Elevator pitch
In the first two decades of the 2000s, Turkey has relied on structural change from traditional to modern sectors on the one hand and educational compositional change on the other hand to create formal employment in the modern sector. In 2000 the share of formally employed salaried employees in total employment was less than 40% for men and 30% for women. By 2021, this ration converged to 60% for men and women. Formal employment has increased for both men and women and the gender gap in formal employment declined substantially until 2020. However, relying on structural change and education to improve job quality has likely run its course. Since Covid-19, time-related underemployment has increased from virtually zero to 10% of the labor force and wages are stagnating if not declining.
Key findings
Strengths
The employment rate for women has increased substantially from 22.3% to 37% between 2005 and 2024.
The share of formal and wage employment has increased.
There was an increase in real monthly wages by almost 70% between 2002 and 2018.
The real wage growth was the highest at the bottom of the wage distribution due to increases in the minimum wage.
Wage inequality has declined noticeably over the period.
Weaknesses
The employment rate for less than tertiary educated women is still only 30% in 2024.
For wage workers, the minimum wage is fast becoming the median wage.
Wages declined in real terms between 2020 and 2022.
The job quality in Turkey is among the lowest of all OECD member states.
The wage premium for university graduates has been declining since 2013 due to stagnant wages.
Migrant workers are almost exclusively informally employed.
Author's main message
Rising but still very low levels of employment, especially for women, is the single most important issue in Turkey’s labor market. One side of low employment is a reserve army of the not-employed. A huge reserve army of not-employed, in turn, means low bargaining power for both the employed and the not-employed that results in (i) a very low share of wages in GDP, (ii) long working hours and increasing time-related underemployment as well as discouraged unemployment, and (iii) very high labor market insecurity. Turkey therefore needs active labor market policies, besides minimum wage increases, to create good jobs in the coming decade.
Motivation
In most recent years, Turkey is characterized by its very high inflation. Since the end of 2021, inflation is officially at 64%, 65%, and 44% in December of 2022, 2023, and 2024 respectively. Many private observers believe these official statistics understate the true value. High inflation is reducing the purchasing power of people on fixed incomes such as wage earners and pensioners. Turkey does not have official series for average or median wages but the wage share of GDP declined by ten percentage points between 2019 and 2022 [1]. Income inequality, as measured by the Gini coefficient, increased by two percentage points between 2020 and 2022, the steepest two-year increase in this century [2]. Low wage earners are somewhat protected, at least during 2023 – an election year – from the impact of inflation but above median wage earners (and also lowest wage earners in the informal sector) probably lost purchasing power.
Another distinguishing factor about Turkey is the very rapid increase in per worker GDP in purchasing power parity (PPP) terms. The growth rate of per worker GDP is one of the highest among OCED countries and per worker GDP is close to Czech levels by 2020. On the other hand, per capita GDP in PPP terms is close to Greece and is only above Latin American countries among OECD member (OECD data available here).
The discrepancy between GDP per capita and GDP per worker is explained by the very low employment levels, especially of women, and long working hours in Turkey. Hence, rapidly increasing employment levels can significantly increase GDP per capita and increase the wage share by strengthening the bargaining power of employees.
Discussion of strengths and weaknesses
Among OECD countries, Turkey has the lowest or second lowest employment rate in the 2000s (Figure 1). The employment rate of men is close to the OECD and Euro area countries’ average for prime-age working men (aged 25-54). On the other hand, the employment rate of women in Turkey is among the lowest within OECD countries (OECD data available here).
Furthermore, the substantial increase in women’s employment between 2014 and 2024 is mostly driven by a rapid expansion of tertiary education. The expansion resulted in a 11-percentage point increase in the share of women with tertiary education, it increased from 12% to 23% for women aged 15 to 64, leading to a significant change in educational composition. This change accounted for two-thirds of the three million increase in the number of employed women during the 2014-2024 period (Figure 2). The rest of the increase in the employment rate is due to women with high school degree. The share of women aged 15 to 64 with a high school degree increased from 16% to 22%.
Another decade of rapid expansion in educational attainment would result in a combined share of high school and tertiary educated women in the 15-to-64-age group to increase by 60% which means that Turkey would approximate the education level of Italy in 2023 (20% with tertiary and 43% with a high school education among all 25-to-64-year-olds) [3]. However, women’s employment rate will be around 40% even in 2034 with the current employment rates by education level.
Thus, the rapid increase in tertiary education is not accompanied by an increase in employment rates for women with tertiary education. Figure 3 shows that the employment rate of women with tertiary education is 61.5 % in 2024, the same as in 2014 (compared to a 3.5% increase for men in the same period). This is a very worrying sign because of the relatively low employment level for highly educated women. The employment rate for women with less than a tertiary degree increased gradually but starting from very low level: from 24.4% to 27.7% for women with less than a high school degree and from 19% to 34.8% for women with high school degree between 2014 and 2024. The authors of one study argue that the conservative cultural grip on employment of women is weakening in Turkey due to increasing education and urbanization [4]. However, even if following their argumentation and allowing for the employment of women with less than tertiary education to continue to increase at the same rate as in the most recent decade, the overall employment rate for women would only be around 43 to 44% by 2034 which is less than Costa Rica’s employment rate of women in 2024. The authors of another study point out that for women with small children employment means a 50% increase in “time- and income-poverty” [5]. That is, the positive consumption effects of higher purchasing power due to new employment is countered by negative consumption effects of time deficits in household production. Currently, public provision of child and eldercare are minimal in Turkey. Low wages for women with less than tertiary education means that they cannot afford market substitutes for household production curtailing their labor force participation.
Employment and the gender employment gap
Turkey experienced a healthy increase in employment among both men and women since 2021. The increase in the employment rate is higher for women, narrowing the gender employment gap (4.7 percentage point for men and 5.3 percentage point for women). However, the decline in the overall gender gap in employment is completely driven by changes in educational composition. Women with tertiary education are disproportionately represented in the labor force and the gender employment gap between men and women with tertiary education is lower than at other education levels. As the share of women with tertiary education in employment increases, the overall gender gap in employment is declining. However, the employment gap by education level has been either stable or slightly increasing between 2019 and 2024 at every education level, as shown in Figure 4.
These employment trends suggest that labor market outcomes in Turkey are dominated by structural change and education policies and that they will probably run their course within a decade when education levels in Turkey catch up to, for instance, Italian levels. Similarly, structural change, especially the exit from agriculture, will likely continue. Between 2014 and 2023, agricultural employment declined by one million. If this trend continues as in the most recent decade, the share of agriculture will be less than 10% by 2034. As a result of the educational composition and exit from the agriculture, continued increase in employment and a decline in employment gender gaps will depend on both active labor market policies to reduce the care burden of all women and the demand for skilled labor, to significantly increase employment of women with high school and tertiary education.
Education and unemployment
For men, the unemployment rate decreases with education whereas for women the unemployment rate is highest for high school graduates (Figure 5). In 2024, the unemployment rate for women with tertiary education was almost twice of men, 13.1% versus 6.2%. Moreover, women with tertiary education constitute 41.4% (32.6%) of all unemployed women despite being 23% (12%) of the 15-to-64-years old women in 2024 (2014). This disproportionate and increasing share of women with tertiary education within the unemployed is due to their much higher level of labor force participation (70% versus less than 40% for all other education categories in 2024) as well due to changes in educational composition. The temporary decline in the employment gap in 2020 reflects a disproportionate employment loss of men due to Covid-19.
Covid-19 and its aftermath: Part-time and discouraged workers
The decline in the headline unemployment rate hides a rapid increase in labor underutilization due to an increase in time-related underemployment as well as for workers who stopped looking for jobs after Covid-19 and after the 2023 shocks (see Figure 1 and Figure 6). For both men and women, the increase in time-related underemployment more than accounted for the decline in unemployment in the post-2021 period.
Job quality
Wage inequality
There are no publicly available wage series for Turkey. As a result, wage inequality studies for Turkey rely on wage series generated from various household surveys. The two key facts in wage studies are (i) the significant equalizing effect of minimum wage increases, especially in 2004 and 2016, and (ii) the stagnation in wages of tertiary degree holders after 2012. Until 2012 wages of tertiary degree holders increased as much as or even more than wages of other education levels resulting in an increasing tertiary education wage premium. After 2012, wage increases disappeared for university graduates, resulting in declining wage premiums. This decline in the university wage premium is partly due to an expansion of skilled labor supply [6] and partly due to a relative lack of demand for skilled labor [7]. The rapid expansion of higher education is singled out for a declining university wage premium but the share of university graduates in employment is still modest in Turkey. Studies on business dynamism [8] and macroeconomic dynamics [9] point to a decline in structural change in Turkey post-2013 which coincides with a peak in the university wage premium.
The real minimum wage increased by approximately 25% between 2004 and 2016, and has significantly reduced most wage inequality measures [10], [7]. For wage workers, the minimum wage is fast becoming the median wage [11]. The negative employment impact of these minimum wage increases was minimal [12]. The employment impact of the minimum wage hike in 2023 (OECD data available here) is likely to be minimal as well given the continued increase of employment at low education levels in 2023 and 2024, however it might lead to increases in part-time employment.
In early 2000s, wage inequality declined significantly after a minimum wage hike in 2004. Between 2005 and 2012 wage inequality slightly increased [10]. After 2012 wage inequality declined parallel to the decline in the education premium beyond secondary school levels [7]. A minimum wage increase in 2016 mostly benefitted less educated workers reducing wage inequality. Between 2004 and 2019 wages for primary school graduates increased by 60% for both men and women whereas for post-secondary graduates’ wages increased roughly by 25% and all of the increase took place by 2013 [7]. In other words, the decline in wage inequality is the combined result of minimum wage increases lifting the bottom of the income distribution and a stagnation of wages for post-secondary educated workers in the post-2012 years.
Lack of demand growth for skilled labor
One recent study explains the increase in the post-secondary education premium between 2004 and 2012 with public sector wage policies [7]. Once real wages in the public sector stagnated in the post-2012 period, the education wage premium started to decline.
The post-2012 period also coincides with a mass entry of new graduates form higher education expansion into the labor market [6]. However, skill-biased technical change does not seem to play a role in increasing wage inequalities in Turkey, unlike in developed countries [10], [7].
The timing of the stagnation in the education wage premium also coincides with structural and macroeconomic issues. One study reports that business dynamism in the manufacturing sector has been declining post-2012 due to a decline in the access to global liquidity [8]. In a parallel vein, the authors of another study focus on the broader economy and show that not only the manufacturing sector but overall economic growth in Turkey dependents on “hot-money” (i.e. the inflow of funds to earn short-term profits on interest rate or anticipated exchange rate shifts) and external debt inflows [9]. They point out that these inflows mostly led to construction-centered economic growth in the 2009-2018 period. Their finding of energy and construction-centered economic growth is consistent with the finding of a structural transformation (from agriculture to construction and services) driving changes in employment composition [7]. The authors of the later study emphasize that skill-biased technological change would manifest itself as within industry shift in relative labor demand and would thus lead to an increase in the university wage premium, not in stagnation [7].
Labor market insecurity and long working hours
Among OECD countries, Turkey is one of the worst performers in terms of job quality. Turkey has one of the highest levels of labor market insecurity defined as expected income loss due to unemployment (Figure 1). Furthermore, there is no improvement between 2015 and 2022 when most OECD countries experienced a sizeable improvement. Furthermore, Turkey has the second longest average usual weekly hours, after Colombia, and the longest working hours defined as 50 hours or more (OECD data available here and here). Nevertheless, the trend in working hours is downward, possibly driven by a structural change towards more formal employment where labor regulations are more likely to be enforced.
Inclusiveness
The decline in the share of informal employment was very slow between 2014 and 2019, and the absolute number of informally employed people did not decline in this period. However, the number of people in informal employment declined by 1.5 million between 2019 and 2020 pointing out that temporary lay-off bans during the initial phase of Covid-19 did not protect this vulnerable group. The increase (decrease) in the formal (informal) employment share during 2020 in Figure 7 (Figure 8) was due to a severe decline in informal employment. For men, most of the decline occurred for salaried or self-employed categories, while for women, the decline was in salaried or unpaid family worker categories. The decline in informal self-employment for men as well as the decline in informal salaried employment for women recovered in subsequent years. Further research is needed to uncover what happened to other jobs that disappeared during Covid-19: whether they disappeared completely or were taken up by informally employed migrant workers.
The number of informally employed worker was in the range of 7.5 and 8 million between 2021 and 2024. The continued decline in the share of informally employed people is due to an increase in overall employment. As expected, informal employment is much more prevalent in agricultural employment (Figure 8). The share of informally employed women is higher than of men throughout the period but informal employment is declining slightly faster for women. This faster decline is driven by structural change: the overall decline in agricultural employment is reducing the share of informally employed women who are much more likely to be employed as unpaid family workers in the agriculture sector.
Turkey received large numbers of migrants in the past 15 years starting with the civil war in Syria. Unfortunately, there are no data on the employment of undocumented migrant workers because labor force surveys do not cover undocumented migrants. There is therefore a dearth of data on undocumented migrants’ employment. The general consensus is that almost all of them are informally employed. The existing studies focus on migrants’ impact on natives’ employment. One study finds that the initial impact of Syrian migrants on natives’ employment is either positive (for high school and above) or zero [13]. The authors find some decline in natives’ wages of less than high school graduates for 2013 (the year of an initial big wave of refugees). In general, their findings suggest that migrants are complementary to most natives instead of substitutes. Besides micro studies, overall employment trends also suggest that if undocumented migrants are substitute for natives, they are substitutes only for the least educated natives in informal jobs.
Limitations and gaps
The 2000s Turkish authorities relied on mainly three policies to improve labor market conditions: real minimum wage increases, structural change, and increasing education level. Minimum wage increases outpaced overall wage growth. As a result, the share of minimum wage earners within wage employees is fast reaching 50%. Increases in the employment rate and in formal employment as well as a decline in wage inequality are the main positive developments in the Turkish labor market.
However, even these positive developments, like the decline in wage inequality, contain bad news: the decline in wage inequality is partly driven by stagnation of wages of workers with tertiary education since 2013. On the employment side, changes in structural and educational composition, instead of active labor market policies, are the main driver. Increasing the share of people with tertiary education accounts for 71% of the overall employment increase between 2014 and 2024. The share of people with tertiary education is likely to approach practical limits within the next decade. A comprehensive agenda to remove employment barriers, especially for women with less than tertiary education, together with policies to create a higher demand for skilled labor to counter stagnation in wages is thus needed. Furthermore, a full-employment perspective is needed to counter these formidable challenges. Policymakers’ long-standing habit of delegating labor market transformation to structural and educational compositional change while propping up below median wages with minimum wage increases is not going to do the task in the next decade.
Summary and policy advice
Turkey still suffers from very low levels of employment even for the prime-age working population. The unemployment rate does not reflect the extent of labor underutilization. Since Covid-2019, the share of discouraged unemployed, time-related underemployment, and official unemployment rates moved in opposite directions, i.e. official unemployment declined while other measures increased. This very large reserve army of the not-employed and underemployed people reduces the bargaining power of both the employed and not-employed. Turkey’s long working hours, low job security, and very low wage share in GDP are further evidence of low bargaining power of workers.
Two distinct issues in Turkey’s labor market can be identified: first, a very low level of women’s employment, and second, a relative lack of demand for skilled labor. These two main issues suggest two distinct policy packages: policies to help increase the employment of women, especially of women with less than tertiary education, as well as policies to encourage the creation of high value-added jobs that can pay high wages.
One study proposes a large investment in social care spending to directly generate jobs and to reduce unpaid work as well as time- and income-poverty [5]. The authors estimate that providing universal public childcare (at a cost of 1.8% of GDP) can generate one million new jobs directly and indirectly. Turkey’s public spending on family benefits including childcare is among the lowest within OECD member countries. An additional 1.8% will mean that Turkey’s family benefit spending will still be around the OECD average (OECD data available here). Some authors propose to increase the support for research and development directed towards immediate competitors of market leaders and policies to reduce the cost of borrowing for such firms in order to revive business dynamism and increase value added [8].
Acknowledgments
The author thanks the IZA World of Labor editors for giving him this opportunity to express his views on this topic. He also thanks the anonymous reviewer for their careful reading and suggestions.
Competing interests
The IZA World of Labor project is committed to the IZA Guiding Principles of Research Integrity. The author declares to have observed these principles.
© Hasan Tekguc
