Access to this full-text is provided by IOP Publishing.
Content available from Environmental Research Letters
This content is subject to copyright. Terms and conditions apply.
Environ. Res. Lett. 13 (2018) 054001 https://doi.org/10.1088/1748-9326/aabb32
LETTER
Climate gentrification: from theory to empiricism in
Miami-Dade County, Florida
Jesse M Keenan1,3, Thomas Hill1and Anurag Gumber2
1Harvard University, Graduate School of Design, 407 Gund Hall, 48 Quincy Street, Cambridge, MA, United States of America
2Harvard University, Kennedy School of, Government, Cambridge, MA, United States of America
3Author to whom any correspondence should be addressed.
OPEN ACCESS
RECEIVED
24 January 2018
REVISED
19 March 2018
ACCEPTED FOR PUBLICATION
3 April 2018
PUBLISHED
23 April 2018
Original content from
this work may be used
under the terms of t he
Creative Commons
Attribution 3. 0 licence.
Any further distribution
of this work must
maintain attrib ution to
the author(s) and the
title of the work, journal
citation and DOI.
E-mail: jkeenan@gsd.harvard.edu
Keywords: climate change, Climate Gentrification, economics, housing, resilience, adaptation, real estate
Supplementary material for this article is available online
Abstract
This article provides a conceptual model for the pathways by which climate change could operate to
impact geographies and property markets whose inferior or superior qualities for supporting the built
environment are subject to a descriptive theory known as ‘Climate Gentrification.’The article utilizes
Miami-Dade County, Florida (MDC) as a case study to explore the market mechanisms that speak to
the operations and processes inherent in the theory. This article tests the hypothesis that the rate of
price appreciation of single-family properties in MDC is positively related to and correlated with
incremental measures of higher elevation (the ‘Elevation Hypothesis’). As a reflection of an increase in
observed nuisance flooding and relative SLR, the second hypothesis is that the rates of price
appreciation in lowest the elevation cohorts have not kept up with the rates of appreciation of higher
elevation cohorts since approximately 2000 (the ‘Nuisance Hypothesis’). The findings support a
validation of both hypotheses and suggest the potential existence of consumer preferences that are
based, in part, on perceptions of flood risk and/or observations of flooding. These preferences and
perceptions are anticipated to be amplified by climate change in a manner that reinforces the
proposition that climate change impacts will affect the marketability and valuation of property with
varying degrees of environmental exposure and resilience functionality. Uncovering these empirical
relationships is a critical first step for understanding the occurrence and parameters of Climate
Gentrification.
Introduction
This article provides a conceptual model for the path-
ways by which climate change could operate to impact
geographies and property markets whose inferior or
superior qualities for supporting the built environment
are subject to a descriptive theory known as ‘Climate
Gentrification’(hereinafter, ‘CG’). To provide empir-
ical resolution to a theory of CG, this article utilizes
Miami-Dade County, Florida (‘MDC’) as a case study
to explore the potential existence of consumer prefer-
ences that are based, in part, on perceptions of flood
risk and/or observations of flooding. These prefer-
ences would be anticipated to be amplified by climate
change in a manner that reinforces the proposition
that climate change will affect the marketability and
valuation of property with varying degrees of exposure
and resilience functionality. It is speculated that com-
paratively high- and low-elevation properties in MDC
will be more or less valuable overtime by virtue of a
property’s capacity to support habitation in the face of
nuisance flooding and relative sea level rise (‘SLR’).
This article tests thehypothesis that the rate of pos-
itive price appreciation in MDC from 1971–2017 is
positively related to and correlated with incremental
measures of higher elevation of the underlying proper-
ties (the ‘Elevation Hypothesis’). As a reflection of an
increase in observed tidal nuisance flooding and SLR
since 2000 (Southeast Florida Regional Climate Change
Compact 2015), the second hypothesis is that the rates
of price appreciation in the lowest elevation cohorts
are below the rates of appreciation of higher elevation
© 2018 The Author(s). Published by IOP Publishing Ltd
Environ. Res. Lett. 13 (2018) 054001
cohorts since 2000 (the ‘Nuisance Hypothesis’). Both
hypotheses are evaluated across MDC, as well as within
various jurisdictions within MDC.5If validated, these
hypotheses would provide partial evidence that market
preferences reward and penalize properties with higher
and lower elevations, respectively. While a validation
of these hypotheses is by no means definitive in estab-
lishing a link between the perception of flood risk and
consumer preferences, the inference of such a relation-
ship would highlight one of multiple pathways b y which
CG may manifest to disrupt economically vulnerable
communities.
The relevance of a theory of CG is defined
by the need to promulgate a broader awareness of
the processes shaping socioeconomic vulnerabilities
and not just physical environmental exposure (F¨
ussel
2007,O’Neill et al 2014). Likewise, it highlights the
dynamic and dependent relationships of elements of
the built environment (e.g. housing, transportation,
public facilities) that may either exacerbate vulner-
abilities associated with climate change impacts or
are themselves exacerbated by such impacts (R¨
as¨
anen
et al 2016,Walkeret al 2016). As climate adapta-
tion planning internalizes the implications of resource
constraints (North and Longhurst 2013)andduepro-
cess (Sovacool and Linn´
er 2016) within the context
of distributive and procedural justice (Bulkeley et al
2013,Shiet al 2016), the onus of the public sector is
to contextualize existing institutional parameters that
define both the vulnerability and exposure of sensi-
tive populations (Anguelovski et al 2016,Chuet al
2017). In this case, understanding the institutional
and economic mechanisms of property markets are
arguably critical for long-term planning. Whether it
is land use or affordable housing planning, the com-
mon denominator is the relative availability and price
of property and real estate. If CG proves to be an accu-
rate description of economic processes and behaviors,
high-elevation property, shaded or wind-cooled prop-
erty, fresh water resourced property, geologically stable
property, ecologically diverse property, pollution-free
property, and property with resiliently design build-
ings will all provide attributes of market valuation
that complicate the existing capacities of society to
house and shelter its most vulnerable populations.
Climate gentrification
While CG has been popularized in the media (Flavelle
2016,Bolstad2017), there has been limited scholar-
ship defining the parameters of this emerging theory
(Keenan and Weisz 2017). CG is based on a simple
proposition: climate change impacts arguably make
some property more or less valuable by virtue of its
5In the US, not all portions of a county are part of a municipal
jurisdiction. As such, unin corporated portions of a count y are entirely
governed and serviced by the county.
capacity to accommodate a certain density of human
settlement and its associated infrastructure. The impli-
cation is that the price volatility associated with rent
seeking, speculative investment, or superior purchas-
ing power is either a primary or a partial driver of
the patterns of urban development that lead to dis-
placement (and sometimes entrenchment) of existing
populations consistent with conventional framings of
gentrification (Slater 2006, Lees et al 2013). While geo-
graphic exposure of property is a primary locational
and environmental attribute of CG, the relative degree
of engineered resilience within buildings and infras-
tructure systems on such property may also provide
a secondary axis of analysis that may explain why
two equally exposed property markets of similar con-
structed attributes may perform differently over the
long-term in the face of climate change (Hollnagel
2014).
CG may arguably manifest in one of several path-
ways, as represented in figure 1.Thefirstpathwayis
what primarily frames this article. It relates to the sub-
stitution of property from an inferior to a superior
location. This may also be viewed as a selection of
properties with superior locational and environmen-
tal attributes among alternative investment options
with inferior qualities. For purposes of representa-
tional simplification in figure 1, it is assumed that
there are only two local options for settlement and
investment, and it is assumed that there are two pop-
ulation wealth cohorts—high-income (i.e. rich) and
lower-to-moderate income (i.e. not rich). Superiority
of one option to another is adjudicated by a property’s
comparatively lower level of physical environmental
exposure or its high level of constructed attributes
for engineered resilience and/or hazard mitigation. In
this article, superiority is informally hypothesized to be
high-elevation geographies (e.g. Little Haiti in Miami)
who are less vulnerable to flooding, in part, because
of a known reliance on gravitational flows to man-
age water in MDC. More fundamental to the theory,
it describes a behavior of moving financial capital to
a geography that offers superior risk-adjusted returns
for accommodating real estate and infrastructure. It
may also offer superior attributes for accommodat-
ing communities and not just assets. This pathway
comes with the caveat that some households may
otherwise be trapped for a lack of resources to relo-
cate (de Sherbinin et al 2011,Blacket al 2013)or
because of outstanding mortgage liabilities (Bricker and
Bucks 2016). This pathway is collectively referenced as
the ‘Superior Investment Pathway.’
As represented in figure 1, the Superior Investment
Pathway is shown within the context of two options. In
reality, there may be many local and non-local options.
It is conceptualized that households—particularly
low-to-moderate income households—would grad-
ually move from the coastal barrier islands (e.g.
Miami Beach) to the mainland of MDC where ele-
vations are significantly higher. However, as economic
2
Environ. Res. Lett. 13 (2018) 054001
Figure 1. Pathways to Climate Gentrification.
productivity and jobs may be undermined by SLR, pop-
ulations may leave MDC altogether (Hauer et al 2016).
As such, CG may operate across multiple scales (i.e.
neighborhoods, municipalities, states, regions, coun-
tries). For instance, SLR impacts in MDC may lead to
CG in central Florida, which is much less physically
exposed. Likewise, Atlanta may be subject to CG stem-
ming from coastal SLR on the east coast of the US
because of its superior labor and housing opportuni-
ties (Hauer 2017). While these networks, transfers and
transitions are difficult to model, emerging research in
demographics suggests that CG may operate at mul-
tiple scales beyond those simplified representations in
figure 1(Curtis and Schneider 2011,Neumannet al
2015). Interview data suggests that speculative prop-
erty investors are already hedging on south Florida’s
gradual exodus to central and north Florida.
The second pathway for CG relates to the dete-
rioration of environmental conditions such that the
overall cost of living can only be feasibly borne by
wealthier and wealthier households, as climate change
impacts manifest in greater frequency and intensity.
Gentrification happens inversely by the fact that vul-
nerable populations are unable to afford to live in situ.
This would be primarily due to the increased costs of
insurance, property taxes, special assessments, prop-
erty repairs, transportation and consumer goods, as
well as a loss in overall productivity (e.g. sitting in traf-
fic in water-clogged streets). For comparatively wealthy
households, prior research has suggested that the ‘risk
of coastal flooding seems inconsequential in determin-
ing property values due to the substantial premiums
that appear to be associated with proximity to coastal
water’(Bin and Kruse 2006, p 137). For those house-
holds who are more sensitive to the carrying costs
associated with such hazards, there may be no alter-
native but to relocate. Those that remain are those who
are either trapped or have invested speculative capital
that they can ‘afford’to lose. An example of this is in
Venice, Italy where environmental conditions, includ-
ing relative SLR and unabated tourism, have resulted
in a total cost-burden that has undermined class diver-
sification (Moretti 2012). This pathway is collectively
referenced as the ‘Cost-Burden Pathway.’
It would be anticipated that over time such a phe-
nomenon would occur on the barrier islands of MDC,
such as Miami Beach. However, research models sug-
gest that adaptation investments in risk mitigation
likely have a threshold by which even informed (and
comparatively wealthy property owners) will eventu-
ally abandon their investments (McNamara and Keeler
2013,Treueret al 2018). As such, it should be qualified
that the pathways to CG are limited in their duration
and intensity, as threshold dynamics are highly unpre-
dictable (Haer et al 2017). Eventually, in the face of
SLR, it can be argued that even the most-wealthy will
likely have to abandon Venice and Miami Beach.
The third pathway relates to the unintended
consequences of making public investments in the
engineered resilience of buildings and infrastructure
(Ayyub 2014,Cer
`
eet al 2017). As a consequence of
these investments, the underlying property increases
in value by virtue of the fact that the positive exter-
nalities associated with performance of the resilience
investments represents a superior outcome to the
status quo—even when netted-out by any costs
associated with the taxes for building and maintain-
ing the resilience infrastructure (Bunten and Kahn
2017). Therefore, any tax consequences associated with
the investments would be absorbed by increases in
3
Environ. Res. Lett. 13 (2018) 054001
property valuation and/or rent payers. This pathway
is a derivative of the well-developed concept of ‘Green
Gentrification,’wherein investments in sustainability
amenities and infrastructure are unevenly distributed
or otherwise associated with gentrification (Checker
2011, Curran and Hamilton 2012,Bryson2013,Sand-
berg 2014, Curran and Hamilton 2017,Gouldand
Lewis 2017,Anguelovskiet al 2018). Although not
widely studied, the exemplar case for this pathway is
the St. Kjeld Climate District in Copenhagen where
a broader resilience strategy to revitalize a neigh-
borhood led to some displacement from increased
rents (Kjaer 2015) and the marginalization of existing
homeowners (Baron and Petersen 2016). This pathway
is referenced as the ‘Resilience Investment Pathway.’
However, there is an alternative hypothetical sce-
nario wherein resilience investments operate to reduce
risk and exposure to such an extent that it reduces
long-term tax and insurance liability. In Copenhagen,
the resilience investment brought the neighborhood
real estate up to ‘market rate.’However, in this
alternative-scenario, the market value becomes more
competitive among alternative substitutes because of
the comparatively lower carrying costs (e.g. taxes
and insurance).
Each of the three pathways represent possi-
ble behaviors that may lead to CG. They do not
independently represent deterministic conditions, as
exogeneity in property markets often defy current
methodologies for pinpointing long-term valuation
trends or preferences. CG is referenced as a descriptive
theory for understanding emerging trends otherwise
referenced as conventional gentrification. Climatic
impacts should be understood within a broader
array of influences driving gentrification, including
historic racial segregation, income inequality, and
the spatial distribution of jobs, transportation, and
housing. However, with CG, it can be argued that
climatic influences will increasingly play an impor-
tant role in the weighted factors driving investment
and locational decisions of households, investors, and
financiers. The empirical portion of this article seeks
to identify potential methodologies and measurements
that may validate the underlying behaviors inherent
in the Superior Investment Pathway.
Research design and methodology
The research design of this article is based on a mixed-
methods approach undertaken in two distinct phases:
(i) theory development and (ii) empirical data anal-
ysis and hypothesis testing (Creswell 2013). In the
first phase, exploratory research was undertaken in
MDC as it relates to vulnerability assessment and the
identification of existing resilience activities and capac-
ities. MDC was selected as a case study based on its
popular and scientifically determined vulnerability to
climate change impacts, including increased nuisance
flooding and SLR inundation (Yin 2013). As part of the
theory development phase, semi-structured interviews
were conducted with numerous (n= 48) local officials,
researchers, real estate developers, investors, financiers,
residents and activists. Interviews suggested a con-
sensus that high-elevation property would increase in
value over the long-term with SLR and that prefer-
ences relating to flood risk (climate change related or
not) were increasingly being recognized among con-
sumers and real estate actors. Interviews confirmed
that speculative investment in certain high-elevation
communities is well underway. The empirical aspects
of this article seek to identify whether a validation
of the hypotheses could partially explain behaviors
consistent with a Superior Investment Pathway.
Detailed property sales information was obtained
from the Miami-Dade County Property Appraiser’s
Office. The dataset contained approximately 800 000
properties and included records for property type, unit
count, lot and building size, property and building
values, year-built, bed and bath counts, market and
property tax assessment values, exemptions, owner
name, address, zoning, and the last three transactions
(the ‘Property Dataset’). Property records with incom-
plete or misregistered values were culled. In order
to understand how patterns might be conditioned or
contextualized by elevation, the analysis involved com-
bining the Property Dataset with elevation data (1/9th
arc-second) for Miami-Dade County sourced by the
United States Geological Survey (‘Elevation Dataset’)
(USGS 2017).
The economic analysis comprised of two prin-
ciple steps. First, a price index was constructed to
allow a comparison of price appreciations of properties
across the entire Property Dataset. This normaliza-
tion of price appreciation allowed for a more resolute
apples-to-apples comparison of price appreciation
by and between different property characteristics.
Second, a linear mixed effects model was con-
structed and coded to understand how the relationship
between elevation and price appreciation varied across
jurisdictions—holding various other explanatory vari-
ables constant (i.e. square footage, sale date, and
construction year). Both the price indexing and the
regression analysis were conducted in parallel using
the programming languages R and Python.
Empirical modeling and findings
From the cleaned Property Dataset, properties contain-
ing single-family homes were isolated. The resulting
Property Dataset was reduced to 107 984 properties.
Single-family homes were selected to the exclusion
of condominium and cooperative properties because
these properties are arguably less sensitive to the nui-
sance and risk of loss from intermediate flooding
because of their varied base floor elevations and insur-
ance structures. Second, condominium development
4
Environ. Res. Lett. 13 (2018) 054001
Figure 2. (a) Range of elevations for municipalities and unincorporated portions of Miami-Dade County. (b) Map of elevations for
municipalities and unincorporated portions of Miami-Dade County.
patterns were spatially concentrated and did not offer
much insight for patterns across time and elevation.
Commercial real estate was also removed because val-
uations are largely dependent on net operating income
and investment cycles (Geltner 2015).
Figure 2represents the range of elevations found in
each of the selected municipalities and the unincorpo-
rated portions of MDC. Not all municipalities in MDC
were selected for analysis because certain municipali-
ties did not have either a meaningful internal variation
in elevation or a robust level of data. With the revised
Property Dataset and the Elevation Dataset, two com-
putational strategies were deployed. As is discussed in
the Supplemental Methodology, the first was to con-
struct a multiplicative price index (Bailey et al 1963,
Hill 2013) and the second was to conduct a linear
mixed effects regression on modified samples within
the subject datasets (Peng and Lu 2012, Reddy 2015).
Rate of appreciation and elevation findings
Figure 3represents a range of jurisdictions wherein
the indexed valuation multiple was decomposed for
elevation cohorts measured in 1 meter increments.
Measurement anomalies below sea level (<0 meter)
were spot-checked and either culled or grouped
into the lowest elevation cohort. The values on the
y-axis are multiples indexed to 1971. The total sam-
ple size of properties broken down by elevation
cohort is found in supplemental table 1 available at
stacks.iop.org/ERL/13/054001/mmedia. For all subject
properties, figure 3(a) demonstrates that properties in
the 2–3 meter and 3–4 meter cohorts have had slightly
higher rates of price appreciation relative to the 1–2
and 0–1 meter cohorts. This finding would be con-
sistent with a validation of the Elevation Hypothesis.
While properties in the 4–5 meter and 5–6 meter ele-
vation cohorts have lagged the group, this finding is
less relevant or impactful because these properties rep-
resent just 1.41% (n= 1518) of the entire sample. This
marginal distribution holds true across all of the eval-
uated jurisdictions. As such, elevation cohorts above 4
meters can generally be ignored.
Figure 3(b) highlights a similar pattern for unin-
corporated parts of MDC, which accounts for 58%
(n= 58 804) of the sample. Unincorporated portions
of MDC suggest a slightly stronger relationship to ele-
vation than the entire sample represented in figure
3(a). As a general observation, the 3–4 meter cohort
has slightly outperformed the 2–3 meter cohort. The
2–3 meter cohort has slightly outperformed the 1–2
meter cohort and the 1–2 meter cohort has outper-
formed the 0–1 meter cohort. This spread has been
particularly pronounced since approximately 2000.
Specific to the City of Miami, figure 3(c)represents
a similar but less conclusive pattern to those found
of figures 3(a)and(b). While the 0–1 meter cohort
has lagged the group for most of the time period,
the relationships between cohorts are less clear than
unincorporated MDC. In particular, there has been a
recent increase in rates of appreciate in the 0–1 meter
5
Environ. Res. Lett. 13 (2018) 054001
Figure 3. Indexed valuation multiple by elevation cohort.
cohort. This might be explained by properties bene-
fiting from their proximate location to a recent boom
in luxury coastal high-rise properties. Overall, the City
of Miami accounts for just 6.70% (n= 7234) of the
sample.
Consistent with a validation of the Nuisance
Hypothesis, the 0–1 meter cohort has significantly
lagged the group since approximately 2000 for all prop-
erties in MDC in figure 3(a). A similar pattern is found
among unincorporated properties in figure 3(b), with
a precipitous drop in price appreciation in approx-
imately 2015 for the 0–1 meter cohort. In addition,
7 of the 12 jurisdictions represented in supplemental
figure 1 all demonstrated a similar pattern wherein the
lowest elevation cohorts (i.e. either 0–1 or 1–2 elevation
cohorts) tracked the general group until approximately
2000, at which point they begin to underperform rela-
tive to the general track of the elevation cohorts.
As represented in figure 3(d), properties in the
City of Miami Beach have expressed a notably negative
relationship between elevation and price appreciation.
This is likely explained by the proposition that spatial
proximity to the water has a positive impact on bot h val-
uation and rate of appreciation, at least as long as those
6
Environ. Res. Lett. 13 (2018) 054001
(a)
(b)
Figure 4. (a) Random effect regression coefficients for elevation effect on price appreciation by jurisdiction. (b) Regression results for
elevation effects on price appreciation by jurisdiction.
bodies of water are deemed to be amenities (McNamara
et al 2015). Supplemental figures 1 and 2 contains sets
of figures for those municipalities that demonstrated
varying degrees of positive and negative relationships
between price appreciation and elevation. Overall,
11 jurisdictions accounting for 76% (n= 82 068) of
the overall sample demonstrated some measure of
positive relationships. By contrast, 5 jurisdictions
accounting for 13% (n=14 014) of the sample were
founded to have some negative relationship. While 17
jurisdictions had either inadequate elevation granu-
larity or inconclusive relationships, these jurisdictions
accounted for 11% (n= 11 798) of the sample.
Regression findings
Utilizing a linear mixed effects model, the price appre-
ciation index was regressed for elevation, construction
year, date of sale, and square footage, within each
of the jurisdictions and the unincorporated portions
of MDC with each variable representing a differ-
ent group. Thereafter, the method sought to obtain
the specific effect of elevation on price appreciation
for each of these jurisdictions, excluding municipal-
ities (n= 6) with less than 200 single-family units
(n=−672). Elevation was found to have a positive
effect on price appreciation in 24 of the 25 jurisdictions
under study. Those 24 jurisdictions represent 98.1%
of the 107 312 properties subject to the regression.
Only North Miami Beach exhibited a negative relation-
ship between elevation and price appreciation, albeit a
weak one.
Figure 4highlights the regression results and the
range of elevation regression coefficients for the subject
jurisdictions. As figure 5represents, the 3 jurisdictions
7
Environ. Res. Lett. 13 (2018) 054001
Figure 5. Map of random effect regression coefficients for elevation on price appreciation by jurisdiction.
with the strongest coefficients are all on the coast.
Overall, 13 (54%) of the 24 jurisdictions with posi-
tive coefficients are land-locked, although nearly all of
the land-locked jurisdictions have significant collec-
tions of lakes and drainage canals. The largest single
jurisdiction represented in the sample, unincorporated
MDC (n= 58 804; 54%), showed a positive corre-
lation. Overall, the sample of all subject properties
showed a positive correlation between elevation and
appreciation when controlling for the aforementioned
variables.
Discussion
It is difficult to identify the effect of elevation on price
appreciation independent of other variables and loca-
tional factors. There are many spatial qualities that
cannot be easily controlled for. The historical devel-
opment patterns of MDC are complex, and uneven
patterns at different elevations runs in contradiction
to many American cities where the historical patterns
of development dictated concentrations of wealth on
high elevations. Since elevation was the only locational
8
Environ. Res. Lett. 13 (2018) 054001
factor, it is possible that the results simply demonstrate
a correlation between location and price apprecia-
tion. However, the jurisdictions that exhibit a positive
relationship between elevation and price appreciation
represent the vast majority of all housing units in MDC.
This overall positive correlative effect provides evi-
dence in support of validating the Elevation Hypothesis.
This evidence is in addition to the observations of
positive relationships between price appreciation and
elevation cohorts in jurisdictions accounting for 76%
(n= 82 068) of the sample population. However, infer-
ential connections between the results of the two mode s
of analysis is inconclusive for some jurisdictions. In the
case of Miami Beach, there was an observed negative
relationship between price appreciation and elevation
cohorts, yet the city had the second highest regres-
sion coefficient. This could be explained by the two
different analytical methods, wherein elevation breaks
in the regression were more precise than the coarse 1
meter cohorts. However, more precise elevation mea-
surements may be inconsequential in the real world
wherein the path of water may not be obstructed by
such nuances in elevation. Future research will need
to find resolution between observations and mean-
ingful breaks and location of elevation. That is to say
that not all elevation represents equal units of risk or
nuisance given the underlying bathymetry and surface
water management capacities of MDC.
There is robust evidence supporting a validation
of the Nuisance Hypothesis. The logic behind the for-
mulation of the Nuisance Hypothesis was based on
the proposition that increased nuisance flooding may
have been negatively impacting low elevation proper-
ties in the market. Interviews with real estate brokers
suggested a certain intelligence about high nuisance
portions of MDC among the brokers. While the find-
ings support the hypothesis, they do not necessarily
speak to a validation of the causal logic. However,
in some areas, lower elevation properties are grossly
underperforming relative to other elevation cohorts.
Likewise, this trend appears to have accelerated in and
around 2000. While measurements of SLR on the East
Coast of the US were observed to accelerate in the
1990s (Miami-Dade Sea Level Rise Task Force 2014,
Davis and Vinogradova 2017), observed incidents of
increased flooding in MDC appear to have accelerated
just after 2000 (Wdowinski et al 2016). This pattern
of acceleration was observed not just in a majority of
the sample, as represented in figures 3(a)(AllProp-
erties), figure 3(b)(unincorporatedMDC),and,toa
lesser extent, in figure 3(c) (Miami), but also in areas
such as El Portal, Miami Shores, and North Miami
Beach, which are subject to ongoing tidal flooding and
King Tides (see supplemental figures 1(b), (d)and(e),
respectively).
The evidence supports a validation of the Eleva-
tion Hypothesis with the broader inference that higher
elevation properties may have a slight advantage in
terms of higher rates of price appreciation that may
be increasing with time. By contrast, the evidence
supporting a validation of the Nuisance Hypothesis
suggests that the lowest elevation properties may be at a
price disadvantage. In relating these find ings to a theory
of CG, the Elevation Hypothesis provides support for
the long-term occurrence of the Superior Investment
Pathway. Over time, it could be argued that higher ele-
vation properties in MDC will become more attractive
because of their superior rates of appreciation.
This may also be viewed within the context of
the Nuisance Hypothesis wherein the lowest elevation
properties are not appreciating at the same rate and
therefore are inferior investments—assuming that rate
of appreciation is a significant factor for investment.
The heuristics of real estate investment suggest that
this rational maximization through long-term appre-
ciation does not always hold (Salzman and Zwinkels
2017). If investors/owners see a relative disadvantage
or opportunity cost to their lower elevation properties,
then this may be one of many other factors that lead to
spatial relocation or the disposition of a particular asset.
Arguably, this may reinforce a Cost-Burden Pathway
if lower-to-moderate income households have more at
stake in terms of their overall net-wealth. The cost bur-
den may be increased by virtue of a cycle of declining tax
rolls and fewer and fewer tax payers. In all cases, this
article provides support for the proposition that cli-
mate change impacts could exacerbate environmental
and locational effects and qualities in property that may
already be reflected to a certain extent in the housing
market.
Uncovering these effects and qualities is a critical
first step for monitoring the incremental occurrence of
CG. What can the public sector do to mitigate the
negative consequences? Land use regulators will be
tasked with evaluating the consequences of relocation
and densification, particularly on higher-elevations
(e.g. inclusionary zoning). As previously theorized,
resilience investments will also have socioeconomic
consequences that should be accounted for. The chal-
lenge for the public sector is to build a sensitivity to the
economic effects of climate change and climate change
adaptation on property markets within existing policy
regimes.
Conclusions
Whether it is through a superior investment among
substitutes; a function of being driven-out through
increased consumer cost-burdens; or, a matter of public
resilience investments that drive up the value of prop-
erty, a theory of CG gives recognition to the various
pathways by which climate change impacts may drive
investment and settlement patterns. In MDC, CG has
been speculated in popular discourse to already explain
gentrification patterns. This article has demonstrated
that the elevation of one’s home in MDC could matter
in terms of long-term price appreciation. The findings
9
Environ. Res. Lett. 13 (2018) 054001
would suggest that a consumer preference may exist
in favor of higher elevation properties. Likewise, lower
elevation properties may be subject to lower rates of
appreciation due to flooding concerns. In light of accel-
erated SLR, these preferences may become more robust
and may lead to more widespread relocations that serve
to gentrify higher elevation communities.
Future research will be tasked with understanding
preferences and heuristics among relevant households
and investors. In particular, there is a need to under-
stand threshold dynamics that shape investment and
relocation decision-making. As such, a diagnostic
understanding of CG provides another step in a long
journey of adaptation that seeks to refine our under-
standing of vulnerability in the name of protecting our
most vulnerable populations from long-term maladap-
tation in human settlements.
ORCID iDs
Jesse M Keenan https://orcid.org/0000-0003-4058-
1682
References
Anguelovski I, Connolly J J, Masip L and Pearsall H 2018 Assessing
green gentrification in historically disenfranchised
neighborhoods: a longitudinal and spatial analysis of
Barcelona Urban Geogr. 39 458–91
Anguelovski I, Shi L, Chu E, Gallagher D, Goh K, Lamb Z and
Teicher H 2016 Equity impacts of urban land use planning for
climate adaptation: critical perspectives from the global North
and South J. Plan. Educ. Res. 36 333–48
Ayyub B M 2014 Systems resilience for multihazard environments:
definition, metrics, and valuation for decision making Risk
Anal. 34 340–55
Bailey M J, Muth R F and Nourse H O 1963 A regression method
for real estate price index construction J. Am. Stat. Assoc. 58
933–42
Baron N and Petersen L K 2016 Understanding controversies in
urban climate change adaptation. A case study of the role of
homeowners in the process of climate change adaptation in
Copenhagen Nord.J.Sci.Technol.Stud.34–13
Bin O and Kruse J B 2006 Real estate market response to coastal
flood hazards Nat. Hazard. Rev. 7137–44
Black R, Arnell N W, Adger W N, Thomas D and Geddes A 2013
Migration, immobility and displacement outcomes following
extreme events Environ. Sci. Policy 27 S32–43
Bolstad E 2017 Higher ground is becoming hot property as sea level
rises (www.scientificamerican.com/article/high-ground-
is-becoming-hot-property-as-sea-level-rises/) (Accessed: 9
April 2018)
Bricker J and Bucks B 2016 Negative home equity, economic
insecurity, and household mobility over the great recession
J. Urban Econ. 91 1–12
Bryson J 2013 The n ature of gentrification Geogr. Compass 7578–87
Bulkeley H, Carmin J, Cast´
an Broto V, Edwards G A S and Fuller S
2013 Climate justice and global cities: mapping the emerging
discourses Glob. Environ. Change 23 914–25
Bunten D and Kahn M E 2017 Optimal real estate capital durability
and localized climate change disaster risk J. Hous. Econ. 36 1–7
Cer`
e G, Rezgui Y and Zhao W 2017 Critical review of existing built
environment resilience frameworks: directions for future
research Int. J. Disaster Risk Red. 25 173–89
Checker M 2011 Wiped out by the ‘greenwave’: environmental
gentrification and the paradoxical politics of urban
sustainability City Soc. 23 210–29
Chu E, Anguelovski I and Roberts D 2017 Climate adaptation as
strategic urbanism: assessing opportunities and uncertainties
for equity and inclusive development in cities Cities 60
378–87
Creswell J W 2013 Research Design: Qualitative, Quantitative, and
Mixed Methods Approaches (Thousand Oaks, CA: Sage
Publications)
Curran W and Hamilton T 2012 Just green enough: contesting
environmental gentrification in Greenpoint, Brooklyn Local
Environ. 17 1027–42
Curran W and Hamilton T 2017 Just Green Enough: Urban
Development and Environmental Gentrification (New York:
Routledge)
Curtis K J and Schneider A 2011 Understanding the demographic
implications of climate change: estimates of localized
population predictions under future scenarios of sea-level rise
Popul. Environ. 33 28–54
Davis J L and Vinogradova N T 2017 Causes of accelerating sea
level on the East Coast of North America Geophys. Res. Lett.
44 5133–41
Flavelle C 2016 Climate change is already forcing Americans to
move (www.bloomberg.com/view/articles/2016-10-31/
climate-change-is-already-forcing-americans-to-move)
(Accessed: 9 April 2018)
F¨
ussel H M 2007 Vulnerability: a generally applicable conceptual
framework for climate change research Glob. Environ. Change
17 155–67
Geltner D 2015 Real estate price indices and price dynamics: an
overview from an investments perspective Annu.Rev.Financ.
Econ. 7615–33
Gould K A and Lewis T L 2017 Green Gentrification: Urban
Sustainability and the Struggle for Environmental Justice (New
York: Routledge)
Haer T, Botzen W J, Moel H and Aerts J C 2017 Integrating
household risk mitigation behavior in flood risk analysis: an
agent-based model approach Risk Anal. 37 1977–92
Hauer M E 2017 Migration induced by sea-level rise could reshape
the US population landscape Nat. Clim. Change 7321–5
Hauer M E, Evans J M and Mishra D R 2016 Millions projected to
be at risk from sea-level rise in the continental United States
Nat. Clim. Change 6691–5
Hill R J 2013 Hedonic price indexes for residential housing: a
survey, evaluation and taxonomy J. Econ. Surv. 27 879–914
Hollnagel E 2014 Resilience engineering and the built environment
Build. Res. Inf. 42 221–8
Keenan J M and Weisz C 2017 Blues Dunes: Climate Change by
Design (New York: Columbia University Press)
Kjaer H L 2015 Climate Change Adaptation in Marginalized
Neighborhoods Masters Thesis (Joint European Master in
Environmental Studies: Cities and Sustainability) (http://
projekter.aau.dk/projekter/files/213985464/artikel.pdf)
Lees L, Slater T and Wyly E 2013 Gentrification (New York:
Routledge)
McNamara D E and Keeler A 2013 A coupled physical and
economic model of the response of coastal real estate to
climate risk Nat. Clim. Change 3559–62
McNamara D E, Gopalakrishnan S, Smith M D and Murray A B
2015 Climate adaptation and policy-induced inflation of
coastal property value PLoS ONE 10 e0121278
Miami-Dade Sea Level Rise Task Force 2014 Report and
Recommendations (http://egov.ci.miami.fl.us/
Legistarweb/Attachments/77672.pdf)
Moretti M 2012 Venice, Italy: balancing antiquity and
sustainability Green Cities of Europe ed T Beatley
(Washington, DC: Island Press)
Neumann B, Vafeidis A T, Zimmermann J and Nicholls R J 2015
Future coastal population growth and exposure to sea-level
rise and coastal flooding-a global assessment PLoS ONE 10
e0118571
10
Environ. Res. Lett. 13 (2018) 054001
North P and Longhurst N 2013 Grassroots localisation? The scalar
potential of and limits of the ‘transition’approach to climate
change and resource constraint Urban Stud. 50 1423–38
O’Neill B C, Kriegler E, Riahi K, Ebi K L, Hallegatte S, Carter T R
and van Vuuren D P 2014 A new scenario framework for
climate change research: the concept of shared socioeconomic
pathways Clim. Change 122 387–400
Peng H and Lu Y 2012 Model selection in linear mixed effect
models J. Multivariate. Anal. 109 109–29
R¨
as¨
anen A, Juhola S, Nygren A, K¨
ak¨
onen M, Kallio M, Monge A M
and Kanninen M 2016 Climate change, multiple stressors and
human vulnerability: a systematic review Reg. Environ. Change
16 2291–302
Reddy S 2015 Residential property value estimation via linear
mixed model methods J. Prop. Tax Assess. Admin. 12 73–93
Salzman D and Zwinkels R C 2017 Behavioral real estate J. Real
Estate Lit. 25 77–106
Sandberg L A 2014 Environmental gentrification in a
post-industrial landscape: the case of the Limhamn quarry,
Malm¨
o, Sweden Local Environ. 19 1068–85
de Sherbinin A et al 2011 Preparing for resettlement associated with
climate change Science 334 456–7
Shi L, Chu E, Anguelovski I, Aylett A, Debats J, Goh K, VanDeveer
S D and Roberts J T 2016 Roadmap towards justice in urban
climate adaptation research Nat. Clim. Change 6131–7
Slater T 2006 The eviction of critical perspectives from
gentrification research Int. J. Urban Reg. Res. 30
737–57
Southeast Florida Regional Compact 2015 Unified sea level rise
projection for Southeast Florida Report (Fort Lauderdale, FL:
SFRC) (www.southeastfloridaclimatecompact.org/
wp-content/uploads/2015/10/2015-Compact-Unified-
Sea-Level-Rise-Projection.pdf)
Sovacool B and Linn´
er B 2016 The Political Economy of Climate
Change Adaptation (New York: Palgrave Macmillan)
Treuer G, Broad K and Meyer R 2018 Using simulations to forecast
homeowner response to sea level rise in South Florida: will
they stay or will they go? Glob. Environ. Change 48 108–18
United States Geological Survey (USGS) 2017 National Elevation
Data Set (https://nationalmap.gov/elevation.html)
Walker G R, Mason M S, Crompton R P and Musulin R T 2016
Application of insurance modelling tools to climate change
adaptation decision-making relating to the built environment
Struct. Infrastruct. Eng. 12 450–62
Wdowinski S, Bray R, Kirtman B P and Wu Z 2016 Increasing
flooding hazard in coastal communities due to rising sea level:
case study of Miami Beach, Florida Ocean Coast. Manage. 126
1–8
Yin R K 2013 Case Study Research: Design and Methods (Thousand
Oaks, CA: Sage Publications)
11
Available via license: CC BY 3.0
Content may be subject to copyright.