The labor market effects of Venezuelan migration to Colombia : reconciling conflicting results

The recent mass migration of Venezuelans to Colombia has become a focal point for economists interested in the labor market effects of migration in developing countries. Existing papers studying this migration wave have consistently found negative effects on the hourly wages of native Colombians, which are most concentrated among less-educated natives working in the informal sector. However, the magnitude and significance of this wage effect varies substantially across papers. I explore the potential specification choices that drive this variation. Differences in how migration is measured are particularly important: exclusion of a subset of migrants from the migration measure, according to characteristics such as time of arrival, amounts to an omitted-variable bias that will tend to inflate the estimated wage effect. In my own analysis based on the total migration rate across 79 metropolitan areas and by using an instrument based on historical migrant locations, I estimate a native hourly wage effect of -1.05% from a 1 percentage point increase in the migrant share or an effect of -0.59% after controlling for regional time trends, alongside little-to-no effect on native employment. Native movements across occupation skill groups and geography are small and do not play a meaningful role in mitigating local wage effects. Wage effects are also larger in cities that have a higher baseline informality rate and lower ease of starting a business.