The paper seeks to explain the influence of the so-called gravity effect on the regional diversity of economic growth in Poland. The gravity effect, which is a reference to Newton’s gravity law, is based on the assumption that regions have some economic influence on one another. The strength of these relationships is proportional to the size of the regional economies and inversely proportional to the distance between them. The applied model of economic growth refers to the neoclassical models of Solow; Mankiw, Romer and Weil; and Nonneman and Vanhoud. The presented model takes into account the gravity effect. The model has one non-trivial stationary point that is asymptotically stable, the authors say, which means that this point is a long-run economic equilibrium point. The paper examines the regional diversity of various macroeconomic indicators in the period from 1999 to 2011. Additionally, parameters of the labor productivity function are estimated at the regional level. The research yielded numerical simulations of two possible scenarios of the economic diversity of Polish regions. In the first scenario, it is assumed that investment rates are different in different regions (mean rates for each region from past periods). In the second scenario, the investment rates are the same for all regions. Both scenarios seem to indicate that the regional diversification of economic growth in Polish regions may decrease significantly, but the process of divergence was stronger under the regime imposed in the first scenario, the authors say. The strongest gravity effects were observed in an area bounded by the cities of Warsaw, Poznań and Wrocław as well as the southern Silesia region, with the possible inclusion of the city of Cracow. Peripheral regions in terms of the gravity effect, according to the authors, are provinces in eastern Poland as well as Świętokrzyskie in the center of the country, Pomorskie and Zachodniopomorskie in the north, and Lubuskie in the west.