The impact of digitalization on youth unemployment in Kazakhstan: Evidence from an ARDL framework

  • 24 Views
  • 6 Downloads

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License

Type of the article: Research Article

Abstract
The purpose of this study is to examine the impact of digitalization on youth unemployment in Kazakhstan over the period 2010–2023, with particular attention to the roles of education expenditure and urbanization, using an autoregressive distributed lag (ARDL). The analysis is based on annual national-level data obtained from the Bureau of National Statistics of the Republic of Kazakhstan and related official sources. Digitalization is proxied by Internet usage rates, while education expenditure, urbanization, and gross regional product are included as control variables. The ARDL bounds testing approach with heteroskedasticity-consistent estimators is employed to capture both short-run dynamics and long-run relationships among the variables. The results indicate a statistically significant short-run effect of digitalization on youth unemployment. Specifically, a 1% increase in Internet penetration is associated with an average reduction of approximately 0.27% in the youth unemployment rate, holding other factors constant. This relationship remains robust across alternative specifications, HAC estimators, and structural break adjustments accounting for the 2015 oil price shock and the COVID-19 pandemic in 2020. In contrast, education expenditure and economic growth exhibit weak or delayed effects on youth unemployment, while evidence of long-run cointegration is borderline. The findings suggest that digitalization contributes to reducing youth unemployment in Kazakhstan primarily through short-term labor market efficiency gains. However, sustaining these effects requires complementary investments in digital skills, education reform, and balanced regional development to ensure inclusive employment outcomes.

view full abstract hide full abstract
    • Figure 1. Research framework
    • Figure 2. Residual diagnostics and influence plots
    • Table 1. Variables, definitions, and data sources
    • Table 2. Descriptive statistics and unit root tests
    • Table 3. ARDL model specification and bounds test results
    • Table 4. Short-run HAC-robust estimates across kernels
    • Table 5. Structural break tests and break-adjusted estimates
    • Table 6. Finite-sample and bootstrap bounds sensitivity test
    • Conceptualization
      Amina Barzhaksyyeva, Yerzhan Amirbekuly, Gulzhikhan Smagulova, Fatih Yucel
    • Data curation
      Amina Barzhaksyyeva, Yerzhan Amirbekuly, Gulzhikhan Smagulova
    • Formal Analysis
      Amina Barzhaksyyeva, Yerzhan Amirbekuly, Gulzhikhan Smagulova
    • Investigation
      Amina Barzhaksyyeva, Yerzhan Amirbekuly, Gulzhikhan Smagulova
    • Resources
      Amina Barzhaksyyeva
    • Visualization
      Amina Barzhaksyyeva
    • Writing – original draft
      Amina Barzhaksyyeva, Yerzhan Amirbekuly, Gulzhikhan Smagulova, Fatih Yucel
    • Writing – review & editing
      Amina Barzhaksyyeva, Yerzhan Amirbekuly, Gulzhikhan Smagulova, Fatih Yucel
    • Methodology
      Yerzhan Amirbekuly, Gulzhikhan Smagulova, Fatih Yucel
    • Project administration
      Yerzhan Amirbekuly, Gulzhikhan Smagulova
    • Supervision
      Yerzhan Amirbekuly, Fatih Yucel
    • Validation
      Yerzhan Amirbekuly, Fatih Yucel