From 2007 to 2009, the U.S. underwent one of the worst recessions in its history, a recession triggered by an unprecedented, international financial crisis that resulted from institutional portfolio concentration in securities backed by home mortgages, and the collapse of that securities market. The period saw a wave of defaults and foreclosures that spared almost no communities in metropolitan areas throughout the country (Bajaj and Story, 2008). Loan defaults and foreclosures, which had tended to be concentrated in lower-income and minority neighborhoods, spread to new and diverse communities, including higher income communities, resulting in broad-based, deep declines in home prices.
The effect of foreclosures and distressed properties remains an issue of much concern to policymakers, community development practitioners, and consumer advocates, and the focus of much research (e.g., Campbell, Giglio, and Pathak, 2011; Baumer, Wolff, and Arnio, 2012; Ergungor and Nelson, 2012; Hartley, 2014; Seo and Mikelbank, 2017). Foreclosures and distressed property sales have enduring repercussions on local housing market values, as well as the pace of market recovery in different local communities. As the overall housing market emerged from the 2008 crisis, many neighborhoods within broader, recovering geographies did not return to market vitality due to (in part) varying geographic concentrations of foreclosures and their disparate effects.
This article reviews the research to date and provides an analysis of local housing market price differentials. We examine the contribution of various factors affecting housing prices, including the structural features of the homes and the characteristics of the neighborhoods in which the properties are located. We also pay particular attention to the effect of distressed property sales on overall home prices in the neighborhoods, post 2007. An important contribution of our exercise is that we are able to derive local housing market price indices. Understanding and measuring house price trends across neighborhoods in various cities has been one of the most challenging, but important topics in housing research recently. Being able to measure house price trends can help housing market stakeholders and policymakers understand which neighborhoods are improving and which continue to struggle, and make strategic decisions about policy development and implementation.
In summary, we find that, well past the height of the 2008 housing financial crisis, in 2017 foreclosures and distressed property sales remained strong contributors of house price depreciation in Cook County. These negative effects are uneven across areas in the county, with places in the city of Chicago (especially low-income areas), seemingly disproportionately affected by housing distress compared to others in suburban Cook County, after controlling for many other characteristics of properties and neighborhoods.
We conduct an examination of the relationship between home value and recovery and income level of the neighborhoods and find that the correlation between income and housing price has increased over time suggesting that the income of the neighborhood has become more important in impacting home values. Home values of neighborhoods in the lower income distribution in the county are still 45 percent or lower than their previous peak. The lack of price recovery in these markets means that homeowners have little or no equity on their homes and many homeowners remain under water. This bifurcation, given high housing costs and increases in much of the metro area, raises concerns about economic mobility of residents, and continued disinvestment in places with relatively low home values.
The remainder of the paper is organized as follows: the next section provides a review of the literature on housing market price. The third section describes the Cook County housing market considered in this case study, and presents the data which we use to analyze local housing market prices. Section 4 discusses the results of our estimates of the effects of various factors on housing price, and examines the trends in price indices in the different submarkets of Cook County in relation to income levels of the areas. Section 5 concludes with a summary and note regarding the implications of our findings.
Factors affecting housing prices
Measuring Housing Prices
Housing has features that make it different than other assets from a valuation standpoint. Unlike other assets, few residences are exactly the same, and accordingly, housing involves a unique pricing structure which is determined by not only its own characteristics (e.g., features of the home, its size, number of bedrooms and baths, and overall interior quality), but also by other contextual factors. These include neighborhood characteristics, such as the quality of local schools; and the location in relationship to other centers of interest or value, such as a central business district (CBD) or other centers of employment. In addition, business cycle effects and macroeconomic conditions, such as interest rates, employment and economic growth, and other factors which determine demand and supply of housing, also influence prices.
In practice, various methods are employed to determine house values, the median sales price being one. Median sales price is the midpoint (value) of all sales taking place in a given geography for a given period of time. The primary strengths of this method are: that data on sales activities and prices are often easily available through local deed transfer recordings or multiple listing services; and finding the median is a straightforward calculation. For these reasons, trends in median sales prices are often used by local realtor groups or the media to discuss area house values and trends. The main limitation of this method is that there is no way to control for changes in the underlying composition of properties selling at any two points in time. This constraint has the potential to create “apples to oranges” price comparisons if there are large differences in the mix of the size and quality of properties selling at two points in time, and can be particularly impactful when sample sizes are small, as could be the case in small geographic areas.
Another method, the repeat sales indices, takes the sales activity on a property at two points in time and measures the change in value over that period. The change is weighted based on the length of time between the two sales, and the average change in sales prices for all properties in a sample are calculated and indexed to an earlier point in time, often the first quarter of 2000. The repeat sales index is an improvement over median sales price in many ways. By only tracking price changes for properties that sell multiple times, a repeat sales index is better able to ensure that the price change being measured is for properties with similar characteristics. Repeat sales indices also have limitations, however. Most importantly, because the sample uses only properties that sell at least twice, it is often difficult to get a large enough sample of property sales for a given period to measure price trends in a small geographic area. Case-Shiller is the best known repeat sales index, and it tracks price trends nationally for a group of large metropolitan areas.
The third method (which we will apply for this study) is derived from a hedonic price model, and is estimated using a regression technique. This method combines data on sales price with property and location characteristics, and controls for factors that might affect sales price. A hedonic model reveals how much influence individual factors have on sale prices, and, by isolating the effects of those variables, allows for the development of an index tracking price changes over a period of time on properties with similar characteristics. Hedonic price models are an improvement over repeat sales technique because they include data on far more sales in a given period for a location, as opposed to just those with previous sales, creating a larger sample in smaller geographic areas, while still controlling for characteristics and location of properties being sold in a given period.
Kain, Quigley (1970), and Rosen (1974) in their seminal works developed a hedonic model to predict house prices. They included structural characteristics of the housing units, neighborhood characteristics, referred to also as the social and natural environment,1 and distance to the central business district. Many subsequent researchers have used distance to the city center as their measure of location (Heikkila et al., 1989).2 The notion of distance to city center in housing market research is drawn from the classical urban economic models, which conceptualize land value, as a negative exponential function of distance (or travel time) from a central business district, assuming a monocentric city (Alonso, 1964; Mills, 1967; Muth, 1969).
Shifts in employment patterns of contemporary urban areas have, however, added complexity to the relationship between housing price, location, and distance. And the hypothesis (in monocentric cities) that prices generally decrease as distance from the CBD increases has tended not to hold (Bender and Hwang, 1985; Coulson, 1991).3 Researchers have therefore also considered various alternative accessibility measures to predict land values and house prices, including distance from multiple employment centers, not just the CBD (e.g., McMillen and MacDonald, 1990; Day et al., 2007). Other researchers have also considered, in addition to distance, travel time to the CBD and to specific points of interests and other employment centers (Song, 1994; Katz and Rosen, 1987; Des Rosiers et al., 2000).
Subsequent to the financial crisis, an emerging line of research has assessed the impact of housing distress or foreclosures on home prices. When a property is foreclosed, it is typically held by a bank as real estate owned (REO) and is left unoccupied for a length of time.4 Foreclosed properties sell at a discount for many reasons, not least that the sellers are operating under a set of unfavorable incentives that may lead to accepting a lower price, even in areas where housing is appreciating (Pennington-Cross, 2006). This could have a cascading or contagion effect – the idea that foreclosures can also negatively impact property values of both the home being foreclosed upon and nearby properties. Immergluck and Smith (2005) made an estimate of the effect of foreclosures in Chicago from 1997 to 1998, and found that property values in Chicago were lowered by more than $598 million or $159,000 per foreclosure. They (Immergluck and Smith, 2006) also found that higher foreclosure rates contribute to higher levels of violent crime in more vulnerable neighborhoods, making them less attractive to prospective buyers and ultimately contributing to more neighborhood decline and lower property value. Later research findings (e.g., Agarwal et al., 2012; Seo and Mikelbank, 2017; Kaplan and Sommers, 2009) confirmed various additional mechanisms, by which the notion of the contagion effects of foreclosures operate including consideration for submarket geographies, racial segregation, lending practices, and the market participants – whether buyers/sellers are individuals or institutions.5
The Cook County housing market and differences in neighborhood income
We focus this analysis on Cook County, the largest and most diverse county in the Chicago MSA, comprising more than 60 percent of the housing units in the MSA. Cook County can be characterized as a “mixed-market” area in the sense that it experienced a more moderate fluctuation in home prices, compared to areas that are well known for having gone through deeper dives in housing prices, such as the Northeast, Florida, and the Southwest in particular, over the recent housing market cycle. The county is particularly well suited for a case study of local neighborhood housing market price differentials post-crash. The area’s main amenity is Lake Michigan, which borders it to the east. Aside from that, there are no other significant natural amenities that could explain strong price differentials within the city. Yet, sharp differences exist between housing/land prices between the north and the high minority populated south sides of Chicago, even as both are bordered by the lake, and commuting cost to the central city (the Loop) is roughly similar (Guerrieri, Hartley, and Hurst, 2010). Housing distress, as measured by the amount of loans in serious delinquency and/or foreclosures, has been higher in the county than in the nation (figure 1).
Figure 1. Serious delinquency or in foreclosures, monthly rate, 2008-2017
We use PUMAs as our unit of measurement of the local housing market. PUMAs, or Public Use Microdata Areas, are geographic areas defined by the U.S. Census, which represent relatively homogeneous areas. There are 33 PUMAs in Cook County, 14 of which encompass suburban communities; the remaining 19 comprise city communities. The PUMA areas are named after the most prominent central municipality or Chicago community area that they contain.
Table 1 reports the median household income in places (PUMA) in Cook County by decreasing order of inflation-adjusted income in 2000. The results make clear the large socioeconomic variations across these local areas. As can be seen, the median income of areas on the lowest end of income spectrum, for example, (Douglas, Grand Boulevard, Oakland, Kenwood, Hyde Park, Washington Park, Woodlawn, and South Shore) is 36 percent of the median income of those areas on the highest end of income in the Northwest Chicago suburbs. Interestingly, both communities in low and upper end of the socioeconomic spectrum have seen some increases/ decreases in median income (adjusted for inflation), reflecting various population and sociodemographic shifts within and across communities in the county. Census data suggest that in recent years, places on the South Side of Chicago have seen strong population declines, especially predominantly black communities,6 while the city of Chicago is a popular destination with educated millennials.7 Meanwhile, some of the suburbs are also seeing lower median household income with a more recent phenomenon dubbed suburbanization of poverty (Kneebone and Holmes, 2016; Kneebone, 2017).8
Table 1. Characteristics of local housing markets in Cook County, Illinois
As table 1 shows, some northern suburbs of Chicago, namely Northbrook, Glenview, Wilmette, Winnetka, Glencoe, and Northfield, have seen the largest increase in income. Also other areas on the North Side of Chicago, and in central, and northwest areas of Chicago have had increases in median income (e.g., Rogers Park, Edgewater, Uptown, Lakeview, Lincoln Park, North Center, Lincoln Square, West Ridge, Forest Glen, North Park, Albany Park, Irving Park, Hermosa, Avondale, Logan Square, West Town, Near West Side, Lower West Side, Near North Side, Loop, and Near South Side). By contrast, most areas on the south and west sides of Chicago, which are on the lowest end of the income spectrum, and many of which are predominantly minority (black and Hispanic) neighborhoods, have seen decreases in income from 2000.
For the formal analysis of this article estimating the hedonic price model, we compile a large dataset made of all detached single family property transactions recorded in Cook County from 1997 to 2017.9 We construct a series of variables from various data sources related to the characteristics and location of properties in our sample. We also construct various measures of accessibility, in addition to distance to the central business district in Chicago, to take into consideration the specific spatial context of the county. Finally, we construct (indicators of) measures of housing distress.
Table 2 lists the variables that we include in the analysis, along with the mean values of these variables. Sales price data on single family sales activity (the dependent variable in the model) was taken from three sources: 1) property transfer records the Cook County Recorder of Deeds via Property Insight;10 2) sales records from Midwest Real Estate Data (MRED); and 3) the Northwest Illinois Multiple Listing Service (MLS). Property characteristics include building structure, square footage, number of bathrooms and bedrooms, and age of the building. These data come from the Cook County Assessor and the Northwest Illinois MLS.
Table 2. Factors affecting housing price
The geographic control variables include distance from properties to Chicago Transit Authority (CTA) rail stations, to Lake Michigan, to any type of publicly-accessible open space, to Metra rail stations, and to a lake or river other than Lake Michigan.11 Spatial data for parcels is obtained annually by the DePaul Institute of Housing Studies (IHS) from the Cook County assessor. Distances to CTA and Metra rail stations were calculated by joining the Cook County road network from the Cook County Data Portal and CTA and Metra rail station locations obtained from the City of Chicago Data Portal. Data on properties' proximity to Lake Michigan, on publicly accessible open space, and bodies of water other than Lake Michigan come from the Chicago Metropolitan Agency for Planning (CMAPs) land use file for 2005.
Indicators of housing market distress include short sale, sale at foreclosure auction, and sale occurring after a property entered REO status. Foreclosure distressed status was determined by identifying the date of a foreclosure filing on a property and tracking subsequent transaction activity. These data come from the Cook County Clerk and Cook County Recorder of Deeds via Property Insight.
Results of empirical estimates
The results of the regression estimates of the hedonic housing price model, which show the effects of the specified characteristics on house price, are reported in Appendix A for Cook County and disaggregated by the city of Chicago and suburban Cook County. (Generally, we were able to explain close to 80 percent of the house price variations within local areas in Cook County using our model, based on the R-Square results.) We illustrate the results in Appendix A in figures 2A, 2B, and 2C for the city of Chicago and suburbs.
The effects of property characteristics are shown in figure 2A. We note that square footage and the lot size are the largest determinative features for housing price. That is, larger homes on average are associated with higher price. To be more precise, square footage contributes more than 30 percent to the value of homes in both the city of Chicago and the Cook County suburbs. Lot size contributes to increasing home price in the city by 16 percent and in the suburbs by 11 percent. In addition, the number of bedrooms, bathrooms, and total number of rooms, have a positive effect on house price. Other amenities, such as a garage, brick exterior, fireplace, and central air conditioning, all have a positive effect on house price. These factors contribute from 1 percent to 8 percent of the value of homes in the city and the suburbs.
Figure 2A. Effects of housing structural characteristics on residential property price
Neighborhood characteristics and distances:
Figure 2B shows the relationship between neighborhood characteristics and proximity to amenities and housing price. Waterfront properties in the suburbs are associated with higher housing price. Proximity to a CTA train stop has a stronger negative effect in the suburbs than in the city of Chicago. Our analysis confirms more recent research, which found that contrary to the classical monocentric model, distance from the CBD is associated with higher housing prices, as opposed to lower. Lake Michigan is associated with higher prices in the suburbs, but lower prices in the city, as previous research of the Chicago market had found (Guerrieri, Hartley, and Hurst, 2010). Waterfront or proximity to other bodies of water (lakes, rivers) increases prices of residential homes in both the city of Chicago and Cook County suburbs.
Figure 2B. Effects of proximity to amenities and of distance from CBD
Sources: Distance variables are based on data from Cook County Data Portal. CTA and Metra rail station locations data are based on data from the City of Chicago Data Portal. Lake Michigan, publicly-accessible open space, and lakes and rivers other than Lake Michigan come from the Chicago Metropolitan Agency for Planning (CMAPs) land use file for 2005. Authors’ calculations.
Distressed sale effects:
Figure 2C shows the effect of distressed sales from each year between 2007 and 2017. The distressed sale/year interaction variables return (highly negative) significant effects on house price. The effects, interestingly, are stronger in the city than the suburbs. To illustrate, before the housing crisis, in 2005 with the distressed sale/year coefficient estimates of -0.06 in the city of Chicago and in the suburbs, this means that assuming a median sale price of $187,500, a distressed sale would barely drop the price of the home. But consider the coefficient estimate in 2009 at the height of the housing market crisis, with a coefficient estimate of -0.64 for the city of Chicago and -0.39 in the suburbs; this means that assuming a median sale price of $187,500, a distressed sale would drop the price of the property to $86,250 in the city of Chicago or to $114,375 in suburban Cook County. For the most recent year for which we conduct this analysis, 2017, again assuming a median sale price of $187,500 with coefficient estimates of -0.43 in the city of Chicago and -0.31 in the suburbs, the distressed sale would drop the price of the property to $105,855 in the city or to $127,892 in the suburbs.
Figure 2C. Effects of distressed housing on residential property price
Home values and low-income markets
As mentioned, the hedonic model estimates allow for the development of an index tracking price and changes over time for specific geographies.12 We use the estimated average price level on the condition of all the control variables from the model presented to derive price indices for Cook County, the city of Chicago, and the housing submarkets (PUMAs). We are particularly interested in understanding how prices have evolved for low-income submarkets relative to high-income submarkets.
Figure 3 shows the differences in the price index for each PUMA, relative to the county in 2017. Places near north and south of the central business district (the Loop), namely, West Town, Hermosa, Avondale, Logan Square, and Lincoln Square have the highest prices relative to the county as a whole. Also, places in northeast, northwest, and north suburbs have relatively higher housing prices than the county. These are areas of higher income (see table 1), and they include Northbrook, Glenview, Wilmette, Winnetka, Glencoe, and Northfield, as well as south suburbs, such as Orland Park, Palos Hills, Palos Park, and Lemont. By contrast, lower-income, South Side neighborhoods of the city of Chicago, namely Auburn Gresham, Roseland Chatham, Burnside, Avalon Park, as well as south Chicago suburbs, such as Chicago Heights, Matteson, Flossmoor, and other Chicago north suburbs like South Holland and Harvey, have lower housing prices, relative to the county (average).
Figure 3. Housing price index differences in Cook County submarkets -- Q2 2017
In figure 4, we report the correlation between income and housing price, and confirm that there is a positive relationship between home price and the income level of the neighborhood. But what is more important to note is the fact that the relationship has increased over time, from being negative in the early 2000s to becoming more positive post the housing market crash. This suggests that housing prices have become even closer to the income level of the area in Cook County, reflecting stronger housing market segregation based on socioeconomic income of the neighborhoods.
Figure 4. Correlation between price and the income level of neighborhoods in Cook County
Variation in price cycles and price shocks from housing market crisis
Figure 5 gives a bird’s-eye view of the annualized growth rate (in price index) for each PUMA; we note the price change ranging from negative or no growth to up to 8 percent annual growth rate in some areas. We focus on specific areas to understand better the variation in the price cycles across the different submarkets by examining the trend in price, as well as measuring the drop in price (price shocks) during the housing market crisis. Prices in Cook County declined by more than 40 percent between 2007 and 2012; but between 2012 and 2017, housing prices increased by more than 30 percent (figure 6).
Figure 5. IHS price index -- year over year change, Q2 2017 Cook County submarkets
Figure 6. Home values in 2017 relative to peak price before housing market crisis
The sets of figures in Appendix B show the price cycles and the drop in housing price covering the housing crisis period (2007-2012), respectively, for the various submarkets (PUMA), grouped by areas (i.e., north, south, etc.). As revealed, the effects of the housing market crash on home prices were uneven across submarkets within each area. Notably, none of the PUMAs in the north had price decline as steep as the county. In the south, all the PUMAs had price decline steeper than the county (with the exception of Ashburn, Beverly, Washington Heights, Morgan Park, and Mount Greenwood).
Variations in housing market recovery across local housing markets
We analyze the extent to which the local housing markets are recovering by examining the change in price in Q4 2017, from peak prices (Q4 2007) in each of the submarkets. The differences across the different submarkets are worth noting (figure 6). Cook County’s home prices have not yet exceeded their previous peak (in nominal terms). In Cook County, home values were still 20 percent less from the previous peak (in Q2 2017), and strong differences exist by submarkets. In some places, prices have surpassed their previous peak significantly. For example: in Hermosa, Avondale, and Logan Square (Chicago city), prices are 12 percent above the previous peak; Near North Side and Near South Side of the Loop (downtown Chicago central business district), prices are 9 percent more than the previous peak; West Town, Near West Side and Lower West Side of the city prices are 15 percent more than the previous peak; and Near North Side of the city, including Lakeview, Lincoln Park, Lincoln Square, and West Ridge, prices are 9 percent above their previous peak. By contrast on the lower end of the income spectrum, in places like South Side neighborhoods of the city – Chicago Lawn, Englewood/West Englewood, and Greater Grand Crossing, prices are 56 percent lower than their previous peak.
Finally, figure 7 shows the share of neighborhoods in different income quartiles and the change in house prices from the previous peak, which makes it clear that the lower the income of the areas in the county, the slower they are recovering.
Figure 7. Income quartile of neighborhoods and home values change from previous peak
Conclusions and implications
We analyzed the determinants of home value and derived price indices for submarkets in Cook County based on a hedonic price model, which takes into consideration characteristics of homes, and of the location of homes, relative to specific amenities in the county. We find results that are consistent with expectations regarding the relationship of housing characteristics and proximity to various amenities and home prices. The model also allows examining the relationship between distressed sales and home prices. We find that the effects of distressed sales are still very potent a decade after the crisis.
According to the 2017 report from the Joint Center for Housing Studies of Harvard University (JCHS) on the State of Housing, home prices in the majority of metros have yet to fully recover from the foreclosure crisis, including in some markets where prices have risen sharply in recent years. For the nation, home prices in real terms were still 9 percent to 16 percent below the mid-2000 peak (as of 2016). (Although, in nominal terms, prices had gained somewhat in 2016 by at least 1.2 percent above the previous peak, according to the main information sources, such as the S&P Corelogic, Case-Shiller, and the Freddie Mac index). The JCHS report signaled that within these metro areas, home prices in low-income areas were slowest to appreciate.
We examined the relationship between the income level of neighborhoods and price, as well as price difference from previous peak to ascertain the extent to which the housing market is recovering and the difference across local areas in the county. We find results that are consistent with national results and those of other metropolitan areas. Since the recession, home prices in Cook County increased by 40 percent; however, home prices were still 10 percent below their peak in the county (in nominal terms). This slow recovery is driven by the deep and increasing disparities in income and socioeconomic conditions. In some of the lowest income quartile areas on the South Side of Chicago in the county, prices were more than 50 percent below their peak.
The relative lack of appreciation of homes in lower-income neighborhoods has several implications. On one hand, slow price appreciation may mean that these places may remain more affordable for low- and moderate-income households. On the other hand, and of greater concern is that the lack of price recovery in some markets also means that homeowners have little or no equity in their homes. In fact, the JCHS (2017) research reported that in Chicago, 12.6 percent of homeowners still had negative equity, which is more than double the national rate. Further, the share of low-income homeowners under water living in some neighborhoods exceeds 40 percent, with no opportunity to refinance or sell without bringing money to the closing table, according to the same report. Given high and increasing housing costs in much of the metro area, the market stagnation in some areas raises concerns about economic mobility of residents and continued disinvestment in places with persistently low home values.
Given the pivotal role of housing in contributing to the financial security and well-being of communities and households, attending to housing challenges remains a priority. The national debate should recognize the diversity within markets and consider the particular challenge that low-income neighborhoods still face, even within a metropolitan area where foreclosure rates have returned to a manageable level. The lingering effects of the housing market crash at the local levels means that policymakers should address specific community needs and marshall resources accordingly.
Appendix A. OLS regression estimates of the hedonic housing price model, which show the effects of the specified characteristics on house price in local housing markets (PUMA) in Cook County.
Appendix B. Housing price index throughout Cook County and price decline throughout Cook County, Q2 2007 - Q2 2012
Appendix B1. Housing price index, Cook County
Appendix B2. Price decline, Cook County Q2 2007 - 2012
Appendix B3. Housing price index, northCook County
Appendix B4. Price decline, north Cook County Q2 2007 - 2012
Appendix B5. Housing price index, northwest Cook County
Appendix B6. Price decline, northwest Cook County Q2 2007 - 2012
Appendix B7. Housing price index, northeast Cook County
Appendix B8. Price decline, northeast Cook County Q2 2007 - 2012
Appendix B9. Housing price index, south Cook County
Appendix B10. Price decline, south Cook County Q2 2007 - 2012
Appendix B11. Housing price index, southwest Cook County
Appendix B12. Price decline, southwest Cook County Q2 2007 - 2012
Appendix B13. Housing price index, southeast Cook County
Appendix B14. Price decline, southeast Cook County Q2 2007 - 2012
Appendix B15. Housing price index, central Cook County
Appendix B16. Price decline, central Cook County Q2 2007 - 2012
Appendix B17. Housing price index, west central Cook County
Appendix B18. Price decline, west central Cook County Q2 2007 - 2012
Appendix B19. Housing price index, south central Cook County
Appendix B20. Price decline, south central Cook County Q2 2007 - 2012
Appendix B21. Housing price index, west Cook County
Appendix B22. Price decline, west Cook County Q2 2007 - 2012
1 See a more detailed review of the literature in John R. Ottensmanna, Seth Paytona, and Joyce Manb (2008); Bowen et al. (2001); Malpezzi (2003).
2 See Heikkila et al. (1989) for a review of studies of determinants of residential property or land values using hedonic models, saying that to the extent they have included location, it has generally been distance to the CBD.
3 Bender and Hwang (1985) review research, which does not lend support to the monocentric concept. Coulson (1991) likewise observed that prior research has had great difficulty in verifying the decline of land prices and land consumption with distance from the CBD, noting in particular that in tests of rent gradients, estimation has often yielded positive or insignificant values.
4 Judicial foreclosures states require the courts to get involved, which substantially slows down the process. By contrast, power-of-sale states allow the bank to sell the property without the court’s supervision. States with Statutory Right of Redemption indirectly delay the resolution of a foreclosure by effectively limiting the demand pool that is willing to buy a foreclosed property. This law allows a foreclosed upon property owner to regain ownership for up to one year, even after it has been sold to someone else.
5 The scope of this study does not include a test of the contagion effects of foreclosures or the mechanisms by which it operates on nearby properties.
7 Chicago is the fifth most popular destination for millennials. See https://urbanmatter.com/chicago/chicago-millennials.
8 According to a report by Brookings Institute, almost every major metropolitan area experienced a significant increase in the suburban poor population from 2000 to 2015. See https://www.brookings.edu/testimonies/the-changing-geography-of-us-poverty.
9 For the purpose of the analysis, we excluded those properties where transactions repeated within 90 days to avoid any potential recording errors and to reduce potential bias in the price index due to frequently traded properties. Additionally, we dropped transactions where we found substantial missing information on essential property characteristics, such as the number of bedrooms, existence of an air conditioning system, or because of errors such as missing property identification numbers, or conflicting sales price information. We end up with a sample reflecting 75.7 percent of the transactions, for this analysis.
11 Geographic variables were calculated using ArcGIS software.
12 Additional details on the technical derivation of the price indices based on this model are provided in DePaul, Institute for Housing Studies, “Description of IHS Hedonic Data Set and Model Developed for PUMA Area Price Index,” May 2015. See https://www.housingstudies.org/media/filer_public/2015/05/12/puma_hedonic_model_technical_paper_d7kh29N.pdf.
Alonso, W., 1964, Location and Land Use, Cambridge: Harvard University Press.
Agarwal, S., B. W. Ambrose, S. Chomsisengphet, and A. B. Sanders, 2012, “Thy Neighbor’s Mortgage: Does Living in a Subprime Neighborhood Impact your Probability of Default?,” Real Estate Economics, Vol. 40, No. 1, pp. 1–22.
Bajaj, V., and L. Story, 2008, “Mortgage Crisis Spreads Past Subprime Loans,” New York Times, February 12.
Baumer, E. P., K. T. Wolff, and A. N. Arnio, 2012, “A Multicity Neighborhood Analysis of Foreclosure and Crime,” Social Science Quarterly, Vol. 93, No. 3, pp. 577–601.
Bender, B., and H-S. Hwang, 1985, “Hedonic Housing Price Indices and Secondary Employment Centers,” Journal of Urban Economics Vol. 17, No. 1, pp. 90–107.
Bowen, W. M., B. A. Mikelbank, and D. M. Prestegaard, 2001, “Theoretical and Empirical Considerations Regarding Space in Hedonic Housing Price Model Applications,” Growth and Change, Vol. 32, No. 4, pp. 466–490.
Campbell, J. Y., S. Giglio, and P. Pathak, 2011, “Forced Sales and House Prices,” The American Economic Review, Vol. 101, No. 5, pp. 2108–2131.
Coulson, N. E., 1991, “Really Useful Tests of the Monocentric Model,” Land Economics, Vol. 67, pp. 299–307.
Day, B., I. J. Bateman, and I. Lake, 2007, “Beyond Implicit Prices: Recovering Theoretically Consistent and Transferable Values for Noise Avoidance from a Hedonic Property Price Model,” Environment and Resource Economics, Vol. 37, No. 1, pp. 211–232.
Des Rosiers, F., M. Thérault, and P-Y Villeneuve, 2000, “Sorting out Access and Neighbourhood Factors in Hedonic Price Modelling: An Application to the Quèbec City Metropolitan Area,” Journal of Property Investment and Finance, Vol. 18, No. 3, pp. 291–315.
Ergungor, O. E., and L. Nelson, 2012, “The Impact of Recovery Efforts on Residential Vacancies,” Federal Reserve Bank of Cleveland, working paper, No. 12-03.
Guerrieri, V., D. Hartley, and E. Hurst, 2010, “Endogenous Gentrification and Housing Price Dynamics,” NBER Working Paper No. 16237.
Hartley, D., 2014, “The Effect of Foreclosures on Nearby Housing Prices: Supply or Disamenity?,” Regional Science and Urban Economics, Vol. 49, pp. 108–117.
Heikkila, E., P. Gordon, J. I. Kim, R. B. Peiser, H. W. Richardson, and D. Dale-Johnson, 1989, “Whatever Happened to CBD-Distance Gradient?: Land Values in a Policentric City,” Environment and Planning A, Vol. 21, No. 2, pp. 221–232.
Immergluck, D., and G. Smith, 2005, “Measuring the Effect of Subprime Lending on Neighborhood Foreclosures: Evidence from Chicago,” Urban Affairs Review, Vol. 40, No. 3, pp. 362–89.
Immergluck, D., and G. Smith, 2006, “The External Costs of Foreclosure: The Impact of Single-Family Mortgage Foreclosures on Property Values,” Housing Policy Debate, Vol. 17, No. 1, pp. 57-79.
Joint Center for Housing Studies of Harvard University (JCHS), 2017, The State of the Nation’s Housing, available at http://www.jchs.harvard.edu/sites/jchs.harvard.edu/files/harvard_jchs_state_of_the_nations_housing_2017.pdf.
Kain, J. F, and J. M. Quigley, 1970, “Measuring the Value of Housing Quality,” Journal of the American Statistical Association, Vol. 65, No. 330.
Kaplan, D. H., and G. G. Sommers, 2009, “An Analysis of the Relationship Between Housing Foreclosures, Lending Practices, and Neighborhood Ecology: Evidence from a Distressed County,” The Professional Geographer, Vol. 61, No. 1, pp. 101–120.
Katz, L., and K. T. Rosen, 1987, “The Interjurisdictional Effects of Growth Controls on Housing Prices,” The Journal of Law and Economics, Vol. 30, No. 1, pp. 149–60.
Kneebone, E., and N. Holmes, 2016, “U.S. Concentrated Poverty in the Wake of the Great Recession,” March 31, available at https://www.brookings.edu/research/u-s-concentrated-poverty-in-the-wake-of-the-great-recession.
Kneebone E., 2017, “The Changing Geography of US Poverty,” February 15, available at https://www.brookings.edu/testimonies/the-changing-geography-of-us-poverty.
Malpezzi, S., 2003, “Hedonic Pricing Models: A Selective and Applied Review,” In T. O’Sullivan and K. Gibb (eds.), Housing Economics and Public Policy, Oxford: Blackwell Science.
McMillen, D. P, and J. F. MacDonald, 1990, “Employment Subcenters and Land Values in a Polycentric Urban Area: The Case of Chicago,” Environment and Planning, Vol. 22, No. 12, pp. 1561–1574.
Mills, E. S., 1967, “An Aggregative Model of Resource Allocation in a Metropolitan Area,” The American Economic Review, May, Vol. 57, No. 2, pp.197–210.
Muth, R. F., 1969, Cities and Housing: The Spacial Pattern of Urban Residential Land Use, Chicago: University of Chicago Press.
Osttensmann, J. R., S. B. Payton, and J. Y. Man, 2008, "Urban Location and Housing Prices within a Hedonic Model,” Journal of Regional Analysis and Policy, Vol. 38, No. 1.
Pennington-Cross, A., 2006, “The Value of Foreclosed Property,” Journal of Real Estate Research, Vol. 28, No. 2, pp.193–214.
Rosen, S., 1974, “Hedonic Prices and Implicit Markets: Product Differentiation in Pure Competition,” Journal of Political Economy, Vol. 82, No. 1, pp. 34–55.
Song, S., 1994, “Modelling Worker Residence Distribution in the Los Angeles Region,” Urban Studies, Vol. 31, No. 9, pp. 1533–1544.
Seo, Y., and B. Mikelbank, 2017, “Spatially and Sequentially Heterogeneous Discounts of Distressed Property Values in Cuyahoga County, Ohio,” Housing Policy Debate, Vol. 27, No. 4, pp. 570–583.