Academic journal article Real Estate Economics

Overbuilding: A Game-Theoretic Approach

Academic journal article Real Estate Economics

Overbuilding: A Game-Theoretic Approach

Article excerpt

Ko Wang [*]

Yuqing Zhou [**]

The persistence of excess vacancy has long been documented in the literature. We propose that, because vacant land does not produce income, there is a tendency for developers to build whenever they can identify a development opportunity. Since developers have to compete with each other for the development opportunity, in the aggregate, developers will supply more units than the demand in the market. In the face of an oversupply, we show that, under certain circumstances, developers will not lower the rental rate to eliminate vacancy space. Our model also has implications for investment decisions dealing with projects that could take advantage of existing but not fully utilized assets.

Over the past few decades, an extensive literature on the persistence of excess vacancy of property markets has been developed. Initially, researchers concentrated on the volatile patterns of housing starts, which Maisel (1963, p. 539) characterized as having "earned the dubious distinction of ranking among the most cyclically volatile industries." Gradually, research interest has shifted to the question of the persistence of excess vacancy in not only the U.S. housing market, but also in international and commercial property markets. [1] However, it is interesting to note that, after a few decades of research and the introduction of numerous theoretical models and empirical evidence, a recent paper of Bar-Ilan and Strange (1996, p. 619) still concludes that the "overbuilding phenomenon is a puzzle."

In the literature, the explanations for oversupply can be categorized into three groups. The first group attributes the oversupply to agency problems, irrationality, and government policies. They argue that as long as developers are able to raise equity funds from syndications (and/or partnerships) or borrow using non-recourse loans, there is no reason for developers not to continue building, since lenders and equity investors will bear the risk. Indeed, high-income investors in the early 1980's were eager to invest in real estate for its favorable tax advantages. For them, the fundamentals of a real estate project may not be the only consideration. Before the bailout of the savings and loan associations, less profitable (or perhaps insolvent) lenders were encouraged to lend on risky real estate investments on which they could charge high interest rates while allowing the government to bear the final risk. This type of explanation (relying partially on irrational behavior) has some merits, but because real es tate cycles are long-lasting, it is safe to conclude that the agency issues of the 1980's cannot be the only explanation for the observed oversupply.

The second group of arguments relies on a construction lag and forecasting errors to explain oversupply and cycles. Because construction takes time, developers have to project future demand before they start construction. Although it is possible that projection errors could cause the observed cycles in real estate markets, this explanation has at least two weaknesses. First, although it is reasonable to expect an individual developer to make projection errors, it is difficult to understand why all developers in the market would make systematic errors so that the errors did not offset each other. This is particularly true when each developer is able to observe the moves of other developers. When one developer observes that another developer is building, she/he should take this into consideration when estimating the net demand of the market. Second, for commercial properties the typical construction period ranges from 1 to 3 years (residential properties have even shorter construction periods). For large cities such as Los Angeles and Houston, it is difficult to understand why the errors accumulated in 1 to 3 years could cause an oversupply with a magnitude of more than a 20% vacancy rate. These two arguments indicate that the construction lag alone cannot fully explain the oversupply phenomenon. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.