2012年-世界发展银行全球_Wages_and_Health_Worker_Retention_in_Ghana___Evidence_from_Public_Sector_Wage_Reforms_38页_2mb

报告摘要

Summary of "Wages and Health Worker Retention in Ghana: Evidence from Public Sector Wage Reforms"

Core Content

This paper investigates the impact of wage increases on the retention of health workers in Ghana, focusing on the causal relationship between public sector wages and migration of skilled health professionals. The study uses a natural experiment created by the 2006 wage reform in the Ghanaian public health sector to estimate the effect of wages on attrition.

Main Findings

Wage Increase and Attrition: A 10% increase in wages is associated with a 1.5 percentage point decrease in annual attrition among 20–35 year-old health workers who are more likely to migrate. This translates to a 1.45 percentage point decrease in attrition rates in the preferred specification, leading to a 10-year survival rate increase from 43% to 51%.
Targeted Impact: The effects of wage increases are concentrated among young health workers (20–35 years) and migratory occupations (e.g., doctors and nurses). No significant effects are observed among older workers or non-migratory categories.
Geographic Concentration: The impact of wages on attrition is strongest in urban areas, but there are no gender differences in the effect.
Migration and Attrition Correlation: The study finds a strong correlation between attrition from the public payroll and aggregate health worker migration, especially to high-income English-speaking countries like the UK and the US.
Wage Gap and Migration: Health workers in Ghana can double their earnings in the UK after adjusting for purchasing power parity, suggesting that wage gaps are a significant factor in migration decisions.
Policy Context: The 2006 wage reform, which introduced the Health Salary Structure (HSS), replaced the Additional Duty Hours Allowance (ADHA) and significantly altered wage levels across different grades and steps. The reform also froze nominal wages from 2006 to 2009, creating a unique opportunity to study the effects of wage changes.

Key Information

Data Sources: The study uses administrative payroll data and migration statistics from the Ghana Nurses and Midwives Council and the UK General Medical Council.
Empirical Strategy: The authors employ a fixed effects model and difference-in-differences estimation to control for time-invariant characteristics and common time shocks. They use wage changes due to the 2006 reform as an instrumental variable to isolate the causal impact of wages on attrition.
Theoretical Framework:
- A simple migration model suggests that higher home wages reduce the probability of migration by decreasing the push effect of wage gaps.
- Credit constraints may also influence migration. If migration costs are high, even higher wages might not lead to more migration unless they can be used to finance the cost of migration.
Limitations:
- The study cannot fully isolate the effect of wages due to limited variation in the data.
- The correlation between attrition and migration is used as a proxy for the causal impact of wages on migration, given the lack of individual-level migration data.

Conclusion

The study provides new empirical evidence that wage increases can effectively reduce health worker attrition, particularly among young and migratory health professionals. The results support the classic push effect of wages on migration, suggesting that raising wages can be an effective tool for improving retention in the public health sector, especially in low-income countries like Ghana.

Key Points

Wage increases are cost-effective for improving retention, especially for young workers.
Attrition is interpreted as a proxy for international migration, given its strong correlation with aggregate migration data.
The 2006 wage reform created a natural experiment to measure the causal impact of wages on migration.
The study highlights the importance of wage differentials in migration decisions and the role of credit constraints in limiting the effectiveness of wage increases.
The empirical approach controls for unobserved heterogeneity and time shocks, using micro data and plausibly exogenous wage variation.

References

Bhargava, A., & Docquier, F. (2008). Physician migration to OECD countries: A new dataset.
Borjas, G. J. (1987). Self-selection in labor migration.
Dovlo, D., & Nyonator, F. (2003). The brain drain and the health sector in Ghana.
Yang, D. (2006). The impact of wage gaps on return migration.
Yang, D., & Choi, J. (2007). Rainfall shocks and migration.
Angelucci, G. (2005). Income and migration.
Gibson, J., & McKenzie, D. (2011). Migration decisions of highly skilled individuals.
Lopez, A., & Schiff, M. (1998). Credit constraints and migration.
Clemens, M. (2007). The economic impact of migration.
Mayda, A. M. (2010). Migration and wages.
Okeke, C. (2009). Physician migration from Africa.