Influences on rural migrant workers' selection of employment location in the mountainous and upland areas of Sichuan, China

This study examines the current employment location selection of rural migrant workers in mountainous and upland areas of Sichuan, China. The analysis employs both representative survey data of 400 households and geographical data calculated using a 30 m Digital Elevation Model (DEM) and Geographic Information System (GIS). A binary and multinomial logistic regression model is used to analyse the influences on employment location selection of rural migrant workers, where the factors considered include personal, household, and community characteristics as well as natural and employment environments. Dividing off-farm employment locations into five categories, we find that 14.98% of rural migrant workers migrated to their home village; 10.98% migrated out of their home village but remained in their home town; 12.81% migrated out of their home town but remained within their home county; 15.47% migrated out of their home county but remained within their home province; and 45.76% migrated out of their home province. Employment location selection of rural migrant workers is found to be significantly influenced by the travel time required to reach a town, the cultivated land area per capita of a worker's household, the worker's age, whether an employer provides housing or meals, and the RDLS (relief degree of land surface) of the worker's home village. Gender is found to affect the likelihood of labourers taking off-farm employment in their home villages but does not appear to influence movement to other migrant locations. A multinomial regression approach is undertaken to analyse rural out-migration to the five migrant locations considered, an approach that reveals considerable heterogeneity that is concealed by the dichotomous approach employed in most previous studies. The study thus contributes to our understanding of rural out-migration in mountainous and upland areas. (C) 2013 Elsevier Ltd. All rights reserved.