Crime and property values.
It is an unfortunate reality for the residents of Memphis that high on the list of local claims to fame--along with Blues, and Elvis--is crime. Memphis ranks 13th in crime per capita on a list of cities with populations of 75,000 and greater. Narrow that list to cities with populations of half a million or more residents, and Memphis jumps to 4th place. And, among Metropolitan Statistical Areas, the eight-county Memphis, TN-MS-AR region is second only to the Detroit area in terms of crime--hardly a silver medal any of us would like to hang from our mantles.
Despite recent reductions in crime--both violent and property crimes were down in January and February 2007 compared to year-earlier numbers, a trend that appears to have begun in late 2006--the city has years of rising crime and a declining population to overcome. Crime inflicts many costs on a city's residents, including feelings of a lack of security and safety, the monetary value of property loss from criminal acts, lost earnings from injury or death, and the impact of crime on property values. It is this last issue that this article addresses.
While no specific research has been done to date on the impact of crime on property values in the Memphis area, there are numerous studies in the economics literature that examine the question. Not surprisingly and nearly unanimously, the literature concludes that crime tends to depress property values, although estimated magnitudes of incidence vary from study to study. As will be discussed later, the implications of these magnitudes are of particular importance from a public policy perspective, vis-a-vis the elasticity of property values with respect to crime.
The theoretical effect of crime on property values is fairly straightforward. Households experience disutility from crime near their homes and as a result would be expected to reduce their bid prices for housing in areas where crime is more prevalent. This theoretical prediction is largely borne out in the existing literature on crime and property values, although the degree to which higher crime affects property values is a matter of debate. Much of the difference in estimates likely depends on estimation techniques, the measurement of crime as well as property values, and the size of the sample studied.
There are two general approaches in the literature to measuring the impact of crime on property values. The first (and earliest) approach regards the value of a house as a function of a flow of services, such as shelter, safety, and so on (Muth, 1969). Crime is regarded here as a disamenity. The second approach treats the value as a function of characteristics of the house: square feet of living space, the number of bathrooms, size of the lot, and so on--the hedonic approach. Frequently included in studies that employ the latter approach are neighborhood characteristics, such as age, income, and racial composition, rates of home-ownership, and job profiles, among others.
The earliest attempt at isolating the effect of crime on property values was made by Thaler (1978). Like later efforts, Thaler used a hedonic price model and regressed the per-acre price of land on a host of neighborhood-specific variables. His conclusion was that a one-standard-deviation increase in the crime rate (property crimes/population) pushed down the per-acre land price by $3,847 and the price of a house by approximately $430. Hellman and Naroff (1979) found similar results, suggesting that a 1.0 percent decrease in the crime rate increased residential property value by $2,300,000.
Rizzo (1979), using crime, property, and rent data from Chicago, found that a 10.0 percent decrease in crime tends to increase property values by between 2.0 and 4.5 percent. Like Hellman and Naroff, Rizzo used an aggregate measure of property value for community areas in Chicago but was the first to employ a two-stage least-squares model, where the crime rate was made endogenous (all previous research had taken crime exogenous). Buck et al. (1993) also found aggregate property values suffered as crime rates increased, but increases in tax rates would ultimately pay for themselves via higher property values.
Three remaining studies, Bowes and Ihlanfeldt (2001), Gibbons (2004), and Linden and Rockoff (2006), are similar in their unit of observation: the sale price of a single-family house; the empirical strategy pursued: hedonics; and the results generated. All three studies found a generally negative relationship property prices and crime. Linden and Rockoff (2006) used the residential location of a convicted criminal as their dependent variable, but found that the proximity of a convicted criminal and property values are negatively related and significantly so.
One important issue to consider in evaluating the literature on crime and property values is the measure of crime used in each study. To date, most research has used some index of criminal acts per capita, such as violent or property crimes. While this approach is the most common, it does ignore the varying impact that crimes of different types have on quality of life, property, earnings, and so on. Only one study, Lynch and Rasmussen (2001),has employed a measure of crime other than per capita incidence. Using a wide range of property-specific and neighborhood variables from home sales in Jacksonville, Florida, the authors estimated the marginal impact on house prices of the crime rate for violent and property crimes but also the "cost of crime;' which is assumed to be higher for more violent crimes, such as murder and rape, than for property crimes, such as burglary and vandalism. This approach has a great intuitive appeal as well as a firm grounding in research.
While there is no single "correct" way to measure the cost of a crime, the authors used estimates from another study, et al. (1995), and arrived at some interesting conclusions about the impact of crime on property values. First, when compared to a per-capita approach to quantifying crime, using the cost of crime generates much lower elasticities of housing prices with respect to crime, implying that the cost of crime has a much lower impact on property values than suggested in earlier research. (Not surprisingly, violent crime a more negative impact on prices than does property crime.) Second, property values are likely to be lower, all else constant, in high crime areas. The authors estimated that all else held constant, the price of an average house in a high crime area would be discounted from $94,000 to $57,000, a declination of nearly 40.0 percent. Further, certain housing characteristics, such as a fence or a large lot, are more highly valued in high crime areas, presumably because they provide a greater barrier from the street. The policy implication of this second conclusion will be discussed later.
The theoretical effect of property values on crime is ambiguous. An area with higher property values could be a more attractive target for criminals, holding transportation expenses and probability of detection constant, and thus could experience a higher crime rate. Alternately, higher property values imply a greater opportunity cost of engaging in crime (i.e., lost wages), which could reduce the incidence of crime in a high property value area. While many of the studies that examined the effect of crime on property values do control for the likelihood of criminal behavior in a given area, (1) there is no little research on which comes first, low property values or crime. However, Bowes (2007) addresses a similar issue: the relationship between retail development and crime. The author examined whether crime is attracted to retail development or discourages such development. When considering all types of crime, the author found that both hypotheses are supported. Retail development both attracts crime and is adversely affected by crime. But when separating crime into categories, the author found that violent crime reduces retail development, where property crime is attracted to it. This supports the notion advanced in Lynch and Rasmussen (2001) that in order to estimate the impact of crime, one must consider not just incidence, but cost as well. Different categories of crime can have different effects on the community at large.
Finally, there are the public policy implications of the literature and property values. Many authors provided estimates of how property values, and in some cases tax revenues. would be impact by a given reduction in crime, as mentioned above. One would expect that higher property taxes would decrease property values and that a reduction in crime would increase property values. The question of which effect dominates is unresolved in the literature. Most of the studies cited here found a relatively inelastic response of housing prices to crime, although Naroff et al. (1980) found a much more elastic responses. Their results indicated that a 1.0 percent decrease in crime rates increased housing values by almost 1.7 percent. This result pointed to the conclusion that a city can realize a net increase in tax revenues from raising tax rates to pay for more police and a reduction in crime, a striking policy implication.
Further, there is the issue of how best to allocate tax revenues in an effort to prevent a loss in property values. Because there is such a marked discount on housing located in crime areas, Lynch and Rasmussen (2001) suggest that tax dollars would best be spent in preventing neighborhoods from "crossing the high crime threshold" and reducing the size of the local tax base.
It is this last point that is perhaps most prescient for Memphis. While many neighborhoods in Memphis have already gone over the edge in terms of crime and other disamenities, the research above suggests that intensively targeting tax dollars in such neighborhoods might be the best strategy in combating crime. The map below is taken from the Memphis Crime Stories website and shows the incidence of crime in the Orange Mound neighborhood on March 28, 2007. Although there is some crime dispersed across the general area, a great amount of crime is concentrated in a very small area. And, while one crime map of one neighborhood on one day does not good policy make, it is clear that crime affects much more than just property values in our city.
Bowes, David R. "A Two-Stage Model of the Simultaneous Relationship Between Retail Development and Crime." Economic Development Quarterly, 21, 1, 79-90, February 2007.
Bowes, David R., and Keith R. Ihlanfeldt. "Identifying the Impacts of Rail Transit Stations on Residential Property Values." Journal of Urban Economics, 50, 1-25, July 2001.
Buck, Andrew J., Simon Hakim, and Uriel Spiegel. "Endogenous Crime Victimization, Taxes, and Property Values." Social Science Quarterly, 74, 2, 334-48, June 1993.
Cohen, Mark A., Ted R. Miller, and Brian Wiersema. Crime in the United States: Victim Costs and Consequences. Final Report to the National Institute of Justice, 1995.
Gibbons, Steve. "The Costs of Urban Property Crime." Economic Journal, 114, 499,441-63, November 2004.
Hellman, Daryl A., and Joel L. Naroff. "The Impact of Crime on Urban Residential Property Values?' Urban Studies, 16, 105-12, February 1979.
Linden, Leigh L. and Jonah E. Rockoff, "There Goes the Neighborhood? Estimates of the Impact of Crime Risk on Property Values from Megan's Laws," Working Paper 12253, National Bureau of Economic Research, May 2006.
Lynch, Allen K., and David W. Rasmussen. "Measuring the Impact of Crime on House Price?' Applied Economics, 33, 1981-1989, December 2001.
Muth, Richard E, The Spatial Pattern of Urban Residential Land Use, The University of Chicago Press, Chicago and London, 1969.
Naroff, Joel L., and Daryl A. Hellman. "Estimates of the Impact of Crime on Property Values." Growth and Change, 22, 24-30, October 1980.
Rizzo, Mario J. "The Effect of Crime on Residential Rents and Property Values." American Economist, 23, 16-21, Spring 1979.
Thaler, Richard. "A Note on the Value of Crime Control: Evidence from the Property Market." Journal of Urban Economics, 5, 137-45, January 1978.
(1) This is typically accomplished by "endogenizing" crime via a two-stage regression model, where the rate of crime is estimated from various area-specific variables and then entered as an explanatory variable in the property value equation.
Douglas A. Campbell, Economics Instructor, joined the University of Memphis Department of Economics faculty in August 2006. Previously, he was a visiting Assistant Professor in the Department of Economics in the Andrew Young School of Policy Studies at Georgia State University, where he obtained his Ph.D. in Economics in 2004. His doctoral research was on the economic incidence of development impact fees, an area in which he is still active. His current working papers focus on inter-jurisdictional impact-fee competition, and impact fees and housing market growth. Lately, he has branched out into research on crime and housing markets.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||influence of crime on real estate industry|
|Author:||Campbell, Douglas A.|
|Date:||Jun 22, 2007|
|Previous Article:||Annexation of Cordova, Tennessee, by Memphis: sale price trends from repeat-sale indexes.|
|Next Article:||The mortgage default crisis: the American dream becomes a nightmare.|