unsafe. For instance, there are farms in the mountains

without roads, and there are farms in areas separated by rivers or creeks without bridges to

connect them to the town. Thus, to sell their produce, the farmer related that they must use a

tyre covered in plastic to carry their harvests across the river because they do not have a boat,

166
The Different Factors of Marketing Channel Choices

and there is no road nor bridge to facilitate their access to the other side of the river. Given how

challenging their situation is, they just usually sell to the only local trader in their village despite

the extremely low monopsonistic buying price.

In this study, the road networks and their quality and the frequency of market and population

density are proxied by the town income class (TownIncClass_O), which showed a significant

positive relationship with choosing direct marketing channels. The findings suggest that the

log odds of selling to direct buyers increase by 0.51, or the probability increases by 1.32% as

the town moves up a rank in category or class. One reason for this is the lower transaction cost

when searching for price or trucking the goods because of better road networks. Generally,

there are also many trader-millers in the area, thereby resulting in better prices and the

accessibility of postproduction services. This variable also mirrors the market frequency (since

a higher town income class would mean a more progressive and populous town) and reflects

the number of milled rice buyers such as the retailers and the population of consumers in the

area. Thus, the higher the town income class, the easier it is to market milled rice, primarily

because of bigger and more frequent markets and higher demand from consumers. The

qualitative data support this, with farmers from these towns affirming that they are able to

directly market their milled rice to their non-farming neighbours or sell to retailers in the public

market.

Sociocultural environment. Unlike the social and cultural values discussed in Sections 6.3

and 6.4, the sociocultural environment referred to here is about the socio-cultural values of the

community that the farmer belongs to. This includes the creditor and traders' self-serving and

unforgiving values once the farmer sells to another buyer. Hence, farmers are afraid of not

selling to them despite having no contract to do so; lest they will be adjudged as “mukhang-

pera” or greedy. This means there will be resentment, and they will lose their sure buyer during

167
The Different Factors of Marketing Channel Choices

bad cropping seasons. In addition, farmers are in an environment where truthful buyers are

rare, as evident by the high deductions, the use of inaccurate weighing scales, and the non-

recording of the exact paddy rice weight. Overall, while farmers practise the culture of shame

and debt of gratitude, most of their buyers are all business. Nevertheless, in some areas, the

farmers still try to keep a smooth relationship by selling their harvests to them despite those ill

practices because that is what their sociocultural environment dictates and expects of them.

They explained that they do not want to deviate from that expectation because they do not want

to feel condemned for their unloyalty to their creditors or buyers. They elaborated that people

in their village or town can be very judgemental, so not abiding by what society expects of

them can cost them their dignity, which they do not want to risk because it is all they have.

Therefore, these farmers sell to their creditors, who are mostly indirect buyers, even with just

a little to no bargaining power.

Natural Environment. Farmers also mentioned the natural environment as another factor

affecting their marketing channel choices. For instance, some farms are isolated by rivers or

located in the mountains, which limit the farmers’ marketing options. They also have limited

drying areas, so they cannot process paddy rice into milled rice, especially during rainy seasons.

While it may be more beneficial to transport their product to the main town and have it

processed or sold there, their location prevents them from doing so. Thus, they just cash it in

immediately after harvest by selling to whoever is accessible, leaving them at the mercy of

whoever goes there or the agent or buying station in the village.

Another important environment factor is the weather because water can largely affect rice

farming in different aspects. For instance, in extreme circumstances, typhoons can also displace

buyers, as experienced by a farmer in the Visayas. She said that they used to sell to a trader-

miller in the town centre before a typhoon ravaged the whole city. When it was rebuilt, the

168
The Different Factors of Marketing Channel Choices

buyer was no longer there, so she just started selling to the local assemblers in their village.

Moreover, farmers cannot dry rice and are left at the mercy of indirect buyers during the rainy

and typhoon seasons ‒ especially when their farms have been flooded. By contrast, droughts

encourage farmers to market to more direct buyers given better drying conditions. Many of

them reported that low drought yields encouraged them to find buyers who offer more

remunerative prices and to improve their marketing strategies to make ends meet. Thus, they

try to haggle better prices; prioritise price over social ties, cultural values, or credit contracts;

dry their crops using the extreme heat; and mill their harvest so they can sell milled rice and

maximise their income. These strategies are also applicable when farmers’ yields are

significantly lowered because of other natural environment factors such as pests and diseases.

6.6 Chapter Synthesis

This chapter shows that many factors are affecting the marketing channel choices of rice

farmers in the Philippines. Based on the quantitative analysis, farmers who are younger, with

fewer years in school, without a fixed income source, with higher off-farm income, with more

training, with indigenous group origin, and with farm traditions are more likely to choose direct

marketing channels such as trader-millers, cooperatives, wholesalers, retailers, and consumers.

The same is true for farmers with smaller farm areas, who own at least one parcel of land, who

harvest more per year, live closer to trader-millers, and live in towns with higher income classes

and provinces with lower rice production volume. By contrast, farmers whose work motivation

is to realise self-fulfilment needs and are inclined to receive a full cash payment and free

trucking services are more likely to sell to indirect buyers such as agents, assemblers, creditor-

traders, and buying stations.

Furthermore, the qualitative data show that while the price is the initial response of farmers on

what shapes their marketing channel choices, there are factors that influence their decisions

169
The Different Factors of Marketing Channel Choices

more. These include cultural factors such as debt of gratitude, loyalty, and ethos of shame, and

social factors such as kinship, friendship, and suki relationship. Health and labour disutility

also impact the farmers’ willingness to do postproduction work, affecting the form of the

product to be sold, which in turn affects their marketing channel options. There are also other

personality factors such as risk avoidance and impulsiveness that influence farmers' decisions

in terms of marketing channel choices.

Finally, the qualitative interviews also revealed that task environment factors such as the

number of buyers, competitors, suppliers, and labour providers directly affect the farmers’

marketing channel choices. By contrast, general environment factors such as natural calamities,

international policies, sociocultural values, economic status, technological equipment, and

legal-political issues affect the farmers’ buyer choices indirectly.

The following summarises all the factors affecting the rice farmers' marketing channel choices

in the Philippines, as revealed by both the quantitative and qualitative analyses (Fig. 4).

Meanwhile, the specific factors that result in a change of buyer and marketing channel level

are discussed in the following chapter.

170
The Different Factors of Marketing Channel Choices

Figure 4. Summary of the factors affecting the rice farmer’s marketing channel choices.

171
Channel Movement and Its Factors

Chapter 7. Channel Movement and Its Factors

Every actor in the market value chain charges a margin, suggesting that a market value chain

that includes more players results in lower margins or share in consumer prices for the farmers

(Crawford, 1997). Thus, to improve the farmers’ marketing gains, it is essential to understand

what motivates them to cut the marketing channel chain shorter or what causes them to do

otherwise.

This chapter discusses the changes in farmers' marketing channel choices and their factors

between the dry and wet seasons of 2017 and the dry season of 2018. The movement analysis

was done by categorising the specific channels into four market value chain levels. These are

the following: (1) Agent/broker; (2) Indirect buyers of paddy rice; (3) Direct buyers of paddy

rice; (4) Direct buyers of milled rice (Fig. 5). Because the farmers can sell to any marketing

channel – from the least to the most direct buyer – their movement can have a value between

negative three to positive three.

Figure 5. The different levels of marketing channels used in this study.

172
Channel Movement and Its Factors

Selling directly to the highest marketing channel level, i.e., consumers, would result in the

largest share in consumer prices and give the highest income. However, farmers usually sell to

indirect buyers. Consequently, their products pass several marketing levels before they get to

consumers. A market value chain involving the four levels would result in the lowest share in

consumer prices. Cutting short the market value chain or selling to a buyer at a higher level is

the ideal scenario, e.g., selling from an agent to a direct buyer of milled rice. However, there

are instances when the farmers who are already selling to more direct buyers sell to a lower

marketing channel level, thereby extending the market value chain. For example, instead of

selling the milled rice directly to a retailer (level 4), some farmers would sell to an assembler

or buying station (level 2), who then would sell to a direct buyer of paddy rice such as a trader-

miller (level 3), who would then sell to a direct buyer of milled rice such as a retailer (level 4).

Overall, the analysis shows that buyer choices had minimal change over time. Specifically,

within the three cropping seasons covered in this study, 39.7%, 41.3%, and 42.7% of the

farmers sold to indirect buyers, including both assemblers and buying stations during seasons

1, 2, and 3, respectively (Fig. 6). The direct buyers of paddy rice such as rice mills and

cooperatives were chosen by 29.7%, 28.7%, and 26.3% of the farmers, closely followed by

agents that were favoured by 24.7%, 24.3%, and 25% of the farmers during seasons 1, 2, and

3, respectively. By contrast, only 6%, 5.7%, and 6% sold to direct buyers of milled rice during

the same period. Overall, the trend shows that more farmers opted to sell to indirect buyers

from seasons 1 to 3. More importantly, the change in marketing channel levels is minimal at

less than one per cent between seasons 1 and 2 and seasons 2 and 3. Nevertheless, it does not

mean that farmers did not move from one channel to another. Notably, the change in marketing

channel levels was more significant when comparing similar seasons, i.e., dry season of 2017

173
Channel Movement and Its Factors

(Season 1) and 2018 (Season 3), where the rate of farmers selling to indirect buyers increased

by 3% and those selling to direct buyers of paddy rice declined by 3.4% (Fig. 6).

Figure 6. Changes in the marketing channel levels of farmers within three cropping seasons.

The factors affecting these movements are discussed in this chapter to better understand

farmers' disposition, which can help improve their market value chain position. Specifically,

the first part shows the factors and specific reasons why farmers are extending the market value

chain by choosing a lower marketing channel level, which is termed in this study as moving

down the chain since they are adding more market players. The second part enumerates the

factors and specific reasons that promote selling to a higher marketing channel level. It is

referred to as moving up the market value chain since they get closer to the end consumers by

cutting out some market players. Each type of movement is described in the later sections since

each type of movement may provide unique insights. Nevertheless, some movements are

highlighted as they provide more useful information, e.g., a one-step cut from level 3 to level

4 may provide more useful information than a one-step cut from level 1 to level 2. Finally, the

174
Channel Movement and Its Factors

last part presents the factors and other aspects that motivate the farmers to remain selling to the

same buyer or stay at the same marketing channel level.

7.1 The Factors Influencing the Extension of the Market Value Chain

The movement analysis demonstrates when the farmers move from more direct to less direct

marketing channels, thereby lengthening or extending the market value chain. Notably, out of

the 600 possible movements, 3.5% resulted in one additional market player, while 1.3% led to

adding two more market players in the value chain (Table 10). Comparing the marketing

channel levels during the same season, the results exhibit a similar pattern with 4% and 1.7%

of the farmers extending the market value chain by one player and two market players,

respectively (Table 11). Overall, the total number of extension cases are lower between

consecutive seasons (4.8%) than across seasons (5.7%), although they are not far apart.

Table 10. The total movement of farmers between marketing levels within seasons 1 to 3.

Marketing Channel Movement Frequency Per cent

Added 2 market players 8 1.3
Added 1 market player 21 3.5
No movement 547 91.2
Cut 1 market player in the chain 22 3.7
Cut 2 market players in the chain 2 .3
Total 600 100.0

Table 11. The movement of farmers between marketing levels within the two dry seasons.

Marketing Channel Movement Frequency Per cent

Added 2 market players 5 1.7
Added 1 market player 12 4.0
No movement 272 90.7
Cut 1 market player in the chain 11 3.7
Total 300 100.0

175
Channel Movement and Its Factors

The most common extension is from a direct buyer of paddy rice to an indirect buyer that

accounted for 2% each between seasons 1 & 2 and 2 & 3, and 3.33% between the two dry

seasons (Seasons 1 & 3). Seasons 2 to 3 registered a 1.33% movement from indirect buyers to

agents and direct buyers of paddy rice to agents, thereby extending the chain by one and two

levels, respectively. The least common change was from selling to direct buyers of milled rice

to indirect buyers, a movement of one level lower, accounting for 0.33% between all compared

seasons (Table 12). Notably, no instance involves a movement of three levels lower.

Table 12. The movement of farmers between Seasons 1 & 2, 2 & 3, and 1 & 3.

Movement Level Change Season 1 to 2 Season 2 to 3 Season 1 to 3

No movement within the market value chain
A to A None 69 23.00% 67 22.33% 70 23.33%
IB to IB None 113 37.67% 115 38.33% 113 37.67%
DB to DB None 80 26.67% 73 24.33% 73 24.33%
MB to MB None 15 5.00% 15 5.00% 16 5.33%
Sub-total 277 92.33% 270 90.00% 272 90.67%
Extended the market value chain
IB to A Added 1 2 0.67% 4 1.33% 1 0.33%
DB to IB Added 1 6 2.00% 6 2.00% 10 3.33%
MB to DB Added 1 2 0.67% 1 0.33% 1 0.33%
DB to A Added 2 2 0.67% 4 1.33% 4 1.33%
MB to IB Added 2 1 0.33% 1 0.33% 1 0.33%
Sub-total 13 4.33% 16 5.33% 17 5.67%
Shortened the market value chain
A to IB Cut 1 4 1.33% 6 2.00% 4 1.33%
IB to DB Cut 1 3 1.00% 5 1.67% 5 1.67%
DB to MB Cut 1 1 0.33% 3 1.00% 2 0.67%
A to DB Cut 2 1 0.33% 0 0.00% 0 0.00%
IB to MB Cut 2 1 0.33% 0 0.00% 0 0.00%
Sub-total 10 3.33% 14 4.67% 11 3.67%
TOTAL 300 100% 300 100% 300 100%

176
Channel Movement and Its Factors

Based on the qualitative inquiry, these movements toward a lower marketing channel level

were influenced by the need for cash, direct buyers' inaccessibility, varietal choice, relationship

with the buyer, convenience, risk avoidance, unfavourable weather, and price. Specifically,

Participant 6 (2019) said that they moved one level lower in the market value chain by selling

to a cooperative instead of selling milled rice to consumers/retailers because of cash

constraints. She explained that their circumstances had changed since their children reached

college. They occasionally require a considerable amount of cash for school expenses, so they

“can no longer depend on irregular income from retailing milled rice”. Thus, they decided to

sell fresh paddy rice to their cooperative instead.

Results also show that some farmers extended the market value chain because of the direct

buyers' unavailability. For instance, two interviewed farmers lengthened the market value chain

because their favourite or suki trader-miller died. One of them explained that he started selling

to a buying station instead because there were no trader-millers left in their area, no

cooperatives, and no NFA. Similarly, the other farmer moved to an indirect buyer, i.e., striker,

because he didn’t like the management of the new rice mill owner, who was not as fair as the

father. In other instances, the movement towards a lower marketing channel level is also

directed by the direct buyer to still benefit from the farmer despite their unavailability. This

was demonstrated in the case of one farmer who was advised by his suki trader-miller to sell to

a buying station since he was away. The suki endorsed the buying station because it was also

his client. Thus, the farmer's harvest still ended up with him, except that it went through an

indirect buyer, so the buying price that the farmer received was lower.

Moreover, some farmers had no choice but to sell to a less direct marketing channel. For

instance, this could be because of their varietal preference, such as the case of Participant 5

(2019), who had to sell to an assembler because “the trader-millers did not like the variety”.

177
Channel Movement and Its Factors

Moreover, some farmers feel the need to sell to close friends and relatives who have suddenly

become rice traders. For example, one farmer moved from selling to a buying station into

selling through an agent because his relative started working as one. Participant 8 (2019)

confessed, “I knew that selling to a more direct buyer would be more beneficial, but I wanted

to help by giving my cousin an income, even at my expense”. Similarly, some farmers admitted

that while they hope to sell to direct buyers, there are times when they end up selling to indirect

buyers because of friendship or amicability. This is especially true when their buyer-friends

come to check on them during harvest time. Participant 14 (2019) explained that “it is hard to

buy a friend”, so despite knowing that their prices are lower than more direct buyers, they are

anxious to refuse them because they are afraid that it would create a rift in their relationship.

Hence, they move from selling to their desired direct buyers to their friend, who owns a buying

station or works as an agent.

In addition, some farmers moved from direct to indirect buyers because of convenience and

risk avoidance. In particular, Participant 18 (2019), a large-scale farmer from the Visayas,

explained that he sometimes sells to a lower marketing channel level, i.e., buying station in

their area, whenever they can afford to buy all his harvest. Although he has trucks to deliver

his crops to huge trader-millers in the neighbouring town, he still believes that “it is better to

sell to buyers nearby because it saves time and effort from trucking, including its risks”. He

explained that their town has “a few militant groups so travelling with crops or money is very

risky”. Hence, given a chance, he would rather sell to several buying stations in their town

“than travel through rough roads and remote areas just to sell to a trader-miller”.

The other farmer who moved down the market value chain also did so because of convenience.

He stated that he sold to an agent instead of a trader-miller because it saves him waiting time.

Likewise, the more convenient process was why some farmers changed from selling milled

178
Channel Movement and Its Factors

rice to consumers to selling paddy rice to indirect buyers. Some farmers also explained that

they moved to indirect buyers, assemblers, and buying stations after having a bad experience

with certain trader-millers. Participant 15 (2019) shared his unpleasant anecdote when he “did

not receive the agreed-upon price after fully delivering their produce”. He emphasised that he

did not want to experience that again; hence, he often just sells to indirect buyers, who pick up

their harvests from the field. Furthermore, some farmers extend the market value chain because

of unfavourable drying conditions. Specifically, the farmers said that sometimes they are not

able to dry because of limited drying facilities. Hence, they just sell fresh paddy rice at the farm

gate to indirect buyers that result in the marketing chain's extension.

Finally, a few farmers (as much as 1.33%) reported that they moved from a more direct (trader-

miller) to an indirect buyer (assembler) because of higher prices. Nevertheless, this was only

the case because of credit – since trader-buyers offer very low prices to their debtors. Hence,

the price of all other buyers, even assemblers, would seem higher compared to it. Similarly,

based on other farmers' experiences, a higher price can be offered by assemblers. Still, it will

not result in higher income because they compensate by making higher weight deductions or

using inaccurate weighing scales. Nevertheless, this claim needs to be investigated further.

7.2 The Factors Influencing the Cut in the Market Value Chain

The results demonstrate that out of the 600 possible movements, only 24 or 4% were

movements to sell to more direct buyers. Of these movements, there were only two instances

(0.3%) when the farmer shortened the market value chain by two players, 22 cases (3.7%) when

they shortened it by one player, and no occurrences that involved cutting three players (Table

10). Comparing their marketing channel choices between the same season type, the data exhibit

that only 11 or 3.7% of the farmers cut a player in the market value chain (Table 11).

179
Channel Movement and Its Factors

The findings illustrate that the most common way farmers shorten the market value chain is by

moving from selling to an agent into an indirect buyer, which accounted for 1.33% between

seasons 1 & 2 and 2% between seasons 2 & 3 (Table 12). The highest movement between

similar seasons (the dry seasons of 2017 and 2018 or seasons 1 and 3) was from selling to an

indirect buyer to a direct buyer of paddy rice, accounting for 1.67%. However, there were only

two recorded instances when the farmers cut the market value chain by two levels: from selling

to an agent to a direct buyer (0.33%) and selling from an indirect buyer to a direct buyer of

milled rice (0.33%). Interestingly, both of which were recorded from Season 1 to 2 (dry to wet

season), which is a transition from favourable to less favourable weather wherein drying

becomes more challenging.

The qualitative data provide insights into the factors that helped shorten the market value chain.

This includes higher disposable income or improved liquidity, better communication means,

trucking services accessibility, favourable weather, manageable harvest volume, and price

priority. Specifically, some farmers were able to sell to a higher marketing channel level

because of better liquidity that helped them avoid borrowing money from creditors, who oblige

them to sell to them. The farmers related that being free from debt enabled them to sell to more

direct buyers such as trader-millers instead of their creditors, who are indirect buyers.

Particpant E (2019), for instance, was able to sell to direct buyers of milled rice instead of just

direct buyers of paddy rice because of “better liquidity brought about by an off-farm income

that gave him the capacity to pay for postproduction expenses such as drying, hauling, and

milling fees necessary to sell milled rice”.

In other instances, farmers cut the market value chain by selling to direct buyers in other areas.

While farmers in small towns typically settle with their local indirect buyers' prices for

convenience, others tend to go further to rice trading hubs when they are dissatisfied with the

180
Channel Movement and Its Factors

local price. They usually travel to the capital of their province or the nearby province. This

gives them more direct options, hence better prices, which shows that access to transportation

results in upward movement in the market value chain.

Favourable drying conditions also result in a marketing channel movement. Specifically, some

farmers were able to sell to more direct marketing channels because this allowed them to dry

their yield and, thus, sell dried paddy rice to trader-millers or milled rice to

wholesalers/retailers/consumers. By contrast, farmers without drying facilities usually extend

the market value chain during the wet season ‒ especially when it is continuously raining. This

was the case for Participant 41 (2019) from Mindanao, who dries paddy rice and sells it as

milled rice during dry seasons because “solar (manual) drying is manageable on the roads”. In

wet seasons, however, he just sells to a trader-miller in a nearby town.

Similarly, one farmer with a limited harvest also cut the market value chain shorter by two

levels. In fact, since a drought resulted in lower harvest volume, Participant 29 (2019) decided

to sell milled rice to consumers instead of selling paddy rice to indirect buyers such as buying

stations or assemblers since “there was only around 5 to 10 sacks of surplus paddy rice, which

was manageable to dry and mill”. This helped him maximise his returns and extend his income

for a few more weeks. The other two-level cut involved a farmer from Mindanao who asked

his assembler-buyer suki to use services to process his harvests into milled rice and market it

directly in their province, instead of selling it as paddy rice to the trader-miller. He suggested

that the buyer can pay him after a few months, provided that the price will be higher by PHP 2

(0.04 USD) a kilogram, which cemented the agreement. Participant 37 (2019) explained that it

happened gradually, taking about “two years of doing business, starting with just a small

volume”. This case is unique given that the farmer still sold to the same person; he just turned

him into a different type of marketing channel.

181
Channel Movement and Its Factors

Finally, farmers' motivation to increase their selling price plays a significant role in cutting or

shortening the market value chain. One strategy they found was to cut the agent. For instance,

some farmers cut agents out and sell to direct buyers after acquiring the trader-miller’s direct

contact details. Participant 22 (2019) shared that he “decided to talk to the rice mill owner to

sell directly after learning about him”. Likewise, one farmer cut the agent and sold directly to

the trader-miller by being the agent of himself, thereby also getting the agent’s commission of

PHP 0.20 (USD 0.004) per kilogram.

Overall, most farmers cut short the market value chain to get higher prices for their goods when

they are motivated by price more than relationships or convenience. For instance, cutting the

market value chain usually happens when the farmers find the courage to refuse their agent-

relative or friend to sell to a trader-miller or when they are willing to work harder to sell milled

rice to consumers.

7.3 The Factors of Same Level Movement

Some farmers extended the market value chain, while others shortened it to get a higher

consumer price share. However, the majority neither extended nor shortened it. Specifically,

results show that farmers did not change their position along the market value chain 91.2% of

the time across seasons (Table 10). After comparing farmers' marketing channel choices during

the same cropping season, the findings registered a slightly lower non-movement at 90.7%

(Table 11). However, this does not necessarily mean that all of them sold to the same buyer or

the same channel. For instance, they could sell to an assembler during one season, then to a

buying station next; or sell to a cooperative first and then move to a trader-miller. Nevertheless,

these are reflected as “no movement” because the marketing channels are within the same level.

182
Channel Movement and Its Factors

While many farmers did not change their market value chain position, it is still interesting to

note why they remained with the same buyer or why they changed their buyers but did not opt

to sell to a higher marketing channel level. Based on the findings, one of the most common

reasons is convenience. Most of the farmers just moved within the same level, between

assemblers and buying stations, because they are the convenient options – they came first.

Hence, farmers would just typically sell to whoever comes first to their field during the harvest

period.

Another crucial factor why farmers remain on the same marketing channel level is credit. Many

farmers move from one indirect buyer to another because they offer faster and easier credit

than direct buyers. Although they charge high-interest rates and offer low buying prices,

farmers buy from indirect buyers ‒ particularly buying stations ‒ because they do not ask for

collateral and are easier to deal with than direct buyers, particularly trader-millers. It was

explained that these indirect buyers only hold on to the farmers’ promises to pay them back

during the harvest time, without interrogation. Participant 45 (2019) said that “he only take my

word for it and I only have to ask once … I don’t have to explain or beg”. This convenience is

very important to farmers, mainly because they do not like “begging” or asking for favours, as

it makes them feel even more lowly. The farmers further explained that buying stations are

usually easier to approach or access, even during off-hours. Participant 16 (2019) also finds

“the high fences and gates of the trader-millers intimidating and unwelcoming”. By contrast,

the farmers who normally borrow from trader-millers replace them with another trader-miller.

This is because they are used to the trader-millers' credit system and policies, and thus, they

are no longer intimidated by them. Overall, these types of changes result in the same level of

marketing channel movement.

183
Channel Movement and Its Factors

The limited buyer option is the final reason for the same-level change. For instance, a farmer

related that his indirect buyer suki discontinued their business. Given there were no direct

buyers in their area, he just chose the best among the available ones. Another example is a

farmer who disclosed that she used to sell to a trader-miller in the centre of the province, but

then the business went bankrupt. Since she still could not afford the cost of selling milled rice,

she had to choose another trader-miller. Finally, some farmers changed from their previous

buyer to another because of a grudge or misunderstanding. Nevertheless, because selling to a

more direct buyer is not feasible, they just chose another buyer who is at the same marketing

channel level.

7.4 The Factors of Invariable Marketing Channels

In contrast to the farmers who moved into a different marketing channel but within the same

level, the majority (71.3%) indeed sold to the same buyer. This is because of the suki or

favoured buyer system, in which the buyer and the seller regularly exchange goods with each

other (Hendriks, 1997).

7.4.1 The Suki

Overall, 71.3% of the farmers consider themselves to have suki or favourite buyers. Buying

stations are the most common suki of farmers, accounting for 37.5% (Table 13). It is not very

surprising considering that buying stations are, most of the time, situated close to the farmer’s

field or home. This is followed by trader-millers with 32.7%, while the third rank is the agents

with 17.8% (Table 13).

184
Channel Movement and Its Factors

Table 13. Distribution of the farmers based on the suki system.

Marketing Frequency of Farmers with suki Suki distribution

Channel farmers with suki (%) among channels (%)
None 86 28.7 -
Buying Station 80 26.7 37.5
Assembler 15 5.0 7.0
Trader-miller 70 23.3 32.7
Cooperative 6 2.0 2.8
Creditor 1 .3 0.4
Agent 38 12.7 17.8
Retailer 4 1.3 1.8
Total 300 100.0 100.0

Results show that farmers consider as few as two consecutive sales to the same buyer as already

being suki, reflecting their desire to sell continuously. Specifically, 40.7% of them are on the

testing and early phase of the suki system with only three to ten repeated sales. By contrast,

23.4% have sold to their suki for years, resulting in as many as 20 repeated sales (Table 14).

Finally, while the average frequency sold to the suki was 22 times, the highest frequency was

96 times. This highest record was by a farmer cultivating rice twice a year for about 52 years,

claiming he only missed selling to the same buyer a couple of times because of crop failure.

Table 14. The frequency that farmers have sold to their suki.

Transactions with Suki Frequency Per cent

3-10 87 40.7
11-20 50 23.4
21-30 23 10.7
31-40 22 10.3
41-50 10 4.7
51-60 10 4.7
61-70 5 2.3
71-80 3 1.4
81-90 2 0.9
More than 90 2 0.9
Total 214 100.0

185
Channel Movement and Its Factors

The ideal suki system involves only one favoured buyer that the farmer should sell to

consistently. However, because the ethos of shame is very strong in the Filipino culture, they

usually do not want to stop selling to their suki even if they find a new buyer who is better.

Participant 30 (2019) related that he ends up “dividing the harvest into two to please the old

suki while also helping themselves receive higher prices”. While this strategy by some farmers

doesn’t seem to be optimal, they think of having two sukis as a better safety net in case one

suki closes down or stops buying and because they don’t want to cause a rift in any of their

relationships.

Finally, while it is generally believed that the benefaction of the suki system results in the

customer receiving special favours (Emmerson, 1982), it has been shown that the depth and

the benefits of such relationships are unique (Hendriks, 1997). Based on the data, only 35.5%

of the farmers with suki received special privileges solely because of the suki relationship,

given that they are not relatives or friends with their buyers (Table 15).

Table 15. The underlying relationships of the farmers to their suki.

Relationship with Suki Frequency Per cent

Family/relative 121 18.8
Friend/acquaintance 107 16.7
Suki/with special privileges 228 35.5
Suki with no special privileges 102 15.9
Supplier/creditor 78 12.1
Others (Landowner) 6 .9
Total 642 100.0

This demonstrates that while the suki system's original idea is to develop loyalty in return for

benefits, notably better prices – it is not the usual case for the rice farmers in the Philippines.

186
Channel Movement and Its Factors

Around 11% confessed that they call them suki because they consistently sell to them, but the

price they get is the same as those who are not suki sellers. Participant 2 (2019) shared that

“suki is just a term I use because I always sell to that buyer, but there is no special price in

return for that”. In fact, sometimes, the price they get can even be lower, especially when they

have outstanding credit with them. This shows that in the Philippines, most of the sellers do

not value relational contracts thinking that the farmers will be worse off without them.

Hence, they think they can take advantage of the farmers, yet they will still do business

with them.

7.4.2 The Factors Keeping the Suki Relationship

It is believed that continuous transactions with the same person build trust over time, resulting

in the suki's embeddedness in the farmers’ networks of social relations (Hendriks, 1997).

Nevertheless, some suki relationships have underlying social or cultural capital that started the

relationship, if not strengthened it. Based on the survey, 18.8% of the farmers with suki had a

blood relationship, 16.7% were friends with their suki, and 12.1% had a debtor-creditor

relationship with their suki (Table 15).

The data also demonstrate that 83.6% of the farmers speak the same dialect as their suki, 55.1%

reported that they probably have the same ancestral roots or origins, 65.9% believe that they

have the same customs/traditions, and 43% are sure that they have the same church or religion

(Table 16). Dialect was notable among the farmers' similarities since they can tell by their

accent if they are from the same region. By contrast, many farmers were not sure if they had

the same ancestral roots or origins, which provides evidence that this might not be a significant

factor in building the relationship.

187
Channel Movement and Its Factors

Table 16. Similarities of the farmers with their suki.

Criteria Yes % No % Unaware %

Dialect 537 83.6 105 16.4 0 0
Ancestral Origins 354 55.1 273 42.5 15 2.3
Customs & Traditions 423 65.9 159 24.8 60 9.3
Church 276 43.0 357 55.6 9 1.4

Overall, the relationships and similarities with the suki are so many that they intertwine or

overlap. Nevertheless, 30.8% of the farmers still attributed their suki relationships to income,

even though they have other strong ties with them, e.g., relatives or friends, apart from being

sukis (Table 17). In fact, some farmers, such as Participant 3 (2019), noted that “the suki

relationship was not intended, but rather coincidental because the suki buyers always happen

to offer the highest prices”. Nevertheless, others are loyal to their sukis because they have a

particular way to get higher income from them, such as by being an agent or allowing the suki

to pay a month later, but at a higher price. Furthermore, farmers such as Participant 30 (2019)

sell to their sukis because they want to keep the suki label, which they can later use “to bargain

for considerations, such as lower deductions during distressful periods”.

Table 17. Reasons why farmers keep their suki relationships.

Reasons for selling to suki Frequency Per cent
Highest benefit 198 30.8
Convenience 57 8.9
Family/relative 57 8.9
Friend/acquaintance 30 4.7
Credit 177 27.6
Buyer’s character/personality 96 15.0
Weighing facility 15 2.3
Others 12 1.9
Total 642 100.0

188
Channel Movement and Its Factors

By contrast, some farmers are selling to their sukis because of plain loyalty. In fact, some are

very loyal in that they do not canvass anymore, trusting that their suki is consistently the highest

buyer they can find in their area. Some farmers canvass rice prices from different buyers, in

some instances, but they still favour or prefer the suki because of familiarity and trust.

Participant 14 (2019) explained that “the canvassed price is used as basis for haggling the price

with them first… only selling to another buyer if they do not match that price or give a higher

offer”.

Some 27.6%, in contrast, have sukis because of credit and debt of gratitude originating from

the fast and easy credit access. Participant 45 (2019) especially appreciates the fact that their

creditor “entertains their requests even during off-hours”, while Participant 13 (2019) said his

creditor was “the only reason he was able to buy a truck with just a verbal commitment to pay,

without any collateral”. The relationship between having a suki and credit is supported by their

correlation coefficient of 0.314 at the 0.01 significance level. The length of the suki relationship

and credit access are also shown to be positively correlated (0.160) at the 0.01 significance

level.

Overall, the farmers call their creditors suki for two reasons: first, for being a suki buyer.

Second, for being a suki creditor, which is a stronger bond that motivates consistent selling.

The suki creditor has different types, which also result in the farmers’ diverse marketing

behaviour. The first and best kind is the suki who offers credit-free interest and fair pricing.

Farmers mostly sell to them without a contract because of their reasonable buying prices and

their good hearts for lending them interest-free money. They feel a debt of gratitude to these

buyers and thus, keep selling to them even when their prices fall a little lower than others. The

second type of creditor suki is the one that offers interest-free credit but takes advantage by

offering non-competitive buying prices. Most of these creditors require that the farmers sell to

189
Channel Movement and Its Factors

them at the price they dictate. Thus, while they do not add interest, farmers are still

disadvantaged because of their lower buying prices. Nevertheless, farmers still sell to them,

even those without a contract to sell, because of the ethos of shame and debt of gratitude for

the interest-free credit. Notably, this type of suki relationship is easily breakable. It can end as

soon as the farmer stops borrowing money or as soon as the buying price difference is beyond

the farmers’ price priority threshold. The third type of creditor suki is the one who imposes a

credit of at least 5% per month but offers a competitive buying price. By contrast, the fourth

type is the creditor with high-interest rates and uncompetitive buying prices.

In general, most farmers remain loyal to the third type of creditor because of the ethos of shame

and debt of gratitude for credit access. Hence, they tend to sell to them even when they do not

owe them anything anymore, provided that their price is the highest or just lower by around

PHP 0.20 (USD 0.004) per kilogram. By contrast, farmers immediately break the suki

relationship with the fourth type of creditor as soon as they have paid their debt. Some indebted

farmers even try to sell half of their harvest to others secretly to get better gains. This is because

most of them do not consider loans from this creditor as a real debt of gratitude, given the high-

interest rates and their low buying price. Nevertheless, some farmers in some areas are left with

no other buyer choice, resulting in the continuous unfavourable suki relationship.

Another reason that encourages the suki system is the buyer's good character or personality,

accounting for 15% of the reasons given by farmers during the interviews (Table 17).

According to Participant 31 (2019), “good personality such as being personable and

approachable, trustworthy, and considerate keeps him selling to his suki”. For many farmers,

having a word of honour is also a top reason to sell to a specific buyer consistently. They value

that the buyer honours the agreed-upon price when they deliver despite the paddy rice's volatile

market price. Farmers likewise keep their suki agents despite their non-crucial role just because

190
Channel Movement and Its Factors

they are kind and show concern for them. Participant 33 (2019) shared that his suki’s “concern

for his welfare and not just of his commission is what keeps their relationship”.

Similarly, there are instances when the suki even goes out of the way to help the farmers. One

example of that was related by Participant 4 (2019) when his suki buyer helped him “get a

tractor inside the ship to transport from Manila to Mindoro without reservations by using his

connections”. Another kind of help was related by Participant 14 (2019), who said that his suki

buyer “supported his project by lending his truck to deliver food”. While farmers are motivated

to sell to these kinds of suki for as long as they can, they also believe that these kinds of suki,

because of their innate goodness, will understand if they ever sell to another buyer who offers

a price they cannot match.

Convenience is another factor for maintaining a suki relationship, accounting for 8.9% (Table

17). The farmers especially appreciate the free transport of their products because they do not

want to arrange for trucking. Most of them do not have transportation means and find it

challenging to commission trucking. Thus, farmers feel greatly relieved when the suki buyer

pays for the fare and delivery fee, or better yet, picks up the harvest from their fields.

Many farmers also keep the suki relationship because of other social bonds with the suki buyer,

such as kinship and friendship, accounting for 8.9% and 4.7%, respectively (Table 17). Farmers

sell to suki relatives because they want to favour them over others, emphasising the need to

help those of the same blood. Participant 1 (2019) said that “blood is thicker than water”. Thus,

she sells to a relative regardless of price to support their business, even at their own expense.

They also think of them as a safety net, more so because they are their relatives. Participant 1

(2019) further expressed her belief that “in return, they (her relative-buyer) will help us when

harvest is bad”. Farmers keep suki friends for the same reasons – because they feel that friends

191
Channel Movement and Its Factors

should help and support each other. Hence, they are willing to forego any additional income

from another buyer to help a friend's business because they likewise assist them in their

endeavours, such as when they need trucks, money, or political support.

In addition, having a good weighing facility was one of the farmers' dominant answers when

asked why they were with their suki (2.3%). Participant 6 (2019) said that “weight accuracy is

very important, even equally important as the price”. She further explained that “a good buying

price is useless if the weighing scale is corrupted or inaccurate”. This is particularly important

when the farmers do not go to the buyer’s location and just entrust the harvests to the agent.

Therefore, they tend to be loyal when they find a trustworthy buyer with an accurate weighing

scale.

Additional factors fuelling the suki relationship were mentioned during the qualitative

interviews. These include risk avoidance. Many farmers confessed that having a suki is like

having a safety net – a sort of sure buyer. Participant 38 (2019) appreciate that because of the

relationship, “the buyer will not refuse the harvest even when the quality is low or when his

storage is full, although they may place higher deductions”. The farmers believe that sticking

with their suki – even if they are getting lower prices – is “worth the assurance of having a

buyer during a bad cropping season”. While this doesn’t seem to be correct at all times, since

some suki buyers refused or took significant advantage of some farmers during stressful harvest

times, many still choose to rely on and keep their suki relationships because of this.

Furthermore, farmers find it risky to sell to other buyers, particularly non-local ones, in lieu of

their suki because it can leave them worse-off if it is not long-term. Specifically, farmers do

not want to damage their suki relationship for a non-local buyer, who may offer a better price

but not come at all the following season. This was experienced by Participant 24 (2019), who

192
Channel Movement and Its Factors

was offered a significantly higher price by a travelling assembler. Thus, he sold to him instead

of his suki, believing that the assembler would come again the next season. However, he

divulged that “the assembler did not come again, and the suki did not want to buy the harvest

anymore because of disloyalty”. Thus, he had no choice but to sell to an indirect buyer who

offered an even lower price than his suki.

Finally, some farmers keep their suki relationships because of amicability and honour. The

farmers explained that their ethos of shame prevents them from selling to a new buyer – more

so if the suki is a relative, friend, or creditor. Similarly, farmers keep selling to their suki,

regardless of price, to prevent a rift or hurt feelings, especially when they have political

aspirations.

7.5 Chapter Synthesis

This chapter demonstrates how farmers are moving from one marketing channel level into

another and the factors affecting those movements or non-movement to understand how to

motivate farmers to improve their position in the market value chain. Overall, the outcomes

show that farmers changed their position within the market value chain in 8.8% of the cases,

while they remained at the same level in 91.2% of the cases. Specifically, farmers extended the

market value chain by adding more players than they used to have 4.8% of the time. These

farmers sell to less direct marketing channels because of the non-availability of the more direct

buyers, varietal issues, convenience, risk-aversion, and credit. Moreover, this study shows that

farmers shortened the market value chain by selling to more direct buyers 4% of the time.

Based on the qualitative data, these changes were influenced by a more manageable/lower

harvest volume, better communication means, transport availability, harvest pooling,

favourable drying weather, and motivation to increase income. However, it is difficult to draw

193
Channel Movement and Its Factors

general conclusions from these given the small number of observations wherein the farmers

changed their marketing channel level.

A great majority (91.2%) of the cases showed that farmers did not change their position within

the market value chain. While 71.3% of them have suki to whom they sold consistently, the

rest of the farmers are likely to have stayed on the same level but moved to a different buyer.

Given these numbers, this study could provide insights, especially in the qualitative analysis,

on the possible factors that could encourage farmers to stay within the same marketing channel

level.

Based on the data, changing the buyer without changing the marketing channel level happens

because of the lack of access to a more direct buyer or other credit providers, convenience,

familiarity, miscommunication or grudge, or the unavailability of the last buyer or suki. By

contrast, reports show that farmers keep their suki relationships because of reasonable price

offers, reliability, easy credit access, debt of gratitude, social ties, risk avoidance, and the

buyer's good personality. Overall, these provide evidence that farmers keep their suki because

of personal, economic, social, cultural, and even external environment factors – which may

even result in farmers selling to more than one suki buyer per cropping season. These

intricacies, along with the other results of this study, are further explained in the following

discussion chapter.

194
Discussion

Chapter 8. Discussion
The first part of this chapter presents the rice farmers' most common marketing practices. The

second part discusses the key findings on the factors affecting the farmers’ choice between

direct and indirect buyers based on the econometric model and compares them with related

studies. It also presents and tries to explain these choices using qualitative data. The third part

deliberates the factors that affect the farmers’ moves towards cutting or extending the market

value chain, which are more specific to help identify detailed conditions that could improve

their current positions. The last part presents the implications of these findings, particularly on

farmers' typology with reference to the price factor.

8.1 The Marketing Practices

There are 112 marketing channels for rice trading in the Philippines (Mataia et al., 2018).

Nevertheless, based on this study's findings, the farmers' most common marketing channels are

agents, assemblers, buying stations, trader-millers, cooperatives, wholesalers, retailers, and

households. A significant majority (94.1%) of the farmers sell to traders or paddy rice buyers.

This supports the findings of Soe et al. (2015), Cazzuffi and McKay (2012), and Abebe et al.

(2018) that many farmers prefer to sell to traders, regardless of whether they are getting a lesser

profit margin from their lower prices and the unfair adjustment of their measuring scales.

Furthermore, the data demonstrate that only a few farmers (5.9%) sold to buyers of milled rice,

yet other studies on marketing channels show higher rates (Cazzuffi & McKay, 2012; Dessie

et al., 2018, 2019). This provides evidence that there is still considerable room for improving

the rice farmers' marketing practices in the Philippines. Particularly, given more favourable

circumstances and fewer constraints, the rate of selling to direct buyers of milled rice can also

be increased.

195
Discussion

Overall, the choice to sell to indirect buyers is mainly because of the preference to sell fresh

paddy rice given their lack of capital, physical ability, time, labour, and facilities to do

postproduction work involved in producing dried and milled rice. The majority of the farmers

selling fresh paddy rice likewise attributed their decision to the additional income from drying

rice, which they perceive to be insignificant. However, based on the data during the qualitative

interviews, the price difference between dried and fresh paddy rice is about PHP 3 (USD 0.06),

so selling dried paddy rice results in an additional net profit of PHP 0.5 (USD 0.01) per

kilogram, after accounting for labour costs. That exact amount was also stipulated by many

farmers in different locations, having computed the cost and benefit of drying. This amount

was very similar to the data of Mataia et al. (2018), showing that the net profit of selling dried

paddy rice is at PHP 0.5 (USD 0.01) per kilogram, given the average procurement price of

fresh paddy rice in the Philippines at PHP 16.31 (USD 0.33) and the selling price at PHP19.37

(USD 0.39). Therefore, if farmers dried their marketable surplus, which is about 70% of their

average yield of 5.07 tons/ha (Mataia et al., 2018), they would have an additional net profit of

PHP 1,772 (USD 35.44). Therefore, contrary to what most of the farmers said during the

interviews, the price difference is not insignificant, given that it is about 5.86% of the wet

season and 7.58% of their dry season incomes.

Nevertheless, selling fresh paddy rice, despite the additional profit offered by selling dried

paddy or milled rice, is not surprising given that the results generally show that the farmers’

choices were not just a function of price. Instead, it is a combination of a lot of other factors

explored in this study, such as labour disutility, social relationships, and cultural values.

196
Discussion

8.2 The Factors of Marketing Channel Choices

This study found that the farmers' buyer choices were influenced by several factors apart from

price and other sociodemographic variables previously used by other studies. Using the non-

separable household model and thus, combining the factors that affect the farmers' decisions

as producers, consumers, and workers, various personal, economic, social, cultural, and

external environmental factors affecting the buyer choices were examined. Notably, non-

sociodemographic factors were found to be relevant in this study because choosing a marketing

channel is more complicated than deciding to participate in the market. While market

participation does not involve considering relationships with another party, the latter proves to

be highly influenced by it. Hence, it shows the viability of the framework ‒ particularly the use

of social and cultural capital.

Firstly, just as with previous studies, the results indicate that personal factors affect the farmers’

marketing channel choices, such as age and education. Particularly, this research’s findings

concur with Mottaleb et al. (2014) ‒ that farmers with higher years of education sell to indirect

buyers. Both studies found that these farmers are usually employed externally, given their

education, limiting their farming time. Thus, they just sell to the fastest and most convenient

buyers that are often indirect buyers, as demonstrated in the quantitative and qualitative data.

By contrast, Soe et al. (2015), in their research on rice, and Dessie et al. (2019), on teff, did not

find any significant relationship between education and marketing channel choices, while the

results of Dessie et al. (2018) on their wheat research showed that higher education increases

the farmers’ probability of selling to wholesalers while decreasing the likelihood of selling to

consumers. They explained that this is because farmers with better education or literacy have

higher skills and knowledge in agricultural marketing that enable them to sell to buyers who

offer more remunerative prices, inferring that wholesaler markets are more profitable than

197
Discussion

consumer outlets. This is, however, not the case for rice in the Philippines where the consumer

outlets typically give the most remunerative prices being the most direct buyers and having

eliminated the middlemen along with their fees (Mataia et al., 2018; Nxumalo et al., 2019).

This disparity in results may be attributed to differences in the methodology. Overall, the effect

of education on the farmers’ income is positive either way since it can result in selling to direct

buyers because of better marketing knowledge and networks or higher liquidity that would also

result in selling to direct buyers.

Furthermore, this study exhibits that higher age is a decisive factor towards less direct

marketing channels, as Cazzuffi and McKay (2012) also showed, owing to the limited extent

of labour that these farmers can perform. However, Dessie et al. (2018) posited that older

farmers are less likely to sell to assemblers ‒ a kind of indirect buyer ‒ because they do not

trust the local buyers' prices. While trust was one reason why some farmers do not sell

specifically to assemblers in this study, the quantitative data did not show a correlation between

not selling to assemblers and age. The qualitative data demonstrate that farmers, both young

and old, doubt the trustworthiness of assemblers ‒ especially the travelling ones, who only

come to the area during harvest season. Moreover, Dessie et al. (2018) defined assemblers

differently from this study. Whereas they qualify assemblers as local, they can either be local

or inter-regional in this study; and rather than location-based, they are defined as traders who

assemble the commodity in the trucking service and have no permanent trading station nearby.

Another possible cause for the contradicting results is the econometric model's difference ‒

particularly their inclusion of credit access. This variable was not included in this study because

of possible endogeneity since farmers in the Philippines can sell to specific buyers because of

credit, which in turn would give them access to credit in the next cropping.

198
Discussion

Interestingly, the research of Dessie et al. (2019) on teff showed that older farmers are less

likely to sell to consumers, and thus, they are more likely to sell to indirect buyers that is similar

to the findings in this study. However, they tried to relate this with literacy – explaining that

older people in Ethiopia are relatively illiterate, so they do not know the additional income they

would get by selling in consumer markets. By contrast, this assumption did not surface in the

results of this study. Instead, based on the qualitative interviews, older farmers prefer to sell to

indirect buyers such as assemblers at the farm gate because they cannot do the postproduction

processes, which pertains to their physical limitations. Some of them cannot read or write well,

but they certainly know how to compute. In fact, having been in the farming profession for

several decades, many of them are aware of the benefits of selling to direct buyers, particularly

consumers. Nevertheless, given their old age and the physical limitations that go with it, they

prefer more convenient channels. They do not want to go through the hassle of selling to

consumer or retail markets because the tedious work could compromise their health. It suggests

that farmers may be less likely to sell to direct buyers because they are not price- or profit-

oriented, as supported by the -0.097 correlation between age and price motivation at the 0.01

significance level. This is exemplified by the cases of the farmers who have the financial

capacity to hire workers to do the postproduction processes but opted not to despite the possible

additional earning of PHP 0.50 (USD 0.01) and the PHP 4 (USD 0.08) per kilogram from

selling dried and milled rice, respectively.

Furthermore, this study's findings that farmers who prefer prompt payment sell to indirect

buyers concur with that of Alene et al. (2008) and Soe et al. (2015) – that those who want

instant cash sell immediately upon harvest to brokers or commission men without much regard

for prices. Cazzuffi and McKay (2012) likewise found that farmers who prefer to receive their

payment for their products immediately sell to traders. While this might seem to be economic-

199
Discussion

related because of liquidity, this study has shown that it can rather be motivated by personality.

The case of the farmer who sold his harvest immediately despite not having credit and the

possibility of getting heavier grains and winning a competition is a good example of their

attitude towards immediate payment, which has nothing to do with credit. Similarly, many of

the farmers, at least those who do not have credit, certainly do not sell immediately because of

the high cost of borrowing money – they sell because of preference for convenience or just

because they are payment anxious. In fact, 90.1% (256 of 284) of the farmers without any credit

still prefer fast cash or to sell at the farm gate to get their money immediately rather than get

credit or process their harvest, even if it gives a higher price. Similarly, 6.8% of those with

credit do not mind receiving a payment that is partial or an instalment. Furthermore, this

variable aims to measure how fast they want the payment – if it is immediately (fast cash as in

1 day) or not immediately (usually 2 weeks - enough time to dry, mill, and sell milled rice

without stocking). Hence, it does not reflect the high cost of borrowing since it doesn’t make a

difference in interest accrual, given that it is less than a month (they are not charged per day;

sometimes, they are even charged per cropping season). Hence, it is more a matter of their

personality – their payment anxiousness.

Similarly, this research suggests that farmers sell to indirect buyers because of personality

factors such as the preference for buyers providing pick-up services, which is evidence of

labour disutility. This is because 76% of the buyers (mostly trader-millers) reported that they

cover the transport cost when they ask the farmers to deliver the harvest to them, while those

who pick up the crop from the farm buy at a lower price to cover their trucking costs (Mataia

et al., 2018). As such, the decision of the majority of the farmers to still sell to indirect buyers

at the farm gate may not be about price but about labour disutility from arranging for trucking.

It is more about the farmers’ preference for convenience ‒ especially since there are many

200
Discussion

instances when the farmer has the capacity to deliver, which they said they know would result

in higher income because of the higher price per kilogram for the paddy rice, yet they choose

not to because they do not want to exert effort even if they have nothing profitable to do. In

fact, based on the quantitative data, 49.7% of the farmers with the capacity to deliver their

products did not opt to, mainly because of labour disutility (average distance to the miller is

6.11 km). Meanwhile, the farmers without means of delivery just rented their transportation

vehicle in 63.8% of the instances when they delivered their products to their buyer (at an

average distance to the miller at 8.03 km). These show that their choices are not necessarily

because of transportation costs and that many of them decide based on labour disutility. Hence

the choices reflect their preference and, thus, their personality. And while this is a crucial

factor, there are no studies that include this variable. Hence, no comparison can be made.

This research likewise provides in-depth information that risk-aversion, in its many forms, also

plays a crucial role in farmers' marketing channel choices that are not widely presented in

previous marketing channel studies. Although their research is not on the marketing channels

of grains, Gabriel and Hundie (2006) also found that postharvest losses, particularly the

possibility of losses during storage, are among the most critical factors that explain why farmers

sell their harvest instantly at the farm gate even at their lowest prices. Furthermore, both the

quantitative and qualitative data show that farmers are likely to sell to direct buyers when their

work motivation is to fulfil basic needs since they are continually struggling to meet their

everyday expenses, such as food, clothing, shelter, and education. Maslow’s theory suggests

that this deprivation can result in a stronger and longer motivation to fulfil these needs

(McLeod, 2007), thus their yearning to sell to more direct buyers for a higher profit.

Nevertheless, this was the first attempt to include this type of variable to reflect the farmers’

motivation. Hence, there is also nothing to compare the results with.

201
Discussion

Secondly, economic factors such as higher household off-farm income, land ownership, and

more frequent harvesting per year increase the probability or odds of selling to direct marketing

channels because they reflect better liquidity that improves market search and allows

postproduction. These results are similar to those of previous studies. For instance, Cazzuffi

and McKay (2012) and Dessie et al. (2018) also found that wealthier households or those with

better liquidity have a lower probability of selling to indirect buyers. This is because they can

use the off-farm income to overcome the financial constraints on postproduction, as also

explained by Woldehanna (2000) and as shown by this study's qualitative data. Studies on

market participation likewise highlighted the importance of working capital on farmers'

engagement in the market (Alene et al., 2008; Cadot et al., 2006; Cuevas & Clarete, 2015;

Stephens & Barrett, 2011).

Moreover, the shorter distance to a trader-miller was also shown to affect the choice for direct

marketing channels positively. It provides evidence that farmers are better engaged in the

market when the transaction cost is lower relative to the income they would get from trading

(De Janvry et al., 1991). This finding is similar to those of previous marketing channel studies

(Dessie et al., 2019; Soe et al., 2015; Yang & Guo, 2011) and even market participation studies

(Alene et al., 2008; Cuevas & Clarete, 2015; Mmbando et al., 2015; Omiti et al., 2009).

Nevertheless, while they all solely attributed this relationship to transaction costs, this study's

qualitative data show that it is also an effect of the social capital developed through proximity

‒ especially of homes. This social capital positively affects the economic transaction since their

social ties encourage both parties to give each other favour – the farmer by selling to the trader-

miller and the trader-miller by providing reasonable prices and fair credit conditions.

Fixed employment was also found to result in selling to indirect buyers because of the limited

capacity to do postproduction work. Although related, this is not highly correlated with the

202
Discussion

education variable, given the 0.224 correlation coefficient between the two at the 0.01

significance level. This can be because many farmers who did not finish tertiary education were

still able to get fixed employment as guards, carpenters, and in other blue-collar jobs. The

negative relationship between fixed employment and selling to indirect buyers could be

attributed to labour disutility in postproduction work because of the disparity between farm and

off-farm wages. As also shown in the qualitative data, the farmers regard the additional income

that they will get from drying as insignificant because they compare it to the rate that they are

being paid at their daily job, which is much higher. For instance, the net profit from farming

per hectare in the CARAGA Region in 2015 was PHP 14,677 (USD 293.54) for six months,

while during the same time, the minimum wage in the said province is PHP 268 (USD 5.36),

which is equivalent to around PHP 35,000 (USD 700) for the same period. Similarly, the

average income per hectare in the MIMAROPA region was 21,000 (USD 420), while its

minimum daily wage is around PHP 245 (USD 4.9) or PHP 32,340 (USD 646.8), both for six

months (DOLE, 2019; Mataia et al., 2018). Therefore, if the total net income from farming is

already small compared to their salary, much more is the additional income from selling

processed rice. Thus, employed farmers usually have labour disutility in terms of

postproduction practices that result in selling to indirect buyers.

Furthermore, the data provided evidence that a larger farm size does not necessarily lead to

selling to more direct buyers but rather the opposite. This is similar to the findings of Cazzuffi

and McKay (2012), who show that consistent sellers with large harvest volume sell to indirect

buyers at the farm gate because they can buy all of their product at once compared with

individual households. They also related this to established relationships, which was likewise

documented in this study's qualitative data. Nevertheless, their data show that generally,

farmers producing larger quantities are less likely to sell to traders and prefer to look for better

203
Discussion

marketing channels, which is similar to the findings of Soe et al. (2015). The results of Dessie

et al. (2018), in contrast, demonstrate slight inconsistency, showing that a bigger farm size

increases the probability of selling to both assemblers and retailers but decreases the likelihood

of selling to consumers. They explained that this is because larger product volume allows the

farmers to sell in bulk, reducing marketing costs compared with consumer outlets.

Nevertheless, their study on teff farmers (Dessie et al., 2019) revealed that a higher quantity of

produce increases the probability of selling in wholesaler and retailer markets; while a bigger

farm size results in a lower probability of selling to the retail market but a higher probability

of selling to the consumer market. In general, these results do not seem to have a unified

direction, partly attributed to dissimilar approaches, products, countries, and econometric

models. Nonetheless, the two studies of Dessie et al. (2018, 2019) showed conflicting results

and even inconsistent conclusions despite all their similarities in methods, sampling, country,

and researchers. These disparities may be evidence of the effect of social factors on farmers'

decisions as found by this study and as shown by Cazzuffi and McKay (2012), highlighting the

need to factor them in, especially the length, kind, and benefits from the relationship.

In addition, while it was expected that price would show a significant effect on choosing direct

buyers, the econometric model did not. This non-significance can be explained by the

complexity of the price variable that was further revealed through analysing the farmers’

qualitative interviews. The data demonstrate that while there are farmers who would always

choose the highest priced buyer, there are also farmers who are not at all after the price. In fact,

some would only select the highest buyer when there are no social, cultural, or personal

constraints. The non-significance of price could also be attributed to the responses of farmers

who claimed to be non-price motivated, yet they sold to direct buyers because of other factors

such as kinship and friendship. More importantly, some of them are selling to direct buyers

204
Discussion

because of credit, which often results in even lower prices than indirect buyers due to the credit

contract. Similarly, some farmers said they were price motivated and selected the buyer with

the highest price offer, yet they sold to agents because they only canvassed among them. Hence,

while they chose the highest buyer among those they asked, it is not the highest possible buyer

in the broader area. This could also be why this variable's contribution to the model is negligible

(0.11 of the Wald chi2).

Thirdly, this study found that social factors such as the number of training sessions attended

and social relationships affect farmers' marketing channel choices as producers, consumers,

and workers. Specifically, attendance at training sessions or other productivity-enhancing

activities results in selling to direct buyers since these can provide a good platform for

knowledge-sharing and linkage to buyers. However, no previous study has included this in

their grains marketing channel model, so no comparisons could be made. Meanwhile, this

study’s data could not affirm the significance of organisational membership on selling to

wholesaler and retailer outlets, as shown by Dessie et al. (2019). This disparity could be

attributed to the non-functionality of farmer cooperatives, associations, and other organisations

in the Philippines. By contrast, this study provides concrete evidence to support the claim of

Cazzuffi and McKay (2012), who explained that the decision of consistent farmer-sellers to

trade with indirect buyers is an effect of social capital ‒ particularly of established relationships

and good reputation as a seller. It also confirms Sobel's (2002) claim that farmers’ economic

transactions are affected by their ties to their networks.

Overall, this study shows that social tie is the most critical social factor in choosing a buyer.

Many farmers proved to prioritise their friends, relatives, or favourite buyers over more

remunerative prices not because it benefits them but because they are part of their personal

alliance system, which they ought to support (Dolan, 1991; Philippines Country Study Guide,

205
Discussion

1999). While this support system should be two-way, the buyers do not always offer the farmers

their highest possible buying price. In fact, many of these buyers give prices that are even lower

than non-related buyers. Yet farmers still choose them because of their obligation, dictated

partly by their sociocultural environment. Hence, even if they want to, this constrains some

farmers from selecting the highest buyer, being wary of the judgements from the people around

them. Thus, while it benefits a few farmers, it is more of a constraint, which only very few are

able to overcome. Therefore, there is a need to address this huge barrier preventing farmers

from selling to the most profitable marketing channel available to them.

Fourthly, the quantitative data suggest that cultural capital through indigenous group origin and

observance of farm traditions lead farmers to sell to direct buyers. This is similar to the findings

of Cazzuffi and McKay (2012), which demonstrate that farmers with indigenous origins are all

less likely to sell at the farm gate. Goetz (1992) postulates that this is because ethnicity and

language increase the cost of observing market prices that lead the farmers to sell milled rice

to their neighbours instead. The qualitative data, however, did not provide this explanation.

While the farmers with indigenous origins were keen on selling milled rice for higher

profitability, saying that they wanted to give their children a good life, they did not associate

this with their indigenous origin, language, or accent. However, some interview data show that

these, later on, encouraged a partial or full gift exchange with the buyer, giving a higher price

or less deduction because of cultural capital such as dialect and similar origins (Throsby 1999).

More importantly, the qualitative data revealed that cultural values such as debt of gratitude,

loyalty, and ethos of shame result in selling to indirect buyers because they are more important

to many farmers than price. For instance, they prioritise their culture of shame over economic

gain because it defines their personality and gives them the ability to face other people with

self-esteem. Baloyi (2010) posits that customer loyalty affects the informal markets' prices,

206
Discussion

which this study affirms. However, this study's data show that some farmers are not always

loyal because of higher prices. It can also be because of convenience or to avoid any

misunderstanding, which further provides evidence that not all farmers prioritise price.

The debt of gratitude, which is a type of embodied cultural capital inherited passively and

acquired consciously (Bourdieu, 1986), likewise showed a powerful influence on buyer

choices, even over price. For the farmers, selling to their creditor even without any form of

contract, regardless of price, is a gesture that acknowledges the favour and is a form of repaying

the person directly or indirectly while demonstrating social responsibility (Rungduin,

Rungduin, Aninacion, Catindig, & JrGallogo, 2016). Interestingly, very few studies on

agricultural marketing involve debt of gratitude or other cultural values despite being one of

the core values that surface when there is an interaction between people (Pe-Pua & Protacio-

Marcelino, 2000) ‒ especially when credit services are involved.

Lastly, external environment factors also affect farmers' marketing options and reinforce some

personal, social, and cultural factors (Kreps, 1990). Therefore, despite having limited studies

involving them, they are essential in understanding rationality (Simon, 2000). Specifically,

living in towns with higher income classes has a positive result on marketing channel choices,

just as on the level of market participation, because of better access to facilities, better road

networks, and good access to direct buyers (Cuevas & Clarete, 2015). By contrast, farmers in

provinces with higher rice production volumes were shown to sell to indirect buyers because

of the more competitive prices of the many assemblers from nearby regions. This highlights

the effect of competition on prices; hence, local policies should encourage, rather than prevent

travelling buyers from entering farming towns. The presence of travelling assemblers in these

provinces also shows that trucking could be cost-effective and practical if farmers can organise

their harvests through a functional group or cooperative.

207
Discussion

The qualitative data additionally revealed that general environment factors, such as

technological equipment, status of the economy, natural calamities, international policies,

sociocultural values, and legal-political issues; and the task environment factors such as the

number of buyers, competitors, suppliers, and the labour market are crucial to the farmers when

deciding to whom to sell their product. Overall, while external environment factors do not

directly affect the marketing channel choices, they can still significantly impact the decision.

For instance, the unavailability of labour help and lack of drying facilities can result in labour

disutility. Moreover, technologies that facilitate the harvesting process through mechanical

reaper or harvester result in grains with a higher moisture content that are heavier. Hence, many

farmers with labour disutility sell them fresh instead of dried, thinking they are getting a higher

income. Nevertheless, Mataia et al. (2018) illustrate that mechanical harvesters lower the paddy

rice’s buying price by PHP 1 (USD 0.2) per kilogram because of the higher moisture content

and the impurities. This means that the buyers account for the additional moisture content and

incorporate it in the pricing either through a lower buying price or deduction.

Likewise, the constraints presented by the natural environment, such as the tropical weather in

the Philippines (with temperature as high as 42 degrees Celsius), rivers separating farms from

the town, and the mountainous locations of some farms make the risk-averse farmers and those

with labour disutility sell to indirect buyers at the farm gate. Similarly, corruption (legal-

political environment) limits the number of buyers in an area or limits access to government

facilities to a few people, resulting in less profit for farmers – especially those with labour

disutility or frail health.

Furthermore, rice trade liberalisation (international environment) has removed farmers’ income

opportunities from selling dried paddy or milled rice during the off-season since imported rice

can come into the country at any time. Hence, this has discouraged many farmers from drying

208
Discussion

and milling their harvest, resulting in selling at the farm gate. Similarly, the socio-cultural

environment also directs farmers' decisions towards what is pleasing to society rather than

economic ones. As previously explained, it reinforces the personal alliance system that results

in the bigger influence of social factors on the marketing channel choices. Thus, most farmers

sell to their relatives, friends, or creditors, resulting in lower buying prices for their rice. This

is mostly because of personality (risk-aversion that they will not find another creditor), social

capital (that they may damage relationships), and cultural values (their high regard for debt of

gratitude and ethos of shame).

The country's poor economic state (economic environment) also results in selling to indirect

buyers. The qualitative investigation support this – with data showing that the lack of facilities

and high crime rates result in labour disutility and risk-aversion, respectively – making farmers

sell to indirect buyers at the farm gate rather than deliver to rice mills. This study also affirmed

the findings of Soe et al. (2015) that poor road conditions also result in selling at the farm gate.

The results illustrate that the general environment can affect the task environment, which

affects the internal environment factors. For instance, positive technological advancements,

e.g., the use of a mechanical harvester, decreases the labour market's constraints that should

facilitate the hiring of workers to dry or haul the harvest ‒ especially for those who are

motivated to maximise their profit. Similarly, the natural environment, such as the farm's

remote location, results in fewer buyers in the area. Hence, farmers sell to whoever is available

to buy at the farm gate because of the disutility of labour. Lastly, a bad economic environment

that results in inadequate market facilities and roads also decreases the number of suppliers and

buyers in an area; thus, making farmers sell without canvassing and remaining with their suki.

209
Discussion

None of the previous studies on marketing channel choices included external environment

factors, yet these are vital considerations of farmers. Although some of these may be assumed

as constraints, they must be considered carefully to better understand farmers' choices. These

cannot be deemed time-invariant since the qualitative interviews show that they vary ‒

especially the task environment factors such as the number of suppliers, competitors, buyers,

and the labour market. Similarly, some general environment factors can also change within one

cropping, affecting the task environment and internal environment, such as a newly constructed

road or newly installed postproduction equipment in the locality.

Overall, sociodemographic factors usually included in the marketing channel choice models,

e.g., age, education, fixed employment, liquidity, volume, distance to market, are found to be

significant in the decision of the farmers on whether to sell fresh paddy, dried paddy, or milled

rice, and consequently their marketing channel choice. Specifically, the findings suggest that

older farmers and those with fixed employment are less likely to sell to direct buyers primarily

because of the labour and time constraints during postproduction. Merely looking at this issue

would mean that easing the drying, hauling, and trucking processes should allow older, even

those with unstable health conditions, and those with limited farming time because of fixed

employment to dry and mill their products and market them to more direct buyers. Similarly,

the negative relationship between larger farm size and more direct marketing channels owing

to the lack of drying facilities suggest the need to establish more drying pavements or more

user-friendly and affordable mechanical dryers. More training and reinforcing the benefits of

selling to more direct buyers, just as traditions are imbibed in a person, should also encourage

farmers to sell to buyers who offer more remunerative prices. Likewise, the results showing

that those living closer to rice mills and in towns with more facilities sell to direct buyers imply

that providing more milling facilities, improving roads, establishing more markets, and giving

210
Discussion

access to direct buyers would significantly improve the marketing income of farmers. Finally,

those with better liquidity are able to sell to more direct buyers, which implies the need to have

accessible and more affordable credit options from government facilities. This would prevent

the farmers from being indebted to their creditors and consequently allow them to choose the

buyer who offers the highest economic benefit.

This study provides evidence that there may be advantages to adding other nominal variables

to capture the effect of crucial personality, social, cultural, and external environment factors

exhibited in the qualitative data, which might have been considered unobservable in the

previous studies. These are evidently very important as farmers cite them as their initial

considerations when deciding whom to sell to. These are critical variables that often determine

where the farmer will sell. For instance, higher buying competition can lead the profit-

maximising farmer towards a more direct buyer, but a farmer with labour disutility towards

selling at the farm gate. Similarly, more suppliers, e.g., credit providers, can help profit-

maximising farmers sell to the highest buyer because there are other creditors if they break

their relationship with the existing lender. However, farmers who give utmost importance to

cultural values such as debt of gratitude, loyalty, or amicable relationships, would not be

affected by the number of credit suppliers. Some factors also predominantly influence the

marketing channel choices that can change the effect of certain variables. Specifically, old

farmers' labour disutility can be countered by their motivation to maximise profit when they

have limited harvests. This is evident when they find ways to dry and even mill their rice to

sell them to direct buyers during bad harvests. Similarly, the effect of the market distance on

buyer choices is also subject to the area safety (economic environment) and the farmers' risk-

aversion (personality). Cazzuffi and McKay (2012) likewise mentioned how the relationship

211
Discussion

between marketing decisions and volume is altered by both the consistency of market

participation and liquidity.

These findings suggest that personality, social, cultural, and external environment factors are

formed like a nexus with the other factors – with some linked to each other and affecting each

other. Thus, while the effect of the sociodemographic factors seems simple and easy to address,

they get more complicated when combined with personality factors, such as the preference for

fast cash, trucking services, or farming motivation. These are also more challenging when

combined with social capital and cultural values, especially when induced by the task and

general environment factors. Thus, this presents the importance of incorporating these variables

in the model.

Finally, among the producer, consumer, and worker utility functions, the worker is most

influential when choosing a buyer since rice marketing can require tedious labour, i.e., drying,

milling, and hauling. Based on the data, many of them prefer convenience over profit. Thus,

farmers do not always act towards profit-maximisation like a business. They are either for

profit-maximisation or not and make their choice between the priority of price or non-price

factors. Most of those who prioritise price sell to the buyers offering the most remunerative

prices; they sell milled rice if there are no constraints such as liquidity, credit bond, or debt of

gratitude. When presented with limitations, they still choose the most feasible profit-

maximising option, e.g., selling dried paddy rice to trader-millers. They likewise try to

overcome task and general environment factors and disregard social and cultural capital.

By contrast, the farmers who are not motivated by price decide based on convenience either

because of fixed employment, old age, heath problems or labour disutility; social factors such

as amicable personality, risk-aversion, or social ties; or cultural values such as debt of gratitude,

212
Discussion

the ethos of shame, or traditions. Their priority is not price, but it does not mean that they do

not want to sell to a more direct buyer who offers them the highest gains. While some do not

care about the price much, many are just constrained by factors such as social ties or cultural

values, which are more important to them than price. In general, the farmers who hold on to

factors such as debt of gratitude are more likely to forego any amount of money. By contrast,

those selling because of a personal alliance, risk avoidance, or labour disutility tend to choose

another buyer given the right price. These farmers have a price priority threshold which, when

met, would result in selecting a buyer because of price.

Thus, this study provides evidence that farmers’ choices may not always be aimed at profit-

maximisation. Still, it is a constant pursuit of self-interest: higher income, self-respect, pride,

stronger social ties, upholding values, and convenience. Consequently, given that not all

farmers want higher incomes, not everyone would opt for the profit-maximising marketing

channel. Hence, while the constraints of the price-motivated farmers can be addressed to help

them maximise profit, these will not affect non-price-motivated farmers' preferences since their

decisions were made without the influence of any constraints. This non-profit-maximising goal

may also be why some farmers disregard many government projects – why they would not use

the mechanical dryers, why they would not find a way to sell to NFA, or why they would not

fully join an association and enjoy its benefits. It is possible that some other factors are

constraining their choices or their choices are already optimal for them.

8.3 The Factors of Movement

The results of the study show that many of the rice farmers in the Philippines are still selling

to indirect buyers and are thus at the start of the long market value chain. While the value chain

of rice in some Asian countries have shortened given the reduced and undermined role of the

213
Discussion

village trader and the increased reach of rice millers and urban wholesalers (Reardon et al.,

2014), the role of middlemen in the Philippines is still strong, given the lack of access to direct

buyers. The many changes in the market value chain in other Asian countries, such as the

increasing commercialisation of paddy farms, increasing disintermediation, and the

disappearance of the tied-output credit market, is yet to be experienced by the Filipino farmers

given the right projects and interventions such as those that could make energy and

transportation more efficient and improve the farmers’ off-farm income (Reardon et al., 2014).

Moreover, because of the lack of access to an affordable and legal system – particularly given

the small size of the informal market, absence of spot markets, and contracting – as is the case

of the farmers in Kathmandu, Nepal, many of the farmers in the Philippines are not moving

much within the market value chain. The majority of them stay within the same type of buyers

or even sell to the same buyer, locally called suki. The suki system is a type of recurrent

relational contract wherein there is a written or verbal contract with implicit or explicit terms

that are open to ex post adaptations to unforeseen circumstances (Bhattarai, Lyne, & Martin,

2013). Many farmers tend to sell regularly to specific buyers through this dyadic relationship

in return for benefits such as higher prices and credit access (Dolan, 1991). However, while

relational contracts result in premiums, bonus payments, and fewer/lower transaction fees

(Bhattarai et al., 2013) and while the suki system's original idea is to develop loyalty in return

for benefits ‒ particularly better prices (Dolan, 1991) – the results show that it is not always

the case for the rice farmers in the Philippines. Although some findings indicate that prices in

the informal markets are set in consideration of the farmers (Baloyi, 2010), this only happens

when there is no credit involved. However, many (27.6%) of the suki farmers in the Philippines

have loans that disadvantage them, obliging them to sell to their suki buyers at their price offer

even if there was no trade contract. Given how significant loyalty, culture of shame, and debt

214
Discussion

of gratitude are for the farmers, these values assure the buyer-creditors that they will sell to

them at whatever price they have to offer. Therefore, these farmer-debtors do not only miss out

the usual benefits from relational contracts, they also lose the opportunity to sell at the most

remunerative price because it limits their options and bargaining power.

In general, the farmers keep the long-term reciprocal interdependency brought about by the

debt of gratitude from credit favours and endure the disadvantages that often come with selling

to their creditor-suki, i.e., lower price or high deductions, because of their high regard for a

smooth relationship that makes them want to avoid confrontation (Dolan, 1991). Sometimes,

they compromise, such as splitting the harvest between two buyers, selling to a particular buyer

every other harvest, or renegotiating the price to show their desire to keep the relationship.

However, this study also found that farmers can break away from the suki relationship when

the sale is blatantly disadvantageous, when the relationship has been damaged, or when the

suki buyer discontinues buying, thereby also confirming the findings of Bhattarai et al. (2013)

‒ that relational contracts are vulnerable to behavioural uncertainties. Overall, farmers try to

be loyal despite the disadvantages unless they lose trust, reflecting the role of trust in the suki

relationship (Dolan, 1991). They also tend to endure the relationship until they find a new

stable buyer, sometimes gradually moving from the old to the new suki while in the new

relationship's testing phase. Nevertheless, when a very offensive trade or circumstance

happens, the farmers, who pride themselves on loyalty and integrity, do not feel guilty for

breaking the relationship at once and selling to another buyer because they were not the ones

who first caused the rift. The farmers also may break the connection when the suki transfers

their business to another manager, proving that cultural capital is non-transferable nor

transmissible (Bourdieu, 1986).

215
Discussion

While credit is one of the main factors that keeps the suki relationship and prevents many

farmers from choosing a buyer freely, the data show that they don’t necessarily select the buyer

with the highest price once they are freed from their obligations. This is also why only a few

farmers moved towards a more direct buyer or shortened the market value chain. Only the

farmers motivated by price or those who can overcome further constraints, such as lack of

drying facilities or expensive hired labour, were able to move to a higher marketing channel

level. Some farmers sold to a less direct buyer that positions them farther from the consumers

in the market value chain.

However, this set of information is based on a very small number of observations wherein the

farmer moved between levels, making it difficult to make general conclusions and infers the

need for further research. Nevertheless, this study's qualitative data suggest that these

movements may result from more constrained circumstances such as the lack of capital or

worse access to drying facilities. More importantly, these can be effects of their motivation to

sell, which is not to maximise profit. For instance, some of them sell to agents to give them

income while also benefiting from its convenience, reflecting how much they regard smooth

relationships more than price. Moreover, the movement can result from a social tie to a new

buyer or because farmers do not want the inconvenience involved in postproduction work or

in searching for the buyer with the highest price. While some interview data also show that a

few farmers extended their market value chain by selling to less direct buyers because of higher

prices, this might just be a misconception. For instance, farmers may have compared the prices

of very small trader-millers in their area with travelling assemblers (indirect buyers), who may

offer higher prices but have higher deductions and or have inaccurate weighing scales. They

usually deduct around 3 to 5 kilograms of fresh paddy rice per sack, in which each sack has 50

kg, on average. Therefore, a buyer can offer a PHP 0.5 (USD 0.01) higher price but deduct

216
Discussion

around 5 kilograms per sack, translating to a discount of around PHP 1.7 (USD 0.034) per

kilogram from the given price, resulting in an overall lower buying price per unit. Given this

tactic, the traders can guise the price they offer as the “highest”, making farmers believe it to

be true. Based on the interviews, this ploy is working well as farmers believe that the NFA, a

government regulatory body, is offering lower prices than the traders because they buy at PHP

17 (USD 0.34) per kilogram. They have discounted the fact that the NFA is not imposing any

weight deduction, so it pays PHP 850 (USD 17) per 50 kilograms, while a trader who buys

even at PHP 18 (USD 0.36) per kilogram only pays PHP 810 (USD 16.2) for the same volume.

The analysis of the farmers' movement along the market value chain presents four implications.

First, these exhibit the possible effects of different factors such as risk-aversion, better

postproduction facilities, and price motivation. In addition, this highlights the critical role of

credit and interpersonal relationships on farmers' movement along the market value chain.

Specifically, farmers selling because of credit can only move to a new buyer when they find a

new credit source.

Second, this study shows that the marketing channel choice is both complementary and a

substitute for other buyers, contrary to what Dlamini-Mazibuko et al. (2019) believes to be just

a substitute; however, this difference may be subject to differences in products (i.e., they

studied vegetable marketing), location, and methods. Dlamini-Mazibuko et al. (2019) defined

that complementarity in marketing channels is displayed when the choice for a particular buyer

is dependent on another buyer, while the substitutability is implied when farmers choose

another buyer because they did not meet the conditions to supply to their first buyer option.

Considering such definitions, the data of this research suggest that the marketing channel

choices of farmers ‒ especially when they are moving to a new buyer ‒ display both

complementarity and substitutability. It is complementary when the first-choice buyer, e.g.,

217
Discussion

suki, is not available and thus suggests an alternative that makes the next choice dependent on

the other buyer. By contrast, it is a substitute when the farmers sell to another buyer because

they cannot sell to consumers or their first choice due to constraints such as drying facilities,

liquidity, or market access. The substitutability of marketing channel choices goes well with

the neoclassical economic theory principle that individuals maximise their profit, and therefore

always choose the buyer with the highest price if they can fulfil the conditions to do so.

Nevertheless, this study demonstrates that not all farmers are profit-maximising. Some of them

decide their buyer preferences based on social ties or cultural values, which increases the

likelihood that their choices are complementary.

Third, this study provides evidence that, while the farmers’ preferences and choices are not

always aimed at profit maximisation, they consistently pursue their welfare or self-interest.

This, however, is not limited to price and could encompass non-price goals such as smooth

relationships, convenience, respect, or risk avoidance. The non-profit maximising attitude of

farmers can be exemplified by those who do not even bother to find a buyer with a more

remunerative price despite minimum constraints after being free from their bond to their

creditor. Some of them did not even compare the product prices, nor did they carefully examine

the pros and cons of their decisions. Instead, they just sold to whoever arrived at the farm first

because of convenience, social ties, or cultural values. These farmers are not after maximising

their income, which was common among the farmer-respondents. In fact, many of the

interviewees confessed that they are aware that they are not even maximising their income from

rice production – that despite knowing how adopting a simple technology can increase their

yield, they choose not to use it because they are content and do not need more money ‒ even if

they are not rich. Similarly, they know that buyers with higher prices can improve their income,

yet they choose to sell because based on convenience, cultural values, or social relationships;

218
Discussion

their goal is not to maximise their profit. Although it is also possible that some of these farmers

still want to improve their buyer choices, the qualitative data show that they will not do it at

the expense of using their rest time, breaking a social tie, or discounting their cultural values.

This attitude of not focusing on money might have something to do with the Philippines'

sociocultural environment characterised by strong religious faith and interpersonal

relationships. These encourage farmers to focus on goodwill and maintain smooth interpersonal

relationships because God will take care of things. (Dolan, 1991).

Lastly, this study shows that farmers understand their options and can account for the best

opportunity using the information they possess. This is contrary to the claim made by Sen

(1993) – that in the process of achieving objectives, which he called “freedom”, an individual

may not find the presence of non-superior alternatives. Sen (1993) believes that sometimes

opportunity-freedom does not take complete orderings because of the dissimilarity between

one choice and another. Nevertheless, this study provides some evidence that farmers can

distinguish a superior option, even in the presence of non-measurable aspects that are totally

different. Specifically, they can weigh the importance and assign value to social ties, cultural

values, risk, and even convenience. Although they find it hard, the farmers can rank their

preferences in complete order. Their choices, however, may not always be optimal, or their

movement from one marketing channel to another may not always be an improvement of their

previous choice, but this does not reflect their inability to identify superior alternatives. First,

as previously explained, this can manifest their non-price motivation, which makes even an

indirect buyer an optimal choice for non-price reasons. Second, it can be because of factors

that constrain the farmers from choosing the best buyer (De Janvry et al., 1991). It can imply

new circumstances or challenges such as external environment risk, credit issue, or the non-

availability or non-access to the preferred buyer. Overall, the data suggest that the rice farmers

219
Discussion

can identify the best option and rank their choices with complete orderings depending on their

motivation, the constraints, the factors, and the price priority thresholds of those factors.

However, it should not be measured in terms of profit because, as was previously stated, it is

not always the objective of farmers.

8.4 The Typology of the Farmers based on Price

Overall, the results show that numerous factors affect farmers' marketing channel choices –

from personal, economic, social, cultural, and external environment factors – illustrating how

complicated the marketing channel choices can be. The numerous possible combinations of

these factors can also result in many farmer typologies. Nevertheless, in line with the

assumption of the neoclassical economic theory that farmers choose their buyers intending to

maximise their profit, the farmers' typology in this study is presented based on price.

The findings suggest that there are three main types of farmers: (1) the Price Prioritiser (PP) or

those who only consider the price as the determining factor when choosing a marketing

channel; (2) the “Unit price prioritiser” (UP) or those who choose based on the unit price

difference; and (3) the “Non-price prioritiser” (NP) or those who disregard price in favour of

non-price factors (Fig. 12).

Figure 7. A simple representation of the farmers who are PP, UP, and NP.

220
Discussion

8.4.1 The Price Prioritiser

The farmers referred to as “Price Prioritiser” are those who choose a marketing channel based

on the price, subject to the constraints presented by other factors. These farmers are computing

whom to get maximum profit from, considering the buying prices, weight deductions, and

costs. If likened to elasticity, these farmers have inelastic price priorities because the net

income is the only deciding factor for them, given they are focused on profit. Thus, this group

are mainly profit-maximising farmers. They do not anchor their decisions on kinship,

friendship, or the suki system, nor do they give more weight to cultural values like debt of

gratitude or ethos of shame. These farmers stand by the belief that business is business, so

nothing personal should go in the way of choosing what would result in profit-maximisation,

even if the additional profit is just a few cents. They live by the belief that if their family or

friends are not the highest buyers, then they were the ones who first broke the personal alliance

system, which dictates that one should support, among others, relatives, friends, and business

partners (Dolan, 1991; Philippines Country Study Guide, 1999). If they did not offer the highest

price possible, which other buyers could offer, these buyers took them just as any other clients

they want to profit from, instead of a relative or friend whom they should support or favour.

Therefore, these farmers believe that they are excused for not helping them as well. In general,

based on the results, these farmers do not place much weight on social or cultural capital

because of their higher schooling years. Therefore, it confirms Dolan's (1991) assertion that

educated Filipinos feel less obligated to decide on buyers based on non-price factors such as

debt of gratitude.

Notably, these PPs are assumed to maximise their profit, given their constraints limiting their

ability to choose the best price (De Janvry et al., 1991). As such, it does not mean that they all

sell milled rice to consumers. For instance, even those who sold paddy rice to the highest

221
Discussion

possible buyer can be considered PP since the farmers’ circumstances only allow that, e.g., the

weather is not favourable and there are no dryers available. Similarly, some farmers are

constrained to sell to their creditor because of credit, yet to maximise their income, they only

sell an amount equivalent to their debt while they sell the rest to the highest buyer accessible

to them. Overall, farmers who maximise their profit given the different constraints they have

are considered PP.

8.4.2 The Unit Price Prioritiser

The “Unit Price Prioritiser” or UP farmers, in contrast, are those who choose price over factors

such as social relationships, cultural values, or convenience but only after meeting the price

priority threshold. This threshold is the rice's unit price that must be exceeded for farmers to

prioritise price over non-price factors. The UP farmers weigh their potential loss or forgone

profit against constraints such as social relationships, cultural values, or convenience.

Specifically, they assess whether the additional profit is sustainable and if it is worth the

possible loss of a friend, dismay of a family member, or the conveniences of selling to a suki

or indirect buyers. Many interviewed farmers disclosed that they sell to their creditors because

of debt of gratitude when the price is below the price priority threshold of PHP 0.20 (USD

0.004). For them, any price difference higher than that is reason enough to forget the debt of

gratitude and sell to the highest buyer ‒ especially if they have other buyer options during the

following season.

Similarly, some farmers who sold because of social relationships, relatives, or friends also

disregarded social capital and sold to the highest buyer when the price met their price priority

threshold of PHP 1 (USD 0.01). They explained that any price difference equivalent to or

higher than that is taking advantage of them by those they regard as family or friends, whom

222
Discussion

they also expect to favour them. Thus, if they sell to another buyer for higher profit and that

family or friend buyer found it offensive, that means that their relationship is only based on

business and could not be classified as part of the real personal alliance system (Dolan, 1991;

Philippines Country Study Guide, 1999). However, they clarified that before selling to the

highest buyer, they would first ask the family or friend buyer to match the highest buyer's

offered price, which shows how much these farmers value their social relationships.

Nevertheless, they only reciprocate the importance that is given to them. More importantly,

these farmers are only willing to forgo prices below their price priority threshold, which differs

per person. However, based on the trend, many farmers would overlook their cultural value of

shame and social relationships for as low as PHP 1 (USD 0.01), but they can accept much

higher foregone income because of debt of gratitude, depending on how deeply rooted it is.

This suggests that many farmers give real debt of gratitude more importance than kinship or

friendship.

8.4.3 The Non-Price Prioritiser

The “Non-price Prioritiser” or NP farmers are those who do not sell to the highest buyer

regardless of the profit margin (Fig. 15). The main reasons why these farmers are not selling

to the highest buyer are detailed in the following:

Strong personal alliance system. These farmers have a solid commitment to the personal

alliance system, which usually dictates that a person’s loyalty goes to the people who belong

in the said system. First, to relatives, whether immediate family members or extended, i.e.,

affinal or consanguineal. The system similarly includes ritual kinship, such as those anchored

through marriage ceremonies, baptism, and confirmation. It likewise covers friendships and

suki relationships. Similarly, it includes people with whom a dyadic bond has been built

223
Discussion

because of concepts such as debt of gratitude (Dolan, 1991; Philippines Country Study Guide,

1999).

Particularly, many farmers are NP because of the suki relationship. In this instance, the buyer

may not be within the initial personal alliance system of the farmer. However, because of the

constant or regular business interaction not because of debt, they start to build a patron-client

bond (Dolan, 1991; Philippines Country Study Guide, 1999) that result in benefits such as good

income, transportation services, reliability, convenience, and sometimes, interest-free credit.

And because many of the farmers have been selling to their suki buyers for decades, many of

them have advanced the relationship into a real friendship or ritual kinship, e.g., the buyer has

become a godfather of the farmer’s child or vice versa. Thus, this study confirms that market-

exchange relationships can develop into genuine relationships beyond business (Dolan, 1991;

Philippines Country Study Guide, 1999).

Overall, the NP farmers prioritise social relationships such as blood relations, friendships, suki

relationships or cultural values such as debt of gratitude over price because of social and

cultural capital. For instance, many farmers sell to their relatives despite their lower prices

because they believe that loyalty is nothing less than a social imperative. More importantly,

they want to avoid anything that would damage their self-esteem, which Filipinos are known

to be sensitive to (Dolan, 1991; Philippines Country Study Guide, 1999).

Risk avoidance. Apart from the personal alliance system, non-price priority can also be a result

of personality. Specifically, risk-aversion is vital in farmers' non-price priority, especially those

with suki creditors and those with low off-farm household income. Many of these farmers tend

to be loyal to their creditors, who also happen to be traders. The main reason for this is the fear

that they will not find another lender if they lose them, given the limited number of informal

224
Discussion

creditors in the area and the lack of the ability to borrow money from formal institutions such

as banks. Hence, despite not having a contract to sell and not being able to bargain for prices,

and no matter the price difference, they still market to the traders who lent them money because

they need a safety net to help them in emergency situations, which encourages them to keep

their social ties.

Strong cultural capital. Other NPs sell to specific buyers because of strong cultural capital,

particularly debt of gratitude. This Filipino trait is very dominant, especially when farmers

receive a considerable favour, such as lending money for hospitalisation that saved a life or

allowed a child to finish college and get a good job. In fact, based on the qualitative interviews,

it can be powerful enough to turn a farmer who used to prioritise price into someone who

prioritises non-price. This establishes that cultural capital, particularly debt of gratitude, can

change a farmer's motivation depending on its roots. Similarly, many farmers sell to their social

alliance or even to the buyer who first offered to buy their harvest, regardless of price, because

of the ethos of shame and the desire not to hurt anybody’s feelings, reflecting how much

farmers regard their reputation and good interpersonal relationship. Overall, these farmers’

longing to maintain a personable trait and a good relationship with others is more important

than price.

Disutility of labour. The decision of the farmers as workers has a lot to do with their marketing

channel choices. In general, NP farmers regard their labour and time as more precious than the

additional profit they would get from selling to more direct buyers. Thus, these farmers are not

willing to exert extra effort to canvass, dry, or transport their product in return for certain

additional net income. For instance, many farmers confessed that they have never dried paddy

rice since they started farming because it is laborious and time-consuming; despite not having

225
Discussion

anything more productive to do, they would still rather sell fresh paddy rice for convenience,

believing that their efforts and time are more precious than what they will gain.

Similarly, some farmers will just sell to whoever comes first to their field during harvest, saying

that it is convenient. They can collect it immediately, and thus, they can avoid wasting their

time waiting for another buyer to pick it up. Still, others will just sell their whole harvest to

their creditor despite very low prices to evade the process of canvassing and looking for higher

buyers. Moreover, several farmers do not even canvass for prices at all, just to avoid the

inconvenience of travelling. Lastly, many farmers are also unwilling to transport their harvest

to the buyer despite additional income, quoting the many inconveniences of trucking. All these

instances lead some farmers to be non-price prioritisers who sell to indirect buyers, whose main

advantage is convenience; thus, showing how much effect the disutility of labour has on

marketing channel choices. In fact, it has such an enormous influence that some farmers would

even choose convenience over social ties or debt of gratitude.

Labour disutility can be attributed to the farmers’ diminishing marginal utility for income,

which starts when they achieve their income goal; after which point, they do not want to exert

much effort just for what seems to be petty cash or an insignificant amount. This makes the

farmers think that the profit from selling fresh and dried paddy rice is the same or that their

price difference is negligible. And while this belief is influenced by personal factors such as

age, health conditions, or personality, it can also be due to fixed employment that is an

economic factor.

Specifically, the labour disutility mainly from having fixed employment can be attributed to

the price difference between the income of farmworkers and other jobs. Many of the employed

farmers put a premium price on postproduction-related labour, such as drying. In fact, the price

226
Discussion

they put is so high that some of them want an additional net income equivalent to at least a

third of their average net income per hectare. This could be because they compare it to the rate

that they are being paid at their daily job, which is much higher. This wage difference could be

referred back to the non-separable household model that presumes that the labour market is

imperfect, leading to the non-separability of the production, consumption, and worker

decisions (Sadoulet & De Janvry, 1995). This market failure results in wage discrimination,

with an unparalleled income between manual dryers and employee salaries. Overall, because

of the seemingly small additional benefit from drying, NP farmers would not even consider

engaging in it because they don’t want the hassle and would rather spend their time resting or

doing other things, which are not necessarily income-generating.

Therefore, this study provides evidence that many of the NP farmers’ maximum utility as a

labourer is not about getting the highest profit as dictated by the neoclassical economic theory

– it is geared towards saving time and effort because of the disutility of labour and the

diminishing marginal utility for money. These are farmers who would not maximise their profit

at the expense of personal convenience, always trading price for it, in one way or another. This

demonstrates that the NP farmer's decision as a worker or labourer dominates whether to sell

fresh paddy, dried paddy, or milled rice, which in turn affects their marketing channel options.

In general, while the findings suggest that there are three typologies of farmers based on price,

it does not imply that a certain farmer's typology cannot change. In fact, the qualitative data

illustrates that non-price priority farmers will not always prioritise non-price factors. They can

change into price-priority farmers given a different situation, such as when they have better

liquidity and given a different set of buyers. This is because the non-price priority is largely

dependent on how important the non-price factors are, e.g., how deep the debt of gratitude is

227
Discussion

or how strong the social tie is. Any social or cultural capital that is not as strong will typically

result in unit-price priority, while certain damage to them can even result in price priority.

Overall, the results provide evidence that individual preferences and choices are always aimed

towards utility maximisation, although that does not necessarily mean profit maximisation.

There is no evidence to show that farmers are not utility maximisers or that their goal is limited

to ensuring “simple reproduction” (Vergopoulos, 1978). Based on the qualitative interviews,

the farmers’ goals can range from simple to complicated, from higher income, self-esteem,

enforcing social ties, upholding values, convenience, or just not stressing out about the choices.

Yet, whatever the farmer’s preference is, it is always geared towards the utility-maximising

option given the information and choice they have at the time of decision. It consistently fulfils

self-interest, which contrasts with Sen's (1993) argument that not all decisions are in pursuit of

one’s own interest or welfare. For instance, selling to agents to provide them with income also

boosts morale or provides more rest time. Similarly, selling to a relative could be done because

of the benefits of the personal alliance system, while marketing to whoever comes first is

convenient and has fewer complications. Therefore, while farmers' marketing channel choices

may not always be geared towards profit maximisation like a business, it is always inclined

towards maximum personal utility.

228
Conclusion

Chapter 9. Conclusion
The first part of this chapter sets out the essential details of this research, along with the

summary of findings. The second part details the practical and policy recommendations using

the new knowledge contributed to the field. Finally, the last part presents suggestions for future

research.

9.1 Summary and Conclusion

The choice of marketing channel is one of the most crucial decisions that farmers must make

because it directly and greatly affects their selling price and profitability. Nevertheless, the

literature on the factors affecting their choices is very limited, while these are crucial insights

needed to create effective projects and policies that could help increase farmers’ income. There

is a knowledge gap on the factors affecting the marketing decisions of grain farmers ‒

especially rice farmers ‒ even though they provide the staple food of the majority of the world’s

population and comprise a large part of the world’s poorest. To help address this knowledge

gap and consequently help improve their marketing income, the factors affecting the marketing

channel choices of the rice farmers in the Philippines, one of the top ten rice-producing

countries in the world, were collected through 300 survey interviews and 45 qualitative in-

depth interviews. The data were gathered following a conceptual framework that uses the non-

separable household model and social and cultural capital theories, laid under the assumption

that choices are made in consideration of the external environment similar to that of an

organisation.

The quantitative data were analysed using descriptive statistics for the farmer typologies,

random-effect logistic regression to determine the significant factors affecting the choice

229
Conclusion

between direct versus indirect marketing channels, bivariate correlations to verify

relationships, and movement analysis to document the change in the marketing channel levels

of farmers. By contrast, the qualitative data were processed using codes that are concept- and

data-driven, and thematic analysis that was both theoretical and inductive. Using the mixed

methods approach, more factors affecting farmers' marketing channel choices were revealed

while allowing triangulation that paved the way to produce more sound results. It similarly

gave the data more breadth and depth that provided a better understanding of farmers'

considerations when choosing a buyer.

Based on the quantitative and qualitative results, it can be concluded that the factors affecting

the marketing channel choices of the rice farmers in the Philippines are numerous. As assumed,

given the non-separability of the producer, consumer, and worker decisions, the factors indeed

range from personal, social, cultural, and external environment factors instead of the usual

sociodemographic factors presented in previous studies. The quantitative model specifically

provides evidence on the significant positive effect of variables such as training, full-time farm

job, landownership, off-farm income, and harvest frequency on choosing direct marketing

channels. Similarly, it demonstrates the negative effect of higher age, education, farm size, and

distance to market on direct buyers' choices. Nevertheless, these only comprise a small part of

the farmers’ decisions.

Variables that were added to cover some aspects of culture such as farm traditions and

indigenous origins; external environment factors, such as higher town income class and lower

provincial rice production volume; and personality, such as work motivation to fulfil basic

needs, and disinclination for fast cash payment and trucking services were also shown to

facilitate the choice of direct buyers. Nevertheless, all these factors still cannot accurately

230
Conclusion

model and explain the farmers' choices. The qualitative data of this study suggest that this is

because a considerable component of buyer choice is attributable to personality traits such as

risk-aversion and labour disutility; social ties such as kinship, friendship, or suki relationship;

cultural values such as debt of gratitude, ethos of shame, and loyalty; and the constraints in the

external environment presented by an adverse task environment, frequent calamities, poor

economic state, punitive sociocultural values, the corrupt legal system, and unfavourable

international policies, among others.

The results of the movement analysis also demonstrate that most farmers stick with the same

buyer because of credit and its cultural implications, risk-aversion, and the effect of the

personal alliance system ‒ particularly the suki relationship. These are some of the reasons why

farmers, even if they want to maximise their profit, maintain their position in the market value

chain. Overall, there are very few farmers who moved to a more direct buyer. In fact, there are

slightly more farmers who extended the market value chain or those who sold to less direct

buyers compared to those who cut the market value chain. Although there is a limited number

of observations showing movement between marketing channel levels, it does not indicate that

farmers cannot identify superior options or are incapable of ranking them completely. The

minimal movement towards a more direct buyer may reflect two aspects. First, farmers are

constrained by some personal, economic, social, cultural, or external environment factors,

forcing them to choose a less direct buyer, yet it can be their best substitute. Second, this

displays that some farmers’ objectives do not focus on maximising profit since others are

centred on maintaining good interpersonal relationships, maintaining self-dignity, and helping

others. While some of these farmers may want to sell to the buyer with the highest price, they

are not willing to do so at the expense of other factors that are more important to them.

231
Conclusion

In conclusion, this study offers evidence that the farmers do not always choose to sell to the

most direct marketing channel because their goal is not always to maximise profit; instead, it

is to maximise utility. Although more farmers still aim to maximise their profit given more

favourable circumstances, many are already happy with their buyer choices because they get

their utility mainly from non-price factors. On the one hand, those who want to maximise their

income sell to more direct buyers and hurdle constraints that can come in the form of personal

factors such risk-aversion and labour disutility; economic factors such as liquidity or

transaction cost; social factors such as kinship, friendship and suki relationship; cultural factors

such as debt of gratitude, loyalty, and ethos of shame; and external environment factors

including both the task and general environments. On the other hand, those who are not profit-

maximising choose their marketing channels based on social ties, cultural values, or

convenience or weigh them first against price. Following that logic, the farmers can generally

be categorised into three groups: the (1) Price Prioritiser (PP) or those with an inelastic

preference for buyers with the highest price; (2) Non-price Prioritiser (NP) or those with an

inelastic preference for buyers owing to social ties, cultural values, or personality factors; and

(3) Unit Price Prioritiser (UP) or those who choose price factors over non-price when the

buying price meets their price priority threshold.

Overall, while not all farmers want to improve their income or their marketing channel choices,

many farmers are still hoping to get better profit; and they are waiting for assistance to be able

to sell to the buyer with the highest price, if not to the most profitable buyer possible.

Addressing the constraining factors found in this study could therefore help them sell to more

direct buyers and achieve their goal to get optimal marketing income.

232
Conclusion

9.2 Practical and Policy Implications

This study aims to identify and better understand the factors affecting the marketing channel

choices of rice farmers in the Philippines to help direct future projects and policies that aim to

help them improve their income. This study's findings are thus useful in guiding the

government strategies and policies in terms of agricultural development, from local to national

level.

Overall, while some farmers are not choosing their buyers based on maximum profit, many are

still eager to sell to more direct buyers. However, they are facing many constraints that only

allow them to sell via the most profitable option possible, which need to be addressed. Among

the constraints is the limited off-farm income, which results in selling to indirect buyers. There

is a need to further intensify the efforts to provide farm households with livestock or vegetable

production and facilities to process agricultural by-products to give them additional income

without limiting their farm time. The income from this will also help protect them in times of

crop failure by having diversified income sources and minimising their risk-aversion that also

leads to selling to indirect buyers. This will likewise save them from borrowing money, if not

help them minimise it, and give them the liquidity to do postproduction processes and to market

the product directly to consumers. The farm household's financial self-sufficiency can impede

the effect of debt of gratitude, the ethos of shame, and risk-aversion on marketing channel

choices since these are all the result of credit or the lack of financial stability. Based on the

experience of other Asian countries, having their own funds from off-farm incomes can

effectively reduce, if not eliminate, “tied output-credit markets” where farmers sell to specific

traders who gave them credit to “lock them in” implicitly (Reardon et al., 2014).

233
Conclusion

Concerned institutions might likewise consider conducting financial literacy training with

marketing and budgeting tips for farmers. This can help guide them when choosing a buyer,

and at the same time, minimise impulsive spending and anxiousness to get paid at the expense

of earning a higher income. Assisting other farm household members to get education and

training that would help them acquire skills will also increase their employability and, thus,

improve the family's financial stability. The government should consider continuing the

conduct of agricultural training sessions that can indirectly help farmers expand their marketing

networks, which could help them find better marketing channels and encourage them to learn

more profitable marketing practices. Moreover, the government can help lower the

transportation costs of rice trading to encourage more sales to direct buyers. This can be done

by continually improving rural roads and partnering with private institutions in creating

railways and improving port facilities to allow more efficiency and lower the cost of grain

transport by land and sea, which is especially important for an archipelago like the Philippines.

Furthermore, the need for marketing support from the government along with price policies to

improve the agricultural marketing environment may be more crucial now than ever,

considering the free trade of rice that was recently enacted. Hence, instead of stopping the

National Food Authority (NFA) price support, it should be strengthened, albeit with

considerable restructuring in management and strategies. Nonetheless, it should still always be

ready to buy the farmers’ harvest at a reasonable price. Moreover, the Local Government Units

(LGUs) who successfully supported the local rice procurement to save farmers from

bankruptcy during the onset of the rice trade liberalisation must be used as marketing models.

Considering the very low buying price of paddy rice after the rice trade liberalisation, the

farmers now have a higher desire to process their rice to give them sufficient income and sustain

234
Conclusion

their means of living. Thus, given better policies and easier procedures, any government price

support will be undoubtedly availed and used by farmers ‒ especially the small-scale ones.

Specifically, as part of the government’s budget to support the farmers’ marketing, the NFA or

LGUs should also function as the rice trader, with the capacity to buy all the farmers’ harvests

in the town. This will set a floor price for paddy rice, thus ensuring that the farmers will get a

good share in the consumer prices apart from having a sure buyer for their product, which could

further counter the effect of risks on their choices. Direct marketing can also be through

cooperatives. Ideally, the cooperative can buy the farmers’ harvest immediately, process it, sell

it as milled rice, and later give them more earnings through dividends. In any case, whether it

be through their cooperatives or LGUs or NFA, this will be shorter and more direct. Thus, it

will prove to be more beneficial for both the farmers and the consumers as having access to

direct buyers and having the assurance of at least one buyer would encourage farmers to sell to

the marketing channel with the most remunerative price, while a shorter value chain would

mean lower milled rice price for the consumers.

At the very least, the government should support the farmers’ marketing by minimising the

constraints of selling to more direct buyers – especially the lack of access to it. It should

facilitate the direct marketing of farmers’ milled rice by linking them to consumers, hotels,

restaurants, or the food industry since the absence of a guaranteed market is also limiting

farmers' marketing income. The village markets of big private subdivisions and farmers’ days

at malls could be a great venue for this if farmers could be assisted with the basic marketing

strategies and selling fees. The Hotel and Restaurants Association of the Philippines (HRAP),

along with their regional groups, can also be tapped to create a sure demand for the local

farmers’ produce. Filling the local demand with local supply will not only address market gaps

235
Conclusion

faster, but it will also make the marketing more efficient and consequently lessen the

transportation required, which could have the added benefit of helping lessen greenhouse gas

emissions in the country. A project to link farmers to direct buyers such as the members of the

HRAP will significantly help farmers and their cooperatives optimise their income from their

farm by selling directly to consumers in bulk, as suggested by the effect of living in towns that

are more populated and with more market. Meanwhile, to ensure that the buyers will buy the

farmers’ planted variety, there should be a partnership between the seed growers who plant the

seeds, ensuring that there is enough in the market, the farmers who will plant them, and the

food industry businesses that will buy them. At the very least, the government should create a

policy allowing the rice farmers to sell milled rice without applying for a retailer’s permit. This

should be their privilege since rice production is their business. This would greatly help many

of them to market their unused rice supply or to sell their harvest to direct buyers without

worrying about getting fines or sanctions. In support of this, the government should also

consider a price advisory for paddy rice; this could be sent to farmers’ registered mobile phones

to inform them of the daily rice prices of the biggest traders in the locality. This will help

increase their market information, giving them better bargaining power and helping them find

the best marketing channel.

More importantly, this study underscores that to help farmers improve their marketing income

effectively, it is crucial to consider the various non-sociodemographic factors, such as

personality, social capital, and cultural values, when creating policy interventions for the

Philippine rice industry. Specifically, efforts should focus on addressing labour disutility, risk-

aversion, and the creditor-buyer tradition.

236
Conclusion

Firstly, it would help to make the postproduction process of rice easier to reduce the negative

effect of personal traits such as labour disutility and thus, allow even farmers who are older or

with health conditions to dry and mill their harvests. The government should continue to create

and implement projects aiming to minimise the demand for labour ‒ especially during the peak

season ‒ to make it more affordable. One way to do that is to implement projects that would

give farmers easy access to efficient dryers, rice mills, and storage. This can be done with more

drying pavements in dry towns and mechanical dryers in wet ones. These, of course, should be

done in consultation with the farmers to identify and understand their real needs and consider

their situation better. The government should also consider installing more storage and milling

facilities. These will ease the physical and financial burden of postproduction. It will also lessen

transaction costs when gathering market information and when marketing, which has also been

shown to result in selling to more direct buyers. The postproduction facilities can be installed

and managed by private institutions to avoid corruption. Compared with cooperatives, these

people would be professionals and trained in business management. The government can

subsidise the equipment for them to be able to offer affordable services from the start since

getting back the cost of investment won’t be factored in the price of service. This way, the

farmers will have higher incentives to use the facilities since their expected income from selling

dried or milled rice will be higher. Alternatively, the government can also help existing private

large rice mills to expand their drying and milling equipment, receive equipment subsidies, or

even get direct state investments, just as has been done in China (Reardon et al., 2014). In

return, rice millers can then offer their postproduction services at very affordable prices to

encourage farmers to dry and mill their harvests or offer higher buying prices for fresh paddy

rice. Moreover, postproduction facilities can be managed by the local government units (LGUs)

to ensure proper accounting and lessen the chances of corruption, if not mitigate it. The

237
Conclusion

facilities in this case, however, should be installed in government lands, rather than seeking

leased lands from private individuals that can later result in the facilities' corruption or

mismanagement. It would also allow its more prolonged use instead of just one decade of leased

contracts. If possible, the NFA should also be devolved and made part of the LGU to manage

these facilities. Being devolved will make the procurement and trading rule more location-

specific and subject to the local situation and farmers' needs. In any case, the NFA has many

facilities strategically located around the Philippines, some of which are disused but can be

made useful again by the LGUs. If possible, the LGUs should also provide transportation

services since it is vital in improving the market access of small farmers, given their lack of

access to it.

Secondly, risk-aversion can be partly addressed by giving the farmers better access to credit

and buyers since this will assure them that they have other credit sources and buyers even if

they break their marketing bonds with their suki. Credit access frequently leads to cultural

factors such as debt of gratitude and the ethos of shame, and the non-profitable type of the suki

system. Thus, it is crucial to provide farmers, especially small-scale ones, with formal lending

sources that have much lower interest rates but offer the same services that informal lenders

give – particularly fast and accessible credit. This will minimise the negative effect of social

factors such as the predatory suki type, ethos of shame, and debt of gratitude and thus, give

farmers bargaining power and increase their marketing income. The government should create

credit programs that are faster and easier as these could ease the farmers from the obligation to

sell to their creditors, allowing them to sell the buyer who provides the most remunerative

price, thereby helping them avoid the ill effects of the predatory type of suki relationship that

comes through credit. This will likewise encourage farmers to cut the market value chain since

238
Conclusion

farmers who are secured of their credit access even during times of calamity are more confident

to choose the most profitable marketing channels. Specifically, to address the credit access

issue, policymakers and government institutions can devise ways to ensure repayment without

collateral since very few farmers can offer securities. This could be done using sureties, just as

how informal lenders are doing it. Since cultural values such as debt of gratitude and the ethos

of shame were enough for the informal lenders to ensure repayment without collateral, the

government can perhaps also use this as their strategy. Specifically, the government can partner

with a private company or a separate association to manage the credit and collections from

farmers, whether it be daily or on a per-semester basis. Private banks and credit institutions can

also be tapped to give low interest rates and facilitate loans for farmers’ cooperatives, or better

yet, individual farmers. It may likewise be useful to incorporate credit into the above proposed

government-trader project. The credit can come both in cash and inputs such as seeds and

fertilisers, which can be deducted from the farmers' pay once they sell. Emphasising their social

obligation to pay and relating it to debt of gratitude and the ethos of shame may help ensure

repayment. This should work well unless a natural disaster or disease hits the farmers, in which

case their credit should be covered by insurance or waived by the government. Improving the

access of farmers to credit will help them sell to more direct buyers and also significantly

improve their marketing income that otherwise goes to paying off the high credit interest of

informal lenders. Overall, without ties created by credit and social relationships that are both

deep-rooted to financial insecurity, the farmers will not be bound to stay with the same buyer,

even at a loss. This should allow many of them to prioritise price when choosing their buyer.

Moreover, given how important to farmers are their relationship with their buyers, the

government institutions tasked to support farmers' marketing should always consider their

239
Conclusion

feelings and treat them with respect. Otherwise, they will not use or avail of the services and

would not feel indebted. Perhaps, this is one reason why the farmers do not engage with the

NFA buyers – because they lack a personal touch and do not really build interpersonal

relationships with them, unlike the informal creditor-traders. This might also be why farmers

have poor credit repayment rates when it comes to government loan projects. Given how vital

interpersonal relationships are with Filipinos, it is easier for the farmers to move from a buyer

with the highest price to a more sympathetic and kinder buyer. Thus, project implementers

should always treat them with care, love, and respect and not act as if the farmers owe it to

them, which is the usual mindset.

Finally, the government should consider the types of farmers in terms of price priority before

designing agricultural policies or projects. These should offer new business models and

interventions that take into account the main factors affecting the buyer choices of farmers.

These projects should be able to minimise the effect of constraints besetting the majority of the

farmers – whether it be coming from personality, social, or cultural factors, among others. By

identifying the type of farmers within the communities, priority projects can be aligned to

address their main constraints.

9.3 Further Research

Production and consumption should both be improved to make the locally-produced rice more

competitive, thus ensuring the sustainability of rice farming in the Philippines. Hence, research

should not only be focused on production but also on improving the marketing income of

farmers. Failure to significantly increase farmers’ income may discourage rice farming, making

the Philippines dependent on imported rice. As such, Filipinos might have to learn not to

240
Conclusion

depend on rice as their main staple when huge rice exporters like Thailand, Vietnam, or

Cambodia refuse to sell their rice because of global constrictions such as during the Covid-19

pandemic. Thus, it is recommended to conduct further studies exploring farmers' marketing

choices and behaviour to help improve their profit.

Specifically, this study highlights the importance of combining mixed methods in

understanding farmers’ marketing channel choices. Given the results, it is suggested that future

studies on agricultural development ‒ particularly marketing channel choices, which involve

factors of relationships, combine both qualitative and quantitative analyses. As was manifested

in this study, the quantitative model can only explain a part of the farmers’ decisions, which

the qualitative data can provide. Although challenging, it may also be useful to consider

including more factors that will reflect the crucial personality, social, and cultural factors

presented in this study.

These factors have rarely been presented in any previous research on the factors affecting

marketing channel choices, not just in the grains industry but in agricultural marketing in

general. Nevertheless, this study has gathered evidence of the factors that are mainly affecting

the farmers’ choices. Moreover, despite the non-dramatic change within-person of these

variables, they are not time-invariant. They change depending on the farmers' circumstances –

their liquidity, other available options, and their labour disutility, among others. Hence, there

is a need to conduct further studies that would incorporate them effectively, not only in the

Philippines but also in other developing countries that are chiefly dependent on rice for

agricultural development. This might also help show the real effect of other variables, e.g., age,

education, and product volume, that have otherwise inconsistent results and similarly verify

the results of this study.

241
Conclusion

Further studies can likewise explore how many farmers are PP, UP, or NP to understand the

potential beneficiaries of a specific project and analyse its costs and benefits better. Similarly,

a study that identifies and explains the different factors affecting the choices of PP, UP, and

NP farmers is also recommended. Using a two-stage Heckman selection model could help

provide reliable and accurate recommendations on addressing each of the typology groups’

constraints. It may also be good to compare its findings with this research to see how the effects

of the variables will change. Likewise, further studies could focus on exploring the effect of

price versus non-price factors on farmers’ decisions and the impact of the buying price on the

farmers' utility to better understand their goals and consequently adjust the policies and projects

accordingly. Modelling the choices of and factors against the price priority threshold of the UP

farmers may also be worth exploring further to understand its flexibility or elasticity. More

importantly, given that many of the farmers have a price priority threshold, it will also help to

study how to measure this. Perhaps a price priority index could be done by having farmers

answer a series of questions that would determine the weight of price versus non-price factors.

This will not only help guide agricultural marketing but also the marketing of other products

and services.

Researchers may also want to compare the price that suki farmers and non-suki farmers get and

confirm the many variables affecting the suki relationship presented in this study. It may

especially be interesting to look more deeply into the effect of reciprocity, trust, and distrust

on social capital and marketing channel choices – perhaps also exploring how farmers put a

value on them. Finally, given the scarcity of studies on farmer movement along the market

value chain and the factors affecting it, there is a need to investigate this further to verify the

results of the current study. Future studies should include a larger number of observations when

242
Conclusion

farmers moved from one marketing channel level to another to provide more generalisable

conclusions on what factors really affect such movements.

Overall, this study's findings and the suggested studies will not provide a miracle solution in

addressing farmers' poverty or solving agricultural development. However, it can help better

understand the topic of agricultural marketing by providing in-depth information on how

different factors influence farmers' decisions. Moreover, it provides perspectives and helps fill

the knowledge gap on how farmers think and act, given the internal and external factors that

usually confront them. This can provide a sound basis for policies and projects, which can help

direct their path and improve their position within the market value chain. Ultimately, it will

contribute to creating effective custom-fit and appropriate market-related projects and policies

for the rice farmers in the Philippines and the other grain farmers in developing countries.

243
Appendices

Appendices

Appendix 1. Pearson correlation of variables with possible multicollinearity.

Correlations
TotalareaW LPareaW
Age Yrsinfarm (hectare) (hectare) SoldkgW
Age Pearson Correlation 1 .743** -0.049 -0.051 -0.057
Sig. (2-tailed) 0.000 0.142 0.125 0.085

N 900 900 900 900 900

Yrsinfarm Pearson Correlation .743** 1 -0.002 0.010 -0.011

Sig. (2-tailed) 0.000 0.947 0.775 0.735

N 900 900 900 900 900

TotalareaW Pearson Correlation -0.049 -0.002 1 .774** .673**
(hectare)
Sig. (2-tailed) 0.142 0.947 0.000 0.000

N 900 900 900 900 900

LPareaW Pearson Correlation -0.051 0.010 .774** 1 .785**
(hectare)
Sig. (2-tailed) 0.125 0.775 0.000 0.000

N 900 900 899 900 900

SoldkgW Pearson Correlation -0.057 -0.011 .673** .785** 1

Sig. (2-tailed) 0.085 0.735 0.000 0.000

N 900 900 900 900 900

**. Correlation is significant at the 0.01 level (2-tailed).

244
Appendices

Appendix 2. The Variance Inflation Factors (VIF) of the independent variables.

Coefficients
Standar
dised
Unstandardised Coeffici Collinearity
Coefficients ents Statistics
Std. Toleranc
Model B Error Beta t Sig. e VIF
(Constant) 0.997 0.215 4.636 0.000
Age (yr) -0.003 0.002 -0.059 -1.231 0.219 0.400 2.499
Yrsinfarm (yr) -0.002 0.002 -0.057 -1.154 0.249 0.373 2.684
Educ (yr) -0.017 0.006 -0.116 -2.944 0.003 0.586 1.705
HHchildren 0.000 0.012 0.000 -0.014 0.989 0.885 1.130
FixedIncomesource -0.037 0.050 -0.025 -0.744 0.457 0.829 1.206
OffFarmIncW1000 0.000 0.000 0.052 1.368 0.172 0.632 1.582
(PhP)
LandOwner 0.128 0.033 0.133 3.866 0.000 0.767 1.304
TotalareaW (ha) -0.005 0.009 -0.029 -0.552 0.581 0.334 2.990
LPareaW (ha) 0.024 0.035 0.048 0.677 0.498 0.180 5.567
SoldKgW10000 (kg) -0.052 0.095 -0.042 -0.547 0.584 0.152 6.566
DMarketW(km) -0.003 0.004 -0.025 -0.670 0.503 0.668 1.498
DMillerW (km) -0.009 0.002 -0.120 -3.543 0.000 0.799 1.252
Irrigation -0.041 0.021 -0.065 -1.918 0.055 0.801 1.249
HarvestFrequency 0.121 0.056 0.068 2.142 0.032 0.897 1.115
NumofOrgMem 0.028 0.020 0.049 1.373 0.170 0.718 1.393
Training 0.037 0.013 0.099 2.882 0.004 0.767 1.304
IndigenousOrigins 0.254 0.063 0.134 4.043 0.000 0.831 1.203
FarmTraditions 0.053 0.035 0.048 1.529 0.127 0.918 1.090
FastCash -0.411 0.058 -0.232 -7.115 0.000 0.855 1.170
PickedUp -0.207 0.045 -0.152 -4.577 0.000 0.822 1.217
MoneyMgt 0.071 0.035 0.067 2.033 0.042 0.845 1.184
MostImportant -0.060 0.025 -0.079 -2.386 0.017 0.842 1.188
PriceMotivation -0.043 0.031 -0.045 -1.405 0.160 0.870 1.149
DrySeason 0.000 0.031 0.000 -0.013 0.990 0.974 1.027
TownIncClass 0.012 0.007 0.057 1.701 0.089 0.815 1.227
Riceprodvolume -0.014 0.007 -0.065 -1.977 0.048 0.843 1.186
100000 (tons/s)
a. Dependent Variable: outlet 2 categories

245
Appendices

Appendix 3. The full STATA results of the random-effects logistic regression.

Random-effects logistic regression Number of obs = 900

Group variable: Farmercode Number of groups = 300

Random effects u_i ~ Gaussian Obs per group:

min = 3
avg = 3.0
max = 3

Integration method: ghermite Integration pts. = 12

Wald chi2(24) = 63.14

Log likelihood = -264.44158 Prob > chi2 = 0.0000

outlet2cat Coef. Std. Err. z P>|z| [95% Conf. Interval]

Age -.1052543 .052279 -2.01 0.044 -.2077192 -.0027894

Educ -.3396555 .1320068 -2.57 0.010 -.598384 -.0809269
HHchildren -.034303 .2838506 -0.12 0.904 -.59064 .5220339
FixedIncome -3.677861 1.241035 -2.96 0.003 -6.110245 -1.245477
OffFarmIncW .00001 4.25e-06 2.36 0.018 1.70e-06 .0000184
Landowner 3.471063 .8354375 4.15 0.000 1.833636 5.108491
LPareaW -.9004685 .4312026 -2.09 0.037 -1.74561 -.055327
Yield -.3015224 .2239565 -1.35 0.178 -.7404691 .1374243
DMillerW -.4087542 .0722119 -5.66 0.000 -.550287 -.2672214
DMarketW -.0017133 .0884039 -0.02 0.985 -.1749817 .1715551
Irrigation -.9722928 .5008291 -1.94 0.052 -1.9539 .0093142
HarvestFrequency 3.756527 1.473297 2.55 0.011 .8689173 6.644136
NumofOrgMem .4439405 .3833045 1.16 0.247 -.3073226 1.195204
Training .6778058 .3407181 1.99 0.047 .0100106 1.345601
IndigenousOrigins 10.01009 1.899222 5.27 0.000 6.287682 13.7325
FarmTraditions 3.807688 .8595323 4.43 0.000 2.123035 5.49234
Fastcash -6.691572 1.332878 -5.02 0.000 -9.303965 -4.079179
PickedUp -3.101601 .8901026 -3.48 0.000 -4.84617 -1.357032
MoneyMgt .9367603 .7420392 1.26 0.207 -.5176099 2.39113
MostImportant -1.432512 .6525364 -2.20 0.028 -2.71146 -.1535645
PriceMotivation .0713176 .5825721 0.12 0.903 -1.070503 1.213138
DrySeason .0923937 .453283 0.20 0.838 -.7960247 .9808121
TownIncClass .5120123 .2202904 2.32 0.020 .0802511 .9437735
RiceProdVolW -4.58e-06 1.72e-06 -2.65 0.008 -7.95e-06 -1.20e-06
_cons 4.122108 5.330763 0.77 0.439 -6.325995 14.57021

/lnsig2u 4.731318 .2664254 4.209134 5.253502

sigma_u 10.65105 1.418856 8.203549 13.82877

rho .9718176 .0072969 .9533934 .9830877

LR test of rho=0: chibar2(01) = 427.15 Prob >= chibar2 = 0.000

246
Appendices

Appendix 4. The average marginal effects of the independent variables.

. margins, dydx (Age Educ HHchildren FixedIncome OffFarmIncW Landowner LPareaW Yield
> DMillerW DMarketW Irrigation HarvestFrequency NumofOrgMem Training IndigenousOrig
> ins FarmTraditions Fastcash PickedUp MoneyMgt MostImportant PriceMotivation DrySea
> son TownIncClass RiceProdVolW)

Average marginal effects Number of obs = 900

Model VCE : OIM

Expression : Pr(outlet2cat=1), predict(pr)

dy/dx w.r.t. : Age Educ HHchildren FixedIncome OffFarmIncW Landowner LPareaW Yield
DMillerW DMarketW Irrigation HarvestFrequency NumofOrgMem Training
IndigenousOrigins FarmTraditions Fastcash PickedUp MoneyMgt
MostImportant PriceMotivation DrySeason TownIncClass RiceProdVolW

Delta-method
dy/dx Std. Err. z P>|z| [95% Conf. Interval]

Age -.0027192 .0013176 -2.06 0.039 -.0053016 -.0001367

Educ -.0087747 .0032412 -2.71 0.007 -.0151274 -.0024221
HHchildren -.0008862 .007331 -0.12 0.904 -.0152548 .0134824
FixedIncome -.0950145 .0290833 -3.27 0.001 -.1520167 -.0380124
OffFarmIncW 2.59e-07 1.07e-07 2.41 0.016 4.86e-08 4.69e-07
Landowner .0896721 .0178743 5.02 0.000 .0546392 .124705
LPareaW -.0232629 .0106199 -2.19 0.028 -.0440775 -.0024482
Yield -.0077896 .0057223 -1.36 0.173 -.019005 .0034258
DMillerW -.0105598 .0012441 -8.49 0.000 -.0129983 -.0081214
DMarketW -.0000443 .0022841 -0.02 0.985 -.004521 .0044324
Irrigation -.0251184 .0125851 -2.00 0.046 -.0497847 -.000452
HarvestFrequency .0970468 .0360368 2.69 0.007 .0264161 .1676775
NumofOrgMem .0114688 .0098144 1.17 0.243 -.007767 .0307047
Training .0175106 .0084572 2.07 0.038 .0009347 .0340864
IndigenousOrigins .2586025 .0355797 7.27 0.000 .1888676 .3283373
FarmTraditions .0983685 .0177361 5.55 0.000 .0636064 .1331306
Fastcash -.1728713 .0282398 -6.12 0.000 -.2282203 -.1175222
PickedUp -.0801273 .0203687 -3.93 0.000 -.1200493 -.0402054
MoneyMgt .0242004 .0188329 1.29 0.199 -.0127114 .0611122
MostImportant -.0370078 .0162689 -2.27 0.023 -.0688942 -.0051214
PriceMotivation .0018424 .0150302 0.12 0.902 -.0276161 .031301
DrySeason .0023869 .0117018 0.20 0.838 -.0205482 .025322
TownIncClass .0132274 .0053983 2.45 0.014 .002647 .0238078
RiceProdVolW -1.18e-07 4.13e-08 -2.86 0.004 -1.99e-07 -3.73e-08

247
Appendices

Appendix 5. Frequency tables of discreet and dummy variables.

Marketing channels
Frequency Per cent Valid Per cent Cumulative Per cent
Valid Agent 222 24.7 24.7 24.7
Assembler 94 10.4 10.4 35.1
Buying station 271 30.1 30.1 65.2

Creditor 5 .6 .6 65.8

Cooperative 16 1.8 1.8 67.6

Rice mill 236 26.2 26.2 93.8

Retailer/consumer 53 5.9 5.9 99.7
Others 3 .3 .3 100.0
Total 900 100.0 100.0

Age
Frequency Per cent Valid Per cent Cumulative Per cent
Valid Below 30 6 .7 .7 .7

30-39 48 5.3 5.3 6.0

40-49 212 23.6 23.6 29.6

50-59 298 33.1 33.1 62.7

60-69 248 27.6 27.6 90.2

70-79 75 8.3 8.3 98.6

80 and above 13 1.4 1.4 100.0

Total 900 100.0 100.0

Distance to trader-miller (kilometres)

Frequency Per cent Valid Per cent Cumulative Per cent
Valid close to zero 15 1.7 1.7 1.7

1-5 522 58.0 58.0 59.7

6-10 207 23.0 23.0 82.7

11-15 93 10.3 10.3 93.0

16-20 18 2.0 2.0 95.0

21-25 15 1.7 1.7 96.7

26-30 18 2.0 2.0 98.7

Above 30 12 1.3 1.3 100.0

Total 900 100.0 100.0

248
Appendices

Distance to market (kilometres)

Frequency Per cent Valid Per cent Cumulative Per cent
Valid close to zero 54 6.0 6.0 6.0

1-5 531 59.0 59.0 65.0

6-10 228 25.3 25.3 90.3

11-15 48 5.3 5.3 95.7

16-20 24 2.7 2.7 98.3

Above 20 15 1.7 1.7 100.0

Total 900 100.0 100.0

Educational level
Frequency Per cent Valid Per cent Cumulative Per cent
Valid 0 3 .3 .3 .3
Primary level 81 9.0 9.0 9.3

Primary graduate 228 25.3 25.3 34.7

Secondary level 96 10.7 10.7 45.3

Secondary graduate 216 24.0 24.0 69.3

Tertiary level 183 20.3 20.3 89.7

Tertiary graduate 66 7.3 7.3 97.0

Postgraduate 27 3.0 3.0 100.0

Total 900 100.0 100.0

Farm traditions
Frequency Per cent Valid Per cent Cumulative Per cent
Valid None 678 75.3 75.3 75.3

Yes 222 24.7 24.7 100.0

Total 900 100.0 100.0

Fast cash
Frequency Per cent Valid Per cent Cumulative Per cent
Valid Credit/Partial/Later 70 7.8 7.8 7.8

FullCash 830 92.2 92.2 100.0

Total 900 100.0 100.0

249
Appendices

Fixed income source

Frequency Per cent Valid Per cent Cumulative Per cent

Valid No other income source 180 20.0 20.0 20.0

With irregular sources of 621 69.0 69.0 89.0

income
With Regular Job 99 11.0 11..0 100.0

Total 900 100.0 100.0

Harvest frequency
Frequency Per cent Valid Per cent Cumulative Per cent
Valid 1.5 24 2.7 2.7 2.7

2.0 777 86.3 86.3 89.0

2.5 48 5.3 5.3 94.3

3.0 51 5.7 5.7 100.0

Total 900 100.0 100.0

Hierarchy of work motivation

Frequency Per cent Valid Per cent Cumulative Per cent
Valid Basic Needs 402 44.7 44.7 44.7

Psychological Needs 429 47.7 47.7 92.3

Self-fulfilment Needs 69 7.7 7.7 100.0

Total 900 100.0 100.0

Irrigated
Frequency Per cent Valid Per cent Cumulative Per cent
Valid Non-irrigated 360 40.0 40.0 40.0
Irrigated 540 60.0 60.0 100.0
Total 900 100.0 100.0

250
Appendices

Land ownership

Frequency Per cent Valid Per cent Cumulative Per cent

Valid Renting 370 41.1 41.1 41.1

With non-rented land 530 58.9 58.9 100.0

Total 900 100.0 100.0

Largest farm parcel (hectares)

Frequency Per cent Valid Per cent Cumulative Per cent
Valid Below 1 311 34.6 34.6 34.6
1 to 1.99 373 41.4 41.4 76.0
2 to 2.99 120 13.3 13.3 89.3
3 to 3.99 44 4.9 4.9 94.2
4 to 4.99 16 1.8 1.8 96.0
5 to 5.99 6 .7 .7 96.7
6 to 6.99 9 1.0 1.0 97.7
8 to 8.99 7 .8 .8 98.4
9 to 9.99 5 .6 .6 99.0
10 and above 6 .7 .7 99.7
12 3 .3 .3 100.0
Total 900 100.0 100.0

Minority membership

Frequency Per cent Valid Per cent Cumulative Per cent

Valid None 840 93.3 93.3 93.3

Yes 60 6.7 6.7 100.0

Total 900 100.0 100.0

Money management
Frequency Per cent Valid Per cent Cumulative Per cent
Valid Not business-minded/Would not invest 651 72.3 72.3 72.3

Business-minded/Would invest 249 27.7 27.7 100.0

Total 900 100.0 100.0

251
Appendices

Number of organisation membership

Frequency Per cent Valid Per cent Cumulative Per cent
Valid 0 264 29.3 29.3 29.3
1 456 50.7 50.7 80.0
2 138 15.3 15.3 95.3
3 33 3.7 3.7 99.0
4 6 .7 .7 99.7
5 3 .3 .3 100.0
Total 900 100.0 100.0

Wants Pick Up_D

Frequency Per cent Valid Per cent Cumulative Per cent
Valid Delivered 127 14.1 14.1 14.1
Prefers Pick Up 773 85.9 85.9 100.0
Total 900 100.0 100.0

Price motivation
Frequency Per cent Valid Per cent Cumulative Per cent
Valid No 480 53.3 53.3 53.3

Yes 420 46.7 46.7 100.0

Total 900 100.0 100.0

Rice production volume (tons/s)

Frequency Per cent Valid Per cent Cumulative Per cent
Valid Up to 150000 479 53.2 53.2 53.2
150001 to 300000 229 25.4 25.4 78.7

300001 to 450000 83 9.2 9.2 87.9

450001 to 600000 20 2.2 2.2 90.1
Above 600000 89 9.9 9.9 100.0
Total 900 100.0 100.0

252
Appendices

Town income class

Frequency Per cent Valid Per cent Cumulative Per cent
Valid 5th class municipality 60 6.7 6.7 6.7
4th class municipality 144 16.0 16.0 22.7

3rd class municipality 168 18.7 18.7 41.3

2nd class municipality 162 18.0 18.0 59.3

1st class municipality 258 28.7 28.7 88.0

6th class city 6 .7 .7 88.7

5th class city 15 1.7 1.7 90.3

4th class city 9 1.0 1.0 91.3

3rd class city 27 3.0 3.0 94.3

2nd class city 27 3.0 3.0 97.3

1st class city 24 2.7 2.7 100.0

Total 900 100.0 100.0

Training
Frequency Per cent Valid Per cent Cumulative Per cent
Valid None 84 9.3 9.3 9.3
1-5 354 39.3 39.3 48.7

6-10 201 22.3 22.3 71.0

11-15 102 11.3 11.3 82.3

More than 15 159 17.7 17.7 100.0

Total 900 100.0 100.0

Delivery to each marketing channel

Frequency Per cent Valid Per cent Cumulative Per cent
Delivered Agent 9 1.0 7.1 7.1
Buying station 44 4.9 34.6 41.7

Cooperative 3 .3 2.4 44.1

Rice mill 51 5.7 40.2 84.3

Retailer/consumer 18 2.0 14.2 98.4

Others 2 .2 1.6 100.0

Total 127 14.1 100.0

Picked Up 773 85.9

Total 900 100.0

253
Appendices

Appendix 6. Selected correlations between certain variables.

Correlation between training, own capital, and total off-farm household income
OwnCapital_D THHIncomeW FixedIncome_D
**
OwnCapital_D Pearson Correlation 1 .187 .120**
Sig. (2-tailed) .000 .000
N 900 900 900
THHIncomeW (PhP) Pearson Correlation .187** 1 .296**
Sig. (2-tailed) .000 .000
N 900 900 900
FixedIncome_D Pearson Correlation .120** .296** 1
Sig. (2-tailed) .000 .000
N 900 900 900
**. Correlation is significant at the 0.01 level (2-tailed).

Correlation between land ownership and the number of times the farmers sold to their suki
Landowner_D Timessold
Landownership Pearson Correlation 1 .079*
Sig. (2-tailed) .045
N 900 642
*
Timessold Pearson Correlation .079 1
Sig. (2-tailed) .045
N 642 642

Correlation between landownership and having own capital

Landowner OwnCapital_D
Landownership Pearson Correlation 1 .086**
Sig. (2-tailed) .010
N 900 900
OwnCapital_D Pearson Correlation .086** 1
Sig. (2-tailed) .010
N 900 900
**. Correlation is significant at the 0.01 level (2-tailed).

254
Appendices

Correlation between the number of harvests per year and having own capital
HarvestFrequency OwnCapital_D
HarvestFrequency Pearson Correlation 1 .072*
Sig. (2-tailed) .030
N 900 900
OwnCapital_D Pearson Correlation .072* 1
Sig. (2-tailed) .030
N 900 900
*. Correlation is significant at the 0.05 level (2-tailed).

Correlation between the number of harvests per year and having a suki.
HarvestFrequency WithSuki
HarvestFrequency Pearson Correlation 1 .069*
Sig. (2-tailed) .038

N 900 900

WithSuki Pearson Correlation .069* 1

Sig. (2-tailed) .038

N 900 900

*. Correlation is significant at the 0.05 level (2-tailed).

Correlation between the distance to the trader-miller (km) and having a good road condition.
DMillerW (km) GoodRoad_D
DMillerW (km) Pearson Correlation 1 -.128**
Sig. (2-tailed) .000

N 900 900
**
GoodRoad_D Pearson Correlation -.128 1
Sig. (2-tailed) .000

N 900 900

*. Correlation is significant at the 0.01 level (2-tailed).

255
Appendices

Appendix 7. Number of farmers with suki, classified based on harvest frequency.

HarvestFrequency Frequency Per cent Valid Per cent Cumulative Per cent
1.5 Valid None 15 62.5 62.5 62.5
Yes 9 37.5 37.5 100.0

Total 24 100.0 100.0

2.0 Valid None 219 28.2 28.2 28.2

Yes 558 71.8 71.8 100.0

Total 777 100.0 100.0

2.5 Valid None 12 25.0 25.0 25.0

Yes 36 75.0 75.0 100.0
Total 48 100.0 100.0

3.0 Valid None 12 23.5 23.5 23.5

Yes 39 76.5 76.5 100.0

Total 51 100.0 100.0

Appendix 8. Movement of farmers with and without suki within marketing levels.

Movement between Seasons 1 and 2 (Dry to wet season)

With Suki Frequency Per cent Valid Per cent Cumulative Per cent
None Valid Added 2 market players 3 3.5 3.5 3.5
Added 1 market player 4 4.7 4.7 8.1
No movement 75 87.2 87.2 95.3
Cut the chain by 1 4 4.7 4.7 100.0
Total 86 100.0 100.0
Yes Valid Added 1 market player 6 2.8 2.8 2.8
No movement 202 94.4 94.4 97.2
Cut the chain by 1 4 1.9 1.9 99.1
Cut the chain by 2 2 .9 .9 100.0
Total 214 100.0 100.0

256
Appendices

Movement between Seasons 2 and 3 (Wet to dry season)

With Suki Frequency Per cent Valid Per cent Cumulative Per cent
None Valid Added 2 market players 2 2.3 2.3 2.3
Added 1 market player 6 7.0 7.0 9.3
No movement 73 84.9 84.9 94.2
Cut the chain by 1 5 5.8 5.8 100.0
Total 86 100.0 100.0
Yes Valid Added 2 market players 3 1.4 1.4 1.4
Added 1 market player 5 2.3 2.3 3.7
No movement 197 92.1 92.1 95.8
Cut the chain by 1 9 4.2 4.2 100.0
Total 214 100.0 100.0

Movement between Seasons 1 and 3 (Between two dry seasons)

With Suki Frequency Per cent Valid Per cent Cumulative Per cent
None Valid Added 2 market players 4 4.7 4.7 4.7
Added 1 market player 7 8.1 8.1 12.8
No movement 71 82.6 82.6 95.3
Cut the chain by 1 4 4.7 4.7 100.0
Total 86 100.0 100.0
Yes Valid Added 2 market players 1 .5 .5 .5
Added 1 market player 5 2.3 2.3 2.8
No movement 201 93.9 93.9 96.7
Cut the chain by 1 7 3.3 3.3 100.0
Total 214 100.0 100.0

257
References

References
Acharya, S. S. (2003). Agricultural marketing in Asia and the Pacific: Issues and priorities.

Bangkok (Thailand): Food and Agriculture Organization.

Adepoju, A. O., Owoeye, I. T., & Adeoye, I. B. (2015). Determinants of market participation

among pineapple farmers in Aiyedaade Local Government Area, Osun State, Nigeria.

International Journal of Fruit Science, 15(4), 392–404.

https://doi.org/10.1080/15538362.2015.1015763

Aguilar, F. V. (2005). Rice in the Filipino Diet and Culture. Makati City: Philippine Institute

for Development Studies (PIDS). Retrieved from http://hdl.handle.net/10419/127896

Alene, A. D., Manyong, V. M., Omanya, G., Mignouna, H. D., Bokanga, M., & Odhiambo,

G. (2008). Smallholder market participation under transactions costs: Maize supply and

fertilizer demand in Kenya. Food Policy, 33(4), 318–328.

Alexander, K. S., Greenhalgh, G., Moglia, M., Thephavanh, M., Sinavong, P., Larson, S., …

Case, P. (2020). What is technology adoption? Exploring the agricultural research value

chain for smallholder farmers in Lao PDR. Agriculture and Human Values, 37(1), 17–

32. https://doi.org/10.1007/s10460-019-09957-8

Allison, P. D. (2009). Fixed effects regression models. Los Angeles: SAGE.

Allison, P. D. (2012). Logistic regression using SAS: Theory and application. Cary, N.C.:

SAS Institute.

Alvero, J. M. (2008). A value chain study of rice in Abuyog, Leyte. (Unpublished

undergraduate thesis). Ateneo de Manila University, Manila, Philippines.

258
References

Antonio, H., & Acpal, C. (2016). Brown rice. Question & Answers. Philippines: Nueva Ecija:

Nueva Ecija: PhilRice.

Arrow, K. J. (2000). Observations on social capital. In P. Dasgupta & I. Serageldin (Eds.),

Social capital: A multifaceted perspective (pp. 3–5). Washington D.C. : The World

Bank.

Arrow, K. J. (2012). Social choice and individual values. Yale University Press.

Azam, M. S., Imai, K. S., & Gaiha, R. (2012). Agricultural supply response and smallholders

market participation: The Case of Cambodia. The School of Economics Discussion

Paper Series. The University of Manchester.

Baloyi, J. K. (2010). An analysis of constraints facing smallholder farmers in the agribusiness

value chain: A case study of farmers in the Limpopo Province. University of Pretoria.

Retrieved from https://repository.up.ac.za/handle/2263/29038

Bardhan, K. (1970). Price and output response of marketed surplus of foodgrains: A cross-

sectional study of some North Indian villages. American Journal of Agricultural

Economics, 52(1), 51–61.

Barrett, C. (2008). Smallholder market participation: Concepts and evidence from eastern and

southern Africa. Food Policy, 33(4), 299–317.

https://doi.org/https://doi.org/10.1016/j.foodpol.2007.10.005

Barrett, Christopher, Clark, M., Clay, D., & Reardon, T. (2005). Heterogeneous constraints,

incentives and income diversification strategies in rural Africa. Quarterly Journal of

International Agriculture, 44(1), 37–60.

259
References

Benfica, R., Boughton, D., Mouzinho, B., & Uaiene, R. (2014). Food crop marketing and

agricultural productivity in a high price environment: Evidence and implications for

Mozambique (Research Paper No. 76E). Directorate of Economics, Ministry of

Agriculture.

Benfica, R., & Tschirley, D. (2012). The dynamics of agricultural market participation and

performance in Central and Northern Mozambique: Evidence from a panel survey.

Directorate of Economics, Ministry of Agriculture.

Bhattarai, S., Lyne, M. C., & Martin, S. K. (2013). Assessing the performance of a supply

chain for organic vegetables from a smallholder perspective. Journal of Agribusiness in

Developing and Emerging Economies, 3(2), 101–118. Retrieved from

http://ezproxy.auckland.ac.nz/login?url=https://www.proquest.com/scholarly-

journals/assessing-performance-supply-chain-organic/docview/1449425116/se-

2?accountid=8424

Bordey, F. H., Moya, P., Beltran, J., & Dawe, D. C. (Eds.). (2016). Competitiveness of

Philippine rice in Asia. Science City of Muñoz, Nueva Ecija: Science City of Muñoz:

Philippine Rice Research Institute. Retrieved from

https://books.google.co.nz/books?id=KYX5nQAACAAJ

Boughton, D., Mather, D., Barrett, C., Benfica, R., Abdula, D., Tschirley, D., & Cunguara, B.

(2007). Market participation by rural households in a low-income country: An asset-

based approach applied to Mozambique. Faith and Economics, 50(1), 64–101.

Bourdieu, P. (1986). The forms of capital. Cultural Theory: An Anthology, 1, 81–93.

260
References

Bourdieu, P. (1990). Artistic taste and cultural capital. Culture and Society: Contemporary

Debates, 205–215.

Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research

in Psychology, 3(2), 77–101. Retrieved from https://www-tandfonline-

com.ezproxy.auckland.ac.nz/doi/abs/10.1191/1478088706qp063oa

Burr, V. (1995). An introduction to social constructionism. London ; New York: New York:

Routledge.

Cadot, O., Dutoit, L., & Olarreaga, M. (2005). How Costly is it for Poor Farmers to Lift

Themselves out of Subsistence? IDEAS Working Paper Series from RePEc.

Carr, E. C. J., & Worth, A. (2001). The use of the telephone interview for research. NT

Research, 6(1), 511–524.

Cazzuffi, C., & McKay, A. (2012). Rice market participation and channels of sale in rural

Vietnam. In International Association of Agricultural Economists (IAAE) Triennial

Conference. Foz do Iguaçu, Brazil.

CGIAR. (2017). Ricepedia. Retrieved July 22, 2017, from http://ricepedia.org/rice-as-a-crop

Chitra, K., & Sreedevi, V. R. (2011). Does personality traits influence the choice of

investment? IUP Journal of Behavioral Finance, 8(2), 47–57.

Clarke, V., & Braun, V. (2013). Teaching thematic analysis. Psychologist, 26(2), 120–123.

Cororaton, C. B. (2014). Rice reforms and poverty in the Philippines: A CGE analysis. ADB.

261
References

Corpuz, P. (2019). Rice tariffication law enacted. Manila, Philippines. Retrieved from

https://gain.fas.usda.gov/Recent GAIN Publications/Rice Tariffication Law

Enacted_Manila_Philippines_2-20-2019.pdf

Crawford, I. M. (1997). Agricultural and food marketing management. Rome: Rome: Food

and Agriculture Organization of the United Nations.

Cuevas, A. C., & Clarete, R. L. (2015). Effect of transaction costs on farmers’ market

participation in the Philippine rice sector. Philippine Journal of Crop Science, 40(1),

35–44.

DA. (2017). Special area for agricultural development. Retrieved July 21, 2017, from

http://www.dasaad.net/about-us/da-saad-briefer/

DA, NEDA, & DBM. (2019). The Implementing Rules and Regulations of Republic Act No.

11203, “An Act Liberalizing the Importation, Exportation, and Trading of Rice, Lifting

for the Purpose the Quantitative Import Restriction on Rice, and for Other Purposes.”

Philippines: http://www.neda.gov.ph/wp-content/uploads/2019/04/IRR-of-RA-No.-

11203-or-the-Rice-Liberalization-Act-RLA_signed.pdf. Retrieved from

http://www.neda.gov.ph/wp-content/uploads/2019/04/IRR-of-RA-No.-11203-or-the-

Rice-Liberalization-Act-RLA_signed.pdf

Daft, R. L. (2008). New era of management (2nd Ed.). Mason, Ohio: USA: Thomson: South-

Western. Retrieved from

https://books.google.co.nz/books?id=RhwMpK19nj0C&printsec=frontcover#v=onepage

&q&f=false

262
References

De Janvry, A., Fafchamps, M., & Sadoulet, E. (1991). Peasant household behaviour with

missing markets: Some paradoxes explained. The Economic Journal, 101(409), 1400–

1417.

Dessie, A. B., Abate, T. M., & Mekie, T. M. (2018). Factors affecting market outlet choice of

wheat producers in North Gondar Zone, Ethiopia. Agriculture and Food Security, 7(1).

https://doi.org/10.1186/s40066-018-0241-x

Dessie, A. B., Abate, T. M., & Mekie, T. M. (2019). Determinants of market outlet choices

by smallholder teff farmers in Dera district, South Gondar Zone, Amhara National

Regional State, Ethiopia: a multivariate probit approach. Journal of Economic

Structures, 8(1). https://doi.org/10.1186/s40008-019-0167-x

Dixon, W. J. (1960). Simplified estimation from censored normal samples. The Annals of

Mathematical Statistics, 31(2), 385–391.

Dlamini-Mazibuko, B. P., Ferrer, S., & Ortmann, G. (2019). Factors affecting the choice of

marketing outlet selection strategies by smallholder farmers in Swaziland. African

Journal of Science, Technology, Innovation and Development, 11(5), 569–577.

https://doi.org/10.1080/20421338.2018.1554323

Dolan, R. E. (1991). Philippines: A Country Study. Washington: GPO for the Library of

Congress. Retrieved from http://countrystudies.us/philippines/

DOLE. (2019). Summary of current regional daily minimum wage rates by region, non-

agriculture and agriculture. Retrieved June 9, 2020, from

https://nwpc.dole.gov.ph/stats/summary-of-current-regional-daily-minimum-wage-rates-

by-region-non-agriculture-and-agriculture/

263
References

Downward, P., & Mearman, A. (2007). Retroduction as mixed-methods triangulation in

economic research: Reorienting economics into social science. Cambridge Journal of

Economics, 31(1), 77–99. Retrieved from www.jstor.org/stable/23601840

EFSIM. (2009). Empowering smallholder farmers in markets: Philippine country paper.

EFSIM. Retrieved from http://www.esfim.org/wp-

content/uploads/Philippines_esfim_paper1_revjan29.doc

Emmerson, D. K. (1982). Rethinking artisanal fisheries development: Western concepts,

Asian experiences. American Anthropologist, 84(3), 696–697.

FAO. (n.d.). Staple foods: What do people eat? Retrieved May 6, 2020, from

http://www.fao.org/3/u8480e/u8480e07.htm

FAO. (2012). FAO Country Programming Framework 2012-2018: Philippines.

FAO. (2015). FAO statistical pocketbook 2015: World food and agriculture. Rome: Rome:

Food and Agriculture Organization of the United Nations.

FAO. (2017). Defining small scale food producers to monitor target 2.3. of the 2030 agenda

for sustainable development (Working Paper Series No. ESS / 17-12). Rome, Italy.

Ferris, S., Robbins, P., Best, R., Seville, D., Buxton, A., Shriver, J., & Wei, E. (2014).

Linking smallholder farmers to markets and the implications for extension and advisory

services. MEAS Brief, 4(10), 13–14.

Francis, J. J., Johnston, M., Robertson, C., Glidewell, L., Entwistle, V., Eccles, M. P., &

Grimshaw, J. M. (2010). What is an adequate sample size? Operationalising data

saturation for theory-based interview studies. Psychology & Health, 25(10), 1229–1245.

264
References

https://doi.org/10.1080/08870440903194015

Gabriel, A. H., & Hundie, B. (2006). Farmers’ post-harvest grain management choices under

liquidity constraints and impending risks: Implications for achieving food security

objectives in Ethiopia. Gold Coast, Australia: AgEcon Search. Retrieved from Paper

presented at the meeting of International Association of Agricultural Economists

Conference, Gold Coast, Australia

Gajjar, N. B. (2013). Factors affecting consumer behavior. International Journal of Research

in Humanities and Social Sciences, 1(2), 10–15.

Gani, B. S., & Adeoti, A. I. (2011). Analysis of market participation and rural poverty among

farmers in northern part of Taraba State, Nigeria. Journal of Economics, 2(1), 23–36.

Gelaw, F., Speelman, S., & Van Huylenbroeck, G. (2016). Farmers’ marketing preferences in

local coffee markets: Evidence from a choice experiment in Ethiopia. Food Policy, 61,

92–102.

Gibbs, G. R. (2008). Thematic coding and categorizing. In Analyzing qualitative data.

London: Sage Publications Ltd.

Goetz, S. J. (1992). A selectivity model of household food marketing behavior in sub-

Saharan Africa. American Journal of Agricultural Economics, 74(2), 444–452.

Granovetter, M. (1985). Economic action and social structure: The problem of

embeddedness. American Journal of Sociology, 91(3), 481–510.

Guest, G., Bunce, A., & Johnson, L. (2006). How many interviews are enough? An

experiment with data saturation and variability. Field Methods, 18(1), 59–82.

265
References

Gummert, M., & Rickman, J. (2011). Paddy drying systems. Retrieved February 28, 2020,

from http://www.knowledgebank.irri.org/training/fact-sheets/postharvest-

management/drying-fact-sheet-category/paddy-drying-systems-fact-sheet

Haessel, W. (1975). The price and income elasticities of home consumption and marketed

surplus of foodgrains. American Journal of Agricultural Economics, 57(1), 111–115.

Heltberg, R., & Tarp, F. (2002). Agricultural supply response and poverty in Mozambique.

Food Policy, 27(2), 103–124.

Hendriks, M. B. (1997). Vegetable Trade in the Central Philippines. An anthropological

study of exchange and credit in a modernizing economy. (PhD thesis, University of

Amsterdam). Retrieved from UvA-DARE (Digital Academic Repository)

Hilbe, J. M. (2009). Logistic regression models (1st ed.). Boca Raton: CRC Press.

Holden, S., Shiferaw, B., & Pender, J. (2004). Non-farm income, household welfare, and

sustainable land management in a less-favoured area in the Ethiopian highlands. Food

Policy, 29(4), 369–392. https://doi.org/10.1016/j.foodpol.2004.07.007

IRRI. (n.d.). Moisture content calculations. Retrieved May 26, 2020, from

http://www.knowledgebank.irri.org/step-by-step-

production/postharvest/harvesting/measuring-moisture-content/moisture-content-

calculations

Isacson, D., & Bingefors, K. (2002). Attitudes towards drugs - a survey in the general

population. Pharmacy World & Science : PWS, 24(3), 104–110. Retrieved from

http://search.proquest.com/docview/71931649/

266
References

Javier, E. Q. (2017). Rice Trade Liberalization. Retrieved July 20, 2017, from

http://www.philrice.gov.ph/rice-trade-liberalization-part-1/

Jegho, L. (2016). Can Indonesia be Self-Sufficient in Rice, Again? Retrieved from

http://www.globalindonesianvoices.com/25204/why-rice-is-expensive-but-farmers-are-

poor-in-indonesia/

Key, N., Sadoulet, E., & Janvry, A. De. (2000). Transactions costs and agricultural household

supply response. American Journal of Agricultural Economics, 82(2), 245–259.

Khanal, N. P., & Maharjan, K. L. (2014). Factors influencing farmers’ behavior in rice seed

selling in the market: A case study in the Tarai region of Nepald. Agricultural and Food

Economics, 2(1), 1–11.

Khanal, N. P., & Maharjan, K. L. (2015). Households’ behavior in selling rice and wheat

seed in the market. In Community seed production sustainability in rice-wheat farming

(pp. 93–105). Tokyo: Springer Japan. https://doi.org/10.1007/978-4-431-55474-5_7

Kherallah, M., Delgado, C., Gare-Madhin, E., Minot, N., & Johnson, M. (2000). The road

half travelled: Agricultural market reforms in Sub-Saharan Africa (No. IFPRI Food

Policy Report 10). Washington: Washington. Retrieved from

https://www.ifpri.org/publication/road-half-traveled-agricultural-market-reform-sub-

saharan-africa

King, A. (2004). Structure and agency. In A. Harrington (Ed.), Modern Social Theory: An

Introduction. Oxford University Press. Retrieved from

https://philpapers.org/rec/KINSAA-4

267
References

Kokic, P. N., & Bell, P. A. (1994). Optimal winsorizing cutoffs for a stratified finite

population estimator. Journal of Official Statistics, 10(4), 419.

Kreps, D. M. (1990). A course in microeconomic theory. New Jersey: Princeton university

press.

Lakner, C., Negre, M., & Prydz, E. B. (2014). Twinning the goals: How can promoting

shared prosperity help to reduce global poverty? The World Bank.

Larkin, P. J., de Casterlé, B., & Schotsmans, P. (2007). Multilingual translation issues in

qualitative research: Reflections on a metaphorical process. Qualitative Health

Research, 17(4), 468–476.

Larson, D., & Deininger, K. (2001). Crop markets and household participation. In R.

Reinikka & P. Collier (Eds.), Uganda’s recovery: The role of farms, firms, and

government (pp. 177–204). World Bank Group. Retrieved from

https://elibrary.worldbank.org/doi/abs/10.1596/0-8213-4664-4#page=185

Löfgren, H., & Robinson, S. (1999). Nonseparable farm household decisions in a computable

general equilibrium model. American Journal of Agricultural Economics, 81(3), 663–

670. https://doi.org/10.2307/1244031

Maguire, M., & Delahunt, B. (2017). Doing a thematic analysis: A practical, step-by-step

guide for learning and teaching scholars. AISHE-J: The All Ireland Journal of Teaching

and Learning in Higher Education, 9(3). Retrieved from

http://ojs.aishe.org/aishe/index.php/aishe-j/article/view/335/553

Marinescu, A. (2016). Axiomatical examination of the neoclassical economic model: Logical

assessment of the assumptions of neoclassical economic model. Theoretical & Applied

268
References

Economics, 23(2), 47–64.

Mataia, A. B., Beltran, J. C., Manalili, R. G., Catudan, B. M., Francisco, N. M., & Flores, A.

C. (2018). Rice Value Chain Analysis in the Philippines. Science City of Muňoz

(Philippines): Science City of Muñoz: Philippine Rice Research Institute.

Mataia, A. B., Beltran, J. C., Manalili, R. G., Catudan, B. M., Francisco, N. M., & Flores, A.

C. (2020). Rice Value Chain Analysis in the Philippines: Value Addition, Constraints,

and Upgrading Strategies. Asian Journal of Agriculture and Development, 17(1362-

2020–1834), 19–42.

Mattos, F., & Zinn, J. (2016). Formation and adaptation of reference prices in grain

marketing: An experimental study. Agricultural Economics, 47(6), 621–632.

Mayers, R. H. (1990). Classical and modern regression with applications (Vol. 2). Belmont,

CA: Duxbury Press.

McElroy, J. C., Hendrickson, A. R., Townsend, A. M., & DeMarie, S. M. (2007).

Dispositional factors in internet use: Personality versus cognitive style. MIS Quarterly,

31(4), 809–820. https://doi.org/10.2307/25148821

McKay, A., Cazzuffi, C., & Perge, E. (2015). Commercialization in agriculture in rural Viet

Nam, 2006-14 (WIDER Working Paper, No. 2015/096). World Institute for

Development Economics Research. Retrieved from

https://www.econstor.eu/handle/10419/126312

McLeod, S. (2007). Maslow’s hierarchy of needs. Simply Psychology, 1, 1–8. Retrieved from

http://www.simplypsychology.org/simplypsychology.org-Maslows-Hierarchy-of-

Needs.pdf

269
References

Mmbando, F. E., Wale, E. Z., & Baiyegunhi, L. J. S. (2015). Determinants of smallholder

farmers’ participation in maize and pigeonpea markets in Tanzania. Agrekon, 54(1), 96–

119. https://doi.org/10.1080/03031853.2014.974630

Montemayor, R. (2020, September 6). Winners and losers from the rice tariffication law.

Retrieved October 1, 2020, from https://newsinfo.inquirer.net/1332019/winners-and-

losers-from-the-rice-tariffication-law

Mottaleb, K. A., Mohanty, S., & Nelson, A. (2014). Strengthening market linkages of farm

households in developing countries. Applied Economic Perspectives and Policy, 37(2),

226–242.

Muthayya, S., Sugimoto, J. D., Montgomery, S., & Maberly, G. F. (2014). An overview of

global rice production, supply, trade, and consumption. Annals of the New York

Academy of Sciences, 1324(1), 7–14. https://doi.org/10.1111/nyas.12540

Nes, F., Abma, T., Jonsson, H., & Deeg, D. (2010). Language differences in qualitative

research: Is meaning lost in translation. European Journal of Ageing, 7(4), 313–316.

NFA. (2011). NFA Annual Report 2011. Quezon City, Philippines: Quezon City: National

Food Authority. Retrieved from http://www.nfa.gov.ph/files/field/pamphlet_en.pdf

NFA. (2017). NFA Procurement Program. Retrieved July 20, 2017, from

http://www.nfa.gov.ph/files/field/pamphlet_en.pdf

Novick, G. (2008). Is there a bias against telephone interviews in qualitative research?

Research in Nursing & Health, 31(4), 391–398.

270
References

Nwokoye, N. G. (1981). Modern marketing for Nigeria. London and Basingstoke: London:

Macmillan Publishers Limited.

Nxumalo, K. K. S., Oduniyi, O. S., Antwi, M. A., & Tekana, S. S. (2019). Determinants of

market channel choice utilised by maize and sunflower farmers in the North West

Province, South Africa. Cogent Social Sciences, 5(1), 1678451.

https://doi.org/10.1080/23311886.2019.1678451

Nyoro, J., Kiiru, M. W., & Jayne, T. S. (1999). Evolution of Kenya’s maize marketing

systems in the post-liberalization era (Food Security Collaborative Working Papers No.

55148). Michigan State University, Department of Agricultural, Food, and Resource

Economics.

OECD. (2017). Agricultural Policies in the Philippines. Paris: Paris: OECD Publishing.

Ojiako, I. A., Tarawali, G., Okechukwu, R. U., & Chianu, J. N. (2017). Household

characteristics and market participation competence of smallholder farmers supplying

cassava to starch processors in Nigeria. International Journal of Agricultural Research,

Innovation and Technology, 6(2), 42–56.

Olga, P., & Monowar, M. (2015). Impact of personality on risk tolerance and investment

decisions: A study on potential investors of Kazakhstan. International Journal of

Commerce and Management, 25(4), 370–384. https://doi.org/10.1108/IJCoMA-01-

2013-0002

Olwande, J., Smale, M., Mathenge, M. K., Place, F., & Mithöfer, D. (2015). Agricultural

marketing by smallholders in Kenya: A comparison of maize, kale and dairy. Food

Policy, 52, 22–32. https://doi.org/http://dx.doi.org/10.1016/j.foodpol.2015.02.002

271
References

Omiti, J., Otieno, D., Nyanamba, T., & McCullough, E. (2009). Factors influencing the

intensity of market participation by smallholder farmers: A case study of rural and peri-

urban areas of Kenya. African Journal of Agricultural and Resource Economics, 3(1),

57–82.

Opdenakker, R. (2006). Advantages and disadvantages of four interview techniques in

qualitative research. Forum Qualitative Sozialforschung/Forum Qualitative Social

Research, 7(4). https://doi.org/http://dx.doi.org/10.17169/fqs-7.4.175

Oseni, G., & Winters, P. (2009). Rural nonfarm activities and agricultural crop production in

Nigeria. Agricultural Economics, 40(2), 189–201. https://doi.org/10.1111/j.1574-

0862.2009.00369.x

Ostrom, E. (1990). Governing the commons: The evolution of institutions for collective

action. Cambridge: Cambridge University Press.

Owach, C., Bahiigwa, G., & Elepu, G. (2017). Farming household food storage, consumption

and sales decision making under price risk in northern Uganda. Journal of Development

and Agricultural Economics, 9(1), 8–15.

Pabuayon, I. M. (2006). Strengthening the Philippine Agricultural Market Information

Service (MIS) at the local level. Metro Manila Commission Professorial Chair Lecture,

College, Laguna (Philippines). Abstract retrieved from https://agris.fao.org/agris-

search/search.do?recordID=PH2007001377. Retrieved from https://agris.fao.org/agris-

search/search.do?recordID=PH2007001377

272
References

Park, A. F. (2006). Risk and household grain management in developing countries. The

Economic Journal, 116(514), 1088–1115. https://doi.org/https://doi.org/10.1111/j.1468-

0297.2006.01124.x

Pe-Pua, R., & Protacio-Marcelino, E. A. (2000). Sikolohiyang Pilipino (Filipino psychology):

A legacy of Virgilio G. Enriquez. Asian Journal of Social Psychology, 3(1), 49–71.

https://doi.org/10.1111/1467-839X.00054

Peduzzi, P., Concato, J., Kemper, E., Holford, T. R., & Feinstein, A. R. (1996). A simulation

study of the number of events per variable in logistic regression analysis. Journal of

Clinical Epidemiology, 49(12), 1373–1379.

Philippine Statistics Authority. (2010). CountrySTAT Philippines. Retrieved May 22, 2020,

from http://countrystat.bas.gov.ph/?cont=factsandfigures

Philippine Statistics Authority. (2015). Marketing costs structure of palay/corn 2013. Quezon

City, Philippines: Philippine Statistics Authority-Bureau of Agricultural Statistics.

Retrieved from https://psa.gov.ph/sites/default/files/marketing_cost_palay2013.pdf

Philippine Statistics Authority. (2016). 2015 Full Year Official Poverty Statistics of the

Philippines. Retrieved May 8, 2018, from https://psa.gov.ph/sites/default/files/2015 Full

Year Official Poverty Statistics of the Philippines Publication_0.pdf

Philippine Statistics Authority. (2017). 2016 Commodity Fact Sheet. Quezon City,

Philippines: Quezon City: Philippine Statistics Authority.

Philippine Statistics Authority. (2018a). Food Availability. Retrieved from

http://philfsis.psa.gov.ph.

273
References

Philippine Statistics Authority. (2018b). Philippine Agriculture In Figures - CountrySTAT

Philippines. Retrieved November 20, 2018, from http://countrystat.psa.gov.ph/?cont=3

Philippines Country Study Guide. (1999) (4th Ed.). Washington, DC: Washington, DC: USA

International Business Publications.

PhilRice. (2003). Philippine Rice Production Training Manual. Science City of Munoz,

Philippines: Philippine Rice Research Institute. Retrieved from

https://www.pinoyrice.com/wp-content/uploads/philippine-rice-production-training-

manual.pdf

PhilRice. (2011). Is marketed surplus of farmers enough to meet their basic needs? (A. B.

Mataia, Ed.). Munoz, Nueva Ecija: Science City of Muñoz: Philippine Rice Research

Institute.

PhilRice. (2015). 2014 National Rice R&D Highlights. Munoz, Nueva Ecija: Science City of

Muñoz: Philippine Rice Research Institute.

PhilRice. (2017). 2016 National Rice R&D Highlights. Munoz, Nueva Ecija: Science City of

Muñoz: Philippine Rice Research Institute.

Poston, B. (2009). Maslow’s hierarchy of needs. Surgical Technologist, 41(8), 347–353.

Rapsomanikis, G. (2015). The economic lives of smallholder farmers: An analysis based on

household data from nine countries. Rome, Italy: Food and Agriculture Organization of

the United Nations.

274
References

Reardon, T., Barrett, C., Berdegué, J. A., & Swinnen, J. F. M. (2009). Agrifood industry

transformation and small farmers in developing countries. World Development, 37(11),

1717–1727.

Reardon, T., Chen, K. Z., Minten, B., Adriano, L., Dao, T. A., Wang, J., & Gupta, S. Das.

(2014). The quiet revolution in Asia’s rice value chains. Annals of the New York

Academy of Sciences, 1331(1), 106–118.

https://doi.org/https://doi.org/10.1111/nyas.12391

Reardon, T., & Timmer, C. P. (2007). Transformation of markets for agricultural output in

developing countries since 1950: How has thinking changed? In R. Evenson & P.

Pingali (Eds.), Handbook of Agricultural Economics (Vol. 3, pp. 2807–2855). Elsevier.

https://doi.org/https://doi.org/10.1016/S1574-0072(06)03055-6

Rios, A. R., Masters, W. A., & Shively, G. E. (2008). Linkages between market participation

and productivity: Results from a multi-country farm household sample. In American

Agricultural Economics Association Annual Meeting, Orlando, Florida, USA (p.

https://ageconsearch.umn.edu/record/6145/files/461).

Rivas, R. (2019, September 2). Farmers seek floor price as palay sells for just P7 per kilo.

Retrieved from https://www.rappler.com/business/239117-farmers-seek-floor-price-

palay

Rungduin, T., Rungduin, D. C., Aninacion, J. G., Catindig, R. B., & JrGallogo, L. S. (2016).

The Filipino character strength of utang na loob: Exploring contextual associations with

gratitude. International Journal of Research Studies in Psychology, 5(1), 13–23.

275
References

Ruppert, D. (2006, August 15). Trimming and Winsorization. Encyclopedia of Statistical

Sciences. https://doi.org/doi:10.1002/0471667196.ess2768.pub2

Sadoulet, E., & De Janvry, A. (1995). Quantitative development policy analysis. Baltimore

and London: Baltimore: Johns Hopkins University Press.

Sandri, S. (2009). On the rationality of the economic actors. In Reflexivity in Economics: An

Experimental Examination on the Self-Referentiality of Economic Theories (pp. 77–

100). Physica-Verlag HD.

Sartwelle, J. D. I., amp, apos, Brien, D. M., Tierney Jr., W. I., & Eggers, T. (2000). The

effect of personal and farm characteristics upon grain marketing practices. Journal of

Agricultural and Applied Economics, 32(1), 95–111.

Schermerhorn, J. R. (2011). Introduction to management (11th Ed). Asia: Iowa: John Wiley

& Sons.

Sen, A. (1993). Markets and freedoms: Achievements and limitations of the market

mechanism in promoting individual freedoms. Oxford Economic Papers, 45(4), 519–

541.

Sharma, K. L., & Gupta, M. P. (1970). Study of farm factors determining marketed surplus of

bajra in Jaipur District. Indian Journal of Agricultural Economics, 25(4), 64–68.

Simon, H. A. (2000). Bounded rationality in social science: Today and tomorrow. Mind &

Society, 1(1), 25–39. https://doi.org/10.1007/BF02512227

Singh, I., Squire, L., & Strauss, J. (1986). Agricultural household models: Extensions,

applications, and policy. Baltimore, MD: The Johns Hopkins University Press.

276
References

Retrieved from

http://documents.worldbank.org/curated/en/621291468739297175/Agricultural-

household-models-extensions-applications-and-policy

Sobel, J. (2002). Can we trust social capital? Journal of Economic Literature, 40(1), 139–

154. https://doi.org/10.1257/0022051027001

Soe, W. (2015). Empirical Studies on Industrial Organization of Paddy Rice in Myanmar.

(Unpublished PhD thesis). Kyushu University, Fukuoka, Japan.

Soe, W., Moritaka, M., & Fukuda, S. (2015). An analysis of the factors influencing marketing

channel choice by paddy rice farmers in Myanmar. Journal of the Faculty of

Agriculture, Kyushu University, 60(2), 535–542.

Solow, R. M. B. T.-H. of M. (1999). Neoclassical growth theory. In Edited by J.B. Taylor

and M. WoodJbrd (Ed.), Handbook of Macroeconomics (Vol. 1, pp. 637–667). Elsevier.

https://doi.org/https://doi.org/10.1016/S1574-0048(99)01012-5

Stávková, J., Stejskal, L., & Toufarová, Z. (2008). Factors influencing consumer behaviour.

Agric. Econ. – Czech, 54(6), 276–284.

https://doi.org/https://www.agriculturejournals.cz/publicFiles/01585.pdf

Stephens, E. C., & Barrett, C. (2011). Incomplete credit markets and commodity marketing

behaviour. Journal of Agricultural Economics, 62(1), 1–24.

https://doi.org/10.1111/j.1477-9552.2010.00274.x

Stern, L., EL–Ansary, A., & Coughlan, A. (1996). Marketing Channels (5th Ed.). New

Jersey: New Jersey: Prentice Hall, Inc.

277
References

Suh, E. E., Kagan, S., & Strumpf, N. (2009). Cultural Competence in Qualitative Interview

Methods with Asian Immigrants. Journal of Transcultural Nursing, 20(2), 194–201.

Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics. Boston: Allyn &

Bacon/Pearson Education.

The World Bank. (2018). Making growth work for the poor : A poverty assessment for the

Philippines. Washington, DC: International Bank for Reconstruction and Development /

The World Bank. Retrieved from

http://documents.worldbank.org/curated/en/273631527594735491/pdf/126194-WP-

REVISED-OUO-9.pdf

Throsby, D. (1999). Cultural Capital. Journal of Cultural Economics, 23(1), 3–12.

https://doi.org/10.1023/A:1007543313370

USTR. (2015). The Philippines. Washington, DC. Retrieved from

https://ustr.gov/sites/default/files/files/reports/2015/NTE/2015 NTE Philippines.pdf

Van Huong, N., Tran Huu, C., & Lebailly, P. (2016). Factors affecting small scale fish

farmers in accessing markets: A case study of fish value chain in HaiDuong Province,

Vietnam. In International Conference on Linkages and Cooperation in Agricultural

Production and Marketing in the Context of International Economic Integration, Hanoi,

Vietnam. Leige, Belgium.

Vergopoulos, K. (1978). Capitalism and peasant productivity. The Journal of Peasant

Studies, 5(4), 446–465.

Vollmer, E., Hermann, D., & Mußhoff, O. (2017). The disposition effect in farmers’ selling

behavior: An experimental investigation (Discussion Papers). Goettingen: Department

278
References

for Agricultural Economics and Rural Development, University of Goettingen.

Retrieved from http://hdl.handle.net/10419/156336

Wickramasinghe, U., & Weinberger, K. (2013). Smallholder market participation and

production specialization: Evolution of thinking, issues and policies. (Working Paper

No. 107) Indonesia: CAPSA-ESCAP.

Woldehanna, T. (2000). Economic analysis and policy implications of farm and off-farm

employment : A case study in the Tigray Region of Northern Ethiopia. (Dissertation,

Wegeningen University and Research Center). Retrieved from

https://library.wur.nl/WebQuery/wurpubs/67022.

Woldeyohanes, T., Heckelei, T., & Surry, Y. (2017). Effect of off-farm income on

smallholder commercialization: Panel evidence from rural households in Ethiopia.

Agricultural Economics, 48(2), 207–218. https://doi.org/10.1111/agec.12327

Wooldridge, J. M. (2012). Introductory econometrics : A modern approach (5th ed.). Mason,

Ohio: South-Western Cengage Learning.

World Rice Production 2017. (2017). Retrieved July 20, 2017, from

https://www.worldriceproduction.com/

WTO. (1999). The Philippines: September 1999. Retrieved from

https://www.wto.org/english/tratop_e/tpr_e/tp114_e.htm

Yang, Z., & Guo, Q. (2011). Positive Analysis of Peasant Households’ Grain Selling

Behavior and Its Influencing Factors. Journal of Jilin Agricultural University, 33(03),

344–348.

279
References

Young, S., Gudjonsson, G. H., Carter, P., Terry, R., & Morris, R. (2012). Simulation of risk-

taking and it relationship with personality. Personality and Individual Differences,

53(3), 294–299. https://doi.org/10.1016/j.paid.2012.03.014

Zafirovski, M. (2000). The rational choice generalization of neoclassical economics

reconsidered: Any theoretical legitimation for economic imperialism? Sociological

Theory, 18(3), 448–471. Retrieved from http://www.jstor.org/stable/223329

280