Zohran Mamdani and the Economics of Well-Intended Harm
Equity in housing requires more than intent; it demands policies grounded in economic reality.
“Next to bombing, rent control is the most effective technique so far known for destroying cities.”
Assar Lindbeck (1930-2020), Swedish economist
New York City mayoral candidate Zohran Mamdani has proposed an aggressive public-housing agenda to address the city’s affordability crisis. His platform calls for tripling the production of city-subsidized housing — roughly 200,000 new permanently affordable units over a decade — as well as doubling capital investment in public housing (NYCHA), staffing up housing agencies, and fast-tracking the development of affordable projects. He also advocates expansive planning reforms — for example increasing zoned residential capacity, eliminating parking minimums, and requiring all new market-rate housing to be rent-stabilized. In effect, Mamdani’s plan combines massive public construction and subsidy of housing with price controls on rents. We examine these policies through the lens of economic theory and empirical evidence. In theory, binding rent ceilings and heavy subsidization have well-known welfare effects on consumer surplus (CS), producer surplus (PS), and deadweight loss (DWL), as well as longer-run incentive problems. In practice, academic studies of rent control and subsidized housing find significant market distortions. As we show below, the net effect of Mamdani’s proposals is likely to be a welfare loss rather than a cure for affordability.
Rent Control in Theory
Static Supply-Demand Model (Short-Run)
Assumptions:
One homogeneous rental‐housing market with demand Qd = D(p) (downward-sloping, D'(p) < 0) and supply Q* = S(p) (upward-sloping, S'(p) > 0).
In a free market, equilibrium (p*, Q*) satisfies D(p*) = S(p*).
Rent control imposes a binding price ceiling pc < p* (a “rent freeze,” as in Mamdani’s proposal).
Sellers (landlords) are profit‐maximizing and may convert or sell buildings if rents are capped.
Free-Market Equilibrium
Without control, equilibrium price p* and quantity Q* solve:
so that Q* = D(p*) = S(p*).
Imposing Rent Control
Now set the rent equal to pc < p*. By monotonicity of D and S:
Hence
meaning supply at the controlled price is strictly less than demand. The market can only transact Qcontrol = S(pc) units (since landlords supply S(pc) but demand is higher). Thus quantity falls from Q* to S(pc) < Q*.
Graphically, a binding ceiling produces a horizontal gap between demand and supply (a shortage).
Supply Reduction
Since pc is fixed below the free‐market rent, some landlords earn lower (or even negative) returns. They respond by restricting supply. For example, they may convert units to other uses (selling as condos or owner‐occupied homes) to capture higher prices. Friedman and Stigler (1946) famously argue that caps induce landlords to sell to owner‐occupants so as to realize the true market value:
“The absence of a ceiling on the selling price of housing means that at present homes occupied by their owners are being rationed by the 1906 method—to the highest bidder. The selling prices of houses is rising as the large and increasing demand encounters the relatively fixed supply. Consequently, many a landlord is deciding that it is better to sell at the inflated market price than to rent at a fixed ceiling price.”
Friedman and Stigler (1946)
Immediate Consequences
1. Rents Rise in Non-Controlled Segments
Increased financial burden on renters in uncontrolled markets, disproportionately affecting newcomers (e.g., migrants, young families).
Accelerated gentrification, as landlords in uncontrolled areas capitalize on scarcity by catering to higher-income tenants.
Reduced affordability in adjacent markets, exacerbating housing inequality as displaced renters compete for limited uncontrolled units.
2. Misallocation and Welfare Loss
Inefficient use of housing: Tenants stay in oversized/undersized units (e.g., empty nesters in family-sized apartments) due to fear of losing rent-controlled leases.
Reduced labor mobility: Workers avoid relocating for better jobs to retain controlled leases, stifling economic productivity.
Deadweight loss: Society loses potential gains from optimal housing matches (e.g., families cramped in small units while singles occupy large ones).
3. Landlord Rent-Seeking
Discriminatory practices: Landlords favor tenants without children, higher-income renters, or those offering side payments (e.g., "key money").
Deterioration of rental quality: With capped returns, landlords cut maintenance, leading to urban blight.
Black markets: Illegal fees or bribes emerge, penalizing honest renters and undermining trust in housing institutions.
Dynamic Housing‐Stock Model (Long Run)
Assumptions:
Let Ht = total housing stock (rental units) at time t.
Housing evolves by investment and depreciation:
\(H_{t+1} = (1-\delta) H_t+I_t,\)where 𝛿 ∈ [0, 1) is the depreciation rate and It is new construction.
Investors build only if expected rent covers costs: let the (net) rental price be Rt, construction cost per unit be c. Assume It is increasing in the profit margin (Rt - c).
Initially at t = 0, free-market rent R* prevails and investment is I* (so Ht would grow to some steady path)
From t = 1 to T, rent is frozen at Rc < R* (Mamdani’s freeze). After t = T, assume control is lifted and rent returns to market level.
No-Control Baseline
If there were never a rent freeze, landlords always expect profit
and thus keep building at rate
For simplicity first set 𝛿 = 0 (no depreciation). Then
where H0 is the initial stock. The stock grows linearly at rate I*.
With Rent Freeze
Under rent control Rc < R*, profit 𝜋c = Rc - c is lower, so investment drops to
Two cases: if Rc barely covers cost, maybe Ic ≈ 0; more generally Ic < I*.
Still with 𝛿 = 0: while control lasts (1 ≤ t ≤ T),
Since Ic < I*, we have
In other words, by the end of the freeze, housing stock is strictly lower than it would have been without control.
After Control is Lifted
For t > T, both scenarios have normal rent R*, so both invest at I* again. However, starting stocks differ. For t > T:
The difference for t ≥ T is
This shows Htc < Htnc for all t ≥ T; indeed the gap (I*-Ic)T stays constant. Even if control is temporary, the lost construction during 1 ≤ t ≤ T never catches up (when 𝛿 = 0).
Including Depreciation
Allow 𝛿 > 0. The difference equation can be solved or examined qualitatively. One finds similarly that if
and
with Ic < I* on t ≤ T, then HTc < HTnc and thereafter Htc < Htnc for a long time. (With 𝛿 > 0, both stocks eventually approach the steady-state I*/𝛿, but Htc remains below Htnc until very far out).
Implications — Long-Run Supply Loss
This simple math shows a persistent shortfall of housing due to the freeze. Even after rent control ends, the city permanently lags the no-control stock by (I*-Ic)T units (assuming no other offsetting policy). In more realistic settings, depreciation or inefficiencies make recovery even slower, and some units may never be replaced (for example, redeveloped as condos).
Thus, “solving” a supply problem by rent control actually shrinks supply over time, a legacy that persists even if controls are later lifted.
Research Findings
Empirical studies confirm these theoretical predictions. For example, rent control’s real effects have been measured in multiple cities. In a landmark study using San Francisco data, Diamond, McQuade, and Qian (2019) estimate the impact of a sudden rent-control expansion. They report that rent control reduced supply of controlled apartments by about 15%, even as it raised the probability of remaining in place by roughly 20% for covered tenants. The reduced supply meant rents in the unconstrained sector actually rose (about 5.1% citywide) — a crowding-out of costs onto other renters. The authors find that the welfare loss from the lost units exceeds the gain to protected tenants, implying net social cost. Similarly, Autor et al. (2014) study Cambridge, MA, where strict rent control was abolished in 1995. They find that ending controls produced a roughly $1.8 billion (approximately +25%) increase in total housing value through 2004, evidence that the controls had been suppressing quality and supply for a decade. In sum, detailed case studies align with textbook theory: rent ceilings help some existing tenants in the short run but exacerbate shortages and reduce housing stock in the long run. Kholodilin (2022) reaches the same conclusion: controls do slow rent increases, but they “lead to a wide range of adverse effects” for landlords and tenants (e.g. reduced maintenance, evictions when tenants must leave, and exclusion of new renters).
Turning to supply-side interventions, the empirical consensus is that heavily subsidized construction crowds out private housing. Using U.S. data, Sinai and Waldfogel (2005) ask whether public and subsidized units raise the total housing stock. They find government-financed units do increase housing, but with diminishing returns: on average three new subsidized units replace two private ones, so the net gain is only one unit per three built. This implies roughly 67% crowd-out. Malpezzi and Vandell (2002) reach a similar finding for the Low-Income Housing Tax Credit (a major program of private affordable construction): their state-level analysis finds no significant correlation between LIHTC units built and the overall housing stock, implying nearly 100% substitution. A spatial econometric study by Eriksen and Rosenthal (2010) also estimates approximately 100% crowd-out for subsidized rental developments across market areas. In other words, when government or tax credits bankroll low-income projects, private builders simply cut back on market-rate construction, leaving total housing about the same.
These findings have practical implications for Mamdani’s plan. A goal of “tripling” city-capacity assumes many units are additional to what the market would supply, but the literature suggests much of that effort would substitute for private development. If rent stabilization is applied to the new units, the substitution effect only intensifies. Importantly, the evidence also points to better alternatives: tenant-based subsidies. Stegman (2009) reviews cost-effectiveness studies and concludes that housing vouchers (tenant subsidies) deliver accommodation of any required quality “at a much lower total cost” than project-based programs. Vouchers or certificates allow low-income families to rent in the private market without displacing an equal amount of private building, because landlords see those as ordinary market tenants (subject to market rents). By contrast, Mamdani’s project-based emphasis risks wasting resources on substitution.
In summary, empirical research on urban housing does not support the optimism of such direct interventions. Cities that tried price caps have seen housing shortages deepen (often invisibly, through decay or “shadow” markets), and places that massively expanded public housing have not reliably improved affordability for the target population. Where studies exist, they suggest market-friendly tools (e.g. vouchers, deregulation of supply constraints) outperform heavy-handed controls. Overall, rent regulation effectively reduces rental prices but often leads to various negative consequences, resulting in an overall net harm unless properly balanced with mitigating measures.
Conclusion
Zohran Mamdani’s housing proposals — large-scale public building combined with broad rent controls — conflict with core economic principles and empirical evidence. Static supply-and-demand analysis predicts shortages, lost units, and a deadweight-loss triangle from rent ceilings, effects corroborated by case studies of rent regulation (Diamond et al., 2019; Autor et al., 2014). Similarly, theory and data on subsidized construction warn of crowd-out: tax-funded building seldom grows the total housing stock one-for-one (Sinai & Waldfogel, 2005; Malpezzi & Vandell, 2002). In practice, these policies tend to benefit a few tenants at the expense of many others and of economic efficiency. A cost-effectiveness perspective even suggests vouchers and market-driven supply expansion achieve affordability with less waste of resources. In short, while aiming to improve housing, Mamdani’s direct interventions are likely to exacerbate the very shortages they target. A rigorous economic assessment thus advises caution: these well-intentioned measures would probably reduce overall welfare rather than resolve the underlying scarcity.
References
Autor, D., Palmer, C., & Pathak, P. (2014). Housing market spillovers: Evidence from the end of rent control in Cambridge, Massachusetts. Journal of Political Economy, 122(3), 661-717.
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), 3365-3398.
Eriksen, E. P., & Rosenthal, S. (2010). Incentives and subsidies: The economics of housing supply. Journal of Housing Economics, 19(2), 95-100.
Friedman, M., & Stigler, G. J. (1946). Roofs or ceilings? The current housing problem. Foundation for Economic Education.
Kholodilin, K. A. (2022). Rent control effects through the lens of empirical research (DIW Roundup 139). DIW Berlin.
Malpezzi, S., & Vandell, K. D. (2002). Does the low-income housing tax credit increase the supply of housing? Journal of Housing Research, 13(1), 117-159.
Sinai, T., & Waldfogel, J. (2005). Do low-income housing subsidies increase the occupied housing stock? Journal of Public Economics, 89(11-12), 2137-2164.
Stegman, M. A. (2009). The cost-effectiveness of alternative methods of delivering housing subsidies. Journal of Urban Affairs, 31(2), 159-179.