2022-06-02-IMF-Monetary_Policy_and_Exchange_Rate_Dynamics_in_a_Behavioral_Open_Economy_Model_44页_1mb

Report Summary: Monetary Policy and Exchange Rate Dynamics in a Behavioral Open Economy Model

Authors: Marcin Kolasa, Sahil Ravgotra, and Pawel Zabczyk
Source: IMF Working Paper WP/22/112

Key Findings

The study extends the standard open economy New Keynesian model by incorporating bounded rationality, specifically using Gabaix’s (2020) cognitive discounting framework. The framework nests rational expectations as a special case (m = 1), where agents discount future deviations from steady state due to "myopia" (controlled by parameter m < 1).

1. Model Extensions

• Cognitive Discounting: Agents discount future expectations (m < 1), shrinking subjective forecasts toward steady state. This affects uncovered interest rate parity (UIP) and policy transmission.
• Incomplete Asset Markets: The small open economy model includes a role for net foreign assets, which becomes persistent under cognitive discounting.

2. Resolution of Exchange Rate Puzzles

• The model resolves several long-standing puzzles like the forward premium puzzle and predictability reversal puzzle by using actual exchange rate expectations (m < 1).
• Behavioral expectations reduce the forward premium anomaly and improve the fit between theoretical UIP and empirical evidence.

3. Monetary Policy Transmission

• Forward Guidance (FGP) Puzzle: Cognitive discounting dampens the effectiveness of forward guidance, as agents underreact to future policy announcements. However, this effect is less pronounced in open economies, partly due to expenditure-switching channels.
• "Low for Longer" Policies: These policies are less effective under behavioral discounting (e.g., m = 0.5), compared to conventional stimulus.

4. Net Foreign Assets and Persistence

• Cognitive discounting exacerbates the unit root problem in small open economies, requiring mechanisms like debt-elastic risk premia to restore stationarity.
• Net foreign assets ((B^*)) and the real exchange rate show higher persistence under myopia, linking to persistent effects of low interest rates.

5. International Monetary Spillovers

• Domestic monetary easing in Home economies reduces spillovers to trading partners, as behavioral agents underreact to policy changes, decreasing the effectiveness of exchange rate adjustments.
• Spillovers are more negative under higher discounting ((m \to 1)) and non-existent under moderate discounting ((m \to 0.5)) in open economies.

6. Comparison with Rational Expectations

• Rational expectations (m = 1) fail to resolve several exchange rate puzzles and forward guidance inefficiencies. Behavioral models provide better empirical alignment and more intuitive policy implications.

Conclusions

The study demonstrates that bounded rationality significantly improves the open economy model by resolving key empirical puzzles (e.g., forward premium and forward guidance puzzles). Policies relying on future commitments (low for longer) are less effective under behavioral discounting. The model highlights that openness moderates some policy effects, while cognitive discounting worsens persistence issues. Overall, the behavioral extension enhances model fit and offers more realistic insights into monetary transmission and exchange rate dynamics.