The Impact of Education Level and Gender on Job Search Duration in Turkey

Abstract

AbstractThis study examines the effects of changes in unemployment rates on job search duration based on education level and gender in Turkey. Using monthly data obtained from the Turkish Statistical Institute for the period January 2005 through February 2013, we investigated the relationship between job search behavior and unemployment among participants with different education levels and genders. Results show that, in general, unemployment rate negatively affected the university graduates' probability of finding a job, though male university graduates' probability of finding a job was affected positively from their unemployment rate. High school graduates were affected positively from an increase in the overall unemployment rate, and when females and males were examined separately, the unemployment rate of the each group showed positive effects. Considering that vocational high school graduates obtain the same degree as traditional high school graduates but are taught required skills for specific jobs, our findings differed for this group compared with traditional high school graduates showing similarities with the results for university graduates. For the illiterate group, unemployment rate was positively related to job-seeking behavior; therefore, unskilled workers found jobs easily during high unemployment periods.Keywords: Job search * Job search duration * Unemployment * Education * Vector Error Correction ModelThere are two fundamental components that determine the reservation wage and the optimum point to conduct a job search: The education level of the job seeker (i.e., the main element of the reservation wage determination) and the market's current unemployment rate. Another important element is the behavioral differences between males and females. In this context, Ashenfelter and Ham (1979), Nickell (1979), Kiefer (1985), and Kettunen (1997) examined the relationship between education and unemployment, while Durand (1975), Pampel and Tanaka (1986), Psacharopoulos and Tzannatos (1989), Schultz (1990), Tansel (2001), and Bildirici, Aykac, Ozaksoy, and Akgul (2012) emphasized the differences of job search behavior, unemployment rate, and reservation wage determination between males and females, particularly during economic crisis periods.A limited number of previous studies have examined this issue specifically within Turkey. Dayioglu and Kasnakoglu (1997), Tunali (1997), Ozar and Senesen (1998), and Dayioglu (2000) evaluated the education level and labor force participation in Turkey, while Tasci and Darici (2009) and Kumas and Caglar (2011) examined the unemployment rate according to the gender gap. Previous research on the topic is limited due to the absence of reliable data on wages, a problem that still exists and causes issues during econometric analysis processes.The relationship between job search behavior and education and unemployment rate has drawn attention in both job search theory and human capital theory.In human capital theory, chosen education level maximizes the utility and income for a person's life. Early empirical research on the subject was conducted by De Wolff and van Slijpe (1973), Willis and Rosen (1979), Garen (1984), and Oosterbeek (1990). Determination of the optimum education level is a crucial issue, as Spence (1973), Hartog (1981; 1986), Duncan and Hoffmen (1981), Tsang and Levin (1985), Rumberger (1981), and Hartog and Oosterbeek (1988) showed that choosing an education level above the optimum point can cause complications in the long run.In job search models, it is assumed that the higher the level of education, the larger the increase in job opportunities. Since job-offer details proposed to individuals differ for the same job, they build up a job description for each individual. Furthermore, it is assumed that as a result of the increase in job opportunities by schooling year, employment offers that fall short of the education level can be accepted, but the ones that exceed it cannot be accepted or proposed. …