Econometric search of undocumented workers in the USA

Abstract

This study seeks to find where undocumented workers reside in the USA. By postulating that the US destination of an undocumented worker is affected by socio-economic factors, an equation is estimated using the standard panel-data methods. A static panel-data estimation (fixed-effects and random effects) does not reveal a meaningful relation between undocumented workers and other factors. The Arellano-Bond estimation, in which the previous level of undocumented workers are used as a proxy for a network, provides statistically significant relations at the 0.05 level; the number of undocumented workers in a state is determined by the state's socio-economic factors such as economic conditions (e.g., unemployment rate and household income) and social factors necessary to form a network (e.g., foreign-born population, Hispanic population, urban population, the number of households having limited English-speaking ability and the previous level of undocumented workers). Among these factors, the previous level of undocumented workers has the largest impact on the current level of undocumented workers.