Remittance patterns in Eastern Europe and the World.

• Author: Parker, Elliott

1. Introduction

   In this study we examine macro-level determinants of migrant remittances for a sample of 93 countries, with a particular focus on how Eastern Europe and the Former Soviet Union might differ from other countries. Our dataset is a balanced panel running from 1995, a time after the formerly socialist economies had significantly liberalized, to 2019, the year before the CoVID-19 pandemic disrupted immigration patterns, economic growth, and the behavior of remittances. In the quarter century in between these events, enormous social changes took place in this region.

   We consider remittance flows both from the perspective of sending and receiving countries. Migrant remittances, defined as financial transfers between migration-sending and receiving countries, represent a significant flow of capital that dwarfs official development aid and are a more stable flow than direct foreign investment (Yang, 2011). According to the World Bank (2020), personal remittances received by all countries totaled 653 billion USD in 2019. Of course, these data only include reported remittances, and migrants may also be able to shift their savings informally, in ways that aren't well measured.

   Remittances are particularly significant for the former socialist economies of Eastern Europe (EE) and the Former Soviet Union (FSU) following the dissolution of the Soviet economic system in the early 1990s and the transition to more market-based economies. International migration flows within the region, following the formation of new nations, the relaxation of border controls, and the integration of many EE countries into the European Union (EU), represent some of the world's largest mass movements of people. These flows can be broadly categorized bi-axially into a Western part of the region to the EU and another axis involving migration from the republics of the FSU, and thus have had a significant impact on economic growth and development (Mansoor and Quillin, 2006).

   In what follows we review the literature on migrant remittances, including both general considerations and trends specific to the EE/FSU region. We follow that with a description of our data and sample, operationalization of key variables, empirical modeling strategy, results, and conclusions.

   Our results show consistently that there is a diminishing effect to remittances; a country with a larger share of emigrants will receive fewer remittances per emigrant, and a country with a larger share of immigrants will pay fewer remittances per immigrant. There is also some evidence that concentration matters, in that remittances diminish when emigrants concentrate in fewer countries, or immigrants come from fewer countries. An overvalued exchange rate leads to more remittances paid, but does not significantly affect remittances received. Finally, there is evidence that the formerly socialist economies of Eastern Europe both receive less and pay less in remittances, once other factors are accounted for, but this does not appear to apply to other republics of the former Soviet Union.

2. Background

   Much of the macroeconomic research on migrant remittances has focused on its consequences for sending countries, such as poverty and inequality reduction (Adams and Page (2005); Mehedintu et al. (2019); Pedovic (2017)) and financial and economic development (Aggarwal et al. (2011); Guiliano and Ruiz-Arranz (2009)), however, in comparison to micro-level research, relatively less research has considered the macrolevel determinants of remittances. What limited research on the macroeconomic determinants of migrant remittances exists focuses on such factors as the number of workers, wage rates and overall economic situation in the host or origin country, the exchange rates and relative interest rate between the sending and receiving country, and transfer costs and exchange rate restrictions (see Hagen-Zanker and Siegel (2007) for review, or individual studies such as Freund and Spatafora (2008)). The determinants of migrant remittances have mostly been considered at the micro-level and include a number of underlying motivations for why such sending exists. Insofar as macro patterns represent the aggregation of micro-level processes, they can be understood from these perspectives.

   2.1. The New Economics of Migration

   A groundbreaking paper by Lucas and Stark (1985) introduced the idea of "tempered altruism" or "enlightened self-interest" to describe migrants' motivation for engaging in remittances exchanges with their origin households. Their approach represented a middle-ground between altruistic approaches to understanding remittances and pure self-interested motivations. From the perspective of the former, migrants derive positive utility from the consumption behavior of their families (in the Beckerian sense) and thus care about such factors as poverty, exogenous shocks, and factors associated with the origin households' well-being. According to the latter, migrants send remittances with the anticipation of gaining something in return, such as the inheritance of property. Thus, remittances can be used to entice migrants to remit back to their home households (de la Briere et al., 2002; Hoddinot, 1994), especially if they possess desirable assets, such as land, property, livestock, and the like.

   Under a "tempered altruism" motive, both altruism and self-interest play a role in remittances, in the sense that migration and remittances are part of a kind of mutually beneficial implicit contract. Such a contract is self-enforcing and takes the form of a coinsurance scheme. For one example, the migrant helps protect the origin household against shocks and economic downturns, and the household protects the migrant from the risk of unemployment and offers a potential return option). This may also involve loan repayments, including investments in the migrant's education (Brown and Poirine, 2005), or an exchange of services such as childcare provided by extended family to absent migrant parents (Cox, 1987).

   Lucas and Stark (1985) added an important element in the development of the so-called "New Economics of Labor Migration" (see Massey et al., 1993; Hagen-Zanker and Siegel, 2007; or Carling, 2008), which has a number of central premises,

1. Decisions about remittances are linked with decisions about migration and made at the household level. This idea contrasts with (but is not mutually exclusive of) other approaches to explaining migration, such as the Harris and Todaro (1970) model of migration, whereby migration is driven by expected (rather than actual) wage differentials between place of origin and destination.

2. Household strategies involving migration are a reaction to market failures (e.g., in credit, insurance, and futures markets) in source countries that prompt migrating household members to move to areas characterized by non-correlated labor markets (e.g., rural vs urban), creating opportunities for investment (i.e., in housing, small businesses and agricultural mechanization which improves farm productivity) and risk diversification.

   Thus, factors such as poor economic conditions at origin (e.g., poverty, high unemployment, low wages, or lack of access to capital) represent the underlying factors encouraging migration, and migrants act as target earners that make up for market failures. Later work in this tradition also linked household remittances to both relative and absolute income considerations (Stark and Taylor, 1989), suggesting that a sense of relative deprivation is an important consideration in household migration decisions.

   2.2. Remittance Decay

   Migration origin (demand-side) factors aside, research has also considered temporal changes in remittance patterns, which inevitably involves considerations about migration-destination (or supply-side) factors. Specifically, a major theme in the remittance literature revolves around the so-called remittance decay hypothesis, the notion that remittances decline with the passage of time. Arguments in favor of remittance decay point to such mechanisms as weakening altruism as family ties loosen over time and distance (Farzanegan et al., 2017), reunification with dependents in the host country resulting in increasingly fewer needy recipients to assist at home (Poirine, 2006), and greater integration and earnings of migrants in the host communities (Cohen, 2011). Empirical evidence for decay is mixed (Carling, 2008), with some research finding no time effect (e.g., Brown and Poirine, 2005) and much of the evidence in favor finding an inverted U-shaped pattern over time (e.g., Makina and Masenge, 2015; Poirine and Dropsy, 2019), whereby remittances increase and peak within the first 6-8 years of absence and then gradually decline.

   However, interpretation of time effects in econometric analysis is sensitive to the inclusion of other variables in analysis (such as controls for transnational family structure and migrant income at destination) and is further complicated by the difficulty of disentangling effects of age, length of stay, and period of migration; although Amuedo-Dorantes and Pozo (2006) were able to differentiate between effects of period of entry and length of residence in their study of Mexican migrants in the United States. Furthermore, as demonstrated by a study of Southeast Asian migrants in the UK (Arun and Ulku, 2011), the decay phenomenon can differ across ethnic groups, with for example, strong evidence for remittance decay for Indian and Pakistani migrants, but not for Bangladeshi migrants.

   The implication for our research is that not only are economic conditions at origin of importance to understanding remittance patterns at the macro level, but also central are destination-context factors such as the size, composition, and human capital endowments of migrants, as well as changes over time.

   2.3. Migrations and Remittances in Eastern Europe and the Former Soviet Union

   We now consider how some of...