Green Issues 1 page assignment
fidelcastro97
greensflhansz_1.pdf
J R E R � V o l . 3 4 � N o . 1 – 2 0 1 2
T h e V a l u e o f ‘ ‘ G r e e n : ’ ’ E v i d e n c e f r o m t h e F i r s t M a n d a t o r y R e s i d e n t i a l G r e e n B u i l d i n g P r o g r a m
A u t h o r s R a m y a R . A r o u l a n d J . A n d r ew H a n s z
A b s t r a c t There has been recent interest in green building and development practices and research. Resulting from growing environmental awareness and concerns, mandatory residential green building programs have been implemented nationally at the municipal level and Texas has passed legislation to create a statewide program. However, the impact of greenness on residential property values has not been rigorously evaluated. This study examines residential transaction prices in two cities and finds a statistically significant premium associated with ‘‘green’’ properties. Additionally, there is evidence of a larger premium associated with green properties located in Frisco, Texas, which has the nation’s first mandatory residential green building program.
Interest in green building is growing. Indeed, the intensity of interest expressed by many sectors of the real estate and the land development industry is a sign of acceptance of green investment and acknowledgement of the need for green building and development practices. There is also concern regarding quality improvement of new home construction from both environmental and sustainability perspectives. With allocation of $5 billion towards weatherizing homes by the American Recovery and Reinvestment Act (ARRA), significant resources will be channeled into green development projects (Committee on Appropriations, 2009). This investment combined with green building programs will further accelerate the American green revolution.
Since zoning, building codes, and most development regulations are enacted and enforced through municipal police powers, local governments are at the forefront in developing and updating green ordinances and policies. Nationwide, a few communities have taken leadership roles and have implemented formal mandatory green building best management practices programs.1 See Exhibit 1 for an overview of selected mandatory green building programs.
Even though smaller municipalities developed the earliest green building policies, the increasing acceptance of green building practices is reflected in the adoption
2 8 � A r o u l a n d H a n s z
E x h i b i t 1 � A Survey of Mandatory Green Building Programs
Jurisdiction State Program Year Enacted / Revised
Frisco TX Residential 2001 / 2007
Boulder CO Residential 2001 / 2008
Marin County CA Residential 2001 / 2008
Austin TX Residential and Commercial 2003
Aspen / Pitkin County CO Residential and Commercial 2003 / 2008
Pleasanton CA Residential and Commercial 2003 / 2006
Arlington County VA Residential and Commercial 2004
Pasadena CA Residential and Commercial 2006 / 2008
Long Beach CA Residential and Commercial 2006
Santa Cruz CA Residential and Commercial 2006
Montgomery County MD Residential and Commercial 2006
Boston MA Residential and Commercial 2007
of policies by cities, counties, and states. For example, the State of Texas has recently adopted legislation to adopt a statewide mandatory green building program to begin January 2012 for single-family residential dwellings.2
A mandatory green building program is a coordinated, systems approach to green development rather than a component-by-component approach to building (Bynum and Rubino, 1998). Since all builders are mandated to follow green building practices in a mandatory program, the community will experience dramatic shifts in sustainable construction practices and an accelerated implementation of green development.
Existing green research has focused on commercial buildings and the impact of voluntary third-party green rating systems. As discussed in the literature review, there has been some evidence that quantifies the economic benefits of green commercial buildings; however, limited research exists on the real estate market’s response to residential green development and building practices. In addition, the market may recognize price premiums for the coordinated systems required of mandatory green building policies. Prior research on green building codes addressed the impact of mandatory household energy and water consumption regulations and did not address green valuation influences or green building programs.
This leads to two research questions. First, do single-family residential markets reflect and capitalize green investment practices in transaction prices? And second, does the market recognize the coordinated and mandatory aspects of a municipal
T h e V a l u e o f ‘ ‘ G r e e n ’ ’ � 2 9
J R E R � V o l . 3 4 � N o . 1 – 2 0 1 2
green development program? These questions are of interest to homeowners, developers, investors, brokers, appraisers, policy makers, and researchers.
This study investigates the price effects of green on residential transaction prices in two similar Texas cities: Frisco and McKinney. Frisco has implemented the country’s first mandatory residential green building program and McKinney has no green mandates, although properties with green features have been voluntarily developed.
The findings of this study reveal a positive and statistically significant effect on the transaction prices of green residential properties. Furthermore, there appears to be a modestly stronger price premium associated with green transaction prices of residential properties associated with the mandatory green building program.
The next section provides background on the study setting, including an overview of Frisco’s mandatory green building program. This section is followed by a review of the extant literature leading to research hypotheses. Next, sample data are described, including the selection criteria used to identify green transactions, and the empirical models are developed as well. Finally, the statistical results are presented, followed by a discussion of the findings and conclusion.
� S e t t i n g For comparison, the cities of Frisco and McKinney, Texas were selected because these adjacent cities are similar in terms of demographics, employment, and housing market activity. The Frisco-McKinney area is one of the fastest developing regions of the Dallas-Fort Worth Metroplex. Land that once was dedicated to the production of wheat and livestock feed is now demanded for new homes, schools, offices, and parks. The populations of both cities have been growing at steady rates for the past two decades, creating stable and steady housing demand (Exhibit 2).
To manage housing growth and promote sustainability, Frisco implemented a residential green building program in May 2001. A primary difference between these two cities is Frisco’s mandatory green building program, as McKinney does not have any green municipal initiative. Therefore, the Frisco-McKinney pairing provides a natural experimental setting for the study of residential transaction prices, green features, and the nation’s first mandatory green building program.
The green development process is a paradigm shift from traditional building practices and developers and builders can be apprehensive, at first, towards mandatory programs. Mandatory programs require developers and builders to deviate from customary construction practices to test new methods and materials that could reduce development feasibility. In fact, this was the initial concern from the local building industry when Frisco’s proposed a mandatory green building program. In response to industry concern, Frisco implemented a performance- based mandatory residential green building program. A performance-based
3 0 � A r o u l a n d H a n s z
E x h i b i t 2 � Demographic, Employment, and Housing Statistics: Frisco and McKinney
Demographic Frisco McKinney
Population 33,714 54,369 Median Age (years) 30.9 30.6 Female Persons 50.50% 49.40% Persons per Household 2.78 2.89
Population in Labor Force Civilian labor force 78.90% 70.10% Employed 77.10% 67.60% Unemployed 1.80% 2.40% % of civilian labor force 2.30% 3.50% Armed forces 0.10% 0.00% Not in labor force 21.10% 29.80%
Mean Travel Time to Work 27.3 mins. 25.4 mins. College / Graduate School 16.40% 16.80% Land Area (square miles) 70 58
Housing Units 13,683 19,462 1-unit, detached 75.80% 73.20% 1-unit, attached 2.40% 1.40% 2 units 1.50% 2.00% 3 or 4 units 1.30% 3.30% 5 to 9 units 3.20% 4.20% 10 to 19 units 8.50% 5.10% 20 or more units 5.30% 8.50%
Note: The source is QuickFacts from the U.S. Census Bureau 2000 (www.infoplease.com / us / census / data / texas).
program sets minimum sustainability standards, but gives developers and builders flexibility in how they meet green development requirements (Witt, Cantrell, Purvis, and Johnston, 2005).
With this initiative, Frisco, Texas became the first city in the United States to adopt a mandatory residential green building program. According to an ordinance passed on May 2, 2001, all residential plats accepted after May 23, 2001 were required to meet or exceed the mandatory green building program standards. After five years, the city restructured and enhanced the residential green building program. The revised program was put into effect for all homes receiving building permits on, or after, July 1, 2007. Frisco’s Residential Green Building Program mandates minimum energy efficiency, water conservation, indoor air quality, and waste recycling standards for all residential construction. Frisco’s assessment of ‘‘green’’ is consistent with the explanation given by Heekin and Meyers (2001).3
Additional details of Frisco’s mandatory green building program are provided in Appendix A.
T h e V a l u e o f ‘ ‘ G r e e n ’ ’ � 3 1
J R E R � V o l . 3 4 � N o . 1 – 2 0 1 2
� L i t e r a t u r e R e v i e w a n d R e s e a r c h Q u e s t i o n s The hedonic pricing method is based on the premise that the value of a good or service can be decomposed into specific benefits / features. If the market recognizes these individual characteristics, then the contributor value of these benefits / features can be quantified (Rosen, 1974). In the real estate literature, hedonic methods have been used to study a wide range of attributes including school districts (Walden, 1990), conservation districts (Diaz, Hansz, Cypher, and Hayunga, 2008), and age-restricted and gated communities (Allen, 1997). Despite a relatively developed body of hedonic real estate literature, there appears to be little published research examining the valuation impact of green, an attribute of residential development that is gaining in importance due to social, environmental, and ethical concerns.
Previous research addressing environmental attributes has primarily focused on the issue of contamination and property values. These studies addressed how proximity to environmental contamination, such as superfund sites, leaking underground storage tanks (LUSTs), landfills, air and water pollution, and others, influence residential property values (Simons and Saginor, 2006). In another meta- analysis, Boyle and Kiel (2001) evaluated 30 hedonic price studies organized into air pollution, water quality, undesirable land uses, and multiple pollution sources. They found that these environmental variables were important influences on residential property values.
Other prior hedonic studies have estimated transaction price discounts that consumers’ demand for properties located proximate to negative environmental factors such as poor air quality, poor water quality, and undesirable land use (Reichert, Small, and Mohanty, 1992; Smolen, Moore, and Conway, 1992; Flower and Ragas, 1994; Benjamin and Sirmans, 1996). This present study differs from these reviewed papers by examining consumer responses to residential green building, a property feature that minimizes the negative impact of human created improvements on the environment resulting in potential operating cost savings to the owner and society.4 In other words, this study evaluates ‘‘green’’ as a building characteristic rather than an environmental externality. The majority of the available published research with this perspective is on commercial buildings and takes a normative, rather than a positive perspective (Ellison, Sayce, and Smith, 2007).
Empirical research of green commercial properties primarily sample the CoStar�
database of commercial and industrial property listings and transactions. These studies evaluated the effects of voluntary green certifications on transaction prices or rental rates (Miller, Spivey, and Florance, 2008; Fuerst and McAllister, 2009, 2011; Eichholtz, Kok, and Quigley, 2010; Wiley, Benefield, and Johnson, 2010). Miller, Spivey, and Florance (2008) is a pioneering study of green features in commercial buildings and investigated the benefits of investing in energy savings and environmental certifications. This study used the U.S.-based ENERGY STAR
3 2 � A r o u l a n d H a n s z
rating system and the LEED-certified office building certifications as ‘‘green’’ metrics and found evidence of sale price premiums of approximately 6% and 11% for ENERGY STAR and LEED certified office building transactions, respectively. A similar effort by Fuerst and McAllister (2011), using the same data, reported rental premiums of approximately 5% for LEED certification and 4% for ENERGY STAR. For sales prices, they found price premiums of 25% for LEED- certified and 26% for ENERGY STAR office buildings.
Wiley, Benefield, and Johnson (2010), also using a hedonic pricing approach, found rental premiums ranging from 7% to 9% for ENERGY STAR qualified buildings and 15% to 18% for LEED certified commercial buildings. Eichholtz, Kok, and Quigley (2010) studied the financial performance of ENERGY STAR labeled and LEED certified commercial buildings and a control group, consisting of all commercial properties located within 1,300 feet of a green property. They find evidence of economic benefits associated with green certified commercial buildings with green rated buildings achieving rental rates that are 3% higher per square foot than comparable but non-green buildings. These studies provide initial evidence of rental and sale price premiums associated with green commercial buildings, defined as building with the ENERGY STAR designation and / or LEED certification, at all levels.
Despite this commercial property research, there is no similar published research on the price impact of green features and development processes on residential transaction prices. This study addresses this literature void by examining and quantifying the market’s response to residential green development. However, the research expectations are rooted in the theory and findings of the green commercial literature.
Fuerst and McAllister (2011) provide a theoretical framework for the market’s expected response to green features, which is displayed in Exhibit 3. The demand for green and non-green properties is given by DG and DNG and the supply for green and non-green properties is given by SG and SNG. Assuming that green properties and non-green properties are almost perfect substitutes, an increase in the demand of green properties will result in a decrease in demand for non-green properties. The demand increase for green properties will result in a price increase from PNG to PG and the short run inelasticity in supply results in a premium (PG � PNG) for green properties. This leads to the first research expectation that green features will result in residential transaction price premiums.
A combination of mandatory government regulations and voluntary industry standards has evolved to promote and regulate green development. As a consequence, general building standards are becoming more rigorous and ubiquitous (Fuerst and McAllister, 2011) and municipalities are now using mandatory building codes to set minimum energy and water consumption requirements for new construction. As a result, a research interest in the relationship between household electricity consumption and building codes has recently developed (Aroonruengsawat, Auffhammer, and Sanstad, 2009; Costa and Kahn, 2009; Jacobsen and Kotchen, 2009).
T h e V a l u e o f ‘ ‘ G r e e n ’ ’ � 3 3
J R E R � V o l . 3 4 � N o . 1 – 2 0 1 2
E x h i b i t 3 � Model of Price Premiums for Green and Mandatory Green
P = Price Q = Quantity G = Green M = Mandatory program NG = Non-green M’ = Mandatory program: long run
QG
PG
PNG
PM
QM QNG QM
Shortage
Price
Quantity
DNG
SG
DG
SM´
DM
SNG
SM
´
Notes: The figure is adapted from Fuerst and McAllister (2011). DG and DNG denote the demand for green and non-green properties. SG and SNG denote the supply for green and non-green properties. An increase in demand for green properties will cause price to increase from PNG to PG, resulting in a premium (PG � PNG) for green properties. A mandatory green building program shifts demand for green properties from DG to DM. In the short run, the supply of green properties (SG) shifts to the right to SM resulting in an increased price PM and will experience a price premium of (PM � PG) above the green properties and a price premium of (PM � PNG) above non-green properties. In the long run, the supply of the green properties will increase to SM’ to match the demand, reducing the price to the level of PG.
Aroonruengsawat, Auffhammer, and Sanstad (2009) examined mandatory state level building codes and identified the impact of these codes on residential electricity consumption. Jacobsen and Kotchen (2009) used residential billing data on electricity and natural-gas consumption and concluded that a mandatory change in energy-code requirements, implemented in Florida in 2002, was associated with a 4% decrease in electricity consumption and a 6% decrease in natural-gas consumption.
Costa and Kahn (2009) concluded that building codes increased the electricity efficiency of dwellings. In this study of building energy efficiency, they
3 4 � A r o u l a n d H a n s z
investigated homeowners’ asset returns to mandatory energy efficiency investment. Using residential transaction prices from Sacramento, California from 2003 to 2009, they found that energy-efficient homes sell at a premium.
A mandatory green building program can dramatically change market standards and accelerate the demand for the green properties. As a result, demand for green properties shifts further to the right in Exhibit 3, from DG to DM and the existing supply of the green properties SG is not sufficient to meet this extra demand. Despite an increase in green real estate development, the short-term supply of new green properties shown as SM lags the change in demand for multiple reasons— from the natural real estate development gestation period to builder hesitation or learning curve associated with the adoption of the green building process. As a result, the supply of green properties (SG) shifts to the right but not to the magnitude as the shift in demand (DG), resulting in an upward shift in prices to PM.
Therefore, in the short run, the price of green properties in a mandatory green building program will experience a price premium of (PM � PG) above the green properties and a price premium of (PM � PG) above non-green properties. The impact of green on residential transaction prices will be enhanced by the presence of a residential mandatory green building program, as has been found in the commercial property literature for the voluntary green building designations and certifications.
In the long run, the supply of the green properties will increase to SM, to match the demand, reducing the price of green properties to the level of PG. Since it becomes mandatory that the new buildings be constructed green, a new market standard is set and the existing supply of non-green buildings is relatively inferior. Essentially, non-green becomes a form of functional obsolescence in a market that recognizes green development as the new market standard. Therefore, there should be a relatively stronger price premium for green properties located within a mandatory green residential program as the mandatory nature of these programs accelerates both the development of green properties within the community and the change to a new, greener market standard.
� M e t h o d Hedonic pricing studies have evaluated the relationships between housing prices, structural features, and situs characteristics [see Sirmans, MacDonald, Macpherson, and Zietz (2006) for an overview of this literature]. The current study uses a hedonic pricing method to estimate the marginal transaction price contribution of green features and development in residential property values. This section provides an overview of the sample, including a description of the sample data and the development of the variable of interest: residential ‘‘greenness.’’
T h e V a l u e o f ‘ ‘ G r e e n ’ ’ � 3 5
J R E R � V o l . 3 4 � N o . 1 – 2 0 1 2
S a m p l e
Residential transaction price data were obtained from the North Texas Real Estate Information System (NTREIS) for the period from January 1, 2002 through July 1, 2009. This sample includes detached single-family residential transactions from the cities of Frisco and McKinney. As previously stated, Frisco enacted the mandatory residential green building program in May 2001 where all residential plats accepted after May 23, 2001 were required to meet or exceed the mandatory green building program standards. Therefore, the first sample (referred to as ‘sample A’ hereafter) consists of newer buildings constructed after the green building program was first enacted and includes 14,055 transaction records. Frisco revised their mandatory green building program in 2007 with the mandate that all homes receiving building permits on, or after, July 1, 2007 follow minimum green building program standards. Therefore the second sample (referred to as ‘sample B’ hereafter) consists of transactions of residential improvements constructed after the green building program was revised. Sample B consists of 867 residential transactions.
D e f i n i n g G r e e n T r a n s a c t i o n s : M a n d a t o r y
All Frisco residential plats accepted after May 23, 2001 are required to build according to the mandatory green building program standards. Frisco maintains a public record of green subdivisions platted after 2001. A matching procedure and an Internet-based GIS identifying all subdivisions were used to identify residential transaction in the 183 identified green Frisco subdivisions. These mandatory green property transactions were identified as both mandatory and green. Because Frisco required that all buildings constructed after July 2007 meet the mandatory green development requirements, all Frisco transactions with improvements constructed after July 2007 were also coded mandatory and green. These coding rules identified 5,991 and 332 mandatory green transactions based on Frisco’s green building program policies in samples A and B, respectively.
D e f i n i n g G r e e n T r a n s a c t i o n s : C h o i c e
Frisco’s 2001 requirement was applicable to subdivisions being platted; however, subdivisions exist that were platted before 2001 but developed after 2001. Therefore, some construction during this time period was not mandated to be green, but could have been constructed green by choice of the developer or owner. For this reason, the Frisco sample was evaluated for transactions of properties that were built between 2001 and 2007 according to green principles due to private initiatives (choice). Also, the McKinney data also contained green construction by developer or owner choice. Therefore, additional database searches were made to identify green by ‘‘choice’’ property transactions. First, the transaction records that had green features in the fields ‘‘green certification’’ and ‘‘green features’’ of
3 6 � A r o u l a n d H a n s z
the MLS database were coded as green by choice transactions. Next, keyword searches were made to identify green construction features. Please refer to Appendix B for the complete list of green key word terms. For sample A, there were 485 and 305 choice green transactions for Frisco and McKinney, respectively. For sample B, there were 57 choice green transactions for McKinney, and no green by choice for Frisco (as all Frisco new construction was mandatory green in this sample).
In conclusion, there were 6,476 residential green property transactions (5,991 mandatory and 485 by choice) in Frisco and 305 green property transactions (all by choice) in McKinney for sample A. For sample B, there were 389 green property transactions, 332 (all mandatory) in Frisco and 57 (all by choice) in McKinney.
C o n t r o l V a r i a b l e s
Model independent variables control for structural, site, quality, and situs features. Appendix C defines the variables and Exhibits 4a and 4b provides descriptive statistics for samples A and B.
Several of the independent variables warrant further discussion. To control for location differences, indicator variables for the county, either Collin or Denton County, control for significant differences in county demographics and other factors affecting residential transaction prices. An indicator variable for city (Frisco or McKinney) is also included but as previously stated, the demographic similarities between these two neighboring cities were an important motivation in selecting this setting and the expectation is that this control variable will not be significant. The model also includes quality variables to identify foreclosure sales, homeowners association (HOA), age, and age2. Note that age and age2 are not anticipated to be significant in sample B due to the limited variability in property ages. Because seasonality effects have been found in residential transaction price data (Goodman, 1992), market trend and seasonality are controlled using annual and quarterly indicator control variables.
� M o d e l s a n d R e s u l t s The following models are used to estimate the effect of green on residential prices:
ln(salesprice) � � � �X � � � �, (1)�0 Green ln(salesprice) � � � �X � � � � � �, (2)�0 Mandatory Choice
where Xi is a (n � k) matrix of traditional structural, site, quality, and situs variables. The statistical models are estimated using ordinary least squares (OLS).
T h e V a l u e o f ‘ ‘ G r e e n ’ ’ � 3 7
J R E R � V o l . 3 4 � N o . 1 – 2 0 1 2
E x h i b i t 4a � Descriptive Statistics: Improvements Constructed after Green Building Program Enactment
(Sample A)
Characteristic
Full Sample (n � 14,055)
Mean Max. Min.
Frisco (n � 6,555)
Mean Max. Min.
McKinney (n � 7,500)
Mean Max. Min.
ln(salesprice) 12.15 12.61 11.52 12.25 12.61 11.70 12.06 12.61 11.52
Green 0.48 1.00 0.00 0.99 1.00 0.00 0.04 1.00 0.00
Mandatory 0.43 1.00 0.00 0.91 1.00 0.00 0.00 0.00 0.00
Choice 0.06 1.00 0.00 0.07 1.00 0.00 0.04 1.00 0.00
lnSqft 7.91 8.54 6.19 7.94 8.54 6.58 7.88 8.46 6.19
Beds 3.72 8.00 0.00 3.84 8.00 2.00 3.62 6.00 2.00
BedsMore 0.60 1.00 0.00 0.59 1.00 0.00 0.60 1.00 0.00
Fullbath 2.36 5.00 1.00 2.46 5.00 1.00 2.27 5.00 1.00
Halfbath 0.41 5.00 0.00 0.43 5.00 0.00 0.40 3.00 0.00
Pool 0.04 1.00 0.00 0.04 1.00 0.00 0.04 1.00 0.00
Age* 3.27 8.00 �1.00 3.36 8.00 0.00 3.20 8.00 �1.00
Age 2 17.49 64.00 0.00 18.08 64.00 0.00 16.97 64.00 0.00
City 0.47 1.00 0.00 1.00 1.00 1.00 0.00 0.00 0.00
County 0.84 1.00 0.00 0.65 1.00 0.00 1.00 1.00 0.00
School 0.54 1.00 0.00 0.95 1.00 0.00 0.19 1.00 0.00
Foreclosure 0.10 1.00 0.00 0.09 1.00 0.00 0.10 1.00 0.00
HOA 0.89 1.00 0.00 0.89 1.00 0.00 0.89 1.00 0.00
Q1 0.22 1.00 0.00 0.21 1.00 0.00 0.22 1.00 0.00
Q2 0.28 1.00 0.00 0.28 1.00 0.00 0.29 1.00 0.00
Q3 0.26 1.00 0.00 0.26 1.00 0.00 0.26 1.00 0.00
Q4 0.24 1.00 0.00 0.25 1.00 0.00 0.24 1.00 0.00
Year 2002 0.03 1.00 0.00 0.03 1.00 0.00 0.03 1.00 0.00
Year 2003 0.15 1.00 0.00 0.16 1.00 0.00 0.14 1.00 0.00
Year 2004 0.16 1.00 0.00 0.16 1.00 0.00 0.16 1.00 0.00
Year 2005 0.18 1.00 0.00 0.18 1.00 0.00 0.18 1.00 0.00
Year 2006 0.17 1.00 0.00 0.16 1.00 0.00 0.18 1.00 0.00
Year 2007 0.14 1.00 0.00 0.14 1.00 0.00 0.14 1.00 0.00
Year 2008 0.12 1.00 0.00 0.12 1.00 0.00 0.12 1.00 0.00
Year 2009 0.05 1.00 0.00 0.05 1.00 0.00 0.05 1.00 0.00
Note: * One record had an age of �1 because it was placed under contract before construction had finished.
3 8 � A r o u l a n d H a n s z
E x h i b i t 4b � Descriptive Statistics: Improvements Constructed After Green Building Program Revision
(Sample B)
Characteristic
Full Sample (n � 867)
Mean Max. Min.
Frisco (n � 332)
Mean Max. Min.
McKinney (n � 535)
Mean Max. Min.
ln(salesprice) 12.17 12.61 11.52 12.29 12.61 11.75 12.09 12.61 11.52
Green 0.45 1.00 0.00 1.00 1.00 1.00 0.11 1.00 0.00
Mandatory 0.38 1.00 0.00 1.00 1.00 1.00 0.00 0.00 0.00
Choice 0.07 1.00 0.00 0.00 0.00 0.00 0.11 1.00 0.00
LnSqft 7.75 8.54 6.19 7.73 8.54 6.19 7.76 8.53 6.19
Beds 3.76 8.00 2.00 3.94 8.00 2.00 3.64 6.00 2.00
BedsMore 0.58 1.00 0.00 0.57 1.00 0.00 0.59 1.00 0.00
Fullbath 2.41 5.00 1.00 2.57 5.00 1.00 2.31 5.00 0.00
Halfbath 0.44 5.00 0.00 0.49 5.00 0.00 0.42 3.00 0.00
Pool 0.01 1.00 0.00 0.01 1.00 0.00 0.01 1.00 0.00
Age* 1.18 3.00 �1.00 1.22 3.00 0.00 1.16 3.00 �1.00
Age 2 2.68 9.00 0.00 2.77 9.00 0.00 2.62 9.00 0.00
City 0.38 1.00 0.00 1.00 1.00 1.00 0.00 0.00 0.00
County 0.84 1.00 0.00 0.59 1.00 0.00 1.00 1.00 0.00
School 0.53 1.00 0.00 0.94 1.00 0.00 0.28 1.00 0.00
Foreclosure 0.08 1.00 0.00 0.08 1.00 0.00 0.07 1.00 0.00
HOA 0.89 1.00 0.00 0.89 1.00 0.00 0.89 1.00 0.00
Q1 0.22 1.00 0.00 0.21 1.00 0.00 0.22 1.00 0.00
Q2 0.26 1.00 0.00 0.25 1.00 0.00 0.26 1.00 0.00
Q3 0.25 1.00 0.00 0.25 1.00 0.00 0.25 1.00 0.00
Q4 0.27 1.00 0.00 0.28 1.00 0.00 0.26 1.00 0.00
Year 2007 0.50 1.00 0.00 0.52 1.00 0.00 0.49 1.00 0.00
Year 2008 0.40 1.00 0.00 0.40 1.00 0.00 0.40 1.00 0.00
Year 2009 0.10 1.00 0.00 0.09 1.00 0.00 0.11 1.00 0.00
Note: * One record had an age of �1 because it was placed under contract before construction had finished.
The dependent variable is the natural log of sale price. Sirmans, Macpherson, and Zietz (2005) discussed the advantages of using the semi-log specification in hedonic modeling. This specification allows for variation in the dollar value of each characteristic and coefficients are interpreted as the percentage change in the
T h e V a l u e o f ‘ ‘ G r e e n ’ ’ � 3 9
J R E R � V o l . 3 4 � N o . 1 – 2 0 1 2
price per unit change associated with each characteristic. This specification also helps to minimize the problem of heteroscedasticity, although heteroscedasticity was found in the sample data, despite the semi-log specification. White’s (1980) heteroscedasticity-consistent standard errors and covariance procedure are used to control for heteroscedasticity.
M o d e l 1
Model 1 uses equation (1). The variable of interest is Green, an indicator variable identifying all green property transactions. Exhibit 5 presents the direct effects model for samples A and B.
With an adjusted R2 of 78% for sample A and an adjusted R2 of 75% for sample B, the explanatory power of both models was acceptable and consistent with published hedonic research of residential transaction prices. Coefficient estimates were of the expected signs. Note that the City independent variables were not statistically significant in both models. Also, the Age and Age2 variables were not significant in Sample B, due to the limited age variability in this sample.
These results demonstrate a price premium associated with green properties. After controlling for price differences associated with structure size, property qualities, and situs, the green variables exhibit coefficients of 2.07% and 2.43% for samples A and B respectively, which were statistically significant at the 1% level.
M o d e l 2
Model 2 uses equation (2). The variables of interest in this model are variables identifying both mandatory and non-mandatory (referred to as ‘choice’) green transactions for samples A and B. Exhibit 6 presents model 2 results.
With adjusted R2s of 78% and 75% for samples A and B, respectively, the explanatory power of both models was acceptable. Coefficient estimates were of the anticipated signs. Again, the City independent variables were not statistically significant in any model and the Age and Age2 variables were not significant in Sample B, due to the limited age variability in this sample.
As compared to model 1, model 2 had similar explanatory power but revealed greater price premiums for green development located in a mandatory green building program and relatively lower price premium for green properties developed by choice. The mandatory green variable coefficients are 4.69% for sample A and 3.03% for sample B. The choice green variable coefficients are 1.05% for sample A and 0.17% for sample B. Since the mandatory variable serves as a proxy for residential green development in Frisco’s mandatory green building program, this does provide initial evidence that the mandatory program may influence residential transaction prices.
4 0 � A r o u l a n d H a n s z
E x h i b i t 5 � Model 1: Hedonic Model with Variable of Interest: ‘‘Green’’
Sample A Sample B
Program Enactment Program Revision
Characteristic � P-value � P-value
Green 0.02 0.000 0.02 0.000
LnSqft 0.89 0.000 0.65 0.000
Beds 0.12 0.000 0.12 0.000
BedsMore �0.04 0.000 �0.05 0.000
Fullbath 0.09 0.000 0.07 0.000
Halfbath 0.01 0.000 0.00 0.000
Pool 0.13 0.000 0.14 0.000
Age 0.00 0.000 �0.01 0.110
Age 2 0.00 0.000 0.00 0.850
City �0.01 0.450 �0.01 0.290
County 0.05 0.000 0.05 0.000
School 0.05 0.000 0.05 0.000
Foreclosure �0.12 0.000 �0.11 0.000
HOA 0.03 0.000 0.05 0.000
Q2 �0.01 0.000 0.01 0.000
Q3 0.01 0.000 0.02 0.000
Q4 �0.01 0.010 �0.01 0.000
Year 2002 �0.04 0.000 — —
Year 2003 �0.02 0.000 — —
Year 2004 �0.04 0.000 — —
Year 2005 �0.01 0.040 — —
Year 2006 0.01 0.050 — —
Year 2007 0.01 0.130 0.01 0.330
Year 2008 0.01 0.130 0.00 0.680
Intercept 5.61 0.000 6.53 0.000
Notes: The dependent variable was the natural logarithm of sale price. Samples A and B consist of 14,055 and 867 residential transactions, respectively. Sample A corresponds to the transactions after the mandatory program enactment (2001) and consists of 6,781 green transactions (6,476 in Frisco and 305 in McKinney) and Sample B corresponds to the transactions after the mandatory program revision (2007) and consists of 389 green transactions (332 in Frisco and 57 in McKinney). For Sample A, R2 � 0.79, Adj. R2 � 0.78, F-stat. � 1,234.33, and P-value � 0.000. For Sample B, R2 � 0.75, Adj. R2 � 0.75, F-stat. � 663.81, and P-value � 0.000.
T h e V a l u e o f ‘ ‘ G r e e n ’ ’ � 4 1
J R E R � V o l . 3 4 � N o . 1 – 2 0 1 2
E x h i b i t 6 � Model 2: Hedonic Model with Variables of Interest: ‘‘Mandatory’’ and ‘‘Choice’’
Sample A Sample B
Program Enactment Program Revision
Characteristic � P-value � P-value
Mandatory 0.05 0.000 0.03 0.000
Choice 0.01 0.000 0.00 0.030
LnSqft 0.89 0.000 0.65 0.000
Beds 0.11 0.000 0.12 0.000
BedsMore �0.07 0.000 �0.05 0.000
Fullbath 0.09 0.000 0.07 0.000
Halfbath 0.02 0.010 0.00 0.000
Pool 0.14 0.000 0.14 0.000
Age �0.06 0.000 �0.01 0.160
Age 2 0.00 0.320 0.00 0.770
City �0.01 0.450 �0.01 0.240
County 0.03 0.000 0.05 0.000
School 0.04 0.000 0.05 0.000
Foreclosure �0.12 0.000 �0.11 0.000
HOA 0.02 0.01 0.05 0.000
Q2 �0.01 0.000 0.01 0.000
Q3 0.01 0.000 0.02 0.000
Q4 �0.01 0.000 �0.01 0.000
Year 2002 �0.03 0.01 — —
Year 2003 �0.01 0.000 — —
Year 2004 �0.01 0.000 — —
Year 2005 0.05 0.000 — —
Year 2006 0.01 0.000 — —
Year 2007 0.00 0.050 0.00 0.610
Year 2008 0.01 0.350 �0.01 0.470
Intercept 5.45 0.000 6.44 0.000
Notes: The dependent variable was the natural logarithm of sale price. Samples A and B consist of 14,055 and 867 residential transactions, respectively. Sample A corresponds to the transactions after the mandatory program enactment (2001) and consists of 6,781 green transactions (6,476 in Frisco and 305 in McKinney) and Sample B corresponds to the transactions after the mandatory program revision (2007) and consists of 389 green transactions (332 in Frisco and 57 in McKinney). For Sample A, R2 � 0.79, Adj. R2 � 0.78, F-stat. � 1,188.78, and P-value � 0.000. For Sample B, R2 � 0.75, Adj. R2 � 0.75, F-stat. � 664.00, and P-value � 0.000.
4 2 � A r o u l a n d H a n s z
� D i s c u s s i o n As a result of environmental and economic concerns, residential green building programs have been enacted in many cities and counties throughout the country. Texas has passed legislation to establish statewide single-family green building requirements starting in 2012. Green development has stepped from the background to the forefront of public policy initiatives.
Even if green building policy is enacted for environmental and social benefits, green development may certainly influence residential transaction prices and property values. In past decades, expenditures in green features were generally written off as no or very low return investments. These features were useful to the owner but the market would not necessarily recognize this investment. However, with increasing energy prices and subsequent changes in attitudes, consumers now consider green in their home-buying process and demand green features and development. As a result, academics, industry professionals, consumers, and policy makers are interested in identifying and measuring any value impacts from green residential development and a mandatory green building program.
However, green research has generally focused on valuation effects on commercial properties and voluntary green initiatives. In general, hedonic studies concerned with commercial green properties have found transaction price and rental premiums for green investment associated with voluntary green designations and certification programs (Miller, Spivey, and Florance, 2008; Eichholtz, Kok, and Quigley, 2010; Wiley, Benefield, and Johnson, 2010; Fuerst and McAllister, 2011). From these results, it does appear that commercial real estate markets are generally responding to and recognizing green investment encouraged by programs such as ENERGY STAR and LEED. The results of this present study parallel the commercial property findings and provide initial evidence that residential markets are recognizing green investment.
It is important to note that this study measures the market’s valuation of green investment and does not quantify green investment costs. There is a perception of high costs associated with green building. For example, Kats (2003, 2005) found that there is an increase of about 2% in green commercial building construction costs. If Kats’s findings can be generalized to residential green construction, then the findings in this present study may suggest that the market is approximately recognizing green construction costs.
However, in order to determine whether or not the value of green equals the cost of green construction, as cost rarely equals value in real estate markets, further research into green construction cost is warranted. Furthermore, green building features provide financial benefits in the form of lowering operating costs over the improvement’s economic life. Reduced operating costs may include reduced energy consumption, water conservation, waste reduction, and better indoor air
T h e V a l u e o f ‘ ‘ G r e e n ’ ’ � 4 3
J R E R � V o l . 3 4 � N o . 1 – 2 0 1 2
quality (RICS, 2005; Kohlhepp, 2006). The true value of cost-saving benefits from energy-saving construction also depends on assumed future energy prices at the time the transaction was negotiated. However, these benefits can be difficult to quantify and therefore difficult for the buyers and sellers to estimate and, therefore, the market to recognize.
Although this study finds a modest price premium for residential green features, perhaps the market is acknowledging initial green construction costs. Going forward, markets may recognize and capitalize the long-term benefits, both financial and non-financial, of green features and building practices. This is a fruitful area for future research.
� C o n c l u s i o n This study examines the value of green in residential properties and the market response to a mandatory green building program in Texas. The results indicate a positive and statistically significant premium for transactions with green residential building features. Additionally, there is evidence of a more robust price premium for green properties located within a mandatory green building program. Although the findings parallel those found in the commercial property literature, this is the first known study to document the valuation impact of green features and development on residential transaction prices.
Specifically, there are three main contributions of the study. First, it introduces ‘‘green’’ into the residential hedonic pricing literature. It is anticipated that greenness will be a common consideration in the hedonic research paradigm. Second, this study is an initial effort to encourage additional research on the effects of green development on residential transaction prices. Wide-scale green development is a relatively recent innovation and there has not been an opportunity to study the impact of green features on property values over an extended time period. The findings show a modest but significant impact on residential transaction prices; however, going forward, green price premiums could increase as markets recognize and capitalize the long-term benefits, both financial and non- financial, of green building ownership. Future research should also consider the costs, as well as the value-creating benefits of green development. This study measured exclusively the value or the market’s reaction to green features and development and not costs. Third, the study examined the impact of a mandatory green development program, as the two Texas cities used in this study provide a unique opportunity to examine the impact of a mandatory residential green building program.
Further research is encouraged to better understand the relationship between residential property values, green development, and green building programs, both mandatory and voluntary. Research on green real estate is in an early stage of a research program with many important questions to consider.
4 4 � A r o u l a n d H a n s z
� A p p e n d i x A �� R e s i d e n t i a l G r e e n B u i l d i n g P r o g r a m G u i d e l i n e s : C i t y o f � F r i s c o
The City of Frisco’s Residential Green Building Program mandates minimum energy efficiency, water conservation, indoor air quality and waste recycling standards for all residential buildings. Energy efficiency, water conservation, indoor air quality, and water recycling are the four categories for which the City of Frisco’s Residential Green Building Program sets forth minimum standards. The minimum requirements in these four categories are provided below:
1. Energy Efficiency In addition to the city’s existing building codes and other relevant regulations, the following must be accomplished: a. The single-family residences should have the Environmental Protection
Agency’s ENERGY STAR designation or a score of 83 or less on the Home Energy Rating Systems (HERS) index.
b. Every home must be tested by a Residential Energy Services Network (RESNET) approved HERS rater annually.
c. Every story in the house should have at least one programmable thermostat.
d. Any room that will be closed with a door should have a return air path. The pressure differential in that door with the door closed and the air handler working should not be more than plus or minus 3 Pascal.
e. All the joints in the air distribution system should be sealed with duct mastic.
2. Indoor Air Quality In addition to the city’s existing building codes and other relevant regulations, the following must be accomplished:
a. Every single-family residence should have a minimum standard of American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) Standard 62.2 or its amendment.
b. Every HVAC unit that supplies air to a bedroom must get all or a portion of outdoor air distribution. Total ventilation rate must be split between each HVAC unit supplying air to a bedroom. Ducts carrying outdoor air must have manual dampers and a filter inside the return.
c. Outdoor air intakes must be located at a minimum of at least 60 inches from all the roofing materials.
d. HVAC plenums on the supply and return side must be constructed using sheet metal or equivalent that is approved by the building officials.
T h e V a l u e o f ‘ ‘ G r e e n ’ ’ � 4 5
J R E R � V o l . 3 4 � N o . 1 – 2 0 1 2
e. Heating and cooling equipment can be used during construction only after manufacturer specified filter is installed.
f. Central vacuums must be expelled outdoors. g. Carpets, cushions, and carpet adhesives must have the carry the Carpet
and Rug Institute (CRI) Green Label. h. Vinyl wallpaper is not permitted to be used on the inside of the exterior
walls and on wet walls. i. Metal drip edge is to be provided at all exposed roof decking.
3. Water Conservation In addition to the city’s landscape ordinance and other relevant regulations, the following must be accomplished. a. Bedding must be mulched to a depth of 2 to 4 inches using recycled
materials that includes wood construction waste. b. Each installed tree must be provided with a portable drip irrigation bag
or zoned bubbler system. 4. Water Recycling
a. In addition to the city’s waste reduction and recycling regulations, brick and wood construction waste from the building site must be taken to a facility legally empowered to accept it for recycling as approved by the county and state in which the facility is located.
The City of Frisco’s Residential Green Building Program content was edited and summarized by the authors.
� A p p e n d i x B �� L i s t o f G r e e n I d e n t i f i e r s
S No. Green Features List
1 ENERGY STAR
2 Greenbuilt
3 HERS
4 LEED
5 NGBP
6 Drought tolerant
7 Energy recovery
8 Air filtration
9 Irrigation control
10 Geo Thermal HVAC
4 6 � A r o u l a n d H a n s z
� A p p e n d i x B ( c o n t i n u e d ) �� L i s t o f G r e e n I d e n t i f i e r s
S No. Green Features List
11 Low flow commode
12 Low flow fixtures
13 Mechanical fresh air
14 Rain freeze sensors
15 Rain water catchment
16 Recirculating hot water
17 Solar
18 Wind power
� A p p e n d i x C �� V a r i a b l e D e f i n i t i o n s
Characteristics Description
ln(salesprice) Natural log of sales price.
Green Dummy variable value equals one if property is green.
Mandatory Dummy variable value equals one if property is in mandatory green program.
Choice Dummy variable value equals one if property is green by choice.
lnSqft Natural log of square foot.
Beds Number of beds.
BedsMore Dummy variable value equals one if number of beds more than four.
Fullbath Number of full baths.
Halfbath Number of half baths.
Pool Dummy variable value equals one if pool is present.
Age Age.
Age 2 Square of Age.
City Dummy variable value equals one if property is in Frisco.
County Dummy variable value equals one if property is in Collin County.
School Dummy variable value equals one if property is in Frisco ISD.
Foreclosure Dummy variable value equals one if property is foreclosed.
HOA Dummy variable value equals one if home owner’s association.
T h e V a l u e o f ‘ ‘ G r e e n ’ ’ � 4 7
J R E R � V o l . 3 4 � N o . 1 – 2 0 1 2
� A p p e n d i x C ( c o n t i n u e d ) �� V a r i a b l e D e f i n i t i o n s
Characteristics Description
Q1 Dummy variable value equals one if sale is in first quarter.
Q2 Dummy variable value equals one if sale is in second quarter.
Q3 Dummy variable value equals one if sale is in third quarter.
Q4 Dummy variable value equals one if sale is in fourth quarter.
Year 2002 Dummy variable value equals one if sale is in 2002
Year 2003 Dummy variable value equals one if sale is in 2003.
Year 2004 Dummy variable value equals one if sale is in 2004.
Year 2005 Dummy variable value equals one if sale is in 2005.
Year 2006 Dummy variable value equals one if sale is in 2006.
Year 2007 Dummy variable value equals one if sale is in 2007.
Year 2008 Dummy variable value equals one if sale is in 2008.
Year 2009 Dummy variable value equals one if sale is in 2009.
� E n d n o t e s 1 Green building programs could be either mandatory or voluntary. This study focuses on
mandatory green building programs. 2 For more information on this legislation please, see the Texas Register, Adopted Rules,
June 4, 2010, 35, 4727–4729. http: / / www.sos.state.tx.us / texreg / pdf / backview / 0604 / 0604adop.pdf.
3 Green building can be identified as development and construction that endorses less energy consumption, promotes water conservation, propagates the best use of building materials, uses renewable resources, encourages competent waste management techniques, preserves the natural environment, and incorporates health and environmental standards (Heekin and Meyers, 2001).
4 The founder of Leadership in Energy and Environmental Design (LEED), the United States Green Building Council, describes green building as the ‘‘design and construction practices that significantly reduce or eliminate the negative impact of buildings on the environment and occupants in five broad areas such as sustainable site planning, safeguarding water and water efficiency, energy efficiency and renewable energy, conservation of materials and resources, and indoor environmental quality,’’ (USGBC, 2001).
� R e f e r e n c e s Allen, M. Measuring the Effects of ‘Adults Only’ Age Restrictions on Condominium Prices. Journal of Real Estate Research, 1997, 14:3, 339–46.
4 8 � A r o u l a n d H a n s z
Aroonruengsawat, A., M. Auffhammer, and A. Sanstad. The Impact of Buildings Codes on Residential Electricity Consumption. Working paper, University of California–Berkeley, 2009.
Benjamin, J.D. and G.S. Sirmans. Mass Transportation, Apartment Rent and Property Values. Journal of Real Estate Research, 1996, 12:1, 1–8.
Boyle, M. and K. Kiel. A Survey of House Price Hedonic Studies of the Impact of Environmental Externalities. Journal of Real Estate Literature, 2001, 9:2, 116–44.
Bynum, R.T. and D.L. Rubino. Handbook of Alternative Materials in Residential Construction. New York: McGraw-Hill, 1998.
Committee on Appropriations. Summary: American Recovery and Reinvestment. Available at: http: / / appropriations.house.gov / pdf / PressSummary02-13-09.pdf, 2009.
Costa, D. and M.E. Kahn. Towards a Greener California: An Analysis of Household Variation in Residential Electricity Purchases. Working paper, University of California–LA, 2009.
Diaz, J., J. Hansz, M. Cypher, and D. Hayunga. The Effects of Conservation Status on Residential Property Values. Journal of Real Estate Research, 2008, 30:2, 225–47.
Ellison, L., S. Sayce, and J. Smith. Socially Responsible Property Investment: Quantifying the Relationship between Sustainability and Investment Property Worth. Journal of Property Research, 2007, 24:3, 191–219.
Eichholtz, P.M.A., N. Kok, and J.M. Quigley. Doing Well by Doing Good: Green Office Buildings. American Economic Review, 2010, 100:5, 2492–2509.
Flower, P.C. and W.R. Ragas. The Effects of Refineries on Neighborhood Property Values. Journal of Real Estate Research, 1994, 9:3, 319–38.
Fuerst, F. and P. McAllister. An Investigation of the Effect of Eco-Labeling on Office Occupancy Rates. Journal of Sustainable Real Estate, 2009, 1:1, 49–64.
——. Green Noise or Green Value? Measuring the Effects of Environmental Certification on Office Values. Real Estate Economics, 2011, 39:1, 1–25.
Goodman, J.L. A Housing Market Matching Model of the Seasonality in Geographic Mobility. Journal of the Real Estate Research, 1992, 8:1, 117–37.
Heekin, D. and D. Meyers. Green Building: Principles and Practice. Working paper, University of Florida Levin College of Law, 2001.
Jacobsen, G.D. and M.J. Kotchen. Are Building Codes Effective at Saving Energy? Evidence from Residential Billing Data in Florida. Working Paper, Yale University, 2009.
Kats, G. The Costs and Financial Benefits of Green Buildings: A Report to California’s Sustainable Building Task Force. Available at: http: / / www.ciwmb.ca.gov / GreenBuilding / Design / CostBenefit / Report.pdf, 2003.
——. Greening America’s Schools: Costs and Benefits. A Capital E Report, Capital E. Available at: http: / / www.cap-e.com / ewebeditpro / items / O59F12807.pdf, 2006.
Kohlhepp, D. The Greening of One Potomac Yard. Crescent Resources LLC, 2006.
Miller, N., J. Spivey, and A. Florance. Does Green Pay Off? Journal of Real Estate Portfolio Management, 2008, 14:4, 385–99.
Reichert, A.K., M. Small, and S. Mohanty. The Impacts of Landfills on Residential Property Values. Journal of Real Estate Research, 1992, 7:3, 297–314.
RICS. Green Value; Growing Assets. London, UK, 2005.
T h e V a l u e o f ‘ ‘ G r e e n ’ ’ � 4 9
J R E R � V o l . 3 4 � N o . 1 – 2 0 1 2
Rosen, S. Hedonic Prices and Explicit markets: Production Differentiation in Pure Competition. Journal of Political Economy, 1974, 82, 34–55.
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., L. MacDonald, D.A. Macpherson, and E.N. Zietz. The Value of Housing Characteristics: A Meta Analysis. Journal of Real Estate Finance and Economics, 2006, 33, 215–40.
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.
Smolen, G.E., G. Moore, and L.V. Conway. Economic Effects of Hazardous Chemical and Proposed Radioactive Waste Landfills on Surrounding Real Estate Values. Journal of Real Estate Research, 1992, 7:3, 283–95.
U.S. Green Building Council. LEED Reference Guide, Version 2.0, Washington, D.C., 2001.
Walden, M. Magnet Schools and the Differential Impact of School Quality on Residential Property Values. Journal of Real Estate Research, 1990, 5:2, 13–31.
White, H. A Heteroskedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroscedasticity. Econometrica, 1980, 48, 817–38.
Wiley, J., J. Benefield, and K. Johnson. Green Design and the Market for Commercial Office Space. Journal of Real Estate Finance and Economics, 2010, 41:2, 228–43.
Witt, J., E. Vasatka, T. Cantrell, J. Purvis, and D. Johnston. Boulder, Colorado and Frisco, Texas: A Tale of Two Green Cities. Available at http: / / www.bouldercolorado.gov / files / Environmental%20Affairs / Green%20Building / usgbc pittsburg 1.pdf, 2005.
We are grateful to Darren Hayunga for comments on an earlier version of this paper, Aaron Smallwood for helpful econometric suggestions, and an anonymous reviewer. Finally, we thank Peggy Swanson for being a mentor.
Ramya R. Aroul, University of Texas, Arlington, TX, 76019 or [email protected].
J. Andrew Hansz, California State University, Fresno, CA 93740-8001 or hansz@ gazarian.info.
