Insights from Small Open Economy HANK models

Monetary policy in a two-country HANK Model

In this paper Sushant Acharya and Paolo Pesenti investigate the global effects of monetary policy in a two-country Heterogeneous Agent New Keynesian (HANK) model. The authors assume two types of households: unconstrained households who can access financial markets and hand-to-mouth households who consume their entire income due to lack of credit access. Their model includes idiosyncratic income risks that vary cyclically with economic conditions as well as sticky wages and nominal rigidities. The authors examine the effects of a contractionary monetary policy shock in one country (Foreign), while the other country (Home) keeps its policy unchanged. Acharaya and Pesenti assume four channels of transmission of monetary policy:

• Precautionary Savings Channel: Uninsurable income risks lead to precautionary savings, affecting consumption and output.
• Real-Income Channel: High MPC households are more sensitive to changes in real exchange rates, amplifying consumption responses.
• Expenditure-Switching Channel: Exchange rate changes shift demand between domestic and foreign goods.
• Intertemporal Substitution Channel: Changes in interest rates influence consumption timing.

Findings:

1. Amplification of Spillovers:
   – Both the precautionary-savings and real-income channels amplify spillovers compared to RANK models.
   – Quantitatively, these channels increase spillovers by 30-60% relative to RANK.
2. Sign Reversal by Precautionary Savings:
   – The precautionary-savings channel can reverse the sign of spillovers. For example, when income risk is countercyclical and sufficiently strong, a Foreign monetary tightening can reduce Home output despite a real exchange rate depreciation—a phenomenon termed “contractionary depreciation.”
3. Role of Real-Income Channel:
   – High MPC households amplify spillovers but do not change their sign relative to RANK. The real-income channel primarily strengthens the impact of exchange rate changes on consumption.
4. Spillbacks:
   Spillbacks refer to how changes in one country’s monetary policy affect its own economy via global linkages. The precautionary-savings channel increases spillbacks by amplifying declines in Foreign output due to heightened consumption risk.
5. Emerging vs Advanced Economies:
   – Spillovers are larger in economies with more high-MPC households or more countercyclical income risks (e.g., emerging markets).
   – Advanced economies experience smaller spillbacks due to relatively lower household vulnerability to income fluctuations.
6. Policy Implications:
   – Flexible exchange rate regimes may not fully insulate economies from external shocks when market incompleteness is considered.
   – The findings suggest that global monetary coordination could mitigate adverse spillover effects.

Spillovers and Spillbacks
Authors: Sushant Acharya, Paolo Pesenti
From: Bank of Canada, Federal Reserve Bank of New York

Open economy HANK vs RANK

Sushant Acharya and Edouard Challe present a tractable Small Open Economy Heterogeneous-Agent New Keynesian (SOE-HANK) model to “understand how uninsured risk and hand-to-mouth behavior affect optimal monetary policy relative to SOE-RANK.” Similar to Acharya and Pesenti above, the authors’ SOE-HANK model includes households that face uninsurable idiosyncratic income risk and unequal access to bond markets. Some households are hand-to-mouth and must consume their current income. Firms produce differentiated goods with sticky prices. Finally, Acharaya and Challe use a Constant Absolute Risk Aversion (CARA) utility for consumption and conditionally normally distributed income risk to achieve traceability.

Findings:

1. SOE-HANK Divine Coincidence: Under Cole-Obstfeld elasticities and acyclical income risk, optimal policy in SOE-HANK mimics that in SOE-RANK, implementing strict domestic producer price stability.
2. New Policy Tradeoffs: Away from those conditions, inefficient fluctuations in consumption inequality generate new monetary policy tradeoffs.
3. Output and Exchange Rate Stabilization: Under plausible calibrations (realistic trade elasticities, elasticity of intertemporal substitution, and countercyclical income risk), the central bank optimally stabilizes output and the exchange rate more in SOE-HANK than in SOE-RANK.
4. Productivity Shocks: After a negative productivity shock, optimal policy in SOE-HANK implements a smaller decline in output and real exchange rate compared to SOE-RANK, tolerating some inflation to mitigate increases in inequality.
5. World Interest Rate Shocks: In response to a fall in world interest rates, optimal policy in SOE-HANK cuts domestic rates more aggressively than in SOE-RANK, resulting in less exchange rate appreciation.
6. Inequality Channels: Both within-group and between-group inequality play important roles in shaping optimal policy, with their relative importance varying depending on the type of shock.

The paper demonstrates that accounting for household heterogeneity and incomplete markets can significantly alter the optimal conduct of monetary policy in small open economies compared to representative agent models.

Inequality and Optimal Monetary Policy in the Open Economy
Authors: Sushant Acharya, Edouard Challe
From: Bank of Canada, Paris School of Economics