Construction Activity Can Signal When Credit Booms Go Wrong

IMF Blog, February 12, 2020

Source details

In Spain, private sector credit as a share of GDP almost doubled between 2000 and 2007; housing prices doubled in real terms over the same period.
Spain experienced a construction-sector employment increase of 47 percent during that boom, compared to an economy-wide employment increase of 27 percent.
The Spanish credit bubble burst in 2008, followed by loan defaults, bank failures, and a prolonged economic slowdown.
New IMF staff research, using a large sample of advanced and emerging market economies since the 1970s, finds that long-lasting credit booms with rapid construction growth never ended well.

Rapid credit growth (“credit booms”) trades off immediate buoyant economic performance against the danger of a future crisis.
The risk of a “bad boom” — a rapid credit growth episode followed by a financial crisis or subpar economic growth — increases when there is also a boom in house prices.
The dangerous combination of credit booms and rapid construction-sector expansion extends beyond the global financial crisis period and across countries.
Signals from construction activity can help distinguish dangerous booms that need to be controlled from buoyant but healthy credit growth (“good booms”).

During booms, output and employment expand faster overall, but growth concentrates in a few industries, specifically construction and, at a distant second, finance.
The industries that gain most during booms experience the most severe downturns during busts, implying credit booms leave few long-term footprints on industrial composition.

Construction is the only sector that consistently behaves differently between good and bad credit booms.
On average, output and employment in the construction sector grow between 2 and 3 percentage points more in bad booms than in good ones.
Possible mechanisms:
- Construction lacks the long-run growth potential of many other industries; overinvestment in construction may divert resources from more productive activities and lower output.
- Temporary boosts in construction employment and relatively low skill requirements may discourage investments in education and skills, with long-lasting effects on output after the boom.
- Construction projects have large up-front financing needs, and final consumers (for example, houses or hotels) tend to borrow to finance purchases, so debt may increase significantly more during construction-led booms.

A 1 percentage point increase in output growth in the construction sector during a boom raises the probability that the boom is bad by 2 percentage points.
A 1 percentage point increase in employment growth in the construction sector during a boom raises the probability that the boom is bad by 5 percentage points.
A 1 percentage point increase in output growth in the construction sector during a bad boom corresponds to nearly a 0.1 percentage point drop in aggregate output growth during the bust.
Construction growth is a stronger predictor of the economic costs of bad booms than many other variables.

If policymakers observe rapid expansion in the construction sector during a credit boom, they should consider tightening macroeconomic policies and using macroprudential tools (such as higher down payments for mortgages).
Policy action may also be triggered by other indicators (for example, house prices or household mortgages), but these indicators can sometimes fail to sound the alarm (for instance, when a construction boom is financed by the corporate sector or by foreigners).
Unusually rapid construction growth could signal the need to impose limits on banks’ exposure to real estate developers and other construction firms.
Because official output and employment data in construction are often available with a lag, higher-frequency indicators such as construction permit applications could provide valuable early signals.
Construction indicators should be included in models that assess risks to future economic activity.

Source: IMF blog post “Construction Activity Can Signal When Credit Booms Go Wrong” (February 12, 2020)

Staff Discussion Note
- Staff Discussion Note (Markdown version)
- Staff Discussion Note (PDF)