Working Papers
Post-Compulsory Schooling of Youth in Turkey: A Case of Pro-cyclical Enrollment (with Meltem Poyraz)
The impact of recessions on school enrollment is ambiguous. On one hand, recessions might increase the likelihood of enrollment due to decreasing opportunity costs of attending school. On the other hand, recessions might discourage enrollment due to reductions households have in funds available for education. In this paper, we empirically analyze how local unemployment rates affect enrollment decisions in Turkey during the period covering the Great Recession of 2008-2009. Our estimates show that the enrollments in general high schools and university undergraduate programs decrease in regions experiencing higher unemployment, whereas the enrollment in vocational high schools increases. Our finding of pro-cyclical enrollment contradicts earlier findings in the context of developed countries. This contrast highlights the variations in the relative importance of the effect of income and in the potential long-term effects of recessions across countries. In particular, the adverse effects of lower income on enrollment might be dominant and recessions might have longer-lasting negative effects in developing countries due to their adverse effect on human capital accumulation. However, counter-cyclicality of enrollments in vocational schools suggests that such negative consequences can be alleviated by providing schooling options that offer more practical courses and allow working while enrolled.
Maternal Education and Early Child Development: The Roles of Parental Support for Learning, Learning Materials, and Father Characteristics (with Yusuf Emre Akgündüz, Pelin Akyol, Abduhrahman B. Aydemir & Murat G. Kırdar)
(also released as EAF WP No 2305)
This paper explores the intergenerational eects of maternal education on the development outcomes of 24- to 59-month-old children in Turkey. As the source of exogenous variation in maternal schooling, we use mothers’ exposure to the 1997 education reform in Turkey, which extended the duration of compulsory schooling from 5 to 8 years. The data come from the 2018 Turkey Demographic and Health Survey, which has a rich special module on early child development. We find a substantial increase in mothers’ educational attainment and a rise in children’s readiness to learn. Our finding is novel because it measures readiness to learn at a very young age rather than cognitive skills at later ages, as the previous studies do. We also find suggestive evidence of a positive impact on children’s social-emotional development. Examining the channels, we find that both mothers and fathers, particularly fathers, spend more time with their children, and the variety of activities parents engage with them rises. In addition, learning materials at home, such as books, rise. Also, exploring father outcomes, we find evidence of reductions in the schooling and age gaps between partners, implying an increase in women’s bargaining power, and suggestive evidence of a rise in fathers’ schooling. These findings about father outcomes are consistent with the significant rise in fathers’ involvement with children.
Publications
Reasons for the Decline in the Population of Turkish Republic Citizens Pursuing Doctoral Education Abroad (METU Studies in Development, 2022)
In this study, declines in both the number of citizens of Turkey who pursued doctoral degrees abroad and their shares among all foreign doctoral students in the OECD (Organisation for Economic Co-operation and Development) countries during the 2013-2016 period and in the United States of America (USA) during the 2005-2015 period are uncovered. The regression analyses show that the declines can be explained by two different factors. The first one is that an increasing number of foreign doctoral students, originating from countries where the scientific production has more rapidly expanded than in Turkey, started studying in the OECD countries and in the USA. The second one is the increase in the capacity of doctoral programs in Turkish universities. The effect of the former factor is found to be larger than the latter one. This finding implies that Turkey fell behind in the increasing global competition for placing students at international doctoral programs. The potential effects of these findings for the higher education system in Turkey are discussed in the last section of the study.
A Practitioner's Guide to Handling Irregularities Resulting from the 2014 Revisions to the Turkish Household Labor Force Survey (Boğaziçi Journal Review of Social, Economic and Administrative Studies, 2021, joint with Meltem Poyraz)
We document the implications of the 2014 revisions to the Turkish Household Labor Force Survey and offer guidance on how to handle the irregularities in population and unemployment statistics that resulted from two particular revisions. First, new population projections were adopted to assign survey weights. Second, a narrower definition of unemployment was adopted. We propose methods to adjust the survey weights for the pre-2014 period in order to discern changes in population statistics by age groups and regions without interruption over time and to calculate the unemployment rates according to both broader and narrower definitions since 2004.
The Sensitivity of Empirical Applications to the 2014 Revisions to the Turkish Household Labor Force Survey (in Theoretical and Empirical Perspectives on Economic and Financial Issues, 2021, joint with Meltem Poyraz)
In this study, we explore the sensitivity of two applications to utilizing the suggested weights in Demirci and Poyraz (2021) to deal with inconsistencies resulting from the 2014 revisions to the THLFS. In the first application, we estimate the effect of Syrian refugees on the employment outcomes of local people to see whether and how the results of empirical studies that rely on regional variation are affected by the 2014 revision. We show that the estimated effect of refugees on the employment rate of the local population is quite similar across weighting schemes because the observed regional variation in refugee ratios is lowly correlated with the changes in the regional statistics that are implied by the correction in Demirci and Poyraz (2021). In the second application, we assess the effect of a placebo treatment that is defined only for the elderly in the post-2014 period. We estimate the effect of the interaction of the dummy variables for the post-2014 period and the group of cohorts older than age 65 on the population of cohorts. We find a statistically significant effect of the interaction variable when the original weights are used, but the effect disappears when the weights suggested by Demirci and Poyraz (2021) are used. This result suggests that researchers should be cautious when working on projects about the elderly population with the THLFS data.
Education Abroad and Post-graduation Location Choices: AnInquiry on Turkish Republic Citizens studying in the United States of America (Journal of Higher Education and Science, 2020)
This study analyzes the education and post-graduation residency choices of students who go abroad from Turkey to pursue higher education. It has been found that most students have chosen to study in countries where the university system is advanced and the quality of academic publishing is better than Turkey. The analysis of students who have taken education in the United States as the most commonly preferred country has shown that the majority of students has attended the master’s- and doctoral-level programs. In addition, it has been observed that Turkish citizens studying at the bachelor’s, master’s, and doctorate levels in the United States were more likely to study in technical fields, such as engineering and computing, compared to university students of the same level in Turkey. It has been also found that a nonnegligible ratio of Turkish students in the United States, such as 21 percent, has enrolled in universities that are considered among the best 100 universities in the world. Although the abundance of Turkish students studying abroad in prominent universities seems to be a richness for Turkey, whether this richness benefits the Turkish economy depends on the rate at which these students return to Turkey after their graduation. To understand the return rate, whether the initial residency choice was the United States or not has been analyzed for all Turkish graduates of American universities who started their studies in the 2005-2015 period. It has been found that most of these students, such as 70 percent of master’s and doctoral degree holders, have continued to stay initially in the United States. Lastly, the long-term residency choices of students who graduated from three American universities have been analyzed, and it has been found that the ratio of those who returned to Turkey 10 years after their graduation has stayed at a low level of 40 percent.