41st Congress of the European Regional Science Association: "European Regional Development Issues in the New Millennium and their Impact on Economic Policy", 29 August - 1 September 2001, Zagreb, Croatia
In this empirical paper we examine whether or not migration decisions can be affected by individual measures of regional policy. To shed light on the question, we study formation of expected earnings and effect of expected earnings on interregional migration decisions in Finland. We consider how this effect varies among individuals, especially among employees in medical and health care sector and in teaching sector. We generated expected earnings in migration and staying alternatives from an endogenous switching regression. These expected earnings were used as additional explanatory variables in a random parameter logit migration model. In the migration model individual can choose between three alternatives: (s)he can either (i) stay in current region, (ii) migrate to a peripheral region, or (iii) migrate to a growth-centre region. The random parameter logit model allows us to specify the impact of expected earning as a function of occupational dummies and other variables, including a random factor. Our estimation results on a one-percent random sample from the Finnish longitudinal census file indicate that a decision to move is influenced by expected earnings, but the impact is dependent, for example, on occupational and educational group and on region of origin.