Geoforum 40 (2009) 589?601

 

:1

Contents lists available at ScienceDirect

GEOFORUM

Geoforum

   

journal homepage: 

 

The production of unequal risk in hazardscapes: An explanatory frame applied
to disaster at the US?Mexico border

Timothy W. Collins

Department of Sociology and Anthropology, University of Texas at El Paso, 500 West University Avenue, El Paso, TX 79968, United States

 

ARTICLE INFO ABSTRACT

 

 

Article history:
Received 23 June 2008
Received in revised form 5 March 2009

This paper advances understanding of how unequal risks to environmental hazards are generated. Mar-
ginalization is the best known explanation for the production of risk. The concept of marginalization was
elaborated through studies of hazards in the global South and connotes how social inequalities constrain
livelihood options of less powerful social groups. Thus, marginal groups are pressured to degrade land-
scapes and occupy hazardous environments while they experience decreased capacities to cope with

 

environmental change. This paper directs analytical attention beyond the material to include the discur-
Vilzrierrability sive realm and demonstrates that the production of unequal risk is contingent upon how hazards are dif?

Risk ferentially perceived, represented, and contested in social spaces. Findings from a ?ood disaster case
study in the El Paso (US)?Ciudad Juarez (Mexico) border metropolis highlight how hegemonic discourses
reinforce material processes of marginalization in ?ood-prone social spaces. Findings also reveal how
socially-powerful geographical groups of people have harnessed institutional resources in their efforts
to externalize risks and capitalize on environmental opportunities in some ?ood?prone areas. The con?
cept of facilitation is used to explain the material-discursive production of socially-elite, ?ood-prone
spaces. Facilitation clari?es how powerful groups are provided privileged access to institutional resources
in their pursuit of environmental rewards, contributing to unjust socioenvironmental outcomes. In con-
clusion, I outline key aspects of how unequal risks are materially and discursively (re)produced within
hazardscapes through relational processes of marginalization/facilitation.

Political ecology
Discourse
United States?Mexico border

 

2009 Elsevier Ltd. All rights reserved.

 

1. Introduction

Between 27 July and 7 September 2006, the Paso del Norte (El
Paso County, USA and Ciudad Juarez, Mexico) received more then
two times its average annual rainfall, resulting in widespread ?ood
damage. Like other disasters, the Paso del Norte ?oods occurred at
the intersection of vulnerable people and hazardous built environ-
ments. Unlike most others, this disaster affected a conurbation
straddling the boundary of two countries. Flood impacts were un-
even. Losses in Juarez were more severe than in El Paso, with
approximately 5000 homes damaged or destroyed, 20,000 resi-
dents left homeless, and extensive loss of physical infrastructure
(Moyer, 2006; Paterson, 2006). Monetary damage in Juarez ex-
ceeded million, more than twice the city?s annual budget
(Paterson, 2006; Johnson, 2006a). The most severe impacts were
experienced by marginal residents of informal settlements within
ju?rez?s rugged western arroyos arid ?uvial landforms that
?ow intermittently) (Collins, in press). In El Paso County, at least
1500 homes, 20 drainage facilities, and 100 roadways were dam-
aged or destroyed by the ?oods (Cook, 2006), with estimated losses
of million (Crowder, 2006a). However, monetary damage

E-mail address: twcollins@utep.edu

0016?7185/$ see front matter 2009 Elsevier Ltd. All rights reserved.


estimates - considered in relation to the disparity in local govern?
mental revenue between El Paso and Juarez suggest that the
severity of the disaster in El Paso County was an order of magnitude
less than that experienced in Juarez. Flood impacts were severe en?
ough to warrant a federal disaster declaration in El Paso County,
where recovery has been most dif?cult for residents of colonias
informal unincorporated settlements) (Collins, in press).
Critical hazards geographers have built a foundation for under?
standing uneven patterns of risk, such as those exposed by the
2006 Paso del Norte ?oods. From a critical hazards perspective, risk
can be de?ned as the combination of people?s exposure to a hazard
and their social vulnerability their capacity to anticipate, re-
spond to, and recover from damage) (Wisner et al., 2004). Critical
hazards scholars have focused on the role of inequalities in the dif-
ferential exposure of socially-strati?ed groups to hazards and the
unequal capacities people have for coping with disaster losses
Bolin and Stanford, 1998; Hewitt, 1997; Watts, 1983; Wisner
et al., 2004). They have also extended beyond materialist modes of
analysis to focus on the role of discourse in differential experiences
of risks, hazards, and disasters. Hazards geographers have ex-
plained how ?blaming nature' for the consequences of disasters
serves to reproduce unjust socioenvironmental arrangements 
Wisner et al., 2004). Mustafa (2002a,b) has argued that different

590 T.W. Collins/ Geoforum 40 (2009) 589-601

forms of power operating materially and discursively shape
people?s access to resources and their vulnerability to hazards.
More broadly, the role of discourse in human?environmental
change has been a focus of study for many political ecologists
(Robbins, 2004).

Still, the role of discourse has not been fully integrated within a
conceptual framework for understanding the production of un?
equal risk, a gap that this paper addresses. A premise of this paper
is that the production of unequal risk is contingent upon how haz?
ards are differentially perceived, represented, and contested in so?
cial spaces. To integrate discursive processes within a case study of
unequal risk in the Paso del Norte, I draw on Mustafa?s (2005) no?
tion of the hazardscape, which expands focus beyond the material
to include the discursive realm. The hazardscape lens focuses on
material?discursive dynamics in the formation, representation,
and transformation of risk. It directs attention to the various fac?
tors operating across spatiotemporal scales that intersect to
(re)produce risks for particular geographical groups of people.
The hazardscape lens focuses the analyst?s view on how claims
about risk connect with power relations and how discourses in?u?
ence material risks in geographical settings. Hazardscapes are
conceived as simultaneous embodiments of material realities
characterized by unequal risks and discursive spatial forma?
tions in which risks are represented and contested through politi?
cal struggles. The hazardscape lens focuses on risk as part of
everyday life, where ?vulnerability. . .[and] exposure to hazards be?
come exemplars of wider struggles over social justice? (Mustafa,
2005, p. 570).

The Paso del Norte study area is a unique context in which to
examine the production of risk because global, national, and local
discourses have all contributed to the formation of a hazardscape
that re?ects unequal power relations. In the case study, I treat
the bi-national El Paso?Juarez conurbation as one hazardscape
where actors and material-discursive processes operating across
a range of scales produced varied sociospatial experiences of the
2006 Paso del Norte ?ood disaster. This approach enables me to sit-
uate claims made about risk particularly those surrounding the
2006 disaster and to reveal how hegemonic discourses privilege
the interests of some sociospatial groups and undermine the well?
being of others.

In the following section, I review hazards and disasters studies
that have spanned the material-discursive divide, and I introduce a
marginalization/facilitation frame for conceptualizing the produc?
tion of unequal risk. Next, I describe the research approach and
introduce the study area. I then use the hazardscape lens to illumi?
nate how situated experiences, perceptions, and representations
shape unequal ?ood risks in the Paso del Norte. The analysis focuses
speci?cally on discursive constructions and material dimensions of
unequal risk in three social spaces experiencing ?ood losses during
the 2006 disaster. I conclude by outlining important dimensions of
how unequal risks are materially-discursively (re)produced within
hazardscapes through relational processes of marginalization/
facilitation.

2. From marginalization to the production of unequal risk

Within the ?eld of hazards geography, marginalization is the
best known explanation for the production of risk. The concept
of marginalization was elaborated through materialist studies of
how the least powerful people in the global South became vulner-
able to socioenvironmental changes. Marginalization connotes
how social inequalities limit livelihood options of the least power?
ful people, leading them to degrade environments and occupy haz?
ardous landscapes, constraining their abilities to cope with
environmental changes (Blaikie and Brook?eld, 1987; Robbins,

2004; Susman et al., 1983). Marginalization illuminates a vicious
cycle, fueled by unequal power relations and reinforced by succes-
sive social and environmental shocks.

The marginalization frame is useful for understanding the pro-
duction of risk in speci?c contexts; however, I have developed two
propositions for expanding the concept. The ?rst proposition is
that marginalization (and the production of unequal risk) is pred?
icated on unequal power relations and, thus, it must be ideologi?
cally legitimated via hegemonic discourses (cf. Gramsci, 1971).
Barnes and Duncan (1992, p. 8) de?ned ?discourses" as ?frame?
works that embrace particular combinations of narratives, con?
cepts, ideologies and signifying practices,? which often vary
among competing interests, although they may coalesce within
hegemonic formations. Emphasizing a limited range of objects
and relations, a discourse privileges certain perspectives at the ex-
pense of others (Feet and Watts, 1996).

A few hazards scholars have examined how hegemonic dis?
courses reinforce material processes of marginalization. Emphasis
has been placed on how expert knowledge systems and techno?
cratic managerial discourses cater to elite groups and undermine
local knowledge and strategies for coping with risks employed by
marginal people (Hewitt, 1983; Mustafa, 2005; Wisner et al.,
1977). In an exemplary study, Mustafa (2005) analyzed the hege?
monic technocratic discourse that shaped state authorities? mate?
rial interventions within Pakistan's Lai River watershed.
Technocrats saw a limited range of choice in reducing ?ood risks.
They were constrained by of?cial mandates and legitimizing dis-
courses (rather than their own cognitive limitations) to focus on
reducing ?ood hazards by evicting ?oodplain residents and
manipulating the river system (Mustafa, 2005, p. 582). Authorities?
?solutions? were prescriptive typically ?driven by donor agen-
das" and ?rarely involved meaningful public participation? 
582). In contrast, those most vulnerable to ?oods who resided
within informal settlements and wielded little power associated
risks with social injustices and saw a much greater range of choice
in dealing with ?ood hazards. They resisted authorities? attempts
to encroach and displace them using a variety of tactics, including
noncompliance, bribery, and social networking to alter local details
of hydrologic engineering projects. However, residents? resistance
must be understood in social context: ?contestation. . .[was] be-
tween differentially powerful social actors and discourses,? where-
in the ?government?s ?ood managers [were] much more effective
in asserting power over the Lai hazardscape than its residents"
582). Marginalization was legitimized through discourses

that systematically exclude[d] the poor from participation in
the management of their living spaces, repose[d] power in a
bureaucracy distant from and hostile to the civil society, and
facilitate[d] alliances between the bureaucracy and the eco-
nomically powerful land ma?as to the detriment of the poor
and powerless (Mustafa, 2005, p. 580).

In sum, Mustafa revealed how mutually reinforcing material?dis?
cursive processes were integral to marginalization.

A second proposition is that while environmental hazards have
been represented through geographical scholarship as constraints
or negative externalities, they are also viewed by people as rewards
opportunities or positive externalities). The marginalization
concept is rooted in a representation of the environment as hazard
(Collins, 2008, in press). Representing landscapes singularly in
terms of the constraints they impose obscures the complexity of
human?environment relations. The representation of the environ-
ment as hazard underpins the hypothesis that the least powerful
people will come ?to live in the most dangerous places? (Susman
et al., 1983, p. 277) through processes of marginalization. However,
empirical studies have revealed associations between elite groups

T.W. Collins/Geoforum 40 (2009) 589?601 591

and highly hazardous (albeit attractive) environments, contradict-
ing this expectation Collins, 2009; Kates and Haarmann,
1992; Rodrigue, 1993). These seemingly contradictory ?ndings
suggest that people and institutions interact in complex ways with
the rewards associated with environmental hazards. Without inte-
grating an analysis of environmental hazards and rewards in a
common framework, the concept of marginalization cannot be
generally applied to understand the production of unequal risk.

Pelling (1999) extended hazards scholarship in a new direction
by incorporating a consideration of the economic rewards associ?
ated with environmental hazards in the analysis of unequal risk.1
He clari?ed the manner in which Guyanans viewed the international
resources that domestic ?ood hazards attracted as ?a potential
source of rents for [state] well as for individuals
and private sector entrepreneurs" 258). He concluded that, due
to pre?existing power differentials, the economic bene?ts experi?
enced by marginal individuals within the informal economic sector
were meager, while local and national political and economic elites
reaped the lion's share of rewards. Resources were co?opted ?by an
entrenched and politicized elite" with the capacity to adapt to
changing institutional arrangements and rebuff challenges to its
control over decision-making about resource distribution. Pelling?s
analysis demonstrated how the uneven allocation of ?ood hazard re-
sources reproduced ?embedded distributions of power and vulnera-
bility? 249); however, his study did not clarify how positive
perceptions of ?ood?prone environments in?uenced the production
of unequal risks.

Other scholars have focused on the material-discursive con-
struction of environments as hazards and rewards. For example,
Cronon (1996, pp. 31?32) observed:

the people who build in exposed locations like these the
locations most susceptible to ?re and mud are often those
with the greatest ability not to do so. . .[This] would seem to
make no sense at all. And yet while standing in the ashes
with scenes of devastation in all directions, one can easily
see why people build here anyway. . .They put themselves
and their families at risk for the simple reason that they
want to be close to nature.

This illuminates how environments shape and re?ect social con?
structions of value and do not conform to one?dimensional
characterizations. It must be recognized that environments simulta?
neously present hazards and rewards and, moreover, that incen?
tives exist for people to create institutional arrangements for
externalizing risks and appropriating rewards. In sum, to more fully
understand the production of unequal risk, multiple in?uences on
decision-making should be recognized, including people's desires
to accumulate wealth, consume environmental rewards, and avoid
risks.

Davis (1998) clari?ed how the consumption of environmental
amenities by elites in hazardous landscapes is facilitated via the
appropriation of institutional resources. He compared the produc-
tion of unequal ?re risks between Malibu and Westlake (down-
town Los Angeles), the richest and poorest landscapes of
Southern California. Af?uent Malibu homeowners were granted
insurance, land-use, and relief/reconstruction subsidies, which
stimulated post-disaster development and upward social succes?
sion within the ?re-dependent ecosystems of the Santa Monica
Mountains. Media coverage of Malibu wild?res represented resi?
dents as ?victims.? In turn, entitled Malibu residents denounced

1 Other scholars have revealed how the institutional resources rewards) that
provide means of social?protection against hazards including property insurance,
?nancial credit, emergency response, disaster relief, and reconstruction assistance 
may be appropriated by elites and withheld from those in greatest need Collins,
2008, in press; Davis, 1998; Klein, 2007; Rodrigue, 1993).

the local state for failing to protect them, demanded costly techno-
logical ?xes to reduce risks, and rejected any ??scal responsibility
for the defense of their precarious habitats? 146). ?Elected of?-
cials, acutely sensitive to Malibu?s national prominence in political
fund-raising" 111) responded with federal disaster declarations,
county tax relief, and a commitment to defend Malibu at all costs.
The structural causes of wild?re risks have been veiled by hege-
monic discourses promulgated by ?media hungry politicians?
131) which have fueled anxieties about the ?incendiary
Other.? Popular concerns about wild?re have been displaced from
anti?socio?ecological real estate pro?teering to the threat of arson
posed by the historically-shifting embodiments of privileged
whites? fears Axis saboteurs, black gangs, Muslim terrorists).
Meanwhile, in Westlake?s inner city tenements, the Los Angeles
building safety department?s conjugal relationship with slumlords
has subjected poor, predominantly Central American residents to a
vicious cycle of inadequate regulation and negligent enforcement,
producing a deadly ?re crisis. Consequently, through material-dis-
cursive processes, Southern Californians have come ?to tolerate
two systems of hazard prevention, separate and unequal" (Davis,
1998, p. 147). Davis? innovation is not simply the observation that
institutionalized power relations serve to marginalize the least
powerful people while facilitating access to resources for elites.
The advance lies in his recognition that environments like Malibu
project mutually constitutive hazards and rewards.

These studies provide the basis for expanding the marginaliza?
tion construct and conceptualizing the production of unequal risk.
They demonstrate the salience of my two propositions, that the
production of unequal risk must be legitimated through
hegemonic discourses and that people associate economic and
environmental rewards with hazards. They illuminate how institu-
tionalized power relations and discursive formations expose the
least powerful social groups to hazards and enable elites to appro-
priate social-protection resources within hazardous, amenity-rich
environments. In theoretical terms, these studies suggest that, in
order to foster an understanding of the production of risk applica-
ble across a range of contexts, it is necessary to expand the concept
of marginalization to include a relational focus on facilitation
(Collins, 2008, in press). Facilitation connotes the institutionally-
mediated process that enables powerful geographical groups of
people to externalize risks and appropriate rewards associated
with hazardous places. Hegemonic discourses are integral to pro-
cesses of marginalization and facilitation.

The marginalization/facilitation frame enables understanding of
how asymmetrical power relations shape social relationships with
environmental hazards and how complex patterns of hazard expo-
sure, vulnerability, and risk may be relationally produced across a
range of scales (Collins, 2008, in press). Whereas the marginaliza?
tion construct is built upon a stable, one?dimensional conception
of the environment as hazard, the marginalization]facilitation
frame draws attention to the existence of dynamic, multi?dimen?
sional social constructions of nature, whereby environments may
at once be viewed as hazards, resources, amenities, and more. This
is important since sources of hazards unstable slopes) and re?
wards views) are often materially inseparable (Collins, 2008,
in press). For the analyst, employing the marginalization/facilita?
tion frame to examine environmental hazards and associated re?
wards while focusing the role of institutions in transferring
risks and allocating protective resources across social spaces -
helps clarify understanding. The case study demonstrates the util?
ity of this frame for understanding the production of unequal risk.

3. Methods

I used multiple methods to analyze the production of unequal
?ood risk in the Paso del Norte. Data were gathered through a study

592 T.W. Collins/ Geoforum 40 (2009) 589-601

of institutions, landscapes, and social groups. Semi-structured key
informant interviews were conducted with 32 representatives of
organizations involved in the 2006 ?ood disaster in El Paso and
juarez using a snowball sampling approach (February?june
2007). Depending on informants? preferences, interviews were
conducted in English or Spanish. I was present at all interviews
and conducted English?language ones. With Spanish-speaking
informants, bilingual student researchers guided interviews. Inter?
views were used to document local areas affected and the speci?c
nature of ?ood impacts, assess the nature and quality of support
provided to ?ood victims, and evaluate the extent to which recov?
ery needs were being met. Based on their local experiences, infor?
mants were asked to assess the recovery capacities of residents in
impacted areas and describe the obstacles to recovery for the pur?
poses of informing cross?site comparisons of vulnerability in the
aftermath of the disaster. In addition, reviews of policy documents
related to area ?ood hazards, regional histories, and print and
internet media reports about the ?ood disaster were completed.
These qualitative methods of data collection allowed for the devel?
opment of an understanding of the material?discursive geogra?
phies of risk within the case study sites as well as the formation
of risk in the Paso del Norte hazardscape.

To situate qualitative data within the ?uvial and social geogra?
phies of the region, secondary data related were culled from multi?
ple sources. I obtained data for 1589 ?ood?impacted residences
throughout El Paso County collected as part of a damage assess?
ment coordinated by the City of El Paso during and immediately
following the 2006 disaster. Households were encouraged through
news media to self-report ?ood damage and a rapid ?eld assess-
ment of all homes with reported damage was conducted to coordi-
nate the delivery of short-term assistance. I used these data to
clarify the role of the ?ood insurance market in generating uneven
patterns of ?ood recovery in El Paso County. Finally, I compiled
data to characterize social vulnerability throughout the study area
from US and Mexican census agencies, and state and local govern-
ment sources. The data and perspective garnered through these
methods enabled analysis of the production of unequal risk.

4. Physical and social aspects of the Paso del Norte hazardscape

The Paso del Norte has experienced uneven development, which
has in?uenced human?environment relations (Heyman, 2007) and
the production of unequal risks. The driving force of the economy
is industrialization, which depends on proximity to US markets
and the availability of low-wage Mexican labor. Industrialization
has fueled rapid growth in population and increased exposure to
?ood hazards, particularly in ju?rez. In 1940, the El Paso?juarez re-
gion was home to about 180,000 people (49,000 in juarez). The
population in the region in 2000 was greater than ten times the
1940 ?gure of?cially 1,884,356 (Table 1) and it is projected
to approach 3.4 million by the year 2020 (2.5 million in juarez).

The Paso del Norte is prone to ?ash ?ooding during intense pre?
cipitation events. Ideally, urban and regional planning would have
prevented the development of ?ood risks, for example, by prohib?
iting residential construction in hazardous areas. In the Paso del
Norte, however, ?ood risks have generally increased through time
as more people have been placed in unsafe conditions without ac?
cess to basic social-protections, adequate housing, civil infra?
structure, and health care (Suarez Toriello and Chavez, 1996;
Ward, 1999).

In terms of physical geography, the western portion of the
conurbation lies within mountainous terrain. The Rio Grande/Bravo
cuts between the rugged Franklin Mountains to the north and Sierra
ju?rez to the south (Fig. 1 The two respective central business dis?
tricts (CBDs) are situated east of a geophysical pinch in the river.

Further to the east a broad alluvial valley forms and both cities
sprawl across this plain.

In social terms, the Paso del Norte is integrated economically
and culturally but divided politically with inequalities apparent
at multiple scales. Social inequalities are evident between juarez
and El Paso, in terms of incomes and levels of education as well
as institutional capacities, indicated by per capita national gross
domestic product (GDP) and local governmental revenue
(Table 1). Based on their geographies of social vulnerability, juarez
and El Paso share important differences and commonalities with
one another (Fig. 2).2 The ?ood?prone western arroyos of El Paso
proper are home to the most socially-elite residents of the Paso del
Norte. In contrast, the western arroyos which ?ow from the Sierra
juarez on the Mexican side are occupied by the region?s most so-
cially?marginal people. El Paso?s barrios adjacent to the central busi-
ness district (CBD) are home to socially?vulnerable US residents,
whereas Mexican elites are concentrated in central juarez.

In juarez and El Paso County, uneven developments have placed
socially?vulnerable people in ?ood?prone peri?urban settlements.
Informal development is a ?normal' part of Mexican life. There is
less popular knowledge to attest to the existence of 1500 colonias
in Texas counties along the Mexican border, which have been per?
mitted to develop clue to the anti-planning culture in this state.
These informally?developed, unincorporated communities are
home to some 400,000 Texas residents (Ward, 1999). Dwellings
in colonias are typically self-built by residents and lack access to
piped water, sewage treatment, publicly-maintained roads, and
other elements of civil infrastructure that are taken-for-granted
by most US residents.

The ability of state institutions to redistribute a portion of the
social surplus toward risk reduction can impart a degree of security
to residents. Disparities in social-protection capacities between
jurisdictions in?uence patterns of unequal risk (Wisner et al.,
2004). El Paso County has more social-protection resources than
juarez, and, within the El Paso County, municipalities have more
resources than unincorporated areas. Disparities in redistributive
capacities are demonstrated by the fact that the per capita invest-
ment by the City of El Paso is roughly 20 times that of the Municipio
de judrez.

There are relatively high levels of livelihood insecurity a key
aspect of vulnerability (Wisner, 1993) among residents of the
Paso del Norte. Socioeconomic statistics offer an indication of levels
of livelihood insecurity (Table 1). El Paso County had lower median
household income (1999 than Texas (1999 
and US (1999 statistics, and a poverty rate over
two times the US rate Livelihood security and capacities
for self?protection are generally reduced in juarez. According to
an unpublished study conducted by El Colegio de la Frontera Norte
in 2006, median household income for juarez was $7260, while
the statistic for Mexico was $8412 (in 1999 

4.1. Case study sites

Uneven developments have shaped distinctive patterns of ?ood
risk in El Paso and juarez. Through case study, three social spaces
impacted by the ?oods became discernable. The groups affected
by Paso del Norte ?ooding include the elite residents of the Westside

2 Poverty status, level of education, US citizenship status, and per capita local
revenue values for each of the 431 block groups that comprise El Paso County were
obtained from various sources (Table 1). These were combined to create the social
vulnerability values for the US depicted in Fig. 2. The values used to represent social
vulnerability differ between El Paso County and juarez due to incompatibilities in the
data collected by the US and Mexican census agencies. Per capita income and level of
education values for each of the 426 (Areas Geoestadisticas Bdsicas) that
comprise Juarez were combined to create the social vulnerability values depicted in
Fig. 2.

T.W. Collins/Geoforum 40 (2009) 589?601

Table 1
Sociodemographic characteristics of the Paso del Norte region and case study sites.

593

 

Sociodemographic metrics

Geographic units

 

 

Ciudad Juarez, Mexico El Paso County, USA Poniente Westside Colonias

Population (total) 1,204,734 679,622 175,472 69,081 68,1 64
Housing units (total) 271,536 209,672 37,177 26,262 16,622
Per capita income 2756 21,781 2302 34,391 17,444
Mean level of education (years) 8.00 11.15 6.35 14.30 9.58
Renter occupied housing units 20.67 36.44 12.51 39.90 16.26
Population not US citizen a 16.00 a 12.29 24.15
National GDP per capita 6114 37,594 6114 37,594 37,594
Local governmental revenue per capita 116 830 116 928 317
Damaged homes with ?ood insurance in 2006 disaster) 3 57.84 a 93.62 20.00

 

Sources: US Bureau of the Census, US Bureau of Economic Analysis, City of El Paso, El Paso County, Mexican Instituto Nacional de Estadistica Geogra?a Informdtica, Municipal

Treasury of Ciudad Juarez.
a Not applicable.

 

COLONIAS

New Mexico

 

El Paso County
Texas 

 
 
   

US Army
Fon Bliss

 

   

 

 

    

i


USA international Boundary CITY OF EL PASO
Mexrco 43?,
4? 0
PONIENTE ?0 9% 08?"

i CIUDAD JUAREZ
I 
i k?

I 5555 \l 
I

i

Municipio de Juarez

 

 

tud Area?: 4"

   

 

 

10

Chihuahua

 

20 Kilometers



 

Fig. 1. The Paso del Norte study area and case study sites. (Sources: US Bureau of the Census, Mexican Instituto Nacional de Estadistica Geografia Informdtica,



of the City of El Paso, and the marginal residents of El Paso County
colonias and Juarez's Poniente.

On the western edge of the City of El Paso, upscale suburbaniza-
tion of the hazardous and scenic upland arroyos that sweep the
slopes of the Franklin Mountains has occurred following World
War 11 (Fig. 1). This has led to the formation of the Westside, the
most exclusive district in a city otherwise poor by US standards.
Westsiders are relatively well-educated and, in relation to El Paso
County as a whole, a higher proportion of them are US citizens
(Table 1).

Colonias of unincorporated El Paso County exist in stark contrast
to the Westside. With the globalization of production, El Paso has
experienced deindustrialization (Spener, 2002). Meanwhile, El
Paso has continued to receive migrants, many of whom are poor.
Deindustrialization in tandem with the growth of the underem-
ployed labor force has produced to a housing crisis (Ward, 1999).
In the search for a home, economic barriers have channeled so-
cially-marginal people into unincorporated colonias (Bath et al.,
1999). Ward (1999) estimated that there were more than 70,000
residents of El Paso County living in over 150 colonias. Per capita

594 T.W. Collins/ Geoforum 40 (2009) 589?601

 

 

 

 

 

 

 

 

 

Lower social vulnerability
El Moderate social vulnerability
- Higher social vulnerability

5

 

Franklin
WESTWAY Mm-
State
Park
US Army
MOWAD Fort Bliss


USA International Boundary SAIPAN I
Mexico CIT-Y OF EL PASO commas
cab - 
PONIENT . 
CBD
- 
Sierra -. 
Juarez 3+5, ,3 I.

. 9 
Municipio de Juarez
Legend

 

10 20 Kilometers


 

Fig. 2. Social vulnerability in El Paso County and Ciudad Juarez. (Source: US Bureau of the Census, Mexican Instituto Nacional de Estadistica Geogra?a Infonndtica,



income among colonia residents is approximately half that of
Westsiders and considerably less than the ?gure for El Paso
County in aggregate; the lower-level of education attained by
colonia adults also re?ects a pattern of low socioeconomic status
(Table 1). In addition, nearly 25% of colonia residents are not US cit?
izens, a statistic that speaks to the character of vulnerability in
colonias.

In the rugged western arroyos of Juarez, directly opposite the
international boundary from El Paso?s Westside, is the third case
study site: the Poniente. While similar in physical geography, the
Poniente stands in social contrast to the Westside, as it is home to
the study area?s most socially?, political-, and economically-mar?
ginal inhabitants (Figs. 1 and 2, Table 1). It contains a patchwork
of informal, irregular neighborhoods asentamientos), which
have developed through decades of squatter settlement. In Juarez,
where af?uent residents lack access to forms of social-protection
considered basic by US elites, hazardous environs like arroyos are
viewed as undesirable places of dwelling. As a result, ?ood-prone
locations in the Poniente are inhabited by socially-marginal people.
While the Municipio the formal local state apparatus) makes it
policy to restrict growth from the Poniente, which contains many
of?cially designated ?high risk? ?ood zones, residents come to live
here because they have no other options in the housing market
(Ward, 1999). Residents even long?term ones often lack (undis?
puted) legal title to the land.

5. The 2006 ?oods

This section applies a hazardscape lens to the 2006 ?ood disas-
ter and examines the material-discursive geographies of unequal
risk in the Paso del Norte. The analysis reveals tensions between
experiences and constructions of risk and it demonstrates how
hegemonic discursive formations have reinforced processes of
marginalization/facilitation, contributing to the reproduction of
unequal risks.

5.1. The transnational production of unequal risk

Hegemonic discourses assert that Mexico generates myriad
transboundary threats from chronic pollution, to natural re-
source depletion, to illegal immigration, to drug-related crime,
and so on - while providing few bene?ts to the US border region
(Urban, 2008). These discourses have played a role in reproducing
unequal ?ood risks between El Paso and Juarez and in generating
?ood impacts in socially?marginal spaces.

During the 2006 ?ood disaster, an antiquated dam (the Dique
Montada) located immediately above Juarez?s Poniente was on the
cusp of catastrophic failure. It was predicted that if the dam
failed a ?ood wall would blast through the Poniente, swash across
the Rio Grande/Bravo, and inundate El Paso's CBD (Gilot, 2007). In
response, the US Army Corps of Engineers prevented the total

T.W. Collins/Geoforum 40 (2009) 589?601 595

failure of the dam. By all accounts including those reported by
the media and others provided through interviews in El Paso and
Juarez this represented the foremost example of cross-border
collaboration to reduce ?ood risk, one involving generous US dona-
tions to protect seemingly incompetent Mexican neighbors. Thus,
people in the Paso del Norte reconciled the Dique Montada incident
based on prior ?knowledge? that had been produced via hegemonic
discourses about transboundary (in)security.

However, viewing the transnational hazardscape from an alter?
native perspective one not ?xated on the Dique Montada episode
it appears that Juarez has provided El Paso with human resources
for ?ood recovery and that there have been impediments to deliv?
ering assistance from El Paso to Juarez. An important transnational
recovery resource ?owing from Juarez to El Paso has been embod?
ied by Mexican laborers, who have engaged in the work of recov?
ery for impacted localities with suf?cient purchasing power 
the Westside). Low-wage Mexican labor has reduced costs of
repairing physical infrastructure, by working for individual house-
holds and businesses and for municipal subcontractors with
the City of El Paso). In this respect, ?ood reconstruction in El Paso
has depended on the exploitation of immigrant labor as it has
along the US Gulf Coast following Hurricane Katrina (Fletcher
et al., 2007).

During the event, Juarez of?cials? requests for public aid were
denied by state Chihuahua) and federal Mexican government
of?cials. Since the US is wealthier than Mexico, one might expect
that recovery assistance from the US would have been directed to
support recovery in juarez. Those with neoliberal ideals, for
example, might have expected that investments by US corpora-
tions in the North American Development Bank, a market-based
development institution, would be made and infrastructure
reconstruction loans provided to Juarez. However, aside from
the Dique Montada incident which was of such concern on the
US side due to the potential impacts in El Paso little evidence
exists of US state or market institutional contributions to ?ood
recovery in Juarez. On the other hand, numerous US-based non-
governmental organizations have attempted to assist Juarenses
with recovery.

Interviews revealed how regulatory intervention impeded orga?
nizations, including the US-branches of the Salvation Army and the
Rotary Club, from providing material aid like clothing and shoes
to Mexico after the ?oods. A representative of the Mexican?
branch of the Salvation Army in Juarez described the problem
(Interview in Spanish, 14 March 2007):

The Salvation Army is very powerful at the worldwide level,
but locally, Juarez is not a strong Army. It doesn?t have a lot
of resources. We base ourselves more on what we can obtain
from the El Paso Army, which is my big brother at my side,
and they are the ones that will provide me with everything
that I might need, if they can do so. Our government unfor-
tunately, the laws are sometimes absurd. Instead of opening
the door so that the resources can ?ow and get to the people
in need, our government says ?Stop, you can't bring that
across." What we need here is more accessibility in order
to get resources to the people in need.

A representative of the Salvation Army in El Paso discussed the
restriction (Interview, 14 March 2007):

We had signi?cant problems with juarez the government 
not Juarez as a place. Early on, like two days into the ?ood?
ing, we got a call requesting material donations for over
there. We said ?No problem. We?d be more than happy to."
We had about two dozen pallets of sorted used clothing.
When we got ready to send the supplies south we called

the consul general and told them that we wanted the Salva-
tion Army in Juarez to be involved; the phone went dead
silent. It is against the law to take used clothing into Mexico.
We went through everyone through the chief customs per-
son, the consul general, the congressman?s of?ce. The
impression that I got from all the representatives that we
tried to work with was that they were more concerned with
offending the government down there than they were in
providing relief for the hurting people in the [asentamien-
tos]. I did an interview [with local media] where I made
the comment that ?It was easier to get drugs from Juarez
into El Paso than it was to get used clothing from here to
there for relief," and that is true.

A Mexican Consulate spokeswoman said that the bottleneck was a
result of North American Free Trade Agreement rules that prohibit
the shipment of bulk clothing and shoes into Mexico because such
materials, even if they were intended as donations, could be used
for trade. She summed up the situation: ?We can work with the Sal?
vation Army, but we cannot get through those regulations" (Meritz,
2006, up).

At the US federal-level, the coupling of national security and
anti?immigration discourses with inadequate funding for ?ood
protection led to (in)actions that contributed to impacts during
the disaster. In one case, concerns about border security directly
led to ?ooding in downtown El Paso?s Segundo Barrio. A concrete-
lined 4-mi1e long segment of the Rio Grande/Bravo was con-
structed in the 19605, which dissects the most urbanized portion
of the Paso del Norte. Connecting laterally with the main channel
are tunnels designed to convey overland ?ow from abutting urban
drainages. These tunnels were reputedly used as entry points by
illegal immigrants. Post?9/11, the Department of Homeland Secu?
rity covered tunnel entrances with a system of grates to stop the
stream of migrants. While the grates may have prevented migrants
from using those speci?c tunnels to enter the US, according to the
City of El Paso (2007a), they also prevented the ?ow of storm?
water to the Rio Grande/Bravo as debris accumulated behind the
grates causing back?ow, ?ooding homes of the working poor in
Segundo Barrio.

Contradicting the of?cial state discourse of concern about bor-
der safety is the fact that there has been inadequate maintenance
of ?ood protection infrastructure in the Paso del Norte, an issue that
has played a prominent role in other US disasters (Comfort, 2006).
Due to lack of funding from the US and Mexico for the International
Boundary Water Commission (IBWC), the bi-national institution
with management authority over the channel of the Rio Grande/
Bravo, more than 1.5 million m3 of silt have accumulated since
the early 19905 in the concrete-lined segment of the river
(City of El Paso, 2007a). Accumulated silt has raised the channel
2.5?3.7 above the design elevation, placing large swaths of the
adjacent urbanized ?oodplain at elevations below that of the river.
The silt-raised channel along with the Homeland Security grates
led to back?ow and ?ooding in low-lying urban areas where storm
drains designed to gravitationally move water to the river 
conveyed ?ows in the opposite direction, inundating neighbor-
hoods and homes (City of El Paso, 2007a).3

In sum, it appears that at the US federal level there is dissonance
between ideal and material aspects of transboundary risk. While

3 Due to heightened recognition of risk along the Rio Grande after the disaster, the
FEMA declared that the entire levee system in El Paso County was not ?certi?able,?
which placed the ?oodplain at ?high risk" in revised FEMA maps. In lieu of receiving
federal funds to upgrade the levees and obtain FEMA certi?cation, tens of thousands
of property owners within the newly designated at?risk areas will be required to
purchase ?ood insurance. There has been no indication at the US federal level that the
resources needed to improve the levee system will be allocated. As a result, the IBWC
has been scrambling to secure funds for levee upgrading and certi?cation.

596 T.W. Collins/ Geoforum 40 (2009) 589-601

security concerns are paramount in of?cial pronouncements, the
materialities of transnational ?ood risk and hazard reduction belie
the discourse. It is important to acknowledge the sociospatial groups
that have been most burdened by this contradictory discursive for-
mation: those most in need in Mexico have been denied transna-
tional resources for recovery while low-lying, socially?marginal
barrios in the US have been subjected to ?ood damage.

5.2. The production of unequal risk across local social spaces

Experiences and discourses of ?ood risk varied in reference to
the case study social spaces. In Juarez?s Poniente, the ?ood disaster
demonstrated residents? vulnerability, which was in?uenced by to
their limited capacities for self?protection and their lack of access
to social-protections (Fig. 3). A Cruz Roja (Mexican Red Cross)
disaster responder from Juarez stated: ?The neighborhoods that
are middle to upper class, nothing happened there. The ones most
affected were those on the western periphery we ran into a lot of
cardboard homes where the poor had squatted in dangerous
zones.? (Interview in Spanish, Cruz Roja, 7 April 2007).

While some informants attributed unsafe conditions in the
Poniente to apolitical pressures rapid growth) others, includ?
ing high-level Municipio of?cials, described how the area was set?
tled through power politics:

All of [the Poniente] is ?invaded? land. Decades ago, the pol-
iticians owned land and they would send people there to
invade and to become votes. Then [the people] would come
to the Municipio and ask for services: water, electricity,
drainage, pavement. Since they are settled in the most
expensive part of the city to provide services, the Municipio
does not have the resources. The general idea is to not give
them services. We need to get them out (Interview in
Spanish, Obras Publicas, 26 April 2007).

For the Municipio, the ?ood ?crisis? signaled an opportunity to get
households in arroyos out and to keep them from coming back:

In normal situations when we attempt move residents away
from arroyos, we cannot because they seek legal protection
against relocation. At the moment the [?ood] tragedy hap?
pened, residents complained about their situation; they said
that it was our fault that they were there. I said ?This is the
moment that we have been looking for" when desperation
will lead residents to make decisions that they would not
under normal circumstances. There were around 400 fami?
lies that were relocated. Their homes were 

 

Fig. 3. ?Self-help' settlement in Ciudad ju?rez?s Poniente.

are trying to ensure by all means that no one will go back
(Interview in Spanish, Protecci?n Civil, 24 May 2007).

As these quotes suggest, juarez authorities attributed Poniente ?ood
damage to ?high risk? squatter settlement, which provided of?cial
justi?cation for evicting residents and razing dwellings. In practice,
however, the pressure placed by the Municipio on Poniente asentam-
ientos has served landed capitalists? ?accumulation through dispos-
session" interests (cf. Harvey, 2006), and it has further marginalized
Poniente residents. The Municipio makes it of?cial policy to restrict
growth from the Poniente due to the rugged terrain, which makes
the expansion of urban infrastructure costly (Ward, 1999). Yet the
Municipio has informally authorized the westward development of
asentamientos in arroyos of the Poniente because these places ex-
isted beyond the margins of the urban space economy. Recent
developments, however, have contradicted the formal growth pol-
icy, as landed capitalist interests have leveraged municipal invest-
ments for westward growth to capitalize on property value
increases. The completion of a costly loop freeway and the expan-
sion of nearby border crossings in New Mexico have greatly in-
creased Poniente land values. In one case, increasing land values
have underpinned a dispute in which a powerful landed family
has placed socially?marginal residents of an asentamiento under
siege by claiming ownership of the land, posting armed guards on
site, and destroying dwellings (Gilot, 2008). In concert with capital-
ist interests, Municipio authorities have accused residents of ?land
speculation" even though many purchased (now disputed) title to
property and have lived there for decades, before the land was im-
bued with exchange value. In sum, the increasing property values
produced through Ju?rez?s uneven development have been detri-
mental to households without clear title to land, of which there
are many in the Poniente. From this viewpoint, the Municipio?s attri?
butions of risk appear to provide an of?cial rationale to legitimate
state-sponsored squatter eviction and Poniente land appropriation
in the disaster?s aftermath.

In El Paso, sharp inequalities in access to resources for ?ood
recovery were evident between incorporated areas and colonias.
The City of El Paso has demonstrated capacity in stimulating recov?
ery (Crowder, 2006b, 2007; Bracamontes, 2006). Incorporated
areas became eligible for Federal Emergency Management Agency
(FEMA) reimbursements totaling 75% of recovery and mitigation
expenses. In contrast, colonias are not eligible for FEMA ?public"
assistance to ?x damaged roads) as infrastructure is consid?
ered ?private? and beyond the purview of the state. Case workers
from Project Hope (Interview, 29 May 2007), which strives to assist
?ood victims with unmet needs, discussed the lack of assistance for
colonia residents:

There was one gentleman that had a lot of rubble that he
stripped out of the house. then another family had adobe
completely piled up. The piles were so huge that they looked
like pyramids. They asked ?why didn?t the city come and
clean that up?" At the beginning [the City of El Paso was]
offering free services to clean up. but they said that it was
the county?s jurisdiction, and then the county said that it
was the city?s, and nothing ever got done. It is like passing
the buck. That left a lot of people to do it themselves.

Unequal access was determined in part by socioeconomic status.
The constraints experienced by households with insecure liveli-
hoods were compounded by the fact that few maintained ?ood
insurance. In many cases of ?ood damage, engineering failures ex-
posed topographically low?lying residential locations outside of
?high risk? areas based on FEMA's of?cial designations to ?oodwa-
ters (FEMA, 2008). Because these were not designated as ?high risk"
areas, residents were not required to maintain ?ood insurance.
Based on unpublished City of El Paso data, 42% of ?ood-impacted

T.W. Collins/Geoforum 40 (2009) 589?601 597

 

 

A b, 5-. Franklin 3
2. Mountains
State Park 
. a . i 
Franklin 05AM),
Mountains ?ne, Font Bliss
State Park .. us:- 
.9 7"
HeaVIly . =A?Elpason52
- f  El Paso - 
ooded USA . .
area v" - 
MEXICO 3511Ciudad Juarez . 

MEXICO Juarez -

. 

 

 

 

 

.

Legend 0 2.5 5 10m 
- Flood damaged properly -

 

Fig. 4. Westside media bias: (A) Reprinted map of the ?heavily ?ooded area" from the Dallas Morning News; (B) Map of widespread ?ood damage in US based on data
collected by the City of El Paso. (Source: The Dallas Morning News, 12 August 2006; City of El Paso.)

households suffered uninsured losses.4 These data indicate that
100% of high?income and 70% of moderate?income households had
?ood insurance to compensate for losses while only 18% of low-in-
come households did. For Westside residents, property repairs may
have absorbed considerable resources. However, ?ood damage did
not undermine basic well-being among typical Westside households.
since material recovery in 94% of such cases was covered by ?ood
insurance (Table 1).

In contrast, only 20% of colonia households maintained ?ood
insurance at the time of the disaster (Table 1). To compound mat-
ters, monetary assets are lacking among colonia residents, meaning
that recovery has depended on capacities to garner external re-
sources. Accessing resources requires social entitlements, which
colonia residents lack in relation to Westsiders. As was the case
in Mexico, civil society organizations many of them church?
and community-based have partially ?lled the social-protection
void in colonias: ?The church groups focused more on the colonias
because they knew that these people didn't have the means to help
themselves. They didn't have insurance. They didn?t feel that they
could have FEMA aid" (Interview, United Way, 23 May 2007).

Meanwhile, popular media attention focused on the plight of El
Paso?s af?uent Westside. Local and national media reports de-
scribed a scene of devastation on the Westside, while they re?
mained relatively silent about ?ood impacts in Juarez, the
remainder of the City of El Paso, and in El Paso County colonias
(Fig. 4A and B). Uneven media attention served to broadcast West-
side residents? admissions of exposure to remediable risks, which
facilitated their access to resources. In accessing resources for ?ood
recovery, the Westside has bene?ted from highly?educated resi-
dents who have drawn on technical knowledge to pinpoint causes
of infrastructure failure and place speci?c demands for remedia-
tion on state institutions demands that have been heard by city
of?cials. In cases of Westside infrastructure failure and ?ood dam-
age, the City of El Paso has deemed system upgrades to be ?feasi-
ble;? many Westside ?ood locations have been placed at the top
of the city's priority list (Crowder, 2006a; Chavez, 2007; City of
El Paso, 2007b). A city councilperson observed that ?The Westside

4 The dataset includes 423 residences on the Westside. Damaged homes outside of
the City of El Paso, however, are underrepresented only 60 homes in
unincorporated areas are included in the dataset). This underrepresentation can be
explained in reference to the factors that in?uence marginality in colonias more
generally fear of deportation, lack of English-language pro?ciency). Hence, it is
likely that the statistic for the proportion of colonia residents with ?ood insurance is
in?ated (Table 1).

seems to get treated better than other parts of the city. If [only
other areas] had ?ooded, would they have spent this amount of
money on it?. . .Realistically no" (El Paso Times Staff, 2008, 

Evidence suggests that, even within a relatively well-endowed
jurisdiction like the City of El Paso, public investment in recovery
can be uneven and favor elite spaces. The cases of two relatively
poor neighborhoods, each located within the City of El Paso, re?ect
how state institutions have negatively affected less powerful geo?
graphical groups through the application of technocratic con-
structs in ?risk reduction? programs. The Mowad and Saipan
neighborhoods both ?ooded clue to civil infrastructure failure
(Fig. 2).

The Mowad neighborhood ?ooded as the result of a diversion
channel failure in August 2006. Residents returned soon thereafter
and began investing themselves (and their money) in reconstruc?
tion. In September 2006, residents were shocked when the City
of El Paso issued a ?nding that Mowad was ?not conducive to res-
idential development? and that there was ?no practical drainage
solution that is cost-effective and where success can be demon-
strated" (City of El Paso, 2006, The city concluded that engi-
neering solutions were infeasible and residents would have to be
moved to safer ground elsewhere. During public meetings resi-
dents protested the city's determination, asserting it was unjust



f-



 

Fig. 5. Mowad residents protested city plans to buy and demolish their ?ood-
damaged homes during a September 2006 El Paso City Council meeting. (Source:
Rudy Gutierrez, El Paso Times, 27 September 2006.)

598 T.W. Collins/ Geoforum 40 (2009) 589-601

(see Fig. 5). Residents argued that the homes they had built and the
sense of community that had developed would be lost with neigh-
borhood demolition and that those values were not re?ected in the
exchange values of their properties estimated for buyout purposes
(Interview, Project Hope, 29 May 2007). The city proceeded with
the involuntary relocation plan. A long?time Mowad resident said:
?It?s an injustice, but what can we do? They (the city) didn?t main?
tain the [diversion channel], and now they want to ?x it by knock?
ing down the houses? (johnson, 2006b, By summer 2007,
Mowad had been demolished.

The Saipan neighborhood was located in central El Paso. In
August 2006, a detention pond maintained by the US Army at Fort
Bliss over?owed, inundating nearly 60 neighborhood homes. In
October 2006, while residents were investing in repairs, the city
was ?determining options for hazard mitigation through an engi?
neered drainage solution? (City of El Paso, 2006, To residents'
dismay, the city concluded that an engineering ?x was infeasible
and, by January 2007, city agents were reviewing properties for
?acquisition" (City of El Paso, 2007b). Technical problems in the buy?
out process arose due to the unique characteristics of Saipan, which
included small homes and very low average property values (2007
Comparable properties elsewhere in the city were val?
ued at 2007 on average. While media reports and politi?
cians branded the relocation program a success (johnson, 2007), the
20 buyout offers accepted by residents as of March 2007 totaled
$1,043,243 an average of about $52,000 per home. Even if the city
had bridged differences in property exchange values, households
would still have experienced a loss of neighborhood use values, a
point made by an informant from Project Hope (29 May 2007):

The communities are very tight and everybody knows each
other. Saipan, for example, I can really understand that it is
not just a street to them. It is a family unit. They considered
their neighbors as a family because their children grew up
together, their children had grandchildren, and they have
been there for over forty years. Everybody knows each other
and everybody helps each other. [From the resident?s per?
spective] it?s like you have replaced my home, but I still lost
the things that bring the memories back to me. You have
also disconnected me from someone that I can walk over
and have tea with. It was de?nitely hard for our clients. It
is so hard to ?nd places that [residents] could get that the
government could say ?This is comparable to what you had.?
Many of them went through selection processes to ?nd out
?No, you don?t qualify for that one. It is twenty thousand dol?
lars too much.? Then many of the people that were relocated
weren't young. They don't drive. It's not like they can say [to
a longtime neighbor] ?I?ll see you tomorrow." I mean, we
take a lot of things for granted.

Why have engineering ?xes been deemed feasible on the Westside
and infeasible in socially?marginal spaces? The cost-bene?t ratio?
nale applied to decision?making advantaged residents of economi-
cally?valuable lands and disadvantaged occupants of property
with relatively less exchange value. In Mowad and Saipan - so?
cially-marginal neighborhoods with great recovery needs - the
low aggregate exchange value of property made the investment of
public resources in infrastructure upgrades unjusti?able based on
cost-bene?t criteria. As a city council representative stated: ?The
buyouts in the Saipan area cost about $5 million, but ?xing the
storm-water management problem there would have cost about
$10 million? (johnson, 2007, On the relatively wealthy West-
side, the higher exchange value of property coupled with media
coverage and citizen pressure made investment in civil infrastruc-
ture economically rational (and perhaps politically necessary). Fur-

thermore, in the Mowad and Saipan cases, the most marginal
households faced the greatest hardships. In each case, relocation
schemes were designed to match relocation funds with appraised
values on a one?to?one basis for property owners. As a result, tenants
received limited dislocation aid and the needs of households with
unveri?ed title to property were not met (Interview, Project Hope,
29 May 2007; Crowder, 2008). These experiences re?ect the injus?
tices embedded within the technocratic cost?bene?t discourse.

Such impositions of external authority have not affected El Paso
County colonia residents. However, colonias have received little
media attention, and their invisibility has translated into reduced
access to aid. In one case, a land?ll operated by the City of El Paso
in an outlying area of the county breached and spewed refuse?la?
den water into a residential area (Johnson, 2006c). Properties were
covered by 60 cm of waste. One resident, who was quoted a price
of $100,000 to remediate damage, said ?We kind of got forgotten
about when the Westside started ?ooding. We're dealing with a
?ood situation and a health situation" Gohnson, 2006c, 

Many of my informants had little involvement with colonias,
and the few who did described desperate situations. For example,
the colonia of Westway northwest of El Paso ?eluded media cov-
erage (Fig. 2). Through interviews it came to light that ?Westway
was pretty much washed away" (Interview, El Paso County Of?ce
of Emergency Management, 6 April 2007). A representative of the
Red Cross in El Paso said:

In Westway you have a lot of extended families and now
instead of having ?ve people living in a home now you have
ten. We see that a lot: a lot of kids, a lot of relatives, all living
under one roof in very cramped conditions. It will take a long
time for them to either replace their homes or rebuild some-
where else" (Interview, 22 March 2007).

In conjunction with their invisibility, colonias have received a dearth
of external aid, with basic needs having gone unmet. Workers with
the El Paso County Health and Environmental District said: ?When
we went to Westway the residents said that it was the ?rst time
that anybody from any government agency had been out to see
the damage" (Interview, 15 May 2007). They added:

It was obvious that they needed ?nancial assistance to get
back on their feet. One lady that had children - her mobile
home was sinking into the ground and she didn't have any?
one to help. When clearing the debris, they were like: ?What
are we supposed to do with this stuff?" The economic bur?
den on them they were having a hard time because they
didn't have any assistance (Interview, 15 May 2007).

Workers from Project Hope discussed colonia residents? awareness
of connections between their invisibility and their lack of access
to resources:

There are areas, like Westway, where they say: ?Mowad got
all the attention and we didn?t. They were taken care of and
they weren't.? You know, the spotlight they believe that
Mowad took all the glory; they were looked after, but all
the colonias like Westway weren't. There is a hostility of cer?
tain areas versus other areas. They are still living in condi?
tions that are not humane. Some of those people still
haven?t found other housing; so they live off their family. If
their family is in Mexico, it is a hard coin to toss between
being in the US and they may not be legal but they can't
go back to Mexico, because if they go, they probably won't
be able to get back over. So they are torn; they are caught
(Interview, 29 May 2007).

As the quote above suggests, the marginalization experienced in
colonias has been encompassed by nationalist and anti-immigration

T.W. Collins/Geoforum 40 (2009) 589?601 599

discourses that hold sway in the US. Legal and political structures
enacted through such discourses serve to further marginalize low?
income and immigrant households who are disproportionately con?
centrated in colonias and other socially?marginal landscapes. A bar?
rier a relatively high proportion of colonia residents encountered
derived from the federal requirements of eligibility for state?subsi-
dized forms of assistance. Federal Emergency Management Agency
(FEMA) requirements restrict receipt of ?individual? assistance
based on legal US residency status. Since legal residents did not
want to sacri?ce the well?being of neighbors, friends, and relatives
(who lacked legal status) while pursuing resources, many forfeited
their entitlements to state aid. As an emergency manager ex-
plained: ?Colonias are like no man?s land. . .A lot of the time you
have people that are in the country illegally who live with family
members who are here legally. They become afraid to ask the gov-
ernment for assistance. . .Those are the people that fall through the
cracks" (Interview, El Paso County Of?ce of Emergency Manage-
ment, 6 April 2007). Workers from Project Hope added: ?When
FEMA was out the ?rst time it was all based on English; there wasn't
anything in Spanishthe people that were affected"
(Interview, 29 May 2007). This especially disadvantaged colonia res-
idents. Hegemonic nationalist and anti-immigration discourses
made these obstacles more dif?cult to surmount:

FEMA did not receive people who were scared to give infor?
mation because of documentation and fear of the govern?
ment. Again, there's that issue of immigration. The people
are afraid because of all the discussion going on at the fed-
eral-level about deporting people that are illegal, about putt?
ing pressure on the borders, about how to treat the ?illegals'
that are already here. There is a huge amount of fear. To a lot
of people the federal government is the federal government;
they don't know the difference between FEMA, the Border
Patrol, and the IRS (Interview, United Way, 23 May 2007).

In summary, the analysis illuminates dissonance between experi-
ences and discourses of ?ood risk. Within the case study social
spaces hegemonic discourses have been central to the reproduction
of unequal risks. Powerful social spaces, such as the Westside, have
been facilitated in ?ood recovery through these discourses, while
vulnerable ones, such as Poniente asentamientos and El Paso County
colonias, have been marginalized.

6. Conclusion: the production of unequal risk

Before highlighting general features of the production of un-
equal risk based on the literature review and case study, the ques-
tion of who is at risk within the Paso del Norte must be addressed.
Many residents of the Westside choose to live in hazardous loca?
tions. They do so under the condition that state and market re?
sources ?ood control infrastructure, ?ood insurance,
emergency aid) are provided to maximize aesthetic and economic
rewards, and minimize environmental risks. While such house-
holds may be exposed to ?ood hazards, they are not socially?vul?
nerable. Accumulated assets and privileged access to institutional
resources facilitate their lifestyle pursuits amidst ?ood hazards.
Other households live in ?ood-prone colonias of El Paso County
and asentamientos of Ju?rez?s Poniente. Such people have few assets
to invest in hazard reduction, insecure livelihoods, and constrained
access to protective resources via institutions Municipio de
Juarez, City of El Paso, FEMA, ?ood insurance and formal housing
markets). They experience acute hazard exposure and social vul-
nerability; hence, they are at?high?risk to ?oods. The case study
illustrates how state and market institutions have played a dual?
role in producing unequal risk by enabling the development of
acute residential ?ood hazards while not making social?protec?

tions available to vulnerable people. The role of these institutions
in neglecting vulnerable Poniente and colonia residents while
providing resources for af?uent households on the Westside exem-
pli?es the mutually constitutive nature of marginalization/
facilitation.

Several important features of the production of unequal risk can
be distilled from the literature review and case study:

1. Environmental hazards are viewed by people as opportuni?
ties. They attract resources that state and market agents (and res?
idents) recognize as a ?source of rents" (Pelling, 1999, p. 258).
Additionally, people see hazardous landscapes as aesthetically-
attractive environments. Facilitation is a process whereby more
powerful groups are granted access to the economic and aesthetic
rewards associated with environmental hazards at the expense of
less powerful groups. Facilitation depends on the creation of par-
ticular desires for views, privacy, seclusion, naturalness) and
material relations the sancti?cation of private property rights,
the formation of ?xed capital, the provision of social-protections)
that allow for excess pro?ts to be realized. For example, high qual?
ity views and attractive (albeit hazardous) arroyo environments
exist on the Westside and in the Poniente. However, they only pro?
vide the basis for accumulation on the Westside, where state-sub-
sidized improvements to the built environment and priority access
to social-protections coincide with biophysical characteristics that
satisfy the view?lot aesthetic, hence swelling rents.

2. Because unjust transfers of wealth (and risk) must be legiti-
mated through ideological means, hegemonic discourses are essen?
tial to relational processes of marginalization/facilitation and the
production of unequal risk. A technocratic managerialist discourse
shapes the contemporary production of risk (Mustafa, 2005). Tech-
nocratic interventions emphasize risk reduction through the
manipulation of biophysical environments. Technocratic manage-
ment fails to reduce risks and produces unequal risks for sev-
eral reasons. The modernist ideology that underpins technocratic
decision-making neglects the reality that social and biophysical
conditions are in constant ?ux, subject to processes of creative
destruction, climate change, land use transformation, and so on.
The assumption that ?ood risks can be isolated and ?solved? with
engineering ?xes is violated by these constantly changing materi?
alities. Additionally, by emphasizing the physical dimension of risk,
technocratic interventions fail to reduce social vulnerability
2003). Technocratic ?solutions? focus instead on relocating
marginal people away from hazardous environments such as
Poniente arroyos which constrains livelihoods and ampli?es vul?
nerabilities. Such hazard reduction programs including those
that are idealized as ?participatory? rarely involve the meaningful
involvement of vulnerable people (Pelling, 1999; Mustafa, 2005).
Even when involuntary relocation programs are underwritten by
the state, as they were by the City of El Paso, they systematically
disadvantage vulnerable groups due to the ?objective' cost?bene?t
criteria they necessarily employ. To reduce vulnerability, institu?
tional programs must meet the needs of the least powerful groups.
Without the implementation of genuine need-based programs,
institutional resources for social?protection are too easily co?opted
by elites those least in need of assistance which ampli?es un?
equal risks.

3. More speci?c regional discourses combine with the prevail?
ing technocratic managerialist one to create hegemonic discursive
formations within hazardscapes. These discursive formations
shape the production of unequal risks through the facilitation of
elite geographical groups of people and the marginalization of oth-
ers. Dissonance between lived experiences and hegemonic regional
discourses is integral to the reproduction of unequal risks. In the
Paso del Norte, national security and anti?immigration discourses
especially those that construct Mexicans as threats have in?u-
enced the production of unequal risks. The institutional structures

600 T.W. Collins/ Geoforum 40 (2009) 589?601

created through such discourses legal residency requirements
for FEMA aid) have excluded many immigrant households 
including those with legal entitlements from access to ?ood
recovery resources. Meanwhile, experiences that contradict these
discourses Mexican laborers reducing ?ood recovery costs
in El Paso, the US neglecting to maintain ?ood protection infra?
structure) have been rendered less legible because they do not con?
form to the hegemonic frame. In the Paso del Norte, contradictory
regional discourses have further marginalized already vulnerable
sociospatial groups in the Poniente, El Paso barrios, and
colonias).

4. The case study provides the basis for classifying local
discourses of risk. In the aftermath of the Paso del Norte ?oods, local
social spaces were subjected to one of two general discourses,
which in?uenced processes of marginalization/facilitation.
Attributions of risk made by state agents (and in some cases pri?
vate interests) involve the denial of responsibility and displace?
ment of blame for unsafe conditions. As hazards scholars have
noted, ?naturalizing? risks attributing them to natural rather
than social causes is a narrative strategy for denying responsibil?
ity. For example, an informant from the City of El Paso (Interview, 1
March 2007) stated:

Normally, we get monsoon storms that move through very
quickly. There might be some rapid runoff and some reports
of minor ?ooding, but nothing terribly serious. But [in 2006]
the weather patterns aligned themselves and the storm con-
tinued. It was a set of conditions that I wouldn't call it a
?perfect storm" but they came together. It was not a 20-
year or a 100-year event. It was a 500-year event.

Alternatively, when risks are attributed to social causes, they are typ?
ically displaced upon vulnerable, stigmatized groups (Pelling, 1999;
Davis, 1998). In the Paso del Norte, attributions have come in multiple
forms: from US assertions about Mexican impositions of transboun-
dary ?ood risks to Juarez of?cials? claims that Poniente ?ood damage
resulted from illegal settlement in ?high risk" zones. Attributions of
risk were made in reference to socially?marginal spaces Mowad,
Saipan, the Poniente), where they justi?ed the imposition of external
authority and from the perspective of many residents induced
negatively perceived changes. In contrast, admissions of risk involve
claims of exposure to remediable risks and demands for assistance.
Because this discourse is founded on a sense of social entitlement,
it is typically deployed by residents of elite social spaces to enlist
institutional resources for local risk reduction. Politicians and media
coverage are instrumental in representing such people as victims, cir?
culating their claims of undue exposure to harm, and warranting the
demands they place upon institutions. Admissions of risk typically
contribute to local empowerment. In the Paso del Norte hazardscape,
these discourses of risk have in?uenced the reproduction of unequal
?ood risks by legitimating the displacement of vulnerable Poniente
residents, the institutional marginalization of people in colonias,
and the over-allocation of social-protection resources to elite groups
on the Westside.

5. People respond to the dynamics of marginalization/facilita-
tion through both compliance and resistance. In the Paso del Norte,
hegemonic global technocratic managerialist), national 
border security, anti-immigration), and local discourses Po-
niente residents create their own vulnerability) have articulated
to produce a hazardscape de?ned by highly asymmetrical power
relations and unequal risks. Hegemonic discourses circulate as
truths that shape the way people see the world, and, thus, they
are invoked even by people who appear to be poorly served by
them. For example, marginal people place faith in modernist
assumptions about the value of expert knowledge and ef?cacy of
technocratic interventions. They may take?for-granted ?natural?

explanations for disasters and they may even accept the notion
they are responsible for creating their own vulnerability. Alterna-
tively, through their experiences, marginal people may understand
that their risks are in?uenced by power dynamics that expose
them to hazards and deny them access to resources (Mustafa,
2005). Some marginal residents of the Paso del Norte like those
displaced from the Mowad neighborhood - have come to see ?ood
risk through the lens of socioenvironmental justice and they have
engaged in acts of resistance. However, they have not successfully
transformed unjust arrangements. This is because contestation in
and over the Paso del Norte hazardscape as Mustafa (2005) ob-
served in Pakistan has been between differentially powerful ac-
tors and discourses, wherein technocrats and elite geographical
groups of people have been more effective in asserting power than
vulnerable residents of ?ood-prone social spaces. Through further
examination of the material-discursive geographies of unequal
risk, a deeper understanding of how hegemonic discursive forma?
tions reinforce processes of marginalization/facilitation can be
achieved. Hopefully, such analyses will contribute to an improved
understanding of how alternative discourses may be effectively de-
ployed to destabilize unjust socioenvironmental arrangements.

Acknowledgements

I am grateful for the insights of Daanish Mustafa on an early
draft of this paper. I also thank Sara Grineski, Cristina Morales,
and Josiah Heyman for their constructive comments and sugges-
tions on various versions of this manuscript. My doctoral advisors
Kevin McHugh and Bob Bolin contributed to the development of
the ideas presented. I also acknowledge the very helpful comments
of three anonymous Geoforum reviewers and editor Scott Prud-
ham. Those who made the ?eld component of this research possi-
ble must be recognized, including my research assistants, Mauricio
Austin and Oscar Morales, as well as the representatives of organi-
zations who made themselves available for interviews. I am also
thankful for the institutional support I received through the
University of Texas at El Paso in the form of a University Research
Institute Grant, without which this work would not have been fea-
sible. Any errors are my own.

References

Barnes, T., Duncan, J., 1992. Writing Worlds: Discourse, Text, and Metaphor in the
Representation of Landscape. Routledge, New York.

Bath, C., Tanski, J., Villareal, R., 1999. The failure to provide basic services to the
colonias of El Paso County: a case of environmental racism? In: Camacho, D.
Environmental Injustices, Political Struggles. Duke University Press,
Durham, NC, pp. 125?137.

Blaikie, P., Brook?eld, H., 1987. Land Degradation and Society. Methuen, London.

Bolin, B., Stanford, L., 1998. The Northridge Earthquake: Vulnerability and Disaster.
Routledge, London.

Bracamontes, R., 2006. City wants to buy out residents in ?ood zone. El Paso Times.
15 September.

Chavez, A., 2007. City drainage improvements take shape slowly. El Paso Times, 29
July.

City of El Paso, 2006. Storm 2006: staff updates, 3 October. Public document
obtained from Records Department.

City of El Paso, 2007a. Storm 2006 update, 6 March. Public document obtained from
Records Department.

City of El Paso, 2007b. Storm 2006 update: accomplishments since the rain stopped,
9 January. Public document obtained from Records Department.

Collins, T., 2008. The political ecology of hazard vulnerability: marginalization,
facilitation and the production of differential risk to urban wild?res in Arizona?s
White Mountains. Journal of Political Ecology 15, 21?43.

Collins, T., 2009. In?uences on wild?re hazard exposure in Arizona?s High Country.
Society and Natural Resources 22, 211?229.

Collins, T., in press. Marginalization, facilitation, and the production of unequal risk:
the 2006 Paso del Norte ?oods. Antipode.

Comfort, L.. 2006. Cities at risk: Hurricane Katrina and the drowning of New
Orleans. Urban Affairs Review 4 (41), 501?516.

Cook, J., 2006. Letter from El Paso Mayor John F. Cook to US. President George W.
Bush, 14 August. 


T.W. Collins/Geoforum 40 (2009) 589-601 601

Cronon, W., 1996. Introduction: in search of nature. In: Cronon, W. 
Uncommon Ground: Rethinking the Human Place in Nature. W.W. Norton
and Co., New York, pp. 23?66.

Crowder, D., 2006a. City repairs take time, money. El Paso Times, 7 September.

Crowder, D., 2006b. Council OKs $25 million in bonds to begin repairs. El Paso
Times, 20 September.

Crowder, D., 2007. City advances utility to manage drainage, ?ood projects. El Paso
Times, 18 May.

Crowder, D., 2008. Neighborhood is razed: 2 families still in Saipan area. El Paso
Times, 2 March.

Davis, M., 1998. The Ecology of Fear: Los Angeles and the Imagination of Disaster.
Metropolitan, New York.

El Paso Times Staff. 2008. Eddie Holguin: why in the next ?ve years? Blog, posted 5
May 2008 (accessed 10.05.2008).

FEMA, 2008. De?nitions of FEMA Flood Zone Designations. 



Fletcher, L., Pham, P., Stover, E., Vinck, P., 2007. Latino workers and human rights in
the aftermath of Hurricane Katrina. Berkeley Journal of Employment and Labor
Law 28 (1), 107?153.

T., 2003. Critical Political Ecology: The Politics of Environmental Science.
Routledge, London.

Gilot, L., 2007. Dam remains vulnerable, Ju?renses say. El Paso Times, 29 July.

Gilot, L., 2008. Group: human rights violated inJuarez land dispute. El Paso Times, 7
March.

Gramsci, A., 1971. Selections from the Prison Notebooks. International Publishers,
New York.

Harvey, D., 2006. Spaces of Global Capitalism: Towards a Theory of Uneven
Geographical Development. Verso, London.

Hewitt, K., 1983. The idea of calamity in a technocratic age. In: Hewitt, K. 
Interpretations of Calamity from the Viewpoint of Human Ecology. Allen and
Unwin, Boston, MA, pp. 3?32.

Hewitt, K., 1997. Regions of Risk: A Geographical Introduction to Disasters.
Longman, Essex, UK.

Heyman, J., 2007. Environmental issues at the US?Mexico border and the
unequal territorialization of value. In: Hornberg, A., McNeill, J., Martinez?
Alier, J. Rethinking Environmental History: World?Systems History
and Global Environmental Change. Altamira Press, Lanham, MD, pp. 327?
344.

Johnson, E., 2006a. Two dry days forecast for El Paso. El Paso Times, 6 September.

Johnson, E., 2006b. Mowad demolition begins: ?ood-damaged homes razed in
buyout. El Paso Times, 29 December.

Johnson, E., 2006c. Levee at Clint Land?ll breached, spreads waste. El Paso Times, 21
August.

Johnson, E., 2007. Residents of Saipan area ?nd new homes: ?ood victims move on.
El Paso Times, 21 June.

Kates, R., Haarmann, V., 1992. Where the poor live: are the assumptions correct?
Environment 34, 4?28.

Klein, N., 2007. The Shock Doctrine: The Rise of Disaster Capitalism. Metropolitan,
New York.

Meritz, D., 2006. Salvation Army aid stymied at border. El Paso Times, 2 September.

Moyer, H., 2006. Texas in ?desperate need?. Disaster News Network: Current News, 7
August.

Mustafa, D., 2002a. To each according to his power? Access to irrigation water and
vulnerability to ?ood hazard in Pakistan. Environment and Planning D: Society
and Space 20, 737?752.

Mustafa, D., 2002b. Linking access and vulnerability: perceptions of irrigation and
?ood management in Pakistan. The Professional Geographer 54, 94?105.

Mustafa, D., 2005. The production of an urban hazardscape in Pakistan: modernity,
vulnerability, and the range of choice. Annals of the Association of American
Geographers 95, 566?586.

Paterson, K., 2006. Recovering from the borderlands little Katrina. Frontera
NorterSur: Online, US?Mexico Border News, 13 August.

Peet, R., Watts, M., 1996. Liberation ecology: development, sustainability, and
environment in an age of market triumphalism. In: Peet, R., Watts, M. 
Liberation Ecologies: Environment, Development, Social Movements.
Routledge, New York, pp. 1-45.

Pelling, M., 1999. The political ecology of ?ood hazard in urban Guyana. Geoforum
3O (3), 249?261.

Robbins, P., 2004. Political Ecology: A Critical Introduction. Blackwell Publishers,
Malden, MA.

Rodrigue, C., 1993. Home with a View: Chaparral ?re hazard and the social
geographies of risk and vulnerability. The California Geographer 33, 29?42.
Spener, D., 2002. The unraveling seam: NAFTA and the decline of the apparel
industry in El Paso, Texas. In: Geref?, G., Spener, D., Bair,J. Free Trade and
Uneven Development: The North American Apparel Industry after NAFTA.

Temple University Press, Philadelphia, pp. 139?160.

Suarez Toriello, E., Chavez, 0., 1996. Per?l del la Frontera M?xico?Estados Unidos/
Pro?le of the United States?Mexico Border. FEMAP, Ciudad Juarez, MX.

Susman, P., O?Keefe, P., Wisner, B., 1983. Global disasters, a radical interpretation.
In: Hewitt, K. Interpretations of Calamity from the Viewpoint of Human
Ecology. Allen and Unwin, Boston, pp. 263?283.

Urban, J., 2008. Nation, Immigration and Environmental Security. Palgrave
MacMillan, New York.

Ward, R., 1999. Colonias and Public Policy in Texas and Mexico: Urbanization by
Stealth. University of Texas Press, Austin.

Watts, M., 1983. On the poverty of theory: natural hazards research in context. In:
Hewitt, K. Interpretations of Calamity from the Viewpoint of Human
Ecology. Allen and Unwin, Boston, pp. 231?262.

Wisner, B., 1993. Disaster vulnerability: scale, power, and daily life. GeoJournal 30,
127?140.

Wisner, B., Blaikie, P., Cannon, T., Davis, I., 2004. At Risk: Natural Hazards, People's
Vulnerability and Disasters, second ed. Routledge, London.

Wisner, B., O?Keefe, P., Westgate, K., 1977. Global systems and local disasters: the
untapped power of people?s science. Disasters 1, 47?57.