Exploring the Linkage between Economic Base,

Revenue Growth, and Revenue Stability

in Large Municipal Governments
Working Paper WP17MO1

Michael Overton
University of North Texas

Robert Bland
University of North Texas

May 2017
The findings and conclusions of this Working Paper reflect the views of the author(s) and have not been
subject to a detailed review by the staff of the Lincoln Institute of Land Policy. Contact the Lincoln Institute
with questions or requests for permission to reprint this paper.
[email protected]

© 2017 Lincoln Institute of Land Policy

 Abstract

The Great Recession had far-reaching consequences for the fiscal health of municipal
governments. The deeply weakened economy has undermined the capacity of local governments
to meet their financial obligations. City leaders face a trifecta of shrinking tax bases and aid from
federal and state governments, increasing demands for public services, and deteriorating
infrastructure. Local governments increasingly find themselves as fiscal islands left to fend for
themselves. Given this reality, it is essential that scholars study how local governments can build
their local economies—their economic base—to more effectively guide public policy. Adding to
the urgency for this inquiry is President Trump’s call for investment in rehabilitating and
expanding public infrastructure as a cornerstone in his goal to rebuild America. Using data from
large U.S. cities, this study explores how a city’s economic base affects its revenue growth and
volatility.
 About the Authors

Dr. Michael Overton is a Research Associate with the Center for Public Management at the
University of North Texas [UNT] and the Grant Manager for the John Peter Smith Foundation.
In May 2015, he earned his Ph.D. in Public Administration from UNT where he also received his
Masters of Public Administration. His research interests include local government management,
public budgeting and finance, and urban governance issues, and has been published in prominent
journals including The American Review of Public Administration, State and Local Government,
and Public Money and Management. In 2015, he was awarded the Toulouse Dissertation Award
in Social Science for best social science dissertation at UNT. Recently, he was selected (1) for
the 2016 Lincoln Scholars Program hosted by the Lincoln Institute of Land Policy and (2) as a
2017 American Society of Public Administration Founders Fellow. As a graduate student, he
was also the recipient of the Hatton Sumners Scholarship.

University of North Texas

Center for Public Management
8812 Chaps Avenue
Fort Worth, TX 76244
[email protected]

Robert L. (Bob) Bland is Professor of Public Administration at the University of North Texas
where he teaches graduate courses in public finance, governmental accounting, and budgeting.
He has been on the faculty at UNT since 1982. He is the author of A Revenue Guide for Local
Government (2nd edition, 2005) and A Budgeting Guide for Local Government (3rd edition, 2013)
both published by the International City/County Management Association, and of several articles
on the municipal bond market, property taxation, and municipal budgeting. On two ocassions he
conducted workshops in Poland on revenue sources for local governments. In 2006 he was
selected as an Honor Professor by the UNT Student Government Association. He was the
recipient of the first Terrell Blodgett Academician Award presented by the Texas City
Management Association in 2007. He also received the Stephen B. Sweeney Academic Award
from ICMA in October 2007. In 2012, he was elected as a fellow in the National Academy of
Public Administration, an independent, non-profit and non-partisan organization that provides
expert advice to Congress and federal agencies.

University of North Texas

Endowed Professor of Local Government and Faculty Director
Center for Public Management
410 Avenue C
Denton, TX 76201
[email protected]
 Table of Contents

Introduction ..................................................................................................................................... 1

Fiscal Health ................................................................................................................................... 2

Revenue Growth and Income Elasticity ......................................................................................... 3

Measuring Revenue .................................................................................................................. 3

Measuring Economic Growth ................................................................................................... 4

Revenue Volatility .......................................................................................................................... 5

Economic Base................................................................................................................................ 6

Aggregate Employment ............................................................................................................ 8

Diversification........................................................................................................................... 8

Clustering .................................................................................................................................. 9

Research Design.............................................................................................................................. 9

Dependent Variables ............................................................................................................... 10

Operationalizing a Municipality’s Economic Base ................................................................ 11

Methods................................................................................................................................... 13

Results .......................................................................................................................................... 15

Conclusion .................................................................................................................................... 23

References ..................................................................................................................................... 25
 Exploring the Linkage between Economic Base, Revenue Growth, and Revenue Stability
in Large Municipal Governments

Introduction

The Great Recession of 2007 to 2009 had far-reaching consequences for the fiscal health of
municipal governments. The resulting weakened national economy stretched the capacity of their
tax bases to meet their financial obligations, forcing many to reevaluate their fiscal strategies. In
balancing current needs with future aspirations, local government managers face a trade-off
between balancing the immediate need for budget stability against investing in future economic
growth (Felix 2008; White 1983).

On the one hand, the adoption of and reliance on stable revenue sources mitigates the negative
consequences of economic downturns. Revenue stability has many benefits including fewer
disruptions in service delivery (Yan 2011), less expenditure volatility driven by economic cycles
(Hendricks and Crawford 2014), and a higher municipal bond rating (Grizzle 2010). On the other
hand, reliance on stable revenues may result in slower long-term revenue growth from the
income inelasticity of these revenue sources. Cities that rely more heavily on income elastic
revenue sources capture more revenue from the growth in their local economy but may sacrifice
the stability of their local revenue streams in return. Both revenue stability and revenue growth
are essential to a city’s long-term fiscal health.

While revenue stability and growth have been examined by many studies, most have explored
how growth in income (Sobel and Holcombe 1996; White 1983) or the fiscal structure of a local
government impacts the volatility or revenue growth (Carroll 2009; Yan 2011). Lacking is an
analysis of the impact that a city’s economic base—its mix of private and public sector
enterprises—has on the growth and stability of its revenue. A city’s economic base is the
foundation from which its revenue is ultimately generated—it is the driver of local prosperity.
This study provides policy leaders a clearer understanding of the link between a city’s economic
base and its long-term fiscal health. No prior study has linked a city’s economic base to its
revenue sources. How do the different types of business sectors that comprise a city’s economic
base impact the volatility of its property tax, sales tax, and own-source revenue in general?

Panel data are used to assess the relationship between a city’s economic base and the growth and
volatility of its revenues. Specifically, this study focuses on the impact of changes in a city’s
economic base on own-source revenue, and then separately on the two most important own-
source revenue streams to local governments sales tax and property tax revenue. As such, proper
attention is given to the impact that a municipality’s economic base has on the fiscal health of a
city through the analysis of own-source revenue and the two most ubiquitous taxing instruments
available to local governments in the United States. Data between 2005 and 2012 for 149 fiscally
standardized cities1 from the Lincoln Institute of Land Policy (Lincoln Institute of Land Policy

1
Data were obtained from the Lincoln Institute of Land Policy. Fiscally Standardized Cities database.
(http://www.lincolninst.edu/subcenters/fiscally-standardized-cities/) Rutland, Vermont is included in the original
150-city FiSC database but was deleted from the sample for this study because it is located in a Micropolitan

Page 1
2016) and industry data from the County Business Pattern Dataset are analyzed using two-way
fixed effects models.

Income elasticity measures the sensitivity of a revenue source to changes in the economy
(Carroll and Goodman 2011). Revenue volatility, on the other hand, measures the difference
between actual revenue and that predicted from a trend. While inelastic revenue sources provide
stable yields, they grow slower than the economy, creating a gap between actual and growth-
adjusted trends. Income elastic sources grow faster than the economy providing a revenue
surplus in good economic times. As the economy slows, so does the revenue growth of income
elastic revenues. As a local government relies on more income elastic revenues, it is likely to see
faster growth in its revenues but experience more revenue volatility over the long term due to the
business cycle. Conversely, as local governments rely more on income inelastic sources of
revenue, they grow slower but have more stable revenue yields over the business cycle.

The paper begins with a brief introduction to strategies for promoting fiscal health and provides
the theoretical basis for connecting a municipality’s economic base to its revenue structure. Next,
drawing on the fiscal health literature, revenue elasticity and volatility are discussed followed by
a discussion of the conceptual and theoretical foundation of a municipality’s economic base. The
last half of the study begins with an elucidation of the methods employed, followed by a
discussion of the research findings and ends with a summary and the implications for public
policy.

Fiscal Health

Fiscal health is the ability of a municipal government to meet current financial, service, and
capital obligations (Carroll and Goodman 2011). Local governments must be able to pay
operational expenses—the annual costs for continuing services—plus one-time capital outlays
for infrastructure to satisfy the full range of citizen demands. Fiscal health ultimately depends on
the capability of a local government to meet its current service and capital demands in addition
to its ability to sustain prosperity in the long term while adapting to external influences
(Hendrick 2011).

Fiscal health is a multidimensional concept that encompasses multiple periods while accounting
for internal capacity and external constraints. In a practical sense, a fiscally healthy local
government must (1) provide ongoing services at a socially acceptable level, (2) meet its capital
needs for infrastructure, (3) meet long-term liabilities such as bond debt and accrued pensions
while (4) simultaneously adjusting its fiscal strategies to changing political and economic
environments.

In their pursuit of fiscal health, local government managers must balance competing forces such
as the current service expectations of constituents with future capital needs of the community.
Revenue growth and volatility are important factors of a local government’s capacity to balance

Statistical Area while the other 149 cities are located in Metropolitan Statistical Areas. Micropolitan statistical areas
are smaller, more rural, and less economically developed than metropolitan statistical areas, and thus not comparable
to the other cities in the study sample.

Page 2
countervailing pressures and maintain fiscal health. While fiscal health has been measured in a
variety of ways, this study uses revenue volatility and growth because they are key determinants
of a municipality’s fiscal health broadly defined (Carroll and Goodman 2011).

Local government managers typically prefer a revenue structure that captures the benefits to
revenue yield from economic growth while simultaneously minimizing revenue volatility.
Unfortunately, these goals are often at odds (Carroll and Goodman 2011; Felix 2009) because
revenue growth occurs with highly elastic revenue systems. However, elastic revenues are
sensitive to shifts in the economy that drive revenue volatility. Gains in revenue through growth
of the local economy leave municipal governments more susceptible to revenue volatility.
Conversely, stable revenues are often associated with limited economic growth because inelastic
revenue sources are less sensitive to shifts in the economy. Officials must strategically manage
these two opposing outcomes to achieve a fiscally healthy revenue structure.

Both outcomes—growth and stability—can promote fiscal health depending on the budgetary
requirements of a local government. If a local government needs to increase services or
infrastructure, a strategy aimed at growing revenue by expanding the tax base gives its leaders
the flexibility to meet those service expectations. Municipalities that are expanding will
inevitably need to fund that expansion and will benefit from a growth-focused revenue strategy.
Conversely, a local government that provides an appropriate level of services will desire more
revenue stability to maintain service levels across business cycles. Bedroom communities and
local governments that have reached buildout or otherwise have limited growth potential will
benefit by pursuing strategies that promote more revenue stability.

Revenue Growth and Income Elasticity

Under normal circumstances, a city experiences revenue growth when its economy expands.
However, the relationship between the two variables varies widely depending on the composition
of the city’s economy and the revenue sources that tap that economic base. The more closely
linked a city’s taxes and fees are to the composition of its economy, the more income elastic the
revenue yield. Local government managers know that they can maximize revenue growth by
relying heavily on income elastic revenue sources.

Analysts face two important decisions when measuring revenue elasticity: how to measure
“revenue,” and how to measure “the economy” (Carroll and Goodman 2011)? Measuring the
income elasticity of taxes and fees is typically accomplished by regressing the natural log of
annual yield from a revenue source on the natural log of the revenue base. Revenue yield is the
amount of revenue levied by a local government, whereas the revenue base is the value of the
objects or transactions to which a tax rate is applied (Bland 2013).

Measuring Revenue

Revenue yield rather than revenue base is used in this study to assess the impact of the
components of the local economy on local budgets. Using revenue base to measure elasticity has
the advantage of isolating the effects of wealth generation from confounding policy decisions

Page 3
such as a change in the tax rate. Growth of a city’s revenue base is also important when
preparing budget forecasts. However, budget decisions are made using revenue yield, not
revenue base. Revenue yield also captures the effect of rate changes in addition to growth or
decline in the base. Finally, given our interest in the impact of economic growth on budgetary
decisions, revenue yield provides a more appropriate measure of that linkage. Given the
distinctive budgetary processes that municipalities typically pursue when making budget
decisions for their tax-supported activities, we also examine two subsets of own-source revenue
yields: the property and sales taxes.

Revenue yield is defined as the total amount of revenue from tax and nontax sources other than
intergovernmental aid. In many cases, local governments adjust tax or service charges rates
annually and occasionally more frequently in the case of some municipal-owned utilities. Many
factors affect rate adjustment decisions that make using a city’s revenue base a less than ideal
way to capture revenue elasticity. First, tax rates are influenced by the tax rates of neighboring
jurisdictions. Local governments look to minimize business relocation and citizen dissatisfaction
by setting tax rates at a favorable level compared to neighboring jurisdictions (Besley and Case
1992; Brueckner 1998, 2003; Buettner 2001; Heyndels and Vuchelen 1998; Ladd 1992; Shleifer
1985). The revenue rates of neighboring municipalities act as an unintentional benchmark that
both businesses and citizens use to evaluate their city’s performance.

Second, revenue rates, particularly for the property tax, are often adjusted to offset changes in the
revenue base (Alm et al. 2011; Doerner and Ihlanfeldt 2011; Ihlanfeldt 2011; Lutz 2008; Lutz et
al. 2011; Vlaicu and Whalley 2011). When a revenue base decreases, local governments often
increase rates upward to balance their budget and vice versa. Some scholars debate the frequency
such offsetting practices occur (Ihlanfeldt 2011; Ihlanfeldt and Willardsen 2014; Ross and Yan
2013), although the budgetary motivations are apparent. Thus, rate adjustments made to balance
a municipal budget are frequently used to offset changes in their respective base.

Third, when preparing budget forecasts, public managers use the expected yield from each
revenue source plus the unassigned fund balance (i.e. that portion above the balance required by
policy) to establish a beginning point for budget negotiations. The unassigned fund balance is
particularly important because it allows local governments to maintain budgeted services and
capital outlays if there is an unexpected fiscal shock resulting in a revenue shortfall.
Furthermore, properly maintained slack resources, such as the unassigned fund balance, are
crucial for maintaining short-term cash solvency (Hendrick 2011). Rate adjustments can occur in
response to changes in a local government’s fund balance or to provide a cushion when slack
resources have been depleted or exceed policy requirements. Thus, using revenue yield rather
than revenue base provides a more accurate connection between a city’s economic base and the
revenue available for budgetary purposes.

Measuring Economic Growth

The other key factor of interest to this study is economic growth. Typical measures include
personal income, GDP (if the unit of analysis is county, state or nation), and unemployment. To
better parse the distinctive effects of the local economic base on revenue yields, this study uses

Page 4
three specifications that measure important dimensions of local economic development:
aggregate local economic growth, industrial diversification, and industrial agglomeration.

To capture the impact of a local government’s economic base on its revenue elasticity, this study
examines not only own-source revenue yield—revenue generated through taxes, fees and
services excluding all intergovernmental revenue—but also the yields from the property tax and
sales tax. These two taxing instruments (1) are widely used by local governments, (2) typically
comprise the largest non-intergovernmental portion of local government revenues, and (3) in the
case of the property tax, are more revenue inelastic and, in the case of the sales tax, are more
revenue elastic, thereby providing contrasting indicators to measure their impact on revenue
volatility.

Property taxes are income inelastic in the short-term because the tax base is fixed at a point in
time and in advance of setting the tax rate. Where local governments have control over the
property tax, the rate is typically adjusted by city councils to bring total revenues into balance
with total expenditures including capital spending, a process referred to as millage offsetting
(Iflanfeldt 2011). While property values fluctuate with the economy, locally set tax rates smooth
out year-to-year revenue yields from the tax. Sales taxes, however, are revenue elastic, and local
governments have less discretion than with the property tax to alter rates prior to or during the
fiscal year.

Revenue Volatility

Revenue volatility is a function of fiscal structure (Afonso 2014; Carroll 2009; Hendricks and
Crawford 2014) and fiscal policy space (FPS) (Hendricks and Crawford 2014; Pagano and
Hoene 2010). Fiscal structure refers to a city’s tax burden, types of revenues used, diversification
of revenue sources, fiscal capacity, and revenue complexity (Carroll 2009; Hendrick and
Crawford 2014). Previous research has focused on fiscal instruments that increase revenue
stability through slack resources and revenue diversification. FPS refers to exogenous
determinants of revenue volatility such as state and local laws that restrict a city’s tax options, its
economic base, and its political culture.

Slack resources are a local government’s “pool of resources in excess of the minimum necessary
to produce a given level of output” (Hendrick 2006, p.15). As a countercyclical measure, slack
resources accumulate when revenues increase and are spent during recessions as governments
work to stabilize expenditures (Bland 2013). While state governments use their slack resources
in this manner (Douglas and Gaddie 2002; Hou 2003, 2005; Hou and Moynihan 2008), local
governments are less inclined to use slack resources purposefully as a countercyclical measure
(Wang and Hou 2010). Yet, recent research indicates that slack resources improve a local
government’s fiscal health (Hendricks 2006; Marlow 2005) and promote revenue stability
(Carroll 2009; Hendricks and Crawford 2014).

Another fiscal tool of interest to public finance scholars is revenue diversification. After the 1978
property tax revolt in California that culminated in passage of Proposition 13, local governments
in other states pressured their legislatures for access to other revenues in addition to the property

Page 5
tax. By diversifying their revenue structures, local governments minimize the impact of a fiscal
shock on one revenue source (Carroll 2009) and capture revenue from individuals or businesses
that otherwise avoid paying the property tax (Ulbrich 1991).

Carroll (2009) finds that revenue diversification decreases revenue volatility in local
governments. But how a municipality diversifies its revenue is critical to achieving revenue
stability (Afonso 2014; Carroll 2009; Hendricks and Crawford 2014; Yan 2011). When cities
substitute revenue from the property tax with either a general sales or income tax, the income
elasticity of their tax base increases and with it the volatility of their revenues (Afonso 2014; Yan
2011). Afonso (2014) finds that counties that rely on more income elastic revenues have
increased revenue volatility. Yan (2011) finds that revenue diversification leads to increased
revenue stability when the employment rate is volatile, and that revenue diversification leads to
decreased revenue stability when the local economic base is stable.

In addition to fiscal structure, a city’s FPS heavily influences revenue and expenditure volatility.
FPS refers to the exogenous parameters that limit the policy options available to city officials
(Hendrick and Crawford 2014; Pagano and Hoene 2010). Five attributes shape the FPS of a city:
its intergovernmental context, locally imposed laws, its political culture, demand for public
goods and services, and its economic base. Hendricks and Crawford (2014) test for the effect of
fiscal policy space on spending volatility and find evidence that population and distance from a
central city decrease spending volatility. But a city’s status as a home rule city has no effect on
spending volatility. While this is a good first step toward understanding the effect a city’s FPS
has on revenue volatility, a more complete understanding would account for the impact of a
city’s economic base on its revenue volatility. This research extends the work by Pagano and
Hoene (2010) and Hendrick and Crawford (2014) to pursue a more detailed theoretical
explanation of the link between a city’s economic base and its fiscal health.

Economic Base

A city’s economic base is the economic resources that produce wealth for residents and
businesses, and ultimately generates revenue for the local government (Hendricks and Crawford
2014; Overton 2016; Peterson 1981). For this study, we define a local government’s economic
base as the aggregate of all public enterprises and private firms located in a city’s borders from
which revenue (property, sales, or other) is directly or indirectly generated (Overton 2016).

Private businesses and public organizations serve many functions that support the economic
prosperity of a local government such as providing income through employment, improving
capital accumulation by constructing buildings, and generating economic activity by producing
and selling goods and services. Some of these activities form the municipality’s revenue base,
which is directly taxed. However, other economic activities are not subject to taxation or other
local charges and fees—like many service industries—but indirectly these economic activities
can improve a city’s fiscal health by providing jobs, positive economic externalities, or
supporting a key local industry as an intermediary. Growth in the economic base normally
produces more jobs, higher personal income, greater consumption, and more property wealth.

Page 6
This economic activity affects a municipality’s revenue bases, revenue rate adjustments, and
revenue yield.

Conceptually, a municipality’s economic base is distinct from its revenue base. An economic
base encompasses all economic activity, while a revenue base is a subset of activities and objects
subject to taxation or other charges and fees. Ideally, the revenue structure of a city is designed
so that increases or decreases in a city’s economic base closely track as increases or decreases in
its revenue base. But the reality of implementing an equitable and efficient tax system results in
an imperfect association between the two (Pagano and Hoene 2010; Tannenwald 2004).

Recently, the use of NAICS data and the evolving economy have exacerbated the differences
between a municipality’s economic base and its revenue base. NAICS provides the most
comprehensive and comparable industrial data available to local government professionals. It
groups together similar businesses for the purposes of providing useful and comparable data. The
NAICS was initially developed in 1997, replacing the previous Standard Industrial Classification
(SIC) system. While the SIC system poorly organized new economic activity, the NAICS
overcame this aggregation defect by adopting a unifying economic concept for grouping
industries—referred to as a supply-side concept. The supply side concept defines and groups
industries based on the similarity of their production processes. While the NAICS created a
robust classification system, it was not created with the purpose of classifying industries by their
contribution to local tax revenue or their conformance to tax codes. The NAICS measures many
aspects of a local government’s economic base well, but not its revenue base. Smaller datasets
exist at the local and state level, but none of these datasets offer national coverage using a
standardized format. The lack of a nationally standardized way to connect local economic and
revenue bases has prevented analysts from studying their interaction.

In addition to data limitations, policy makers have been unable to create taxing instruments and
update tax legislation at the same pace that the economy is evolving. The fiscal structures of
local governments were designed for older economic institutions, and the fiscal structures that
once worked have failed to keep pace with emerging economies (Tannenwald 2004). First, the
US economy has been increasingly shifting toward a service economy. While the trend away
from manufacturing started as far back as 1939 (Short 2011), the national economy continues
shifting toward service-providing industries. In the 10-year period between 2004–2014, gross
employment in the service sector grew by 9% and made up 80.1% of all employment, while
goods-producing industries declined in gross employment by 12.1% making up only 12.7% of all
employment in 2014 (Bureau of Labor Statistics 2015). The consumption of intangible services
has increased as a percentage of total consumer consumption. Unfortunately, tax bases tend to
rely on the consumption of tangible goods. Industries that provide services are difficult to tax,
and often exempt from sales taxes, due to cross-border concerns.

Second, technology, the internet, and social media have created new industries while
simultaneously reshaping the way commerce occurs. Local tax structures have not kept pace with
the high-tech revolution and fail to distribute the tax burden fairly across newer versus older
industries. The emergence of e-commerce, the internet of things, the “sharing” economy, big
data, machine learning, and the subsequent loss in state and local sales tax revenue is evidence

Page 7
that technological developments have out-grown state and local fiscal structures. Tax policies
and instruments have failed to keep pace with the evolving economy.

The NAICS data and the changing economy have contributed to expanding the divide between a
municipal government’s economic base and its revenue bases. The purpose of this study is to
determine how a local government’s economic base impacts revenue elasticity and volatility.
Previous studies into revenue volatility and growth either (1) do not include economic base
variables or (2) use broad aggregate outcome measures—such as unemployment rate—that do
little to differentiate between different components in a city’s economic base. This study
proposes measuring a local government’s economic base using three conceptual measures:
measures of aggregate economic base employment, diversification, and clustering.

Aggregate Employment

The overall strength of a local economy is frequently measured using the city’s unemployment
rate. Certainly, a low unemployment rate is an indication of a strong local economy where most
residents have an income, are self-sufficient, and are likely to spend money, which promote
economic growth. Unemployment is an important concept when determining the strength of a
local government’s economic base, yet it fails to explain why a city’s economic base is strong or
resilient to economic shocks.

Another aggregate employment measures is unemployment volatility—the difference between a

long-run unemployment trend line and actual unemployment. As unemployment volatility
increases, a city’s economic base is more susceptible to noncyclical economic shocks indicating
that the economic base as a whole lacks stability (Yan 2011). Unlike unemployment rate that
signals the strength of a local economy, unemployment volatility is an indicator of an economic
base’s resilience to economic shocks.

This study contends that the strength of an economic base is partially dependent on the mix of
industries comprising that base. Businesses interact with each other in a dynamic economic
system where they are supported by intermediaries and threatened by competitors. The actual
industry type is less important than how those industries are organized, giving rise to the
concepts of economic diversification and clustering.

Diversification

Diversification is a two-dimensional concept concerned with (1) the breadth of different

component parts and (2) the balance between these options (Goodman 2016). Industrial diversity
is the number of different industries and the intensity of each industry’s utilization in the local
economy. Diversified local economies, those economies with a broad and balanced mix of
industries, show more stable growth in the long-run because their growth is not dependent on any
single industry (Chinitz 1961). In addition, diversified economies create robust knowledge
spillover networks (Chinitiz 1961; Jackson 2016; Jacobs 1969), enhance local capital investment
(Chinitz 1961), and improve firm survival rates (Renski 2011).

Page 8
Modern portfolio theory provides insight into the effect of diversification on fiscal health. It
argues that an investor owning a broad number of stocks and bonds will have far more stable
returns in the long-run than a portfolio containing one stock. Because the growth of stocks and
bonds is not perfectly associated, stocks dropping in value will not result in total loss of the
stockholder’s investment (Brealey and Myers 1991). From the viewpoint of modern portfolio
theory, diversification mitigates risk and stabilizes growth.

However, modern portfolio theory assumes that there is an available and accessible mix of stocks
and bonds whose prices and risk do not perfectly correlate. This assumption is relatively safe
given the global supply of investment opportunities. Yet, this assumption does not universally
apply to local or regional economies because the different industries in a diversified economic
base are interdependent. Some industries provide supporting services for other industries similar
to the relationship between mining and manufacturing and manufacturing and
transportation/warehousing. Some industries do not compete for a global market at the local
level—industries like retail trade and food services rely on local foot traffic and the synergistic
gains from their location. Despite the industrial interdependence in a local economy, research
shows that a more diversified economy has more stable long-run growth, but not necessarily
higher long-run economic stability (Chinitz 1961).

Clustering

Clustering occurs when firms that support the output of a central industry locate geographically
in proximity to each other. Clustering, also called agglomeration, gives firms a competitive
advantage through lower production costs (Porter 2000). It creates competitive advantages from
the increased productivity of complementary firms, increased innovation among businesses in
that industry, and new business formation beyond what occurs outside the cluster (Porter 2000).
The clustering of similar primary and intermediary industries is likely to improve the economic
condition of a local government because it improves productive efficiency of the economic base.
Efficient production matters because it enables a city’s economic base to compete in a global
market and is more likely to weather economic downturns than non-clustered competitors.

Research shows that clustered industries have lower supply costs and increased innovation.
Clustered firms gain competitive advantages by sharing infrastructure (Burchfield, Overman,
Puga, and Turner 2006), sharing a labor market pool (Overman and Puga 2010), improving
proximity to suppliers (Amiti and Cameron 2007), increasing the likelihood of employee-
employer matching (Costa and Khan 2000), and encouraging knowledge spillovers (Audretsch
and Feldman 1996).

Research Design

Two measures are used to gage the impact of economic base on revenue yield: revenue elasticity
and revenue volatility. Revenue elasticity and revenue volatility represent a vector of dependent
variables. To understand the connection between economic base and revenue yield this study
uses the revenue elasticity and volatility for own-source revenue, property tax revenue, and sales

Page 9
tax revenue. By exploring these individual revenue streams, a more accurate measure of the
linkages between various industries and individual revenue streams can be provided.

Revenue elasticity (𝑅𝑅𝑅𝑅𝑖𝑖𝑖𝑖 ), and revenue volatility (𝑅𝑅𝑅𝑅𝑖𝑖𝑡𝑡 ) are modeled using the following
specifications:

𝑅𝑅𝑅𝑅𝑖𝑖𝑖𝑖 = ∝ + 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝑖𝑖𝑖𝑖 + 𝐸𝐸𝐸𝐸𝐸𝐸𝑖𝑖𝑖𝑖 + 𝐸𝐸𝐸𝐸𝐸𝐸𝑖𝑖𝑖𝑖 + 𝑅𝑅𝑅𝑅𝑖𝑖𝑖𝑖−1

𝑅𝑅𝑅𝑅𝑖𝑖𝑖𝑖 = ∝ + 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝑖𝑖𝑖𝑖 + 𝐸𝐸𝐸𝐸𝐸𝐸𝑖𝑖𝑖𝑖 + 𝐸𝐸𝐸𝐸𝐸𝐸𝑖𝑖𝑖𝑖 + 𝑅𝑅𝑅𝑅𝑖𝑖𝑖𝑖−1 + REit

Revenue elasticity (𝑅𝑅𝑅𝑅𝑖𝑖𝑖𝑖 ) and revenue volatility (𝑅𝑅𝑅𝑅𝑖𝑖𝑖𝑖 ) are functions of aggregate economic base
employment (𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝑖𝑖𝑖𝑖 ), economic base diversification (𝐸𝐸𝐸𝐸𝐸𝐸𝑖𝑖𝑖𝑖 ), economic base clustering
(𝐸𝐸𝐸𝐸𝐸𝐸𝑖𝑖𝑖𝑖 ), and the revenue elasticity (𝑅𝑅𝑅𝑅𝑖𝑖𝑖𝑖−1) and volatility (𝑅𝑅𝑅𝑅𝑖𝑖𝑖𝑖−1 ) of the previous year.
Economic base employment, diversification, and clustering are all used to understand not only
the impact certain industries have on revenue elasticity and revenue volatility but also the impact
that diversification and clustering within those industries has on a local government’s fiscal
health.

Annual data from 2005 to 2012 were obtained for 149 fiscally standardized cities from the
Lincoln Institute’s FiSC database, from the U.S. Census Bureau’s County Business Pattern
database, and from the Bureau of Economic Analysis. The final set of observations constitutes a
balanced dataset of 1043 observations.

The unit of analysis for this study is a fiscally standardized city [FiSC]. A FiSC is a fictitious city
created to facilitate fiscal comparisons among local government service areas. One problem with
using cities as the unit of analysis is that there is no standard package of services. While some
cities provide most, if not all, public services, others provide only a small portion while the
overlapping counties and special districts pick up the balance. FiSCs are constructed by taking
the financial data of 149 large cities and adding that to the financial data of overlapping
governments. The key benefit of using FiSCs is that it allows for cross-city comparison of total
revenue burden and services received. Currently, FiSCs provides annual data through 2012, but
the industry data used to test the economic base hypotheses are only available in their current
form starting in 2005. Thus, our sample is limited to the years 2005–2012. 2

Dependent Variables

To fully explore the impact of a city’s economic base on revenue elasticity and revenue
volatility, three different revenue sources are used as dependent variables: own-source revenue,
property tax revenue, and sales tax revenue. Own-source revenue is the aggregate of all revenue
generated by the local government excluding state and federal aid. Property and sales tax
revenue are subsets of a local government’s own-source revenue. Revenue elasticity is
operationalized as the natural log of these various revenue streams divided by the natural log of
economic base.

2
The first year of the North American Industrial Classification Systems (NAICS) was 1998. Previously, the
Standard Industrial Classification (SIC) system was used. While there is some comparability between the two, it is
limited—especially at the level of detail of this study.

Page 10
Revenue volatility is operationalized as the absolute value of the predicted revenue minus the
actual revenue divided by the predicted revenue to standardize the variable. We employ a two-
step procedure adapted from previous studies to create our revenue volatility measures (Hou
2003; Marlowe 2005; Carroll 2009; Wang and Hou 2009; Carroll and Goodman 2013; Hendricks
and Crawford 2014). The first step estimates a long-run revenue growth trend regression (Carroll
2009).

R 𝑖𝑖𝑖𝑖 = exp(∝ + 𝛽𝛽1 𝑖𝑖 + 𝛽𝛽2 𝑡𝑡)

Revenue for city i in year t, R 𝑖𝑖𝑖𝑖 , is modeled as a series of dichotomous variables, which indicate
the year and each municipality in the dataset. 3 After estimating revenue or expenditure volatility,
the next step is to determine the deviation of actual revenues and expenditures from the values
derived from the first step. The absolute value of the difference between predicted and actual
values is a measure of budget volatility, our dependent variable. This value is then divided by the
predicted revenue. As this value approaches zero, revenue volatility decreases. Volatility
measures are calculated for all revenue sources: own-source revenue, property tax revenue, and
sales tax revenue.

Operationalizing a Municipality’s Economic Base

Economic base variables are divided into three categories: aggregate employment measures,
diversification, and clustering. The primary independent variable of interest in this study is the
city’s economic base. While the Economic Census provides the data needed for this study at the
city level, it has two limitations. First, the Economic Census only comes out every 5 years, which
would allow estimating the model using data from only three time periods (i.e., 2002, 2007, and
2012). Second, the Economic Census of 2002 and 2007 only provide data for cities with a
population over 5000, and the 2012 economic census provides data for cities with populations
over 2500. While this sampling procedure has no direct effect on the FiSCs used in the dataset, it
does limit inference of various spatial factors.

To overcome these limitations, county-level data from the County Business Patterns dataset are
used. This dataset provides data for every county for every year between 2005 and 2012. We
conduct an analysis of the connection between a city’s economic base and its budget volatility.
The County Business Patterns dataset uses the North American Industry Classification System
(NAICS), which groups businesses with similar products and services into the same category.
The classification is divided into NAICS codes with various levels of classification detail. Two
digit NAICS codes (21 different classifications) are the broadest classification and groups
businesses by sector (i.e. manufacturing, retail, entertainment, etc.) while three digit NAICS
codes (82–87 different classifications depending on the year) classify businesses by subsectors.
The economic base measures present detail based on the two digit NAICS codes. Although there
are significantly greater levels of industrial classification, predictive power is lost using these
narrower industrial classifications (Billings and Johnson 2012).

3
Although the primary sample only uses data from 2005-2012, this budget growth trend regression model uses all
financial data available; 1977-2012. By using a greater range of data, a more accurate growth trend regression model
can be estimated.

Page 11
In order to capture the overall economic climate in a municipality, unemployment rate and
unemployment volatility are included in the analysis. Unemployment rate is the percent of the
population in the labor force that is actively seeking employment but not currently employed.
This measure captures overall productive efficiency, utilization of the labor force, and individual
employee economic distress. Unemployment volatility is calculated using the same two-step
procedure to calculate revenue volatility and is used to capture overall volatility in the local
economy. This variable captures local economic shocks that are difficult to operationalize in
addition to accounting for the overall economic stability of the local economy. Unemployment
data are collected at the county level.

Economic base diversification is measured in two ways using the Hirschman Herfinadahl index
(HHI): one employment-based and one firm-based. The employment based measure tests for the
impact of industrial diversification on a city’s revenue elasticity and volatility. While this
employment-based measure accounts for the mix of industries, it does not account for the market
power of firms in those industries. If a town has an equally proportional industrial mix, but each
of those industries is dominated by one or a few firms, then the city might be more vulnerable
than the industrial diversity measure indicates. To account for this potential influence, a firm
based HHI is included that measures how concentrated employment is among firms across
industries. The HHI is calculated using the following formula

� 𝑆𝑆𝑖𝑖𝑖𝑖2
𝑖𝑖𝑖𝑖

The HHI is the sum of each 2-digit industry’s share of either the number of employees or the
number of firms in county j.

This study uses two measures to capture clustering: industrial agglomeration and industry
productivity. A location quotient measures the agglomeration of an industry in a city relative to
the regional sorting of that industry. Specifically, a location quotient uses industry employment
data to calculate the ratio of the share of the focal city’s employment comprised of a specific
industry compared to the state’s share of an industry’s employment relative to total state
employment.

The LQ of industry i in city j is calculated using the following formula:

(𝑋𝑋𝑖𝑖𝑖𝑖 ⁄𝑋𝑋∗𝑗𝑗 )
𝐿𝐿𝐿𝐿𝑖𝑖𝑖𝑖 =
(𝑋𝑋𝑖𝑖∗⁄𝑋𝑋∗∗ )

where 𝑋𝑋𝑖𝑖𝑖𝑖 is the total number of full time employees in industry i in city j, 𝑋𝑋∗𝑗𝑗 , is the total
number of full time employees in all industries * in city j; 𝑋𝑋𝑖𝑖∗ , is the total number of full time
employees in industry i in state *; 𝑋𝑋∗∗ , is the total number of full time employees in all industries
* in state *. When a city’s LQ is greater than 1, then industry i is considered geographically
clustered in that city compared to other cities in the state.

While the LQ can determine the relative density of businesses in a region, an important
component of industrial clusters. However, it does not account for the actual economic strength

Page 12
of that industry. An industrial productivity proxy is used to help indemnify industries that drive
the overall regional economy (Hill and Brennan 2000). Productivity can be thought of as the
value added to an economy through every hour of work. Since no direct measures exist, a proxy
is created [PP].

𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖
PP𝑖𝑖𝑖𝑖 = [𝐺𝐺𝐺𝐺𝐺𝐺𝑖𝑖𝑖𝑖 ∗ ] / 𝐸𝐸𝐸𝐸𝐸𝐸𝑖𝑖𝑖𝑖
𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑖𝑖𝑖𝑖

P𝑖𝑖𝑖𝑖 is the productivity of industry i in county j. We know the GDP of an industry at the state
level. However, we want to know its productivity at the county level, which requires a proxy
measure. The PP is calculated by multiplying the GDP of industry i in state t by the payroll of
industry i in county j by payroll for industry i in state t. After assigning a portion of an industry’s
GDP to a county, this is divided by employment in industry i in county j.

Methods

To empirically examine the research questions, six two-way4 fixed-effects panel data models are
estimated. For all models, Hausman tests indicated that FE models are the appropriate estimation
technique. Specification tests also showed the presence of heteroscedasticity 5 and serial
correlation. 6 Robust standard errors with Andrews weights are used to address heteroscedasticity
and serial correlation (Andrews, 1991; Andrews and Monahan, 1992). Furthermore, all
independent variables that vary annually are lagged by one year in order to overcome possible
endogeneity (Wooldridge, 2010). The sample for regressions where sales tax revenue was the
dependent variable required that we drop 63 observations because sales taxes were not collected
in nine cities in the FiSC dataset.

4
F-tests for time and individuals effects were statistically significant (P<0.01) for all three models.
5
Breusch-Pagan test for heteroskedasticity was statistically significant at a 95 percent confidence level for all
models, which suggests the presence of heteroscedasticity.
6
The Breusch-Godfrey test for panel models was statistically significant at a 99 percent confidence level for all
models indicating the presence of serial correlation (Wooldridge 2012).

Page 13
Table 1: Descriptive Statistics

Page 14
Table 1 Continued: Descriptive Statistics

Results

Table 1 presents the descriptive statistics for variables used in the analysis. Table 2 presents the
results of the regressions with elasticity as the dependent variable. Table 3 presents the results of
the regressions using revenue volatility as the dependent variables. Overall, all six models
presented in table 2 and table 3 exhibit excellent goodness of fit statistics. Every F-statistic is
statistically significant at a 99.9% confidence level indicating that each model is jointly
significant. The adjusted R-squared values for each model are also excellent. All three elasticity
models presented in table 2 have an adjusted R-squared at or above .9831. The three volatility
models in Table 3 also have excellent adjusted R-squared values. Overall, the fixed effects
specifications have strong model diagnostics.

Page 15
Table 2: Elasticity Regression Results

Page 16
Table 3: Volatility Regression Results

Page 17
The aggregate measure variables, unemployment rate and unemployment volatility, in table 2
and table 3 were included in the models to account for overall economic conditions. A
municipality’s unemployment rate is negatively associated with own-source revenue elasticity
and property tax revenue elasticity, but no statistically significant relationship was found
between unemployment and sales tax revenue elasticity. These findings suggest that as
unemployment increases, own-source revenue and property tax revenue have lower elasticities.
These findings are intuitive because higher unemployment indicates a less productive local
economy, which leads to lower value being added to the community and consequently lower tax
revenues. The lack of a statistically significant relationship between the unemployment rate and
sales tax revenue elasticity is likely occurring because community wealth and consumption
habits are being captured by the municipal fixed effects.

The results indicate that there is a complex relationship between unemployment rate and revenue
volatility. First, the local unemployment rate has no statistically significant impact on own-
source revenue volatility. Since own-source revenue includes sales and property tax revenue,
perhaps the opposite coefficients negate each other when revenue is aggregated. Second, the
unemployment rate has a statistically significant relationship with the volatility of both sales tax
and property tax revenue. However, an increase in the local unemployment rate affects sales tax
volatility and property tax volatility in opposite ways. As the local unemployment rate increases,
the revenue volatility of sales tax yield increases and property tax yield decreases. An intuitive
theoretical proposition is that higher unemployment destabilizes the local economy by providing
less disposable income to consumers, and with it a loss of property value, particularly for retail,
food, and entertainment establishments.

Third, the unemployment rate is negatively associated with property tax revenue volatility. This
finding is less intuitive, but still logical. As the unemployment rate increases, property tax
volatility decreases because prices in the real estate market stabilize or even decline as demand
for property eases. Local government managers adjust millage rates to ensure that the property
tax yield balances the budget (Alm et al. 2011; Doerner and Ihlanfeldt 2011; Ihlanfeldt 2011;
Lutz 2008; Lutz et al. 2011; Vlaicu and Whalley 2011). However, millage rate adjustments are
sensitive to the local political climate. When citizens are out of work, city officials are less
inclined to risk the political backlash associated with increasing property tax rates. By holding
millage rates constant or raising the rates only marginally, property tax stability increases during
times of high unemployment.

Unemployment volatility is a measure of economic base stability used in previous studies (Yan
2011). However, in all six models, unemployment volatility, the difference between the expected
and actual unemployment rate, is statistically insignificant at a 90 percent confidence level.
Separate regressions were run (1) without the year fixed effects, (2) without municipal effects,
(3) without the clustering and diversification measures, and (4) on a single year rather than
across the whole panel to see if any of the variables were capturing economic instability. 7 In
every iteration across all six dependent variables, unemployment volatility was statistically
insignificant suggesting that aggregate economic base instability has no impact on revenue
growth and volatility.

7
These regressions will be provided upon request.

Page 18
Economic base diversity was expected to increase elasticity and decrease stability across the
three revenue sources. However, diversity based on the number of firms (i.e., establishment HHI)
was statistically insignificant in all six models and diversity based on employment (i.e.,
employment HHI) was only statistically significant for property tax revenue elasticity and
volatility. An increase in employment diversification 8 results in an increase in property tax
elasticity and volatility. As predicted, diversification of a municipality’s industrial base results in
greater long-run elasticity in property tax revenue.

A diversified economic base has two distinct advantages for increasing property tax growth.
First, local governments can better “pick and choose” which industries to attract to their
jurisdiction to maximize property values (Fischel 2009). A diversified economic base implies
that a local government has multiple businesses that wish to locate within the jurisdiction. An
array of industries enables local officials to maximize revenue through zoning laws that prevent
firms associated with negative externalities (i.e., sex shops and manufacturing firms that pollute)
and enable relocation of firms with positive externalities (e.g., shopping centers, warehouse
space in a manufacturing heavy municipality, etc.). Second, a diversified economic base allows
industrial synergies to emerge in the city. These positive externalities are capitalized into higher
property values (Bogart and Cromwell 1997; Clark and Herrin 2000; Fischel 2009; Geoghegan et
al. 2003; Hilber and Mayer 2009; Oates 1969).

An increase in employment diversification also results in an increase in property tax volatility.

While modern portfolio theory predicts that employment diversification creates greater revenue
stability, local industries are intrinsically linked and therefore a decrease in volatility is unlikely
to occur. While it is fairly clear why diversity did not decrease volatility, it is not entirely clear
why employment diversity is positively associated with property tax revenue volatility. One
possible explanation is that diversity lends itself to local networks, where the failure of one
business ripples out and affects other businesses more than if the local economy was less
diversified.

8
Due to the way the HHI is calculated, an increase in diversification is a decrease in the HHI.

Page 19
Table 4: Clustering Coefficient Result Summary

Clustering was measured using two specifications: LQ and PP. While the detailed results can be
found in table 2 and table 3, table 4 summarizes the clustering variable results in the elasticity
and volatility models. A “+” sign indicates that the LQ, PP, or both are statistically significant
and positively related to the corresponding dependent variable. A “-” sign indicates a statistically
significant and negative relationship with the corresponding dependent variable. A blank cell
indicates that there was no statistically significant relationship at a 90 percent confidence level. 9

9
A “+/-” or “-/+” indicates that the LQ and PP had different coefficient signs.

Page 20
Table 4 reveals several interesting patterns. First, utility industries behave counter intuitively
because clustering is negatively associated with sales tax revenue elasticity, and positively
related to both own-source and property tax revenue volatility. Clusters should (1) increase
growth due to the economic advantages they bring, and (2) decrease volatility due to global
supply advantages. However, the utility industry is structured differently than other industries.
For example, utility firms are often accompanied by strict land-use regulations to ensure public
safety. These regulations limit (to a degree) the number of retail firms or other industries with
foot traffic. A shopping mall would not be wise to locate next to a water treatment plant where
chlorine gas is a byproduct of the treatment process. Increases in property tax revenue volatility
(and subsequently own-source revenue volatility) probably occur by offsetting changes in
millage rates.

Often local governments own utilities, and some of the more entrepreneurial local governments
sell their excess supply on the open market. Revenue from utilities is subject to volatile
commodity costs, and thus, can be unpredictable. Therefore, offsetting this volatility with
adjustments in the millage rate is more likely to occur when local governments rely on utility
clusters for revenue.

To test the proposition that the uncharacteristic direction of the coefficients can be explained by
municipal ownership of utilities, Pearson correlation tests were run between Utility LQ (LQ and
not the PP was statistically significant for volatility), and a variable not included in the analysis:
Municipal Revenue from Utilities. The results in table 5 indicate a statistically significant and
negative correlation between Utility LQ and Utility Revenue. Increases in utility clustering are
associated with more revenue volatility—the opposite of what was expected. However, this
finding suggests a different explanation: that increases in utility clustering are the result of
increased private/non-profit utilities, and subsequently, more utility competition for local
governments.

Table 5: Pearson Correlations

Utility
Revenue
Utility LQ -0.19 ***
1. Signif. codes: ‘***’ 0.001

Unlike the utility clusters, the estimates for the 19 other industrial classifications show broad
patterns of significance that have policy significance for local government officials and scholars
alike. The first important pattern that emerges is that every statistically significant cluster in each
of the elasticity models is positively associated with the respective model’s dependent variable. 10
In other words, an increase in clustering is associated with an increase in own-source revenue,
sales tax revenue, and property tax revenue elasticity. Of the 19 clusters, only education services
is statistically significant across all three elasticity models. Information (sales tax and property
tax models), accommodation and food services (own-source and property tax models), and retail
trade (own-source and sales tax models) are statistically significant in two elasticity models.

10
Excluding the utility industry, which as previously discussed is an oddity.

Page 21
Only service-producing industrial clusters are consistently significant across two or more
models. Another important pattern emerges with the goods-producing industries. First, none of
the goods-producing industries affect own-source revenue elasticity. While industrial clusters
from goods-producing sectors impact individual revenue streams like sales taxes or property
taxes, they ultimately do not impact overall revenue elasticity. The shrinking importance of
goods-producing sectors in the national economy extends to local government revenue elasticity.

Second, all goods-producing sectors, except agriculture, forestry, fishing, and hunting, have
statistically significant (90% confidence interval) positive relationships with sales tax elasticity.
Since, tangible goods produced by these sectors are easily subject to a sales tax, local
governments have successfully captured the economic activity of these sectors into their sales tax
base.

Third, the goods-producing sector, agriculture, forestry, fishing, and hunting, has a statistically
significant and positive relationship with property tax elasticity. Since activities in this sector
require designated swaths of land set aside for economic purposes, it is easier to apply land-value
appraisals that reflect their economic value.

Another important pattern for public policy is the statistically significant cluster variables in each
of the three volatility models that have the same direction of signs for their estimated
coefficients, but not across all three models. Sales tax revenue volatility is positively associated
with eight statistically significant cluster industries. This finding suggests that clusters actually
make sales tax revenue less stable. Yet, property tax revenue volatility is negatively associated
with ten statistically significant industrial clusters. Unlike sales tax revenue volatility, an increase
in an industrial cluster is associated with a decrease in property tax volatility. Clusters stabilize
property tax revenue and destabilize sales tax revenue. The consistency of the coefficient sign
direction within each volatility model and across clusters suggests that this is not a data artifact
but a robust pattern.

Why do clusters increase sales tax revenue volatility and destabilize property tax revenue
volatility? The answer lies in the attributes of each tax base. The sales tax is levied on individual
economic transactions. Conversely, a property tax is a tax on wealth levied annually. Unlike
sales taxes, property tax rates are more easily adjusted (i.e., millage offsetting), and often used to
balance budgets in response to changes in the property tax base or shortfalls in other revenue
sources. But clusters interact differently with the sales tax base than with the property tax base.
Clusters lower supply costs and give the clustered industry a competitive global advantage. This
global advantage not only increases the industries revenue generation but also increases the
ability of firms to survive economic shocks thus minimizing the chances of business bankruptcy.
This ability to survive has a two-fold impact on property tax revenue stability. First, the lower
chance of bankruptcy and competitive advantages of locating in an industrial cluster minimize
the chances of firms relocating. On the margin, a business in a cluster will remain in their current
location, thus keeping a steady property tax base. Second, the decreased chance of clustered
firms relocating also decreases the chance of employees relocating. The productive advantages
of industrial clusters create a natural economic incentive for businesses to remain in their current
location.

Page 22
However, clusters do not stabilize sales tax revenues. For businesses, the primary draw of
participating in an industrial cluster is that, through the supply/demand advantages, their goods
and services will be more competitive in a larger market. This larger market can result from
increased centralization of foot traffic, similar to what occurs in shopping centers where
consumers can walk and shop, or the larger market can, in fact, be a global rather than regional
market. Clusters expand and externalize the market for goods and services produced in a
municipality. An internalized market has less growth potential, but is predictable and stable. As
the market for a region’s products expands, the aggregate transactions (i.e., the sales tax base)
become susceptible to a wider range of factors outside the control of the firm.

The third and final important pattern that emerges is one that is most important to local officials
and managers attempting to integrate economic development and simultaneously promote greater
municipal fiscal health. The literature on elasticity and volatility suggests a trade-off exits
between the two—growth leads to greater volatility. The analysis suggests that clustering in fact
improves own-source revenue/property tax revenue growth while simultaneously decreasing
volatility. Conversely, clustering improves sales tax growth, while simultaneously increasing
sales tax volatility.

Conclusion

This study investigated how different types of industries impact revenue elasticity and volatility
in a municipality. A municipality’s economic base is the foundation from which tax and fee
revenue is ultimately generated. No previous studies have explored the connection between a
municipality’s economic base and its various revenue streams. While local governments have
access to a wide variety of taxes, fees, service charges, and the income from enterprise activities,
this study examined the elasticity and volatility of three revenue streams: own-source revenue,
sales tax revenue, and property tax revenue. This study contributes to the current public finance
literature by providing a conceptual and empirical framework that can be used to develop our
understanding of a local government’s economic base.

Specifically, this study found that economic base diversity had no impact on own-source
revenue/sales tax revenue elasticity and volatility. However, the analysis indicated that
employment diversity was associated with increased property tax revenue elasticity, and
volatility suggesting that industrial diversity could be an effective strategy for local governments
(1) looking to grow and (2) heavily reliant on the property tax.

In addition, economic base clustering was associated with increased revenue elasticity for own-
source, sales, and property tax revenue. Economic base clustering was also associated with
decreasing own-source and property tax revenue volatility, but increasing sales tax revenue
volatility. These findings suggest that pursuing industrial clusters can be an effective way to
increase own-source, sales tax, and property tax revenue, but those local governments heavily
reliant on sales tax revenue will see decreases in stability. The analysis suggests that the
development of industrial clusters can be a particularly good strategy for improving municipal
fiscal health when (1) a local government relies mostly on property tax revenue, (2) limits utility

Page 23
clustering, and (3) heavily recruits businesses in educational services, accommodation and food
services, retail trade, and wholesale trade.

Our analysis suggests that clustering does the impossible for local governments concerned with
fiscal health—certain clusters can simultaneously promote revenue growth and stability when the
local government relies on the property tax. Of particular interest to development and public
finance scholars are the findings that industrial clustering increases growth in both property and
sales tax, but decreases volatility only for the property tax while increasing volatility in sales tax
revenue. Clusters create robust production regions where businesses are likely to thrive resulting
in great property tax stability. But clusters also connect the local economy to an increased
number of distant markets that ultimately decrease sales tax stability.

This study fills an important gap in public finance, local governance, and economic development
scholarship by presenting a multi-dimensional conceptualization of a municipalities economic
base, providing local government officials a framework for developing a cohesive
development/finance strategy to improve the fiscal health of the city, and illustrating the
importance of industrial clusters in public finance studies. While the findings are important,
more work needs to be conducted on the intersection of economic development and public
finance. Particularly, studies are needed that use national samples, rural cities, different measures
of diversification and clustering, and studies that explore longer time horizons. Researching this
area has the potential to increase the capacity of our local government’s ability to meet the
challenges of the future, and create better communities.

Page 24
 References

Afonso, W. B. 2013. “Diversification Toward Stability? The Effect of Local Sales Taxes on Own
Source Revenue.” Journal of Public Budgeting, Accounting and Financial Management
25(4): 649–674.
Alm, J., R. Buschman, and D. Sjoquist. 2011. “Rethinking Local Government Reliance on the
Property Tax.” Regional Science and Urban Economics 41(4): 320–331.
Amiti, M., and L. Cameron. 2007. “Economic Geography and Wages.” The Review of
Economics and Statistics 89(1): 15–29.
Andrews, D. 1991. “Heteroskedasticity and Autocorrelation Consistent Covariance Matrix
Estimation.” Econometrica: Journal of the Econometric Society 59(3): 817–858.
Andrews, D., and J. Monahan. 1992. “An Improved Heteroskedasticity and Autocorrelation
Consistent Covariance Matrix Estimator.” Econometrica: Journal of the Econometric
Society 60(4): 953–966.
Audretsch, D., and M. Feldman. 1996. “R&D Spillovers and the Geography of Innovation And
Production.” American Economic Review 86(3): 630–40.
Besley, T., and A. Case. 1992. “Incumbent Behavior: Vote Seeking, Tax Setting and Yardstick
Competition.” American Economic Review 85(1): 25–45.
Billings, S., and E. Johnson. 2012. “The Location Quotient as an Estimator of Industrial
Concentration.” Regional Science and Urban Economics 42(4): 642–647.
Bland, R. 2013. A Budgeting Guide for Local Government. 3rd ed. Washington, DC: ICMA
Press.
Bogart, W., and B. Cromwell. 1997. “How Much More is a Good School District Worth?”
National Tax Journal 50(2): 215–232.
Brealey, R., and S. Myers. 1991. Principles of Corporate Finance. New York, NY: McGraw-Hill.
Brueckner, J. 1998. “Testing for Strategic Interaction among Local Governments: The Case Of
Growth Controls.” Journal of Urban Economics 44(3): 438–67.
———. 2003. “Strategic Interaction among Governments: An Overview of Empirical Studies.”
International Regional Science Review 26(2): 175–88.
Buettner, T. 2001. “Local Business Taxation and Competition for Capital: The Choice of the Tax
Rate.” Regional Science and Urban Economics 31(2): 215–45.
Burchfield, M., H. Overman, D. Puga, and M. Turner. 2006. “Causes of Sprawl: A Portrait from
Space.” Quarterly Journal of Economics 121(2): 587–633.
Bureau of Labor Statistics. “Industry Employment and Output Projections to 2024.” December
2015, accessed January 10, 2017. https://www.bls.gov/opub/mlr/2015/article/industry-
employment-and-output-projections-to-2024.htm
Carroll, D. 2009. “Diversifying Municipal Government Revenue Structures: Fiscal Illusion or
Instability?” Public Budgeting and Finance 29(1): 27–48.

Page 25
Carroll, D., and C. Goodman. 2011. “The Effects of Assessment Quality on Revenue Volatility.”
Public Budgeting and Finance 31(1): 76–94.
———. 2013. “Understanding and Measuring Elasticity, Volatility, and Implications for Local
Government Fiscal Health.” Handbook of local government fiscal health. 225–50.
Carroll, D., and K. Stater. 2009. “Revenue Diversification in Nonprofit Organizations: Does it
Lead to Financial Stability.” Journal of Public Administration Research and Theory 19(3):
947–966.
Chinitz, B. 1961 “Contrasts in Agglomeration: New York and Pittsburgh.” American Economic
Review 51(2): 279–289.
Clark, D., and W. Herrin. 2000. “The Impact of Public School Attributes on Home Sale Prices in
California.” Growth and Change, 31(3), 385–407.
Costa, D., and M. Kahn. 2000. “Power Couples: Changes in the Locational Choice of the
College Educated, 1940–1990.” Quarterly Journal of Economics 115(4): 1287–315.
Doerner, W., and K. Ihlanfeldt. 2011. “House Prices and City Revenues.” Regional Science and
Urban Economics 41(4): 332–342.
Douglas, J., and R. Gaddie. 2002. “State Rainy Day Funds and Fiscal Crises: Rainy Day Funds
and the 1990–1991 Recession Revisited.” Public Budgeting and Finance 22(1), 19–30.
Felix, R. 2008. “The Growth and Volatility of State Tax Revenue Sources in the Tenth District.”
Economic Review—Federal Reserve Bank of Kansas City 93(3): 63–88.
Fischel, W. 2001. The Homevoter Hypothesis. Cambridge, MA: Harvard University Press.
Gamage, D. 2010. “Preventing State Budget Crises: Managing the Fiscal Volatility Problem.”
California Law Review 98(3): 749–812.
Geoghegan, J., L. Lynch., and S. Bucholtz. 2003. “Capitalization of Open Spaces into Housing
Values and the Residential Property Tax Revenue Impacts of Agricultural Easement
Programs.” Agricultural and Resource Economics Review 32(1): 33–45.
Goodman, C. 2016. “On the Measurement of Revenue Diversity.” Paper presented at the annual
conference for the Association of Public Budgeting and Finance (October 1–3).
Grizzle, C. 2010. “The Impact of Budget Stabilization Funds on State General Obligation Bond
Ratings.” Public Budgeting and Finance 30(2): 95–111.
Hendrick, R. 2006. “The Role of Slack in Local Government Finances.” Public Budgeting and
Finance 26(1): 14–46.
———. 2011. Managing the Fiscal Metropolis: The Financial Policies, Practices, and Health of
Suburban Municipalities. Washington, DC: Georgetown University Press.
Hendrick, R., and J. Crawford. 2014. “Municipal Fiscal Policy Space and Fiscal Structure: Tools
for Managing Spending Volatility.” Public Budgeting and Finance 34(3): 24–50.
Heyndels, B., and J. Vuchelen. 1998. “Tax Mimicking among Belgian Municipalities.” National
Tax Journal 51(1): 89–101.

Page 26
Hilber, C. and C. Mayer. 2009. “Why Do Households Without Children Support Local Public
Schools? Linking House Price Capitalization to School Spending.” Journal of Urban
Economics 65(1): 74–90.
Hill, E., and J. Brennan. 2000. “A Methodology for Identifying the Drivers of Industrial Clusters:
The Foundation of Regional Competitive Advantage.” Economic development
quarterly 14(1): 65–96.
Holcombe, R., and R. Sobel. 1996. “The Stability of International Coalitions in United Nations
Voting from 1946 to 1973.” Public Choice 86(1–2): 17–34.
Hou, Y. 2003. “What Stabilizes State General Fund Expenditures in Downturn Years—Budget
Stabilization Fund or General Fund Unreserved Undesignated Balance?” Public Budgeting
and Finance 23(3): 64–91.
———. 2005. “Fiscal Reserves and State Own-Source Expenditure in Downturn Years.” Public
Finance Review 33(1): 117–144.
Hou, Y., and D. Moynihan. 2008. “The Case for Countercyclical Fiscal Capacity.” Journal of
Public Administration and Research Theory 18(1): 139–159.
Ihlanfeldt, K. 2011. “How Do Cities and Counties Respond to Changes in Their Property Tax
Base?” The Review of Regional Studies 41(1): 27–48.
Ihlanfeldt, K., and K. Willardsen. 2014. “The Millage Rate Offset and Property Tax Revenue
Stability.” Regional Science and Urban Economics 46: 167–176.
Jackson, R., 2015. “Are Industry Clusters and Diversity Strange Bedfellows?” The Review of
Regional Studies 45(2): 113.
Jacobs, J. 1969. The Economy of Cities. New York, NY: Vintage.
Jordan, M. 2003. “Punctuations and Agendas: A New Look at Local Government Budget
Expenditures.” Journal of Policy Analysis and Management 22(3): 345–360.
Ladd, H. 1992. “Mimicking of Local Tax Burdens among Neighboring Counties.” Public
Finance Review 20(4): 450–67.
Lincoln Institute of Land Policy. “Fiscally Standardized Cities Database.” Accessed May 5,
2016, http://datatoolkits.lincolninst.edu/subcenters/fiscally-standardized-cities/.
Lutz, B. 2008. “The Connection between House Price Appreciation and Property Tax
Revenues.” National Tax Journal 61(3): 555–572.
Lutz, B., R. Molly, and H. Shan. 2011. “The Housing Crisis and State and Local Government
Tax Revenue: Five Channel.” Regional Science and Urban Economics 41(4): 306–319.
Marlowe, J. 2005. “Fiscal Slack and Counter‐ Cyclical Expenditure Stabilization: A First Look
at the Local Level.” Public Budgeting and Finance 25(3): 48–72.
Oates, W. 1969. “The Effects of Property Taxes and Local Public Spending on Property Values:
An Empirical Study of Tax Capitalization and the Tiebout Hypothesis.” The Journal of
Political Economy 77(6): 957.

Page 27
Overman, H., and D. Puga. 2010. “Labor Pooling as a Source of Agglomeration: An Empirical
Investigation.” In Agglomeration Economics, edited by Edward Glaeser. 133–150.
Chicago, IL: University of Chicago Press.
Overton, M., 2016. “Sorting Through the Determinants of Local Government Competition.” The
American Review of Public Administration, p.0275074016651143.
Pagano, M., and C. Hoene. 2010. “States and the Fiscal Policy Space of Cities.” In The Property
Tax and Local Autonomy, edited by Michael Bell, David Brunori, and Joan Youngman,
243–284. Cambridge, MA: Lincoln Institute of Land Policy.
Porter, M. 2000. “Location, Competition, and Economic Development: Local Clusters in a
Global Economy.” Economic development quarterly 14(1): 15–34.
Renski, H. 2011. “External Economies of Localization, Urbanization and Industrial
Diversity and New Firm Survival,” Papers in Regional Science 90(3): 473–502.
Ross, J. and W. Yan. 2013. “Fiscal Illusion from Property Reassessment? An Empirical Test of
the Residual View.” National Tax Journal 66(1): 7–32.
Shleifer, A. 1985. “A Theory of Yardstick Competition.” Rand Journal of Economics 16(3):
319–27.
Short, D. 2011. “Charting the Incredible Shift from Manufacturing to Services in America.”
Business Insider. September 5, 2011, accessed January 20, 2017,
http://www.businessinsider.com/charting-the-incredible-shift-from-manufacturing-to-
services-in-america-2011-9
Tannenwald, R. 2004. Are State and Local Revenue Systems Becoming Obsolete? Washington,
DC: National League of Cities.
Ulbrich, H. 1991. Nonproperty Taxes. In Local Government Finance, ed. J. Petersen and D.
Strachota, 113–33. Chicago: Government Finance Officers Association
Vlaicu, R. and A. Whalley. 2011. “Do Housing Bubbles Generate Fiscal Bubbles?” Public
Choice 149(1–2): 89–108.
Wang, W., and Y. Hou. 2012. “Do Local Governments Save and Spend Across Budget Cycles?
Evidence from North Carolina.” The American Review of Public Administration 42(2):
152–169.
White, F. 1983. “Trade-Off in Growth and Stability in State Taxes.” National Tax Journal 36(1):
103–114.
Wooldridge, J. 2010. Econometric Analysis of Cross Section and Panel Data. MIT press.
———. 2012. Introductory Econometrics: A Modern Approach. Cengage Learning.
Woodward, D., and P. Guimarães. 2009. “Porter’s Cluster Strategy and Industrial Targeting.” In
Targeting Regional Economic Development, edited by Stephan Goetz, Steven Deller, and
Thomas Harris, 68–83. New York, NY: Routledge.
Yan, W. 2011. “The Interactive Effect of Revenue Diversification and Economic Base on US
Local Government Revenue Stability.” Public Money and Management 31(6): 419–426.

Page 28