Content uploaded by V. Kelly Turner
Author content
All content in this area was uploaded by V. Kelly Turner
Content may be subject to copyright.
1167
Urban Geography, 2011, 32, 8, pp. 1167–1188. http://dx.doi.org/10.2747/0272-3638.32.8.1167
Copyright © 2011 by Bellwether Publishing, Ltd. All rights reserved.
THE IMPACT OF HOMEOWNERS ASSOCIATIONS ON RESIDENTIAL
WATER DEMAND MANAGEMENT IN PHOENIX, ARIZONA1
V. Kelly Turner2 and Dorothy C. Ibes
School of Geographical Sciences and Urban Planning
Arizona State University
Abstract: In regulating residential landscaping and maintenance practices, homeowner asso-
ciations (HOAs) are potentially important institutional actors in urban climate adaptation and
water demand management. One view posits that HOAs will use outdoor landscaping and scarce
water resources to maintain the aesthetic appearance and hence property values of homes in their
domain. An alternative view from commons theory suggests that they will behave in a variety of
ways and have diverse environmental resource management outcomes. This research compares
water consumption in single-family residential communities with and without HOAs in Phoenix,
Arizona. Findings reveal there is not a significant correlation between water consumption and
presence or absence of HOAs, after accounting for other relevant variables. HOAs did not co-opt a
disproportionate share of urban water use, lending credence to the claims of commons theory that
similar institutional types may result in diverse resource outcomes, depending upon geographi-
cal and social context. Results suggest the potential to utilize HOAs as an entree-point for water
demand reduction strategies. [Key words: water management, residential landscapes, common
properties, homeowners associations.]
INTRODUCTION
Water resources are the paramount environmental concern in the arid cities of the U.S.
Southwest, where the coupled pressures of population growth (Gammage et al., 2008) and
climate change increasingly stress limited existing supplies (Karl et al., 2009; Balling and
Goodrich, 2010). Recently, there has been heightened awareness that the supply-side strat-
egies currently favored for balancing supply and demand have reached a point of diminish-
ing returns. For example, Colorado River supplies are overallocated, so the construction of
larger dams does not appear to be a viable solution for increasing supply to meet current
demands (Gober et al., 2010; Morehouse, 2000). Instead, new and innovative demand
management strategies will be necessary if Southwest cities are to meet the water needs
required to sustain large metropolitan regions (Gober et al., 2010; Gober and Kirkwood,
2010). Such innovation requires greater recognition of the relationship between water
consumption and urban form, identifying the co-benefits of coordinating urban planning
with water management. For example, implementing the urban growth boundary (UGB)
in Portland, Oregon generates density—at least within the UGB—associated with lower
1This article is based on work supported by the National Science Foundation under Grant No. SES-0345945,
Decision Center for a Desert City (DCDC). Any opinions, findings, and conclusions or recommendation
expressed in this material are those of the author(s) and do not necessarily reflect the views of the National
Science Foundation (NSF).
2Correspondence concerning this article should be addressed to V. Kelly Turner, School of Geographical Sciences
and Urban Planning, Arizona State University, P.O. Box 878209, Tempe, Arizona 85287-8209; telephone: 480-
965-3367; fax: 480-965-8383; email: vkturner@asu.edu
1168 TURNER AND IBES
water consumption (Chang et al., 2010). In the past, demand management strategies in
Phoenix have targeted individual households as well as municipal providers but, in so
doing, have neglected the wide range of institutional actors that make decisions that impact
urban water consumption. One set of neglected actors are the Homeowners Associations
(HOAs) that have come to dominate new residential developments. This research explores
the implications of the presence of HOAs for single-family residential water demand man-
agement in Phoenix, Arizona.
Reducing residential water consumption is a critical component of any demand man-
agement program. In the City of Phoenix, residential uses account for approximately half
of all municipal water demand (Wentz and Gober, 2007), 70% of which can be attributed
to outdoor uses, largely for pools and yard maintenance (City of Phoenix, 2010). In the
past, passive demand management strategies such as educational campaigns and tax incen-
tives (ibid.) have been favored over more aggressive policies such as water use bans and
pricing mechanisms (Gober, 2006; Hirt et al., 2008). Residential water demand manage-
ment policies and programs are predominantly voluntary and target individual households.
In reality, decision-making occurs at multiple scales and among a variety of institutional
actors, including HOAs, which heavily influence decision-making at the neighborhood
scale.
HOAs have emerged as a dominant institution in the residential landscape across
the United States. The presence of HOAs in the U.S. has increased dramatically—from
500 in the 1960s to 305,000 in 2009, a 600-fold increase in five decades (Community
Associations Institute, 2010). The majority of new residential developments in Phoenix
since 1985 have been governed by HOAs, defined as “a common interest organization to
which all the owners of lots in a planned community … must belong” (Arizona, 2010).3
Property owners are dues-paying members of the HOA and are contractually obliged to
follow the rules and regulations specified in a set of governing documents. These docu-
ments include covenants, codes, and restrictions (CCRs) that are legally enforceable, and
the provisions contained in their text run permanently with the land. Modifying a CCR is
difficult because changes usually require a supermajority of all property owners—not just
those that actually vote—which is particularly problematic given investor-owners that do
not actually reside in the community (McKenzie, 1994). In practice, therefore, CCRs are
virtually immutable.
HOAs developed as a mechanism to control individual home aesthetics and preserve
property values; as such, their CCRs may stipulate landscape structure and maintenance
practices. For example, CCRs often contain clauses that require, permit, or prohibit a
specific percentage and type of ground cover, particular plant species, and weed-free or
green lawns (Cook et al., 2011). Such rules have implications for the water required to
maintain particular structures. A non-random survey of HOAs (N = 35) in the Phoenix
metropolitan area found that the number of CCRs containing clauses relating to land-
scape structure and management has increased since the 1960s; of the HOAs sampled,
77% regulated vegetation and pest management, 69% regulated water management, and
38% regulated species composition (Larson et al., 2008). HOAs, therefore, appear to exert
3The process has been spurred in Phoenix by the desire of local developments to maintain a distinctive archi-
tectural appearance, thereby distinguishing them from neighboring areas from which they may be separated by
walls, gates, and fences (e.g., Romig, 2010, p. 1067).
HOMEOWNERS ASSOCIATIONS AND RESIDENTIAL WATER DEMAND 1169
influence on landscape design and resource uses beyond the purview of public institutions
and personal preference.4
Systematic assessments of the impact of HOAs on resource uses and the environment at
large, especially as compared to non-HOA residences, are few. This situation is somewhat
surprising given: (1) the large body of research in the social sciences linking HOAs—
and the types of communities they govern—to the capture of a disproportionate share of
municipal resources (Kennedy, 1995; Barry, 2002; Vesselinov et al., 2007); and (2) the rise
of social-ecological perspectives in the study of urban residential landscapes (Robbins,
2007; Cook et al. 2011; House-Peters and Chang 2011). Each research domain offers a set
of empirical and theoretical insights that implies the effects of HOA rules on environmen-
tal resource consumption. From urban social geography, we may conclude that the HOA is
a monolithic institution created by and for the elite to protect property values through the
control of municipal resources including, by extension, environmental resources. Recent
developments in environmental commons research offer an alternative perspective, in
which resource management reflects the complexity of the coupled systems of which they
are a part (Ostrom, 2007, 2009). The so-called “commons perspective” predicts variability
in environmental management and hence water demand among HOAs. This study offers a
systematic, quantitative comparison of water demand by residences within and outside of
HOAs in Phoenix. It asks: Do HOA households generate larger water demand than non-
HOA households? Furthermore, it attempts to synthesize the advances of two relatively
discrete domains of knowledge in the pursuit of a more robust theoretical guide for urban
social-ecological research.
HOAs AND ENVIRONMENTAL RESOURCE MANAGEMENT
With few exceptions, scholarly attention has ignored the role of HOAs as environmental
resource management institutions. The authors are aware of only one published empirical
example. Martin et al. (2003) found that residences with CCRs in Phoenix tended to have
landscape designs associated with lower water use, although nearly half of the respondents
in the associated HOAs expressed a preference for landscaping that included some pro-
portion of turf (ibid.). More generally, others have written about the wide range of actors
and influences that shape residential outdoor space. Most notably, Robbins (2007) argues
that the normative values of homeowners and the lawn care industry mutually construct a
political ecology of lawn care with sometimes dubious environmental repercussions. Cook
et al. (2011) synthesize the broad array of human and ecological factors that exert influ-
ence on residential landscapes as social-ecological systems, noting the potential impact of
formal institutions, such as HOAs, on landscape management practices. In the absence of
research that specifically addresses the environmental resource implication of HOAs, we
turn to two discrete conceptual perspectives—social geography and environmental com-
mons—to guide our empirical analysis of HOAs and water use in Phoenix.
4Although in theory, by deciding to purchase housing in HOA x or y one is exhibiting a preference, it is common
for individuals to purchase a home without reading the CCRs of an HOA carefully. Furthermore, as fewer and
fewer residences do not have HOAs, individuals are increasingly constrained in their choices in this matter.
1170 TURNER AND IBES
HOAs and Public Goods
A wealth of literature across the social sciences and urban planning posits that HOAs,
and the types of communities they govern, organize the urban landscape in ways that
facilitate the capture of municipal resources. They do so by organizing residential areas
in ways that enhance property values and offer protection from the volatility of the city at
large. Flight from urban blight is certainly not a new phenomenon—social scientists have
chronicled the ways in which the wealthy have sought to spatially distance themselves
from urban ills since the beginning of the industrial revolution. Blakely and Schneider
(1999) posit that private communities in general, and gated communities in particular,
are the most recent and extreme manifestation of this recurrent theme. Private communi-
ties enhance segregation by controlling access, privatizing amenities, and erecting physi-
cal barriers in the pursuit of exclusivity (ibid.), protection of property values (McKenzie,
1994), and escape from perceived danger (Low, 2003).
Social scientists have documented several spillover effects associated with private
residential communities, including the creation of informal subsidies, urban decline, and
secession from civic responsibility. Often built at the city periphery, planned communi-
ties require an extension of municipal infrastructure and services such as roads, plumb-
ing, firefighting, police, and schools, the cost of which is not always recovered through
property taxes. In effect, the tax dollars of the inner city subsidize the development of
these communities (Kennedy, 1995; Duany et al., 2000). This filtering of the wealthy
into the periphery often leads to urban decline, as cities lack a sufficient tax base to pro-
vide the funds to maintain infrastructure and municipal services (Vesselinov et al., 2007;
Duany et al., 2000). Increasingly, this conceptualization of the urban core declining as
the wealthy move to the suburbs is too simplistic. Private communities are often situated
adjacent to unprivatized properties; however, the effect on the distribution of municipal
resources is similar. Members of private communities are taxed via HOA dues which
fund services and amenities that financially strapped municipalities are less equipped
to provide. Some municipalities actively encourage the development of private com-
munities as surrogate providers of amenities such as parks and open space, yet there is
a fundamental difference in that such places exclude the public at large and potentially
discourage the development and maintenance of public civic spaces (Kennedy, 1995;
Webster, 2001). Furthermore, a secessionist spirit has been documented among indi-
viduals in private communities, in which they retreat from civic responsibilities such
as paying taxes and coalesce under the auspices of an HOA to rally against undesirable
municipal decisions (Purcell, 1997).
By extension, it is reasonable to suppose that an additional, if not yet documented,
effect might be that these communities command a disproportionate share of a city’s envi-
ronmental resources as well. The HOA, acting as a monolithic arbitrator for the wealthy,
facilitates the capture of resources for their needs. From a social geographic perspective,
HOAs can be expected to positively correlate with resource use, commanding large quan-
tities of municipal resources to maintain consumptive living conditions associated with
wealth. HOAs, therefore, might be expected to demand a higher than average share of
municipal water resources to maintain water-consumptive residential landscapes in order
to elevate property values.
HOMEOWNERS ASSOCIATIONS AND RESIDENTIAL WATER DEMAND 1171
HOAs as Regulators of the Commons
Research on the governance of the commons suggests a more complex relationship
between HOAs and environmental resources. HOAs can be conceptualized as managers
of the commons. An environmental commons is a “natural or man-made resource system
that is sufficiently large as to make it costly (but not impossible) to exclude potential
beneficiaries from obtaining benefits from its use,” (Ostrom, 1990, p. 30). In this instance,
the resource system is the municipal water supply, which is distinct from the resource
units, which are quantities of water. There may be many individuals or firms that are
beneficiaries, or “appropriators” of resource units, with access to the resource system.
In Phoenix, these beneficiaries include industrial, agricultural, and residential users with
access to municipal water resources. As a manager of the commons, HOAs influence the
quantity of water appropriated to a particular beneficiary, the residential community, within
the broader set of beneficiaries with access to the municipal water resource system.
Environmental commons research has examined a wide variety of resource systems
operating at different scales with very different social and biophysical contexts and
managing institutions, although the bulk of the foundational research is associated with
topics such as rangelands (Swallow and Bromley, 1995) and fisheries (Berkes, 2009) in a
local, rural, or developing-nation context, and global climate change research (Ostrom et
al., 1999). Application of commons theory to local urban resource systems in developed
nations—a less frequent, nevertheless emergent literature (Schlager and Blomquist, 2001;
Acheson and Brewer, 2003; Hanna, 2003)—indicates that commons theory is generaliz-
able to such resource systems.
Systematic research on the environmental commons has revealed that no particular
institutional type guarantees a positive or negative outcome; there is no “one-size-fits-all”
strategy that works best for all commons (Ostrom, 2007, 2009). Rather, environmental
resource management outcomes are facilitated by institutions that are imbedded in par-
ticular contexts and facilitated by a variety of “deeper tier” variables (ibid.). In contrast,
other work postulates that small-scale environmental decisions made in the absence of a
collective agenda (e.g, commons or HOA) may lead to a “tyranny of small decisions” that
generate resource depletion and other unintended consequences at broader scales (Odum,
1982, p. 782; Cooper et al., 2007). This process may seem particularly applicable in urban
landscapes that tend to be highly fragmented in terms of urban form, ownership, and man-
agement (Grimm et al., 2008a, 2008b).
The view that HOAs represent a diverse set of environmental resource management
goals and outcomes compliments “green” residential planning and design practices and
movements such as Smart Growth, New Urbanism (Duany et al., 2000), transit-oriented
development, Leadership in Energy and Environmental Design (LEED) certifications
(Green Building Council, 2010), biodiversity and open-space conservation subdivisions
(Arendt, 1996), and the incorporation of organic agriculture as a subdivision amenity
(Applebaum, 2009). Such planned development schemes also often utilize CCRs to
codify sustainability goals. Unfortunately, little work has linked the environmental and
resource implications of these interests to the resource management potential of HOAs
more broadly. Empirical evidence of environmental claims is limited and is not yet sub-
stantiated by long-term monitoring. Furthermore, empirical research connecting “green”
projects to environmental resource theory is limited and ambiguous at best. For example,
1172 TURNER AND IBES
Hurley and Halfacre (2009) found that the impact of “green” planning schemes can be
both a benefit to some while a detriment to other users of the resource system. Their
case study of exurban growth in South Carolina illustrates the ways in which a conser-
vation of sweetgrass via subdivision development limited access to a common resource
through privatization, thus excluding the local rural poor from harvesting basket-making
resources, jeopardizing both their livelihoods and the preservation of a long-standing cul-
tural practice. Several authors offer a critical perspective on the emergent phenomenon
of sustainable urban development, arguing that such projects focus on local communities
at the expense of a larger environmental project (Bjelland et al., 2006), fail to consider
the broader socio-economic processes that facilitate decision-making (Veninga, 2004),
and deploy a limited conceptualization of environmental services and amenities closely
associated with middle-class American values (Zimmerman, 2001; Hurley and Halfacre,
2009). Such research illuminates the ambiguous outcomes associated with environmental
resource management in private communities—intentionally striving to achieve sustain-
able outcomes or not.
In this paper we explore the extent to which the presence of HOA governance has influ-
enced water demand in residential subdivisions in Phoenix, Arizona. Our primary question
asks: Does single-family residential water use (annual household consumption) among
residences in HOAs in Phoenix differ significantly from use in non-HOA residences and,
if so, is this difference registered in overall outdoor water use? To address this question
we examined the spatial distribution and extent of HOAs in Phoenix in relation to a set of
social and biophysical determinants of water use.
DATA AND METHODS
Study Area
The Phoenix Metropolitan Area is located in the arid northern Sonoran Desert of the U.S.
Southwest. The City of Phoenix, the largest of the 28 municipalities, covers some 1,228
sq. km and houses over 1.5 million residents (U.S. Census, 2006). The region receives an
average of 21.1 cm of precipitation annually (NOAA, 2010). This supply is augmented by
groundwater, local surface water from the Salt and Verde rivers, and by the Central Arizona
Project Canal that delivers water from the Colorado River. Trends in population growth,
sprawling development, reliance on non-renewable groundwater supplies, high water use
rates, and a historic preference for water-intensive landscaping (e.g., in residential yards)
have stressed the region’s already scarce water resources (Hirt et al., 2008). Residential
landscapes in the region are typically classified as mesic (high water-use vegetation and
turf grass), xeric (low water-use plants on granite), oasis (a mixture of mesic and xeric),
or native (native Sonoran vegetation) (CAP-LTER, n.d.). The water-consumptive species
of mesic landscaping are often associated with elevated water demand. In some instances,
however, inefficient water management practices elevate water consumption in other resi-
dential landscape types (Stabler and Martin, 2000). In this context, the rising number of
HOAs that regulate residential landscape management practices raises concerns about
their impact on water use in the region.
HOMEOWNERS ASSOCIATIONS AND RESIDENTIAL WATER DEMAND 1173
Data
This study focused on HOAs zoned solely as single-family residential, as this dwell-
ing type accounts for nearly three times the water demand of multi-family units (Wentz
and Gober, 2007). Thousands of non-profit corporations are registered as HOAs in the
City of Phoenix, but current listings of HOAs are not spatially explicit. Further, available
contact information usually refers to the location of off-site management companies that
are often responsible for multiple HOAs in different residential subdivisions. This lack
of spatial information has therefore prevented the analyses of HOA locations in relation
to patterns of household water demand. Therefore, the first task of this research required
mapping the locations and extent of HOA communities across the region. Data used for
this step included a GIS data layer, obtained through the Maricopa County Recorder’s
Office, depicting subdivision boundaries for residential communities with and without
HOAs as well as other subdivided land uses across the county (Table 1). Subdivisions with
CCRs filed with the Maricopa County Recorder’s Office were tagged as HOAs.
Previous research has linked household water demand to household size (number of
people that occupy home), lot size, the presence of pools, and type of landscaping (Wentz
and Gober, 2007). The most consumptive households are those with relatively more people
on larger lots that maintain a private pool and mesic yard (grass lawns and non-native veg-
etation, mostly shrubs and trees). Affluence also has been linked to increased household
water consumption (Harlan et al., 2009; see also Balling et al., 2009), although this may
be an indirect correlation as wealthier households tend to maintain swimming pools and
larger lots (Wentz and Gober, 2007). Furthermore, HOAs have been linked empirically (La
Goix, 2005) and qualitatively (McKenzie, 1994) to wealthy, white populations.
Based on these studies, we analyzed the following independent variables as determi-
nants of our dependent variable, water demand: average household size (i.e., number of
people living in a single residence), income, race, lot size, presence of pools, and landscap-
ing type (Table 1). Home age was added because newer homes often possess drip irrigation
systems that may significantly reduce water demand for landscaping (Gleick, 2000). The
TABLE 1. s~êá~ÄäÉ=^ííêáÄìíÉë
Variables Source Spatial scale Temporal scale
Percent Hispanic U.S. Census Bureau Census block group 2000
Percent white U.S. Census Bureau Census block group 2000
Median income U.S. Census Bureau Census block group 2000
Mean household size U.S. Census Bureau Census block group 2000
Mean home age Maricopa County Parcel 2006
Mean lot size Maricopa County Parcel 2006
Percent pools Maricopa County Parcel 2006
Mean NDVIaCAP-LTER, n.d., [2011?],
Buyantuyev, 2007
n.a. 2005
HOA coverage original: digitized n.a. 2009
aNormalized Difference Vegetation Index.
1174 TURNER AND IBES
first three variables were drawn at the level of the census block group from the 2000 United
States census. Data for home age, lot size, and pools were obtained through the Maricopa
County Assessor at the parcel level, and later converted to the census block group level.
Landscaping type was drawn from extant examinations of NDVI (Normalized Difference
Vegetation Index) derived from a 2005 Landsat TM image (Buyantuyev, 2007).5 NDVI is a
measure of “greenness” capturing distinctions between areas dominated by grass, shrubs,
and trees from, for example, desert, bare soil, and impervious surfaces.
Water demand data used in this study were obtained from the City of Phoenix and rep-
resent the average amount of water used (in CCF, or cubic hundred feet) by single-family
households during 2006 by census block group (n = 997). Demand data do not distinguish
in- and outdoor use, although 70% of single-family residential water use involves outdoor
uses such as landscaping and pools. The role of indoor water use is captured in the deter-
minants of water demand, as there is evidence of a high correlation between the number of
people in a household and indoor water use (R2 = 0.9994) (Mayer et al., 1999, p. 90); thus
the inclusion of the variable household size can be said to act as a control for indoor use.
Although nearly 40% of residential water consumption occurs during the summer months
(Wentz and Gober, 2007), annual single-family water demand data capture total demand
on water supply, inclusive of seasonal variations.
Methods
Based on the spatially explicit dataset of all subdivision boundaries in Maricopa County,
all single-family residential subdivisions with CCRs were coded as belonging to an HOA,
while single-family subdivisions without a CCR were coded as non-HOA in kind. All
other types of subdivisions (e.g., commercial, industrial, mixed-use, schools, churches)
were removed from the dataset. The independent variable used in the regression model was
the percentage area of HOA and non-HOA single-family residences within each census
block group. All other data also were aggregated to the common scale of the census block
group to allow for valid comparisons among variables.
Analysis
Our analysis consisted of several steps: Pearson’s correlations, factor analysis, stepwise
regression, and validity tests. Again, all variables, including percent coverage of HOAs,
were analyzed at the census block group level. We first ran two sets of bivariate correla-
tions to determine if there was a relationship between the independent variables and (1)
percent of the census block group zoned as single-family residential and belonging to an
HOA, and (2) mean annual water demand (during 2006) for single-family households.
Next, we conducted a factor analysis in order to address multicollinearity issues among
our independent variables. We then created a stepwise regression model to quantify the
added predictive capacity, if any, of HOAs in determining water consumption. Finally, we
conducted a series of validity tests based on our data manipulation, aggregating all data to
the census block group scale.
5NDVI values (range –1 to 1) were computed as follows: (NIR – RED)/(NIR + RED), where NIR (near infrared)
is the TM band 4 (0.76–0.9 micrometers) and RED is Band 3 (0.78–0.82 micrometers).
HOMEOWNERS ASSOCIATIONS AND RESIDENTIAL WATER DEMAND 1175
Next, to address our principal question—do HOAs capture a larger share of water
resources than their non-HOA counterparts—we employed a stepwise multiple regression
model to examine the added predictive capacity of incorporating HOAs into a list of deter-
minants of water demand. The dependent variable was defined as the mean annual water
demand for single-family households within a given census block unit. The nine inde-
pendent variables included percent HOA coverage, median income, mean NDVI value,
mean home age, mean lot size, percent of homes with a pool, and the mean household size
by census block group. For the purpose of analysis, all variables were aggregated to the
census block. To avoid issues of multicollinearity, we ran a factor analysis on all of the
independent variables excluding percent HOA coverage, because our primary purpose was
to uncover its added predictive capacity. We then ran a second regression model using the
extracted factors.
Recognizing that one of our data limitations involved translating the geographic bound-
aries of the HOAs into percentages per census block, the final step of the analysis was
a validity test to determine if this data manipulation altered our results for the City of
Phoenix as a whole. To test this we compared descriptive statistics for census blocks with
100% HOAs versus those without any HOAs in order to determine if our results were
affected by the presence of HOA and non-HOA communities in the same census block.
If we establish that mean water consumption in census blocks with 100% HOAs and no
HOAs reflects our findings for all census blocks, we will have added confidence that our
manipulation of the data did not impact the integrity of our results.
RESULTS
This study provides the first mapping of single-family HOA subdivisions for the City
of Phoenix (Fig. 1), along with the following descriptive information. The City of Phoenix
has approximately 6,000 single-family residential subdivisions, some 27% (n = 1,642) of
which are governed by HOAs. While HOAs comprise a minority of these subdivisions in
number, they constitute a majority of the residential land area. Single-family residential
subdivisions consume approximately 37% of the total land area in the City of Phoenix,
64% of which is governed by HOAs. Spatially HOAs tend to be located on the periphery
of the city (Fig. 1). This study also classified census block groups by the percentage of
residential subdivisions governed by HOAs (Fig. 2). Again, larger, more recently devel-
oped areas around the periphery of the city tended to contain relatively more HOA subdivi-
sions than did more centrally located block groups. This observation raises concerns that
the percent HOA variable is affected by spatial autocorrelation, which could amplify the
observed relationships between independent variables and annual household consumption.
In order to address this concern, we tested for spatial autocorrelation among census block
groups with HOAs by calculating Moran’s I. We found that spatial clustering of census
block groups by percent HOA was not statistically significant at a 0.05 significance level
(z-score 0.397, p-value 0.691).
In a Pearson’s correlation test, percent HOA-governed residential subdivisions in each
census block group (CBG) had a weak but significant correlation with annual single-
family household water consumption (0.063), a moderate correlation with percent white
(0.261) and median income (0.299), and a weak correlation with household size (0.186).
1176 TURNER AND IBES
Percentage HOA area was negatively correlated with percent Hispanic (–0.292) and mean
home age (–0.495) (Tables 2 and 3). The strongest relationship was the negative correla-
tion between HOAs and older homes. These findings suggest that members of HOAs tend
to be wealthier, white, non-Hispanic populations that live in newer homes with pools.
The weak correlation between HOAs and more water-intensive lifestyles suggests that
the relationship is less straightforward. HOAs were not significantly correlated with
NDVI or “greenness,” nor were lots in HOAs significantly larger than lots in non-HOA
neighborhoods.
Fig. 1. Single-family residential subdivisions with and without HOAs.
HOMEOWNERS ASSOCIATIONS AND RESIDENTIAL WATER DEMAND 1177
In addition to the weak positive correlation with percent HOA (0.063), single-family
household water consumption had a strong positive correlation with median income (0.687)
and a moderate positive correlation with percent white (0.321), percent pools (0.264), and
lot size (0.175). Demand was negatively correlated with percent Hispanic (–0.297), and the
age of the home and household size were not significant determinants of household water
consumption. These results indicate that higher water demand tends to be a characteristic
of higher income, white populations with greener lawns, more pools, and larger lots, while
lower demand is more characteristic of Hispanic populations (Tables 2 and 3).
Fig. 2. Census blocks by percentage HOA.
1178 TURNER AND IBES
Many of our independent variables are correlated. Therefore, to address issues of multi-
collinearity, we ran a factor analysis on all of our independent variables excluding percent
HOA, because our primary research question seeks to discover its added predictive capac-
ity. Our analysis generated two factors that appear to cluster those variables associated
with socioeconomic status and also those associated with outdoor home features. Factor 1
included percent Hispanic (0.923), income (–0.775), household size (0.631), and home age
(0.558) (Table 4). Factor 2 included lot size (0.651) and NDVI (0.503). Percent pools was
not included in either factor. Next, we ran an additional regression model using Factor 1,
Factor 2, percent pools, and percent HOA. We found that Factor 1 and Factor 2 explained
TABLE 2. mÉ~êëçåÛë=ÅçêêÉä~íáçåëW=mÉêÅÉåí=ÅçîÉê~ÖÉ=çÑ=el^ë=~åÇ=ëçÅá~ä=~åÇ=
ÄáçéÜóëáÅ~ä=ÇÉíÉêãáå~åíë=çÑ=ï~íÉê=ÇÉã~åÇ=Äó=ÅÉåëìë=ÄäçÅâ=Öêçìé=En=Z=VVOF
Variable Percent HOA Household water demand
Percent HOA 1 0.063*
Household water demand 0.063* 1
Household size 0.186** –0.018
Income 0.299** 0.687**
NDVI 0.033 0.268**
Home age –0.495** –0.034
Lot size –0.058 0.175**
Percent pool 0.165** 0.264**
Percent Hispanic –0.292** –0.297**
Percent white 0.261** 0.321**
a*p > 0.05; ** p > 0.01.
TABLE 3. mÉ~êëçåÛë=r=`çêêÉä~íáçåë=Ñçê=`Éåëìë=_äçÅâ=dêçìéë=En=Z=VVOFa
Variable Pct.
HOA
Household
water
demand
HH size Income NDVI Home age Lot size Pct. pool
Pct. HOA 1 – – – – – – –
Household
water demand
0.063* 1 – – – – – –
Household
size
–0.009 –0.060 1 – – – – –
Income 0.299** 0.687** –0.174** 1 – – – –
NDVI 0.033 0.268** –0.103** 0.288** 1 – – –
Home age –0.495** –0.034 0.186** –0.405** –0.058 1 – –
Lot size –0.058 0.175** 0.037 0.080* 0.126** 0.042 1 –
Pct. pool 0.165** 0.264** –0.066* 0.403** 0.162** –0.093** 0.058 1
a*p > 0.05; ** p > 0.01.
HOMEOWNERS ASSOCIATIONS AND RESIDENTIAL WATER DEMAND 1179
a majority (65%) of the variance. Percent pools and our variable of interest, percent HOA,
provided nominal explanatory value (less than 2 and 1 percent, respectively) (Table 5).
Our findings are potentially affected by the conversion of our variable of interest, HOA,
into percent HOA per census block group unit. In doing so, our results may no longer
reflect the influence of HOAs but may be more a reflection of characteristics associated
with the block group boundary or heterogeneity caused by the presence of HOAs and non-
HOAs together in a CBG. Therefore, to verify that our results with regard to household
water demand were not skewed by the conversion from subdivision boundary to census
block groups, we ran descriptive statistics and created histograms for two subsets of the
data: (1) census block groups in which all residential subdivisions were governed by HOAs
(n = 129); and (2) census block groups without any residential HOAs (n = 102) (Figs. 3 and
4). Only minor variations were found. The n-values and mean water use for both sets of
data were similar and demand was highly variable in both cases. Mean water use in CBGs
with 100% HOAs was 183 gallons per capita per day (GPCD) with a standard deviation
of 43.9, and demand in non-HOA block groups was 168 GPCD with a standard deviation
of 41.2. Water demand in non-HOA block groups were slightly more skewed and among
HOA block groups the minimum unit of consumption (n = 87) was more than twice that of
non-HOA block groups (n = 42). The minor variations in these two datasets therefore add
validity to our findings with regard to the entire dataset.
DISCUSSION
Despite positive correlations between income and HOAs, income and water consump-
tion, and the fact that income explains a large percentage of the variation in water demand,
HOAs did not exhibit a strong significant correlation with water demand. This finding may
be a function of differences between those determinants of water demand associated with
wealth and those associated with HOAs. High income is associated with newer homes,
more pools, and “greenness”; however, HOAs are only weakly correlated with more pools
and are not correlated with NDVI, although they are correlated with newer homes. The
weaker correlation between HOAs and pools compared to income and pools may reflect
the prevalence of community pools—as opposed to private, backyard pools—in develop-
ments with HOAs. Second, unlike income, HOAs were not significantly correlated with
NDVI.
This finding indicates a diversity of landscaping treatments throughout the City of Phoe-
nix. Such diversity reflects an increase in the popularity of desert landscaping aesthetics
beginning in the late 20th century, as well as land use and water legacies. For example,
some neighborhoods are able to maintain green lawns year round through flood irrigation
or “grandfathered” water rights from previous, agricultural land uses. Furthermore, mid-
century desert oasis (green lawns) aesthetics are protected in historical neighborhoods.
These older neighborhoods may utilize voluntary neighborhood associations that are
institutionally different than HOAs and were not included in our study. Alternatively,
communities with HOAs may utilize common space as private parks in lieu of lawns on
individual lots. Different correlations with pools and NDVI, both of which are determi-
nants of water demand, may account for a lack of correlation between HOAs and water
use via wealth.
1180 TURNER AND IBES
Yet another potential explanation lies in the lack of association between HOAs and lot
size. Larger lots are associated with higher water consumption. However, HOAs have a
weak, but significant, negative correlation with lot size. This finding reflects the diversity
of the built environment and urban forms now developed in communities with HOAs.
TABLE 4. mêáåÅáé~ä=`çãéçåÉåí=^å~äóëáë=`çãéçåÉåí=j~íêáñ
Variable
Component
1 2
Percent Hispanic 0.923 0.245
Median income –0.775 0.278
Mean household size 0.631 0.456
Home age –0.558 0.183
Percent pools –0.488 0.433
Lot size –0.004 0.651
NDVI –0.381 0.503
Percent variance 36.833 17.747
Determinant 0.098
KMO 0.510
TABLE 5. píÉéïáëÉ=jìäíáéäÉ=RÉÖêÉëëáçå=Ñçê=mêáåÅáé~ä=`çãéçåÉåí=c~Åíçêë==
~åÇ=líÜÉê=RÉäÉî~åí=s~êá~ÄäÉë
Model
(Adjusted R2)Variables t
Unstandardized coefficients Standardized
BStandard error Beta*
1 (0.180) (Constant) 79.112 187.831 2.374
Factor 1 –14.779 –35.106 2.375 –0.425
2 (0.309) (Constant) 86.182 187.831 2.179
Factor 1 –16.100 –35.106 2.181 –0.425
Factor 2 13.633 29.727 2.181 0.360
3 (0.325) (Constant) 64.759 198.827 3.070
Factor 1 –16.420 –42.080 2.563 –0.510
Factor 2 14.474 35.907 2.481 0.435
Percent pools –5.054 –0.825 0.164 –0.173
4 (0.331) (Constant) 45.759 208.972 4.567
Factor 1 –16.644 –44.397 2.667 –0.538
Factor 2 14.437 35.686 2.472 0.432
Percent pools –5.010 –0.820 0.164 –0.172
Percent HOA –2.991 –17.947 6.000 –0.083
*p > 0.01.
HOMEOWNERS ASSOCIATIONS AND RESIDENTIAL WATER DEMAND 1181
Furthermore, developers may have an incentive to create communities with smaller lot
sizes in order to increase the number of units available for sale and, therefore, profit.
Smaller lots may be supplanted with common space like pocket parks and community
pools, therefore contributing to the lack of association between HOAs and NDVI (see
above).
The strong relationship between HOAs and newer homes as well as income and newer
homes, paired with a lack of significant correlation between new homes and water use,
points to further distinctions. The average home in an HOA is younger than those in non-
HOAs, and thus they may be more likely to possess water-efficient infrastructure, such
as drip irrigation technology, as opposed to inefficient surface irrigation. The benefits of
water-efficient technology, however, may be tempered by trade-offs with other factors that
increase water consumption. Our results indicate that there is a preference for new homes,
in HOAs, among wealthier individuals, but that high water demand in newer homes is more
a function of wealth than location within an HOA. Further, water-saving technologies in
newer homes may be offset by high water use among wealthier households as well as the
fact that outdoor water use constitutes the majority of residential demand. Additionally,
newer developments, especially those built on the periphery of the city, may be more likely
to utilize native desert landscaping because of shifting consumer preferences, proximity
to open-space desert preserves, and state and municipal regulations on plant material and
land cover. On the other hand, desert landscapes also require water inputs and overwater-
ing may be commonplace (Stabler and Martin, 2000). It is also possible that newer homes
tend to be larger than older homes and therefore require more indoor water use. These
questions offer potential for further research.
Fig. 3. Histogram of block groups in which 100% of the subdivisions are governed by HOAs.
1182 TURNER AND IBES
HOAs consume a larger percentage of single-family residential land than non-HOAs
and tend to be concentrated at the city periphery. Intuitively, this may reflect the fact that
larger, less expensive tracts of land are available for development at the periphery than in
the older, central city. Logic dictates that the typical residential development at the periph-
ery is newer and wealthier, with larger lots and more amenities like pools and green space
relative to typical residential areas in the central city. These would be expected to be older
homes on small lots with fewer amenities such as pools and green space. There are several
notable exceptions to such one-dimensional descriptions of urban form that may explain
the weak correlations that create ambiguity in our findings. Examples of non-HOAs in
the periphery include older, lot-by-lot “stick-built,” communities, retirement communi-
ties, large single-owner estates, and platted subdivisions that have not yet been completed
and lack registered CCRs. Within the central city, there are 18 historic districts and many
neighborhood associations that are voluntary and institutionally distinct from HOAs, but
may function informally in ways that impact water demand. For example, in historic neigh-
borhoods with a legacy of green lawns, residents may follow informal “rules” to maintain
them rather than convert to desert landscaping in order to reduce water consumption. Such
exceptions add to the complexity of the residential landscape in the city.
The possibility that some HOAs manage residential landscapes in ways that provide
preferred outcomes through efficient use of water suggests that the HOA may be an impor-
tant institution in water demand management in the future. For example, policies may be
introduced that encourage the participation of HOAs in informational campaigns, retro-
fitting landscaping and irrigation infrastructure, halting winter seeding, and other water
demand reduction strategies. While the potential for HOAs to act as an entry point for
Fig. 4. Histogram of block groups in which 0% of the subdivisions are governed by HOAs.
HOMEOWNERS ASSOCIATIONS AND RESIDENTIAL WATER DEMAND 1183
residential water demand management is significant, there are several barriers. The most
significant barrier, perhaps, is the difficulty in changing the CCRs commensurate with
needed improvements or changes in resource and landscape use. Supermajority votes are
likely to be especially difficult in changes involving financial costs, such as retrofitting
lawns to desertscapes. The coupled effect of immutable CCRs and the legacy of the origi-
nal “yardscapes” may constitute a significant obstacle for water demand reduction strate-
gies. Finally, water demand management in HOAs potentially affects water supplies for
the city at large. HOAs tend to be home to wealthy, white populations more capable of
absorbing water costs than less wealthy populations, the lower socio-economic strata of
which do not live in HOAs. This raises significant social and environmental justice ques-
tions.6 The City of Phoenix has a relatively diverse water portfolio, and water availability
and price has not yet become a significant impending threat. In peripheral communities
with less diverse portfolios and less availability, however, the water consumption habits of
wealthy inhabitants, HOA members or not, is a potential concern.
This study found that census blocks with high percentages of HOAs are correlated with
higher water use, but the correlation was weak and the variable did not surface in the step-
wise regression. Furthermore, census blocks comprised entirely of HOAs exhibited high
deviance from the mean water demand. These differences may reflect heterogeneity in the
institutional structure and dynamics of different HOAs, as well as the importance of the
broader social-environmental context in which landscaping and management decisions are
made. Our finding that census blocks comprised entirely of non-HOAs also exhibit devi-
ance from mean water demand may reflect a legacy of uneven development in historic,
central Phoenix (Bolin et al., 2005). Downtown Phoenix is characterized by large, irrigated
turf lots in wealthier neighborhoods and a lack of landscaping altogether in the less afflu-
ent South Phoenix neighborhoods. More research is necessary to characterize the diversity
within both HOAs and non-HOAs and identify the mechanisms driving these differences.
Data Limitations
We recognize that our research was limited by several data and analytical barriers. We
were unable to use the physical bounds of the HOAs to determine correlations with the
other variables, primarily because the scale at which they were reported was mismatched
with our subdivision boundaries. It was not feasible to translate the parcel- and raster-level
data to the scale of the subdivision given the large n-value for our HOAs and non-HOAs.
Furthermore, water use represents politically sensitive information in the Southwest and,
much like census data, reporting of water demand must be aggregated to protect the iden-
tity of the users. We compensated for these shortcomings by separating out those census
blocks that were entirely composed of HOAs or non-HOAs. Our study is also limited by
our scope. HOAs are a prominent fixture in the suburbs of sprawling metropolitan areas
like Phoenix and differences between communities with and without HOAs may be more
6While it is beyond the purview of this paper, it is worth noting that our findings empirically connecting HOAs
to white, wealthy populations raises concerns about public service and amenity distribution. Municipalities often
encourage the development of HOA communities as a mechanism to subsidize public services such as trash col-
lection and pools via privately collected HOA dues. These services provided through HOAs are not necessarily
extended to the broader community.
1184 TURNER AND IBES
pronounced there. Further, this analysis focused only on single-family residential HOAs,
excluding analysis of mixed-use or multi-family residential HOAs that may exhibit differ-
ent relationships among the variables. Finally, given the variety of HOA governing docu-
ments and potential incompleteness of the Maricopa County Recorder’s Office records,
our definition of HOAs as subdivisions with CCRs may have led to the exclusion of some
HOAs.
CONCLUSION
Our findings appear to reject a conceptualization of HOAs as monolithic institutions
driving high levels of resource consumption as evidenced in the weak, but significant,
correlation between HOAs and water use. Our findings suggest that HOAs may not neces-
sarily lead to increased water use by single-family home residents, despite speculation that
they do, and despite significant relationships with determinants of water demand relative to
non-HOA units. In order to explain this apparent contradiction we look to a conceptualiza-
tion of HOAs as managers of the environmental commons in a coupled social- ecological
system. In operation, HOAs are a diverse set of institutional actors that negotiate a com-
plex set of trade-offs and synergies associated with the determinants of water demand
in residential landscapes in the City of Phoenix. The ways in which specific rules and
regulations associated with different types of HOAs facilitate or constrain water demand
management merits further research.
The importance of HOAs as environmental resource management institutions on the
residential landscape cannot be understated. Indeed, HOAs are spatially dominant in
newly developed areas on the periphery of Phoenix and virtually all new residential devel-
opments utilize CCRs, providing an additional layer of governance to municipal and state
regulations, including those applied to landscaping and maintenance practices. As manag-
ers of a common resource, our research suggests that, in aggregate, HOAs do not neces-
sarily significantly alter water consumption as compared to communities without HOAs.
This finding is consistent with recent developments in commons theory that state that no
institutional type in of itself guarantees a particular management outcome. A more robust
understanding of management outcomes lies in understanding a “deeper” set of variables
associated with social-ecological systems. Despite the uncertain findings of our research,
HOAs are nonetheless influential water demand management institutions as they govern
some 64% of residentially zoned land in Phoenix (nearly 786 km2) and will, most likely,
continue to proliferate and include landscaping management clauses in their CCRs. It
is critical, therefore, to explain the mixed results of this study: an increased percentage
of HOA area only weakly correlates with higher levels of residential water demand and
did not increase the predictive capacity of regression models in determining residential
water demand. We argue that in order to explain this variance it will be critical to continu-
ing incorporating insights from commons and social-ecological theory into the study of
HOAs—a topic that has largely remained in the social science research domain—in order
to unpack the specific rules and regulations that influence demand for water resources.
We offer several points of departure in this endeavor. Commons theory has identified a
suite of variables related to resources, resource systems, resource users, and management
as well as the ways in which they interact that have consistently proven important to under-
stand environmental resource management outcomes in empirical research (Ostrom, 2009).
HOMEOWNERS ASSOCIATIONS AND RESIDENTIAL WATER DEMAND 1185
The CCRs pertaining to landscaping and management of HOAs are an important resource
for such analysis. Additionally, commons scholars and social-ecological research more
broadly have identified adaptive management strategies as the best institutional “fit” for
resource management under prevailing uncertainty. Presuming water stress will continue
to unfold in the southwestern U.S., the apparent uncertainty associated with water-climate
models, and the relative difficulty in altering community CCRs, the quantification of their
impacts on water consumption are of paramount concern. In this institutional climate, a
greater understanding of the trade-offs associated with water consumption in residential
landscapes may be necessary in order for individual HOAs to identify and target efforts
toward the demand management strategies most applicable to their needs.
In addition, our finding that HOAs do not ubiquitously command a disproportionate
share of water resources may reveal the untapped potential to utilize HOAs for neighbor-
hood-scale water demand management. Municipalities could facilitate such opportunities
in new developments by creating financial incentives for developers to incorporate water
demand management into new CCRs. Because HOA developments are already dominant
in Phoenix and in many other parts of the country, it is imperative to find creative ways
to retrofit existing HOAs with water-efficient landscaping and infrastructure and to over-
come legal barriers by redrafting CCRs where appropriate. Furthermore, HOAs might
be targeted for partnerships with government agencies already involved in water demand
management. Finally, HOAs can adopt principals associated with existing programs such
as the Sustainable Sites Initiative (SSI) or Leadership in Energy Efficiency Design for
Neighborhood Development (LEED-ND) that guide sustainable land use and management
practices. These opportunities require that HOAs move beyond a narrow goal of protecting
property values through aesthetics, and incorporate other quality-of-life components into
their management agendas.
REFERENCES
Acheson, J. M. and Brewer, J. F., 2003, changes in the territorial system of the Maine
lobster industry. In N. Dolsak, and E. Ostrom, editors, The Commons in the New Mil-
lennium: Challenges and Adaptation. Cambridge, MA: MIT Press, 37–60.
Appelbaum, A., 2009, Growing with the crops, nearby property values: Organic farms as
subdivision amenities. New York Times, July 1.
Arendt, R. G., 1996, Conservation Design for Subdivisions. Washington, DC: Island
Press.
Arizona State Senate Issue Brief, 2010, Homeowners association. Retrieved August 24,
2010 from http://www.azleg.state.az.us/briefs/Senate/HOMEOWNERS%20ASSOCIA
TIONS.pdf
Balling, R. and Cubaque, H. C., 2009, Estimating future residential water consumption in
Phoenix, Arizona based on simulated changes in climate. Physical Geography, Vol. 30,
308–323.
Balling, R. and Goodrich, G. B., 2010, Increasing drought in the American Southwest? A
continental perspective using a spatial analytical evaluation of recent trends. Physical
Geography, Vol. 31, 293–306.
1186 TURNER AND IBES
Barry, B., 2002, Social exclusion, social isolation, and the distribution of income. In J.
Hills, J. Le Grand, and D. Piachaud, editors, Understanding Social Exclusion. Oxford,
UK: Oxford University Press, 13–29.
Berkes, F., 1985, Fishermen and “the tragedy of the commons.” Environmental Conserva-
tion, Vol. 12, 199–206.
Bjelland, M. D., Maley, M., Cowger, L., and Barajas, L., 2006, The quest for authentic
place: The production of suburban alternatives in Minnesota’s St. Croix Valley. Urban
Geography, Vol. 27, 253–270.
Blakeley, E. J. and Schneider, M. G., 1999, Fortress America: Gated Communities in the
United States. Washington, DC and Cambridge, MA: The Brookings Institute/Lincoln
Land Institute.
Bolin, B., Grineski, S., and Collins, T., 2005, The geography of despair: Environmental
racism and the making of South Phoenix Arizona, USA. Research in Human Ecology,
Vol. 12, 156–168.
Buyantuyev, A., 2007, CAP-LTER data resources: NDVI (Normalized Difference Vegeta-
tion Index) of the 2005 Landsat Thematic Mapper image. Retrieved June 25, 2010 from
http://caplter.asu.edu/data/?path=/exist/rest/db/datasets/util/xquery/getDatasetById.
xql?_xsl=/db/datasets/util/xslt/datasetHTML.xsl&id=knb-lter-cap.371.1
CAP-LTER: Central Arizona–Phoenix Long-Term Ecological Research, n.d., Research
highlight 2: North Desert Village experiment. Retrieved August 1, 2011 from http://
caplter.asu.edu/research/research-projects/research-home/research-highlight-2/
City of Phoenix, 2010, Outdoor water use. Retrieved August 24, 2010 from http://phoenix
.gov/waterservices/wrc/home/outdoor/index.html
Chang, H., Parandvash, G. H., and Shandas, V., 2010, Spatial variations of single- family
residential water consumption in Portland, Oregon. Urban Geography, Vol. 31,
953–972.
Community Associations Institute, 2010, Retrieved August 24, 2010 from http://www
.caionline.org/info/research/Pages/default.aspx
Cook, E. M., Hall, S. J., and Larson, K. L., 2011, Residential landscapes as social-
ecological systems: A synthesis of multi-scalar interactions between people and their
home environment. Urban Ecosystems [doi: 10.1007/s11252-011-0197-0].
Cooper, C. B., Dickinson, J., Phillips, T., and Bonney, R, 2007, Citizen science as a tool
for conservation in residential ecosystems. Ecology and Society, Vol. 13. Available at
http://www.ecologyandsociety.org/vol13/iss2/resp1
Duany, A., Plater-Zyberk, E., and Speck, J., 2000, Suburban Nation: The Rise of Sprawl
and the Decline of the American Dream. New York, NY: North Point Press.
Gleick, P. H., 2000, A look at twenty-first century water resources development. Water
International, Vol. 25, 127–138.
Gammage, G., Jr., Hall, J. H., Land, R. E., Melnick, R., and Welsch. N., 2008, Megapoli-
tan: Arizona’s Sun Corridor. Tempe, AZ: Arizona State University, Morrison Institute
of Public Policy.
Gober, P., 2006, Metropolitan Phoenix: Place Making and Community Building in the
Desert. Philadelphia, PA: Philadelphia University Press.
Gober, P. and Kirkwood, C. W., 2010, A vulnerability assessment of climate-induced water
shortage in Phoenix. Proceedings of the National Academy of Sciences USA, Vol. 107,
21,295–21,299.
HOMEOWNERS ASSOCIATIONS AND RESIDENTIAL WATER DEMAND 1187
Gober, P., Kirkwood, C. W., Balling, R. C., Elliz, A. W, and Deitrick, S., 2010, Water plan-
ning under climatic uncertainty in Phoenix: Why we need a new paradigm. Annals of
the Association of American Geographers, Vol. 100, 356–372.
Green Building Council, 2010, Retrieved August 24, 2010 from http://www.usgbc.org/
DisplayPage.aspx?CategoryID=19
Grimm, N. B., S. H. Faeth, N. E. Golubiewski, C. L. Redman, J. Wu, X. Bai, and J. M.
Biggs, 2008, Global change and the ecology of cities. Science, Vol. 319, 756–760.
Grimm, N. B., Grove, J. M., Pickett, S. T., and Redman, C., 2008, Urban ecology: Inte-
grated approaches to long-term studies of urban ecological systems. Bioscience, Vol.
50, 571–584.
Hanna, S., 2003, “Transition in the American fishing commons: Management problems
and institutional design challenges” In N. Dolsak, and E. Ostrom, editors, The Com-
mons in the New Millennium: Challenges and Adaptation. Cambridge, MA: MIT Press,
61–76.
Harlan, S. L., Yabiku, S. T., Larsen, L., and Brazel, A. J., 2009, Household water consump-
tion in an arid city: Affluence, affordance, and attitudes. Society & Natural Resources,
Vol. 22, 691–709.
Hirt, P., Gustafson, A., and Larson, K., 2008, The mirage in the Valley of the Sun. Environ-
mental History, Vol. 13, 482–513.
House-Peters, L. A. and Chang, H. C. 2011, Urban water demand modeling: Review of
concepts, methods, and organizing principles. Water Resources Research, Vol. 47,
W05401 [doi:1029/2010WR009624].
Hurley, P. T. and Halfacre, A. C., 2009, Dodging alligators, rattlesnakes, and backyard
docks: A political ecology of sweetgrass basket-making and conservation in the South
Carolina Lowcountry, USA. GeoJournal. Retrieved August 24, 2011 from http://www
.springerlink.com/content/cjnr23k35871302w/fulltext.html
Karl, T. R., Melillo, J. M., and Peterson, T. C., editors, 2009, Global Climate Change
Impacts in the United States: A State of Knowledge Report from the U.S. Global Change
Research Program. New York, NY: Cambridge University Press.
Kennedy, D. J., 1995, Residential associations as state actors: Regulating the impact of
gated communities on nonmembers. Yale Law Journal, Vol. 105, 727–760.
Larson, K., Hall, S., Cook, E., Funke, B., Strawhacker, C., and Turner, V. K., 2008, The
Socioecology of Residential Landscapes. Tempe, AZ: CAP-LTER Workshop Report.
La Goix, R., 2005, Gated communities: Sprawl and social segregation in Southern
California. Housing Studies, Vol. 20, 323–343.
Low, S. M., 2004, Behind the Gates: Life, Security, and the Pursuit of Happiness in Fortress
America. New York, NY: Routledge.
Martin, C., Peterson, K., and Stabler, L., 2003, Residential landscaping in Phoenix,
Arizona, U.S.: Practices and preferences relative to covenants, codes, and restrictions.
Journal of Arboriculture, Vol. 29, 9–17.
Mayer, P. W., DeOreo, W. B., Opitz, E., Kiefer, J., Dziegielewski, B., Davis, W., and Nelson,
J. O., 1999, Residential End Uses of Water. Denver, CO: American Water Works Asso-
ciation Research Foundation.
McKenzie, E., 1994, Privatopia: Homeowner Associations and the Rise of Residential
Private Government. New Haven, CT: Yale University Press.
1188 TURNER AND IBES
Morehouse, B. J., 2000, Climate impacts on urban water resources in the Southwest:
The importance of context. Journal of American Water Resource Association, Vol. 36,
256–277.
NOAA, 2010, NOWData: NOAA online weather data, Phoenix precipitation. Retrieved
July 25, 2010 from http://nowdata.rcc-acis.org/PSR/pubACIS_results
Odum, W. E., 1982, Environmental degradation and the tyranny of small decisions. Bio-
Science, Vol. 32, 728–729.
Ostrom, E., 1990, Governing the Commons: The Evolution of Institutions for Collective
Action. New York, NY: Cambridge University Press.
Ostrom, E., 2007, Going beyond panaceas. Proceedings of the National Academy of
Sciences, Vol. 39, 15,176–15,178.
Ostrom, E., 2009, A general framework for analyzing sustainability in social-ecological
systems. Science, Vol. 24, 419–422.
Ostrom, E., Burger, J., Field, C. B., Norgaard, R. B., and Policansky, D., 1999, Revisiting
the commons: Local lessons, global challenges. Science, Vol. 284, 278–282.
Purcell, M., 1997, Ruling Los Angeles: Neighborhood movements, urban regimes, and the
production of space in Southern California. Urban Geography, Vol. 18, 684–704.
Robbins, P., 2007, Lawn People: How Grasses, Weeds, and Chemicals Make Us Who We
Are. Philadelphia, PA: Temple University Press.
Romig, K., 2010, Community and social capital in upper-income neighborhoods: An
investigation in metropolitan Phoenix. Urban Geography, Vol. 31, 1065–1079.
Schlager, E. and Blomquist, W., 2001, Water resources: The Southwest United States. In
Protecting the Commons: A Framework for Resource Management in the Americas.
Washington, DC: Island Press, 133–160.
Swallow, B. M. and Bromely, D. W., 1995, Institutions, governance, and incentives in
common property regimes for African rangelands. Environment and Resource Econom-
ics, Vol. 6, 99–118.
Stabler, L. B. and Martin, C. A., 2000, Irrigation regimens differentially affect growth
and water use efficiency of two Southwest landscape plants. Journal of Environmental
Horticulture, Vol. 18, 66–70.
U.S. Census Bureau, 2006, Phoenix (city) quickfacts: 2006 population estimates. Retrieved
June 25, 2010 from http://quickfacts.census.gov/qfd/states/04/0455000.html
Veninga, C., 2004, Spatial prescriptions and social realities: New Urbanism and the pro-
duction of Northwest Landing. Urban Geography, Vol. 25, 458–482.
Vesselinov, E., Cazessus, M., and Falk, W., 2007, Gated communities and spatial inequal-
ity. Journal of Urban Affairs, Vol. 29, 109–127.
Webster, C., 2001, Gated cities of tomorrow. The Town Planning Review, Vol. 72, 149–
170.
Wentz, E. and Gober, P., 2007, Determinants of small area water consumption for the City
of Phoenix, Arizona. Water Resources Management, Vol. 21, 1849–1863
Zimmerman, J., 2001, The “nature” of urbanism on the New Urbanist frontier: Sustainable
development, or defense of the suburban dream? Urban Geography, Vol. 22, 249–267.
