AN ANALYSIS OF UNEMPLOYMENT DURATION IN TURKEY: EVIDENCE FROM INDIVIDUAL LEVEL DATA

Aysıt TANSEL (Middle East Technicanl University)
H.Mehmet TAŞÇI (Middle East Technicanl University)

In this study we will provide a micro-perspective on the Turkish labor market by using the Household Labor Force Survey data for the year of 2000. The object of this study is to investigate the determinants of unemployment duration among working-age groups. In the analyses, we will use non-parametric (Kaplan-Meier survival function) as well as parametric (Weibull, Exponential and etc.) and semi-parametric (Cox partial likelihood) estimation methods, and control for unobserved heterogeneity. In the parametric estimation part we will compare the models using the Akaike Information Criterion (AIC) to choose the best fitting model. Such analyses identify the statistical effect of explanatory variables, such as personal characteristics (education and age), family background, demographic characteristics, and local labor market conditions, on the exit rate out of unemployment. This part of the study will provide the main characteristics of the short-term and long-term unemployed people in Turkey. It is expected to find that age and education level has a negative effect on unemployment duration. Also, individuals with higher level of education (i.e. graduated from a university) are not expected to have longer unemployment duration than those with lower education (i.e. primary, middle and high school graduates). These analyses will be carried out for men and women separately.