XI ENCONTRO DA ABCP

Curitiba, 31 de julho – 3 de agosto de 2018

Área Temática: Política Internacional

MIGRATION AND HUMAN DEVELOPMENT IN LATIN AMERICA.

THE LONGITUDINAL EFFECTS OF LOW-SKILLED AND HIGH-SKILLED EMIGRATION IN

THE SENDING COUNTRIES

Ana Magdalena Figueroa

University of Sao Paulo

1
 Abstract

This research analyzes the effect of low skilled and high skilled migration on the human
development of the sending countries. This paper is not concerned with why migrants choose to
move. Instead, the concern here is to know what happens to the human development of the
sending countries when they actually move out of their home countries.

This work is based on the New Economics of Labor Migration theory and uses Latin
America as the sample of sending countries. The data are analyzed through Feasible Generalized
Least Squares and Panel Corrected Standard Errors. 5-year-averages are employed in order to
calculate the long-run effect of migration on human development. The period covered in this
research goes from 1970 to 2015. The results show mixed short-term effects of migration on the
human development of the sending countries. However, in the long-run, the effects of migration
on human development are mostly positive.

Keywords: Latin America, United States, International Migration, Human Development, Sending
Countries

2
Introduction

Data shows that there exist vast differences in human development across and within
countries in Latin America, and migration has been found to be a powerful tool (sometimes the
only tool) through which many people can expand their choices and capabilities. The UNDP has
recognized the potential that migration has to effectively increase income, health and education
prospects, not only for the migrants, but also for their families left behind, and that “being able to
decide where to live is a key element of human freedom” (UNDP, 2009. p.1). This research
adopts Amartya Sen’s approach to human development that considers human development as
a process of enlarging the choices that people have reason to value. (UNDP, 1990). In this sense,
this author believes that people have good reason to value their choice to migrate in order to help
enlarging the choices of their families back home. Hence, the importance of this research in the
Latin American region stems not only from its economic, political and social outcomes. It is
important for policy makers to know how emigration impacts the sending countries after losing
their people at the national, regional and international scale. Additionally, it is important to know
how low-skilled emigration differentiates from high-skilled emigration in terms of its effects on the
sending countries.

This work is based on the new economics of labor migration theory (NELM), which holds
that households are part of larger groups, like communities and countries. Because of this,
households related to migrants transmit the impacts of migration to other members of those
groups, and even households that are not related to migrants are affected by migration when
they interact with migrant related households. Therefore, it is very likely that the impacts of
migration can be found even outside of the households that send migrants and receive the
remittances directly, extending to the whole economy (TAYLOR, 1999; FIGUEROA, 2016).

Research questions

This research is driven by the interest of investigating how does the human development
of the sending countries vary after thousands of people leave every year. The product of
migration, remittances, is not the focus here. Instead, the focus are the migrants themselves and
how they can affect the sending countries after exercising their right to exit. In other words, how
does the movement of people out of the country per se affect the sending countries? This is a
shift from the common perspective where the effect of migration is studied through remittances
and is one of the main contributions of this research. Additionally, it is studied what are the short-
3
run and long-run effects of this movements of people. Another contribution of this research is to
investigate if high-skilled emigration affects the sending countries in the same was as low-skilled
emigration. If not, how do the sending countries react to these two different types of citizens
leaving? This research makes an exploration into the three areas that human development is
usually divided in order to find out how low-skilled and high-skilled emigration affects each area.
That is, it is studied how low-skilled and high-skilled emigration impact the freedom to access a
healthy life, education and a decent standard of life. In general, the leading hypothesis is that
migration is positively correlated with human development in the sending countries. Mixed effects
are expected in the long-run, while more positive effects are expected in the long-run.

Methodological strategy and data

The empirical analysis done in order to assess the impact of migration on the human
development of the sending countries comprises data for 21 Latin American countries whose
population is greater than 500,000 for a period from 1970 to 2015. The data are organized in a
time-series cross-section panel that presents 970 country-year observations. All the variables
used in the models were logged and analyzed for unit roots finding that most of them had to be
differenced once in order to make the time series stationary. In general, many of the time series
datasets are first-difference stationary.

With regard to the measurement of migration, data is always a challenge as the desirable
measurements are not available for all the countries and years in the sample. For this analysis,
migrants who are accepted as such in United States have been classified into low-skilled and
high-skilled workers. The most common method to define a high or low-skilled worker is by his or
her education level. However, this research failed to find these kind of data for all the countries
and all the years included in the analysis. Therefore, this work uses the second criterion:
occupation (PARSONS et al., 2015). These data were taken from the Yearbook of Immigration
Statistics of the Department of Homeland Security 1 of the United States that reports the
profession/occupation of each immigrant at the time of being accepted in the country. The criteria
for the classification of high-skilled and low-skilled were applied following the International
Standard Classification of Occupations (ISCO) 2 , which groups occupations based on a two-

1 Yearbook of Immigration Statistics. Department of Homeland Security. Available at

http://www.dhs.gov/yearbook-immigration-statistics
2 The International Standard Classification of Occupations was developed by the International Labor

Organization in 1957. It has gone through three revisions, being the last revision the one published in 2008.
4
dimensional concept of skill. That is, the skill level and skill specialization required to proficiently
perform the tasks of every occupation is used in the classification. Data for remittances comes
from the World Bank’s World Development Indicators3. Remittances as percentage of a given
country’s GDP is used in order to account for the significance of the received remittances for a
country given the magnitude of its economy.

In order to address human development, first we consider the area of a decent standard
of life, four indicators are used. The first one is a measure of civil liberties from Freedom House
4
. The second and third indicators are the labor market, understood as employment and wages.
For the employment variable, the Penn World Table 5 dataset was used, which measures the
employment ratio relative to the working age population. This variable was chosen instead of the
more commonly used measure of unemployment, due to the lack of the complete data for all the
countries and all the years in the sample. Data on wages comes from the Occupational Wages
around the World Database, which is derived from the International Labor Organization October
Inquiry Database6. Additionally, the national income has also been found to contribute, even
though it has been claimed that for this end, national income should be well distributed among
the population (ANAND et al 2000).

In order to address the area of access to a long and healthy life, three indicators are used:
life expectancy at birth, government expenditure in health, and births attended by health skilled
personnel. Data for life expectancy at birth comes from the World Development Indicators and
measures the average number of years that a newborn is expected to live if the patterns of
mortality at the time of their birth stay equal throughout their life. Data for government expenditure
in health are derived from the World Development Indicators and it measures the amount of
money that a government destines for health in a given country at a given year. The amount
presented is the government expenditure as percentage of the total government expenditure and
it includes recurrent and capital spending from government budgets, external borrowings and

The resolution concerning updating the ISCO is available at

http://www.ilo.org/public/english/bureau/stat/isco/docs/resol08.pdf Last accessed 2/25/2016.
3 World Development Indicators Database is available on
http://databank.worldbank.org/data/reports.aspx?source=world-development-indicators Last visited on
August 20, 2017.
4 Freedom House provides annual data on political rights and civil liberties around the world. It is available

on https://freedomhouse.org/report-types/freedom-world Last visited on August 20, 2017.

5 Feenstra, Robert C., Robert Inklaar and Marcel P. Timmer (2015), "The Next Generation of the Penn

World Table" forthcoming American Economic Review, Available at www.ggdc.net/pwt

6 The Occupational Wages around the World Database provides data on wages for 161 occupations in

171 countries. The database calibrates the data into a normalized wage rate for each occupation hourly
and monthly. It is available on http://www.nber.org/oww/ Last visited on August 20, 2017.G
5
grants, donations from international agencies and nongovernmental organizations, and social
health insurance funds7. Data for the births attended by skilled health personnel is derived from
the World Health Organization. The data represents the percentage of live births that are
attended by a person who is considered to be a health professional, such as a midwife, doctor or
nurse, and who has received education and training in the skills needed to manage a normal
pregnancy, childbirth and postnatal period. It excludes traditional birth attendants8.

In order to address the area of access to education, enrollment in primary, secondary and
tertiary education according to gender parity index is used in order to take into consideration the
education of girls vis-à-vis boys, as well as government expenditure in education. Data on the
enrollment in school enrollment in general and as gender parity index is derived from the World
Development Indicators. School enrollment in primary secondary and tertiary education is the
ratio of total enrollment in a given country and a given year. School enrollment as gender parity
index is the ratio of girls to boys enrolled at a primary, secondary, and tertiary level in public and
private schools. Data on government expenditure on education are derived from the World
Development Indicators and it measures the general government expenditure on education as
percentage of the GDP. It includes recurrent and capital spending from government budgets, as
well as transfers from international sources. 9 The following variables are controls that are
considered to affect the human development. Major episodes of political violence (MEPV), GDP
growth, population, net migration, IMF conditionality, trade, regime type, FDI, economic
sanctions, and homicide rate.

Methodology

The data were analyzed through Time Series Cross Sectional (TSCS) Feasible
Generalized Least Squares in order to test the impact of high-skilled, low-skilled emigration, and
remittances on the human development of the sending countries. This methodology was chosen

7 Definition provided by the World Bank, available on

http://data.worldbank.org/indicator/SH.XPD.PUBL.GX.ZS Last visited on August 20, 2017.
8 Definition provided by the World Health Organization, available on
http://www.who.int/gho/maternal_health/skilled_care/skilled_birth_attendance_text/en/ Last visited on
August 20, 2017.
9 Definition provided by the World Bank, available on
http://data.worldbank.org/indicator/SE.XPD.TOTL.GD.ZS Last visited on August 20, 2017.
6
as it allows for an auto regressive term of nature AR(1) in the model, and it assumes correlation
among the independent variables and panels, as well as heteroscedasticity10.

However, the general base model is as follows:

'Yit = αt + ' β1yit-1 + 'β2xit +'β3x2it + FEt+ ϵit (1)

Where yit is each variable representing the concept of human development which includes
a lagged measure on the right side of the equation in order to account for the previous effects. Xit
is each variable representing the concept of migration and its effect of on human development.
X2 represents the control variables among which there is a dummy variable that controls for the
1990 shock. FEt represent year fixed effects. Given that one-differenced variables are used,
country fixed effects are not considered. Panel-specific autocorrelation structure is used as it is
assumed that each country has errors that follow a different AR1 process11. As a robustness
check, the models were also run using Panel Corrected Standard Errors, using the option for AR1
panel-specific autocorrelation structure given that the coefficient of the AR1 process would not be
the same for each panel. It is assumed, with PCSE, that the disturbances are heteroskedastic
and correlated across panels12.

Endogeneity

The data used for this research are annual. However, migration is a process that may take
some time for it to affect the human development of the sending countries in a meaningful way.
In other words, migrants usually do not establish and find a job immediately in the receiving
country. Therefore, it would be difficult for them to remit immediately. In order to show the effect
of migration on human development of the sending countries in the long-run, this research re-
estimated the models using 5-year-averages. In addition to showing a long-run effect, this
methodology would also help with the inherent problem of endogeneity. The models were
estimated using Feasible Generalized Least Squares and Panel Corrected Standard Errors in
order to compare the results with those bases on a yearly basis. The models were estimated with
period fixed effects and panel-specific autocorrelation structure.

10 Stata Manuals, 2015. Last accessed October 9, 2015. Available on

http://www.stata.com/manuals13/xtxtgls.pdf
11 Idem
12 Stata Manuals, 2015. Last access November 27, 2017. Available on
https://www.stata.com/manuals13/xtxtpcse.pdf
7
Results
Migration and the access to a decent standard of life

The variables analyzed here are wages, employment, GNI, and freedom. Each of the
independent variables (high-skilled emigration, low-skilled emigration, and remittances) is
analyzed for its effect as a first difference and as a one-year-lag in order to see the effect of the
previous period. Table 1 shows the results for the effect of emigration on the access to a decent
standard of life. The first four columns show the results of the analysis through Feasible
Generalized Least Squares (Model 1). The following columns show a re-estimation through Panel
Corrected Standard Errors (Model 2). Models 3 and 4 correspond to models 1 and 2 re-estimated
through the 5-year-average, respectively.

The results presented are for the effect of high-skilled and low-skilled emigration, as well
as remittances13. The results obtained are, in general, consistent across all the methods used. In
the case of wages, for example, high-skilled and low-skilled emigration show positive short-run
effects in models 1 and 2. These results imply that wages go up as high-skilled and low-skilled
emigration increases. In other words, theoretically, if the demand for labor is kept equal in the
sending countries after some emigration of workers, the labor supply should decrease as
unemployment decreases, and wages should increase as a response to the lowest productivity
after emigration, due to labor scarcity. If the demand for labor increases, as any expanding or
growing economy, but emigration continues to take workers away, labor supply would be even
lower than the pre-migration point and wages should increase even more. Nevertheless, if a given
migrant sending country has a surplus of labor, production would not be affected given that the
required labor would still be in the country, and even though unemployment should decrease,
wages would not be affected because there is no labor scarcity, and therefore, productivity would
not be hurt (FIGUEROA, 2016).

In the case of Latin America, the analysis shows that the emigration of both, high-skilled
and low-skilled workers helps increase wages and employment in the sending countries.
Therefore, theoretically, freer migration would cause an increase in wages in Latin America
triggering a redistribution of income and inequality would very likely decrease in the long run
(FIGUEROA, 2016). Additionally, migration can have positive effects by reducing competition for
limited jobs when there are high levels of unemployment (Koser, 2007).

XZ
GControl variables are not presented on the tables in order to save space. Full results are with the author.
8
 Table 1 - Effect of Migration on the Access to a Decent Standard of Life
XTFGLS (Model 1) XTPCSE (Model 2) XTFGLS (5-year-average) (Model3) XTPCSE (5-year-average) (Model4)

Wages Employ GNI Freed Wages Employ GNI Freed Wages Employ GNI Freedom Wages Employ GNI Freed

High-Skilled_d
0.067*** 0.042** -0.023*** -0.045 0.066*** 0.004* -0.003 - 0.017** 0.101** 0.005** 0.009** 0.013* 0.100*** 0.006*** 0.010** 0.013*
(0.03) (0.02) (0.01) (0.04) (0.02) (0.00) (0.00) (0.01) (0.00) (0.00) (0.00) (0.01) (0.01) (0.00) (0.00) (0.01)
High-Skilled_l
0.036 0.057*** -0.033*** -0.075** 0.035*** 0.003*** -0.005** -0.002 0.036** 0.002** 0.005** 0.007** 0.035*** 0.003*** 0.005** 0.006
(0.02) (0.01) (0.01) (0.03) (0.01) (0.00) (0.00) (0.00) (0.01) (0.00) (0.00) (0.00) (0.01) (0.00) (0.00) (0.00)

Low-Skilled_d
0.160** 0.046*** -0.022 -0.309** 0.160*** 0.004 -0.002 0.004 0.080 0.006*** 0.011*** 0.014*** 0.091 0.004* 0.013** 0.014*
(0.07) (0.01) (0.03) (0.16) (0.04) (0.00) (0.00) (0.01) (0.06) (0.00) (0.00) (0.01) (0.54) (0.00) (0.00) (0.01)
Low-Skilled_l
0.221*** 0.025*** -0.044** -0.116 0.221*** 0.002* -0.001 0.002 -0.060 0.000 0.005*** 0.005 -0.082 0.000 0.005** 0.006*
(0.05) (0.01) (0.02) (0.10) (0.03) (0.00) (0.00) (0.00) (0.11) (0.00) (0.00) (0.00) (0.11) (0.00) (0.00) (0.00)

Remittances_d
-0.028 0.015*** -0.017** 0.002 -0.028* 0.000 -0.006* 0.021*** 0.028** 0.007** 0.000 0.017*** -0.004 0.007** 0.002 0.022**
(0.03) (0.00) (0.01) (0.04) (0.02) (0.00) (0.00) (0.01) (0.01) (0.00) (0.00) (0.01) (0.02) (0.00) (0.01) (0.01)
Remittances_l
-0.029* 0.012*** -0.017*** 0.037 -0.029*** 0.000 -0.003*** 0.003*** 0.014 -0.004 0.007** 0.002 0.008 -0.004 0.007** 0.004**
(0.02) (0.00) (0.00) (0.03) (0.01) (0.00) (0.00) (0.00) (0.01) (0.00) (0.00) (0.00) (0.01) (0.00) (0.00) (0.00)
Standard errors in parentheses. *p<0.10,**p<0.05, ***p<0.01

9
 10

In other words, competition for scarce jobs reduces and the high-skilled workers that
remained in the sending countries might experience an upward in wages. Low-skilled emigrants,
on the other hand, were expected to behave similarly, especially if a given sending country
experiences surplus of labor. Therefore, low-skilled emigration might also help reach an increase
in the wages of the fellows behind. That is, with less population, even low-skilled jobs would see
an increase in their wages as labor becomes scarcer (ROSENZWEIG, 2005).

These results would mean that migration works as a way out for those countries that have
high unemployment rates and those that have low wages. The findings in this research are similar
to those of Figueroa (2016a), who studied Central America and Mexico and found that an increase
in high-skilled and low-skilled emigration leads to higher employment. Additionally, the author
found that an increase in high-skilled emigration leads to higher wages in her sample. However,
remittances have a negative short-run effect in models 1 and 2 on wages, and the contemporary
short-run effect was not statistically significant. It seems like remittances do not necessarily
translate into higher wages immediately. Some time would probably be needed for adjustment
between lower productivity and the increase of wages caused by labor scarcity. In the case of
employment, remittances show a positive effect in model 1. Indeed, remittances could help start
new small companies that create more employment in the sending countries.

With respect to GNI, there seems to be short-run negative effects in all the three measures
of migration in models 1 and 2. This might be due to the negative shock that the economy faces
when the work force decreases as a result of emigration, high-skilled or low-skilled, given that
those people who emigrate stop, at least temporarily, contributing to their economy. On the other
hand, even though the negative short-run impact of remittances is counterintuitive, it could be
explained by time itself. In other words, it would take some time for remittances to be spent in
consumption and make a positive impact on the national income of the sending countries when
there are priorities such as health care and education in the household. And even when
investment in health and education could lead to a higher national income, this would only be in
the long-run. Furthermore, there is research that points out that remittances can cause the “Dutch
disease” decreasing competitiveness of the national economy and increasing the economic costs
on exports (AMUEDO-DORANTES et al 2009), therefore, an increase in remittances could
adversely affect national income.

In the case of freedom, it was somewhat striking that the high-skilled and low-skilled
emigration showed a negative effect in models 1 and 2. High-skilled people usually push hard for
changes in civil and political freedoms, given their high knowledge and skills, and therefore, there
 11

could be a negative impact when these people stop doing it in order to emigrate. On the other
hand, the low-skilled ones also push for changes in civil and political rights through protests and
diffusion of information, and when thousands of low-skilled workers emigrate, the short-run
negative effect could be explained. For both of them, voice could operate as the preferred way of
action, being exit the second one. Once abroad, they both can influence their relatives and friends
back home, in an attempt to continue to fight for better civil and political freedoms, and even
democracy. On the other hand, remittances show a positive impact in all the models, having
statistical significance in model 2. This result implies that remittances can be used in order to
advance civil and political rights, through individuals or migrant organizations, having an important
immediate effect.

With respect to models 3 and 4, it can be observed that even the negative short-run effects
turned positive in the long-run, as is the case of GNI and freedom. This could mean that in the
long-run, emigration ends up affecting positively the living standards of the sending countries. In
the case of wages, the results showed positive effects for high-skilled emigration and remittances.
On the other hand, in the case of the low-skilled emigration, the effect is positive but without
statistical significance. This could mean that in the long-run, wages end up adjusting for the high-
skilled sector, but not for the low-skilled one, which has only short-run positive effects. One
explanation for this difference could be the nature of the national labor market. For example, those
countries with a high rate of unemployment and a minimum low-skilled emigration would not
experience an increase in wages because there would never be labor shortage. Therefore,
companies would still offer the minimum wage and there would always be people willing to accept
it. On the other hand, high-skilled emigration is usually more valuable in the sending countries,
and when faced with the risk of losing valuable high-skilled workers, companies would be willing
to pay more in order to avoid this brain drain. Additionally, the negative short-run effect of
remittances is transformed into a long-run positive effect, probably through the creation of new
jobs.

In the case of employment, both high-skilled and low-skilled emigration showed positive
effects in the long-run, which could be evidence of employment adjusting after five years after
both types of emigration. On the other hand, remittances show a positive and statistically
significant effect in the long-run. With respect to GNI, the short-run negative effects of models 1
and 2 became positive in the long-run in models 3 and 4, evidencing that GNI would need some
extra time in order to adjust and absorb the positive effects of remittances. With respect to
freedom, both high-skilled and low-skilled emigration overcome the short-run negative effects and
 12

become positive in the long-run. Remittances also show a consistent positive effect on freedom
in the long-run. As discussed above, once emigrants are in the receiving country, both high-skilled
and low-skilled emigrants can influence their relatives and friends back home, in an attempt to
continue to fight for better civil and political freedoms, and even democracy (ESCRIBÀ-FOLCH;
MESEGUER; WRIGHT, 2015). Indeed, as emigrants acquire new knowledge about countries with
more civil and political freedom, they transmit it to their relatives and friends in the sending
countries through several communication channels. These social and political remittances can
trigger a desire for change in the sending countries that can be translated into more demands for
the incumbent government and protests that lead to expanded freedoms.

Migration and the access to a long healthy life

The variables analyzed here are the life expectancy at birth, births attended by skilled
health personnel, and government expenditure on health. Each of the independent variables
(high-skilled emigration, low-skilled emigration, and remittances) is analyzed for its effect as a first
difference and as a one-year-lag. Table 2 shows the effect of migration on the access to a long
healthy life. The first four columns show the results of the analysis through Feasible Generalized
Least Squares (Model 1). The following columns show a re-estimation through Panel Corrected
Standard Errors (Model 2). Models 3 and 4 correspond to models 1 and 2 re-estimated through
the 5-year-average, respectively.

In the first place, it was surprising to see a negative short-run effect of the three measures
of migration on the whole area of a long health life. Even though the results are mixed, there are
several negative short-run effects that were unexpected. In the case of life expectancy, for
example, high-skilled and low-skilled emigration show a negative short-run effect. Remittances
show a similar negative short-run effect. One explanation for this could be the high amounts of
stress both on the emigrants and on their family that directly affects their health, and therefore,
life expectancy. Naatus (2013) found a similar negative effect on his study about remittances in
El Salvador. The author explains that this negative effect might be because as people in the
sending countries become dependent on remittances their habits change. For example, people
who used to work on intense jobs such as agriculture, have no more need of this kind of job and,
overtime, they become less physically active. In addition, their diet changes from basic grain foods
to more unhealthy foods (e.g. junk food) and more consumption of alcohol.
 13

Table 2 - Effect of Migration on the Access to a Long Healthy Life

XTFGLS (Model 1) XTPCSE (Model 2) XTFGLS (5-year-average) (Model3) XTPCSE (5-year-average) (Model4)
Life Health Health Life Health Health Life Health Health Life Health Health
Expect Pers Expend Expect Pers Expend Expect Pers Expend Expect Pers Expend
High-Skilled_d
-0.005** -0.003 -0.021* 0.000 -0.003** -0.016*** 0.002*** 0.057*** 0.000 0.002*** 0.054*** 0.0004*
(0.00) (0.00) (0.01) (0.00) (0.00) (0.01) (0.00) (0.01) (0.00) (0.00) (0.01) (0.00)
High-Skilled_l
-0.002** 0.002 -0.016*** 0.000 0.001 -0.01*** 0.006*** 0.020*** 0.001* 0.005*** 0.019*** 0.001***
(0.00) (0.00) (0.01) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00)

Low-Skilled_d
0.000 -0.009** -0.010 -0.002* -0.008*** -0.008** 0.006*** -0.014*** 0.004 0.002*** -0.014*** 0.009
(0.00) (0.00) (0.01) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.01)
Low-Skilled_l
-0.002* -0.002 -0.015*** 0.000* 0.00*** -0.01** 0.005*** 0.025*** 0.00 0.002*** 0.026*** 0.00
0.000 -0.009** -0.010 (0.002) (0.008) (0.008) (0.00) (0.00) (0.00) (0.00) (0.00) (0.02)

Remittances_d
0.000 -0.010 -0.010 0.000 -0.008 -0.008 0.001*** -0.017*** 0.003** 0.002*** -0.015*** 0.0003**
(0.00) (0.02) (0.02) (0.00) (0.01) (0.01) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
Remittances_l
-0.003*** 0.000 0.000 -0.002*** 0.001 0.001 0.004*** 0.029*** 0.000 0.004*** 0.027*** 0.000
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.00) (0.01) (0.00)
Standard errors in parentheses *p<0.10,**p<0.05, ***p<0.01
 14

Additionally, the author explains that the children of emigrants are usually raised by their
grandparents, which puts more pressure on the elderly (NAATUS, 2013). Emigrating entails an
initial investment that can serve for the journey, and for the first days in the receiving country when
immigrants would be jobless. In other words, it might take some time before the new immigrants
find a job, and they should be financially prepared for this. Sometimes, people who are planning
on emigrating save money for some time or take loans in order to finance this endeavor. This is
money that could be used on a better nutrition or medicines, but it is being used for emigration.
This is expected to negatively affect the families of the emigrants and their life expectancy in the
short-run.

High-skilled emigration showed a negative short-run effect on births attended by highly

skilled personnel, being statistically significant in model 2 only, while low-skilled emigration also
showed negative short-run significant effects in models 1 and 2. One speculation for this result
could be that emigration of the high-skilled ones can work as an incentive for other high-skilled
workers to feel the urge to emigrate as well, probably attracted by higher wages, possibly
evidencing some brain drain effects. On the other hand, emigration of the low-skilled workers
could work on the same direction when their children prefer to emigrate instead of spending
several years pursuing a health degree. In fact, Bundred and Levitt (2000) describe the emigration
of highly skilled health personnel as a “medical carousel” in which health personnel continuously
moves to countries with a higher living standard. For example, in the case of Latin America, the
migration of health personnel is on the rise and sending countries are hiring foreign-trained health
personnel in order to compensate the emigration of their own (CARPIO; BENCH, 2015).
Remittances, on the other hand, showed a positive effect in the lagged variable in Model 5 only,
implying that there is some positive effect from the previous year. This suggests that remittances
could be used in order to get health degrees, loosing some pressure on the urge to emigrate and
on the brain drain.

Finally, high-skilled and low-skilled emigration show negative short-run effects on the
government expenditure on health. Remittances, on the other hand, do not show a statistically
significant effect in all the models. One explanation for this could be that the emigration of the
workforce, either high-skilled or low-skilled, relaxes the burden of governments of investing in
health more and more each year, which would be expected as a normal result of the annual
increase in population and the needed medical improvements in health. In other words, it is
possible that less people in a given country work as an incentive for the government to invest less
in the health system and more in other areas.
 15

Nevertheless, when seeing models 3 and 4, it can be seen that some of the negative short-
run effects turned positive in the long-run. With respect to life expectancy, the negative short-run
effects turn positive in the long run in the three measures of migration. In the case of health
expenditure on health, the negative short-run effects of the high-skilled and low-skilled turned
positive. Additionally, the effects of remittances, which was not statistically significant in the yearly
basis, turned positive and statistically significant. This could mean that remittances need some
time in order to make a positive effect. On the other hand, in the case of births attended by highly
skilled personnel, which presented negative short-run effects, turned positive in the case of the
high-skilled emigration. This could be due to the influence that the high-skilled emigrants can have
on the migrants-to-be in the sending countries. In other words, the high-skilled emigrants can
somehow convince people in the sending countries to pursue a health degree, either because
they would be more helpful at home, or because a health degree would be more helpful in case
people still want to emigrate. In the case of the low-skilled emigration, which has a negative short-
run effect, continue to have a long-run negative effect. Again, it is possible that the low-skilled
emigrants influence people at home in order to emigrate, instead of pursuing a health degree for
so many years. Another reason for this could include the negative effect of the absence of the
emigrant parents. Research has found that there are negative effects of the absence of parents
on school attainment, including that the teenager is in age of working and has to stop school on
order to help provide at home or the lack of authority of the relatives that are taking care of the
migrant’s children (HAGEN-ZANKER, 2015).

Lastly, remittances, which had no statistically significant effect on births attended by highly
skilled personnel, reached a negative long-run effect. A speculation for this could be that people
in the sending countries use remittances for other things, even for migrating, and less people use
remittances for the pursuing of a health degree. In other words, in some cases, an increase in
remittances might not lead to a higher school outcome (HAGEN-ZANKER, 2015).

Migration and the access to education

The variables analyzed here are the enrollment ratio in primary, secondary, and tertiary
education according to gender parity index, which is used in order to take into consideration the
education of girls vis-à-vis boys, as well as the government expenditure on education. Each of
the independent variables (high-skilled emigration, low-skilled emigration, and remittances) is
analyzed for its effect as a first difference and as a one-year-lag. Table 3 shows the effect of
migration on the access to education. The first four columns show the results of the analysis
 16

through Feasible Generalized Least Squares (Model 1). The following columns show a re-
estimation through Panel Corrected Standard Errors (Model 2). Models 3 and 4 correspond to
models 1 and 2 re-estimated through the 5-year-average, respectively.

To begin with, the results show no statistically significant effects from the high-skilled
emigration on government expenditure on education. Low-skilled emigration, on the other hand,
shows a negative short-run effect of the lagged variable in model 2, while remittances show a
positive effect of the lagged variable in model 1. In his study, Batista et al (2012) found that there
is a positive relationship between the probability of migrating and the investment on more
education, and that a decrease in the probability of migrating leads to lower levels of education.
In the case of Latin America, it seems like the probability of high-skilled emigration does not trigger
a positive effect in government spending on education. Furthermore, it seems like the probability
of low-skilled emigration has a negative effect in government spending on education in the short-
run.

Likewise, when seeing models 3 and 4, it can be observed that high-skilled emigration
continues to show no statistically significant effects in the long-run, while low-skilled emigration
showed a negative long-run effect. One speculation for this could be that governments see that
there is no return to their investment in education of their citizens over decades, and therefore,
government expenditure on education does not increase as it should. On the other hand, it is
interesting that remittances have a long-run positive effect. Perhaps, remittances could serve as
an incentive for governments to invest on a more skilled workforce with the primary purpose of
increasing human development at home, or with the secondary purpose of “exporting” more highly
qualified labor, which could remit even more and to serve as means for trade and investment.
Even though remittances are used by the receivers and not by the government, the injection of
this extra money into the system helps ease the burden that governments have with respect to
investments.

In other words, given that people now have some extra income, they are able to provide
some services by themselves, such as better consumption items, more access to health and
private schools. This gives room to the government for investing in the areas of more priority at a
certain time. Therefore, even though the idea of losing people might not be very appealing to the
governments of the sending countries because they lose very necessary high-skilled and low-
skilled workers, the fact of having extra money in the economy represents a temporary solution
that is not the ideal policy, but it can work.
 17

Table 3 - Effect of Migration on the Access to Education

XTFGLS (Model 1) XTPCSE (Model 2) XTFGLS (5-year-average) (Model3) XTPCSE (5-year-average) (Model4)
Educ Primary Second Tertiary Educ Primary Second Tertiary Educ Primary Second Tertiary Educ Primary Second Tertiary
Exp Exp Exp Exp
High-Skilled_d
0.025 0.000 0.003 0.094 0.008 -0.003 0.003 -0.007 0.013 0.005*** 0.009*** 0.004 0.024 0.005*** 0.009*** -0.006
(0.07) (0.00) (0.00) (0.13) (0.01) (0.00) (0.00) (0.01) (0.03) (0.00) (0.00) (0.01) (0.04) (0.00) (0.00) (0.01)
High-Skilled_l
0.000 0.003 0.003 0.038 0.002 0.003*** -0.001 -0.002 -0.014 0.003*** -0.001 -0.012*** -0.006 0.003*** 0.000 -0.008*
(0.06) (0.00) (0.00) (0.07) (0.00) (0.00) (0.00) (0.01) (0.03) (0.00) (0.00) (0.02) (0.03) (0.00) (0.00) (0.00)

Low-Skilled_d
0.230 0.021* -0.003 0.069 0.035 -0.001 0.003 -0.011 0.034 0.004*** 0.007** -0.008 0.032 0.005*** 0.008** -0.006
(0.23) (0.01) (0.01) (0.15) (0.03) (0.00) (0.00) (0.01) (0.07) (0.00) (0.00) (0.01) (0.07) (0.00) (0.00) (0.01)
Low-Skilled_l
0.014 0.005 0.004 0.183*** -0.008*** 0.002** -0.008*** -0.006 -0.231** 0.007*** 0.008** 0.048*** -0.230** 0.008*** 0.009** 0.052***
(0.14) (0.00) (0.01) (0.07) (0.00) (0.00) (0.00) (0.00) (0.11) (0.00) (0.00) (0.01) (0.11) (0.00) (0.00) (0.01)

Remittances_d
0.018 0.004* -0.002 0.039 0.000 0.000 0.005 -0.010 0.069*** 0.002*** 0.009*** 0.021*** 0.069*** 0.002*** 0.009*** 0.021***
(0.05) (0.00) (0.00) (0.16) (0.01) (0.00) (0.01) (0.01) (0.01) (0.00) (0.00) (0.00) (0.01) (0.00) (0.00) (0.00)
Remittances_l
0.092* -0.001 0.002 0.067** 0.003 0.001** -0.005*** 0.000 -0.022 0.001*** 0.010*** 0.049*** -0.032* 0.002*** 0.010*** 0.049***
(0.05) (0.00) (0.00) (0.03) (0.00) (0.00) (0.00) (0.00) (0.02) (0.00) (0.00) (0.00) (0.02) (0.00) (0.00) (0.00)
Standard errors in parentheses *p<0.10,**p<0.05, ***p<0.01
 18

With respect to primary education, high-skilled emigration showed a positive short-run

effect of the lagged variable in model 2. Low-skilled emigration, on the other hand, show a positive
short-run effect in models 1 and 2. Remittances also show positive short-run effects in the two
models. From these results, it can be observed that enrollment in primary education, for girls vis-
à-vis boys, is one of the variables with most positive effects. One of the main motivations for people
to emigrate is to give their kids better opportunities in life, and girls are being educated more than
ever before. In this sense, it is important for emigrating parents that their kids go to school and
basic education is very important. This could explain why enrollment in primary education has
several positive results in the statistical analysis, as well as the positive effect of remittances. In
fact, when looking at models 3 and 4, it can be observed that enrollment in primary education has
stronger effects in the long-run, implying that both high-skilled and low-skilled emigration, as well
as remittances, serve as an incentive to enroll at least in basic education in the sending countries.
These positive results are also influenced by the constant efforts of Latin American and Caribbean
governments to make primary education universal (OCAMPO; VALLEJO, 2012). Still, even in spite
of these efforts and even when in most of the cases primary education is universal, in some
countries parents still have to deal with expenses such as uniforms, shoes of a certain style, and
school supplies in order to enroll their kids in schools. Furthermore, in rural areas, some
transportation is required. Remittances are of much help in these type of situations.

In the case of secondary education, high-skilled emigration does not show any significant
effects. Low-skilled emigration, on the other hand, shows a negative short-run effect of the lagged
variable in model 2 only. Remittances, on the other hand, show a negative short-run effect of the
lagged variable in model 2. These recurrent negative effects might be explained by the urge to
migrate that some teenagers can have, given that their parents are already in the United States
and they are probably being raised by other relatives. Through the constant communication with
their parents, from social media, and from television, teenagers can have a glimpse of the way of
life in the country where their parents are, such as the United States for example, and might
consider to emigrate. Remittances, on the other hand, can be used for this endeavor, reducing the
enrollment in secondary education. Another possible explanation is the high number of teenage
pregnancies that Latin America has. Indeed, according to UNICEF, Latin America is the region
with more teenage pregnancies in the world14. If teenage girls grow up without their parents, who
would be abroad, there would be less discipline at home and a pregnancy would be more likely to

14UNICEF, 2016. See more details on ››—šaVVžžžU—™Œ•šˆUŠ–”Všˆ“œ‹† †ŠŒ•ŠˆVh”Œ™ŠˆTsˆ›•ˆT

jˆ™‰ŒTŒ”‰ˆ™ˆ¡–T—™ŒŠ–¡†W†[\`Y\[W^Y\U›”“ Last accessed on December 18, 2017.
 19

happen. Nevertheless, looking at models 3 and 4, it can be observed that these short-run negative
effects on the enrollment on secondary education are reversed and become positive in the long-
run for the three measures of migration.

Finally, with respect to tertiary education, high-skilled emigration does not show any
statistically significant effect in any of the models, while low-skilled emigration shows a positive
short-run effect in model 1 only. Remittances, on the other hand, show a positive short-run effect
in model 1. In this respect, it could be safe to draw on the research by Boucher, Stark and Taylor
(2009) that discusses how the probability of emigration works as an incentive to acquire more
education and more skills. Indeed, even though in this case it is the low-skilled emigration that
has stronger positive effects, as citizens aim for a better economic future, they could be
encouraged to get a better education in order to be eligible for a job abroad. This would increase
the freedoms of both education and economic development in the long-run, leading to brain gain
as the return to education. Additionally, the notion of “brain exchanges” that characterizes
globalizing economies can also act as an incentive for sending countries to insert themselves in
the global labor market and boost education enrollment of their girls and eventually, an overall
national well-being.

In models 3 and 4, high-skilled emigration show a negative long-run effect from the lagged
variable. These negative long-run effect could be caused by young people who are influenced by
the high-skilled emigrants in order to get higher education abroad. Indeed, more and more
international organizations are fully funding scholarships abroad for people in the developing
countries, and they give particular importance to the education of girls in the developing countries.
This would reduce a little the enrollment in tertiary education in the sending countries15. Some
governments from the most developed countries are also trying to attract young people from Latin
America to study with partly or fully funded scholarships in their countries. These are great news
for Latin American governments as young people are being the focus of investment for a future
higher human development. However, sometimes these young people do not go back to their
countries of origin due to several reasons, including insecurity or lack of opportunities for their
obtained higher education. In spite of this, if these newly educated people decide to go back home,
they will eventually help increase the human development of their countries. If they decide to stay

15One of these international organizations is the Organization of American States, who annually gives
hundreds of fully funded scholarships for the people of the Americas. More information available on
››—aVVžžžU–ˆšU–™ŽVŒ•VšŠ–“ˆ™š—šV Last visited December 13, 2017.
 20

abroad, they have a chance to get a higher wage and help the sending countries through economic,
social or political remittances.

Another reason for this could be that, having information on wages and standard of life in
the receiving country through information given by the high-skilled emigrants can trigger the urge
to emigrate on people in the sending countries. When considering that higher wages and a higher
standard of life can be achieved by emigrating, regularly or irregularly, people might consider that
the cost of pursuing tertiary education (time and money) is too high. Therefore, emigration could
sound as the immediate solution with the highest return. There is research that evidences this point
in the case of Mexico. The authors find that emigration to the United States decreases the
probability of educational attainment (MCKENZIE; RAPOPORT, 2006). For this to be avoided,
high-skilled emigrants could clarify that higher wages come as return for higher education.
Nevertheless, in most cases, even low-skilled emigrants can earn higher wages in the receiving
country when compared to the achievable wages in the sending countries.

On the other hand, low-skilled emigration and remittances show strong positive effects in
the long-run. I believe that, at least with respect to education, there should not be great differences
in the effects that high-skilled and low-skilled emigration can have. This is because for the high-
skilled emigrants, it is important that their children get as much education as possible, because
they already had a glimpse at the competitiveness that their children will face. Low-skilled
emigrants, on the other hand, expect that their children have a better education than the one they
had, because they expect that education will give their children, and girls in particular, all the things
they were not able to achieve. In this sense, even when the motivations are different, their aim is
the same: give their children as much education as possible, and educating girls is becoming more
and more important in Latin America.

Final considerations
From the results obtained through this research, it can be said that the impacts of
emigration on human development are heterogeneous, and that depend on the type of emigration
(high-skilled or low-skilled) and on time (short-run or long-run). The findings of this study suggest
that migration plays a very important role, mostly positive, in the improvement of the human
development of the sending countries. These findings contradict the arguments of Delgado and
Covarrubias (2006), who explain that migration does not generate opportunities for growth or
social development in the sending countries, and that it creates unemployment, stagnation,
inflation, and economic dependency. Even though this research is concerned with human
 21

development only, it can be said that the results found here present a more positive picture in the
long-run for the countries in the sample. I also found statistical evidence that suggests that there
are important short-term benefits of migration on the access to a decent living standard and to
education. Specifically, wages and employment seem to benefit from short-term positive effects,
as well as the enrollment in primary and tertiary education. However, there are also some worrying
negative short-term effects on national income, freedom, government expenditure in education,
and enrollment in secondary education. Nevertheless, most of these negative short-run effects
revert in the long-run and become more positive.

The results point that the area of a long and healthy life is the most affected by short-term
negative effects, which become positive in the long-run in the case of life expectancy and
government expenditure in health. However, there is a persistent negative effect also in the long-
run from the emigration of the low-skilled workers on the number of babies delivered by highly
skilled personnel. In the long-run, it seems like high-skilled emigration has more positive effects
on living standards and on a long healthy life than the low-skilled emigration. On the other hand,
in the case of education, it seems like low-skilled emigration has more positive long-run effects.
This is important because, first, it shows that migration is a process that can take some time in
order to show a positive effect. Second, it helps establish a difference between low-skilled and
high-skilled emigration in term of their effects on specific areas of human development. Third,
even though the contributions of the high-skilled can be greater in the long-run, it became evident
in this study that low-skilled emigrants are more in absolute numbers, which makes them reach
great contributions that deserve to be recognized as much as those of the high-skilled ones. And
given that most of the times they have less benefits in the receiving country, the governments of
the sending countries must ensure that these emigrants are offered all the protection and benefits
they need so that their capabilities expand even more.

REFERENCES

ACOSTA, P. A.; LARTEY, E. K. K.; MANDELMAN, F. S. Remittances and the Dutch disease.
Journal of International Economics, v. 79, n. 1, p. 102–116, 1 set. 2009.

AMUEDO-DORANTES, C.; POZO, S. Workers’ Remittances and the Real Exchange Rate: A
Paradox of Gifts. World Development, v. 32, n. 8, p. 1407–1417, 1 ago. 2004.

ANAND, S.; SEN, A. The income component of the human development index. Journal of human
development, v. 1, n. 1, p. 83–106, 2000.
 22

BATISTA, C.; LACUESTA, A.; VICENTE, P. C. Testing the ‘brain gain’hypothesis: Micro evidence
from Cape Verde. Journal of Development Economics, v. 97, n. 1, p. 32–45, 2012.

BOUCHER, S.; STARK, O.; TAYLOR, J. E. A gain with a drain? Evidence from rural Mexico on
the new economics of the brain drain. In: Corruption, development and institutional design.
[s.l.] Springer, 2009. p. 100–119.

CARPIO, C.; BENCH, N. S. The Health Workforce in Latin America and the Caribbean: An
Analysis of Colombia, Costa Rica, Jamaica, Panama, Peru, and Uruguay. World Bank
Publications, 2015.

DELGADO WISE, R.; COVARRUBIAS, H. The Mexico-United States migratory system: Dilemmas
of regional integration, development, and emigration. Migración y Desarrollo, n. 7, 2006.

ESCRIBÀ-FOLCH, A.; MESEGUER, C.; WRIGHT, J. Remittances and democratization.

International Studies Quarterly, v. 59, n. 3, p. 571–586, 2015.

FIGUEROA, A. Migration from Central America and Mexico to the United States: How Does the
Movement of People Affect the Labor Market of the Sending Countries? Encuentro
Latinoamericano, v. 3, n. 2, p. 31–47, 2016. 2017.

HAGEN-ZANKER, J. Effects of remittances and migration on migrant sending countries,

communities and households. Economic and Private Sector, v. 1, p. 1–29, 2015.

MCKENZIE, D. J.; RAPOPORT, H. Can Migration Reduce Educational Attainment? Evidence

from Mexico. Rochester, NY: Social Science Research Network, 1 jun. 2006.

NAATUS, M. K. The socio-economic impact of migrant remittances on life expectancy and

education in El Salvador. International Journal of Business and Social Science, v. 4, n. 8, 2013.

OCAMPO, J. A.; VALLEJO, J. Economic growth, equity and human development in Latin America.
Journal of Human Development and Capabilities, v. 13, n. 1, p. 107–133, 2012.

PARSONS, C. R.; ROJON, S.; SAMANANI, F.; WETTACH, L. Conceptualising International

High-Skilled Migration.

RANIS, G.; STEWART, F. Strategies for success in human development. Journal of human
development, v. 1, n. 1, p. 49–69, 2000.

ROSENZWEIG, M. R. Consequences of migration for developing countries. In: United Nations

Expert Group Meeting on International Migration and Development, Population Division, July,
Proceedings .2005.

TAYLOR, E. J. The New Economics of Labour Migration and the Role of Remittances in the
Migration Process. International Migration, v. 37, n. 1, p. 63–88, 1 mar. 1999.

UNDP. Human Development Report 1990 | Human Development Reports.

UNDP. Human Development Report 2009 | Human Development Reports.