世界发展银行-Effects-of-Public-Sector-Wages-on-Corruption---Wage-Inequality-Matters_41页_1mb

Summary of "Effects of Public Sector Wages on Corruption: Wage Inequality Matters"

Core Content

This paper investigates the relationship between public sector wages and corruption, emphasizing the role of wage inequality in shaping the effectiveness of anti-corruption policies. Using a new panel dataset from the Worldwide Bureaucracy Indicators (WWBI), the authors analyze how variations in public sector wages and wage inequality affect corruption levels across countries.

Main Findings

• Public sector wages and corruption: On average, public sector wages have no significant impact on corruption levels. However, in countries with relatively compressed wage distributions, higher wages may reduce corruption.
• Wage inequality and corruption: In countries with highly unequal public sector wages, increasing wages can actually encourage corruption. This is because wage compression may lead to non-random sorting of employees, affecting the overall level of corruption.
• Interaction effect: The interaction between the public-private wage differential and wage compression in the public sector is significant. When wage compression is low, increasing public sector wages reduces corruption; but when compression is high, the opposite occurs.
• Robustness of results: The findings are robust to various empirical specifications, estimation methods, and distributional assumptions. The use of the share of contracts in the private sector as an instrument helps address reverse causality and omitted variable bias.

Key Variables and Data

• Corruption indicators:

  • Corruption_WGI: Derived from the Worldwide Governance Indicators (WGI), measures perceptions of both petty and grand corruption.
  • Corruption_WEF: From the World Economic Forum (WEF), reflects perceptions of government effectiveness in preventing illegal fund diversion and the frequency of unofficial payments.
  • Corruption_TI: From Transparency International (TI), a composite index of perceived public sector corruption.

• Public sector wage premium (WP):

  • Measured as the difference in wages between public and formal private sector workers.
  • Also considered for all workers (formal and informal) in the private sector.

• Wage compression ratio (WC):

  • Defined as the ratio of the 90th to the 10th percentile wages in the public sector.
  • Used to capture wage inequality within the public sector.

• Control variables:

  • Rule of law: From WGI, indicates the effectiveness of law enforcement in penalizing corrupt behavior.
  • Quality of bureaucracy: From ICRG, reflects the ability of institutions to deliver public services under political pressure.
  • GDP per capita: Used as a standard control variable in cross-country studies.
  • Government final consumption expenditure as a share of GDP: Reflects the size of the public sector and potential link to corruption.

Methodology

• The authors employ a fixed effects (FE) regression model to estimate the relationship between public sector wages and corruption.
• They also use a more complex specification that includes the interaction between wage differential and wage compression.
• To address endogeneity and omitted variable bias, they use the share of private sector contracts as an instrument for the public-private wage differential.
• The paper includes robustness checks using alternative models, including pooled OLS, random effects, and extreme bound analysis.

Policy Implications

• Cost-effective anti-corruption strategies: Increasing public sector wages may not always reduce corruption. It depends on the existing level of wage inequality in the public sector.
• Wage compression matters: In countries with compressed wage distributions, higher wages can reduce corruption. However, in countries with highly unequal wages, wage increases may exacerbate corruption.
• Combining policies: Policymakers should consider combining wage increases with measures to decompress public sector wages to achieve effective anti-corruption outcomes.

Limitations and Considerations

• Data quality: The paper highlights the importance of using micro-level data to accurately estimate wage differentials and compression ratios, as macro-level data often leads to measurement errors.
• External validity: Findings from single-country studies may not be generalizable due to the multidimensional nature of corruption and the influence of unobservable factors such as cultural norms.
• Reverse causality: There is a potential for reverse causality, where corruption may affect public sector wages rather than the other way around.

Conclusion

The study provides new insights into the complex relationship between public sector wages and corruption. It emphasizes the importance of wage inequality in determining the effectiveness of wage-based anti-corruption policies. The results suggest that wage reforms should be carefully designed, taking into account the existing wage distribution in the public sector to avoid unintended consequences.