2012年-BIS国际清算银行_Systematic_monetary_policy_and_the_forward_premium_puzzle_42页_581kb

Summary of "Systematic Monetary Policy and the Forward Premium Puzzle"

Core Content

This working paper by Demosthenes N. Tambakis and Nikola A. Tarashev investigates whether systematic monetary policy rules can explain the forward premium puzzle (FPP), which is the empirical observation that high-interest-rate currencies tend to appreciate, thus violating Uncovered Interest Parity (UIP). The authors analyze the relationship between monetary policy and UIP violations using a theoretical model and empirical data from seven small open economies and the euro area.

Main Findings

• UIP Violations are Common: The paper finds that six out of seven currencies (AUD, CAD, CHF, SEK, USD, NZD, and EUR) experience strong UIP violations, with slope coefficients in the Fama regression ranging from -0.5 to -3.2.
• Inflation-Targeting and UIP Violations: The degree of UIP violations is found to change with the adoption of explicit inflation targeting. For example, CAD and SEK show stronger violations after inflation targeting, while GBP-USD violations occur only before the implementation of inflation targeting.
• Only Forward-Looking Rule Explains Strong UIP Violations: The authors find that only a forward-looking monetary policy rule based on CPI inflation can account for the observed strong UIP violations. This is because such rules generate a negative covariance between the interest rate differential and expected nominal depreciation, which is essential for the sign condition of UIP violation.
• Sign Condition and Magnitude Condition: The sign condition requires that the covariance between the interest differential and expected depreciation is negative, while the magnitude condition requires that this covariance is sufficiently large relative to the variance of the interest rate differential to produce a large negative slope coefficient in the Fama regression.

Key Policy Rules

The paper considers six monetary policy rules, grouped based on whether they target CPI inflation or domestic inflation:

1. Strict Rules:

   • CPI-based (IT): Targets a zero CPI inflation rate.
   • Domestic-based (ITd): Targets a zero domestic inflation rate.

2. Taylor-Type Rules:

   • CPI-based (TR): Responds to current CPI inflation and output gap.
   • Domestic-based (TRd): Responds to current domestic inflation and output gap.

3. Forward-Looking Rules:

   • CPI-based (FW): Responds to expected CPI inflation and output gap.
   • Domestic-based (FWd): Responds to expected domestic inflation and output gap.

Only the CPI-based forward-looking rule (FW) satisfies both the sign condition and the magnitude condition, leading to strong UIP violations. The other rules, particularly those based on domestic inflation, fail to produce strong violations because they do not respond to foreign shocks, leading to a positive slope coefficient in the Fama regression, consistent with UIP.

Model and Empirical Analysis

• Model Structure: The model is a reduced-form open economy New Keynesian model with two exogenous shocks: a domestic shock to the natural real interest rate (NIR) and a foreign shock to the real exchange rate (RER). These shocks follow AR(1) processes.
• Empirical Approach: The authors use the Fama (1984) regression structure to estimate UIP violations, incorporating structural breaks related to the adoption of inflation targeting.
• Key Equation: The paper derives the model-implied slope coefficient in the Fama regression as a function of the covariance between expected depreciation and the interest rate differential, as well as the variance of the interest rate differential.
• CPI vs Domestic Inflation: The paper highlights that CPI-based rules, especially forward-looking ones, are more effective in explaining UIP violations, as they capture the impact of foreign shocks on the exchange rate and inflation dynamics.

Implications

• Monetary Policy Influence: Systematic monetary policy, particularly forward-looking rules, can significantly influence the exchange rate and interest rate dynamics, thereby affecting UIP violations.
• Market Microstructure: The paper also draws parallels with market microstructure studies, which suggest that arbitrage opportunities in foreign exchange markets are limited, justifying the focus on monetary policy rather than private sector behavior.
• Policy Design: The findings imply that monetary policy rules should be forward-looking and based on CPI inflation to better account for the observed UIP violations in small open economies.

Conclusion

The authors conclude that the forward-looking CPI-based monetary policy rule is the only rule that can fully explain the strong UIP violations observed in the data. This suggests that monetary policy design plays a crucial role in shaping exchange rate dynamics and the behavior of currency risk premiums. The paper provides both theoretical and empirical support for this conclusion, emphasizing the importance of considering the interplay between domestic and foreign shocks in policy formulation.