2014年-IMF国际货币组织全球_A_Hybrid_Approach_to_Estimating_the_Efficiency_of_Public_Spending_on_Education_in_Emerging_and_Developing_Economies_24页_649kb

Summary of "A Hybrid Approach to Estimating the Efficiency of Public Spending on Education in Emerging and Developing Economies"

Core Content

This paper explores the efficiency of public spending on secondary education in emerging and developing economies using a hybrid method that combines elements of non-parametric and parametric approaches. The study addresses the limitations of traditional methods like Data Envelopment Analysis (DEA) and Stochastic Frontier Analysis (SFA), particularly their inability to account for exogenous factors such as economic development and institutional quality.

Main Views and Key Information

1. Challenges in Measuring Education Spending Efficiency

• Traditional Techniques Limitations:
  • DEA struggles with incorporating multiple explanatory variables.
  • SFA, while allowing for more variables, often faces issues with statistically insignificant relationships between spending and outcomes.
  • Both methods assume no measurement error and rely on hypothetical comparators, which can introduce bias.

2. Hybrid Methodology

• Grid Search Approach:

  • The first stage of the hybrid method uses a "grid search" to identify efficient data points.
  • A caliper of size $c$ and step size $s$ (both set to 10 or 0.1 standard deviation) is used to determine efficiency.
  • Only 14% of the 310 data points were labeled as efficient, indicating significant inefficiency in education spending.

• Stochastic Features:

  • The second stage applies a locally weighted scatter plot smoothing (LOWESS) to reduce measurement error.
  • This helps in estimating a smooth frontier based on actual observations rather than hypothetical ones.

• Exogenous Constraints:

  • The hybrid approach accounts for exogenous constraints such as GDP per capita and governance effectiveness.
  • The sample is split into lower-income and higher-income groups to construct group-specific efficiency frontiers.

3. Empirical Findings

• Inefficiency in Education Spending:

  • Secondary education spending is inefficient in many emerging and developing economies, especially in Africa.
  • The average potential gain in net enrollment rate for lower-income economies is 36 percentage points, compared to 7 percentage points for higher-income economies.

• Factors Affecting Efficiency:

  • High student-to-teacher ratios and high rural population shares are associated with lower efficiency.
  • Governance effectiveness (measured by the World Bank's Governance Indicators) and income inequality (measured by the Gini coefficient) significantly influence efficiency.

• Efficiency Scores and Potential Gains:

  • The average potential gain across 89 economies is 23 additional enrolled students per 100 students.
  • Efficiency scores vary widely, with higher-income economies averaging 90% and lower-income ones averaging 55%.

• Regional Differences:

  • European and Western Hemisphere economies are more efficient.
  • African economies are among the least efficient, highlighting the need for targeted interventions.

4. Sensitivity Analysis

• The results are sensitive to the choice of caliper and step size.
• Larger calipers and steps increase the potential gain due to fewer efficient data points.
• The surfaces of potential gains are bumpy, and reducing caliper and step size leads to a significant drop in potential gain scores, likely due to reduced sample size.

5. Policy Implications

• Improving Efficiency:
  • Reducing student-to-teacher ratios and improving institutional quality (governance effectiveness) can enhance education efficiency.
  • Easing access to education facilities and reducing income inequality are also key to improving efficiency.
• Targeted Interventions:
  • Policies should be tailored to address the specific inefficiencies in lower-income economies, where the potential gains are the largest.
  • The hybrid approach provides a more accurate and robust measure of efficiency by accounting for development levels and using actual data points.

Conclusion

The paper concludes that the hybrid method offers a more refined and accurate way to measure the efficiency of public spending on education in emerging and developing economies. It emphasizes the importance of considering exogenous factors such as economic development and governance in efficiency analysis. The results suggest that substantial efficiency gains are possible, particularly in lower-income economies, through targeted spending reallocation and institutional improvements.