Research Highlights Article
November 20, 2025
Market segmentation and housing prices
Did relaxed credit standards drive the housing boom that led to the Great Recession?
Source: l i g h t p o e t
It is generally accepted that the housing boom and bust of the 2000s was a key factor leading to the Great Recession. But economists have struggled to reach a consensus on the role that credit availability played in driving house prices during this period. Some researchers have argued that loosened credit standards, such as mortgage lending restrictions, explained most of the housing cycle, while others have found that credit played little role.
“Ten years after the crash, we still didn’t know the answer to what seems like a fundamental question,” economist Daniel L. Greenwald told the AEA.
In a paper in the American Economic Review, Greenwald and coauthor Adam Guren took a step toward resolving this puzzle by identifying a previously overlooked factor that explains these contradictory findings. The authors demonstrated that the degree of segmentation between rental and owner-occupied housing significantly determines how credit affects house prices. If rental properties cannot easily be converted to owner-occupied homes, expanding credit drives up prices. But if landlords are willing to sell to households at the present value of rents, easier credit primarily increases homeownership rates rather than prices.
First, the authors examined the response of price–rent ratios and homeownership rates to credit supply shocks. They observed that the relative magnitude of these two responses reveals the slope of what they term the "tenure supply curve," which captures how willing landlords are to sell rental properties as prices rise. A steeper curve indicates a more inelastic supply—and hence strong segmentation—while a flatter curve indicates a more elastic supply—suggesting a more frictionless market.
Price–rent ratio versus homeownership rate
The chart below shows the evolution of the price–rent ratio and homeownership rate since 1965. The gray dots represent years before the housing boom, the blue dots represent years during the boom, and the red dots represent years after the housing boom.
To quantify the actual slope of the tenure supply curve, and hence the degree of segmentation, the researchers applied three different identification strategies from the existing literature. The first exploits differential city-level exposure to changes in conforming loan limits—a standard used by Fannie Mae and Freddie Mac to buy mortgages. The second uses the 2004 federal preemption of state antipredatory lending laws, which affected areas differently based on their level of exposure to banks that are regulated by the Office of the Comptroller of the Currency. And the third leverages variation in exposure to the expansion of private label securitization through differences in bank funding sources across regions.
To improve measurement precision, the researchers also constructed a new local homeownership rate measure using microdata on property ownership records and individual address histories for US housing units. Their new dataset covers 390 metropolitan areas and provides substantially less noisy estimates than the Census Housing and Vacancy Survey, which covers only 75 areas and suffers from small sample sizes.
Despite using different sources of variation affecting different parts of the mortgage market, all three empirical strategies yielded similar results. Credit supply shocks significantly increased house prices and price–rent ratios but had much smaller, often statistically insignificant, effects on homeownership rates. The price–rent ratio responded at least 3.8 times more strongly than the homeownership rate across all specifications, with many substantially larger, suggesting a steeply sloped tenure supply curve. This finding suggests that US housing markets, overall, have strong frictions—it's not easy to quickly convert rental properties to owner-occupied homes, so easier credit mostly drives up prices rather than expanding homeownership.
To apply these findings to the housing boom, the authors developed a dynamic equilibrium model that incorporated both extremes of perfect segmentation and frictionless markets. When calibrated to match the empirical estimates, the model indicated that relaxing credit standards alone explained 32 to 50 percent of the rise in price–rent ratios during the boom. When combined with the 2 percentage point decline in mortgage rates during this period, credit factors explained 70 percent of the observed price increase. For comparison, the model suggested that under no segmentation, credit conditions explained only 2 percent of the boom, while under full segmentation, they explained 77 percent of the boom.
In a world with no segmentation, where credit doesn't affect the house price at all, a lot of these policies would be much less effective. If you want to design macroprudential policies that work not only directly on credit, but indirectly through the price of housing, these results are pretty important.
Daniel Greenwald
These findings may have important implications for macroprudential policies that tighten credit standards. Policies involving loan-to-value or debt-to-income limits can effectively dampen housing cycles, but only in the presence of significant market segmentation. In the calibrated model, preventing the credit expansion during the boom would have reduced the rise in price–rent ratios by 53 percent. However, in a counterfactual frictionless market, the same policies would have had a minimal effect on prices, reducing their overall effect on mortgage volumes.
“In a world with no segmentation, where credit doesn't affect the house price at all, a lot of these policies would be much less effective,” Greenwald said. “If you want to design macroprudential policies that work not only directly on credit, but indirectly through the price of housing, these results are pretty important.”
The authors’ research highlights the importance of market structure in determining how financial shocks propagate through the economy.
♦
“Do Credit Conditions Move House Prices?” appears in the October 2025 issue of the American Economic Review.