View all Articles
Commentary By Ingrid Gould Ellen

Housing America’s Cities: Promising Policy Ideas for Affordable Housing

Editor's note: The following is the first chapter of The Next Urban Renaissance (2015) published by the Manhattan Institute.


New York City mayor Bill de Blasio recently announced an ambitious plan to build or preserve 200,000 units of affordable housing in the next ten years. More than 100 pages in length, the plan lays out a long list of strategies to increase the supply of affordable housing. Not all the ideas involve spending more money; but overall, the plan calls for considerable resources, proposing to spend over $8 billion in city funds and nearly $3 billion in state and federal funds.

Cities around the country would be hard-pressed to match this financial commitment, and a significant increase in federal spending on housing is highly unlikely. But even without spending billions, cities can still do much to enhance the affordability of housing.

This essay suggests three such reforms. First, cities could incentivize construction and development—and thereby increase the supply of housing—by more heavily taxing land than property. Such a “split-rate” tax would encourage development of underutilized land by reducing the added tax burden that standard property taxes impose on improving buildings. Second, cities could reduce (or even eliminate) minimum parking requirements that significantly increase the cost of housing. Finally, cities could shift some of the public funds currently spent on homeless shelters to time-limited rental subsidies for those at risk of homelessness. None of these ideas is new, but each deserves serious reconsideration as housing affordability problems mount around the country, especially in high-demand, coastal cities.

I. Adopt Land-Value Taxation

Henry George first proposed the idea of a land tax in his 1879 classic, Progress and Poverty.1 Economists ever since have celebrated the land tax as the most efficient, least distortionary way that governments can raise money. Unlike taxing income, which can discourage work, or taxing buildings, which can discourage property investment and maintenance, taxing land will not inhibit supply because the quantity of land is literally fixed.

Further, a land tax would discourage speculators from hoarding undeveloped land and incentivize them to develop their parcels to the full extent allowable. Regardless of whether a parcel sits vacant, houses a partially occupied, one-story retail strip, or holds a 30-story apartment tower, the annual tax bill would be the same. By switching to a land tax, a city could therefore increase the supply of housing and, by doing so, reduce prices across the board.

Land taxes offer other benefits. They should appeal to environ- mentalists by encouraging denser development in cities, thus reducing suburban sprawl.2 George also emphasized the normative appeal of a land tax. He stressed that increases in the value of a parcel of land are essentially windfall gains from urban growth, as any appreciation is driven by private and public investment in the surrounding area, not by any actions of the owner of the particular parcel. Thus, he argued that owners should be taxed on the value of their land. In the words of George, if you own land, “you need do nothing more. You may sit down and smoke your pipe; you may lie around like the lazzaroni of Naples or the leperos of Mexico; you may go up in a balloon, or down a hole in the ground; and without doing one stroke of work, without adding one iota to the wealth of the community, in ten years, you will be rich!”3

Despite the clear appeal of a land tax, surprisingly few localities have adopted it, perhaps in part because of the political challenges posed by transitioning to such a tax. That said, the idea is not completely untested. About 700 cities around the world (in 30 countries) currently use what is called a split-rate tax system, in which the tax rate levied on the value of land is higher than the rate on the value of improvements.4

A split-rate tax would likely be more viable politically because it would involve a less dramatic shift for most jurisdictions; yet it would still offer many of the same advantages as a land tax. By taxing improvements at lower rates than land, owners would face a reduced disincentive to invest in improvements that increase the value of the structures on their property. In the U.S., only 16 municipalities in Pennsylvania and two counties in Hawaii have adopted a split-rate property tax.5 The Connecticut and Virginia state legislatures recently granted cities the authority to implement a split-rate tax, but no cities in either state have yet adopted one.6

Admittedly, there is little empirical work on the effects of a split-rate tax, partly because of its relatively limited reach to date. But what little evidence we have suggests promise. Wallace Oates and Robert Schwab studied a tax reform in Pittsburgh, which increased taxes on land to more than five times the rate on structures. The authors found that the higher taxes on land—and correspondingly lower taxes on im- provements—stimulated building activity in the city.7

Perhaps the most commonly voiced criticism of a split-rate tax is concern about feasibility and, in particular, reliable assessments. Property-tax assessors are widely criticized for failing to provide timely and accurate assessments of property values in our current property-tax systems, and a system that requires valuing land separately from buildings or improvements would be even more challenging. While 29 states require land and improvements to be assessed independently for property-tax purposes, assessors in these states feel little pressure to accurately price the underlying land, as any disputes that arise center on total assessments, not the division of value between structure and land.

One recent survey8 suggested that most assessors in these states simply estimate land values from the sales of vacant land, an approach that may yield highly inaccurate estimates in jurisdictions where such sales are rare. But in this age of Big Data, it is hard to imagine that we cannot develop more reliable ways to value land. Property-tax assessors have already created new computer models to arrive at more accurate and timely assessments of the value of individual components of properties, including land; further improvements are surely possible.

Simpler approaches can work, too. Several recent studies show that teardowns, or situations when properties are sold and then fairly quickly demolished, can be used to value land, at least in large cities.9 Further, most housing in the U.S. is now produced through planned unit developments, with a minimal variety of homes. Thus, variation in prices of such comparable homes sold across areas should be explained largely by variation in the price of land.

The larger challenge is the politics. Any proposal for tax-system reform will inevitably create winners as well as losers, and the losers will have an incentive to resist change. In the case of a split-rate tax, the losers will be owners of parcels with high land-to-building value ratios, or owners of small buildings on valuable, centrally located parcels, who will likely see an increase in their tax bills after the switch to a split-rate tax.

Jurisdictions can mitigate the increased burden on selected property owners—and the political opposition they invite—by phasing in the split-rate tax rate slowly. They should also be sure to adopt timely and state-of-the-art assessment practices so that both structure and land values are assessed as fairly and accurately as possible. Finally, cities might introduce a tax credit to help reduce the burden on lower-income landowners, or owners of affordable housing, who may otherwise have difficulty paying the increased tax bill.

To be sure, on its own, a split-rate tax may not be sufficient to dramatically transform the level of housing production in a city. But switching to a smarter tax-policy regime is an important piece of the solution.

II. Eliminate Minimum-Parking Requirements

Nearly every U.S. city mandates that developers include a minimum number of parking spaces in their developments. New York City’s government, for example, started requiring that new residential development include off-street parking in 1950. The 1961 Zoning Resolution later increased those requirements.

Today, developers in New York are required to provide an average of 43 new off-street parking spaces for every 100 new housing units they construct. The minimum requirements differ dramatically across boroughs, ranging from an average of 122 spaces per 100 units on Staten Island to just five spaces in Manhattan (most of which is exempt).

Around the country, requirements are typically even more onerous; the median municipality in the U.S. requires that developers set aside 1.5 parking spaces for each two-bedroom unit. Given that a parking space requires 300 to 400 square feet of building area, these regulations typically add about 50 percent to the floor area needed to build a 900-square-foot apartment.10 Of course, many—indeed, most—developers might choose to provide parking, anyway; but recent research suggests that in many jurisdictions, the requirements are forcing developers to build more parking spaces than they would otherwise build.11

“Even without spending billions, cities can do much to enhance the affordability of housing.”

Consider that in New York City, the average five-to-nine-unit development that qualified for a waiver from parking requirements built just 0.5 parking spaces; developments that failed to qualify for a waiver included 5.3 spaces, almost exactly the average requirement for a project of this size. For developments that include ten to 14 units, those that qualified for a waiver built fewerthan one parking space, while those that did not included more than seven spots—again, roughly the average number required by the city.12 (An additional concern is that requirements that vary by building size may create perverse incentives for developers to subdivide lots and build multiple smaller buildings in order to evade parking requirements.)

Proponents of minimum parking requirements argue that, without them, developers would under-provide parking, as they would fail to take into account the burden that new residents without parking spaces would impose on existing residents—car owners, in particular—in the neighborhood.

While some car owners would surely be willing to pay a premium for the privilege of off-street parking, others might choose to park in nearby garages or on neighborhood streets, imposing an external congestion cost on neighbors. The additional time that drivers would then have to spend traversing the neighborhood to look for parking would increase congestion and local air pollution. Further, city officials may worry that more competition for parking will make it harder for them to retain middle-class households, most of which own cars.13

While these externality arguments have some merit, the provision of parking itself generates negative externalities. For one thing, off-street parking may encourage both car ownership and driving, which increases congestion and emissions. For another, off-street parking typically comes with unsightly street-front entrances that undermine a streetscape and eliminate at least one curbside parking space.

Thus, the benefits of internalizing this externality of increased competition for parking must be weighed against the externalities generated by the provision of parking, as well as the enormous cost that such parking requirements impose on housing. By requiring that developers set aside roughly 450 more square feet of additional space for every residential unit, these regulations add significantly to construction costs and increase the price that residents must pay for housing. Donald Shoup estimates the cost of a Los Angeles parking space at over $31,000.14

Significantly, these costs are passed on to all city residents, not just car owners. Even apartment dwellers without cars are effectively forced to pay for the cost of a parking space because the cost of parking provision makes development more expensive. There is no way for residents to reduce what they pay for parking by driving less or owning fewer cars. Their only option is to move to another jurisdiction that does not have binding requirements.

If a jurisdiction wishes to address parking-congestion externalities that may exist, there are more efficient ways to address them, such as charging for on-street parking to balance supply and demand. To make such parking fees more politically palatable, Shoup has proposed the establishment of Parking Benefit Districts that would dedicate revenues raised from curbside parking charges to pay for local neighborhood im- provements.15 Given improvements in technology, the transaction costs of charging for parking (and adjusting fees by time of day) are now relatively low.

While some city leaders might still blanch at the political risks of removing minimum parking requirements, some leading cities have done so without any major revolts or reductions in growth rates. In the early 2000s, London ended its minimum parking requirements and actually replaced them with maximum parking standards. Similarly, through a series of ordinances passed in the late 1990s and 2000s, San Francisco eliminated minimum parking requirements in much of the city.

While cities may shy away from complete elimination of requirements and replacement with maximum parking standards, they might at least consider eliminating requirements for buildings constructed within a certain distance of transit. Alternatively, or in addition, cities might eliminate requirements for affordable housing developments, especially in areas near public transit, as fewer low-income renters own cars and those who do drive less than other car owners.16

High construction costs frustrate efforts to create more affordable housing around the country. While there is a limit to what cities can do to reduce those costs, they should do whatever they can to wrestle with the aspects of the costs they can control. Parking requirements are one such lever and, as such, should be carefully considered for reform.

III. Experiment with Short-Term Rent Subsidies

Federal assistance programs currently help approximately 5 million low-income households pay for housing. The largest assistance program is the Housing Choice Voucher program, which has been shown to be effective in reducing household crowding, lowering the risk of doubling up, and preventing homelessness. However, only about one in four renter households eligible for federal housing assistance actually receives it. Thus, in most cities across the country, demand for housing assistance far outstrips supply.

The lucky few essentially receive a lifetime subsidy if their incomes remain below eligibility thresholds; the unlucky many receive nothing and continue to struggle to make ends meet. Estimates from the U.S. Department of Housing and Urban Development show that two-thirds of income-eligible but unassisted renters paid more than half of their income on rent, leaving little for other critical needs.17

Policy reform at the federal level would be needed to fundamentally address this inequity, and the Bipartisan Housing Commission recently offered several proposals.18 But reform is not coming anytime soon. In the meantime, cities have some ability, in partnership with their states, to experiment with time-limited subsidies that would allow them to serve more households. Indeed, a number of cities have recently moved in this direction in their approach to assisting homeless families and individuals.

Rather than placing families in need in shelters, they have provided them with what is called “rapid rehousing” help, or temporary assistance to move into permanent housing as quickly as possible, typically in the private market. Government officials provide housing search counseling, help with landlord negotiation, short-term rental assistance, and other services as needed to help participants stay in their homes. Because assistance is time-limited, cities have been able to serve many more households than they would by relying on traditional housing subsidies. These time-limited subsidies have also been shown to be cheaper than assigning people to temporary shelters.19

“High construction costs frustrate efforts to create more affordable housing. Cities should do whatever they can to wrestle with those costs.”

Preliminary evidence suggests that these rapid rehousing programs have been effective in reducing spells of homelessness and in helping people find and keep housing. Most people participating are able to find housing, and few appear to return to shelters. Rapid rehousing programs may also promote long-term self-sufficiency, as shelters are not only costly for taxpayers but also stressful and disruptive for families and children.20 It is highly difficult for adults to find and maintain a job while living in a shelter.

There has not yet been a rigorous evaluation of the long-term impacts of rapid rehousing programs, and many questions still remain: Will landlords be willing to accept them? How will families manage the transition when the subsidy ends? Will such time-limited subsidies make any meaningful difference in an individual’s long-term well-being? Still, the promise of initial evidence supports further exploration of this approach.

Further, this evidence is sufficiently promising to call for pilot programs to provide short-term housing assistance to low-income households more generally. For example, cities might introduce or expand efforts to provide temporary assistance to help renters weather shortterm setbacks, such as a job loss or an unexpected health care expense.

Existing housing programs are useless in providing such short-term assistance, designed, as they are, to address permanently low incomes rather than volatility. In many cities, wait lists for housing assistance are so long that applicants have to wait five to ten years to receive subsidies, rendering them completely unable to address temporary setbacks. Thus, cities might experiment and test different models of providing emergency assistance, such as developing rental insurance programs that provide one-time assistance for rent shortfalls. We need more evidence to understand what models are most cost-effective in stabilizing families at risk of losing their homes, and cities can help.

Some charge that, in high-cost cities like New York, time-limited subsidies aimed at either preventing eviction or rapid rehousing would be too modest to meaningfully help households and would merely forestall the inevitable. Yet this is far from a foregone conclusion. Even in high-cost cities, short-term subsidies might help many more households to get back on their feet after losing their homes or to avoid homelessness altogether. With three out of four low-income households not receiving any housing subsidy at all, the dramatic need for broader housing assistance warrants more exploration of ways to stretch scarce subsidy dollars further.


To be sure, these proposed reforms would not be sufficient to solve the housing affordability problems that so many cities around the country face. But both theory and existing research suggest that they would be promising additions to a city’s policy toolbox.


1. Henry George, Progress and Poverty: An Inquiry into the Cause of Industrial Depressions and of Increase of Want with Increase of Wealth—The Remedy (New York: Doubleday, Page, 1920).
2. Jan K. Bruckner, “Property Taxation and Urban Sprawl,” in Property Taxation and Local Government Finance: Essays in Honor of C. Lowell Harriss, ed. Wallace E. Oates (Cambridge, Mass.: Lincoln Institute of Land Policy, 2001), 153–72.
3. George, Progress and Poverty, book 5, chap. 2,
4. Jeffrey P. Cohen and Cletus Coughlin, “An Introduction to Two-Rate Taxation of Land and Buildings,” Federal Reserve Bank of St. Louis Review 87, no. 3 (2005): 359–74.
5. Sally Kwak and James Mak, “Political Economy of Property Tax Reform: Hawaii’s Experiment with Split-Rate Taxation,” American Journal of Economics and Sociology 70 (2011): 4–29.
6. Richard F. Dye and Richard W. England, “Assessing the Theory and Practice of Land Value Taxation,” Lincoln Institute of Land Policy, 2010.
7. Wallace Oates and Robert Schwab, “The Impact of Urban Land Taxation: The Pittsburgh Experience,” National Tax Journal 50, no. 1 (1997): 1–21.
8. Dye and England, “Assessing the Theory and Practice of Land Value Taxation.”
9. Richard F. Dye and Daniel P. McMillen, “Teardowns and Land Values in the Chicago Metropolitan Area,” Journal of Urban Economics 61, no. 1 (2007): 45–63; and Ingrid Gould Ellen and Michael Gedal, “Valuing Urban Land: Comparing the Use of Teardown and Vacant Land Sales,” Furman Center, working paper, 2012.
10. Jenny Xie, “Exposed: America’s Totally Inconsistent Minimum Parking Requirements,”
The Atlantic CityLab, August 23, 2013.
11. W. Bowman Cutter and Sofia Franco, “Do Parking Requirements Significantly Increase the Area Dedicated to Parking? A Test of the Effect of Parking Requirements Values
in Los Angeles County,” Transportation Research Part A: Policy and Practice 46, no. 6 (2012): 901–25.
12. Furman Center, “Searching for the Right Spot: Minimum Parking Requirements and Housing Affordability in New York City,” March 2012.
13. Ibid. Even in New York City, 62 percent of households with incomes that are at least 150 percent of the median own a car.
14. Donald C. Shoup, The High Cost of Free Parking (Chicago: American Planning Association, 2011).
15. Idem, “The Trouble with Minimum Parking Requirements,” Transportation Research Part A: Policy and Practice 33, no. 7 (1999): 549–74.
16. According to the 2013 American Community Survey, about 20 percent of all renters did not own a car, and the proportion is likely higher among low-income renters.
17. U.S. Department of Housing and Urban Development, “Worst Case Housing Needs,” 2009.
18. The Bipartisan Housing Commission recently advocated that the federal government guarantee assistance to all households earning less than 30 percent of their area median income that apply, while also providing one-time, temporary assistance to households earning 30–80 percent of area median income.
19. National Alliance to End Homelessness, “Rapid Rehousing: A History and Core Components,” Solutions Brief, April 22, 2014, rapid-re-housing-a-history-and-core-components.
20. A Washington State Department of Commerce study found that working-age adults who received rapid rehousing were 25 percent more likely to be employed over the year following receipt of assistance and earned about $420 more than comparison adults who did not receive rapid rehousing assistance, ms/rda/research/11/185.pdf.