In this paper, I use longitudinal data from the 1998 and 2001 Living Standard Measurement Surveys in Nicaragua to examine the impact of the emigration of household members on the household labor market integration and poverty. The main findings of the paper are that households from which an emigrant left had a reduction in members, a reduction in working members, a reduction in labor income than otherwise similar households. However, those households also had a reduction in poverty. This finding is a result of the different patterns of migration from Nicaragua to the United States and Costa Rica. Households with emigrants to Costa Rica tend to be poorer, to have emigrants that were working prior to migration, and to have the greatest relative improvements in poverty following emigration.

The focus on short-term macroeconomic factors, including unemployment and wages, is insufficient to explain international migration. Institutional factors, bound to change only in the long run, can potentially have a large impact on migration flows. To illustrate this, we analyze Mexico-U.S. migration focusing on social security coverage, an important indicator of job formality. Using retrospective longitudinal data from the Mexican Migration Project, we find that workers are more likely to migrate to the United States when they lack social security coverage, suggesting that job formality discourages international migration. By old age, a history of short-term or moderate migration does not seem to significantly improve a worker’s prospects of exiting the labor force. However, substantial migration experience (10 years or more) does help workers without social security contributions match the retirement prospects of nonmigrants with social security coverage, indicating that long-term migration experience effectively acts as a substitute for social security.

We explore the time pattern of remittances using data on return migrants from the Mexican Migration Project. Some of these return migrants have settled in the U.S. and are returning to Mexico to visit family and friends, whereas others are temporary migrants returning home after a working spell in the U.S. We find that the dollar amount remitted first increases with time spent in the U.S. to later on decline after five and a half years of U.S. experience, lending support to the existence of an inverted “U” time pattern in migrant remittances. Furthermore, as expected, remittance decay occurs at a faster rate for migrants maintaining weaker ties to Mexico and up to three years later for their counterparts with spouses back in their origin communities.

We seek to further understand the factors that determine per emigrant remittances using data from 23 Latin American and Caribbean countries over the 1980-2003 period. We find that emigrants avoid remitting when the exchange rate is under pressure. This finding is consistent with the notion that remitters strive to reduce their exposure to exchange rate losses by taking into account the expected future value of current flows to the home country. Such a finding is important because it implies that remittances are not necessarily a stable source of external finance. Our result is robust to corrections for endogeneity with respect to the exchange rate variable. We also find that geography in the form of distance helps predict the flow of per emigrant remittances.

International migrants are persons who cross national borders and remain outside their countries of birth or citizenship for 12 months or more, regardless of the reason for being abroad or legal status while abroad. According to UN estimates, the number of international migrants was 191 million in 2005, and half were in the labor force of the destination country. International labor migration affects both receiving and sending countries, and most of the impacts of emigration on sending countries are captured in the so-called 3-R channels, viz, recruitment, remittances, and returns. Recruitment in Mexico and Turkey began with US and German government approval of the movement of unskilled workers, In both countries, labor flows got larger and lasted longer than anticipated because of distortion and dependence: some employers made investment decisions that assumed migrants would continue to be available and some migrants and regions became dependent on foreign jobs and remittances. There was a decade pause in Mexico-US migration between the mid-1960s and mid-1970s, while there was a shift from labor migration in the early 1970s to family unification in the late 1970s in the case of Turkey-Germany migration. The governments of both Mexico and Turkey see emigration as a short-term safety valve, but the two countries have different policies toward current and future migrants. The Mexican government, which often says that it represents 105 million Mexicans in Mexico and another 25 million abroad (half children born to Mexicans abroad), anticipates continued emigration and rising remittances, and has made policies to improve conditions for Mexicans abroad a top priority. The Turkish government, on the other hand, downplays potential emigration as it seeks entry into the European Union (EU), emphasizing that few Turks are leaving for foreign jobs and that the infusion of foreign and local investment with EU entry will speed up economic and job growth and encourage most Turks to stay home when they have the freedom to seek jobs elsewhere in the EU.