2013年-IMF国际货币组织全球_Waste_Not_Want_Not_The_Efficiency_of_Health_Expenditure_in_Emerging_and_Developing_Economies_26页_677kb

Summary of "Waste Not, Want Not: The Efficiency of Health Expenditure in Emerging and Developing Economies"

Core Content

This working paper by Francesco Grigoli and Javier Kapsoli examines the efficiency of public health expenditure in 80 emerging and developing economies between 2001 and 2010. The goal is to assess whether increasing health spending alone is sufficient to improve health outcomes or if greater efficiency in spending is necessary. The authors use a stochastic frontier analysis (SFA) to measure efficiency, which accounts for socioeconomic determinants of health outcomes, such as education, income inequality, and access to clean water and sanitation.

Main Points

1. Public Health Spending and Outcomes

• Public health spending is lower in emerging and developing economies compared to advanced economies.
• Health outputs and outcomes (e.g., life expectancy, infant mortality, immunization rates) are systematically worse in these economies.
• The paper highlights that increasing spending alone may not yield significant improvements in health outcomes if efficiency is low.

2. Efficiency Gaps

• African economies have the lowest efficiency in health spending.
• At current spending levels, African countries could increase life expectancy by up to five years if they adopted best practices.
• A 10% increase in public health spending in these economies would only result in a two-month gain in life expectancy, indicating that efficiency is a key determinant of health outcomes.

3. Efficiency Measurement Techniques

• Two main approaches are used to measure efficiency: non-parametric (e.g., FDH, DEA) and parametric (e.g., SFA).
• Non-parametric methods are criticized for not accounting for socioeconomic factors and lag effects of health spending.
• SFA is argued to be a more robust method, as it allows for the inclusion of covariates and controls for random noise and measurement errors.

4. Key Socioeconomic Determinants

• Education levels (years of schooling) are lower in emerging and developing economies, which is associated with worse health outcomes.
• Income inequality (measured by the Gini coefficient) is higher in these economies.
• Access to sanitation and clean water is more limited.
• Alcohol consumption is higher in advanced economies, contributing to lower life expectancy.
• Tuberculosis and HIV prevalence is higher in emerging and developing economies, further affecting health outcomes.

5. Methodology and Findings

• The paper estimates a stochastic frontier model where:
  • $ y_i $: log of HALE (Health-Adjusted Life Expectancy) for the 2006–10 period.
  • $ x_i $: log of public and private health expenditures for the 2001–05 period.
  • $ z_i $: log of socioeconomic covariates (e.g., GDP, education, population density, etc.).
• Multicollinearity diagnostics show that real GDP per capita is highly correlated with other variables and is therefore excluded from the model to avoid bias.
• The SFA model is estimated in two steps:
  • First, the log-likelihood function is maximized to estimate parameters.
  • Second, the inefficiency scores are calculated from the conditional distribution of the error term.

Key Information

• Efficiency scores are derived using a stochastic frontier model that incorporates a wide range of socioeconomic determinants.
• The average potential gain from reaching the regional efficiency average is presented in Table 4, with African countries showing the highest gains.
• The impact of health expenditure is shown to be stronger in public spending than in private spending.
• Life expectancy is the main health outcome used in the analysis, and the results indicate that efficiency plays a crucial role in achieving better outcomes.

Conclusion

• The study concludes that efficiency in health spending is critical for improving health outcomes in emerging and developing economies.
• African economies lag far behind in efficiency, and improving this could have a substantial impact on life expectancy.
• Parametric methods, particularly SFA, are preferred over non-parametric ones due to their ability to control for random variation and socioeconomic factors.
• The results are robust across different model specifications, lag structures, and assumptions about the error term distribution.

Tables and Figures

• Table 1 provides average health and socioeconomic indicators for advanced and emerging/developing economies.
• Table 2 shows multicollinearity diagnostics, indicating the need to exclude real GDP per capita.
• Table 3 presents SFA regression results, showing the positive and significant impact of public health spending on HALE.
• Table 4 outlines the potential gains from reaching the regional average of efficiency.
• Figure 1 illustrates the relationship between public health expenditure and health outcomes, showing a positive correlation with HALE and a negative correlation with mortality rates.

References and Authors

• Authors: Francesco Grigoli and Javier Kapsoli
• Email: FGrigoli@imf.org; JKapsoli@imf.org
• JEL Classification: H51, I12, I18
• Keywords: Health expenditure, efficiency, emerging economies, developing economies