Microeconomic Estimates of Housing Depreciation

Abstract

Measuring economic depreciation in residential real estate is important for deepening our understanding of housing market behavior and for assessing a variety of public policies. Independent estimates of depreciation rates for many metropolitan areas provide a rich data base for analyzing the interaction between structure depreciation and, for example, suburbanization. Also, public policies such as rent control and tax incentives for construction could be better evaluated and designed with improved estimates of depreciation. For example, in an attempt stimulate construction in 1981 Congress cut allowable depreciation for tax purposes from something like 30 years 15 years. Then in 1986 in response apparent over-building, Congress raised the depreciation rate back 27.5 years. Better estimates of depreciation rates could have contributed better targeting of incentives. A literature survey indicates significant variation in estimates of depreciation rates for residential real estate. The variation in reported results seems be due differences in the definition of depreciation employed, in the models and data used estimate depreciation rates and, in some cases, the condition of the real estate market examined at the time of the study. A primary objective of this paper is establish that rates of economic depreciation vary spatially. By using the same hedonic model and representative data from each of 59 metropolitan areas, we are able eliminate differences in models and data as the only source of variation in estimated depreciation rates and attribute some of the observed variation differences in housing markets. This paper reports estimates of metropolitan area specific rates of economic depreciation for residential real estate obtained using the hedonic pricing methodology. Separate h donic equations were estimated for owneroc upied and renter-occupied properties for each of 59 metropolitan areas. With the exception of variables that identified particular locations within the urban area, the same hedonic equation was used estimate tenure spec f c depreciation rates. That is, the same hedonic specification was used for each of the 59 metropolitan areas estimate depreciation rates for owner-occupied and for renteroccupied housing. We define economic depreciation to be the decline in asset price (or shadow price) due aging (Hulten and Wykoff 1981, 85). Our hedonic equation for renter-occupied housing therefore examines how the current market price of the flow of (rental) housing services varies with dwelling age. Rents might be expected change as the flow of housing services decreases as the unit deteriorates. The market value of owner-occupied residential real estate, on the other hand, reflects the present value of future housing services as well as current services. Hence, changes in any of the determinants of present value will have a significant impact on the current market value of owner-occupied housing.