2015年-BIS国际清算银行_When_is_macroprudential_policy_effective_24页_332kb

Summary of "When is macroprudential policy effective?" by Chris McDonald

Core Content

This BIS Working Paper by Chris McDonald investigates the effectiveness of macroprudential policy tools, specifically loan-to-value (LTV) and debt-to-income (DTI) limits, in stabilizing the housing market. The study focuses on how these policies perform at different stages of the housing cycle and evaluates whether tightening or loosening measures have symmetric or asymmetric impacts.

Main Points

• Macroprudential policies are increasingly used to manage house price volatility and financial imbalances.
• Tightening LTV and DTI limits is generally more effective in reducing housing credit growth and house price inflation than loosening them.
• The effectiveness of these policies depends on the phase of the housing cycle.
• Tightening measures are more impactful when:
  • Credit is expanding quickly.
  • House prices are high relative to income.
• Loosening measures tend to have smaller or negligible effects, especially during downturns.
• The house-price-to-income ratio is a key indicator of when macroprudential measures are most effective.
• Policy changes are estimated using panel regressions and counterfactual analysis, with the results showing that the impact of tightening is more pronounced than that of loosening.

Key Findings

• Tightening LTV limits reduces housing credit by 4–6% and house prices by 5–9%.
• Tightening DTI limits reduces housing credit by 2–3%, but has insignificant effects on house prices.
• Loosening LTV and DTI limits does not significantly boost housing credit or prices, with effects often being negative or neutral.
• The difference in effects between tightening and loosening is negligible during downturns, suggesting that the asymmetry in effectiveness is primarily observed during booms.
• The house-price-to-income ratio is used as a key cyclical measure, with higher ratios indicating greater effectiveness of tightening measures.
• Other cyclical measures such as annual housing credit growth and house price inflation also influence the effectiveness of these policies.

Methodology

• The study uses panel data from 17 economies (including Asia-Pacific and developed economies).
• It estimates the effects of policy changes using generalised method of moments (GMM).
• The four-quarter effect and before/after difference are used to measure the short-term and long-term impacts of policy changes.
• Dummy variables are used to represent policy changes, with values of 1 indicating tightening or loosening and 0 otherwise.

Data and Variables

• Data Sources:
  • Housing credit data from CEIC, official statistics, and central banks.
  • House price indices from CEIC and the BIS property price database.
  • Control variables include real interest rates, real disposable income, and lagged dependent variables.
• Cyclical Measures:
  • House-price-to-income ratio (both relative to mean and absolute).
  • Annual housing credit growth.
  • Annual house price inflation.
  • GDP and income gaps.

Policy Implications

• Tightening measures are more effective in curbing credit and house price growth during expansions.
• Loosening measures are less effective, particularly during downturns, and may even exacerbate financial imbalances.
• The timing of policy changes is crucial for their effectiveness, as they are often implemented during periods of high house prices and strong credit growth.
• Regulators should consider the housing cycle stage when designing and implementing macroprudential policies.

Conclusion

The paper concludes that the effectiveness of macroprudential policies is asymmetric across the housing cycle, with tightening measures being more impactful in booms and loosening measures having limited or negligible effects, especially during downturns. This suggests that macroprudential tools should be applied strategically based on the current state of the housing market to ensure their maximum impact.