2014年-IMF国际货币组织全球_Regional_Labor_Market_Adjustments_in_the_United_States_and_Europe_37页_1mb

Summary of "Regional Labor Market Adjustments in the United States and Europe"

Core Content

This working paper by Mai Dao, Davide Furceri, and Prakash Loungani analyzes the patterns and mechanisms of regional labor market adjustments in the United States and Europe over the past 40 years. The study uses state-level and regional data to examine the dynamic responses of employment, unemployment, participation rates, and net migration to labor demand shocks. It also investigates the evolution of these adjustment mechanisms and the factors influencing them, such as spatial dispersion of labor market conditions and endogeneity of labor demand shocks.

Main Findings

United States

1. Long-term Employment Response

   • The long-run effect of a state-specific labor demand shock on employment levels has decreased over time, indicating a reduced role of net migration in adjusting to regional shocks.
   • The long-term employment loss due to a 1% negative labor demand shock is around 1.2%, compared to a larger decline in earlier periods.

2. Short-term Adjustments

   • A 1% negative labor demand shock leads to a significant increase in unemployment (0.22 percentage points in the first year, peaking at 0.29 after two years) and a decrease in participation rates (0.24 percentage points in the first year, peaking at -0.34 after two years).
   • Net migration has become a weaker adjustment mechanism in the short term, with a 1% shock reducing the working-age population by 0.1% instead of 0.4% through net migration.

3. Role of Participation Rates

   • The participation rate has increasingly acted as an absorber of regional shocks, particularly in the short term, compared to earlier decades.

4. Cyclical Patterns

   • There are strong cyclical patterns in the response of labor market variables to shocks. During downturns, more workers adjust through unemployment and participation, while migration is less responsive.

5. Spatial Dispersion

   • The dispersion of labor market conditions across states has declined over time, which is linked to the secular decline in the migration response to regional shocks.

6. Endogeneity and Instrumental Variables

   • The paper tests the identification assumption of the original BK (Blanchard and Katz, 1992) study using two instrumental variables (IVs):
     • Industry mix variable (based on state-specific industrial composition and national industry growth): Stronger correlation with employment growth and more robust in the first stage.
     • Oil price variable (based on changes in oil and gas industry employment relative to aggregate employment): Weak instrument, leading to potential bias in estimates.
   • The use of 2SLS (Two-Stage Least Squares) with the industry mix variable leads to different results compared to OLS, particularly in the short run, suggesting that labor demand shocks are not exogenous.

Key Insights

• Migration as an Adjustment Mechanism:

  • Migration has historically played a key role in adjusting to regional labor market shocks, but its importance has diminished over time.
  • The response of net migration to regional shocks is stronger during aggregate downturns than in booms.

• Role of Participation and Unemployment:

  • Participation rates have become a more significant absorber of regional shocks, especially in the short term.
  • Unemployment rates show a stronger response in the short run, but this has weakened over time.

• Spatial Dispersion and Adjustment:

  • Spatial dispersion of labor market conditions increases during recessions and decreases during booms.
  • The declining trend in dispersion is associated with the reduced migration response to regional shocks.

• European Comparison:

  • In Europe, the short-term response of participation rates to labor demand shocks is typically larger than in the US.
  • However, the immediate response of net migration in Europe has increased over time, suggesting a shift in adjustment mechanisms.

Methodology

• Data Sources:

  • State-level employment, unemployment, and participation data from the Bureau of Labor Statistics (BLS).
  • Net migration data from the American Community Survey (ACS).
  • Regional data from 21 European countries.

• Estimation Approach:

  • Panel VAR (Vector Autoregression) models are used to estimate the dynamic responses of labor market variables.
  • The paper tests the identification assumption using instrumental variables (IVs), including the industry mix and oil price variables.
  • 2SLS estimation is employed to address endogeneity, with the industry mix variable showing superior performance.

• Key Equations:

  • A system of panel VAR equations is used to estimate the dynamic response of employment, unemployment, and participation rates to labor demand shocks.
  • The model assumes that the current employment growth is driven by labor demand shocks and is weakly exogenous to the unemployment and participation equations.

Conclusion

The paper highlights a shift in the adjustment mechanisms of US labor markets over the past few decades, with a reduced role of net migration and increased reliance on participation and unemployment rates. These changes are linked to the declining spatial dispersion of labor market conditions. In Europe, while participation rates respond more strongly to shocks in the short term, net migration has become more responsive over time. The results contribute to understanding the evolving nature of labor market flexibility and its implications for macroeconomic policy.