How remittances contribute to Human Capital Formation in Latin America and Mexico

(1)

The Money from the other Side

How Remittances contribute to Human Capital Formation

in Latin America and Mexico

Master of Science in

International Development Studies Graduate School of Social Sciences University of Amsterdam

Christine CM Affolter Jan Van Galenstraat 335 1061 AZ Amsterdam Student No. 10861076 affolterchristina@gmail.com

10.08.2015

Supervision

Dr. Enrique Gomez Llata Cazares Dr. Niels Beerepoot

(2)

Acknowledgements

I would like to take this opportunity to thank everyone who has contributed to the success of this Master's project. First of all, I would like to express my gratitude to my parents that have supported my studies in many ways. Thank you for your confidence in me, my project and for your unconditional support.

I am grateful to my mentor Alexander, who had helped to overcome every single challenge I encountered. I highly appreciate his advice and his encouraging attitude towards my project, not to forget his valuable assistance throughout this Master's thesis.

I would also like to thank Alexandra for introducing me to her incredibly rich Mexican culture. It is through her that my fieldwork turned into a fantastic, unique and informative experience. Alexandra, thank you for making me love Mexico.

Another thanks also to Livia and Max, who went through the effort of proofreading this thesis. Your remarks contributed valuably to the final version.

I owe special thanks to Santi, who has made this Master more of an adventure than anything else. I am grateful for numerous enlightening, inspiring conversations and moments that lead to an efficient and fruitful thesis writing process.

Querida Rosy, muchísimas gracias para tu apoyo en mi proyecto en Puebla y San Andres Azumiatla. Gracias para haberme introducido a tus abuelitos y haber compartido la confianza que tienen en tí. Tu apoyo es un gran enriquecimiento para mi tesis y un símbolo valioso de gran amistad.

Muchas gracias a todos las madres, hijas, hijos, hermanas, esposos y esposas para haber dado muchas informaciones y para haber confiado en mí. Espero que se puedan ver a sus queridos prontísimo y que no tengan que irse otra vez. Que sepan, que hay una persona al otro lado del mundo que piensa en sus destinos.

(3)

Abstract

Twenty years ago, when remittances outran Official Development Assistance, migrant money has hit the agenda of International Development Studies. The amount of received remittances

in the year 2014 is referred to as the New Record High in the Latin American region.1_With

more than one third, Mexico is still receiving the lion's share of the region's remittances. Naturally, the considerable amount of external capital brings along important development implications for migrants and their families left behind. The purpose of this study is to explore these implications and the stimulation of remittances inflow on human capital formation in Latin America. The study pays close attention to Mexico, as it is the Latin American country that receives the highest remittances share.

Different actors have acknowledged the impact of remittances in Latin America; Economists and social scientists cannot decide on whether remittances stimulate development positively or negatively. This scientist community also fails to consent on which form of remittances expenditure prevails: consumption or investment. In the meantime, the government of Mexico has recognised the economic potential and channels the external capital into national and local migrant programmes. The study takes a clear position on these controversies and contributes valuably to the current discussion.

Building on a theoretical basis of human capital theory and transnationalism, the two-level analysis questions the relationship between remittances and human capital anew. As a combination of panel analysis and a case study of Mexico, the mixed methods approach aims to answer how remittances promote human capital formation in Latin America and Mexico. The panel analysis provides solid results that remittances and human capital increase simultaneously. While the qualitative analysis zooms in and sheds light onto household budgets and living contexts of migrant families in Puebla, Mexico. The case study confirms, that the relation remittances and human capital is more complex: Remittances expenditure in Puebla is influenced by family solidarity, regularity of remittances income and group affiliation. Moreover, both the quantitative and qualitative analyses conclude that the developmental impact of remittances on human capital is visible middle- to long-term.

Keywords: Remittances, Migration, Human Capital, Human Capital Formation, Human Capital Theory, Transnationalism, Latin America, Mexico, Puebla, Cholula

(4)

Abbreviations

BBVA Banco Bilbao Vizcaya Argentaria

BUAP Benemérita Universidad Autónoma de Puebla

CIA Central Intelligence Agency

DF Distrito Federal - Federal District of Mexico City

EFA Education for All Global Monitoring (Report)

GDP Gross Domestic Product

GFMD Global Forum on Migration and Development

H Hypothesis

HDI Human Development Index

IFI International Financial Institution

ILO International Labour Organisation

LAC Latin American Countries

MAMSL Meters above sea level

MASA Mexican American Student Association

MDG Millennium Development Goals

MIF Multilateral Investment Fund

MXN Mexican Pesos

MN Minnesota

N Number, Cases

NAFTA North American Free Trade Agreement

NC North Carolina

ODA Official Development Assistance

OECD Organisation for Economic Co-operation and Development

OLS Ordinary Least Squares (Regression)

PISA Programme for International Student Assessment

R Respondent

SEDESOL Secretaría de Desarrollo Social - Ministry for Social Development

SQ Sub question

UDLAP Universidad de las Americas Puebla

UN United Nations

UNDP United Nations Development Programme

UNESCO United Nations Educational, Scientific and Cultural Organisation

US United States of America

USD US-Dollars

WEF World Economic Forum

(6)

List of Visualisations

Chart I: Operationalisation ... 23

Chart II: Results on Hypothesis I ... 31

Chart III: Transparency in Research ... 41

Chart IV: Family Remittances Typology ... 50

Figure 1: Remittances and ODA sent to Latin American (1990 - 2012) ... 17

Figure 2: Remittances and ODA sent to Latin America, without Mexico (1990 - 2012) ... 17

Figure 3: Visualisation of Research and Conceptualisation ... 21

Figure 4: Remittances and ODA sent to Mexico (1990 - 2012) ... 23

Figure 5: Promotion of Human Capital Formation ... 43

Map 1: Research Location ... 42

Map 2: The top 10 states with the greatest remittances income (as % of state GDP) in 2013 46 Picture 1: Advertisement for Money Transfer Services ... 35

(7)

1. Introduction

Twenty years ago, remittances flows have outrun official development assistance (ODA). This trend has been detected in different developing countries around the world. Around 580 billion US Dollars (USD) remittances have officially been transferred in the year 2014 (World Bank 2014). In that year, the amount sent to Latin America and the totalled 65.3 billion USD (Multilateral Investment Fund 2015). With this share, Latin America is currently among the top remittances-receiving areas in the world. Unsurprisingly, remittances and possible development effects in Latin America have spawned considerable interest of scholars in International Development Studies (e.g. Ratha 2013, Lopez Cordova 2005 or Calero et al. 2009). This study will examine the effect of remittances on human capital in Latin America and specifically to Mexico as this country receives more than a third of remittances sent to the region.

Migrant remittances are personal cash transfers from migrants living abroad (external) or moved to another part of the country of origin (internal). These transfers root in financial compensation, salaries, wages that individuals or families earn and send to households in the country (or city) of origin.2 They are understood as an important source of capital that has constantly been on the rise for the past years. Therefore, remittances have been on the agenda of many financial institutions and organisation for a long time already. According to the Multilateral Investment Fund MIF (Maldonado & Hayem 2013) the United States are responsible for three quarters of the entire remittances income to Latin American countries (LAC). Due to the proximity to the United States, Mexico is amongst the biggest remittances receiving countries in Latin America. After India, China and the Philippines, Mexico even receives the highest share of remittances globally.

Naturally, this vast amount of money awakens the interest of many actors: while international institutions track the exact amount of transferred money with Argus eyes and national governments channel the external capital for their benefit, remittances alleviate the poverty level and loosen financial constraints for many remittances recipients. A broad scientific community consisting out of economists, educational specialists or social scientists, theorises

(8)

on the scope of the developmental impact of remittances. Research on remittances produces knowledge about impacts of this global phenomenon and can lead to a better understanding on how to turn remittances or the foregoing migration, into benefits for development. The goal of an international dialogue on migration and development (e.g. Global Forum on Migration and Development) is, to steer the money into sustainable development projects.

It is a common concern of International Development Studies to find out, what enhances a country's development and how it can be controlled. International Development Studies are generally described as a multidisciplinary field and like other Social Sciences, it has been subject to many ideological changes. Hettne (1995, p.64) argues that theories of development in Social Sciences are rather additive than exchangeable. In other words, new theories, new ideologies, knowledge or political events have constantly shaped International Development Studies. The theoretical discussion of Jones also builds up on two foregoing strands and provides an explanation why remittances are developmentally controversial. Jones' discussion displays two opposing views that both link remittances to development (1998, p.4): The first

strand holds the view of structuralism.3 In that view, emigrant regions suffer from economic

dependency and therewith-related economic decline. Remittances in these regions are more likely to be spent on consumptions than on investments. Simultaneously, these regions are confronted with increasing interfamilial inequalities. Haas (2007) argues the paradigm shift in Social Sciences challenges the optimistic views on migration in the 1960s. Instead of eroding old structures and changing for the better, migration is believed to reinforce problems of the underdeveloped (Haas 2007, p.4). The main argument for the migration-pessimists is the loss of human capital, the "brain-drain", that would leave these communities in a dependent, deprived, non-productive state (Haas 2007, p.5). Jones (1998) also mentions that migration-pessimists make migration responsible for emerging inequalities. Furthermore, the migration-pessimists attribute only detrimental effects to remittances. Additionally, money transfers from other countries, mostly more developed and industrialised countries support the dependency theory. More radically, the "capitalist penetration" was viewed not only detrimental to development but the very cause of underdevelopment (Haas 2007, p.5).

The second strand embraces the arguments of the functional school. According to functionalists, remittances trigger and enhance investments. Further, they catalyse economic

(9)

freedom and decrease interurban inequalities. Countercyclical to the migration-pessimists' belief, remittances decrease inequalities because everyone ends up benefiting from the recycling of extra capital. These arguments are reminiscent of the migration-optimistic view that capital transfers into developing countries. The capital finally enables developing countries to jump on the bandwagon of industrialisation (Haas 2007, p.3).

Notwithstanding this migration-pessimists and -optimist discussion, remittances are believed to be a valuable stimulation for development as they appear to be a stable source of external finance (Haas 2007, p.8). As many remittances are also transmitted in unofficial channels, their stabilising and insuring effect is even higher than official reports document. Additionally, remittances often cover important shares of a countries' trade deficit nationwide (Haas 2007, p.9), whereas on a micro level, remittances enable households and families as agents of change. Remittances are also useful to insure households and to protect them from economic shocks (Calero et al., 2009). Chapter 2.3 introduces other positive effects such as better nutrition, better health or wellbeing in general.

Out of all the impacts remittances inflow may have, effects on human capital have gained special attention by Social Sciences. Eventually, a strong human capital is expected to yield benefits to a country's development or even a strengthened economy. The relationship between remittances and human capital has been recognised and vastly explored to this date. However, the findings on this relationship are mixed. Many studies point out, that the connection of migrant money with human capital is only discoverable in a specific context. Either the relationship applies especially to children (Ratha 2013), especially to girls in rural areas (De & Ratha 2012) or it is only true for specific countries (Salas 2014). Some findings do not confirm a positive effect of remittances on human capital but discover a negative correlation on educational attainment (Acosta et al. 2007). Lastly, there is also some evidence that remittances enhance self-employment (Acosta 2006).

Similarly to the migration optimistic view, this study assumes that remittances and human capital in Latin America and Mexico are positively interrelated. Mexico is the country that receives the highest remittances share in Latin America and is therefore chosen as the country of fieldwork. In 2013, the inflow of remittances reached 22 billion USD (Cohn et al. 2013). The study assumes that human capital investment, as a form of remittances expenditure,

(10)

contributes to stronger sustainable development. The research question of this study tests these assumptions:

How do remittances contribute to human capital formation in Latin American countries and Mexico?

The research question and its four sub questions are answered on two levels. The study questions the relationship between remittances and human capital with a mixed methods approach consisting of a panel analysis and a case study. The study is applied to Latin America as an interesting remittances receiving area with a special focus on Mexico, the fourth biggest remittances recipient globally. A regression analysis on 20 Latin American countries for the years 1990 - 2012 yields quantitative data and allows certain generalisations for the alleged relationship. A different operationalisation of human capital approaches the relation between remittances and human capital from a new angle. The analysis concludes that more remittances go along with increased secondary completion and labour market participation. After the quantitative analysis on LAC, a case study of Puebla completes the research with interview analysis and leads to a two-dimensional study. During my fieldwork stay in Puebla, I was able to acquire 35 interview respondents and contextual material about migration. Puebla is amongst the top ten remittances receiving areas in Mexico and therefore serves as an excellent fieldwork location. Together with the interviews, the material provides information on the remittances and human capital relation in Puebla. The contextual material not only shows that the Mexican government has already acknowledged the leverage of migrant money, but also that there are many actors involved in a warm about remittances investments (e.g. civil society). The case study and goes beyond a causal relationship and highlights personal experiences and living conditions of remittances recipients, while revealing the complexity of the migration phenomenon on local level. The two methods yield contrasting results and allow a macro and micro perspective on the same research question. The quantitative analysis gathers generalizable results that are completed by qualitative findings that reveal personal insights from Pueblan migrant families. The two-dimensionality and the new operationalisation of human capital are considered as the main singularities of this study.

(11)

limitations of this study. Chapter 3 presents the chosen mixed methods approach and contrasts the advantages and disadvantages of both methods. In the following (4), the panel analysis on twenty Latin American countries proves that increased remittances go along with stronger human capital. Finally, the qualitative analysis (5) builds the pièce de résistance of this thesis and informs about the experiences of my ten weeks fieldwork stay in Cholula, Puebla (Mexico). It also answers the sub questions and the overarching research question of this study. The conclusion (6) brings the two method strands together, whereas the outlook for further research closes the study.

(12)

2. Theoretical Discussion

The following chapter provides the theoretical basis and elaborates on the two main concepts, namely remittances and human capital. The theoretical discussion introduces the main concepts and embraces the presumed connection of remittances and human capital. In a second step, the chapter considers existing literature, formulates important assumptions and discusses the rationale behind the research. In order to derive the research question, the end of the chapter introduces the situation in Latin America and Mexico.

For the analysis of the relationship between remittances and human capital formation, it is necessary to embed the research question into a theoretical context. The research is situated in the transnationalism approach and human capital theory. The following section discusses the perspective of those two theories on remittances and human capital and why they are important variables in my research. The discussion also aims to provide information on the distinction between human capital and human development.

2.1 Transnationalism

Transnationalism is a flexible term and embraces many subtopics. It covers cross-border ties and sheds light on consequences of global migration and multiculturalism. Transnationalism investigates migrants’ behaviour transnationally, across borders of nation states. The notion of

transnationalism is useful to underpin central questions of social and political change and

transformation (Faist 2010, p.7). It focuses on individuals, groups, networks, firms and institutions that are engaging globally. Therefore, it is closely interrelated with globalisation (The New School of Social Research 2010). Traditionally, transnationalism has a strong economic component. One of its underlying assumptions is, that people migrate in order to pursue financial improvement, either for themselves or for their relatives. Transnationalism is capable of identifying economic processes that go beyond national borders and thus happen in an “economic space” (Ban 2012).

Transnationalism also observes social engagement that goes across the borders of the nation

state and therefore also doubts the existence of the state normatively (Ban 2012). Among others, social engagement within the transnational context embraces reciprocity and solidarity

(13)

(Faist 2010, p.10). Thetransnationalism approach is highly suitable to understand remittances

processes: Remittances flows are a product of cross-border migration flows. Whereas

transnationalism traditionally includes different social formations as mentioned before, this

thesis is limited to relevant actors of remittances reception.

Remittances are seen as a product of transnationalism. The knowledge on remittances is independent from national borders but yet heavily dependent on national policies and multinational factors. The motives and consequences of remittances cannot be understood without adopting the outlined transnational perspective.

2.2 Human Capital Theory

The origins of human capital theory date back to Adam Smith and his Wealth of Nations (1776). For the first time, education was viewed as an investment that would enable people to behave productively. From the idea that skills, education, experience and knowledge are investments and therefore indirect capital, the notion human capital was created.

Traditionally, economists tend to differ between two types of economic activities: consumption and investment. Whereas the first embraces immediate benefits and satisfied needs, the latter creates capacities for future income (Woodhall 2001). In the 1960s the concept of human capital gained new connotations. Thomas Schultz outlines such an understanding of human capital in his seminal work Investment in Human Capital (1961). He points out, that the quantity of workers as an economic factor is meaningless. It is rather the quality of the labour - the human capital - that is valuable to productivity. Human capital becomes an irremissible contribution to economic processes and economic growth. Therefore, Schultz focuses on rates of return yielded by human capital. Subsequently, economists have associated the term human capital with education, as it contributes to the productivity and efficiency by increasing the level of cognitive stock (Olaniyan & Okemakinde 2008, p.479).

Human capital theory emphasises the importance and contribution of education to economic

productivity. According to Olaniyan & Okemakinde (2008, p.480) it is commonly believed that human capital is a key factor to national, social and economic development. Further, it is believed to be more important than bare capital or material resources. Thus, the study at hand

(14)

builds upon the assumption that human capital is an indispensable component to a country’s development. The term human capital does not only signify for a factor of production but also for a multifaceted pivotal component of a country. Given the significance of human capital for development, it is important to understand which factors impact its formation.

Social research predominantly attributes developmental progress to human capital, as they believe that investments in human capital are also investments in development. The economic argument seems to be clear: improvements in human capital, or in the labour force of a country, will yield return. It is easier for a well-educated labour force to adapt to technological change and therewith increases productivity. Nevertheless, financial

development does not necessarily ensure human development. In the course of this study it is

important to define where human capital ends and human development begins, as these terms are not interchangeable. By creating a knowledge paradigm, Mehrotra (2005) critically assesses this relationship. The author (2005) shows that the human capital theory originally underlay human development notion. In that sense, international financial institutions (IFIs) argued that with economic progress there comes human development. However Mehrotra (2005) argues that human development is a much broader notion than human capital and that the former entails human capabilities, values, and social change whereas the latter has a strong economic orientation. Finally, Mehrotra (2005, p.301) holds another interesting argument against human capital or human capital theory. The author sees the theory threatened by the "black box" that stands for a technical relationship of input and outputs. According to Mehrotra (2005) the black box ignores the structure of human capital. On the example of education, the black box ignores the level of schooling, attainment, enrolments and quality. The circumstance that the inner structure of the "black box" is ignored, makes Mehrotra conclude that human capital does not contribute to development. The study at hand does not agree with this assumption. As already mentioned before, I agree with the idea from Hettne (1995) that theories of development add up instead of mutually exchange. I believe that the understanding of human capital has changed so much that human capital now entails

more than only economic productivity. Even though human capital theory might not replace

human development, human capital has evolved and there is more behind it. I am convinced

(15)

2.3 How Remittances turn into Human Capital

On the basis of transnationalism and human capital theory, the following section locates the relation between remittances and human capital into the current social scientific discussion. By discussing the literature on the two main concept of the study, the section introduces each concept and establishes the link between remittances and human capital in order to derive the research and sub questions of this study. On the one hand, the vast existing body of knowledge implies great interest in the topic and therefore justifies its academic relevance. On the other hand, it is difficult to holistically situate the research topic into the entire existing literature. Notwithstanding, by describing relevant literature to understand the concepts, the section shows that many disagreements on the relationship between remittances and human capital remain. Moreover, it formulates the academic contribution of the study and how the latter fills the mentioned research gaps. At the end of the section, the relationship of remittances and human capital is applied to Latin America and Mexico.

Importance of Remittances

There is a common "migration pessimistic" assumption, that a country loses valuable parts of its labour force due to migration. Migrants who benefit from high education in their home country therefore contribute to the “brain drain” phenomenon, when leaving the country. The consequence would be a reduction of a country’s production along with a worsened business environment (Ratha et al. 2011, p.7). Nevertheless, to this “brain drain” there is also a “brain gain” as a country is able to offset some of the output losses through remittances (Ratha 2005, p.164). Additionally, Ratha describes that a country’s economy may also profit from accumulated assets or increased small-business investments through remittances. Other social scientists also conclude that remittances mainly have positive economic effects on receiving countries. Calero et al. (2009) for example, focus on remittances and human capital investments. The authors (2009) provide evidence that remittances affect human capital through relaxing household conditions. An additional component in their study focuses on vulnerability to economic shocks and how remittances can act as mitigation mechanisms.

Overall, there is no consensus on how remittances interact with - for example - economic growth. Neither is there a consensus on what exactly remittances are spent on. The social scientific community agrees with the economists that remittances are either spent on consumption or investment. Therefore, one general hypothesis implies that remittances are

(16)

being spent on consumption goods, because remittances as an additional source of income reduce financial constraints on households. In that case, remittances do not show a pivotal developmental impact. Combes & Ebeke (2010) for example analyse remittances and househould consumption instability and detect a negative relationship. Remittances therefore release consumption constraints. Remittances are attributed an insuring role by preventing recipients from GDP instability. Another study in favor of the consumption hypothesis by Zhu et al. (2014) suggests, that Chinese recipients are more likely to spend remittances income on nonhousing expenditure. Interestingly, they do not find evidence that remittances income is productively invested at all. Although some studies argue in favour of this first hypothesis (Fajnzylber & López 2007) the following paragraph shows, that there is more plausible evidence that remittance income is invested.

Among the different effects that remittances money can have on a country's economy or on individual households, the impact on human capital has earned closer attention. Assuming that remittances are invested for human capital expenditure, Le (2011) finds a positive effect of remittances and human capital. With a model that examines motives for sending remittances, Le (2011) observes a business encouraging effect of remittances that also stimulates investment. His findings suggest that remittances and financial development can be complementary because remittances in form of investments mobilise productive activities. More specifically, Adams & Cuecuecha (2010) divide Guatemalan receivers into groups of internal and international (external) remittance income in order to detect different spending behaviour. The authors discover that regardless where remittances come from (either externally or internally), they are more often spent on education. These findings are consistent with the study from Salas (2014) on Peruvian children. Salas explores the impact of remittances on the quality of education. She observes a significant positive effect of remittances because parents invest remittances in private schooling that provide better education prospects. Acosta et al. (in Fajnzylber & López 2008) also investigate the household behaviour of remittances recipients. Their findings indicate that remittances have positive effects on educational and health expenditures. The impact of remittances varies by gender and geographical location (rural or urban) (in Fajnzylber & López 2008, p.151). However this positive trend is limited to upper income classes only. Furthermore, the authors conclude that remittances positively affect school attainment. Interestingly, this effect tends to be higher when the schooling of the parents is low (in Fajnzylber & López 2008, p.154).

(17)

Education of the parents seems to be an important external factor that contributes to human capital of the children.

Yang (2005) offers a different perspective: His study focuses on the exchange rate evolvement and its impact on investment outcomes in the Philippines. Yang’s dichotomous research sheds light into remittances management in sending and receiving countries. He discusses effects of exchange rate shocks onto different investment outcomes. Among others, he focuses on human capital investment. He deduces that more favourable exchange rate shocks trigger more child-schooling, less child labour and higher household investments in education.

Many studies mainly focus on an educational element of human capital. They look into the formation of a country's labour force. In recent years other components have demanded closer attention: Health and labour market participation have been considered as valuable elements of human capital formation. The study by Lopez Cordova (2005) examines the development impact of remittances in Mexican municipalities. The author not only investigates schooling, but also pays close attention to health outcomes. In a large cross-section of Mexican municipalities, Lopez Cordova provides evidence of a positive effect of remittances on a decrease of infant mortality. Hildebrandt & McKenzie (2005) also find migration to have a positive impact on children's health outcomes. With the help of a national demographic survey the authors provide evidence that migration leads to lower rates of infant mortality and higher birth weights. They conclude that these outcomes are caused by increased income and wealth. Hildebrandt and McKenzie indirectly include an educational aspect and argue that migration also allows for better health knowledge that improves health measures taken in a household.

In an earlier project, Acosta (2006) looks into economic effects of remittances on labour supply and school attendance. With a household survey from El Salvador, Acosta's estimates suggest ambiguous results. Firstly, child wage and female labour supply are negatively affected by remittance income. Secondly, labour market participation of men in El Salvador increases when remittance income rises. The author suggests that remittances create more opportunities for self-employment, particularly for middle-aged men. Similarly to this project, Woodruff & Zenteno (2007) attribute great importance to gender, as migration might affect

(18)

female and male labour participation differently. In their survey of 6,000 Mexican self-employers, the authors do not find any robust relationship between migration networks and male participation in the labour force. Whilst for females, the OLS regression provides evidence for a negative relationship between migration and work force participation.

The literature review demonstrates that there is already a broad state of knowledge of remittances and their effects. Among others, the presented authors discuss effects on country (Ratha 2011) and on household level (Combes & Ebeke 2010), what remittances are spent on (Zhu 2014), how remittances are generally invested (Le 2011) or invested in human capital (Salas 2014). Despite the extensive body of knowledge, reviewing the literature also shows that many controversies remain. A consensus can be found on the positive impact of remittances investments. The current literature argues that remittances predominantly have a positive developmental effect on migrants and their families.4 Based on this widespread assumption, I argue that remittances spent on investments have a stronger sustainable, developmental effect. Due to the positive developmental effect on migrants and their human capital formation and the argument that strong human capital is the basis for human development (Chapter 2.2), this study argues that remittances have a positive effect on human development. Even though this study cannot reinvent the wheel, it wants to provide knowledge on remittances and human capital. By examining the relation between remittances and human capital, the study emphasises the importance of remittances to development. As remittances money increases globally, there are many actors who are interested in channelling migrant money into development projects that benefit countries, regions, communities or individuals. Therefore, this study intends to produce knowledge that eventually contributes to a contemporary migration and development dialogue.

Even though these studies focus on many different impacts of remittances on human capital, some research gaps can be identified. There are no recent studies that have a broad, multi-dimensional understanding of human capital. The authors usually focus on one component of human capital. Furthermore, the majority of the introduced studies base their research on pre-existing data and have not gathered the data themselves. To the best of my knowledge, there is no study that applies a macro and micro perspective (mixed methods approach) to support

(19)

their argument on the impact of remittances on human capital. Finally, as opposed to other studies, my qualitative analysis provides information that goes beyond an econometric model;

the case study on Puebla informs about the complexity related to local remittances inflows.

Human Capital

Reviewing literature on the relation remittances and human capital shows that the latter is more than just economic productivity. The papers being discussed focus on different aspects of human capital. Social scientists conduct research on various affected levels: Educational attendance and attainment, school enrolment, years of schooling, child labour and investments, secondary education, health and labour force in general. In other words, many studies operationalise human capital into educational variables. However, Stroombergen & Nana (2002, p.12) propose that education (in human capital) is also understood as skills, knowledge or competitiveness of a country. Further, they see a challenge in a time horizon that measures the impact of education in human capital adequately. The quantitative analysis of this study acknowledges this time horizon challenge and takes its limitations into consideration (Chapter 4.5).

Similarly to Stroombergen & Nana, this study also assumes that human capital is more than just technical inputs and outputs (blackbox). This research builds upon the assumption that remittance income is developmentally more sustainable when invested than consumed. Among the many developmental effects that remittance income may have on economics, institutions or politics, it focuses on human capital. Chapter 2.2 introduced the idea that human capital has evolved and that it reaches into different disciplines. Human capital has been approached from many angles; the Organisation for Economic Co-operation and Development (OECD) holds a holistic but general view of the concept. The OECD report (OECD 2013a) on measuring human capital identifies 4 pillars: Education, Health & Wellness, Workforce & Employment and Enabling Environment pillar. Using these four pillars the OECD constitutes the Human Capital Index. This index designs effective measures for workforce planning and also serves as a basis of dialogue for the World Economic Forum (WEF) (OECD 2013a, p.3). The index has recently been established and therewith accesses recent data. However it is not able to the show development over a longer period of time. The extensive project embraces 51 indicators. With this index, the OECD acknowledges the importance of human capital formation. However the index has a strong economic component

(20)

as it aims to disclose the evolvement of a global economy, prosperity and competitiveness of nations (OECD 2013a, p.3). Further, within the scope of my analysis, the 51 indicators provide redundant and farfetched information. Therefore the Human Capital Index will not be incorporated as such. Nevertheless, its division into pillars of education, health, employability and enabling environment is inspiring. The first three of these pillars serve as three dimensions of the concept of human capital. The fourth pillar of enabling environment will not be taken into consideration because firstly, there were no adequate variables found for operationalising this pillar (for the relevant time period). Secondly, this study does not believe that its influence is fundamental for the human capital formation.

The educational attribute of human capital appears to be the most well known. As outlined before, education is commonly used as a direct translation for human capital. People that are educated and that are going to elementary, secondary or higher level of schooling yield economic returns to their community, village or country. Apart from the indicators mentioned in the literature review, years of schooling and enrolment are popular indicators for education. One of the Millennium Development Goals (MDG) that run out this year has aimed towards universal primary education. This remains an important goal alongside expanding secondary and tertiary enrolments. Among the regions where there are huge enrolment gains in secondary education, Latin America's achievement from 38% (1975) to 90% (2005) was the highest (Perkins et al. 2013). Nevertheless, years of schooling and gross enrolment do not indicate the quality of education. How well pupils are taught and how much they learn is crucial to how long they attend school (EFA Global Monitoring Report 2005).

Another measurement for education and human capital is education attainment. It refers to the levels of education completed (primary, secondary and tertiary education). Educational attainment is used to identify individuals that hold a broader knowledge base, more specialised skills in order to estimate the potential of a (future) labour force (OECD 2013b). The variable that is used in the quantitative analysis of my study is the secondary school completion rate. Firstly, it aims to target the quantity of pupils that attend secondary school. Secondly, the primary school is more likely to be compulsory and therefore sheds less relevant results for this study. Unfortunately, the issue of school quality is not addressed in my study, as data on teachers' qualification, PISA results or internet access in school is scarce or non-existing.

(21)

Another pillar of human capital and associated with education is labour and employability. Education is responsible for attaining the skills and knowledge of human capital. Labour makes individuals specialise their abilities and apply the knowledge to return profits. Naturally, employment and labour are connected to many economic benefits. However employment is also a strong contribution to individuals' well-being. The labour market participation as part of human capital is therefore included in the quantitative analysis.

A third pillar mentioned by the 2013 OECD report is health. This pillar is also suitable for the conceptualisation of human capital. Health contributes to the sustainability of human capital. In the human capital discussion, health has earned less significance than labour and education. This is partially due to the external influence that health is believed to have on human capital. Grossmann (1972, p.224) finds that the stock of knowledge affects the market or nonmarket productivity, whereby the stock of health defines the amount of time of productivity. Healthier workers tend to be more productive because they are more energetic and they lose fewer workdays due to illness (Perkins et al. 2013, p.319). As mentioned in the literature review, health is a pivotal aspect of human capital. Health is therefore also included in my operationalisation of human capital.

Remittances and Human Capital in Latin America

As outlined in the introduction, Latin America is an important remittances receiving region. Remittances income in Latin America has suffered during the global financial crisis in 2008/2009 and is now recovering again. The total remittance income in Mexico with 22 billion USD (in 2013) was still around 30% lower that in 2006 (Cohn et al. 2013). The following Figure 1 illustrates the remittances income in comparison with the Official Development Assistance to Latin America. After the decline for one two years (after the global financial crisis), both curves increase again. Figure 2 shows that Mexico merely increases the remittances and ODA. The high remittance income of Mexico does not distort the evolvement of the curves. When comparing the y-axis (USD in Millions) of the two figures, it becomes clear how big the Mexican share of total remittances actually is. Further, the figures illustrate the huge difference between remittances and the ODA.

Notwithstanding the fall of total Mexican remittances income within the past decade, the country remains the biggest remittances recipient of entire Latin America (40% of Latin

(22)

American remittances income). Mexico's high share of emigrants (11.9 million) is responsible for the biggest migration corridor Mexico-USA worldwide (World Bank 2011). According to PewHispanic (Cohn et al. 2013) it is due to the proximity to the United States of America that there are high emigration and large remittance income numbers in Mexico. Unsurprisingly, Mexico globally ranks fourth on the list of top remittance-receiving countries (Multilateral Investment Fund 2015). In regard of the total remittances share, Mexico is - with the exception of India, China and the Philippines - the most interesting case when focusing on migration and remittance income. This considerable amount of money and the latent migration raise many developmental questions in this region. It is for this reason, that the

(23)

Figure 2: Remittances and ODA sent to Latin America, without Mexico (1990 - 2012) Figure 1: Remittances and ODA sent to Latin American (1990 - 2012)

(24)

2.4 Hypothesis and Sub questions

The research question on how remittances contribute to human capital formation in Latin America and Mexico will be answered applying a mixed methods approach. The quantitative analysis contributes a answer on a macro level, whereas the case study answers the question on a micro level. Many of the discussed studies argue, that remittances are more so invested

than merely consumed. Consequently, these studies declare that remittance contribute to

sustainable development when invested in human capital. Due to this convincing evidence, I

side with this argument by introducing the first sub question of this study:

SQ1: How do remittances contribute to a stronger human capital formation in Latin American countries?

The panel analysis will answer this sub question with the help of the following hypothesis:

H1: The higher the amount of remittance income, the stronger the human capital formation in Latin American countries.

This hypothesis applies to Latin American countries and through a cross-country analysis it yields information on the relation between remittances and human capital formation. The analysis uses a similar conceptualisation as the Human Capital Index but it applies a timeline of 22 years (1990 - 2012). Furthermore, qualitative data on Mexico helps to go beyond identifying potential causalities and specific effects. The second part of the study introduces individual opinions and experiences on the complex matter.

As outlined in the literature review, the effects of remittances trigger controversies among the scientific community. The fact that scientists cannot agree on either consumption or investment will be utilised in this study, by taking a clear side in this discussion. First of all, this study approaches this debate from a new angle; a combination of a macro and micro angle questions the developmental sustainability of remittances. However the controversy on remittance usage implies that other influencing variables must be included (control variables). The additional benefit from the case study will embrace further influences such as the socio-economic situation that is believed to mediate the relationship between remittances and human capital. On the macro level, the study assumes, that remittances foster human capital.

(25)

On the micro level (case study), more explicit sub questions address this remittances ambiguity and explore the phenomenon of remittances income influencing human capital:

SQ2: How does remittances income contribute to improved human capital formation of Pueblan remittances recipients?

SQ3: What externalities influence remittances expenditure of remittances recipients in Puebla?

SQ4: How strong is the developmental effect of remittances on the human capital of remittances recipients in Puebla?

2.5 Limitations of the Study

This thesis contributes to the developmental dialogue of International Development Studies. I am highly aware that the study and its impacts are limited, for example when considering the data collection. Accessing already existing data has high potential but also drawbacks. I was not able to verify the conditions under which the quantitative data has been collected (World Bank data, UN data). Furthermore, the data about remittances and migration only yields information on registered or officially known figures. Even though migrants tend to be involved in the informal economy, "grey data" is not included in the study, which ultimately affects the explanatory power of the thesis. Additionally, the quantitative and qualitative analyses are subject to multiple limitations. The latter is discussed in chapter 5.4. Firstly, as for the panel analysis, the gender dimension is not included. Some studies (Acosta et al. 2007; Calero et al. 2009) point out that remittances money has different effects on women than on men. For data availability reasons, the gender specification cannot be considered. Secondly, I support the assumption that the remittances expenditure varies for regular and irregular remittances recipients. It is possible to include this in the qualitative analysis. Whereas the quantitative data fails to include this discrepancy as it is impossible to collect large data on regular and irregular remittances income. Thirdly, the quantitative analysis suffers from potential endogeneity; the direction of causality is assumed to go from the independent variable (remittances income) to the dependent variable (human capital). However it is important to note that human capital can also influence migration (negatively) and consequently remittances income (negatively). Also, increased remittances go along with enhanced human capital until reaching turning point when human capital can be no further increased. The relationship develops along an inverted U-shaped curve.

(26)

3. Methodology

The research question of this study can be best answered by applying mixed methods. Elaborating on the strengths and weaknesses of either the methods and the advantages of mixed approach ascertains triangulation. Combining the strengths of both the methods allows also the outweighing of their weaknesses. According to Johnson & Onwuegbuzie (2004) pragmatic quantitative research has - among others - the following advantages: Quantitative research traditionally receives more attention due to more credibility of numerical data. Furthermore, the quantitative results are independent from the researcher, as they are gathered from other sources (databases). The data used has been collected in costly in wide scale projects that could not have been conducted in the course of this study. Finally as the research question aims to detect phenomena across a large number of people, a cross-country analysis is a clear advantage. Nevertheless, Johnson & Onwuegbuzie (2004) call attention to the weaknesses of quantitative research, it fails to include local customs for example. Quantitative research tends to generalise and is not useful to obtain individual information. Additionally, the knowledge produced is more applicable to abstract mechanisms rather than specific contexts (Johnson & Onwuegbuzie 2004, p.19). In order to zoom in on local conditions a qualitative approach is more informative. Qualitative research with the case study of Puebla allows for in-depth insights, describing idiosyncrasies and serves the purpose of obtaining detailed information. Qualitative research focussing on a single case can answer the question of how a phenomenon evolves.

Transferred to the mixed methods approach of this study, it means that the quantitative part reproduces and supplements a strong theoretical argument discussed in the previous section. On macro level, the quantitative analysis questions the relation of remittances and human capital anew. The independence from the data ensures high objectivity on the regression results. The quantitative analysis manages to pin down the general process of remittances and the impact on human capital for many countries. However in alignment with the mixed methods approach, the quantitative results are not enough to entirely understand the complexity of the phenomenon. Therefore, it is important to include the case study that approaches the research question from a different angle. It includes perspectives, stories and experiences from Pueblan migrants that complement the quantitative analysis. The case study provides valuable micro information and contextualises the quantitative results.

(27)

The composition of both methods contributes to the validity and accuracy of the results. The mixed methods approach allows multiple insights on the same object of research. Moreover, stronger evidence for a conclusion through the convergence of different findings is a clear strength for Johnson & Onwuegbuzie (2004). The complete knowledge produced through both studies is a valuable contribution to the current scientific debate.

Figure 3 serves as visualisation of the research and the main concepts of this study. It presents

the overarching research question on how remittances contribute to human capital formation

in Latin America and Mexico. The vertical line in the middle separates the quantitative from

the qualitative analysis. To the left, the quantitative analysis in shape of a grey arrow, tests SQ1 (H1), whether a higher amount of remittances goes along with stronger human capital formation. Human capital is presented on the top of three pillars (1,2,3) that stand for the trident of human capital. The outer left lists the six control variables that are included in the analysis. On the side of Mexico, the second grey arrow substitutes SQ2, on how remittances

(28)

income contributes to human capital formation in Puebla. To the right, a smaller white arrow stands for SQ3 that wants to find out what externalities influence remittances expenditure. Lastly, a blue arrow in the bottom right presents SQ4, that questions the strength of the developmental effect of remittances on human capital in Puebla. The arrow heads almost outside of the illustration: It shows that the amplitude of the developmental effect is unknown.

3.1 Operationalisation

The operationalisation is divided into two parts: One part describes describes variables for the quantitative analysis, and the other explains indicators for the qualitative research. Chart 1 operationalises the two level of analysis and the concepts into dimensions and variables. When operationalising the concept human capital, the variables cover the dimensions discussed in chapter 2.4. There are many possible indicators for education (see chapter 2.3). Inspired by the studies from Zhunio et al. (2012) and Gyimah-Brempong & Asiedu (2009), I choose secondary completion rate as an indicator.

Both studies conclude remittances increase human capital formation (in 69 low- and middle-income countries and Ghana). Nevertheless, the authors measure the influence of remittances on human capital in secondary enrolment and primary completion rates. I am strongly convinced that the influence on remittances money can still be measured in secondary school and therefore decide to measure the influence on secondary completion rate. Moreover, Zhunio et al. (2012, p.4609) argue that secondary school is less likely to be compulsory, which could imply that school costs are not covered by governments but need to be paid privately. Subsequently, I expect remittance money being spent on secondary school.

As discussed in the literature review, Hildebrandt & Mc Kenzie (2005) and Lopez Cordova (2005) find a positive effect of remittances on health. In this analysis, health is translated into private health expenditure, as this is the most straightforward indicator. The assumption underlying this operationalisation is that remittances can alleviate financial constraints and therefore flow directly into health expenditure.

(29)

Level of Analysis

Concept Dimension Variable Indicator

Panel Analysis Remittances Recipients Remittances Income Remittances income (% of GDP)

Human Capital Education Secondary Completion Rate Lower secondary completion rate, total (% of relevant age group), weighted average

Health Private Health Expenditure Health Expenditure private in % GDP, weighted average

Employability Labour Participation Labour force participation rate, total (% of total population ages 15+)

Case Study Remittances Money Regular Remittance Income Frequency of Remittances Income Irregular Remittances Income Frequency of Remittances Income Mean of Transferral Private or Public Institutions

(Bank)

Mediating Actors and Institutions*

Families Family members involved in Remittances Transferral Money Transfer Institutions, Private Companies

Use and Purpose* _Income _{Form of Expenditure}

Savings No Remittances Expenditure

Development Impact* _{Poverty Alleviation} _{Changes in socio-economic Status}

Social Development Expenditure according to Kinship Logic*

Human Capital Life in Mexico Education Expenditure for Education Health Expenditure for Health Labour Expenditure for Labour

Development Impact* Human Capital Investments Expenditure for Human Capital related Investment

Context for Migration

Family Family Circle Family Members of Remittances Recipient

Kinship Logic Family Structure, Closeness to other Family Members Communication Frequency and Means of

Communication

Life in the US Group Affiliation Returning to Mexico / Staying abroad

Time Period Frequency and Duration of Stay abroad

Citizenship US / Dual Citizenship

Migration Reasons & Causes Origination of Migration Legal Status No / Residence Permit

*_{adapted from Goldring Typology (2004)}

(30)

Lastly, remittances allegedly create more opportunities for self-employment. In alignment with the findings from Acosta (2006) and Woodruff & Zenteno (2007), I expect remittances money to have a positive influence on labour force participation. The indicator for this dimension is therefore labour force participation rate.

As for the case study, the concepts are also split up into human capital and remittances. Remittances are divided into four dimensions. "Money" as the central theme of the interview guide, provides information on the regularity of remittances income. Regular and Irregular remittances income are expected to influence human capital formation differently. The remaining three dimensions ("Mediating Actors and Institutions", "Use and Purpose", "Development Impact") are listed in anticipation of the Goldring Typology (2004, p. 812), introduced in Chapter 5. Questions on the first dimension yield information about actors and institutions involved in remittances transferral. The dimension "Use and Purpose" covers

questions on remittances expenditure as either income or savings.5 The remittances dimension

"Development Impact" targets the developmental effect of remittances from sub question 4. In contrast to the panel analysis, the case study operationalises human capital into two dimensions. The theme "Life in Mexico" embraces questions on the education, health and labour of the interview respondents, analogue to the conceptualisation in the panel analysis. This study assumes that investments are developmentally more sustainable than consumption. The dimension "Development Impact" of the concept human capital therefore seeks to identify a developmental effect through human capital investments.

The additional concept "Context for Migration" splits up into three topics of "Family", "Life in US" and "Migration". These three dimensions are expected to influence the relation remittances and human capital and are therefore relevant to the operationalisation. The dimensions are necessary to understand the background and living conditions of the respondents. The concept "Context for Migration" covers three of the five interview guide themes. The dimension "Family" includes questions on the family structure of the respondents. "Kinship Logic" and "Communication" reveal further relevant information and highlight the importance of interfamilial relations of Pueblan respondents. "Life in the US" as another contextual dimension, embraces "Group Affiliation", "Time Period" and

5_{Even though "savings" is the opposite of remittances expenditure, it is assigned to this category. Remittances} expenditure embraces how recipients deal with the money. In other words, where the remittance money goes

(31)

"Citizenship". These three variables offer a realistic view of the life of Mexican migrants in the US. Lastly, the dimension "Migration" with the two variables "Reasons & Causes for Migration" and "Legal Status" targets questions on the migration condition for remittances consignors. The qualitative analysis shows that the introduced dimensions are closely interlinked.

(32)

4. Stronger Human Capital Formation

For the quantitative part a regression analysis on Latin American countries is applied. The analysis aims to detect a relation of remittances and human capital. The analysed time period embraces the years 1990 until 2012, when the latest evaluation was published.

4.1 Data

International institutions such as World Bank and United Nations Organisation keep annual record of development data and financial transactions globally. Remittances and migration are important topics in development debates, unsurprisingly there exists vast knowledge and the available data is extensive. As this study is built upon two methods, two different sets of data are necessary. The data collection for the quantitative method was already conducted by precedent projects. The variables are retrieved from the most complete or most significant data sets of the two databases from the World Bank and the UN. Additionally, the Center for Systemic Peace provides the Polity IV data. The data for the regression analysis is therefore a combination of the best available variables from the three databases.6_{An overview on all} variables for Mexico in the relevant time period is in the appendices (Appendix V).

4.2 Cases

For the statistical analysis the cases of the following 20 Latin American countries are used:

Argentina, Bolivia, Brazil, Chile, Costa Rica, Colombia, Dominican Republic, Ecuador, El Salvador, Guatemala, Guyana, Haiti, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, Uruguay, Venezuela. To approach the analysis holistically, it originally included all

countries of the Latin American region defined by the UNESCO (2015).7 However the list is

adapted to include only countries that yield data on all variables of interest.

4.3 Variables

The variables for the regression analysis are selected according to the operationalisation scheme. Whereas the operationalisation of remittances is a straightforward process, a translation of human capital into variables is more demanding. The empirical model includes

6_{Using data from different databases does not pose a problem - World Bank and UN mutually exchange their}

data.

7_{Data for the following countries was missing in either one or more of the relevant databases: Antigua and}

(33)

three indicators for the concept. In order to represent each dimension of human capital the following dependent variables are used.

Dependent Variables

Education is measured in lower secondary completion rate (variable name: seccompletion). This variable is obtained from the World Bank database and it is the quotient of the new entrants in the last grade of lower secondary education (regardless of age) in relation to the population at the entrance age for the last grade of lower secondary education (World Bank 2015). In other words, it is the number of all new entrants in the last grade of lower secondary education of all the same-aged population in a country. The ratio scaled variable displays the data in percentages. This variable brings along a significant number of missing values in the data set. The missing values for the seccompletion variable obstruct certain measuring methods, hamper highly representative results obtained with this variable and cause a loss of information. However a comparison to the UNESCO database reassures the choice of World Bank data: UNESCO neither provides education completion data before 1999, nor secondary completion rates at all. Missing values in data sets are a common problem therefore many solutions are provided. A popular solution is multiple imputation that is applied for one of control variables (variable name: migration). This is not an option for seccompletion as the missing values are not regularly distributed (or regularly missing).

Employment, as another dimension that measures human capital is represented as labour participation (variable name: laborpart). As well as the variable for education, labour participation data is retrieved from the World Bank database. However it is remodelled International Labour Organzation (ILO) data. Laborpart informs about the labour force participation of country populations older than 15 years. It displays the data on the supply of available labour for producing goods and services in an economy. It includes people who are currently employed or first-time job-seekers (World Bank 2015). The dataset is complete and does not entail any missing values. The ratio scaled variable appears in percentages of the active populations in the labour market.

Health, as the next dimension that measures human capital, is represented as private health expenditure in percentage to the total GDP (variable name: privatehealth). The World Bank database retrieves the data from the World Health Organization (WHO). Privatehealth provides information on private health expenditures in relation to the GDP of a country. The variable is ratio scaled and it includes expenditure on direct household spending, private

How remittances contribute to Human Capital Formation in Latin America and Mexico

The Money from the other Side