The geographic mobility of labor : An investigation of the role of wages and unemployment rates in the migration process

While research on the geographic mobility of labor has as its basic objective the evaluation of the labor market's efficiency in reallocating manpower resources, previous economic empirical research has concentracted on identification and quantification of the determinants of the volume of movement among areas. There is fairly wide agreement among economists that people move in response to economic incentives, but there is no concensus on the relationships themselves. Previous empirical research has suggested that clarification is needed with respect to the role of wages and unemployment rates in the migration decision. Some researchers have suggested that the migration process is asymmetrical, i.e., out-migration is a demographic phenomenon; and economic factors enter only the decision which affect the migrant's choice of a destination. But the results of previous research are not clear about how wages and unemployment affect these decisions: directly as the classical theory of factor mobility suggests, indirectly through the potential migrant's expectation of the supply of labor in alternative destinations or not at all as the job vacancy analysis suggests. The purpose of this investigation was to clarify and quantify the relationship between migration and local labor market conditions, i.e., wages and unemployment rates, by systematically reclassifying migrants into more homogeneous, noncompeting groups. The procedure was suggested by recent developments in human capital theory which have developed the costs and returns analysis. This analysis, which considers the migration decision as an investment, reconciles the individual decision framework with the characteristics of specific groups. It also suggests that the migration responses among groups may differ so extensively that the analysis of aggregate data will yield biased if not meaningless results. The analytical technique consisted of fitting gross in- and out-migration data for migrants classified into noncompeting groups to a series of linear single equation models. The data employed were cross-sectional and referred to the events during the interval 1955-60. Observations were obtained for 89 of the largest 101 standard metropolitan statistical areas for which migration data and labor market data were available. The role of wages and unemployment rates in the migration process has not been resolved by this investigation, although the results provide some argument for the classification of migrants into more homogeneous groups. The basic conclusions were: (1) the findings confirm the existence of an aggregation problem in migration analysis. Economic factors, e.g,, wage levels and unemployment rates, enter only the decisions which affect the individual's choice of a destination. The importance of these factors, however, varies so extensively among particular migrant groups that aggregation of these groups presents problems which cannot be ignored. These findings render suspect the behavioral implications of those investigations which have analyzed aggregate data. (2) The findings also confirm the asymmetry of the migration process. Rather than depending on economic factors, the decision to leave, even for fairly well-defined groups, appears to be largely a function of age. This finding replicates the results of other researchers with a slightly different body of data.