A Note on Industry Changes and Migration

Abstract

Despite the practical and theoretical importance of the phenomenon, differences in community growth rates have yet to be explained satisfactorily. Only a few studies have examined the relation between growth rates and other characteristics of communities, and the results have been uniformly disappointing, meaning that no close relationships have been discovered.' We are not, however, completely ignorant of the factors that produce differential community growth rates. For one thing, migration is now generally regarded as the key factor, at least in certain countries. As an illustration consider the relation between the 1950-60 total growth rate and the two components of the total growth rate-net migration and natural increase-for the 254 counties in Texas. Whereas the product-moment coefficient of correlation between the total growth rate and the natural increase rate is only .65, it is .95 between the total growth rate and the net migration rate.2 Further, since the correlation between net migration and natural increase is .40, there is reason to believe that migration had some influence on the 1950-60 natural increase rates in Texas counties. Finally, differences in migration rates are not as unpredictable as one might believe. Although no close relations between characteristics of communities and subsequent migration rates have been discovered, there is general agreement that the migration rate is closely linked to the expansion of a community's industries (i.e., an increase in employment) during the growth period. This observation may fall short of an adequate theory,3 but it does pinpoint the proximate cause of growth, at least of the migration component. Moreover, a close relation between the expansion of industries and migration can be demonstrated. As an illustration, let us consider non-metropolitan counties and Standard Metropolitan Statistical Areas in Texas over the years 1950-60. The coefficient of correlation between the net migration rates and the ratio of the 1960 to the 1950 labor force for these territorial units (N=246) is .919. Given the inevitable possibility of errors and deficiencies in migration rates or labor force data, or in both, the coefficient is as high as one could reasonably expect. Thus, the result in this case is consistent with the belief that migration and industry expansion are closely related, and it clearly suggests that an adequate explanation of industry expansion would be a major contribution to the further explanation of differences in migration rates.