# RECORD 02b5387f9679dc6d340d34b2c07dc617e35c96d3
# TITLE: Property Values, Water Quality, and Benefit Transfer: A Nationwide Meta-analysis
# YEAR: 2021.0  VENUE: Land Economics  DOI: 10.3368/le.98.2.050120-0062r1
# AUTHORS: D. Guignet; M. Heberling; M. Papenfus; O. Griot
# PAGES: 28  CANDIDATE TABLE PAGES (1-based): [7, 11, 14, 17]
# ABSTRACT: We construct a comprehensive, publicly available meta-dataset based on 36 hedonic studies that examine the effects of water quality on housing values in the United States. The meta-dataset includes 656 unique estimates and entails a cluster structure that accounts for price effects at different distances. Focusing on water clarity, we estimate reduced-form meta-regressions that account for within-market dependence, statistical precision, housing market and waterbody heterogeneity, publication bias, and methodological practices. Although we find evidence of systematic heterogeneity, the out-of-sample transfer errors are large. We discuss the implications for benefit transfer and future work to improve transfer performance.


===== PAGE 1 (score 10, numeric-paren 0, table-words 0, var-words 10) =====

Land Economics • May 2022 • 98 (2): 191–218
DOI:10.3368/le.98.2.050120-0062R1
ISSN 0023-7639; E-ISSN 1543-8325
© 2022 by the Board of Regents of the
University of Wisconsin System
O    This open access article is distributed under the terms of the CC-BY-NC-ND license (http://creativecommons. 
org/licenses/by-nc-nd/4.0) and is freely available online at: http://le.uwpress.org 
S  Appendix materials are freely available at http://le.uwpress.org and via the links in the electronic version of 
this article.
191
Property Values, Water Quality, and Benefit Transfer: 
A Nationwide Meta-analysis  O  S
Dennis Guignet  Assistant Professor, Department of Economics, Appalachian State University, Boone, 
North Carolina; guignetdb@appstate.edu
Matthew T. Heberling  Research Economist, Office of Research and Development, U.S. Environmental 
Protection Agency, Cincinnati, Ohio; Heberling.Matt@epa.gov
Michael Papenfus  Environmental Economist, Office of Research and Development, U.S. Environmental 
Protection Agency, Corvallis, Oregon; Papenfus.Michael@epa.gov
Olivia Griot  Senior Analyst, Abt Associates, Inc., Cambridge, Massachusetts; olivia_griot@abtassoc.com
ABSTRACT  We construct a comprehensive, 
publicly available meta-dataset based on 36 
hedonic studies that examine the effects of 
water quality on housing values in the United 
States. The meta-dataset includes 656 unique 
estimates and entails a cluster structure that 
accounts for price effects at different dis­
tances. Focusing on water clarity, we estimate 
reduced-form meta-regressions that account 
for within-market dependence, statistical 
precision, housing market and waterbody 
heterogeneity, publication bias, and method­
ological practices. Although we find evidence 
of systematic heterogeneity, the out-of-sample 
transfer errors are large. We discuss the im­
plications for benefit transfer and future work 
to improve transfer performance. (JEL Q51, 
Q53)
1. Introduction
The hedonic literature examining the effects 
of surface water quality on residential prop­
erty values began over 50 years ago with 
David’s (1968) report. Since then, the litera­
ture has evolved significantly. To assess this 
literature’s suitability to support manage­
ment decisions related to water quality, we 
use meta-­analytic methods to synthesize and 
draw key conclusions from 36 unique studies 
in the United States. There are several exist­
ing meta-­analyses of hedonic property value 
studies, including applications to air quality 
(Smith and Huang 1993, 1995), contaminated 
sites (Messer et al. 2006; Kiel and Williams 
2007; Schütt 2021), open space (Mazzotta, 
Besedin, and Speers 2014), and noise (Nelson 
2004). To our knowledge, this study is the first 
comprehensive and rigorous meta-analysis of 
the hedonic literature examining surface wa­
ter quality.1
The results from meta-analyses can help 
make predictions for benefit transfer—where 
an analyst uses the predicted outcomes to in­
fer ex ante or ex post effects of some policy 
action, in lieu of conducting a new original 
study. Analyses of public policies often rely 
on benefit transfer because original studies 
require a lot of time and money or are infeasi­
ble because of data constraints. In fact, benefit 
transfer is one of the most common approaches 
used to complete benefit-cost analyses at the 
U.S. Environmental Protection Agency (U.S. 
EPA 2010; Newbold et al. 2018a). Improving 
benefit transfer, as well as combining limited 
but heterogeneous information for surface 
water quality changes, remains a priority for 
policy makers (Newbold et al. 2018a).
Several meta-analyses of stated preference 
studies on water quality have been published 
1 There are three notable unpublished studies. In her M.A. 
thesis, Fath (2011) conducted a limited meta-analysis of 13 
hedonic studies. Ge et al. (2013) conducted a meta-analysis 
that combined contingent valuation, travel cost, and hedonic 
studies. They estimated one meta-regression using only he­
donic studies for comparison (10 studies with 127 observa­
tions). Abt Associates (2015) estimated the capitalization 
effects of large-scale changes in water clarity of lakes using 
a simple weighted-average across nine hedonic studies. We 
are aware of one published meta-analysis that focused on 
urban rivers and property values, but it did not examine mea­
sures of water quality (Chen et al. 2019).
at BIBL DE L UNIVERSITE LAVAL on October 20, 2026. Copyright 2021
Downloaded from 


===== PAGE 7 (score 29, numeric-paren 16, table-words 1, var-words 4) =====

98(2)
197
Guignet et al.: Property Values and Water Quality
Table 1
Mean Elasticity Estimates of the Three Most Frequently Examined Water Quality Measures
Water Quality Measure
Unweighted 
Mean
(1)
Cluster-Weighted 
Mean
(2)
Variance-Adjusted 
Cluster-Weighted Mean
(3)
RES Cluster-Adjusted 
Weighted Mean
(4)
n
Studies
Clusters (Kd)
Chlorophyll a (mg/L)
Waterfront
0.737*
0.324*
−0.023***
−0.026***
18
  3
  3
[−0.044, 1.517]
[−0.036, 0.684]
[−0.028, −0.019]
[−0.031, −0.021]
Nonwaterfront within 500 m
0.005
0.01
0.008***
0.009***
18
  3
  3
[−0.201, 0.211]
[−0.085, 0.105]
[0.005, 0.010]
[0.006, 0.012]
Fecal Coliform (count per 100 mL)
Waterfront
−0.018***
−0.037
−0.028
−1.3E-4***
36
  4
  4
[−0.026, −0.011]
[−0.088, 0.014]
[−0.079, 0.023]
[−1.8E-4, −0.7E-4]
Nonwaterfront within 500 m
−0.020***
−0.058*
−0.061*
−0.052**
20
  3
  3
[−0.034, −0.006]
[−0.127, 0.010]
[−0.129, 0.008]
[−0.096, −0.008]
Water Clarity (Secchi disk depth, m)
Waterfront
0.158
0.206
0.191
0.109***
177
18
66
[−6.099, 6.416]
[−16.575, 16.987]
[−16.590, 16.972]
[0.099, 0.118]
Nonwaterfront within 500 m
0.028***
0.042***
0.041***
0.026***
83
  6
19
[0.020, 0.036]
[0.025, 0.059]
[0.024, 0.058]
[0.017, 0.034]
Note: Confidence intervals at the 95% level are displayed in brackets. Only elasticity estimates pertaining to the three most commonly examined water quality measures in the hedonic literature are pre­
sented, but the full suite of mean elasticity estimates are presented in Appendix B.2. We present the respective units for each water quality measure in parentheses just as a reference but emphasize that the 
elasticity estimates are unit-less. RES, random effect size.
* p < 0.1; ** p < 0.05; *** p < 0.01.
at BIBL DE L UNIVERSITE LAVAL on October 20, 2026. Copyright 2021
Downloaded from 


===== PAGE 8 (score 7, numeric-paren 0, table-words 1, var-words 5) =====

Land Economics
198
May 2022
same waterbody and housing market, so the 
weight given to those estimates must be re­
duced accordingly (Mrozek and Taylor 2002). 
We next present cluster-weighted means, 
where we define each cluster as a unique study 
and housing market combination. Meta-obser­
vations estimated from a common transaction 
dataset in terms of the study area, time period, 
and waterbodies are really just different esti­
mates of the same underlying “true” elasticity. 
No matter how many estimates are provided 
by a study for a specific location, each cluster 
as a whole is given the same overall weight. 
This holds regardless of whether the estimates 
for different clusters (i.e., housing markets) 
are from the same study or a different study. 
For example, Boyle, Poor, and Taylor (1999) 
estimate the effects of lake clarity on four dif­
ferent housing markets in Maine. These esti­
mates are each given a weight of one and thus 
are weighted the same as if they were esti­
mates for four different housing markets from 
four different studies.12
More formally, let εˆidj denote elasticity es­
timate i, at distance d, for cluster j, and dj
k  is 
the number of elasticity estimates for distance 
bin d in each cluster j. The cluster-weighted 
mean elasticity for each distance bin d is
{
}
ε
ε
=
=
= ∑
∑
1
1
1
,
ˆ
d
dj
dj
K
k
idj
k
j
i
d
d
K
 
[1]
where the same ω
= 1
dj
idj
k  weight is given to 
each meta-observation in cluster j and for that 
distance bin. The total number of clusters in 
the meta-dataset for distance bin d is Kd.
The cluster-weighted mean elasticities are 
presented in Table 1, column (2). The results 
are generally similar, suggesting a marginally 
significant increase in waterfront home values 
in response to an increase in the concentration 
of chlorophyll a, and again an insignificant 
effect on the value of nonwaterfront homes. 
The negative price elasticity among water­
front homes with respect to fecal coliform is 
12 Ara (2007) statistically identified and separately ana­
lyzed several submarkets when estimating the housing price 
effects around Lake Erie. These submarkets sometimes 
overlap because of different statistical strategies, and so in 
our meta-analysis we treat all estimates from Ara (2007) as 
being from the same broader housing market.
now insignificant. The mean elasticity with 
respect to fecal coliform counts for nonwa­
terfront homes is only marginally significant 
but larger in magnitude, suggesting a 0.06% 
decrease in value due to a 1% increase in fecal 
coliform counts. The cluster-weighted mean 
elasticities with respect to water clarity are 
similar to the unweighted means, suggesting a 
positive but insignificant effect on waterfront 
home prices, and a significant 0.04% increase 
in nonwaterfront home prices due to a 1% in­
crease in Secchi disk depth.
We propose an adjustment to the above 
cluster weights that redistributes the weight 
given to each observation within a cluster and 
distance bin. Consider the incorporation of a 
reallocation parameter (ridj) to the above clus­
ter weights, as follows:
ω
= 1
,
dj
idj
idj
k r
 
[2]
where 
≤
≤
0
idj
dj
r
k . This is a generalization 
of the above cluster weights, where 
=1
idj
r
. 
This is also a generalization of the approach 
taken by some meta-analysts who select a 
single “preferred” estimate from each clus­
ter, in which case 
=
idj
dj
r
k  for the preferred 
estimate and zero otherwise. As discussed, 
choosing a single preferred or best estimate 
can be challenging, disregards information, 
and introduces additional subjectivity to the 
meta-analysis.
Instead, we propose an adjustment to the 
cluster weights based on the inverse variance, 
or fixed effect size (FES) weights commonly 
used in the meta-analysis literature (Nelson 
and Kennedy 2009; Borenstein et al. 2010; 
Nelson 2015; Havránek, Horvath, and Zey­
nalov 2016; Vesco et al. 2020; Schütt 2021). 
Under the FES framework, each meta-obser­
vation is considered a draw from the same 
underlying population distribution, which 
makes sense in the context of multiple esti­
mates from the same cluster (i.e., housing 
market). Our proposed variance-adjusted 
cluster (VAC) weights give more weight to 
more precise estimates within a cluster while 
ensuring that equal influence is given to each 
cluster (or housing market) examined in the 
literature. For example, Walsh, Milon, and 
Scrogin (2011) provide six elasticity estimates 
at BIBL DE L UNIVERSITE LAVAL on October 20, 2026. Copyright 2021
Downloaded from 


===== PAGE 11 (score 28, numeric-paren 0, table-words 6, var-words 10) =====

98(2)
201
Guignet et al.: Property Values and Water Quality
were derived from models that did not use 
econometric methods to account for spatial 
dependence (i.e., spatial fixed effects, spatial 
lag of neighboring house prices, or accounting 
for spatial autocorrelation via a formal spatial 
autocorrelation coefficient or cluster-robust 
standard errors). Although the majority of pri­
mary studies use in situ Secchi disk measure­
ments (12 studies), we do see that 61% of the 
elasticity observations are based on hedonic 
regressions that used clarity measures other 
than in situ measures.14 About 22% of the ob­
14 This large proportion is primarily due to Ara (2007) 
and Walsh et al. (2017), who provide 30 and 112 elasticity 
served elasticities with respect to water clarity 
are from hedonic regressions that also con­
trol for other measures of water or ecological 
quality. A time trend variable, as reflected by 
the last year of transaction data in the primary 
study, is also included and ranges from 1994 
to 2014. This is converted into an index repre­
senting the number of years since 1994, which 
estimates for homes around Lake Erie and the Chesapeake 
Bay, respectively. Those studies, along with Guignet et al. 
(2017), use water clarity values based on spatial interpola­
tions. Two other studies use measurements predicted from 
water clarity models (Boyle and Taylor 2001; Liu et al. 
2014), and one study uses satellite data (Horsch and Lewis 
2009).
Table 2
Descriptive Statistics of Observations Pertaining to Water Clarity
Variable
Mean
Std. Dev.
Min.
Max.
Dependent Variable
Elasticity
0.1167
0.2543
−0.6478
1.7198
Study Area/Commodity Variable
Waterfronta
0.6808
0.4671
0
1
Mean clarity (Secchi disk depth, m)
2.34
1.97
0.38
6.45
Lake or reservoira
0.5615
0.4972
0
1
Estuarya
0.4385
0.4972
0
1
Waterbody sizeb (sq. km)
6.8846
5.3156
0.0002
20.8858
Median income (thousands, 2017$)
59.078
14.144
37.865
91.174
College degree (% population)
0.1366
0.0414
0.0768
0.2734
Population density (households/sq. km)
49.91
58.38
1.41
227.96
Mean house price (thousands, nominal $)
211.314
131.341
31.287
675.364
Northeasta
0.2885
0.4539
0
1
Midwesta
0.1923
0.3949
0
1
Southa
0.4846
0.5007
0
1
Westa
0.0346
0.1832
0
1
Methodological Variable
Elasticity variance
1,228.159
18,704.52
9.03E-06
301,448.5
Unpublisheda
0.1500
0.3578
0
1
Not in situ measurea
0.6115
0.4883
0
1
Other water quality variablesa
0.2192
0.4145
0
1
Assessed valuesa
0.0538
0.2261
0
1
Study time period (years)
10.27
3.82
3
24
Time trend (0=1994 to 20=2014)
8.59
6.17
0
20
No spatial methodsa
0.3808
0.4865
0
1
Double-loga
0.4308
0.4961
0
1
Linear-loga
0.3077
0.4624
0
1
Lineara
0.0385
0.1927
0
1
Log-lineara
0.2231
0.4171
0
1
Note: Unweighted descriptive statistics are presented for n = 260 elasticity estimates in the meta-dataset per­
taining to water clarity. Estimates are based on 18 primary hedonic studies, corresponding to 66 study-housing 
market clusters.
aIndependent variables that are dummy variables.
bSize of the waterbody (or waterbodies) examined in the primary studies was only available for n = 79 of the 
260 observations in the meta-dataset pertaining to water clarity. If multiple waterbodies were examined in the 
primary study hedonic regression models, the average waterbody size is reported.
at BIBL DE L UNIVERSITE LAVAL on October 20, 2026. Copyright 2021
Downloaded from 


===== PAGE 12 (score 10, numeric-paren 0, table-words 1, var-words 8) =====

Land Economics
202
May 2022
corresponds to the first study of water clarity 
in the meta-dataset.
We were able to identify and include three 
unpublished hedonic studies examining water 
clarity (15% of the observations), but publi­
cation bias is still a concern given our goal 
of obtaining an accurate estimate of the true 
underlying elasticities for purposes of benefit 
transfer. Following recommendations by Stan­
ley and Doucouliagos (2012), we examine 
a series of funnel plots and implement more 
formal funnel-asymmetry and precision-effect 
tests. In doing so, we find clear evidence that 
publication bias is a concern with the meta­
data collected from this literature (see Appen­
dix B.3 for details). In the meta-regression 
models discussed next, we include the inverse 
of the elasticity variances as a right-hand-side 
covariate to minimize such publication bias 
(Stanley and Doucouliagos 2012, 2014).
3. Meta-regression Methodology
Function transfers based on meta-regressions 
can be a useful approach for benefit transfer 
(Nelson 2015). The approach takes advantage 
of the full amount of information provided 
by the literature while accounting for key di­
mensions of heterogeneity in the outcome of 
interest. Consider the following reduced-form 
meta-regression model:
ε
β
=
+
+
+
0
1
2
ˆ
,
idj
idj
e
x
z
β
β
idj
idj
 
[6]
where the parameters to be estimated are β0
1
, β , 
and 
2
β . The right-side moderator variables 
include a vector of characteristics of the pri­
mary estimate, study area, and corresponding 
waterbody (xidj) and a vector of methodolog­
ical variables (zidj), which describe attributes 
of the primary study and model assumptions. 
The error term eidj is assumed to be normally 
distributed and is allowed to be correlated 
within clusters.
The vector xidj includes indicators of 
whether the elasticity estimate corresponds to 
waterfront homes, water quality in an estuary 
(as opposed to freshwater lakes),15 and the 
15 As described in Section 2, the meta-dataset includes 
price elasticities corresponding to freshwater lakes, 
mean baseline water clarity level correspond­
ing to the respective waterbody or portion of 
the waterbody. The vector xidj also includes 
characteristics of the study area and housing 
market, such as median income, proportion 
of the population with a college degree, mean 
house prices, and indicators denoting each of 
the four broad U.S. regions: the Northeast, 
Midwest, South, or West.
The vector zidj captures differences in elas­
ticities due to estimate quality and method­
ological choices made by the primary study 
authors. If particular values of zidj denote best 
practices, then such information can be ex­
ploited when predicting values for purposes of 
benefit transfer (Boyle and Wooldridge 2018). 
The vector zidj includes the variance of the 
corresponding elasticity estimate. Under the 
assumption that better-quality estimates have 
a lower variance, Stanley and Doucouliagos 
(2012, 2014) argue that this attribute should 
be set to zero in any subsequent benefit trans­
fer exercise.
In addition, zidj includes indicator vari­
ables denoting unpublished studies, whether 
a study used assessed values (as opposed to 
actual transaction prices), different functional 
forms, and when the model did not account 
for spatial dependence. If a primary study 
model did not include spatial fixed effects, a 
spatial lag of housing prices, nor account for 
spatial autocorrelation in some fashion, then 
the no spatial methods indicator is set to one, 
and is zero otherwise.
We also include a study year trend vari­
able to possibly reflect changes in empirical 
methods, data, tastes and preferences, and 
awareness of water quality over time (Rosen­
berger and Johnston 2009). Time trends in 
meta-analyses of stated preference studies are 
typically based on the year the primary study 
survey was conducted, which is different from 
the year of publication (e.g., Van Houtven, 
Powers, and Pattanayak 2007; Rosenberger 
and Johnston 2009; Johnston, Besedin, and 
Holland 2019). For a hedonic meta-analysis, 
the choice is not as clear because the observed 
revealed preference data in a primary study 
estuaries, rivers, and small rivers and streams. However, the 
hedonic studies in the meta-dataset that examine water clar­
ity focus solely on lakes and estuaries.
at BIBL DE L UNIVERSITE LAVAL on October 20, 2026. Copyright 2021
Downloaded from 


===== PAGE 14 (score 97, numeric-paren 70, table-words 6, var-words 4) =====

Land Economics
204
May 2022
in water clarity leads to a 0.04% increase in 
the price of nonwaterfront homes within 500 
m of the waterbody. As expected, the price 
elasticity with respect to clearer waters is sig­
nificantly higher among waterfront homes (by 
0.1457 percentage points). Together these es­
timates suggest that, on average, a 1% increase 
in water clarity leads to a 0.19% increase in 
waterfront home prices. The statistically sig­
nificant coefficient corresponding to elasticity 
variance suggests that publication bias is a 
concern, but setting this variable to zero when 
predicting values for function transfers con­
trols for such bias (Stanley and Doucouliagos 
2012).
Model 2 in Table 3 includes indicator 
variables denoting the four regions of the 
United States, with the Northeast being the 
omitted category. The negative region coef­
ficients suggest that housing price elasticities 
with respect to water clarity in the Midwest, 
South, and West tend to be less than those 
in the Northeast, but such differences in this 
model are only significant in the South. For 
example, a 1% increase in water clarity in the 
Northeast would lead to an average increase 
in value of 0.27% for waterfront homes and 
0.25% for nonwaterfront homes within 500 
m. In the South, the results suggest that a 
1% increase in clarity corresponds to much 
smaller but still significant 0.04% and 0.02% 
increases in price among waterfront and non­
waterfront homes. Considering the average 
house price of $211,314, this suggests that a 
Table 3
Variance-Adjusted Cluster-Weighted Least Squares Meta-regression Results
Variable
(1)
(2)
(3)
(4)
(5)
(6)
Waterfronta
0.1457***
0.0170
−0.0677
0.1010**
−0.1126
0.1251***
(0.046)
(0.023)
(0.052)
(0.039)
(0.086)
(0.042)
Midwesta
−0.1056
−0.1488*
−0.2059
−0.2561***
−0.3156***
(0.087)
(0.082)
(0.132)
(0.096)
(0.116)
Southa
−0.2248***
−0.2797***
−0.3209*
−0.2889***
−0.3945***
(0.076)
(0.084)
(0.171)
(0.108)
(0.131)
Westa
−0.1803
−0.1803
−0.1747
−0.3745**
−0.3772**
(0.120)
(0.121)
(0.147)
(0.180)
(0.174)
Estuarya
−0.0401**
−0.1223***
(0.016)
(0.039)
Waterfront × estuary
0.1274**
0.1721*
(0.055)
(0.088)
Mean clarity 
0.0375
0.0799**
(0.030)
(0.035)
Waterfront × mean clarity 
−0.0684**
−0.1061**
(0.029)
(0.050)
Elasticity variance
8.97E-07***
6.26E-07**
6.26E-07**
5.10E-07
6.02E-07**
5.02E-07
(1.52E-07)
(2.56E-07)
(2.57E-07)
(3.30E-07)
(2.56E-07)
(3.65E-07)
Time trend
0.0183***
0.0223***
(0.006)
(0.006)
Linear-loga
0.1973
0.1987
(0.161)
(0.159)
Lineara
0.2198**
0.1159
(0.086)
(0.083)
Log-lineara
−0.0020
−0.0059
(0.008)
(0.008)
Constant
0.0414*
0.2474***
0.3320***
0.3092*
0.1681
0.0493
(0.021)
(0.077)
(0.090)
(0.182)
(0.122)
(0.180)
Observations
260
260
260
260
260
260
Adjusted R-squared
0.080
0.148
0.148
0.156
0.170
0.176
Note: The dependent variable is the home price elasticity with respect to water clarity (Secchi disk depth). Clustered-robust standard errors are 
in parentheses and are clustered according to the K = 66 study-housing market combinations. Weighted least squares regressions are estimated 
using the “regress” routine in Stata 16 and defining analytical weights equal to the variance-adjusted cluster weights (see equation [3]).
aIndependent variables that are dummy variables.
* p < 0.1; ** p < 0.05; *** p < 0.01.
at BIBL DE L UNIVERSITE LAVAL on October 20, 2026. Copyright 2021
Downloaded from 


===== PAGE 15 (score 14, numeric-paren 0, table-words 2, var-words 9) =====

98(2)
205
Guignet et al.: Property Values and Water Quality
1% increase in water clarity (an average of 
2.34 cm or just under 1 in.) would increase 
the value of a waterfront and nonwaterfront 
home in the Northeast by $559 (p = 0.000) 
and $522 (p = 0.001), respectively. This same 
improvement for otherwise similar waterfront 
and nonwaterfront homes in the South would 
be only $84 (p = 0.061) and $48 (p = 0.041).
Models 3 and 4 assess potential hetero­
geneity in the housing price effects based 
on characteristics of the environmental com­
modity, in this case, the type of waterbody 
and whether the waters are already relatively 
clear. Model 3 includes an indicator denoting 
whether the elasticity estimates correspond to 
an estuary (as opposed to a freshwater lake or 
reservoir) as well as a corresponding interac­
tion term with the waterfront indicator. The 
negative and statistically significant coeffi­
cient on estuary suggests that an increase in 
water clarity has a smaller effect on the price 
of homes near an estuary, compared with an 
increase in lake water clarity. Such a finding 
seems reasonable given that surrounding res­
idents may not generally expect the water to 
be clear in estuaries because brackish waters 
are often naturally opaque. However, the op­
posite is found among waterfront homes, as 
suggested by the positive and statistically sig­
nificant 0.0873 (p = 0.049) sum of the estuary 
and waterfront × estuary coefficients.
To better illustrate the implications of 
model 3, consider a 1% increase in clarity in 
an estuary in the Northeast. This would lead 
to a 0.35% increase in the value of a bayfront 
home and a 0.29% increase in the value of a 
nonbayfront home that was still within 500 m 
of the estuary. In contrast, the same 1% in­
crease in water clarity would lead to a 0.26% 
increase among lakefront homes in the North­
east and a larger 0.33% increase among non­
lakefront properties that are within 500 m. Al­
though it may seem counterintuitive that the 
elasticity among lakefront properties is less, 
this does not necessarily violate the intuition 
that those closest to the resource should hold 
a higher value for an improvement. Baseline 
house values tend to be much larger among 
waterfront homes, so a smaller elasticity does 
not necessarily translate to a smaller marginal 
implicit price. For example, Walsh, Milon, 
and Scrogin (2011) report a mean sale price of 
$452,646 and $199,982 (2002$) among lake­
front and nonlakefront homes, respectively. 
Applying these mean values to the aforemen­
tioned elasticities suggests an implicit price 
for a 1% increase in clarity of $1,197 for lake­
front homes and $664 for nonlakefront homes.
Model 4 explores whether the house price 
elasticity with respect to clearer waters tends 
to be systematically different depending on 
baseline water clarity.18 The model includes 
the mean clarity level corresponding to each 
primary study estimate and an interaction 
term between mean clarity and the waterfront 
indicator. The main effect is statistically in­
significant, as is the sum of the main effect 
and the coefficient on the waterfront × mean 
clarity interaction term. This suggests that the 
price elasticity for nonwaterfront and water­
front homes, respectively, does not vary with 
baseline water clarity.
In subsequent models not reported here, 
we find little evidence of systematic hetero­
geneity with respect to the other study area 
and commodity variables reported in Table 2, 
including household median income, percent 
of population with a college degree, mean 
house prices, and waterbody size. Although 
mean house prices and waterbody charac­
teristics are taken from the primary studies, 
the lack of significant findings pertaining to 
household median income, education, and 
other census-derived sociodemographics may 
be partly attributed to the spatial coarseness 
of the county- or state-level variables and the 
resulting measurement error.
Models 5 and 6 build on the previous 
two models by controlling for methodolog­
ical characteristics of the primary studies. 
The time trend variable is added, as well as 
indicators denoting the functional form of 
the hedonic price function that was assumed 
by the primary study authors (double-log is 
the omitted category). The positive and sig­
nificant time trend suggests that, all else 
constant, the elasticity estimates have been 
increasing over time. The specification indi­
cators are largely insignificant, with the ex­
ception of linear in model 5. This suggests 
18 Because of collinearity concerns, we do not account for 
both waterbody type and baseline water clarity in the same 
model.
at BIBL DE L UNIVERSITE LAVAL on October 20, 2026. Copyright 2021
Downloaded from 


===== PAGE 17 (score 92, numeric-paren 64, table-words 6, var-words 5) =====

98(2)
207
Guignet et al.: Property Values and Water Quality
proportions across the remaining three speci­
fication indicators. More specifically, among 
the remaining 250 observations that used 
one of these specifications to estimate the 
price elasticity with respect to water clarity, 
we see that 45% were based on a double-log 
(the omitted category), 32% on a linear-log, 
and 23% on a log-linear specification. To 
infer an elasticity that is based on the most 
recent methods and data possible, the value 
for the time trend index is set to 20 (which 
(2007), where the meta-analyst identifies the set of all possi­
ble combinations of methodological variable values, assigns 
a probability to each, and then predicts the benefit transfer 
estimates for each methodological variable combination and 
takes the average across all combinations.
corresponds to 2014, the most recent year ob­
served in the metadata).
This illustrative exercise yields an “av­
erage” elasticity for waterfront homes of 
0.2698, suggesting that a 1% increase in 
Secchi disk depth (an increase of 2.34cm, on 
average) leads to an average increase in wa­
terfront home values of 0.27% (p = 0.000). 
A slightly smaller 0.2564 elasticity is esti­
mated for the “average” nonwaterfront home 
(p = 0.000). Based on the average home price 
from Table 2, these results translate to a mean 
implicit price of $570 (p = 0.000) and $542 
(p = 0.000) for waterfront and nonwaterfront 
homes, respectively. Overall, the literature 
yields plausible and statistically significant 
Table 4
RES Cluster-Adjusted Weighted Least Squares Meta-regression Results
Variable
(1)
(2)
(3)
(4)
(5)
(6)
Waterfronta
0.0828***
0.0374*
−0.0356
0.0715*
0.0080
0.0829**
(0.018)
(0.022)
(0.050)
(0.036)
(0.044)
(0.031)
Midwesta
−0.0204
−0.0462
−0.0475
−0.1565***
−0.1476***
(0.040)
(0.036)
(0.048)
(0.032)
(0.039)
Southa
−0.0833***
−0.1451***
−0.1096
−0.2600***
−0.2495***
(0.031)
(0.043)
(0.081)
(0.037)
(0.044)
Westa
−0.0607
−0.0607
−0.0595
−0.3200***
−0.4216***
(0.097)
(0.098)
(0.107)
(0.111)
(0.077)
Estuarya
−0.0181
−0.0534***
(0.020)
(0.020)
Waterfront × estuary
0.1019*
0.0582
(0.055)
(0.050)
Mean clarity 
0.0247
0.0601***
(0.037)
(0.022)
Waterfront × mean clarity 
−0.0332
−0.0317
(0.032)
(0.024)
Elasticity variance
2.22E-05
2.05E-05
2.01E-05
2.01E-05
1.91E-05
1.86E-05
(2.18E-05)
(2.04E-05)
(2.01E-05)
(2.01E-05)
(1.93E-05)
(1.89E-05)
Time trend
0.0121***
0.0158***
(0.002)
(0.002)
Linear-loga
−0.0371
−0.0953*
(0.040)
(0.049)
Lineara
0.0807
0.0493
(0.091)
(0.052)
Log-lineara
−0.0023
−0.0001
(0.005)
(0.005)
Constant
0.0257
0.1025***
0.1755***
0.1086
0.1577***
0.0034
(0.016)
(0.031)
(0.055)
(0.098)
(0.034)
(0.063)
Observations
260
260
260
260
260
260
Adjusted R-squared
0.101
0.146
0.159
0.148
0.201
0.222
Note: The dependent variable is the home price elasticity with respect to water clarity (Secchi disk depth). Clustered-robust standard errors are 
in parentheses and are clustered according to the K = 66 study-housing market combinations. Weighted least squares regressions are estimated 
using the “regress” routine in Stata 16 and defining the analytical weights equal to the random effect size (RES) cluster-adjusted weights (see 
Section 2).
aIndependent variables that are dummy variables.
* p < 0.1; ** p < 0.05; *** p < 0.01.
at BIBL DE L UNIVERSITE LAVAL on October 20, 2026. Copyright 2021
Downloaded from 


===== PAGE 18 (score 15, numeric-paren 0, table-words 6, var-words 3) =====

Land Economics
208
May 2022
estimates of how water clarity is capitalized 
in surrounding home values, even after empir­
ically controlling for key dimensions of het­
erogeneity, publication bias, and methodolog­
ical assumptions.
Best-Performing Model for Benefit Transfer
To examine out-of-sample transfer error and 
assess the performance of the various models 
and weighting schemes, we iteratively leave 
out observations corresponding to the 66 hous­
ing market study clusters, and then reestimate 
the mean unit values and meta-­regression mod­
els using the remaining sample. The predicted 
elasticities are then estimated for the excluded 
cluster. This is repeated by excluding the 66 
clusters one at a time. After completing all 66 
iterations, we calculate the median absolute 
transfer error. Similar out-of-sample transfer 
error exercises have been implemented in the 
literature (e.g., Lindhjem and Navrud 2008; 
Stapler and Johnston 2009).
We conduct this out-of-sample trans­
fer exercise in two ways. In the first ap­
proach, we construct a synthetic observa­
tion for each distance bin d in cluster j and 
then compare the elasticity value for this 
synthetic observation to the predicted elas­
ticity from the meta-­regression models. The 
synthetic observation is constructed us­
ing the same VAC weights (i.e., an inverse 
variance weighted mean across all elastic­
ity estimates for distance bin d in cluster j):
ε
ε
=
=




=






∑
∑
1
1
1
1
.
ˆ
ˆ 
idj
dj
dj
idj
v
k
s
idj
dj
k
i
v
i
The corresponding right-hand-side variables 
for the synthetic observations are calculated 
in the same fashion. Those variable values 
are then plugged into the estimated meta-
regressions to yield a predicted elasticity εˆ
s
dj, 
which is compared to the “actual” elasticity 
for each synthetic observation εˆs
dj. A similar 
exercise is done using the mean unit values, 
where the waterfront or nonwaterfront means 
are calculated each iteration, and then used as 
the unit value prediction εˆ
s
dj for the excluded 
observations. The transfer error is calculated 
as the absolute value of the percent difference:
ε
ε
ε
−




=
×




ˆ
ˆ
ˆ
%
100 .
s
s
dj
dj
s
dj
dj
TE
 
[7]
 
Our synthetic observation approach for 
measuring 
out-of-sample 
transfer 
error 
weights the “actual” observed elasticity es­
timates and the sample used to parameter­
ize the meta-regression models in the same 
way. When dealing with a panel- or cluster-­
structured meta-dataset, the more common 
practice of comparing predicted and observed 
elasticity estimates for all left-out observa­
tions within each iteration (e.g., Londoño and 
Johnston 2012; Fitzpatrick, Parmeter, and 
Agar 2017; Subroy et al. 2019) potentially 
inflates the transfer error. The parameterized 
meta-regressions, and hence the predicted 
elasticities εˆidj, would discount less precise 
estimates, but the excluded elasticity observa­
tions εˆidj that these are compared with in each 
iteration would all be treated equally when 
assessing the transfer error. This inconsistent 
weighting across the predicted and observed 
elasticities automatically puts the predictive 
performance of the meta-regression models at 
a disadvantage. Nonetheless, we carry out our 
transfer error exercise using this conventional 
approach and find similar results.
The median absolute transfer error results 
for each model and weighting scheme are 
presented in Table 5. The top panel shows 
the median transfer errors using our out-of-
sample synthetic observation approach. The 
lower panel shows the median transfer errors 
when all excluded observations are treated 
equally and used for comparison. The results 
suggest a median absolute transfer error of 
76%–119% under the synthetic observation 
comparison versus 83%–131% when compar­
ing all excluded observations.
Although errors of this size are not un­
heard of, the transfer errors for this study are 
in the high range. Kaul et al. (2013) examined 
1,071 transfer errors reported by 31 studies 
and report that the absolute value of the trans­
fer errors ranged from 0% to 7,496%, with a 
median of 39%. Rosenberger (2015) summa­
rized the results for 38 studies that statistically 
at BIBL DE L UNIVERSITE LAVAL on October 20, 2026. Copyright 2021
Downloaded from