Getting Past “Yes”: A Q&A on the Affordability Crisis (Part 2)

The City of Yes zoning text amendment proposal continues its reckless march through the public review process, improbably announcing “next stop: housing affordability,” but really inspiring zero confidence about where the hell it’s taking us. Part 1 of this series refuted the claim that we are confronting a generalized housing crisis and answered questions about the nature of the city’s affordable housing crisis. Today, we tackle in greater detail the City’s argument that there is widespread consensus among housing scholars and advocates that addressing the affordability crisis requires a housing supply increase at all income levels.

The extent of this overstatement makes our job easy, allowing us to rebut the City’s assertion with a single dissenting view. As it happens, though, housing researchers disagree about the relationship between housing supply and affordability at every step of the way. What follows is less a literature review than a brief overview of research areas where scholarly disagreement undermines the neat supply-side narrative that rationalizes the City’s current deregulatory campaign.   

Aren’t low vacancy rates what’s driving up rent prices? If that’s the case, wouldn’t you want to increase vacancy rates through new construction? 

First of all, as pointed out in part 1, substantial evidence forecasts a softening of the market. In addition, the lowest vacancy rates are disproportionately located at the bottom of the market, which is not reached by market-rate new construction. With regard to the main question, however, it is actually not clear whether high vacancy rates drive down rent prices. In theory, they should. But empirical findings have long called this relation into question. Eubank and Sirmans’ (1979) study of rental housing price changes in four metropolitan markets finds vacancy rates to have no significant effect on rent prices after controlling for operating expenses. An earlier study by de Leeuw and Ekamen (1971) reaches similar results, concluding that they have no significant effect after controlling for household income, costs, and price levels in the area. 

A coincidence between rising rental prices and rising vacancy rates is actually not so rare an occurrence. Belsky and Goodman (1996) studies the persistence of this phenomenon throughout the 1980s, offering as its explanation factors that include changes in the rent-setting behavior of landlords, an increase in the natural vacancy rate, and distortions in the measurement of the vacancy rate due to high levels of new construction. A more recent study by McClure (2018) looks at vacancy rates in the 455 metropolitan statistical areas in the United States and finds them to be “not even weakly significant in [their] capacity to explain variation in rent.” (ibid p.20).

Wouldn’t relaxing zoning restrictions help avoid a possible housing shortage by incentivizing new construction?

It’s actually not clear that it would. Upzonings, by definition, increase the development potential of rezoned areas. That’s why they typically boost land value and garner the enthusiastic support of land owners and developers. As any underachiever might tell you, however, potential possessed is different from potential realized. Empirically, numerous studies have found limited association between upzonings and new development. Freemark’s (2019) analysis of the short-term local impacts of two Chicago upzonings, for instance, discovers no increase in housing construction within five years of the zoning change, despite transactional evidence of increased property values within two. A Portland study by Dong (2021) of rezoning effects over a longer period, fifteen years, does identify an association between zoning increases and a higher probability of development. This amounted, though, to only a 2.5% difference between the percentage of upzoned parcels that saw any development and the percentage of those that did in a synthetic control group. 

More broadly, a review by Freeman (2023) of all peer-reviewed empirical English-language research on housing-related, land-use regulatory change as of the spring of 2023 concludes that studies into the impacts of deregulation on housing production at a neighborhood scale show mixed results. Some show an increase in housing development; some find none. It further determines that, at a regional level, there is not enough evidence to discern whether findings of increases in new construction result from upzonings or merely from a reallocation of development activity from non-upzoned to upzoned areas (ibid. p. 21). 

One of the possible explanations for the failure of upzonings to generate new housing construction is the existence of sufficient pre-upzoning buildable capacity to accommodate the appetite for new development. As pointed out in part 1, the current zoning in New York City would allow the construction of 1.8 billion square feet of residential development.

The one thing we do know is that relaxing zoning restrictions, like City of Yes proposes, can allow individual development projects to be bigger and taller than they now can be and to be built in open spaces where they now can’t. So looser zoning restrictions may not always result in more housing units being built overall. But they almost certainly increase the likelihood of out-of-context development that sticks out and is shoehorned into spaces where it otherwise wouldn’t go.

Wouldn’t relaxing zoning constraints lower housing prices by allowing suppliers of housing to better respond to the pent up demand for it?

Sure…

Great!

… if you assume perfect information, zero transaction costs, among other conditions that allow demand and supply curves to meet in happy equilibrium on classroom chalkboards.

Oh

Look, you could theorize a scenario in which an upzoning does exactly that, increases residential development and lowers housing costs. But you could also theorize a scenario in which the upzoning unleashes a wave of real estate speculation that leads to the demolition of older relatively affordable housing units, or in which it triggers a rush of new development that enhances neighborhood amenities and induces demand from affluent households that then bid up residential prices, and so forth. The highly contingent relation between zoning changes and housing prices explains the range of disparate findings among studies on the matter. Freemark’s (2023) review of the literature concludes that upzonings tend to be associated with increased property values on affected sites and that the evidence on whether new construction off-sets this effect on a per-unit basis is mixed. His own study of Chicago (2019), for one, finds an association within two years of the rezoning action between a 20% increase in allowable density and increases in property values of between 15% and 23.3%. Studies at a more regional level have also yielded results that diverge from basic economic theory. Limb and Murray (2022), for one, find no relationship between substantial increases in zoning capacity and housing prices in Brisbane over a twenty year period. This suggests that private development will not necessarily respond to price signals and mitigate residential costs even in the absence of zoning constraints. Housing prices, in fact, rose in Brisbane during the study period more rapidly than even in famously regulated, tight markets like San Francisco. 

But aren’t there studies showing that as housing supply grows housing prices drop?

To be sure, some studies show increases in housing supply exerting downward pressure on housing prices. But basing affordable housing policy on those findings would be a bit like deciding to stop the forward motion of your car by digging your bare heels into the ground Fred-Flinstone-style. You could do it. But it probably wouldn’t be your first (or second) choice. In part 3 of this series, we will examine why.

References:

Belsky, E., & Goodman, J. (1996). Explaining the Vacancy Rate-Rent Paradox of the 1980s. Journal of Real Estate Research, 11(3), 309–323. 

de Leeuw, F., & Ekanem, N. F. (1971). The Supply of Rental Housing. The American Economic Review, 61(5), 806–817.

Dong, H. (2024). Exploring the Impacts of Zoning and Upzoning on Housing Development: A Quasi-experimental Analysis at the Parcel Level. Journal of Planning Education and Research, 44(1), 403–415.

Eubank, A. A., Jr., & Sirmans, C. R. (1979). The Price Adjustment Mechanism for Rental Housing in the United States. The Quarterly Journal of Economics, 93(1), 163–168. 

Freemark, Y. (2020). Upzoning Chicago: Impacts of a Zoning Reform on Property Values and Housing Construction. Urban Affairs Review, 56(3), 758–789.

Freemark, Y. (2023). Zoning Change: Upzonings, Downzonings, and Their Impacts on Residential Construction, Housing Costs, and Neighborhood Demographics. Journal of Planning Literature, 38(4), 548–570.

McClure, K. (2019). The allocation of rental assistance resources: The paradox of high housing costs and high vacancy rates. International Journal of Housing Policy, 19(1), 69–94.

Murray, C., & Limb, M. (2023). We Zoned for Density and Got Higher House Prices: Supply and Price Effects of Upzoning over 20 Years. Urban Policy and Research.