Public education expenditure and crime among educated individuals: Time-series evidence from Azerbaijan

  • 13 Views
  • 4 Downloads

Creative Commons License DMCA.com Protection Status
This work is licensed under a Creative Commons Attribution 4.0 International License

Type of the article: Research Article

Abstract
This study aims to examine the relationship between public education expenditure, student enrollment, and crime among educated individuals in Azerbaijan over the period 1998–2023. Using annual time-series data, the analysis investigates whether education-related public investment and educational participation are associated with changes in criminal activity within the educated population. The empirical findings reveal the existence of a stable long-run relationship among the variables, indicating that education expenditure, student enrollment, and crime dynamics evolve in an interconnected manner over time. The results further demonstrate that both public education expenditure (Wald = 11.356, p = 0.003) and student enrollment (Wald = 27.576, p < 0.001) possess statistically significant predictive power in explaining variations in crime among educated individuals. In particular, the stronger statistical effect observed for student enrollment suggests that broader participation in the education system may play a particularly important role in shaping long-term crime patterns. Overall, the findings indicate that education policy extends beyond its conventional function of human capital development and may contribute to broader social outcomes, including the reduction of crime-related vulnerabilities. By providing empirical evidence from Azerbaijan as a transition economy, this study contributes to the growing literature examining the wider societal implications of education policy and its role in promoting long-term social stability and sustainable development.

view full abstract hide full abstract
    • Figure 1. Time series plots of study variables
    • Figure 2. Impulse response functions with confidence intervals from the VAR model
    • Table 1. Descriptive data of the variables
    • Table 2. Unit root test
    • Table 3. VAR lag order selection criteria
    • Table 4. Roots of the characteristic polynomial
    • Table 5. VAR residual serial correlation LM tests
    • Table 6. VAR residual normality tests
    • Table 7. VAR residual heteroskedasticity test
    • Table 8. Johansen cointegration test
    • Table 9. Toda–Yamamoto Granger causality test results
    • Conceptualization
      Ramil Hasanov, Tarana Aliyeva
    • Data curation
      Ramil Hasanov, Fidan Safarova
    • Formal Analysis
      Ramil Hasanov, Tarana Aliyeva, Fidan Safarova, Tarana Shirvanova, Laszlo Vasa
    • Investigation
      Ramil Hasanov, Fidan Safarova, Tarana Shirvanova, Laszlo Vasa
    • Methodology
      Ramil Hasanov, Laszlo Vasa
    • Project administration
      Ramil Hasanov, Tarana Aliyeva
    • Resources
      Ramil Hasanov, Tarana Aliyeva
    • Software
      Ramil Hasanov
    • Supervision
      Ramil Hasanov, Tarana Aliyeva, Fidan Safarova, Tarana Shirvanova, Laszlo Vasa
    • Visualization
      Ramil Hasanov
    • Writing – original draft
      Ramil Hasanov, Tarana Aliyeva
    • Validation
      Tarana Aliyeva, Fidan Safarova, Tarana Shirvanova, Laszlo Vasa
    • Writing – review & editing
      Fidan Safarova, Tarana Shirvanova, Laszlo Vasa