University of Southern California Dissertations and Theses

Testing the entrepreneurial city hypothesis: a study of the Los Angeles region

(USC Thesis Other)

Testing the entrepreneurial city hypothesis: a study of the Los Angeles region

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
TESTING THE ENTREPRENEURIAL CITY HYPOTHESIS: A STUDY OF THE LOS
ANGELES REGION

by

Ajay Agarwal

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
(PLANNING)

August 2009

Copyright 2009 Ajay Agarwal

ii

Dedication
This dissertation is dedicated to my loving parents who live more than 10,000 miles
away, but are always there for me whenever I need them.

iii

Acknowledgements

This work would not have been possible without the unrelenting love and support of my
wife Sukriti. Throughout the past three years, my adorable son Arnav (who turned 3 last
week) kept me playfully engaged just enough to be able to complete my dissertation and
yet have a life.

I will remain eternally grateful to my advisor Prof. Genevieve Giuliano, for her advice,
mentoring, and guidance. She has watched out for me in more ways than I could express
in words here.

I had most wonderful and supportive friends and colleagues who were extremely patient
with my endless whining while I was writing my dissertation. Most notably Abhishek
Mamgain who, despite struggling with his own dissertation, more than occasionally took
me out for a Single Malt to cheer me up.

iv

Table of Contents
Dedication ii

Acknowledgements iii

List of Tables vi

List of Figures ix

Abstract x

Chapter 1: Introduction 1

Chapter 2: Literature Review 5
2.1 What is an Employment Center? 5
2.2 Theoretical Determinants of Employment Center Growth 11
2.3 The Role of Local Governments 14
2.4 Empirical Support 18
2.4 Summary 23

Chapter 3: Research Approach And Methodology 24
3.1 Examining Employment Center Growth 26
3.2 Examining Employment Growth across Cities 39
3.3 Case Studies 39

Chapter 4: Quantitative Analysis 41
4.1 Descriptive Analysis 41
4.2 Regression Analysis of Employment Center Growth 52
4.3 City Level Analysis 59
4.4 Discussion 69

Chapter 5: Qualitative Analysis 72
5.1 Selection Criteria 72
5.2 Framework of Analysis 74
5.3 Case Study 1: The City of Burbank 76
5.4 Case Study 2: The City of Pasadena 94
5.5 Lessons from the Case Studies 111

Chapter 6: Summary Of Findings And Conclusions 114
6.1 Findings from the Quantitative Analysis 114
6.2 Findings from the Qualitative Analysis 115
6.3 Discussion 116
6.4 Future Research 121

v

Bibliography 123

vi

List of Tables

Table 2.1: Select Empirical Studies of Urban Form 19

Table 2.2: Share of jobs in centers, CBD and other centers by metro area,
Lee’s min density centers 20

Table 4.1: Select Characteristics of Employment Centers in the Los Angeles
CMSA, 2000 44

Table 4.2: Employment Centers: Change in Employment, 1990-2000 45

Table 4.3: Means of Control Variables 46

Table 4.4: Number of Centers by their Growth Promotion and Growth
Management Policies 47

Table 4.5: Number of Centers by Employment Center Growth (1990-2000)
and Local Growth Management Policies 49

Table 4.6: Number of Centers by Employment Center Growth (1980-2000)
and Local Growth Management Policies 49

Table 4.7: Number of Centers by Employment Center Growth (1990-2000)
and Local Growth Promotion Policies 50

Table 4.8: Number of Centers by Growth category and Annual Per Capita
Expenditure on Developmental Activities in the 1990s 51

Table 4.9: Number of Centers by Growth category and Average Local
Taxes/License Fees 52

Table 4.10: Variable Definitions with Means and Standard Deviations 53

Table 4.11: OLS Estimates, Control Variables Only 55

Table 4.12: OLS Estimates Model 1 56

Table 4.13: OLS Estimates Model 2 57

Table 4.14: Employment Growth inside Cities 60

vii

Table 4.15: Number of Cities by their Growth Promotion and Growth
Management Policies 61

Table 4.16: Number of Cities by Employment Growth and Local Growth
Management Policies 61

Table 4.17: Number of Cities by Employment Growth and Local Growth
Promotion Policies 62

Table 4.18: Cities by Growth Category and Annual Per Capita Expenditure on
Developmental Activities in the 1990s 63

Table 4.19: Cities by Growth Category and Local License Fees/ Taxes on
Businesses 64

Table 4.20: Variable Definitions with Means and Standard Deviations 65

Table 4.21: OLS Estimates, Control Variables Only 66

Table 4.22: OLS Estimates, Add Local Government Variables 68

Table 5.1: Selected Characteristics of Burbank and Pasadena 74

Table 5.2: General Population Characteristics, Burbank and LA County, 2000 78

Table 5.3: Housing Characteristics, Burbank and LA County, 2000 79

Table 5.4: Economic Characteristics, Burbank and LA County, 2000 80

Table 5.5: Employed Civilian Population 16 years and over by Industry,
Burbank, 2000 81

Table 5.6: Burbank Revenues 82

Table 5.7: Jobs by Industry, Burbank, 2000 82

Table 5.8: General Population Characteristics, Pasadena and LA County,
2000 96

Table 5.9: Housing Characteristics, Pasadena and LA County, 2000 97

Table 5.10: Economic Characteristics, Pasadena and LA County, 2000 98

viii

Table 5.11: Employed Civilian Population 16 years and over by Industry,
Pasadena, 2000 98

Table 5.12: Pasadena Revenues 100

Table 5.13: Jobs by Industry, Pasadena, 2000 100

ix

List of Figures

Figure 2.1: Land Rent Gradients in a Polycentric city 6
Figure 3.1: Los Angeles CMSA 25
Figure 4.1: Employment Centers in the Region, 2000 42
Figure 4.2: Cities Included in the Analysis 59
Figure 4.3: Planning Districts in the City of Los Angeles 64
Figure 5.1: The City of Burbank 77
Figure 5.2: The Burbank Media District Overlay Zone 88
Figure 5.3: Suggested Parking Design in the Media District Specific Plan 90
Figure 5.4: The City of Pasadena 95
Figure 5.5: Old Pasadena 105
Figure 5.6: Metro Gold Line Route 110

x

Abstract

This dissertation examines the role of local governments in the evolution of metropolitan
spatial structure, particularly with respect to the growth of employment centers --
locations with significant concentration of economic activity and hence employment.
The study is conducted in two parts: one, quantitative analysis of employment center
growth in the Los Angeles region between 1990-2000; and two, detailed qualitative case
studies of two cities: Pasadena and Burbank.

For the quantitative analysis, employment center growth is regressed as a function of
local development policies and a set of control variables. The Los Angeles region is ideal
for study: more than 150 cities and many employment centers make a systematic
analysis possible. The findings from the analysis suggest that economic forces prevail
over local development policies in affecting the emergence and growth of employment
centers. Overall, employment center growth appears to be a part of the larger
decentralization phenomenon. Firms value access to the labor force and hence jobs
follow people. As population decentralizes, so do jobs. There is some indication that
local government policies tend to be largely reactive. For example, locations
experiencing high employment growth have higher incidence of strong growth control
policies whereas locations facing employment losses have high incidence of strong
growth promotion policies.

xi

The case studies substantiate the findings from quantitative analysis. The case studies
indicate that there may be differences between enactment of growth control policies and
their actual enforcement. Local governments may choose to selectively enforce growth
controls by way of granting conditional use permits that override growth control
ordinances. Cities want to pursue growth of sales tax revenues because property taxes
are constrained by Proposition 13. The case studies also indicate that often a local
government’s actions are conditioned by unexpected market conditions over which
municipalities have no control.

1

Chapter 1: Introduction

This dissertation examines the role of local governments in the evolution of metropolitan
spatial structure, particularly with respect to the growth of employment centers –
locations with significant concentration of economic activity and hence employment.
The evolution of metropolitan spatial structure is of interest to urban researchers across
many disciplines. Understanding the factors that influence metropolitan trends also has
important policy implications for urban planning, infrastructure investment, and local
economic development as it would help implement policies that are more effective at
both reversing and reinforcing such trends.

Contemporary metropolitan form is characterized by significant decentralization of
population and employment from the metropolitan core and formation of multiple
employment centers i.e. a polycentric spatial structure. There is an extensive literature on
the evolution of metropolitan spatial structure. Decades of research by economists,
geographers, historians, and others have generated both well-accepted and contentious
explanations for the emerging metropolitan spatial structure (e.g. Jackson 1985, Cervero
1989, Chinitz 1991, Anas et al 1998, Muller 2004). In urban economic literature,
polycentric spatial structure is described as an outcome of the interplay between the
centripetal forces of decentralization and centrifugal forces of agglomeration. Factors
associated with decentralization include 1) economic restructuring, 2) reduced transport
and communications costs, 3) transportation, housing, and tax policy; 4) per capita
incomes; 5) American cultural preferences (e.g. Scott, 1988; Castells 1989, Graham and

2

Marvin, 1996, Nivola 1996, Cairncross 1997, Kloosterman and Musterd 2001, Muller
1981, Bruegmann 2005). Agglomeration economies remain the critical factor associated
with concentration (Rosenthal and Strange 2003); other factors include access to the
transportation network and longevity of capital stock, i.e. path dependence (Batty 2001).

As a metropolitan region grows, the existence of multiple employment centers is more
likely. The presence of multiple employment centers as a norm rather than exception is
widely documented for not only large, but also medium size metropolitan regions (e.g.
Giuliano and Small 1991, McMillen and MacDonald 1997, 1998, 2000, Cervero and Wu
1997, Bogart and Ferry 1999, Anderson and Bogart 2001, McMillen and Smith 2003,
Coffey and Shearmur 2002, Small and Song 1994). Using the example of Los Angeles,
Gordon and Richardson (1996) argue that metropolitan regions may have already moved
“beyond polycentricity.”

The empirical literature on evolution of urban spatial structure does not explain why an
employment center emerges in a particular place. Most of the empirical research on
employment trends has been conducted at the scale of metropolitan areas, and hence does
not address spatial variation within metropolitan areas. While existence of cities offers
compelling evidence of agglomeration economies, yet empirical studies conducted at
metropolitan level aggregate data to a single point, which camouflages the nuances of
development that play out at a finer geography. Indeed, there is empirical evidence that
agglomeration economies matter at a geographic scope far smaller than the metropolitan

3

area (Redfearn et al forthcoming). Hence using employment centers as a unit of analysis
is more reasonable than using cities or metropolitan areas for testing hypotheses
grounded in urban economic theory.

Favorable conditions for employment concentration (e.g. available land, transport access,
labor force access) may exist in many locations. Within a single metropolitan region
there are multiple local jurisdictions, which often compete with one another for economic
growth. There may be significant inter-jurisdictional differences in terms of development
policies and the ensuing regulatory environment, making some jurisdictions more
conducive to growth than others. These differences could explain the emergence and
growth of employment centers at some locations and not others within the same
metropolitan region. Despite an extensive literature in geography, economics and
regional science on metropolitan spatial structure and its evolution, there has been little
systematic analysis of the role of local government in spatial organization. Through
zoning controls, infrastructure investment, local tax policy and other mechanisms, local
governments have significant potential to shape the landscape within their jurisdictions.

This dissertation explicitly examines the association between development policies
enacted by the local governments and employment center growth.
1
The remaining
dissertation is organized in five chapters. Chapter 2 reviews literature to examine

1
Local government policies focus not just on development but also on redistribution, social equity, safety
and security, etc. While all aspects of local governance may influence “development” the focus here is
policies that explicitly influence development or growth. It is posited that such developmental policies are
associated with employment center growth.

4

theoretical determinants of employment center growth and establishes
conceptual/theoretical foundation for potential role of local governments in employment
center growth. Chapter 3 describes research approach and methodology. Chapter 4
presents quantitative analysis. Chapter 5 presents qualitative case studies. Finally,
Chapter 6 summarizes the research findings, discusses their policy implications, and
concludes with potential future research on the topic.

5

Chapter 2: Literature Review

2.1 What is an Employment Center?
An employment center can be understood a location with substantial concentration of
economic activity and hence employment. Researchers have given several names to such
activity centers or locations of substantial employment concentration, including for
example employment subcenter (Giuliano and Small, 1991), suburban employment center
(Cervero, 1989), edge city (Garreau, 1992), job concentration (Forestall and Greene,
1997), employment pole (Coffey and Shearmur, 2002). In this dissertation I shall use the
term ‘employment center’ to denote a site of significant geographic concentration of
economic activity.

In economic terms, an employment center is a cluster of activity of sufficient magnitude
to influence land prices and hence spatial form. In theory, identifying employment
centers is straightforward: any cluster that independently influences land values
constitutes a center. Thus if we have two approximately equal clusters, we estimate their
respective land value or density gradients, and the area beneath the gradient is the center.
If the centers are close enough to overlap, the intersection of the gradients defines the
boundary. If we have one large and one small cluster, the small cluster is a center only if
it has some independent influence, e.g. some portion of its gradient dominates that of the
larger center. Figure 2.1 illustrates.

6

Figure 2.1 Land Rent Gradients in a Polycentric city

The reality of metropolitan areas is far more complicated than these simple theoretical
concepts. Metropolitan areas have many clusters of employment, from isolated suburban
office parks to the downtown. Nor are clusters convenient circles or ellipses. In some
cases major freeways define linear concentrations, in others a cluster might be broken up
by a river or canyon. It is therefore not surprising that in empirical research employment
centers have been defined in many different ways.

Employment centers have been defined in both qualitative and quantitative terms.
However, in order to conduct systematic empirical analysis, employment centers must be
defined using some quantitative measure. Hence I focus on quantitative methods only.
Methods used to identify centers can be grouped into two major categories: 1)
Parametric, and 2) Non-parametric.

7

Parametric Approach
As the name suggests, in the parametric approach, employment centers are assigned
certain parameters, which could either be absolute and relative. In absolute terms, an
employment center must meet some minimum criteria, such as employment density
and/or total number of jobs, irrespective of other areas. In relative terms, the parameters
assigned to employment centers are specified are comparative, e.g. substantially higher
employment density than nearby areas where “substantially higher” could again be
defined in different ways. There are several examples of both approaches in the
literature.

In an early attempt, Cervero (1989) identifies an employment center as a site with a
minimum of 2,000 employees and 1 million square feet of office space. Employment
center boundaries are defined by property lines or local planning agencies.

Giuliano and Small (1991), identify an employment center as a set of contiguous analysis
zones
2
such that each have a certain minimum employment density D and together have a
certain minimum total employment E. In the same study they used values of 10 jobs per
acre and 10,000 jobs for D and E to identify employment centers in the Los Angeles
Metropolitan Area (using 1970 analysis zones and 1980 employment data). They do not
use the criterion of relative employment density as it would exclude some larger centers
in the core area. However, a relative criterion might identify some smaller centers in the

2
Analysis zones are spatial units approximately the size of census tracts.

8

outer counties (Ventura, San Bernardino, and Riverside). To address the latter issue, they
lower the threshold value of total employment E from 10,000 to 7,000 for the outer
counties only and successfully identify one center in each. Their methodology has been
used in several subsequent studies, for example Anderson and Bogart (2001), Bogart and
Ferry (1999), Small and Song (1994). McMillen (2001) criticizes this approach arguing
that it contains an element of arbitrariness as assigning values to D and E is not based on
any particular criteria.

Other methods are more complicated. Forstall and Greene (1997) identify an employment
center as a set of contiguous census tracts with combined employment to resident worker
ratio of 1 or greater and at least one tract of which has employment to resident worker
ratio of 1.25 or greater. Employment to resident worker ratio of 1 or greater implies net
inflow of workers into center. Similarly, Coffey and Shearmur (2002) identify an
employment center as a set of contiguous census tracts which together have at least 5,000
jobs and minimum employment to resident worker ratio of 1. Gordon and Richardson
(1996) identify employment centers in terms of trip densities and not simply job densities
arguing that different activity centers with the same number of jobs generate different
levels of traffic depending on industry sector composition.

Non-parametric Approach
Non-parametric methods use sophisticated statistical methods and algorithms to identify
employment centers, typically as locations with significantly higher employment density

9

than their surroundings. Advances in computing technologies help the advance of
nonparametric approach (Redfearn, 2007).
Gordon, Richardson and Wong (1986) used t-values on population and employment
density gradients to determine the locations and number of subcenters. Heikkila et al.
(1989), McDonald and Prather (1994), and Small and Song (1994) applied similar
approaches to identify potential centers.

McMillen (2001) proposes a two-stage non-parametric procedure to identify employment
centers, which in this case are sites with significantly higher employment density than the
spatial trend. In the first stage, potential center candidates are identified as significant
positive residuals in a smoothed employment density function. LWR is used to smooth
the natural logarithm of employment density over space, giving higher weights to nearer
tracts. In this particular method, McMillen gives weight to the nearest 50 percent of the
observations. In the second stage, a semi-parametric employment density function is
estimated to determine if potential employment center candidates have significant local
effect on employment density. This model is innovative in that it uses objective
statistical criteria to determine whether local employment peaks are large enough to be
counted as employment centers. Similarly, Craig and Ng (2001) used quantile smoothing
splines to identify employment subcenters.

McMillen (2003) proposes a simple algorithm for identifying subcenters, using standard
properties of contiguity matrices. He defines two sites as “contiguous” if they are within

10

1.25 miles of distance. The advantage of this approach is that sets of contiguous
candidate subcenter sites can be combined without using local knowledge. McMillen
indicates that this procedure is easier than visual inspection of maps for large
metropolitan areas. McMillen also claims that the algorithm could be utilized in
identifying potential employment centers through minimum employment density cut-offs.

Redfearn (2007) utilizes spatial econometric techniques to generate a smooth
employment density surface, identify local maxima on the density surface (which are
potential center candidates), and calculate density gradients in several directions from the
maxima. A maximum is identified as an employment center if all the gradients are
negative. An information criterion is then used to set the center boundaries such that the
set of tracts in the center is significantly denser than those that surround it.

There is no consensus on a single definition and the corresponding method of
employment center identification. However, based on the empirical research on
metropolitan form, it could be concluded that the monocentric model is no longer
sufficient to explain the emerging urban spatial structure. Existence of multiple
employment centers in not only large, but also medium sized metropolitan regions has
now been unambiguously established. The following section discusses theoretical
determinants of employment center formation and growth.

11

2.2 Theoretical Determinants of Employment Center Growth
There is an extensive literature explaining the evolution of metropolitan spatial structure
in economic terms (e.g. Mills 1967, Fujita 1989, Anas et al 1998). The existence of
employment centers, such as the Central Business District (CBD), is explained on the
basis of external economies of scale in production (agglomeration economies). It is
argued that firms co-locate inside the CBD to benefit from external economies of scale,
both pecuniary and technological. These include labor force access, knowledge
spillovers, and input sharing. To the extent that agglomeration benefits outweigh
agglomeration diseconomies, such as traffic congestion, high land rents, etc., firms
continue to locate inside the existing centers. Over time, however, an existing center may
grow to a point where the negative externalities of locating inside it outweigh the
benefits, at least for some firms. As firms seek locations outside the CBD, agglomeration
benefits could lead to the emergence of an employment center at another location.

Within this broad framework, researchers have offered several theories regarding the
emergence and growth of centers at multiple locations. One set of theories is based on
traditional arguments of economies of scale in production and diseconomies in
transportation and congestion. Helsley and Sullivan (1991) argue that development of an
employment center outside the CBD begins when transportation diseconomies reduce the
social value of labor inside the CBD to a point at which the social value of labor in the
employment center exceeds the social value of labor in the CBD by the fixed cost of
employment center infrastructure. Strong external scale economies inside the CBD and

12

high infrastructure costs may delay employment center formation, whereas greater
congestion diseconomies may accelerate employment center formation. Chen (1996)
proposes that an exogenous change in transportation technology that lowers transport cost
may lead to the formation of an employment center, as reduced transport cost decreases
agglomeration economies.

The role of decision-making agents is considered in the literature. Local government’s
influence via tax and land-use policy in promoting local growth has been posited (Fujita,
1989; Sullivan, 1986; Zhang and Sasaki, 1997, 2000). Henderson and Mitra (1996) note
the role of private developers in facilitating migration of firms, and hence in the creation
of employment centers. Wieand (1987) argues that a center emerges due to the concerted
efforts either of a large private developer or a city development agency. Brasington
(2001) suggests decisions of several small developers can create an employment center.
Anas et al (1998) argue that both developers and local government play a critical role in
formation of new employment centers. The underlying assumption here is that there are
several rival developers, each competing for some strategic location to develop.
Government intervention then could become the key factor in deciding the new location.
Intervention could come in form of land use regulation, providing infrastructure at certain
specific locations, or providing subsidies to developers and/or to firms for relocation at
specific locations, etc.

13

Some theorists ascribe center formation to location decisions of large firms. Fujita and
Thisse (2002) posit that an employment center may emerge when a large firm moves to a
distant location away from the CBD, where several smaller firms may be present. The
large firm moves far enough to take advantage of lower land rents and cheaper labor, but
close enough to the CBD to take advantage of information flows and other urbanization
economies. Finally there is the random element: location of a firm may depend on
idiosyncratic preferences of entrepreneurs, knowledge-workers, chief executive officers,
or others involved in decision making (Anas et. al. 1998).

Although there is a large body of empirical literature on metropolitan spatial trends, the
theoretical expectations regarding growth of employment centers per se have been
subjected to very little empirical testing. There is extensive empirical evidence that
establishes productivity, competitiveness, propensity to innovate, and quality of life as
factors favorable to regional economic growth in general (e.g. Carlton 1983, Glaeser et al
1992, Henderson 2003, Rosenthal and Strange 2001). There is also evidence that
agglomeration economies have positive influence on regional productivity and hence
employment growth (Rosenthal and Strange 2003). It is argued that firms seek profit-
maximizing locations and are drawn to the most productive regions (where
agglomeration economies are at work) resulting in the growth of firms in those regions.
Using case based evidence, Porter (1990) argues that clusters (geographic entities
conceptually similar to employment centers albeit more specialized in terms of
constituent industries and firms) enhance firms’ competitiveness and offer productivity

14

advantages, and that local competition in clusters encourages innovation. According to
Porter, emergence and growth of a cluster depends in part on the institutional and
regulatory response to its development needs.

All of the explanations above are plausible, but none explain why an employment center
emerges in a particular place. Most of the empirical research has been conducted at the
scale of metropolitan areas, and hence does not address spatial variation within
metropolitan areas. Favorable conditions for employment concentration (e.g. available
land, transport access, labor force access) may exist in many locations. Within a single
metropolitan region there typically are multiple local jurisdictions. In this fragmented
Tiebout world, cities may compete with one another for economic growth. There may be
significant inter-jurisdictional differences in terms of development policies and the
ensuing regulatory environment, making some locations more conducive to growth than
others. These differences could explain the emergence and growth of employment
centers at some locations and not others within the same metropolitan region.

2.3 The Role of Local Governments
What is the role of governments in growth and development of cities? In the Tiebout
(1956) framework, cities compete with one another to attract residents via various
packages of taxes and services. Since households value access to jobs and other
economic activities, the availability of economic activity is one dimension of the
competition. Indeed the most important issues on a city’s agenda involve maintaining or

15

improving the city’s fiscal base (Peterson 1981). City officials favor developmental
policies that “contribute to the economic well-being of the city” and “can be expected to
yield economic benefits that will protect the community’s fiscal resources” (page 131).
Altshuler and Luberoff (2003) explain the history of investment in “mega-projects” (e.g.
transit systems, sports stadiums, airports) in part as the result of political leaders seeking
investment and economic development. Frieden and Sagalyn (1992) describe several
decades of effort for city revitalization, and used the term “entrepreneurial cities” to
describe cities that sought to reinvent themselves as entertainment and shopping
destinations in an effort to compete with suburban communities for economic
development. I adopt the Frieden and Sagalyn term to mean cities using their regulatory
powers to influence economic development. While Frieden and Sagalyn were concerned
with aging central cities responding to competition from the suburbs, I am concerned
with cities in general and the competition among them for development or tax revenue.

Inter-jurisdictional competition translates into cities making themselves attractive to
potential investors: homebuyers in predominantly residential suburbs and business
investors in employment centers (Altshuler and Luberoff 2003). Cities use two major
types of instruments to enhance their business competitiveness: 1) inducements such as
zoning concessions, tax abatements, low-interest loans, the use of eminent domain to
assemble large parcels of land, etc.; 2) direct investment in “mega-projects” such as
transit systems, convention centers, sports stadiums, and airports (Altshuler and Luberoff
2003).

16

Cities may also seek to obtain various public benefits on behalf of residents by taking
advantage of the “development rights” they control (Sagalyn, 1997). Development rights
may be exchanged for public projects (e.g. local infrastructure or affordable housing
funds). Various forms of public-private partnerships may be employed, with public
contributions in form of public land, development rights, or a combination of both while
the private developer is responsible for investing and developing both public-benefit and
for-profit components of the project. The scope of negotiations in public-private
partnerships far exceeds that in generic regulatory approvals (Sagalyn 1997). Altshuler
and Gomez-Ibanez (1993) argue that land use policies are driven largely by local fiscal
considerations. Local governments and citizens increasingly believe that new
developments require more in public spending than they produce in tax revenues. New
development is increasingly expected to “pay its own way”, hence the growth in
exactions observed over the previous decades. Furthermore, cities have increasingly
enacted “fiscal” zoning regulations that aim to develop a tax-base of predominantly
expensive homes and commercial/industrial properties, which have low service needs and
generate high tax revenues (Hanushek and Quigley 1990).

There are several ways in which a local government could influence employment
concentration and growth. All local development, both residential and
commercial/industrial, is subject to a variety of local land use policies and regulations,
such as the zoning code or general plan. Local governments may facilitate development
at particular locations by investing in infrastructure, such as streets and sewers, at those

17

locations. Capital infrastructure is highly durable and once built has long life. Capital
investments therefore influence the subsequent development(s) for a long time (path
dependence). Labor force supply would depend on residential attractiveness – price and
availability of housing, amenities, low incidence of crime, etc. Thus local governments
may also indirectly influence employment location through residential land use policy
and public service provision (Gottlieb 1994, 1995). A local government could facilitate
development by offering incentives, e.g. subsidies, tax breaks, or density bonuses in
certain areas, to firms for locating inside its jurisdiction.

A local government could also create barriers to growth, at some or all areas within its
jurisdiction, by enacting explicit growth controls such as permit caps, density limits,
growth boundaries, etc. A local government could also implicitly make the permitting
process cumbersome and lengthy thereby delaying and inflating the cost of some or all
projects (Feiock 2004). Researchers have presented different motivations for enactment
of restrictive land use regulations and growth controls, e.g. environmental concerns
(Deakin 1989), preservation of property values (Fischel 2001), and so on. Restrictive
land use regulations, including zoning and growth controls, reduce the supply of
developable land which should in turn results in higher land prices (Fischel 1991, Knapp
and Nelson 1992). The empirical evidence in this regard is mixed. While a substantial
number of land use and growth control studies find strong association between growth
management regulation and land/housing prices, several studies also show little or no
effect of regulation on price, implying that sometimes, local regulation is symbolic,

18

ineffectual, or only weakly enforced (Quigley and Rosenthal 2005). Ihlandfeldt (2004)
notes that higher prices may reflect the amenity value of restricted growth as well as the
effect of reduced supply. Finally, U.S. land use policy is criticized as leading to
population and employment decentralization (sprawl) and contributing to spatial
mismatch (Stoll 2005).

2.4 Empirical Support
There is a rich empirical literature on urban spatial structure. Researchers have
extensively studied spatial trends of population and employment over time. The
literature establishes substantial decentralization of population from the core and general
flattening of population density gradients over time. However, as this dissertation is
primarily concerned with employment center growth, I focus on the empirical literature
on employment trends.
The empirical literature on employment centers is largely concerned with identification
of employment centers in metropolitan regions. These studies unambiguously establish
presence of multiple employment centers in not only large but also medium size
metropolitan regions (see Table 2.1). However, most of the studies have tended to be
cross-sectional and not necessarily addressed employment center growth. One exception
is a recently published study by Lee (2007), which compares employment trends between
1990 and 2000 for major metropolitan areas in the U.S.

19

Lee (2007) identifies employment centers using a combination of non-parametric/
parametric method, and by a minimum density method.
3
Table 2.2 presents a summary
of his results using the minimum density method. The table gives shares of total
employment in the CBD, in other centers, and the total share of employment in centers.
Los Angeles and San Francisco stand out for their small share of CBD employment,
while New York, Boston and Philadelphia have relatively larger CBD shares. Results for
Portland are mixed. Comparing 1990 and 2000, CBDs lost employment share, but only
in the case of Portland is there a large change. Other centers also lost share in every case.
Lee’s analysis shows employment growth is faster outside of center, and hence an
increasing share of employment is located outside centers.

3
Discussion on these methods is mentioned in Chapter 3

20

Table 2.1 Select Empirical Studies of Urban Form
Author Study
Period
Data Source Method Used Study Area # of
Centers
Giuliano and
Small (1991)

1980 Southern
California
Association of
Governments
Parametric
Employment Density ≥ 10
jobs/acre and
Total Employment ≥10,000
Los Angeles
CMSA
35
McMillen &
McDonald
(1997)
1980 CTPP Non-parametric
Locally weighted regression
(LWR)
Suburban Chicago
MSA (excludes
city of Chicago)
15
Anderson
and Bogart
(2001)
1990 NA Parametric
Employment Density ≥
5,000 jobs/ square mile &
Total Employment ≥ 10,000
Cleveland 9
Indianapolis 11
Portland 11
St. Louis 10
McMillen
and Smith
(2003)
1990 CTPP Combination of Non-
parametric and Parametric
LWR to identify potential
centers, then apply a
minimum total employment
criterion of 10,000 jobs to
select final centers.
Chicago, IL 12
Dallas, TX 12
Detroit, MI 8
Los Angeles, CA 46
New York, NY 38
Portland, OR 3
San Diego, CA 6
San Francisco, CA 12
Seattle, WA 14
Washington, DC 10
Giuliano et
al (2007)
1980 SCAG Parametric
Employment Density ≥ 10
jobs/acre and
Total Employment ≥ 10,000
Los Angeles
CMSA
36
1990 46
2000 48
1980 SCAG Employment Density ≥ 20
jobs/acre and
Total Employment ≥ 20,000
Los Angeles
CMSA
10
1990 13
2000 15
Redfearn
(2007)
2000 SCAG Non Parametric
LWR and statistical
algorithms
Los Angeles
CMSA
41
Lee (2007) 1990 CTPP, CMSA Revised McMillan and
Smith
New York 34
Los Angeles 44
Boston 10
San Francisco 22
Portland 3
Philadelphia 14
2000 CTPP, CMSA Revised McMillan and
Smith
New York 35
Los Angeles 42
Boston 8
San Francisco 18
Portland 3
Philadelphia 11
Source: Giuliano, Agarwal, and Redfearn (2008)

21

Table 2.2 Share of jobs in centers, CBD and other centers by metro area, Lee’s
minimum density centers
NY LA Boston Portland SF Philadelphia
Share in CBD
1990 21.8 15.3 22.0 26.8 14.7 20.9
2000 21.2 13.0 21.8 19.2 14.2 15.9
Share in other centers
1990 8.5 22.3 5.1 7.8 15.7 10.7
2000 7.5 19.3 2.0 9.3 19.7 6.7
Total share in centers
1990 30.3 37.6 27.0 34.6 30.4 31.6
2000 28.7 32.3 23.8 28.6 33.8 22.7
Total employment (millions)
1990 9.0 6.8 2.2 0.7 3.1 2.3
2000 9.4 6.7 2.3 1.1 3.4 2.4
Source: Lee (2007), pp. 501-507

There is an extensive empirical literature on the association between land use regulations
and housing markets (e.g, Malpezzi 1996, Wolch and Gabriel 1981, Schwartz et al 1984);
and the influence of growth controls on development pattern in general and housing in
particular (e.g. Brueckner 1990, Fischel 1989, Glickfeld and Levine 1992, Downs 2004).
However, there is little empirical work on the influence of local development
policies/regulations on employment growth, or on the association between city-level
regulations and growth of employment centers.

One reason for the lack of empirical research on the topic is data limitations. Often the
number of employment centers in a metropolitan region is not large enough for a
systematic statistical analysis. Los Angeles, with 48 centers in 2000, is perhaps the only

22

exception in the U.S. Other large metropolitan regions such as New York, Chicago, etc.
all have fewer centers (refer Table 2.1).

I am aware of only one prior empirical study (Giuliano and Small 1999) that investigates
a series of hypothesis related to the determinants of employment center growth in the Los
Angeles region in the 1970s. Using 1970 and 1980 data, Giuliano and Small (1999)
empirically investigate a series of hypothesis to explain the determinants of growth of
employment centers (between 1970 and 1980) in the Los Angeles region. Their
hypotheses are related to 1) economic productivity; 2) labor force accessibility; and 3)
access to the region’s transportation facilities.

Some of the Giuliano and Small (1999) results are counterintuitive. For example, they
found no significant relationship between center growth and accessibility to labor force or
access to the highway system. Their explanation is that accessibility to labor force and
good highway access were so prevalent across the Los Angeles region that these forces
“did not exert any discernible effects on differential growth rates” of employment centers
(page 190). They did, however, find evidence that proximity to large airports influences
urban center formation. Their conclusion is that “unique location factors, including
zoning and fiscal policies as well as airport access and land availability may be more
relevant” in explaining intra-metropolitan spatial trends (page 199).

23

A lot has changed in the Los Angeles region since 1980. The region has added more than
1.8 million jobs and 4.6 million persons. The number of employment centers (using the
Giuliano and Small definition) has grown from 32 to 48. The region’s highways and
airports have become ever more congested. Given changes of this magnitude, and that
several of Giuliano and Small results were quite counterintuitive, it is reasonable to
revisit the question re determinants of employment center growth using data from the Los
Angeles region. A higher number of observations in 2000 (48 as compared to 32) is
likely to yield more robust results. Furthermore, Giuliano and Small’s conclusion
regarding “unique local factors” could be systematically examined.

2.4 Summary
In theory, local governments could influence employment center growth in a number of
ways. Local government efforts could potentially facilitate as well as discourage
employment concentration and growth. The association, however, has not been subject
to systematic empirical scrutiny so far.

24

Chapter 3: Research Approach and Methodology

This dissertation explicitly examines the association between development policies
enacted by the local governments and employment center growth. The research is
conducted in three complementary stages: 1) quantitative analysis of the association
between local development policies and employment center growth; 2) quantitative
examination of the association between local development policies and employment
growth more generally; and 3) because the many ways that local government efforts
could influence employment growth cannot be fully quantified, detailed case studies of
two cities to examine local government efforts and their outcomes.

The dissertation uses 1990-2000 data from the five county Los Angeles Consolidated
Metropolitan Statistical Area (CMSA), as defined by the U.S. Bureau of Census (see
Figure 3.1). The analysis is restricted to the urbanized area portion of the CMSA as
defined by the bureau of census. The non-urbanized portion of the CMSA consists of
vast but largely uninhabited tracts of mountains and deserts and could not be reasonably
expected to influence employment center growth. Polycentricity of Los Angeles Region
is well-documented (Giuliano and Small 1991, Forestall and Greene 1997, Redfearn
2007), and the region has a large enough number of employment centers to make a
systematic analysis possible.

25

Figure 3.1 Los Angeles CMSA

The research approach is guided by some general considerations and data constraints.
First, the literature is unclear on lags in the development process. Since I am interested in
employment concentrations, rather than a particular project or development, I posit that
local policies likely have impacts over many years. Ideally I would trace policies over
many decades, but longitudinal data sources are limited. I therefore select the longest
period for which I could obtain comparable data. Second, empirical analysis requires a
reasonably large number of observations. The Los Angeles region has by far the largest
number of employment centers among US metropolitan areas – 48 in 2000 – yet the
number is just sufficient. Because this empirical analysis is limited by the number of
centers, qualitative case studies will help to inform the results. Third, the number of
ways in which local governments might influence employment growth is quite large. It is
not possible to collect data on all aspects of local government regulation and policy (the

26

data are limited in time and scope), and hence I focus on key proxies for infrastructure
investment, growth inducements and controls, and tax policies. Finally, the relationship
between local government efforts and growth outcomes may not be clear. It is possible
that cities with low rates of employment growth would invest more in attracting growth
while cities with rapid growth may be more likely to impose growth controls. Thus the
analysis must incorporate controls for the overall local economic environment.

3.1 Examining Employment Center Growth
Examining the role of local governments in employment center growth requires the
following: 1) working definition of employment center, 2) a model of employment
center growth that controls for other relevant factors.

Identifying an employment center
As discussed earlier, researchers have identified employment centers in many ways, from
qualitative measures based on local perception (e.g. Garreau, 1991) to simple quantitative
parameters (e.g. Giuliano and Small, 1991, Forestall and Greene 1997, Gordon and
Richardson 1996), to more complex specifications that attempt to consider irregularities
in the spatial distribution (e.g. McMillen 2003, Redfearn, 2007).

Giuliano and Small (1991) use an absolute definition: an employment center is a
relatively compact geographic area containing a “sizeable” employment base. They
identify an employment center as a set of contiguous analysis zones (spatial units

27

approximately the size of census tracts) such that each has a certain minimum
employment density and together have a certain minimum total employment. They
choose an absolute measure, arguing that a relative measure would exclude some larger
centers in the core area. A major criticism of the Giuliano and Small approach is its
arbitrariness: the more stringent the cutoff, the fewer centers will be identified. That
said, the approach has held up quite well. McMillen (2003) argues “non-parametric
estimation procedure allows the cut-off points for employment density to vary both
within and across cities, whereas researchers using the Giuliano and Small procedure
typically let local knowledge and a priori expectations determine the choice of cut-off
points” …yet…“their procedure has so far proved the most popular” (pp 57, 58).

The Redfearn (2007) approach utilizes spatial econometric techniques to smooth the
employment density surface, identify local maxima on the density surface, and calculate
density gradients in several directions from the maxima. A center is identified if all
gradients are negative. An information criterion is then used to set the center boundaries
such that the set of tracts in the center is significantly denser than those that surround it.

All definitions of employment centers contain some aspect of subjective decisions made
by researchers (Giuliano et al, 2007). Each has its advantages as well as limitations and
there is no consensus on the right method. My understanding is that Redfearn (2007)
does a better job of identifying emerging centers (in areas with relatively flat employment
density surface). This is because the method utilizes a comparative definition of

28

employment centers and some of the emerging centers may not meet pre assigned
parameters. This also means that the Redfearn method may not be the best to study
growth of employment centers over time because the way his algorithm is set up the
center boundaries may shift significantly in response to growth that may have occurred in
the neighboring tracts. The Giuliano et al (2007) centers do a better job of identifying
center growth over a time series because tracts add or drop from the centers based on set
parameters. In other words, all employment centers in the Giuliano and Small method
will meet the same minimum criteria in any given year, e.g. all centers will have at least
10 jobs per acre whether using data from 1980, 1990, or 2000. In the Redfearn approach,
there may be variations in center characteristics identified using data from 1980, 1990,
and 2000. Hence for the purpose of this study, I have chosen to use the Giuliano et al
centers.

A Model of Employment Center Growth
The conceptual model is straightforward: employment center growth is estimated as a
function of local government actions and other relevant factors:
( )
C L
X X f Y , = (1)
where X
L
is a vector of local government activities and X
C
is a vector of control
measures. Giuliano and Small (1999) establish a series of hypotheses based on the
theoretical determinants of employment center growth: economic productivity, labor
force accessibility, and transportation network accessibility. I use their model as a point

29

of departure for this study. The data is drawn from the Los Angeles region, and will be
described in a later section.

Unit of Analysis
The unit of analysis is an employment center, each of which comprises of one or more
census tracts. Giuliano et al (2007) identify 48 employment centers in the region in 2000,
using the Giuliano and Small (1991) method. Holding constant the center boundaries, I
examine the employment change inside these centers between 1990 and 2000. The year
2000 is the latest year for which census employment data is available. It is reasonable to
use employment center boundaries in 2000 and examining determinants of growth inside
the same. The other possibility is the opposite—using 1990 center boundaries and
examining growth inside those. This would exclude a) employment centers that were
formed between 1990 and 2000 and b) census tracts that were added to the existing
employment centers because of employment growth. As the purpose of this study is to
explain determinants of growth, using 1990 center boundaries will therefore be
unreasonable. A third possibility is using a pooled sample. But, using a pooled sample
in this case does not present substantial advantages because the center locations remain
fairly stable between 1990 and 2000.

Dependent variable
Employment center growth could be expressed as absolute growth, i.e. change in total
employment inside a center, or growth rate, i.e. percentage change in the number of jobs

30

inside a center, or more generally: ∆E/E
λ
where 0≤ λ ≤ 1. Absolute growth and growth
rate then become special cases where λ=0 and λ=1, respectively.

Measures of Local Government Influence
A challenge in such empirical work involving historical time-series data is identifying
measures that are both valid and available. As discussed previously, a local government
may use a wide gamut of policy instruments to influence employment center growth.
Ideally, I would like to measure the influence of all such policies on employment center
growth. A more pragmatic approach, however, is to use reasonable proxies that are
correlated with “entrepreneurial governance” and allow me to accomplish the set task in a
reasonable time frame. After much deliberation and running several pilot studies, I
selected four broad measures of local government activities 1) Expenditure on
developmental activities; 2) Growth management index; 3) Growth promotion index; and
4) Index of local taxes and license fees on businesses.

Expenditure on Development Activities
I expect an entrepreneurial city to invest larger amounts of funds (either its revenues or
state/federal grants) in promoting development. I acknowledge that often (and
increasingly after Prop 13 in California) private developers and not cities invest in public
projects including cultural amenities, open spaces, and infrastructure. Nevertheless, cities
that desire the most growth are likely to invest higher amounts in development. It is very
difficult, if not completely impossible, to ascertain the true value of private investments

31

in public facilities or services made as part of development negotiations. The detailed
case studies of select cities will allow analysis of the development process at a finer
detail. Detailed methodology of case studies is discussed later.

The State Controller of California publishes annual expenditure reports for all the cities
in the state. From the reports, I calculate per capita expenditure on development for each
city in sample for each year between 1990 and 1995. Using this time period allows for
some time lag between investment and its outcome. The literature is not clear on how
much time lag should be allowed for between investment in development and its
outcomes. It may be possible that expenditure on development made during years prior
to 1990 is relevant too. However, I am positing development expenditure as a flow
measure. Hence cities that consistently spend more per capita on development should,
theoretically, generate more economic growth.

Using per capita figures makes them comparable across cities, which vary significantly in
their size and population. I count expenditure on transportation, infrastructure, and
utilities as developmental. Often significant expenditure on developmental activities is
done by the local community redevelopment agency (CRA). CRA revenues and
expenditures are reported separately and are not included in the city’s annual reports. I
add each redevelopment agency’s expenditure (excluding administrative expenses) to its
respective city’s development expenditure.

32

All figures are adjusted for inflation to reflect their real value in 1990 dollars. In cases
where an employment center stretches across more than one city, I weight each city’s per
capita expenditure in proportion to its respective share of jobs in that center. The
weighted figures are then aggregated to achieve the final number for that employment
center. The final index is the average annual per capita expenditure on development
between 1990 and 1995.

Indices of Growth Management and Growth Promotion
Growth management policies cover a wide gamut of regulations or rather restrictions
including permit caps, growth moratoria, growth boundaries, etc. Presence of growth
controls may reflect two things: one, the city is experiencing or is anticipating high
growth; and two, the city does not desire this growth, at least not all of it. Researchers
have suggested several motivations for growth controls including environmental reasons,
preservation of property values, or simply NIMBYsm. The exact motivations for growth
control may be debatable, but for the purpose of this study the incidence of act itself is
more relevant.

Glickfeld and Levine (1992) perhaps accomplished the most exhaustive of all studies of
growth control measures in California. They studied growth control measures in 443
jurisdictions in California. They surveyed each jurisdiction for fourteen types of growth
control measures adopted by the local government including permit caps, adequate public
facilities ordinances (APF), commercial square footage limits, industrial square footage

33

limits, commercial/industrial infrastructure limitations, commercial/industrial
downzoning, commercial height restrictions, growth management elements of general
plans, and Urban Growth Boundaries (UGBs) or greenbelts.

Glickfeld et al (1998) expand on Glickfeld and Levine (1992) study described in the
preceding paragraph. They supplement the prior study with a survey of growth
promotion policies enacted in all the cities in the Southern California region. The authors
then assign a score to each jurisdiction surveyed in Southern California based on the
combined strength of its growth management policies. The scores categorize
jurisdictions in five groups ranging from those with very strong growth management
policies to insignificant growth management policies. The authors admit that the
presence of a policy on paper, however, had little bearing on its actual enforcement.
Actual implementation of growth policies indeed varied widely from jurisdiction to
jurisdiction. The results are based solely on the enactment of growth policies and does
not account for the level of their enforcement or implementation.

The authors assign scores for growth promotion policies in a similar way. They survey 9
growth promotion instruments including fast track permit-processing, financial
incentives, and subsidies. The jurisdictions surveyed are then categorized in five groups
ranging from those with very strong growth promotion policies to those with insignificant
growth promotion policies.

34

I use the two publications described above to construct individual indices of growth
promotion and growth management for each employment center. Each index is a score
ranging between 1 and 3, with 1 reflecting weakest policies and 3 reflecting the strongest.
In cases where an employment center stretches across more than one city, I weight each
city’s score in proportion to its respective share of jobs in that center in 1990. All
weighted scores are then aggregated to achieve the final score for that employment
center.

Local Cost of Doing Business
A pro-growth city might be expected to impose lower costs of doing business inside its
jurisdiction in terms of local license fees and taxes. Kosmont Realty Corp has published
an annual Cost of Doing Business Survey for all cities in Southern California since 1995.
From the surveys, I calculate average local taxes/license fee on manufacturing,
wholesale, retail, professional office, and general office business categories in 1995. The
taxes/fees are calculated for the first $10 million in gross receipts. Local tax and
licensing fee rates do not tend to change much over time and are construed to be
indicative of a city’s attitude in general towards economic growth.

Control Variables
The control variables, have been adapted from Giuliano and Small (1999).

35

Size (in the base year)
Size of the employment center, in terms of total jobs, is a proxy to measure the influence
of agglomeration. Large centers might grow faster because of agglomeration economies,
but they may also grow slower because of diseconomies in congestion. I consider total
employment to be a better indicator of the level of economic activity as compared to say
commercial/industrial space. Space could lie vacant or be used at different levels of
intensity, which may distort the true levels of economic activity.

Employment Density (in the base year)
Employment density of an employment center is a proxy for congestion/ land scarcity
inside the center. Lower density centers may have more room to expand whereas the
opposite would be true for highly dense centers. Amount of land zoned for commercial
or industrial use, or commercial/industrial vacancy rates would be better measures of land
availability, but such data are not available for prior years.

Industry Composition
Employment centers that have a larger proportion of industry sectors that experienced
rapid regional growth could be expected to grow faster and vice versa. I allocate total
number of jobs in each center into 10 industry sectors: manufacturing, specialized
manufacturing, wholesale, retail, finance-insurance-real estate (FIRE), public
administration, entertainment, health, services, and business services. I exclude all
employment in agriculture, mining, and construction from the data. It is not reasonable

36

to use construction employment to explain determinants of growth as construction itself
is likely to be a result or outcome of growth or anticipation of future growth by
developers. As this study deals with urbanized area, it is not reasonable to expect
significant presence of agriculture and mining activities in the study area anyways.

Using regional growth rates of these industries, I calculate predicted employment change
inside each center:
ΔEp = Σi E
i
ĝ
i
(2)
where E
i
is beginning of period employment in industry sector i, and ĝ
i
is the sector’s
regional growth rate. This shift share analysis allows me to control for extra-regional
policies that may differentially affect growth or decline of industries such as decline of
defense related aerospace industry in the late 80s early 90s.

Relative Location in the Overall Region
“If economies of urbanization are important, then a [sub]center’s growth prospects may
depend on the centrality of its location within the overall region” (Giuliano and Small
1999, page 194). One of the reasons firms cluster is to take advantage of external
economies of scale. Some unique services may only be available in inside or in close
proximity of the CBD. Hence proximity to the CBD may facilitate growth of
employment centers. But, the CBD may also compete with other employment centers for
attracting both the resident labor force and employers. Furthermore, a more central
location may also come at the expense of higher congestion and land prices, thereby

37

offsetting some of the growth. I measure an employment center’s centrality as its
distance from the Los Angeles Downtown.

Labor Force Accessibility (in the base year)
Labor force accessibility is fundamental to the emergence and growth of employment
centers. In theory, locations with higher number of resident labor force (higher labor
force accessibility) would be able to hire workers for lower wage rates and therefore
would have more scope of growth. Accessibility is generally specified as an interaction
of an attractiveness factor and a cost factor. I calculate a center’s relative labor force
accessibility:
B
m
= Σ L
j
(E
m
e
-βrjm
/Σ E
k
e
-βrjk
) (3)
where B
m
is the relative labor force accessibility of employment center m and may be
viewed as attaching to each member of the labor force a probability, based solely on
commuting distance, of choosing to work in the employment center in question. Hence it
reflects the competitive locational advantage of the center’s firms. L
j
is the resident labor
force in tract j, r
jm
is the road network distance from tract j to the highest density tract of
employment center m; r
jk
is the road network distance from tract j to tract k; E
m
and

E
k
are
the total employment in tracts m and k, respectively; and β is a parameter. The parameter
β measures the commuting distance over which the attractiveness declines to a fraction e
-
1
of its peak value. The value of β is set equal to the regional average commuting
distance of 9.81 miles (in 1990). All 2475 urbanized census tracts of the region are
included in labor force accessibility calculations.

38

Proximity to Major Regional Airports
Many firms now have a national or even an international outlook (Giuliano 1998).
Proximity to airports may be more important to such firms than to the regional freeway
network, especially in mature metropolitan areas with extensive freeway network. Hence
proximity to a major airport should facilitate center growth. Proximity to airports is
measured in two parts: one, distance from the center to LAX, the region’s largest airport;
and two, distance from the center to the nearest of region’s three other major airports:
Ontario (in San Bernardino County), John Wayne (in Orange County), and Burbank (in
Los Angeles County).

Limitations of the Model
The challenge in any such empirical work is finding reasonable measures given the
complexity of multiple local policies and regulations that may directly or indirectly affect
employment growth. There is always the issue of omission of significant variables, some
of which are simply too difficult to measure, e.g. the model does not capture the
investments in infrastructure made by private developers, which according to some
theories may be an important determinant of employment center growth. There may be
differences in the on-book land use regulations and their actual enforcement, e.g. in some
municipalities growth controls may be duly enacted but seldom enforced strictly. The
model will not be able to account for such discrepancies. The model does not capture the
influence of historical path dependence arising out of previously built capital stock.

39

3.2 Examining Employment Growth across Cities
Substantial employment exists outside the employment centers. An argument could be
made that perhaps local government policies influence employment growth more
generally, and not just inside employment centers. To test the validity of this argument,
I conduct quantitative analysis of employment growth with city as the unit of analysis. I
include 78 cities from the urbanized portion of the Los Angeles CMSA for which
comparable data is available. The four local government influence variables used in
stage 1 described above remain the same. The model is slightly modified in its
specification and is described later.

3.3 Case Studies
There are several aspects of entrepreneurial governance that are not readily captured by a
purely quantitative approach. Hence, to supplement the quantitative analysis, I conduct a
complementary qualitative analysis – detailed case studies – of two cities. These case
studies illustrate the more nuanced efforts of local governments to induce employment
growth within their jurisdiction. Such efforts include, but are not limited to, forging
alliances/partnerships with private developers/businesses for purposes of development,
attracting firms by offering incentives and/or subsidies, inducing investment in local
infrastructure, creating special districts or specific plan areas for a more concerted
development effort in that location, and so on. The case studies allow delving further
back in time to explore lags between policy actions and impacts on growth. I study two

40

cities: Pasadena and Burbank in Los Angeles County. Detailed methodology of case
studies is described in Chapter 5.

41

Chapter 4: Quantitative Analysis

This chapter describes quantitative analysis of employment center growth in the Los
Angeles region. The chapter is structured in four parts. The first part presents
descriptive analysis of the data; the second part presents regression analysis of
employment center growth between 1990 and 2000; the third part presents regression
analysis of employment center growth between 1980 and 2000; and the fourth part
presents regression analysis of employment growth more generally (not restricted to
employment centers). The chapter concludes with a summary of findings from the
quantitative analysis.

4.1 Descriptive Analysis
Giuliano et al (2007) identify 48 employment centers in the region in 2000. Out of these
48 centers, I drop four as potential errors due to data conversion. All four centers are
located on isolated single census tracts, and lack of economic activity there has been
verified. I exclude two other centers from the analysis: one is located in the City of
Industry and the other in the City of Vernon. Both these cities are outliers in terms of
their extremely low population and a big employment base. City of Industry had a 2000
census population of 777 and Vernon had a 2000 census population of only 91! Because
per capita figures have been used for several local government activity measures, these
two cities would skew the results significantly. Hence the final number of centers used in

42

this analysis is reduced to 42. Consistent center boundaries have been used throughout
the analysis.

Figure 4.1 shows spatial distribution of the 42 employment centers in the region. Except
for the one in Riverside County, all centers are located inside Los Angeles and Orange
Counties. Most of the centers are located in close proximity of a major limited access
expressway (locally called freeway). However, not all freeway junctions necessarily
have an employment center. There are employment centers near each of the region’s four
major airports (marked with a star), except for the Ontario airport.

Figure 4.1 Employment Centers in the Region, 2000

43

Table 4.1 gives information on selected characteristics of the region’s employment
centers. The centers are ranked in the order of their size, i.e. Center 1 had the largest
number of jobs in 2000 whereas Center 42 had the lowest. There is a “rank size” effect,
i.e. a few very large and many smaller centers. The largest center had more than half
million jobs in 2000, whereas the smallest center had just a little over 10,000 jobs, a
result of the center identification criteria used (cut-off limit of 10,000 jobs).

The largest center in terms of area, LA Downtown, was spread over nearly 18,000 acres
while the smallest, Newport Beach, was spread across just over 600 acres. There was
also a large variation in employment density. The densest center, LA Downtown, had an
employment density of approximately 30 jobs per acre in 2000 whereas each of the least
dense centers (centers 6, 30, and 37) had employment density of 10 jobs per acre, just
above the cutoff.

Between 1990 and 2000, not all centers grew, some also lost employment (see Table 4.1).
Center 3 (Santa Ana-Costa Mesa-Irvine) was the biggest gainer, whereas Center 19
(Long Beach) was the biggest loser in terms of total number of jobs. Center 36
(Burbank) had the highest negative growth rate whereas Center 18 (Irvine-Lake Forest)
had the highest positive growth rate. There is a good mix of centers that lost jobs and the
ones that gained jobs to test the hypotheses re the determinants of employment center
growth.

44

Table 4.1 Select Characteristics of Employment Centers in the Los Angeles CMSA,
2000
Table 4.1 Select Characteristics of Employment Centers in the Los Angeles CMSA, 2000
ID Location Area
Total
Jobs
2000
Total
Jobs 1990
Job
Density
2000
Job
Growth
90-00
Percent
Growth
90-00
1 LA Downtown 17,949 539,645 563,717 30.1 (24,072) (4.3)
2 West LA-Santa Monica 13,773 421,049 394,691 30.6 26,358 6.7
3 Santa Ana- Costa Mesa-Irvine 16,648 291,673 249,354 17.5 42,319 17.0
4 Burbank-Glendale 6,786 132,149 105,578 19.5 26,571 25.2
5 Anaheim-Orange 7,202 123,462 108,840 17.1 14,622 13.4
6 Whitter-Santa Fe Springs 7,060 69,891 69,053 10.0 838 1.2
7 Los Angeles 2,781 70,896 50,403 25.5 20,493 40.7
8 Pasadena 2,823 58,424 59,687 20.7 (1,263) (2.1)
9 El Segundo 2,993 53,432 45,707 17.8 7,725 16.9
10 Anaheim 2,671 40,114 34,693 15.0 5,421 15.6
11
Los Alamitos-Cypress-Garden
Grove
2,662 37,687 28,608 14.2 9,079 31.7
12 Long Beach 1,726 37,125 45,524 21.5 (8,399) (18.4)
13 Los Angeles 1,695 34,267 24,549 20.2 9,718 39.6
14 Los Angeles 967 31,674 19,463 32.7 12,211 62.7
15 Inglewood-Los Angeles 1,257 32,865 34,932 26.2 (2,067) (5.9)
16 Downey 2,299 31,262 32,099 13.6 (837) (2.6)
17 El Monte-South El Monte 1,936 30,622 29,563 15.8 1,059 3.6
18 Irvine 2,606 29,382 16,546 11.3 12,835 77.6
19 Long Beach 2,475 30,312 59,428 12.2 (29,116) (49.0)
20 Torrance 2,123 29,817 28,808 14.0 1,009 3.5
21 Los Angeles 2,110 28,507 31,243 13.5 (2,736) (8.8)
22 Fullerton 1,964 23,934 25,683 12.2 (1,749) (6.8)
23 Los Angeles 1,915 25,440 32,958 13.3 (7,518) (22.8)
24 Culver City-Los Angeles 1,269 23,934 32,451 18.9 (8,517) (26.2)
25 Santa Ana 1,262 25,232 21,040 20.0 4,192 19.9
26 Los Angeles-Culver City 1,926 21,074 21,428 10.9 (354) (1.7)
27 Huntington Beach 1,317 20,623 24,637 15.7 (4,014) (16.3)
28 Los Angeles 1,657 27,772 26,224 16.7 1,548) 5.9
29 New Port Beach 610 15,492 9,100 25.4 6,392 70.2
30 Carson 1,488 14,707 20,323 10.0 (5,616) (27.6)
31 Torrance 1,356 14,262 17,340 10.5 (3,078) (17.8)
32 Covina 807 13,842 8,770 17.1 5,072 57.8
33 Riverside 658 14,604 15,322 22.2 (718) (4.7)
34 Los Angeles 947 13,471 13,389 14.2 82 0.6
35 Gardena 687 12,867 12,005 18.7 862 7.2
36 Burbank 1,055 12,938 26,790 12.3 (13,852) (51.7)
37 Monrovia 1,162 11,378 12,624 10.0 (1,246) (9.9)
38 Los Angeles 1,053 11,607 13,190 11.0 (1,583) (12.0)
39 Los Angeles 801 10,696 8,670 13.3 2,026 23.4
40 West Covina 953 11,390 10,738 12.0 652 6.1
41 Los Angeles 969 10,046 10,206 10.4 (160) (1.6)
42 Hermosa Beach - Redondo 773 10,020 7,725 13.0 2,295 29.7

45

Table 4.2 presents change in employment inside employment centers between 1990 and
2000. Centers that lost jobs are categorized as declining centers whereas centers that
gained jobs are categorized as growing centers. There is an almost even split in the
number of centers that gained and the ones that lost jobs. Between 1990 and 2000, on
average, the declining centers lost about 16 percent jobs while the growing centers gained
about 6 percent jobs. In terms of average total change, the declining centers lost more
than 6200 jobs each whereas the growing centers gained more than 9600 jobs each.

Table 4.2 Employment Centers: Change in Employment, 1990-2000
N Minimum Maximum Mean Std. Deviation
Declining Centers 20
-29,116
(51.7)
-186
(1.7)
- 6,269
(16.1)
7,891
(14.9)
Growing centers 22
82
(0.6)
42,319
(77.6)
9,621
(25.8)
10,853
(23.2)
All centers 42
-29,116
(51.7)
42,319
(77.6)
2,054
(5.8)
12,390
(28.8)
Note: Percent change in employment between 1990 and 2000 in parentheses

Control Variables
Table 4.3 presents means of control variables in the two growth categories. On average,
the growing centers had lower employment density in 1990 as compared to the declining
centers. The growing centers were located further from Downtown Los Angeles and
further from the Los Angeles International Airport (LAX) as compared to the declining
centers. Growing centers were in close proximity of a major airport (excluding LAX) as
compared to the declining centers. Note that only ‘employment density 1990’ and

46

‘proximity to the nearest airport’ variables have significantly different means in the two
growth categories.

Table 4.3 Means of Control Variables
Variable All
Centers
Declining
Centers
Growing
Centers
F
stat.
P
value
Employment 1990 57,217 55,290 58,967 0.012 0.912
Employment Density 1990 16.4 18.5 14.4 5.655 0.022
Distance to LA Downtown 17.6 16.5 18.6 0.464 0.500
Distance to LAX 19.7 17.9 21.4 0.857 0.360
Proximity to the Nearest
Airport Excluding LAX
(1/distance)
0.09 0.07 0.11 4.296 0.045
Relative Labor Force
Accessibility
26,753 22,749 30,394 1.239 0.272

Measures for Local Government Activities
This section presents descriptive statistics for of local government activity measures.
Overall, there is no clear pattern of association between employment center growth and
these measures. Indeed a very mixed picture emerges with local governments often
enacting counter-active policies, e.g. concurrent strong growth management and strong
growth promotion policies.

Growth Governance
Glickfeld et al (1999) have categorized 167 jurisdictions in the five-county Southern
California region based on the relative strength of growth promotion policies, enacted as
of 1992. The categories are very strong, strong, moderate, weak, and insignificant. I

47

combine cities in very strong and strong categories as a single category “strong.” I also
combine cities in weak and insignificant categories as a single category “weak.” Hence I
end up with three categories of growth promotion policies: strong, moderate, and weak.
Similarly, cities are categorized as strong, moderate, and weak for their growth
management policies. As the authors of the source study note, these categories are based
on policies that were enacted. There may be significant differences among jurisdictions
in terms of their enforcement. As there is no way of measuring what was enforced and
what was not at the time, I assume that all growth management and growth promotion
policies were equally enforced.

Table 4.4 Number of Centers by Growth Promotion & Growth Management
Policies
Category Weak
Promotion
Moderate
Promotion
Strong
Promotion
All
Centers
Weak Growth Management 2 0 3 5
Moderate Growth
Management
5 1 6 12
Strong Growth Management 2 5 18 25
All Centers 9 6 27 42

Table 4.4 presents cross-tabulation results of local growth promotion and growth
management indices of employment centers. Sixty percent of all centers were located in
cities with strong growth management policies whereas 65% were located in cities with
strong growth promotion policies. One would anticipate that cities that enact strong
growth promotion policies would have weak growth management and vice-a-versa.
Interestingly, however, several centers are located in places with both strong growth

48

management as well as strong growth promotion policies. There could be several
explanations as to why local governments would enact such counteractive policies.

One explanation is that local governments may want to restrict residential growth and
pursue commercial/industrial growth. However, this is already ruled out by Glickfeld et
al (1999) who offer two other possibilities: one, different departments within each city’s
bureaucracy are responsible for growth management and growth promotion and there is a
general lack of coordination among them; two, these policies were enacted during
different economic climates, i.e. promotion policies were enacted during economic
recession and growth controls enacted at the time of strong local growth. A third
explanation is also possible: local governments may enact such growth management
policies that give their jurisdiction an environmentally responsible (or responsive) face.
Similarly, local politicians could indulge in growth promotion as a part of their social
justice and affirmative action agenda. Enforcement of both types of policies, however,
would remain subjective. All these explanations are speculative and their validity cannot
be ascertained based on the survey data. However, it is reasonable to infer that the cities
with both strong growth promotion and growth management policies are indicative of
very active local governments.

Table 4.5 reports the crosstabs between the two growth categories of employment center
growth (between 1990 and 2000) and the three categories of local growth management
policies. Majority of the employment centers had strong growth management policies

49

whereas only 12% of the centers had weak growth management policies. Similar
proportion of both the growing centers (59%) the declining centers (60%) had strong
growth management policies.
4
This is not surprising as presence of growth management
policies is likely in locations experiencing high growth pressures. This becomes clear
with the cross-tabulation of employment center growth from 1980–1990 (the preceding
decade) and growth management policies (surveyed in 1989). Most of the centers with
moderate to strong growth management policies in 1992 experienced growth pressures
during the preceding decade (see Table 4.6).

Table 4.5 Number of Centers by Employment Center Growth (1990-2000) and
Local Growth Management Policies
Category Weak Moderate Strong Total
Declining Centers 2 6 12 20
Growing Centers 3 6 13 22
All Centers 5 12 25 42

Table 4.6 Number of Centers by Employment Center Growth (1980-1990) and
Local Growth Management Policies
Category Weak Moderate Strong Total
Declining Centers 0 1 6 7
Growing Centers 5 11 19 35
All Centers 5 12 25 42

Table 4.7 reports the crosstabs between the growth categories of centers and the
categories of growth promotion policies. Similar to growth management policies,

4
Note that because several cells had small counts a reliable test statistic of these crosstabs (such as the chi-
square) is not possible.

50

majority of employment centers had moderate to strong growth promotion policies. A
larger proportion of the declining centers (80%) had strong growth promotion policies as
compared to the growing centers (50%). Indeed a much higher proportion of the growing
centers (36%) had weak growth promotion policies as compared to the declining centers
(5%). The table does reveal any clear association between growth promotion policies
(surveyed in 1992) and employment center growth during the preceding decade.

Table 4.7 Number of Employment Centers by Center Growth (1990-2000) and
Local Growth Promotion Policies
Category Weak Moderate Strong Total
Declining Centers 1 3 16 20
Growing Centers 8 3 11 22
All Centers 9 6 27 42

Expenditure on Developmental Activities
The mean per capita annual expenditure on developmental activities during the 1990s
was approximately $750 whereas it was $1,215 in the 1980s (adjusted for inflation).
There was a large variation in expenditure: the lowest amount spent was $142 (Center 17
El Monte-South El Monte) whereas the highest was $1,636 (center 10, Anaheim). In
1990, 19 centers had developmental expenditures below the sample mean while 23
centers had expenditures above the sample mean.

Table 4.8 reports cross-tabulation between growth categories of centers and categories of
annual per capita expenditure on developmental activities in the 1990s. The low
expenditure category represents locations that spent less than the sample mean. The

51

opposite is true for the high category. A slightly smaller proportion of centers with low
expenditure on development declined as compared to the centers with high expenditure
on development (42 % versus 58%). Not all centers with high expenditure on
development experienced high growth. Indeed 48 % of the centers with low expenditure
experienced growth. Table 4.7 does not reveal any clear association between expenditure
on development and employment center growth.

Table 4.8 Number of Employment Centers by Growth Category and
Annual Per Capita Expenditure on Developmental Activities in the 1990s
Category Low High All Centers
Declining Centers 8 (42 %) 12 (52 %) 20
Growing Centers 11 (58 %) 11 (48 %) 22
All Centers 19 (100 %) 23 (100 %) 42

Local License Fees/ Taxes on Businesses
The range of local business license fees/ taxes is even more remarkable than the
expenditure on developmental activities. City of Irvine levied the lowest average license
fee on businesses (flat rate of $50) whereas the city of Los Angeles levied the highest
($29,404)!
5
The variable had a sample mean of 11,330. Twenty six centers had taxes
below the sample mean whereas 16 centers had local taxes above the sample mean.

5
This measure counts only the license fee and taxes on businesses levied by municipalities. There may be
other costs imposed on businesses by municipalities (e.g. utility user tax, transient occupancy tax, etc.) that
are not accounted for here. In some cases, such as the city of Irvine, other charges not reported here could
substantially increase the cost of doing business in that city.

52

Table 4.9 shows the association between growth of employment centers and the local
taxes on businesses. The low tax category represents locations that levied local taxes
below the sample mean of $11,330 whereas the opposite is true for the high category. Not
all centers with high taxes necessarily declined or the ones with low taxes grew.
Apparently, there is no clear association between the two variables.

Table 4.9 Number of Employment Centers by Growth Category and
Average Local Taxes/ License Fees
Category Low High All Centers
Declining Centers 11 (42 %) 9 (56 %) 20
Growing Centers 15 (58 %) 7 (44 %) 22
All Centers 26 (100 %) 16 (100 %) 42

The various descriptive statistics of local government activities described above do not
reveal any conclusive association between the four local government activities measures
and employment center growth. It could be argued that there are fifteen centers (ten are
completely within the Los Angeles city jurisdiction and five are partially inside) inside
the city of Los Angeles, which could potentially confound the results. I repeated the
above exercise after excluding all fifteen centers that were wholly or partially inside Los
Angeles, which did not change the results substantially.

4.2 Regression Analysis of Employment Center Growth
I use the variables described in the previous section to estimate regression models of
determinants of employment center growth. Two dependent variables are used: one,

53

absolute change in employment between 1990 and 2000, and two, percent change in
center’s employment between 1990 and 2000.

Table 4.10 Variable Definitions with Means and Standard Deviations
Variable Description Mean Std Dvn
E
00
– E
90
Absolute change in employment between
2000 and 1990 (dependent variable 1)
2,692 11,826
(E
00
– E
90
)/E
90
Percent change in employment between 1990
and 2000 (dependent variable 2)
6.09 29.10
Emp_90 1990 employment 44,862 69,469
Dens_90 1990 density (jobs/acre) 15.9 5.4
Dist_LA Distance to CBD (miles) 18.0 9.9
Pred_Gr Predicted growth based on average regional
industry sector growth (shift share)
1,508.
4
6,114.8
Pred_Gr_Rate Predicted growth rate based on average
regional industry sector growth (shift share)
1.9 9.2
Dist_LAX Distance to LAX (miles) 19.9 12.2
Prox_Air 1/Distance (in miles) to the nearest airport
excluding LAX
0.09 0.06
Rel_LFA Relative labor force accessibility 24,823 18,681
Exp_90 Average annual per capita expenditure on
developmental activities in 1990s
744 424
Control Growth Control Index - -
Promo Growth Promotion Index - -
Taxes Local taxes/ license fees on businesses 10,959 12,355

Table 4.10 presents variable names, definition and descriptive statistics. Note that this
analysis does not include the CBD. Urban theory on polycentricity is about why new
centers form in a metropolitan region implicitly implying that the dynamics of growth of
the “subcenters” is different from that of the main center. Furthermore, Los Angeles
downtown, the region’s CBD, enjoys numerous unique benefits such as an exceptionally
large government employment base, vested corporate interests, and long established
economic activities. Hence I exclude LA Downtown from this analysis.

54

Employment Center Growth 1990-2000
Table 4.11 presents the results from the ordinary least square (OLS) estimates of the base
models. The unit of analysis is an employment center. Results are reported for both
dependent variables. As noted previously, Los Angeles Downtown, the historical CBD
of the region, is excluded from the regression analysis.

Regression results for both model 1 and Model 2 presented in Table 4.11 are consistent
with the theoretical understanding of employment center growth. In Model 1, base year
employment is not significantly associated with absolute employment growth (p value =
0.172) but has a positive sign, which is consistent with the theory—larger centers present
larger agglomeration benefits and may therefore result in higher growth. As expected,
employment density is significantly and negatively associated with employment growth –
higher density implies agglomeration diseconomies and results in lower growth.
Distance to the CBD is significantly and negatively associated with employment center
growth implying higher growth in the periphery as compared to the region’s core.
Distance to LAX international airport is not significant. This is somewhat surprising, but
could be explained on the basis of substantial job losses during the early 1990s in the
aerospace industry located in close proximity of the LAX airport. Proximity to a major
airport, labor force accessibility, and predicted growth (shift-share) are all significantly
and positively associated with employment center growth, which is consistent with the
theory.

55

Table 4.11 OLS Estimates, Control Variables Only
Unit of Analysis = Employment center
N = 41 (sample excludes LA Downtown)
Model 1
Dep = E
00
– E
90

Model 2
Dep = (E
00
– E
90
)/E
90

Variable Beta Coeff. T Beta Coeff. T
Constant 1.986 4.574
Ln (Emp_90) 0.144 1.357 -0.229 1.659
Ln (Den_90) -0.467 5.273 -0.477 3.638
Ln (Dist_LA) -0.330 2.987 -0.558 3.505
Dist_LAX 0.168 1.535 0.379 2.376
Prox_Air 0.393 4.489 0.334 2.640
Rel_LFA 0.388 4.549 0.451 3.637
Pred_Gr 0.472 4.964 - -
Pred_Gr_rate - - 0.260 2.266
Adj R squared 0.771 0.515
Bold = significant at p < 0.01; Italics= significant at 0.01 < p < 0.05
Underline = significant at 0.05 < p < 0.10

Signs and significance of the control variables in Model 2 are consistent with those from
Model 1, total base year employment being the only exception. Total employment is
significant and has a negative sign implying larger centers have lower grower rates. This
is not surprising as inside larger centers agglomeration diseconomies may outweigh
agglomeration benefits, thereby resulting in slower or even negative growth rate.

56
Table 4.12 OLS estimates, Model 1
Unit of Analysis = Employment center; N = 41 ; Dependent Variable = E
00
– E
90

Add Exp_90 Add Control Add Promo Add Taxes All Var.
Variable Beta
Coeff.
T Beta
Coeff.
T Beta
Coeff.
T Beta
Coeff.
T Beta
Coeff.
T
Constant 1.968 1.526 1.984 1.883 0.403
Ln (Emp_90) 0.161 1.528 0.184 1.610 0.152 1.400 0.142 1.281 0.342 2.790
Ln (Den_90) -0.448 5.072 -0.502 5.231 -0.460 5.062 -0.466 5.150 -0.528 5.873
Ln (Dist_LA) -0.363 3.256 -0.332 2.999 -0.332 2.971 -0.331 2.909 -0.407 3.796
Dist_LAX 0.181 1.668 0.196 1.725 0.155 1.355 0.166 1.408 0.325 2.653
Prox_Air 0.400 4.625 0.364 3.901 0.390 4.389 0.395 4.235 0.291 3.125
Rel_LFA 0.411 4.795 0.379 4.419 0.374 4.127 0.388 4.472 0.417 4.713
Pred_Gr 0.460 4.890 0.459 4.772 0.461 4.673 0.473 4.873 0.389 4.081
Exp_90 -0.115 1.382 -0.320 5.686
Control 0.087 0.953 0.259 2.390
Promo -0.045 0.482 -0.011 0.108
Taxes -0.006 0.066 0.107 0.942
Adj R squared 0.777 0.770 0.765 0.763 0.802
Bold = sig. at p < 0.01; Italics= sig. at 0.01 < p < 0.05; Underline = sig. at 0.05 < p < 0.10

57
Table 4.13 OLS estimates, Model 2
Unit of Analysis = Employment center; N = 41; Dep = (E00 – E90)/E90
Add Exp_90 Add Control Add Promo Add Taxes All Var.
Variable Beta
Coeff.
T Beta
Coeff.
T Beta Coeff. T Beta
Coeff.
T Beta
Coeff.
T
Constant 4.502 4.046 4.438 4.286 3.391
Ln (Emp_90) -0.228 1.623 -0.190 1.290 -0.230 1.640 -0.226 1.563 -0.154 0.904
Ln (Den_90) -0.476 3.507 -0.517 3.642 -0.484 3.508 -0.479 3.555 -0.528 3.503
Ln (Dist_LA) -0.560 3.394 0.557 3.479 -0.558 3.452 -0.555 3.378 -0.588 3.378
Dist_LAX 0.380 2.336 0.411 2.477 0.386 2.313 0.384 2.248 0.455 2.275
Prox_Air 0.334 2.598 0.297 2.190 0.336 2.606 0.330 2.429 0.280 1.819
Rel_LFA 0.453 3.525 0.441 3.519 0.459 3.430 0.450 3.567 0.466 3.159
Pred_Gr_rate 0.258 2.181 0.254 2.196 0.271 2.060 0.261 2.217 0.240 1.710
Exp_90 -0.009 0.072 -0.121 0.626
Control 0.102 0.465 0.028 0.186 0.172 0.964
Promo 0.031 0.095 0.025 0.141
Taxes 0.004 0.022
Adj R square 0.499 0.509 0.500 0.499 0.465
Bold = sig. at p < 0.01; Italics= sig. at 0.01 < p < 0.05; Underline = sig. at 0.05 < p < 0.10

58

Table 4.12 shows that all four measures for local government activities bear statistically
insignificant association with both dependent variables of employment center growth.
Furthermore, measures of per capita annual expenditure on development activities and
growth promotion policies have a negative sign implying inverse relationship with
employment growth. This alludes to the efforts being made by the local governments to
promote growth at economically declining locations. Similarly, the growth control policy
measure has a positive sign implying stricter growth controls at locations experiencing
higher growth. The local tax variable also has a positive sign, which is somewhat
surprising and indicates a rather inconsequential relationship between local economic
growth and the prevailing tax rates.

Table 4.13 presents regression results for Model 2, where the dependent variable is
percent employment change between 1990 and 2000. The results from Model 2 are
consistent with those from Model 1 described above and show that the local policy
variables are not significantly associated with employment center growth rate. The
control variables remain consistent with theoretical expectations. The average annual per
capita expenditure on development variable (exp_90) bears a negative sign and the
growth control variable (control) bears a positive sign, both being opposite of what is
theoretically expected.

The regression analysis of the models presented above show that the local development
policies as measured in this study do not bear statistically significant association with

59

employment center growth. Implications of these results are discussed later in this
chapter.

4.3 City Level Analysis
During the study period, substantial employment growth has occurred outside the
employment centers. An argument could be made that the local government policies
might have been instrumental in promoting employment growth in general and not
particularly inside the employment centers. If this were true, we should be able to test
our hypotheses using city as our unit of analysis instead of an employment center. This
section presents quantitative analysis of employment growth inside the 77 cities (out of
180 total cities) in the Southern California Region for which comparable data was
available for the study period.

Figure 4.2 Cities included in the analysis

60

Including less than half the cities may be problematic if the exclusion is systematic.
However, such is not the case. Figure 4.2 shows the cities that are included in this
analysis. Cities included in the analysis are fairly well spread across the urbanized
portion of the Los Angeles CMSA. The total 1990 employment of all the cities in the
sample is 5.2 million and the total 1990 population is9.6 million, which represents 75.6 %
and 68.5% of the region’s total employment and population, respectively.

Table 4.14 Employment Growth inside Cities
Category Count Percentage Average Change in
Employment
Declining Cities 38 49 -7,505
Growing Cities 39 51 9,378
All Cities 77 100 1,045

Between 1990 and 2000, 38 cities lost employment while 39 gained employment (see
Table 4.14). Cities that gained jobs are categorized as growing cities and the ones that
lost jobs are categorized as declining cities. On average, the declining cities lost about
5,357 jobs each, while the growing cities gained about 8,702 jobs each.

Descriptive Analysis of Local Government policies
Growth Governance
Table 4.15 presents the association between local growth promotion policies and local
growth management policies. There is almost even split across the three growth
management categories. However, a slightly higher number of cities had strong growth

61

promotion policies in place as compared to those with weak or moderate growth
promotion policies. Interestingly, 14 cities had both strong growth management policies
as well as strong growth promotion policies.

Table 4.15 Number of Cities by their Growth Promotion and Growth
Management Policies
Category Weak
Management
Moderate
Management
Strong
Management
All Cities
Weak Promotion 8 7 6 21
Moderate Promotion 9 10 6 25
Strong Promotion 9 8 14 31
All Cities 26 25 27 77

Table 4.16 Number of Cities by Employment Growth and Local Growth
Management Policies
Category Weak Moderate Strong All Cities
Declining Cities 15 (39.5) 11 (28.9) 12 (31.6) 38 (100)
Growing Cities 11 (28.2) 14 (35.9) 14 (35.9) 39 (100)
All Cities 26 (33.8) 25 (32.5) 26 (33.8) 77 (100)
Pearson Chi-Square Value= 0.572 df = 2 Asymp. Sig. (2-sided ) = 0.572
Note: Parentheses contain row percentages

Table 4.16 presents the association between the growth category of cities and the relative
strength of local growth management policies. A slightly lower percentage of declining
cities (31.6 %) had strong growth promotion policies as compared to the growing cities
(35.9 %). The table does not reveal any clear association between employment growth
and local growth promotion policies.

62

Table 4.17 presents the association between the growth category of cities and the relative
strength of local growth promotion policies. While more than 46% of all the growing
cities had strong growth promotion policies, 34.2% of declining cities too had strong
growth promotion policies. The table reflects the inconclusive association between local
growth promotion policies and employment growth.

Table 4.17 Number of Cities by Employment Growth and Local Growth
Promotion Policies
Category Weak Moderate Strong All Cities
Declining Cities 13 (34.2) 12 (13.6) 13 (34.2) 38 (100)
Growing Cities 8 (20.5) 13 (33.3) 18 (46.2) 39 (100)
All Cities 21 (27.3) 25 (32.5) 31 (40.3) 77 (100)
Pearson Chi-Square Value= 2.024 df = 2 Asymp. Sig. (2-sided ) = 0.363
Note: Parentheses contain row percentages

Expenditure on Developmental Activities
The mean per capita annual expenditure on developmental activities during the 1990s
was approximately $207. There was a large variation in expenditure: the lowest amount
spent was $45 (city of Bell) whereas the highest was $1,472 (city of Irwindale). The
declining cities had a mean development expenditure of $214 whereas growing cities had
a mean development expenditure of $202. Fifty-three cities had developmental
expenditures below the sample mean while 24 spent amounts greater than the sample
mean (see Table 4.18).

Table 4.18 reports cross-tabulation between growth categories of cities and categories of
annual per capita expenditure on developmental activities during the 1990s. The low

63

expenditure category represents the cities that spent less than the sample mean. The
opposite is true for the high category. Substantially higher proportions of cities that were
both declining and growing had low annual per capita expenditure on development
whereas. Not all cities with high expenditure on development experienced growth.
Indeed a higher number of cities with high expenditure on development were declining
rather than growing. This is indicative of an inconclusive association between the two.

Table 4.18 Cities by Growth Category and Annual Per Capita
Expenditure on Developmental Activities in the 1990s
Category Low High All Centers
Declining Cities 25 (65.8) 13 (34.2) 38 (100)
Growing Cities 28 (71.8) 11 (28.2) 39 (100)
All Cities

53 (68.8) 24 (31.2) 77 (100)
Pearson Chi-Square Value= 0.324 df = 1 Asymp. Sig. (2-sided ) = 0.569
Note: Parentheses contain row percentages

Local Taxes on Businesses
Table 4.19 shows the association between employment growth of cities and the local
taxes on businesses. The low tax category represents cities that levied local taxes below
the sample mean of $ 3,811 whereas the opposite is true for the high category. Fifty
seven cities had low tax rates whereas 20 cities had high tax rates. The declining cities
had a mean tax rate of $4,658 whereas the growing cities had a mean tax rate of $2,986.
A substantially higher percentage of growing cities had local taxes lower than the sample
mean, but so did declining cities.

64

Table 4.19 Cities by Growth Category and Local License Fees/ Taxes on
Businesses
Category Low High All Centers
Declining 26 (68) 12 (32) 38 (100)
Growing 31 (80) 8 (20) 39 (100)
All 57 (74) 20 (26) 114 (100)
Pearson Chi-Square Value=1.22 df =1 Asymp. Sig. (2-sided ) = 0.268
Note: Parentheses contain row percentages

The descriptive analysis of cities does not indicate any clear association between local
government measures and employment growth.

Figure 4.3 Planning Districts in the city of Los Angeles

Regression Analysis of Employment Growth Inside Cities, 1990-2000
Los Angeles, the region’s central city, is by definition the largest. With its immense size,
the city could potentially skew the results. For example, distance to airports, distance to

65

the CBD, and labor force accessibility variables are measured from the centroid of each
city. Since Los Angeles is a very large spatial unit, distance to its centroid may not be a
good proxy for these measures across the entire city. Hence I divided Los Angeles into
the 38 “planning districts” defined by the city (see Figure 4.3). Each planning district
would now be treated as an independent city. While the local government variables such
as expenditure per capita on development etc. will remain the same for all these 38
districts, control variables such as distance to the nearest airport, labor force accessibility,
etc. will be calculated separately for each. This takes the number of observations in
regression analysis to 114.

Table 4.20 Variable Definitions with Means and Standard Deviations
Variable Description Mean Std Dvn
E
00
– E
90
Absolute change in employment between
2000 and 1990 (dependent variable 1)
562 11,273
E
00
/ E
90
Percent change in employment between 1990
and 2000 (dependent variable 2)
1.03 0.25
Emp_90 1990 employment 62,940 191,409
Dens_90 1990 density (jobs/acre) 6.77 9.67
Dist_LA Distance to CBD (miles) 20.75 15.10
Pred_Gr Predicted growth based on average regional
industry sector growth (shift share)
3227 11,851
Dist_LAX Distance to LAX (miles) 24.19 17.13
Prox_Air 1/Distance (in miles) to the nearest airport
excluding LAX
0.09 0.11
Rel_LFA Relative labor force accessibility 117,702 329,347
Exp_90 Average annual per capita expenditure on
developmental activities in 1990s
243 15091
Control Growth Control Index
Promo Growth Promotion Index
Taxes Local taxes/ license fees on businesses 13865 15091
Pop_grt Population growth between 1980 &1980 24851 60648
Pop_gr Ratio of 1990 and 1980 populations 1.41 0.72
Med_val Median value of a single family home in 1990 223367 90627

66

Table 4.20 presents description, mean, and standard deviation of the variables used in
regression analysis. Table 4.21 and 4.22 present results from the ordinary least square
(OLS) estimates. The unit of analysis is a city. The dependent variable is change in total
employment inside cities between 1990 and 2000. Table 4.21 presents control variables
only. Total employment, employment density, labor force accessibility, and predicted
growth (shift-share), proximity to a major airport, and LA County dummy are all
statistically significant. Distance to Los Angeles CBD, distance to LAX, and proximity to
a major airport are all not significant.

Table 4.21 OLS Estimates, Control Variables Only
Unit of Analysis = City; N = 114 ; Dep = E
00
– E
90

Control
Variables
Add Population
Growth 80-90
Add LA County
Dummy
Variable Beta
Coeff.
T Beta
Coeff.
T Beta
Coeff.
T
Constant 1.631 1.410 1.070
Emp_90 -1.377 1.978 -1.368 1.935 -1.489 2.132
Den_90 -0.192 1.822 -0.197 1.857 -0.190 1.814
Dist_LA 0.126 0.552 0.127 0.545 -0.087 0.345
Dist_LAX 0.054 0.224 0.050 0.198 0.020 0.081
Prox_Air 0.044 0.495 0.041 0.459 0.014 0.154
Rel_LFA 0.223 1.929 1.438 2.084 1.538 2.259
Pred_Gr 0.200 2.130 0.202 1.929 0.178 1.717
Pop_gr 0.011 0.110 -0.009 0.090
LA County
Dummy
-0.299 2.072
Adj R
square
0.222 0.218 0.243
Bold = significant at p < 0.01
Italics= significant at 0.01 < p < 0.05
Underline = significant at 0.05 < p < 0.10

67

Table 4.22 adds local government variables to the model. Only Growth Promotion
variable is statistically significant, but has a negative sign, opposite of what is anticipated.
This implies local government efforts in promoting local economic development in cities
that were losing jobs. All other measures for local government activities remain
statistically insignificant.

These results could be interpreted in two ways. First interpretation may be that the
measures used for local government efforts are inadequate. Inadequacies might be in not
allowing sufficient time-lags, or simply inappropriate choice of policy variables. The
other interpretation is the opposite: the measures of local government efforts used here
really don’t matter. It would be safe to conclude that local government policies, as
measured in this study, are ineffective in promoting local employment growth.

68
Table 4.22 OLS Estimates, Add Local Government Variables
Unit of Analysis = City; N = 114 ; Dep = E
00
– E
90

Add Exp_90 Add Promo Add Control Add Taxes
Variable Beta
Coeff.
T Beta
Coeff.
T Beta
Coeff.
T Beta
Coeff.
T
Constant 0.921 1.696 1.728 2.014
Emp_90 -1.590 2.251 -1.432 2.042 -1.437 2.040 -1.469 2.086
Den_90 -0.191 1.828 -0.192 1.861 -0.187 1.793 -0.196 1.872
Dist_LA 0.109 0.431 -0.123 0.493 -0.083 0.307 -0.138 0.501
Dist_LAX 0.030 0.120 0.076 0.308 0.045 0.174 0.109 0.413
Prox_Air 0.006 0.071 -0.011 0.128 -0.001 0.014 -0.006 0.060
Rel_LFA 1.631 2.371 1.469 2.148 1.478 2.151 1.526 2.219
Pred_Gr 0.179 1.718 0.192 1.867 0.197 1.891 0.180 1.710
Pop_gr 0.001 0.011 -0.023 0.233 -0.030 0.296 -0.037 0.373
LA County
Dummy
-0.324 2.210 -0.304 2.099 -0.299 2.045 -0.352 2.292
Exp_90 0.086 0.957 0.110 1.240 0.110 1.228 0.093 1.026
Promo -0.166 1.946 -0.148 1.522 -0.181 1.779
Control -0.041 0.383 -0.103 0.855
Taxes 0.150 1.109
Adj R
square
0.213 0.262 0.256 0.258
Bold = significant at p < 0.01
Italics= significant at 0.01 < p < 0.05
Underline = significant at 0.05 < p < 0.10

69

4.4 Discussion
The models presented above indicate a general lack of effectiveness of local government
efforts in promoting employment growth, both inside employment centers and within city
limits. The results also show that economic factors indeed play a significant role in both
employment center and city growth. These results are on the one hand counterintuitive to
our expectations from a planning perspective, yet on the other hand they are consistent
with our knowledge of local economic development efforts.

The results from this study could be interpreted in several ways. One, emergence and
growth of employment centers is a part of the larger decentralization phenomenon in
general. As population decentralizes, so do jobs. Agglomeration economies are indeed
important to some firms, which leads them to cluster and form multiple employment
centers. Access to the airports matter for growth of employment centers and hence there
are employment centers in close proximity of all the region’s major airports. Within this
largely economic explanation there is little scope for the local governments to influence
employment growth.

Second, local government policies are reactive instead of being proactive. There is some
indication that locations with high growth have higher incidence of growth controls
whereas locations facing employment losses have high incidence of strong growth
promotion policies and tend to spend higher amounts on developmental activities. It may

70

be more difficult to achieve desired success when policies have to work against the
prevailing economic environment.

Third, although the number of observations is limited, one would think that having a
good mix of centers that grew and declined would produce interesting results. However,
this didn’t turn out to be the case. Furthermore, the fact that nearly one third of the
employment centers were wholly or partially located in a single city (Los Angeles) was
also a limitation of the sample. While all these centers shared the same local government
policies, as measured here but not likely in reality, almost half of them gained jobs while
the other half lost jobs during the study period.

Finally, the proxies for local government activities used in this study may be inadequate.
This could be on account of several reasons: one, policies may require longer time
periods to bear results than the one used in this study. This is a valid argument but with
the available data a longer time series test is not presently possible. Two, the
developmental expenditure variable does not capture investment made by private
agencies in public infrastructure, which could be substantial. However, there is no data
available on the same presently. Three, the study takes policies enacted on their face
value, ignoring the differences in terms of level of enforcement. This is true, but there is
no way of ascertaining what was actually enforced and what wasn’t. Such inadequacies
of local government proxies are limitations of the quantitative study. Some of these

71

issues are addressed in the detailed case studies of two cities in the region, Pasadena and
Burbank, discussed in the next chapter.

72

Chapter 5: Qualitative Analysis

This chapter presents case studies of two cities: Burbank and Pasadena, both in Los
Angeles County. The quantitative analysis presented in the previous chapter suggests
that local government interventions (as measured in this study) did not have significant
influence on employment center growth in the Los Angeles region. These case studies
offer a detailed qualitative analysis of local government efforts, and allude to a more
nuanced picture of employment center growth. The case studies confirm that economic
forces indeed play a key role in employment center growth. However, the case studies
also suggest that local governments may play a role in employment center growth albeit
in very context-specific ways.

5.1 Selection Criteria
Selection of case studies needs some careful thinking. The goal of this qualitative
analysis is not verification or falsification, but explanation of the ways in which a local
government may affect employment growth within their jurisdiction and how those
efforts are conditioned by the market/ economic forces. According to Flyvbjerg (2004,
page 420):
When the objective is to achieve the greatest possible amount of information on a
given problem or phenomenon, a representative case or a random sample may not
be the most appropriate strategy. This is because the typical or average case is
often not the richest in information. Atypical or extreme cases often reveal more
information because they activate more actors and more basic mechanisms in the
situation studied.

73

Based on this reasoning, the city of Burbank is chosen as one case study. The city faced
an economic catastrophe in 1990 when Lockheed, its then biggest private employer, shut
down. The city, however, recovered robustly from large economic losses over the
following decade, not only because of its recovery efforts thereafter, but also its
economic development planning before (and with no relation whatsoever to) the fact.
Indeed, as the case study reveals, this economic catastrophe aided the local government
to achieve several economic development goals, which may not have been possible had
Lockheed not shut down. The city implemented a specific plan that appeased the
residents (overcoming Not In My Back Yard syndrome) while still allowing for the
motion picture industry’s growth in the city: it promoted the development of a large scale
retail center in the city’s downtown; and the 300 acre parcel vacated by Lockheed
became an asset for the city allowing development opportunities previously not possible.

The other case study, the city of Pasadena, is equally interesting and extreme. The city’s
employment center, the city’s downtown, had become extremely blighted and run down
in the post-war period, between 1950 and 1980. The city’s downtown revitalization
strategy not only employed common and popular economic development strategies of
those times, such as large scale developments using tax increment financing, but also a
series of innovative and coordinated steps between 1980 and 2000 (Kotin and Szalay
2007). Among other strategies, the city implemented specific plans to preserve its
historic architectural character, implemented physical planning strategies to make the

74

downtown pedestrian friendly, and adopted a parking policy that not only contributed to
city revenues, but also paid for the much needed public improvements in the downtown.
The two cities share several common characteristics in terms of their age, relative
location in the region, access to the transport network, etc. Each of the two cities had at
least one employment center in 2000 (see Table 5.1). In both cases, the local government
was spurred into action by the anticipation of extremely dire economic prospects if
nothing was done, i.e. reaction to adverse economic circumstances. However, the two
cities allude to two very different sets of local government efforts aimed at promoting
local economic development. In the case of Burbank, the local government efforts were
conditioned by unexpected events and unintended consequences—things that nobody
really anticipated would happen and therefore did not “plan” for. In the case of
Pasadena, a very active city in the region, much of city’s local economic development
efforts were rather unconventional for the times.

Table 5.1 Selected Characteristics of Burbank and Pasadena
Burbank Pasadena
Year Incorporated 1911 1886
Area (sq. miles) 17.5 23.2
Population 1990 93,643 131,591
Population 2000 100,316 133,936
Distance from Downtown Los Angeles (miles) 12 10
Number of Employment Centers 2 1

5.2 Framework of Analysis
As stated previously, a key purpose of this analysis is to explain in some detail the ways
in which a local government may affect employment growth within their jurisdiction.

75

+The case studies are meant to supplement the findings from the quantitative analysis
presented in Chapter 4. The case studies attempt to address three limitations of the
quantitative analysis:
1. Time Lags: The study period of the quantitative analysis was ten years – 1990-2000.
It could be argued that ten years may not be long enough for several local
government efforts to produce their desired effects. Due to data limitations, it was
not possible to do a systematic quantitative analysis over a longer period. However,
it is possible to do a qualitative analysis over a longer period to address the issue.

2. A more nuanced explanation: Another potential criticism of the quantitative analysis
is that it reduces the local government efforts to four proxies. It could be argued that
the story of local economic growth, and hence employment center growth, is more
nuanced and much richer than that captured in the reduced econometric model. A
qualitative analysis would allow the researcher to overcome some of the data
limitations of the quantitative analysis, thereby enabling him to present a narrative of
economic growth that is not only more nuanced, but also more “complete.”

3. The enforcement problem: One of the limitations of quantitative variables was that
they could not capture the differences between enactment of policies and their
enforcement by the local governments. For example, several cities have enacted
strong growth control policies but seldom enforce them strictly (Glickfeld et al
1999). Detailed case studies may be able to reveal more about such discrepancies
and allude to their effect on employment growth.

76

4. Finally, the quantitative analysis does not capture several aspects of the
“entrepreneurial city” hypothesis. For example, Sagalyn (1997) discusses “deal-
making” and “negotiating” as two key features of entrepreneurial cities, things that
cannot be precisely quantified. The case studies would allow the researcher to
address such aspects of local government efforts that were not captured in the
quantitative analysis.

5.3 Case Study 1: The City of Burbank
The city of Burbank, one of the early twentieth century suburbs of the Los Angeles
region, was incorporated in 1911. The city is located in Los Angeles County, about 12
miles north-west of Downtown Los Angeles. Burbank has been a home for motion
picture industry for long – First National Pictures located in Burbank during the early
1920s and was soon taken over by another company founded by the four Warner
brothers. Several large motion picture studios including Warner Bros., NBC Universal,
and The Walt Disney Company are located in Burbank.

Burbank is very well-connected to the region’s transportation network. The Golden State
Freeway (Interstate-5) bisects Burbank while the State Route 134 Freeway cuts through
the southern tip of the City (see Figure 1). One of the region’s major passenger airports –
the Burbank/Glendale/Pasadena Airport – is located in the north east of the city. More
than 70 domestic flights operate every day from the airport, ferrying 4.9 million travelers

77

each year (city of Burbank’s official website). Metrolink, the region’s commuter rail, has
a stop in the city.

The city has two employment centers: one located in the city’s north-west area adjacent
to the Burbank Airport (center # 36), and the other stretching along the southeast
boundary of the city (center # 4). The latter contains the city’s motion picture industry
cluster, ‘The Media District,’ and a shopping mall ‘The Burbank Center’ (see Figure 5.1).
While center #36 lost approximately 14,000 jobs between 1990 and 2000, center #4
gained more than 18,000 jobs during the same period.

Figure 5.1 The City of Burbank

78

The city of Burbank is governed by an elected mayor and a 5 member city council. The
city employs a City Manager to professionally manage city’s affairs. The city has a
Community Development Department, which is responsible for physical planning,
transportation planning, processing licenses and permits, promoting housing and local
economic development, etc.

Demographic Profile
Table 5.2 General Population Characteristics, Burbank and L.A. County
Burbank L.A. County
1990 2000 2000
Total Population 93,643 100,316 9,519,338
Percent Male 48.60 48.5 49.4
Percent Female 51.40 51.5 50.6
Percent White 82.6 72.2 48.7
Percent African American 1.7 2.1 9.8
Percent Asian 6.8 9.2 11.9
Percent Other Races 8.4 16.5 29.6
Percent Hispanic (of any race) 22.6 24.9 44.6
Median Age (in years) -- 36.4 33.9
Source: U.S. Census Bureau

Burbank had census 2000 population of 100, 316. Between 1990 and 2000, Burbank’s
total population increased by more than 7 percent (see Table 5.2). Nearly 52 percent of
the city’s population was male and 48 percent female. The city’s population has
remained predominantly white, although the share of white population dropped
significantly between 1990 and 2000. The share of Asian population in the city
increased steadily during the same period. Hispanics (of any race) were almost one

79

quarter of the city’s total population in 2000. The city had a much higher percentage of
white population and much lower shares of African Americans and Hispanics as
compared to L.A. County (see Table 5.2). The city’s population had a median age of
36.4 years in Census 2000, slightly higher than the County’s median age of 33.9 years.

Table 5.3 compares housing characteristics for the city of Burbank with LA County.
Burbank had a higher percentage of its total housing stock occupied (not vacant) than LA
County. More than 56 percent of the city’s occupied housing units were renter occupied.
The city had a higher median value of a single family home and higher median rent than
LA County.

Table 5.3 Housing Characteristics, Burbank and L.A. County 2000
Burbank L.A. County
Total Housing Units 42,216 3,270,909
Percent Occupied Housing Units 97.11 95.8
Percent Owner Occupied 43.53 47.9
Percent Renter Occupied 56.47 52.1
Percent Vacant Housing Units 2.89 4.2
Median Value of Single Family Home 256,400 209,300
Median Rent 778 704
Source: U.S. Census Bureau

Table 5.4 presents data on key economic characteristics for the populations of Burbank
and Los Angeles County. City of Burbank had a higher proportion of its population aged
16 years and older in the labor force as compared to LA County. Furthermore, Burbank
had a lower unemployment rate than the County. On average, Burbank’s labor force had

80

a slightly shorter commute than that of LA County. The city residents were more affluent
as compared to the whole County with significantly higher median household income and
per capita income. The poverty rate in the city was less than half of that of the County.
Burbank’s population had better educational attainment than the County with more than
83 percent of the city’s population aged 25 years or older achieving high school or higher
level of education.

Table 5.4 Economic Characteristics, Burbank and L.A. County, 2000
Burbank L.A. County
Population 16 years or older 80,339 7,122,525
Percent Population 16 years or older in the Labor Force 65.7 60.5
Percent Population 16 years or older in the Labor
Force and employed
65.6 55.5
Percent Population 16 years or older in the Labor
Force and unemployed
4.1 5.0
Percent Population 16 years or older not in the labor
force
34.3 39.5
Mean Travel Time to Work (in minutes) 25.1 29.4
Median Household Income 47,467 42,189
Per capita Income 25,713 20,683
Percent Population Below Poverty Level 8.1 17.9
Population 25 years and older 70,523 5,882,948
Percent High School Graduate or higher 83.1 69.9
Percent Bachelor’s degree or higher 29.0 24.9
Source: U.S. Census Bureau

Table 5.5 presents shares of the employed labor force in the city of Burbank by industry
of employment. These industry sectors are based on the North American Industry
Classification System (NAICS). As the city’s population is relatively educated, it is
reasonable that a higher proportion of the the city’s labor force is employed in skilled

81

professions. The highest proportion of Burbank’s labor force was employed in
educational, health, and social services (almost 16%). Information industry employed
14.3 percent of Burbank’s labor force. Note that in NAICS the “information” sector
includes motion picture and sound recording industries, which have a big presence in the
city of Burbank.

Table 5.5 Employed Civilian Population 16 years and over by Industry, 2000
Burbank L.A. County
Industry Sector Number Percent Number Percent
Agriculture, forestry, fishing and
hunting, and mining
100 0.2 10,188 0.3
Construction 2,126 4.3 202,829 5.1
Manufacturing 4,959 10.0 586,627 14.8
Wholesale trade 1,547 3.1 184,369 4.7
Retail trade 5,120 10.4 416,390 10.5
Transportation and warehousing, and
utilities
1,894 3.8 198,375 5.0
Information 7,079 14.3 213,589 5.4
FIRE 4,238 8.6 272,304 6.9
Professional, scientific, management,
administrative, and waste management
services
5,663 11.5 455,069 11.5
Educational, health and social services
7,844 15.9 722,792 18.3
Arts, entertainment, recreation,
accommodation and food services
4,578 9.3 332,753 8.4
Other services (except public
administration)
2,568 5.2 233,193 5.9
Public administration 1,683 3.4 124,937 3.2
Source: U.S. Census Bureau

Local Revenue
Table 5.6 presents data on the city of Burbank’s revenues. Functional revenues are
revenues that are either generated from direct services or associated with a specific

82

service such as grant conditions or statutory requirements; general revenues are revenues
that cannot be associated with a specific service. These revenues do not reflect special
assessment funds, internal service funds, pension trusts, or agency transactions. The table
shows that between 1990 and 2005, the city’s general revenues and sales tax revenues
declined in real terms while the city’s functional revenues and property tax revenue
increased.

Table 5.6 Burbank Revenues
1984-85* 1990-91 1994-95* 2004-05*
General Revenue 50,048,162 68,623,383 62,347,572 62,350,727
Functional Revenue 118,422,418 148,801,519 150,533,762 257,361,766
Property Tax 9,352,484 12,684,238 9,860,561 13,444,036
Sales Tax 10,941,931 14,048,799 13,528,182 13,646,835
*Adjusted for inflation; Constant 1990 Dollars

Table 5.7 Jobs by Industry, Burbank, 2000
Industry Sector Number Percent
Manufacturing 6441 9.55
Wholesale trade 2903 4.31
Retail trade 7128 10.57
Transportation and warehousing, and utilities N.A. N.A.
Information 26,034 38.61
FIRE 1,620 2.40
Professional, scientific, & technical services N.A. N.A.
Administrative, and waste management services 7,865 11.66
Educational services 187 0.27
Health and social services 6,075 9.00
Arts, entertainment, recreation 1,492 2.21
Accommodation and food services 5,377 7.97
Other services (except public administration) 2,308 3.42
Total 67,430 100.00
Source: Economic Census 2000
Note: These numbers are not consistent with the EDD employment data presented in Chapter 4.

83

Table 5.7 presents data on the number and shares of jobs in different industry sectors in
the city of Burbank. The information sector, which includes the motion picture industry,
was the largest employer in the city in 2000 followed by the retail trade and public
sectors.

The Local Economy, 1980-2000
During the 1980s, aerospace defense manufacturing and motion picture industry sectors
had significant presence in Burbank. Lockheed, an aircraft manufacturer (majorly
defense aircrafts), employed approximately 15,000 workers while other firms associated
with Lockheed employed another 3,000-5,000 workers. Several motion picture studios
had located their soundstages, studios, and other production facilities in Burbank. The
motion picture industry was rapidly expanding in Burbank in the early 1980s- twelve new
buildings totaling 2.2 million square feet were either constructed or approved for
construction by the city during the early to mid 1980s.

In 1990, Burbank was hit by a major economic catastrophe when Lockheed announced
closure of its production facility in the city, resulting in an estimated loss of
approximately 15,000 – 18,000 jobs. More than 300 acres of prime property in the city,
owned and occupied by Lockheed, became a source of potential blight overnight.
Although the city continued to earn the property tax from Lockheed, yet other tax
revenue losses were substantial– approx. 1 million annually or 0.5 % of its total revenue.

84

How did Burbank manage to recover from the rather strong economic shock? In the
words of Jack Lynch, the deputy Housing & Redevelopment Manager of the city “we
were just lucky.” As this case study reveals, “luck” implicitly implies favorable market
conditions accompanied by a robust growth in the motion picture industry. Favorable
market conditions include a diverse local economic base (see Table 5.7), close proximity
to two freeways (Golden State and Ventura freeways), proximity to a major airport
(Burbank-Glendale-Pasadena airport), and a healthy regional economy.

Indeed Burbank’s remarkable resilience was a result of the combined influence of the
market forces and local government efforts. Two separate local government efforts need
to be mentioned here: one, development of The Burbank Center and implementation of
The Media District Specific Plan. Interestingly, none of these two were motivated by
Lockheed’s closure. The two were entirely exclusive efforts, initiated well before
Lockheed’s closure. But, the timing of the two couldn’t have been better for the city.

In addition, Lockheed took upon itself to sell more than 300 acre of property it owned in
Burbank, which benefitted from motion picture industry’s growth in the city. Substantial
part of the Lockheed property was purchased and converted into animation studios,
soundstages, and other production facilities by private developers. A big private
developer, Zelman Development Company, built a retail shopping mall “Burbank Empire
Center” on a 103 acre lot. The Burbank-Glendale-Pasadena airport purchased
approximately 100 acres for its expansion. It took more than a decade before the entire

85

Lockheed property was sold and/or redeveloped. Although one would assume that the
city would have played a positive role in these transactions, it did not. In fact, the city
was opposed to both large transactions- the airport purchase and the Empire Center Mall-
for various political reasons, e.g. The Empire Center directly competed against the city
promoted Burbank Center shopping mall. The following section discusses the two key
projects that the city undertook.

The Burbank Center
The Burbank Center opened as a major regional shopping center in downtown Burbank
in 1991. The center is a traditional automobile-oriented shopping mall, with three large
department stores, smaller mall shops, a multi-screen cinema theatre, a free standing
furniture retail store, structured parking, and restaurants. The center has a total area of
approximately 42 acres. The motivation underlying the promotion of the mall was a)
revitalizing the downtown; and b) increasing the local sales tax revenue base.

The land for the project was assembled by the local Community Redevelopment Agency
(CRA), which conveyed the site to Alexander Haagen Company Inc., a private developer,
though a ground lease for a term of 55 years. The developer agreed to pay in return
$1,000 per year, plus 50% of all net cash flow revenues generated from the project. The
net cash flow includes all annual cash flow from the operation of the center after debt
service and 50% of all revenues from the sale or refinancing of any project component.

86

CRA leased land to the furniture store (IKEA) separately and independently. CRA
conveyed the ground lease to IKEA practically free in return for a sales tax guarantee.
IKEA agreed to a guaranteed minimum sales tax of $500,000 for the first 3 years,
$700,000 for the 4
th
through 10
th
year, and $1 million for the 10
th
through 15
th
year. The
ground lease agreement was open to negotiation or termination at the end of the 15
th
year.

The project was expected to raise estimated revenue of approx. $ 49 million (net present
value in 1989) over a period of 55 years. CRA incurred total site assembly and
development costs of approximately 55.5 million (in 1989) dollars. This includes
administrative costs, both on-site and off-site public improvements such as parking, but
excludes the residual cost of land at end of the ground lease term (estimated at 30 million
dollars).

Clearly, development of the Burbank Center was made possible by CRA’s active efforts
over a long period. CRA explored several projects and negotiated with several potential
developers before reaching a deal with the Alaxender Haagen Company Inc. in 1989. For
example, in May 1987, CRA entered into an agreement with the Walt Disney Co. for
developing an entertainment complex on the same site. The complex would have
comprised of movie back lot tours, theatres, shopping areas, a hotel, a planetarium, rides
and nightclubs. After toying with the proposal for nearly 11 months, the Walt Disney Co.
backed out citing the project as being economically unviable. The CRA and Walt Disney
Co. agreement, however, was also facing a legal challenge from MCA Inc. on grounds

87

that it had been unfairly excluded from making a competing proposal. Incidentally,
MCA Inc. was the parent company of Universal Studios and operated a successful movie
back lot tour not very far from Burbank. The CRA also evaluated proposals submitted by
Triple Five Development Enterprises, Price Co. and Kornwasser & Friedman Shopping
Center Properties, before finalizing the one made by Alaxender Haagen Company Inc.

The Media District Specific Plan
The media district specific plan was adopted by Burbank in January 1991. The plan was
motivated by the concerns about rapid expansion of the media industry in Burbank in the
early 1980s- twelve new building totaling 2.2 million square feet were either constructed
or approved for construction in the Media District during the early to mid 1980s. The
city planners presented the specific plan as a growth control strategy “designed to
dramatically reduce the amount of development” (page i). One of the key features of the
plan was the imposition of density limits on the media district. Prior to the adoption of
the specific plan, the media district had virtually no limit on the either the Floor Area
Ratio (FAR) or the building heights.

The media district specific plan did not provide explicit incentives or subsidies to the
motion picture industry, but provide an explicit assurance to the industry re the future
development pattern inside the media cluster, i.e. the area would remain exclusive to the
motion picture industry. It also quelled the residents’ concerns regarding traffic

88

congestion, noise, and other externalities associated with the growth of motion picture
industry in Burbank.

Figure 5.2 The Burbank Media District Overlay Zone

Figures 5.2 shows land uses within the media district, with the industrial use (read related
to motion picture industry) being predominant. The plan explicitly excludes non-film
and television related production facilities and businesses from locating within the media
district. The industrial land use category in the specific plan permits only media related
uses such as media office, workshop, sound stage, etc. Certain other uses such as
childcare, health facility could be permitted with a conditional use permit. The intent
here is to create a distinct character, and also to prevent any land uses that may conflict
with the motion picture industry from locating within the district.

89

Key features of the specific plan include:
1. Floor Area Ratio (FAR) restricted to 1.1
2. Stair-step height restrictions:
a. 25 feet maximum height within 50 feet of the closest R-1 lot line
b. 35 feet maximum within 150 feet of the closest R-1 lot line
c. 50 feet maximum from 150 feet to 300 feet of the closest R-1 lot line
d. 70 feet maximum from 300 feet to 500 feet of the closest R-1 lot line
3. All building with 35 feet or more height to receive a conditional use permit,
requiring public notice to the surrounding property owners and a hearing before
the planning board
4. 15 story maximum height limit; but a building may exceed the 15 stories if “it
provides extraordinary offset measures such as subterranean parking, large
amounts of open space, etc.” The absolute maximum height permissible under
any circumstance is 25 stories.
5. Several traffic calming, congestion mitigation, and transportation improvements
planned in the adjacent neighborhoods.

The specific plan lays emphasis on creating a pedestrian friendly commercial
environment, reduction of traffic through travel demand management, and incentives for
employers to reduce peak hour traffic. The plan also lays a great deal of emphasis on
creating an inviting image through a mix of urban design features (see Figure 5.3) and
encouraging retail along major arterial routes.

90

Figure 5.3 Suggested Parking Design in The Media District Specific Plan

(Source: City of Burbank (1990) Figure 3-10, page 70)

While the specific plan claims to control density, yet it offers great flexibility to the
studios and media production related facilities to build at high densities. The three major
studios – NBC, Warner Brothers, and Disney Productions – own large parcels designated
as Planned Developments (PD) in the overlay zone. The three studios were afforded a
great deal of flexibility within the PD, with several specific plan recommendations being
suggestive and not really binding:
• “Encourage Warner Bros. Studios to recycle the commercially zoned
land…” (page 38);
• “Require NBC to adopt procedures which will encourage people to use
the on-site parking provided for them” (page 40);
• “Require Disney to follow a master plan that provides adequate off-street
parking and maximum flexibility to employees for ingress and egress so
that employees do not resort to parking on residential streets……A
balanced distribution of parking is encouraged” (page 43).

91

Furthermore, a number of items, especially transportation demand management, in the
plan are merely suggestions, i.e. they are not legally enforceable, e.g. “employers may
actively foster and monitor carpool formation” (page 99) or “employers may choose to
promote programs to increase public transportation ridership” (page 99).

The specific plan was appreciated by both local residents and major studios. While the
residents were pleased with the density limits, traffic congestion mitigation measures,
etc., for the motion picture studios it removed a great deal of uncertainty regarding the
future of growth in the media district. Following the implementation of the specific plan
in 1991, between 1991 and 2000, the Burbank City Council approved three major
planned developments proposed by the three largest property holders in the media
district: NBC, Warner Bros., and Disney.

In November 1991, Walt Disney Co. submitted plans to the city of Burbank for a $600
million expansion of its studio during the next two decades that could double the space it
then occupied. The proposed development adds 1.9 million square feet of buildings, new
soundstages and a helipad. The plan was in compliance with the new specific plan and
comprised of heavily landscaped buffer zone along the border. The Burbank City Council
issued a preliminary, unanimous approval to the plan in October 1992. Disney’s
expansion plans received local welcome and support, especially as it came soon after
Lockheed’s closure.

92

In 1994, Warner Bros. submitted master plan for approval for its 104 acre property in the
media district. The ambitious $800 million expansion was approved by the Burbank City
Council in October 1995 despite few traffic and noise concerns raised by local residents.
The plan sought to add 2.2 million square feet of space inside the media district and
another 1.1 million at its nearby Studio Ranch facility. The planned expansion was
expected to bring 8,000 new jobs to the city over the next 20 years.

In March 1997, the Burbank City Council approved NBC's plan to more than double the
size of its studio over the next 25 years. The project was expected to add 4,300 new jobs
and $ 2.5 million in annual tax revenue to the city. The NBC plan added 1.5 million
square feet of studio and office space inside the media district. To address the
neighborhood residents’ traffic congestion concerns, NBC agreed to add traffic signals
and reconfigure traffic lanes around its complex.

Clearly, the specific plan made it possible for the studios to propose and the city council
to approve such large scale developments without much public opposition. The single
biggest concerns that the neighborhood residents had in all the three proposals was an
increase in traffic and the ensuing congestion, parking spillovers, and noise issues.
Traffic related issues were explicitly addressed in the specific plan, and the developers
voluntarily adopted the specific plan’s recommended mitigation measures.

93

Economic activity outside the employment centers
The city of Burbank has aggressively pursued sales tax revenues. For example, the city
has been able to attract Fry’s Electronics, a discount superstore near the Burbank airport
by using a mix of incentives and a minimum sales tax guarantee. Similarly, Costco
wholesale, another big sales tax generator, has located to the north of The Burbank
Center. Apart from the two major employment centers, economic activity in much of the
city is an indiscriminate mix of strip malls, auto repair facilities, and small mom and pop
businesses. The two employment centers are somewhat like islands on an otherwise quite
barren landscape of rundown buildings.

Discussion
The Media District Specific Plan is a classical example of a reactive growth management
strategy motivated by a local growth spurt—in this case the motion picture industry in the
early 1980s. The specific plan is also a good example of how the local government tried
to balance its own need for tax revenues and resident homeowners’ desire to preserve
their property values. While the city could not afford to lose the motion picture industry
revenues, it also could not afford to displease the resident homeowners.

In the end, it was really the combined effect of the market forces and local government
efforts that contributed to Burbank’s economic growth in the 1990s and continued growth
of the media cluster. The market forces converted the vacant Lockheed property into an
asset rather than liability for the city of Burbank. Suddenly, there was this enormous

94

room for the media industry to expand, which was available at a convenient location in
close proximity of the industry’s existing facilities.

Burbank’s local government was entrepreneurial in its implementation of a specific plan
that a) strategically managed to prevent active opposition to the cluster’s growth and b)
made the district particularly conducive to the media industry’s growth. Although the
specific plan may lack strong policy recommendations, yet it served its (unsaid) purpose
well enough. Similarly, city’s effort to develop The Burbank Center was quite successful
as it helped buoy local sales tax revenues.

Although none of these two local government efforts were motivated by Lockheed’s
closure, yet they helped the city tremendously in not sinking into a local economic
depression. Both the efforts contributed to employment center growth in the city. The
timing for both couldn’t have been better. Perhaps Mr. Jack Lynch was right, Burbank
indeed was “lucky.”

5.4 Case Study 2: The City of Pasadena
The city of Pasadena, an earlier twentieth century suburb of the city of Los Angeles, was
incorporated in 1886. The medium size city is located in Los Angeles County,
approximately eight miles northeast of downtown Los Angeles. Pasadena is famous for
its annual new year’s day Rose Parade, and the Rose Bowl football game. California
Institute of Technology (Caltech), which houses the internationally renowned Jet

95

Propulsion Laboratory (JPL), is located in Pasadena. The city is ethnically diverse, well-
educated, and has higher per capita and household incomes as compared to the Los
Angeles County.

Pasadena’s employment center overlaps partially with the boundaries of the city’s
downtown including Old Pasadena. Major commercial development of the city is
concentrated on Colorado Boulevard and Lake Avenue (see Fig. 5.4).

Figure 5.4 The City of Pasadena

96

Demographic Profile
Pasadena had census 2000 population of 133, 936. Between 1990 and 2000, Pasadena’s
total population increased by 1.8 percent. Nearly 49 percent of the city’s population was
male and 51 percent female (see Table 5.8). The city’s population has remained
predominantly white, although the share of white population dropped slightly between
1990 and 2000. The share of Asian population in the city increased during the same
period. Hispanics (of any race) were more than a third of the city’s total population in
2000. The city had a higher percentage of white population, a substantially higher share
of African American population, and a much lower percentage of Hispanics as compared
to L.A. County. The city’s population had a median age of 34.5 years in Census 2000,
slightly higher than the County’s median age of 33.9 years.

Table 5.8 General Population Characteristics, Pasadena and L.A. County
Pasadena L.A. County
1990 2000 2000
Total Population 131,591 133,936 9,519,338
Percent Male 48.9 51.1 49.4
Percent Female 51.4 51.5 50.6
Percent White 57.3 53.4 48.7
Percent African American 20.0 14.4 9.8
Percent Asian 8.1 10.0 11.9
Percent Other Races 14.6 22.2 29.6
Percent Hispanic (of any race) 27.3 33.4 44.6
Median Age (in years) -- 34.5 33.9
Source: U.S. Census Bureau

97

Table 5.9 Housing Characteristics, Pasadena and L.A. County, 2000
Pasadena L.A. County
Total Housing Units 54,312 3,270,909
Percent Occupied Housing Units 95.4 95.8
Percent Owner Occupied 45.8 47.9
Percent Renter Occupied 54.2 52.1
Percent Vacant Housing Units 4.6 4.2
Median Value of Single Family Home 286,400 209,300
Median Rent 1,294 704
Source: U.S. Census Bureau

Table 5.9 compares housing characteristics for the city of Pasadena with LA County.
Pasadena had almost similar occupancy rate of its total housing stock (not vacant) as that
of L.A. County. The city had a higher median value of a single family home and higher
median rent than LA County.

Table 5.10 presents data on key economic characteristics for the populations of Pasadena
and Los Angeles County. According to Census 2000, 63.8 percent of Pasadena’s
population below 16 years of age was in labor force as compared to 60.5 percent in LA
County. On average, Pasadena’s labor force had a slightly shorter commute than that of
LA County. The city residents were more affluent as compared to the whole County with
higher median household income and per capita income. According to Census 2000, 14.9
percent of Pasadena’s households were below poverty level. Pasadena’s population had
better educational attainment than L.A. County. More than 41 percent of the city’s
population aged 25 years or older achieved a bachelor’s degree or higher as compared to
24.9 percent in L.A. County.

98

Table 5.10 Economic Characteristics, Pasadena and L.A. County, 2000
Pasadena
L.A.
County
Population 16 years or older 105,882 7,122,525
Percent Population 16 years or older in the Labor Force 63.8 60.5
Percent Population 16 years or older in the Labor
Force and employed
59.5 55.5
Percent Population 16 years or older in the Labor
Force and unemployed
4.3 5.0
Percent Population 16 years or older not in the labor
force
36.2 39.5
Mean Travel Time to Work (in minutes) 25.9 29.4
Median Household Income 46,012 42,189
Per capita Income 28,186 20,683
Percent Population Below Poverty Level 15.9 17.9
Population 25 years and older 90,934 5,882,948
Percent High School Graduate or higher 79.5 69.9
Percent Bachelor’s degree or higher 41.3 24.9
Source: U.S. Census Bureau

Table 5.11: Employed Civilian Population 16 years and over by Industry,
Pasadena, 2000
Industry Sector Number Percent
Agriculture, forestry, fishing and hunting, and mining 125 0.2
Construction 2,565 4.1
Manufacturing 4,782 7.6
Wholesale trade 1,696 2.7
Retail trade 5,336 8.5
Transportation and warehousing, and utilities 2,152 3.4
Information 3,681 5.8
FIRE 5,440 8.6
Professional, scientific, management, administrative,
and waste management services 10,320 16.4
Educational, health and social services 15,179 24.1
Arts, entertainment, recreation, accommodation and
food services 5,374 8.5
Other services (except public administration) 4.117 6.5
Public administration 2,337 3.7

99

Table 5.11 presents shares of the employed labor force in the city of Pasadena by industry
of employment. As the city’s population has relatively high levels of educational
attainment, it is reasonable that a higher proportion of the city’s labor force is employed
in skilled professions. The highest percentage of Pasadena’s labor force was employed
in educational, health, and social services (24.1 %). Professional, scientific,
management, administrative, and waste management services employed 16.4 percent of
Pasadena’s labor force.

Local Revenue
Table 5.12 presents data on the city of Pasadena’s revenues. The table shows that the
city’s general revenues and functional revenues declined (in real terms) between 1984-85
and 2004-05. While the city’s property tax revenue increased by almost 50 percent (in
real terms) between 1984-85 and 2004-05, its sales tax revenue increased only marginally
during the same period. This is noteworthy because much has been written about
resurgence of Pasadena’s downtown and success of the local economic development
efforts in the city. Substantial increases in Old Pasadena’s sales tax revenues have been
reported (Kotin and Szalay, 2007). However, such reports often do not adjust for
inflation. Furthermore, while Old Pasadena’s sales revenues may have increased, the
city’s overall sales tax revenues have declined in real terms between 1985 and 2005.

100

Table 5.12: Pasadena Revenues
1984-85* 1990-91 1994-95* 2000-01* 2004-05*
Total
Revenue
227,777,365 322,707,768 275,574,743 464,423,914 352,161,887
General
Revenue
85,904,036 85,569,713 69,561,441 76,073,888 94,264,384
Functional
Revenue
141,873,329 237,138,055 206,013,302 388,350,026 257,897,503
Property
Tax
10,669,007 17,708,083 13,233,309 15,907,684 15,727,392
Sales Tax 16,831,671 19,813,405 19,006,003 22,072,683 17,254,077
*Adjusted for inflation to reflect real value in 1991 dollars

Table 5.13: Jobs by Industry, Pasadena, 2000
Industry Sector Number Percent
Manufacturing 1,750 2.58
Wholesale trade 2,340 3.45
Retail trade 9,520 14.03
Transportation and warehousing, and utilities N.A. N.A.
Information 6,383 9.41
FIRE 1,734 2.56
Professional, scientific, management, administrative,
and waste management services 10,622 15.65
Administrative & support and Waste management &
remediation services 7,366 10.86
Educational services 345 0.51
Health care and Social assistance 13,382 19.72
Arts, entertainment, and recreation 1,886 2.78
Accommodation and food services 9,139 13.47
Other services (except public administration) 3,385 4.99
Total 67,852 100.00
Source: Economic Census 2000
Note: These numbers are not consistent with the EDD data used in calculation of employment
center presented characteristics in Chapter 4.

101

Table 5.13 presents data on the number and shares of jobs in different industry sectors in
the city of Pasadena. Health care sector was the biggest employer in the city in 2000.
Retail sector, professional services, and the hospitality industry were the other large
employers in the city in addition to the public sector. Indeed share of local jobs in the
retail industry is higher in Pasadena as compared to the city of Burbank – 14% in
Pasadena versus 11 % in Burbank.

Local Economic Development: Initial Efforts
Between 1950 and 1980, like many inner ring cities in the period, downtown Pasadena
lost businesses to newer suburbs. The city suffered not only revenue losses, but also
experienced urban blight on account of vacant commercial properties, rise in poverty
levels and crime. During these three decades, private community groups and the local
government of Pasadena made several attempts to revitalize the local economy.

Perhaps one of the earliest local economic revitalization efforts in Pasadena was initiated
in 1956, when a group of local merchants and business owners formed the Pasadena
Central Improvement Association (Palmer 2003). As its first attempt, the Association
formed a parking district and constructed a large parking lot adjacent to Colorado
Boulevard after demolishing several buildings. This was not surprising as the biggest
concern was how to stop losing business to the more auto friendly outer suburbs. Five
years after this privately led effort, Community Redevelopment Agency was formed in
Pasadena in 1961. The Agency commissioned the first major study on downtown

102

revitalization in 1966, and creation of the Downtown Redevelopment Area was approved
by the redevelopment agency in 1970 (Koblik 1981).

In 1971, property owners in the redevelopment area together with the local Chamber of
Commerce formed a new organization called ‘Pasadena Now’ and agreed to fund a 20
year action plan being promoted by the redevelopment agency called “The Pasadena
Central District Improvement Program.” The plan called for immediate rebuilding of 369
acres in the core of the city. Property owners on the West Colorado Boulevard, however,
chose to not participate in the program. Pasadena Now and the redevelopment agency
dismissed these properties as being too run down and instead promoted in their location
an “Old Town” along the lines of San Francisco’s Ghirardelli’s Square. These efforts to
create a boutique like old town never really took off and the Old Town area remained
more or less untouched by the initial redevelopment efforts of the city.

Between 1970 and 1975, several projects were either completed or were well under way
including a Conference Center, the Pasadena Hilton Hotel, several corporate
headquarters, and the Plaza Pasadena mall. Indeed, the redevelopment agency’s efforts
during these five years were quite entrepreneurial in terms of their persistence and
creativity in negotiating deals with the developers and community groups, use of tax
increment financing tool to fund future development, etc. (Frieden and Sagalyn 1989).

103

According to Frieden and Sagalyn (1989) the successful completion of the Plaza
Pasadena Mall in 1980 was instrumental in Pasadena’s subsequent economic revival.
The entrepreneurial manner in which the city dealt with the contentious issues involved
in financing and building the mall sent a positive signal to other investors and developers.
The project also helped the city’s campaign to attract corporate headquarters in to the city
and several firms including Parsons Engineering, Pacific Bell, Kaiser Permanente
(re)located in downtown Pasadena.
6,7
However, unlike the older buildings in the area
that exuded nostalgic charm owing to their historic character these new developments
were contrastingly modern and glossy. As a result the redevelopment efforts were not
much appreciated by the local community (Koblik 1981).

Beginning in 1975, under pressure from various local community groups, the
redevelopment agency started rethinking the redevelopment principles it had adopted in
1971. Instead of demolition and replacement of the existing structures, the new thinking
focused on their rehabilitation. The redevelopment agency realized that preservation and
rehabilitation of old historic buildings had the potential to be economically viable and
even preferable, particularly with federal assistance available for such conversions
(Marston 1975).

6
For a more detailed discussion on this see Chapter 5 Pasadena: No Bed of Roses in Frieden and Sagalyn
(1989) Downtown Inc Cambridge, MA: MIT Press, pp87-106

7
Ultimately, Plaza Pasadena failed and was demolished in 1999. A mixed use (residential-retail-
commercial) development “Paseo Colorado” was built on the same location.

104

The vision in Pasadena now involved not only historic preservation, but also formation
and promotion of a distinct civic identity (Palmer 2003). Commissioned to convert this
vision into a workable plan by the city and local community groups, the Arroyo Group of
Architects produced a planning document in 1978 called “A Plan for Old Pasadena.” The
plan was funded by the City, Pasadena Heritage, the Pasadena Central Improvement
Association, and the National Endowment for the Arts. In this new plan the Old Town,
which was shunned a few years back as too run down, became the cornerstone of
community economic revitalization. The plan not only identified properties for
preservation, but also the ones where appropriate modifications and additions (invariably
of parking) were possible and feasible.

In the decade following the creation of A Plan for Old Town, between 1980 and 1990, the
city efforts were focused on both historic preservation and economic revitalization of
downtown Pasadena. Several creative adaptive reuse projects were promoted, but two
projects were largely the turning points in Old Town’s revival. The first key milestone in
Old Town’s revival was the opening of United Theatre in the end of 1986. The theatre’s
popular success brought to the area much needed patronage, not only to the theatre, but
also to local dining and shopping (discussions with Marsha Rood and Robert Fazio,
retired local planners and activists). The second milestone was reached in 1993 with the
successful completion of One Colorado, one of the earliest successful redevelopment
projects in the Old Town. During the latter half of the 1980s, Pasadena invested in
several public parking facilities, which have been one of the key factors in attracting

105

shoppers and diners to the area. These factors responsible for revival of Old Pasadena
have been widely documented (e.g. Kolozsvari and Shoup 2003, Fader 2001, Kotin and
Szalay 2007).

One Colorado, a 275,000 square feet shopping and entertainment venue, was inaugurated
in 1993. One Colorado is an approximately 300,000 square feet block bounded by
Colorado, Fair Oaks, Union, and De Lacey streets (see Figure 5.5).

Figure 5.5 Old Pasadena

(Source: Official website of the Old Pasadena Management District http://www.oldpasadena.org)

106

The One Colorado project was earlier conceived by another developer as specialty retail.
The developer had plans to demolish all the existing structures (except the street façade)
and develop a single mall type building. The developer’s plans met considerable
resistance from the Pasadena Heritage and the city residents. Ultimately the developer
abandoned the project in 1988 citing reasons of inadequate financing (Kotin and Szalay
2007)

A new developer, the Stitzel Company, redesigned the project, retaining much of the
historic shell of the buildings and creating an outdoor circulation system using five alleys
and a plaza to connect the buildings and uses. With numerous conditions attached by the
city, the project was approved in 1990. The project was successfully completed towards
the end of 1992. Some of its tenants included national retail chains such as Banana
Republic, Victoria’s Secret, Gap, and a multi screen Cineplex. Commercial success of
One Colorado sent a positive signal to other investors and business owners and was
catalytic in Old Town’s revival during the 1990s.

Perhaps the major public effort instrumental in Old Town’s revival has been the city’s
creative parking policy. Two specific strategies are discussed below: development of
public parking garages and curbside metered parking policy.

107

Public Parking Garages in Old Pasadena
Between 1985 and 1987, the city of Pasadena developed three parking garages in the Old
Town with a total of approximately 2,300 parking spaces (see Fig. 5.5 – parking garages
are marked by the letter P). The first garage was financed with industrial development
revenue bonds, and the second and third with certificates of participation. Debt service
was paid from the net operating income of the parking operations; retail and restaurant
rents; tax increment funds; fees from zoning-credit contracts; and in the initial years of
the bonds, general funds revenues (Fader 2001).

A key hurdle in conversion of historic buildings for retail or commercial purposes is lack
of parking. Much of the building stock in the Old Town consists of older buildings built
on narrow lots without any provision for on-site parking space. In such circumstances,
provision of public parking garages at strategic locations within the Old Town has proven
to one of the keys to successful local economic revitalization.

The garages were provided as an incentive to business owners, particularly small
retailers. Space was reserved on the ground floor for restaurants and neighborhood
services such as convenience stores. The strategic locations of these garages fostered
substantial pedestrian traffic in the Old Town, which has been vastly instrumental in
promoting a lively and vibrant street atmosphere. The pedestrian vitality in turn
benefitted the local retailers and restaurants immensely.

108

Curbside Metered parking
Kolozsvari and Shoup (2003) attribute success of Old Town partially to its curbside
metered parking policy. Until 1993, Old Town did not have any parking meters on street.
The city’s staff proposed installing parking meters at the curb, but faced stiff opposition
to the idea from local business owners. Businesses feared that the lack of free parking
would further drive away the already scarce customers to shopping malls like Plaza
Pasadena that provided ample free parking. Proponents of parking meters argued that
much of parking space is occupied by local employees rather than customers and
restricted term meters would discourage such employees from hogging parking space.
Parking space thus freed would then be used by customers who want to shop in Old
Town and local businesses will thus benefit.

The debate dragged on for two years before the city arrived at a compromise with the
merchants and property owners. The city agreed to spend all the revenue generated by
parking meters on public improvement projects within Old Town. Furthermore, only the
blocks with parking meter revenues would receive the added services financed by the
parking meter revenue. In return the merchants and property owners not only agreed to a
high parking meter rate ($1 an hour), but also to operating the meters on evenings and
Sundays.

The parking meter policy benefitted both the city and the business owners as it paid for
the much needed public investments in Old Town. In 2001, the Old Town’s 690 parking

109

Memorial Park Gold Line Station, Pasadena

Across from the Memorial Park Station

Transit Oriented Development in Pasadena has
fostered residential developments in close proximity
of the train stations, but has not been able to
generate commercial activity.
meters yielded $1.2 million in net parking revenue (excluding collection costs) to fund
additional services. Additional services include sidewalk and street maintenance and
improvement, additional foot patrol officers in the area, marketing of the Old Town, etc.

Outside Old Pasadena
While much has been written about Old Pasadena, area outside Old Pasadena has not
received much attention of researchers.
One reason is that outside the Old
Pasadena, Pasadena’s landscape does
not differ much from other inner ring
cities with similar affluence. For
example, along its major arterials such
as Arroyo parkway and Lake Avenue,
there are numerous strip malls and
retail stores with ample on-site parking.

The city on the whole has not done
very well in terms of retail sales as
reflected in the sales tax numbers and
evident from visual survey of the
condition of these storefronts.
Furthermore, the chic pedestrian

110

friendly environment of Old Town remains limited to Old Town and is not reflected
elsewhere in the city. Indeed the newer developments – such as the up market
Wholefoods grocery store on Arroyo Parkway – are not different from their any other
location elsewhere. Hence the markets don’t seem to be cuing up to Old Pasadena’s
theme.

Figure 5.6 Metro Gold Line Route

Pasadena has also tried to implement “transit-oriented development” at several locations
within the city around Metro Gold Line train stations (see Figure 5.6). While several
residential developments have been completed in close proximity of these stations,
commercial development around stations is still insignificant. The city has enticed
developers into building housing at or near stations by offering incentives such as
reduced parking requirements, density bonuses, etc. However, the overall development

111

pattern around these stations does not differ significantly from any other comparable
suburban development.
8

Discussion
Pasadena presents several ways of how an active city may influence employment center
growth. Old Pasadena is a result of a mix of publicly financed projects and private
investment. However, perhaps the most important factor in the city’s revival was its use
of somewhat unconventional approach (at the time) of recognizing local cache of historic
architecture and adapting it to the popular need, i.e. ability to drive. However, Pasadena
has not been able to replicate its Old Town strategy to other parts of the city, which are
not very different from any other affluent inner ring suburb.

5.5 Lessons from the Case Studies
Local governments have some potential to influence employment growth in general, and
employment centers in particular. However, market forces are more effective
determinants of employment center growth than local governments. At times, market and
government work together to form public-private partnerships and joint ventures, e.g.
development of Plaza Pasadena. Although, intuitively such combined forces should be
more potent than either the markets or the government by itself, yet these case studies
indicate otherwise.

8
For a detailed discussion on this see Gold Line Corridor Study prepared by Ralph and Goldy Lewis
center for Regional Policy Studies, University of California Los Angeles, March 2007.
Available online: http://www.scag.ca.gov/publications/pdf/2007/Gold_Line_FINALReport_040907.pdf

112

Cities focus on discrete projects, often with the intention of generating additional sales
tax revenues. These projects tend to be large scale developments and evolve over time,
often requiring some market support, i.e. a private developer willing to take risk,
howsoever minimal, of investing and building the project. More than often cities attempt
to attract private developers by underwriting a large portion, if not all, of the financial
risk using public money.

Targeting a particular project or a particular area for development implies lower public
investment in other parts of the city. Even if the project is funded using tax increment
financing, or other projected future revenue streams from the project itself to service
debt, targeting particular projects may result at the expense of other areas in the city. For
example, when the city of Pasadena focused on developing Plaza Pasadena during the
early 1980s, Old Pasadena did not receive city’s attention. Once old Pasadena became
focus of city’s development during the 1990s, other parts of downtown were deprived of
as much municipal investment.

Often employment centers emerge and grow over large periods of time. Market
conditions may change over time, and hence it is important for the promoters (private and
public) to be able to adapt to the changing market conditions. For example, the city of
Burbank adapted its “growth control” strategies to promote motion picture industry’s
growth and Pasadena rewrote the way core city revitalization is done. This also implies
that the periods in which municipal efforts are evaluated are important. For example,

113

Pasadena’s efforts during the 1990s to redevelop Old Pasadena look much better than if
the revenue streams in 200-01 are considered whereas the effort does not look as good if
2004-05 revenue streams are considered.

Finally, employment centers are but artifacts of numerous market conditions, only a part
which may be affected by the present municipal policies—others may be a result of local,
state, and national policies that were or have been in place for long periods of time in the
past.

114

Chapter 6: Summary of Findings and Conclusions

This chapter presents a summary of findings from the quantitative and qualitative
analysis and discusses policy implications of this research. The chapter concludes with a
brief discussion of possible next steps in future research on the topic.

6.1 Findings from the Quantitative Analysis
1. There is no statistical evidence that the local development policies (as measured
in this study) are associated with employment growth in general, and employment
center growth in particular.
2. Economic factors such as agglomeration, labor force accessibility, proximity to a
major airport, and industry mix are associated with employment center growth.
3. There is evidence that local development policies tend to be reactive, i.e. cities
experiencing economic downturn have a high incidence of strong growth
promotion policies whereas cities facing growth pressures have a high incidence
of growth control policies.
4. Several cities have implemented mutually counteractive growth promotion and
growth control policies.

115

6.2 Findings from the Qualitative Analysis
1. There may be differences between enactment of growth control policies and their
actual enforcement. Local governments may choose to selectively enforce growth
controls by way of granting conditional use permits that override growth control
ordinances (e.g. Burbank’s selective enforcement of the Media District Specific
Plan—an explicit growth control strategy).
2. Cities want to pursue growth of sales tax revenues, at least older cities, because
property taxes are constrained by Proposition 13. Much of a local government
economic development effort, both conventional and innovative, is motivated by
its pursuit of sales tax revenue and not necessarily employment growth.
3. Often a local government’s actions are conditioned by external economic forces
and unexpected market conditions over which municipalities have no control.
4. There are substantial time lags between enactment of a policy, and its outcomes.
Often it takes years before a policy is finalized—involving multiple stakeholders,
political interest, etc.—and involves negotiating and deal making at several levels.
Economic environment could have changed since a policy was enacted, which
may further influence the ways in which that policy is enforced.
5. Cities could be entrepreneurial in more than their negotiating and deal making—
they are also entrepreneurial in how they manage often competing interests of
residents, business owners, and markets.

116

6.3 Discussion
In theory, local governments could potentially influence employment center growth (in
either direction). The quantitative analysis presented in Chapter 4, however, does not
find any statistical evidence that supports this theory. Local development policies, as
measured in this study, do not bear statistically significant association with employment
growth in general and employment center growth in particular.

Growth Governance
Growth governance includes local growth promotion and growth control policies. This
research finds that growth governance policies are not associated with employment center
growth. The research also finds that growth governance policies tend to be largely
reactive. Cities that are experiencing economic decline tend to implement strong growth
promotion policies while cities that are experiencing growth pressures tend to implement
strong growth controls. Growth control efforts may often be led by local residents who
are skeptical about the influence of increasing economic activity on their quality of life,
e.g. fear of traffic congestion, air and noise pollution, and above all negative impact on
their property values – all constituents of Not In My Backyard (NIMBY) syndrome.

One of the reasons that growth controls are not significantly associated with employment
center growth is (lack of) their enforcement. The case studies confirm that the growth
control policies are selectively enforced. A local government may have enacted explicit
growth control policies to appease local residents and voters, but subsequent enforcement

117

of such policies remains subjective. In instances where the local government wants to
attract private investment, it may loosen growth controls by granting conditional use
permits, variances, etc. Permitting of such projects is a collaborative effort in the sense
that the private investor agrees to pay for environmental impact mitigation efforts to
pacify local residents’ potential opposition. The gamut of such efforts is large and
depends largely on local circumstances.

The key goal of local growth promotion policies is increasing the local sales tax revenue
base and not increasing the number of jobs. Property taxes are restricted by Proposition
13 in California and hence sales tax is one of the key components of general tax revenue
that a city could raise. This leads to local governments’ pursuit of retail stores and
shopping malls. But, by its very nature the retail industry has limited agglomeration
benefits and tends to disperse with population. It is therefore not surprising that local
growth promotion policies are not associated with centers’ employment growth.

Perhaps most intriguing finding regarding growth governance is concurrent enactment of
mutually counteractive growth control and growth promotion policies by several cities in
the Los Angeles region. There are several possible explanations for this:
1) There are spatial variations in terms of target areas for growth promotion and
growth controls. Cities are often large spatial units and it is quite possible that
there are certain locations where the local government may want to discourage

118

growth and vice-a-versa. The surveys on which this study is based did not
account for spatial variations.
2) As stated previously, growth control policies are selectively enforced. Hence it is
entirely possible that the two sets of policies exist on paper, but only one is really
effectively enforced.
3) A third explanation, suggested by Glickfeld and Levine (1992), is that the two
sets of policies were enacted at different times. While strong growth promotion
policies may have been enacted at a time when the city was losing tax revenues,
growth control policies may have been enacted when the opposite was happening.
It is quite possible that the latter was enacted without repealing the former.
4) Another explanation, also suggested by Glickfeld and Levine (1992), is that
growth controls and growth promotion are under the purview of different
departments within a city’s bureaucracy and the two pursue their agenda without
co-ordination. This is possibly true for cities with large bureaucracy (such as Los
Angeles) but likely not for the smaller ones.

All the above explanations are possible, but the first two are apparently the most
plausible. The case studies also allude to the same.

Expenditure on Developmental Activities
The quantitative analysis does not find any association between expenditure on
developmental activities and employment center growth. As explained in Chapter 4, this

119

variable does not include investment in public works made by private developers (as
exactions, environmental impact mitigation measures, or otherwise). In the post-Prop 13
California, the total value of these private investments is substantial. While no exact
numbers on this are available, anecdotal estimates place such investments to be anywhere
from a few million dollars to more than several hundred annually within Southern
California region alone! This limitation aside, it is not surprising that the public
expenditure on developmental activities is not associated with employment center
growth.

Case studies show that the bulk of public expenditure on development is targeted at
downtown revitalization. The city of Burbank invested millions of dollars in assembling
and providing infrastructure for developing a shopping mall in its downtown. The
investment was accompanied by other incentives in return for minimum sales tax
guarantee. The project was successful in parts, i.e. some retailers were able to meet their
sales targets and the others were not. Success here implies sales tax dollars and not the
number of jobs.

This may be true elsewhere too. In Pasadena, for example, the local government again
invested substantial amounts and offered significant incentives to develop a downtown
shopping mall that failed (Plaza Pasadena). The city also invested substantial amounts in
Old Pasadena’s revitalization. While Old Pasadena could be considered a success in
terms of creating a vibrant shopping and entertainment environment, yet Old Pasadena is

120

a small part of both downtown Pasadena and the city’s employment center. Furthermore,
retail industry and restaurants (which form the bulk of jobs in Old Pasadena) often
employ a large number of part time employees and do not create as many regular full-
time jobs as say the FIRE sector. Hence, the public investment does not necessarily
translate to employment center growth per se.

Local Taxes/License Fees on Businesses
This study does not find any evidence that there is an association between employment
center growth and local taxes. The study also reveals that there are large variations in
local taxes/license fees on businesses across the cities. While some cities levy a small
flat rate license fee, others have a more complex taxation formula based on the size of a
business based on gross receipts and/or the number of employees. In addition to these
direct fees, which are used in this study, cities may also impose a number of indirect fees,
e.g. a utility user tax. In terms of total cost of doing business at a location, however,
these taxes are often very small as compared to the cost of renting/owning real estate.
But, it would have been inappropriate to include real estate prices as a determinant of
employment center growth. Real estate prices are driven by demand to some extent,
which is endogenous to employment center growth. Finally, a number of taxes are
written off as incentives and subsidies by the local governments, especially when the
local government wants to attract a large private employer.

121

Economic Forces
In contrast to the local government efforts, this study finds several control variables,
economic factors theoretically associated with employment center growth, to be
statistically significant. These factors include agglomeration economies, congestion
diseconomies, access to the nearest airport, and labor force accessibility. This study does
not find any association between a center’s employment growth and its relative location
in the region alluding to the weakening of the economic ties between the region’s
historical center (downtown Los Angeles) and other employment centers. Contrary to the
Giuliano and Small (1999) study, this study does not find evidence that distance to LAX
international airport influences employment center growth. This could be explained on
the basis of substantial employment losses during the late 1990s in the aerospace industry
located in close proximity of LAX.

6.4 Future Research
This study raises several interesting questions that could be systematically explored in the
future. One, why do local governments implement counteractive growth governance
policies? One of the key theses of this dissertation is that employment centers are an
outcome of the overall decentralization phenomenon—firms value access to skilled labor
force, and hence jobs follow people. Local government efforts, as measured in this study,
do not significantly influence employment center growth. Based on these findings, local
governments could perhaps be more effective in promoting center growth by facilitating
local labor force accessibility. Is that something a local government could do? If yes,

122

what could be some of the policy instruments that a local government could use for this
purpose?

This study has not addressed the issue of historical path dependence in employment
center growth as raised by several researchers. The case studies provide some evidence
re longevity of the built stock and path dependence in employment center growth. Hence
this is a valid hypothesis and could be systematically examined.

This study uses data from the Los Angeles region, which may be unique in several
respects. An obvious extension of the study will be to examine whether the results hold
up using data from other metropolitan regions in the U.S.

123

Bibliography

Altshuler, A. and D. Luberoff (2003) Mega-Projects: The Changing Politics of Urban
Public Investment.Washington, DC: Brookings Institution Press
Anas A., A. Richard, and K. Small (1998) “Urban spatial structure,” Journal of Economic
Literature, 36 (3), 1426-1464.
Anderson N and W Bogart (2001) “The Structure of Sprawl: Identifying
andCharacterizing Employment Centers in Polycentric Metropolitan Area,” American
Journal of Economics and Sociology, 60(1), 147-169.
Ashiru O., J. Polak and R. Noland (2003) “Development and Application of an Activity
Based Space-Time Accessibility Measure for Individual Activity Schedules,”
Presented at the Annual Meeting of the European Regional Science Association,
Jyvaskyla, Finland
Barabasi, A.L., 2002, Linked: The New Science of Networks. Perseus, Cambridge, MA
Batty M (2001) “Polynucleated urban landscapes” Urban Studies Vol 38 (4) pp 635-655
Bogart, W. and W. C. Ferry (1999) “Employment Centres in Greater Cleveland:
Evidence of Evolution in a Formerly Monocentric City,” Urban Studies, 36(X), 2099-
2110.
Brasington D M (2001) “A model of urban growth with endogenous suburban production
centers,” The Annals of Regional Science, Vol. 35 (3), pp. 411 - 430.
Brueckner J K (1990) “Growth controls and land values in an open city” Land Economics
Vol. 66, pp 237-248
Burns, L. D. (1979) Transportation, Temporal, and Spatial Components of Accessibility,
Lexington, MA: Lexington Books.
Cairncross, F. (1997) The Death of Distance: How the Communications Revolution Will
Change Our Lives. Cambridge, MA: Harvard Business School Press.
Cervero R (1989) America’s Suburban Center: the Land Use Transportation Link,
Boston, MA: Unwin Hyman

124

Cervero R. (1989) “Suburban Employment Centers: Probing the Influence of Site
Features on the Journey-to-Work” Journal of Planning Education and Research,
Vol.8 (2), pp. 75-85
Cervero, R. and K-L Wu (1997) “Polycentrism, commuting, and residential location in
the San Francisco Bay Area,” Environment and Planning A, 29(5), 865-886
Chen, H. (1996) “The Simulation of a Proposed Non-linear Dynamic Urban Growth
Model”, Anals of Regional Science, 30(3), 305-319.
Chinitz, B. (1991) “A framework for speculating about future urban growth patterns in
the US,” Urban Studies, 28(6), 939-959.
City of Burbank (1991) The Burbank Media District Specific Plan City of Burbank
City of Burbank (1997) Burbank Center Plan City of Burbank
City of Psadena (2006) Pasadena Pedestrian Plan City of Pasadena Department of
Transportation
Coffey, W. and R. Shearmur (2002) “Agglomeration and dispersion of high-order service
employment in the Montreal Metropolitan Region, 1981-96, Urban Studies, 39(3),
359-378.
Deakin E (1989) in Brower D, D Godschalk, and D Porter eds. (1989) “Understanding
Growth Management: Critical Issues and A Research Agenda” Cambridge, MA:
Lincoln Institute of Land Policy
Downs A. ed. (2004) “Gowth Management and Affordable Housing: Do They Conflict?”
Washington, D.C.: Brookings Institution Press
Fader S. (2001) “Old Pasadena” in Bayard M. editor (2001) Developing Retail
Entertainment Destinations 2
nd
edition Urban Land Institute
Feiock R (2004) "Politics, Governance, and the Complexity of Local Land Use
Regulation," Urban Studies, 41 (2), 363-77.
Fischel W (1989) “Do Growth Controls Matter? A review of empirical evidence on the
effectiveness and efficiency of local government land use regulation” Cambridge,
MA: Lincoln Institute of Land Policy

125

Fischel W (1991) “What do economists know about growth controls: A research review”
in Understanding Growth Management: Criitical Issues ansd A Research Agenda (D
Bower, D Godschalk, and D Porter eds.) Washington DC: Urban Land Institute

Forestall,R. and R. Greene (1997) “Defining job concentrations: the Los Angeles case,”
Urban Geography, 18(8), 705-39.
Fotheringham, A.S., 1983b, Some theoretical aspects of destination choice and their
relevance to production-constrained gravity models, Environment and Planning A,
15, pp. 1121-1132.
Frieden, B. and L. Sagalyn (1992) Downtown, Inc: How America Rebuilds Cities.
Cambridge, MA: MIT Press.
Fujita M. and J-F Thisse (2002) Economies of agglomeration: Cities, Industrial Location,
and Regional Growth, Cambridge, U.K.: Cambridge University Press
Fujita, M. (1989) Urban Economic Theory: Land Use and City Size, Cambridge, UK:
Cambridge University Press.
Garreau, J. (1991) Edge City: Life on the New Frontier. New York: Doubleday.
Geertman, S.C.M., Eck, J.R.R., 1995, GIS and models of accessibility potential: an
application in planning, International Journal of Geographical Information systems,
9(1), pp. 67-80.
Giuliano G., A. Agarwal, and C. Redfearn (2008) Metropolitan Spatial Trends in
Employment and Housing: Literature Review. Report prepared for the TRB
Committee on Land Use, VMT and Energy.
Giuliano G and K Small (1999) “The determinants of growth of employment subcenters,”
Journal of Transport Geography, 7(3), 189-201.
Giuliano, G and C. Redfearn, with A. Agarwal, C. Li, and D. Zhuang (2007)
“Employment concentrations in Los Angeles, 1980-2000.” Environment and
Planning A, vol.39, pp 2935-2957
Giuliano, G. (1998) "Information technology, work patterns and intrametropolitan
location: A case study," Urban Studies, 35(7), 1077-1095.
Giuliano, G. (1999) “Land use and transportation: Why we won’t get there from here,”
Transportation Research Circular, 492, 179-198.

126

Giuliano, G. and K. Small (1991) “Subcenters in the Los Angeles region,” Regional
Science and Urban Economics, 21(2), 163-182.
Giuliano, G. and K. Small (1993) "Is the journey to work explained by urban structure?"
Urban Studies, 30(9), 1485-1500.
Glickfeld, M. and Levine, N. (1992) Regional Growth…Local Reaction: The Enactment
and Effects of Local Growth Control and Management Measures in California .
Cambridge, MA: Lincoln Institute of Land Policy.
Glickfeld M., W. Fulton, G. McMurran, N. Levine (1999) Growth Governance in
Southern California. Working Paper, Claremont Graduate University Research
Institute.
Gordon, P. and H. Richardson (1996) “Beyond polycentricity: the dispersed metropolis,
Los Angeles, 1970-1980,” Journal of the American Planning Association, 62(3), 289
- 295.
Gottlieb, P. (1994) “Amenities as economic development tools: Is there enough
evidence?” Economic Development Quarterly, 8, 270-285.
Gottlieb, P. (1995) “Residential amenities, firm location and economic development,”
Urban Studies, 32(9), 1413-1436.
Guldmann, J.-M., 1999, Competing destinations and intervening opportunities interaction
models of inter-city telecommunication flows, Regional Science, 78, pp. 179-194.
Helseley, R. and A. Sullivan (1991) “Urban subcenter formation,” Regional Science and
Urban Economics, 21, 255-275.
Henderson, J.V. (2003), “The Urbanization Process and Economic Growth: The So-What
Question,” Journal of Economic Growth 8: 47-71.
Henderson,V. and A. Mitra (1996) “The new urban landscape: Developers and edge
cities, Regional Science and Urban Economics, 26, 613-643.
Horner M. (2007) “A multi-scale analysis of urban form and commuting change in a
small metropolitan area (1990–2000)” Annals of Regional Science, Vol. 41, pp 315-
332
K. C. Koblik, (1981) “City Recycling: Old Pasadena-A Case Study” MA thesis,
California State Polytechnic University, Pomona

127

Kloosterman, R. and S. Musterd (2001) “Polycentric urban region: as a research
concept,” Urban Studies, 38(4), 619-629.
Knapp G and A Nelson (1992) “The Regulated Landscape: Lessons on State Land Use
Planning from Oregon” Cambridge, MA: Lincoln Institute of Land Policy
Kolozsvari D and D Shoup (2003) “Turning Small Change into big Changes” Access vol.
23
Kotin A and K Szalay (2007) “Old Pasadena, California: A Downtown Redevelopment”
in R. Peiser and A. Schmitz editors (2007) Regenerating Older Suburbs Washington,
D.C.: Urban land Institute
Kwan, M.P., and Hong, X.D. 1998, Network-based constraints-oriented choice set
formation using GIS, Geographical Systems, 5, pp. 139-162.
Kwan, M.-P., Murray, A.T., O’Kelly, M.E., and Tiefelsdorf, M., 2003, Recent advances
in accessibility research: Representation, methodology and applications, Journal of
Geographical Systems, 5, pp. 129-138.
Lee B. (2007) “Edge or Edgeless Cities? Urban Spatial Structure in U.S. Metropolitan
Areas, 1980 to 2000” Journal of Regional Science Vol.47 (3), pp 479-515
Lee S., J. Seo, and C. Webster (2006) “The Decentralizing Metropolis: Economic
Diversity and Commuting in the US Suburbs” Urban Studies Vol. 43 (13), pp 2525-
2549
Lenntorp, B. (1978) “A time-geographical simulation model of activity programmes,” in
T. Carlstein, D. Parkes, N. Thrift (eds.) Timing Space and Spacing Time Volume2:
Human Activity and Time Geography, 162-180.
Loukaitou-Sideris A (2007) Gold Line Corridor Study Report prepared by Ralph and
Goldy Lewis center for Regional Policy Studies, University of California Los
Angeles. Available online:
http://www.scag.ca.gov/publications/pdf/2007/Gold_Line_FINALReport_040907.pdf
Malpezzi S (1996) ‘Housing prices, externalities, and regulation in U.S. metropolitan
areas” Journal of Housing Research Vol 7 (2), 209-241.
McDonald, J. and P. Prather (1994) “Suburban employment centres: The case of
Chicago,” Urban Studies, 31(2), 201-218.
McMillen, D. (2001) “Nonparametric employment subcenter identification,” Journal of
Urban Economics, 50(3), 448-473.

128

McMillen, D. (2003) “Identifying sub-centers using contiguity matrices,” Urban Studies,
40(1), 57-69.
McMillen, D. and J. MacDonald (1997) “A nonparametric analysis of employment
density in a polycentric city,” Journal of Regional Science, 37(4), 591-612.
McMillen, D. and J. MacDonald (1998) “Suburban subcenters and employment density
in metropolitan Chicago,” Journal of Urban Economics, 43(2), 157-180 .
McMillen, D. and J. MacDonald (2000) “Employment subcenters and subsequent real
estate development in suburban Chicago, Journal of Urban Economics, 48(1), 135-
157.
McMillen, D. and S. Smith (2003) “The number of subcenters in large urban areas,”
Journal Of Urban Economics, 53(3), 321-338.
Mills E. (1967) “An aggregative model of resource allocation in a metropolitan area” The
American Economic Review 57(2) 197-210
Muller, P. (1981) Contemporary Suburban America. Englewood Cliffs, NJ: Prentice-
Hall.
Murray, A.T., Wu, X., 2003, Accessibility tradeoffs in public transit planning, Journal of
Geographical Systems, 5, pp. 93-107.
Palmer C (2003) “Historic Preservation and Urban Cultural Geography in southern
California” XXX, pp 59-71
Peterson P (1981) City Limits Chicago, IL: The University of Chicago Press
Quigley J M and L A Rosenthal (2005) “The Effects of Land Use Regulation on the Price
of Housing: What do we know? What can we learn?” Cityscape: A Journal of Policy
Development and Research Vol. 8(1), pp 69-138
Recker W., C. Chen, and M. McNally (2001) “Measuring the impact of efficient
household travel decisions on potential travel time savings and accessibility gains”
Transportation Research Part A: Policy and Practice, 35(4), pages 339-369
Redfearn, C. (2007) “The topography of metropolitan employment: Identifying centers of
employment in a polycentric urban area,” Journal of Urban Economics, 61(3), pp
519-541.
Rosenthal S and Strange W C (2001) “The Determinants of Agglomeration” Journal of
Urban Economics Vol. 50, pp 119-229

129

Rosenthal, S. and W. C. Strange (2003), "Geography, Industrial Organization, and
Agglomeration," Review of Economics and Statistics, 85 (2): 377-393.
Sagalyn L (1997) “Negotiating for Public Benefits: The Bargaining Calculus of Public-
Private Development” Urban Studies, Vol 34 (12), 1955-70.
Scott, A. (1988) Metropolis: From Division of Labor to Urban Form. Berkeley, CA:
University of California Press.
Shen, Q., 1998, Spatial technologies, accessibility, and the social construction of urban
space, Computers, Environment and Urban Systems, 22(5), pp. 447-464.
Small, K. and S. Song (1994) Population and employment densities: structure and
change, Journal of Urban Economics, vol 36, 292-313.
Stoll M (2005) “Job Sprawl and the Spatial Mismatch between Blacks and Jobs”
Washington DC: The Brookings Institution Survey Series
Stouffer, S.A., 1940, Intervening opportunities and competing migrants, Journal of
Regional Science, 2, pp. 1-26.
Sullivan, A. (1986) “A general equilibrium model with agglomerative economies and
decentralized employment, Journal of Urban Economics, 20, 55-74.
Thomas, I., Hermia, J.-P., Vanelslander, T., and Verhetsel, A., 2003, Accessibility to
freight transport networks in Belgium: A geographical approach, Tijdschrift voor
Economische en Sociale Geografie , 94(4), pp. 424–438.
Tiebout, C. (1956) “A pure theory of local expenditures,” Journal of Political Economy,
64, 415-424.
TransCAD (2004) Travel Demand Modeling with TransCAD 4.7, Caliper Corporation.
Wachs, M., and Kumagai, T.G., 1973, Physical accessibility as a social indicator, Socio-
Economic Planning Science, 7, pp. 437-456.
Wang, F., and Minor, W.W., 2002, Where the jobs are: Employment access and crime
patters in Cleveland, Annals of the Association of American Geographers, 92(3), pp.
465-486.
Weber, J., and Kwan, M.P., 2002, Bringing time back in: A study on the influence of
travel time variations and facility opening hours on individual accessibility, The
Professional Geographer, 54, pp.226-240.

130

Wee, B., Hagoort, M. and Annema, J.A., 2001, Accessibility measures with competition,
Journal of Transport Geography, 9, pp. 199-208.
Weibull, J.W., 1976, An axiomatic approach to the measurement of accessibility,
Regional Science and Urban Economics, 6, pp. 357-379.
Wieand K. (1987) “An extension of the monocentric urban spatial equilibrium model to a
multicenter setting: The case of the two center city,” Journal of Urban Economics,
21, 259-271.
Wolch, Jennifer R., and S.A. Gabriel, 1981, .Local Land Development Policies and
Urban Housing Values, Environment and Planning A, Vol.13, 253-1276.
Yang, J. (2008) “Policy implications of excess commuting: Examining the impacts of
changes in US metropolitan spatial structure,” Urban Studies, 45(2), 391-405.
Zhang, Y. and K. Sasaki (1997) “Effects of subcenter formation on urban spatial
structure,” Regional Science and Urban Economics, 27(3), 297-324.
Zhang, Y. and K. Sasaki (2000) “Spatial structure in an open city with a subcenter,”
Annals of Regional Science, 34(1), 37 - 53.

Abstract (if available)

Abstract This dissertation examines the role of local governments in the evolution of metropolitan spatial structure, particularly with respect to the growth of employment centers -- locations with significant concentration of economic activity and hence employment. The study is conducted in two parts: one, quantitative analysis of employment center growth in the Los Angeles region between 1990-2000

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

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

PDF

Urban spatial transformation and job accessibility: spatial mismatch hypothesis revisited

PDF

Urban spatial structure, commuting, and growth in U.S. metropolitan areas

PDF

Essays on congestion, agglomeration, and urban spatial structure

PDF

Beyond spatial mismatch: immigrant employment in urban America

PDF

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

PDF

Satisfaction with local "public goods" and services: the effects of household income and privatization in southern California

PDF

The relationship between women's household responsibilities and commute lengths: a study on women in the US and Great Britain

PDF

Essays on economic modeling: spatial-temporal extensions and verification

PDF

Children’s travel behavior in journeys to school

PDF

Linking participation, program design and outcomes; voluntary air quality programs at the ports of Los Angeles and Long Beach

PDF

Choice neighborhoods: a spatial and exploratory analysis of Housing Authority City of Los Angeles public housing

PDF

The institutionalization of nonprofit management: emergence, development, and legitimization

PDF

Delimiting the postmodern urban center: an analysis of urban amenity clusters in Los Angeles

PDF

Recent experience in the utilization of private finance for American toll road development

PDF

Governing public goods: how representation and political power in local and regional institutions shape inequalities

PDF

A study on the impact of congestion pricing: Benefits and revenue redistribution for the case of Seoul and Los Angeles

PDF

Jump -starting Main Street: A case study of the Los Angeles Neighborhood Initiative (LANI)

PDF

The geography of Black commerce and culture: Los Angeles, California, and beyond

PDF

Social construction of the experience economy: the spatial ecology of outdoor advertising in Los Angeles

Asset Metadata

Creator Agarwal, Ajay (author)

Core Title Testing the entrepreneurial city hypothesis: a study of the Los Angeles region

School School of Policy, Planning, and Development

Degree Doctor of Philosophy

Degree Program Planning

Publication Date 07/01/2009

Defense Date 04/09/2009

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

Tag employment center,growth management,OAI-PMH Harvest,urban spatial structure

Place Name Burbank (city or populated place), Los Angeles (city or populated place), Pasadena (city or populated place)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Giuliano, Genevieve (committee chair), Moore, James Elliott, II (committee member), Redfearn, Christian L. (committee member), Richardson, Harry W. (committee member)

Creator Email ajayagar@usc.edu,foragarwal@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2321

Unique identifier UC1443088

Identifier etd-Agarwal-2891 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-564516 (legacy record id),usctheses-m2321 (legacy record id)

Legacy Identifier etd-Agarwal-2891.pdf

Dmrecord 564516

Document Type Dissertation

Rights Agarwal, Ajay

Type texts

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

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu