«reNtal HousiNg Policy iN tHe uNited states Volume 13, Number 2 • 2011 U.S. Department of Housing and Urban Development | Office of Policy ...»
Paciorek, Andrew, and Todd Sinai. 2010. “Does Home Owning Smooth the Variability of Future Housing Consumption?” NBER Working Paper 16531. Cambridge, MA: National Bureau of Economic Research.
Poterba, James. 1984 (November). “Tax Subsidies to Owner-Occupied Housing: An Asset Market Approach,” Quarterly Journal of Economics 99 (4): 729–752.
Poterba, James, and Todd Sinai. 2011 (June). “Revenue Costs and Incentive Effects of the Mortgage Interest Deduction for Owner-Occupied Housing,” National Tax Journal 64 (2): 531–564.
———. 2008. “Tax Expenditures for Owner-Occupied Housing: Deductions for Property Taxes
and Mortgage Interest and the Exclusion of Imputed Rental Income,” American Economic Review:
Papers and Proceedings 98 (2): 84–89.
Saiz, Albert. 2010 (August). “The Geographic Determinants of Housing Supply,” Quarterly Journal of Economics 125 (3): 1253–1296.
Shiller, Robert. 2008. “Derivatives Markets for Home Prices,” NBER Working Paper 13962.
Cambridge, MA: National Bureau of Economic Research.
Sinai, Todd. 2009. “Spatial Variation in the Risk of Home Owning.” In Housing Markets and the
Economy, Risk Regulation, and Policy, edited by Edward Glaeser and John Quigley. Danbury, CT:
Lincoln Institute of Land Policy.
Sinai, Todd, and Nicholas Souleles. 2009 (October). “Can Owning a Home Hedge the Risk of Moving?” NBER Working Paper 15462. Cambridge, MA: National Bureau of Economic Research.
———. 2008. “Net Worth and Housing Equity in Retirement.” In Recalibrating Retirement Spending and Saving, edited by John Ameriks and Olivia S. Mitchell. Oxford, United Kingdom: Oxford University Press.
———. 2005. “Owner-Occupied Housing As a Hedge Against Rent Risk,” Quarterly Journal of Economics 120 (2): 763–789.
Sinai, Todd, and Joel Waldfogel. 2005. “Do Low-Income Housing Subsidies Increase the Occupied Housing Stock?” Journal of Public Economics 89 (11–12): 2137–2164.
Voicu, Cristian. 2007. “Optimal Portfolios With Housing Derivatives.” Harvard Business School mimeo. Cambridge, MA: Harvard University.
Abstract Current rental housing assistance programs are not designed to provide a safety net for people whose lives are volatile, nor are they designed to encourage low-income people to live in good locations. These deficiencies can be corrected. The U.S. Department of Housing and Urban Development (HUD) should establish a program of rental insurance—like mortgage insurance, but for renters. Low-income housing assistance formulas should be revised to reward good neighborhood features, and punish bad.
Introduction The rental housing market for lower income Americans could work a lot better than it has been working. Tenants (and the landlords who rent to them) could have more security; fewer people could experience homelessness; and subsidies could do more for the people who receive them, and for their neighbors. This article explains how. The key is thinking about security and externalities in 21st century terms—not in Great Depression terms.
This article analyzes two broad functions that government interventions in the rental housing market can perform: (1) providing a safety net and (2) generating external benefits. Our current policies perform both functions poorly: they move too slowly and too arbitrarily to insure people against most of the risks they face, and are least available precisely when most people need them most; and most of the external benefits they generate are the ones that were important in the first half of the 20th century, not the ones that matter in the first half of the 21st. This article shows how both functions can be performed better.
The perspective in this article is overwhelmingly classical. This article identifies market failures and steps that could correct them. It also identifies failures in current policies. Low-income rental housing markets would have serious problems without significant government involvement, and they can still have serious problems if that involvement is not wise.
Cityscape 127 Cityscape: A Journal of Policy Development and Research • Volume 13, Number 2 • 2011 U.S. Department of Housing and Urban Development • Office of Policy Development and Research O’Flaherty A Better Safety Net How It Works Now Governments perform a useful function when they provide people with valuable insurance that markets would otherwise fail to provide. The major shocks that affect low-income people today are loss of income, loss of health, and loss of relationships (see O’Flaherty, 2009). Families insure against these risks poorly (Bentolila and Ichino, 2008; Dynarski and Gruber, 1997). Governments insure against income shocks very little (especially for adults not accompanied by children and those who have used up Temporary Assistance for Needy Families eligibility), and insurance against health shocks is far from perfect (especially against income losses that sometimes ensue from health shocks). Neither governments nor private insurers offer much meaningful protection against relationship shocks.
All these shocks indirectly affect housing markets. Low-income people primarily spend money on food, clothing, transportation, and housing; rent is generally the largest single bill they receive in a month. Rent is hard to adjust quickly and relatively easy to borrow against. Gas stations and supermarkets do not let customers leave without paying, but landlords cannot evict tenants costlessly when they miss a payment. Thus, landlords become insurers of last resort.
This ability to borrow against rent has consequences. Landlords are not good insurers, and most lack means to spread risk. So they charge too much for this insurance, they screen tenants too strictly, and they under-insure. (To the extent that landlord-tenant law forces landlords to provide more insurance than they want to, landlords charge even more and screen even more strictly.) Even large landlords with wealthy tenants find acting as an insurance company burdensome; evidence of this unwillingness is the apparent success of Insurent®, a private rental insurance company in New York City that specializes in large, high-end buildings. Tenant payment problems are correlated with the business cycle and so confront landlords exactly when those landlords face the most severe cashflow problems from vacancies and tight money. Most tenants, moreover, live in apartments owned by small landlords. Structure size and ownership do not correlate perfectly, but only 18 percent of tenants in 2009 lived in structures of 20 or more units (U.S. Census Bureau, 2011).
Current housing subsidy programs do little to mitigate the risks that low-income households face.
For households who are already subsidized, rents fall when incomes fall or medical expenses rise, and so this group has good insurance. But lags are often long. Current year’s rent is based largely on retrospective income, and reporting takes time. The New York City Housing Authority (NYCHA), for instance, begins gathering retrospective income information 5 months before lease renewal (NYCHA, 2008). And, of course, rents are then fixed for the term of a lease. Thus, by the end of a lease, income many months ago has more impact on current rent than income last month.
NYCHA has “emergency procedures,” but even these are slow. Rent can be reduced during the term of a lease for a tenant who loses a job, but only after 13 weeks.
But most low-income households are not subsidized, and the connection between their misfortunes and any housing assistance they get is approximately zero. Only if a shock lasts the very long time needed to get to the top of the relevant queue does the shock bring housing assistance, and only if the household was foresighted enough to apply early (and does not contain someone with a criminal record).
128 Rental Housing Policy in the United States Rental Housing Assistance for the 21st Century Indeed, households are least likely to get housing assistance when they are most likely to need it.
A household is most likely to need help when a recession hits—either nationally or just locally. But that is when queues are most likely to be longest and help least likely to be forthcoming. (Queues in recessions are likely to be long both because many people need help and because few assisted households are moving up and making way for others. See Ambrose , Hungerford , and Olsen et al. .) Homeless shelters are probably the most effective safety net in the housing market now, and in locations with a rich array of shelters or a right-to-housing, they may perform this function well.
But they are expensive and often demeaning. Like hospital emergency rooms, shelters are necessary but they are not a good substitute for a sound social insurance system.
What To Do Many economists (for example, Olsen, 2008) have argued for a needs-based housing assistance entitlement program, and on many grounds such a program would be a huge improvement over the current set-up. An entitlement program could also offer much better insurance since households in distress would not have to queue for assistance, and funding would automatically expand in recessions so that access would not become harder.
Entitlement programs have several drawbacks, however. Bureaucratically, they would be ill equipped to deal with volatile income, health, and relationships, although perhaps good information technology and careful design could create a nimble and fast system. Current rationed housing assistance programs have small work disincentive effects (Ludwig and Jacob, 2008; Olsen et al., 2005;
Shroder, 2002; Susin, 2005; Tatian and Snow, 2005), but if program entrance were more closely tied to income, these effects might be larger. Current programs probably have significant sharingdisincentive effects (Ellen and O’Flaherty, 2007; Sinai and Waldfogel, 2005), and an entitlement program would have these effects, too, unless it were restructured to be sharing neutral.1 And entitlement programs could be expensive (although Olsen  showed some controversial ways that their costs could be considerably reduced, but does not require sharing-neutrality, which any sensible program would have).
Rental Insurance Basics The most straightforward way to provide insurance is to provide insurance, and it is likely to be the cheapest way too. The federal government already insures mortgages through the U.S. Department of Veterans Affairs (VA) and the Federal Housing Agency (FHA), and mortgage-backed securities through the government-sponsored enterprises (GSEs) (although this last role may not be continuing). Under some circumstances, this insurance costs the government nothing, although, of course, those are not the circumstances that obtain now. Rental insurance markets might be a good way to reduce risks in this market.
Programs have sharing disincentives when they give larger per person subsidies to smaller households. Almost all existing HUD programs have sharing disincentives. Programs are sharing neutral when the size of the subsidy is not affected by the number of adults one shares housing with. See He, O’Flaherty, and Rosenheck (2010) for a discussion of possible reasons for sharing disincentives and evidence that these reasons are not supported empirically.
The basic idea is that a tenant pays a small insurance premium when she leases an apartment. If some predetermined event like a job loss, divorce, or major illness occurs during the term of the lease and makes it impossible for her to keep up with the rent, the insurance kicks in and pays the landlord a fixed amount (say $500) for a fixed number of months (say 6 months). After the tenant stabilizes her life, the government might seek some repayment, but probably not full repayment.
As mentioned, at least one private profit-making company, Insurent in New York, sells rental insurance now, and so the concept is not totally infeasible.
What are the benefits of such insurance? Landlords could relax their screening criteria, require smaller deposits, and charge lower rents; all tenants would benefit. Tenants who encountered setbacks would have breathing room to resolve their problems, or to reduce their consumption deliberately. Social agencies and homelessness prevention services could receive informative advance warnings and possibly target their activities better. Children might not have to move so often, and the length of assistance could even be targeted to the school year. Shelters would see fewer families who were temporarily down on their luck.
This rental insurance program would not insure against rapid rises in the market price of rental housing. The moral hazard problems would be too great. Fortunately, rent shocks do not appear to be a major risk that poor people face or a major direct precursor to homelessness (O’Flaherty, 2009).
Why Not Just Make Other Safety-Net Programs Better?
Is rental insurance just a weak and politically expedient substitute for a more generous social insurance system, one with larger cash payments and more sensitive triggers? No. Two sets of market failures make rental insurance desirable in itself.
The first and probably most important market failures are the external costs of homelessness and of instability for children whose parents do not fully internalize their academic progress. Because of these external costs, the government has an interest in maintaining a tenancy even when a household head would prefer to use the same amount of money for some other purpose, even ex ante.