The effect of remittances on social expenditures in Central and Eastern Europe

(1)

The effect of remittances on social expenditures

in Central and Eastern Europe

A comparative study between Hungary and the Czech Republic

By: Alexandra Vágó

S2085364

Supervisor: Dr. Alexandre Afonso

Second reader: Dr. Johan Christensen

Leiden University,

The Institute of Public Administration,

Public Administration MSc,

Economics and Governance track,

(2)

2

Figures and tables

Graph 1: Development of emigrations in Hungary and the Czech Republic (1950-2018). ... 36

Graph 2: Evolution of total population in Hungary and the Czech Republic (1950-2018). ... 38

Graph 3: Population growth rate in Hungary and the Czech Republic from (1960-2018). ... 38

Graph 4: Personal remittances received in Hungary and the Czech Republic as a percentage of GDP 42 Graph 5: Inflow of remittances in Hungary and the Czech Republic in billion current USD per year (1995-2018). ... 43

Graph 6: Percentage of expenditure on social protection in Hungary and the Czech Republic as a share of GDP. ... 68

Graph 7: Social protection benefits per inhabitant in Hungary and the Czech Republic. ... 49

Graph 8: Unemployment spending as percentage of total government spending in Hungary and the Czech Republic. ... 69

Graph 9: Unemployment rate in Hungary and the Czech Republic. ... 51

Figure 1: Hungarian’s approval of leadership of the European Union and Viktor Orbán ... 52

Graph 1: Development of emigrations in Hungary and the Czech Republic (1950-2018) ... 56

Graph 2: Evolution of total population in Hungary and the Czech Republic (1950-2018). ... 59

Graph 3: Population growth rate in Hungary and the Czech Republic (1960-2018). ... 59

Graph 4: Personal remittances received in Hungary and the Czech Republic as a percentage of GDP 63 Graph 5: Inflow of remittances in Hungary and the Czech Republic in billion current USD per year (1995-2018). ... 64

Graph 6: Percentage of expenditure on social protection in Hungary and the Czech Republic as a share of GDP. ... 68

Graph 7: Social protection benefits per inhabitant in Hungary and the Czech Republic. ... 69

Graph 8: Unemployment spending as percentage of total government spending in Hungary and the Czech Republic. ... 69

Graph 9: Unemployment rate in Hungary and the Czech Republic. ... 70

(4)

4

Introduction

In 2019, the annual flows of remittances to low- and middle-income countries are expected to amount up to 550 billion dollars, which is larger than foreign direct investment and official development assistance in low- and middle-income countries (World Bank Group, 2019). The flow of remittance increased from 2017 to 2018, with 9.6% per cent, particularly in Europe and Central Asia and South Asia (World Bank Group, 2019). For developing countries, remittances represent a source of financial inflow that has become a primary source of external finance (Page & Sonia, 2018; Piracha & León-Ledesma, 2004). Due to the extent of these remittance flows, researchers and governments from developing and developed countries have been conducting considerable research on the impact of remittances in the receiving and sending countries (Page & Sonia, 2018). Many scholars argue that remittances have poverty reducing effects and stimulate GDP growth, while other scholars have found no direct link between the inflow of remittance and GDP growth. Particularly, Doyle (2015) concluded from his research on 18 Latin American countries between 1990 and 2009, that as the inflow of remittances increases support for leftist politics and social expenditures decrease. Thus, Doyle (2015) argued that countries receiving repeated inflows of remittances, will experience an increase in support for right-wing policies, which over time would reduce social expenditures.

Even though after the Iron Curtain fell and the following quarter century was characterized by continuous east-west migration flows, the majority of the research on remittances has been mainly focused on Latin America. During the regime change in Central and Eastern Europe, masses of jobs disappeared, which led to high unemployment and might have pushed labour emigration (Organisciak-Krzykowska, 2017; Orenstein, 2013). As a result, outmigration started to appear, particularly due to free movement of labour opportunities created by the European enlargement (Organisciak-Krzykowska, 2017). The largest European enlargement took place in 2004, when 10 countries joined the European Union; Estonia, Cyprus, Lithuania, Latvia, Malta, Poland, Slovenia, Slovakia, Hungary and the Czech Republic (Organisciak-Krzykowska, 2017). Bulgaria and Romania joined the European Union in 2007 and Croatia in 2013. Many of the emigrants were highly educated and seem to be permanent, while only a limited number of emigrants return to their home country. (IMF, July 2016). While economic factors such as a higher income per capita drive migration from the post-Soviet states, European enlargement and therefore the freedom of movement has had a significant impact on the social

(5)

5 and economic effects of the migrant sending and receiving countries (The World Bank, 2007; Goschin, 2014).

As there already been significant research done on the effects of remittances on poverty, economic growth and automatic stabilization, this study attempts to bridge the gap in research by focusing on the Central and Eastern European region. Therefore, the goal of this study is to try to answer the main research question: how do remittances affect social expenditures in Central and Eastern Europe? Central and Eastern European countries share historical and political experiences, because of direct or indirect influence of the Soviet Union (Romaniuk & Szromek, 2016). As the Soviet Union fell, countries in this region started to transform politically, economically and socially by implementing broadly similar directions of reforms, that do vary in terms of range and depth across countries (Romaniuk & Szromek, 2016).

The first chapter of this study will introduce the concepts utilized in this paper, followed by a literature review on these concepts as well as the hypothesized relationships developing from the literature.

The second chapter will explain in more detail this study’s case selection and research design. This chapter will clarify in a detailed manner how this study’s variables will be measured and collected. The last section of this chapter will discuss the study’s limitations.

Chapter three of this study will include a brief overview of the Central and Eastern European welfare state, as well as emigration patterns and remittance growth. This chapter will give the reader more context regarding the case selection and the similarities (as well as differences) before and after the transition period, due to the shared history and importance of this history in terms of social reforms.

Chapter four will describe the Hungarian case, starting with a historical overview of emigration developments and the population decline. This chapter will furthermore trace the inflow of remittances over time. The third section of this chapter includes a historical overview of social reforms and social expenditures, as well as the unemployment rate and unemployment spending. The last section of this chapter will try to grasp public and political discourse in order to be able to answer the second hypothesis.

Chapter five in turn will explore the Czech case, beginning with an in depth explanation of how the Czech Republic changed from a migrant sending country into a migrant receiving country

(6)

6 and how this affected the population. The second section of this chapter will track inflow of remittances, followed by a section on social reforms and social expenditures. The fourth section additionally includes unemployment spending as well as the unemployment rate. The fourth section will include qualitative information on the current public and political discourse. The findings from chapter four and five will be analysed in chapter six, in order to confirm or disconfirm this study’s hypotheses and to draw conclusions about this study’s research question. The relation between emigration, population and unemployment will be analysed in the first section of chapter six. The second section includes an analysis on the first hypothesised relationship between remittances and social expenditures, while the last section will analyse the possible relationship between remittances and support for right-wing policies. At last, a concluding chapter will examine and review the whole study, discuss limitations and suggest possible further research on this subject.

(7)

7

Chapter 1: Theoretical framework

This chapter will introduce the most important concepts of this paper and continue with a literature review on the discussion of the current literature on these concepts and the expectations developing from this. In the first part of this chapter the concepts of remittances, the welfare state and social expenditures will be explored in more detail, in order to be able to answer the research question of this paper. The second part of this chapter will review multiple theories on these concepts and will serve as a theoretical foundation of the possible causal relationship between the concepts researched in this paper: the inflow of remittances and the receiving country’s social expenditures.

1.1 Conceptualisation

1.1.1 Remitttances

Personal remittances are defined by the World Bank as “the sum of personal transfers and compensation of employees” (the World Bank, n.d.). Personal transfers contain any transfer made in cash or in kind, independent of the relationship between the sender and receiver and the origin of the sender’s income (the World Bank, n.d.). Secondly, compensation of employees includes the salary of workers who are seasonal, border or short-term employed in a foreign economy (the World Bank, n.d.). The International Monetary Fund (IMF) defines remittances as household income that stems from the temporary or permanent movement of people to foreign economies (IMF, 2009). Remittances can furthermore be interpreted as a stable form of cross-border financial flow, that in developing countries can serve as an important function of earnings and insurance for households receiving remittances (Doyle, 2015). According to Page and Plaza (2018), in most developing countries the volume of remittances exceeds the volume of foreign aid and investments. The development of remittances could therefore be expressed as a new and extending financial connection between developed and developing countries (Page & Plaza, Migration remittances and development: a review of global evidence, 2018). Remitting behaviour depends on many different factors such as age, education, employment, gender, household size, credit access and the number of years since migration (Page & Plaza, Migration remittances and development: a review of global evidence, 2018). Mosley and Singer (2015), more explicitly explain how people active in foreign economies act as agents in the global stream of finance. When workers temporarily or permanently move to another country, money stemming from income generated in the foreign economy is frequently sent back to their

(8)

8 home countries (Mosley & Singer, 2015). These remittances function not only as financial support for their families but have considerable effects on the remittance receiving country’s aggregate level (Mosley & Singer, 2015). Remittances differentiate from other capital flows, because remittances do not create debt or any other obligations and thereby have the ability to provide institutions with financial access that might otherwise not be available (The World Bank, 2006). Remittances are furthermore a considerable and stable source of income for many households in the Eastern-European region (The World Bank, 2006). The macro-economic effects of remittances on economic development and economic stability are widely discussed in the current literature and will be explored in more detail in the literature review.

1.1.2. The welfare state

After the second world war, social security programs that provided housing, health care and income security became fundamental aspects of advanced industrial democracies (Pierson, 1994). The welfare state played an essential role in the exceptional economic growth in all the advanced industrial democracies, that lasted for a quarter of a century (Pierson, 1994). Moreover, Pierson (1994) argues that the welfare state did not only improve the economy, but also caused political changes. Social expenditures became an essential tool of macro- and microeconomic policy and the assurance of social protection increased the legitimacy of Western democracies (Pierson, 1994). Such assurances of social protection assisted workers to adapt to the developing market conditions and additionally strengthened wage restraint (Pierson, 1994). Esping-Andersen (1996) additionally argues that the welfare state meant more than the enhancement of social policies in post-war Europe. Therefore, Esping-Andersen (1996) defines the post war European welfare state in both economic and moral terms:

“In economic terms, the extension of income and employment security as a citizen's right meant a deliberate departure from the orthodoxies of the pure market. In moral terms, the welfare state promised a more universal, classless justice and solidarity of 'the people'; it was presented as a ray of hope to those who were asked to sacrifice for the common good in the war effort.” (Esping-Andersen, 1996, p. 2)

Daram Ghai, director of the United Nations Research Institute for Social Development (1996), defines the welfare state as a result of centuries long striving for social security and social protection in industrialized countries. Ghai states that this striving may even be seen as “one of their proudest achievements in the post-war period” (Ghai, Daram; UNRISD, 1996, p. 7).

(9)

9 According to Lauzaddyte-Tutliene, Balezentis and Goculenko (2018), the welfare state’s main task is to establish social protection as well as high standards of living. Depending on how countries determine to establish these goals and which policies are therefore realized, distinctive welfare regimes can be noted (Lauzaddyte-Tutliene et al., 2018). These distinctive welfare regimes and typologies that have been introduced by multiple scholars and researchers, will be explained in more detail in the literature review.

Nevertheless, according to Esping-Andersen (1996) the present-day welfare state is struggling with two specific challenges. The first challenge consists of the continuing division between the developing society’s demands and liabilities, and the social welfare schemes in place to meet them (Esping-Andersen, 1996). The demands and liabilities regarding social protection schemes can change, due to for example the changing family structure and occupational structure (Esping-Andersen, 1996). The second challenge the welfare states faces, consists of external forces, such as economic and demographic trends, that influence the welfare state’s viability in the future (Esping-Andersen, 1996).

1.1.3. Social expenditures

The OECD (2003) defines social expenditures as expenditures that offer support, by providing social benefits to households and individuals, when circumstances occur that negatively affect the households’ or individuals’ welfare. Social benefits can be offered by public and private institutions, but do not include market transactions or transfers between households and individuals (OECD, 2003). John Baldock (2018) explains that redistribution is an essential condition to the OECD’s definition of social expenditures. Redistribution can mean for example, to increase income equality or to increase the access to education or healthcare (Baldock, 2018). Redistribution can furthermore be aimed at market failures, where social expenditures ascertain the redistribution of resources that due to the market failure(s) are currently allocated inefficiently (Baldock, 2018). Currently (2018) among the OECD countries, the highest share of social spending is spent on pensions and healthcare (Baldock, 2018). According to Dahl and van der Wel (2013), social expenditures are expected to improve the population’s health and thereby decrease health inequalities. Government social expenditures furthermore increase social integration and participation rates in lower social-economic groups and educational levels (Dahl & van der Wel, 2013). Moreover, health inequalities between men and women could also be influenced by social expenditures (Dahl & van der Wel, 2013). Lastly,

(10)

10 their study concluded that increasing social expenditures would lower educational inequalities for men and women (Dahl & van der Wel, 2013).

1.2 Literature review

1.2.1 Welfare state typologies

In the year 1974, R. M. Titmuss was one of the first scholars to develop a welfare state classification (Titmuss, 1974; Lauzaddyte-Tutliene, Balezentis, & Goculenko, 2018). Titmuss (1974) divided the welfare state classification in three different models that represent different considerations regarding work ethic, social welfare institutions and the role of the family:

▪ Model A: the residual welfare model

▪ Model B: the industrial achievement-performance model ▪ Model C: the institutional redistributive model

Model A assumes that an individual’s social need can be met either by the private market or the family, and that social policies should only be introduced when one of these two channels fail to supply an individual’s social need (Titmuss, 1974). In model B on the other hand, social welfare institutions play an important part as partners of the economy, in terms of performance and productivity, and social services are provided based on work performance, productivity and integrity (Titmuss, 1974). Lastly, model C offers social services on a universalistic basis, build upon the principle of social equality (Titmuss, 1974).

Esping-Andersen (1995) in turn developed a welfare state typology based on the degree of decommodification and social stratification. Decommodification can be defined as the degree to which one does not have to rely on the market in order to maintain a certain social standard of living, while stratification indicates the severity of redistribution and solidarity enforced by the welfare state (Esping-Andersen, 1995). Esping-Andersen (1995) divided countries in three theoretical models that describe the relationship of the state, family and the market: the liberal model, the conservative-corporatist model and the social-democratic model. The liberal model includes a low level of decommodification and noticeable stratification, since social benefits are generally low and only aimed at poor individuals, due to the market preference (Esping-Andersen, 1995). In the conservative-corporate model, on the other hand, there is a considerable priority for the family and church, which results in less importance of social services, a

(11)

11 moderate level of decommodification and stratification of society (Esping-Andersen, 1995). Stratification was additionally based on gender, as women were mainly responsible for supporting their traditional family at home (Esping-Andersen, 1995). The social democratic model seeks to establish high universal living standards and widely pursues the concept of equality (Esping-Andersen, 1995). Therefore, it includes a high level of decommodification and little stratification (Esping-Andersen, 1995). Yet, the high levels of social benefits additionally result in high taxes (Esping-Andersen, 1995). Esping-Andersen tested these three models by analysing the pensions, sickness benefits and unemployment benefits across 18 countries of the Organization for Economic Co-operation and Development (OECD).

Even though the welfare state typology by Esping-Andersen has been highly influential, it still received some criticism. Bambra (2005) argues that typologies focused on cash benefits, such as Esping-Andersen’s, ignore the actual delivery of social services, such as health care services, and the variation of emphasis of countries regarding cash benefits or welfare state services. Bambra (2005) analysed the same 18 OECD countries as Esping-Andersen, but included health care services in her study. Her analysis showed that after including health care services to Esping-Andersen’s analysis, not all countries could fit according to the three traditional models and therefore the typology had to be expanded to five welfare state models by including two subgroups: the liberal subgroup and the conservative subgroup (Bambra, 2005).

McMenamin (2003) conducted the first research that systematically analysed Central and Eastern European capitalist democracies. According to McMenamin (2003), existing studies provide variables and samples of countries that are unfortunately not able to offer a credible analysis of the uniqueness of Central and Eastern European countries. Therefore, McMenamin measured social welfare regimes, political institutions and economic systems across 22 capitalist democracies (including Hungary, Poland and the Czech Republic) (McMenamin, 2003). Interestingly, McMenamin’s research developed a consistent grouping for Central and Eastern European countries. When the capitalist democracies are grouped into two separate models, Central and Eastern European countries are grouped among the continental European countries, in liberal model (McMenamin, 2003). In the case of three or four separate models, Central and Eastern European countries belong with Greece, Portugal, Spain and Ireland (Mosley & Singer, 2015). If capitalist democracies are grouped into six or more models, Central and Eastern European countries form their own model (Mosley & Singer, 2015).

Fenger (2007) additionally argues that the traditional typology proposed by Esping-Andersen is not applicable to Central and Eastern European welfare states and therefore subdivided these

(12)

12 countries into three groups. Fenger argues that there are six types of welfare state models: the conservative-corporatist model, the social-democratic model, the liberal model, the former-USSR model, the post-communist European model and the developing welfare states model (Fenger, 2007). The former-USSR model includes Belarus, Estonia, Latvia, Lithuania, Russia and Ukraine and is characterized by the relatively low level of trust and social situation compared to Western European-models (Fenger, 2007). The post-communist European model resembles the former-USSR model, but differs in economic development, social situation and life expectancy scores, which are higher compared to the former-USSR model (Fenger, 2007). Additionally, according to Fenger (2007), post-communist European countries are generally more egalitarian than former-USSR countries. Bulgaria, Croatia, Czech Republic, Hungary, Poland and Slovakia belong to the post-communist European model (Fenger, 2007). Finally, the developing welfare states model includes Georgia, Romania and Moldova and symbolizes countries that are not yet mature welfare states and therefore still developing (Fenger, 2007). This is depicted by the relatively low life expectancy and social situation compared to other countries in the Central and Eastern European region (Fenger, 2007).

Lauzaddyte-Tutliene et al. (2018), moreover studied the distinctiveness of Central and Eastern European welfare state models. Their study analysed 29 European countries and eventually developed a model of five clusters: the Mediterranean welfare model, small European states welfare model, old European countries welfare model, the Eastern Europe welfare model and the Central Europe welfare model (Lauzaddyte-Tutliene et al., 2018). Central and Eastern European countries have been divided into two separate models, where the Eastern Europe welfare model includes Latvia, Estonia, Lithuania, Bulgaria and Romania, and the Central Europe welfare model includes Hungary, Czech Republic, Slovakia, Slovenia, Croatia and Poland (Lauzaddyte-Tutliene et al., 2018). According to Lauzaddyte-Tutliene et al. (2018), the Central Europe welfare model is more similar to the old European countries and Mediterranean countries, compared to the Eastern Europe welfare model. The Eastern Europe welfare model namely involves lower social contributions (as a share of GDP), leading to less labour market flexibility and more gender differences in unemployment, compared to the Central European welfare model (Lauzaddyte-Tutliene et al., 2018).

Bohle and Greskovits (2018) additionally define a separate welfare state model for the Visegrád countries (Hungary, the Czech Republic, Poland and the Slovak Republic), that is characterized by their orientation on social transfers for especially pensioners, instead of investment oriented. According to Bohle and Greskovits (2018), the Visegrád’s strong dependence towards

(13)

13 consumption-oriented welfare programs has on the one hand created a loyal voter base, but on the other hand generated social expenditures that hindered public budgets and caused unpopular retrenchment. Meanwhile, education, skill formation and research and development are neglected and hinder foreign direct investment (Bohle & Greskovits, 2018).

Furthermore, Arts and Gelissen (2002) argue that Esping-Andersen’s typology does not cover Mediterranean countries and neglects the gender-dimension by failing to mention the family’s place in the welfare state and the degree to which women are included in the labour market. Additionally, according to Arts and Gelissen (2002), the Antipodean countries (Australia and New Zealand) should form a separate group, since their social policies are more inclusive than the liberal welfare state. Therefore, Arts and Gelissen have added two models to the typology: the Latin rim (including Portugal, Italy and Spain) and the radical model (including the Antipodean countries) (Arts & Gelissen, 2002). Lastly, Arts and Gelissen (2002) argue that in real life countries are expected to show hybrid forms of the proposed models. The Netherlands is an example of a hybrid form between the first three models proposed by Esping-Andersen (Arts & Gelissen, 2002).

1.2.2. Remittances and economic development

As stated by Addison (2004), director of research at the Bank of Ghana, remittances positively influence the integration of developing countries into the global economy. For multiple generations, remittances have been a significant source of financial support for the migrant’s family in their home country, particularly in Africa (Addison, 2004). Yet, as migration continues to increase, the inflow of remittances increases as well and becomes a critical stream of foreign currency for many developing countries (Addison, 2004).

Comes et al. (2018) analysed the effects of remittances on economic growth in seven Central and Easter-European countries (Romania, Bulgaria, Croatia, Czech Republic, Hungary, Slovakia and Slovenia) and found that remittances have a positive impact on economic growth in these seven countries. Their most significant effect of remittances on GDP was found in the Czech Republic, Romania and Hungary (Comes et al., 2018). Gianetti et al. (2009) furthermore linked remittances to economic growth, since remittance flows increase consumption patterns in the migrant worker’s home country that stimulate the demand for jobs and infrastructures that eventually may improve economic growth in the community. Goshin (2014) moreover constructed two growth models to test whether remittances when treated as capital flows, would

(14)

14 have macroeconomic growth potential. Goshin (2014) tested both models on aggregate data covering ten Central and Eastern-European countries in the time period of 1996 – 2011. The main empirical result has been the significant positive influence of remittances on absolute and relative GDP growth in the selected ten Central and Eastern-European countries (Goschin, 2014). Even though the potential GDP in sending countries will decline due to the emigration of potential workers, the overall net effect remains positive. Thus, the amount of received remittances compensates for the loss of workforce in Central and Eastern-European countries (Goschin, 2014). Doyle (2015) furthermore argues that remittances contribute to an important source of foreign exchange in developing countries, where remittances lead to a growth of the overall income levels and economic security. Similarly, Addison (2004) argues that remittances have the ability to positively affect the receiving country’s economy through savings, investments, poverty relief and income distribution. According to Mosley and Singer (2015), remitances have an overall positive impact on the economic development in the migrant workers’ home country. Remittances are frequently viewed as a form of aid in developing countries that have troubles generating domestic jobs, because remittance inflows may increase spending, leading to an increasing demand in local service industries and construction (Mosley & Singer, 2015). Addison (2004) furthermore states that remittance inflows are becoming a critical element in most developing countries with flawed credit markets.

On the other hand, several studies have reported small or insignificant impact of remittances on

the migrant workers’ home country’s economy. Haller et al. (2018) have conducted a

comparative study on Romania and Bulgaria between 1990 and 2015. Their study aimed to search determinants of international migration within the 28 European Union countries and to analyse the impact of remittances on income inequality and economic growth (Haller et al., 2018). Haller et al. (2018) discovered no direct link between remittances received per capita and the growth of GDP per capita neither in Romania nor Bulgaria. Barajas et al. (2009) furthermore studied 84 remittance receiving countries in the period of 1970 till 2004 and found negative or insignificant growth effects of remittances. Insignificant growth effects can be explained by the growth-enhancing influence of remittances in receiving countries, flowing through multiple paths (Barajas et al., 2009). Because these remittances affect economic growth through different paths, remittances can positively and negatively influence long-term economic growth (Barajas et al., 2009). Rao and Hassan (2012) estimated growth effects of 40 remittance receiving countries and found small positive growth effects. This small indirect growth effect of remittances can be explained by four main channels: “these are volatility in

(15)

15 output, investment rate, developments in the financial sector and the exchange rate” (Rao & Hassan, 2012, p. 367). The largest effect on growth was affected by volatility, whereas the smallest influence has been the exchange rate (Rao & Hassan, 2012). Similarly, even though Piracha and León-Ledesma (2004) argue that remittances alleviate poverty for migrant households in Central and Eastern-European countries, their results found no economic impact of remittances on economic growth. Lastly, Chami et al. (2003) discovered that remittances introduce a moral-hazard problem, where migrant households who receive income from remittances may be motivated to reduce their own work effort because of their financial inflow from remittances. Furthermore, migrant households interact under economic uncertainty and information asymmetries (Chami et al., 2003). This moral-hazard problem may be severe enough to reduce economic activity that may lead to negative effect on economic growth (Chami et al., 2003). Therefore Chami et al. (2003) conclude that remittance do not constitute to a significant source of capital for economic development.

However, the positive effects of remittances on macroeconomic development may also be conditional to the receiving country’s economic and political situation. Giuliano and Ruiz-Arranz (2009) studied 100 developing countries between 1975 and 2002 and found that remittances are connected to economic growth, especially in less developed financial countries. In countries where the financial sector is less developed, remittances have evolved into an alternative for ineffective or even inexistent credit markets (Giuliano & Ruiz-Arranz, 2009). Due to this development, remittances contribute to the allocation of capital in developing countries in order to enhance economic growth (Giuliano & Ruiz-Arranz, 2009). Barajas et al. (2009) additionally support the importance of stable domestic institutions in remittance receiving countries, in order to be able to channel remittances into growth-enhancing activities. Barajas et al. (2009) explain the negative or insignificant growth effects their study found, by suggesting that many developing countries do not have stable institutions in place yet to channel remittances into growth-enhancing activities (Barajas et al., 2009). As reported by a World Bank study focused on migration in Eastern Europe (2006), the quality of political and economic institutions in the receiving country, conditions the impact of remittances on economic growth and development. In general, remittances are estimated to have a positive effect on the receiving country’s economic growth but are particularly conditioned on the quality of existing institutions (The World Bank, 2006).

(16)

16 1.2.3. Poverty reducing effects of remitt ances

While the welfare state functions to provide social protection and increases redistribution, by lowering income inequality and providing access to education or healthcare, remittances seem to fulfil similar functions, according to the following authors. Gianetti et al. (2009) examined the effects of remittances in Central-Eastern European countries (Slovenia, Poland, Czech Republic and Hungary) as well and found that remittances lead to a reduction of poverty. Remittances increased the standard of living at the household level and improved education, health, housing and family welfare (Gianetti et al., 2009). Chami et al. (2009) furthermore argue that remittances alleviate poverty in developing countries by providing an important source of income to households receiving remittances. León-Ledesma and Piracha (2004) additionally studied the effect of remittances on employment performance in Central and Eastern-European economies between 1990 and 1999 and found that remittances have proven to ease poverty for migrant households. Haller et al. (2018) found a positive relationship between received remittances and income inequality as well in Romania and Bulgaria. In 2015, the International Fund for Agricultural Development (IFAD) studied remittances in Europe and concluded that remittances increase living standards of the receiving families (IFAD, 2015). Remittances have the ability to take care of the migrant’s families in receiving countries and provide a form of social security net in times of emergencies (International Fund for Agricultural Development, June 2015). Remittances have furthermore been substantial in reducing poverty and equality and thereby reducing government incentives towards structural reforms (IMF, July 2016).

Addison (2004) additionally argues that remittances have a significant impact on poverty reduction and income and distribution effects. Remittances aid poverty relief by the remittance inflows from the migrants to their family in their home country (Addison, 2004). On the other hand, Addison (2004) argues that this macroeconomic effect of remittances aiding poverty relief may also generate dependence on the remittance inflows in the migrants’ home country. This dependence on remittance inflows could hinder people to alleviate poverty by education and by joining workforce (Addison, 2004). Nevertheless, this distributive effect of remittance inflows contributes to a significant dimension of the developmental effects of remittances on the receiving country’s economy, because remittances are able to increase the income levels of low-income families and eventually alleviate those families from poverty (Addison, 2004). Addison (2004) concludes that according to the Ghana Living Standard Survey, remittances have positively affected Ghana by reducing rural poverty.

(17)

17 According to Doyle (2015), remittances lead to a decrease in welfare and social expenditures, but meanwhile also lead to an increase in the overall levels of government expenditures. Doyle explains that countries who receive repeated remittance inflows will experience an increase in income and economic security (Doyle, 2015). This increase in the income risk will result in an increase in partisan politics, because households receiving remittances will moderate their support for taxation in order to fund welfare programs and thereby reduce their support for leftist parties (Doyle, 2015). Over time, the support for leftist parties and social expenditures will weaken and eventually social expenditures will be reduced via the electoral channel in democratic countries (Doyle, 2015). Doyle (2015) tested this hypothesis over a sample of 18 Latin American countries, between 1990 and 2009. His results indicated a strong support for this hypothesis, meaning that as remittances grow larger in Latin America, increasingly the support for social expenditures weakens as well (Doyle, 2015). Thus, countries that receive remittances will have lower social spending and a smaller welfare state due to an increase in partisan politics, that was created by the reduced income risks (Doyle, 2015). Even though remittances drive this “bottom-up causal mechanism where remittances alter or change individual preferences towards redistribution and taxation, democratic leaders, in fiscally constrained developing world countries, conscious of the substitution effect of remittances, may expedite this effect” (Doyle, 2015, p. 789).

1.2.4. Automatic stabili sers

Automatic stabilisers can be defined as “elements of fiscal policy which mitigate output fluctuations without discretionary government action” (Dolls, Fuest, & Peichl, 2012, p. 179). Automatic stabilisers do not require deliberate intervention by government and are therefore not subject to implementation lags (in't Veld, Larch, & Vandeweyer, 2013). In’t Veld et al. (2013) furthermore state that views very much diverge about which elements of the budget actually provide automatic stabilisation over the cycle. Nevertheless, according to in’t Veld et al. (2013), there is a notional understanding that automatic stabilizers involve budgetary arrangements that help smooth output without the explicit intervention of a country’s fiscal authority.

Two types of studies dominate the empirical literature on automatic stabilisers: macro and micro data studies (Dolls, Fuest, & Peichl, 2012). Dolls et al. (2012) explain how for example the IMF utilises macroeconomic data in order to assess automatic stabilisation, as for instance

(18)

18 revenue and expenditure to GDP ratios. Furthermore, studies that apply a more complex approach measure automatic stabilisation by examining the elasticity of for example social security contributions, unemployment benefit levels as well as multiple forms of taxes (Dolls, Fuest, & Peichl, 2012).

There are two ways to analyse automatic stabilisers, namely by examining their size and degree of output smoothing (in't Veld, Larch, & Vandeweyer, 2013). The size of automatic stabilisers is defined as the change in budget resulting from a change in economic activity and can be measured by two indicators: budgetary sensitivity and semi-elasticity (in't Veld, Larch, & Vandeweyer, 2013). Cottarelli and Fedelino (2010) additionally argue that automatic stabilisers can be assessed by examining the government size as an indicator. Therefore, the automatic stabilisers are dependent on the adaptation of revenues and spending as a response to shifts in the GDP (Cottarelli & Fedelino, 2010). This adaptation can furthermore be defined as the elasticity of revenues and spending in regard to the GDP rate (Cottarelli & Fedelino, 2010). As stated by Dolls et al. (2012), whether automatic stabilisers alleviate income shocks on household demand is dependent on two factors. The first factor is the tax and transfer system in a country, that regulates to what extent an economic shock to the gross income will affect the disposable income (Dolls, Fuest, & Peichl, 2012). Dolls et al. (2012) argue that a progressive tax will have considerable stronger stabilising effects compared to a proportional income tax. Secondly, automatic stabilisation furthermore depends on the connection between the present disposable income and the demand for goods and services (Dolls, Fuest, & Peichl, 2012). Unfortunately, examining automatic stabilisers in such detail is beyond the scope of this paper and will therefore not be utilised in this thesis.

1.2.5. The welfare state as automatic stabili ser

The welfare state has furthermore been treated as a “powerful countercyclical tool, producing deficits during recessionary periods and (at least in theory) surpluses during boom times” (Pierson, 1994, p. 2). In’t Veld et al. (2013) associate the size of the government with automatic

stabilisers as well. Often, the majority of government expenditures are not reduced during economic downturns or crises, or increased during economic upturns. This stability of government expenditures has an overall stabilising effect on the total output and therefore provide a stabilising effect (in't Veld, Larch, & Vandeweyer, 2013). Similarly, Cottarelli and Fedelino (2010) argue that when government size is greater, automatic stabilisation is more

(19)

19

extensive as well (Cottarelli & Fedelino, 2010). Starke, Klaasch and van Hooren (2014) similarly agree that smaller welfare states have small automatic stabilisers and therefore crisis responses in smaller welfare states are shaped by partisan politics, compared to more generous welfare states with highly developed automatic stabilisers where the direction of policy change is not subject to debate.

In the European Union, automatic stabilisers absorb about 38% of income shocks, however there is great heterogeneity among automatic stabilisers within Europe (Dolls, Fuest, & Peichl, 2012). Particularly, automatic stabilisers in Eastern and Southern European countries are lower compared to the Continental and Northern European countries (Dolls, Fuest, & Peichl, 2012). Dolls et al. (2012) explain, that this heterogeneity could have two possible explanations. Firstly, countries with lower incomes per capita often have smaller public sectors and therefore weaker automatic stabilisers, while countries with higher incomes will in turn have more automatic stabilisers (Dolls, Fuest, & Peichl, 2012). A second potential explanation could be that open economies tend to have weaker automatic stabilisers (Dolls, Fuest, & Peichl, 2012). However, Dolls et al. (2012) have not found a correlation between stabilisation and trade openness (trade to GDP ratios).

1.2.6. Stabilising effect of remittances

According to the World Bank (2006), remittances are the main source of income for many of the poorest countries in Eastern Europe and serve as a ‘cushion’ against economic and political instability. This means that remittances have the ability to work counter-cyclical during economic downturns (The World Bank, 2006). The World Bank (2006) has developed statistical estimations on the macroeconomic impact of remittances and their findings propose “a mild positive impact on long-term patterns of macroeconomic growth, while evidence on their impact on the distribution of poverty is mixed” (The World Bank, 2006, p. 61). However, a possible negative side-effect of remittances could be that remittances may increase pressure on the exchange rate and decrease the competitiveness of exports (The World Bank, 2006).

Mosley and Singer (2015) additionally argue that remittances are a stable form of external finance, that during economic downturns tend to cushion the economic effects in migrant households. Likewise, remittances could function as a safety net, where remittance inflows are increased when there is an economic downturn or household crisis (Mosley & Singer, 2015).

(20)

20 In a similar manner, Page and Plaza (2018) state that remittances can act counter-cyclically in the remittance receiving country. Often migrants increase the flow of remittances as a response to negative shocks or economic downturns and exchange rate developments (Page & Plaza, 2018). Therefore, their study concluded that, “the greater stability of remittance flows and their anti- cyclicality may contribute to the stability of recipient economies by compensating for foreign exchange losses due to macroeconomic shocks” (Page & Plaza, 2018: p. 283).

Addison provides us with another point of view on how remittances can have stabilising effects on the remittance receiving country’s economy. According to Addison (2004), remittances positively affect a country’s balance of payments account and can furthermore lessen any balance of payments crisis (Addison, 2004). Addison (2004) explains this mechanism by the growth effects of remittance inflows on national savings and investment rates, and because remittances contribute to the receiving country’s foreign exchange that can be used to finance crucial imports. The inflow of remittances is received by the remittance receiving countries as foreign savings and are therefore complementary to the national savings, leading to an increase of the total available resources for investment purposes (Addison, 2004). Particularly developing countries experience the positive effects of remittances, by for example serving the purpose of financing social investment activities such as the building of schools or clinics and the development of infrastructural projects (Addison, 2004). In 2003, the Bank of Ghana estimated that remittances were, after exports, the most significant resource of inflow (Addison, 2004). Furthermore, remittances in Ghana are also used to create small-businesses and thereby creating employment opportunities that are of particular importance for the youth in rural areas (Addison, 2004).

Another study found that remittances are a remarkably stable source of income and seem to be compensatory and countercyclically in nature (Chami, Hakura, & Montiel, 2009). Chami, Hakura and Montiel (2009) studied the impact of remittances on output stability and analysed a sample of 70 countries whose income depended on remittance inflows. The compensatory nature of remittances can be explained as the reaction of the flow of remittances to the general economic downturns or crises. Meaning, that the inflow of remittances rises when the home country’s economy suffers a downturn. Therefore, Chami et al. (2009) argue that remittance flows are macroeconomically stabilising in the same manner as countercyclical fiscal policy would be. Furthermore, Chami et al. (2009) found that remittances have a stabilising effect for all remittance receiving countries on average, however this stabilising effect was conditioned

(21)

21 on the size of remittance inflows. Countries that had received considerably large remittance inflows also enjoyed the stabilising effect of remittances, but this effect was weakened when the remittance inflows were considerably large (Chami, Hakura, & Montiel, 2009). That remittance inflows tend to be stabilising on the receiving countries’ economies on average, is especially of importance for developing countries, because remittance inflows could potentially stabilize GDP growth in these developing countries (Chami, Hakura, & Montiel, 2009).

1.3. Hypotheses

The welfare state and remittances seem to fulfil similar functions, especially when analysing automatic stabilisation and the reduction of poverty. When countries have high levels of social spending, the majority of these expenditures are generally not lowered during economic downturns or increased when the economy is growing (in't Veld, Larch, & Vandeweyer, 2013). According to in’t Veld et al. (2013) this unresponsiveness of social spending to economic down- or upturns provides a stabilising effect on a country’s national economy during economic developments. Unfortunately, examining the effect of remittances on automatic stabilisation is beyond the scope of this thesis and will therefore serve as an example on how the welfare state and remittances can fulfil similar functions in the remittance receiving country. However, since remittances positively effect living standards in the receiving countries and thereby reduce poverty and equality, government incentives of applying structural reforms or increasing social expenditures decrease. Remittances sustain migrant households in imperfect economical areas but could also result in cutbacks on social expenditures (Mosley & Singer, 2015; Doyle, 2015). This mechanism generates this thesis’ first hypothesis:

H1: Countries receiving larger flows of remittances have fewer social expenditures

Moreover, remittances may increase the support for social welfare cutbacks as well as support for right-wing policies too (Doyle, 2015). Doyle’s study (2015) concluded that repeated flows of remittances will lead to a reduction in support for social expenditures. From Doyle’s study the following expectation develops:

H2: Countries receiving repeated flows of remittances will experience an increase in support for right-wing policies

(22)

22

Chapter 2: Research design and methods

2.1 Case selection

It has been chosen to focus on the Central and Eastern European geographic area, because the majority of migrants from this region migrate to Western European countries and the development of migration has been unique to this region due to conditions of economic transitions and political and social liberalization (The World Bank, 2007). The OECD (2001) defines Central and Eastern European Countries, “as an OECD term for the group of countries comprising Albania, Bulgaria, Croatia, the Czech Republic, Hungary, Poland, Romania, the Slovak Republic, Slovenia and the three Baltic states: Estonia, Latvia and Lithuania” (OECD, 2001).

For this comparative case study two countries from Eastern- and Central Europe have been chosen in order to analyse the possible hypothesised causal influence of the explanatory variable on the outcome variable. These two specific cases are Hungary and the Czech Republic. Both countries have been chosen, because the explanatory variable, the level of received remittances, must differ across the cases in order to control for selection bias. Furthermore, a Most Similar Systems Design I utilises the logic of conditioning by a blocking strategy, meaning that the effect of a confounding variable will be blocked when observations have very similar characteristics, but different values of the main explanatory variable are matched and analysed (Toshkov, 2016). Since Hungary was the third largest receiver of remittances in Central and Eastern Europe (3.3% of GDP in 2017) and there is an advantage of Hungarian language skills by the researcher, Hungary was the first country to be selected. Because the percentage of received remittances must vary across cases in order to eliminate selection bias, Slovenia, Poland and the Czech Republic were considered for the comparative analysis. In order to control for the effect of confounding variables, the cases must have similar characteristics. The Czech Republic had the most similar percentage of social expenditures as Hungary as a percentage of the GDP, namely 18.9% (Hungary 19.2%) (Eurostat, 2018). While the percentage of social expenditures in Poland lies at 20.3% and 23.3% in Slovenia (Eurostat, 2018). On some countries, such as Albania and Croatia, data on social expenditures was unfortunately unavailable or consisted of considerable gaps. Moreover, both Hungary and the Czech Republic experienced the fall of the Soviet Union, joined the NATO in 1999, the European Union in 2004, share the same major religion (Christianity), have approximately the

(23)

23 same number of inhabitants (Hungary 9.6 million and Czech Republic 10.6 million) and both do not use the Euro as their currency (Worldometers, n.d.). The Most Similar Systems Design I will be explained in more detail in the next section.

2.2 Research design

This research follows a deductive logic, where a theory and multiple hypotheses will be tested by researching a selection of cases. A comparative research design has been chosen in order to analyse the possible causal relationship between remittances and social expenditures. A small-N comparative research or within-case analysis would be more appropriate when the aim would be to analyse a specific case or a considerably broad causal relationship, because the available observations and diversity would be limited. Moreover, large-N research designs gather quantitative data on a large number of cases but analyse relatively few aspects of each case. Therefore, a comparative research design will combine both research designs by analysing particular aspects of a small set of cases, while still trying to infer causation from the observations. Nevertheless, a limitation of small-N comparative research is generalization, because the research will only be conducted on a set of cases instead of probability samples. However, the goal of this study is to research and test theory on a possible causal relationship of remittances and social expenditures in the two cases, rather than generalization. Because cross-case analysis of comparative research can be rather weak in distinguishing association from causation, this study will utilize two levels of analysis: within-country analysis on remittances, emigration, social expenditures and discourse, as well as a cross-case analysis. Moreover, this research design will involve longitudinal data, because repeated observations over periods of time will be used for the same variables (remittances, emigration and social expenditures).

In this research a Most Similar Systems Design I will be utilized, where cases that vary in regard to the explanatory variable will test hypothesized causal relationships. This research design has been found the most appropriate for this study, given that it serves a deductive as well as a comparative purpose with a focus on one or a small number of hypotheses. The dependent variable in this study is the level of social expenditures and the independent variable is remittances, while unemployment is included as a confounding variable and emigration as a control variable. Because a Most Similar Systems Design is not able to accommodate complex

(24)

24 relationships, this type of research design will be combined with within-case analysis for both cases.

2.3 Measuring variables

This study’s independent variable, remittances, will be measured in two ways: by tracking received remittances as a percentage of Gross domestic product (GDP), as well as the received remittances in current USD from 1995 until 2018. Remittances relative to GDP will be utilized in this study, because it shows the dependence of remittances in the countries. Data on remittances in current USD has been included in this study, in order to analyse whether the development of remittances as a percentage of GDP is a result of the changing GDP or incoming remittances.

The dependent variable, social expenditures, will be measured by tracking both social protection benefits as a percentage of total government expenditures and the social protection benefits in Euro per inhabitant over time. The last measure is included in this study, because an increase in social protection benefits relative to total government expenditures, may be due to a decrease of total government expenditures or an increase in social protection benefits.

Unemployment, is a confounding variable in this study, since unemployment could also influence the dependent variable; social expenditures. In the absence of data on social expenditures per unemployed worker, the effect of unemployment on social protection expenditures will be analysed by including both unemployment spending as a percentage of total government spending and the unemployment rate in this study. The unemployment rate will be utilised complimentary to the data on unemployment spending, because a change in unemployment spending may be caused by a fluctuating unemployment rate or austerity measures.

Emigration is this study’s control variable and will be measured by tracking the number of emigrations over time, together with the total population and the population growth rate. Population data has been included in the analysis of emigration data, because the Central and Eastern European region has been characterized by a decrease of the population. The decreasing population may influence emigration numbers, while in turn emigration affects the total population as well. The population growth rate has been included in this study, because by

(25)

25 including the growth rate, one is able to quickly observe whether the population has been growing or decreasing and by how much.

2.4 Data collection

This study will collect two types of data, namely statistical data and in depth qualitative information. The first part of this subchapter will describe how and why this study will collect its statistical data. The second part will in turn explain how the qualitative data is gathered and analysed.

2.4.1 Statistical data collection

Data on emigration will be collected from multiple sources, as unfortunately no database contains emigration numbers from the time period 1950 – 2018. Therefore, the United Nations Database for emigration numbers from 1990 – 2013 will be utilised for comparable emigration data cross the Central and Eastern European region. In order to analyse emigration developments for the cases, data will be gathered from the Czech National Statistical Office and the Hungarian Statistical Office and combined with the United Nations data on emigrations. Unfortunately, the Hungarian Statistical Office only contains data on emigration from 2010 until 2018. In order to cover the period from 1950 – 1990, the data utilised by Péter Tóth (2003) in his paper on International Migration – Major Trends in Demography: A Hungarian Example will collected. Unfortunately, a gap in Hungarian emigration data still exists for the period 1989 - 1994. Data on total population for the period 1950 – 2019 (measured on the 1st_{of July) will} be collected by the United Nations Database. The population growth rate for the period 1960 – 2018 will be gathered from the World Bank, because the World Bank collects population data from multiple sources, such as the United Nations Population Division, national statistical offices and Eurostat.

Aggregate data on remittances from the World Bank’s World Development Indicator database from 1992 – 2018 will be collected in order to analyse the independent variable. Even though emigration will be analysed from 1950 onwards, data on remittances for both cases is unfortunately only available from 1995 onwards. The World Bank data on remittances is collected from the IMF Balance of Payment Statistics database and gap filled by World Bank

(26)

26 staff estimates (The World Bank, n.d.). The World Bank defines remittances as: “Personal remittances is the sum of personal transfers and compensation by employees” (The World Bank, n.d.). Compensation by employees is furthermore defined as “income of border, seasonal and other short-term workers who are employed in an economy where they are not resident and of residents employed by non-resident entities” (The World Bank, n.d.).

Aggregated data on social expenditures will be collected from Eurostat, as this database contains the most extensive data on European governmental expenditures. Social expenditure data from the Czech Republic is available from 1995 until 2017, while for Hungary this data is only available from 1999 (until 2017). Both unemployment spending as percentage of total government spending (1995-2017), as well as the unemployment rate (1999-2018) will also be gathered from Eurostat.

2.4.2. Qualitative data collect ion

The qualitative data in this study will be collected through secondary sources and includes literature written by academic researchers, historians, economists, political experts and policymakers who have researched Central Eastern European emigration, remittances and social expenditures. Information on the historical development of the Central and Eastern European welfare state and social expenses is vital to discover possible patterns of path dependency and to understand the context of the reforms that will be analysed in this study. For the collection of qualitative information on public discourse, secondary sources from the Factiva database have been utilised.

2.5 Limitations

The main limitation of this study is data collection. Particularly data on emigration has often at times been found unreliable, because it was incomplete or underestimated. The SEEMIG report, which is a European Union funded report analysing historical migratory developments, stated that after the 1980s, the reliability of national emigration statistics has deteriorated and therefore the process became impossible to measure (Godri, Soltész, & Bodacz-Nagy, 2013). Drbohlav (2004) additionally pointed out that the data from the Czech statistical office and the United Nations data differ, for the years they overlap. The data from the Hungarian statistical office also differ from the United Nations data for the years where there is overlap. Thus, even though

(27)

27 this study has tried to combine multiple data sources into one graphical data for the development of Hungarian and Czech emigration, the unstable reliability of the emigration data is important to keep in mind when analysing the data.

Additional gaps in data exist for social expenditures as a share of GDP and the social protection benefits per inhabitant in Euro for Hungary, for the period 1995-1999. This gap in data, limits the researcher’s ability to analyse and compare data on these issues with Czech data, which is available from 1995 onwards.

(28)

28

Chapter 3: A broad Central and Eastern European overview

This chapter will serve as an overview regarding the history of the Central and Eastern European welfare state, the transition period and social security reforms in Central and Eastern Europe. The second part of this chapter will describe migration patterns and the European accession in 2004, while the third and last section will focus on the development of remittances and remittance growth.

3.1 Short history of the Central and Eastern European welfare state

3.1.1 The old communist welfare regime

The old communist welfare regime represented three pillars: full and quasi-obligatory employment, broad and universalistic social insurance and a company-based system of services and benefits (Standing, 1996). Reforms that had taken place after the communist regime had fallen, deteriorated the first and the third pillar (Standing, 1996). Social protection in the old communist welfare regime included both a universalistic and employment-based form, where all citizens could benefit from consumer price subsidies and where social protection was secured when a person was employed through the system of ‘work collective’ (Standing, 1996). Standing (1996) explains that the ‘work collective’ system was a form of social policy based on full employment, where employment was guaranteed, contributed to the social norm and was regulated at a highly formal level in specific labour codes. Unemployment did exist but was not officially (Standing, 1996). The ‘work collective’ system eventually failed, because it lacked dynamic efficiency, due its heavy bureaucratic mechanisms that dispersed social services, and was therefore not able to adjust to the change according to the demand for social security (Standing, 1996).

3.1.2. The neo-liberal transition

After the Soviet Union fell, the countries in Central and Eastern Europe started adopting reforms in order to cope with the global economy (Standing, 1996). Since the old communist regime was based on the three pillars, most Central and Eastern European countries faced troubles with global competitiveness, low levels of labour productivity and outworn technologies (Standing, 1996). Standing (1996) described the reforms that took place as ‘shock therapy’, where prices were liberalised, a social security system was developed, and

(29)

29 privatisations took place on a considerable large scale. According to Orenstein (2013), international institutions, such as the G-24, IMF and World Bank, together with Western governments helped Central and Eastern European countries in order to implement neo-liberal ideologies during the transition period. Adascalitei (2012) furthermore states that international actors such as the IMF and World Bank have been one of the key influences on national policy reforms in Central and Eastern European countries. For example, the World Bank had been advocating the restructuring of old welfare programs and additionally assisted national governments financially in order to be able to do so (Adascalitei, 2012). These neo-liberal reforms started by the liberalisation of prices, liberalisation of international trade and deregulating the market price (Orenstein, 2013). This transition additionally led to one of the largest decreases in life expectancy (Orenstein, 2013). According to Orenstein (2013), neo-liberal policies, such as the elimination of food subsidies and mass privatisation, have contributed to the extensive decrease in life expectancy as well as the concern of political corruption. In turn, poverty and mortality rates accelerated and migration increased (Standing, 1996). The reforms resulted in a short transition recession due to large-scale unemployment, but eventually led to economic growth (Orenstein, 2013). According to Orenstein (2013), the Central and Eastern European region became one of the biggest regions for inward investment. While the economy started to grow and the aggregate living standards increased, many people however did not get the chance to share this as inequality and corruption started to increase (Orenstein, 2013). In many countries in Central and Eastern Europe criminal organisations and mafias infected the economic and political spheres in these countries (Orenstein, 2013). However, the period of economic growth ended when the global economic crisis in 2008 occurred and Central and Eastern Europe experienced one of the longest recessions and weakest recoveries of any developing region (Orenstein, 2013).

According to Orenstein (2013), politics in Central and Eastern Europe began to develop after the ‘shock therapy’, as interest groups started to oppose neo-liberal ideas. The first group of opponents of the liberal ideas consisted of right-wing nationalists, who perceived the neo-liberal reforms as a change in dictatorship, where they would be too dependent on Western countries (Orenstein, 2013). The second group advocated partial reforms, since these actors were able to profit from imperfect markets during the transition period by for example buying products for a low price due to the subsidising and selling these products abroad at international prices (Orenstein, 2013). The third and final group consisted of social democrats and socialists, who believed that the state should have a bigger role in regulating labour unions and welfare

The effect of remittances on social expenditures in Central and Eastern Europe