The effects of migration on income: a micro study with Canadian data 1965-71.

Abstract. This paper presents a test of the human capital model applied to migration. Micro data are used to estimate earnings functions for 1971 which include variables associated with personal and labour market characteristics and also earnings at the start of the data period (1965 or 1966) to represent prior human capital. Migration activity is restricted to single moves which differentiate between short- and long-distance moves and between moves made early and late in the 1965-71 period. The testing process and the final estimates show that it is very difficult to detect a significantly positive effect of migration on income within a five-year time horizon. The immediate payoffs appear to be negative, but even some years after the move the average payoffs are extremely small. Married women experience strongly negative payoffs from long-distance moves made in the early part of the period. Married and single males, and persons in the lower half of the income distribution register positive payoffs after a few years. These weak results are not inconsistent with earlier work, but they do raise some important questions about the strong conclusions often drawn from empirical work on migration patterns. [ABSTRACT FROM AUTHOR]