Search results

Policies to deter illegal entry and reduce the number of undocumented immigrants have a complex impact on migration patterns, border crossings, and duration. However, studies generally assume the method of crossing into the United States is exogenous with respect to migration duration. Using data from the Mexican Migration Project, this paper finds that the migrant's decision to hire a coyote (smuggler) to cross the border is endogenous with respect to duration. Instrumental variable estimates provide evidence that migrants who incur the cost of hiring a coyote have longer migration durations as they need to work longer in the United States. The migrants most likely to hire coyotes have less education, little migration experience, and/or come from rural communities. Results suggest that continuing to increase guest worker programs could actually decrease the number of Mexican immigrants in the United States by eliminating the need for coyotes which would reduce migration durations. This would better utilize the immigrant population in the United States by encouraging immigrants to stay while employed and to migrate home when unemployed, with the knowledge they can later return. Reducing coyote use would also reduce income flowing to Mexican cartels which have profited from human smuggling. (JEL O15, J61, J64)

The rapid growth of international migration and remittances has led to a wealth of research examining these trends. One of these lines of research attempts to determine the relationship between migration, remittances and the adoption of "Green Revolution" high yield variety (HYV) seeds. This literature proposes that migration and remittances may increase HYV use by reducing household risk and credit constraints. Given the large scale of both migration and remittances, getting an accurate measurement of these impacts is crucial to designing and implementing policies in rural areas. This paper examines the relationship between migration, remittances and agricultural technology. The credit and risk hypotheses are tested using data from the Mexican Migration Project. The main focus of the paper is examining the issue of endogeneity with respect to migration and remittances. When the probit approach is tested, evidence of endogeneity bias with respect to migration and remittances is found. The risk and credit hypotheses are then tested with two-stage and three-stage analyses, in order to address the problem of endogeneity. The results are found to be significantly different when methodologies are employed to deal with endogeneity; suggesting this is an issue that needs to be addressed. Overall, the results find some evidence to support both the credit and risk hypotheses; although the results for the risk hypothesis are more mixed.

This paper presents a model of migrant behavior in which remittances are treated as both a consumption transfer to households and as an alternative savings mechanism for migrants. This approach leads to the implication that migrants' remittance/savings behavior should be affected by the relative rate of return on their savings and on the savings of their remittance receiving households. Statistical support for this hypothesis is found using data from Mexican workers in the United States. The paper finds this evidence using selectivity corrected remittance and savings equations. These results hold when the equations are estimated separately or jointly. These results imply that developing access to better savings and investment mechanisms for households in Mexico may increase remittance inflows from household members that are migrants.