Whether Upzoning Lowers Rents Shouldn’t Matter
Or why more people need to realize prices aren’t the only thing that matters in housing markets.
Almost all of the debates about housing policy over the last several years have revolved around prices. This is natural but unfortunate, as it has left the most important benefits of building more housing out of the public eye. More construction would indeed lower prices, but more construction is not good because prices would fall.
As the most obvious example of this general phenomenon, New York City mayor and democratic socialist Zohran Mamdani’s efforts to “freeze the rent,” as well as the resulting backlash, are all focused on prices. But so is the “YIMBY” vs. “NIMBY” cage match over zoning regulations. The dividing line between the two camps is invariably whether someone believes that upzoning and repealing other restrictive regulations will lower rents, or they believe that rents would stay the same or even go up.
Building is good. (Because it lowers prices.) Or building is bad. (Because it raises prices.) Even when nary a soul would be displaced by a proposed development, like the Hudson Yards project in New York or a multifamily proposal meant for a parking lot here in the small town I stay in as a PhD student, pushback usually comes in the form of a couple insistent questions: How many units will really be “affordable”? And what will the average monthly rent be?
The Argument’s Jerusalem Demsas, in her recent column making the case that housing regulations really do matter, reasonably chose to focus mostly on how these regulations affect the cost of housing. While rebutting anti-YIMBY arguments for the Roosevelt Institute, Ned Resnikoff, also a great policy wonk, again focused on claims about whether increasing supply would meaningfully reduce prices.
This focus on prices does have an obvious, proximate cause. The most salient and painful part of any contract for housing—whether a mortgage or rental agreement—is the monthly payment. Once a month, every month, a large sum of money leaves your bank account. That sucks. Like me, you probably get an email from your bank or credit union confirming that your checking account balance is now appreciably lower. That also sucks.
Despite this, focusing just on a single aspect of the housing market, even one as important as prices, requires eliding the most interesting parts of housing markets, the factors that make renting a new place to live different (and more difficult) than buying a bag of dried lentils at your local grocery. As understandable and even rational as our current price monomania may be, we cannot actually understand the benefits of increasing the housing supply unless we all take a collective second to step back and think about what it really means to purchase a home or rent an apartment.
Almost invariably, when you make a major purchase you aren’t really buying different quantities of one discrete thing. You’re considering a lot of “bundles” of distinct qualities and then trying to pick the best bundle given your budget. Think about the experience of purchasing a new laptop and scrolling through a store’s inventory. You’re probably checking each model’s storage, RAM, CPU specs, and screen size and resolution, at minimum.
In quantitative economics, this is the core insight underlying so-called hedonic pricing models. First pioneered by economist Andrew Court in the late 1930s, hedonic pricing models recognize that the value of a good is really determined by a combination of several characteristics. Court, as an employee of the Automobile Manufacturers Association, set about trying to explain why the Bureau of Labor Statistics estimated car prices had risen more than 40 percent over a decade. The types of automobiles being sold had obviously changed between 1925 and 1935, he observed, but the BLS indices implicitly assumed cars of a given model were comparable, and at the same time ignored some components that had gone from add-ons to defaults.
“In the case of passenger cars, if the relative importance to the customer of horse-power, braking capacity, window area, seat width, tire size, etc., could be established, the data reflecting these characteristics could be combined into an index of usefulness and desirability,” Court wrote in his 1939 article. “Prices per vehicle divided by this index of Hedonic would yield valid comparisons in the face of changing specifications.”
Since the revival of hedonic pricing by economist Zvi Griliches in the 1960s, the descendants of Court’s early models have now been used extensively in real estate economics. After all, the same logic that governs shopping for a laptop applies when people are participating in the housing market. This becomes patently obvious when browsing Zillow.
Every unit in every housing market is defined by a very important set of characteristics: the number of bedrooms, the number of bathrooms, the square footage, and, most importantly, the location, inter alia. When the median person enters the housing market as a consumer, they aren’t looking for the cheapest roof and four walls currently on the market or else everyone would be moving to Todd County, South Dakota. Consumers of housing weigh monthly rent or mortgage costs against all of these other characteristics and then try to find the best match for their own preferences and budget.
The lack of attention paid to these other non-price determinants of demand is the single factor which has most impoverished the public’s understanding of why upzoning and other regulatory changes are important. An ocean of digital ink has been spilled debating the precise impacts of policy changes on prices. Scholars try to measure both quantity and price changes in their work, but podcasts and articles about YIMBYism and its critics usually begin with notes like “Housing prices have been soaring,” or “For years, housing activists couldn’t agree on what was driving prices higher, let alone how to fix it.” Even academic articles on both sides of the divide are fairly regularly published with ‘catchy’ titles like “We Zoned for Density and Got Higher House Prices.” The conversations on X, né Twitter, are naturally far worse.
Instead of accepting the premises of this interminable back-and-forth, policy makers and housing advocates ought to heed Court and his successors and recognize that any policy change which increased the housing supply but kept prices constant would still have significant public benefits. If a city adds more units, that can be good independently of what happens to prices.
Let’s imagine, if you will, the NIMBY’s worst nightmare: A magic slider that allows a city planner to frictionlessly increase the number of units in a given area. Move the slider and you can instantaneously double the number of units in some local housing market, keeping the composition of housing characteristics constant. (I.e., if 10 percent of units were studio apartments before, 10 percent are studio apartments after the slider; if 5 percent of units rented for below $700 a month before, 5 percent of units rent for below $700 a month afterwards.)
If your understanding of housing markets has been entirely shaped by the foci of recent debates, then this magic device would appear utterly pointless. Sure, it increases the supply of housing, but by its very construction the slider has absolutely no effect on prices. In reality, though, this sort of adjustment would be miraculous.
As I highlighted above, one of the most important determinants of demand for housing is location. Really it’s location, location, location that explains why a mansion way out in the boonies costs a fraction of what an apartment in New York does, an observation core to the foundational model of modern urban economics. And what increasing the supply of housing in an area does, in addition to any price effects, is massively increase the number of options facing consumers in the housing market. (In economics jargon, it adds to their “choice set.”)
No one can move into a unit that is already fully occupied by another household. And while it is possible to reproduce almost all of the bundle of characteristics that define someone’s dream home just about anywhere in the world, construction companies, no matter how hard they try, cannot manufacture points in physical space.
But what a construction company can’t literally do, it can metaphorically do by increasing density. By providing more options at a specific location, increasing density grants more families the opportunity to live in a certain neighborhood. Increasing density lets more workers live within a reasonable commuting distance from their employers and allows more people to find a spot next to their preferred bundle of spatial amenities, like a lovely view of the Pacific Ocean outside their window. And every single one of the people who would choose to live in these more dense neighborhoods would have their lives improved.
Many of my colleagues here at UMass Amherst critique the standard economic model of a consumer as a “rational utility maximizer” but it is self-evident that when folks choose to make a major decision like buying a house, they choose to make the transaction that offers them the most benefit out of all options available to them that they were aware of. By which I simply mean that whenever someone chooses to rent or purchase a unit, that decision was invariably the best, most-preferred option for their situation.
So if someone moves into a newly built unit, then that unit must have better catered to their tastes given their budget than all of the other units out on the market while they were searching for a new place to live. The people who move into their old place must have decided that that now vacant unit was the best fit for them, and so on for the people who move into the unit left empty by that move. Supposing that that initial unit of new construction had never been built, ceteris paribus, the would-be resident and everyone else in that chain of movers would have been forced to move into a less preferred unit and been left worse off.
Every new unit that is built and occupied helps someone just by existing. On Earth-2, where the housing market is centrally planned by HUD and all prices are set in stone, building more housing still improves people’s lives.
Personally, I do in fact believe that dramatically increasing the supply of housing would lower rents. But the excessive focus on this one particular question has unfortunately resulted in an understanding of increasing the housing stock as an instrumental goal, as only a method to reduce prices. Meanwhile, a relatively neglected question of equal or greater import is how policy reforms can best increase the supply of units regardless of price effects. More eyes need to be directed toward the Center for Public Enterprise’s reports on housing supply tools or FRED graphs of housing starts, and fewer on the latest catchily titled op-ed about why YIMBYism won’t make San Francisco affordable.
America needs much more housing: Every day, renters and prospective homeowners across the country are forced to settle and spend their lives in subpar units. When YIMBYs and housing policy analysts are making the case for giving everyone more choices of where to live, price effects alone simply cannot fully capture the benefits of their proposals.