Document View

Private governments such as homeowners associations and condominium cooperatives provide all manner of collective consumption goods, from road maintenance, trash collection, and snow removal to transportation, policing, and medical care. These organizations were practically unheard of in 1960, but today some 54.6 million people in the US live in a various neighborhood associations. Homeowners associations (HOAs) would seem to have passed the market test, but not without controversy. In an effort to determine if HOAs have a positive or negative effect on home values, sales data from five ZIP codes in the Washington, DC, suburban area of Prince William County, VA, from the years 2000-2004 were collected. Local private governments are expanding in number both in the US and around the world. Data indicate that houses in HOAs in Northern Virginia are worth, on average, more than 5% more than similar houses in the same neighborhood but outside of HOAs.

"PRIVATE GOVERNMENTS" SUCH AS HOMEowners associations and condominium cooperatives provide all manner of collective consumption goods, from road maintenance, trash collection, and snow removal to transportation, policing, and medical care, These organizations were practically unheard of in 1960, but today some 54.6 million people in the United States live in various neighborhood associations. That figure continues to rise each year because a majority of new housing units in rapidly growing urban areas are privately governed. Local private governments are also becoming common in Britain, Spain, Brazil, and even in several former and current communist nations, including Russia and China.

Homeowners associations (HOAs) would seem to have passed the market test, but not without controversy. Some residents chafe at restrictions imposed by associations on home improvements, pets, the number and age of residents, and so forth. Others are upset that associations need not operate under the rule of one-person, one-vote, but instead often impose property requirements (one home, one vote) and other limitations on the franchise.

A NET BENEFIT?

Constraint is the essence of all government. The important question is whether constraining others is worth the price of constraining oneself. If that constraint is valuable in the case of HOAs, then homes within HOAs will sell for more than comparable homes outside of HOAs.

Theory does suggest that HOAs increase home value. The rights and responsibilities that come with an HOA affect a home's value, just as do the number of bedrooms and the quality of the local schools. A great house in a community with a poorly designed or badly managed HOA is a great house in a bad neighborhood. Developers, therefore, have an incentive to write constitutions for their HOAs that optimally balance constraints and costs with benefits. Homeowners, as well, have incentives to ensure that their HOA is well-managed and pursues the community's best interests.

Surprisingly little work has been done to test this theory. One of the few empirical attempts to ascertain the value of HOAs appeared in a recent issue of Regulation ("Accountability and Private Governments," Spring 2005). In that article, authors Laura Langbein and Kim Spotswood-Bright ask the right question: "Do homeowners associations affect property values?" Unfortunately, they fail to answer that question.

Enlarge 200% Enlarge 400%

Enlarge 200% Enlarge 400%

Langbein and Spotswood-Bright argue on theoretical grounds that the governing boards of HOAs can become dominated by special interest groups and thus fail to maximize the interests of most owners. While their arguments are well-taken, all governments, public and private, face the same problems. Nor do they acknowledge that developers of HOAs have incentives to minimize special interest politics and can do so in part by designing constitutions with property restrictions, supermajority voting, and quorum rules.

In their empirical work, Langbein and Spotswood-Bright use data from only six homeowners associations and no data on comparable homes or condominiums outside of HOAs. Thus, they are unable to answer the fundamental question.

Instead of testing whether HOAs raise or lower sales prices, the authors look at the effect of association fees on sales prices, and they find that higher fees are associated with lower prices. We believe their model is misspecified and misinterpreted. Without going into those problems in detail, consider that Langbein and Spotswood-Bright find that associations that provide more services have lower sales prices, even after holding fees constant. We should be suspicious of any empirical model that produces results at such variance with economic theory and common sense.

ANOTHER EFFORT

In an effort to determine if HOAs have a positive or negative effect on home values, we collected sales data from five ZIP codes in the Washington, D.C., suburban area of Prince William County, Va., from the years 2000-2004. The ZIP codes we examined were 22192, 20155, 22026, 20112, and 20181, representing the Lake Ridge, Gainesville, Dumfries/Montclair, Manassas, and Nokesville areas, respectively.

Prince William County is an ideal testing ground because it contains a mix of homes in and out of HOAs. Our data include information on house characteristics such as age, lot size, and the number of bathrooms, bedrooms, fireplaces, levels, and style (colonial, contemporary, rambler etc.). We also have information on sales prices and whether the house is within an HOA. Internal house characteristics will help us to control for differences in units inside and outside of HOAs.

However, as any real estate agent will tell you, there is a factor that influences house prices even more than the number of bedrooms, fireplaces, and levels-location, location, location. Our data also allow us to locate homes by subdivision. ZIP code areas are quite small; ZIP code 20112, for example, contains just over 6,000 housing units. But subdivisions are much smaller; within ZIP code 20112 alone we have data on nearly 250 subdivisions. In our empirical work, we will compare houses in HOAs with houses outside of HOAs but within the same subdivision. Because the subdivisions are small, intra-subdivision comparison will control for a wide variety of factors that could influence house prices. Such factors include the quality of local schools, local crime rates, noise levels, travel time to the nearest shopping malls and major highways, neighborhood composition, and so forth. Residents within HOAs, of course, will have access to some property value-enhancing facilities such as swimming pools that residents outside of HOAs do not, but it is precisely the value of those amenities and other HOA-associated rights and responsibilities that we wish to measure. By focusing on houses within the same subdivision, we can be confident that we are measuring effects that are due to HOAs and not to locational advantages not associated with HOAs.

Enlarge 200% Enlarge 400%

Enlarge 200% Enlarge 400%

SECESSION OF THE SUCCESSFUL? In Table 1, we show that the mean sales price for houses within HOAs is actually lower than mean price for houses without HOAs. Of course, this comparison does not control for house characteristics or location, so the difference in value is not a good estimate of the effect of HOAs.

The difference, however, does explode a common myth. In his book Privatopia, Evan McKenzie characterizes HOAs as being "privatization for the few." Similarly, Robert Reich, secretary of labor under President Bill Clinton, refers to the HOA movement as "secession of the successful." Yet, in Northern Virginia and many other growing areas of the country, neither of those characterizations is correct. Most houses sold during this time period were sold within HOAs. Thus, although HOAs do represent privatization, it is a privatization for the many, not the few. Similarly, HOAs can hardly be said to represent "secession of the successful" when homes within HOAs are often less expensive than relevant alternatives.

HOA VALUE Houses within HOAs may differ from houses without HOAs in various valuable characteristics, e.g., the average number of bedrooms. Thus, in Table 2, we compare houses within and outside of HOAs after controlling for a variety of housing characteristics. This comparison will give us a more precise estimate of how much HOAs contribute to house value.

In the first column of Table 2, we control for internal housing characteristics and the year of sale. We log the sales price so that the coefficients can be interpreted as the percentage change in the sales price caused by a change in the specified variable. We find, for example, that houses in HOAs are 6.1 percent more valuable than similar houses located outside of HOAs. In the second and third columns, we introduce further controls for house style and location effects (ZIP code and subdivision). The coefficient on HOA is robust across the columns. In the third column, which contains the most controls, we find that HOAs raise house value by 5.4 percent.

The typical house within an HOA during the time period we studied sold for $255,000. We thus estimate that membership in an HOA increased house value by nearly $14,000. In an efficient market, the annual fee for HOAs would be capitalized in the sales price, so the $14,000 increase in home value is the net value of living within an HOA, i.e., the increase in value after all the benefits and costs of HOAs are summed.

Examining the other coefficients, we can express the value of an HOA in another way. Holding all else equal, we estimate that consumers value a three-bedroom home in an HOA about as much as a four-bedroom home without an HOA. We suspect that most people consider this a significant increase in value, which helps to explain why HOAs have become so popular.

The increase in house value caused by HOAs is especially remarkable when one considers that residents of HOAs pay twice for many local services-once in taxes and then again in HOA fees. It is also worth noting that we find that HOAs increase value by at least 5-6 percent. Competition tends to push prices down to marginal costs, so as the privatization of local government continues, the estimated price-premium may fall even as consumer benefits from HOAs increase.

CONCLUSION

Local private governments are expanding in number both in the United States and around the world. Local private governments are also expanding in scope, with many offering private security and a few even offering services such as day care, schools, and courts to arbitrate homeowner disputes.

Our data indicate that houses in HOAs in Northern Virginia are worth, on average, more than 5 percent more than similar houses in the same neighborhood but outside of HOAs. Given those large advantages, it is not surprising that HOAs are growing rapidly.

We have shown that HOAs increase house value, but many questions remain unanswered. Do HOAs increase house value because they offer better-quality services than local government? Or do HOAs increase house value because they offer services that local governments cannot (e.g., more restrictive zoning) or do not (e.g., greater security)? How does the governance structure of HOAs impact house value? Can we find, for example, what people are willing to pay for term limits or supermajority rules? Studying HOAs may also help us to identify how and why local governments are failing to maximize returns for their residents. Ideally, the knowledge provided by private governments could be used to improve services for everyone.