Adbi wp872

McIntosh, Craig; Mansini, Caio Scuarcialupi
Working Paper
The use of financial technology in the agriculture sector
ADBI Working Paper, No. 872
Provided in Cooperation with:

Asian Development Bank Institute (ADBI), Tokyo
Suggested Citation: McIntosh, Craig; Mansini, Caio Scuarcialupi (2018) : The use of financial
technology in the agriculture sector, ADBI Working Paper, No. 872, Asian Development Bank
Institute (ADBI), Tokyo
This Version is available at:

https://hdl.handle.net/10419/190293
Standard-Nutzungsbedingungen: Terms of use:
Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Documents in EconStor may be saved and copied for your personal
Zwecken und zum Privatgebrauch gespeichert und kopiert werden. and scholarly purposes.
Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle You are not to copy documents for public or commercial purposes, to
Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich exhibit the documents publicly, to make them publicly available on the
machen, vertreiben oder anderweitig nutzen. internet, or to distribute or otherwise use the documents in public.
Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen If the documents have been made available under an Open Content
(insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, Licence (especially Creative Commons Licences), you may exercise
gelten abweichend von diesen Nutzungsbedingungen die in der dort further usage rights as specified in the indicated licence.
genannten Lizenz gewährten Nutzungsrechte.
https://creativecommons.org/licenses/by-nc-nd/3.0/igo/
ADBI Working Paper Series
THE USE OF FINANCIAL TECHNOLOGY

IN THE AGRICULTURE SECTOR
Craig McIntosh and

Caio Scuarcialupi Mansini
No. 872
September 2018
Asian Development Bank Institute

Craig McIntosh is a professor, and Caio Scuarcialupi Mansini is a graduate student, both
at the School of Global Policy and Strategy of the University of California, San Diego.
The views expressed in this paper are the views of the author and do not necessarily reflect
the views or policies of ADBI, ADB, its Board of Directors, or the governments they
represent. ADBI does not guarantee the accuracy of the data included in this paper and
accepts no responsibility for any consequences of their use. Terminology used may not
necessarily be consistent with ADB official terms.
Working papers are subject to formal revision and correction before they are finalized and
considered published.
The Working Paper series is a continuation of the formerly named Discussion Paper series;
the numbering of the papers continued without interruption or change. ADBI’s working papers
reflect initial ideas on a topic and are posted online for discussion. Some working papers may
develop into other forms of publication.
The Asian Development Bank recognizes “China” as the People’s Republic of China.
In this report, “$” refers to United States dollars.
Suggested citation:
McIntosh, C. and C. S. Mansini, 2018. The Use of Financial Technology in the Agriculture
Sector. ADBI Working Paper 872. Tokyo: Asian Development Bank Institute. Available:
https://www.adb.org/publications/use-financial-technology-agriculture-sector
Please contact the authors for information about this paper.
Email: ctmcintosh@ucsd.edu, cmansini@ucsd.edu
Asian Development Bank Institute

Kasumigaseki Building, 8th Floor
3-2-5 Kasumigaseki, Chiyoda-ku
Tokyo 100-6008, Japan
Tel: +81-3-3593-5500
Fax: +81-3-3593-5571
URL: www.adbi.org
E-mail: info@adbi.org
© 2018 Asian Development Bank Institute

ADBI Working Paper 872 McIntosh and Mansini
Abstract
The finance sector has a key role to play in allowing agriculture to contribute to economic
growth and poverty reduction. A rapidly evolving technological landscape is opening up new
possibilities to target and price credit, to share risk, and to harness information technology to
expand agricultural productivity. At the same time, many obstacles are not technological, so it
is important to look for strategic places where policy and investment can help to improve
outcomes for agricultural households. Our analysis first situates agricultural finance in the
Asian context, and then discusses the role of financial technology (FinTech) in driving new
products in credit and risk markets. We examine linkages to mobile money, financial literacy,
national identification systems, and blockchain technology. The paper concludes with a
discussion of policy takeaways for FinTech in agriculture to promote growth, enhance financial
inclusion, and improve regional economic integration.
Keywords: financial technology, agriculture, credit, risk, financial literacy
JEL Classification: Q16, O16, O33

Contents
1. INTRODUCTION ......................................................................................................... 1
2. THE CONTEXT ........................................................................................................... 3
3. CREDIT AND SAVINGS PRODUCTS ...................................................................... 13
3.1 Microfinance .................................................................................................. 14

3.2 Credit Reporting ............................................................................................ 14
3.3 Asset Registries and Warehouse Receipt Systems ...................................... 16
3.4 Savings Products .......................................................................................... 16
4. INSURANCE AND RISK ........................................................................................... 18
4.1 Risk in the Agricultural Finance System ........................................................ 18

4.2 Micro-insurance ............................................................................................. 18
4.3 FinTech and Risk Sharing in the Agricultural System ................................... 20
5. COMPLEMENTARY INFRASTRUCTURE AND THE ENABLING ENVIRONMENT 21
5.1 Mobile Money ................................................................................................ 21

5.2 Financial Literacy and FinTech ..................................................................... 22
5.3 National and Biometric Identification Systems .............................................. 22
5.4 Blockchain Technologies ............................................................................... 24
6. POLICY RECOMMENDATIONS ............................................................................... 26
6.1 FinTech Policies to Promote High-Value Agriculture..................................... 26

6.2 Financial Inclusion for the Agriculture Sector ................................................ 27
6.3 Cross-Cutting FinTech investments .............................................................. 28
REFERENCES ..................................................................................................................... 29
1. INTRODUCTION
Credit and risk are pivotal dimensions of agriculture everywhere in the world. Two core
features of agricultural production are the long time lag between input investment
and profit realization, and the large covariate risks imposed on agricultural production by
weather shocks. These two dimensions create a set of interlocking problems both on the
supply side (financial institutions face large and systemic risks in providing
credit to agriculture) and on the demand side (farmers face many risks beyond their
control in trying to finance the investments necessary to increase productivity).
Fortunately, the technological landscape for the provision of financial services is shifting
quickly, and the developing world appears poised to leapfrog legacy systems in a number
of exciting ways. This paper summarizes recent advances in the technology that can be
used to underwrite credit and risk in agriculture, places in context the gaps in coverage
in Asia, and concludes with a set of policy recommendations as to the types of
interventions that appear most promising across the highly varying national contexts of
Asia.
The potential for digital financial services to increase growth in Asia, particularly among
excluded segments of the population, is substantial. Financial technology (FinTech) is
generating new ways to target and collateralize credit, to price and spread risk, and
to organize agricultural value chains. A 2017 ADB report finds that digital payment
systems could close 40% of the unmet need for payment services and 20% of the need
for credit. The same report finds that widespread implementation of digital financial
services could increase GDP growth in Indonesia and the Philippines by 2%–3% per
year, and in Cambodia by as much as 6% (ADB and Oliver Wyman 2017). Indeed,
worldwide we see that innovative financial technologies often take off precisely in
economies that have certain enabling features but do not have well-developed legacy
systems. Thus, microfinance has taken off in Indonesia and mobile banking in the
Philippines, both of which have relatively poorly developed formal financial systems, and
digital currencies dominate in the People’s Republic of China (PRC) where low credit
card penetration does not permit credit card use in online commerce. Therefore, FinTech
represents a space where innovation can be made to serve the marginalized in ways
that generate both welfare and economic growth.
Information and communication technology is changing agriculture in many dimensions
beyond financial services. Clearly, global access to mobile phones is fundamentally
changing the way that farmers access price information, search for buyers, and build
brands as they attempt to move up the value chain. That said, a large number of rigorous
studies conducted on the use of specific technology platforms to transmit price
information or to conduct agricultural extension have arrived at surprisingly mixed results.
Recent innovations in risk sharing, such as the use of index insurance, have largely fallen
flat due to lack of demand. Some of the excitement around novel FinTech solutions for
agriculture, such as the use of the blockchain, is still largely unproven. Hence, in this
paper we also provide a summary of the recent rigorous empirical evidence from field
studies as to the success of FinTech innovations, and try to point the way forward for the
most promising financial technologies.
The centrality of credit and risk may be clearest when we consider agriculture as a part
of the overall economy, and consider the central role played by farming in the broader
Agricultural Transformation and the subsequent Structural Transformation. Most
developing economies begin with a very large share of the population engaged in
smallholder agriculture, farming small plots with low capital intensity and trading little
of their output. To become a direct contributor to economic growth, agriculture must
1
become more capital intensive so as to allow it to bolster exports and contribute to overall
productivity. This process requires heavy investments to be made in the farming sector
which will in general amplify the financial risks faced by farming households. To permit
these investments to be made, then, we must consider both the access to finance
enjoyed by agricultural households and the tools at their disposal to control the risks they
face in making production more capital intensive. Agricultural land itself is the most
important store of value that can be used to collateralize this investment, which creates
an integral tie between land ownership rights and the apportionment of default risk in
agriculture. FinTech is being used both to enhance the ability of farmers to use collateral
and to permit new forms of more flexible, uncollateralized credit.
What is the particular role played by agriculture that makes it a sector of unique interest
in terms of economic development? Two particularly important and quite distinct
motivations are apparent. The first is its “macro” role in the AT/ST: in order to liberate the
labor that drives urbanization and industrialization, agriculture must engage in a set of
labor-saving investments that boost overall productivity. For poorer countries, the
development of agricultural processing and export businesses typically represents a
critical step on the path toward a modern economy. The second is its “micro” role as the
sector in which the large majority of the low-income population works, meaning that
aggregate poverty and inequality are likely to be driven strongly by changes to the nature
of agricultural production, particularly at low levels of overall development.
Both the “macro” and “micro” roles of agricultural development are served when a
country undergoes a successful structural transformation over the long term, but when
we seek to adjust agricultural policy at a moment in time, the levers for these two
purposes may be quite distinct. To promote agriculture’s role as a sector in the overall
economy, policy should be focused on improving value added, export markets, labor
productivity, diversification, and the transmission of labor toward more productive
sectors. For these purposes, FinTech needs to drive mechanization and processing
capacity, as well as manage a complex set of risks that accompany engaging with
the world market (quality certification, handling exposure to global commodity price
fluctuation, external demand shocks). If instead we focus on the critical micro-level
welfare implications of agriculture as a sector employing most of the population below
the poverty line, quite a different set of objectives may emerge, particularly in the short
term. Here, we may see smallholder farming more as a necessary reality to be
confronted, and hence interventions that generate marginal improvements in risk-
adjusted profits for households can have substantial welfare consequences, even in the
absence of any macro-level transformation. Indeed, GDP growth emanating from
agriculture has three times the effect on increasing the income of the bottom 40% of the
distribution as growth emanating from other sectors (de Janvry and Sadoulet 2009). In
focusing on agriculture’s pro-poor dimension, the policy focus will tend toward
micro-credit and micro-insurance, as well as interventions that can target and expand
access to financial services for previously marginalized groups. To use the language of
Dorward et al. (2009), the macro policies involve “stepping up” and “moving out” of
agriculture while the micro-level policies help farmers to “hang on” as smallholders.
In certain dimensions, these two objectives face common obstacles, so we can achieve
win-wins: both agendas can be promoted by improving agricultural productivity, shifting
weather and global price risks off of farmers, and enhancing household-level income
diversification. In other dimensions, they may be at odds: transfers in the form of
subsidies to prices, interest rates, or the cost of agricultural risk may enhance welfare in
the short term yet retard the movement toward the consolidation of land and movement
of labor that is typically associated with economic development at the national level.
FinTech appears to be a particularly attractive approach to agriculture when seen in this
2
light because it embeds a strong logic for the win-win propositions of better targeting of
credit, better pricing of risk, and a shift of aggregate risk to better diversified parties. All
these changes should be beneficial to both macroeconomic efficiency and the welfare of
smallholder farmers.
FinTech plays a number of critical roles in driving the ability to provide mass-scale
agricultural finance, particularly in developing countries where access may be far from
universal. The array of digital technologies has dramatically decreased the cost of
providing services on the margin, allowing them to be offered in smaller packages to
poorer customers. This holds out the promise that less developed countries and remote
regions could leapfrog legacy systems and use mobile/digital technology to drive
agricultural productivity in novel ways. Big data tools allow institutions to target credit
more precisely, thereby reaching better borrower pools and expanding access to
uncollateralized credit. Better measurement of climate shocks using various types of
remote sensing permit a shifting of covariate risk within the agricultural system (although
progress has been slower here than with credit).
This paper provides an overview of the use of FinTech for agriculture. We begin our
analysis by providing context for levels of financial penetration, agricultural productivity,
and stages in the agricultural transformation globally and within Asia. We then move to
a detailed discussion of credit products, and the way in which novel technologies can
target and extend uncollateralized credit in new ways, can allow new types of assets to
serve as collateral, and allow for an expansion of agricultural productivity. We then move
to the use of FinTech to tackle risk in agriculture, focusing on the weather index insurance
products that have been extensively piloted over the past decade. Following that, we
discuss three specific forms of FinTech that are critical enablers of changes
to the operation of agricultural supply chains—mobile money, biometric identification,
and the blockchain—as well as the role financial literacy plays in generating financial
inclusion. We conclude with a set of policies and products that seem to have particularly
strong theoretical or empirical justification, and the ways in which the context and policy
objectives alter the critical areas of focus.
2. THE CONTEXT
Seen from a macroeconomic context, agriculture is playing a declining role in production
and employment in ADB members. From 2000 to 2016, employment in agriculture
among the ADB members fell from 46% to 30%, and the contribution of agriculture to
total value added declined from 21% to 12.7% (Table 1). In 1990, the population of ADB
members was 70% rural and only 30% urban, while by 2017 the share living in rural and
urban areas had equalized (Figure 1). In keeping with an agricultural transformation that
is well underway, these changes were accompanied by a large increase in agricultural
productivity, with cereal yields per hectare increasing on average from 2,848 kg in 2000
to 3,637 kg in 2016, an improvement of 28% in only 16 years. Figure 2 shows the
evolution of cereal yields over this interval by country, demonstrating particularly
impressive improvements in the PRC and Indonesia.
In aggregate, then, Asia is clearly seeing the mass-scale movement of population out of
rural areas and the mechanization of agriculture that is critical to a structural
transformation in the process of overall economic development. However, workhorse
theoretical models such as the Lewis model have suggested for decades that increases
in the welfare of urban workers will ultimately be limited by the level of immiseration in
the rural areas, and therefore finding the policy and financial tools to allow for an increase
in wages and productivity in agriculture are key to allowing economic development to
3
benefit workers. Improving the flow of agricultural financing and technology is critical in
allowing countries to escape a low value-added poverty trap (Manova and Yu 2013),
even as the region as a whole moves away from agriculture.
Table 1: The Role of Agriculture in ADB Members, 2000–2016

(% of Total Value Added,a % of Total Employment,b c % of Total Exports,d Kg/Hectaree f)
% of Value Added Employment Share Employment Share
from Agriculturea in Agricultureb in Industryc
ADB Members 2000 2016 2000 2016 2000 2016
Afghanistan 33.75 20.97 76.83 61.77 5.50 6.72
Armenia 27.11 15.66 46.80 34.20 16.92 15.91
Australia 3.20 2.60 4.55 2.61 21.56 19.28
Azerbaijan 16.73 5.62 41.83 36.87 11.45 14.23
Bangladesh 21.99 13.73 59.92 40.10 11.66 20.96
Cambodia 37.75 24.06 74.48 27.07 7.56 27.03
PRC 15.36 8.35 43.40 18.07 28.66 26.73
Georgia 24.95 7.38 53.01 41.08 9.40 12.43
Hong Kong, China 0.08 0.08 0.31 0.22 20.56 13.12
India 22.01 15.87 59.66 43.09 16.56 23.75
Indonesia 16.20 13.31 44.06 31.50 18.10 21.72
Japan 1.47 1.15 5.05 3.50 31.15 25.60
Kazakhstan 9.14 4.49 36.76 17.97 15.87 20.79
Korea, Rep. of 3.96 1.94 10.16 4.90 28.90 24.84
Kyrgyz Republic 35.84 12.58 48.82 26.74 13.24 22.13
Lao PDR 35.77 16.72 82.47 61.67 4.06 9.60
Mongolia 26.82 11.05 51.37 30.39 11.61 19.01
Myanmar 52.38 25.82 74.86 50.61 10.67 16.45
Nepal 36.92 28.10 75.79 72.01 7.13 7.95
New Zealand 7.31 8.66 6.57 23.46 20.20
Pakistan 22.96 23.05 45.41 42.14 18.95 23.66
Papua New Guinea 34.91 66.52 20.63 4.38 7.50
Philippines 15.46 9.66 38.51 26.47 16.07 17.60
Sri Lanka 18.31 7.59 38.40 27.10 24.62 25.55
Tajikistan 26.46 20.38 59.98 51.18 16.36 16.47
Thailand 9.10 8.58 47.55 33.05 19.79 22.66
Timor-Leste 7.37 11.35 60.40 24.89 7.85 14.66
Turkmenistan 20.15 22.81 8.34 38.68 44.95
Uzbekistan 27.69 16.72 38.01 21.93 31.20 37.61
Viet Nam 24.22 15.83 63.56 41.37 13.73 24.92
Unweighted Average: 21.18 12.69 46.00 30.27 16.86 20.13
continued on next page
4
Table 1 continued
Export Share
in Agricultured Fertilizer Usee Cereal Yieldsf
ADB Members 2000 2016 2000 2016 2000 2016
Afghanistan 16.36 3.86 1,318.83 1,981.70
Armenia 3.27 0.59 27.97 1,809.08 3,076.10
Australia 6.04 2.86 47.39 1,909.33 2,074.30
Azerbaijan 4.43 0.28 10.85 2,162.83 3,004.70
Bangladesh 1.65 179.33 3,149.29 4,628.90
Cambodia 2.15 2.08 5.79 1,990.36 3,459.90
PRC 1.03 0.42 402.14 4,906.07 6,029.20
Georgia 2.86 1.25 33.04 1,994.86 2,517.20
Hong Kong, China 0.37 2.95 461.20 2,000.00 2,000.00
India 1.45 1.27 112.10 2,286.27 2,992.80
Indonesia 4.72 5.07 132.57 4,053.31 5,405.50
Japan 0.52 0.75 342.59 6,002.18 4,975.50
Kazakhstan 1.74 0.30 1.23 937.18 1,347.70
Korea, Rep. of 1.00 0.89 512.42 6,226.77 6,795.20
Kyrgyz Republic 13.06 3.30 19.52 2,515.39 3,104.40
Lao PDR 3.24 2,966.76 4,626.70
Mongolia 21.81 6.93 6.30 682.78 1,279.40
Myanmar 2.60 10.26 3,041.03 3,607.40
Nepal 0.73 3.55 8.33 2,096.52 2,605.40
New Zealand 13.91 12.25 2,148.70 6,360.53 8,383.80
Pakistan 2.44 0.91 108.18 2,256.48 3,064.20
Papua New Guinea 4.90 175.22 3,792.42 4,737.80
Philippines 0.71 0.76 157.78 2,591.38 3,529.00
Sri Lanka 1.87 2.65 276.51 3,298.01 3,897.40
Tajikistan 12.59 35.88 1,523.00 3,348.70
Thailand 4.10 3.87 126.03 2,747.82 3,031.80
Timor-Leste 0.13 1,541.90 2,454.40
Turkmenistan 16.79 2,192.81 1,075.60
Uzbekistan 2,946.42 4,613.10
Viet Nam 2.21 1.41 335.86 4,135.12 5,448.00
Unweighted Average: 4.86 3.19 218.50 2,847.82 3,636.53
ADB members with population over 1 million.
Source: World Development Indicators Database 2018.
5
Figure 1: Evolution of Rural/Urban Population
Source: World Indicators 2018.
Figure 2: Cereal Yields (kg per hectare for wheat, rice, maize, barley, oats, rye, millet,
sorghum, buckwheat, and mixed grains) by Country and Year
Cereal Yields, by Country and Year

1000 2000 3000 4000 5000 6000
Cereal yields per hectare
1970 1980 1990 2000 2010 2017

Year
CHN AUS
KHM THA
IND IDN
Source: FAO Stat.
Further, this picture of a strong aggregate decline in the importance of agriculture masks
an enormous degree of cross-country differentiation. Indeed, in this respect, Asia is the
most heterogeneous area of the world. This can be seen in the difference between the
development level of countries like Japan and the Republic of Korea, which together
account for almost 6% of the world gross domestic product (GDP), and that of Myanmar
and Nepal, which combined do not represent 0.40% of the world’s GDP (World
Development Indicators Database 2018). Even in the most recent data, countries such
as Afghanistan, the Lao People's Democratic Republic (Lao PDR), and Nepal still have
more than 60% of their population engaged in agriculture, while this fraction is below 5%
for Australia; Hong Kong, China; and Japan. Given the importance of export markets as
6
a vehicle for value added in agriculture, this diversity in development levels within Asia
should be seen as an opportunity for less developed countries in the region to exploit the
purchasing power of their wealthier neighbors to their own benefit. Also, agriculture can
continue to play an important role in the economies of even very developed economies,
as witnessed by the fact that the farming sector represented 12.25% of New Zealand’s
exports in 2016, second only to Afghanistan within Asia.
Table 2: Changes in Welfare, 2000–2017

(Purchasing Power Parity,a Kcal/dayb)
% of Population
GDP Per Capitaa Rural Food Deficitb
1995– 2016– 1995– 2016– 1995– 2016–
ADB Members 2005 2017 2005 2017 2005 2017
Afghanistan 1,090 1,803 78.70 72.65 311.45 173.00
Armenia 3,308 8,489 35.14 37.49 143.09 41.00
Australia 34,828 44,571 12.89 10.38
Azerbaijan 4,820 15,924 48.23 44.96 144.00 12.00
Bangladesh 1,651 3,422 76.09 64.59 196.00 116.00
Cambodia 1,413 3,555 81.53 78.94 201.55 97.00
PRC 3,875 14,703 63.70 42.85 138.36 74.00
Georgia 3,446 9,511 47.15 46.07 176.82 55.00
Hong Kong, China 34,414 55,234 0.00 0.00
India 2,520 6,261 72.21 66.66 135.09 109.00
Indonesia 6,163 10,978 58.56 45.18 118.00 51.00
Japan 33,847 38,642 19.50 5.88
Kazakhstan 11,016 23,751 44.45 46.77 26.64 18.00
Korea, Rep. of 20,755 35,479 20.26 17.35 12.09 4.00
Kyrgyz Republic 2,072 3,352 64.48 64.07 103.09 40.00
Lao PDR 2,516 6,235 77.86 59.84 297.36 128.00
Mongolia 4,860 11,601 41.59 26.80 324.18 156.00
Myanmar 1,415 5,448 72.93 65.07 429.18 103.00
Nepal 1,519 2,370 86.76 80.81 153.55 51.00
New Zealand 27,938 35,931 14.29 13.66
Pakistan 3,539 4,946 66.79 60.54 168.64 172.00
Papua New Guinea 2,844 3,820 86.59 86.94
Philippines 4,288 7,416 52.32 55.74 151.82 93.00
Sri Lanka 5,391 11,557 81.56 81.56 261.00 192.00
Tajikistan 1,295 2,830 72.94 73.05 260.82 250.00
Thailand 9,802 15,992 67.31 47.89 163.82 53.00
Timor-Leste 6,294 6,934 73.69 66.29 227.00 184.00
Turkmenistan 5,216 16,019 54.08 49.42 56.00 22.00
Uzbekistan 2,548 6,146 62.55 63.45 66.36 29.00
Viet Nam 2,615 6,005 75.47 65.44 218.73 83.00
Unweighted average: 8,243 13,964 56.99 51.34 179.39 92.24
ADB member countries with population over 1 million.
7
The deep structural changes to Asian economies have been accompanied by a very
dramatic reduction in poverty. Table 2 shows that average GDP per capita among larger
ADB members has risen from just over $8,000 to almost $14,000 in purchasing power
parity terms during the 18 years from 2000 to 2017, and over the same interval the
average food deficit fell from 180 to 92 kilocalories per person per day. This halving of
nutritional shortfalls over such a short period is impressive, and given the large
decreases in the share of the population in rural areas over the same period might be
thought to have resulted from urbanization. However, Figure 3, which plots the declines
in poverty rates at the national level over the same period, tells quite a different story. It
shows that subsequent to 2002 almost all of the reduction of aggregate poverty that has
taken place in Asia has done so in rural areas, emphasizing the critical role that the
agriculture sector plays in employing the poor. In this sense, while agriculture may play
a declining role in overall macroeconomic importance over time, it is the sector in which
growth has the strongest role in reducing poverty (de Janvry and Sadoulet 2009).
Figure 3: Evolution of National Poverty Rates, ADB Members
Evolution of National Poverty Rates: Rural and Urban Population

Asian Development Bank Members - 2000 to 2016
60%
Poverty Headcount Ratio
20% 0% %40
11
00
01
02
03
04
05
06
07
08
09
10
12
13
14
15
16
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
Rural Population Urban Population
Source: World Development Indicators, 2018.
Clearly, access to financial services will play a critical role in enhancing agricultural
productivity and thereby generating the benefit from its poverty-reducing abilities. In this
regard as well, we see both tremendous growth in Asia over recent decades and a great
deal of heterogeneity remaining at the present moment. From 2000 to 2017, overall credit
to the private sector rose from 42.5% to 62.8%, the number of bank branches per
100,000 people rose from 13 to 16.5, and the number of depositors per 1,000 adults rose
from 1,350 to 1,550. Perhaps most impressive, there was a reduction of more than one-
third in the average interest reported in the WDI, from an APR of 15.4% to just over 10%.
In general, there has been a very meaningful improvement
in overall financial depth over this period. Again, however, this masks enormous
heterogeneity. Afghanistan, Kazakhstan, the Lao PDR, Myanmar, and Viet Nam each
have fewer than five bank branches per 100,000 people, while Georgia, Japan,
Mongolia, and Uzbekistan each have more than 30.
8
Table 3: Financial Depth

Credit to the Depositors with
Private Sector, Bank Branches Commercial Banks
% of GDP per 100,000 pop (per 1,000 adults) Interest Rates
1995– 2016– 1995– 2016– 1995– 2016– 1995– 2016–
ADB Members 2005 2017 2005 2017 2005 2017 2005 2017
Afghanistan 3.42 0.48 2.22 188.98 15.00
Armenia 7.24 47.52 11.57 23.10 289.91 1,567.77 34.48 15.88
Australia 87.29 141.71 30.65 27.80 7.50 5.33
Azerbaijan 4.85 21.40 6.64 2,021.38 17.77 16.46
Bangladesh 23.13 46.25 6.89 8.44 345.99 724.90 12.66 9.97
Cambodia 6.31 83.97
PRC 107.05 156.47 8.78 34.04 6.95 4.35
Georgia 7.26 57.68 10.21 32.67 324.18 1,867.42 21.13 12.05
Hong Kong, China 152.19 203.80 23.40 21.43 7.43 5.00
India 29.09 49.55 8.95 14.06 607.31 1,731.27 12.80 9.59
Indonesia 33.80 32.77 5.22 17.39 497.12 1,055.58 20.07 11.48
Japan 144.68 102.98 34.50 34.10 7,907.62 7,190.45 2.21 1.04
Kazakhstan 14.60 28.33 3.84 2.96
Korea, Rep. of 83.00 143.88 17.07 16.26 4,121.82 5,212.49 8.59 3.42
Kyrgyz Republic 6.05 20.75 5.22 8.38 44.73 22.18
Lao PDR 8.09 3.04 579.16
Mongolia 12.89 52.24 40.83 70.37 324.79 1,333.64 35.40 19.87
Myanmar 8.15 22.72 1.78 3.41 114.42 251.84 15.67 13.00
Nepal 26.17 80.92 2.56 9.58 657.04 10.81
New Zealand 104.49 35.08
Pakistan 24.72 16.69 7.71 10.36 163.55 380.10 8.16 8.48
Papua New Guinea 15.69 17.19 1.83 14.29 8.76
Philippines 38.95 46.24 8.11 8.87 364.95 577.09 12.41 5.63
Sri Lanka 30.43 45.56 8.93 10.94 10.49
Tajikistan 10.76 15.10 4.96 12.49
Thailand 119.17 112.56 8.01 12.37 1,144.14 1,549.30 10.10 4.44
Timor-Leste 5.84 8.37 2.11 5.04 13.67
Turkmenistan
Uzbekistan 39.58 36.14
Viet Nam 35.51 127.24 3.87 963.38 12.09 7.18
Unweighted 42.50 64.82 13.04 16.55 1,350.48 1,549.21 15.39 10.15
average:

What, then, are the levers that can be used to enhance financial access? FinTech
services can play a crucial role here. Table 4 shows summary statistics for three
important underlying institutions that enable financial deepening, namely credit bureaus,
asset registries, and access to mobile services. In the period 1995–2005, only four ADB
members had meaningful credit bureau coverage (Australia; Hong Kong, China; the
Republic of Korea; and New Zealand), and none of them had credit registry coverage for
more than 5% of the population. By 2016–2017, this picture had changed substantially;
the number of countries in which more than half the population was covered by credit
bureaus had more than doubled, and credit registries had become more widespread.
Australia; Hong Kong, China; Japan; the Republic of Korea; and New Zealand all report
universal coverage by credit bureaus in the most recent data. As indicated in Figure 4,
9
credit bureau coverage is highly correlated with the overall level of economic
development of a country. Interestingly, credit registries have taken off in a set of less
developed economies (members with more than 40% coverage are the PRC, Indonesia,
Mongolia, and Viet Nam), suggesting that the development of credit registries has been
driven more by policy than emerging as a natural feature of overall economic
development. As might be expected, mobile phone coverage has skyrocketed during this
period, with the number of mobile subscriptions rising more than tenfold, from 12 per 100
people to 114.7. More than half of the countries in the region have more than one mobile
phone subscription per adult, and the lowest rate of penetration, in Papua New Guinea,
still indicates that nearly half of all adults have mobile phones.
Table 4: Presence of Financial Infrastructure

Mobile Subscriptions
Private Credit Bureau Public Credit Registry (number of
Coverage Coverage subscriptions
(percent of population) (percent of population) per 100 people)
1995– 2016– 1995– 2016– 1995– 2016–
ADB Members 2005 2017 2005 2017 2005 2017
Afghanistan 0.0 0.0 0.0 0.9 0.75 62.34
Armenia 0.0 74.2 1.3 0.0 2.33 117.43
Australia 97.7 100.0 0.0 0.0 48.46 110.05
Azerbaijan 0.0 0.0 0.2 37.0 7.82 104.77
Bangladesh 0.0 0.9 0.98 83.45
Cambodia 0.0 47.0 0.0 0.0 2.39 126.35
PRC 21.3 92.5 10.62 97.25
Georgia 0.0 92.2 0.0 0.0 7.86 140.95
Hong Kong, China 63.0 100.0 0.0 0.0 72.77 240.80
India 32.5 0.0 1.65 85.17
Indonesia 9.2 53.6 5.02 147.66
Japan 100.0 0.0 48.28 130.61
Kazakhstan 0.0 53.2 0.0 0.0 6.55 141.96
Korea, Rep. of 80.7 100.0 0.0 0.0 46.73 120.68
Kyrgyz Republic 0.1 33.8 0.0 0.0 1.87 127.84
Lao PDR 0.0 0.0 0.0 11.1 1.76 58.57
Mongolia 0.0 0.0 3.5 43.6 7.04 111.24
Myanmar 0.0 0.0 0.08 95.65
Nepal 0.1 1.8 0.0 0.0 0.17 110.83
New Zealand 96.8 100.0 0.0 0.0 43.65 124.44
Pakistan 6.3 9.7 1.43 70.65
Papua New Guinea 0.0 6.6 0.0 0.0 0.29 46.78
Philippines 3.6 9.1 0.0 0.0 14.43 109.37
Sri Lanka 2.1 46.1 0.0 0.0 4.56 124.03
Tajikistan 0.0 37.6 0.0 0.0 0.62 107.61
Thailand 16.7 54.8 0.0 0.0 16.48 173.78
Timor-Leste 0.0 0.0 0.0 5.6 3.22 117.61
Turkmenistan 0.38 151.43
Uzbekistan 0.0 34.0 0.0 0.0 0.72 73.98
Viet Nam 0.0 17.3 1.0 46.4 2.41 127.53
Unweighted average: 16.4 37.1 0.3 10.4 12.04 114.69
10
Figure 4: Credit Bureau Coverage and GDP, 2017
Source: WDI.
How important are financial services in promoting the growth of agriculture in Asia? To
get a picture of this, we must move to more specialized data. Information from the World
Bank’s Global Financial Inclusion Database (FinDex) database allows us to disaggregate
the use of financial services only in rural areas. Figure 5 shows that Asia and the Pacific
overall has similar credit depth in rural areas as Latin America and the Caribbean (LAC)
(about 20% of the rural population have loans from a financial institution), that over 70%
of the rural population is banked in having access to a debit card, and roughly half of the
rural population uses digital payments, but that only just over 10% of the rural population
has taken a loan to start a business. In general, Asia and the Pacific shows rates of rural
financial penetration that are between those of the developed world (North America and
Europe) and the developing world (Latin America and the Caribbean, Middle East and
North Africa, and Sub-Saharan Africa).
Figure 5: Use of Financial Services in Rural Areas
11
Use of Financial Services within Rural Areas, 2017

Percent of Rural Population, By Region
Asia
EUR
LAC
MENA
NAm
SSA
0 20 40 60 80 100
Borrowed from an FI Borrowed to Start Business

Has a Debit Card Uses Digital Payments
Source: FinDex/WDI.
We conclude the overview of context by suggesting that huge possibilities exist for
financial development to push agricultural development in Asia. Data from FinDex show
that the largest concentrations of unbanked individuals are in Asia, not just in
the PRC and India but in Indonesia, Viet Nam, and the Philippines. Similarly, a total of
235 million individuals worldwide are unbanked and receiving agricultural payments in
cash, and again Asian countries feature prominently in this number. The intersection of
these two facts suggests that mobile payments have the potential to revolutionize the
way that agricultural transactions take place. Figure 6 shows that even for farmers who
have access to financial accounts, 80% of individuals in the PRC, India, and Thailand
received at least some of their payments in cash, and in these countries between 5%
and 20% of banked farmers received all of their payments in cash. We now move on to
a more detailed discussion of the role of FinTech in extending credit and protecting
against risk in agriculture.
Figure 6: Payments in Cash for Agricultural Products
12
Reprinted from World Development Indicators, Global FinDex Database:

Gallup World Survey, 2007, Chapter 6, Figure 6.11.
Source: Global FinDex database.
3. CREDIT AND SAVINGS PRODUCTS

Agricultural lending can be seen as a way of understanding the marginal returns to capital
in the agriculture sector. In contexts where supply chains are not well developed and
poor infrastructure makes input prices high and output prices low, a set of recent
experimental studies attempting to extend credit to farmers have typically seen quite
weak demand. A microfinance project in a rural area of Morocco facing virtually no
competition from other lenders saw takeup of 17% of households (Crepon et al. 2015).
Experiences from Sierra Leone and Mali have seen takeup in the range of 21%–25%.
The contrast of these experiences from SSA with Asia is quite stark. In Asia, where
agriculture is quickly transforming, transportation and processing infrastructure is good,
and high-value exports are growing quickly, takeup of agricultural loans has typically
been much higher. Asia has a number of pioneering examples of large-scale
microfinance institutions that have substantively changed access to credit both within
agriculture and more broadly, including Bank Rakyat Indonesia, the Thai Village fund,
and the Bank for Agriculture and Agricultural Cooperatives (BAAC) which reports
providing access to credit to 95% of farming households in Thailand.
Historically, much of the credit provided to the agriculture sector has either come from
state lending institutions or been underwritten by implicit or explicit guarantees from
public actors. Given the relatively high probability that such a system will be hit by
systematic default driven by weather shocks, public credit institutions have struggled to
balance the political pressures to provide a safety net against the financial pressures
of long-term solvency. In particular, providing the blend of insurance against unavoidable
aggregate shocks while retaining the ability to collect loans under normal circumstances
13
has proved challenging. For private-sector institutions as well the undiversifiable nature
of large-scale agricultural shocks difficult to manage, and is therefore likely to restrict the
flow of credit to the sector (Carter, Cheng, and Sarris 2016). Hence farmers in many
countries face real barriers to access to credit arising from the intersection of returns and
risk. We will argue that state actors have a critical role to play in creating the enabling
environment for healthy credit markets, but that ultimately the goal should be to crowd
private commercial capital into agricultural investments.
3.1 Microfinance
Ironically, some of the fastest growth in credit to rural areas has come via a modality that
was in many ways designed to prevent borrowers from using it for standard agricultural
investment. The microfinance revolution, which has brought $102 billion of investment
into developing countries and extended credit to 123 million borrowers worldwide (BNP
Paribas, 2017), has generally employed a payment model that is explicitly unsuited for
agriculture. Borrowers take loans in cycles (most typically four months) with frequent
repayment beginning almost immediately after disbursement of the loan. This is an ideal
repayment schedule for the cash flow of retail businesses and can with some
diversification be made to work for fast turnover agriculture (such as vegetable gardens),
but in general cannot be used to finance long-cycle agricultural investment. The picture
with regard to microfinance in Asia, however, is quite distinct. Borrowers of the Thailand’s
Village Fund are not only poor but are disproportionately agricultural (Boonperm,
Haughton, and Khandker 2013). Several major Asian microfinance institutions have
explicitly created products for agriculture, such as Thailand’s BAAC, for which loans to
non-agricultural households are capped at 20% (Terada and Vandenberg 2014). In this
sense, then, Asia appears to have been uniquely successful at pushing microfinance
institutions to serve the agriculture sector.
Several institutional innovations gave rise to the microfinance revolution. First, the
concept of joint liability (Besley and Coate 1995; Ghatak and Guinnane 1999) allowed
borrowers to be mutually responsible for each other’s debts, and consequently to
substitute social collateral for physical collateral. This serves to combat both adverse
selection (because members will only agree to be jointly liable for those they trust)
and moral hazard (because members are incentivized to exert suasion to ensure
repayment by their group members), thereby helping to resolve one of the core
asymmetric information issues that bedevils credit markets (Stiglitz and Weiss 1981).
Second, microfinance lenders typically use high-powered dynamic incentives, whereby
borrowers are started with very small loan sizes, and their ability to work their way up to
larger loans is predicated on successful repayment of each subsequent loan. Finally,
microfinance represents an early form of FinTech in that all MFIs look for ways to
decrease costs on the margin by using contracts and technology to be able to offer small
loans with minimal fixed costs. Examples include the use of group lending contracts,
disbursements and/or payments via mobile money, as well as the use of rapid
diagnostics implemented via tablet surveys to target credit toward good or deserving
borrowers.
3.2 Credit Reporting

A core linkage between FinTech and microfinance lending is the nature of credit
information sharing. Credit bureaus can serve as a critical borrower discipline device
(Jappelli and Pagano 1999). Sharing information on total levels of indebtedness may be
particularly important in microfinance markets where many borrowers receive loans
smaller than their actual demand for credit (due to capped loan sizes necessary to
14
maintain dynamic incentives), and so may try to borrow from multiple lenders (McIntosh
and Wydick 2005). Recent empirical work shows that FinTech innovations can improve
targeting of credit even for populations with very little credit history; Bjorkegren and
Grissen (2017) illustrate that mobile phone call data records are strongly predictive of
default behavior in Rwandan microfinance. Despite the obvious theoretical justification
behind the formation of credit bureaus, the actual sharing of credit information presents
substantial strategic risks to banks and MFIs (Padilla and Pagano 1997). The institutions
that are most important to include are the largest lenders, but these organizations also
potentially have the most to lose by sharing (because they reveal more information than
they learn), so in practice the path toward the establishment of credit bureau coverage
has been a slow and uneven one in the developing world
(de Janvry et al. 2010). This combination of potentially large welfare gains with serious
incentive problems in the formation of bureaus appears to make this a central area of
focus for regulatory policy, since efficient sharing systems may not emerge without
governmental requirements.
Institutional details in the design of credit bureaus are critical. Vercammen (1995)
presents a theoretical model that shows that while retention of credit information is
important, imposing a statute of limitations on the time window of data recorded in the
bureau is also important. Without this, individuals may develop reputations that are either
so good or so bad that they become relatively immune to recent changes in behavior,
thereby dulling incentives for continuous good repayment. Bureaus can also exist at
several levels of informational granularity; the most basic of these is a simple list of
defaulting individuals; such bureaus are relatively easy to establish and impose some
check on asymmetric information problems with respect to default, but do not allow
lenders to price risk in any sophisticated way. More complete bureaus contain real-time
information on current indebtedness levels of borrowers, meaning that at the time of
applying for a loan the new lender can accurately price the risk of default by incorporating
the pre-existing debt load. The data and technological requirements for establishing real-
time data sharing are formidable for many smaller MFIs who still operate most accounts
using spreadsheet programs, but as advanced management information systems and
high-speed internet connectivity become more ubiquitous, these obstacles are falling
away.
This leaves the regulatory hurdles as a core enabler or obstacle to the formation of
information sharing bureaus in microfinance markets. Key regulatory issues in the
creation of bureaus include (i) which types of institutions are required to share
information, (ii) the exact nature of the information to be shared, (iii) the circumstances
under which financial institutions are permitted to query the credit of a potential borrower,
(iv) exactly what information financial institutions can observe about queried borrowers,
and (v) what the recourse is for borrowers who find that the bureau contains incorrect
information on them.
The empirical literature suggests that the introduction of credit bureaus can have a
substantial impact on microfinance markets. De Janvry et al. (2010) find that the
introduction of a credit bureau into Guatemalan microfinance markets led to a sharp
increase in client turnover in the six months after the bureau was introduced. In the month
that the bureau was introduced to a branch of the MFI, more than 60% of the pre-existing
clients applying for new loans were checked, and 11% of pre-existing borrowers were
refused new loans as a result. Those refused were more likely to have defaulted on
exterior loans (37% of the refused had defaulted while only 21% of those given loans
again had defaulted) and were disproportionately the types of individuals with low
repayment (males and more educated clients) and those with more variable repayment
15
(males). Counterbalancing this exodus of troubled pre-existing borrowers was a huge

influx of new individual borrowers who were extended credit; on aggregate, the bureau
led to an increase in the number of borrowers of 27% within six months, and generated
no deterioration in repayment performance. Despite this strong evidence of enhanced
profits for participating microfinance institutions, the Guatemalan credit bureau did not
receive strong government backing and has struggled to maintain membership because
of the incentive issues described above.
3.3 Asset Registries and Warehouse Receipt Systems

Because much agricultural investment involves purchasing durable equipment (tractors,
pumps, tools, combines, water tanks, etc.) it can be self-financing in the form of leasing.
For financial institutions, however, leasing presents an identification problem akin to the
one addressed by bureaus at the borrower level: when offering a lien on an asset, how
can the lender be assured that the same asset has not been promised as security to
anyone else? Asset registries may be particularly critical in increasing the flow of credit
to smaller producers (De la Campa 2011; Jack et al. 2016). Two policy reforms that have
been highlighted as particularly important in the literature are the writing of updated
secured transaction laws and the formulation of electronic asset registries that, like
sophisticated credit bureaus, allow for the real-time verification of data from the universe
of other lending institutions.
A specific form of asset registry that has received a great deal of policy attention is a
warehouse receipt (WR) system. This permits farmers to borrow money against grain
stored in a warehouse, thereby permitting them to realize cash immediately at harvest
time without the grain being sold immediately. Such a system is motivated most strongly
in shallow markets with poor producers, where prices tend to be lowest at harvest time
but appreciate strongly and predictably in the months thereafter, but where producers
lack the capital necessary to delay selling until prices are higher. As long as the rate of
price appreciation exceeds the sum of the interest and spoilage rates over the months
subsequent to harvest, WR systems should permit a win-win-win, in which farmers
achieve better prices on average, financial intermediaries make money on the loans, and
consumers benefit via better smoothed agricultural prices (Burke, Bergquist, and Miguel
2018). Unfortunately, WR systems have proven quite complex to build and maintain.
They require credible registration of the collateralized grain, as well as deeply liquid
warehouse operators who have the ability to extend a large amount of credit. While
companies such as Yes Bank in India and Quedancor in the Philippines have worked to
establish WR systems (Miller 2008), in general their growth has been limited by issues
such as the lack of warehouse capacity, lack of liquidity for warehouse operators, and
lack of understanding of the system among smallholder farmers (Shalendra, Haque, and
Anu Peter 2016). In addition, WR systems expose intermediaries to substantial
commodity price risk (Miranda, Mulangu, and Kemeze 2017). In any case, WR systems
can only deal with the limited issue of post-harvest price variation and are not in general
a solution to the overarching problem of the lack of financing for long-cycle agricultural
investment during planting time.
3.4 Savings Products

Another financial tool to address liquidity and risk constraints is savings, which permits
investment as well as enables households to take more risk if consumption can
be protected using savings as a buffer. Like credit, enhanced access to formal
micro-savings has the potential to improve productivity through both macro-level
intermediation effects (Diamond 1984) and micro-level benefits for households
16
(Rutherford 2000). Recent experimental literature on micro-savings products has

found effects ranging from moderate (an increase in formal savings but no overall
changes in savings, consumption, or income; see Dupas et al. 2018), all the way up to
large and durable impacts of temporary savings products on household expenditures
multiple years later (Prina 2015; Schaner 2017). Specific efforts have been made to link
micro-savings products to agriculture by aligning them with the input purchase calendar,
and “labelling” deposits in accounts as being for the purpose of input use. Brune et al.
(2016) use a commitment savings experiment to encourage households to increase
investments in agricultural inputs, and find not only savings but investments, agricultural
profits, and consumption to be elevated as a result (although they find
the increase in investment to be larger than the savings effect, suggesting that the
intervention increased the salience of input investment in ways other than the savings
account). Carter et al. (2016) find that when households receive subsidies, savings
products actually compete with agricultural investment, meaning that households with
access to savings accounts transfer less of a windfall into future input investment (using
it instead to buffer). Hence, access to savings interacts in complex ways with agricultural
investment.
FinTech innovations have been used in numerous ways to enable micro-savings. When
government transfer programs move from cash payouts to the use of deposits to formal
accounts with debit cards, evidence from Mexico suggests that household savings will
rise over time as a result (Bachas et al. 2017). No small part of this effect appears to
arise from the ability to easily check account balances and verify deposits, thereby
raising the credibility of the formal banking system in environments in which trust may be
a major obstacle (Callen et al. 2014). Nonetheless, despite the substantial empirical
evidence suggesting that micro-savings are beneficial and that FinTech enables these
products to scale and to serve clients in new ways, mass-scale micro-savings accounts
have been slow to propagate. Interest rates on these accounts is typically zero or very
low, regulatory barriers to deposit-taking are substantial, and profits to financial
institutions may be limited.
17
Micro-savings products appear to have strong justification in terms of increased welfare

and income among borrowers, as well as in improving overall intermediation and
financial depth in the economy. A major obstacle to the scaling of FinTech for savings is
that, quite rightly, financial regulators have set strict accounting and liquidity
requirements not only for banks but for non-bank financial institutions that wish to begin
intermediating savings. This has prevented many smaller, more nimble institutions from
being able to capture savings, yet is based on a justified desire to protect citizens from
potential loss of savings to careless or unscrupulous deposit-taking institutions. Hence,
progress in this area requires careful coordination between policymakers and innovative
institutions to allow formal savings to grow in a safe, carefully regulated environment.
4. INSURANCE AND RISK

4.1 Risk in the Agricultural Finance System
Agricultural incomes are subject to substantial covariate shocks, particularly when
farming is rain-fed. Recent empirical evidence from Thailand by Samphantharak and
Townsend (2018) illustrates that households are more sensitive to correlated
(undiversifiable) risks on the margin than they are to idiosyncratic (and hence potentially
diversifiable risks). This correlated risk sits at the heart of the agricultural financing
system, and the core question is how to structure ownership of this risk in a manner that
generates as few distortions to agricultural efficiency as possible. If it remains on the
shoulders of farmers, they may rationally underinvest in inputs and thus slow down the
Agricultural Transformation. If it moves to the agricultural banking system, it will lead to
an under-provision of credit due to the difficulty of hedging this large covariate shock for
any but the largest and best diversified banks. Conceptually, we might prefer to shift it to
the shoulders of the highly diversified international re-insurers, but this is not always
straightforward for reasons we will discuss. Similarly, price risk is an inherent feature of
agricultural trade, and smallholder farmers moving into integration with international
markets may lose the implicit revenue insurance provided by shallow output markets
(when local harvests are poor, local prices in shallow markets are high). In general, the
Agricultural Transformation requires farmers to make a set of investments that
dramatically increase the financial risks of farming, making issues of risk and insurance
paramount.
4.2 Micro-insurance
The most important advance in the provision of micro-insurance in recent years has been
index insurance. While the insured party typically desires complete insurance
as provided by indemnity policies, assessing losses is an expensive process, and
indemnity insurance can introduce issues of adverse selection and moral hazard for
insurers. Instead, index insurance seeks to find a cheaply collectible proxy for the
covariate shock inherent in agriculture and to insure only the component of variation
correlated with this index. The most typical examples in practice have been rainfall
and Normalized Difference Vegetation Index (NDVI) products, but in monopsony
environments where a single buyer has accurate yield information on producers
(such as cotton) it has also been possible to build indexes based on area yields or
aggregate livestock mortality. The promise of index insurance is that it becomes possible
to offer very small insurance contracts profitably because there is no need to loss-adjust
each contract separately, and that insurance can be provided with no adverse selection
or moral hazard because the object against which insurance is written is beyond the
18
control of the insured party. Given that much theoretical literature in economics suggests
that it is precisely the correlated weather shock that informal risk pooling should be least
effective in dealing with (Townsend 1994), there is a strong argument from first principles
that index insurance addresses a risk that is not otherwise easy for rural communities to
diversify.
Unfortunately, the experience of a large number of index insurance pilots that have
attempted to introduce these projects into agricultural markets over recent years has
been quite negative (Cole et al. 2013). Not a single study has found robust private
demand for index insurance at market prices, and of the many pilots conducted, not a
one has moved to scale as a sustainable private market product (Jameel Poverty Action
Lab 2016). Several studies have found that interlinking an insurance product with credit,
far from untying the Gordian knot at the intersection of credit and risk, actually leads to
a decrease in the demand for credit (relative to a standalone credit product; Giné and
Yang 2009; Banerjee, Duflo, and Hornbeck 2014). A meta-demand curve estimated
across multiple randomized controlled trials in South Asia and Africa suggests that at
market prices (120% of actuarially fair price), market demand is below 10% of the
potential market, and that in order to push demand over 50% of the market, it would be
necessary to subsidize index insurance to cost only about 40% of the actuarially fair price
(Figure 7). Hence, there seems to be little prospect that micro-insurance will become a
large-scale, private-sector means of protecting the agricultural system from weather risk.
Figure 7: Demand Curves for Index Insurance
Reprinted from Jameel Poverty Action Lab, Make It Rain, 2016, Policy Bulletin, Massachusetts Institute of Technology.
The core problem is a lack of demand. Why do farmers not want to pay for this product
that theory suggests should be so attractive? One obvious reason is that index insurance
is only partial insurance; the imperfect correlation between the index and the actual yields
experienced on the farm exposes farmers to “basis risk” (Barnett, Barrett, and Skees
2008). In the context of weather index insurance, basis risk is defined as the difference
between the variation in actual yields at the farm level (the quantity the farmer wishes to
insure) and the variation defined by the index (the quantity actually insured), and may
arise either because yields are imperfectly described by the quantity defined by the index
(such as rainfall), or because the index measurement is taken
in a place other than the farm (hence geographic variation in weather contributes to basis
risk). A large stream of literature suggests that the demand response to partial insurance
is substantially more complex than perfect insurance (Eekhoudt, Gollier, and Schlesinger
1996), and that the presence of basis risk can introduce non-monotonicity into the way
19
that insurance demand changes with risk aversion (Clarke 2016). More recent research
has focused on behavioral reasons for stagnating demand; the most influential of these
has been the idea of “ambiguity aversion,” whereby individuals may dislike paying for
products when they do not perfectly understand the distribution of shocks and payouts
(Gilboa and Schmeidler 1989; Bryan 2010). Additional behavioral explanations are the
probabilistic nature of basis risk driving an overweighting of the probability of contract
failure (McIntosh, Povell, and Sadoulet, forthcoming) as well as the failure to correctly
reduce the compound lotteries inherent in the failure of index insurance products (Elabed
and Carter 2015).
In terms of micro-insurance, many of the open avenues for private market product
development lie squarely within the FinTech space. First, there is the use of advanced
technology to improve the indexes themselves: by pushing down their spatial and
temporal level of granularity, it may be possible to squeeze basis risk out of the index.
Next, since the credibility and timeliness of the insurance payout seem to be major
obstacles, there are opportunities to use scanning technology and the immediacy of
mobile money to structure novel insurance products (Prashad et al. 2014). For example,
the Kilimo Salama product fielded by the Syngenta Foundation in Kenya links the
purchase of a bag of fertilizer to a mobile phone number through a scratch card, and
makes an automated payment to the farmer via mobile money if the index pays out, with
the cost of the premium embedded in the price of fertilizer.
4.3 FinTech and Risk Sharing in the Agricultural System

Technological innovation is potentially directly conducive to the shifting of risks within the
agricultural system simply because it enhances our ability to objectively quantify variation
in a fine-grained way. Forward and futures markets play a well-established role in
allowing producers and intermediaries to hedge price risk in agriculture. Remotely
operated weather monitoring stations, high-resolution satellite-based monitoring of
yields, and the use of mobile phones to capture and transmit sentinel agricultural price
information all provide novel early-warning systems for weather-driven shocks.
Classification of drought events within rich data on weather patterns is the
type of problem that big data learning algorithms excel at solving, so there is real reason
to think that our ability to build highly predictive indexes is improving rapidly.
Unfortunately, in practice, new sources of measurement present the statistical
conundrum that to understand their risk properties we need long historical time series,
so building a re-insurable index off of a completely novel source of data can prove
challenging. Innovation in the financial field, in this sense, is likely to lag innovation in the
technical problem of building indexes.
Another conundrum that has been encountered in the development of index insurance
products is that, while changing climate only makes the need to insure against correlated
shocks more pressing in a welfare sense, it also makes weather risk harder to price. This
problem has manifested itself through global re-insurance companies, which understand
well that the climate in places like the Horn of Africa and the Pacific coast of South
America is changing and are inclined to subject re-insurance in these locations to an
additional risk-loading cost. The ability to pass weather risk onto the global re-insurance
system is largely one that is priced at private market rates, and such insurance becomes
more expensive to acquire as the distribution of global weather risk becomes more
unpredictable.
The lack of demand for index insurance may appear less surprising if we examine
the way that agricultural risk markets work in developed countries. While future and
20
forward contracts are commonly used to handle price risk, few completely private
agricultural insurance systems exist to serve farmers in OECD countries. Typically, even
in countries such as the US that are typically committed to market mechanisms, the
government plays a significant role in protecting farmers from weather risk, and these
programs are usually heavily subsidized. Other crop insurance systems such as India’s
National Agricultural Insurance Scheme have reached large scale via a government
mandate: in India, any farmer receiving a private-sector agricultural loan is required to
insure it through NAIS.
Ultimately, in many contexts it is inevitable that the state will bear many of the costs
of agricultural risk protection, whether through the presence of ex-ante commitments
(through safety net programs) or through disaster relief. Some evidence suggests that
these public safety nets prove an obstacle to the development of private insurance
markets (Duru 2016). If this is to be the case, and if demand at the private level is weak
anyway, then a natural alternative arrangement would be to use index insurance to
protect governments and banks that bear most of the current exposure to this risk. In this
sense, given that these entities are likely to be highly liquid, index insurance can
be thought of as a loan loss reserve fund (banking) or “rainy day” fund (government)
through which budgets can be smoothed and systematic tail risks managed. Examples
of this approach include Mexico’s CADENA program (de Janvry, Ritchie, and Sadoulet
2016), which uses international re-insurance to underwrite a variety of area-based yield
and index insurance programs that state and local governments can decide to purchase
for their citizens. In principle, this is an attractive way to handle the systemic risks
inherent to agricultural credit systems without the patronage and agency problems that
emerge if the government handles tail risks by making budget constraints soft.
A closely related development is the issuance of “catastrophe bonds” that make
payments in the event a pre-specified disaster occurs. These instruments, widely used
in the US, are now being offered in Singapore, and the actively developing market for
insurance-linked securities may help to lower the price of transferring risks, as well as
developing the bond market in an Asian region that still relies heavily on bank finance
(Ralph 2017).
5. COMPLEMENTARY INFRASTRUCTURE
AND THE ENABLING ENVIRONMENT
5.1 Mobile Money
Mobile phones and mobile money can provide the infrastructural backbone to
provide FinTech services to otherwise marginalized populations. Mobile phones give
individuals access to information about prices and business opportunities, thus improving
spatial arbitrage (Jensen 2007). Mobile technology also provides a novel impetus to
interact with the written word and so can provide a platform to promote literacy (Aker,
Ksoll, and Lybbert 2012). The advent of mobile money further extends the possibilities
of a mobile phone, providing potential for savings (Mbiti and Weil 2011), and eases the
sending of remittances and risk pooling within social networks (Jack and Suri 2014).
21
The PRC is a standout example of the ways in which mobile payment systems can serve
as the backbone for a rapid expansion of a variety of FinTech services. Due to the
explosion in e-commerce without a well-developed pre-existing credit card payment
infrastructure, the PRC has become the premiere example of how e-commerce and
mobile payments systems can reinforce each other. Trading platforms such as Alibaba
and Tencent have given rise to a number of different mobile payment systems (Alipay
and TenPay together have over 90% of the market share of Chinese mobile payments,
although numerous smaller competitors exist). Mobile payments account for almost
three-quarters of all online purchases in the country, and are expected to continue to
grow by more than 60% in the upcoming year (Chen 2017).
5.2 Financial Literacy and FinTech

FinTech products open up exciting new possibilities, but can also present unfamiliar
options to households and require that people be willing to undertake substantial learning
costs. Innovations provide households with new ways to organize and discipline their
financial lives, but can also open up new danger of over-indebtedness. This tension
motivates the importance of financial literacy in the context of FinTech. Over the longer
term, financial literacy plays a strong role in allowing households to solve life cycle
problems (Lusardi and Mitchell 2014), an issue that will be increasingly important in Asia,
a region that combines an aging population with retirees who are being asked to take on
a greater responsibility for retirement savings (Yoshino, Morgan, and Wignaraja 2015).
Financial literacy has a gender dimension as well, with women having lower levels of
financial knowledge in many countries (Xu and Zia 2012). Therefore, financial literacy is
a key part of generating true inclusion with FinTech products. The Bayesian justification
for FinTech training programs is strong, since a shifting landscape provides individuals
with many new choices about which they have not yet had a chance to learn.
While financial literacy is highly correlated with good economic outcomes, there is reason
to think that the decision to acquire financial skills is correlated with otherwise positive
financial behavior (such as patience; see Meier and Sprenger 2013). Programs
attempting to increase financial literacy in an experimental way have generated mixed
results, including in some cases substantial improvements in savings and decreases
in consumption of temptation goods (Calderone et al. 2018). Simplicity appears to
be strongly desirable in training programs tailored to the poorest groups, and “rule of
thumb” interventions were been found to be more effective than more complex trainings
in the Dominican Republic (Drexler, Fischer, and Schoar 2014). Also, trainings that are
tailored by gender can help improve the substantial knowledge gaps that remain. A core
point in the welfare analysis of FinTech innovations is that the marginal welfare gain from
the use of even relatively primitive communication services may exceed the gain from
the more sophisticated services used in more developed countries because they
represent such an improvement relative to what was previously available. FinTech
investments should be motivated by closing the gaps that prevent marginalized
populations from taking full advantage of tools to which they otherwise lack access.
5.3 National and Biometric Identification Systems

Any attempt to build a mass-scale transfer program or credit bureau in a developing
country will quickly run up against the problem of how to accurately identify individuals
in the population. In countries without robust national identification systems, this turns
out to be a challenging task. To move toward full convertibility of mobile money into
normal bank savings, countries must satisfy the Know Your Customer (KYC) laws,
established by the international system to prevent money laundering. KYC laws place
22
strict requirements on the types of identification that can be used to create accounts that
permit resources to move from mobile wallets into the banking system, and hence
identification systems are key to the ability to fully integrate FinTech solutions into the
broader economy (ADB 2017).
This issue can also drive substantial leakage from public transfer systems (which may
become rife with ghost beneficiaries), and is prohibitive to the construction of robust
national credit reporting architecture. This is an area that has seen substantial recent
policy work. Countries that lack any clear national ID system have been striving to
establish them, and some countries such as Indonesia have been striving to move
toward biometric identification (that country’s e-KTP identification system covers 86% of
the population and is now being used for digital payments in government transfer
programs). Countries with paper-based national ID cards may still see substantial
improvements in welfare by moving to the use of more sophisticated biometric
identification. Muralidharan, Niehaus, and Sukhtankar (2016) analyze the introduction of
smartcards into India’s massive National Rural Employment Guarantee system
(the largest public transfer system in the world), and find that these ID cards substantially
improve state capacity, making payments timelier, improving targeting, and combatting
corruption in the program. In a subsequent paper, the authors show that the impact of
the smartcards on improving the living standards of the poor was so strong as to increase
real private sector wages by 6%, driving up consumption in poor households by 13%
(Muralidharan, Niehaus, and Sukhtankar 2017).
Even in countries without national ID systems, biometric identification can improve credit
market outcomes. Giné, Goldberg, and Yang (2012) implement a fingerprinting system
in a single agricultural lending bank in Malawi and find that the use of the technology
decreases repayment problems among borrowers who had high ex ante risk of default.
This decrease is accompanied by smaller loan sizes and greater concentration of loan
resources on business expenditures within this group for the treatment than for the
control, consistent with both moral hazard and adverse selection being at play. That
biometric technology used by one lender on farmers of a single crop (paprika) can alter
credit market outcomes should be cause for optimism; this suggests that it may not be
necessary to set up universal, real-time biometric credit bureaus in order to have this
technology lead to real improvements in credit market outcomes. Because biometric
identification systems can be KYC-compliant even in countries without strong national
ID systems, biometrics provide a way to leapfrog legacy paper-based systems and
created integrated digital financial services across platforms.
The SIM cards used in mobile phones are becoming the de facto form of personal
identification in many less developed economies, as mobile money leapfrogs legacy
government identification systems. This is a form of identity that bears resemblance to a
number of online marketplaces in more developed countries. When reputation hinges on
a form of identity that can be shed and restarted (by buying a new phone number, or
creating a new login to a trading platform such as eBay or Alibaba), we cannot expect
users to ever tolerate a reputation worse than starting over. This has implications for the
way that tenure on the system will be scored: when “new” users in a financial platform
are in steady state largely made up of past deadbeats re-entering the system, then we
expect a long-established reputation (even if checkered) to be strongly preferred over
someone with no history. Because of the many services that can be connected to mobile
money or other account identifiers, it may be the case that private sector-led modalities
for individual identification can function as an effective substitute in states that lack the
capacity to identify individuals themselves.
At the other end of the spectrum are efforts to link reputation across many different
domains that are currently underway in the PRC. Certainly, applying standard models of
23
the effects of asymmetric information in credit markets to a more general set of problems
(such as crime) indicates that this effort is likely to be effective at controlling specific
forms of social misbehavior. However, even setting aside the fundamental
civil liberty questions, the linking of reputation across domains has echoes to the
long-standing debate in agricultural economics over the role of interlinking in the
provision of financial services. In a model in which producers are competitive in all
markets but limited in their ability to offer products due to asymmetric information, the
interlinking of contracts enhances the ability to offer financial services and will result in
an improvement in consumer welfare. If, on the other hand, a single monopolistic
provider is able to interlink across multiple markets, that provider can then push the client
back to a less advantageous reservation utility and welfare will fall. Given
this ambiguity, the improvement of identification systems within one domain (such as
credit repayment) has a stronger welfare foundation than the linking of behavior
across domains.
5.4 Blockchain Technologies

Blockchain technology has been promoted as the panacea for a wide range of
technological problems in recent years. It may be reasonable to approach these
proposals with some skepticism in the context of developing-country agriculture, where
the use of the blockchain is as yet unproven, and many other institutions may be absent,
so the technical dimension of how to build databases is relatively less important. We
provide a brief technical background on blockchain technology and then discuss its
application to three FinTech problems: blockchain currency as a form of exchange, the
use of the ledger in credit bureaus and asset registries, and the uses of blockchain
technology in supply chains.
The blockchain is a structured peer-to-peer communication network that uses a specific
system of hashes, digital signatures, and timestamp servers to make reliable
transactions over a network, eliminating the necessity for an intermediary (Nakamoto
2008). The intuition behind this complex technology of is straightforward: to reduce
transactions costs and to allow two unconnected parties to trade in the absence of a
trusted intermediary. The blockchain makes use of cryptographic proofs to eliminate this
intermediary (typically a bank or other financial institution), allowing two parties to
transact directly.
In an example transaction, the receiver of a payment would receive the transfer along
with a digital signature that allows the recipient to observe in the hash of the transaction
the information of every node that was engaged in the chain. The receiver can verify the
signatures of the block to examine the authenticity of the payments, or the chain of
ownership of the money in that transaction, and then verify purchase if they are
satisfactory. This distributed ledger provides a mechanism that is at once highly secure
and easily accessible in a way that many proprietary commercial data networks are not.
While much of the emphasis on the blockchain to date has centered on the role of
anonymity (as in crypto-currencies such as BitCoin), if the blockchain is to take off in the
context of agricultural FinTech, it is likely to be its accessibility that proves its most
attractive feature.
24
Given that digital currency is by far the best established use of the blockchain, it is natural
to ask whether this type of currency could serve as a medium of exchange for agricultural
trade. Within a country, it would appear that the only justification for using cryptocurrency
for exchange would be to evade law enforcement or paying taxes, so is hardly to be
recommended as a matter of national policy. For international exchange, it is not hard to
imagine digital currency providing an attractive medium in which to conduct business,
since with relatively low barriers on both sides of the transaction, it provides a secure
way to transact and would be amenable to a variety of “buyer verifies quality” types of
contractual arrangements that are important in agricultural trade. In this sense, the
marginal value of the blockchain perhaps appears smallest in countries that already have
well-functioning legacy systems serving farmers, and the greatest potential gain among
those whom current systems do not serve well. However, the enormous gyrations in the
value of existing cryptocurrencies such as BitCoin make them an extremely unattractive
medium of transaction for intermediaries who already hold substantial price risk in
agriculture and will not want to hold this form of currency risk as well. Hence, the
maturation and stabilization of cryptocurrency markets may be a precondition for their
more widespread use as a standard medium of exchange.
5.3.1 Blockchain Technology in Credit Reporting

The problem of building asset and credit registries, as has been noted by many technical
experts, appears to be an ideal application of distributed ledger technology. A blockchain
registry network would have the same features as a common distributed ledger network
technology, but the blockchain feature could also be used to ensure that the historical
record of ownership as well as the transfer of assets, lands, automobiles, homes, and
land was current and unambiguous. Although the idea of a transparent registry that
records each link in the credit process is promising, several challenges must be taken
into consideration, including privacy and the incentives for those institutions that currently
possess credit information to share it.
Underlining the challenges to building such a system at scale, the US title insurance
industry took in $18 billion in 2006 (Woolley 2006). This provides evidence of the costs
imposed by uncertainty over legal title in asset registries, even in a developed country
that has a well-functioning integrated reporting system. Most asset registry systems are
proprietary and overseen by a single commercial entity, which may well be necessary
given the costs of establishing and operating a standard reporting infrastructure.
The technological capacity to use a public distributed ledger for asset registries might
result in a restructuring of the industrial organization of this sector, and lead to lower
costs on the margin. As the barriers to using such a system fall, it is precisely in poorer
clients with less valuable assets that we would expect to see dramatic changes in
financial access.
5.3.2 Blockchain Ledgers in Agricultural Supply Chains

A key feature of the blockchain—the sequential validity of transactions—is of the highest
importance in terms of traceability in agricultural supply chains. Such technology not only
can trace back the origin of contaminated food at a much faster pace than conventional
methods (del Castillo 2017; Popper and Lohr 2017) but may
be the ideal way of structuring the traceability infrastructure that is necessary for the
“farm to fork” model of high-value agricultural production (Kim and Laskowski 2018). The
blockchain enables transactions in which the buyer places funds for the transaction in a
secure escrow, the seller verifies the funds are present and undertakes trade, and then
the buyer removes the escrow conditional on being satisfied with
the delivery of the contract. So much of agricultural trade involves fast-moving spot
25
trades on commodities whose quality is difficult to verify and in which trust between
buyers and sellers may be low. The attempt to bring greater depth to these markets with
ICT-driven trading platforms often runs into trouble over trust and verification
when trading changes from being personalized to conducted with strangers. Hence, seen
simply as a highly accessible and credible environment in which to conduct exchange
with multiple stages of transaction and verification, public distributed ledgers may be
useful.
Several large-scale pilots of this idea are underway. IBM and Walmart have created a
trial blockchain system for supply chain tracing to allow for better oversight of food safety,
and the Chinese giant Alibaba is engaged with PricewaterhouseCoopers in a blockchain
trial to monitor food imports from Australia and New Zealand. Indonesian seafood
exports are being tracked by Provenance with a blockchain system (Kim and Laskowski
2018). Nonetheless, serious obstacles exist to the full-scale implementation of
blockchain traceability: first, these systems require a degree of supply chain
sophistication and origin custody tracing that is unusual in developing country agriculture.
Proper origin labelling, RFID tags, secure packaging, and integrated intermediary
networks may be more important obstacles than the database technology used. Second,
profit margins at each stage in the agricultural supply chain may be quite small, diffusing
the incentives to invest in systemic solutions even if overall benefits would be large.
Finally, blockchain systems are not typically interoperable, meaning that these systems
may struggle to achieve in reality the scalability that their technology provides in theory.
Nonetheless, this appears to be the most attractive area for the use of the blockchain in
agriculture, with a number of advanced pilots already underway in Asia.
6. POLICY RECOMMENDATIONS
Asia is a region with tremendous internal heterogeneity in the macroeconomic
importance of agriculture as well as the level of financial development. It is therefore
important to tailor policy recommendations in a context-specific manner. To this end, we
propose three distinct ways of thinking about the role that FinTech can play in promoting
agricultural development in Asia.
6.1 FinTech Policies to Promote High-Value Agriculture

For agriculture to play a constructive role in overall economic growth, the sector must not
simply improve overall yields but must achieve the quality control and phytosanitary
certifications that will be required to serve demanding urban consumers and to
feed into international markets. These are very rarely improvements that can be made
at the level of the individual producer, but must instead involve entire value chains
that become sufficiently organized to establish “farm to fork” custody certification. In this
sense, the key actors may not be farmers but rather processors, intermediaries in the
value chain, and certification entities that can allow quality premiums to be captured in
final output markets and translated all the way back up the supply chain to the original
producers.
26
Core policy innovations for this objective include:

• Permitting large-scale collateralized credit to agriculture, which will typically
require creation of asset registries that permit this lending to take place.
Emphasis on financial deepening among input suppliers to permit them to offer
capital on lease.
• Well-organized risk markets that can protect the solvency of financial institutions
making large-scale investments in agricultural capacity in case of shocks to
output. Policies: government and finance institutions use re-insurance markets
for tail risks (“risk layering”), catastrophe bonds, explicit underwriting of extreme
risk to permit private sector lenders to deepen exposure to agriculture.
• Conceptualizing large-scale financing to agriculture as “value chain financing”;
organized, systematic investments across a set of actors to create vertically
integrated supply chains feeding high-value urban markets (supermarkets) and
export markets.
• Establishing and financing certification entities (phytosanitary, organic, fair trade,
etc.) to permit value added in agriculture.
• Contracting innovations to try to push the quality/price gradient that exists in
international markets all the way down to original producers; organized
intermediaries who can segregate output of different qualities throughout the
supply chain.
6.2 Financial Inclusion for the Agriculture Sector

A distinct set of policies is apparent if we think of agriculture as a sector that supports
the majority of the world’s working poor. Here, financial inclusion is the key concept, in
that we must find ways of organizing agricultural production such that they benefit the
smallholder farmer as much as possible. In this respect FinTech provides a particularly
exciting set of opportunities in that recent advancements in technology may provide
enormous benefits to those who have been excluded from legacy systems.
• Shifting agricultural payment systems, financial services from banks, G2P
payments, and input/output suppliers to function using mobile money
infrastructure rather than cash.
• Building credit bureau infrastructure that permits better targeting and
performance of uncollateralized credit markets, giving smallholder producers
without collateral access to credit.
• A coordinated regulatory push on FIs to require information sharing, particularly
on smallholder loans. Emphasis should be on the “ladder of credit,” requiring
sharing between small (microfinance) and large (private commercial) lending
institutions so that the entrepreneurial poor can use credit reputation as a tool for
economic mobility.
• Financial literacy programs to ensure that individuals are aware of new FinTech
options, able to interact with systems and price compare, and using tools to plan
for the future. Gender is an important part of financial literacy and programs
tailored to specific demographics are more effective.
27
• ICT-driven agricultural extension services (such as Olam, Precision Agriculture

for Development, etc.) that allow smallholder farmers to improve quality, improve
profits, and feed output into higher-value supply chains.
• ICT-driven trading platforms to give smallholders access to price information and
deeper markets. Shallow markets inhibit smallholders from investing in
productivity enhancement because prices fall when output increases.
• Land titling and land registries to permit farmers to collateralize land into credit.
Land-collateralized credit may need to be paired with explicit credit insurance to
allow borrowers to avoid “risk rationing” on the demand side. Providing
agricultural insurance to banks is not sufficient if the banks do not extend this
conditionality explicitly in their loans.
6.3 Cross-Cutting FinTech investments

Policymakers at the regional level (such as ADB) can achieve win-wins in the promotion
of agriculture by looking for sector investments that enhance both the productivity and
inclusivity of the sector. In particular, to the extent that technological innovations can be
developed in one context and scaled to others, regional development banks should be
looking for “best case” scenarios in which to pilot and develop scalable technologies that
can then be reproduced regionally.
• Regional regulatory harmonization to permit efficient international trade. Key
areas:
o International harmonization of mobile money and digital payment systems to
enhance financial exchange.
o Harmonization of certification requirements for quality grading, organic
designation, food safety, etc.
• Regional approach to risk management so as to allow pooling over as large an
area as possible without paying loading costs of international commercial re-
insurance. Combine publicly administered risk pools with index-style triggers to
avoid political-economy pitfalls over payouts.
• Centralized investments in technology backbone for agriculture:
o Blockchain-based supply chain architecture. Having been developed in one
context this may prove to be highly scalable across countries.
o Technology to create robust national ID systems and convert them to
biometric identification if necessary.
28
REFERENCES
Aker, J. C., C. Ksoll, and T. J. Lybbert. 2012. Can Mobile Phones Improve Learning?
Evidence from a Field Experiment in Niger. American Economic Journal:
Applied Economics 4(4): 94–120.
Asian Development Bank (ADB) and Oliver Wyman. 2017. Accelerating Financial
Inclusion in South-East Asia with Digital Finance. https://think-asia.org/
bitstream/handle/11540/7500/financial-inclusion-se-asia.pdf?sequence=1,
accessed August 6 2018.
Bachas, P., P. Gertler, S. Higgins, and E. Seira. 2017. Banking on Trust: How Debit
Cards Enable the Poor to Save More. No. w23252. National Bureau of
Economic Research Working Paper.
Banerjee, A., E. Duflo, and R. Hornbeck. 2014. Bundling Health Insurance and
Microfinance in India: There Cannot Be Adverse Selection if There Is No
Demand. American Economic Review 104(5): 291–97.
Barnett, B. J., C. B. Barrett, and J. R. Skees. 2008. Poverty traps and index-based risk
transfer products. World Development 36(10): 1766–1785.
Besley, T., and S. Coate. 1995. Group Lending, Repayment Incentives, and Social
Collateral. Journal of Development Economics 46: 1–18.
Björkegren, D., and D. Grissen. 2017. Behavior Revealed in Mobile Phone Usage
Predicts Loan Repayment. arXiv preprint arXiv:1712.05840.
BNP Paribas. 2017. Microfinance Barometer 2018. https://group.bnpparibas/en/news/
microfinance-barometer-2017-global-trends-sector, accessed August 10 2018.
Boonperm, J., J. Haughton, and S. R. Khandker. 2013. Does the Village Fund Matter in
Thailand? Evaluating the Impact on Incomes and Spending. Journal of Asian
Economics 25: 3–16.
Brune, L., X. Giné, J. Goldberg, and D. Yang. 2016. Facilitating Savings for Agriculture:
Field Experimental Evidence from Malawi. Economic Development and Cultural
Change 64(2): 187–220.
Bryan, G. 2010. Ambiguity and Insurance. Unpublished manuscript.
Burke, M., L. F. Bergquist, and E. Miguel. 2018. Sell Low and Buy High: Arbitrage and
Local Price Effects in Kenyan Markets. No. w24476. National Bureau of
Economic Research.
Calderone, M., N. Fiala, F. Mulaj, S. Sadhu, and L. Sarr. 2018. Financial Education
and Savings Behavior: Evidence from a Randomized Experiment among Low-
Income Clients of Branchless Banking in India. Economic Development and
Cultural Change, 66(4).
Callen, M., S. De Mel, C. McIntosh, and C. Woodruff. 2014. What are the Headwaters
of Formal Savings? Experimental Evidence from Sri Lanka. No. w20736.
National Bureau of Economic Research.
Carter, M. R., L. Cheng, and A. Sarri. 2016. Where and How Index Insurance Can
Boost the Adoption of Improved Agricultural Technologies. Journal of
Development Economics 118: 59–71.
Chen, Wingyi. 2017. China Mobile Payment Report, 2017. https://walkthechat.com/
china-mobile-payment-report-2017/ (Accessed 2 August 2018.).
29
Clarke, Daniel J. 2016. A Theory of Rational Demand for Index Insurance. American
Economic Journal: Microeconomics 8(1): 283–306.
Cole, S., X. Giné, J. Tobacman, P. Topalova, R. Townsend, and J. Vickery. 2013.
Barriers to household risk management: Evidence from India. American
Economic Journal: Applied Economics, 5(1), 104–35.
Crépon, B., F. Devoto, E. Duflo, and W. Parienté. 2015. Estimating the Impact of
Microcredit on Those Who Take It Up: Evidence from a Randomized
Experiment in Morocco. American Economic Journal: Applied Economics
7(1): 123–50.
De Janvry, A., E. R. Ritchie, and E. Sadoulet. 2016. Weather Index Insurance and
Shock Coping: Evidence from Mexico’s CADENA Program. World Bank
Working Paper.
De Janvry, A., and E. Sadoulet. 2009. Agricultural Growth and Poverty Reduction:
Additional Evidence. World Bank Research Observer 25(1): 1–20.
De Janvry, A., C. McIntosh, and E. Sadoulet. 2010. The supply-and demand-side
impacts of credit market information. Journal of development Economics,
93(2), 173–188.
De la Campa, A. A. 2011. Increasing Access to Credit through Reforming Secured
Transactions in the MENA Region. World Bank Working Paper.
Castillo, Michael del. 2017. “Walmart, Kroger & Nestle Team with IBM Blockchain to
Fight Food Poisoning – CoinDesk.” 2017. https://www.coindesk.com/walmart-
kroger-nestle-team-with-ibm-blockchain-to-fight-food-poisoning/. (accessed
on July 18 2018).
Diamond, D. W. 1984. Financial Intermediation and Delegated Monitoring. Review of
Economic Studies 51(3): 393–414.
Dorward, A., et al. 2009. Hanging In, Stepping Up and Stepping Out: Livelihood
Aspirations and Strategies of the Poor. Development in Practice 19(2): 240–47.
Drexler, A., G. Fischer, and A. Schoar. 2014. Keeping It Simple: Financial Literacy and
Rules of Thumb. American Economic Journal: Applied Economics 6(2): 1–31.
Dupas, P., D. Karlan, J. Robinson, and D. Ubfal. 2018. Banking the Unbanked?
Evidence from three countries. American Economic Journal: Applied
Economics, 10(2), 257–97.
Duru, M. J. 2016. Too Certain to Invest? Public Safety Nets and Insurance Markets in
Ethiopia. World Development 78: 37–51.
Eeckhoudt, L., C. Gollier, and H. Schlesinger. 1996. Changes in Background Risk and
Risk-Taking Behavior. Econometrica 64(3): 683–689.
Elabed, G., and M. Carter. 2015. Basis Risk and Compound-Risk Aversion: Evidence
from a WTP Experiment in Mali. Journal of Economic Behavior and
Organization 118(C): 150–66.
Ghatak, M., and T. Guinnane. 1999. The Economics of Lending with Joint Liability:
Theory and Practice. Journal of Development Economics 60: 195–228.
Gilboa, I., and D. Schmeidler. 1989. Maxmin Expected Utility with Non-unique Prior.
Journal of Mathematical Economics 18(2): 141–153.
30
Giné, X., J. Goldberg, and D. Yang. 2012. Credit Market Consequences of Improved
Personal Identification: Field Experimental Evidence from Malawi. American
Economic Review 102(6): 2923–54.
Giné, X., and D. Yang. 2009. Insurance, Credit, and Technology Adoption: Field
Experiment Evidence from Malawi. Journal of Development Economics
89: 1–11.
Jack, W., M. Kremer, J. De Laat, and T. Suri. 2016. Borrowing Requirements, Credit
Access, and Adverse Selection: Evidence from Kenya. No. w22686. National
Bureau of Economic Research Working Paper.
Jack, W., and T. Suri. 2014. Risk Sharing and Transactions Costs: Evidence from
Kenya's Mobile Money Revolution. American Economic Review 104(1):
183–223.
Jameel Poverty Action Lab. 2016. Make It Rain. Policy Bulletin. Cambridge, MA:
Massachusetts Institute of Technology.
Jappelli, T., and Marco Pagano. 1999. Information Sharing, Lending and Defaults:
Cross-Country Evidence. Journal of Banking and Finance 26(10): 2017–45.
Jensen, R. 2007. The Digital Provide: Information (Technology), Market Performance,
and Welfare in the South Indian Fisheries Sector. Quarterly Journal of
Economics 122(3): 879–924.
Kim, H. M., and M. Laskowski. 2018. Toward an Ontology‐Driven Blockchain Design
for Supply‐Chain Provenance. Intelligent Systems in Accounting, Finance and
Management 25(1): 18–27.
Lusardi, A., and O. S. Mitchell. 2014. The Economic Importance of Financial Literacy:
Theory and Evidence. Journal of Economic Literature 52(1): 5–44.
Manova, K., and Z. Yu. 2013. Firms and Credit Constraints along the Value Chain:
Processing Trade in China. Unpublished manuscript, Stanford University.
Mbiti, I., and D. N. Weil. 2011. Mobile Banking: The Impact of M-Pesa in Kenya.
No. w17129. National Bureau of Economic Research.
McIntosh, C., F. Povell, and E. Sadoulet (Forthcoming). Utility, Risk, and Demand for
Incomplete Insurance: Lab Experiments with Guatemalan Cooperatives.
Economic Journal.
McIntosh, C., and B. Wydick. 2005. Competition and Microfinance. Journal of
Development Economics 78: 271–98.
Meier, S., and C. D. Sprenger. 2013. Discounting Financial Literacy: Time Preferences
and Participation in Financial Education Programs. Journal of Economic
Behavior & Organization 95: 159–74.
Miller, C. 2008. A Baker's Dozen Lessons of Value Chain Financing in Agriculture.
Enterprise Development and Microfinance 19(4): 273–80.
Miranda, M. J., F. M. Mulangu, and F. H. Kemeze. 2017. Warehouse Receipt Financing
for Smallholders in Developing Countries: Short on Logic, Long on Imagination.
Working Paper.
Muralidharan, K., P. Niehaus, and S. Sukhtankar. 2016. Building State Capacity:
Evidence from Biometric Smartcards in India. American Economic Review
106(10): 2895–929.
31
Muralidharan, K., P. Niehaus, and S. Sukhtankar. 2017. General Equilibrium Effects of

(Improving) Public Employment Programs: Experimental Evidence from India.
No. w23838. National Bureau of Economic Research.
Nakamoto, S. 2008. Bitcoin: A Peer-to-Peer Electronic Cash System. www.bitcoin.org,
https://doi.org/10.1007/s10838-008-9062-0 (accessed on 19 July 2018).
Padilla, J., and M. Pagano. 1997. Endogenous Communication among Lenders and
Entrepreneurial Incentives. Review of Financial Studies 10(1): 205–36.
Popper, N., and S. Loher. 2017. “Blockchain: A Better Way to Track Pork Chops,
Bonds, Bad Peanut Butter? – The New York Times.” 2017.
https://www.nytimes.com/2017/03/04/business/dealbook/blockchain-ibm-
bitcoin.html. (accessed on July 19 2018).
Prashad, P., J. Leach, A. Dalal, and D. Saunders. 2014. Mobile Phones and
Microinsurance. Enterprise Development and Microfinance 25(1): 72–86.
Prina, S. 2015. Banking the poor via savings accounts: Evidence from a Field
Experiment. Journal of Development Economics 115(7): 16–31.
Ralph, O. 2017. Singapore Seeks to Muscle in on Market for Catastrophe Bonds.
Financial Times. https://www.ft.com/content/55229b4c-befd-11e7-b8a3-
38a6e068f464 (accessed 1 August 2018).
Rutherford, S. 2000. The Poor and Their Money. New York: Oxford University Press.
Samphantharak, K., and R. M. Townsend. 2018. Risk and Return in Village
Economies. American Economic Journal: Microeconomics 10(1): 1–40.
Schaner, S. 2017. The Persistent Power of Behavioral Change: Long-Run Impacts of
Schaner, S. (2018). The persistent power of behavioral change: Long-run
impacts of temporary savings subsidies for the poor. American Economic
Journal: Applied Economics, 10(3), 67–100.
Shalendra, M. S., E. Haque, and V. Anu Peter. 2016. Issues Limiting the Progress in
Negotiable Warehouse Receipt (NWR) Financing in India. Agricultural
Economics Research Review 29(1). 53–60.
Stiglitz, J., and A. Weiss. 1981. Credit Rationing in Markets with Imperfect Information.
American Economic Review 71(3): 393–410.
Terada, Y., and P. Vandenberg. 2014. Thailand’s State-Led Approach to Financial
Inclusion. Financial Inclusion in Asia 89. 89–110.
Townsend, R. M. 1994. Risk and Insurance in Village India. Econometrica: Journal of
the Econometric Society. May 1. 539–91.
Vercammen, J. 1995. Credit Bureaus and Sustainable Reputation Effects in Credit
Markets. Economica 62(248): 461–78.
Woolley, S. 2006. Real Estate's Richest Racket-Title insurance firms rake in USD18
billion a year for a product that is outdated, largely unneeded-and protected
by law. Forbes, 148.
Xu, L., and B. Zia. 2012. Financial Literacy around the World: An Overview of the
Evidence with Practical Suggestions for the Way Forward. World Bank
Working Paper.
Yoshino, N., P. Morgan, and G. Wignaraja. 2015. Financial Education in Asia:
Assessment and Recommendations. Working Paper.
32

Adbi wp872

Uploaded by

Copyright:

Available Formats

Adbi wp872

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Adbi wp872

Uploaded by

Copyright:

Available Formats

McIntosh, Craig; Mansini, Caio Scuarcialupi

ADBI Working Paper, No. 872

Provided in Cooperation with:

This Version is available at:

Standard-Nutzungsbedingungen: Terms of use:

THE USE OF FINANCIAL TECHNOLOGY

Craig McIntosh and

Asian Development Bank Institute

In this report, “$” refers to United States dollars.

Please contact the authors for information about this paper.

Email: ctmcintosh@ucsd.edu, cmansini@ucsd.edu

Asian Development Bank Institute

© 2018 Asian Development Bank Institute

Keywords: financial technology, agriculture, credit, risk, financial literacy

JEL Classification: Q16, O16, O33

2. THE CONTEXT ........................................................................................................... 3

3. CREDIT AND SAVINGS PRODUCTS ...................................................................... 13

3.1 Microfinance .................................................................................................. 14

4. INSURANCE AND RISK ........................................................................................... 18

4.1 Risk in the Agricultural Finance System ........................................................ 18

5. COMPLEMENTARY INFRASTRUCTURE AND THE ENABLING ENVIRONMENT 21

5.1 Mobile Money ................................................................................................ 21

6. POLICY RECOMMENDATIONS ............................................................................... 26

6.1 FinTech Policies to Promote High-Value Agriculture..................................... 26

Table 1: The Role of Agriculture in ADB Members, 2000–2016

Figure 1: Evolution of Rural/Urban Population

Source: World Indicators 2018.

Cereal Yields, by Country and Year

1970 1980 1990 2000 2010 2017

Source: FAO Stat.

Table 2: Changes in Welfare, 2000–2017

Figure 3: Evolution of National Poverty Rates, ADB Members

Evolution of National Poverty Rates: Rural and Urban Population

Rural Population Urban Population

Source: World Development Indicators, 2018.

Table 3: Financial Depth

ADB member countries with population over 1 million.

Table 4: Presence of Financial Infrastructure

Source: World Development Indicators Database 2018.

Figure 4: Credit Bureau Coverage and GDP, 2017

Figure 5: Use of Financial Services in Rural Areas

Use of Financial Services within Rural Areas, 2017

Borrowed from an FI Borrowed to Start Business

Figure 6: Payments in Cash for Agricultural Products

Reprinted from World Development Indicators, Global FinDex Database:

3. CREDIT AND SAVINGS PRODUCTS

3.2 Credit Reporting

(males). Counterbalancing this exodus of troubled pre-existing borrowers was a huge

3.3 Asset Registries and Warehouse Receipt Systems

3.4 Savings Products

(Rutherford 2000). Recent experimental literature on micro-savings products has

Micro-savings products appear to have strong justification in terms of increased welfare

4. INSURANCE AND RISK

Figure 7: Demand Curves for Index Insurance

4.3 FinTech and Risk Sharing in the Agricultural System

5.2 Financial Literacy and FinTech

5.3 National and Biometric Identification Systems

5.4 Blockchain Technologies

5.3.1 Blockchain Technology in Credit Reporting