Assessing the Risk of the Next Housing Bust

IMF Blog, April 4, 2019

Source details

Authors: Claudio Raddatz Kiefer, Nico Valckx
Date: April 4, 2019
Core insight: The IMF developed a tool, House Prices at Risk, that gauges the likelihood of severe downturns in home prices and feeds into the Fund’s growth-at-risk model to link financial conditions to the danger of an economic downturn.
Geographic/sample coverage: data from 22 advanced economies, 10 emerging-market economies, and the major cities in those countries.

Housing construction and related spending account for one-sixth of the US and euro-area economies.
Mortgages and housing-related lending constitute a large share of banks’ assets in many countries, linking house-price declines to banking-sector health.
More than two-thirds of financial crises in recent decades were preceded by a boom-bust cycle in home prices.

Five conditions used to assess future large drops in home prices:
- past price growth
- economic growth
- credit booms
- overvaluation
- financial conditions
Predictive question: Can large drops in home prices (those that occur roughly once every 20 years — that is, those that have a 5 percent probability) be predicted by changes in the five conditions?
Example scenario logic: If financial conditions tighten, households are heavily indebted, and house prices are stretched, the model predicts more instances of very large declines in home prices.

At end-2017, weighted by GDP, the odds of a big drop in inflation-adjusted house prices were lower in most advanced economies in the sample than 10 years earlier but remained above the historical average.
In emerging markets, riskiness was higher in 2017 than on the eve of the global financial crisis.
Downside risks to house prices remain elevated in:
- more than 25 percent of these advanced economies
- nearly 40 percent in emerging markets in the study
China stands out among emerging-market results, especially its Eastern provinces.
House prices in major cities around the world move in tandem, increasing the chance that a shock in one country affects housing markets elsewhere (synchronized swings).

A reading of minus 12 on the IMF gauge — corresponding to a 5 percent probability of a 12 percent drop in prices — indicates:
- a 31 percent probability of a financial crisis two years later in advanced economies
- a 10 percent probability of a financial crisis two years later in emerging markets

Do not target house prices directly; instead, strengthen resilience of households, banks, and the economy through targeted policies.
Macroprudential policy (recommended when the economy is strong and housing is booming):
- restrict loan-to-value ratios (amount of a home loan as a proportion of the property’s value)
- limit debt-service-to-income ratios (size of monthly mortgage payments as a proportion of income)
Monetary policy:
- Cutting the central bank’s interest rate can mitigate near-term risks — up to a few quarters ahead — and appears effective mainly in advanced economies.
Capital flow measures:
- Manage capital flows when surges in capital inflows increase downside risks to house prices (relevant for affected countries).
Structural and fiscal options to consider alongside cyclical tools:
- longer-term structural policies to increase housing supply or impose zoning restrictions
- fiscal measures such as property taxes

The House Prices at Risk tool provides an early-warning, predictive capability that can help policy makers take timely steps to limit damage and reduce the likelihood of crises similar to 2008.
Short-term easing can support prices now but may increase long-term vulnerability by fueling overborrowing; policymakers should weigh immediate support against longer-term risk accumulation.

Source: IMF blog post “Assessing the Risk of the Next Housing Bust,” April 4, 2019.