38th Congress of the European Regional Science Association: "Europe Quo Vadis? - Regional Questions at the Turn of the Century", 28 August - 1 September 1998, Vienna, Austria
This paper reports the results of two years of research on the estimation of regional occupational employment migration rates and their influence on estimates of future job openings by occupational group. The first section provides a general description of state and metropolitan area migration by occupation and other demographic variables. The descriptive statistics demonstrate that the overall level of out-migration rates and their variation among demographic and occupational groups are quite large relative to current employment estimates and estimates of future job openings that currently do not take into account regional out-migration. The second section describes the construction, and use of, estimated adjusted out-migration rates. The adjusted out-migration rates are created by using incomplete data methods to statistically combine data from the 1990 Census and the 1987 Occupational Mobility Current Population Survey. This hybrid data set contains complete information on occupational migration and mobility and allows us to isolate the out-migration rate that reflects changes in state-of-residence but not changes in occupation. These adjusted rates eliminate potential double counting that would be introduced using unadjusted occupational out-migration rates. This application of incomplete data methods is "tested" by generating a set of adjusted migration rates for a region of the United States and applying these rates to estimates of base year regional occupation employment to produce a set of estimated job openings that take into account regional out-migration. The results show that the total number of estimated job openings by occupation have to be revised significantly upwards when out-migration is taken into account.