Extended Summary

Youth unemployment has emerged as one of the most pressing challenges of contemporary societies, not only due to its economic dimensions but also because of its social and demographic implications. In the Turkish context, since the early 2000s, both youth unemployment and the share of young people not in education, employment, or training (NEET) have remained consistently above the OECD and EU averages, making this issue a persistent structural concern. Against this backdrop, this study examines the macroeconomic and socio-demographic determinants of NEET in Turkey over the period 2000–2023 by employing the ARDL approach. The main aim is to capture both the short- and long-run dynamics shaping young people’s participation in the labor market and persistence in education.

The data set is constructed from annual series obtained from the World Bank database. The dependent variable is the NEET rate, while the independent variables include a set of macroeconomic indicators (real GDP per capita, total unemployment rate, inflation, exports), socio-demographic factors (unemployment with advanced education, government expenditure on education, female labor force participation, youth dependency ratio, internet usage), and two global shocks (the 2008 global financial crisis and the COVID-19 pandemic). The selection of these variables is grounded in the NEET literature, which emphasizes multidimensionality, while also reflecting the specific structural conditions of Turkey.

Methodologically, the ARDL framework is employed due to its flexibility in accommodating regressors with different orders of integration (I(0) and I(1)), its robustness in small samples, and its ability to estimate both short- and long-run dynamics within a single specification. The optimal lag lengths were determined using the Schwarz Information Criterion, while high intercorrelations among variables were considered in model construction to minimize potential multicollinearity. Stationarity was tested through unit root tests, cointegration relationships were verified with bounds tests, and short-run adjustments were captured through the Error Correction Model (ECM). Diagnostic tests (Breusch-Godfrey, Breusch-Pagan, Ramsey RESET) confirmed the statistical soundness of the models.

Empirical findings highlight that NEET rates are sensitive to both macroeconomic and socio-demographic factors in the long run. Real GDP per capita exhibits mixed effects: in some specifications it reduces NEET, while in others it shows a limited increasing effect, pointing to the non-inclusive nature of economic growth. Education-related variables emerge as strong predictors: higher government expenditure on education significantly lowers NEET, whereas unemployment with advanced education raises it. This underscores that education policies need to address not only quantity but also quality.

Among macroeconomic indicators, exports exert a reducing effect on NEET, while inflation demonstrates a more complex role, with positive or negative coefficients depending on the specification. This suggests that price instability influences youth labor market outcomes in non-linear ways. Socio-demographic variables also play a crucial role: female labor force participation reduces NEET rates, while a higher youth dependency ratio increases them, signaling that demographic pressures amplify vulnerabilities in youth labor markets. Internet usage, meanwhile, is associated with lower NEET rates, highlighting the importance of digitalization in fostering youth labor market integration.

Short-run results reveal that NEET is highly responsive to shocks. The 2008 global financial crisis exerted a strong and positive effect on NEET, while the impact of the COVID-19 pandemic appears heterogeneous: in some models it significantly increased NEET, while in others its effect was limited. This variation likely reflects sectoral, gendered, and regional heterogeneity in how the pandemic affected youth. The error correction term (CointEq(-1)) is consistently negative and significant, confirming the presence of long-run equilibrium and indicating that deviations from the equilibrium NEET level are corrected over time.

The contribution of this study lies in its integrated macro-level approach to analyzing NEET in Turkey. While earlier studies have mainly focused on micro-level or regional data, this research systematically evaluates structural determinants within a long-term time series framework. By incorporating external shocks such as the global financial crisis and the COVID-19 pandemic, it also illustrates how Turkey’s youth labor market vulnerabilities are shaped by both domestic and global dynamics.

Overall, the findings suggest that NEET in Turkey cannot be addressed solely through growth-oriented policies. Instead, a comprehensive strategy is required, combining macroeconomic stability with targeted social and labor market interventions. Policies should enhance the quality of education, facilitate school-to-work transitions, support female employment, and expand digital infrastructure. Addressing demographic pressures and building resilience against global shocks are also essential to reducing NEET rates and ensuring more inclusive labor market outcomes for young people.