The research program is putting a lot of effort and resources in generating new data on household and migrant characteristics and on remittances. This is being done through household surveys that include a migration and remittance module. Household surveys
These are being used to assess empirically the impact of remittances on poverty and inequality in migrant-sending countries, as well as on expenditures on education, health, housing, entrepreneurship, and more. So far, the empirical work in this area is being based on household surveys that were carried out by Bank staff (e.g., for Pakistan), by consultants (e.g., for Mexico, Norway and Albania), by official statistical agencies (e.g., for the Philippines), and by the Bank’s Living Standards Measurement Studies (e.g., for Guatemala and Ghana), while new household surveys are also being conducted for Morocco, New Zealand and Samoa. Endogeneity issues associated with estimating the effects of migration and remittances were dealt with in most studies, either by instrumental variable estimation (IV) or by selecting a unique experiment to examine these effects. The latter includes examining the impact of the Asia crisis and the exogenous changes in exchange rates (Philippines) and of a lottery system (for migration from Samoa to New Zealand).
Selected Studies
A few studies are described below. This page covers only some of the initial papers prepared by consultants or Bank staff participating in the project. Additional papers are being prepared for several of the countries being examined. Impact of migration, remittances on poverty and spending structure of households (Guatemala)Determinants of migration, impact of migration on inequality (Mexico)  Impact of migration, remittances on poverty, income distribution, and local development (Morocco)Impact of New Zealand 'lottery' immigration policy on accepted, unaccepted households and households that did not apply (Samoa, New Zealand)
Impact of migration, remittances on poverty and spending structure of households (Guatemala)
Using data from the 2000 national household (hh) survey conducted by the Instituto Nacional de Estadistica in Guatemala, Richard Adams assesses the impact of migration and remittances on poverty and on the spending structure of hhs by comparing the marginal expenditure patterns of remittance-receiving households with the non-receiving households. A limitation of the data is that they do not include information on migrant characteristics. The author corrects for hh selection bias by using a two-step Heckman-type selection correction procedure. He finds that, though both internal and international remittances reduce the level of poverty (measured by poverty headcount), the reduction in the severity or depth of poverty (measured by the poverty gap or squared poverty gap) is much stronger. He also finds that, compared to non-receiving hhs, those receiving remittances spend at the margin less on consumption goods and more on housing and education. Compared to non-receiving hhs, those receiving internal remittances spend at the margin 45% more on education and those receiving international remittances spend 58% more. Adams also examined the impact of both migration and remittances in a study of 71 developing countries (paper co-authored with John Page) , and finds that both reduce poverty. The results of OLS and IV estimation are very similar. A 10% increase in international migrants (as a share of the origin country’s population) lowers the share of those living on less than a dollar per person per day by 2.1%, while a 10% increase in per capita remittances lowers the share of those living on less than a dollar per person per day by 3.5%. Back to selected studies Determinants of migration, impact of migration on inequality (Mexico)Â
Jorge Mora and Edward Taylor are preparing a study based on data on rural Mexico from the 2003 Mexico National Rural Household Survey that collected information on migrant, hh and community characteristics, as well as on internal and international remittances. Their first paper provides preliminary results on the determinants of migration as a function of the three sets of characteristics. Their second paper provides preliminary findings which indicate that international remittances slightly increase rural income inequalities (measured by the Gini correlation of international remittances with total income), while internal remittances are income equalizers. However, both types of remittances have an equalizing effect on incomes in high-migration areas. Moreover, international remittances reduce rural poverty (measured by a modified Foster-Greer-Thorbecke poverty index) by a greater amount than internal remittances. The larger the share of hhs with migrants in a region, the more favorable the effect that changes in remittances have on rural poverty. Back to selected studies Impact of migration, remittances on poverty, income distribution, and local development (Morocco)Since very little work has been done on international migration and remittance issues in the Middle East/North Africa, Richard Adams is currently working with the Moroccan government to implement a nationally-representative survey of 5000 households in rural and urban Morocco. The purpose of this survey is to identify the impact of international migration and remittances on poverty, income distribution and local development in a major labor-exporting country in the Middle East. One important aspect of this work will be to identify how the receipt of remittances affects the level of household expenditures on education, health and investment in local business. Back to selected studies Impact of New Zealand 'lottery' immigration policy on accepted, unaccepted households and households that did not apply (Samoa, New Zealand)
David McKenzie and co-authors are preparing hh surveys in Samoa and New Zealand that exploit a unique experiment. New Zealand immigration policy allows entry of an annual fixed quota of Samoans who fulfill minimal age and English language restrictions. All qualifying Samoans can apply and the selection occurs through a lottery. The research project consists of collecting hh and individual data of: a) households with family member(s) accepted under the quota, b) hhs which applied for the quota but were not successful, and c) hhs that did not apply. The project will also collect hh and individual data of Samoan migrants in New Zealand and Australia to assess the impact of migration on the migrants themselves. Back to selected studies
Improvement of remittance records in national statistics In addition to this microeconomic work on remittances, the World Bank is taking the lead jointly with the IMF of a working group of statistical agencies to better record remittances in national statistics. This working group has been formed at the request of the G8. Improving statistical standards in this area is a long-haul project. In the short-term, teams are trying to get a better grasp of the extent of remittances by surveying central banks. The World Bank's Development Prospects Group initially published some work on the evolution of the size of remittance flows to developing countries (see latest Global Development Finance report] and this work is ongoing. Moreover, work that had been initiated in 2003 to assess the macroeconomic effects of remittances is being continued, focusing on four aspects: a) the sustainability of a country’s creditworthiness and access to international capital markets; b) the stability brought by remittances to the balance of payments; c) the effect on export competitiveness through the real exchange rate changes; and d) the cyclicality of remittances. This work will be presented in the next Global Economic Prospects report (to be published in November 2005). The GEP will also include some policy conclusions on how to reduce the costs of sending remittances and increasing the transparency of the financial sector.
Documents
The policy research working papers below are drawn from the World Bank's archives. |
RSS