recent

How AI-empowered ‘citizen developers’ drive digital transformation

Our top 5 ‘Working Definitions‘ of 2024

How to manage two types of generative AI

Credit: iStock

Ideas Made to Matter

Finance

How credit conditions affect housing prices — lessons from the ’00s

By

How might credit conditions affect this year’s housing market? A look to the past can offer some clues.

Between 1997 and 2006, housing prices soared 74%. And even though two decades have passed since that boom and subsequent bust that led to the Great Recession, researchers have had trouble reaching consensus on the role that credit played in housing prices.

A new paper from MIT Sloan brings researchers one step closer to settling the debate — and may provide valuable guidance if the frenzy continues in the current housing market. The paper, “Do Credit Conditions Move Housing Prices?,” was co-authored by MIT Sloan finance professorand Boston University economics professor Adam Guren.

“I started to notice there was a lot of disagreement in the economic literature about the role of credit in the housing boom/bust,” Greenwald said. “As I read more, it just became clear that people are coming to all kinds of different conclusions on this topic.” In fact, two papers published in the “Journal of Political Economy,” published by the University of Chicago Press, “came to totally different, opposite conclusions,” with changes in credit explaining either most or none of the boom, depending on the paper.

“I thought, ‘This is really weird. We need to get to the bottom of this,’” Greenwald said.

Credit conditions are commonly cited as a driving force of house-price fluctuations. But one of the problems researchers face is that the role of credit cannot be pinned down using data alone.

“What you really need to know is what would have happened in a fictional universe where credit didn’t change,” Greenwald said, “and the data from that universe doesn’t exist.”

Solving the problem required building an economic model.

The role of the rental market

Greenwald and Guren’s work showed that existing models came to different conclusions largely because of assumptions made about the rental market. In some models, landlords and investors stand ready to trade vast quantities of housing at stable prices, absorbing increases in housing demand spurred by easy credit, thereby preventing a run-up in house prices. In other models, landlords and investors are basically irrelevant, and changes in credit conditions can spur wild fluctuations in house prices.

To pin down where the world lies between these two extremes, the researchers applied several real-world shocks to credit availability, outlined in previous work, including changes in the Fannie Mae/Freddie Mac Conforming Loan Limit, regulatory policy changes by the Office of the Comptroller of the Currency, and geographic variation in the funding structure of banks.

The key idea the researchers were testing: The more landlords absorb demand, the bigger the resulting change in the homeownership rate as they sell their holdings to households. The less that landlords absorb demand, the more house prices should change instead.

“Credit conditions matter” — then and now

To measure the bottom-line effect of credit during the 2000s boom and bust, Greenwald designed a model that combines these estimates of landlord flexibility with scenarios that capture changes in credit limits and interest rates observed over the boom period.

The results — estimated using data from the CBRE Economic Advisors, CoreLogic, and the U.S. Census — show that the market responds to credit changes mostly through house prices rather than homeownership. This implies that landlords are largely unable to absorb changes in household demand, and that credit conditions are capable of driving big swings in house prices.

Related Articles

This niche mortgage might ease the next housing bust
Here’s how much the 2008 bailouts really cost
Rethinking how the housing crisis happened

Results show that 34% of the rise in house prices (relative to rents) over the boom period can be directly attributed to relaxations in credit standards, while 72% can be attributed to the combination of changing credit standards and interest rates. All told, these results imply that credit conditions played a dominant role during the 2000s housing boom period, and remain central for investors today.

“For households and investors, credit conditions matter,” Greenwald said. “If you’re interested in the cyclicality of house prices, and what your risks are as an investor, it means that changes in credit conditions or the interest rate can have big effects.”

Tightening credit

Greenwald said that if policymakers wanted to curb growth in either house prices or credit during a boom period, keeping credit limits tight would have “a big effect on house prices.” Because the amount you can borrow against your house depends on its value,

“If you can stop house prices from going up, you have an indirect effect where you also stop people from borrowing more because their house is worth more," Greenwald said.

In the future, the authors hope to create their own estimates of local homeownership rates by using detailed address history data to improve the precision of their results.

“We’ve made a lot of progress in understanding why the models get different results,” Greenwald said. While he expects the debate to continue over how big of a role credit actually plays, he believes this new methodology will remain “the right way to think about the problem.”

Read next: 7 principles for the ‘perfect portfolio’

For more info Tracy Mayor Senior Associate Director, Editorial (617) 253-0065