Labour in Turkey

Contributors

Abreg S. Çelem received his Ph.D. in Econometrics with his dissertation entitled “New Open Economy Macroeconomics; Exchange Rates and Current Account: An Empirical Analysis” from Gazi University, Ankara, in 2011. His fields of expertise are time series analysis, panel data analysis and microeconometrics. He works at the Department of Economics, Ondokuz Mayıs University, Samsun, since 2005. E-mail: ascelem@gmail.com

Pelin Akçagün is Assistant Professor in the Department of Economics at Ondokuz Mayis University. She received her B.Sc., M.Sc. and Ph.D. from Middle East Technical University, Department of Economics. As a visiting research scholar, she studied at University of Illinois at Urbana-Champaign during 2013–2014. Her Ph.D. thesis on Spatial Econometric Analysis of Regional Growth and Employment Convergence in Turkey received Middle East Technical University Best Ph.D. Thesis Award and Central Bank of the Republic of Turkey Academic Encouragement Award. Her research interests include theoretical econometrics and applied econometrics with special focus on regional development and labour economics. E-mail: pakcagun@gmail.com

Nihan Ciğerci Ulukan is Assistant Professor in Labour Economics and Industrial Relations Department at Ordu University in Turkey. She has a B.A., M.A and Ph.D. in Labour Economics and Industrial Relations from Marmara University, Istanbul, Turkey. Her Ph.D. dissertation is entitled “Migrants and Labour Market: Bulgarian Migrants in Bursa”. Throughout her academic career, she has been mainly concentrating on the migration and labour relations in Turkey. Her current research focuses on circular migration and seasonal migrant labour. E-mail: nculukan@odu.edu.tr

Umut Ulukan is Assistant Professor in Labour Economics and Industrial Relations Department at Ordu University in Turkey where he has been a faculty member since 2009. He has a B.A. and M.A in Labour Economics and Industrial Relations from Marmara University, Istanbul, Turkey. He holds a Ph.D. in Development Economics from the same university. His Ph.D. dissertation entitled “Structural Change of Turkish Agriculture and Contract Farming” was published in 2009 (in Turkish) and was also awarded by Turkish Social Sciences Association. His current research focuses on the new patterns of social differentiation among small-scale fishers in Turkey. E-mail: umutulukan@odu.edu.tr

Gülçin Taşkıran received Ph.D. degree in the Department of Labour Economics with her thesis “Alternative Organizations of Precarious Workers: Working Class Organization Practices in Service Sector” from Marmara University in 2014. She has many articles and book chapters on education-employment relationship, unemployment, informal employment and labour organizations in Turkey, and she has published many editorial books on gender and labour economics. He is currently working as an Associate Professor of Social Policy at Altınbaş University, Faculty of Economics, Administrative and Social Sciences. E-mail: gulcin.taskiran@altinbas.edu.tr

Güner Yazıcıoğlu Akyüz is Research Assistant at the Department of Sociology in Ondokuz Mayıs University, Samsun. She received B.A. in Sociology Department from Mimar Sinan Fine Arts University and continues to her Ph.D. dissertation in Anadolu University. Her main areas of research are surveillance, gender and urban studies. E-mail: guneryaz@gmail.com

Emek Yıldırım-Şahin was born in Ankara. As the secondary education, she was graduated from the Mehmet Emin Resulzade Anatolian High School, Ankara. She took her B.A. degree from the Department of Philosophy at the Middle East Technical University (METU), the minor degree from the Department of History at METU, M.Sc. degree from the Department of Political Science and Public Administration at METU and Ph.D. degree from the Department of Political Science and Public Administration at Ankara University as well. Between 2008 and 2010, she has been in pre-PhD research at the Peoples’ Friendship University of Russia (РУДН/RUDN) in Moscow. Since 2012, she is working in the Department of Political Science and Public Administration at Artvin Çoruh University. Her language skills are as English at advance level, Russian at intermediate level and German at beginner level. E-mail: emekyildirim@artvin.edu.tr

Fatih Çağatay Cengiz is a political scientist and lecturer at Ondokuz Mayıs University in Samsun. Dr. Cengiz received his M.A. from London School of Economics and Political Science (LSE) in 2010 and doctorate degree from School of Oriental and African Studies (SOAS) in London in 2016. His articles were published in the British Journal of Middle Eastern Studies and Politikon: South African Journal of Political Sciences. His research focuses on state theory, democratization and Islamic fundamentalism. E-mail: cagatay.cengiz@omu.edu.tr

Ferhat Akyüz is an economist and a lecturer at Samsun University. Dr. Akyüz received his Ph.D. from Marmara University, Istanbul, Turkey. Akyüz is author of the 2011 book Construction of Social Insecurity and Capitalism in Turkey (in Turkish). His articles have appeared in Mülkiye Dergisi and Praksis. His research ←12 | 13→currently focuses on the history of public debt in Turkey. E-mail: akyuzferhat@yahoo.co.uk, ferhat.akyuz@samsun.edu.tr

Polat S. Alpman works currently as Associate Professor in the Department of Social Work at Yalova University, Turkey. He received a Ph.D. in Sociology from Ankara University, Turkey, with the dissertation entitled “The Effects of Ethnicity on Social Inequality”. Dr. Alpman’s main areas of interest are the state, class, social inequality, discrimination, migration, citizenship and urban issues. Alpman, in addition to several book chapters and articles, published his book titled “Esmer Yakalılar: Kent, Sınıf, Kimlik ve Kürt Emeği” (Esmer Collar: Urban, Class, Identity and Kurdish Labour) with İletişim Publishing in 2016. E-mail: polatalpman@gmail.com

Özkan Öztürk works currently as Assistant Professor in the Department of Sociology at Karabük University, Turkey. He received a Ph.D. in Sociology from Ankara University, Turkey, with the dissertation entitled “Environmentalism and the Content of Environmental Discourses in Turkey”. Dr. Öztürk’s research interests and recent articles published in several international and national journals lie within the domains of environmental sociology, sociology of literature and modernization theories. E-mail: ozkanozturk2003@gmail.com

Yeliz Sarıöz Gökten is an economist. Dr. Gökten received Ph.D. degree from the Gazi University, Ankara, Turkey. She is a lecturer in the Department of Economics at Niğde Ömer Halisdemir University. Her research currently focuses on international political economy, hegemonic transition and new international institutions. Dr. Gökten’s doctoral dissertation was published by Notabene Publications as “The Past and Future of Hegemony Relations, A Neo-Gramscian Approach”. E-mail: yelizsarioz@ohu.edu.tr

Özgür Narin is Assistant Professor of Economics at Ordu University. He graduated from the Electrical & Electronics Engineering Department of Middle East Technical University. His Ph.D. is from Development Economics Department of Marmara University. He studied on the capitalist production of Science and Technology, particularly innovation and the changing scientific labour process. As a visiting scholar at New York University, he studied on changing labour processes under crisis and technological development. His current research is on Artificial Intelligence, “General Intellect” and the alternative reorganization of social production and society. He is also a member of IstanbuLab (https://stsistanbul.org), an independent platform engaged in the critical and constructive studies of Science, Technology and Society (STS) in Turkey. E-mail: narinozgur@yahoo.com

Abreg S. ÇELEM

Structural Change in the Employment Effects of Sectoral Growth in Turkey

Abstract: As employment generation capacity of an economy is one of the determinants of macroeconomic performance, most of the empirical studies carried out for Turkish economy reveal that the economic growth in Turkey doesn’t create the desired employment, which is known as the jobless growth phenomenon in economic theory. In this study, as two contributions to the above inference, I aim to (i) investigate the employment-generating capacity of economic growth in Turkey in terms of sectoral output and (ii) investigate the number, the exact dates and the effects on the parameters of structural breaks led by the global crisis in 2008. For this purpose, I apply Bai and Perron’s approach for estimating linear regression models under structural breaks with unknown numbers and dates. In the first step of the estimation process, a single structural break demonstrating the global crisis is dated as the third quarter of 2009. Under this break date, the results of the analysis indicate that the statistically significant and high employment elasticity of the agriculture sector turns to a lower and insignificant one after the crisis; industry sector turns to be an employment-generating sector after the global crisis, but with a poor statistical evidence; the services sector has a negative and insignificant employment elasticity. The employment generation capacity of output growth is significantly high only in construction sector in both periods. The study concludes that the jobless growth phenomenon in Turkey, exhibited by many studies, seems to be a major problem of the labour market in Turkey as the sole employment-generating sector is construction, a sector of short-term contracts and activities.

Keywords: employment in Turkey, employment elasticities of sectoral growth, structural breaks, Bai-Perron test

JEL Codes: E24, J21, C22

1. Introduction

Employment growth in Turkey has remained sluggish, in an adverse direction with the output growth in recent decades. Consequently, the economy has suffered from high unemployment rates. The unemployment rates in the country, which has exhibited seasonal fluctuations between 8 % and 11 % till the global crisis in 2008, have peaked up to 14.70 % during the most aggravated days ←15 | 16→of the crisis. The bearish trend after the crisis lasted only until 2012. Then the upward trend started again and the unemployment rate reached 13.80 % at the end of 2019. This level is just one percentage point below the one in 2008-crisis era (see Figure 1).

When this pessimistic view of unemployment rates is evaluated together with the growth trend of the country’s economy in recent years, the following question comes to the fore: How high is the employment generation capacity of output growth in Turkey? In fact, this is one of the determinants of macroeconomic performance as it is an indicator of the power of an economy to create labour force. In economics literature, this concept is generally quantitated as the employment intensity of growth or elasticity of employment with respect to output. Basically, though different calculation methods are used, this elasticity shows the relative employment growth associated with one percentage point of output growth (Kapsos, 2005). Contrary to the “output elasticity of labour” which can be defined as the response of output with respect to labour input, employment intensity of growth refers to the causality from output growth to labour.

Most of the empirical studies that aim to unveil this interaction in Turkey reveal a pessimistic picture in this regard (for instance, Yamak & Yamak (1999); Demirgil (2010); Barışık et al. (2010); Korkmaz & Yılgör (2010); Muratoğlu ←16 | 17→(2011); Murat & Yılmaz Eser (2013); Ertuğrul & Uçak (2013); Altunöz (2015); Özdemir & Yıldırım (1)2015); Arı (2016); Ulusoy et al. (2016); Demiral & Çelem (2020)). These studies reach the conclusion that the economic growth in Turkey doesn’t create the desired employment, which is known as the jobless growth phenomenon in economic theory. Jobless growth in Turkey has been associated with many factors. A prominent one is the increasing labour productivity led by technological innovation. Within this aspect, the advances in technological level, and hence, the labour productivity will occur dissimilarly in different sectors of the economy. In this case, it would be more appropriate to examine this problematic via sectoral data instead of the aggregated ones. However, only a few studies (e.g. Abdioğlu & Albayrak (2017); Tuncer & Altıok (2012); Aksoy (2013); Altuntepe & Güner (2013)) have focused on the employment effects of sectoral growth in Turkey. In order to illuminate this area of concern, I investigated the employment generation capacities of output growths in agriculture, industry, services and construction sectors in this study using empirical methods. On the other hand, as the related elasticities are likely subject to transformations or breaks, classical linear model is employed by a structural break model in order to reveal the number and the periods of possible structural changes. In the study, Section 2 outlines the econometric method for investigating structural breaks in linear models. Section 3 presents the dataset and the empirical analysis and Section 4 concludes.

2. Linear Regression Model with Break Points

The parameters of a standard linear regression model are assumed not to vary over time. However, economic crises and structural changes, as the determining factors in economy and the scope of many case studies, have crucial role in time series analysis. In one of the prominent studies that focus on the effects of structural change in linear regression, Chow (1960) proposes a test for investigating regime change at a known breaking date using an F-statistic. Along with it, Quandt (1958, 1960) develops likelihood ratio tests for structural change on a known single break date.

Afterwards, in more comprehensive contributions that considered the breaking date unknown, Andrews (1993) proposed tests for parameter instability and structural change with a single unknown breaking point in non-linear parametric models and Andrews and Ploberger (1994) improved these tests by allowing the – unknown – breaking point both to lie in a particular interval or remain completely uncertain. In the meanwhile, Perron (1989) studied the issue within the stationarity concept, applying modifications on unit root tests, that consider ←17 | 18→a single and known breaking date. Subsequently, Zivot ve Andrews (1992) proposed another unit root test in which break date is estimated rather than fixed.