«reNtal HousiNg Policy iN tHe uNited states Volume 13, Number 2 • 2011 U.S. Department of Housing and Urban Development | Office of Policy ...»
Other work has looked at the effect of land use control indices on population and local economic growth. For example, Nandwa and Ogura (2010) found a negative correlation among several land use indices and area growth. The Wharton Index correlation with growth is negative but not significantly so, perhaps because the index is correlated with hard-to-measure variables that are positively correlated with demand.
Glaeser and Kahn (2010) looked at the relationship between the Wharton Index and carbon emissions for new households across metropolitan areas. I found that the more restrictive areas have less energy use, primarily because of their temperate climates. Disproportionately restricting the growth of more environmentally friendly areas, such as coastal California, presumably shifts new development toward less environmentally friendly, but more pro-development areas, such as Houston.
Malpezzi (1996) played an important role in developing the index.
Conclusion: Toward a New Housing Policy What would an ideal national housing policy look like? A purist might hold that because the citizenship benefits of homeownership have not been conclusively proven, it makes sense to have no pro-homeownership bias whatsoever. Perhaps the tax system should move away from an earnings tax toward a consumption tax, which would favor all forms of investment, but the purist’s position would avoid any pro-homeownership bias whatsoever.
A less extreme position is to accept that some benefits from homeownership do come from greater investment in social capital. In this case, it becomes reasonable to subsidize homeownership in some way. But even with a belief in subsidizing homeownership, the current system still seems poorly targeted, highly regressive, and excessively engineered to encourage borrowing money and buying big homes.
One feasible path to reform would be to gradually lower the upper limit on the home mortgage interest deduction by $100,000 per year over the next 7 years. Currently, the mortgage interest deduction is capped at $1 million. President George W. Bush’s tax reform panel suggested that this should be cut substantially. To avoid creating too much change in the housing market too quickly, the cap could be cut by $100,000 per year for the next 7 years. This gradualist approach would be unlikely to significantly roil markets and it would have little effect on the vast majority of Americans whose mortgages are below $300,000.
By lowering the cap on the deduction, the extreme incentives to borrow money to buy big houses would be reduced. The incentives to buy houses that cost more than $300,000 would be substantially reduced. Some of the worst distortions associated with the home mortgage interest deduction would be eliminated. The artificial incentive to borrow would be capped at $300,000, and the artificial incentive to buy would drop off at somewhat more than that amount. A lower cap on the home mortgage interest deduction would achieve some of the benefits of eliminating the deduction altogether with a much less drastic policy shift.
After America achieved a lower tax deduction, it would be possible to envision proposing an alternative situation in which all homeowners—or all homeowners below a certain income threshold—receive a flat homeowner’s tax credit that is independent of the size of the home and the size of the mortgage. This annual tax credit would go to any homeowner who occupied his or her home. The flat tax credit would appeal to people with smaller mortgages and it would reduce the benefit to borrow and buy big. For example, assume that a tax credit of $2,000 per year was available to all homeowners. This credit would be appealing to people whose marginal tax interest rate obligations were less than $66,000 per year and who had a marginal tax rate of less than 30 percent. If the prevailing interest rate were 6.6 percent, every homeowner who had a mortgage of less than $100,000 would be eligible for the tax credit. The tax credit could be made market specific so that higher tax credits were available in more expensive markets.
Even this flat tax policy would have the unfortunate side effect of encouraging people to leave cities that are dense with apartment buildings. If we wanted to encourage homeownership but limit the density-reducing effect of homeownership policies, it would be possible to slightly offset
the ownership tax credit with a density tax credit. For example, the ownership credit would be higher in places where density levels were higher. Renters could also receive a tax credit for living in high-density areas.
Alternatively, the tax credit could be tied to the area’s cost of living instead of the density level.
Giving a positive tax credit in places that are more expensive to live would reduce the tax incentive to leave those high-cost areas. Moreover, because housing is more expensive near the city center, this tax credit would reduce the incentive to flee the core.
Having a homeowner’s tax credit and accurately taxing the costs of driving and low-density living may be the best solutions. If individuals paid for the full social cost of their energy use (and traffic congestion), then we would not need to create further distortions that would push people toward dense living environments. Better energy pricing would also induce more efficient energy usage along other margins.
Although these suggested proposals for policy change may be one economist’s ideal, they will most likely not make any political headway. These proposals are too far from the status quo to have a chance of being implemented. Because I do not think it makes sense to let the perfect be the enemy of the good, I will suggest a policy path that is also difficult but less obviously impossible than a wholesale eradication of one of Washington’s most popular policies.
This suggested path would move the United States toward a system in which homeownership were handled with a flat tax credit that creates no incentives to borrow or buy big. Moreover, after we adopted a flat credit, we could then reasonably ask whether homeownership actually delivers social capital that is worth $2,000 per year. That debate could have the positive effect of pushing legislators toward adopting policies that are more soundly grounded on real evidence about the magnitude of the externalities of homeownership.
Although reducing the cap on the home mortgage interest deduction would create a fairer, less biased system, it would do little to make life better for renters except allow them to pay a slightly smaller share of the federal tax burden. Indeed, reducing the incentive to own may increase the demand for rental units and push rental costs up, at least in the short run. Ideally, we would pass legislation to ease the rental burden to offset this effect.
In the previous section on rental housing, structure type, and local barriers to building, I argued that limiting new construction artificially boosts housing prices. Unfortunately, it is quite difficult for the federal government to meaningfully influence local land use decisions. One strategy, borrowed from the U.S. Department of Education’s “Race to the Top,” is to offer a reward to localities that make meaningful changes that will increase the supply of multifamily rental housing units. If a pot of aid, which could be used for local housing-related infrastructure, were tied to changes in the permitting environment, then this incentive could both serve as a demonstration project and provide a meaningful push toward new development.
One reasonable way to proceed would be to allocate a pot of money that would then be allocated after 48 months based on new units produced relative to historic trends and changes in the land use and permitting procedures. The allocations could be paid for with the increased tax revenues coming from changes to the home mortgage interest deduction or with the low-income housing
tax credit. Indeed, one way of structuring the program would just be to allocate LIHTC funds to localities based on past performance in the construction of multifamily units and changes in the permitting process.
Another approach would be for the U.S. Department of Housing and Urban Development (HUD) to undertake lawsuits against communities that excessively restrict new development. Recently, HUD successfully undertook a case in Westchester, New York, that led to some regulatory reform.
The court system invariably involves a fair amount of randomness and unpredictability, but as evidenced by the Mount Laurel system,10 it can achieve powerful results. Mount Laurel itself offers an attractive model in which exclusive communities essentially pay a price for their restrictiveness, which then promotes more affordable housing elsewhere—this serves as a tax on NIMBYism.11 Final Words For too long, America has had privileged homeownership without fully internalizing the consequences of its programs. This policy’s specific structure has had adverse consequences: we subsidize borrowing and thereby encourage people to over-leverage themselves to bet on the vicissitudes of the housing market. One adverse consequence, however, would occur in any pro-homeownership policy. Because ownership and structure are so tightly linked, promoting homeownership means promoting single-family detached homes at the expense of multiunit dwellings.
This subsidy pushes people away from cities and toward sprawling suburbs. This push has adverse consequences for the environment and the economy. We need to rethink our pro-homeownership stand and, at the very least, eliminate some of the most extreme features of the home mortgage interest tax deduction. Slowly lowering the cap on the interest deduction would be quite reasonable. At the same time, we should also consider policy decisions that discourage the tendency of many localities to ardently oppose new development of multiunit structures.
Appendix: Model In this appendix, I examine the investment decisions of renters, owners, and landlords. I first focus on an investment that improves the physical condition of the house, which could be cleaning the gutters or fighting mold. I assume that this investment is noncontractible so that the renter cannot be paid for it by the owner or vice-versa. If the renter undertakes the maintenance action, there is a cost of T units of time and the cost of this in foregone after-tax earnings is (1-t)WT, where W is the tenant’s wage and t is the tax rate, plus M, any cash costs.
A tenant will undertake this improvement investment if and only if the cost (1-t)WT is less than the consumption benefit of the investment during the resident’s time in the house, denoted B, minus any increase in the rental payments charged by the landlord, which we denote rR. The In a 1975 court case, the Burlington County, New Jersey National Association for the Advancement of Colored People sued the Mount Laurel Township because the township’s zoning ordinances restricted lower income people from finding affordable housing.
The term NIMBYism is derived from the phrase, “Not in my back yard.”
house is more attractive as a result of the maintenance, and the landlord certainly has an incentive to charge more. All costs, benefits, prices, and rents are assumed to be time zero discounted real values for algebraic simplicity.
In some cases, a landlord may be willing to make the investment. We assume that the cost to the landlord is dWT + M, in before-tax dollars. The term d is meant to capture the possibility that the landlord has access to either more efficient or less expensive labor. I also assume that there is a potential legal penalty, denoted K, for failing to engage in the maintenance, which reflects regulations that may apply to home maintenance. This legal penalty is presumably an expected value because maintenance will often be hard to establish in a court of law. Because the landlord will typically be taxed on all earnings and be able to deduct expenses, the landlord will undertake the action if and only if rP + rR + K dWT + M. (2) In this case, the maintenance is more likely to be performed if the rent impact, the price impact, or the legal penalty is high.
I assume that the unit experience N possible opportunities for maintenance. These opportunities have identical costs but differ in their benefits B, which equal + m, where m is characterized by a density function f(m). I assume that rP = qpB, and rR = qRB, where qR ≤ 1, and I calculate benefits by adding together the benefits received by the landlord and tenant in after-tax dollars. I further assume that, if the benefits of a maintenance action are positive for the landlord, then the landlord will take the maintenance action. The demand for homeownership will rise if the maintenancerelated benefits of owning the home are larger relative to the benefits of renting it. With these assumptions, the following proposition follows.
(a) If, then renters will engage in some maintenance themselves, and, in this case, owner-occupied units always have more maintenance than rental units.
(b) If, then in rental units, the tenants will perform no maintenance and more maintenance will occur in rental units than in owner-occupied units if K is greater than some value K*, which is decreasing with qR, increasing with t and d, falling with W and T if and only if d (1 – t), and increasing with M if and only if 1 qR.
If d is large, then I denoted building size as U*, at which owner-occupiers and landlords receive the same benefits. For building sizes above U*, owner-occupancy is less attractive than renting and, for sizes below U*, owner-occupancy is more attractive. The value of U* is falling with d and qR and if f (m) is uniform, U* is independent of and rising with M.