The Determinants of Remittances: Evidence from Moldova

Abstract

This article explores the factors that account for the receipt of remittances by households in Moldova that have members who have migrated abroad, mostly temporarily. We approach our research question from the perspective of the recipient household and use it to interpret the econometric findings on the determinants of remittances. Our results show that a combination of household and migrant characteristics and some community-level variables are the key elements in explaining remittance behaviour. Drawing on these estimates, we argue that the results point towards altruism and investment as the two possible motives behind remittance flows to Moldova, which significantly affect the country's development prospects.

Notes

¹ According to Luecke et al., (2007), only 14% of the Moldovan labour migrants plan to settle abroad permanently.

² The total population of the country was around 3.6 million in 2008.

³ According to Dujisin (2006), some villages lost up to 40% of their active population. The sight of villages with only children and elderly became increasingly common in the independent Moldovan state.

⁴ It is important to note that some of the exhibited increase in remittance receipts may be a result of improvements in the measurement and reporting of remittances at various points in time.

⁵ The cost of migration to the EU in 2006 was $3584 compared with only $74 and $132 for CIS migrants in the construction industry and other sectors, respectively (Luecke et al., 2007).

⁶ See Docquier & Rapapport (2003) for a detailed review of the remittances motivations literature.

⁷ Once again, note that our analysis is based on the data collected in Moldova, i.e. from the recipients' point of view. We, therefore, use remittance receipts to interpret the determinants of remittance behaviour.

⁸ The sampling scheme is described in the Appendix.

⁹ Ex-household members are working-age migrants who are no longer a part of the household in Moldova, and may have since formed their own household abroad.

¹⁰ These include washing machine, microwave oven, television, cable television, TV satellite dish, home telephone, mobile phone, computer and gas.

¹¹ If the person responded that he or she trusted more than two financial institutions then that household was labelled as having trust in the institutions of the home country.

¹² We would like to emphasise here that there are several endogeneity issues that we are not able to capture in our data. Therefore, the results presented in this section should be interpreted as being based on correlation rather than causality. Even with this shortcoming, we argue that these correlations provide a useful insight into the different determinants of remittances, which could then be used to make some conclusions about possible motives behind these transfers.

¹³ We acknowledge that the altruism hypothesis makes rather precise predictions about the derivatives of transfers with respect to income; derivatives cannot be reckoned with the data here. However, an argument, at least on the basis of correlation, could be made about the possibility of the presence of the altruism motive here.

¹⁴ However, another way to interpret this result is that perhaps big families are more likely to have members with low human capital and are also more likely to send out more than one migrant. This could possibly generate the negative relationship that we observe.

¹⁵ The presence of networks helps the migrant to settle in smoothly in the host country and reduces the various costs (settlement costs, job-search costs, and period of job search) and hence risks in the host country. As a result, the migrant is able to save more and hence remit more.