Abstract:

In this study, the validity of the wage curve is examined by using the individual data obtained from the Home People’s Workforce Research (TÜIK) in Turkey in the period 2004-2013 and 2014-2017 in 26 regions (level 2) according to the region, age and gender groups. The wage curve defines the negative relationship between the local unemployment rate and the wage level. In previous studies for Turkey, it has not been investigated whether wage unemployment flexibility shows differences in the regions. As indicated in the Card (1995) the unemployment rates are used specific to the group. It is estimated a wage curve for Turkey using micro-level wage data. The survey contains a total of 6,520,839 individual observations. From these data, 1,018,202 individuals with full-time, continuous employee and wage answered more than 0 to the question were included in the work. The data is analyzed using the FE-2SLS and EC-2SLS models. According to Hausman’s test results, meaningful findings that make consistent predictions are assessed in the model. The forecast consistent for subgroups is mostly EC-2SLS. In Turkey, regional forecasts have a higher wage flexibility than general findings. However, there are also areas where wages do not respond to regional unemployment specific to the group. According to this study, in regional estimates, the unemployment flexibility of real wages in Turkey varies from region to region in different age and gender groups.

Keywords: