Regional income convergence: An econometric analysis

The income inequality in Turkey is one of the major problems of economists as well as politicians. This study aims to approach the problem from a regional perspective. There are two major objectives in this study. The first objective is to determine the level of convergence and/or divergence and the second objective is to determine the spatial autocorrelation and in income clusters in Turkey. In addition to these major objectives, the validity of Kuznets? inverted U hypothesis and the relation between regional income inequality and the growth rate of national economy are also interest areas of this thesis. The analyses are performed at two aggregation levels: provincial and regional. The provincial level analysis considers the 67 conventional provinces, whereas, at the regional level analyses, NUTS Level 2 regions are used. The convergence analysis includes absolute and conditional ß-convergence, the ?-convergence and analyzing the time trend of inequality measures such as Gini, Atkinson and Theil indexes. The spatial autocorrelation analysis includes a global measure of spatial autocorrelation, Moran?s I, and local indicators of spatial association. Results of convergence analysis indicate that neither a monotonic convergence nor a monotonic divergence occurred in between 1970 and 2001. Sub-period analysis of inequality measures revealed that during the first sub-period from 1970 to 1986 there is an increase in regional inequalities, whereas, during the second sub-period from 1987 to 2001, these inequalities have decreased or at least remained close to their 1987 value. The spatial autocorrelation analysis indicated that from 1970 to 2001, the provinces and regions in Turkey became more dependent to their neighbors. Moreover, the local indicators revealed that the poor and rich provinces and regions are not distributed randomly across the country but instead follows a specific pattern. Poor provinces are clustered at the east and rich provinces are clustered at the west. Furthermore, these two major clusters are becoming sharper.