University of Southern California Dissertations and Theses

Location of warehouses and environmental justice: Three essays

(USC Thesis Other)

Location of warehouses and environmental justice: Three essays

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)

Location of Warehouses and Environmental Justice:
Three Essays

By

Quan Yuan

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA

In Partial Fulfillment of the
Requirements for the Degree

DOCTOR OF PHILOSOPHY
URBAN PLANNING AND DEVELOPMENT

August 2018

Copyright 2018 Quan Yuan
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)

Copyright © 2018 by Quan Yuan

All rights reserved. Printed in the United States of America. No part of this dissertation
may be reproduced or transmitted in any manner or by any means whatsoever without
written permission from the author, except in the case of brief quotations embodied in
critical articles or reviews.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)

I dedicate this dissertation to my parents, my wife, and my son. You are my sunshine.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
i

Acknowledgement
I would like to express my deep appreciation and gratitude to my advisor and
dissertation committee chair, Professor Genevieve Giuliano, for her generous support and
guidance throughout the five-year Ph.D. program. Professor Giuliano is such a great
mentor, who has taught me how to think critically and independently, and stay motivated
and patient as an academic person. I would also like to thank my dissertation committee
members, Professor Marlon Boarnet, and Professor Manuel Pastor for their thought-
provoking suggestions and valuable help over these years. I am indebted to my
colleagues Sanggyun Kang, Yuting Hou, Xize Wang, Eunjin Shin, Nathan Hutson, and
Yougeng Lu for their help during my years of study at the University of Southern
California.
I give my special thanks to my father, Wei Yuan, my mother, Yuanfeng Liu, my
father-in-law, Qining Duan, and my mother-in-law, Ying Lei, who have always been
supportive and encouraging. Without their wholehearted support, I would not be able to
pursue my dream of becoming a social scientist. I met my wife Qianyao at the USC Price.
She is my best friend, a good listener, and a magician who can create beauty, harmony,
and happiness. My son Evian was born in the fourth year of my Ph.D. program. He has
brought us so much joy and helped me learn to better balance life and work.

Quan Yuan
August 2018
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
ii

Abstract of the Dissertation
This dissertation examines a rising environmental justice problem in warehousing
location. Following a stepwise approach, the three-essay research identifies the
socioeconomic processes that contribute to the problem and the relevant agents involved
in the processes, evaluates the relationship between warehousing facilities and socially
advantaged population using empirical data and econometric models, and provides policy
implications on how to mitigate the environmental inequity. Results show that
warehouses are significantly more likely to be located in minority neighborhoods, and the
disproportionate siting of warehouses is the primary drive for the spatial coincidence. In
addition, local public policies greatly affect the location of warehouses and may make a
clear difference in reducing the disparate distribution. The dissertation justifies the
significance of the environmental inequity in warehousing location and implies that
relevant agents including governments, warehousing developers, and local residents need
to work together in a regional collaboration framework to address the problem.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
iii

Table of Contents

Acknowledgement .............................................................................................................. i
Abstract of the Dissertation ............................................................................................. ii
List of Tables ..................................................................................................................... vi
List of Figures ................................................................................................................. viii
CHAPTER 1 Introduction .............................................................................................. 1
1. Dissertation Overview ............................................................................................. 1
2. Contributions of the Dissertation ............................................................................. 5
3. Organization of the Dissertation .............................................................................. 7
CHAPTER 2 Location of Warehouses and Environmental Justice ............................ 9
Abstract ........................................................................................................................... 9
1. Introduction ............................................................................................................ 10
2. Literature Review................................................................................................... 12
2.1. Location choice of warehousing facilities ...................................................... 12
2.2. Warehouses, externalities and environmental justice ..................................... 14
3. Research Approach ................................................................................................ 20
3.1. Models ............................................................................................................ 20
3.2. Data sources .................................................................................................... 22
3.3. Data description .............................................................................................. 25
4. Results .................................................................................................................... 28
4.1. Model 1 ........................................................................................................... 29
4.2. Model 2 ........................................................................................................... 30
4.3. Model 3 ........................................................................................................... 31
5. Conclusion ............................................................................................................. 35
CHAPTER 3 Mega Freight Generators in My Backyard:
A Longitudinal Study of Environmental Justice in Warehousing Location .............. 39
Abstract ......................................................................................................................... 39
1. Introduction ............................................................................................................ 41
2. Literature Review................................................................................................... 43
2.1. Contradictions in longitudinal studies on environmental justice .................... 43
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
iv

2.2. Location choices of warehousing facilities and minority people ................... 48
3. Research Approach ................................................................................................ 50
4. Data ........................................................................................................................ 59
5. Results .................................................................................................................... 62
5.1. Summary statistics .......................................................................................... 62
5.2. Simultaneous Equation Model: minorities as a whole ................................... 64
5.3. Simultaneous Equation Model: different race groups .................................... 70
5.4. Simultaneous Equation Model: ethnic churning ............................................ 75
6. Conclusion ............................................................................................................. 79
CHAPTER 4 Environmental Justice in Warehousing Location across Cities:
Institutional Perspectives ............................................................................................... 83
Abstract ......................................................................................................................... 83
1. Introduction ............................................................................................................ 85
2. Literature Review................................................................................................... 88
2.1. Warehouses as locally undesirable land uses? ................................................ 88
2.2. The role of local public policies in warehousing location choice .................. 92
3. Data and Approach ................................................................................................. 95
3.1. Research approach .......................................................................................... 95
3.2. Study area and observation period .................................................................. 97
3.3. Selecting the sample for comparative case studies ....................................... 100
3.4. Data description ............................................................................................ 105
4. Results .................................................................................................................. 107
4.1. Land use policies .......................................................................................... 107
4.2. Job related policies ........................................................................................110
4.3. Financial incentives .......................................................................................112
4.4. Environmental regulations .............................................................................115
5. Discussion ............................................................................................................ 120
5.1. A synthesis of policies .................................................................................. 120
5.2. Tradeoffs made to make policies .................................................................. 124
5.3. It is about the history; yet it is also about the future ..................................... 125
6. Implications on Environmental Justice Issues ..................................................... 128
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
v

CHAPTER 5 Conclusions ........................................................................................... 131
References ...................................................................................................................... 136

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
vi

List of Tables
Table 1 Descriptive statistics of variables ......................................................................... 27
Table 2 Regression analysis results ................................................................................... 32
Table 3 Expected relationship between the dependent variable and control variables
in each equation of the model ........................................................................................... 58
Table 4 Descriptive statistics of variables ......................................................................... 61
Table 5 Comparing tracts with and without adjacent W&Ds in the Los Angeles
region, in 2000 and 2010 respectively .............................................................................. 62
Table 6 Comparing values in selected variables in 2000 between tracts that received
adjacent W&Ds during 2000-2010 and other tracts in the Los Angeles region ............... 63
Table 7 Comparing value changes in selected variables between tracts with adjacent
W&Ds in 2000 and other tracts in the Los Angeles region .............................................. 64
Table 8 Regression analysis result of SEM Equation 1 (Dependent variable:
Changes in warehousing activity densities 2000-2010) .................................................... 67
Table 9 Regression analysis result of SEM Equation 2 (Dependent variable:
Changes in percentage of minorities 2000-2010) ............................................................. 68
Table 10 Regression analysis result of SEM Equation 1 for different race groups
(Dependent variable: Changes in warehousing activity densities during 2000-2010) ..... 72
Table 11 Regression analysis result of SEM Equation 2 for different race groups
(Dependent variable: Changes in percentage of race group during 2000-2010) .............. 73
Table 12 Regression analysis result of SEM Equation 1 (Dependent variable:
Changes in warehousing activity densities 2000-2010) .................................................... 77
Table 13 Regression analysis result of SEM Equation 2 (Dependent variable:
Ethnic churning during 2000-2010) .................................................................................. 78
Table 14 Municipalities within each subregion .............................................................. 101
Table 15 Summary of socioeconomic characteristics and indicators for
warehousing development of case study municipalities ................................................. 104
Table 16 Municipalities and institutions that interviewees are from .............................. 107
Table 17 Number of land parcels, total land area and share of land zoned as
industrial in the case study cities in 2008 ....................................................................... 108
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
vii

Table 18 Number of industrial-zoned parcels with area of more than 10 acres
in the eight case study cities in 2008 ............................................................................. 109
Table 19 Tax rates in effect of case study cities ...............................................................113
Table 20 A synthesis of warehousing related public policies adopted in case
study cities ...................................................................................................................... 121

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
viii

List of Figures
Figure 1 Spatial distribution of warehouses and diesel PM emissions in the
Los Angeles region (urbanized areas) ............................................................................... 17
Figure 2 The year built of warehousing facilities in the Los Angeles region .................. 24
Figure 3 Distribution of warehouses and freight generators in the Los Angeles
region ............................................................................................................................... 26
Figure 4 Spatial distribution of warehouses and different types of neighborhoods ......... 28
Figure 5 Year built of existing warehousing facilities in the Los Angeles region ............ 53
Figure 6 Illustration of the spatial relationship between a specific census tract
and its adjacent warehouses .............................................................................................. 54
Figure 7 Changes in the percentage of minority population and location of
warehouses built during 2000-2010 in the Los Angeles region ........................................ 66
Figure 8 Changes in the percentage of Black population and location of
warehouses built during 2000-2010 in the Los Angeles region ........................................ 74
Figure 9 Total added square footage (in millions) of warehousing facilities by
municipality in the municipalities from during 1996 to- 2016 ......................................... 99
Figure 10 Sub-regions in the Los Angeles CSA ............................................................. 101
Figure 11 The exterior of the Kroger Company Food distribution center in
Compton ...........................................................................................................................118
Figure 12 Accumulated built-up area of newly built warehouses in Carson
and Compton until 2016................................................................................................. 122
Figure 13 Land use patterns in Rancho Cucamonga and Upland in 2008 ...................... 123
Figure 14 Accumulated built-up area of newly built warehouses in selected
cities during 1966-2016 ................................................................................................. 127

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
1

CHAPTER 1
Introduction

1. Dissertation Overview
Global trade has been greatly changing the world in both positive and negative ways.
While the improved transportation infrastructure and emerging technologies enable
expeditious international goods movement, the freight system that makes this happen
nonetheless generates significant environmental and social impacts on local communities.
The society-wide benefits in contrast with the localized externalities have attracted
growing attention from the government, the public and the academia. Warehouses and
distribution centers (W&Ds), for instance, play a vital role in the freight system and at the
same time contribute to the concentration of freight-related impacts in the host
communities. Unfortunately, these impacts have not been carefully studied yet.
In response to the increasing yet changing freight demand, logistics service
providers rely on W&Ds to make supply chains more efficient and effective. The
strategic placement of these facilities allows the positioning of products and services
close to major markets and customers (Ecklund, 2010). To accommodate large volumes
of goods and serve widespread customers, warehousing developers choose to build more
facilities, especially in the major metropolitan areas where the supply networks are
complicated and the demand is strong (Bergqvist, 2016). The proliferation of warehouses,
however, leads to the spatial expansion of freight-related environmental externalities. As
intermediates connecting suppliers and customers along the supply chains, warehouses
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
2

act like magnets that attract a large number of truck trips. Truck movement generates air
pollution (Kim et al., 2004; Kozawa, et al., 2009), noise (Seto et al, 2007), pavement
damages (Cidell, 2015), and traffic safety concern (Roudsari, et al., 2004; Clifton,
Burnier, and Akar, 2009). Apart from these impacts, warehousing facilities as huge flat-
roof structures also pose environmental threats such as heat island effects (Voogt, 2007).
In addition to the spreading of the externalities, the uneven distribution of these
externalities is equally discernible. A small proportion of communities nevertheless
receive the majority of warehouses, and residents in those host communities have begun
to realize such disparities. The localized but uneven impacts have then caused concern
about a potential environmental justice problem. However, there has been very limited
research—either theoretical understanding or empirical evidence—on this problem.
To explore the environmental inequity in warehousing location, I would focus on
the following interdependent research questions and seek answers to these questions
through reviewing existing research progress and analyzing empirical data.
Research Question 1: Where are warehouses spatially distributed?
Research Question 2: Are warehouses disproportionately located in socially
disadvantaged neighborhoods?
Research Question 3: What are the socioeconomic processes that contribute to the
environmental justice problem in warehousing location?
Research Question 4: What are the probable mechanisms behind the problem
given the behaviors of relevant agents involved in the socioeconomic processes?
Research Question 5: Do local public policies play a significant role in creating
the environmental justice problem?
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
3

Research Question 6: Given the role of local public policy making, what can local
governments do to address the problem?
The three essays of the dissertation examine these research questions in a step-
wise sequence. They are logically interdependent and mutually supported, making the
dissertation a comprehensive and cohesive whole.
The Essay One, Location of Warehouses and Environmental Justice, which bears
the same name of the dissertation, lays the theoretical foundation of the entire dissertation
by connecting warehousing location to the concept of environmental justice. It then
examines the spatial distribution of warehouses using cross-sectional data for the Los
Angeles region. Multivariate statistical models are developed to test whether warehouses
are more likely to be located in socially disadvantaged (defined as poor or minority-
dominated) neighborhoods. Results show that medium-income minority and low-income
minority neighborhoods receive significantly more warehousing activities. Percentage of
the minority population is positively associated with warehousing location as expected,
but the effects of household income are less clear. This essay confirms the existence of
environmental inequity in warehousing location, but the cross-sectional analysis does not
provide much information on how the inequity is created.
The Essay Two, Mega Freight Generators: A Longitudinal Analysis of
Environmental Justice in Warehousing Location, is built on the Essay One while
extending the discussion of the environmental justice problem to a longitudinal
examination. In this essay, I develop a two-equation Simultaneous Equation Model to
estimate the interdependent relationship between the location choices of warehouses and
minority population. With controlling for other location factors, the model demonstrates
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
4

that the siting of warehouses follows minority population but not vice versa. This finding
supplements the results in the Essay One with the observation on the dynamic
socioeconomic processes. The longitudinal analysis provides useful evidence on the
mechanism of the environmental inequity in warehousing location and suggests the
necessity to regulate excessive warehousing development. Public policies apparently play
a critical role in facilitating such regulations. But the Essay One and Two, which are both
quantitative, failed to carefully study the role of public policy making, due to certain
limitations in research design.
In response to this gap in the first two essays, I conduct a qualitative study on how
public policies affect warehousing location and the subsequent environmental justice
problem in my Third Essay, Environmental justice in warehousing location across cities:
institutional perspectives. To obtain the information on the effects, I interview city
planners, public agency staff, warehousing developers, and local residents, and study
public documents such as general plans and zoning codes. By comparing pairs of
municipalities that have similar socioeconomic characteristics but different trajectories of
warehousing development, I am able to identify how major public policy elements
contribute to the siting of warehouses in various ways. At the end of the article, I
conclude that although the current environmental justice patterns in warehousing location
are less relevant to public policies adopted in the last decades, these policies in reverse
can greatly reshape the landscape of warehousing development. If local governments can
make the best use of those policies, the environmental justice problem can be effectively
alleviated.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
5

Therefore, the three essays of the dissertations are within a consistent theoretical
framework but respectively focus on separate key issues related to the research topic.
Through the three essays, I articulate the research questions, review the relevant
literature, develop effective methods, and test hypotheses that help understand the
patterns and mechanism of the environmental inequity in warehousing location.
2. Contributions of the Dissertation
The dissertation explores a research topic that has not been seriously studied despite the
growing public concern. There has been very limited research on the theoretical
understanding, empirical testing, and policy solutions of the environmental justice
problem in warehousing location. The dissertation provides a comprehensive overview of
the problem, relates the problem to the existing literature, evaluates the effectiveness of
various methodologies, tests several key hypotheses using data from consistent sources,
and discusses the policy implications of the empirical findings.
In the Essay One, I link the on-going logistics restructuring to the spatial
expansion of warehousing facilities and related environmental impacts, and further raise
the central research question: is warehousing location an environmental justice problem
given the localized environmental externalities? To explore this question, I use both
spatial techniques and econometric models to identify the spatial relationship between
warehouses and socially disadvantaged neighborhoods. While the conceptual model is
not essentially different from traditional ones, I use three ways to measure the distribution
of warehouses: a dummy variable showing whether or not a spatial unit contains at least
one warehouse, a discrete variable showing how many warehouses are located in a spatial
unit, and a continuous variable showing the intensity of warehousing activities (aggregate
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
6

warehousing space). Due to the limited availability of data, most existing studies
nonetheless failed to measure the sizes of locally undesirable facilities. Those three forms
of indicators in the Essay One can therefore more comprehensively measure the
distribution of environmental hazards, not only the facilities. The results showed that in
spite of these variations in measuring warehousing location, the environmental justice
problem is significant in the way that minority population bear more warehousing related
externalities.
In the Essay Two, I examine the mutual relationship between the location choices
of warehousing facilities and minority population. By connecting the two dynamic
socioeconomic processes together, I emphasize the necessity of using longitudinal data to
decouple the interdependent relationship as a means of understanding the mechanism
behind the environmental justice problem. Before proceeding to empirical testing, I point
out several important variations in research design and discussed how these variations
could lead to different or even contradictory results. These discussions are in particular
critical for accurately estimate how the siting of warehouses and the movement of
minority people affect each other. In the empirical test section, I estimate the
interdependent relationships between warehousing development and minority relocation
in three dimensions: minority as a whole, three different minority groups (African
American, Latino, and Asian), and ethnic churning (which will be explained in the Essay
Two). These dimensions, especially the last two, which have been largely neglected in
the literature, can provide a broad scope for evaluating the causes of the environmental
justice problem.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
7

In the Essay Three, I stress the crucial role of local public policies in warehousing
location and the emergence of environmental justice problem, considering the
development of warehouses is essentially a land use decision process. As most of the
environmental justice studies are from the fields of sociology and environmental studies,
policy implications are not their main focuses. This essay evaluates the environmental
justice problem from the perspective of planning and public policy making. I conduct
four pairwise comparative case studies and examine a few major policy elements in
relation to warehousing location. By linking local policy making to environmental
inequity, I demonstrate a largely ignored aspect of the environmental justice inquires—
why and how local public policies can contribute to the solution of the problem.
Apart from the research topic, the dissertation also contributed to the broader
literature of environmental justice by introducing warehouses into the general framework
and clarifying a few methodological choices. For instance, in the Essay Two, I argue that
the results in some previous studies rejected the existence of environmental inequity,
probably because that the siting of certain locally undesirable land uses in minority or
poor neighborhoods, as a “rare event”, is unlikely to infer statistical significance in
econometric models. The statistical insignificance does not necessarily rule out the
possibility that socially disadvantaged population bears more environmental threats.
3. Organization of the Dissertation
As stated above, the dissertation consists of three separate but interdependent essays.
Each essay stands alone as a research paper with an independent structure. As the three
essays focus on the same central research problem with different focuses and
perspectives, their literature review sections partially overlap. All three essays have been
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
8

submitted to peer-reviewed journals or presented at academic conferences. Among them,
the Essay One has been submitted to the Journal of Planning Education and Research,
and the Essay Two has been accepted for publication by the Land Use Policy. After
presenting the three essays, the dissertation will end with a general conclusion and
discussion on policy implications based on the findings in the essays.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
9

CHAPTER 2
Location of Warehouses and Environmental Justice

Abstract
Warehousing activities generate substantial environmental externalities that affect
surrounding neighborhoods. Using data from the Los Angeles region, this study tests the
relationship between the spatial distribution of warehouses and neighborhoods with
different demographic and socioeconomic characteristics. The results show that
warehouses are disproportionately located in both low-income and medium-income
minority neighborhoods. The distribution of warehousing facilities and activities is highly
related to the percentage of minorities as expected, but its relationship with household
income is nonetheless mixed. In the LA region, low-income neighborhoods are not
always attractive to warehouse developers because they are not convenient for
warehousing development.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
10

1. Introduction
The rapid growth of industrial production and household consumption in the recent
decades has driven up the demand for freight movement. The volume and frequency of
freight flow across borders has increased substantially in the era of globalization and E-
commerce. In response to the growing freight demand, the expansion of warehouses and
distribution centers (WDCs, defined as large commercial buildings primarily for storage
of goods) in many metropolitan areas was impressive. Between 2006 and 2016,
employment in the warehousing and storage industry in the U.S. increased by 47%,
compared to a 3% increase in the retail industry and a 13% decrease in the manufacturing
industry (Bureau of Labor Statistics, 2017). Today the spatial distribution of warehousing
facilities also remarkably differs from ten years ago. Several important changes in the
warehousing industry jointly lead to the “Logistics Sprawl” phenomenon, which has been
found in many gateway cities including Los Angeles and Atlanta (e.g. Dablanc, Ogilvie
and Goodchild, 2014; Dablanc and Ross, 2012). The new generation of warehouses and
distribution centers were built larger and consumed more land space. Among all WDCs,
the share of large size ones (over 500,000 square feet) has increased significantly
(Andreoli, Goodchild and Vitasek, 2010). Urban cores, which have been well developed,
could hardly accommodate these new warehouses. Due to inflated land values and
increased congestion, the central areas are no longer the best locations for organizing
logistics services. Meanwhile, the evolution of supply chain management gives
warehousing service providers more incentives to think regionally and locate their
facilities in places with better access to an enlarged market area. The trends of
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
11

consolidation and decentralization largely shape the new patterns of logistics facility
siting (Giuliano et al., 2013).
On the other hand, warehousing facilities attract a lot of truck trips, which in turn
generates air pollution, noise and GHG emissions (Dablanc et al., 2013), and contributes
to pavement damage and traffic safety threats in adjacent neighborhoods. Following the
spatial redistribution of WDCs, these environmental impacts have increasingly become
regionally decentralized but locally concentrated. Such a tendency has attracted growing
interest, however, little is known about the characteristics of the neighborhoods with
concentrated environmental burdens. Are disadvantaged neighborhoods with limited
economic and political power more vulnerable to these environmental burdens? If
warehouses are disproportionately located in socially disadvantaged neighborhoods, this
spatial pattern suggests an environmental justice problem in warehousing location.
This paper examines the spatial dimensions of the environmental justice problem
and tests the spatial relationship between warehousing facilities and disadvantaged
neighborhoods using empirical data. In Section 2, I discuss how the location choice of
warehouses is affected by various factors, and to what degree these factors are subject to
the innovations in the warehousing industry. I also review studies on warehousing related
environmental impacts and consider how warehousing location can be fit into the
framework of environmental justice. I present the research approach and explain how
econometric models are used to estimate the relationship between warehouses and
socially disadvantaged population in Section 3. Results from the models are shown in
Section 4. The paper closes with conclusions and policy implications in Section 5.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
12

2. Literature Review
2.1. Location choice of warehousing facilities
Studies on the location choice of warehouses, especially those in the area of operational
management, emphasize the roles of transport access (Bowen, 2008; Verhetsel et al.,
2015), labor costs (Sivitanidou, 1996; Demirel et al., 2010), land rent (Durmus and Turk,
2012), and economies of agglomeration (Warffemius, 2007; McKinnon, 2009). While
these factors are subject to changes in infrastructure capacity, urban structure, and
industry mix, several trends have been systematically affecting the way warehousing
activities are spatially organized.
Warehousing services play an essential part of any industrial supply networks,
on which warehouses are closely linked to customers including wholesalers,
manufacturers, importers, and exporters. Given that customers increasingly benefit from
worldwide production and distribution, warehousing service providers have to respond to
demand from more widespread locations than two decades ago when cross-border goods
movement was less prevalent (Rodrigue, 2008). Both the market and supply chain
fragmentation have greatly influenced the spatial distribution of warehousing facilities.
The location choice making of warehouses is, therefore, more related to regional, inter-
state or even international connections. New warehousing establishments are no longer
closely attached to locations in the immediate vicinity to local centers, but more likely to
be found in suburban areas with good access to regional markets and resources (Hesse,
2007).
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
13

With the extensive development of transportation infrastructure, transportation
costs have dropped substantially during the last decades (Giuliano, 2004). While the city
cores are heavily congested, suburbs are realistic alternatives that offer good access to
regional markets and national networks (Hesse, 2002; Holl and Mariotti, 2016). In
addition, warehousing firms expand their space and capacity by consolidating old
facilities and building larger new ones to accommodate the increased freight demand
(Hesse and Rodrigue, 2004). Suburban locations with low land rent, large land parcels,
and flexible land use policies are popular among warehousing developers. Recent
empirical evidence demonstrates a pattern of decentralization from urban cores to
suburban clusters in quite a few metropolitan areas (Dablanc and Ross, 2012).
Existing studies on the warehousing location choice provide limited information
on institutional and social factors. Zoning ordinances and land use policies imply how
local governments favor different industries, and they can greatly affect the location
choice of warehouses (Dablanc and Rakotonarivo, 2010; Christensen Associates et al.,
2012). Dablanc (2013) discussed how municipal governments’ policies influenced the
growth of warehousing industry in the Los Angeles region. Municipalities, especially
those in a financial crisis, would be more likely to attract warehouses for the purpose of
job and tax revenue generation. Governments may also provide conveniences for
developing warehousing facilities in designated places like business parks (Verhetsel et
al., 2015). The interactions between warehousing location choice and local communities,
on the other hand, have caused increasing concern among the public, but empirical
research on this topic is largely absent.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
14

2.2. Warehouses, externalities and environmental justice
The traditional functions of warehouses and distribution centers include receiving,
checking, storing, loading and shipping goods and materials. In particular, the storage
function had been central to the entire warehousing service chain. With the rapid
evolution of supply chain management in the recent two decades, many conventional
warehousing service providers, however, have been transforming into third-party
logistics providers (3PLs). Modern warehousing facilities are more and more involved in
many value-adding services including inventory management, packaging and return
handling (Bowen, 2008; Cidell, 2011; Rodrigue, 2008). Automated systems are
introduced to fulfill these functions efficiently, and warehousing facilities with rentable
built-up areas (RBA) larger than half a million square feet have become increasingly
popular. In the Los Angeles region, the average RBA of the warehouses built in the last
ten years (2007-2017) reached 140,000 sqft, which is more than twice of that for
warehouses built before 2007 (52,912 sqft), according to CoStar data (2017). As major
intermediates and terminals for goods movement, WDCs generate a large number of
truck trips. Many of these trips are made by combination trucks. According to a 2011
survey of 31 warehouses by Kunzman Associates, Inc. for NAIOP Inland Empire
(Kunzman Associates, 2011), the average truck trip generation rate was 0.3 truck
trip/day/1,000 sqft, and out of all trucks involved, 70% were five plus axle combination
trucks. Another study by the South Coast Air Quality Management District (2014)
estimated the truck trip generation rate at 0.66 truck trip/day/1,000 sqft. Thus a half-
million-square-feet warehouse could bring about as many as 300 truck trips a day to
adjacent communities.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
15

Warehouses and distribution centers have caused growing concerns over the
externalities that affect the surrounding communities (Newman, 2012; Esquivel, 2015).
Among these externalities, air pollution from truck emissions has received most of the
attention. In the State of California, heavy-duty vehicles contribute to 90% of the diesel
PM10, 17% of the SO2, and 53% of the NOx from all on-road emission sources
(National Emissions Inventory, 2014). Figure 1 below shows a clear spatial correlation
between the distribution of warehouses and diesel PM emissions in the Los Angeles
region. This map confirms the notable contribution of warehousing activities, especially
truck activities, to this toxic air contaminant (also refer to Dessouky, Giuliano, and
Moore, 2008). Diesel emissions can cause various health problems. More than 40 diesel
exhaust components including arsenic, benzene, nickel, 1, 3-butadiene, and formaldehyde
are listed by the United States Environmental Protection Agency as hazardous air
pollutants, and many of them can incur cancer risks (US EPA, 2017; OEHHA, 2002).
Studies in the field of public health have examined the relationship between the exposure
to truck-related emissions such as nitrogen oxides, and PM 2.5, and corresponding health
outcomes in areas with high densities of truck footprints (e.g. Bluffstone and Ouderkirk,
2007; Kozawa, Fruin and Winer, 2009; Perez et al., 2009). Apart from air pollution,
trucks also generate noise (Seto et al., 2007; Bai et al., 2009; Dong et al, 2014), pavement
damage (Cidell, 2015), as well as traffic safety concerns (particularly traffic fatalities, see
Chang and Mannering, 1999; Roudsari, et al., 2004; Clifton, Burnier, and Akar, 2009).
For instance, Seto et al., (2007) found that the noise a heavy truck generates is equivalent
to more than 22 automobiles. Other than truck related impacts, warehousing facilities as
large-size buildings with impervious surfaces, substantially affect local built
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
16

environment, exacerbate urban heat island effects (Aniello et al., 1995; Voogt, 2007), and
potentially increase the problems with stormwater capture and related runoff to
surrounding areas (Yang and Li, 2013). Externalities related to warehousing operation
affect the local environment, property values and the quality of life. The fact that local
residents treat warehouses and distribution centers as locally undesirable suggests the
possibility that environmental injustice exists in warehousing location. In spite that a
growing number of studies have documented how disadvantaged communities are
exposed to disproportionate truck traffic and diesel emissions (Lena et al., 2002;
Houston, Krudysz and Winer, 2008; Marshall, Swor, and Nguyen, 2014), none of them
has systematically linked the distribution patterns of warehouses to environmental justice
using empirical data.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
17

Figure 1 Spatial distribution of warehouses and diesel PM emissions in the Los Angeles
region (urbanized areas) (data source: CalEnviroScreen 3.0)
The environmental justice problem emphasizes the disproportionate
environmental burden in disadvantaged communities due to various social, economic and
political factors (Bullard, 1996; Mohai and Saha, 2007; Mohai, Pellow and Roberts,
2009). The majority of research on environmental justice focuses on the siting of locally
undesirable land uses (LULUs) near poor and minority populations. Why are these
people more vulnerable to the environmental burden associated with LULUs? First,
developers of locally undesirable land uses tend to locate their facilities in places with
low land rents and unskilled labor. These places are more likely to be neighborhoods
dominated by poor and minority people. This spatial coincidence could also result from
the housing market dynamics following the siting of LULUs. Moreover, the resulting
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
18

externalities can negatively impact housing values, causing wealthier residents to move
out and poorer residents to move in (Mohai and Bryant, 1992). Second, developers of
LULUs intentionally prefer locations with the least effective political resistance. The
strength of collective voices in a neighborhood largely determines whether locally
unwanted facilities could be sited there easily. Political resources and power are
nonetheless not evenly distributed; people of color or lower socioeconomic classes have
structural disadvantages so that their communities are more likely to receive LULUs.
(Hamilton, 1993; Bullard, 1996; Pastor et al., 2001; Tajik and Minkler, 2006). Third,
racial discrimination, which can be found in the public policy making, land use decisions
and housing dynamics, contributes to environmental inequity as well (Silver, 1997; Cole
and Foster, 2001). The path dependence of racism can partly explain how some
neighborhoods have received significantly more environmental burden than others
(Mohai, Pellow and Roberts, 2009).
Given the environmental externalities generated by warehousing activities,
warehouses and distribution centers are probably locally undesirable. However, they are
different from traditional LULUs including toxic facilities and landfills in several ways.
First, most of the externalities are from truck activities around warehouses instead of
operating activities within the facilities. Trucks move in and out of warehouses, polluting
the air, generating noise, and damaging road pavement in the surrounding neighborhoods
(Aljohani and Thompson, 2016). In spite of some evidence on how designated truck
routes and truck operation time control schemes can help mitigate truck related impacts
(e.g. Anderson, Allen and Browne, 2005; Allen, Thorne and Browne, 2007; Holguí n-
Veras et al., 2011), these impacts remain significant in most neighborhoods with
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
19

warehouses in close vicinity. Second, warehousing facilities can be found in many urban
and suburban locations where a large number of people live nearby, given that local
policy making has not adequately considered their growing environmental impacts.
Traditional locally undesirable land uses such as landfills and toxic facilities nonetheless
have more restricted land use regulations (e.g. buffering) and environmental standards
away from residential areas (e.g. USEPA, 2015). With the expansion of the warehousing
industry, warehousing related impacts extend to more neighborhoods, especially
suburban ones. The exposure of local residents to warehousing related pollution is thus an
increasing concern. Environmental and planning agencies including California Air
Resources Board have been considering the pollution when updating the South Coast Air
District's Air Quality Management Plan and CARB's State Implementation Plan (SIP)
(California Air Resources Board, 2017 ). However, more attention and efforts are needed
to monitor and regulate such pollution, particularly from the perspective of the
governments and public agencies.
Studies on the link between environmental justice and transportation have
emerged in the past two decades. There are three major concerns: the process of
transportation decision-making, the distribution of transportation benefits, and the burden
of transportation costs (Schweitzer, 2004). Compared to other transportation facilities,
warehouses and distribution centers probably generate more costs than benefits for local
residents. Warehousing businesses have lower job densities than many other industries
(Dablanc, 2013), and many of these jobs are low-pay and temporary ones (Kirkham,
2015; Kitroeff, 2016). Only a small proportion of local residents can financially benefit
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
20

from warehouses, yet to a limited degree, while all of them have to suffer from
warehousing related externalities.
Warehouses and distribution centers may create even higher overall
environmental impacts than traditional LULUs and other transportation facilities. To
answer whether environmental inequity exists in the location of warehouses, this paper
examines the spatial relationship between warehousing facilities and socially
disadvantaged communities.
3. Research Approach
Warehouses and distribution centers consume a large amount of land space to move,
store, and process goods. The recent trend of consolidation in the warehousing industry
provides further incentives for the developers to seek for low land rents (Hesse, 2008;
McKinnon, 2009). Warehouse developers may prefer places where political pressure and
resistance is low. Locations with low-skilled labor pools and local policies favoring
warehousing development are also attractive to the developers. Taking all these factors
together, such neighborhoods are more likely to be socially disadvantaged
neighborhoods, normally neighborhoods with high shares of poor or minority
populations. This study aims to test the hypothesis that warehouses are disproportionately
located in disadvantaged neighborhoods, and such neighborhoods are more likely to have
higher warehousing activity intensity.
3.1. Models
Whether a certain neighborhood contains warehousing facilities has a lot to do with the
nature of the neighborhood. This study will focus on the status quo; it will investigate the
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
21

current spatial relationship between warehousing facilities and disadvantaged
neighborhoods. The study assumes that the warehousing location is dependent on various
characteristics of the neighborhood itself. In the conceptual model (see the equation
below), the dependent variable measures the spatial distribution of warehousing facilities
in each neighborhood. The primary independent variables of interest are population
characteristics, including race and socioeconomic status of residents in the geographic
unit. Control variables include transport access, population and employment densities and
other economic attributes of the neighborhood.
𝑌 𝑖 = 𝑓 𝑃 𝐶 𝑖 ,𝐶 𝑉 𝑖

, where Y = warehousing location, PC = population characteristics variables, and CV =
control variables.
The dependent variable, the warehousing location, is measured using binary,
discrete and continuous indicators: whether a neighborhood contains at least one
warehouse, the number of warehouses, and warehouse activity intensity in a
neighborhood. While many previous studies measure the location of LULUs using binary
variables (e.g. Anderton et al., 1994; Goetz and Kemlage, 1996), this set of indicators
describes how warehousing facilities and activities are concentrated in a more
comprehensive way. Population characteristics include the share of minority population
and median household income. The minority population is defined as all non-white
people including African Americans, Asians, Hispanics and so forth. These two
indicators can help identify different types of neighborhoods, including socially
disadvantaged ones. Transport access contains a vector of indicators measuring the
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
22

accessibility of a certain neighborhood to freeway ramps and major freight generators
such as airports, seaports and intermodal facilities. Variables including population density
and employment densities in manufacturing, wholesale and retail industries provide
proxies for land use and industry mix patterns. The patterns imply the preference of local
authorities over different land use types, which can be highly related to the location
choice of warehousing facilities. The model also includes the density of residents
working in the Transportation, warehousing and utilities industries, as well as the median
housing rents. These two variables are reasonable proxies for labor and land costs for
warehousing operation.
3.2. Data sources
The study area is the Los Angeles Combined Statistical Area, which is the second largest
metro and the largest trade gateway in the US. Give the strong demand for freight
movement, the region has a long history of warehousing development. According to
Costar data (2017), the oldest warehouses and distribution centers were built mainly in
Downtown LA, Southeast LA and along the I-5 corridor (see Figure 2). In the recent two
decades, a large proportion of new warehousing facilities were built in the hinterland
instead. The Inland Empire has risen as a major hot spot in the region. The dynamic
changes in warehousing distribution in the LA region provide a good opportunity to study
its spatial relationship with local communities. Another reason for selecting the LA
region as the study area is that there is a relatively comprehensive database for
quantitative analysis. A detailed database of warehousing buildings is provided by the
CoStar Realty Information Inc. The up-to-date database contains information including
but not limited to location, rentable built-up area and year built, from which we know
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
23

how large the facilities are and when they were built. Facilities in the Warehousing and
Storage Industry (based on CoStar’s definition, four relevant subcategories are included:
warehouses, distribution centers, intermodal warehouses and refrigerated warehouses)
with rentable built-up area of at least 30,000 sqft are identified as “warehousing
facilities” in this study. Under this definition, self-storage units or mini warehouses are
excluded from the observations in this study. Population data is from the 2010-2014
American Community Survey 5-Year Estimates, and employment data is from the 2014
Longitudinal Employer-Household Dynamics Workplace Area Characteristics. The two
data sets offer the most recent socioeconomic and demographic indicator estimates.
Federal Aviation Administration (FAA) and National Geospatial-Intelligence Agency
(NGA) offer locations of freight generators in this region. I choose census tract as the
geographic unit for analysis. There are 3,775 census tracts in the urbanized area, which
does not include the census tracts with population and employment density below the
one-tailed 1.96 standard deviations of the mean of the natural log form of the variables
(Giuliano et al., 2015).
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
24

Figure 2 The year of construction of warehousing facilities in the Los Angeles region (data
source: Costar Inc.)
Household income and the share of minorities are the two major independent
variables of interest, and they are found highly correlated in the study region. Low-
income neighborhoods are very likely to be neighborhoods with high shares of
minorities. Such spatial correlation of the two key variables was not taken into
consideration by many previous empirical studies; they include these two variables,
household income and percentage of minorities, in multivariate models without fully
ruling out multicollinearity (e.g. Been and Gupta, 1997; Mohai and Saha, 2006). The
failure of decoupling the correlation may significantly affect the accuracy of estimated
coefficients and the interpretation of model results. Particularly for the region of Los
Angeles where the poverty-minority overlap is evident, further processing of the data to
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
25

distinguish the effects of both variables is needed. In order to disentangle the spatial
covariation of these two indicators, I divide all census tracts into groups by minority
dominance and household income levels. If a census tract has more minority residents
than white residents, it is categorized into the minority-dominant group; otherwise, it
belongs to the white-dominant group. If a census tract has a household income that is at
least one standard deviation below (or above) the overall mean, it is categorized into the
low (or high)-income group. All the other census tracts fall into the medium-income
group. The intersection of these two categorization standards creates six groups: high-
income minority, medium-income minority, low-income minority, high-income white,
medium-income white, and low-income white. Given that the low-income white group
contains less than 1% of the observations, I combine the medium-income white and low-
income white together to generate the reference group. The following analysis will test
whether the other four groups of census tracts would differ from the reference group with
regard to warehousing distribution patterns.
3.3. Data description
There are 5,845 warehousing facilities in the study area and they are highly clustered.
They are concentrated along major freight corridors such as I-5, and I-710, and in the
vicinity of major freight generators including airports, seaports, and intermodal terminals
(see Figure 3). Out of all 3,775 census tracts, 3,131 are without any warehouses, 201 are
with one warehouse, and 106 contain at least ten warehouses.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
26

Figure 3 Distribution of warehouses and freight generators in the Los Angeles region (data
source: Federal Aviation Administration and National Geospatial-Intelligence Agency)
The average distance from all census tracts to the nearest freeway ramp is 2.2 km,
which is rather low given the geographic scale of the region. It demonstrates that the
entire urbanized territory in the LA region has relatively ubiquitous freeway access.
Accessibility to the freight generators nonetheless varies a lot across the zones, given the
limited locations of airports, seaports and intermodal facilities. For instance, the average
distance from all census tracts to the nearest seaport is 54 km, indicating that only a small
proportion of the zones are highly accessible to the seaports.
Table 1 below shows the descriptive statistics of variables in the models.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
27

Table 1 Descriptive statistics of variables
Variables Definition (Unit) Obs. Mean SD
P wh
Whether a census tract contains at least one warehouse
3775 0.17 0.38
Num wh
Number of warehouses in a census tract
3775 1.49 10.29
Density wh
Warehousing density in a census tract (in terms of rentable
building area) (sqft/km
2
)
3775 43,234 179,195
kmHwy Distance to nearest highway ramp (km) 3775 2.19 2.96
kmSea Distance to nearest seaport (km) 3775 53.15 34.84
kmair Distance to nearest airport (km) 3775 23.88 24.28
kmint Distance to nearest intermodal facility (km) 3775 25.33 23.69
Popdensity Population density (person/km
2
) 3775 4110 3787
Manuden Employment density in Manufacturing industry (job/km
2
) 3775 64.00 211.63
Wholeden Employment density in Wholesale industry (job/km
2
) 3775 47.78 162.10
Retailden Employment density in Retail industry (job/km
2
) 3775 134.99 324.67
WHresidensity
Density of residents in Transportation, warehousing and utility
industry (person/km
2
)
3775 91.61 101.08
Medrent Median housing rents (dollars) 3775 1554 562
HighincMinor Whether the census tract is high-income and minority-dominant 3775 0.04 0.19
MedincMinor
Whether the census tract is medium-income and minority-
dominant
3775 0.55 0.50
LowincMinor Whether the census tract is low-income and minority-dominant 3775 0.12 0.32
HighincWhite Whether the census tract is high-income and white-dominant 3775 0.12 0.32

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
28

4. Results
To largely normalize the distribution of variable values, I use the natural log forms of all
variables except for percentages and dummies. According to the existing environmental
justice literature, compared to the reference group, neighborhoods with lower household
income levels and higher percentages of minorities are expected to have higher
probabilities of containing warehousing facilities, all else equal. Figure 4 presents an
overview of the spatial relationship between warehousing facilities and different types of
neighborhoods.

Figure 4 Spatial distribution of warehouses and different types of neighborhoods
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
29

4.1. Model 1
I use the logistic regression method for the Model 1 on the probability of containing at
least one warehouse in a census tract, a dummy dependent variable. The regression
results (see Table 2-Model 1) show that out of four neighborhood groups, the medium-
income minority group and low-income minority group are both found to have
significantly higher probabilities of containing at least one warehouse than the reference
group. The difference between the two coefficients—medium-income minority (1.155)
and low-income minority (1.147)—is nonetheless not statistically significant (Prob.>chi-
square=0.97).
Among transport access variables, only the access to the nearest airport is
marginally significant. Given the growing demand for air freight, warehousing facilities
that are specialized in postal and express delivery services have increased a lot in the
surrounding areas of major airports. The distance to the nearest freeway is not a
statistically significant factor, probably because the entire region has a generally
ubiquitous freeway access. There are two close-by seaports—the Port of Los Angeles and
the Port of Long Beach—in the region and the majority of warehousing facilities are
located far from it. The area next to the seaports can no longer accommodate more
warehousing facilities due to limited land availability. This tendency might partly explain
the positive sign of the coefficient for the distance to the nearest seaport.
All the other variables are statistically significant and with expected signs. The
neighborhoods with higher population densities and employment densities in retail are
less likely to have at least one warehouse, but those with higher employment densities in
manufacturing and wholesaling are otherwise more likely to contain them. These findings
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
30

suggest that warehouses are located close to their customers including manufacturers and
wholesalers while away from areas with high population density and low land
availability. Warehouses are more likely to be in the neighborhoods with higher densities
of residents in the transportation, warehousing and utility industry, and lower median
housing rents. These neighborhoods can provide lower costs of hiring employees and
obtaining land. The overall fit of the model is good given the pseudo R-squared value.
4.2. Model 2
The dependent variable for the Model 2 is the number of warehouses in a census tract and
this indicator offers more details for describing the distribution of warehousing facilities
than the one in the Model 1. I test different econometric regression methods for discrete
dependent variables including Poisson Regression, Negative Binomial Regression, Zero-
inflated Poisson Regression and Zero-inflated Negative Binomial Regression. The
comparison indicates that Zero-inflated Negative Binomial Regression theoretically fits
best given the skewed distributed values of the dependent variable and it provides the
largest log-likelihood value among the methods . The results of the Model 2 slightly
differ from the Model 1 (see Table 2-Model 2). The results show that medium-income
minority and low-income minority groups both have significantly higher numbers of
warehousing facilities relative to the reference group. The magnitude of the coefficient
for the low-income minority group (0.593) is, however, lower than that for the medium-
income minority group (0.759), suggesting an unclear relationship between household
income levels and warehousing distribution.
Distance to nearest freeway ramp is the only transport access variable with an
insignificant coefficient. The coefficient for median housing rents is not statistically
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
31

significant. The signs and significances for other control variables do not differ from the
Model 1.
4.3. Model 3
In the Model 3, I calculate the warehousing activity intensity as the aggregate
warehousing square footage divided by the square kilometer of each census tract’s area.
The warehousing related environmental externalities are closely associated with the
warehousing space used. In contrast to the EJ literature in which most studies did not
consider the size of LULUs, this indicator accounts for the size of the warehouses and
may be the most accurate measurement of the warehousing location. I use the OLS
Regression method for Model 3 on the warehousing activity intensity in a census tract, a
continuous dependent variable. Similar to the two models above, medium-income
minority and low-income minority neighborhoods have higher warehousing activity
intensities than the reference group (see Table 2-Model 3). Meanwhile, the distance to
nearest freeway appears to be significant, indicating that when the model accounts for the
size of warehouses, host neighborhoods are more sensitive to freeway access. The other
results for the Model 3 are similar to those for the Model 1 and 2. Finally, the Model 3
also has a reasonable level of explanatory power.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
32

Table 2 Regression analysis results

Model 1 Model 2 Model 3
Dependent variable
Probability of containing at least one
warehouse
Number of warehouses Warehousing activity intensity

Coefficient S.E Coefficient S.E Coefficient S.E
High-income minority
0.367 (0.410) 0.224 (0.249) 0.153 (0.363)
Medium-income minority
1.155*** (0.208) 0.759*** (0.160) 1.359*** (0.165)
Low-income minority
1.147*** (0.301) 0.593*** (0.219) 1.531*** (0.253)
High-income white
0.437 (0.312) 0.212 (0.234) 0.325 (0.243)
Distance to freeway
-0.095 (0.058) -0.034 (0.040) -0.116** (0.055)
Distance to airport
-0.188* (0.103) -0.166*** (0.063) -0.323*** (0.096)
Distance to seaport
0.259* (0.147) 0.173* (0.090) 0.126 (0.124)
Distance to intermodal terminal
-0.060 (0.093) -0.212*** (0.059) -0.116 (0.086)
Population density
-1.722*** (0.131) -1.571*** (0.078) -1.865*** (0.106)
Manufacturing employment density
0.493*** (0.041) 0.373*** (0.028) 0.728*** (0.037)
Wholesaling employment density
0.757*** (0.059) 0.713*** (0.040) 0.896*** (0.049)
Retail employment density
-0.259*** (0.047) -0.151*** (0.033) -0.339*** (0.040)
WH resident density
0.535*** (0.100) 0.338*** (0.061) 0.486*** (0.083)
Median housing rents
-0.593* (0.311) 0.189 (0.200) -0.941*** (0.274)
Constant
10.347*** (2.662) 5.753*** (1.715) 19.324*** (2.323)
Pseudo R-squared 0.458

Log Likelihood
-2,308
Adjusted R-squared
0.426
Sample Size 3,636
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
33

Putting all the results together, I find a consistent pattern of environmental
inequity in warehousing location across models. First, all regression results confirm that
medium-income minority and low-income minority neighborhoods contain significantly
more warehouses and warehousing activities than the reference group, the medium-
income and low-income white neighborhoods. It is consistent with my hypothesis that
warehouses are disproportionately located in minority-dominant neighborhoods. Second,
low-income minority neighborhoods nonetheless do not have significantly higher
concentration of warehousing facilities or activities than medium-income minority ones.
Compared to the minority dominance, the way the household income level in a
neighborhood affects the warehousing location is less explicit. These results are
consistent with quite a few previous studies (e.g. Pastor et al., 2001; Ringquist, 2005),
which found that environmental inequity is more about race instead of socioeconomic
class.
In general, low-income minority neighborhoods can provide cheaper land, but
why don’t they attract more warehousing facilities than medium-income minority
neighborhoods? It is probably associated with the spatial distribution of low-income
minority population in the Los Angeles region. A high proportion of low-income
minority neighborhoods are located in the old Central and South LA, where residential
density is high and land availability is strictly limited (see Figure 4). These areas are not
among the best choices across the region for developers, especially at the moment
warehousing facilities are growing substantially in sizes and consuming a huge amount of
land. In addition, among other low-income minority neighborhoods, most of them scatter
in the periphery of the region, where the access to customers and transportation
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
34

infrastructure is relatively poor. Therefore, the majority of low-income minority
neighborhoods do not provide certain necessities for warehousing development. In
contrast, it is not difficult to identify that many medium-income neighborhoods have
clusters of warehousing facilities (see Figure 4). For instance, a large number of zones in
the Inland Empire provide conveniences for warehousing development including cheap
land, good regional access, and favorable local policies, and they will probably remain
popular among developers in the near future.
Besides the conditions related to warehousing development, multiple social and
institutional factors could also contribute to environmental injustice. The long-term path
dependence of zoning regulations and high variances in land use policies across
municipalities may greatly affect the distribution of warehousing facilities and local
residents (Dablanc, 2013). Land use regulations and policies on warehousing and relevant
industries like manufacturing have been in effect for many decades. Municipalities have
long-standing strategies in land use and industrial development. These could be an
important reason why some low-income neighborhoods are largely free from
warehousing facilities but nearby medium-income ones have enormous of them. Finally,
warehouses hire blue-collar workers, many of whom live in medium-income
neighborhoods. Proximity to labor pool is another reason that warehousing developers
prefer medium-income neighborhoods to low-income ones. Due to these factors, in the
LA metro, low-income minority neighborhoods do not usually have a higher probability
to be targeted for warehousing development than medium-income minority ones.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
35

5. Conclusion
This study presents a theoretical framework that links warehousing location to
environmental justice. Warehousing facilities are locally undesirable as they generate
increasing environmental impacts on local neighborhoods. I examine the factors that may
affect warehousing distribution and hypothesize that warehousing facilities are
disproportionately located in disadvantaged neighborhoods. I use data for the Los
Angeles region and estimate three models using different indicators of warehousing
location. In spite of these difference measurements, the results across these models are
generally consistent. The results confirm that transport access, land use development and
economic attributes of a certain zone are closely associated with warehousing intensity in
that zone. With all these variables controlled, the models show that warehouses and
distribution centers are disproportionately located in both low-income and medium-
income minority neighborhoods. But low-income minority neighborhoods do not have a
higher concentration of warehousing development than medium-income minority ones.
Overall, the environmental inequity exists in the distribution of warehousing facilities and
the results concur with the conventional understanding of environmental justice (e.g.
Pastor et al., 2001). The distribution of warehousing facilities and activities is highly
related to the percentage of minorities as expected, but its relationship with household
income is mixed. The mixed effects of income may partly result from the nature of the
warehousing industry. In the LA metro, low-income neighborhoods are not always
attractive to warehouse developers as they could not provide adequate conveniences
including land availability, transport access and labor pools for warehousing
development. As warehousing facilities are increasingly moving to suburbs, the findings
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
36

point out a trend of the suburbanization of environmental inequity subsequent to the
suburbanization of poverty and minority population.
Admittedly, the conclusions may be subject to the spatial organization and
demographics of the study area, the Los Angeles region. Like many other American
metros, the Los Angeles region has a strong mix of minority and poor populations,
making it essential to spatially decouple the two factors. Low-income neighborhoods are
primarily concentrated in the historical central locations and the periphery of the region.
This pattern can be easily found in many metros such as Philadelphia and Atlanta. Such
spatial organization makes it difficult for warehousing developers to find desirable
locations in many low-income urban neighborhoods. Meanwhile, I follow the literature
and define the minority as non-white people, while in the Los Angeles region Hispanic
population are increasingly dominant and no longer de facto “minority”. In the recent
studies, Hispanics remained to be considered as politically deprived and their ethnic and
racial identity was still associated with historic marginalization (Pulido, 2000), thus it is
reasonable to assume that they are vulnerable to disproportionate environmental burdens.
In fact, the spatial distribution of warehousing facilities clearly shows many hotspots in
Hispanic dominant cities such as La Puente, Commerce, and San Fernando. Finally, the
Los Angeles region functions as a national trade gateway and the volumes of goods
coming into and out of the region supply the whole nation. The demand-driven expansion
of the warehousing industry could essentially lead to more significant and prevalent
environmental justice inequity across the region. More evidence from other metropolitan
areas, especially those with different demographic patterns and roles in international
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
37

trade, would help to further justify the conclusions. I will analyze data for other metros to
test whether the findings still hold without the assumptions above.
Although this paper shows clear evidence of the environmental justice problem in
warehousing location, mechanisms behind the findings are largely unknown. For
instance, the study does not answer whether the co-location of warehouses and minority
population is attributable to the disproportionate siting of warehouses or the movement of
the minorities subsequent to the siting process. This interdependence issue can potentially
be addressed through the Simultaneous Equation Model using longitudinal data. On the
other hand, the quantitative tests in this paper do not incorporate the institutional factors
since these factors are difficult to quantify at the neighborhood level. Local public
policies including zoning ordinances, tax incentives, and industrial strategies may play a
vital role in the spatial distribution of warehouses, and thus of the warehousing related
externalities. Further qualitative data would help explain why residents in some cities
suffer from more severe environmental burden than those in other cities.
This study examines environmental justice from a new perspective of
warehousing distribution. It can provide the governments and planners a general
overview of the disproportionate distribution of warehousing related externalities. State
governments, and regional planning agencies, which have research capacities and
frameworks for regional collaboration, could monitor the spatial distribution of
environmental impacts associated with warehouses, and provide guidance to local
authorities on how to mitigate these impacts. County and city governments, on the other
hand, have much stronger influences on the warehousing location choice. Through land
use, building and environmental regulations, these governments may effectively attract
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
38

warehousing development or on the contrary, push such development to neighbors. To
simultaneously achieve sustainable industrial growth and maintain a just environment for
residents, the governments have to develop policy packages in line with their
backgrounds and long-term visions.
With the information revealed in this study, the disadvantaged population would
become more aware of environmental inequity they could encounter. Environmental
advocates may consider including warehousing facilities in the examination of
environmental justice and help the disadvantaged people avoid disparate environmental
burdens. If local residents are well organized to fight for justice, many neighborhoods
probably would not suffer from negative environmental and social consequences due to
massive warehousing development.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
39

CHAPTER 3
Mega Freight Generators in My Backyard: A Longitudinal
Study of Environmental Justice in Warehousing Location

Abstract
Environmental impacts of warehousing activities have attracted increasing attention from
the governments, the public and the academia. While a few studies have confirmed the
cross-sectional spatial coincidence between warehousing facilities and minority
population, little is known about the causal relationship behind the co-location pattern.
Using data of the Los Angeles Combined Statistical Area in 2000-2010, this paper
estimates a two-equation simultaneous model of the location choices of warehousing
facilities and the minority population. Results show that, all else equal, changes in the
percentage share of minorities significantly and positively affect the changes in
warehousing activity density during the same period, but not vice versa. Thus, the
environmental justice problem in warehousing location is found to be solely from the
disproportionate siting of warehouses in minority-dominated areas, rather than from the
movement of minority population towards warehousing. Furthermore, the variants of the
model suggest that contrary to Latinos and Asians, the inflow of Blacks into a
neighborhood would not lead to an increase in warehousing activities. And a
neighborhood with more ethnic churning would be more likely under the pressure of
warehousing expansion. The government and the public need to work together to
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
40

facilitate an effective regulation of warehousing related externalities and a fair
distribution of related environmental disamenities.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
41

1. Introduction
With the explosion of global trade, people and firms in the world today are highly
interconnected regardless of physical distance. After one clicks on the “Place Order”
button at home, the items the customer just bought will possibly be sent out from a
warehouse located thousands of miles away. The rise of E-commerce further stimulates
the demand for goods movement and drives the proliferation of warehousing facilities.
Warehouses and distribution centers (W&Ds) have been extensively developed in the
major metro areas, especially in those regional trade gateways. These facilities consume
large tracts of land, attract high volumes of truck movement, and greatly affect the built
environment. As local residents become increasingly aware of the environmental
externalities generated by logistics activities, the environmental justice (EJ) problem in
warehousing location arises. Warehouses are found disproportionately located in
communities dominated by minorities in a few recent quantitative studies (e.g. Author,
2018), but the research on this topic is still very limited, and the mechanisms behind such
environmental injustice are largely unknown and untested. Longitudinal research, in
particular, could help improve our understanding of environmental disparities related to
warehousing location and further provide policy implications for mitigating these
disparities.
Why is longitudinal analysis important? Although the literature has extensively
discussed the causes of environmental inequity, more longitudinal analyses are needed to
track the dynamics that result in the inequity and justify these causes, especially in the
case of warehousing location. As Mohai and Saha (2015a) argued, much of the EJ
research tests the existence of environmental disparities using cross-sectional data, but
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
42

lacks a careful examination of the processes by which the disparities are created.
Specifically, most existing studies focus on whether locally undesirable land uses
(LULUs) are disproportionately located in minority or poor neighborhoods at a certain
point in time. They nonetheless fail to show how the hypothesized socioeconomic
dynamics contribute to such co-location patterns. Therefore, cross-sectional snapshots
may help identify the spatial imbalance of environmental burdens, but provide no
answers to the classic “which came first” question. First raised in the 1990s (Hamilton,
1993; Been, 1994), this question asks whether LULUs are sited before people of color or
poor population reside in the same neighborhoods or after that. By testing the sequence
of these two processes, researchers would be able to understand the behaviors of the
agents involved and identify who or what social structures are responsible for the
problem.
This study aims to explore how socioeconomic dynamics may result in the
coincident relationship between warehousing facilities and minority population using
longitudinal data. The Los Angeles region, where a massive growth in the logistics
industry has been occurring for decades, is selected as the study area. The period of the
Year 2000-2010, when globalization drove another wave of warehousing expansion, is
selected as the observation period. Using the Simultaneous Equation Model, this study
shows how different factors are involved in the location choices of both W&Ds and
people of color, and on top of that, whether W&Ds follow people of color or vice versa.
Results reveal that the disproportionate siting of warehouses in minority neighborhoods is
the only dominant process that leads to the environmental inequity. The findings verify
the existence of environmental justice problem in warehousing location, and further
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
43

suggest that the disparities are more related to unregulated logistics expansion, rather
than the housing choice of vulnerable population groups.
The paper is organized as follows. Section 2 contains a literature review of recent
research progress on relevant topics. Section 3 covers research approach and
methodology. Data is described in Section 4, results are presented and discussed in
Section 5, and conclusions are included in the final section.
2. Literature Review
2.1. Contradictions in longitudinal studies on environmental justice
Before examining the environmental justice problem in warehousing location, I chose to
conduct a careful review of the broader literature on environmental justice to see how
longitudinal studies have helped understand the problem. The literature has extensively
discussed the three categories—economic, sociopolitical and racial—of explanations for
environmental inequity (see review papers: Mohai, Pellow and Roberts, 2009; Mohai and
Saha, 2015b). For example, the path dependence of zoning regulations (Cole and Foster,
2001), racial discrimination in the housing market (Bullard, Grigsby and Lee, 1994), or
simply the low land rent (Ringquist, 2003) all possibly contribute to the spatial disparities
in environmental impacts. The explanations are closely linked to multiple socioeconomic
processes including firm location choice, housing location choice, community collective
actions, and public policy making. These processes demonstrate why environmental
burdens may be disproportionately placed in certain neighborhoods. While the spatial
relationship between LULUs and different types of communities, as a status quo, has
been explicitly identified in the previous studies (see Ringquist, 2005), it is nonetheless
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
44

more difficult to distinguish the roles of those various socioeconomic processes when
they are likely to be interdependent.
To address this endogeneity problem, researchers have more frequently used
longitudinal data in the environmental justice research. Indeed, longitudinal data may be
used to observe the behaviors of relevant agents in the socioeconomic processes and
further justify the theoretical explanations. However, the findings in many longitudinal
studies are nonetheless contradictory, suggesting the necessity to systematically evaluate
the contexts, research design, and data sources in those studies. Such an evaluation is
particularly helpful for developing a reliable methodology to assess the environmental
disparities in warehousing activities across places.
Locally undesirable land uses impose substantial environmental impacts on host
neighborhoods. The siting of LULUs is found to follow the “path of least resistance” and
would probably end up in neighborhoods where the opposition to the siting is weak
(Taylor, 1992; Bullard and Wright, 1993; Bullard, 2000). The political power to avoid
LULUs is usually subject to the racial composition or the socioeconomic status, so the
host communities are more likely to be low-income or minority-dominated ones (Mohai,
Pellow and Roberts, 2009). The increasing environmental burden on those vulnerable
communities thus may be ascribed to both the spatial expansion of environmental
externalities and the disparities in political and organizational resources. As Saha and
Mohai (2005) pointed out, the behaviors of relevant agents in these dynamics including
facility developers and public policy makers have been changing over time. The
environmental justice patterns are therefore highly subject to the historical context of
siting. For instance, the patterns of disparate siting of hazardous waste facilities in
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
45

Michigan were only significant after 1970, at which time uneasiness about these facilities
began to arise (Saha and Mohai, 2005).
Moreover, it could be difficult to statistically recognize environmental disparities
when the siting of the LULUs is a “rare event”. Most quantitative analyses require
statistical interpretation of data, and the verification of hypotheses largely relies on the
statistical significance of coefficients. However, even if real disparities exist in the spatial
distribution of LULUs, such disparities may not be of statistical significance when the
observed siting of locally undesirable facilities is very infrequent. For example, a study of
Oakes et al. (1996), among many others (e.g. Hamilton, 1993; Anderson et al., 1994;
Pastor et al., 2001), examined environmental inequity related to the location of
Hazardous Waste Treatment, Storage and Disposal Facilities (TSDFs). In this nationwide
study, out of all 35681 tracts, there are only 473 tracts that contain waste facilities. With
over 98% of the observations receiving a value of “0” in the dependent variable (DV) of
the logistics model, it is methodologically unlikely for the authors to statistically identify
the existence of environmental inequity in this case.
Apart from these considerations on research contexts, researchers use diversified
methods in the longitudinal studies and the various ways they understand and manipulate
data may also lead to contradictory conclusions. First, statistical results are sensitive to
the choice of geographic scope. Compared to the nationwide longitudinal studies,
regionwide ones generally provide more support for the existence of environmental
disparities (Mohai and Saha, 2015a). Why do they generate different results? Nationwide
data hardly covers detailed localized factors such as zoning and land use regulations. The
failure to control for these factors might undermine the validity of model estimates.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
46

Studies focusing on a single region, however, may have more methodological choices to
address such problems. For example, dummies that represent detailed jurisdictional
variances can help control localized factors to a large degree. Second, the estimated
relationship between environmental hazards and population groups is equally subject to
spatial units researchers choose for analysis. According to Mohai and Saha (2007,
2015b), the distance-based method has been increasingly prevalent, as it measures this
relationship more accurately than the unit-hazard coincidence method. The distance-
based method effectively accounts for all spatial units (e.g. census tract) within a certain
distance to an undesirable facility, while the unit-hazard coincidence method may not
sufficiently estimate the impacts of the LULU, especially when the facility is located near
the boundary of the unit. Third, some of the recent studies tested both location choice
hypothesis and the demographic change hypothesis, but few of them considered the
interrelationship between both hypotheses. In these studies (Oakes et al., 1996; Been and
Gupta, 1997; Shaiku and Loomis, 1999), the changes in the distribution of LULUs and
the changes in the concentration of minority or poor population were estimated in
separate models, and thus their potential mutual effects are ignored. Pastor et al. (2001),
on the contrary, emphasized the necessity to account for these effects. All of these
distinctions in research design contribute to the inconsistent results in the literature. Thus
a carefully designed longitudinal analysis that considers the advantages and
disadvantages of various research settings and methods would be helpful for clarifying
the confusing contradictions.
A growing number of studies have examined the environmental threats to local
population from extensive warehousing development. Widespread warehousing
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
47

structures with large flat roofs can contribute to strong urban heat island effects (Voogt,
2007) and risks of stormwater runoff (Yang and Li, 2013). The frequent truck movement
to and from warehouses generates substantial air pollutants (Dablanc, 2013), noise (Dong
et al, 2014), and pavement damage (Cidell, 2015). Subsequent health outcomes such as
asthma and respiratory allergy have been found in various cities with intensive truck
activities (Kim et al., 2004; Kozawa, et al., 2009). More and more residents have become
aware of these environmental threats (Newman, 2012; Esquivel, 2015). Compared to
traditional LULUs, W&Ds are different in several aspects, which may significantly affect
the empirical testing of the EJ problem in warehousing location. First, warehousing
establishments are more spatially widespread, yet more clustered than toxic facilities and
waste landfills, which have been extensively studied in environmental justice research.
Second, due to zoning regulations, warehouses are generally located closer to residential
areas than many locally undesirable facilities, so their location choice may be more
intimately related to the local housing market and demographic changes. It would be
easier to observe these changes, if any, in statistical models. In other words, if the
interrelationship between warehousing siting and the inflow of minority population
exists, the spatial data of warehousing facilities provides adequate conveniences for
identifying it. Third, in contrary to many traditional LULUs, warehousing facilities are
more footloose as they require lower investment in infrastructure and land use permits,
making their location choice more responsive to socioeconomic factors such as land rent
and industrial connections. The above three differences are well linked to the last yet the
most crucial one: W&Ds are less regulated and monitored in the current land use
planning system, given the conventional views on their environmental consequences.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
48

Given the increasing evidence on the warehousing related externalities (e.g. McKinnon,
2010), a careful reassessment of the planning system is needed.
2.2. Location choices of warehousing facilities and minority people
Given the spreading environmental impacts, the warehousing location has caused
increasing concerns about its environmental justice implications. While there is yet
limited understanding of how such a problem emerges, the broader literature of
environmental justice (e.g. Mohai, Pellow and Roberts, 2009) suggests a start from the
behaviors of related agents: warehousing developers and minority people. It would be
helpful to examine the factors in the location choice of both warehousing facilities and
minority households. Warehousing location choice is not only still subject to traditional
industrial location factors including proximity to transportation infrastructure and skilled
labor force, but also more and more affected by various changes in supply chain
management, logistics services and urban structure (Jakubicek, 2010; Author, 2018). In
major gateway cities, suburban areas with cheap land, good transport access to regional
markets, and favorable political and institutional environment become increasingly
popular among logistics developers (Giuliano et al., 2016). Among these factors, race and
socioeconomic status are much less studied, but they are of importance in two ways. On
one hand, warehousing developers would find it less expensive and time-consuming
when siting their facilities in neighborhoods dominated by minority and poor residents,
who have limited ability to collectively resist LULUs. On the other hand, local authorities
and political leaders in these neighborhoods may be more willing to promote
warehousing as a development strategy, especially when the cities are in a financial crisis
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
49

(Husing, 2010). Thus the dynamic changes in the demographics of a neighborhood could
be a noteworthy factor in the warehousing location choice.
The housing location choice is affected by employment opportunities, commuting
costs, housing rents, local amenities and environment, and community preferences (e.g.
Weisbrod, Lerman and Ben-Akiva, 1980; Bayoh, Irwin and Haab, 2006; Schirmer, van
Eggermond and Axhausen, 2014). Comparably, the minority population nonetheless has
narrower choices than the Whites, given their disadvantages in the housing market.
Warehouses are associated with high volumes of truck flows, which contribute to air
pollution, traffic congestion and pavement damage (e.g. Dessouky, Giuliano, and Moore,
2008; Dablanc, 2013; Cidell, 2015), and thus lead to the degradation of the local
environment. As these negative impacts are capitalized into land values, housing rent
drops and the host communities may subsequently attract more minority households.
Therefore, the firm location choice of warehousing facilities and housing location
choice of minority households are likely to be mutually dependent and causally related.
There is, however, no research that has tested the interdependent relationships so far.
Such interdependent relationships may even differ across minority groups. Would Black,
Asian, and Latino population bear different levels of environmental hazards? The
literature has presented positive answers to this question, although results are not fully
consistent across existing studies. Much research showed that the Black population was
more likely than Latinos and Asians to live in environmentally hazardous neighborhoods
(e.g. Alba et al., 1999; Crowder and Downey, 2010). Despite this traditional view, quite a
few recent studies nonetheless denied the particular vulnerability of the Blacks. For
instance, Bullard et al., (2008) found that percentages of African Americans,
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
50

Hispanics/Latinos, and Asians/Pacific Islanders in host neighborhoods of TSDFs were
1.7, 2.3, and 1.8 times greater than non-host ones. Hipp and Lakon (2010) concluded the
disparities in toxic air emissions exposures were highest for Hispanics, followed by
Asians and then by African Americans after analyzing data between 1990 and 2000.
Mohai and Saha (2015) examined the data of TSDFs between 1991 and 1995 and
demonstrated increasing vulnerability of Asians to TSDF siting and decreasing
vulnerability of Blacks and Hispanics in that time period. Given the mixed results, the
inter-race differences in exposure to environmental hazards may be largely subject to
research contexts and more efforts are needed to clarify these differences. On the other
hand, do African Americans, Latinos and Asians react to environmental hazards
differently? Compared to the other two minority groups, Asians were found to have a
higher likelihood of leaving highly polluted neighborhoods due to their relatively higher
levels of education and household income (Crowder and Downey, 2010). Whether this
contrast in reactions between minority groups can be applied to warehouses is, however,
unknown yet.
3. Research Approach
This article aims to investigate how the environmental justice problem in warehousing
location occurs, using longitudinal data and statistical models that estimate
interdependent relationships. Given what I have found in the literature review, I would
like to highlight several features of the research settings. First, I hypothesize that the
location choice process and the demographic change process are interdependent, and use
the Simultaneous Equation Model (SEM) to test the interrelationship. While the two
processes are estimated in equations with separate sets of determinant factors controlled,
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
51

the dependent variable in one equation essentially reacts to that in the other equation. For
example, the location choice of warehouses is not only subject to variables such as
transport access, but also likely to be affected by the demographic changes during the
same period of time. A simultaneous equation model is estimated to identify the latter
effects, which have been ignored in most of the previous studies.
Second, I select the Los Angeles Combined Statistical Area (LA CSA) as my
study area, in contrast with previous EJ studies that covered multiple metro areas. The LA
CSA is located in Southern California and contains five counties including Los Angeles
County, Orange County, Riverside County, San Bernardino County, and Ventura County.
The five-county metropolitan area functions as a single housing and industrial real estate
market. By confining my analysis to the region, I could reduce the variances in local
circumstances that affect the location choices and thus effectively diminish the biases that
result from uncontrolled variances. Meanwhile, the LA CSA has a long history of
warehousing development; the oldest batch of warehouses in the region was built in the
city of Los Angeles in the 1910s (CoStar, 2016) when the population of the city was only
about 320,000. The region is also famous for its demographic diversity; White, Black,
Latino and Asian respectively account for 32%, 6%, 46% and 13% of the total
population. Each of these races plays a substantial role in making LA a demographically
diverse region. Given the warehousing development tradition and mixture of multiple
races, the LA CSA is perfect for observing the location choices of warehousing facilities
as well as groups of people in different races.
Third, I observe the period from the Year 2000 to 2010. This period witnessed a
substantial expansion of warehousing facilities in the LA CSA, especially in the suburban
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
52

areas like Inland Empire (the cities of western Riverside County and southwestern San
Bernardino County). As revealed in the literature review, different choices of the
observation period may result in contradictory outcomes in the longitudinal analysis. The
first decade of this century witnessed the intensification of globalization. In response to
the growing freight demand, the LA CSA, as the biggest trade gateway and a major
consumption market in North America, experienced a massive spatial expansion in the
warehousing industry during the period (see Figure 5). Simultaneously, local residents
had begun to realize the spreading environmental threats associated with the siting of
warehouses, although people had not been organized to fight against it until recently
(Newman, 2012; Esquivel, 2015). The intensification of warehousing related
environmental externalities and different reactions to these externalities across race
groups during the observation period might result in environmental inequity. Therefore,
the period 2000-2010 could be a good candidate for observing and understanding
environmental justice in warehousing location.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
53

Figure 5 Year built of existing warehousing facilities in the Los Angeles region (data source:
Costar Inc.)
Fourth, I account for the size of a warehouse, for instance, in terms of rentable
square footage, in measuring the spatial distribution of warehouses. The capacity that a
warehousing facility stores and processes goods is closely associated with the truck
traffic and land use impacts the facility generates. With regard to the existing EJ studies,
most of them failed to account for the scale of activities in LULUs, probably due to data
unavailability.
Finally, the environmental impacts associated with warehousing facilities are
easily extended from the host neighborhoods to nearby neighborhoods via the movement
of trucks. It is therefore necessary to link each neighborhood to all adjacent warehouses
within a certain distance, not limited to those within its boundary, in the spatial analysis. I
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
54

chose census tract as my basic unit of analysis and treated all tracts that spatially intersect
with a half-mile buffer of each warehouse as the area environmentally impacted by the
facility. Reversely, a warehouse linked to one tract in this way was regarded as an
“adjacent warehouse” to that tract (see the Figure 6). In dense urban areas, a half-mile
buffer zone could cover as many as ten tracts, while in suburban areas, it may include the
host tract only. Such a distance-based definition could help avoid the possible spatial bias
when using the unit-coincidence method, especially in the case that a warehouse is
located close to the boundary of the host tract.

Figure 6 Illustration of the spatial relationship between a specific census tract and its
adjacent warehouses
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
55

With the research settings included, the general conceptual model is a
Simultaneous Equation Model of firm location choice of warehousing facilities and
housing location choice of the minority population. According to Pastor et al. (2001) and
Mohai and Saha (2007), minorities are likely to be more vulnerable to environmental
inequity than low-income people. Thus in this paper, the minority population is selected
as the representative group of the vulnerable population. The general model (see
Equations 1 and 2) is,
{
∆𝑊𝐻
𝑖 = 𝑓 (∆𝑀𝑖𝑛𝑜𝑟 𝑖 , 𝐶𝑉
1
𝑖 ) 𝐸𝑞𝑢𝑎𝑡𝑖𝑜𝑛 1
∆𝑀𝑖𝑛𝑜𝑟 𝑖 = 𝑓 (∆𝑊𝐻
𝑖 , 𝐶𝑉
2
𝑖 ) 𝐸𝑞𝑢𝑎𝑡𝑖𝑜𝑛 2

, where ΔWH = changes in warehousing activity density, Δ Minor = changes in
percentage of minorities in the entire population, CV1 = control variables in firm location
choice equation, and CV2 = control variables in housing location choice equation.
The selection of control variables is key to obtaining accurate estimates of the
interdependent relationships. According to the literature, I expect new warehousing
facilities would concentrate in places with low household income, land rent and
population densities, as well as high employment densities in the wholesale industry. I
intentionally select the employment density in the wholesale industry (instead of other
industries) as a control variable because wholesalers on one hand consume warehousing
services and likely to be spatially correlated with warehouses, but on the other hand, they,
in theory, do not generate activities that significantly affect the movement of the minority
population. Comparatively, other customers of warehouses such as retailers and
manufacturers may nonetheless be positively or negatively related to the inflow of
minorities into a neighborhood. The employment density of wholesalers can, therefore,
better serve as a valid exogenous variable, which is essential for solving the SEM.
Warehouses are also expected to locate near intermodal terminals to more conveniently
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
56

access their services. While I control for the pre-existing levels (in the Year 2000) of
warehousing activity density in the Equation 1, I include a quadratic term for
warehousing activity density in 2000 as well. The curvilinear relationship is possible
because of intra-industry competition and agglomeration benefits. Clustering W&Ds
compete for land, customers and labor force, and at the same time, they can reduce costs
by sharing infrastructure and resources such as freight handling equipment (McKinnon,
2009; Sheffi, 2010). Given that land availability varies tremendously across the five
counties in the Los Angeles region, dummies for each county are incorporated into the
equation to control for that heterogeneity. Finally, local public policies are nonetheless
not accounted for due to the difficulty to accurately quantifying such variations across
census tracts. In the Equation 1, employment density in wholesale industry and distance
to the nearest intermodal terminal are exogenous variables (or predetermined variables,
which are theoretically not correlated with the dependent variable in the Equation 2, i.e.
the changes in percentage of minorities). These two variables can be used as instruments
for the DV in the Equation 1.
In the Equation 2, household income, housing values, distance to nearest
employment center and density of park space are among the major control variables. Due
to various constraints in searching for desirable housing, the minority population is likely
to concentrate in neighborhoods with low household income and housing values, and
poor access to job opportunities and green space. Similar to the Equation 1, the pre-
existing levels of warehousing activity density, and the quadratic terms of household
income and percentage of minorities in 2000 are controlled. While minorities are less
competitive in bidding for housing in white-dominant and rich neighborhoods, they may
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
57

still prefer not to reside in communities in extreme poverty or with very high percentages
of minorities (Jarvis, 2015). These constraints and preferences suggest possible quadratic
relationships in the equation. As counties have generally different patterns of racial
composition, dummies are included in the Equation 2 as well. Distance to employment
center and density of park space are instruments for the DV in the Equation 4, given they
are theoretically uncorrelated with changes in warehousing activity density.
Simultaneous equations 3 and 4 display the complete form of the SEM. Both
dependent variables are measured as changes in the corresponding indicators from 2000
to 2010. All control variables are measured as values at the beginning of the study period,
or Year 2000, except for the time-invariant ones (county dummies). Except for dummies,
percentages and index variables, all other variables are transformed into the natural
logarithm form.
{

∆WH = 𝛼 0
+ 𝛼 1
𝑊𝐻
𝑡 −1
+ 𝛼 2
∆𝑀𝑖𝑛𝑜𝑟 + 𝛼 3
𝑀𝑖𝑛𝑜𝑟 𝑡 −1
+𝛼 4
𝐼𝑛𝑐𝑜𝑚𝑒 𝑡 −1
+𝛼 5
𝐼𝑛𝑑𝑢𝑠𝑡𝑟𝑦 𝑡 −1
+ 𝛼 6
𝐼𝑛𝑡𝑒𝑟𝑚𝑜𝑑𝑎𝑙 𝑡 −1
+ 𝜀 ⋯⋯⋯⋯⋯𝐸𝑞𝑢𝑎𝑡𝑖𝑜𝑛 3
∆Minor = 𝛽 0
+ 𝛽 1
𝑀𝑖𝑛𝑜𝑟 𝑡 −1
+ 𝛽 2
∆𝑊𝐻 + 𝛽 3
𝑊𝐻
𝑡 −1
+𝛽 4
𝐼𝑛𝑐𝑜𝑚𝑒 𝑡 −1
+𝛽 5
𝑃𝑎𝑟𝑘 𝑡 −1
+𝛽 6
𝐸𝑚𝑝𝐶𝑒𝑛𝑡𝑒𝑟 𝑡 −1
+ 𝜔 ⋯⋯⋯⋯⋯⋯𝐸𝑞𝑢𝑎𝑡𝑖𝑜𝑛 4

, where WH= warehousing activity density, Minor= percentage of minorities in the entire
population, Income=household income, Industry=industrial connection variables,
Intermodal=access to the nearest intermodal terminal, Park=density of park space,
EmpCenter=access to the nearest employment center, t stands for time, and ε and ω are
estimation errors.
The expected relationships between the dependent variable and control variables
in each equation are listed in Table 3 below.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
58

Table 3 Expected relationship between the dependent variable and control variables in each
equation of the model
Equation 3: firm location choice of warehousing facilities
Dependent variable
Explanatory variables Changes in warehousing activity density
Changes in percentage of minorities +
Pre-period warehousing activity density curvilinear
Pre-period percentage of minorities +
Pre-period household income curvilinear
Pre-period land rent -
Pre-period population density -
Pre-period employment density in
wholesaling
+
Pre-period distance to the nearest
intermodal terminal
-
Equation 4: housing location choice of minority population
Dependent variable
Explanatory variables Changes in percentage of minorities
Changes in warehousing activity density +
Pre-period percentage of minorities curvilinear
Pre-period household income curvilinear
Pre-period warehousing activity density either
Pre-period land rent -
Pre-period population density +
Pre-period density of park space -
Pre-period distance to the nearest
employment center
+

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
59

4. Data
The five counties in the LA CSA have been developed to different degrees. Apart from
mountainous areas, Los Angeles County and Orange County are largely developed, while
the other three counties still have large tracts of suburban land undeveloped. The LA
CSA is the second largest CSA and the largest trade gateway in the USA. The region has
long been popular among warehousing developers due to its significant role in goods
import/export, production, and consumption during the last decades. The recent trend of
warehousing development shows a moderate growth in the Gateway Cities (a region
located in southeastern Los Angeles County between the City of Los Angeles, Orange
County, and the Pacific Ocean) and a dramatic expansion in the Inland Empire (Giuliano,
et al., 2016) (see the Figure 5). A detailed database of warehousing buildings is provided
by the CoStar Realty Information Inc. Originally retrieved from real estate listings, the
data does not contain those facilities that disappeared during the study period. It is
possible that the warehousing activity density could be overestimated. But after
comparing the data with the Zipcode Business Patterns (ZBP), I found the
overestimation, if any, was only marginal and would hardly affect the results. The
database contains information including but not limited to geospatial location, rentable
built-up area and year built, from which we know how large the facilities are and when
they were built. The geocoded data confirmed that the newest generation of warehouses
was mostly in suburbs of the region, especially in the Southwest San Bernardino County.
Population data is from the Census 2000 100-Percent data and Sample data, the Census
2010 100-Percent data, and the 2006-2010 American Community Survey 5-Year
Estimates. Employment data are from the 2010 Longitudinal Employer Household
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
60

Dynamics (LEHD) Workplace Area Characteristics and the 2000 Southern California
Association of Governments (SCAG) employment estimates. The California Department
of Transportation (Caltrans) offers the locations of intermodal terminals.
The observation units in the model are census tracts. There are 3,775 census tracts
in the LA CSA’s urban area in 2010, which does not include the census tracts with
population and employment density below the one-tailed 1.96 standard deviations of the
mean of the natural log form of the variables (Giuliano, et al., 2017). After eliminating
those tracts with no values for some variables, a total of 3,674 census tracts were put into
the Simultaneous Equation Model. All variable values collected for the Year 2000 were
recalculated into the 2010 tract shapes, to eliminate any data mismatch due to the changes
in tract boundaries during the ten-year period. The conversion between 2000 and 2010
tract shapes was conducted by aerial apportioning.
For each observation unit in the model, the measurement of several variables
needs further clarification. Given the aforementioned method to determine the
relationship between each tract and its “adjacent warehouses”, the half-mile searching
ended up with 1641 tracts, or 44% with at least one “adjacent warehouse” in 2000. On
top of that, warehousing activity density was measured as aggregate rentable built-up
areas of all “adjacent warehouses” divided by the square kilometers of the census tract
area. As a continuous variable, warehousing activity density accounts for the sizes of
warehouses and functions as an adequate indicator for warehousing concentration and
subsequent environmental impacts. Percentage of minorities was measured as non-white
population (including mainly Latino, African-American and Asian people) divided by all
population. Median housing values were used as a proxy for land costs for warehousing
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
61

development. The density of park space was defined as aggregate park space within a
census tract divided by the land area of the tract. Distance to the nearest employment
center was measured based on employment centers that were identified using data in
2000 and methods in Giuliano, et al. (1991).
Table 4 contains the descriptive statistics of all variables in the model.
Table 4 Descriptive statistics of variables
Variable Obs. Mean Std. Dev.
Changes in warehousing activity density during 2000-2010 (square
feet/square km)
3674 47,999 184,630
Warehousing activity density in 2000 (square feet/square km) 3674 286,672 798,088
Changes in percentage of minorities during 2000-2010 (%) 3674 3.6 7.9
Percentage of minorities in 2000 (%) 3674 60.4 28.3
Changes in percentage of Latino during 2000-2010 (%) 3674 5.5 8.6
Percentage of Latino in 2000 (%) 3674 39.7 27.4
Changes in percentage of Black during 2000-2010 (%) 3674 -0.4 4.0
Percentage of Black in 2000 (%) 3674 7.6 12.8
Changes in percentage of Asian during 2000-2010 (%) 3674 2.3 4.9
Percentage of Asian in 2000 (%) 3674 10.3 12.3
Median household income in 2000 (dollars) 3674 43,477 25,957
Median housing values in 2000 (dollars) 3674 201,973 146,713
Population density in 2000 (person/sqkm) 3674 3,917 3,620
Employment density in wholesaling in 2000 (employee/sqkm) 3674 75 266
Distance to nearest intermodal terminal (km) 3674 25.3 23.4
Density of park space in 2000 3674 0.018 0.054
Distance to nearest employment center in 2000 (km) 3674 57.7 69.4

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
62

5. Results
5.1. Summary statistics
Table 5 shows the differences in the values of selected variables between tracts with at
least one adjacent warehouse and those without, in 2000 and 2010 respectively. From the
table, we can find that in 2000, the tracts with at least one adjacent warehouse were
significantly different from others, especially in the percentage of minorities, median
household income, and median housing values. Such differences in 2010 were not as
great as in 2000. For instance, compared to all the 3,674 tracts, those tracts with at least
one adjacent warehouse in 2000, on average, had about 31% lower median housing
values. That difference dropped to 21% in 2010.
Table 5 Comparing tracts with and without adjacent W&Ds in the Los Angeles region, in
2000 and 2010 respectively
Variable
2000 2010
Average
With adjacent W&Ds
(difference)
Average
With adjacent W&Ds
(difference)
% Minority
60.4 20.3*** 64.0 19.8***
% Latino
39.7 19.3*** 45.3 19.3***
% Black
7.6 1.0*** 7.2 0.5
% Asian
10.3 0.4 12.6 0.3
Household income
43,477 -8,427*** 63,693 -12,579***
Housing values
201,973 -62,658*** 483,430 -103,405***
Population density
3,917 29 3,800 -106
Number of obs. 3,674 1,615 3,674 1,674

Table 6 shows the values in selected variables between the tracts that received at
least one adjacent warehouse in 2000-2010 and other tracts. Compared to other tracts,
those tracts that received adjacent warehouses during the period had higher percentages
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
63

of minority (mainly Latinos), lower median household income and housing values, and
higher population density and employment density in wholesaling. All these differences
were statistically significant at 1% level. Therefore, the tracts receiving adjacent W&Ds,
in various aspects, were significantly different from others prior to the siting of
warehouses. Particularly, these tracts had much higher percentages of minorities in 2000.
Table 6 Comparing values in selected variables in 2000 between tracts that received adjacent
W&Ds during 2000-2010 and other tracts in the Los Angeles region
Variable
Values in 2000
Average
Receive adjacent W&Ds vs.
others
(difference)
% Minority
60.5 14.1***
% Latino
39.7 14.7***
% Black
7.6 0.1
% Asian
10.3 0.3
Household income
43,477 -6,164***
Housing values
201,973 -62,917***
Population density
3,917 1,015***
Number of obs. 3,674 656
On the other hand, the characteristics of neighborhoods may change following the
siting of warehouses (see Table 7). I observed tracts with adjacent W&Ds in 2000,
benchmarking against those without. Changes in the percentage of minorities as a whole
between these two groups of tracts are not significantly different, suggesting that there
might not be “move-in” effects after the siting of warehouses. Tracts with adjacent
W&Ds in 2000 had even experienced a significantly stronger outward migration of Black
population, compared to other tracts.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
64

Table 7 Comparing value changes in selected variables between tracts with adjacent W&Ds
in 2000 and other tracts in the Los Angeles region
Changes in variables during 2000-2010 Average
Tracts with adjacent W&Ds in
2000 vs. others
(difference)
Changes in % Minority 3.6 0.1
Changes in % Latino 5.5 0.0
Changes in % Black -0.4 -0.4***
Changes in % Asian 2.3 0.0
Changes in Household income 20,216 -4,121***
Changes in Housing values 281,456 -33,408***
Changes in Population density -116 -13
Number of obs. 3,674 1,615

5.2. Simultaneous Equation Model: minorities as a whole
The statistics tests above indicate that the neighborhoods receiving W&Ds during 2000
and 2010 tend to have higher percentages of minorities at the beginning of the period.
Meanwhile, no significant “move-in” effects on demographic changes are found during
the period. However, these tests, which not only assume oversimplified pairwise
relationship, but also fail to account for simultaneous interactions between dependent
variables, can hardly warrant any firm conclusions. As discussed earlier, the location
choices of warehouses and minority population are likely to be interdependent and
simultaneous. Separately running models for the two processes would probably end up
with biased results. To further support what I found in Tables 5 to 7, I estimated the
system of simultaneous equations using three-stage least squares (3SLS). According to
the Hansen-Sargan test result, the model is significantly over-identified, thus it is safe to
use three-stage least squares to estimate coefficients in the SEM. Tables 8 and 9 show the
regression results. Changes in the percentage of minorities during 2000 and 2010 are a
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
65

significant factor in the changes in warehousing activity density during the same period
but not vice versa. Changes in warehousing activity density would increase by 24 % if the
changes in the percentage of minorities increase by 1 unit (e.g. from 1 percent to 2
percent). The relationship can also be verified by the Figure 7 below, in which the
coincidence of minority inflow and new warehousing development during 2000-2010 is
evident. The results, on one hand, confirm the existence of the environmental justice
problem, while on the other hand suggest that the problem results from the location
choice process, not the demographic change process. It is generally consistent with the
existing literature (e.g. Pastor et al., 2001; Morello-Frosch et al., 2002), although this
model does not find any “move-out” effects associated with the siting of warehousing
facilities.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
66

Figure 7 Changes in the percentage of minority population and location of warehouses built
during 2000-2010 in the Los Angeles region (data source: Costar Inc., Census 2000 and
Census 2010)

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
67

Table 8 Regression analysis result of SEM Equation 1 (Dependent variable: Changes in
warehousing activity densities 2000-2010)
Changes in warehousing activity densities 2000-2010

Coefficient S.E Sig.
Changes in percentage of minorities during 2000-2010 28.130 (4.618) ***
Warehouse RBA density in 2000 0.118 (0.017) ***
Square of warehouse RBA density in 2000 0.085 (0.006) ***
Percentage of minorities in 2000 1.123 (0.400) ***
Median Household income in 2000 -0.546 (0.337)
Square of median household income in 2000 0.016 (0.036)
Median Housing values in 2000 0.536 (0.335)
Population density in 2000 -0.764 (0.092) ***
Employment density in wholesaling in 2000 0.262 (0.049) ***
Distance to nearest intermodal terminal -0.730 (0.102) ***
Census Tract in Orange County 0.355 (0.227)
Census Tract in Riverside County 0.007 (0.331)
Census Tract in San Bernardino County 1.357 (0.315) ***
Census Tract in Ventura County 0.938 (0.376) **
Constant 2.509 (2.385)
Chi-square statistics 1,605
Pseudo R-squared 17.0%
Sample Size 3,674

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
68

Table 9 Regression analysis result of SEM Equation 2 (Dependent variable: Changes in
percentage of minorities 2000-2010)
Changes in percentage of minorities 2000-2010

Coefficient S.E Sig.
Changes in warehouse RBA density during 2000-2010 0.001 (0.001)
Warehouse RBA density in 2000 0.000 (0.000)
Percentage of minorities in 2000 -0.024 (0.006) ***
Square of percentage of minorities in 2000 -0.151 (0.018) ***
Median Household income in 2000 0.036 (0.004) ***
Square of median household income in 2000 -0.002 (0.001) ***
Median Housing values in 2000 -0.031 (0.004) ***
Population density in 2000 0.000 (0.002)
Distance to employment center in 2000 0.013 (0.001) ***
Density of park space in 2000 -0.034 (0.019) *
Census Tract in Orange County 0.017 (0.004) ***
Census Tract in Riverside County 0.034 (0.005) ***
Census Tract in San Bernardino County 0.022 (0.005) ***
Census Tract in Ventura County -0.015 (0.006) **
Constant 0.051 (0.037)
Chi-square statistics 1,106
Pseudo R-squared 23.1%
Sample Size 3,674
Meanwhile, all the other explanatory variables in both regressions, in general,
have expected coefficients in terms of both signs and significances. The relationship
between the changes in warehousing activity density during 2000 and 2010 and its pre-
period (in 2000) value is found to be nonlinear. Two effects might be considered here:
clustering and dispersing. The clustering effects primarily come from agglomeration
benefits and zoning policies. Warehousing facilities are located near each other to enjoy
the economies of agglomeration. Meanwhile, by developing industrial parks, some local
governments tend to keep warehousing land uses spatially concentrated in order to better
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
69

regulate and manage land use development (Peddle, 1993). This tendency rationalize the
fact that many new warehouses are found in pre-existing warehousing clusters. The
dispersing effects are nonetheless linked to the land availability and intra-industry
competition. When a census tract is already filled with many warehouses, newcomers
may find it difficult to compete for labor force and customers with the preexisting
warehousing firms. Results in Table 8 indicate that when pre-period warehousing activity
density increases, its changes during 2000 and 2010 increase as well, suggesting the
clustering effects are stronger than the dispersing effects. On top of that, more future
research can help further clarify the relationship.
New warehousing facilities are more likely to be sited in areas with lower
population densities, higher wholesale employment densities, and good access to
intermodal terminals. For instance, a 1% increase in wholesaling employment density is
associated with a 0.2% increase in changes in warehousing activity density, all else equal.
Such a close relationship is expected, as warehousing developers tend to locate their
facilities near wholesalers to enjoy easy access to these customers. The results also show
that compared to the Los Angeles County, new warehouses are significantly more
concentrated in San Bernardino and Ventura Counties, suggesting that warehousing
facilities had been decentralizing from the old core, LA County where land availability is
comparatively limited. This finding resonates with the spatial mapping and supports the
“logistics sprawl” hypothesis (e.g. Dablanc, Ogilvie and Goodchild, 2014). Finally, the
housing values variable, as a proxy of land rent, is not a statistically significant factor in
the warehousing location equation. This somewhat unexpected finding may be partly due
to the inadequate measurement of land costs using housing values, but I have to admit
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
70

that it was the best proxy for industrial land rent at hand. There is also a possibility that
land rent is indeed not among the most influential factors, especially in the LA CSA
where industrial land at a moderate rent is widespread. In many cases, warehousing
developers may be willing to trade higher rent for better access to transport terminals and
customers and favorable local policies. Results from other models below would further
validate these explanations.
On the other hand, changes in the percentage of minorities during 2000 and 2010
are highly related to the percentage of minorities, household income, housing values,
density of park space and distance to employment center at the beginning of the period.
For example, a 1% decrease in the density of park space in 2000 is associated with a 0.03
unit increase (e.g. from 1 percent to 1.03 percent) in changes in the percentage of
minorities, all else equal. The results suggest that minority people are at a distinct
disadvantage in competing for housing that is close to local amenities including parks and
employment opportunities. Meanwhile, compared to the LA County, minority population
flow more into all counties except for the Ventura County, which is a popular region
among white residents. These findings are all consistent with expectations.
5.3. Simultaneous Equation Model: different race groups
In the next step, I examined how race groups including Whites, Latinos, Blacks, and
Asians behave differently in these dynamics, by running variants of the model above.
These SEM regression results (see Table 10 and 11) indicate that those neighborhoods
with Whites moving in had seen warehousing moving out, neighborhoods with inflows of
Latino and Asian population had experienced substantial warehousing development,
while neighborhoods receiving Black population had nonetheless witnessed no
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
71

significant changes in warehousing intensity. The contrast between these race groups
suggests race exactly plays a crucial role in the warehousing location choice. From the
perspective of environmental justice, whether a community suffers from disparate
hazards could highly depend on the strength of political resistance there. The regression
results demonstrate a likely pattern in which Latinos and Asians, among minorities, are
politically weaker and thus more vulnerable to warehousing related environmental
impacts than Blacks. Why is that pattern? As environmental justice advocates have
historically paid more attention to Black population (see studies on “Put It in Blacks’
Backyards”, e.g. Bullard, 2001), it is possible that the disparities in environmental
hazards are nonetheless less significant in Black neighborhoods (for instance, in Mohai
and Saha 2015b, percent black is found to be less influential in the siting of TSDFs than
other minority groups). Admittedly, empirical evidence for this hypothesis has still been
scarce so far. Such a distinction between minority groups could also be partly explained
by Black population’s unique migration pattern, in which they most likely move to the
periphery of the urban area (see Figure 8). These areas have limited access to freight
infrastructure as well as major customers and may not be the best choices for
warehousing developers. The limited access of the Black population to warehousing
facilities also indicates their reduced exposure to warehousing jobs. This actually brings
about another interesting question: if moving away from polluting facilities means
moving away from job opportunities, how should we evaluate the loss and gain of those
people? Research on this question is unfortunately very limited.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
72

Table 10 Regression analysis result of SEM Equation 1 for different race groups (Dependent
variable: Changes in warehousing activity densities during 2000-2010)
Changes in warehousing activity densities during 2000-2010

White Latino Black Asian
Changes in percentage of certain race group
during 2000-2010
-27.130*** 16.492*** -0.077 29.931**
Warehouse RBA density in 2000 0.118*** 0.118*** 0.125*** 0.116***
Square of warehouse RBA density in 2000 0.085*** 0.088*** 0.086*** 0.091***
Percentage of certain race group in 2000 -1.123*** 0.298 0.183 -3.752
Household income in 2000 -0.546 0.013 0.967*** 1.067***
Square of median household income in 2000 0.016 -0.039 -0.081*** -0.059*
Housing values in 2000 0.536 -0.096 -1.019*** -1.022***
Population density in 2000 -0.764*** -0.938*** -0.806*** -0.625***
Employment density in wholesaling in 2000 0.262*** 0.268*** 0.247*** 0.245***
Distance to nearest intermodal terminal -0.730*** -0.663*** -0.644*** -0.707***
Census Tract in Orange County 0.355 0.885*** 0.734*** 0.362
Census Tract in Riverside County 0.007 0.097 0.957*** 1.255***
Census Tract in San Bernardino County 1.357*** 1.245*** 2.107*** 2.316***
Census Tract in Ventura County 0.938** 0.506 0.770** 1.081***
Constant 3.632 6.934*** 8.655*** 5.555**
Chi-square statistics 1,605 1,785 1,932 1,690
Pseudo R-squared 17.0% 28.1% 34.5% 25.2%
Sample Size 3,674

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
73

Table 11 Regression analysis result of SEM Equation 2 for different race groups (Dependent
variable: Changes in percentage of race group during 2000-2010)
Changes in percentage of certain race group during 2000-2010

White Latino Black Asian
Changes in warehouse RBA density during
2000-2010
-0.001 0.000 0.000 -0.001**
Warehouse RBA density in 2000 0.000 0.000 0.000 0.000*
Percentage of race group in 2000 -0.027*** 0.042*** -0.202*** 0.182***
Square of percentage of race group in 2000
0.151*** -0.217*** 0.224*** -0.152***
Median Household income in 2000 -0.036*** 0.049*** 0.002 -0.007***
Square of median household income in 2000 0.002*** -0.001 0.000 -0.001***
Median Housing values in 2000 0.031*** -0.036*** -0.005** 0.002
Population density in 2000 0.000 0.006*** -0.003*** -0.007***
Distance to employment center in 2000 -0.013*** 0.008*** 0.001* 0.003***
Density of park space in 2000 0.034* -0.062*** 0.014 0.007
Census Tract in Orange County -0.017*** -0.006 -0.004** 0.013***
Census Tract in Riverside County -0.034*** 0.051*** -0.003 -0.006**
Census Tract in San Bernardino County -0.022*** 0.051*** 0.003 -0.003
Census Tract in Ventura County 0.015** 0.008 -0.011*** -0.011***
Constant -0.026 -0.110** 0.069*** 0.105***
Chi-square statistics 1,106 902 662 636
Pseudo R-squared 23.1% 19.6% 15.3% 13.9%
Sample Size 3,674
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
74

Figure 8 Changes in the percentage of Black population and location of warehouses built during
2000-2010 in the Los Angeles region (data source: Costar Inc., Census 2000 and Census 2010)
While changes in the percentage of minorities as a whole are not significantly
related to changes in warehousing activity density (see Table 9), I find that the siting of
warehouses actually leads to “move-out” effects on Asian population, but not on other
groups of minorities (see Table 11). Oakes et al. (1996) and Pastor et al. (2001) reported
similar findings, although these studies do not distinguish minority groups. Different
consumer tastes across minority groups might help explain this pattern. The move-out
effects suggest that Asian residents may be more sensitive to warehousing-related
environmental impacts and relocate in response to the siting of warehouses.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
75

5.4. Simultaneous Equation Model: ethnic churning
Per the discussion on Table 10 above, the literature has not adequately shown whether
political resistance would be the major reason why Blacks are unique among minorities
in the dynamics of disproportionate warehousing development. The hypothesis of
political weakness has not been adequately verified using longitudinal data. Ethnic
churning, an indicator developed by Pastor et al. (2001), can further help test whether
political power or civic engagement matters in resisting disproportionate siting of locally
unwanted land uses such as warehouses. Ethnic churning, which is defined as the
absolute sum of ethnic changes, distinguishes, for example, a tract with 5% Blacks
moving out and 5% Latinos moving in (ethnic churning equals 10%) from a tract with
30% Blacks moving out and 30% Latinos moving in (ethnic churning equals 60%). While
these two types of ethnic changes both indicate no changes in the percentage of
minorities, they may represent two distinct circumstances of the local civic organization.
Pastor et al. (2001) argued that the ethnic churning would weaken neighborhood social
capital and increase its vulnerability to siting LULUs. Incorporating ethnic churning into
the SEM, therefore, can yield results that help understand how differences in political
power would contribute to environmental inequity.
Table 12 shows that ethnic churning is indeed significantly related to changes in
warehousing activity densities; in other words, neighborhoods with more ethnic churning
had received more warehousing development. The results on other explanatory variables
are consistent with those in Table 6 except for household income and housing values. It
turns out that income has first a positive then a negative effect on warehousing location in
this model. This finding, though not seen in the first SEM results (see Table 8), resonates
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
76

with quite a few earlier studies (e.g. Boer et al., 1997; Morello-Frosch et al., 2002) where
income was found to have a nonlinear relationship with the siting of LULUs. The poorest
neighborhoods may not be most vulnerable to disparate warehousing related hazards;
instead, they, in some cases, receive fewer facilities than other neighborhoods. In
addition, the coefficient for housing values confirms that warehousing developers tend to
locate their facilities in places with lower land rent. This finding supplements the
unexpected results in Table 8 and reaffirms the significance of land rent in the location
choice of warehouses.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
77

Table 12 Regression analysis result of SEM Equation 1 (Dependent variable: Changes in
warehousing activity densities 2000-2010)
Changes in warehousing activity densities 2000-2010

Coefficient S.E Sig.
Ethnic churning during 2000-2010 5.594 (2.839) **
Warehouse RBA density in 2000 0.120 (0.016) ***
Square of warehouse RBA density in 2000 0.089 (0.006) ***
Percentage of minorities in 2000 0.487 (0.349)
Median Household income in 2000 0.946 (0.205) ***
Square of median household income in 2000 -0.066 (0.030) **
Median Housing values in 2000 -0.664 (0.240) ***
Population density in 2000 -0.784 (0.088) ***
Employment density in wholesaling in 2000 0.237 (0.049) ***
Distance to nearest intermodal terminal -0.563 (0.101) ***
Census Tract in Orange County 0.801 (0.195) ***
Census Tract in Riverside County 0.943 (0.256) ***
Census Tract in San Bernardino County 2.191 (0.262) ***
Census Tract in Ventura County 0.798 (0.346) **
Constant 2.798 (2.994)
Chi-square statistics 1,927
Pseudo R-squared 34.1%
Sample Size 3,674

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
78

Table 13 Regression analysis result of SEM Equation 2 (Dependent variable: Ethnic churning
during 2000-2010)
Ethnic churning during 2000-2010

Coefficient S.E Sig.
Changes in warehouse RBA density during 2000-2010
-0.002 (0.001) **
Warehouse RBA density in 2000
0.000 (0.000)
Percentage of minorities in 2000
0.018 (0.007) **
Square of percentage of minorities in 2000
-0.372 (0.022) ***
Median Household income in 2000
-0.006 (0.005)
Square of median household income in 2000
0.001 (0.001)
Median Housing values in 2000
-0.034 (0.005) ***
Population density in 2000
-0.015 (0.002) ***
Distance to employment center in 2000
-0.004 (0.001) **
Density of park space in 2000
0.002 (0.024)
Census Tract in Orange County
-0.011 (0.004) ***
Census Tract in Riverside County
0.006 (0.005)
Census Tract in San Bernardino County
-0.002 (0.006)
Census Tract in Ventura County
-0.004 (0.007)
Constant
0.740 (0.041) ***
Chi-square statistics
1,106
Pseudo R-squared
22.2%
Sample Size
3,674
In accordance with Pastor et al. (2001), warehousing siting has a negative effect
on ethnic churning, or in other words, the more warehouses are sited in a neighborhood,
the less ethnic churning would occur in the neighborhood. This is another piece of
evidence against the demographic change hypothesis; the colocation of warehouses and
minority population does not come from the demographic changes following the siting of
warehouses.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
79

6. Conclusion
This study employs longitudinal data in the Los Angeles region to test whether
warehousing facilities are disproportionately sited in minority neighborhoods, or
reversely, minority population move into neighborhoods that receive warehouses, on the
assumption that these two processes are interdependent. This empirical test would, on one
hand, provide more robust evidence on the existence of environmental inequity in
warehousing location, and on the other hand, illustrate how such inequity is created.
Simultaneous equation model results suggest that the inflow of minorities is positively
associated with the siting of warehouses, which, however, does not lead to any move-in
of the minority population afterwards. That is to say, with other factors controlled,
minorities attract warehouses, but not vice versa.
Different minority groups, however, are in separate positions in the dynamics of
disproportionate warehousing development. Unlike Latinos and Asians, the influx of
Black population into a neighborhood may not induce the siting of warehouses
afterwards. This difference may be partly ascribed to stronger civic organization among
the Blacks or the distant location of the communities with Black people moving in. In
addition, model results also show that ethnic churning, an indicator of social stability, is
closely linked to warehousing development. Neighborhoods with strong ethnic churning
are more vulnerable to growing warehousing related impacts.
This research is one of the earliest studies that explore the mechanisms behind the
environmental justice problem in warehousing location. Admittedly, there are some
limitations due to the lack of better data sets. The dataset of warehousing establishments
in this study does not cover the “death” of warehouses, making it difficult to precisely
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
80

measure the net gain of warehousing intensity. Meanwhile, this study fails to control
many localized factors such as public policies and zoning regulations as it is difficult to
collect detailed data for these factors and further quantify them in the SEM. More future
research on this topic is needed to help policy makers understand the relationship
between local policies and warehousing-related environmental disparities.
In spite of the limitations, the study explicitly points out that the disproportionate
siting of warehouses, rather than the housing market dynamics, is the dominant causal
factor of the environmental justice problem in warehousing location. Policy implications
of the conclusions are profound. Given that the disparate siting of warehouses is largely
responsible for the environmental inequity, governments and planning agencies should
step in and evaluate how local land use and environmental standards can help mitigate
such inequity. As argued by Pastor et al. (2001) and Morello-Frosch et al. (2002), a
facility-by-facility search is not an efficient way to address the problem. Instead,
developing and enforcing strict and consistent standards to regulate excessive
warehousing development would be more realistic and effective, especially in the long
term. Local authorities need to carefully examine how proposed warehousing projects
would affect the environment and the residents’ lives, after fully considering the existing
developments and environmental conditions.
The consistency in standards across local municipalities is in particular important
for controlling the impacts of mobile sources (Morello-Frosch et al., 2002; Pastor,
Morello‐Frosch, & Sadd, 2005) such as trucks, so it would be rather helpful if state and
local governments as well as public agencies can work together to establish regional
collaboration networks for combating disparities in the distribution of freight-related air
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
81

pollutants and noise. Such practices so far have been rare but the public sector is noticing
the necessity. The California Air Resources Board (CARB) has initiated a program called
“Minimizing Community Health Impacts from Freight”, through which the organization
aims to develop proposals to reduce freight impacts on local communities. Several
workshops and community meetings were held to get stakeholders involved in the
program and such communication and collaboration would greatly help identify where
the major conflicts are and how to resolve them.
The minority population is found to be vulnerable to disproportionate siting of
warehousing facilities due to their less effective resistance to the siting decision. How can
the public sector empower those people to fight against the disparities? First, minority
people should be better informed of adequate information about the hazards as well as the
political process of land use decision making. In some neighborhoods with extensive
warehousing land uses, minority residents can easily observe the presence of
environmental hazards. For instance, a Latino resident complained about the “black soot
blowing out of truck exhausts” at one of the community meetings held in Long Beach by
CARB. But many residents stated that they were not aware how (much) these impacts
would affect their health, living space, and property values. In many cases, neither were
they notified of new warehousing projects and the environmental consequences these
projects could bring to their neighborhoods. The minority population deserves the
disclosure of more information, with which they can better get involved in the land use
decision-making process.
To get their voices heard in the political process, minority population also need
more support on public organization and participation. As this study has demonstrated,
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
82

social stability and cohesion matter a lot in organizing effective resistance to the
disproportionate warehousing development. To encourage civic engagement and provide
support for the political organization would be a good strategy for public agencies to
empower minority residents to stand out and speak for themselves. Once warehousing
related environmental hazards are incorporated into the agenda of environmental justice
movement by minority residents, the changes in public policy making and the regulatory
environment are expected to arrive soon. As Bullard and Wright (1987) argued, minority
communities could use their institutions such as civic clubs and political organizations to
develop a network of advocates and experts. Such a network can not only help minority
population better access and understand the information about environmental hazards, but
also function as a base for well-organized social movement.
After all, warehousing location is an emerging and significant environmental
justice problem facing the major metropolitan areas. Compared to other LULUs,
warehousing facilities have been proliferating thanks to the unprecedented goods
movement demand. To avoid potentially damaging environmental and social
consequences, the government and the public need to work together to facilitate an
effective regulation of warehousing related externalities and a fair distribution of related
environmental disamenities.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
83

CHAPTER 4
Environmental Justice in Warehousing Location across Cities:
Institutional Perspectives

Abstract
An increasing number of recent research revealed that warehouses are disproportionately
located in minority neighborhoods, and it becomes a rising environmental justice problem
that calls for a systematic examination. The existing research, however, failed to present
how municipal institutional factors play a role in the dynamics that cause the
environmental inequity. In this study, I first identified local public policy elements that
affect the location choice of warehousing facilities by reviewing relevant literature and
interviewing planners, warehousing developers, and regional agency staff. I then
evaluated the relationship between these policy elements and warehousing development
using qualitative city-level data for the Los Angeles region. In this step, I decoupled the
local institutional factors from other factors in the warehousing location choice by
comparing municipalities which have similar socioeconomic characteristics but different
trajectories of warehousing development. Results showed that land use policies (e.g.
industrial zoning, land parcel division rules), job related policies (e.g. job creation
initiative, job density requirements), financial incentives (e.g. tax rates, financial
incentives), and environmental regulations (e.g. building design, buffering between
conflicting land uses, landscaping) are the major policy elements that impact
warehousing development. There is, however, no evidence showing that local policies
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
84

have played a major role in linking low-income and minority population with
warehousing related externalities so far. However, as the growing warehousing related
environmental threats have caused growing concern among local residents, local policies
would potentially become a major drive towards environmental inequity in warehousing
location.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
85

1. Introduction
In response to the growing demand for goods movement, the warehousing industry has
experienced an unprecedented expansion. Such expansion is particularly remarkable in
contrast with other major industries. During 2007-2016, the number of employees in the
warehousing and storage industry increased by 43% in the United States, while the
percentage changes in manufacturing, wholesale, and retail sectors during the same
period were -12%, -1% and 3% respectively (Bureau of Labor Statistics, 2017). The
warehousing boom is prevalent across many large metros, especially those with abundant
inexpensive suburban land. Among the largest eight metropolitan areas in the U.S., six of
them, Los Angeles, Chicago, Dallas, Houston, Philadelphia and Miami, all gained at least
20% increases in the number of establishments in the warehousing and storage industry
between 2003 and 2015 (Longitudinal Employment-Household Dynamics, 2016).
The steady and widespread growth in the warehousing industry has profound
environmental and social implications. Warehousing facilities generate considerable
environmental impacts on local communities. Truck activities are the major sources of
these externalities. When trucks frequently move in and out of warehouses, the vehicles
emit air pollutants including NOx, SO2, CO and particulate matters, damage road
pavement, generate noise, and pose safety threats to pedestrians and bicyclists in the
neighborhoods. Therefore, warehouses are increasingly regarded undesirable by residents
living nearby, particularly when the new generation of warehousing facilities are built
much larger than the old ones. People begin to worry about the explicit environmental
threats and seek support from local governments and regional planning organizations.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
86

In addition to the spatial expansion of warehouses, the uneven distribution of
these facilities causes equal concerns. Recent research (Author, 2018) revealed that
warehouses are disproportionately located in minority neighborhoods, therefore
warehousing becomes a rising environmental justice (EJ) problem that calls for a
systematic examination. It is increasingly evident that the spatial inequity in warehousing
location exists not only at the neighborhood level, but also between municipalities.
During the same period and in the same region, some cities received a large number of
new warehousing facilities while their neighbors were surprisingly free from any
warehousing move-in. Citizens in those warehousing hot spots therefore suffer much
more from the environmental externalities. The dramatic variations across cities imply
strong effects of city-level factors.
What factors lead to the spatial inequity in warehousing activity intensity? The
environmental justice literature has proposed the classic threefold—economic,
sociopolitical and racial—explanations (e.g. Mohai, Pellow, and Roberts, 2009), but the
majority of the research supported these explanations using neighborhood-level
indicators and data only. Whether city-level factors, especially institutional factors play a
major role in the dynamics is, however, largely unknown. After all, local governments
make land use decisions, and their interventions and policies have strong and long-term
effects on land use patterns. These effects are too critical to be ignored in evaluating the
environmental justice problem in warehousing location. More empirical evidence is
particularly needed to shed some light on this topic.
The emergence of warehousing related environmental justice problem largely
derives from the warehousing location choice process (Yuan, 2018). To study the
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
87

relationship between local institutional factors and warehousing location choice is key to
understanding how local policies may contribute to the environmental inequity. Several
empirical studies (Yuan, 2018a, 2018b) measured the variations in warehousing activity
intensity among neighborhoods and studied the factors that are readily measurable and
quantifiable at the neighborhood level, such as socioeconomic characteristics, transport
access and industrial connections. But the role of local municipalities in attracting or
preventing warehousing siting has not been addressed yet. Measurability and data
availability may be two main reasons for this gap in the literature. It is difficult to collect
data for public policies such as zoning and land use regulations and meanwhile figure out
which policies are actually enforced. At the neighborhood level—usually defined as
census tract, block group, or zip code in empirical studies—, such difficulty is even more
visible as the differences in policy enforcement between neighborhoods are quite subtle.
As a result, to measure local institutional factors at the neighborhood level is not
theoretically operational.
The prior quantitative work has confirmed the coincident location between
warehouses and minority population but failed to clarify the role of local institutional
factors due to methodological constraints. In response, this study turns to qualitative case
studies to understand the dynamics underlying the spatial patterns that have not been
fully explained. In this study, I first identified local public policies that affect the location
choice of warehousing facilities by reviewing relevant literature and interviewing policy
makers. Then I evaluated the relationship between these policies and warehousing
development using qualitative city-level data for the Los Angeles region. With this
information, I finally discussed how local policy making may lead to the disproportionate
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
88

distribution of warehousing related environmental hazards, and furthermore can help
address this problem.
The paper contains 6 sections. Section 2 contains a literature review of recent
research progress on relevant topics. Section 3 covers research approach and data
sources. Results are presented in Section 4, and detailed discussions are included in the
final two sections.
2. Literature Review
2.1. Warehouses as locally undesirable land uses?
With respect to warehousing related environmental impacts, research that systematically
examines these impacts using empirical data is scarce, but studies on the externalities of
truck movement and large-scale structures can help us understand how warehouses may
be environmentally unfriendly and locally undesirable. Warehouses attract and generate a
large volume of truck trips, and trucks are a major emitter of various air pollutants
including NOx, PM10 and PM2.5. For instance, in the State of California, trucks account
for 33% of the NOx, 14% of the PM10 and 17% of the PM2.5 in the entire transportation
sector (National Emission Inventory, 2014). The contribution of trucks to air pollution
appears to be more dramatic when the fleet sizes of different vehicles are taken into
consideration. Among all registered vehicles in California, 73% are automobiles and 16%
are trucks; however, the share of NOx emitted by autos, 28%, is even lower than that by
trucks. A rough calculation based on these numbers reveals that one truck can generate
nitrogen dioxides five times as much as one auto. Similar contrast can be found when
measuring noise generation. According to Seto et al, (2007), a medium truck on average
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
89

can generate noise nine times as much as an automobile, and that ratio increases to
twenty-two when it comes to a combination truck, which has been widely used in goods
movement. Air pollutants and noise can cause a wide range of health problems including
asthma, respiratory allergy, and hearing despair (e.g. Passchier-Vermeer and Passchier,
2000; Kim et al., 2004; Kozawa, et al., 2009). On the other hand, warehouses as sizable
structures can aggravate the urban heat island effects (Voogt, 2007) and stormwater
runoff threats (Yang and Li, 2013), although empirical studies exclusively focusing on
warehousing facilities are quite inadequate. It is worth noting that these environmental
impacts are all highly related to the sizes of warehouses. The average size of warehouses
built in the Los Angeles Combined Statistical Area during 2006-2016 reached 263
thousand square feet, 2.3 times as many as twenty years ago (during 1986-1996, the
average number was 114 thousand square feet) (Costar, 2016). In the era of globalization
and E-commerce, developers tend to build megawarehouses as a means to seek
economies of scale. The proliferation of these facilities greatly boosts the expansion of
subsequent environmental impacts.
Another factor that makes warehouses less locally desirable is their shrinking
employment opportunities and benefits. Warehouses used to create quite a lot of blue
collar jobs whose work ranged from loading and unloading trucks, stacking items on
pallets, checking damages, managing inventory, and so forth. However, a large
proportion of these jobs have been taken by robots, automated systems and computers. In
some warehouses, as high as 75% of the warehousing jobs have been replaced by robots
(Yakowicz, 2017) and such trend is found all over the globe (McNeice, 2017). The
introduction of robots and automated systems as substitutes of low-skilled labor not only
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
90

depresses the job demand but also places pressure on the wages. A study by the National
Bureau of Economic Research (Acemoglu and Restrepo, 2017) found that for each new
robot added per 1,000 workers, wages in the surrounding area would fall between 0.25
and 0.5 percent. Warehousing jobs are likely among the most vulnerable ones given the
recent technological advances in the warehousing robotic systems, which can now not
only take over a lot of human work but also run 24/7 (Pooler ,2017).
The growing environmental externalities and diminishing contribution to quality
job creation of warehousing activities make the facilities locally unwanted. The growing
complaint from local residents about these externalities further implies that warehousing
location might cause an environmental justice problem. As stated above, empirical
studies have justified the existence of such problem but the role of local institutional
factors has not been carefully studied yet. How do the aforementioned three explanations
relate to local institutional factors? The economic explanation ascribes the disparate
siting of locally undesirable land uses (LULUs) to the behaviors of the industry to seek
cheap resources and ready conveniences (Mohai and Saha, 1994). Local land use,
economic development and financial policies have significant effects on the production
and allocation of these economic resources and conveniences, and thus on the siting of
those land uses. According to the sociopolitical explanation, local governments are
criticized for making biased land use policies and decisions against minorities (Mohai,
Pellow, and Roberts, 2009). Affluent white communities have political resources to exert
stronger influence on local policy and decision making, so communities with less
effective political power thus end up with more intensive environmental hazards. Racial
discrimination explanation, on the other hand, highlighted the discriminatory zoning
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
91

made by local governments (Roberts and Toffolon-Weiss, 2001; Cole and Foster, 2001).
But such behaviors were in particular typical in the early 1900s (Cole and Foster, 2001).
After all, the environmental justice problem essentially results from racially biased local
policies which greatly affect the location choice of LULUs.
In the case of warehousing location, are these effects still valid? Maybe not at this
moment. Dating back to two decades ago, warehouses were smaller, generated fewer
truck trips, and hired more blue-collar workers. Therefore, warehousing related
environmental threats were much less significant at that time. Meanwhile, the research
and knowledge on the environmental implications of warehousing activities had been
very limited until the recent decade or so. In the old days when the negative
environmental impacts of warehousing activities were largely uncertain, discrimination in
local land use policy making based on the impacts would be less likely to take place. But
the tendency of warehousing developers to look for advantageous economic resources
would be common even half a century ago. It actually recalls an important discussion in
the EJ literature. Saha and Mohai (2005), examined the temporal patterns of the disparate
siting of waste facilities using data from 1950 to 1990 in the State of Michigan. They
found that the disparate siting of the facilities was only significant after 1970, and the
patterns varied due to changes in anxieties about hazardous waste, social group political
participation in siting decisions, and their effects on facility siting outcomes. By referring
to the history of industrial development, they thus concluded that the emergence of
environmental justice problem is highly subject to the historical context. Such finding can
also be applied to the case of warehousing location. Whether and how local governments
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
92

behave in favor of certain population groups would largely depend on the time period in
which the warehousing location choice process is observed.
In summary, local public policies may significantly affect the disproportionate
distribution of warehousing related environmental impacts. Such effects are likely to vary
over time and are highly subject to the race-based bias in the policy making process,
which plays an important role in the warehousing location choice.
2.2. The role of local public policies in warehousing location choice
Warehousing developers make location choices to maximize their profits. When
developers compare available sites to locate their facilities, they consider how those
locations are related to potential land costs, transportation costs, labor costs, and
operating costs (Korpela and Tuominen, 1996; Sivitanidou, 1996; Verhetsel et al., 2015).
The literature has well documented the traditional way developers make tradeoffs
between these costs (e.g. Hesse and Rodrigue, 2004; Demirel et al., 2010). However, the
recent restructuring of the warehousing industry in response to the changing logistics
demand has greatly reshaped the way these costs are weighted. In particular, local public
policies are becoming increasingly important in influencing the attractiveness of locations
for warehousing development.
Compared to two decades ago, warehouses are built in larger sizes and consume
larger parcels of land, thanks to increased storage capacity and the introduction of
automated warehousing equipment which demands ample interior space (Andreoli,
Goodchild and Vitasek, 2010). Warehousing developers thus have to spend more on land
lease, acquisition, and consolidation if the local land use ordinances and patterns are not
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
93

compatible (Hesse, 2004). Municipalities that provide large parcels of industrial-zoned
land and flexible land use regulations are readier to accommodate these large-size
warehouses (McKinnon, 2009).
As for warehousing establishments, transportation costs are related to the
accessibility to its suppliers and consumers along the supply chain. The proximity to
these businesses, and the connectivity to transport infrastructure such as freeways,
railways, and intermodal terminals can both significantly influence the speed and
reliability of freight deliveries (Bowen, 2008). The on-going market and supply chain
fragmentation eases the attachment of warehouses to the city cores (Rodrigue, 2008). As
a result, suburban cities with good regional access gain the redefined locational
advantages and also popularity among warehousing developers. Some of these cities
invest in local transport infrastructure or adopt favorable transport policies to support
warehousing activities. The practices can give warehousing operators additional
incentives to develop facilities in those cities.
Warehouses primarily hire workers with skills such as truck driving and cargo
handling (Husing, 2016). To save labor costs, warehousing developers are in favor of
locations where abundant blue collar workforce with relevant skills are available. Some
cities develop training programs and skill-learning plans to help local residents better
fulfill warehousing job requirements (Husing, 2016). Meanwhile, cities can adjust labor
policies such as minimum wage, labor union requirements, and employment benefits
regulations to affect the labor costs, although such practices targeting the warehousing
industry are not common.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
94

Tax policies are another critical factor in warehousing location choice.
Warehousing businesses are land-intensive, and property taxes are dominant on their tax
bills. Some facilities, especially those providing refrigerating storage, are charged a fair
amount of utility user taxes, if any, while paying for utilities. In the cases where some
warehousing businesses process and sell goods, sales tax rates are also a relevant tax
policy element. Therefore, among other financial incentives or disincentives, tax policy
elements including tax rates, incentives (discounts, credits, etc.), and disincentives are
possible traditional tools local governments use to encourage or discourage warehousing
development (Ackerman, 2012).
Unfortunately, the effects of local public policies on all these location choice
factors have not been adequately examined using empirical evidence. The existing
research only provides anecdotal descriptions or even hypothetical statements. Some
other factors such as environmental regulations have not been carefully discussed in the
literature. Therefore, a systematic examination of all relevant factors and their effects is
needed. The next question is from the perspective of local governments, what does
warehousing development bring about?
Among the few relevant studies, Dablanc (2014) argued that for local
governments, job opportunities and tax revenues are two primary benefits from
warehousing development. However, compared to other industries, the warehousing
industry has no advantages in generating either benefits. Warehousing facilities have
comparably lower job densities (Dablanc, 2014) and a considerable proportion of
warehousing jobs are low-paid and temporary (Kirkham, 2015; Kitroeff, 2016). With
latest technological innovations in warehouse robots and automated vehicles, the new
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
95

generation of warehouses further shrink their labor demand (McNeice, 2017). With
regard to tax revenues, most warehousing businesses still do not generate sales taxes at
all, despite that warehousing facilities become increasingly specialized in value-adding
services (Bowen, 2008; Cidell, 2011; Rodrigue, 2008).
Warehousing related social costs, on the other hand, are growing thanks to the
rapid expansion of warehouses. Air pollution, pavement damage, noise, and traffic safety
threats are all notable environmental problems facing local policy makers. The tradeoff
between these benefits and costs would largely determine the attitudes of local
governments. To understand the way local governments view warehousing development,
researchers have to carefully evaluate how policy makers weigh the subsequent benefits
and costs. Empirical evidence on this topic is nonetheless highly limited in the literature.
3. Data and Approach
3.1. Research approach
According to the literature review, local policies may be a major cause of warehousing
related environmental inequity. To test this hypothesis, this study examines how local
policies affect the warehousing location choice and further discusses whether and how
these policies become favorable or unfavorable to warehousing location. As the process
of warehousing location choice is subject to a large group of economic, social and
political factors, I had to decouple the local institutional factors from others. I compared
municipalities which have similar socioeconomic characteristics and accessibility to
transport infrastructure but different trajectories of warehousing development. I assumed
that these control variables are more subject to public policies that take effect at the sub-
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
96

regional level rather than localized warehousing-related policies. For instance, industrial
land rent, as a key variable in the location choice function of warehousing developers, is
more likely to be an outcome of the sub-regional real estate market dynamics. The
variations in industrial land rent within a real estate sub region may be less significant
than across sub regions. These control variables include population size, land area,
average land rent, racial composition, housing stock age, household income,
unemployment rate, and share of residents in transportation and warehousing occupation.
Among them, the racial composition suggests the political and social structures, and the
housing stock age indicates the urban development history—since when and how much
the land has been developed. The control variables demonstrate the socioeconomic
conditions of each municipality from different perspectives. Spatial location is another
important dimension that I controlled for before conducting the comparison. Where a city
is located is not only related to many socioeconomic environments such as labor and real
estate environments, but also suggests how far away each city is from freight generators
including airports, seaports, and intermodal terminals. The proximity to these freight
generators is expected to have great impacts on the warehousing location choice.
The comparison was made in pairs. A pair of municipalities are qualified as
candidates in the comparison, if they have generally similar socioeconomic
characteristics and are in close vicinity but had experienced different trajectories of
warehousing development during the observation period in spite of these similarities. To
determine the qualified pairs, I started with a large number of potential pairs, each of
which was located in the same sub-region, and then narrowed the list down based on the
socioeconomic characteristics. After the sample for comparison was drawn, I surveyed
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
97

local public policy elements that are theoretically relevant to warehousing development,
ranging from land use, economic (industrial) development, environmental assessment, to
tax, labor force and job creation. The collected data for each pair of municipalities in the
sample were incorporated together, summarized and analyzed. I then further made
connections between these policy elements and warehousing development trajectories
and concluded with a discussion on how policy elements may contribute to environment
inequity.
3.2. Study area and observation period
I chose the Los Angeles Combined Statistical Area (CSA) as my study area given its long
history of warehousing development, and rich information on local policy making. The
region contains five counties including Los Angeles County, Orange County, Riverside
County, San Bernardino County, and Ventura County, and a total of 184 municipalities
and 145 Census-designated (unincorporated) places. These municipalities vary a lot in
ages, sizes, land values, political structures, and policy making processes. They present
much information on how variations in policy making across municipalities lead to
different behaviors of warehousing developers.
I examined the policy elements in effect during the last two decades, 1996-2016,
when freight demand had been driven by the tide of globalization and the rise of E-
commerce. Within the period, local governments, especially those at favorable regional
locations, may sense the growing demand pressure for logistics land use development.
Some of them had responded to the changing demand by making new strategies, while
warehousing development had still been affected by many long-standing policies. Both
groups of public policies are considered in the analysis.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
98

The spatial expansion of the warehousing during the observation period was
dramatic, and the intensity of new warehousing development varied a lot across
municipalities. Figure 9 displays the total added square footage (in millions) of
warehousing facilities by municipality from 1996 to 2016. Because of its large size, the
city of Los Angeles is not included. Other areas in light grey tone did not have new
warehouse construction during the period. Logistics sprawl (Dablanc, Ogilvie, and
Goodchild, 2014) was evident during the period; new warehousing facilities were largely
clustered in the cities in the San Bernardino County and the Riverside County, two large
suburban counties in the region. Municipalities in the core counties, Los Angeles and
Orange County were generally no longer the most popular destinations of warehousing
establishments, although several cities in these two counties still stood out with active
warehousing development. The city of Los Angeles, as the largest municipality in the
region, had experienced a constant decline of warehousing development during the last
decades. The average annual square footage of warehousing facilities in the city of LA
decreased from 821 thousand fifty years ago (Year 1966-1976) to only 66 thousand in the
last ten years (Year 2006-2016).
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
99

Figure 9 Total added square footage (in millions) of warehousing facilities by municipality in
the municipalities from during 1996-2016 (the city of Los Angeles is not included given its
much larger size; other municipalities that are not included in the map do not have any
warehousing development during the period)
I chose the Year 2000, a data-abundant year at the beginning of the observation
period, to observe the pre-existing socioeconomic attributes of cities. By examining these
socioeconomic attributes, I was able to ensure municipalities in pairs selected for case
studies were comparable in terms of warehousing location factors other than institutional
ones. The Census 2000 provides data for control variables including population size,
median housing rent, racial composition, median year built of housing stock, household
income, employment rate, and the share of residents in the transportation and
warehousing occupation in 2000. Among the variables, median housing rent is a proxy
for the industrial land rent level. The data for warehousing development comes from the
Costar Inc., which provides up-to-date industrial real estate listing information. The
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
100

Costar data, which contains the spatial location, the square feet size, and the year built of
each warehousing facility in the region, can offer a detailed coverage of warehousing
development in the municipalities. The data can be used to track when, where, and how
large the warehousing facilities were built.
3.3. Selecting the sample for comparative case studies
To select the appropriate candidates of municipalities into the sample for the comparable
case studies is the key to adequately evaluate the relationship between local policy
making and warehousing development. I first divided the entire region into several sub-
regions according to spatial location, boundaries of political jurisdictions and divisions
(counties, and congressional districts), the relationship with transport infrastructure
(freeways, and rail lines), and regional organization and Councils of governments (e.g.
Gateway Cities Council of Governments). These criteria help group cities with
approximately comparable socioeconomic and political conditions together. After
excluding a few sub-regions with no warehousing development (e.g. West LA, and
Northern LA County) during the observation period, the municipalities in the entire LA
CSA were categorized into thirteen sub-regions (see Figure 10 and Table 14). In general,
the municipalities within each sub-region have relatively similar attraction for
warehousing development, such as land availability, land rent, land use patterns, distance
to freight generators, access to transport infrastructure, and so forth.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
101

Figure 10 Sub-regions in the Los Angeles CSA
Table 14 Municipalities within each subregion
County Sub-region Municipalities
Los Angeles
South Bay 14 municipalities (incl. Rancho Palos Verdes,
Redondo Beach, etc.)
Gateway cities-south 7 municipalities (incl. Carson, Compton,
Long Beach, etc.)
Gateway cities-north 19 municipalities (incl. Huntington Park,
Commerce, Pico Rivera, etc.)
San Gabriel Valley 29 municipalities (incl. Arcadia, Rosemead,
Covina, etc.)
Orange
Orange County-north 12 municipalities (incl. Anaheim, Santa Ana,
etc.)
Orange County-south 8 municipalities (incl. Huntington Beach,
Cypress, etc.)
San Bernardino
San Bernardino County-
west
7 municipalities (incl. Rancho Cucamonga,
Rialto, etc.)
San Bernardino County-
east
7 municipalities (incl. Colton, Redlands,
Loma Linda, etc.)
Riverside
Riverside County-north 7 municipalities (incl. Moreno Valley,
Riverside, Perris, etc.)
Riverside County-south 6 municipalities (incl. Temecula, Murrieta,
etc.)
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
102

I then searched in each sub-region and selected those pairs of municipalities with
very different levels of warehousing development. After I double checked the
socioeconomic characteristics of these first-round candidates and eliminated the
unqualified ones—those pairs with certain highly dissimilar socioeconomic
characteristics. For instance, Commerce and Montebello, two neighboring municipalities,
nonetheless have a wide gap in population density, because Commerce has long been an
industrial city while Montebello is more residential-oriented. Divergent industrial
development traditions make these two cities less comparable in terms of policy making.
Therefore, this pair of municipalities was not qualified as case study candidates although
they share many socioeconomic characteristics and indeed experienced different levels of
warehousing development during the observation period. The search ended up with four
pairs of municipalities: Compton vs. Carson (in the Gateway cities-south sub-region),
Pico Rivera vs. San Fe Springs (in the Gateway cities-east sub-region), Upland vs.
Rancho Cucamonga (in the San Bernardino County-west sub-region), and Redlands vs.
Yucaipa (in the San Bernardino County-east sub-region) (see Figure 9).
A summary of socioeconomic characteristics as well as indicators for
warehousing development of these case study municipalities is presented in Table 15.
The table demonstrates that each pair of municipalities share generally similar
socioeconomic characteristics while municipalities across sub-regions are somewhat
different. For instance, cities in Los Angeles County all have older housing stock and
lower percentages of white population than those in San Bernardino County given the
former group had been developed earlier while the latter group had been suburban areas
with lots of white settlers. The general homogeneity within each pair and heterogeneity
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
103

across the pairs ensure the comparability between cases within each pair and the
representation of different types of cities in the region. Table 3 also indicates that each
pair of cities had remarkably different level of warehousing development during the
observation period, raising the question: why do two cities with similar socioeconomic
characteristics widely differ in warehousing development? The eight city cases would
allow me to explore this question in various urban contexts ranging from traditional
industrial space near the port complex, to rising suburban hotspots of warehousing
facilities. Assuming that each pair of municipalities have largely homogeneous physical
and socioeconomic conveniences for the warehousing location choice, I would examine
how local public policies contribute to the variations in warehousing development.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
104

Table 15 Summary of socioeconomic characteristics and indicators for warehousing development of case study municipalities
Pair Pair 1 Pair 2 Pair 3 Pair 4
Subregion Los Angeles County-
Gateway cities South
Los Angeles County-
Gateway cities North
San Bernardino County-West San Bernardino County-
East
City Carson Compton Santa Fe
Springs
Pico Rivera Rancho
Cucamonga
Upland Redlands Yucaipa
Built-up area in warehousing
during 1996-2016 (1,000 sqft)
8,791 180 9,305 253 13,882 155 10,145 0
Population (2000) 89,730 93,493 17,438 63,428 127,743 68,393 63,591 41,207
Land area (acre) 10,161 5,071 4,781 4,739 20,436 8,092 16,923 16,048
Housing rent(2000, $) 754 597 747 700 872 710 689 610
Racial
composition
(2000)
White 12% 1% 19% 8% 54% 55% 63% 76%
Black 25% 39% 3% 1% 8% 7% 4% 1%
Asian 22% 0% 3% 2% 6% 7% 5% 1%
Hispanic 35% 57% 72% 88% 28% 27% 24% 19%
Median year built of housing
stock
1965 1954 1956 1959 1987 1976 1974 1976
Household income
(2000, $)
52,284 31,819 44,540 41,564 60,931 48,734 48,155 39,144
Unemployment rate (2000, %) 7.9 13.7 6.8 7.2 5.4 5.7 6.4 6.4
Share of residents in
transportation and
warehousing occupation (2000,
%)
9.5% 8.0% 8.1% 8.1% 6.7% 5.5% 4.9% 5.8%
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
105

3.4. Data description
To understand the role of public policies in warehousing location choice, I would rely on
qualitative data from different perspectives including city planners, public agency staff,
real estate developers, and local residents and different sources comprised of interviews,
public documents, and field visits. In the interviews with city planners or specialists, I
mainly focused on the history of land use development, the adoption and enforcement of
policy elements that are related to warehousing development, and the attitude of the city
government towards the warehousing industry. Public agency staff and local residents, on
the other hand, provided their opinions on the relationship between institutional factors
and warehousing development. Real estate/property developers were asked about how the
local public policies and regulations affect their location choice making. Public
documents including general plans, zoning ordinances, land use data files, labor related
policies, financial incentives, and environmental standards were studied to examine the
details of various policy elements. The recent general plans of each surveyed
municipality were included in the data and I paid special attention to the sections
regarding economic development, land use, and environmental conservation in those
plans. Regarding the other documents, I in particular focused on the ones mentioned in
the interviews and identified whether there were gaps between the written and perceived
policy elements. With regard to each policy element, I collected as comprehensive and
comparable information as possible for each municipality to avoid any bias in the
comparative case studies. To understand the details of historical land use development, I
obtained land use data files for all surveyed municipalities from the Southern California
Association of Governments (SCAG). For each surveyed city, I paid several visits to its
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
106

industrial (if any warehousing) areas and observed how these areas spatially related to
other land uses. Finally, I reviewed the reactions of the local media outlets that covered
relevant issues including newspapers and websites.
I contacted each municipality in the case studies as well as several major regional
public agencies such as Southern California Association of Governments, the South
Coast Air Quality Management District and Gateway Cities Council of Governments to
set up interviews. I chose two real estate developers, each of which represents a separate
type: the Watsons Land Company as a professional industrial developer which develops
warehousing facilities for renters, and the Kroger Company as a retailing company which
develops warehousing facilities for its own logistics uses. When I visited the industrial
site in the surveyed municipalities, local residents nearby were randomly interviewed. An
interview protocol was created for each category of interviews. After all, seven city
planners, two real estate/property developers, three public agency staff, and three local
residents were interviewed to collect data on how the effects of local policies on
warehousing development were produced and understood from their personal and
organizational perspectives. Table 16 shows where the interviewees are from. Although
not all municipalities and public agencies responded to the interview invitations and
offered relevant information, the outcomes of interviews provided much data for
analyzing the role of public policies on warehousing development. Other data sources,
especially public documents, supplement the interview records by providing objective
written information.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
107

Table 16 Municipalities and institutions that interviewees are from
Category Municipality or institution that interviewees are from
City planners or specialists
Carson, Compton, Santa Fe Springs, Pico Rivera, Upland, Rancho
Cucamonga, and Redlands
Public agency staff
Southern California Association of Governments, South Coast Air
Quality Management District, Gateway Cities Council of Governments
Real estate developer The Watson Land Company, the Kroger Company
Local residents Compton, Rancho Cucamonga

4. Results
4.1. Land use policies
Whether a warehousing cluster develops in a city to a large degree depends on its land
use policies. Land uses, as a gradual urban process, show substantial inertia. The existing
land use patterns largely depend on what had been developed in the past, so these patterns
were more affected by the previous policy making and land use decisions rather than the
current policies. The path dependence of land use development affects various factors in
the warehousing location choice, including land availability, land parcel size, industrial
connections, and redevelopment costs. Like the majority of cities in the region, all the
eight municipalities do not create a separate category for warehousing land uses in the
zoning codes; instead, they simply regard warehousing as one of industrial or
manufacturing types. All the land zoned as industrial is theoretically available for
warehousing development except for any designated or restricted zones. Therefore, the
total land area zoned for industrial is an effective proxy for the land availability for
warehousing development. Table 17 displays the number of land parcels, total land area
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
108

and share of land zoned as industrial in the surveyed cities in 2008
1
. As expected, all
warehousing intensive cities have more land parcels and total land area zoned as
industrial, and thus can provide warehousing developers more options than their
neighbors. Land use patterns, which are an important legacy from the early times of
municipal incorporation, greatly influence the likelihood of warehousing siting.
Table 17 Number of land parcels, total land area and share of land zoned as industrial in the
case study cities in 2008
In addition, a city, which had divided its land into relatively fragmented pieces, is
less likely to provide large land parcels, an essential requisite for the development of
new-generation warehousing facilities. A search in the Costar dataset of warehousing
establishments indicated that the average land area of a warehouse built in the Los
Angeles region during 1996-2016 was around 10 acres. The following Table 18 shows
the number of land parcels that are zoned as industrial and have an area of more than 10
acres in the eight case study cities in 2008. From the table, we can see that Carson, Santa
Fe Springs, Rancho Cucamonga and Redlands all have a significantly higher number or
share of land parcels that could meet the 10-acre threshold than their counterparts in the

1
Year 2008 is regarded as a midpoint of the observation period and the land use data in that year
is regarded as an average status of land use development during the period. I referred to the land
use data in 2008 provided by SCAG.
Pair City
Number of land
parcels zoned as
industrial
Total land area
zoned as industrial
(acre)
Share of industrial-
zoned land
Pair 1
Carson 1013 3,555 35.0%
Compton 838 1,184 23.3%
Pair 2
Santa Fe Springs 1812 3,184 66.6%
Pico Rivera 267 447 9.4%
Pair 3
Rancho Cucamonga 1101 2,464 12.1%
Upland 408 987 12.2%
Pair 4
Redlands 308 938 5.5%
Yucaipa 151 146 0.9%
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
109

pairs. Large industrial-zoned land parcels therefore can earn these cities advantages to
attract warehousing developers, especially in the era of megawarehouses.
Table 18 Number of industrial-zoned parcels with area of more than 10 acres in the eight case
study cities in 2008
The land use patterns result from long-term land use practices and policies. The
significance of historical land use development is confirmed by local planners. Richard
Rojas, Senior Planner in Carson, ascribed the popularity of land with warehousing
developers to the way land was divided and developed during the early times. Before
Carson was incorporated, a large proportion of the city’s land was divided and owned by
the Dominguez family, who had much stronger control of the land development than the
city council then. At that time, the land was open to almost any type of development. The
extensive development of industrial land uses, especially oil extraction land uses, had
largely shaped the land use patterns in which industrial, including warehousing uses, had
a dominant position. Thanks to the land use legacy, the city had a hard time adjusting its
policies to control industrial development, not to mention regulate warehousing related
externalities. To the contrary, Upland, since its birth, had grown up from an agriculture
area to a commercial and residential community. The tradition of encouraging retail and
commercial development also has deeply rooted in the economic and land use
Pair City
Number of
industrial-zoned
land parcels
Number of industrial-
zoned parcels with area
of more than 10 acres
Percentage of
industrial-zoned
parcels with area of
more than 10 acres
Pair 1
Carson 1013 60 5.9%
Compton 838 11 1.3%
Pair 2
Santa Fe Springs 1812 39 2.2%
Pico Rivera 267 7 2.6%
Pair 3
Rancho
Cucamonga
1101 52 4.7%
Upland 408 15 3.7%
Pair 4
Redlands 308 21 6.8%
Yucaipa 151 0 0.0%
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
110

development strategies, according to Jerry Guarracino, Senior Planner of Upland.
Therefore, the history of a city’s development not only largely determines the physical
conveniences for developing warehouses, but also affects the way local planners
understand and propose future developments (Dablanc, 2014).
4.2. Job related policies
Job creation is a key objective that city managers pursue when they make land use
policies and decisions. Warehouses have several characteristics related to employment
generation. First, the majority of employees in warehouses are low- or medium-skilled
workers. These jobs do not require high education, they are not paid well and many of
them are temporary (Kirkham, 2015). Second, warehousing facilities usually have low
job densities than many other industry sectors. An early but highly relevant study
prepared for SCAG by The Natelson Company (2001) showed that the average job
density of warehousing is only around 11 employees/acre, while average job densities of
light manufacturing and heavy manufacturing are 15 and 24 employees/acre respectively.
Average job density of office land uses is even much higher. Third, the gap in job density
between warehousing and other industry sectors may further widen as warehouses are
getting more and more automated. Employment opportunities in warehouses are
increasingly replaced by robots and automation systems. Such transformation is
especially prevalent in suburban areas where large parcels of land is available and
modern automation equipment can be integrated into the warehousing buildings. All in
all, warehouses do generate job opportunities but the quality and security of those jobs
are questionable.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
111

Given these characteristics, local governments nonetheless have different or even
contradictory job related policies on warehousing development. In the interviews,
planners in several cities including Carson, and Santa Fe Springs acknowledged the
importance of warehousing jobs in the economy. Politicians in many Inland Empire cities
are more enthusiastic. They believe “the rapidly evolving industry will increasingly
demand higher-level skills and pay higher salaries” and “have invited the growth of the
industry (of warehousing)” (Kirkham, 2015). Although their belief may not be supported
by the recent trend, the warehousing industry had still gained the favor of many local
policy makers. Those cities in favor of the job opportunities from warehouses had made
policies that help realize these potential benefits. Dean Jones, Senior Economic
Development Specialist of Compton, said the city appreciated that warehouses bring
about low and moderate-skilled jobs such as trucking drivers and had been engaged to
provide qualified residents with employment referrals to these companies. Such policy
can help match jobs with the workforce and reduce the labor costs of warehousing
businesses.
A few other cities paid more attention to job quality. Policies promoting high job
density development turn out to be increasingly popular and influential. Half of the case
study cities (Rancho Cucamonga, Upland, Yucaipa, and Pico Rivera) proposed or
adopted policies that discourage industries with low job density or low pay/stability.
These policies provide cities guidelines to rule out industries like warehousing that are
not consistent with their economic development goals. Rancho Cucamonga, for instance,
encouraged the conversion of warehousing land uses into “higher end” manufacturing
uses, which employ more workers per acre (City of Rancho Cucamonga, 2010).
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
112

Similarly, Upland favors “high quality businesses and job opportunities” and targets
“businesses that provide a high job density in terms of the number of jobs per square foot
of built space to generate the maximum amount of employment” (City of Upland, 2014).
Its general plan even puts an example to illustrate the job density-based strategy, “a
research and development business which employs 10 people over 10,000 square feet
would be prioritized over a warehouse, which might employ 3 people over 50,000 square
feet”. Upland and Yucaipa also combined the orientation towards high job density
development with financial incentives, which will be discussed below.
4.3. Financial incentives
Local governments create financial incentives or disincentives including tax rate
adjustments, tax deductions, exemptions, rebates, and credits, impact fees, and financial
assistance and support to encourage or discourage certain categories of industrial
development. The financial incentives are effective policy tools used to increase or
decrease the costs facing warehousing developers.
Although tax rates are largely determined by high levels of governments (federal
and states), the differences in tax rates between municipalities in the same subregion
could still reflect how different developers are subject to tax liabilities. A comparison in
tax rates between the case study cities illustrates tax rates are likely to contribute to
variations in warehousing development (see Table 19). For instance, compared to
Compton, Carson has lower sales tax, property tax and utility user tax rates. Siting
warehousing facilities in Carson can thus save developers monetary costs on tax
expenses. The comparison between Santa Fe Springs and Pico Rivera reveals a similar
pattern though in a less significant way. The four cities in San Bernardino County have
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
113

the same tax rates in general. Do these cities have any different financial incentives or
disincentives for warehousing development? Upland provided tax rebates, fee reductions
and other financial incentives to prioritize the development of businesses with high job
densities. Yucaipa also offered financial assistance to businesses that lack capital to
expand facilities and operations, but similar to Upland, Yucaipa focused its public
resources on supporting “the retention, expansion, and attraction of businesses that
generate a higher number of jobs per acre and higher net municipal revenues per acre”
(City of Yucaipa, 2016). As stated in the last section, warehouses have significantly low
job densities, thus these financial instruments are essentially disincentives for
warehousing development.
Table 19 Tax rates in effect of case study cities
Pair Pair 1 Pair 2 Pair 3 Pair 4
City Carson Compton Santa Fe
Springs
Pico
Rivera
Rancho
Cucamonga
Upland Redlands Yucaipa
Sales tax rate
8.75% 10.25% 8.75% 9.75% 7.50% 7.75% 7.75% 7.75%
Property tax
rate
1.18% 1.53% 1.10% 1.12% 0.81% 0.81% 0.81% 0.81%
Utility user
tax rate
2.00% 10.00% 5.00% 5.00% 0.00% 0.00% 0.00% 0.00%
Another type of disincentives, impact fees are also found in the policy package of
surveyed cities. Impact fees are a common financial disincentive but very few cities had
developed designated impact fees for warehousing development. With the growth in
warehousing related externalities, cities have been urged to adopt impact fees to repair
the local roads that are damaged by frequent truck movement (Slowik, 2017). Among the
case study cities, Santa Fe Springs is the only one that requires warehousing developers
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
114

to pay one-time traffic impact and street resurface fees (Cuong Nguyen, Senior Planner of
Santa Fe Springs). These fees, however, are still inadequate to cover the long-term
impacts of warehousing activities, according to Cuong Nguyen.
The interviews with local planners and warehousing developers further revealed a
critical question: how do local governments understand the tax contribution of
warehouses and make relevant policies? The tax contribution nature of warehouses
affects not only the way tax rates and other financial incentives function in warehousing
location choice, but also the willingness of local governments to make tradeoffs in
making other policies. Different types of warehouses have different tax liabilities and
such differences are becoming more complicated as the functions of warehouses are
changing. It is traditionally understood that warehousing facilities, which typically
provide storage services, create no sales taxes and overall fairly low per-acre tax revenues
(Dablanc, 2014). Such perception resonated with the opinions of quite a few local
planners. For instance, Senior Planner of Santa Fe Springs, Cuong Nguyen mentioned
that the city did not accept the proposal of warehousing development particularly for the
purpose of tax generation, as “that type of projects usually are not major tax
contributors”. The rise of E-commerce, however, has been changing the views on the
relationship between warehousing and tax revenue. Many distribution or fulfillment
centers not only store items but also process the sale of products and directly connect
sellers and customers, and thus they have sales tax liabilities in states including California
(Taxjar, 2018.) and New York. The rapid increase in the number of fulfillment centers
creates incentives for municipalities to reevaluate their tax policies so as to bolster sales
tax revenue from these new facilities (City of Yucaipa, 2015). Santa Fe Springs also
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
115

encouraged fulfillment center-like warehouses that provide sales in addition to storage
services so that the city can benefit from the sales taxes, according to Cuong Nguyen.
Therefore, similar to job opportunities, the changing views of local authorities on tax
revenue from warehousing development would impact the way they make and adjust
public policies.
4.4. Environmental regulations
Environmental regulations are sometimes ignored by researchers when they evaluate the
factors for the warehousing location choice. As local residents have become more and
more aware of the environmental impacts associated with warehousing activities, they
have been nudging the governments into solving the problem. Residents in the city of
Moreno Valley, a rising “superstar” of warehousing development in the Los Angeles
region, have been opposing the construction of the World Logistics Center, a 40-million-
square-foot warehouse project (Emma Foehringer Merchant., 2017). The negotiation
between residents and the city council has continued for years and a consensus that
satisfies all stakeholders has not been achieved yet. Meanwhile, environmental agencies
also get involved in this issue and persuade policy makers to take actions. In the World
Logistics Center case, regional agencies including South Coast Air Quality Management
District (SCAQMD), and Riverside County as well as environmental groups including
EarthJustice, the Center for Biological Diversity, and the Coalition for Clean Air have all
stepped in to challenge the project (Forsyth, 2015). Other cities (e.g. Carson, and
Compton) also received policy guidance letters regarding new warehousing development
from the SCAQMD (Richard Rojas, and Dean Jones). Under the pressure, how do local
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
116

planners perceive the environmental issues? What have they done about the rapid
expansion of warehousing related environmental externalities?
Unfortunately, the majority of the case study cities so far have not taken seriously
the excessive warehousing development as a major environmental problem; they either
failed to realize the environmental threats of warehousing activities,or overlooked the
role of local policy making in causing this environmental problem. The apathy of local
officials has been found to be one of the major barriers for maintaining the quality of the
environment and achieving sustainable development (Saha and Paterson, 2008).
According to Richard Rojas, there has been very limited understanding of environmental
externalities associated with warehousing development and “a lot of cities do not know
the impacts or what other options could be”. As a result, local policy makers have been
indifferent to warehousing development from the perspective of environmental
consequences.
In spite of the generally passive attitudes, some cities have adopted strategies for
regulating the warehousing related impacts. First, to restrict or discourage new
development of warehousing land uses would be the most straightforward approach. To
create special zones for restricting warehousing uses is an effective policy alternative.
Pico Rivera has been exploring the likelihood of developing a truck intensive overlay
zone to “further manage the location and concentration of trucking uses so as to better
mitigate noise, traffic and circulation, air pollution and other impacts to adjacent or
nearby sensitive land uses” (City of Pico Rivera, 2014). Second, to improve the
environmental performance of warehousing facilities and equipment can also effectively
reduce the impacts. Compton, as an outstanding example, encourages eco-friendly
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
117

warehouses (See Figure 11). The city had been working with neighboring cities to
promote the use of clean trucks in warehouses (ComptonHerald, 2015), after it was fined
for diesel truck and equipment violations in 2011 (California Air Resource Board, 2011).
Third, to physically isolate warehousing activities from other land uses especially
residential land uses could minimize the conflicts. Most cities have municipal codes that
require buffers or separation of residential/commercial land uses from industrial uses,
especially those generating environmental hazards. These regulations are particularly
crucial for minimizing the exposure of local population to the environmental hazards.
However, quite a few cities have not included warehouses in the category of industrial
uses that are locally detrimental, or have failed to updated their standards given the ever-
growing size of warehousing facilities and the subsequent environmental impacts. For
instance, Richard Rojas admitted that the current buffering regulations in the municipal
codes of Carson were outdated as the buffer distance required for warehousing facilities
from residential uses is only 100 feet, while the recommended standard by the SCAQMD
is 1,000 feet (e.g. Santa Fe Springs had adopted the 1,000-feet buffer in their codes).
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
118

Figure 11 The exterior of the Kroger Company Food distribution center in Compton (Photo
credit: Quan Yuan)
The interviews demonstrated that some cities have realized the necessity of
adjusting policies to address warehousing related impacts, but they have nonetheless
found it difficult to make a real difference in the short term. Carson is among the few
municipalities that are reconsidering their warehousing development history and
searching policy options to prevent potential negative environmental consequences,
although it is still struggling to work out an immediately effect way to do that. Carson,
which was incorporated in 1968, is a very young city in Los Angeles County. Even since
its incorporation, the city had been dealing with quite a few LULUs including garbage
dumps and waste treatment plants. These facilities were “Not In My Back Yard”
(“N.I.M.B.Y.”) legacy from the old times when Carson was an unincorporated area and
lacked the political clout to resist their sitings. Even after many years’ efforts to replace
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
119

those facilities, other contaminated industrial sites still act as a hurdle to many new
developments (City of Carson, n.d.). In these circumstances, much of the land in Carson
had remained used as industrial and warehousing had been growing rapidly in the recent
decades. As with the LULUs in the early days, local residents noticed the environmental
threats of the warehousing activities. The pressure from environmental agencies
including the SCAQMD also drove the local authority to step in. In front of these policy
makers were the strong demand for warehousing development, the increasing conflicts
between warehousing and residential land uses, as well as the limited information on
resolving the conflicts. The city had attempted to learn from other cities with intensive
warehousing development, but successful cases turned out to be scarce. Therefore, the
city decided to take a step back and do some research first. To make sure things would
not go worse, the city adopted a 45-day Temporary Moratorium on Truck Yards,
Logistics Facilities, Hazardous Materials, or Waste Facilities, Container Storage, and
Container Parking in 2016 (City of Carson, 2017). According to the moratorium, new
warehousing development was banned during the period. The temporary ban was later
extended to March 2018 as city leaders needed more time to work out “new, permanent
ground rules to protect residents from heavy truck traffic and pollution” (Mazza, 2017)
2
.
Among all case study cities, Carson is the only one which had adopted designated
restrictions on warehousing development, and more actions are expected to follow.
Most municipalities have remained silent about the environmental implications of
warehouses. Without effective interventions from local governments, the warehousing

2
https://www.dailybreeze.com/2017/05/03/carson-extends-ban-on-industrial-growth-through-
2018/
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
120

related impacts would not only continue spreading out, but also disproportionately
concentrate in certain cities.
5. Discussion
5.1. A synthesis of policies
Compiling all the results together, I found that the significant gap in warehousing
development between two cities in each pair is likely to be caused by their different sets
of local public policies. If each city has its clear-cut preferences over warehousing
development, its policy making would be greatly affected by such preferences. But
according to the case studies, most surveyed cities did not have a distinct attitude towards
warehousing, given their mixed opinions on the subsequent benefits and costs from
massive warehousing development. Therefore, the policies adopted in each city are likely
to result from discrete considerations of warehousing related benefits or costs, instead of
a consistent strategy on supporting or regulating this type of land uses. In terms of
attracting or preventing warehousing location, the policies do not have to all work in the
same direction. Nor do they produce effects to the same degree. Table 20 presents all
aforementioned local policy elements by city and policy category. The overall picture
supports the hypothesis that the gaps in warehousing development levels are largely
caused by differences in local policies; when pro-warehousing policies dominate in a
city, it probably develops into warehousing hot spots. Among all the policy categories,
land use policies appear to be the most influential one but, in some cases, other policies
can dominate the overall policy effects as well.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
121

Table 20 A synthesis of warehousing related public policies adopted in case study cities
Pair Pair 1 Pair 2 Pair 3 Pair 4
Policy category Carson Compton Santa Fe
Springs
Pico
Rivera
Rancho
Cucamonga
Upland Redlands Yucaipa
Land use policies
Pro-warehousing (High
availability of industrial-zoned
land, large land parcels, etc.)
• • • •
Anti-warehousing (Low
availability of industrial-zoned
land, small land parcels, etc.)
• • • •
Job related
policies
Pro-warehousing (Job-matching
policies, etc.)
•
Anti-warehousing (Job density
requirements, etc.)
• • • •
Financial
incentives
/disincentives
Pro-warehousing (Low tax rates,
tax deductions, etc.)
• •
Anti-warehousing (High tax rates,
impact fees, etc.)
• • • • •
Environmental
regulations
Pro-warehousing
(Outdated standards, etc.)
• • • • • • •
Anti-warehousing (truck intensive
overlay zones, etc.)
• • •
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
122

A close examination of paired cities may provide more explicit evidence on the
findings above. For instance, the gap in warehousing development level between Carson
and Compton in fact did not emerge until late 1990s (see Figure 12). Apart from certain
socioeconomic factors such as crime rates, local public policies are found to be one of the
major causes of the gap. In particular, due to its legacy of land use policies, Carson has
more industrial-zoned land and many large land parcels, which are important
conveniences for developing large-size modern warehouses. Comparably, its neighbor
Compton has much fewer industrial sites with large land parcels. Such difference, in
addition to differences in other policy elements, might to a significant degree explain the
divergence in warehousing development intensity between the two cities since late 1990s
when megawarehouses became increasingly popular.

Figure 12 Accumulated built-up area of newly built warehouses in Carson and Compton until
2016 (data source: Costar Inc.)
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
123

Figure 13 shows the land use patterns in Upland and its counterpart Rancho
Cucamonga in 2008. Located in the same sub-region, the two cities have similar overall
patterns except for industrial land uses. In both cities, industrial land uses including
warehousing account for 12% of all land uses. However, 5% out of the 12% in Rancho
Cucamonga is warehousing, while warehousing accounts for almost none of the 12% in
Upland. This sharp contrast in land use patterns suggests that Upland apparently favors
other industrial land uses over warehousing. Such a significant difference may not be
fully explained by industrial zoning and size of land parcels, provided that these two
cities do not differ from each other as much as in other pairs (see Table 4 and 5). Instead,
the policies promoting high job density could rather be a key to understanding the
difference. The minimized share of warehousing in the entire industrial land uses in
Upland is probably related to the perception of the city that warehousing is the least
efficient type of land use in terms of job generation. Corresponding financial incentives
and other policies drive the inclined industrial development.

Figure 13 Land use patterns in Rancho Cucamonga and Upland in 2008 (data source:
Southern California Association of Governments)

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
124

5.2. Tradeoffs made to make policies
Interviews with local planners and the examination of general plans both revealed that
cities share some common objectives when they make policies; two major goals are job
opportunities and tax revenue. According to the interviewees, prior to policy making,
municipalities need to evaluate their own advantages and disadvantages in attracting
different businesses, assess their capabilities of promoting certain sectors, and consider
the opinions of stakeholders including land owners, residents, nearby business owners
and so forth. Given the limited economic, spatial, and political resources, cities have to
make tradeoffs—in many case identify priorities—when creating the strategies and
policies for economic and land use developing. Moreover, cities differ in these resources,
so they may end up with different arrangements of priorities.
For instance, Upland and Yucaipa, both of which were famous for agriculture
industry, especially fruit planting in the early-20
th
Century. During the post-World War II
period, they, like other cities in the San Bernardino County, replaced citrus and apple
groves into residential areas. But unlike many of their neighbors, Upland and Yucaipa
had kept their community identity—focusing on local residents’ quality of life and small-
town character. To maintain such identity, these two cities had prioritized industry sectors
that provide the most benefits and opportunities to local residents. The warehousing
industry with low job density was not? welcomed, in spite of the strong demand over the
past two decades. While the two cities give up the job opportunities and tax revenue from
warehouses, they avoid the environmental impacts as well.
As the cases of Upland and Yucaipa suggest, local residents matter when
governments make tradeoffs. It also applies to cities where warehousing related
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
125

environmental threats have caused increasing complaints from nearby residents. The
temporary pause for warehousing development in the City of Carson was originally
driven by a small segment of residents who urged the city to take actions on the growing
externalities. The pause to consider the next steps definitely helps, but the major
challenges remain. One of the challenges is the monetary and time costs of updating the
current codes, ordinances and policies. Per Richard Rojas, updating the general plan and
zoning codes to better regulate warehousing related impacts would require quite a few
years and cost around one million dollars. These costs are substantial to the city, which
has an annual revenue of around 2 million now. Given the difficulties, the process of
formally developing new policies turns out to be complicated and slow. To go through a
discretionary process to create some handy rules is a more realistic way, but it also needs
a stronger support from local communities and regional public agencies. After all, the
city has to make tradeoffs between short-term and long term goals, as well as between the
benefits and costs of generating policies to regulate warehousing development.
5.3. It is about the history; yet it is also about the future
When talking about local public policy making, I frequently referred to history, traditions
and path dependence. The effects of public policies are long-term, and sometimes inertial
and lagged. Therefore, the current land use and industrial development patterns are the
heritage of policies that were adopted and in effect during the last several decades. For
example, Carson and Rancho Cucamonga are both outstanding warehousing hot spots at
the moment. What do they have in common in terms of land use development
trajectories? They are both relatively young municipalities in the region; Carson was
incorporated in 1968 and Rancho Cucamonga became a city in 1977. Before those dates,
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
126

the two places as unincorporated areas suffered from uncontrolled development of the so-
called “society needs facilities” such as garbage dumps. Without adequate representation
and effective organization of residents under the umbrella of an incorporated city, these
developments, were much more likely to be located in unincorporated areas. Once
incorporated, the two cities had to cope with the consequences of historical land use
practices. As a matter of expedience, they had much of the land used by the early LULUs
replaced by less polluting industrial land uses instead of residential or commercial, due to
the remaining environmental concerns. The intensive warehousing development may be
partially ascribed to the policies following the land use legacy. With the experience of
dealing with the locally undesirable land uses, Carson and Rancho Cucamonga in fact
had recently considered adjusting policies to regulate excessive warehousing
development. Compared to Redlands and several other cities in the Inland Empire, these
two cities apparently slowed down the expansion of warehousing facilities during the last
ten years (see Figure 14). Although both cities still have a very large stock of
warehousing space, this could be a sign of policy reorientation that targets better
management of this type of land uses.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
127

Figure 14 Accumulated built-up area of newly built warehouses in selected cities during 1966-
2016 (data source: Costar Inc.)
The cases of Carson and Rancho Cucamonga indicate that history matters a lot in
evaluating how local policies affect warehousing development. Those stories also suggest
local governments could start from now to consider making appropriate policies to
change the future. If they stay indifferent and passive, and make no actions to emerging
environmental threats, things may get worse than they think. In only twenty years
between 1996 and 2016, more than 10 million square feet warehousing space had been
added to many cities, thanks to the rising demand for warehousing services. Subsequent
environmental concerns arise in quite a few of those warehousing intensive cities.
Another important lesson from the longitudinal analysis on the relationship
between warehousing development and local policies is the ever-changing nature of
industries and their environmental and social implications. Whether warehouses deserve
particular attention may depends on the environmental threats of these facilities and the
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
128

exposure of local residents to these threats. Both of the two factors change over time and
are greatly affected by local policies. If local governments can keep track of the way
industries change and generate externalities that impact nearby communities, they would
be able to develop effective policies so as to address any potential negative consequences.
6. Implications on Environmental Justice Issues
The environmental justice problem is one of those potential negative consequences facing
local municipalities. If people of color or lower socioeconomic status suffer from more
warehousing related environmental externalities, what role do local policies play in
creating such spatial inequity? How do the policy elements discussed above relate to the
environmental justice problem?
The results from the collected qualitative data do not indicate a significant causal
relationship between local policies and the environmental justice problem in warehousing
location during the observation period. There is no evidence showing that local policies
have played a major role in linking low-income and minority population with
warehousing related externalities. Although cities have adopted various policy elements
that strongly influence the warehousing development, income and race are found
generally irrelevant to policy making across cities. This finding has much to do with the
neutral attitude of local governments towards warehousing during the observation period.
Many planners admitted that their municipalities had been generally indifferent to
warehousing relative to other industrial types, and there had been very limited
understanding of warehousing related environmental implications among these
interviewees. Given their neutrality, intentionally biased policy and land use decision
making had not been frequently seen in warehousing siting.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
129

However, the null results may be a matter of timing. As the growing warehousing
related environmental threats have caused more and more concerns among local
residents, local policies can potentially become a major drive towards environmental
inequity in warehousing location. According to the discussion above, local policy making
is highly subject to residents’ point of view and political actions. With increasing
awareness of environmental externalities from warehouses, more affluent and educated
residents are more likely to organize and lobby the local authorities for anti-warehousing
policies. Anti-warehousing policies can largely alter the socioeconomic environment
facing the warehousing developers. Local resident groups may persuade governments to
reduce industrial zoning, convert industrial land to more fragmented parcels that support
residential and commercial, add job density thresholds on industrial development, levy
impact fees, and require more stringent rules on facility design and landscaping. These
policy options would discourage warehousing development by reducing availability and
adding costs.
A city with stronger political will and power thus is more likely to avoid
warehousing related externalities. Variations in demographic and socioeconomic
characteristics including income, race, and education attainment may eventually be
translated into variations in warehousing related policy making and intensity of
warehousing development. The currently significant gap in warehousing externalities
between municipalities may even widen, making the environmental justice problem more
severe. Then what should local governments do to cope with the problem after all? First,
they should get familiar with how warehousing development is affected by local policies,
and how warehousing causes environmental hazards. By linking the knowledge together,
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
130

local authorities would be able to identify where the most critical issues lie in their cities.
Second, local governments need to review the current policies, evaluate their
socioeconomic resources and capacities, and make practical plans to balance the benefits
and costs from warehousing development. Third, local governments may need to work
together to reduce the spatial inequity in warehousing location. Regional collaborations
can help address the problem. More policy restrictions in one municipality may drive the
demand for warehousing development to its neighbors, aggravating the environmental
inequity. A consensus on regulating warehousing related externalities at the regional level
is thus particularly in need. Cities can learn from each other and adopt the most
appropriate policies according to their own circumstances. Finally, local governments
may consider establishing a policy discussion forum where residents, environmental
agencies, regional organizations, and the private sector can sit together to generate
policies that are both acceptable to all parties and effective in achieving the shared goals.
By mixing the voices from different perspectives, such a forum can inform local
authorities where the conflicts of interests are and how to resolve them. More transparent
communication would particularly help cities with limited economic and political
resources to fight against disproportionate warehousing siting.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
131

CHAPTER 5
Conclusions
This three-essay dissertation, as an exploratory research series, aims to provide
theoretical thinking, empirical evidence, and policy discussion on a rising problem in the
major metropolitan areas—the environmental inequity in warehousing location. The
essays examine the problem following a coherent theoretical framework, test hypotheses
using different research methods and data, and generate results that are inherently
consistent and mutually supported. In this chapter, I will summary the findings as an end
of the dissertation.
In the Essay One, I started from the changing location factors of warehouses and
subsequent expansion of warehousing related environmental externalities. Given the
externalities, I proposed the hypothesis that warehousing facilities are disproportionately
located in social disadvantaged neighborhoods. To test whether the environmental
inequity has become a significant phenomenon, I developed a set of econometric models
using data for the Los Angeles region, which is a major trade gateway in the North
America. Considering warehousing facilities are widespread and differ in sizes, I used
three different forms of indicators to measure warehousing distribution. Results showed
that warehouses are more likely to be located in minority neighborhoods, with variables
including transport access, industrial connections, and neighborhood economic attributes
controlled. However, the relationship between warehousing location and household
income is less significant. These results confirmed the environmental inequity in
warehousing location and demonstrated that minority population are particularly
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
132

vulnerable to such inequity. While the Essay One examined the cross-sectional patterns
of warehousing distribution in relation to minority and low-income communities, the co-
location patterns only displayed the socioeconomic outcomes of the interactions between
related agents including warehousing developers and those vulnerable people. Therefore,
a careful examination of the dynamic processes that lead to these outcomes was needed to
discover the causes of environmental inequity.
In the Essay Two, I took a step further to investigate how the environmental
inequity in warehousing location is created. I focused on the location choice behaviors of
warehousing developers and minority population and used the Simultaneous Equation
Model to estimate the interdependent relationship between the two location choice
processes. Results showed that neighborhoods with minority inflows would receive a
higher level of warehousing development, but neighborhoods with intense warehousing
development nonetheless would not attract minority population. The findings
demonstrated that the spatial coincidence between warehouses and the minorities
principally if not solely results from the disparate siting of warehouses instead of the
move-in of minority population. While these results explained how warehouses are
disproportionately located in minority neighborhoods, they have strong policy
implications. In particular, to properly regulate warehousing development turns out to be
an effective strategy for mitigating the environmental inequity. The public sector,
especially local governments who primarily make land use decisions should take the
responsibility. How local governments can create, or on the contrary alleviate the
environmental justice problem is, however, largely unknown yet. In the Essay One and
Two, I explored the roles of various location factors in warehousing location choice, but
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
133

local public policies were not incorporated into the models due to the lack of adequate
and accurate data. It would be rather difficult to quantitatively measure the effects of
public policies. Alternatively, a qualitative study that examines such effects could be
rather helpful to shed some light on the question.
Thus, in the Essay Three, I shifted my focus from neighborhoods to municipalities
and relied on qualitative data to explore how local public policies may contribute to the
environmental inequity in warehousing location or help address the problem. I selected
four pairs of municipalities in the region of Los Angeles to conduct a comparative case
study. Two municipalities in each pair have similar socioeconomic characteristics and are
located next to each other, but they differ substantially in the warehousing development
trajectories. By controlling other factors as much as possible, I was able to identify the
relationship between local policies and warehousing development. Land use policies, job-
related policies, financial incentives, and environmental regulations are found to be major
policy elements that affect warehousing development, though in different ways and to
different degrees. I further attempted to link the local policy making to warehousing
related environmental inequity but concluded that the current patterns of disproportionate
warehousing distribution had little to do with local governments. This is primarily
because most local authorities so far have been generally neutral about the warehousing
industry and they have limited understanding of the environmental and social
consequences following massive warehousing development. However, given the trend of
disparate warehousing distribution revealed in the first two essays, local public policy
making may become an increasingly critical part of the solutions to warehousing related
environmental justice problem.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
134

The solutions to the problem depend on the collaboration of all relevant agents
including different levels of—federal, state, and local—governments, regional public
agencies, warehousing developers, industry associations, and local residents. The three
essays had provided a set of policy implications based on the empirical findings.
First, all these agents need to better understand and recognize the problem. The
environmental inequity in warehousing location has attracted increasing attention, but the
efforts to formally and systematically evaluate the problem from the perspective of the
public sector have been inadequate. Federal or state governments may consider
integrating warehousing location into the traditional environmental justice monitoring
framework and offer support for relevant education and training. To alleviate the
environmental inequity is crucial to avoiding land use conflicts and achieving
environmental and social sustainability. Each of the related agents can significantly
benefit from it.
Second, information access is highly important in combating environmental
inequity. Local residents, especially those vulnerable population should be fully informed
of warehousing development and subsequent environmental consequences. This is an
effective way to empower the socially disadvantaged people. The elimination of any
information asymmetry between agents would also be a prerequisite to effective
communication and problem solving. A public forum that includes the private sector, the
public sector, and local residents can help establish a network of information sharing and
exchange.
Third, regional collaboration is required to address the warehousing related
environmental inequity, which is essentially a regional problem. The disproportionate
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
135

distribution of warehousing facilities in minority neighborhoods may largely result from
the disparities in economic, social, and political resources within a region. Therefore, in
order to mitigate such disparities, the public sector have to create region-wide consistent
policies or regulations on warehousing development. A lack of coordination between
municipalities and communities could end up with even more severe spatial inequity.
Higher levels of governments and public agencies can work as coordinators to facilitate
the regional cooperation.
Finally, innovations in technologies and policy practices are also helpful. There
are many ways to mitigate the environmental inequity in warehousing location by using
new techniques or adopting innovative policies. Creating isolation and buffering by
redesigning the road networks, encouraging environmental-friendly warehousing
structures and sites, and upgrading to green vehicles are all possible options. The public
sector and the private sector need to work together to find a correct path towards better
management of warehousing related externalities and the minimization of the impacts on
local communities.

Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
136

References
Acemoglu, D., & Restrepo, P. (2017). Robots and jobs: Evidence from US labor markets.
NBER Working Paper No. 23285. Available at http://www.nber.org/papers/w23285
Ackerman, K. B. (2012). Practical handbook of warehousing. Springer Science &
Business Media.
Aljohani, K., & Thompson, R. G. (2016). Impacts of logistics sprawl on the urban
environment and logistics: Taxonomy and review of literature. Journal of Transport
Geography, 57, 255-263.
Alba, R. D., Logan, J. R., Stults, B. J., Marzan, G., & Zhang, W. (1999). Immigrant
groups in the suburbs: A reexamination of suburbanization and spatial assimilation.
American sociological review, 446-460.
Allen, J., Thorne, G., & Browne, M. (2007). BESTUFS good practice guide on urban
freight transport.
Allen, J., Browne, M., & Cherrett, T. (2012). Investigating relationships between road
freight transport, facility location, logistics management and urban form. Journal of
Transport Geography, 24, 45-57.
Anderson, A. B., Anderton, D. L., & Oakes, J. M. (1994). Environmental equity:
Evaluating TSDF siting over the past two decades. Waste Age;(United States), 25(7).
Anderson, S., Allen, J., & Browne, M. (2005). Urban logistics––how can it meet policy
makers’ sustainability objectives? Journal of transport geography, 13(1), 71-81.
Andreoli, D., Goodchild, A., & Vitasek, K. (2010). The rise of mega distribution centers
and the impact on logistical uncertainty. Transportation Letters, 2(2), 75-88.
Aniello, C., Morgan, K., Busbey, A., & Newland, L. (1995). Mapping micro-urban heat
islands using Landsat TM and a GIS. Computers & Geosciences, 21(8), 965-969.
Bai, Y., Schrock, S. D., Mulinazzi, T. E., Hou, W., Liu, C., & Firman, U. (2010).
Estimating highway pavement damage costs attributed to truck traffic.
Bayoh, I., Irwin, E. G., & Haab, T. (2006). Determinants of residential location choice:
How important are local public goods in attracting homeowners to central city locations?
Journal of Regional Science, 46(1), 97-120.
Been, V. (1994). Locally undesirable land uses in minority neighborhoods:
Disproportionate siting or market dynamics? Yale Law Journal, 1383-1422.
Been, V., & Gupta, F. (1997). Coming to the nuisance or going to the barrios? A
longitudinal analysis of environmental justice claims. Ecology law quarterly, 24(1), 1-56.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
137

Bergqvist, R. (2016). Transport and logistics facilities expansion and social
sustainability: A critical discussion and findings from the City of Gothenburg, Sweden.
Bluffstone, R. A., & Ouderkirk, B. (2007). Warehouses, trucks, and PM2. 5: Human
health and logistics industry growth in the eastern Inland Empire. Contemporary
Economic Policy, 25(1), 79-91.
Boarnet, M. G., Hong, A., & Santiago-Bartolomei, R. (2017). Urban spatial structure,
employment subcenters, and freight travel. Journal of Transport Geography, 60, 267-276.
Boer, J. T., Pastor, M., Sadd, J. L., & Snyder, L. D. (1997). Is there environmental
racism? The demographics of hazardous waste in Los Angeles County. Social Science
Quarterly, 78(4), 793-810.
Bowen Jr, J. T. (2008). Moving places: the geography of warehousing in the US. Journal
of Transport Geography, 16(6), 379-387.
Browne, M., Allen, J., Nemoto, T., Patier, D., & Visser, J. (2012). Reducing social and
environmental impacts of urban freight transport: A review of some major cities.
Procedia-Social and Behavioral Sciences, 39, 19-33.
Bullard, R. D. (1996). Environmental justice: It's more than waste facility siting. Social
science quarterly, 77(3), 493-499.
Bullard, R. D. (2001). Environmental justice in the 21st century: Race still matters.
Phylon (1960-), 49(3/4), 151-171.
Bullard, R. D. (2008). Dumping in Dixie: Race, class, and environmental quality.
Westview Press.
Bullard, R. D., Grigsby, J. E., & Lee, C. (1994). Residential apartheid: the American
legacy. Los Angles: UCLA Center for Afro-American Studies
Bullard, R. D., Mohai, P., Saha, R., & Wright, B. (2008). Toxic wastes and race at
twenty: Why race still matters after all of these years. Environmental Law, 371-411.
Bullard, R., & Wright, B. H. (1987). Environmentalism and the politics of equity:
emergent trends in the black community. Mid-American Review of Sociology, 12(2), 21-
37.
Bullard, R. D., & Wright, B. H. (1993). Environmental Justice for all: Community
Perspectives on Health and Research. Toxicology and industrial health, 9(5), 821-841.
Bureau of Labor Statistics. (2016) Industries at a Glance: NAICS Code Index. Available
at http://www.bls.gov/iag/tgs/iag_index_naics.htm
Bureau of Labor Statistics (2017). Employment, Hours, and Earnings – National.
Available at https://www.bls.gov/data/
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
138

California Air Resources Board. (2011). City of Compton fined $48,000 for diesel truck
and equipment violations. Available at
https://www.arb.ca.gov/newsrel/newsrelease.php?id=194
California Air Resources Board. (2017). Implementation of March 2017 Board Direction
on Reducing the Community Health Impacts from Freight Facilities. Available at
https://www.arb.ca.gov/gmp/sfti/reducing_the_community_health_impact.pdf
Chang, L. Y., & Mannering, F. (1999). Analysis of injury severity and vehicle occupancy
in truck-and non-truck-involved accidents. Accident Analysis & Prevention, 31(5), 579-
592.
Christensen Associates, Grow, Bruening, Pett, K.H.S., (2012, April). Preserving and
Protecting Freight Infrastructure and Routes. In 2012 Joint Rail Conference (pp. 533-
542). American Society of Mechanical Engineers.
Cidell, J. (2011). Distribution centers among the rooftops: the global logistics network
meets the suburban spatial imaginary. International Journal of Urban and Regional
Research, 35(4), 832-851.
Cidell, J. (2015). Distribution centers as distributed places. Cargomobilities: Moving
Materials in a Global Age. New York: Routledge, 17-34.
City of Carson. (n.d.) Growing Pains of a Young City. City of Carson Website. Available
at http://ci.carson.ca.us/AboutCarson/GrowingPains.aspx
City of Carson. (2017) Logistics Facilities Moratorium. City of Carson Website.
Available at http://ci.carson.ca.us/CommunityDevelopment/AdHocMoratorium.aspx
City of Pico Rivera. (2014) Pico Rivera General Plan. Available at http://www.pico-
rivera.org/depts/ced/planning/plan.asp
City of Rancho Cucamonga. (2010). Rancho Cucamonga General Plan. Available at
https://www.cityofrc.us/cityhall/planning/genplan.asp
City of Upland. (2014). Upland General Plan. Available at
https://www.ci.upland.ca.us/#General_Plan_&_Map
City of Yucaipa. (2016) Yucaipa General Plan. Available at
http://yucaipa.org/development/general-plan/
Clifton, K. J., Burnier, C. V., & Akar, G. (2009). Severity of injury resulting from
pedestrian–vehicle crashes: What can we learn from examining the built environment?
Transportation research part D: transport and environment, 14(6), 425-436.
Cole, L. W., & Foster, S. R. (2001). From the ground up: Environmental racism and the
rise of the environmental justice movement. NYU Press.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
139

Compton Herald. (2015). Brown, other mayors target truck freeway pollution.
Comptonherald.org. Available at http://comptonherald.org/mayors-target-truck-freeway-
pollution
Crowder, K., & Downey, L. (2010). Interneighborhood migration, race, and
environmental hazards: Modeling microlevel processes of environmental inequality.
American Journal of Sociology, 115(4), 1110-1149.
Dablanc, L. (2013). Commercial goods transport, Paris, France. Case study prepared for
global report on human settlements.
Dablanc, L. (2014). Logistics sprawl and urban freight planning issues in a major
gateway city. In Sustainable urban logistics: Concepts, methods and information systems
(pp. 49-69). Springer, Berlin, Heidelberg.
Dablanc, L., Giuliano, G., Holliday, K., & O'Brien, T. (2013). Best practices in urban
freight management: Lessons from an international survey. Transportation Research
Record: Journal of the Transportation Research Board, (2379), 29-38.
Dablanc, L., Ogilvie, S., & Goodchild, A. (2014). Logistics sprawl: differential
warehousing development patterns in Los Angeles, California, and Seattle, Washington.
Transportation Research Record: Journal of the Transportation Research Board, (2410),
105-112.
Dablanc, L., & Rakotonarivo, D. (2010). The impacts of logistics sprawl: How does the
location of parcel transport terminals affect the energy efficiency of goods’ movements in
Paris and what can we do about it?. Procedia-Social and Behavioral Sciences, 2(3), 6087-
6096.
Dablanc, L., & Ross, C. (2012). Atlanta: a mega logistics center in the Piedmont Atlantic
Megaregion (PAM). Journal of transport geography, 24, 432-442.
Demirel, T., Demirel, N. Ç., & Kahraman, C. (2010). Multi-criteria warehouse location
selection using Choquet integral. Expert Systems with Applications, 37(5), 3943-3952.
Dessouky, M., Giuliano, G., & Moore II, J. E. (2008). Selected papers from the National
Urban Freight conference. Transportation Research Part E: Logistics and Transportation
Review, 44(2), 181-184.
Dong, Q., Huang, B., Shu, X., Zhou, C., & Maxwell, J. (2014). Use of Finite Element
Analysis and Fatigue Failure Model to Estimate Costs of Pavement Damage Caused by
Heavy Vehicles. Transportation Research Record: Journal of the Transportation Research
Board, (2455), 54-62.
Durmuş, A., & Turk, S. S. (2014). Factors influencing location selection of warehouses at
the intra-urban level: istanbul case. European Planning Studies, 22(2), 268-292.
Ecklund, D. (2010). Warehousing Efficiency and Effectiveness in the Supply Chain
Process. Available at
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
140

http://www.scmr.com/article/warehousing_efficiency_and_effectiveness_in_the_supply_
chain_process
Emma Foehringer Merchant. (2017). How the ‘Orange Empire’ became the land of
megawarehouses. Wired.com. Available at https://www.wired.com/2017/05/orange-
empire-became-land-megawarehouses
Esquivel, P. (2015). Massive Warehouse Plan has Moreno Valley Divided over Jobs,
Environment. LATimes.com. Available at
http://www.latimes.com/business/realestate/commercial/la-me-world-logistics-20150628-
story.html
Forsyth E. (2015). Coalition Calls for EPA to Protect the Health of Southern California
Residents. Earthjustice.org. Available at https://earthjustice.org/blog/2015-april/coalition-
calls-for-epa-to-protect-the-health-of-southern-california-residents
Freight Analysis Framework (n.d.). Available at
http://faf.ornl.gov/fafweb/Extraction0.aspx
Giuliano, G. (1995). The weakening transportation-land use connection. ACCESS, 1(6),
3-11.
Giuliano, G. (2004). Land use impacts of transportation investments. The geography of
urban transportation, 3, 237-273.
Giuliano, G. (2013). Synthesis of freight research in urban transportation planning (Vol.
23). Transportation Research Board.
Giuliano, G., Kang, S., & Yuan, Q. (2015). Using proxies to describe the metropolitan
freight landscape. Urban Studies, 0042098017691438.
Giuliano, G., Kang, S., & Yuan, Q. (2016). Spatial dynamics of the logistics industry and
implications for freight flows (No. NCST-20160600). Available at
https://ncst.ucdavis.edu/wp-content/uploads/2014/08/06-17-2016-NCST-Spatial-
Dynamics-Draft-Final-Report.pdf
Giuliano, G., & Small, K. A. (1991). Subcenters in the Los Angeles region. Regional
science and urban economics, 21(2), 163-182.
Hamilton, J. T. (1993). Politics and social costs: estimating the impact of collective
action on hazardous waste facilities. The RAND journal of economics, 101-125.
Hesse, M. (2002) Location matters. ACCESS Magazine 1, no. 21
Hesse, M. (2004). Land for logistics: locational dynamics, real estate markets and
political regulation of regional distribution complexes. Tijdschrift voor economische en
sociale geografie, 95(2), 162-173.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
141

Hesse, M. (2007). The System of Flows and the Restructuring of Space Elements of a
Geography of Distribution (Das System der Strö me und die Re-Strukturierung des
Raumes. Elemente einer Geographie der Distribution). Erdkunde, 1-12.
Hesse, M. (2016). The city as a terminal: The urban context of logistics and freight
transport. Routledge.
Hesse, M., & Rodrigue, J. P. (2004). The transport geography of logistics and freight
distribution. Journal of transport geography, 12(3), 171-184.
Hipp, J. R., & Lakon, C. M. (2010). Social disparities in health: Disproportionate toxicity
proximity in minority communities over a decade. Health & place, 16(4), 674-683.
Holguí n-Veras, J., Ozbay, K., Kornhauser, A., Brom, M., Iyer, S., Yushimito, W., ... &
Silas, M. (2011). Overall impacts of off-hour delivery programs in New York City
Metropolitan Area. Transportation Research Record: Journal of the Transportation
Research Board, (2238), 68-76.
Holl, A., & Mariotti, I. (2017). The geography of logistics firm location: the role of
accessibility. Networks and Spatial Economics, 1-25.
Houston, D., Li, W., & Wu, J. (2014). Disparities in exposure to automobile and truck
traffic and vehicle emissions near the Los Angeles–Long Beach port complex. American
journal of public health, 104(1), 156-164.
Husing, J. (2010). Dirt Theory: How Southern California’s New Subregions Mature
Economically and Its Implications for the Location of Distribution Facilities”. In
Comprehensive Regional Goods Movement Plan and Implementation Strategy Industrial
Space in Southern California: Future Supply and Demand for Warehousing and
Intermodal Facilities. Available at
http://www.freightworks.org/DocumentLibrary/Comprehensive%20Regional%20Goods
%20Movement%20Plan%20and%20Implementation%20Strategy%20-%20Reigonal%20
Warehousing%20Needs%20Assessment%20Final%20Report.pdf
Husing, J. (2016) Business Outreach Report Governor’s Innovation Award. Economics &
Politics, Inc. Available at
http://www.johnhusing.com/Husing%20Research%20Reports/Interview%20Briefing%20
July%202016%20%20v.%205.pdf
Jakubicek, P. (2010). Understanding the Location Choices of Logistics Firms (Master's
thesis, University of Waterloo).
Jarvis, B. F. (2015). Matching Residential Mobility to Raced-Based Neighborhood
Preferences in Los Angeles: Implications for Racial Residential Segregation. University
of California, Los Angeles.
Kim, J. J., Smorodinsky, S., Lipsett, M., Singer, B. C., Hodgson, A. T., & Ostro, B.
(2004). Traffic-related air pollution near busy roads: the East Bay Children's Respiratory
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
142

Health Study. American journal of respiratory and critical care medicine, 170(5), 520-
526.
Kirkham, C. (2015) Inland Empire sees surge in warehouse jobs, but many are low-pay,
temporary. Los Angeles Times. Available at www.latimes.com/business/la-fi-inland-
empire-warehouses-20150419-story.html
Kitroeff, N. (2016). California's Inland Empire reels after losing hundreds of blue-collar
jobs August 29, 2016. LA Times. Available at www.latimes.com/business/la-fi-ashley-
furniture-inland-empire-20160829-snap-story.html
Korpela, J., & Tuominen, M. (1996). A decision aid in warehouse site selection.
International Journal of Production Economics, 45(1-3), 169-180.
Kozawa, K. H., Fruin, S. A., & Winer, A. M. (2009). Near-road air pollution impacts of
goods movement in communities adjacent to the Ports of Los Angeles and Long Beach.
Atmospheric Environment, 43(18), 2960-2970.
Kunzman Associates, Inc. (2011). Trip generation analysis for high-cube warehouse
distribution center land use. Prepared for NAIOP, THE Commercial Real Estate
Development Association, Inland Empire Chapter. Available at
http://www.moval.org/misc/pdf/wlc/deir/append/L-21appS.pdf
Lena, T. S., Ochieng, V., Carter, M., Holguí n-Veras, J., & Kinney, P. L. (2002).
Elemental carbon and PM (2.5) levels in an urban community heavily impacted by truck
traffic. Environmental Health Perspectives, 110(10), 1009.
Longitudinal Employment-Household Dynamics. (2016). LEHD Origin-Destination
Employment Statistics (LODES)- Workplace Area Characteristics (WAC). Available at
https://lehd.ces.census.gov/data/#qwi
Marshall, J. D., Swor, K. R., & Nguyen, N. P. (2014). Prioritizing environmental justice
and equality: Diesel emissions in Southern California. Environmental science &
technology, 48(7), 4063-4068.
Mazza, S. (2017). Carson extends ban on industrial growth through 2018. Dailybreeze.
Available at https://www.dailybreeze.com/2017/05/03/carson-extends-ban-on-industrial-
growth-through-2018/
McKinnon, A. (2009). The present and future land requirements of logistical activities.
Land Use Policy, 26, S293-S301.
McKinnon, A., Browne, M., Whiteing, A., & Piecyk, M. (Eds.). (2015). Green logistics:
Improving the environmental sustainability of logistics. Kogan Page Publishers.
McNeice, A. (2017) Robots poised to take over warehouse jobs around the globe.
Chinadaily.com. Available at http://www.chinadaily.com.cn/world/2017-
09/28/content_32570957.htm
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
143

Mohai, P., & Bryant, B. (1992). Environmental injustice: weighing race and class as
factors in the distribution of environmental hazards. U. Colo. L. Rev., 63, 921.
Mohai, P., Pellow, D., & Roberts, J. T. (2009). Environmental justice. Annual Review of
Environment and Resources, 34, 405-430.
Mohai, P., & Saha, R. (1994, August). Environmental injustice: examining alternative
explanations. In Meeting of the American Sociological Association, August, Los
Angeles, California.
Mohai, P., & Saha, R. (2006). Reassessing racial and socioeconomic disparities in
environmental justice research. Demography, 43(2), 383-399.
Mohai, P., & Saha, R. (2007). Racial inequality in the distribution of hazardous waste: A
national-level reassessment. Social problems, 54(3), 343-370.
Mohai, P., & Saha, R. (2015). Which came first, people or pollution? A review of theory
and evidence from longitudinal environmental justice studies. Environmental Research
Letters, 10(12), 125011.
Mohai, P., & Saha, R. (2015). Which came first, people or pollution? Assessing the
disparate siting and post-siting demographic change hypotheses of environmental
injustice. Environmental Research Letters, 10(11), 115008.
Morello-Frosch, R., Pastor Jr, M., Porras, C., & Sadd, J. (2002). Environmental justice
and regional inequality in southern California: implications for future research.
Environmental health perspectives, 110(Suppl 2), 149.
National Emission Inventory. (2014). 2014 National Emissions Inventory (NEI) Data.
Available at https://www.epa.gov/air-emissions-inventories/2014-national-emissions-
inventory-nei-data
Newman, P. (2012) Inland Ports of Southern California–Warehouses, Distribution
Centers, Intermodal Facilities Impacts, Costs and Trends. Center for Community Action
and Environmental Justice. Available at http://caseygrants.org/wp-
content/uploads/2012/08/Inland+Ports+of+Southern+California+-
+Wareshouses+Distribution+Centers+and+Intermodal+Facilities+-
+Impacts+Costs+and+Trends.pdf.
Oakes, J. M., Anderton, D. L., & Anderson, A. B. (1996). A longitudinal analysis of
environmental equity in communities with hazardous waste facilities. Social Science
Research, 25(2), 125-148.
Office of Environmental Health Hazard Assessment, California Environmental Protection
Agency. (2002). Health Effects of Diesel Exhaust. Available at
https://oehha.ca.gov/media/downloads/calenviroscreen/indicators/diesel4-02.pdf
Passchier-Vermeer, W., & Passchier, W. F. (2000). Noise exposure and public health.
Environmental health perspectives, 108(Suppl 1), 123.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
144

Pastor, M., Sadd, J., & Hipp, J. (2001). Which came first? Toxic facilities, minority
move-in, and environmental justice. Journal of urban affairs, 23(1), 1-21.
Pastor, M., Morello‐Frosch, R., & Sadd, J. L. (2005). The air is always cleaner on the
other side: Race, space, and ambient air toxics exposures in California. Journal of urban
affairs, 27(2), 127-148.
Peddle, M. T. (1993). Planned industrial and commercial developments in the United
States: a review of the history, literature, and empirical evidence regarding industrial
parks and research parks. Economic Development Quarterly, 7(1), 107-124.
Regional Air Quality Task Force. (2005) Good Neighbor Guidelines For Siting New
and/or Modified Warehouse/Distribution Facilities. Western Riverside Council of
Governments. Available at http://www.moval.org/misc/pdf/prologis/final/AppA/D-3-
09JohnsonAttachmentH-GoodNeighborGuidelines.pdf
Perez, L., Kü nzli, N., Avol, E., Hricko, A. M., Lurmann, F., Nicholas, E., ... &
McConnell, R. (2009). Global goods movement and the local burden of childhood asthma
in southern California. American Journal of Public Health, 99(S3), S622-S628.
Pooler, M. (2017). Amazon Robots Bring a Brave New World to the Warehouse. Ft.com.
Available at https://www.ft.com/content/916b93fc-8716-11e7-8bb1-5ba57d47eff7
Ringquist, E. J. (2005). Assessing evidence of environmental inequities: A meta‐analysis.
Journal of Policy Analysis and Management, 24(2), 223-247.
Roberts, J. T., & Toffolon-Weiss, M. M. (2001). Chronicles from the environmental
justice frontline. Cambridge University Press.
Rodrigue, J. P. (2008). The Thruport concept and transmodal rail freight distribution in
North America. Journal of Transport Geography, 16(4), 233-246.
Roudsari, B. S., Mock, C. N., Kaufman, R., Grossman, D., Henary, B. Y., & Crandall, J.
(2004). Pedestrian crashes: higher injury severity and mortality rate for light truck
vehicles compared with passenger vehicles. Injury Prevention, 10(3), 154-158.
Saha, D., & Paterson, R. G. (2008). Local government efforts to promote the “Three Es”
of sustainable development: survey in medium to large cities in the United States. Journal
of Planning Education and Research, 28(1), 21-37.
Saha, R., & Mohai, P. (2005). Historical context and hazardous waste facility siting:
Understanding temporal patterns in Michigan. Social problems, 52(4), 618-648.
Schirmer, P. M., van Eggermond, M. A., & Axhausen, K. W. (2014). The role of location
in residential location choice models: a review of literature. Journal of Transport and
Land Use, 7(2), 3-21.
Schweitzer, L., & Valenzuela Jr, A. (2004). Environmental injustice and transportation:
the claims and the evidence. Journal of planning literature, 18(4), 383-398.
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
145

Seto, E. Y. W., Holt, A., Rivard, T., & Bhatia, R. (2007). Spatial distribution of traffic
induced noise exposures in a US city: an analytic tool for assessing the health impacts of
urban planning decisions. International journal of health geographics, 6(1), 24.
Shaikh, S. L., & Loomis, J. B. (1999). An investigation into the presence and causes of
environmental inequity in Denver, Colorado. The Social Science Journal, 36(1), 77-92.
Sivitanidou, R. (1996). Warehouse and distribution facilities and community attributes:
an empirical study. Environment and Planning A, 28(7), 1261-1278.
Slowik, T. (2017). Slowik: Warehouse developers bringing trucks to Will County owe us
better roads, should pay impact fee. Chicago Tribute. Available at
http://www.chicagotribune.com/suburbs/daily-southtown/opinion/ct-sta-slowik-
warehouse-moratorium-st-1130-20171129-story.html
Smith, L. T., Sax, T., Bode, R., & Shimp, D. (2001) Community Scale Estimation of
Toxic Air Pollutants from Stationary Sources. California Air Resources Board, California
Environmental Protection Agency. In 10th International Emission Inventory Conference.
Available at https://www3.epa.gov/ttnchie1/conference/ei10/toxics/smith.pdf
South Coast Air Quality Management District. (2014). Warehouse Truck Trip Study Data
Results and Usage.” Mobile Source Committee. Available at
http://www.aqmd.gov/docs/default-source/ceqa/handbook/high-cube-warehouse-trip-rate-
study-for-air-quality-analysis/finaltrucktripstudymsc072514.pdf?sfvrsn
Tajik, M., & Minkler, M. (2006). Environmental justice research and action: a case study
in political economy and community-academic collaboration. International quarterly of
community health education, 26(3), 213-231.
Taxjar. (2018) Do I have sales tax nexus in California if I sell on Amazon FBA?
Taxjar.com. Available at https://www.taxjar.com/states/california-sales-tax-online/#how-
to-collect-sales-tax-in-california
Taylor, D. (1992). The environmental justice movement. EPA J., 18, 23.
The Naterlso Company, Inc. (2001). Employment density study summary report.
Prepared for Southern California Association of Governments. Available at
https://www.mwcog.org/file.aspx?A=QTTlTR24POOOUIw5mPNzK8F4d8djdJe4LF9Ex
j6lXOU%3D
U.S. Environmental Protection Agency. (2015). Introduction to Municipal Solid Waste
Disposal Facility Criteria. Available at https://www.epa.gov/sites/production/files/2015-
09/documents/mswlf05.pdf
U.S. Environmental Protection Agency. (2017). List of Hazardous Air Pollutants.
Available at https://iaspub.epa.gov/triexplorer/tri_text.list_chemical_hap
Location of Warehouses and Environmental Justice: Three Essays Yuan (2018)
146

Verhetsel, A., Kessels, R., Goos, P., Zijlstra, T., Blomme, N., & Cant, J. (2015). Location
of logistics companies: a stated preference study to disentangle the impact of
accessibility. Journal of Transport Geography, 42, 110-121.
Voogt, J. (2007). How researchers measure urban heat islands. In United States
Environmental Protection Agency (EPA), State and Local Climate and Energy Program,
Heat Island Effect, Urban Heat Island Webcasts and Conference Calls.
Warffemius, P. (2007). Modeling the Clustering of Distribution Centers around
Amsterdam Airport Schiphol: location endowments, economies of agglomeration,
locked-in logistics and policy implications.
Weisbrod, G., Lerman, S. R., & Ben-Akiva, M. (1980). Tradeoffs in residential location
decisions: Transportation versus other factors. Transportation Policy and Decision-
Making, 1(1), 13-26.
Yakowicz, W. (2017). Robots Replaced 75 Percent of Jobs at This Warehouse but Not
One Employee Was Cut. Inc.com. Available at https://www.inc.com/will-
yakowicz/boxed-didnt-fire-one-employee-after-robots-replaced-75-percent-warehouse-
jobs.html
Yang, B., & Li, S. (2013). Green infrastructure design for stormwater runoff and water
quality: Empirical evidence from large watershed-scale community developments. Water,
5(4), 2038-2057.

Abstract (if available)

Abstract This dissertation examines a rising environmental justice problem in warehousing location. Following a stepwise approach, the three-essay research identifies the socioeconomic processes that contribute to the problem and the relevant agents involved in the processes, evaluates the relationship between warehousing facilities and socially advantaged population using empirical data and econometric models, and provides policy implications on how to mitigate the environmental inequity. Results show that warehouses are significantly more likely to be located in minority neighborhoods, and the disproportionate siting of warehouses is the primary drive for the spatial coincidence. In addition, local public policies greatly affect the location of warehouses and may make a clear difference in reducing the disparate distribution. The dissertation justifies the significance of the environmental inequity in warehousing location and implies that relevant agents including governments, warehousing developers, and local residents need to work together in a regional collaboration framework to address the problem.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Unraveling decentralization of warehousing and distribution centers: three essays

PDF

Urban air pollution and environmental justice: three essays

PDF

Environmental justice in real estate, public services, and policy

PDF

Productive frictions and urbanism in transition: planning lessons from traffic flows and urban street life in Ho Chi Minh City, Vietnam

PDF

The impact of demographic shifts on automobile travel in the United States: three empirical essays

PDF

Who learns where: understanding the equity implications of charter school reform in the District of Columbia

PDF

Essays on congestion, agglomeration, and urban spatial structure

PDF

Property and labor formalization in the age of the sharing economy: Airbnb, housing affordability, and entrepreneurship in Havana

PDF

The long-term impact of COVID-19 on commute, employment, housing, and environment in the post-pandemic era

PDF

The built environment, tour complexity, and active travel

PDF

The demand for reliable travel: evidence from Los Angeles, and implications for public transit policy

PDF

Spatial and temporal expenditure-pricing equity of rail transit fare policies

PDF

Active travel, outdoor leisure, and neighborhood environment: path analysis, Los Angeles County

PDF

Population and employment distribution and urban spatial structure: an empirical analysis of metropolitan Beijing, China in the post-reform era

PDF

Essays on the economics of cities

PDF

Why go green? Cities' adoption of local renewable energy policies and urban sustainability certifications

PDF

Alternative ways, locations, and partners to meet the recreational needs of underserved communities: the case of Florence-Firestone

PDF

The development implications of China’s Belt and Road Initiative for Russia, Kazakhstan and Belarus

PDF

Cities on the environmental justice frontline: the intractability of hazards and the governability of amenities

PDF

Household mobility and neighborhood impacts

Asset Metadata

Creator Yuan, Quan (author)

Core Title Location of warehouses and environmental justice: Three essays

School School of Policy, Planning and Development

Degree Doctor of Philosophy

Degree Program Urban Planning and Development

Publication Date 07/29/2018

Defense Date 05/02/2018

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag environmental justice,minority population,mixed methods,OAI-PMH Harvest,Public Policy,warehousing location

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Giuliano, Genevieve (committee chair), Boarnet, Marlon (committee member), Pastor, Manuel (committee member)

Creator Email quanyuan@usc.edu,uscjackq@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-39561

Unique identifier UC11671516

Identifier etd-YuanQuan-6553.pdf (filename),usctheses-c89-39561 (legacy record id)

Legacy Identifier etd-YuanQuan-6553.pdf

Dmrecord 39561

Document Type Dissertation

Format application/pdf (imt)

Rights Yuan, Quan

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA