Wednesday, October 24, 2012

When Rent Control Ended in Cambridge, Mass.

Every intro class teaches about price ceilings, and I suspect that 99% of them use rent control laws as an example. Of course, the standard lesson from a supply-and-demand diagram is that price ceilings lead to a situation where the quantity demanded exceeds the quantity supplied, and so while the price of rent-controlled apartments is lower, good luck in finding a vacancy!

The slightly more sophisticated insight is what I call in my own intro textbook the problem of "many margins for action." (Of course, if you are teaching an intro econ class, I encourage you to take a look at my Principles of Economics textbook, a high quality and lower-cost alternative to the big publishers, available here.)  Landlords who face rent control legislation can skimp on maintenance, or hunt for ways to force the renter to bear additional fees or costs. If a large number of landlords act in this way, the feeling of the neighborhood and property values for homes that are not rentals may be affected, too.

Cambridge, Massachusetts, has a rent control law in place from 1970 to 1994. It was ended by a statewide vote that barely squeaked out a 51%-49% majority--and ended despite the fact that Cambridge residents favored the continuation of the law by a 60%-40% majority. The law placed limits on rents for all rental properties in Cambridge built in 1969 or earlier. In "Housing Market Spillovers: Evidence from the End of Rent Control in Cambridge, Massachusetts," David H. Autor, Christopher J. Palmer, and Parag A. Pathak look at what happened. (The paper is published as NBER Working Paper 18125. These working papers are not freely available on-line, but many in academia will have access through institutional memberships. Full disclosure: David Autor is editor of my own Journal of Economic Perspectives, and thus my boss.) Autor, Palmer, and Pathak have data on rents and prices in both controlled rental buildings, uncontrolled rental buildings, and owner-occupied housing. They can also make comparisons to neighboring suburbs that did not have rent controls in place. Here are a few of their more striking findings:

-- The rent-controlled buildings in Cambridge, Mass., typically had rents 25%-40% below the level of uncontrolled rental buildings nearby. However, the maintenance of rent-controlled building was often subpar, with a higher incidence of issues like holes in walls or floors, chipped or peeling paint, loose railings, and the like. More broadly, owners of rent-controlled properties had no incentive to do any major fix-ups or renovations, because they would be unable to recoup the costs.

-- Rent control laws are still easy to find, if not exactly widespread, in the United States. For example, "New York City’s system of rent regulation affects at least one million apartments, while cities such as San Francisco, Los Angeles, Washington DC, and many towns in California and New Jersey have various forms of rent regulation."

-- Not surprisingly, the end of rent control in 1995 meant that prices of the buildings that had formerly been rent-controlled rose. "Our statistical analysis also indicates that rent controlled properties were valued at a discount of about 50 percent relative to never-controlled properties with comparable characteristics in the same neighborhoods during the rent control era, and that the assessed values of these properties increased by approximately 18 to 25 percent after rent control ended."

--  More surprising, it turns out that the end of rent control raised the value of all the non-controlled properties in Cambridge, too. Properties that were in a neighborhood with a higher percentage of rent-controlled properties increased in value by more than those in neighborhoods with a lower percentage of rent-controlled properties. Indeed, when rent control ended, the gains to owners of  uncontrolled properties were greater in total than the gains to the owners of rent-controlled properties. "The economic magnitude of the effect of rent control removal on the value of Cambridge’s housing stock is $1.8 billion. We calculate that positive spillovers from decontrol added $1.0 billion to the value of the never-controlled housing stock in Cambridge, equal to 10 percent of its total value and one-sixth of its appreciation between 1994 and 2004. Notably, direct effects on decontrolled properties are smaller than the spillovers. We estimate that rent control removal raised the value of decontrolled properties by $770 million, which is 25 percent less than the spillover effect."

Taking all of this together, it seems to me like the way to think about rent control--at least in the form that it was enacted in Cambridge, Mass.-- is that it creates a situation of low-quality and poorly-maintained housing stock, which then rents for less than uncontrolled properties. If the goal of public policy is to create lower-quality and more affordable housing, there are other ways to accomplish that goal. For example, zoning laws could require that rental complexes include a mixture of regular and small-sized rental apartments, so that the small-sized (and thus "lower quality") apartments would rent for less. Or those with lower incomes could just receive housing vouchers.

But when rent control is enacted in a way that leads to degradation of a substantial portion of the housing stock, the costs are not just carried by landlords of those rent-controlled apartments. In fact, a majority of the costs may be as a result of spillover effects to real estate that isn't rent-controlled. When a substantial proportion of the houses in a neighborhood are not well-maintained, everyone's housing prices will suffer.