2015年-IMF国际货币组织全球_Some_Misconceptions_about_Public_Investment_Efficiency_and_Growth_37页_783kb

Summary of IMF Working Paper: "Some Misconceptions about Public Investment Efficiency and Growth"

Core Content

This IMF Working Paper challenges common misconceptions about the relationship between public investment efficiency and economic growth. It argues that the growth impact of public investment spending is not dependent on the level of efficiency, a result that contrasts with some earlier policy analyses and empirical findings.

Main Points

1. Definition of Public Investment Efficiency

• Public investment efficiency is defined as the ratio of the actual increment to public capital to the amount spent on it.
• It is typically assumed to be less than 1, meaning that not all public investment spending translates into productive public capital.

2. The Invariance Result

• The growth impact of public investment spending is invariant to the level of efficiency.
• This result is derived from a standard Cobb-Douglas production function and a capital accumulation equation.
• The key insight is that efficiency has two offsetting effects on the rate of return:
  • A higher efficiency increases the capital per unit of investment spending.
  • It also lowers the capital/output ratio, which reduces the marginal productivity of capital.
• In the Cobb-Douglas case, these two effects exactly cancel each other, resulting in the same growth impact regardless of efficiency.

3. Empirical Evidence

• A regression of GDP per capita on the public capital stock (with control for private capital) shows no significant correlation between efficiency and the growth impact.
• The paper notes that empirical studies often measure efficiency as a time-invariant index (e.g., PIMI), which aligns with the theoretical invariance result.
• The paper also discusses variations in empirical results due to different measures of efficiency and different assumptions in the model.

4. Policy Implications

• The invariance result suggests that blanket recommendations against increasing public investment in inefficient countries may be misguided.
• It emphasizes that both efficiency and the capital stock need to be considered when evaluating the impact of public investment.
• Structural reforms that improve efficiency (i.e., "investing in investing") can have powerful growth effects.

5. Limitations and Extensions

• The paper considers several extensions to the basic model:
  • CES production function: Efficiency still has no effect on the growth impact of investment.
  • Private capital: The inclusion of private capital does not change the invariance result.
  • Waste or corruption: These factors can reduce the amount of capital built but do not affect the growth impact in the same way as efficiency.
  • Investing in investing: Improving efficiency through structural reforms is crucial for growth, as it enhances the rate of return on public investment.

6. Key Findings

• The growth effect of public investment is independent of efficiency.
• The marginal product of capital (MPK) is not a direct measure of the growth impact of public investment.
• Empirical estimates of the rate of return to public investment often implicitly account for inefficiency.
• Efforts to infer efficiency from GDP growth and production function assumptions are misguided.

Conclusion

The paper concludes that the invariance result is a robust finding that has important implications for policy. It suggests that the efficiency of public investment is not the only factor to consider when evaluating the growth impact of public investment. Instead, the rate of return and capital stock levels are also critical. The paper also highlights the importance of structural reforms that improve efficiency, as they can lead to significant growth benefits.

Key Information

• Authors: Andrew Berg, Edward F. Buffie, Catherine Pattillo, Rafael Portillo, Andrea Presbitero, and Luis-Felipe Zanna
• Date: December 2015
• JEL Codes: O40; O43; H54
• Keywords: Public investment; Growth; Efficiency; Low-Income Countries
• Model Used: Standard exogenous growth model (Solow, 1956; Mankiw, Romer, and Weil, 1992)
• Methodology: Theoretical analysis with empirical support using the Common Correlated Effects Mean Group (CCEMG) estimator
• Data Sources: Penn World Tables, Gupta et al. (2014), Dabla-Norris et al. (2012), and Global Competitiveness Report (GCR)

Figures Mentioned

1. Efficiency and the output impact of public capital stock – Shows no significant correlation between efficiency and growth impact.
2. Public investment spending and investment efficiency – Illustrates the inverse relationship between efficiency and investment-to-GDP ratio.
3. The growth effect of additional investment spending in the CES case – Supports the invariance result.
4. The elasticity of output with respect to the true capital stock – Highlights the role of capital stock in growth.
5. Impulse Responses for High and Low-Efficiency Countries – Demonstrates the growth effect of efficiency changes.
6. Impulse Responses for High and Low Private Adjustment Costs – Shows how private capital affects the results.
7. Corruption: Impulse Responses Without Private Capital Accumulation – Highlights the role of corruption in public investment.
8. Corruption: Impulse Responses With Private Capital Accumulation – Compares different investment scenarios.
9. Improving Efficiency: Impulse Responses – Demonstrates the benefits of efficiency improvements.

Appendix and References

• Appendix A: Discusses the concept of efficiency and its formalization in the model.
• Appendix B: Explores the role of corruption in the consumption function.
• References: Includes works by Pritchett (2000), Caselli (2005), Gupta et al. (2014), and others, highlighting the academic context of the paper.