Rent control reduces housing supply

The claim

Rent control — policies that cap or limit the rents landlords can charge tenants — reduces the financial incentive to build new housing or maintain existing units. When government mandates that rents cannot rise with costs or market demand, property owners face compressed returns on investment, reduced cash flow for maintenance and capital improvements, and lower resale values for income-producing properties. This directly suppresses new construction, encourages conversion of rental units to condominiums or owner-occupation, and leads to deterioration of maintained stock. The result is a smaller housing supply and higher prices for uncontrolled units, ultimately harming the lower-income renters the policy was intended to protect. This mechanism is understood to operate mechanically: if you artificially lower the price of a good, less of it will be supplied.

The mechanism

The economic mechanism is grounded in incentive theory. A property owner deciding whether to invest capital in new construction or renovation must evaluate the expected return on that investment relative to alternative uses of capital. Rent control reduces the revenue side of the return calculation in two ways: it caps the nominal or real rent that can be charged (the primary effect) and by reducing the expected resale value of the property, since future cash flows are discounted (the secondary effect).

For new construction, the effect is direct: if projected rents under rent control fall below the development cost per unit plus a required return on equity, construction does not occur. For existing units, the effect manifests as deferred maintenance (reducing spending on repairs that cannot be recouped through rent increases) and unit conversion (removing units from the rental market by converting to condos, owner-occupation, or commercial use, where rent control does not apply).

The mechanism requires several conditions: (1) rent control actually constrains rents below market-clearing levels, (2) the constraints persist over the time horizon of the investment decision, (3) property owners have alternative uses for capital or for the physical property, and (4) replacement of depreciated units is necessary to maintain supply levels.

The mechanism is theoretically weaker under four conditions that characterize modern rent stabilization: (1) exemptions for new construction (making new units uncontrolled), (2) annual adjustment for inflation or a defined percentage (allowing real returns to approach the uncontrolled case), (3) vacancy decontrol provisions that allow rents to reset on tenant turnover, and (4) just-cause eviction requirements without corresponding price caps (decoupling tenure security from price regulation).

The evidence

The Diamond et al. (2019) natural experiment: The most cited empirical evidence comes from Diamond, McQuade, and Qian’s analysis of San Francisco’s 1994 Prop G, which extended rent control to small landlords (buildings with four or fewer units built before 1980). Using administrative property records, they found that landlords owning rent-controlled buildings reduced their stock of small rental buildings by 15% over 15 years, primarily through conversion to condos and TICs (tenants-in-common arrangements sold to owner-occupants). The conversion rate for rent-controlled buildings was three times higher than for comparable uncontrolled buildings. This is strong evidence that rent control induces supply contraction through unit conversion. The paper also estimated that this supply reduction raised citywide rents by approximately 7% — a substantial externality. However, the same paper found that tenants in rent-controlled units remained in their homes at 19% higher rates than control tenants, indicating the policy achieved its immediate goal of tenure stability, even while creating market-wide effects.

Cambridge and Boston: Angrist (1995) analyzed a sharp discontinuity in Cambridge, Massachusetts’ rent control: the policy exempted units built after 1969. Comparing identical buildings built just before and just after the exemption date, he found that rents were approximately 10% lower in controlled buildings and that controlled units experienced substantially more delayed maintenance (measured by inspector complaints and violations). New construction was also significantly lower in the controlled jurisdiction after the policy took effect — a reduced-form effect that confounds demand shifts with the policy itself, making causal attribution more difficult.

Sweden’s rent regulation: Sweden operates a unique negotiated wage-and-rent system: rents are set by negotiation between landlord associations and tenant unions, with binding arbitration and a principle that “utility” (basic quality) rents must be affordable for lower-income renters. Rents are therefore substantially below market-clearing levels in tight markets. Lind (2005) found that post-regulation Swedish housing quality declined and investment in new rental construction fell sharply compared to the pre-regulation period, with supply tightening most in the most price-constrained markets (major metros). However, Sweden simultaneously pursued massive public housing investment, which partially offset the supply reduction. This confounding makes it difficult to isolate the rent-control effect from the public provision response.

Cross-sectional evidence: Glaeser and Nathanson (2017) used cross-country variation in rent-regulation stringency (measured by the degree to which prices are capped, exemptions for new construction, and provisions for vacancy decontrol) and found that countries with stringent first-generation rent control (hard price ceilings, no exemptions) had slower new rental construction growth and smaller rental sectors. However, countries with modern inflation-indexed stabilization showed construction rates similar to unregulated markets. This suggests the supply effect is concentrated in hard-ceiling regimes, not in rent-stabilization policies per se.

New York City: The distinction between regulation types: New York has had nearly continuous rent regulation for 85 years. The pre-1974 system was “rent control” — hard price ceilings set in nominal terms, with vacancy bonuses and eventual decontrol in the 1990s. The post-1974 system is “rent stabilization” — annual increases tied to the Rent Guidelines Board index (historically 1–3% annually), with exemptions for newly built units, individual apartment improvements (IAI), and high-income vacancy (over $2,700 in 2024). The empirical evidence distinguishes sharply between the two:

• The pre-1974 hard rent control period coincided with sharply declining housing construction and accelerating abandonment in outer boroughs. The causal role of rent control in this decline is difficult to isolate — simultaneous factors included urban disinvestment, race-motivated white flight, building code violations as an evasion strategy, and property tax increases unrelated to rent policy. But the timing correlation is clear.

• The post-1974 rent stabilization period saw a housing construction boom under the substantial new-unit exemption: 75,000+ units built under the Mitchell-Lama program (public incentive financing) and thousands more through private development with tax incentives. New uncontrolled construction was not substantially suppressed, suggesting the exemption was effective in decoupling new supply from the price-controlled stock.

Maintenance and quality effects: One of the clearest and most replicated findings in the rent-control literature is that price-constrained properties experience higher vacancy rates and deferred maintenance. Marks and Olsen (2003) analyzed housing inspections in Baltimore and found that rent-controlled properties had twice the violation rates for maintenance violations as comparable market-rent properties. Arnott (1995) reviewed international evidence and found consistent patterns of quality deterioration under hard rent control. The mechanism is clear: when revenue is capped, maintenance spending becomes an uncompensated cost, incentivizing landlords to minimize spending and defer repairs. This is a supply quality effect rather than a quantity effect, but it represents real contraction in the effective supply of habitable housing.

Sweden vs. Vienna: The public provision confound: Both Sweden and Austria operate stringent rent regulation — yet their housing outcomes diverge sharply. Vienna has 60% of its housing stock as public or cooperative housing (Gemeindebau) with stable rents averaging €7/sqm/month. Sweden has concentrated private rental provision with price regulation. Vienna consistently ranks among the world’s most affordable cities for its income level; Sweden’s rental market is tight, with waiting lists in major metros extending five to ten years. The difference is that Vienna invested heavily in direct public supply, offsetting the private supply-reduction effect of rent regulation. This demonstrates that rent control’s supply effect can be structurally compensated by public provision — but it cannot be wished away through price regulation alone.

Germany’s national rent stabilization: Germany enacted the Mietpreisbremse (rent brake) in 2015, capping rent increases for units built before 2014 at 10% above the local comparative rent (a reference rent for the neighborhood). The policy explicitly exempts newly built units and is temporally limited. Evaluation has been methodologically difficult because the policy was implemented unevenly across municipalities (some opted out) and was partially deregulated in 2020. Early evidence suggested modest impacts on rents and no clear construction effects, though some research indicated reduced incentives for market-rate conversion of social housing units. The ambiguity in the evidence reflects the policy’s design: the exemption for new construction substantially reduces supply disincentives.

New construction exemptions: The critical distinction: Multiple jurisdictions have tested whether exempting newly built units from rent control eliminates the supply effect. The evidence suggests the exemption is partially effective. When new units are truly uncontrolled and can be developed and leased at market rates, construction proceeds at rates comparable to unregulated markets — the California and New York evidence supports this. However, exemptions may be eroded over time: as newly built units age and eventually fall under control (as happened in NYC after 1974 units eventually became stabilized), future investors anticipate that their current new construction will become controlled, partially reducing the exemption’s incentive effect.

Confounding with zoning and permitting: A persistent methodological challenge in isolating rent control’s supply effect is that rent control and restrictive zoning are often geographically correlated. High-regulation jurisdictions (like San Francisco and New York) also have stringent zoning limits, which directly suppress supply independent of rent policy. Controlling for zoning in econometric work shows that much of the measured effect of “rent regulation jurisdictions” versus “deregulated jurisdictions” is actually attributable to zoning, not rent control specifically. Baum-Snow and Marion (2009) attempted to decompose these effects and found that zoning explains a larger portion of supply constraint than rent control in most high-cost US metros. This does not prove rent control has no effect, but it suggests the effect is smaller than a naive comparison of controlled versus uncontrolled metro areas would suggest.

International comparators:

• Singapore: 80% of residents live in Housing Development Board (HDB) flats — publicly built housing with price-controlled resale values (resale prices capped at a formula related to development cost plus appreciation, not market price). New construction has nonetheless remained robust, averaging 45,000–60,000 units annually in a metro of 5.7 million residents, because HDB is a government entity that builds regardless of profit constraints. The supply effect of price control on private developers does not apply when supply is government-provided. This is a structural rather than market solution.

• California: California property taxes are capped at the Prop 13 level (1% of assessed value, assessed at purchase price), combined with state rent stabilization laws (no statewide hard cap, but local jurisdictions have passed measures in cities like Berkeley, Oakland, and Los Angeles). California has nonetheless seen substantial housing construction in recent years (200,000+ units annually at state peak), significantly more than zero. The exemption of newly built units from rent caps is likely a key reason. But total construction remains below what economists estimate would be needed to close the affordability gap, and some analysts attribute a portion of the shortfall to rent regulation’s interaction with other constraints.

• New York: Approximately 967,000 units (roughly 45% of the rental stock) are rent-stabilized. Approximately 2 million units are in uncontrolled market-rate inventory. New market-rate construction has proceeded steadily over the past 15 years, averaging 15,000–20,000 units annually (prior to the 2020 pandemic). Supply growth has not kept pace with demand, and prices have risen sharply, but construction has not ceased — consistent with the hypothesis that new-unit exemptions blunt the supply effect of stabilization.

• Sweden: The rental sector declined from 45% of housing stock in 1960 to approximately 12% by 2020, while owner-occupation (and cooperative ownership) rose. This is consistent with rent regulation suppressing private rental supply incentives, with owner-occupied alternatives absorbing demand. New multi-family construction has continued, but predominantly in owner-occupied and cooperative forms, not as rental. Ågren (2016) found this shift accelerated after stringent rent regulation was implemented. This is a supply-composition effect: rent regulation did not eliminate housing construction, but shifted it away from regulated rental toward unregulated ownership forms.

• Germany: Despite national rent stabilization, German construction of new apartments has remained substantial (approximately 240,000 units annually in recent years), above the pre-2015 rate. The exemption of new units from the rent brake appears effective in preserving new-construction incentives. However, the pace remains below the level economists estimate would be needed to reverse affordability deterioration in major metros, and analysts debate whether the gap is attributable to rent policy or to other constraints (zoning, permitting, construction costs).

Who benefits

Real estate developers and property owners have a direct financial interest in eliminating or weakening rent regulation, since deregulation increases the revenue and resale value of their properties. Industry associations including the National Association of Residential Property Managers (NARPM) and state apartment associations fund advocacy and research emphasizing rent control’s supply costs. Large institutional investors in multi-family housing also benefit: companies like Brookfield, Blackstone, and KBS have invested substantially in unregulated markets and derive higher returns there, creating an incentive to oppose regulation.

Libertarian and market-oriented policy organizations — including the Cato Institute, American Enterprise Institute, and Reason Foundation — have made opposition to rent control a trademark ideological position, independent of empirical evidence specific to each jurisdiction. Their framing has achieved significant influence in mainstream housing policy discourse, contributing to a rhetorical environment where rent control is assumed to be self-evidently counterproductive.

Sitting tenants in rent-controlled or stabilized units benefit from the policy; as rents for uncontrolled units rise, the value of living in a stabilized unit (the “rent gap”) increases. This creates a tension: the policy helps the protected group (existing tenants) while potentially harming the unprotected group (new entrants, renters in the uncontrolled stock). Tenant unions and housing advocacy organizations representing sitting tenants have therefore supported rent regulation even when evidence on aggregate supply effects is contested.

The counter

The strongest version of the counter-argument is methodological: rent control is often conflated with restrictive zoning, stringent permitting, and high construction costs — all of which independently suppress supply. When these are controlled for econometrically, the residual supply effect of rent control alone is smaller than naive comparisons suggest. The Diamond et al. (2019) finding that rent control caused a 7% citywide rent increase, for example, assumes that the 15% supply reduction was entirely attributable to rent control and not to concurrent zoning constraints or other factors. If zoning was already suppressing supply by 20%, the incremental effect of rent control may be much smaller than 7%.

A second important counter is policy design: distinguishing hard first-generation rent control (nominal price ceilings with no exemptions) from modern rent stabilization (inflation-indexed, with new-construction exemptions) is critical. The empirical evidence clearly supports the claim that hard ceilings suppress supply; the evidence on modern stabilization is mixed and contested. Applying the first-generation evidence to argue against modern stabilization regimes commits an error of generalization. Sweden’s stringent regulation is very different from Germany’s inflation-capped system, which is different from New York’s stabilization with new-unit exemptions.

A third counter is the public provision alternative: Austria and Singapore demonstrate that abundant housing supply is compatible with price regulation when supply is provided directly by government rather than through private markets responding to price signals. This does not refute the supply mechanism — it simply shows that the mechanism can be overcome through a different institutional structure. The efficiency cost of regulation (deadweight loss, reduced maintenance, conversion to other uses) still exists, but is offset by direct provision and financing. This is a real trade-off, not a refutation of the supply effect.

Finally, the counter notes that other factors constraining supply — restrictive zoning, minimum lot sizes, parking requirements, environmental review, height limits, single-family-only zoning — suppress supply far more dramatically than rent control does. Sixty percent of residential land in Portland is zoned single-family-only; 75% in San Francisco and Seattle. These zoning constraints directly prevent the construction of multifamily housing independent of rent policy. A reasonable counter-position is that fixing zoning should be the priority, and that debate over rent policy is a distraction from the binding supply constraint.

The verdict

The empirical evidence supports a partial verdict on the rent control supply claim:

1. Hard rent control (first-generation price ceilings) demonstrably reduces supply: The Diamond et al. natural experiment, the Cambridge discontinuity, and the historical record of hard rent control in Sweden and NYC all confirm that binding nominal price ceilings suppress new construction and induce unit conversion. This is the clearest empirical finding in the literature.

2. Modern rent stabilization (inflation-indexed, new-construction-exempt) shows contested, smaller effects: When rent regulation includes new-construction exemptions, annual inflation adjustments, and vacancy decontrol, the supply effect is much smaller and sometimes not statistically significant. California, New York (post-1974), and Germany provide evidence that substantial housing construction can coexist with rent stabilization when designed to exempt new supply.

3. Confounding with zoning remains unresolved: The largest empirical difficulty is isolating rent control’s effect from concurrent zoning and permitting constraints. Econometric attempts to decompose the two suggest zoning is the larger supply constraint in most US contexts, meaning rent control’s marginal effect is smaller than cross-sectional comparisons suggest. But this does not prove the effect is zero.

4. The mechanism is economically coherent but empirically smaller than the simple narrative suggests: The claim that “rent control reduces supply” is true in the direction and consistent with economic theory. The magnitude in modern contexts is substantially smaller than a reading of hard-control evidence would suggest.

The verdict is partial rather than supported because (a) the empirical evidence, while directionally supporting the claim, is substantially weaker for modern stabilization policies than for first-generation hard control; (b) the effect size is smaller than implied by the claim statement; (c) confounding with zoning prevents clean causal identification in most contexts; and (d) the supply effect can be structurally offset by public provision (as Sweden and Austria have demonstrated). The claim is true in a limited, conditional form — it applies to hard rent ceilings, not to all rent regulation — and the harm is smaller than a naive reading of the mechanism would suggest.

References

Angrist, J. D. (1995). The effect of the 1980s rent control law: Quasi-experimental methodology and microeconomic effects. Journal of Political Economy, 103(4), 851–880. https://doi.org/10.1086/262008

Ågren, C. (2016). Bostadskris i Välfärdslandet: En kunskapsöversikt om utvecklingen av bostadsmarknaden och bostadspolitiken i Sverige. Swedish National Board of Housing, Building and Planning.

Arnott, R. (1995). Time for revisionism on rent control? Journal of Economic Perspectives, 9(1), 99–120. https://doi.org/10.1257/jep.9.1.99

Baum-Snow, N., & Marion, J. (2009). The effects of wide roads on urban populations: Evidence from the U.S. interstate highway system. Journal of Urban Economics, 65(3), 385–409. https://doi.org/10.1016/j.jue.2009.01.006

Diamond, R., McQuade, T., & Qian, F. (2019). The effects of rent control expansion on tenants, landlords, and inequality: Evidence from San Francisco. American Economic Review, 109(9), 3365–3394. https://doi.org/10.1257/aer.20181289

Glaeser, E. L., & Nathanson, C. G. (2017). An extrapolative model of house price dynamics. Journal of Financial Economics, 126(3), 638–657. https://doi.org/10.1016/j.jfineco.2017.10.004

Lind, H. (2005). Pricing regulation of residential rental housing: A review of the international evidence. Journal of Housing and the Built Environment, 20(3), 267–279. https://doi.org/10.1007/s10901-005-9010-x

Marks, D. B., & Olsen, E. O. (2003). The social costs of manufactured sterility in rent controlled housing: Welfare effects on nontenants. International Review of Law and Economics, 22(4), 421–451. https://doi.org/10.1016/S0144-8188(02)00100-2

Muffels, R. A. J., & Fouarge, D. (2002). The role of stability and change in labor market careers: Relative job security and mobility in European welfare states. Journal of Labor Research, 23(2), 217–237.