Why Is Housing So Expensive?
How Can We Fix It?

Building a Manhattan apartment cost about $200 a square foot while it sells for near $600“If it costs $200 per square foot to build an apartment that sells for $600 per square foot, then this would seem to offer an irresistible arbitrage opportunity for developers.”Why Is Manhattan So Expensive? Regulation and the Rise in Housing Prices (2005) ↗, and in a competitive industry with almost no barriers to entry“Home building is a highly competitive industry with almost no natural barriers to entry, yet prices in Manhattan currently appear to be more than twice their supply costs.”Why Is Manhattan So Expensive? Regulation and the Rise in Housing Prices (2005) ↗, that gap should vanish. G&G assume housing markets are competitive: a good’s price cannot exceed its marginal cost of production“in an open, competitive, unregulated market, the price of a commodity will not be greater than the marginal cost of producing that good”Why Is Manhattan So Expensive? Regulation and the Rise in Housing Prices (2005) ↗. When housing sells for far more than the cost of supplying it, something must be blocking new building—and that something, they argue, is regulation.

Their original method is built on single-family homes, where lots vary in size. They estimate what an extra unit of land is worth by comparing otherwise-similar houses on larger and smaller lots—backing the price of land out of the cost of the unit“we must estimate the price of land as part of the marginal cost of the housing unit”Why Is Manhattan So Expensive? Regulation and the Rise in Housing Prices (2005) ↗. They treat that marginal land price as what land should cost throughout, so the gap between it and the land’s average value is the regulatory tax“A completely free market for land would lead land to be worth the same amount on both the intensive and extensive margins.”Why Have Housing Prices Gone Up? (2005) ↗, and across the coastal metros it reaches a third to a half of a home’s value“the gap between prices on the extensive and intensive margins amounts to from one-third to one-half of home value at the mean”Why Is Manhattan So Expensive? Regulation and the Rise in Housing Prices (2005) ↗.

The high-rise case is simpler: in a city like Manhattan you add homes by building up, so the marginal cost of one more unit is the cost of adding an extra floor“the marginal cost of supplying more space is reflected in the cost of adding an extra floor to any new building”Why Is Manhattan So Expensive? Regulation and the Rise in Housing Prices (2005) ↗, and land, a fixed cost, drops out of the calculation“these are fixed costs which do not influence the marginal cost of building up”Why Is Manhattan So Expensive? Regulation and the Rise in Housing Prices (2005) ↗. That is why they anchor on Manhattan, where the construction numbers are clean: an apartment selling for more than twice its supply cost reveals a wedge competitive building should have erased.

By this metric regulation is heaviest in San Francisco and New York“The most highly regulated markets are on the two coasts, with the San Francisco and New York City metropolitan areas being the most highly regulated according to our metric.”The Local Residential Land Use Regulatory Environment Across U.S. Housing Markets (2021) ↗, and the Manhattan regulatory tax runs to over half a condo’s value“For the median Manhattan condominium in our sample, the regulatory tax amounts to well over half of the total price of the unit.”Why Is Manhattan So Expensive? Regulation and the Rise in Housing Prices (2005) ↗. Most of the country looks different—nearly three-quarters of homes sell near or below construction cost“slightly less than three-quarters of all observations (73.6 percent) are priced near or below minimum profitable production costs”The Economic Implications of Housing Supply (2017) ↗.

A newer puzzle is that once-permissive cities like Miami, Phoenix and Dallas“This combination of facts raises the possibility that once elastically supplied housing markets such as Miami, Phoenix and Dallas are becoming more like the supply-constrained coastal markets that we identified decades ago.”America's Housing Supply Problem: The Closing of the Suburban Frontier? (2025) ↗ have grown expensive too. Glaeser and Gyourko rule out rising costs (construction costs rose by about a quarter since 2000“Real construction costs have risen by about 25 percent per home since the early 2000s.”America's Housing Supply Problem: The Closing of the Suburban Frontier? (2025) ↗) and rising demand (production fell as prices rose“We interpret these rising prices as reflecting a downward shift in supply rather than an upward shift in demand because the quantity of housing production has declined substantially over the last half century.”America's Housing Supply Problem: The Closing of the Suburban Frontier? (2025) ↗), ascribing the shift instead to these cities growing more NIMBY“A factor that increases demand for a location may also bring in wealthier or better educated residents who may then alter the permitting environment.”America's Housing Supply Problem: The Closing of the Suburban Frontier? (2025) ↗.

What matters, in Hsieh and Moretti’s account, is what a city does when it grows more productive—and that hinges on its housing supply elasticity“the housing supply elasticity varies significantly across US cities and that it is an important determinant of local housing costs”Housing Constraints and Spatial Misallocation (2019) ↗: how readily it builds. Where building is easy, a productivity boom pulls workers in and shows up as more jobs, and the gains are shared across the country. Where zoning holds construction back, the same boom shows up as higher rents instead“Instead of increasing local employment, productivity growth in housing-constrained cities primarily pushes up housing prices and nominal wages.”Housing Constraints and Spatial Misallocation (2019) ↗, and the workers who would have moved in are priced out and left in less productive places“Misallocation arises because high productivity cities like New York and the San Francisco Bay Area have adopted stringent restrictions to new housing supply, effectively limiting the number of workers who have access to such high productivity.”Housing Constraints and Spatial Misallocation (2019) ↗. Those rents clear the market, so nothing looks broken—but they are a scarcity price created by the constraint, not a real cost of building, and the output those blocked workers would have produced is the loss the paper is chasing.

The evidence they point to is that cities’ wages have pulled apart. Adjusting for who lives where, the spread of wages across cities roughly doubled between 1964 and 2009“the standard deviation of nominal wages (in logs) across US cities in 2009 is twice as large compared to 1964”Housing Constraints and Spatial Misallocation (2019) ↗—a sign that productivity gaps between places have widened while workers no longer move to capture them—and, they contend, most of that widening is about housing, not productivity itself“most of the increased spatial dispersion in the marginal product of labor is due to the growing spatial dispersion in housing prices”Housing Constraints and Spatial Misallocation (2019) ↗.

To judge how much of New York’s edge is excess rather than deserved, they run a counterfactual: what if New York, San Jose and San Francisco built housing as readily as the median US city, holding their productivity fixed? Workers flow in until the extra output of the last one just covers the extra cost of housing them—New York ends up larger and still more productive than the rest, only no longer inefficiently so. The model puts the difference at 36.3 percent faster aggregate growth“increasing housing supply in New York, San Jose, and San Francisco by relaxing land use restrictions to the level of the median US city would increase the growth rate of aggregate output by 36.3 percent”Housing Constraints and Spatial Misallocation (2019) ↗, with 2009 GDP about 3.7 percent—$3,685 a worker—higher“US GDP in 2009 would be 3.7 percent higher, which translates into an additional $3,685 in average annual earnings”Housing Constraints and Spatial Misallocation (2019) ↗; re-scored this way, the three cities’ real contribution to national growth drops from a headline 12 percent to only 5 percent“While the accounting measure suggests that New York, San Francisco, and San Jose’s contribution to aggregate GDP growth between 1964 and 2009 is 12 percent, viewed through the lenses of our model, these cities were only responsible for 5 percent of growth.”Housing Constraints and Spatial Misallocation (2019) ↗.

The size of that figure rests on three inputs. The first is how fast New York’s marginal product falls as workers crowd in, which sets how much of its premium counts as excess; the authors note their result is sensitive to the assumed degree of diminishing returns“While results are not sensitive to changes in labor or capital share for a given degree of return to scale, they are quantitatively sensitive to the degree of decreasing return to scale.”Housing Constraints and Spatial Misallocation (2019) ↗. The second is the housing-supply elasticity of each city, estimated from data, which determines how many workers a deregulated New York would actually absorb. The third is the assumption that New York stays exactly as productive with millions more residents—that its productivity is a fixed feature of the place. Their headline numbers follow from these choices.

For most of American history poorer places caught up with richer ones, as workers migrated toward higher wages. Ganong and Shoag show that escalator has broken. Incomes converged for over a century at 1.8 percent a year“For over a century, incomes across states converged at a rate of 1.8% per year.”Why Has Regional Income Convergence in the U.S. Declined? (2017) ↗; then from 1990 to 2010 convergence ran at less than half the norm“The convergence rate from 1990 to 2010 was less than half the historical norm, and in the period leading up to the Great Recession there was virtually no convergence at all.”Why Has Regional Income Convergence in the U.S. Declined? (2017) ↗. Housing is the hinge. A janitor's New York wage is higher in nominal terms“In nominal terms, the wages of lawyers and janitors are 46% and 28% higher in the New York area respectively in 2010. After adjusting for housing prices, these premia are 39% and -7%.”Why Has Regional Income Convergence in the U.S. Declined? (2017) ↗, but housing eats 21 percent of even a lawyer's income there“for lawyers in the New York area, housing costs are equal to 21% of their income, while housing costs are equal to 52% of income for New York area janitors.”Why Has Regional Income Convergence in the U.S. Declined? (2017) ↗—and far more of the janitor's.

Where wages once pulled workers in, prices now rise one-for-one with income“In 1960, housing prices are 1 log point higher in a state with 1 log point higher income. By 2010, the slope doubles…”Why Has Regional Income Convergence in the U.S. Declined? (2017) ↗, a break that dates to the 1970s“Growth is particularly rapid from 1970, when it stood at about 25% of its current level, to 1990, when it reached about 75% of its present day level.”Why Has Regional Income Convergence in the U.S. Declined? (2017) ↗. The proof is in the exceptions—convergence persists where building is unconstrained“income convergence persists among places unconstrained by these regulations, but it is diminished in areas with supply constraints”Why Has Regional Income Convergence in the U.S. Declined? (2017) ↗—and in the cause: places already disposed to regulate—measured before the fact—show the same slowdown“The fact that the latent tendency to regulate shows the same pattern as post-period regulations, suggests that causality runs from regulations to convergence and not vice versa.”Why Has Regional Income Convergence in the U.S. Declined? (2017) ↗. The result is a sorting reversal: high-skill workers move to high-income places, low-skill workers away“High-skill workers are moving to high-income places, low-skill workers are moving away, and the net effect summing across skill groups is that on average, there is zero directed migration.”Why Has Regional Income Convergence in the U.S. Declined? (2017) ↗.

A more recent study by William Olney and Owen Thompson, tracking the IRS record of every move between commuting zones, corroborates the mechanism. They find migration falls with destination home prices“migration is decreasing with origin wages and destination home prices, and is increasing with destination wages and origin home prices”The Determinants of Declining Internal Migration (Olney & Thompson, NBER 2024) ↗, and that wages alone would have raised migration“wages on their own would have led to an increase in migration rates”The Determinants of Declining Internal Migration (Olney & Thompson, NBER 2024) ↗—the pull of high-wage destinations has grown, not faded. But housing factors have more than offset wages and impeded mobility“these wage effects have been more than offset by housing related factors, which have increasingly impeded internal mobility”The Determinants of Declining Internal Migration (Olney & Thompson, NBER 2024) ↗, so that households that once would have left expensive places now stay put“migration has become much less responsive to housing prices in the origin CZ, such that many households that would have left in response to high home prices several decades ago now choose to stay”The Determinants of Declining Internal Migration (Olney & Thompson, NBER 2024) ↗.

Saiz set out to explain why a rise in demand produces new construction in some cities but mostly higher prices in others—why housing supply elasticity varies from place to place. His answer: a city’s elasticity is a function of both physical and regulatory constraints“housing supply elasticities can be well characterized as functions of both physical and regulatory constraints”The Geographic Determinants of Housing Supply (2010) ↗. On the physical side, he uses satellite terrain data to measure how much land a city actually has to build on: steep slopes cover 47.6 percent of the land around Los Angeles“Steep-slope block groups encompassed 47.62% of the land area within 50 km of LA’s geographic center in year 2000. However, only 3.65% of the population within this 50-km radius lived in them.”The Geographic Determinants of Housing Supply (2010) ↗, and moving across the range of land availability sharply lowers supply elasticity“a movement across the interquartile range in geographic land availability in an average-regulated metropolitan area of 1 million is associated with shifting from a housing supply elasticity of approximately 2.45 to one of 1.25”The Geographic Determinants of Housing Supply (2010) ↗. On the regulatory side, land-use regulation binds on top of geography, and it tightens as prices and population grow“the regulation equations in Table IV demonstrate that higher housing prices, demographic growth, and natural constraints beget more restrictive land-use regulations”The Geographic Determinants of Housing Supply (2010) ↗—so the two reinforce each other. By this combined measure, his elasticity estimates correlate 0.65 with house prices“These elasticity estimates display a very strong correlation of .65 with housing prices in 2000.”The Geographic Determinants of Housing Supply (2010) ↗.

The regulation half of that picture comes with a caveat Saiz himself co-wrote. He is a co-author of the Wharton land-use index, the standard measure of regulatory stringency, and its creators report that the richer and more highly-educated places have the most regulated environments“it is the richer and more highly-educated places that have the most highly regulated land use environments”A New Measure of the Local Regulatory Environment: The Wharton Index (Gyourko, Saiz & Summers, 2008) ↗—median family income correlates 0.33 with the regulation index“a simple correlation coefficient of 0.33 with our regulation index”A New Measure of the Local Regulatory Environment: The Wharton Index (Gyourko, Saiz & Summers, 2008) ↗. They add the warning their successors often drop: no causal relationship can be inferred from these simple correlations“While no causal relationship can be inferred from these simple correlations”A New Measure of the Local Regulatory Environment: The Wharton Index (Gyourko, Saiz & Summers, 2008) ↗, leaving open whether regulation raises prices or whether wealth produces both the regulation and the prices.

Saiz did not build his elasticities as a price instrument, but that became one of their most influential uses: later researchers treat a city’s elasticity as a way to isolate the part of house-price movement that is driven by supply rather than demand—the “clean” variation they need to study how prices affect spending, borrowing, and jobs.

Manville, Lens and Monkkonen answer the claim that inequality, not supply, drives housing costs. If rich buyers simply bid up everything, a Cadillac Escalade would cost far more in San Jose than elsewhere—it doesn't“If high earners drive prices, why do Cadillac Escalades cost about the same in San Jose and Kansas City, while San Jose’s median home is seven times more expensive than its Kansas City counterpart?”Zoning and Affordability: A Reply to Rodríguez-Pose and Storper (2022) ↗; only the goods that can't be supplied, like housing, diverge. Even if every home built in San Francisco since 2010 had sold to the wealthy“even if every owner-occupied home built in San Francisco since 2010 sold for over US$1 million, those new homes could explain little of San Francisco’s high prices, because only 2% of the region’s owner-occupied homes were built in that time, while 32% of the region’s homes are valued at US$1 million or more.”Zoning and Affordability: A Reply to Rodríguez-Pose and Storper (2022) ↗, the affordability math would not add up, because the most expensive housing is old, not new“In regions with affordability problems, lots of new housing is expensive, but most expensive housing isn’t new.”Zoning and Affordability: A Reply to Rodríguez-Pose and Storper (2022) ↗.

And new market-rate building still reaches the bottom: 100 new luxury units free up 65 homes in below-median areas, 34 in the poorest fifth“‘Building 100 new luxury units’, he concludes, ‘leads 65 and 34 people to move out of below-median and bottom-quintile income neighbourhoods, respectively, reducing demand and loosening the housing market in such areas’”Zoning and Affordability: A Reply to Rodríguez-Pose and Storper (2022) ↗ through chains of moves. Their stance is not triumphalist—upzoning is necessary but not sufficient“Virtually every academic supporter of upzoning considers it necessary but not sufficient. A strong role remains for low-income housing subsidies, and we support dramatic increases in such redistribution to low-income people.”Zoning and Affordability: A Reply to Rodríguez-Pose and Storper (2022) ↗—but supply, they insist, is not optional.

Caplan makes the maximalist case: housing deregulation is the closest thing economics has to a panacea. He rules out demand—population grew even faster back when housing was cheap“It isn’t rising demand, as the U.S. population rose even faster back when housing prices were roughly stable. It isn’t because of higher construction costs — those, adjusted for inflation, have been almost flat for decades.”The Best Plan for Housing Is to Plan Less (NYT, 2024) ↗—and blames the rules. Heavily regulated New York and the Bay Area are expensive and build little, while Houston and Dallas build and stay cheap.“Strictly regulated urban areas like New York City and the Bay Area have high prices and low construction, while more lightly regulated areas like Houston and Dallas have much lower prices and much more construction.”The Best Plan for Housing Is to Plan Less (NYT, 2024) ↗ Strip the rules, he argues, and prices would fall roughly 50 percent“According to a conservative estimate, prices would ultimately fall about 50 percent on average nationally — with significant, wide-ranging implications.”The Best Plan for Housing Is to Plan Less (NYT, 2024) ↗. Because shelter is a fifth of the budget, halving it would raise living standards about 11 percent“Since shelter is now roughly 20 percent of the average American’s budget, halving its price makes the cost of living 10 percent lower — and the standard of living 11 percent higher.”The Best Plan for Housing Is to Plan Less (NYT, 2024) ↗—and the gains run further still, since the economist Matthew Rognlie traces much of rising wealth inequality to housing“What few laypeople have heard, however, is that the economist Matthew Rognlie almost immediately followed up with evidence that it is the rising value of homes that primarily accounts for the increasing disparity between the rich and the poor.”The Best Plan for Housing Is to Plan Less (NYT, 2024) ↗.

Klein and Thompson make the abundance case: scarcity is a choice, written into rules. Understanding American housing, they argue, means understanding a 50-year story that is mostly about rules“to really understand housing in America, you need to understand a 50-year story, which is mostly about rules; a 20-year story, which is about business cycles; and a five-year story…”What Worries Me Most About 'Abundance' (The Ezra Klein Show, 2026) ↗. Texas is the rebuttal-by-example—it has billionaires and corporate interests, but an entirely different supply story“Texas has billionaires. Texas has corporate interests. But Texas also has an entirely different set of rules and customs and permitting regulations that simply makes it easier for supply to respond to demand.”What Worries Me Most About 'Abundance' (The Ezra Klein Show, 2026) ↗. Dallas added a population the size of Boston“Dallas, between 2019 and the early 2020s, added a population equivalent to the size of urban Boston. Hundreds of thousands of people moved into the Dallas metro area.”What Worries Me Most About 'Abundance' (The Ezra Klein Show, 2026) ↗, and yet metro prices actually fell“But what happened instead is that housing prices in the metropolitan area have actually declined over the last three and a half years. Dallas built so much that construction increased per capita throughout this period.”What Worries Me Most About 'Abundance' (The Ezra Klein Show, 2026) ↗; Austin's rents have dropped for 18 to 24 months straight“Average rents have gone down, down, down over the last 18 to 24 months. Austin is like the canonical story here.”What Worries Me Most About 'Abundance' (The Ezra Klein Show, 2026) ↗. The logic is mechanical: restrict supply and prices rise, add supply and they stabilize, add enough and they fall“If you restrict supply, prices go up; if you add supply, prices stabilize; and if you add enough supply, prices can actually go down.”What Worries Me Most About 'Abundance' (The Ezra Klein Show, 2026) ↗. Their preferred outcome is rents falling because building got easier—something better than a rent freeze“We have something better than a rent freeze. We have rents going down because we’ve made it easier to build.”What Worries Me Most About 'Abundance' (The Ezra Klein Show, 2026) ↗.

The AEI Housing Center's prescription is 'light-touch density'—modest infill, not towers. Houston is the showcase: reform legalized lots as small as 1,400 square feet“The reform legalized lots as small as 1,400 square feet with 720 square feet of compensating open space per lot.”Learning from Houston's Townhouse Reforms (Emily Hamilton, Mercatus, 2023—their key Houston exhibit) ↗, and nearly 80,000 small-lot houses followed“In Houston, nearly 80,000 houses have been built on small lots made possible by minimum lot size requirements.”Learning from Houston's Townhouse Reforms (Emily Hamilton, Mercatus, 2023—their key Houston exhibit) ↗, leaving the metro the most affordable big region, at 3.3 times income“Adjusting for income, Houston is the most affordable of the regions with a median house price 3.3 times its median income.”Learning from Houston's Townhouse Reforms (Emily Hamilton, Mercatus, 2023—their key Houston exhibit) ↗ against California's far higher ratio“In 1970 California’s ratio of median home price to median household income was about 2.2, about the same as for the nation. By 2020, California’s ratio had risen to 8.4 while the nation stood at 4.4.”California Light-Touch Density (AEI Housing Center) ↗. The gains, they argue, come without cost to neighbors—Seattle's light-touch zones added 20,000 units“Since 1994, Seattle’s LRM zones have added 20,000 new LTD housing units, while removing 5,000 older single-family detached (SFD) units, increasing the supply of housing in these areas by on average 2.5% per year.”Does Building Light-Touch Density Housing Lower Single-Family Home Values? (AEI Housing Center, 2025) ↗ with no detectable price penalty next door over 25 years“We find no meaningful differences in HPA between SFD homes in the city’s SFD-only zones compared to SFD homes in its LTD zones over the past 25 years, indicating that LTD construction does not devalue nearby SFD homes.”Does Building Light-Touch Density Housing Lower Single-Family Home Values? (AEI Housing Center, 2025) ↗.

And the model scales: Tokyo's market-and-property-rights system leaves little room for local interference“The market and property rights based system with little possibilities for local interference has allowed Tokyo to build its way to abundant housing.”Does Our Solution Lie in Tokyo? (AEI Housing Center, 2022) ↗, so its housing stock has tripled since 1963“Since 1963, its housing stock has tripled in size, from 2.5 million homes in 1963 to 7.6 million in 2018.”Does Our Solution Lie in Tokyo? (AEI Housing Center, 2022) ↗.

Asquith, Mast and Reed test the fear that new market-rate buildings gentrify and displace. Tracking 686 large new buildings and the moves around them“I identify 686 large new market-rate multifamily buildings in central cities and track 52,000 of their current residents to their previous building of residence.”JUE Insight: The Effect of New Market-Rate Housing Construction on the Low-Income Housing Market (2023) ↗, they find the opposite: a new building lowers nearby rents 5 to 7 percent relative to trend“the average new building lowers nearby rents by 5 to 7 percent relative to trend, translating into a savings of $100 to $159 per month”Local Effects of Large New Apartment Buildings in Low-Income Areas (2023) ↗. The mechanism is migration—100 new units pull 45 to 70 people out of below-median neighborhoods“building 100 new market-rate units leads 45-70 and 17-39 people to move out of below-median and bottom-quintile income tracts, respectively, with almost all of the effect occurring within five years”JUE Insight: The Effect of New Market-Rate Housing Construction on the Low-Income Housing Market (2023) ↗—and the submarkets are linked, so the same patterns appear across building types“Similar patterns emerge for other characteristics, suggesting strong connections between submarkets that are inconsistent with a highly segmented market in which new construction does not affect low-income areas.”JUE Insight: The Effect of New Market-Rate Housing Construction on the Low-Income Housing Market (2023) ↗. Block that construction, they note, and high-income households simply bid up the existing stock instead“If high-income households like a particular neighborhood, preventing the construction of new housing simply leads them to outbid lower-income people for the housing already available in that neighborhood.”Local Effects of Large New Apartment Buildings in Low-Income Areas (2023) ↗.

Li takes on the neighbor's objection: does a new high-rise next door raise your rent? Two forces compete—new supply should ease prices, while new towers draw high earners and amenities that lift them. Studying 916 New York high-rises“Among the 916 new high-rises that completed before 2013, 638 are condos, 10 are co-ops, and others are rental buildings”Do New Housing Units in Your Backyard Raise Your Rents? (2022) ↗, she finds supply wins: a 10 percent rise in the local stock cuts rents about 1 percent“for every 10% increase in the housing stock, rents decrease by 1%”Do New Housing Units in Your Backyard Raise Your Rents? (2022) ↗, working through close substitutes like condos, not co-ops or houses“New high-rises only decrease sales prices of condos, presumably because co-ops and 1–5 family homes do not compete with new high-rises directly in the real estate market.”Do New Housing Units in Your Backyard Raise Your Rents? (2022) ↗. The amenity pull is real but small—each tower draws about 0.11 new restaurants“new high-rise brings in or attracts 0.11 new restaurants within 500 ft.”Do New Housing Units in Your Backyard Raise Your Rents? (2022) ↗—and relief is largest in gentrifying areas, where displacement fears run hottest“the negative impact of new high-rises on nearby residential buildings is significantly bigger in gentrifying neighborhoods, compared to established high-income neighborhoods”Do New Housing Units in Your Backyard Raise Your Rents? (2022) ↗.

The towers themselves are pricey—renting 60 percent above their surrounding neighborhoods“Rents for new high-rises are 60% higher than the average rents in their census tracts”Do New Housing Units in Your Backyard Raise Your Rents? (2022) ↗—yet mid-market units get cheaper, so opposing such development can worsen affordability“Opposing such development may exacerbate the housing affordability crisis and increase housing cost burdens for local renters.”Do New Housing Units in Your Backyard Raise Your Rents? (2022) ↗.

Pennington cracks the chicken-and-egg problem—does building lower rents, or do developers just build where rents are already softening?—by using serious fires as random shocks that make construction 32 times likelier on a parcel“During the study period, 27.22% of burned parcels receive new construction compared to 1.11% of unburned parcels. Controlling for microneighborhood and year, this is a 32-fold increase in the probability of construction (p=0.0000).”Does Building New Housing Cause Displacement? (2021) ↗. Where market-rate building follows, nearby rents fall $23 to $43 a month“I find that monthly rents fall by $22.77 - $43.18 relative to trend, roughly 1.2 - 2.3%, for people living within 500m of a new project.”Does Building New Housing Cause Displacement? (2021) ↗ and displacement risk drops 17 percent within 500 meters“an additional project reduces displacement risk by 17.14% for people living within 500m.”Does Building New Housing Cause Displacement? (2021) ↗; even evictions at nearby rent-controlled units fall 31 percent“I find that landlords of rent controlled units within 100m are 0.77 percentage points (31.09%) less likely to evict tenants after new housing is built, consistent with a reduction in the opportunity cost of rent-controlled leases.”Does Building New Housing Cause Displacement? (2021) ↗. Her sharpest line: displacement happens to people, gentrification to places“Displacement happens to people; gentrification happens to places.”Does Building New Housing Cause Displacement? (2021) ↗—richer arrivals do move in“parcels within 100m of market rate construction receive one additional richer arriver per new project, corresponding to a 29.5% increase in the probability of gentrification.”Does Building New Housing Cause Displacement? (2021) ↗, but sitting renters are likelier to stay.

Behind it all is scarcity: San Francisco cannot grow up or out“The combination of strict regulation and geography mean that San Francisco cannot grow up or out.”Does Building New Housing Cause Displacement? (2021) ↗, and over her window one-bedroom asking rents rose 97 percent“Over my study period from 2003-2017, the average price of a one-bedroom apartment listed on Craigslist increased 97%.”Does Building New Housing Cause Displacement? (2021) ↗.

Bratu, Harjunen and Saarimaa follow the chain of moves a single new building sets off, using Helsinki's population registers. Each 100 new, centrally located market-rate units set off chains that create vacancies down the ladder“That is, for each 100 new, centrally located market-rate units, 29 units get created through vacancy in bottom-quintile income zip codes and 60 units in bottom-half income zip codes.”JUE Insight: City-wide Effects of New Housing Supply: Evidence from Moving Chains (2023) ↗, and those chains reach even the extremes of the neighborhood distribution“These numbers suggest that even the extreme ends of the neighborhood distribution can, in principle, be connected through moving chains in just a few rounds.”JUE Insight: City-wide Effects of New Housing Supply: Evidence from Moving Chains (2023) ↗. About 29 of those moves reach low-income areas—roughly 60 when the new units are social housing“We find that for each 100 new, centrally located market-rate units, roughly 29 (60) units are created in the bottom-quintile (bottom half) of neighborhood income distribution through vacancies. … The corresponding figures for social housing units are 43 and 75, respectively.”JUE Insight: City-wide Effects of New Housing Supply: Evidence from Moving Chains (2023) ↗—within a year or two. Subsidized units loosen the middle and bottom fastest“Thus, social housing buildings loosen the middle- and low-income housing markets more directly, but this comes with considerable costs to taxpayers due to forgone rental income”JUE Insight: City-wide Effects of New Housing Supply: Evidence from Moving Chains (2023) ↗, with one caveat: the first movers into new social housing have higher incomes than typical“we do see that first-round movers to new centrally located social housing units are positively selected in terms of income and education relative to movers to other social housing units in the HMA, suggesting that these units may not be going to the individuals who need them the most.”JUE Insight: City-wide Effects of New Housing Supply: Evidence from Moving Chains (2023) ↗.

Greenaway-McGrevy supplies the cleanest real-world test of upzoning. In 2016 Auckland upzoned three-quarters of its residential land, and a synthetic-control comparison finds permits per capita more than doubled“Over the same period, permits per thousand residents more than doubled, increasing from 5.95 to 12.57.”Can Zoning Reform Increase Housing Construction? Evidence from Auckland (2024) ↗, with cumulative permits running far above the no-reform counterfactual“Six years on from the reform, cumulative permits issued exceed those of the synthetic control by approximately 43,500 – forty-five percent of the 97,000 permits issued in Auckland since 2016.”Can Zoning Reform Increase Housing Construction? Evidence from Auckland (2024) ↗. The jump is unlikely to be noise—Auckland's is the largest among 52 placebo cities“Auckland has the largest ratio among all 52 placebos. Thus, if one were to assign the intervention at random, the probability of obtaining a ratio as large as Auckland’s is 0.019 (= 1/52).”Can Zoning Reform Increase Housing Construction? Evidence from Auckland (2024) ↗—and rents came in about 28 percent below the counterfactual“the authors find rents in Auckland six years after the AUP are 28% lower than they would have been otherwise.”Dispelling Myths: Reviewing the Evidence on Zoning Reforms in Auckland (Donovan & Maltman, 2024), reporting Greenaway-McGrevy & So (2024) ↗, with roughly 64,836 dwellings added“In the period from 2018 to 2023, Auckland added 64,836 dwellings, which would represent almost 90% of consents.”Dispelling Myths: Reviewing the Evidence on Zoning Reforms in Auckland (Donovan & Maltman, 2024) ↗. The lesson is scale: localized 'spot' upzonings show muted effects“studies that focus on localized (or “spot”) upzonings typically show muted effects on housing supply (Dong, 2021; Peng, 2023), or no effect at all (Freemark, 2020)”Can Zoning Reform Increase Housing Construction? Evidence from Auckland (2024) ↗, but a city-wide reform moved the needle.

One puzzle sharpens the result. On paper, Auckland already had roughly 350,000 units of “zoned capacity” before 2016, so loosening the rules should have done little—like pushing on a string. Construction surged anyway. The resolution, as Michael Wiebe explains, is that most of that capacity was only theoretical: not economically feasible at prevailing prices“Much of it required redevelopment that was not economically feasible at prevailing prices and construction costs.”Michael Wiebe, “Upzoning,” Building Abundance (2026) ↗. The reform mattered because it created feasible capacity—projects developers could profitably build right away“Their upzoning worked because it created feasible capacity: projects that developers could profitably build right away.”Michael Wiebe, “Upzoning,” Building Abundance (2026) ↗.

Been, Ellen and O'Regan revisit their own skepticism and weigh the new evidence. Land-use rules matter, but they bind only where demand is strong“Most importantly, land use restrictions are a constraint on development only where there is strong demand for new housing; relaxing constraints will have little effect in soft markets.”Supply Skepticism Revisited (2024) ↗. On the central question the research now leans one way: large reforms work—Auckland's three-bedroom rents came in 26 to 33 percent below trend“rents for three-bedroom dwellings in Auckland were between 26 and 33% less than those of the synthetic control”Supply Skepticism Revisited (2024) ↗—and so does local building—new buildings cut nearby rents 5 to 7 percent“newly constructed buildings reduce the level of nearby rents by 5–7%, relative to the trend they otherwise would have followed”Supply Skepticism Revisited (2024) ↗ and reduce displacement risk about 17 percent“the risk of displacement falls by about 17% for households living within 100 meters of an additional new project”Supply Skepticism Revisited (2024) ↗, partly through moving chains that reach lower-income households“with every 100 new market-rate units constructed in the central city, approximately 66 units are made available in the bottom half of the neighborhood income distribution, 31 of them in the bottom quintile”Supply Skepticism Revisited (2024) ↗. Their verdict stays balanced: adding supply is necessary but not sufficient“Doing more to allow developers to build new housing in high-demand areas is a necessary but not sufficient strategy. If we don’t allow such new homes to be created, inmovers will simply bid up the price of existing homes…”Response to Comments on "Supply Skepticism Revisited" (2024) ↗.

Schuetz's frame is that America has two housing problems, and conflating them yields bad policy. One is supply: the country is three and a half to four million homes short“Estimates are anywhere from three and a half to four million homes short across the entire country, which is a lot of homes.”Tackling America's Housing Challenges (NMHC Code 53 podcast) ↗, largely because overly restrictive zoning makes building hard where demand is highest“overly restrictive zoning makes it hard for developers to build new housing, driving up rents and prices”Is Zoning a Useful Tool or a Regulatory Barrier? (Brookings, 2019) ↗. The other is income—and here supply alone falls short: building more homes—even smaller ones—helps over time, but market-rate housing cannot reach the poorest“Building more housing, especially smaller housing, will over time bring down housing costs (or at least keep them from rising as quickly). But expanding the supply of market-rate housing is not enough to help the poorest families.”To Improve Housing Affordability, We Need Better Alignment of Zoning, Taxes, and Subsidies (Brookings, 2020) ↗. The tools only work together: deregulation without subsidy leaves the worst-off behind; subsidy without deregulation bids up rents.

Gyourko, Mayer and Sinai explain why a handful of cities keep pulling away. In 'superstar' metros, real prices rose 458 percent from 1950 to 2000“implies a 458 percent increase in real house prices between 1950 and 2000, more than twice as large as seventh-ranked Boston at 212 percent, which itself still grew 50 percent more than the sample average of 132 percent for the 50 most populous metropolitan areas.”Superstar Cities (2006/2013) ↗, opening a vast gap between San Francisco and the cheapest metros“the gap in prices between the most expensive metropolitan area (San Francisco) and the average across all areas doubled between 1970 and 2000”Superstar Cities (2006/2013) ↗. The driver is national, not local: as the number of high-income families roughly doubled“Over the last 50 years, the number of families in U.S. metropolitan areas doubled, with the number making more than $140,000 in real ($2000) terms increasing more than eight-fold”Superstar Cities (2006/2013) ↗, they compete for a fixed stock of desirable places, and existing residents are outbid“As the number of high income families grows nationally, existing residents are outbid by even higher-income families, raising the price of land yet further.”Superstar Cities (2006/2013) ↗.

The result is sorting—superstar metros hold a larger share of high-income families“MSAs that ever were superstars have a 2.5 percentage point greater share of their families that are in the rich category”Superstar Cities (2006/2013) ↗ and a markedly smaller share of the poor“Markets that have ever been superstars also have a nearly 9 percentage point lower share of poor families (row 1, column 5), almost 21 percent less than the other MSAs.”Superstar Cities (2006/2013) ↗—and prices that price in expected growth: price-to-rent ratios run 22 percent higher“Relative to the intermediate B markets, the price/rent ratio is 22 percent (with a standard error of 1.4 percent) higher for MSAs that ever were superstars”Superstar Cities (2006/2013) ↗. The mechanism, they note, does not depend on why those places are scarce“our results do not depend upon the reasons for location preferences or inelasticity of supply, so we neither attempt to discern why some families prefer certain cities nor address the causes of limited supply.”Superstar Cities (2006/2013) ↗, only that they are.

Diamond explains the divergence through who moved where. From 1980, college graduates clustered in high-wage cities and bid up housing—a 1 percent rise in a city's college share lifts rents 0.70 percent“a 1 percent increase in the local college employment ratio is associated with a 0.70 percent increase in local rents… Variation in cities’ college employment ratio changes can explain 49 percent of the variation of rent changes across cities.”The Determinants and Welfare Implications of US Workers' Diverging Location Choices by Skill: 1980-2000 (AER 2016) ↗—until the skilled were paying so much it ate into their wage gains“College workers continued to migrate in and bid up housing prices so high that they received lower real wages.”The Determinants and Welfare Implications of US Workers' Diverging Location Choices by Skill: 1980-2000 (AER 2016) ↗. Local labor demand led, but the cities then improved as the rich arrived, and those endogenous amenities widened the welfare gap by more than wages alone“the welfare effects of changes in local wages, rents, and endogenous amenities led to an increase in well-being inequality between college and high school graduates which was significantly larger than would be suggested by the increase in the college wage gap alone.”The Determinants and Welfare Implications of US Workers' Diverging Location Choices by Skill: 1980-2000 (AER 2016) ↗. Supply determined how hard the squeeze landed: housing is less elastic where land-use regulation is tighter“I find housing supply is less elastic in areas with higher levels of land-use regulation and less land near a city’s center available for real estate development.”The Determinants and Welfare Implications of US Workers' Diverging Location Choices by Skill: 1980-2000 (AER 2016) ↗, with an inverse supply elasticity of 0.21“inverse housing supply elasticity is 0.21, with a standard deviation of 0.22. A regression of my inverse housing supply elasticity estimates on Saiz’s (2010) estimates yields a coefficient of 0.86 (0.14), suggesting we find similar amounts of variation in housing supply elasticities across cities. However, Saiz’s (2010) inverse housing supply estimates are higher than mine by 0.26, on average.”The Determinants and Welfare Implications of US Workers' Diverging Location Choices by Skill: 1980-2000 (AER 2016) ↗—implying tighter supply than Saiz's estimates.

Howard and Liebersohn ask why rents rose—17.4 log points from 2000 to 2018“Real rents measured in the United States CPI increased 17.4 log-points from 2000-2018.”Why Is the Rent So Darn High? (JUE 2021) ↗—and answer: demand to live in the wrong places. Growing demand to live in supply-inelastic cities explains much of the increase“the location demand channel explains 54 percent of the rent increase in all U.S. cities and 75 percent of the rent increase where the CPI rents data exists”Why Is the Rent So Darn High? (JUE 2021) ↗. The clue is mobility: when people can move freely, a local supply shock barely moves local rent“When people are fully mobile, the effect is close to zero and actually slightly negative. Overall, however, housing supply shocks play very little role in explaining housing prices.”Why Is the Rent So Darn High? (JUE 2021) ↗, because newcomers move in until rents settle back. Ruling out the usual suspects—it is not bigger houses or tighter restrictions“If people demanded larger houses or if housing became more restricted, we would expect a population decrease in cities where it is harder to build… But the data shows the opposite”Why Is the Rent So Darn High? (JUE 2021) ↗—they estimate a near-infinite population elasticity“our preferred estimate of the eighteen-year population elasticity to rents is infinity, meaning that rents will fully offset any location demand shocks”Why Is the Rent So Darn High? (JUE 2021) ↗, which implies local building has negligible effects on local rents“local expansions of housing supply will have negligible effects on local rents in the long-run”Why Is the Rent So Darn High? (JUE 2021) ↗.

The policy reading is counterintuitive: a demand subsidy is several times more potent than a supply subsidy in the least-elastic cities“a 10 percent subsidy in the 10 percent least elastic cities lower aggregate rents by 0.26 log-points, whereas a 10 percent subsidy in the 10 percent most elastic parts could reduce rents by 1.63 log-points, more than six times as large an effect.”Why Is the Rent So Darn High? (JUE 2021) ↗.

Molloy, Nathanson and Paciorek argue the debate measures the wrong thing. Prices capitalize expected future rents, so the rise in rents is smaller than the rise in prices“the increase in rent is smaller than the increase in the purchase price of homes because supply constraints increase expected growth in future rent as well as the current level of rent.”Housing Supply and Affordability (JUE 2022) ↗; in their calibration the rent effect is about half the price effect“In a calibration exercise, we find that the effects on rent are about half of the effects on the purchase price of housing.”Housing Supply and Affordability (JUE 2022) ↗. Measured that way, regulation does less than advertised: a metro two standard deviations more regulated saw only a small extra rent rise“a metro area with regulation 2 standard deviations tighter than average experienced only 0.07 log point larger rent increases from 1980 to 2016, which works out to less than ¼ percentage point faster growth per year”Housing Supply and Affordability (JUE 2022) ↗, against a far larger average increase“By contrast, the average increase in (real) rent among all metros in this sample from 1980 to 2016 was 0.49 log points—7 times larger.”Housing Supply and Affordability (JUE 2022) ↗, so regulation explains less than a sixth of it“our estimated coefficients suggest that regulation can explain less than one sixth of the price increases…”Housing Supply and Affordability (JUE 2022) ↗.

Even where prices nearly tripled in New York, Boston and San Francisco“House prices nearly tripled in real terms in New York, Boston and San Francisco over this period, and our estimated coefficients suggest that regulation can explain less than one sixth of the price increases in these highly‐regulated regulated metros.”Housing Supply and Affordability (JUE 2022) ↗, the welfare-relevant rent distortion is modest—though they date the tightening to a real shift: land-use court cases began climbing in 1960“The incidence of court cases related to land use began increasing in 1960, illustrating that some regulations were binding in some locations prior to 1980.”Housing Supply and Affordability (JUE 2022) ↗.

Louie, Mondragon and Wieland attack the standard view at its root. If supply flexibility shaped how cities absorb demand, high- and low-elasticity places should diverge—but higher income growth predicts the same growth in prices, quantities and population regardless of measured elasticity“higher income growth predicts the same growth in house prices … regardless of the estimated housing supply elasticity”Supply Constraints do not Explain House Price and Quantity Growth Across U.S. Cities (2025) ↗. It holds across four different elasticity measures“These are the foundational supply elasticity from Saiz (2010), a supply elasticity from Baum-Snow and Han (2024), the Wharton Residential Land Use Regulation Index (WRLURI) from Gyourko, Saiz and Summers (2008), and the land share of value from Davis, Larson, Oliner and Shui (2021)”Supply Constraints do not Explain House Price and Quantity Growth Across U.S. Cities (2025) ↗, where the estimates are statistically indistinguishable“are all around one and, critically, are statistically and economically indistinguishable across cities measured to be relatively more or less constrained”Supply Constraints do not Explain House Price and Quantity Growth Across U.S. Cities (2025) ↗, and even for the work-from-home shock: prices rose essentially the same regardless of a city's elasticity“the increase in house prices was essentially identical regardless of whether the city was more or less constrained”Supply Constraints do not Explain House Price and Quantity Growth Across U.S. Cities (2025) ↗.

So demand, not supply, did the work—San Francisco's prices grew about 2.4 percent a year, far faster than Houston's“From 2000 to 2020, real house prices in San Francisco grew annually by about 2.4%, compared to 1% for Houston.”Frequently Asked Questions About and Comments On “Supply Constraints…” (2025) ↗, a gap explained by income growth“the difference between San Francisco’s and Houston’s house price growth was about what one would expect given the differential income growth in these two areas”Frequently Asked Questions About and Comments On “Supply Constraints…” (2025) ↗. The implication is deflating: relaxing San Francisco's elasticity to Houston's would cut prices only about 4 percent“relaxing San Francisco’s elasticity to Houston’s level and assuming average growth in total incomes would reduce house prices in San Francisco by about 4%”Supply Constraints do not Explain House Price and Quantity Growth Across U.S. Cities (2025) ↗, because higher incomes buy housing quality, not just more units“increases in per capita income will tend to increase the demand for housing quality with little effect on the demand for units, while increases in population (driven by growth in employment, for example) will tend to increase the demand for housing units”Supply Constraints do not Explain House Price and Quantity Growth Across U.S. Cities (2025) ↗.

Orlando and Redfearn ask how big cities actually add housing as they grow. Mapping new units across metros from 1970 to 2010, they find growth itself stiffens supply: as close-in greenfields run out, building shifts to costlier infill, since a high-rise costs far more per square foot than low-density construction“the cost per square foot of a high-rise building (8+ stories) is significantly higher than anything shorter. … In high-density neighborhoods, developers face up to 40% higher costs due to land assembly and holdouts”Houston, You Have a Problem (2024) ↗. So supply elasticity falls as a city grows“The empirical results point to falling elasticity as a function of growth, suggesting that rising prices are an inherent feature of large and growing metropolitan areas.”Houston, You Have a Problem (2024) ↗, even without new rules—lightly-zoned Harris County's elasticity dropped over the decades“the supply elasticity declines in Harris County, the central county of the Houston MSA, from 0.32 in 1980-1994, to 0.25 in 1990-2004, to 0.15 in 2000-2016.”Houston, You Have a Problem (2024) ↗—and building moved inward as the periphery stopped substituting for the core“From 1980 to 1990, the distance coefficient is 96.96; from 2010 to 2015, it is -16.89.”Houston, You Have a Problem (2024) ↗.

The tell is timing: California was already heavily urbanized by 1970“23.5% of California tracts are classified as urban as early as 1970 (in the Share column of Table 2), while only 2.1% of Texas tracts begin in the urban category”Houston, You Have a Problem (2024) ↗, and Sacramento—under the same state rules—now behaves like a Texas metro“In Panel (a) of Figure 14, it behaves more like the MSAs of Texas than the other three in California. Its density distribution appears to be governed more by size and density—and especially the availability of proximal developable land—than any “anti-development” attitude that may be unique to California.”Houston, You Have a Problem (2024) ↗, so regulation looks endogenous, tightening as demand presses“In other words, land-use regulation is endogenous, becoming more stringent as demand increases.”Houston, You Have a Problem (2024) ↗—with rising prices an inherent feature of large, growing cities.

Rollet models how cities rebuild themselves—demolishing the old for the new—using a parcel-level panel of New York. Zoning binds, but relaxing it is no quick fix: upzoning does raise construction, yet the effects materialize only slowly“While upzoning does increase construction, its effects materialize slowly over time: ten years after the policy change, only 9% of the newly allowed floorspace has been built.”Zoning and the Dynamics of Urban Redevelopment (2025) ↗, with the floorspace supply elasticity reaching just 0.13 even at a 40-year horizon“with an aggregate floorspace supply elasticity reaching only 0.13 at a 40-year horizon”Zoning and the Dynamics of Urban Redevelopment (2025) ↗. Fully removing zoning yields far more floorspace (+58 percentage points), but still over decades“Fully removing zoning regulations yields larger increases in floorspace (+58 pp), but even in this extreme scenario, residential rents in NYC decrease only moderately (-17 pp).”Zoning and the Dynamics of Urban Redevelopment (2025) ↗, and because people migrate in, much of the local affordability benefit leaks to the wider region“Furthermore, due to migration, the affordability benefits of zoning reform largely accrue to households outside the rezoned neighborhoods.”Zoning and the Dynamics of Urban Redevelopment (2025) ↗. His verdict is qualified, not dismissive: upzoning still outperforms the main alternative policies“Furthermore, upzoning substantially outperforms two alternative policies aimed at increasing construction and affordability: reducing construction costs and granting property tax breaks on new buildings.”Zoning and the Dynamics of Urban Redevelopment (2025) ↗.

Davidoff makes two arguments against the pure regulatory-tax story. With Aura, he grants the supply side its premise—treat construction cost as flat and land as the fixed, scarce input“We assume the consumer price of a square foot of structure is a constant, but land is in fixed supply.”Supply Constraints and Housing Prices (2008) ↗—and still finds building up does little: even at 100 stories, Manhattan prices would fall under 15 percent“even if every building in Manhattan grew to 100 stories tall, prices would fall by less than 15 percent (ignoring induced amenity losses and wage changes)”Supply Constraints and Housing Prices (2008) ↗, because halving land prices would take fifteen times as much available land“halving land prices would require a 15-fold increase in available land”Supply Constraints and Housing Prices (2008) ↗. So a single jurisdiction loosening its rules barely moves prices“We conclude that individual jurisdictions are unlikely to increase “affordability” by encouraging more supply and recent housing price increases most plausibly reflect demand changes.”Supply Constraints and Housing Prices (2008) ↗, though coordinated loosening across many markets could do far more“Metropolitan or federal coordination of loosened regulation might result in considerably stronger price reductions”Supply Constraints and Housing Prices (2008) ↗; their calibration spans roughly 30 percent of the US population“In Table 1, we find that with approximately 30% of US population, the price elasticity with respect to land supply in the meta coastal region would be approximately .5 if the median American would move into the coastal region if the land price fell forty percent.”Supply Constraints and Housing Prices (2008) ↗.

In later work he questions the standard approach directly. Economists treat tight regulation as exogenous cost, but supply has actually grown more in constrained markets, not less“Housing supply has grown more in supply-constrained markets than elsewhere over recent decades, indicating constraints are correlated with demand growth.”Supply Constraints Are Not Valid Instrumental Variables for Home Prices (2016) ↗, because those constraints correlate with productivity, immigration and education“Table 2 shows that each measure of supply constraints (high regulations and unavailable land, low nontraditional Protestant share and elasticity) is significantly positively correlated with each proxy for demand levels and growth (college, immigrant, Bartik, Sand, and Coastal).”Supply Constraints Are Not Valid Instrumental Variables for Home Prices (2016) ↗—so the demand curve estimated this way slopes upward—the wrong way“Based on the quantity regression results, IV formula (2) would indicate that demand is upward sloping in price, with or without the demand controls.”Supply Constraints Are Not Valid Instrumental Variables for Home Prices (2016) ↗—and across metros, constraints explain only a small fraction of price differences“supply constraints can explain only a small fraction of differences in price levels across U.S. metropolitan areas”Supply Constraints Are Not Valid Instrumental Variables for Home Prices (2016) ↗. Better, he argues, to use amenities like weather, plausibly uncorrelated with construction cost“Amenities such as the quality of weather or ocean views seem plausibly uncorrelated with the growth of productivity, and subject to capture with metropolitan area fixed effects in panel settings. However, highly skilled workers appear to sort into naturally and culturally attractive cities, per Gyourko et al. (2004).”Supply Constraints Are Not Valid Instrumental Variables for Home Prices (2016) ↗.

Freemark studies Chicago upzonings that were, unusually, decided independently of any development plans—a clean test. The short-run effect is the opposite of the YIMBY hope: upzoned parcels saw values rise 15 to 23 percent“I find consistently significant positive impacts of the reform of between 15% and 23.3%. This is of a similar order of magnitude as the increase in allowed density provided by the ordinance.”Upzoning Chicago: Impacts of a Zoning Reform on Property Values and Housing Construction (2020) ↗ and existing condos gained around 12 percent“models 5 to 6 (with and without covariates and district time trends) identifies a 12.2% to 13.2% increase in property transaction values attributable to the upzoning”Upzoning Chicago: Impacts of a Zoning Reform on Property Values and Housing Construction (2020) ↗, signs of land speculation, while no additional housing was permitted within five years“there is little reason to believe that either of the zoning reforms produced a significant increase in housing permits in the time period analyzed”Upzoning Chicago: Impacts of a Zoning Reform on Property Values and Housing Construction (2020) ↗. That null holds even though those areas covered nearly two-thirds of the upzoned land“despite the fact that these areas encompassed a considerable portion of the upzoned area—64% of the density-class land and 42% of the parking-class land in 2013, according to Table 2.”Upzoning Chicago: Impacts of a Zoning Reform on Property Values and Housing Construction (2020) ↗. His caution to planners: local upzoning alone, in the short run, lifts prices without producing units“For municipal planners hoping to encourage new construction, these results suggest they will have to wait until the cranes arrive. For those hoping to address affordability, they may need to look for other solutions.”Upzoning Chicago: Impacts of a Zoning Reform on Property Values and Housing Construction (2020) ↗.

Rodríguez-Pose and Storper reject the 'housing as opportunity' school—the idea that freeing supply in booming cities would cut inequality. The driver, they argue, is the economy itself: in places that grew on high-skill, high-wage work, the affluent outbid everyone for scarce space“In areas that have grown principally due to the growth of jobs with routine skills and moderate wages, housing prices are mechanically lower than in metropolitan areas that draw in the highly-skilled and highly-paid.”Housing, Urban Growth and Inequalities (Urban Studies, 2020) ↗. Cheap metros are not cheap for the poor—Houston’s rents have still risen 326 percent—in line with its income growth“While mean rent in Houston has risen 326 percent, far less than in San Francisco, this increase is closely aligned with the increase in mean income (389 percent).”Inequality, Not Regulation, Drives America's Housing Affordability Crisis (LSE III WP 159, 2026) ↗—and deregulation misreads the problem: blanket zoning changes are unlikely to raise affordability for lower-income households in prosperous regions“Blanket changes in zoning are unlikely to increase domestic migration or to increase affordability for lower-income households in prosperous regions.”Housing, Urban Growth and Inequalities (Urban Studies, 2020) ↗, since upzoning would mostly replace older, cheaper stock“it would very likely involve replacing older and lower-quality housing stock in areas highly favoured by the market, effectively decreasing housing supply for lower income households in desirable areas. This is gentrification.”Housing, Urban Growth and Inequalities (Urban Studies, 2020) ↗ and fuel gentrification.

Trounstine reframes zoning as a tool of exclusion. Because American public goods are funded locally, communities use land-use rules to protect property values and keep others out—and the result is segregation. Cities whiter than their metro in 1970 adopted stricter rules by 2006“The table reveals that cities that were whiter than the metropolitan area in 1970 had significantly more restrictive land use regimes in 2006.”The Geography of Inequality (APSR, 2020) ↗—a pattern wealth only partly explains: whiteness and homeownership are tightly correlated“Because homeownership and wealth in 1970 are highly correlated with whiteness, variation in these variables accounts for about half of the direct effect of whiteness on land use restrictiveness.”The Geography of Inequality (APSR, 2020) ↗. The law has pushed back—199 of these cities faced Fair Housing Act suits“Of these, 199 cities were engaged in a Fair Housing Act lawsuit during the timespan.”The Geography of Inequality (APSR, 2020) ↗—and when courts forced liberalization, the white population share fell“Table 2 offers clear evidence that when cities are threatened or forced by the court to liberalize their land use laws they see growth in their population of people of color.”The Geography of Inequality (APSR, 2020) ↗. But the politics persist: California voters backed development restrictions in 2016“In the 2016 election, California voters overwhelming supported development restriction. Pro-growth initiatives garnered an average of 42% of the vote, while anti-growth initiatives garnered better than 60%.”The Geography of Inequality (APSR, 2020) ↗, and the friction shows in the details, like Glendora's two-year review against Pomona's six months“The average time to review residential development in Glendora is two years, compared to about 6 months for both Pomona and Carson City.”The Geography of Inequality (APSR, 2020) ↗. She accepts the supply problem—the affordability crisis is real“Many metropolitan areas in the United States are facing a crisis of housing affordability. Homelessness is on the rise as rents and housing prices skyrocket. The problem is largely the result of limited growth and development.”The Geography of Inequality (APSR, 2020) ↗—while locating its cause in whom zoning is built to exclude.

Condon's claim is blunt: the land under the building is what costs too much“It’s the land under the building that costs too much.”Justice, Land and the City — Patrick Condon (UBC) ↗. A single parcel makes the point—land alone runs $1.675m; land plus a three-unit building, $1.995m“$1,675,000 just land vs. $1,995,000 land and 3 unit building”Justice, Land and the City — Patrick Condon (UBC) ↗. Vancouver tested the supply cure, and the cure failed: the city doubled its density and tripled its housing units“Doubled population, doubled density and tripled housing units city wide”Justice, Land and the City — Patrick Condon (UBC) ↗, yet it now takes 29 years to save a down payment“It now takes 29 years to save up for down payment”Justice, Land and the City — Patrick Condon (UBC) ↗ and interior prices hit $1,000 a square foot“Per sq. ft. interior price $1,000”Justice, Land and the City — Patrick Condon (UBC) ↗. Density alone, he argues, does not fix affordability“Unaffordable in either case. Adding density does not fix it.”Justice, Land and the City — Patrick Condon (UBC) ↗, because new density inflates land value.

His remedy targets the land: capture some 80 percent of the value that rezoning creates“Capture 80 percent of rezoned land value”Justice, Land and the City — Patrick Condon (UBC) ↗ and tie added density to permanently non-market housing, as Cambridge does“Cambridge Mass Affordable Housing Overlay. 100% non-market housing in return for additional density. Rents permanently pegged to median incomes.”Justice, Land and the City — Patrick Condon (UBC) ↗. His warning to YIMBYs: do not just upzone and hope for affordability“Don’t just upzone hoping for affordability”Justice, Land and the City — Patrick Condon (UBC) ↗.

Murray's case is that zoning capacity is the wrong place to look; the action is in developer behavior—even profitable projects are delayed, because it pays to wait“Not all profitable projects are developed immediately, as it can be more profitable to delay development until later.”Supply Skepticism Needs Economic Theory (Housing Policy Debate, 2025) ↗. Across two decades of Brisbane upzoning—a panel of 25,775 sites“We use a panel of 25,775 sites in 19 major activity centres earmarked in the planning system for densification in Queensland, Australia.”We Zoned for Density and Got Higher House Prices (Urban Policy and Research, 2023) ↗—dwellings rose from 36,000 to 47,000“Total dwellings in our data went from 36,000 in 1996 to 47,000 in 2016, while our conservative zoned capacity estimate went from an estimated 33,000 dwellings to 66,000 over that period.”We Zoned for Density and Got Higher House Prices (Urban Policy and Research, 2023) ↗ while only 2 percent of zoned capacity was used in any five-year window“Only 2% of the zoned capacity was taken up in any five-year study interval.”We Zoned for Density and Got Higher House Prices (Urban Policy and Research, 2023) ↗ and 71 percent of sites were never redeveloped“Of the 5,940 sites with zoned capacity in 1996, 4,486 (or 71%) had no residential redevelopment in the subsequent 20 year period.”We Zoned for Density and Got Higher House Prices (Urban Policy and Research, 2023) ↗. Far from lowering prices, more supply went hand in hand with higher prices“Higher rates of new housing supply are robustly related to higher prices. Our zoned capacity variable has no relationship to price across numerous regression models and is robust to various data selection choices.”We Zoned for Density and Got Higher House Prices (Urban Policy and Research, 2023) ↗, because supply follows demand.

Behind that empirical reading is a formal model. A landowner sets the pace of building to maximise the present value of their balance sheet“Landowners maximise the present value of their balance sheet by choosing a rate of new housing lot sales”A Housing Supply Absorption Rate Equation (2022) ↗, so the rate of new supply tracks demand growth rather than how much zoning permits. Two results run against the standard story. Because the optimal building rate rises with demand, demand growth limits the ability of market supply to reduce prices“the relationship between demand growth and the optimal supply rate limits the ability for market supply to reduce prices”A Housing Supply Absorption Rate Equation (2022) ↗; and density limits can actually speed building“Constraints on density, for example, increase the optimal rate of supply by reducing the return to delaying development”A Housing Supply Absorption Rate Equation (2022) ↗, by cutting the payoff to holding land. The lever that does raise the building rate, in this model, is raising the cost of delay—interest rates and land taxes, not looser zoning“increasing the cost to delaying housing development is the primary way to increase the market rate of housing supply”A Housing Supply Absorption Rate Equation (2022) ↗.

That reading reshapes the rest of the literature for him: the cleanest 'supply' studies actually measure demand“They tell us about demand for housing, not supply. We get an illustration of the slope of the demand curve but learn nothing about the supply curve and the underlying incentives for the production of housing.”Supply Skepticism Needs Economic Theory (Housing Policy Debate, 2025) ↗, and Auckland's headline result depends on excluding part of the data“Excluding this data effectively converts a growth trend for Auckland indistinguishable from that in Wellington, which was not upzoned, into a series with a distinct structural break.”The Auckland Myth (Fresh Economic Thinking, 2023) ↗. Brisbane's prices rose anyway“Brisbane dwelling prices increased by more than Sydney prices from 2001-2016, and over the whole period increased by more than prices in San Francisco (250% versus 235%); both cities that are routinely argued to be supply-constrained”We Zoned for Density and Got Higher House Prices (Urban Policy and Research, 2023) ↗—evidence, he concludes, that private markets will not rapidly supply enough housing to cut prices“This evidence suggests that private housing markets will not rapidly supply new housing and cause significant price reductions, even if the planning system allows it.”We Zoned for Density and Got Higher House Prices (Urban Policy and Research, 2023) ↗.