2014年-IMF国际货币组织全球_Hysteresis_in_Unemployment_and_Jobless_Recoveries_37页_667kb

Summary of "Hysteresis in Unemployment and Jobless Recoveries" by Dmitry Plotnikov

Core Content

This paper explores the phenomenon of hysteresis in unemployment and the jobless recoveries observed in the U.S. post-Great Recession and throughout the postwar period. The author proposes a general equilibrium rational expectations model with search frictions and multiple equilibria, where aggregate shocks have a permanent effect on the unemployment rate.

The model incorporates search in the labor market, CES production technology, and adaptive expectations for household wealth. It distinguishes between demand shocks (self-fulfilling belief shocks) and supply shocks (productivity shocks), showing that both play a role in explaining employment dynamics and unemployment persistence.

Main Views

• Hysteresis in unemployment is path-dependent, as in the work of Blanchard and Summers (1986, 1987).
• Wealth changes significantly affect the speed of employment recovery. Large wealth losses during recessions lead to prolonged unemployment.
• The Great Recession and other post-1990 recessions are characterized by jobless recoveries, which can be explained by negative wealth changes.
• Multiple equilibria allow the model to generate feedback effects from the real economy to expectations, which is absent in standard models.
• The model includes an externality term that depends on the aggregate unemployment rate, making the labor market incomplete.

Key Information

1. Empirical Observations

• Figure 1 shows that employment recovery after the Great Recession was unusually slow, taking over 66 months to return to pre-recession levels.
• Table 1 provides evidence that wealth losses are correlated with slow employment recovery across all postwar recessions.
• Before 1990, employment recoveries were relatively quick, but since then, they have become jobless, indicating a structural change in the labor market.

2. Model Structure

• The model is a dynamic stochastic general equilibrium (DSGE) model.
• It includes households, firms, and labor market search.
• The CES production function allows for variable capital-labor substitution.
• The labor market is incomplete due to the lack of a market for unemployment inputs.
• Adaptive expectations are used to model household wealth expectations, which influence aggregate demand and thus unemployment.

3. Model Equilibrium

• The equilibrium is defined by the following equations:
  • The Euler equation for consumption:
    $$
    \frac{1}{c_t} = \beta E_t\left[ \frac{1}{c_{t+1}} \left(1 - \delta + a \left(\frac{y_{t+1}}{k_{t+1}}\right)^{1 - \rho} \right]
    $$
  • The budget constraint:
    $$
    c_t + I_t = y_t
    $$
  • The capital accumulation equation:
    $$
    I_t = k_{t+1} - (1 - \delta)k_t
    $$
  • The production function with externality:
    $$
    \bar{y}_t = \left(a \bar{k}_t^\rho + b s_t^\rho \bar{l}_t^\rho \left(1 - \frac{\bar{l}_t^{\frac{1 - \theta}{\theta}}}{\Gamma} \right)^\rho \right)^{\frac{1}{\rho}}
    $$
  • The wage equation with externality:
    $$
    w_t = b s_t^\rho \left(1 - \frac{l_t^{\frac{1 - \theta}{\theta}}}{\Gamma} \right)^\rho \left( \frac{y_t}{l_t} \right)^{1 - \rho}
    $$

4. Model Dynamics

• The model generates a continuum of steady-state equilibria, each associated with a different employment level.
• Steady-state employment is determined by the externality term and the matching technology.
• TFP shocks have a protracted effect on the economy, leading to persistent unemployment.
• Belief shocks (non-fundamental) also contribute to unemployment persistence by reducing aggregate demand.

5. Estimation and Results

• The model is estimated using Bayesian methods.
• The posterior estimates show that both demand and supply shocks are important in explaining postwar macroeconomic dynamics.
• The impulse response functions demonstrate that TFP and belief shocks together can reproduce the observed unemployment dynamics, including the Great Recession recovery.
• The model's performance is compared with the RBC model, and it is shown to better explain the jobless recoveries.

Conclusion

The paper concludes that hysteresis in unemployment is a significant feature of the U.S. economy, especially after the Great Recession. The model provides a framework that integrates both demand and supply shocks, allowing for jobless recoveries and persistent unemployment. The externality term and adaptive expectations are central to the model's ability to capture real-world dynamics that standard models fail to explain.

The model is calibrated using Equation (18), which defines the socially optimal employment level. This helps to identify the scale parameter $ \Gamma $ in the estimation process. The paper also highlights that wealth changes are a key determinant of employment recovery speed, and that non-fundamental shocks (such as belief shocks) can have long-lasting effects on the economy.

This work contributes to the understanding of business cycles and the role of hysteresis in shaping unemployment dynamics.