Essays on Migration and Remittances in the Philippines

Abstract

This thesis compiles three essays on migration and remittances in the Philippines. The first essay investigates the role of remittances as an informal risk-coping mechanism for families during a disaster, and explores the interaction between remittances, relief, and a region's infrastructure and institutions. The study employs a unique identification strategy by tracking the regions that were directly lying on the path of Typhoon Haiyan, one of the most devastating typhoons in Philippine history. Using a difference-in-difference setup, the remittance response was found to be an increase in incidence (number of households) but not necessarily in level. The extended analysis only provided suggestive evidence that remittances and relief generally act as substitutes but serve as complements during disasters. Moreover, remittances are higher in regions with good facilitators, but remittances tend to insure households against poor institutions during disasters. This study establishes remittances as both aid and insurance during times of disaster. The second essay delves into how culture, migrant networks, and immigration policies may influence the labor market assimilation of Filipino immigrants. This study departs from the conventional framework used in studying assimilation by focusing only on immigrants from one sending country but in different host countries. This emphasizes the effects of the host country institutions on wage and employment assimilation. It was found that larger similarity in culture, a larger migrant network, and more selective immigration policies reduce the wage gap relative to natives and improves the transferability of human capital, thus facilitating assimilation. However, assimilation is hampered in countries with greater culture diversity and fractionalization. The third essay estimates the spatial spillover effects of migration on education. This study uses a novel way in representing proximity and neighborhood effects by borrowing techniques from geography and spatial statistics. While causality is not argued, it is shown that the location of migrants and non-migrants within districts is quasi-random after controlling for sorting and correlated effects. The distance to the nearest migrant is calculated for each non-migrant household, and interspatial autocorrelation was used to identify whether non-migrants are parts of migrant clusters (hotspots). The results show that closer proximity to a migrant and being identified as part of a migrant cluster is associated with higher tertiary enrolment. This suggests that the prospect of migration is enough to generate a brain gain for communities in general. Findings are only suggestive that community remittances may be a channel. Show more