Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The Impact of Shuttered Golf Courses on Property Values

The Impact of Shuttered Golf Courses on Property Values The Impact of Shuttered Golf Courses on Property Values Stephanie R. Yates and Lary B. Cowart A u t h o r s A b s t r a c t We measure the impact of a golf course as a residential amenity on surrounding home values in several communities in Shelby County, Alabama. We compare the values of homes in golf course communities (GCCs) and non-golf course communities, as well as the values of homes within these communities before and after the golf course closes. Using a methodology similar to Bond, Seiler, and Seiler (2002), we examine the sales prices of homes within GCCs both before and after a golf course closure to see how the closure affects the sales prices of homes and test for the significance of that difference. We calculate the difference in value for homes in GCCs before and after the golf course is closed and test for the significance of that difference. We estimate the degree to which specific factors explain the variance in home prices in these communities before and after the golf course closed. We find that homes in GCCs sell at a 9% premium compared to homes in non-GCCs. We also find that home prices in GCCs decrease by 17% after the related golf course closes; home prices for properties adjacent to a golf course diminish as well. K e y w o r d s golf courses, golf course communities, value diminution, residential amenities As noted in a July 2019 Wall Street Journal article (Bauerlein, 2019), golf has declined in popularity in recent years, resulting in a number of golf course closures. This leaves developers and homeowners in a difficult position and scrambling to preserve home values. We seek to quantify the impact of this trend using a sample of home sales in Shelby County, Alabama. Earlier research confirmed the widely held belief that houses in golf course communities (GCCs) were more highly valued and sold for higher prices than otherwise equivalent houses in non-golf course communities. Most of that research occurred in the 1990s and early 2000s when both golf play and housing prices were rising. Then, the Great Recession became an inflection point for both golf and the housing market. According to the National Golf Foundation (2018), the high of 30.6 million U.S. golfers in 2003 was reduced to 24.7 million by 2014. Furthermore, as the number of golfers between the ages of 18 and 34 had declined 30% over the past 20 years, there was concern regarding the impact this might have on the sport despite the number of baby boomers nearing retirement age. There were more than 13,700 golf courses nationwide in 2015, according to the National Golf Foundation, down from 16,000 in 2009, of which 3,200 had T h e I m p a c t o f S h u t t e r e d G o l f C o u r s e s o n P r o p e r t y V a l u e s 3 communities built around them. That is a 14% drop in the total number of golf courses within a six-year period. If houses in a GCC command a premium, what happens to that premium when the golf course closes? R e v i e w o f t h e L i t e r a t u r e Location has long been a key factor in determining the value and desirability of residential real estate. Many published works consider the impact of available amenities on housing values. Researchers have largely subdivided those amenities into environmental amenities and residential amenities. Environmental amenities are the features associated with the surrounding neighborhood and residential amenities specific to the homes themselves. The most commonly studied environmental amenities are views. Several studies consider the impact of views on home value; therefore, we further subdivide this literature into those articles that analyze water views versus non-water views. Amenities Davies (1974) uses 1968 data from the town of Nottingham to analyze environmental amenities and accessibility based on the theory that consumers seeking to maximize their utility derive it from residential characteristics. This model indicates that the number of bedrooms, central heating, plot size, open space in the immediate vicinity, the absence of industrial nuisance, and seclusion are positive factors, whereas poor dwelling quality and proximity to a main road are negative factors. Sirmans, Macpherson, and Zietz (2005) review studies using hedonic modeling to estimate house prices. Hedonic modeling identifies the factors that contribute to prices. The factors the authors identify as positive and statistically significant in this study are slanted versus flat roof, sprinkler system, garden bath, separate shower stall, double oven, and gated community, while not having attic space negatively affects selling price. Many studies have examined the impact of externalities on property values. Bourassa, Hoesli, and Sun (2005, 2006) studied the prices of three specific aesthetic externalities: a water view, the appearance of nearby improvements, and the quality of landscaping in the neighborhood in a sample of residential sales from 1986 to 1996 in Auckland, Christchurch, and Wellington New Zealand. The impact of the factors that they identify changes with the real estate cycle but that the premium placed on water views is highest in areas where the supply is lowest. Simons and Saginor (2006) hypothesize that markets can internalize proximity to positive factors and this effect can be determined. Their meta-analysis using 620 observations from 17 peer-reviewed articles indicates that properties located near a positive attribute such as beach frontage, water views, parks, and golf courses command a 25% premium. Nelson (2010) estimates hedonic prices for summer and winter rentals for vacation houses located near a lake and ski-golf resort in rural western Maryland. There, access to recreation impacts rental offers. Specifically, lakefront locations command a $1,000–$1,200 weekly premium and ski-slope access is valued at $500–$600 per week. These premiums amount to J O S R E V o l . 1 1 2 0 1 9 u u 4 Y a t e s a n d C o w a r t 42%–44% of the total average value for lakefront locations and 27%–32% of total average value for ski-slope access. In Birmingham, Alabama, Rauterkus, and Miller (2011) find that land values increase with walkability but that this result reverses as neighborhoods become more car-dependent. That is, in generally walkable areas, values increase with walkability. However, in generally car- dependent areas, values decrease with walkability. Grudnitski (2003) employs a semi-logarithmic form of regression to study the impact on sales price of residential properties in different types of GCCs. He compares the sales prices of residential properties that were not located in GCCs and found 12.5% price premiums for houses located in private GCCs, 6% for houses located in semiprivate GCCs, and 5.7% for houses located in public GCCs. Fraser and Allen (2017) study 241 condominium transactions occurring between 2011 and 2015 in three southwest Florida communities and find that after controlling for unit age, size, bedrooms, bathrooms, garage spaces, and time trend, proximity to a cart path may reduce transaction prices by 5.1%. The authors define proximate paths as those where the golf cart path is routed on the dwelling side of the teeing area, fairway, or green of the condominium unit. The authors then further analyze the golf condominium units that are proximate to cart paths and differentiate between those that are located at a tee box versus those that are located near a green. They find evidence suggesting an even greater discount such that units with proximate cart paths near greens sell for a discount of 11.6% and those near tees sell for discounts of 8.6%. View Lots Rodriguez and Sirmans (1994) use multiple regression analysis to estimate that a good view in Fairfax County, Virginia adds about 8% to the value of a single- family house. Do and Grudnitski (1995) find that a golf course location adds 7.6% to a property’s sales price. Ming and Hian (2005) find construction that would obstruct the view of existing units depresses the prices of the obstructed development by around 8% in the long term. Poudyal, Hodges, Fenderson, and Tarkington (2010) find that increasing the size of forest area visible from a house by one acre increased the house price by $30. Wyman and Sperry (2010) find a hierarchy in pricing with lakefront lots commanding the highest premiums (124%– 287%), followed by lake views (94%–133%) and then golf course views (42%– 85%). Baranzini and Schaerer (2011) use hedonic modeling to assess the value of view and landscape use in the Geneva, Switzerland rental market and find that accessibility to various environmental amenities, landscape use, and size, as well as the view of them have a statistically significant impact on rents. In addition, they find that proximity, size, and view of water amenities increase rents. However, a view of urban parks and building-covered areas negatively affect rents. Wen, Zhang, and Zhang (2015) find that parks, mountains, rivers, and lakes have significant positive effects on housing price. Mittal and Byahut (2019) use a geographically-weighted regression model that indicates an average 3.4% price premium on the mean value of homes in the study area. Mothorpe and Wyman (2017) define an interior lot as one without lake frontage, golf frontage or a mountain view and find a collapse in the value of interior lots in GCCs compared to interior lots elsewhere in Pickens County, South Carolina from 2000 to 2016. T h e I m p a c t o f S h u t t e r e d G o l f C o u r s e s o n P r o p e r t y V a l u e s 5 Our review of the literature indicates that residential and environmental amenities have a positive and significant impact on residential housing prices. Our focus is on one particular environmental amenity—golf courses. Earlier researchers found that while golf courses do not command as much of a premium as lakefront lots and lake views, they may command a price premium of 8% or more. Further, they find that obstructing a view that was previously unobstructed may depress prices by 8%. We estimate the impact of eliminating a positive amenity (proximity to a golf course) using a sample of homes in Shelby County, Alabama. D a t a a n d M e t h o d o l o g y We analyze data related to six GCCs in Shelby County, Alabama. Shelby County is one of seven counties that form the Birmingham-Hoover MSA and has been the fastest growing county in Alabama every decade since 1970. The county had an estimated population of 213,605 in 2017. Founded in 1818, Shelby County covers a total area of 810 square miles. Economically, the median household income for Shelby County was $68,380 according to the 2010 Census and only 7.4% of the population was below the poverty line compared to $59,039 and 14.5% nationwide respectively. There are 288 golf courses in Alabama. Nine of these courses are located in Shelby County. The golf courses included in the study are Ballantrae Golf Club, Eagle Point Golf Club, Greystone Golf and Country Club, Inverness Country Club, Montevallo Golf Club, and Shoal Creek Golf Club. Our golf course and associated GCC of interest is Eagle Point Golf Club. Eagle Point Golf Club was a 71-par, 6,470-yard golf course that opened in 1990 and closed May 1, 2016. The three golf courses in Shelby County that we did not study do not have associated communities. We collected sales data for the subdivisions in which these golf courses are located from the MLS according to the dates in Exhibit 1. We remove all transactions involving quit claim deeds and any other transactions that are unqualified such that our dataset contains only transactions that occurred in an open, competitive market between typically informed and motivated buyers and sellers. We also remove all sales of vacant lots. This leaves only improved residential real property in our final dataset, which are parcels of real estate having an inhabitable structure located on it. In order to address the potential for heteroscedasticity in our model, we removed several outliers. Specifically, we removed 38 transactions with a sales price of less than $100,000, as well as those with a sales price greater than $1,000,000. We anticipate that the closure of a golf course within a residential community negatively impacts home prices. Additionally, we anticipate, in accordance with Bond, Seiler, and Seiler (2002), that sale price is positively related to the number of bathrooms, living space, lot size, and frontage. In contrast to Bond, Seiler, and Seiler, however, we anticipate a negative correlation between sales price and age given the high demand for new construction in this area. The data are summarized in Exhibit 2. J O S R E V o l . 1 1 2 0 1 9 u u 6 Y a t e s a n d C o w a r t Exhibit 1 u Sample Communities Golf Course Community Transactions Mean Sale Price Eagle Point Golf Club 1990–2016 42 $389,027 Heatherwood Golf Club 1986–2009 2016–present 80 $419,017 Greystone Golf and Country Club 1991–present 766 $391,105 Inverness Country Club 1973–present 123 $248,862 Montevallo Golf Club 1955–present 1 $300,000 Shoal Creek Golf Club 1977–present 37 $482,004 Ballantrae Golf Club 2004–present 345 $279,307 Brook Highland Never 342 $352,661 Meadowbrook Never 223 $266,514 Total 1,959 Mean $344,357 Notes: This table presents information regarding the communities in our sample. Heatherwood Golf Club reopened as Heatherwood Hills Country Club on October 8, 2016. This club has two associated courses: Founders Course and Legacy Course. We define our variables as follows: PRICE 5 Market sales price of the home; FORMER 5 Qualitative variable indicating whether or not the sale occurred in a neighborhood that was formerly a GCC; AGE 5 Age of the home (in years); SIZE 5 Size of the home. We use two different variables to measure size: BATHROOMS 5 Number of bathrooms and LIVING AREA 5 Total heated square footage of living space; LOT 5 Size of the lot measured as either: LOT SIZE 5 Total square footage of the lot or LOT DEPTH 5 Horizontal distance between the front and rear property lines measured in feet; LOCATION 5 Location of the home measured as either SCHOOLRANK 5 The Great Schools ranking of the associated high school (1– 10) or GOLFCOURSEHOME 5 Qualitative variable indicating whether the home is located adjacent to a golf course. TREND 5 Monthly time trend variable that captures general market price changes over the study period as in Fraser and Allen (2017). This approach is a means of de-trending the data and thereby controlling for the linear nature of inflation and its impact on sale prices. DEMOGRAPHICS 5 Block group-level demographics of the residents from the 2017 American Community Survey: T h e I m p a c t o f S h u t t e r e d G o l f C o u r s e s o n P r o p e r t y V a l u e s 7 Exhibit 2 u Descriptive Statistics N Min. Mean Max. Std. Dev. Panel A: Continuous variables AGE 1,951 0 18 101 8.76 BATHROOMS 1,909 0 2 6 1.16 LIVINGAREA 1,951 992 2,776 7,367 994.30 LOTDEPTH 1,959 0 152 878 85.97 LOTSIZE 1,957 0 18,674 367,211 22,940.89 MEDIANAGE 1,954 30.1 43 54.2 5.34 MEDIANHHI 1,954 29,481 105,068 147,774 27,621.50 PRCNTBCHLR 1,954 10.22 39 50.74 7.30 SALEPRICE 1,959 100,000 344,356 985,000 161,259 SCHOOLRANK 1,959 3 9 10 0.51 Panel B: Binary variables 1 0 N Frequency Percentage Frequency Percentage FORMERGCC 1,959 78 3.98% 1,881 96.02% GOLFCOURSEHOME 1,959 145 7.40% 1,814 92.60% RECESSION 1,955 89 4.55% 1,866 95.45% Notes: This table presents the details of our sample of homes in nine different Shelby County, Alabama golf course communities. MEDIANAGE 5 Measured as age (in median years); MEDIANHHI 5 Income (median household income in dollars); PRCNTBCHLR 5 Educational attainment (as % of the population that has earned a bachelor’s degree); and RECESSION 5 Qualitative variable indicating whether the sale occurred during a recession. We identify recessions in our sample as the period from December 2007 to June 2009. To begin our examination of the impact of a residential amenity—a golf course— on home values, we examine the transactions in our sample where either there was never an associated golf course or the golf course never closed. We conduct an analysis of variance (ANOVA) test to compare the mean values of our key variables across the three main subsets of properties. These three subsets are homes located in communities that: (1) have always been GCCs; (2) were never GCCs; and (3) were previously GCCs. We refer to the first group as GCCs, the second group as non-GCCs, and the third group as closed GCCs. We exclude 24 J O S R E V o l . 1 1 2 0 1 9 u u 8 Y a t e s a n d C o w a r t Exhibit 3 u Comparison of Golf Course Communities to Non-Golf Course Communities and Former Golf Course Communities Difference Difference Closed GCCs and GCCs and GCCs Non-GCCs GCCs Non-GCCs Closed GCCs AGE 16.31 23.52 21.69 7.21*** 5.38*** BATHROOMS 2.27 2.57 3.82 0.30*** 1.55*** LIVINGAREA 2,782.35 2,646.37 3,415.34 2135.98** 632.99*** LOTDEPTH 139.36 172.49 199.10 33.13*** 59.74*** LOTSIZE 16,322.57 21,123.76 34,837.14 4,801.19*** 18,514.60*** MEDIANAGE 44.99 39.73 41.46 25.27*** 23.53*** MEDIANHHI 113,686.40 83,558.74 117,767.90 230,127.60*** 4,081.50 PRCNTBCHLR 38.73 42.53 34.36 3.80*** 24.36*** SALEPRICE 349,600.20 318,659.70 402,702.70 230,940.50 53,102.50** SCHOOLRANK 9.60 9.00 9.00 20.60*** 20.60*** Observations 1,272 565 80 Notes: This table presents the mean values for the golf course community (GCC) homes the non-golf course community homes (non-GCCs), and homes in former GCCs. *Statistically significant at 90%. **Statistically significant at 95%. ***Statistically significant at 99%. home sales that occurred in a GCC prior to golf course closure and 17 sales that occurred in a GCC after the corresponding golf course reopened. We present the results of this comparison in Exhibit 3. We find statistically significant differences in nearly all of our key variables for GCCs versus non- GCCs. Specifically, we find in GCCs, the living area, age, household income, price, and school rank are statistically significantly higher than in non-GCCs. Conversely, we find evidence that in non-GCCs, the homes are older, have more bathrooms, and larger lots. We focus on the statistically significant difference in the mean SALEPRICE for these two subsets of our data. As shown in Exhibit 3, we find that homes in GCCs sell at a $30,940.50 premium compared to homes in non-GCCs. Next, we consider the impact of the removal of a residential amenity. We first analyze golf course adjacent properties where we most expect to see price diminution after the removal of a residential amenity. The results are in Exhibit 4. In our full sample, the mean SALEPRICE for golf course adjacent properties is $556,305.00 compared to $328,794.40 for non-golf course adjacent properties, which results in a 41% premium. However, when we differentiate between homes adjacent to open versus closed golf courses, we find a significant difference in T h e I m p a c t o f S h u t t e r e d G o l f C o u r s e s o n P r o p e r t y V a l u e s 9 Exhibit 4 u Comparison of Golf Course Adjacent Within Current and Former Golf Course Communities All Golf Course Difference GCA Adjacent (GCA) GCA Homes GCA Homes in Homes Open and Homes in Open GCCs Closed GCCs Closed GCCs AGE 22.13 21.22 26.52 5.30*** BATHROOMS 3.54 3.50 3.72 0.22*** LIVINGAREA 4,016.25 4,110.17 3,562.96 2547.21*** LOTDEPTH 214.86 219.37 193.09 226.28*** LOTSIZE 34,976.85 35,671.15 31,473.77 24,197.38*** MEDIANAGE 47.52 48.13 44.58 23.55*** MEDIANHHI 120,116.10 118,324.70 128,761.70 10,437.00 PRCNTBCHLR 38.98 39.73 35.38 24.35*** SALEPRICE 556,305.00 586,426.10 410,937.70 2175,488.40** SCHOOLRANK 9.53 9.64 9.00 20.64*** Observations 134 111 23 Notes: This table presents the mean values for the golf course community (GCC) homes the non-golf course community homes (non-GCCs), and homes in former GCCs. *Statistically significant at 90%. **Statistically significant at 95%. ***Statistically significant at 99%. sale prices. If the golf course is operating, the mean SALEPRICE is $586,426.10. If the golf course has ceased operations, the mean sale price falls to $410,937.70. This statistically significant price difference of $175,488.40, or 30%, suggests that the value of golf course adjacent properties suffer when the adjacent golf course closes. Using our full sample, we seek to determine the factors that explain the variation in house prices. In the context of the removal of a residential amenity, we hypothesize that a key factor is the closure of a golf course within a community. Our explanatory variable of interest is FORMERGCC, which is a binary variable if the transaction occurred in a community that was formerly a GCC. We estimate the following model: SALEPRICE 5 a 1 b FORMERGCC 1 b AGE 1 b SIZE 1 i 1 1 2 2 3 3 1 b LOT 1 b LOCATION 1 b TREND 4 4 5 5 6 6 1 b DEMOGRAPHICS 1 b RECESSION 1 «. 7 7 8 8 J O S R E V o l . 1 1 2 0 1 9 u u 1 0 Y a t e s a n d C o w a r t Exhibit 5 u Correlation between Housing Variables AGE BATHS FRMR GCA LIVING DEPTH SIZE RANK AGE 1.0000 BATHS 0.0746*** 1.0000 FRMRGCC 0.0619*** 0.2086*** 1.0000 GCA 0.1062*** 0.2652*** 0.1827*** 1.0000 LIVING 0.0373* 0.7645*** 0.1308*** 0.3388*** 1.0000 LOTDEPTH 0.3354*** 0.4315*** 0.1133*** 0.1973*** 0.4444*** 1.0000 LOTSIZE 0.2158*** 0.4237*** 0.1458*** 0.1919*** 0.5434*** 0.5811*** 1.0000 RANK 0.1326*** 0.0079 20.1553*** 0.0756*** 0.1916 20.0086 20.0606*** 1.0000 Notes: This table presents pairwise correlation coefficients for the housing variables. *Statistically significant at 90%. **Statistically significant at 95%. ***Statistically significant at 99%. T h e I m p a c t o f S h u t t e r e d G o l f C o u r s e s o n P r o p e r t y V a l u e s 1 1 Exhibit 6 u Full Regression Results Beta t-Stat. Tolerance Intercept 2393,002.90 210.26*** FORMERGCC 258,669.16 25.25*** 0.7963 AGE 22,332.29 210.28*** 0.7027 BATHROOMS 30,917.07 12.80*** 0.3601 LIVINGAREA 94.39 29.79*** 0.2768 LOTDEPTH 10.61 0.40 0.5356 LOTSIZE 0.83 7.56*** 0.4766 GOLFCOURSELOT 63,741.95 8.64*** 0.7892 SCHOOLRANK 36,226.88 8.33*** 0.5644 TREND 165.96 2.86*** 0.7414 MEDIANAGE 567.04 1.48 0.6654 MEDIANHHI 20.12 21.50 0.5599 PRCNTBCHLR 964.49 3.41*** 0.6626 RECESSION 11,650.12 1.30 0.7764 N 1,898 Adj. R 0.7981 F 577.93 Notes: In this table, we present the results from a regression of our independent variables against the transaction-based dependent variable, market sales price. *Statistically significant at 90%. **Statistically significant at 95%. ***Statistically significant at 99%. Before estimating our regression equation, we check for correlation. The results are in Exhibit 5. We find significant levels of correlation between several variables. R e s u l t s Univariate and simple multivariate regressions not shown here indicate that all of our alternative measures of size, lot, and location have significant explanatory power over sales price. Based on these results, we use all of these measures in our full model. Thus, we find that both the number of bathrooms and the overall size of the home have significant explanatory power over sales price. In addition, with regard to the size of the lot, overall lot size is significant but not lot depth. Finally, given that location matters, we identified golf course adjacent lots and the ranking of the affiliated public high school. Both of these location-based variables have significant explanatory power over sales price. J O S R E V o l . 1 1 2 0 1 9 u u 1 2 Y a t e s a n d C o w a r t Exhibit 7 u Full Regression Results with Only Statistically Significant Variables Variable Beta Std. Error t-Stat. Intercept 2363,315.00 35,158.55 210.33*** FORMERGCC 263,851.92 10,772.65 25.93*** AGE 22,277.12 215.01 210.59*** BATHROOMS 31,403.72 2,379.88 13.20*** LIVINGAREA 94.41 3.16 29.84*** LOTSIZE 0.82 0.19 8.44*** GOLFCOURSELOT 65,114.78 7,173.07 9.08*** SCHOOLRANK 35,410.93 3,751.37 9.44*** TREND 129.78 50.95 2.55** PRCNTBCHLR 891.31 271.33 3.28*** N 1,899 Adj. R 0.7989 F 834.01 Notes: In this table, we present the results from a regression of our remaining, statistically significant independent variables against the transaction-based dependent variable, SALEPRICE. *Statistically significant at 90%. **Statistically significant at 95%. ***Statistically significant at 99%. We present the results of our full regression of our independent variables against the transaction-based dependent variable SALEPRICE, in Exhibit 6. Due to the significant correlation between explanatory variables as shown in Exhibit 5, we check for multicollinearity among the explanatory variables by calculating tolerance as the inverse of the variance inflation factor. Given that all variables have a tolerance greater than 0.10, we do not detect multicollinearity in the model. The significant variables are FORMERGCC, AGE, BATHROOMS, LIVINGAREA, LOTSIZE, GOLFCOURSELOT, SCHOOLRANK, TREND, and PRCNTBCHLR. The beta coefficient for FORMERGCC, our variable of interest, is negative and statistically significant. This indicates that the closing of a golf course results in value diminution. This is a key result as our results also support earlier studies suggesting that golf courses are a value adding residential amenity. This new finding suggests that the removal of a residential amenity can have an opposite effect. As in Bond, Seiler, and Seiler (2002), we estimate the regression again with just the remaining significant independent variables included in the model in order to find an unbiased estimate of the true beta values. Exhibit 7 presents the results of our analysis. We find unbiased beta estimates for nine independent variables: FORMERGCC, AGE, BATHROOMS, LIVINGAREA, LOTSIZE, GOLFCOURSELOT, SCHOOLRANK, PRCNTBCHLR, and TREND. These results T h e I m p a c t o f S h u t t e r e d G o l f C o u r s e s o n P r o p e r t y V a l u e s 1 3 indicate that buyers de-value homes in communities where the community closed its associated golf course. The negative correlation between price and the qualitative variable indicating that a sample transaction occurred in a community that was previously a GCC provides evidence of this buyer sentiment. Recall that the average home in our sample is 18 years old, has two bathrooms, 2 2 2,776 ft located on a 18,674.37 ft (0.43 acre) lot, where the schools are ranked 9 out of 10 and 39% of the residents hold at least a bachelor’s degree. Given these parameter estimates, we estimate that the average home adjacent to an operating golf course would sell for $380,236.30. That same home adjacent to a closed golf course would sell for $316,384.40. This results in a $63,851.90 or 17% value diminution due to the removal of a residential amenity. We find evidence that in GCCs in Shelby County, Alabama, the closing of the golf course, age and size of the home, and the size and location of the lot partially explain the variability in market prices after controlling for block group demographics and economic trends. As homes age, they command a lower market price and larger homes command higher market prices. This result may be due, in part, to the significant amount of new construction in the county throughout the sample period. Additionally, these newer homes tend to be larger than existing homes. C o n c l u s i o n This study is unique because we analyze the impact of the removal of an environmental amenity. Several studies have considered the impact of a golf course and views of a golf course on surrounding home values, but we are the first to our knowledge to analyze the impact of the closure of a golf course. We find evidence that the closure of a golf course reduces subsequent sales prices by 17%. In addition, we find that the age, size of the home, size and location of the lot, and block group demographics are significant determinants of market value. We considered whether sustainability factored into our results by analyzing the disposition of the land in our sample of golf courses. We find that the ranking of the affiliated public school and proximity to the golf course affect the sale price. This result indicates that while developers consider green spaces in their planning processes, removing environmental amenities can have a value-destroying effect on neighboring homes. Redevelopment and its impact on property values is an important consideration when golf courses close. In some communities, zoning limitations eliminate the possibility of constructing homes on closed fairways. This was the case for one of the GCCs in our sample, which eventually reopened due to limited redevelopment options, among other issues. In other communities, GCCs are being ‘‘repurposed’’ into smaller, upgraded courses, new construction, and public use projects, such as parks and ball fields. J O S R E V o l . 1 1 2 0 1 9 u u 1 4 Y a t e s a n d C o w a r t We also compare GCC properties to non-GCC properties and find that homes in GCCs sell at a premium relative to homes in non-GCCs when there is no change in the status of the affiliated golf course. Future research should compare the removal of other environmental amenities to determine if this result is unique to golf courses. The study serves as a complement to earlier work assessing the value of certain amenities as it relates to housing values, as well as the value of interior lots versus non-interior lots. Our finding that closing a golf course in a residential community is a value-destroying event could have implications for real estate developers and golf course owners facing a dip in rounds played and country club memberships. Other researchers have implied that there is a ‘‘tipping point’’ whereby golf courses are more valuable for the redevelopment potential of the land where they are located as opposed to the revenues generated by the course itself. However, some developers have faced backlash from property owners concerned about a decline in property values upon closure of their neighborhood golf course. Our study provides evidence to homeowners and developers that support this fear. Ultimately, a decline in the number of golfers over recent years has led to the closure of many golf courses throughout the country. While we suggest that this removal of a residential amenity can lead to a decrease in home value, the greater issue is one of sustainability. That is, one could argue that a goal of sustainable real estate and sustainable property development is to provide long-term benefits to the community while minimizing environmental impacts. Changes in consumer preferences appear to indicate that golf courses may no longer provide such a benefit in some communities. In those communities, the challenge is to determine how to repurpose GCCs to provide long-term benefits that match consumer preferences. Future research in sustainability should consider how to address a community’s sustainability needs in the face of changing preferences. R e f e r e n c e s Baranzini, A. and C. Schaerer. A Sight for Sore Eyes: Assessing the Value of View and Landscape Use on the Housing Market. Journal of Housing Economics, 2011, 20:3. Bauerlein, V. The Most Pressing Question in the Suburbs: What Do You Do When the Golf Course Shuts Down? The Wall Street Journal, July 20, 2019, p. 3. Bond, M.T., V.L. Seiler, and M.J. Seiler. Residential Real Estate Prices: A Room with a View. Journal of Real Estate Research, 2002, 23:1 / 2, 129–38. Bourassa, S.C., M. Hoesli, and J. Sun. The Price of Aesthetic Externalities. Journal of Real Estate Literature, 2005, 13:2, 167–87. ——. The Price of Aesthetic Externalities. The Appraisal Journal, 2006, 74:1, 14–29. Davies, G. An Econometric Analysis of Residential Amenity. Urban Studies, 1974, 217– Do, A.Q. and G. Grudnitski. Golf Courses and Residential House Prices: An Empirical Examination. The Journal of Real Estate Finance and Economics, 1995, 10:3, 261–70. Foundation, T.N . The 2016 U.S. Golf Economy Report. Jupiter, FL: The National Golf Foundation, 2018. T h e I m p a c t o f S h u t t e r e d G o l f C o u r s e s o n P r o p e r t y V a l u e s 1 5 Fraser, S.P. and M.T. Allen. Golf Course Design and Real Estate Values: The Impact of Cart Paths on Condominium Prices. The Appraisal Journal, 2017, 96–103. Grudnitski, G. The Effect of Course Type on Housing Prices. The Appraisal Journal, 2003, 71:2, 145–49. Ming, Y.S. and C.C. Hian. Obstruction of View and its Impact on Residential Apartment Prices. Pacific Rim Property Research Journal, 2005, 11:3, 299–315. Mittal, J. and S. Byahut. Scenic Landscapes, Visual Accessibility and Premium Values in a Single Family Housing Market: A Spatial Hedonic Approach. Environment and Planning B: Urban Analytics and City Science, 2019, 46:1. Mothorpe, C. and D. Wyman. Collapse: The Decline and Fall of Master Planned Golf Course Communities. Journal of Property Investment and Finance, 2017, 35:6, 638–51. Nelson, J.P. Valuing Rural Recreation Amenities: Hedonic Prices for Vacation Rental Houses at Deep Creek Lake, Maryland. Agricultural and Resource Economics Review, 2010, 39:3, 485–504. Poudyal, N ., D.G. Hodges, J. Fenderson, and W. Tarkington. Realizing the Economic Value of a Forested Landscape in a Viewshed. Southern Journal of Applied Forestry, 2010, 34: 2, 72–78. Rauterkus, S.Y. and N .G. Miller. Residential Land Values and Walkability. Journal of Sustainable Real Estate, 2011, 3:1, 23–43. Rodriguez, M. and C.F. Sirmans. Quantifying the Value of a View in Single-Family Housing Markets. Appraisal Journal, 1994, 62. Simons, R.A. and J.D. Saginor. A Meta-Analysis of the Effect of Environmental Contamination and Positive Amenities on Residential Real Estate Values. Journal of Real Estate Research, 2006, 28:1, 71–104. Sirmans, G.S., D.A. Macpherson, and E.N . Zietz. The Composition of Hedonic Pricing Models. Journal of Real Estate Literature, 2005, 13:1, 3–43. Wen, H., Y. Zhang, and L. Zhang. Assessing Amenity Effects of Urban Landscapes on Housing Price in Hangzhou, China. Urban Forestry and Urban Greening, 2015, 14:4, 1017–26. Wyman, D. and S. Sperry. The Million Dollar View: A Study of Golf Course, Mountain and Lake Lots. Appraisal Journal, 2010, 78:2. All articles published in JOSRE are distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Stephanie R. Yates, University of Alabama at Birmingham, Birmingham, AL 35294 or sryates@uab.edu. Lary B. Cowart, University of Alabama at Birmingham, Birmingham, AL 35294 or lcowart@uab.edu. J O S R E V o l . 1 1 2 0 1 9 u u http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Sustainable Real Estate Taylor & Francis

The Impact of Shuttered Golf Courses on Property Values

Loading next page...
 
/lp/taylor-francis/the-impact-of-shuttered-golf-courses-on-property-values-8JnlBg6yeV

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Taylor & Francis
Copyright
© 2019 American Real Estate Society
ISSN
1949-8284
DOI
10.22300/1949-8276.11.1.2
Publisher site
See Article on Publisher Site

Abstract

The Impact of Shuttered Golf Courses on Property Values Stephanie R. Yates and Lary B. Cowart A u t h o r s A b s t r a c t We measure the impact of a golf course as a residential amenity on surrounding home values in several communities in Shelby County, Alabama. We compare the values of homes in golf course communities (GCCs) and non-golf course communities, as well as the values of homes within these communities before and after the golf course closes. Using a methodology similar to Bond, Seiler, and Seiler (2002), we examine the sales prices of homes within GCCs both before and after a golf course closure to see how the closure affects the sales prices of homes and test for the significance of that difference. We calculate the difference in value for homes in GCCs before and after the golf course is closed and test for the significance of that difference. We estimate the degree to which specific factors explain the variance in home prices in these communities before and after the golf course closed. We find that homes in GCCs sell at a 9% premium compared to homes in non-GCCs. We also find that home prices in GCCs decrease by 17% after the related golf course closes; home prices for properties adjacent to a golf course diminish as well. K e y w o r d s golf courses, golf course communities, value diminution, residential amenities As noted in a July 2019 Wall Street Journal article (Bauerlein, 2019), golf has declined in popularity in recent years, resulting in a number of golf course closures. This leaves developers and homeowners in a difficult position and scrambling to preserve home values. We seek to quantify the impact of this trend using a sample of home sales in Shelby County, Alabama. Earlier research confirmed the widely held belief that houses in golf course communities (GCCs) were more highly valued and sold for higher prices than otherwise equivalent houses in non-golf course communities. Most of that research occurred in the 1990s and early 2000s when both golf play and housing prices were rising. Then, the Great Recession became an inflection point for both golf and the housing market. According to the National Golf Foundation (2018), the high of 30.6 million U.S. golfers in 2003 was reduced to 24.7 million by 2014. Furthermore, as the number of golfers between the ages of 18 and 34 had declined 30% over the past 20 years, there was concern regarding the impact this might have on the sport despite the number of baby boomers nearing retirement age. There were more than 13,700 golf courses nationwide in 2015, according to the National Golf Foundation, down from 16,000 in 2009, of which 3,200 had T h e I m p a c t o f S h u t t e r e d G o l f C o u r s e s o n P r o p e r t y V a l u e s 3 communities built around them. That is a 14% drop in the total number of golf courses within a six-year period. If houses in a GCC command a premium, what happens to that premium when the golf course closes? R e v i e w o f t h e L i t e r a t u r e Location has long been a key factor in determining the value and desirability of residential real estate. Many published works consider the impact of available amenities on housing values. Researchers have largely subdivided those amenities into environmental amenities and residential amenities. Environmental amenities are the features associated with the surrounding neighborhood and residential amenities specific to the homes themselves. The most commonly studied environmental amenities are views. Several studies consider the impact of views on home value; therefore, we further subdivide this literature into those articles that analyze water views versus non-water views. Amenities Davies (1974) uses 1968 data from the town of Nottingham to analyze environmental amenities and accessibility based on the theory that consumers seeking to maximize their utility derive it from residential characteristics. This model indicates that the number of bedrooms, central heating, plot size, open space in the immediate vicinity, the absence of industrial nuisance, and seclusion are positive factors, whereas poor dwelling quality and proximity to a main road are negative factors. Sirmans, Macpherson, and Zietz (2005) review studies using hedonic modeling to estimate house prices. Hedonic modeling identifies the factors that contribute to prices. The factors the authors identify as positive and statistically significant in this study are slanted versus flat roof, sprinkler system, garden bath, separate shower stall, double oven, and gated community, while not having attic space negatively affects selling price. Many studies have examined the impact of externalities on property values. Bourassa, Hoesli, and Sun (2005, 2006) studied the prices of three specific aesthetic externalities: a water view, the appearance of nearby improvements, and the quality of landscaping in the neighborhood in a sample of residential sales from 1986 to 1996 in Auckland, Christchurch, and Wellington New Zealand. The impact of the factors that they identify changes with the real estate cycle but that the premium placed on water views is highest in areas where the supply is lowest. Simons and Saginor (2006) hypothesize that markets can internalize proximity to positive factors and this effect can be determined. Their meta-analysis using 620 observations from 17 peer-reviewed articles indicates that properties located near a positive attribute such as beach frontage, water views, parks, and golf courses command a 25% premium. Nelson (2010) estimates hedonic prices for summer and winter rentals for vacation houses located near a lake and ski-golf resort in rural western Maryland. There, access to recreation impacts rental offers. Specifically, lakefront locations command a $1,000–$1,200 weekly premium and ski-slope access is valued at $500–$600 per week. These premiums amount to J O S R E V o l . 1 1 2 0 1 9 u u 4 Y a t e s a n d C o w a r t 42%–44% of the total average value for lakefront locations and 27%–32% of total average value for ski-slope access. In Birmingham, Alabama, Rauterkus, and Miller (2011) find that land values increase with walkability but that this result reverses as neighborhoods become more car-dependent. That is, in generally walkable areas, values increase with walkability. However, in generally car- dependent areas, values decrease with walkability. Grudnitski (2003) employs a semi-logarithmic form of regression to study the impact on sales price of residential properties in different types of GCCs. He compares the sales prices of residential properties that were not located in GCCs and found 12.5% price premiums for houses located in private GCCs, 6% for houses located in semiprivate GCCs, and 5.7% for houses located in public GCCs. Fraser and Allen (2017) study 241 condominium transactions occurring between 2011 and 2015 in three southwest Florida communities and find that after controlling for unit age, size, bedrooms, bathrooms, garage spaces, and time trend, proximity to a cart path may reduce transaction prices by 5.1%. The authors define proximate paths as those where the golf cart path is routed on the dwelling side of the teeing area, fairway, or green of the condominium unit. The authors then further analyze the golf condominium units that are proximate to cart paths and differentiate between those that are located at a tee box versus those that are located near a green. They find evidence suggesting an even greater discount such that units with proximate cart paths near greens sell for a discount of 11.6% and those near tees sell for discounts of 8.6%. View Lots Rodriguez and Sirmans (1994) use multiple regression analysis to estimate that a good view in Fairfax County, Virginia adds about 8% to the value of a single- family house. Do and Grudnitski (1995) find that a golf course location adds 7.6% to a property’s sales price. Ming and Hian (2005) find construction that would obstruct the view of existing units depresses the prices of the obstructed development by around 8% in the long term. Poudyal, Hodges, Fenderson, and Tarkington (2010) find that increasing the size of forest area visible from a house by one acre increased the house price by $30. Wyman and Sperry (2010) find a hierarchy in pricing with lakefront lots commanding the highest premiums (124%– 287%), followed by lake views (94%–133%) and then golf course views (42%– 85%). Baranzini and Schaerer (2011) use hedonic modeling to assess the value of view and landscape use in the Geneva, Switzerland rental market and find that accessibility to various environmental amenities, landscape use, and size, as well as the view of them have a statistically significant impact on rents. In addition, they find that proximity, size, and view of water amenities increase rents. However, a view of urban parks and building-covered areas negatively affect rents. Wen, Zhang, and Zhang (2015) find that parks, mountains, rivers, and lakes have significant positive effects on housing price. Mittal and Byahut (2019) use a geographically-weighted regression model that indicates an average 3.4% price premium on the mean value of homes in the study area. Mothorpe and Wyman (2017) define an interior lot as one without lake frontage, golf frontage or a mountain view and find a collapse in the value of interior lots in GCCs compared to interior lots elsewhere in Pickens County, South Carolina from 2000 to 2016. T h e I m p a c t o f S h u t t e r e d G o l f C o u r s e s o n P r o p e r t y V a l u e s 5 Our review of the literature indicates that residential and environmental amenities have a positive and significant impact on residential housing prices. Our focus is on one particular environmental amenity—golf courses. Earlier researchers found that while golf courses do not command as much of a premium as lakefront lots and lake views, they may command a price premium of 8% or more. Further, they find that obstructing a view that was previously unobstructed may depress prices by 8%. We estimate the impact of eliminating a positive amenity (proximity to a golf course) using a sample of homes in Shelby County, Alabama. D a t a a n d M e t h o d o l o g y We analyze data related to six GCCs in Shelby County, Alabama. Shelby County is one of seven counties that form the Birmingham-Hoover MSA and has been the fastest growing county in Alabama every decade since 1970. The county had an estimated population of 213,605 in 2017. Founded in 1818, Shelby County covers a total area of 810 square miles. Economically, the median household income for Shelby County was $68,380 according to the 2010 Census and only 7.4% of the population was below the poverty line compared to $59,039 and 14.5% nationwide respectively. There are 288 golf courses in Alabama. Nine of these courses are located in Shelby County. The golf courses included in the study are Ballantrae Golf Club, Eagle Point Golf Club, Greystone Golf and Country Club, Inverness Country Club, Montevallo Golf Club, and Shoal Creek Golf Club. Our golf course and associated GCC of interest is Eagle Point Golf Club. Eagle Point Golf Club was a 71-par, 6,470-yard golf course that opened in 1990 and closed May 1, 2016. The three golf courses in Shelby County that we did not study do not have associated communities. We collected sales data for the subdivisions in which these golf courses are located from the MLS according to the dates in Exhibit 1. We remove all transactions involving quit claim deeds and any other transactions that are unqualified such that our dataset contains only transactions that occurred in an open, competitive market between typically informed and motivated buyers and sellers. We also remove all sales of vacant lots. This leaves only improved residential real property in our final dataset, which are parcels of real estate having an inhabitable structure located on it. In order to address the potential for heteroscedasticity in our model, we removed several outliers. Specifically, we removed 38 transactions with a sales price of less than $100,000, as well as those with a sales price greater than $1,000,000. We anticipate that the closure of a golf course within a residential community negatively impacts home prices. Additionally, we anticipate, in accordance with Bond, Seiler, and Seiler (2002), that sale price is positively related to the number of bathrooms, living space, lot size, and frontage. In contrast to Bond, Seiler, and Seiler, however, we anticipate a negative correlation between sales price and age given the high demand for new construction in this area. The data are summarized in Exhibit 2. J O S R E V o l . 1 1 2 0 1 9 u u 6 Y a t e s a n d C o w a r t Exhibit 1 u Sample Communities Golf Course Community Transactions Mean Sale Price Eagle Point Golf Club 1990–2016 42 $389,027 Heatherwood Golf Club 1986–2009 2016–present 80 $419,017 Greystone Golf and Country Club 1991–present 766 $391,105 Inverness Country Club 1973–present 123 $248,862 Montevallo Golf Club 1955–present 1 $300,000 Shoal Creek Golf Club 1977–present 37 $482,004 Ballantrae Golf Club 2004–present 345 $279,307 Brook Highland Never 342 $352,661 Meadowbrook Never 223 $266,514 Total 1,959 Mean $344,357 Notes: This table presents information regarding the communities in our sample. Heatherwood Golf Club reopened as Heatherwood Hills Country Club on October 8, 2016. This club has two associated courses: Founders Course and Legacy Course. We define our variables as follows: PRICE 5 Market sales price of the home; FORMER 5 Qualitative variable indicating whether or not the sale occurred in a neighborhood that was formerly a GCC; AGE 5 Age of the home (in years); SIZE 5 Size of the home. We use two different variables to measure size: BATHROOMS 5 Number of bathrooms and LIVING AREA 5 Total heated square footage of living space; LOT 5 Size of the lot measured as either: LOT SIZE 5 Total square footage of the lot or LOT DEPTH 5 Horizontal distance between the front and rear property lines measured in feet; LOCATION 5 Location of the home measured as either SCHOOLRANK 5 The Great Schools ranking of the associated high school (1– 10) or GOLFCOURSEHOME 5 Qualitative variable indicating whether the home is located adjacent to a golf course. TREND 5 Monthly time trend variable that captures general market price changes over the study period as in Fraser and Allen (2017). This approach is a means of de-trending the data and thereby controlling for the linear nature of inflation and its impact on sale prices. DEMOGRAPHICS 5 Block group-level demographics of the residents from the 2017 American Community Survey: T h e I m p a c t o f S h u t t e r e d G o l f C o u r s e s o n P r o p e r t y V a l u e s 7 Exhibit 2 u Descriptive Statistics N Min. Mean Max. Std. Dev. Panel A: Continuous variables AGE 1,951 0 18 101 8.76 BATHROOMS 1,909 0 2 6 1.16 LIVINGAREA 1,951 992 2,776 7,367 994.30 LOTDEPTH 1,959 0 152 878 85.97 LOTSIZE 1,957 0 18,674 367,211 22,940.89 MEDIANAGE 1,954 30.1 43 54.2 5.34 MEDIANHHI 1,954 29,481 105,068 147,774 27,621.50 PRCNTBCHLR 1,954 10.22 39 50.74 7.30 SALEPRICE 1,959 100,000 344,356 985,000 161,259 SCHOOLRANK 1,959 3 9 10 0.51 Panel B: Binary variables 1 0 N Frequency Percentage Frequency Percentage FORMERGCC 1,959 78 3.98% 1,881 96.02% GOLFCOURSEHOME 1,959 145 7.40% 1,814 92.60% RECESSION 1,955 89 4.55% 1,866 95.45% Notes: This table presents the details of our sample of homes in nine different Shelby County, Alabama golf course communities. MEDIANAGE 5 Measured as age (in median years); MEDIANHHI 5 Income (median household income in dollars); PRCNTBCHLR 5 Educational attainment (as % of the population that has earned a bachelor’s degree); and RECESSION 5 Qualitative variable indicating whether the sale occurred during a recession. We identify recessions in our sample as the period from December 2007 to June 2009. To begin our examination of the impact of a residential amenity—a golf course— on home values, we examine the transactions in our sample where either there was never an associated golf course or the golf course never closed. We conduct an analysis of variance (ANOVA) test to compare the mean values of our key variables across the three main subsets of properties. These three subsets are homes located in communities that: (1) have always been GCCs; (2) were never GCCs; and (3) were previously GCCs. We refer to the first group as GCCs, the second group as non-GCCs, and the third group as closed GCCs. We exclude 24 J O S R E V o l . 1 1 2 0 1 9 u u 8 Y a t e s a n d C o w a r t Exhibit 3 u Comparison of Golf Course Communities to Non-Golf Course Communities and Former Golf Course Communities Difference Difference Closed GCCs and GCCs and GCCs Non-GCCs GCCs Non-GCCs Closed GCCs AGE 16.31 23.52 21.69 7.21*** 5.38*** BATHROOMS 2.27 2.57 3.82 0.30*** 1.55*** LIVINGAREA 2,782.35 2,646.37 3,415.34 2135.98** 632.99*** LOTDEPTH 139.36 172.49 199.10 33.13*** 59.74*** LOTSIZE 16,322.57 21,123.76 34,837.14 4,801.19*** 18,514.60*** MEDIANAGE 44.99 39.73 41.46 25.27*** 23.53*** MEDIANHHI 113,686.40 83,558.74 117,767.90 230,127.60*** 4,081.50 PRCNTBCHLR 38.73 42.53 34.36 3.80*** 24.36*** SALEPRICE 349,600.20 318,659.70 402,702.70 230,940.50 53,102.50** SCHOOLRANK 9.60 9.00 9.00 20.60*** 20.60*** Observations 1,272 565 80 Notes: This table presents the mean values for the golf course community (GCC) homes the non-golf course community homes (non-GCCs), and homes in former GCCs. *Statistically significant at 90%. **Statistically significant at 95%. ***Statistically significant at 99%. home sales that occurred in a GCC prior to golf course closure and 17 sales that occurred in a GCC after the corresponding golf course reopened. We present the results of this comparison in Exhibit 3. We find statistically significant differences in nearly all of our key variables for GCCs versus non- GCCs. Specifically, we find in GCCs, the living area, age, household income, price, and school rank are statistically significantly higher than in non-GCCs. Conversely, we find evidence that in non-GCCs, the homes are older, have more bathrooms, and larger lots. We focus on the statistically significant difference in the mean SALEPRICE for these two subsets of our data. As shown in Exhibit 3, we find that homes in GCCs sell at a $30,940.50 premium compared to homes in non-GCCs. Next, we consider the impact of the removal of a residential amenity. We first analyze golf course adjacent properties where we most expect to see price diminution after the removal of a residential amenity. The results are in Exhibit 4. In our full sample, the mean SALEPRICE for golf course adjacent properties is $556,305.00 compared to $328,794.40 for non-golf course adjacent properties, which results in a 41% premium. However, when we differentiate between homes adjacent to open versus closed golf courses, we find a significant difference in T h e I m p a c t o f S h u t t e r e d G o l f C o u r s e s o n P r o p e r t y V a l u e s 9 Exhibit 4 u Comparison of Golf Course Adjacent Within Current and Former Golf Course Communities All Golf Course Difference GCA Adjacent (GCA) GCA Homes GCA Homes in Homes Open and Homes in Open GCCs Closed GCCs Closed GCCs AGE 22.13 21.22 26.52 5.30*** BATHROOMS 3.54 3.50 3.72 0.22*** LIVINGAREA 4,016.25 4,110.17 3,562.96 2547.21*** LOTDEPTH 214.86 219.37 193.09 226.28*** LOTSIZE 34,976.85 35,671.15 31,473.77 24,197.38*** MEDIANAGE 47.52 48.13 44.58 23.55*** MEDIANHHI 120,116.10 118,324.70 128,761.70 10,437.00 PRCNTBCHLR 38.98 39.73 35.38 24.35*** SALEPRICE 556,305.00 586,426.10 410,937.70 2175,488.40** SCHOOLRANK 9.53 9.64 9.00 20.64*** Observations 134 111 23 Notes: This table presents the mean values for the golf course community (GCC) homes the non-golf course community homes (non-GCCs), and homes in former GCCs. *Statistically significant at 90%. **Statistically significant at 95%. ***Statistically significant at 99%. sale prices. If the golf course is operating, the mean SALEPRICE is $586,426.10. If the golf course has ceased operations, the mean sale price falls to $410,937.70. This statistically significant price difference of $175,488.40, or 30%, suggests that the value of golf course adjacent properties suffer when the adjacent golf course closes. Using our full sample, we seek to determine the factors that explain the variation in house prices. In the context of the removal of a residential amenity, we hypothesize that a key factor is the closure of a golf course within a community. Our explanatory variable of interest is FORMERGCC, which is a binary variable if the transaction occurred in a community that was formerly a GCC. We estimate the following model: SALEPRICE 5 a 1 b FORMERGCC 1 b AGE 1 b SIZE 1 i 1 1 2 2 3 3 1 b LOT 1 b LOCATION 1 b TREND 4 4 5 5 6 6 1 b DEMOGRAPHICS 1 b RECESSION 1 «. 7 7 8 8 J O S R E V o l . 1 1 2 0 1 9 u u 1 0 Y a t e s a n d C o w a r t Exhibit 5 u Correlation between Housing Variables AGE BATHS FRMR GCA LIVING DEPTH SIZE RANK AGE 1.0000 BATHS 0.0746*** 1.0000 FRMRGCC 0.0619*** 0.2086*** 1.0000 GCA 0.1062*** 0.2652*** 0.1827*** 1.0000 LIVING 0.0373* 0.7645*** 0.1308*** 0.3388*** 1.0000 LOTDEPTH 0.3354*** 0.4315*** 0.1133*** 0.1973*** 0.4444*** 1.0000 LOTSIZE 0.2158*** 0.4237*** 0.1458*** 0.1919*** 0.5434*** 0.5811*** 1.0000 RANK 0.1326*** 0.0079 20.1553*** 0.0756*** 0.1916 20.0086 20.0606*** 1.0000 Notes: This table presents pairwise correlation coefficients for the housing variables. *Statistically significant at 90%. **Statistically significant at 95%. ***Statistically significant at 99%. T h e I m p a c t o f S h u t t e r e d G o l f C o u r s e s o n P r o p e r t y V a l u e s 1 1 Exhibit 6 u Full Regression Results Beta t-Stat. Tolerance Intercept 2393,002.90 210.26*** FORMERGCC 258,669.16 25.25*** 0.7963 AGE 22,332.29 210.28*** 0.7027 BATHROOMS 30,917.07 12.80*** 0.3601 LIVINGAREA 94.39 29.79*** 0.2768 LOTDEPTH 10.61 0.40 0.5356 LOTSIZE 0.83 7.56*** 0.4766 GOLFCOURSELOT 63,741.95 8.64*** 0.7892 SCHOOLRANK 36,226.88 8.33*** 0.5644 TREND 165.96 2.86*** 0.7414 MEDIANAGE 567.04 1.48 0.6654 MEDIANHHI 20.12 21.50 0.5599 PRCNTBCHLR 964.49 3.41*** 0.6626 RECESSION 11,650.12 1.30 0.7764 N 1,898 Adj. R 0.7981 F 577.93 Notes: In this table, we present the results from a regression of our independent variables against the transaction-based dependent variable, market sales price. *Statistically significant at 90%. **Statistically significant at 95%. ***Statistically significant at 99%. Before estimating our regression equation, we check for correlation. The results are in Exhibit 5. We find significant levels of correlation between several variables. R e s u l t s Univariate and simple multivariate regressions not shown here indicate that all of our alternative measures of size, lot, and location have significant explanatory power over sales price. Based on these results, we use all of these measures in our full model. Thus, we find that both the number of bathrooms and the overall size of the home have significant explanatory power over sales price. In addition, with regard to the size of the lot, overall lot size is significant but not lot depth. Finally, given that location matters, we identified golf course adjacent lots and the ranking of the affiliated public high school. Both of these location-based variables have significant explanatory power over sales price. J O S R E V o l . 1 1 2 0 1 9 u u 1 2 Y a t e s a n d C o w a r t Exhibit 7 u Full Regression Results with Only Statistically Significant Variables Variable Beta Std. Error t-Stat. Intercept 2363,315.00 35,158.55 210.33*** FORMERGCC 263,851.92 10,772.65 25.93*** AGE 22,277.12 215.01 210.59*** BATHROOMS 31,403.72 2,379.88 13.20*** LIVINGAREA 94.41 3.16 29.84*** LOTSIZE 0.82 0.19 8.44*** GOLFCOURSELOT 65,114.78 7,173.07 9.08*** SCHOOLRANK 35,410.93 3,751.37 9.44*** TREND 129.78 50.95 2.55** PRCNTBCHLR 891.31 271.33 3.28*** N 1,899 Adj. R 0.7989 F 834.01 Notes: In this table, we present the results from a regression of our remaining, statistically significant independent variables against the transaction-based dependent variable, SALEPRICE. *Statistically significant at 90%. **Statistically significant at 95%. ***Statistically significant at 99%. We present the results of our full regression of our independent variables against the transaction-based dependent variable SALEPRICE, in Exhibit 6. Due to the significant correlation between explanatory variables as shown in Exhibit 5, we check for multicollinearity among the explanatory variables by calculating tolerance as the inverse of the variance inflation factor. Given that all variables have a tolerance greater than 0.10, we do not detect multicollinearity in the model. The significant variables are FORMERGCC, AGE, BATHROOMS, LIVINGAREA, LOTSIZE, GOLFCOURSELOT, SCHOOLRANK, TREND, and PRCNTBCHLR. The beta coefficient for FORMERGCC, our variable of interest, is negative and statistically significant. This indicates that the closing of a golf course results in value diminution. This is a key result as our results also support earlier studies suggesting that golf courses are a value adding residential amenity. This new finding suggests that the removal of a residential amenity can have an opposite effect. As in Bond, Seiler, and Seiler (2002), we estimate the regression again with just the remaining significant independent variables included in the model in order to find an unbiased estimate of the true beta values. Exhibit 7 presents the results of our analysis. We find unbiased beta estimates for nine independent variables: FORMERGCC, AGE, BATHROOMS, LIVINGAREA, LOTSIZE, GOLFCOURSELOT, SCHOOLRANK, PRCNTBCHLR, and TREND. These results T h e I m p a c t o f S h u t t e r e d G o l f C o u r s e s o n P r o p e r t y V a l u e s 1 3 indicate that buyers de-value homes in communities where the community closed its associated golf course. The negative correlation between price and the qualitative variable indicating that a sample transaction occurred in a community that was previously a GCC provides evidence of this buyer sentiment. Recall that the average home in our sample is 18 years old, has two bathrooms, 2 2 2,776 ft located on a 18,674.37 ft (0.43 acre) lot, where the schools are ranked 9 out of 10 and 39% of the residents hold at least a bachelor’s degree. Given these parameter estimates, we estimate that the average home adjacent to an operating golf course would sell for $380,236.30. That same home adjacent to a closed golf course would sell for $316,384.40. This results in a $63,851.90 or 17% value diminution due to the removal of a residential amenity. We find evidence that in GCCs in Shelby County, Alabama, the closing of the golf course, age and size of the home, and the size and location of the lot partially explain the variability in market prices after controlling for block group demographics and economic trends. As homes age, they command a lower market price and larger homes command higher market prices. This result may be due, in part, to the significant amount of new construction in the county throughout the sample period. Additionally, these newer homes tend to be larger than existing homes. C o n c l u s i o n This study is unique because we analyze the impact of the removal of an environmental amenity. Several studies have considered the impact of a golf course and views of a golf course on surrounding home values, but we are the first to our knowledge to analyze the impact of the closure of a golf course. We find evidence that the closure of a golf course reduces subsequent sales prices by 17%. In addition, we find that the age, size of the home, size and location of the lot, and block group demographics are significant determinants of market value. We considered whether sustainability factored into our results by analyzing the disposition of the land in our sample of golf courses. We find that the ranking of the affiliated public school and proximity to the golf course affect the sale price. This result indicates that while developers consider green spaces in their planning processes, removing environmental amenities can have a value-destroying effect on neighboring homes. Redevelopment and its impact on property values is an important consideration when golf courses close. In some communities, zoning limitations eliminate the possibility of constructing homes on closed fairways. This was the case for one of the GCCs in our sample, which eventually reopened due to limited redevelopment options, among other issues. In other communities, GCCs are being ‘‘repurposed’’ into smaller, upgraded courses, new construction, and public use projects, such as parks and ball fields. J O S R E V o l . 1 1 2 0 1 9 u u 1 4 Y a t e s a n d C o w a r t We also compare GCC properties to non-GCC properties and find that homes in GCCs sell at a premium relative to homes in non-GCCs when there is no change in the status of the affiliated golf course. Future research should compare the removal of other environmental amenities to determine if this result is unique to golf courses. The study serves as a complement to earlier work assessing the value of certain amenities as it relates to housing values, as well as the value of interior lots versus non-interior lots. Our finding that closing a golf course in a residential community is a value-destroying event could have implications for real estate developers and golf course owners facing a dip in rounds played and country club memberships. Other researchers have implied that there is a ‘‘tipping point’’ whereby golf courses are more valuable for the redevelopment potential of the land where they are located as opposed to the revenues generated by the course itself. However, some developers have faced backlash from property owners concerned about a decline in property values upon closure of their neighborhood golf course. Our study provides evidence to homeowners and developers that support this fear. Ultimately, a decline in the number of golfers over recent years has led to the closure of many golf courses throughout the country. While we suggest that this removal of a residential amenity can lead to a decrease in home value, the greater issue is one of sustainability. That is, one could argue that a goal of sustainable real estate and sustainable property development is to provide long-term benefits to the community while minimizing environmental impacts. Changes in consumer preferences appear to indicate that golf courses may no longer provide such a benefit in some communities. In those communities, the challenge is to determine how to repurpose GCCs to provide long-term benefits that match consumer preferences. Future research in sustainability should consider how to address a community’s sustainability needs in the face of changing preferences. R e f e r e n c e s Baranzini, A. and C. Schaerer. A Sight for Sore Eyes: Assessing the Value of View and Landscape Use on the Housing Market. Journal of Housing Economics, 2011, 20:3. Bauerlein, V. The Most Pressing Question in the Suburbs: What Do You Do When the Golf Course Shuts Down? The Wall Street Journal, July 20, 2019, p. 3. Bond, M.T., V.L. Seiler, and M.J. Seiler. Residential Real Estate Prices: A Room with a View. Journal of Real Estate Research, 2002, 23:1 / 2, 129–38. Bourassa, S.C., M. Hoesli, and J. Sun. The Price of Aesthetic Externalities. Journal of Real Estate Literature, 2005, 13:2, 167–87. ——. The Price of Aesthetic Externalities. The Appraisal Journal, 2006, 74:1, 14–29. Davies, G. An Econometric Analysis of Residential Amenity. Urban Studies, 1974, 217– Do, A.Q. and G. Grudnitski. Golf Courses and Residential House Prices: An Empirical Examination. The Journal of Real Estate Finance and Economics, 1995, 10:3, 261–70. Foundation, T.N . The 2016 U.S. Golf Economy Report. Jupiter, FL: The National Golf Foundation, 2018. T h e I m p a c t o f S h u t t e r e d G o l f C o u r s e s o n P r o p e r t y V a l u e s 1 5 Fraser, S.P. and M.T. Allen. Golf Course Design and Real Estate Values: The Impact of Cart Paths on Condominium Prices. The Appraisal Journal, 2017, 96–103. Grudnitski, G. The Effect of Course Type on Housing Prices. The Appraisal Journal, 2003, 71:2, 145–49. Ming, Y.S. and C.C. Hian. Obstruction of View and its Impact on Residential Apartment Prices. Pacific Rim Property Research Journal, 2005, 11:3, 299–315. Mittal, J. and S. Byahut. Scenic Landscapes, Visual Accessibility and Premium Values in a Single Family Housing Market: A Spatial Hedonic Approach. Environment and Planning B: Urban Analytics and City Science, 2019, 46:1. Mothorpe, C. and D. Wyman. Collapse: The Decline and Fall of Master Planned Golf Course Communities. Journal of Property Investment and Finance, 2017, 35:6, 638–51. Nelson, J.P. Valuing Rural Recreation Amenities: Hedonic Prices for Vacation Rental Houses at Deep Creek Lake, Maryland. Agricultural and Resource Economics Review, 2010, 39:3, 485–504. Poudyal, N ., D.G. Hodges, J. Fenderson, and W. Tarkington. Realizing the Economic Value of a Forested Landscape in a Viewshed. Southern Journal of Applied Forestry, 2010, 34: 2, 72–78. Rauterkus, S.Y. and N .G. Miller. Residential Land Values and Walkability. Journal of Sustainable Real Estate, 2011, 3:1, 23–43. Rodriguez, M. and C.F. Sirmans. Quantifying the Value of a View in Single-Family Housing Markets. Appraisal Journal, 1994, 62. Simons, R.A. and J.D. Saginor. A Meta-Analysis of the Effect of Environmental Contamination and Positive Amenities on Residential Real Estate Values. Journal of Real Estate Research, 2006, 28:1, 71–104. Sirmans, G.S., D.A. Macpherson, and E.N . Zietz. The Composition of Hedonic Pricing Models. Journal of Real Estate Literature, 2005, 13:1, 3–43. Wen, H., Y. Zhang, and L. Zhang. Assessing Amenity Effects of Urban Landscapes on Housing Price in Hangzhou, China. Urban Forestry and Urban Greening, 2015, 14:4, 1017–26. Wyman, D. and S. Sperry. The Million Dollar View: A Study of Golf Course, Mountain and Lake Lots. Appraisal Journal, 2010, 78:2. All articles published in JOSRE are distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Stephanie R. Yates, University of Alabama at Birmingham, Birmingham, AL 35294 or sryates@uab.edu. Lary B. Cowart, University of Alabama at Birmingham, Birmingham, AL 35294 or lcowart@uab.edu. J O S R E V o l . 1 1 2 0 1 9 u u

Journal

Journal of Sustainable Real EstateTaylor & Francis

Published: Jan 1, 2019

Keywords: golf courses; golf course communities; value diminution; residential amenities

There are no references for this article.