Does history fully determine the spatial distribution of human capital?

In the United States, regions with more human capital tend to attract skilled workers (e.g., see Glaeser and Berry, 2005), and as a result, convergence between regions does not occur (e.g., see Barro and Sala-i-Martin, 1992). Presently, many of the most productive European workers try to migrate to the United States. As a consequence, economic growth in Europe could be affected by such migrations (e.g., see G Saint Paul, 2008). Futhermore, Indian and Chinese entrepreneurs are recently coming back to their home countries despite higher wages in the United States. The main explanations are the lack of economic opportunities in the US and the costs of the labour force (e.g., see Saxenian et al., 2011). To comprehend these problems, we develop a theoretical Economic Geography model with heterogeneous skills for workers. It is an extension of Krugman's famous model "History versus expectations". This two regions model describes an economy with one input assuming two levels of skills for labor force, where workers have the opportunity to migrate. The question is whether history completely determines the final equilibrium or whether is it possible to attract skilled workers to areas with less human capital. This is a dynamic model, which is able to explain the location choices of workers between countries or within countries. The model could be extended to more complex new economic geography model with utility functions instead of wages (e.g., see Ottaviano, Tabuchi and Thisse, 2002). With the heterogeneity of workers, several equilibriums appear that were not present in Krugman's model. Dispersion of human capital is now a possibility where all the skilled workers are located in one area and all the unskilled workers are in the other region. Our results suggest that history does not determine the final outcome even with high interest rates conditional on economies scales that are sufficiently high contrary to Krugman. At least one equilibrium path emerges that was not present in Krugman's model. Under certain conditions, the final equilibrium is stable. This finding contradicts previous papers, which demonstrated that heterogeneity was a stabilizing force (e.g., see Morris and Shin, 2006; Herrendorf et al., 2000).