«reNtal HousiNg Policy iN tHe uNited states Volume 13, Number 2 • 2011 U.S. Department of Housing and Urban Development | Office of Policy ...»
to structure-specific maintenance or excessive use. Homes require a fair amount of care and they suffer when abused. Tenants have only a limited incentive to invest in their homes, and, as has been empirically well documented, rental homes depreciate more swiftly than owner-occupied housing (Shilling, Sirmans, and Dombrow, 1991). As Henderson and Ioannides (1989) wrote, assigning ownership to the occupier provides more incentives for taking better care of the housing units.
The model in the appendix explores this intuition. At its heart, the model follows Grossman and Hart (1986) and examines the connection between ownership-structure and some noncontractible investment, such as cleaning the gutters in a house or fighting mold. If the unit is owner occupied, the owner undertakes such investments; if the unit is rented, either the landlord or the tenant undertakes the investments.
Owners are more likely than renters to undertake maintenance operations for two major reasons.
First, the owner captures the long-term price effects of the maintenance, which the renter does not.
Second, the renter will actually suffer if better maintenance leads the landlord to raise rents. Both forces suggest that owner-occupiers will take better care of homes than renters.
It is less obvious, however, that owner-occupiers will take worse care of their homes than landlords. After all, the landlords have the same long-run price incentives and they benefit from the higher rent that they can charge as a result of maintenance.
The model in the appendix emphasizes two reasons why owner-occupiers may do more than landlords to invest in apartments and two reasons why landlords may invest more than owneroccupiers. One reason compelling landlords to invest less is that long-term rent contracts at a fixed rate reduce the connection between maintenance and rent and dilute the landlords’ incentive to invest. Frankena (1975) made this point in the rent control literature. Even if the rent is flexible, if it is decided by ex post bargaining, the landlords will under-invest unless they have complete bargaining power.
A second force leading landlords to invest less money in their properties is that owners may use their own time to maintain the home, while landlords often hire outside help and must pay taxes on the wages earned by that help. Assume that an investment of 50 person-hours will increase the resale value of the unit by $1,000 dollars (ignoring discounting). Assume further that the market wage of both the owner-occupier and a maintenance worker hired to care for the unit equals $25 per hour and that the tax rate is 30 percent, so that the average tax wage rate is $17.50. The owner-occupier gains $1,000 in extra housing value (the capital gains of the home are untaxed) at a cost of $875. The landlord gains the same $1,000 in housing value but must pay $1,250. As a result of this price discrepancy, the owner-occupier is more likely to engage in the improvement.
In fact, it is quite possible that the owner-occupier will engage in too much maintenance, at least relative to working outside the home, because wages are taxed but the benefits from home maintenance are not. Of course, if externalities from home maintenance exist, then private maintenance will be too low without the tax wedge advantage.
An added advantage from owner-occupier maintenance is that the owner-occupier lives in the home, which means the person providing the maintenance has more information about the state of the home and fewer travel costs. Some of the advantages of owner-occupancy versus absent landlord vanish when the landlord also lives in the building and provides maintenance personally.
This situation is not unusual in lower density dwellings, such as Boston’s triple decker (three-story) apartment buildings, where the owner often lives in the home and does not need to travel. A tax wedge still exists, however, because the higher rents that come from landlord maintenance get taxed, while the quality-of-life benefits that the owner-occupier receives from maintenance do not.
Other factors challenge the tenant’s advantage in providing maintenance and lead to landlord maintenance. The landlord may either have or be able to hire specialized human capital in home maintenance. If external maintenance providers are significantly more efficient than the owner-occupier is, then hiring such providers may eat into the tax wedge advantage of personal maintenance. If external provision is much better, then the owner-occupier may hire external help as well, which eliminates the tax wedge altogether. As long as the owner-occupier can hire the same outside talent at the same wage that the landlord pays, this scenario will not lead to less maintenance by the owner-occupier. If the landlord has more knowledge of the outside market for maintenance talent or if there are returns to scale in hiring such talent, then the landlord may indeed have a maintenance edge. A second force that may induce more maintenance among landlords is regulations that may require a certain level of maintenance.
As Henderson and Ioannides (1989) argued, maintenance-related benefits help explain homeownership levels and in what situations to expect more homeownership. The model in the appendix unsurprisingly predicts that higher tax levels will lead to more homeownership, because the value of the tax wedge is higher and because people who are particularly unskilled at home maintenance, relative to landlords, will be more likely to rent. Assuming that landlords do provide less maintenance because rents do not adjust closely to investments, then more ownership of structures will occur in situations in which maintenance is more valuable.
The model also considers multifamily units, in which only two types of maintenance exist. Some maintenance investments are specific to the unit, but others are specific to the building, such as maintaining the roof or the boiler. The basic logic of such interventions is similar to the algebra already discussed, except insofar as the maintenance action creates externalities to neighbors either because the actual maintenance creates negative externalities while it is going on, or because unitspecific maintenance makes the block of units more attractive as a whole. Although it is possible for individual owner-occupiers in a multifamily building to undertake such actions on their own, it is more natural for them to coordinate in some fashion with a condominium or cooperative association.
Once an association makes a maintenance decision, however, a great deal can go wrong. Coordination costs, especially those involving the time of owners, may be substantial. Delegating out management can be very expensive as well. The model assumes that these coordination costs scale up with the size of the building.
That assumption leads to the prediction that owner-occupancy will be more common for smaller structures than for bigger structures. For smaller structures, the tax wedge and the fact that owneroccupiers internalize all the benefits from maintenance dominate and encourage ownership. For larger structures, the costs of coordinating activities dominate and as a result, it is more efficient to have a single landlord, even one who is under-investing in maintenance. This logic suggests that ownership should be less common in bigger buildings.
14 Rental Housing Policy in the United States Rethinking the Federal Bias Toward Homeownership Finally, the model turns to social capital investments, which are defined as investments that affect the quality of the neighborhood but not the structure. These investments require time but not money and only the residents themselves can make them. These investments may also create externalities. In general, we should expect to see more investments from homeowners than from renters, both because homeowners internalize the benefits of future price increases and they do not lose from increases in rents. This tendency for homeowners to make social capital investments provides one justification for the correlations between homeownership and social capital, as found by DiPasquale and Glaeser (1999) and others.
This connection is particularly relevant if investments in social capital create positive externalities.
In that case, this connection becomes the primary justification of national pro-homeownership policies. I now assume that the social benefits from these investments are proportional to the private benefits and ask what conditions might increase the size of the homeownership externality.
One particular question is whether structure type or individual characteristics will connect with these investments. For example, it is natural to assume that both the benefits and costs of investment are a function of the size of the building. It might be easier to connect with others in large dense structures. Alternatively, the benefits of social capital might be lower or higher in denser areas. Social capital might be less valuable if people are more connected to the streets in shorter buildings or it might be more valuable if the community associations are needed to make blocks safer.
The model shows that the relationship between building size and the cost of developing social capital is critical in determining whether the effect of homeownership on social capital increases with building size. If bigger buildings reduce the cost of interactions, then the social capital-related benefits of ownership, and any externality from that social capital, will actually be larger in owneroccupied buildings.
Another way in which individual characteristics or structure type might affect the decision to subsidize social capital by subsidizing homeownership is to influence the size of the externality relative to private benefits. Any variable that increases the size of the external benefits from these investments without affecting the private decision to invest will naturally make public subsidy more appealing. The question then becomes empirical: What do we know about the differential effect of homeownership on forms of social capital investment and what do we know about the size of the externalities?
The model appendix also suggests that the link between social capital and homeownership can be reversed if large, dense dwellings act to lower the costs of social interactions sufficiently. If people who live near one another in dense neighborhoods find it much easier to interact than do people who are spread out at long distances, then the subsidization of homeownership can perversely lower social capital, even though homeownership—holding structure constant—increases investment in social capital. We will discuss the evidence on this possibility in the next section.
Evidence on Homeownership, Structure, and Policies In this more empirical section, I explore five different topics and discuss the policy issues that surround them. The first subsection recalls the basic facts about homeownership and structure type. The second subsection looks at the connections among structure type, ownership, cities, and density. Subsidizing homeownership does seem to mean subsidizing suburbia. The third section focuses on the pollution and congestion externalities associated with subsidizing ownership and single-family detached homes. The fourth section focuses on the economic effect of subsidizing people to live in metropolitan areas where single-family detached homes are more common. The fifth section returns to the social consequences of these policies.
Homeownership and Structure Type Exhibit 1 shows the breakdown of ownership by structure type using the 1970 and 2000 Census Individual Public Use Microsample. The results are as follows: first column, the entire population in 2000; second column, the top quintile of the population by income in 2000; third column, the lower quintile of the population by income in 2000; and fourth column, 1970. I have excluded mobile homes and other nonpermanent structures.
The first row gives results for single-family detached housing. In the United States as a whole in 2000,
86.8 percent of these units were owner-occupied and 13.2 percent were rented in 2000. Among the richest quintile of Americans, the ownership rate in this structure type was an astonishing
95.6 percent. Even among the poorest quintile, the ownership rate was 73.3 percent. In 1970, the ownership rate for single-family detached houses was 81.6 percent. The rental of a single-family detached unit is relatively rare and concentrated among poorer Americans.
Single-family attached homes are significantly more likely to be rented. In this group, the total homeownership rate falls to 65 percent. The homeownership rate is 84 percent among the richest quintile but only 46 percent among the poorest quintile. In 1970, the ownership rate for attached single-family homes was 58 percent. This is clearly a middle category in which ownership and renting both are attractive. Two-family homes occupy the same place, resulting in a substantial mixture between owner-occupied and rental units, and renting is, as always, higher among poorer Americans.
When we consider three- or four-family units, however, the rental rate skyrockets. In the United States as a whole, the ownership rate in three- or four-unit structures is only 13.7 percent. This figure has changed little since 1970, when the rate was 14.1 percent. Even among the richest quintile of Americans, only 36 percent own if they live in three- or four-unit structures.
Among even larger structures, the ownership rate is typically even lower. In structures with 5 to 50 units, the rental rate is more than 88 percent. This figure is lower than it was in 1970, which perhaps reflects the fact that the home mortgage interest deduction provides a large incentive to own expensive condominiums.
The tight connection between structure type and ownership also holds across places. Exhibit 2 shows the -.68 correlation between the share of households that are single-family detached and the share of households that are owner occupied. The large differences in structure type across Exhibit 2 Correlation Between Share of Owner-Occupied and Single-Family Detached Housing Units, by MSA