Determinants of Underemployment in Turkey: A Micro Evidence from HLFS Data

Abstract

In the literature there have been a large number of theoretical and empirical studies on labour market. In this study we examine the main determinants of underemployment in Turkey using the individual level data obtained from Household Labor Force Surveys of 2006-2007 and 2008. In the empirical part of the study we use probit and logit models. We find that there is a statistically significant difference between males and females and likelihood of being underemployed is lower for females than males. With regards to the effect of residence, we find that the coefficient estimate of urban variable is only significant for females and has a negative effect on the likelihood of being underemployed. A further finding is that being married declines the probability of being underemployed for both males and females. Moreover, regardless of gender difference, individuals with primary school and over education degree are less likely to be underemployed compared to illiterates. There is an inverse-U shaped relation between age and likelihood of being underemployed. Finally, individuals in other occupation groups are more likely to be underemployed compared to legislators, senior, officials and managers.