Fiscal impacts of land development a critique of methods and review of issues

[From the U.S. Government Printing Office, www.gpo.gov]

0 17

Coastal Zone
Information
center

FISCAL IMPACTS
OF LAND
DEVELOPMENT

A Critique of Methods
and Review of Issues

Thomas Muller

f7t

HJ
9105
.M78 THE URBAN INSTITUTE
1975

JUL 2 6 196

FISCAL IMPACTS
OF LAND
U Q DEPARTMENT OF COMMERCE NOAA DEVELOPMENT
COASIAL SERVICES CENTER
234 SOUTH HOBSON AVENUE A Critique of Methods
HARLESTON , SC 29405-2413
and Review of Issues

Thomas Muller
ft,01p,9gty of CSC Library

1\4
Of
op @04s@ \
?vk -Npl@-5 vixoc wo, 'k
5 00- xS0,

\0 ?(0
(\O PC

TIT--

C,4" LL4

The research for this report was made possible through a research
grant from the Office of Policy Development and Research of the
U.S. Department of Housing and Urban Development under the
provisions of Section 701(b) of the Housing Act of 1954, as
amended, to The Urban Institute. The publication of this report
was supported by the Ford Foundation. The findings and con-
clugions presented in this report do not represent official policy
of the Department of Housing and Urban Development, the Ford
Foundation, or The Urban Institute.

THE URBAN INSTITUTE

Library of Congress Catalog Card Number 75-03682

U.I. 176-214-3

ISBN 87766-137-5

REFER TO URI 98000 WHEN ORDERINGC..'e,,@:tj;@j Z;

A/75/5M

Available from:

Publications Office
The Urban Institute
2 100 M Street,'N.W.
Washington, D.C. 20037

List Price $2.95

Printed in the United States-of America

FOREWORD

This is the second in a series of reports by The Urban Institute's Land Use Center concerned with
evaluating the impacts of land development. The first report, Measuring Impacts of Land Development:
An Initial Approach, suggested a wide range of social, economic, and environmental concerns which
should routinely be assessed in the course of deciding among alternative land uses. It also described vari-
ous procedures for actually making impact measurements which might be used by local governments to
achieve a better understanding of land use consequences both before and after a developer's plans are given
the go ahead.
This report discusses in much greater depth and detail one of the most hotly debated questions of im-
pact measurement: how development affects the revenues and expenditures of local government. Will new
development inflate the demand for public services and facilities in excess of the tax revenues it generates?
Are the nation's communities on an unending quest for new development and new sources of revenue to
ease the tax burdens associated with developments already in place which do not pay their own way? Are
cities and suburbs, and suburban communities themselves pitted against each other in obtaining additional
tax dollars and excluding tax users?
For many years, it was assumed that almost any growth was good for the community and the local
fisc. But, this perspective is rapidly changing. The growing volume of fiscal impact or cost-revenue studies
alone, by both government and developers, attests to the fact that many communities are taking a second
look. The impact of development on the fiscal health of local governments is now a widely recognized cri-
terion for judging the acceptability of a land use proposal.
It is timely, therefore, to step back from the claims and counterclaims of developers, environmentalists,
no-growth or slow-growth advocates, and others and take stock of the state of the art which underlies their
bottomline calculations of fiscal impact. This report does that. A large number of fiscal studies were ex-
amined to determine why findings differ, to identify the major analytical issues, and to synthesize a judg-
ment about the utility of fiscal impact analysis as a tool for better land use planning and management.
Of course, land development impacts extend far beyond the question of public revenues and expendi-
tures. Subsequent reports will focus on private economic effects, the environment, and social concerns.
Sound public policy requires an ability to responsibly weigh the outcomes of different land uses in rela-
tion to community objectives which are frequently in conflict. Taken together, this series of reports should
help contribute to that end.

WORTH BATEMAN, Executive Director
Land Use Center
The Urban Institute

CONTENTS

Page
Foreword ------------------------------- ---------- -------- ---------------------------------------------------------------------------------- iii
Acknowledgments ------------------------------------------------------------------------------------------------------------------------ vii
Advisory Group ----------------------------------------- ---------------------------------------------------------------------------------- Viii
1. INTRODUCTION ---------------------------- ...................................................................................I
Overview ------------------------------------------ ----------------------------------- ...............................................1
Perspectives of Fiscal Studies-As Related to Sponsorship ................................................1
11. COST-REVENUE ANALYSIS-STATE OF THE ART _ ...............................................3
Study Categories ------------------------------------------------------------------------------------------------------------------3
Individual developments -------- _1 -----------------------------------------------------------------------------------3
Similar developments in several locations --------------------------------------------------------------------4
Communitywide growth studies ..................................................................................4
Regional studies ------------------------ ...................................................................................4
Public service perspective -------- -----------------------------------------------------------------------------------4
Data and Methodologies ------------------------------------- ................................................................4
Sources of Data ------------------------------------- ----------------------------------------------------------------------5
Household surveys ------------------ -----------------------------------------------------------------------------5
Review of detailed agency records ........................................................................5
Aggregate data from local, state, and federal agencies ------------------------------------------ 6
Basic Methods --------------------------------------------------------------------------------------------------------------6
Estimates by local officials ------------------------------------------------------------------------------------6
Community standards ..........................................................................................6
'Share allocation ----------------------------------------------------------------------------------------------------7
Accountant's approach ........................................................................................7
Methods Using Individual Household Data (Microlevel Analysis) ------------------------------ 7
Demographic patterns ------- ----------------------------------------------------------------------------------7
Income patterns _: --------------------------------------------------------------------------------------------------8
Other characteristics ----------- ..................................................................................9
Methods Using Aggregate Data (Macrolevel Analysis) ---------------------------------------------- 9
Cross-sectional analysis ----------------------------------------------------------------------------------------9
Time-series analysis .............................................................................................. 10
Econometric models ---------------------------------------------------------------------------------------------- 10
Fiscal flow analysis ------------------------------------------------- ---------------------------------------------- 11
Analysis of Physical Characteristics ------------------------------------------------------------------------------ 11
Alternative development patterns .......................................................................... 11
Characteristics of land ------------------------------------------------------------------------------------------ 12
Limitations of Alternative Techniques Discussed -------------------------------------------------------- 12
111. WHY FINDINGS DIFFER ------------- ----------------------------------------------------------------------- ---------- 15
Differences in Revenue Assumptions _. ----------------------------------------------------------------------- .......... 15
Choice of Allocation Approach -------------- ............................................................................. 16
State-Local Fiscal Structure ------------- --------------------------------------------------------- ------------------------ 18
Intra-Community Location ------------------_- --------------------------------------------------------------------------- 18
Spatial Scope of Analysis ------------------------------------------------------------------------------------------------------ 18
IV. FISCAL IMPACT ANALYSIS-MAJOR ISSUES ............................................................ 19
The Fiscal Effects of Changes in Population and Density .................................................... 19
Capital outlays -------------------------------------------------------------------------------------------------------------- 20
Operating outlays ---------------------------------------------------------------------------------------------------------- 22
Revenues ---------------------------------------------------------------------------------------------------------------------- 24
Impact of outmigration --------------------------------------------------------------------------------------- .......... 24
Notes of caution ------------------------------------------------------------------------------------------------------------ 25
Shifts in Preferences for Services and Taxes ------------------------------------------------------------------------ 25
Shift in service demand ------------------------------------------------------------------------------------------------ 26
Shift in revenue sources ------------------------------------------------------------------------------------------------ 26
Importance of preference shifts to jurisdictions ------ ----------------------------------------------------- 26

Page
Distributional Effects --------------------------------------------------------------------------------- -------------------------- 26
Changes in housing mix ------------------------------------------------------------------------------------------------ 27
Efficient location of households ------------------------------------------------------------------------------------ 27
Income distribution ---------- ------------------------------------------------------------------------------------------- 28
Secondary Impacts ----- ---------------------------------------------------------------------------------------------------------- 28
Changes in value of surrounding property -------------------------------------------------------------------- 28
Increased demand for goods and services resulting from residential development ---------- 29
Increased demand for housing and services to households as a
result of added industrial development -------------------------------------------------------- --------- 29
Increased demand for housing and services resulting from commercial development .... 30
Increased activity on the periphery of development ------------------------------------------------------ 30
Public sector capital investments ---- ----------------------------------------------------------------------------- 30
Time Horizon ---------------------------------------- ----------------------------------------------------------------------------- 31
Revenues ---------------------------------------- ----------------------------------------------------------------------------- 31
Expenditures -------------------------- --------------------------------------------------------------------------------- 31
Life cycle of household -------------------- ----------------------------------------------------------------------------- 32
Fiscal Interactions Among Jurisdictions __. ----------------------------------------------------------------------------- 32
Factors Which Limit the Scope of Fiscal Impact Studies ---------------------------- ----------------------- 33
Lack of conceptual framework ------------------------------------------------------------ ....................... 33
Insufficient analyses of empirical data ------------------------------------------------ ----------------------- 33
Shortages of resources -----------------I --------------------------------------------------------------------------------- 33
Sponsor objectives --------------------------- ---------------------------------------------------- ----------------------- 33
V. IMPLEMENTATION AND USE OF FISCAL IMPACT ANALYSIS .......... ................. 35
Guidelines for Initiating Studies -----_ ------- ---------------------------------------------------------------------------- 35
General approach -------------------------------------------------------------------------------- ----------------------- 35
When to implement an analysis of individual developments ---------------------------------------- 36
Nontechnical issues ------------------------------------------------------------------------------------------------------ 37
Present Uses and Potential Misuses -------------------------------------------------------------------------------------- 37
Alternative options for dealing with expectations of fiscal deficits ------------_---------------- 38
VI. SUMMARY-STUDY FINDINGS ------------------------------------------------------------------------------------ 41
Expanding Uses of Fiscal Impact Studies ---------------------------------------------------------------------------- 41
Methodologies-State of the Art ------------------------------------------------------------------------------------------ 41
Why Findings May Be Inconsistent --------- ............................................................................ 42
Fiscal Impact of Different Development Types -------------------------------------------------------------------- 42
Fiscal Impact Analysis-Looking to the Future .................................................................. 42
Effects of increasing population size and density ----------- -------------------------------------------- 42
Tax and service preferences ------------- ---------------------------------------------------------------------------- 42
Spatial and income distribution effects ---- _------------------------------------------------------------------ 43
Secondary impacts ------- ------------------- ---------------------------------------------------------------------------- 43
Areas for Further Study ------------------------------------------------------------------------------------------------------ 43
Industrial developments -------------------- --------------------------------------------------------------------------- 43
Pattern of development ----------------------- ------------------- ---------------------------------------------------- 43
Redevelopment projects ------------------- __ ------------ ------------------------------------------------------------ 43
Retrospective evaluation of new communities, PUDs, and large developments ------------ 43
Regional studies ---------------------------------------------------------------------------------------------------------- 44
Policy effects of fiscal impact studies ----_------------------- ------------------------------------------------- 44
Bibliography --------------------------------------------- ---------------------------------------------------------------------------------- 45
Appendixes
A. Measuring Fiscal Impact-A General Approach ------------------------------------------------------------ 51
Estimating net changes in government fiscal flow -------------------------------------------------- 51
Additional references ----------- ---------------------------------------------------------------------------------- 57
B. Attendees, Fiscal Cost-Revenue Analysis Conference, 1974 ---------------------------------------- 59

TABLES
1. Per Capita Outlays for Selected Municipal Functions, 1970-71 ............................................ 21
2. Outlays for Municipal Services as a Percentage of Income, 1969-70 ........................................ 25

ACKNOWLEDGMENTS

The Office of Policy Development and Research of the U.S. Department of Housing and Urban De-
velopment sponsored this study as one part of a broader research effort on state and local government eval-
uation of land development. The encouragement of Wyndham Clarke, Director, Division of Community
Planning, Development and Conservation, and the specific suggestions of his deputy, James Hoben, as well
as the review of the draft by Kenneth Kenney of the same division, are greatly appreciated.
A significant contribution to the study was made by participants of a small conference held in January
1974 in Washington, D.C., sponsored jointly by The Urban Institute and HUD. The written contributions
of Bruce Drenning, Barton-Aschman Associates; John P. Fowler, City of San Diego; Robert W. Glover,
Livingston and Blayney; Alan Mallach, New Jersey County and Municipal Study Commission; and Harold
Phipps, Northern Virginia Planning Commission; and the informal comments by other attendees of the con-
ference, are gratefully acknowledged. All participants are listed in Appendix B.
Worth Bateman, Executive Director of the Land Use Center and George E. Peterson, Philip S.
Schaenman, and Grace Dawson, all of The Urban Institute, provided valuable insights and comments.
Walter Rybeck edited the report and contributed substantially to its organization and content.-T.M.

vii

ADVISORY GROUP

The Advisory Group provided guidance for the overall project of which this report is one part. This report, however,
is the sole responsibility of the author.

TIMOTHY A. BARROW / Mayor, Phoenix, Arizona ROBERT H. PASLAY / Planning Director, Planning Commis-
KURT W. BAUER / Executive Director, Southeast Wisconsin sion, Nashville, Tennessee
Regional Planning Commission, Waukesha, Wisconsin RICHARD A. PERSICO / Executive Director, Adirondack
FRANK H. BEAL / Deputy Director for Research, American Park Agency, Ray Brook, New York
Society of Planning Officials, Chicago, Illinois JAMES R. REID / Director, Office of Comprehensive Planning,
MELVIN L. BERGHEIM / Councilman, Alexandria, Virginia, Fairfax County, Virginia
and National League of Cities-U.S. Conference of Mayors E. JACK SCHOOP / Chief Planner, California Coastal Zone
Conservation Commission, San Francisco, California
RICHARD F. COUNTS / Zoning Administrator, Planning De-
partment, Phoenix, Arizona DUANE L. SEARLES / Special Counsel on Growth and En-
vironment, National Association of Home Builders, Wash-
CARL D. GOSLINE / Director of General Planning, East ington, D.C.
Central Florida Regional Planning Council, Winter Park,
Florida PHILIP A. STEDFAST / Planning Director, Department of
City Planning, Norfolk, Virginia
BERNARD D. GROSS / Glover Park Citizens Association, and
New Communities Program, HUD, Washington, D.C. DAVID L. TALBOTT / Director of Planning, Falls Church,
Virginia
HARRY P. HATRY / Director, State and Local Government RICHARD E. TUSTIAN / Director of Planning, Maryland
Research Program, The Urban Institute, Washington, D.C. National Capital Parks and Planning Commission, Silver
TED KOLDERIE / Executive Director, Citizens League, Min- Spring, Maryland
neapolis, Minnesota F. ROSS VOGELGESANG / Director, Division of Planning
F. DENVER LINDLEY / Commissioner, Bucks County, and Zoning, Indianapolis, Indiana
Doylestown, Pennsylvania THORNTON K. WARE / Planning Director, Rensselaer
JACK LINVILLE, JR. / Assistant Executive Director, Ameri- County, Troy, New York
can Institute of Planners, Washington, D.C. JOSEPH S. WHOLEY / Chairman, Arlington County Board,
Arlington, Virginia, and Director, Program Evaluation
ALAN H. MAGAZINE / Supervisor, Fairfax County Board, Studies, The Urban Institute, Washington, D.C.
Fairfax, Virginia, and Project Director, Contract Research
Center, International City Management Association, Wash- FRANKLIN C. WOOD/ Executive Director, Bucks County
ington, D.C. Planning Commission, Doylestown, Pennsylvania

viii

1. INTRODUCTION

OVERVIEW penditures associated with new development have con-
The purpose of fiscal impact studies is to determine centrated on estimation and computational tools. The
whether developments will generate enough new taxes state of knowledge, therefore, is most advanced in this
to pay for the added public services they will require. area. Less attention has been given to determination of
This is not merely an accounting problem since future the proper scope of an analysis. Which approach or
tax levels can affect public services and hence the combination of approaches is most appropriate for a
character of the community. The growing interest in the given circumstance? Should economies or diseconornies
use of fiscal analysis to evaluate the impact of land of scale, for example, be incorporated into an analysis?
development projects on local government reflects the These and other issues are discussed and guidance given
concern that some land uses may have adverse fiscal to those sponsoring or reviewing work in the fiscal im-
effects. During 1973 and early 1974, the volume of pact area. The attempt is to show the best of current
fiscal studies exceeded all such analyses identified in practice while indicating some of the gaps or defects that
the previous five years (see Bibliography). Increas- require the special attention of analysts.
ingly, communities are also undertaking studies of al- The emphasis of this report on fiscal impact is not
ternative future growth policies, and assessing fiscal intended to imply that these are more important than
effects as part of the evaluation of available policy other types of impacts. Sometimes they may be; at
options. In addition to local interest in this subject, other times they clearly should be accorded a low
some state governments, such as Florida and Vermont, priority, for reasons discussed later in connection with
have adopted land use or environmental legislation re- the uses and misuses of fiscal analyses. Yet at no time
quiring that fiscal effects be considered for certain kinds can a community safely disregard entirely these fiscal
of projects. impacts on the local treasury and thus on the local
These expanding uses of fiscal studies make it ad- government's ability to provide services.
visable to take stock of the present state of the art in Detailed procedures are not specified because a num-
this field. How can local and other governments use ber of previous studies, cited in this report, illustrate
fiscal studies to best advantage? What factors cause computation methods and data formats that may be
results to differ? How can local officials avoid confu- followed. However, Appendix A outlines a general ap-
sion as contradictory impact findings are cited by proach that may serve as a guide to those who are
groups proposing and opposing new development? Pre- initiating fiscal impact analyses of individual develop-
liminary answers to these and other questions are of- ments.
fered in this report as it tries to present an objective PERSPECTIVES OF FISCAL STUDIES-
perspective of the strengths and weaknesses of the AS RELATED TO SPONSORSHIP
many recent studies that have focused on the fiscal
aspects of land use change. The major objectives of those sponsoring studies
Past attempts at estimating the revenues and ex- vary according to their perspectives. Some want to win

support for proposed developments, others want to The overall interest of local jurisdictions is to de-
block or modify developments, and still others want to termine the effect of new development on the welfare
provide local decision makers with information on the of their residents. Local governments, often through
economic effects of development, to create more knowl- the planning or zoning boards, are frequently faced
edge about the intricacies of growth, or to aid in the with the need to evaluate, in the context of broad eco-
formulation of new local, regional, and national growth nomic, environmental, and social issues, the results
policies. of privately sponsored cost-revenue studies submitted
The objectives of study sponsors determine, to a to them. Because of their need to respond to specific
considerable extent, the scope and approach of analyses proposals, local governments look to fiscal impact
undertaken. Sponsors can be grouped into private in- studies which develop general methodologies to aid
terests (primarily representing developers), antidevel- them in the assessment of reports prepared by private
opinent groups, nonprofit organizations, and the pub- groups. Communities may also carry out their own
lic sector, which may include local governments and impact analyses of individual developments or sponsor
regional as well as federal agencies. studies aimed at estimating aggregate fiscal and other
Privately sponsored cost-revenue studies of pro- changes due to larger community growth patterns.
posed development are most often initiated to convince These uses of impact studies reflect the dual role of
a public body to grant approval for a project. The the local decision maker: to consider the short-term
results of such studies are presented to counteract the fiscal implications of individual developments, such as
concern expressed by communities that the develop- their impact on tax rates and on, service demands, and
ments in question may cause an increase in the tax to consider the long-term implications of growth as
burdens of its present residents. these relate to implicit or explicit goals of the com-
Those opposed to new development, for either en- munity. The first role relates to existing legal constraints
vironmental or other reasons, may sponsor studies to and institutional arrangements which require a specific
show that, in addition to other problems, a project is response to each application for a land use change,
not fiscally beneficial@ They tend to emphasize what with emphasis on the immediate impact of a decision
economists refer to as "negative externalities"; that is, on the community. This short-term perspective also re-
impacts of a development which will adversely affect flects the view of those residents who are more con-
residents outside its immediate boundaries, such as cerned with the likely effect of a new development on
traffic congestion and overcrowding of public facil- the tax rate in the next year or two than with the long-
ities. term fiscal effects of a development on the community.
Nonprofit organizations such as universities are pri- Estimating the gross short-term effects may not require
marily concerned with advancing the state of the art, a comprehensive analysis. The second role involves an
improving methodologies to refine estimation proce- understanding of more complex issues, such as scale
dures, and examining policy implications of findings. economies, the effects of constraining development on
The public sector has a somewhat different role. fiscal flows, and shifts in service preferences as a result
Bodies representing local governments at the regional of inmigration.
level are particularly interested in fiical flows among
jurisdictions, economies or diseconomies of scale asso-
ciated with consolidation of services regionally, the The current state of the art is discussed in Chap-
differential fiscal impact of growth among communities, ter 2. It reviews studies directly or indirectly linked to
and the most efficient location of new households cost-revenue analysis and discusses methodologies ap-
within the region. plied to estimate fiscal effects. Chapter 3 notes factors
The federal government has two major objectives which influence study results.
in sponsoring fiscal impact analysis studies: first, to Chapter 4 presents a discussion that is more closely
provide information on the fiscal effects of certain fed- linked to the frontiers of the state of the art of fiscal
eral programs, particularly in the areas of housing and impact analysis. It takes into account the long-term and
economic development, and second, to develop general broader fiscal effects of growth. Chapter 5 provides
methodologies and guidelines for use by local govern- guidelines for implementing studies and notes current
ments which do not have the resources to sponsor this uses. Chapter 6 summarizes study findings and sug-
kind of research. gests areas for further research.

2 Fiscal Impacts of Land Development

11. COST-REVENUE ANALYSIS
STATE OF THE ART

Cost-revenue or fiscal impact studies range in scope Individual Developments
from analyses of individual developments which con- The type of land use change acted on most fre-
Sider only revenues and operating outlays to studies quently by local officials is the request for public ap-
which examine the fiscal, effects of alternative growth proval of an individual development.
policies in large urban areas. Taken narrowly, cost- Many private studies related to such proposals, par-
revenue studies deal with computing the amount of ticularly those involving relatively small developments,
revenue contributed directly by a new development, and have so narrow a focus and contribute so little to fiscal
the cost of providing services directly linked to the new methodologies that they have been generally excluded
development. . from this survey.
A few fiscal impact studies, usually sponsored by In addition to private efforts, a number of studies
public agencies, also explicitly or implicitly incorporate sponsored by the Department of Housing and Urban
one or more of the following considerations: Development and nonprofit organizations have been
* Secondary effects, such as the impact of inmigra- completed recently which are aimed at aiding local gov-
tion attracted by new industrial development. ernments in the evaluation of predominantly residential
* The likely economies or diseconornies of scale as- developments.
sociated with increasing the number of units Nonresidential developments are usually not the sub-
(households and business enterprises) serviced. ject of privately sponsored impact analyses directed to
9 Preferences of new residents for public services local governments since, in terms of direct impact, they
which may differ from those of the base popula- are likely to show a fiscal surplus and, thus, their fiscal
tion. effects are not a major concern to local government.'
Studies of large, planned, multi-use developments-
* Distributional fiscal effects. particularly PUDs (Planned Unit Developments) and
new communities-form the majority of individual de-
STUDY CATEGORIES velopment studies examined in this review. New com-
The fiscal studies examined are of two broad types, munities and PUDs are of considerable interest to both
those concerned specifically with the cost-revenue im-
plications of proposed or expected development and 1. Among exceptions are the state of Florida, where large
commercial and other developments defined as having regional
those which examine the demand for public services as impact have to file applications which specify their likely eco-
a function of population characteristics. The studies are nomic and environmental effect on the region. In addition,
also grouped into categories which reflect their geo- studies may be sponsored by developers in certain communities
sensitive to new growth such as Montgomery County, Mary-
graphic scope, starting with individual developments land, where opposition to nonresidential development focuses
and leading up to regional growth patterns. on environmental rather than economic concerns.

local governments and regional associations, since their rates are examined in a number of studies as aids in
potential population and economic activity frequently formulating land use policies for the community. In
exceed that of the existing community and in any event addition, alternative patterns of growth (high and low
are likely to result in significant fiscal changes. These population densities) are compared in some studies to
studies have been sponsored by local and regional as determine how sensitive fiscal results are to these de-
well as federal agencies, most notably by HUD. In velopment patterns. In most cases, it is assumed that
addition, developers have sponsored such studies .2 historical patterns will continue and that the socio-
In some instances, comparative fiscal analyses of economic and demographic characteristics of new resi-
alternative development patterns on the same site are dents will follow past trends. As a result, the effects
included. Specifically, differences between planned of differences in characteristics of new residents on the
growth, sprawl development, or no development are fiscal balance are frequently not explicitly considered.
considered in a number of cost-revenue studies. This is one of the limitations of many communitywide
studies reviewed.
Similar Developments in Several Locations
From a research perspective, the study of one large Regional Studies
building or a few structures provides little insight into a A small group of studies examine fiscal impact from
more general fiscal pattern associated with a particular a regional perspective. Their objective is to compare
form of development. For example, a finding that one fiscal characteristics of communities within a region by
highrise apartment building or one industry produces a size, density, and type of land use and identify causes
fiscal surplus may not be representative of other apart- for differences in revenue and expenditure patterns. In
ments or firms in the same community. A number of addition, fiscal flows among the communities may be
studies are now being undertaken which examine the computed and causes for household locational patterns
fiscal impact of similar land uses in several locations. identified [33]. However, these studies have not con-
Both San Francisco, California and Arlington, Virginia sidered techniques to determine, from the public sector
have undertaken studies to examine the fiscal and fiscal perspective, the most efficient location of new
other impacts of predominantly commercial highrise households within the region.3
developments already completed in these communities
[20, 21].* The fiscal impact of apartments and town- Public Service Perspective
houses in a large number of New Jersey communities A number of studies examine the demand for local
and the fiscal impact of new industrial facilities in a public services as a function of population and other
number of small Kentucky and Oklahoma communities community characteristics. Most of these studies are not
have also been the subject of recent studies [16, 22]. directly aimed at evaluating new development and do
Communitywide Growth Studies not fit the geographic categories previously cited. How-
ever, since their objective is to determine how the de-
The third major group of studies focuses on total mand for and the cost of public services varies as a
communities and has as its primary objective estimat- function of community size, density, and location, they
ing the collective fiscal impact of development on an are directly relevant to the broader scope of fiscal
area. Since the objectives of a community are more impact analyses as defined in this report. Many of these
than economic, communitywide studies funded by local studies appear in professional economic journals and
governments tend to incorporate environmental and related publications. They include econometric models
other concerns. The fiscal parts of these studies are for projecting the cost of public services and studies of
frequently aimed at responding to the question: Does the fiscal impact of annexation or consolidation.
growth collectively pay its way or do existing residents
subsidize new residents? The general approach is com- DATA AND METHODOLOGIES
parative; the fiscal and nonfiscal implications of no Various approaches are used to estimate costs and
growth, low-level growth rates, or high-level growth
revenues associated with new development. Most fiscal
2. Studies of multi-use developments, particularly those spon- studies of residential developments consider the impact.
sored by public agencies, frequently include nonfiscal consider- of a projected number of housing units with site plans
ations-such as the impact of large-scale projects on housing showing their location, configuration, and density. From
for various income groups, employment, and the physical en-
vironment. However, the focus of this study is on the fiscal 3. The Urban Institute housing model appears to be a use-
parts of these projects. ful tool for examining the impact of new housing on the exist-
Numbers in brackets refer to studies cited in Bibliography. ing housing stock in a region. See: [641.

4 Fiscal Impacts of Land Development

this and other information, the likely characteristics of residents compared to the larger households residing
future households are estimated. These data, in turn, in detached housing. A survey of households in Fairfax
are used to project the additional consumption of public County, Virginia, was aimed at relating the consump-
services and the cost of providing these services. tion of educational services as a function of such factors
Another group of studies examines the cost of services as housing type, occupation, and education of house-
for the same number of housing units given alternative hold members and their length of residence in the
physical patterns of development on a specific site. county., [8 11.
Comparatively little effort is taken in either group of School children or their parents are frequently sur-
studies to estimate future revenues. Frequently, only veyed by local governments, as in Prince George's
property taxes are computed directly, with estimates County, Maryland, to determine the number of students
based on property values projected for the new devel- from various housing types, including detached housing
opment. Other revenue estimates are obtained by as- units and apartments [27].
surning that new residents will pay the same taxes, on The major limitation of personal interviews is the
a per capita basis, as present residents. high cost. Mailed questionnaires tend to have a low
. We will look first at the sources of data gathered for response rate and are frequently unrepresentative. The
the studies and then at alternative methods of analyzing major advantage of personal interviews is less chance
the data and applying the results. In addition to dis- of misinterpreting questions. Telephone interviews often
cussing techniques and their limitations, the more inter- represent a satisfactory compromise between the in-
esting study findings will also be noted; however, a sufficient response rate from mail surveys and the high
comprehensive evaluation of results is beyond the scope cost of personal interviews. The resistence of some
of this paper. Although only one approach may be citizens to questions about their education, employ-
noted in a discussion of a specific study, the studies ment or income can limit household surveys [81].
almost invariably use multiple means to gather and An alternative approach to obtain individual house-
analyze relevant data. The methods discussed can be hold data on residents of new developments is to survey
utilized in studies of new development as well as those mortgage applications filed with financial institutions
which more generally examine public service demand. and rental applications filed with managers of rental
units. This approach was applied to estimate the fiscal
Sources of Data impact of a large PUD by obtaining data on income,
Data sources for fiscal impact studies include house- age distribution and size of household for both owners
hold surveys, local government agency records, and and renters [13]. The advantages of this data source
more aggregate data collected by local, state and federal are the low cost and the reliability of information, par-
agencies. ticularly on income. However, most developers and
financial institutions consider such data as confidential,
Household Surveys limiting access to this information source.
One means of obtaining micro-level information Review of Detailed Agency Records
about the usage of services, not simply generalized aver- Data on service demand or tax payments on an in-
ages, but data that reveal variation according to the dividuat household basis can be obtained by examining
type of household or firm, is through direct surveys records maintained for specific service functions by most
conducted by mail, telephone, or personal interview. agencies of local governments. For example, future pub-
A sample of local residents drawn by the use of ac- lic safety expenditures in a partially completed new
cepted statistical procedures is selected to represent the community were estimated using crime rates for already
many factors which may explain differences in taxes occupied units [9]. It was found that the frequency of
paid and service consumption, such as income, housing crime in these initial developments approximated the
type, and location within the community. An example community average. Similarly, the frequency of fire by
of this approach was a mailed questionnaire used in housing types was obtained by tabulating the logs main-
Montgomery County, Maryland, as part of a study to tained at fire stations by address [271.4 In a number of
determine the fiscal impact of various economic activi- studies, property taxes paid by a particular building or
ties as a function of income, occupation and housing group of housing units have been determined by ex-
type [61 ]. The study shows little difference between types
of housing in the frequency of service calls for police 4. Service costs may not vary in direct proportion to fre-
quency of use. For example, the cost of maintaining a fire
and fire protection. Yet, the data illustrate that, on a per station may not vary significantly whether it handles a great
capita basis, service usage is greater for apartment many or very few service calls.

Cost-Revenue Analysis-State of the Art 5

amining assessment records, such as are maintained by or analysis. These techniques have the advantage of
most communities. The advantage of this method is low cost, but are frequently unreliable in predicting
that the actual service consumption level for a geo- service demand.
graphic area or type of housing unit can be obtained
without the necessity of involving the recipients directly. Estimates by Local Officials
The major drawback is that it is sometimes difficult to One recent study suggests estimating expenditures by
obtain access to data, especially records which are not asking those responsible for the provision of specific
in the public domain. services what changes in demand they would project
as a result of a specific development [2]. For example,
Aggregate Data from Local, State, and the police chief and fire chief are asked what addi-
Federal Agencies tional personnel and equipment they will require as a
Some communities aggregate detailed records of result of new development. A drawback of this ap-
services, with data on the level and the unit cost of proach is that estimates may reflect short-term esti-
services, and make them available to residents as part mates rather than long-term incremental expenditures.
of annual budgets or other reports. The Richmond, It is difficult to assess to what degree the level and
Virginia annual budget, for example, shows the unit quality of services to a community may change as a
cost of providing almost all local services, such as the result of the projected manpower assignments. If a
average cost of responding to fire alarms, conducting community has a tight budget, no personnel may be
electrical inspections of buildings, and investigating added as a result of the new development. Presumably,
major crimes for the current year, with cost estimates however, the distribution of existing personnel to pro-
provided for future years. Other communities issue an- vide services to new areas would have detrimental
nual reports for various departments which show the effects on older development.
level of services provided to residential housing, com- This use of estimates by local officials and depart-
mercial, and industrial facilities. The 1970 Census of mcnt heads to project future service costs has been
Population and Housing is a frequently used source of shown to be extremely inaccurate in a retrospective
information relevant to cost-revenue analysis. For ex- analysis of the fiscal impact of a large development
ample, the socioeconomic and demographic character- [10]. Estimated costs, even taking inflationary pressures
istics of inmigrants to a region and some communities into account, sharply underestimated the increase in the
can be identified. These data can be used to estimate community budget.
tax revenues from new residents and their likely de- The reliability of local agency estimates for a large
mand for services such as education. The data which area has also been examined as part of a study of the
relate house value and rent payments to income in fiscal impact of annexation. Additional personnel were
areas (census tracts) of new development are useful in projected by responsible departments for all major city
estimating the income of future residents, if the market services prior to annexation. Several years following
value of projected housing units is known. The Census the annexation, these values were compared to the
of Governments issues data on asses sed-to-market value actual number of personnel added [721. The data
ratios for property by type of land use every five years, showed that, for some services, the estimates varied
which are needed to estimate property taxes. Data on widely from the actual change. The additional number
public expenditures and revenues by function for all of personnel hired appeared to be strongly influenced
large cities by the Census, grouped by population size, by immediate budget considerations rather than by pro-
are helpful in comparing changes in outlays and rev- jected shifts in service demand.
enues as population increases.
These and similar aggregate data sources have sev- Community Standards
eral limitations. Unless the data are updated frequently, Another rudimentary means of estimating expendi-
the information becomes obsolete. With the exception tures likely to result from new developments is for a
of data on school attendance, little information is pro- community to develop and apply service standards. For
vided on service consumption as a function of house- example, a number of communities have established
hold characteristics. per capita standards, such as a certain number of police
and fire personnel per 1,000 population, specified acres
Basic Methods of open space per 10,000 residents, a minimum number
A number of techniques have been applied in cost- of library personnel and books per 1,000 residents, and
revenue studies which require minimal data gathering so many hospital beds per 1,000 residents. A review of

6 Fiscal Impacts of Land Development

standards for various services and suggested standards the premise that if servicing new residential growth
for a suburban community are noted in two recent requires additional city personnel or equipment, such
growth studies [30, 36]. The number of employees per additions would be spelled out in the budget. In the
capita has been used to estimate public costs in other absence of such ' identified outlays, the increments in
fiscal impact reports. expenditures for the city are assumed not to be asso-
A major limitation of this approach as it is currently ciated with growth. The use of this approach resulted
applied is its input orientation. Differential needs as- in showing that the incremental cost of new housing is
sociated with the characteristics of new population are relatively small, since the San Diego budget did not
not taken into account. Since the use of community directly identify all costs associated with new housing.
standards is essentially a per capita approach, it as-
sumes that demand is not sensitive to the income levels Methods Using Individual Household
and other socioeconomic characteristics of new resi- Data (Microlevel Analysis)
dents, to revenue receipts, or to the addition of indus- The individual household analysis approach empba-
trial and commercial property. Because of demand sizes variations in the characteristics of the population,
shifts caused by these factors, service quality projec- particularly between residents of new development and
tions may be distorted by applying per capita standards, the balance of the population. Typically, the house-
Share Allocation hold size, age distribution, and income of new residents
are estimated on the basis of projected housing charac-
A technique similar to the community standards ap- teristics such as number of bedrooms and cost per unit.
proach is to allocate additional expenditures for the The anticipated demand and consumption for public
anticipated new popul 'ation according to the present per services, once household characteristics are projected,
capita cost of services in the community. This was one should be based on- data obtained from surveys of re-
basis of allocating the cost of noneducational local cently completed developments. Service consumption,
functions for a PUD in New Jersey [10]. Since 45 per- in turn, is readily translated into local expenditures.
cent of the population increase over a given time period This form of analysis is applicable to studies of both
was attributable to the PUD, 45 percent of the addi- individual developments and communities.
tional cost for services was assigned to the development.
The major drawbacks of this essentially per capita Demographic Patterns
approach closely parallel those involved in the use of A strong relationship exists between demographic
community standards: the approach does not distin- characteristics of new residents and the type and size
guish differences between new and old residents, or of housing units provided in a new development. The
between new residents living in various types of devel- size of a household and the expected number of school-
opment. The share allocation technique has some vir- age children can be estimated using data provided
tues when it is used for comparative purposes, but it on the type of housing unit and the number of bed-
remains unsatisfactory as the basis for projecting the rooms to be constructed. The relationship between the
costs of new growth. number of school children and type of housing (de-
Accountant's Approach tached single-family, townhouse, garden apartment,
highrise apartment) has been established in a number
In applying the "accountant's approach," annual of studies. These efforts show the relationship between
budgets are examined for year-to-year expenditure the number of bedrooms per dwelling unit and school-
changes by function that can be directly attributed to age children. However, as the subsequent discussion in-
population growth. The major limitation of this ap- dicates, the development pattern can influence demo-
proach is that accounting procedures are not aimed at graphic patterns. It has also been shown that, in general,
specifying incremental costs. In most cases, at least residents of new detached housing units have more
some operating costs associated with new development school-age children than the community average. Since
are aggregated with service costs for existing residents public education is usually the largest single local out-
in municipal budgets, so that development-related in- lay, comprising 60 to 80 percent of local budgets in
creases are difficult to identify. many suburban areas, the number of school-age children
In a cost-revenue analysis of new housing develop- has a considerable impact on fiscal flows. Demographic
ments in San Diego, a comparison of the city's budget data are also needed to project service demands for
for two years was used in projecting fiscal impact [24]. noneducational services.
This applied the accountant's approach, based on Data for demographic analysis are frequently derived

Cost-Revenue Analysis-State of the Art ' 7

from previous studies; in some cases, communitywide Income Patterns
averages from the Census of Population or local sources
have been used. However, for purposes of estimating Since almost all major local revenues derived from
the impact of new development, demographic charac- households are income-elastic (rising or falling as in-
teristics need to be estimated from surveys of similar comes rise or fall), data on the income of new residents
developments. Unfortunately, the use of values derived are critical in estimating taxes from new residential
from other communities may be misleading. For ex- development [761.
ample, a survey of Reston, a new community in Fairfax The importance of income on the level of services
County, Virginia, shows 1.6 school children per de- provided to residents of a new development depends
tached housing unit, compared to less than 1.2 in the greatly on the ability of the income groups in question
balance of Faiifax County for similar housing [9]. to influence allocation of services. For example, a new
Another new community, Columbia, Maryland, has high-income development in a generally moderate-in-
about 1.2 scbool-age children from each detached hous- come community may demand and receive services ex-
ing unit or more than 20 percent below Reston's aver- ceeding the jurisdiction average. If residents of a new
age [661. The demographic profile of PUDs (Planned development represent a dominant proportion of a
Unit Developments) in New Jersey indicates an average community's population, there is little doubt that their
household size of 4.1 with 1.9 children for single-family preferences for services will be reflected in the budg-
households [101. The proportion of school-age children etary and allocation process because of their political
in new communities and PUDs tends to be higher than clout.
the proportion found in similar housing within conven- Income analysis was one method used in a case study
tional developments. of Albermarle County, Virginia, where the demand for
Most studies find that the number of school-age selected services was estimated based on the income
children in suburban housing is closely related to the anticipated for residents of the new development as
number of bedrooms in each unit [27]. In the absence compared with incomes of residents in the balance of
of other data, the bedroom count can be used as a the community [5]. For example, it was suggested that
crude proxy for household size. 5 the demand for library service and recreation would be
Data on the number of school-age children likely to above the community per capita average, while the
attend elementary and secondary grades need to be demand for health and welfare service would be below
adjusted to reflect non-public school attendance. Among the average. Projections also took account of the ten-
the factors which influence private school enrollment are dency of affluent communities to demand higher-than-
the perceived quality of the public school system and the average public school services. The previously cited
religious affiliation of residents. Household income does study in Montgomery County, Maryland incorporated
not appear to be a significant variable in determining survey data on the demand for public services as a
public school attendance in a suburban community [61 ]. function of income and housing type [611. This study
Demographic analysis is also a useful tool in estimat- showed that demand for public tennis courts, golf
ing the change in demand for noneducational public courses, libraries, recreation centers, and public park-
services since consumption of health care, libraries and ing lots increased with rising income. For example,
recreation is concentrated among certain age groups. households with income under $8,000 used tennis
For example, a survey in Fairfax County, Virginia shows courts, golf courses and libraries only half as frequently
that over 80 percent of persons using selected recrea- as those earning over $15,000. However, the use of
tional facilities are between the ages of 5 and 34 al- police, fire, and health services was not related to house-
though this age group comprises less than 60 percent hold income. Cross-sectional studies also support the
of the total population [68]. view that many public expenditure outlays are income-
Changes in the demographic profile of residents as related [351.
housing units in a neighborhood age also may prove A community's future aggregate expenditure levels
significant, but such changes have not been identified in are likely to be closely linked to revenues which can
the studies reviewed. be anticipated. In estimating communitywide effects of
growth, it is useful to compute the share of personal
5. Among exceptions are new residents of subsidized two- income which is collected by local government to pay
bedroom rental housing who tend to have the same number of for its services. If real income is projected to increase,
school-age children as larger nonsubsidized detached housing the level or quality of services can be increased without
units [27, 361. Housing built to accommodate a particular
group, such as singles and the elderly obviously do not follow changing the tax rate. If the same level of services is
this pattern. maintained, tax rates can be reduced.

8 Fiscal Impacts of Land Development

The income level of inmigrating households is likely addition, differences in income and land use should be
to be somewhat above the existing population average. considered.
As shown in a recent study, higher-income residents in An example of the use of this technique is a recently
both a central city and all its suburbs contribute sub- completed study by the Rutgers University Center for
stantially more revenue than they receive in services Urban Policy Research [54]. It uses regression analysis
[33]. As a result, the ability to attract higher-income to examine expenditure patterns in New Jersey juris-
residents influences all future fiscal projections. Growth dictions for all major local services between 1960 and
policies, in turn, influence the'likely income level of 1970. Communities are grouped by their population
future households. size and rate of growth, and aggregate expenditure data
are compared to determine how they vary among these
Other Characteristics groups. Since the study includes almost all municipali-
Population characteristics other than age and income ties in the state, both the methodology and the results
which affect service demand and consumption include are of interest. The authors state that previous research
educational level, previous place of residence, and had shown that community growth usually resulted in a
ethnic background of inmigrants. For example, moder- lower level of expenditures per capita when compared
ately sized communities, such as Palo Alto, California with expenditure levels of relatively stable communi-
and Ann Arbor, Michigan have higher per pupil ex- ties.", However, this study finds that growing cities
penditures than communities of similar size and income spend proportionately more per capita for municipal
where the average educational attainment of residents services than cities with more stable populations. The
is lower. New Jersey study shows further that both cities gaining
It is likely that persons whose previous place of and losing in population have rising costs for public
residence was a small town or farm will demand fewer services, and that cities with declining populations have
public services than those who previously resided in an even more rapid rise in the unit cost of public
metropolitan areas. There also appear to be regional services than cities which are growing.'
differences in service demand which reflect, in part, the New Jersey data were also used in a cross-sectional
ethnic composition and cultural values of residents. fiscal impact study of a new community [3]. In this
Information on such differences is limited, and it study, communities in the state are ranked by annual
often may not be cost-effective for a community to growth rates and population size. Costs for specific
incorporate these factors in their fiscal calculations. services are then estimated as a function of population
size. Cross-sectional analysis was also the technique
Methods Using Aggregate Data used to project expenditures for public services in
(Macrolevel Analysis) Michigan cities with a population of over 10,000 resi-
Aggregate or macro data analysis is most useful at dents [1]. These studies, with far less clear-cut results,
the community or subcommunity level rather than in show that even when population size and density are
examining individual developments. held constant, considerable variations in per capita out-
lays for local services remain.
Cross-sectional Analysis As part of a community growth study, a cross-sec-
tional analysis of growing California cities examined
Cross-sectional analysis consists of comparing rev- the relationship between population size and commu-
enue and expenditure data for observations taken dur- nity characteristics affecting local revenues and ex-
ing the same time interval. In most studies, the basic penditures [41]. The data show that as the size of the
observation elements are communitywide data. For ex- city increases, assessed property values, sales tax re-
ample, expenditures for services in the same year are ceipts, and median income per capita all drop, while
compared for communities, grouped by population size, the per capita cost of public services, number of crimes
to determine, by the use of regression analysis, if satis- per capita, and salaries for municipal workers rise.
tically significant relationships exist between their per Regional fiscal impact studies, which are concerned
capita outlays and their rate of growth. Expenditures with differences in the distribution of characteristics of
by community size, holding rate of growth constant, are
also compared. In order to effectively use cross-sec- 6. Interestingly, an independent survey of the literature
tional analysis in a fiscal impact study, it is necessary to yielded the same conclusions as the New Jersey study [33].
have a sufficiently large sample so that data on com- 7. This is not surprising, since large declining cities in New
munities can be stratified by groups of communities Jersey, such as Newark, have continuous outmigration of
middle-income households and, thus, are left with a dispropor-
with similar total populations and rates of growth. In tionate share of low-income families. See also: [75],

Cost-Revenue Analysis-State of the Art 9

the population, housing, and land use in a region as however, subject to criticism because potentially im-
well as with intra-regional differences in expenditures portant factors have not been considered."
and revenues, also utilize cross-sectional techniques. Based on these findings, the Fairfax study concludes
A cross-sectional study was made of 226 communi- that expenditures per capita are not as sensitive to
ties in the Pennsylvania part of the Philadelphia SMSA change in population as are revenues.
(Standard Metropolitan Statistical Area), including 94 A similar time-series analysis technique is applied to
contiguous municipalities with a density of more than estimate the incremental cost of additional students for
500 persons per square mile and defined as suburban a number of residential developments in California [8].
[35]. This study finds that, holding wealth constant, in- It is concluded, as in Fairfax County, that the incre-
creases in density correlate positively with increases in mental per pupil cost of education is substantially be-
local expenditures. No relationship was found between low average costs. However, a subsequent review of this
per capita income and expenditures for most public study shows that the methodology applied is ques-
services, although this finding did not include the tionable, since it assumes that the marginal cost of an
social services which are particularly sensitive to in- additional student did not increase during an eight-year
come. Industrial and commercial communities-those period of rising expenditures [14].
in which at least one-third of the property base is com- A time-series study in Boulder, Colorado [251 for
prised of industrial and commercial property-have a period of 21 years (1950-1970) shows a rise in
high police, fire, and general government outlays com- per capita outlays in real terms over this period. The
pared with predominantly residential communities. The study examines the rise in real personal income, and
study found that library outlays correlate positively with then estimates, local expenditures as a percentage of
social rank, while general government and fire service personal income. Adjusted as a percent of income, little
outlays are lower where single-unit dwellings predomi- difference in aggregate expenditure is shown for the
nate. Police outlays were higher in areas of increased time period studied. General government costs de-
density and parks and recreation expenditures increased creased, while library, parks and recreation, and sewer
with the value of housing. outlays increased. The model projects rising per capita
expenditures as a percent of per capita income for the
Time-Series Analysis 1972-1990 time period, assuming a growth in popula-
Time-series analysis involves the use of observations tion from 67,000 in 1970 to 122,000 in 1990.
of a variable or set of variables at different points in Econometric Models
time. For example, expenditure data for the same serv- Econometric analysis utilizes a mathematical formu-
ice in the same community can be compared for suc-
cessive years. A major limitation of the time-series lation of economic theory and statistical procedures to
studies that were examined is that usually only one ex- measure theoretical relationships and to verify (or re-
planatory variable-population-was considered. Other ject) given hypotheses. Econometric analysis applies
changes in community characteristics during the time techniques such as factor analysis and more advanced
span, possibly crucial ones, were not captured. A statis- statistical methods. Therefore, to perform an econo-
metric study, the analyst must have in-depth data on
tical limitation of time-series data is that they are not services and considerable skill in structuring models.
"independent," since the dollar allocation for a par- Thus the applicability of this technique is limited to
ticular service is closely related to the outlay level in comparatively few jurisdictions.
the previous year. A revenue/expenditure model applied to the City of
Time-series techniques have been applied in a num- New Haven, Connecticut, is an example of econometric
ber of studies. The Fairfax County Five Year County- analysis which takes a considerable number of vari-
Wide Development Program [29] uses historical pat- ables into account, many simultaneously.0 In addition.
terns to examine the impact of alternative population to population, salary and service levels are incorpo-
growth rates, making the assumption that population rated into the model. Specific services are examined to
characteristics are independent of the growth rate. Uti-
lizing time-series data, the study indicates that, as the 8. The methodology did not explicitly incorporate inflation
general population and student population increase, rates and salaries which were increasing at the time student en-
public services such as education, health, and welfare rollment showed only modest increases. In addition, lags in
show reduced average costs, while public safety and adjusting for changes in enrollment due to institutional rigidity
and contractual obligations were not considered.
general government show increasing average costs. The 9. See: [82]. This book provides data on the resources and
equations applied to estimate incremental costs are, computer costs associated with using the model.

10 Fiscal Impacts of Land Development

determine what proxies can be used for service demand, annexed low-density suburban area, but operating costs
and to what degree changes in the supply of services for such functions as public safety and social services
account for changes in outlay over time. Included in the were lower. On balance, there was a positive fiscal flow
analysis are property assessment and other submodels, from the suburban area to the central city.
and an analysis of demand for police and fire services
as a function of neighborhood characteristics. Analysis of Physical Characteristics
As examples of the results obtained, the study finds The physical configuration of development, the mix
a high incidence of fires in areas which have crowded of land uses, population density, and the natural en-
dwelling units of poor quality, and a low incidence in vironment are among factors which influence both pub-
areas which have a high percentage of owner-occupied lic and private service costs, Studies which focus on
units. Police manpower is shown to increase with the this aspect are briefly noted, and limitations of this
rise in the crime index. However, the relationship be- approach as applied in recent reports outlined.
tween changes in expenditures and changes in the
demand for and supply of services does not follow a Alternative Development Patterns
consistent pattern. The study projects city expenditures Local communities, counties and federal agencies
based on alternative policy assumptions regarding serv- have been among sponsors of studies to determine what
ice levels. . pattern of development results in the lowest per capita
Fiscal Flow Analysis cost of providing equivalent public services.
Fiscal flow analysis is a means of comparing aggre- A comprehensive study published in 1955 estimates
the fiscal effects of additional residential growth in
gate public sector costs and revenues among spatial suburban communities considering service costs for two
areas or population groups. A positive fiscal flow de- patterns of development in alternative locations [46].
notes an excess of revenues over expenditures, with the The likely annual cost of major capital outlays to serve
excess flowing, in the case of spatial analysis, to a diff er- the same population-given low, medium, and high
ent subcommunity area or jurisdiction, and, in the case density development on a hypothetical site-is the sub-
of households, from one income group to another. ject of a detailed analysis published in 1957 [38]. These
Fiscal flow data by intrajurisdictional area, land use, early studies are aimed at estimating the costs of growth
and income provide a framework for examining the patterns representative of residential developments
impact of both individual developments and aggregate prevalent in the 1950s.10
development. However, this technique requires a more The emphasis from the mid-1960s to the early 1970s
substantial cross-sectional data base of revenues and on Planned Unit Developments (PUDs) and new com-
expenditures than many communities have. munities initiated a new group of studies. The first of
In a study of Prince George's County, Maryland these efforts identified, which simulates public costs of
[,27], fiscal flows are estimated by type of housing unit land development alternatives in a semirural area, was
and intrajurisdictional district. The county was divided sponsored by Howard County, Maryland as part of
into districts which were selected on the basis of popu- their analysis of the new town of Columbia [37]. This
lation growth rates, population density, and income. report projects sewer, water, highway, park, and school
Revenues for each district were computed from assess- costs and concludes that the per capita cost of serving
ment records and income characteristics; costs for serv- the same population is reduced if population density is
ices, such as education, police, fire, and welfare, were increased. The Howard County report anticipated costs
derived from county records. The study found that both of providing utilities and other services from engi-
single-family dwelling units and garden apartments, in neering handbooks and similar sources. Other studies
the aggregate, produced a deficit. As to fiscal flows, elsewhere utilize a similar approach.
there was a drain from districts having moderate-in- One recent report estimates cost implications of
come housing into both lower-income and above- providing major public services and certain private
average-income areas. services (such as utilities and housing) given a num-
A similar fiscal flow approach was used in a study ber of hypothetical alternative development patterns
in Richmond, Virginia, which compared costs and rev- at the neighborhood and community level. The study
enues from the core central city with a newly annexed concludes that substantial savings can be achieved
suburban area consisting primarily of relatively new in public and private capital outlays as a result
residential developments [76]. This study showed that,
on a per capita basis, capital outlays were higher in the 10. For reviews of these and similar studies, see Kain [39].

Cost-Revenue Analysis-State of the Art

of what is defined as "planned" higher density de- open space are considered in only one study.
velopment. Most of the savings are attributed directly * Distinctions between private costs, public services
to increasing the number of housing units per acre. paid by user charges, and public costs are not
Data in this report partly represent simulated values explicitly stated.
and composite statistics from various sources and lo-
cations. The conclusions regarding differences in capital e Shifts between private and public costs, as for
and operating costs are not tested empirically by a com- recreation, transportation, and utilities, as density
parison of actual data obtained from planned commu- changes are usually not considered.
nities or large PUDs.11 * Differences in private and public construction
Neither this study nor the Howard County effort in- costs on a square foot basis have not been taken
clude estimates of the revenues likely to accrue to local into account in some reports.12
governments, which would also vary with density, hous- * Maintenance costs (such as repairs to utility
ing mix, and urban form, and which would therefore lines) are not accounted for.
have a bearing on the fiscal impact of different housing
patterns. Characteristics of Land
The implications for expenditures given alternative
development patterns are also examined in a study of Capital outlays, and to a lesser degree operating
St. Charles Community, a projected new community in costs, are influenced by the physical location of a pro-
St. Charles County, Maryland [361. Capital and operat- posed development.
ing outlays associated with a planned community and For example, topographic characteristics determine
with the existing growth trend (sprawl) are compared. the extent of grading necessary for private as well as
The study, which also incorporates the impact of socio- public construction, while soil characteristics determine
economic characteristics on per capita expenditures for the foundation necessary to assure structural stability.
services, shows that the planned community will bring The need for sewerage and water treatment facilities is
savings in capital outlays and school transportation costs. a function of both density and land characteristics.
The impacts of alternative density patterns on major A large-scale effort was underway in 1975 to estimate
public service costs are considered in a study of Santa costs of providing public facilities as a function of land
Rosa, California [41]. It shows little difference in op- characteristics. 13 Since the proportion of facility and
erating costs but projects lower capital outlays if higher service costs 'financed by the public and private sector
density developments were to be encouraged. vary considerably between and among states, the im-
Fiscal and other implications of alternative develop- portance of this factor in fiscal impact analysis depends
ment densities are the subject of a report focusing on a on the location of a proposed development.
semirural area of Connecticut [42] which estimates costs
and revenues if higher density "planned" growth takes LIMITATIONS OF ALTERNATIVE
place compared to trend or sprawl development. Al- TECHNIQUES DISCUSSED
though the two growth patterns would result in differ- A major limitation of most techniques described in
ences in demographic and socioeconomic characteristics, the previous sections, particularly those which estimate
such as income, the study assumes that the per capita changing expenditure levels, is that attempts are rarely
costs for almost all services are indifferent to variations made to adjust for differences in the quality and scope
in population characteristics. of services.
The approach taken in this group of studies has cer- It may be that higher per capita outlays in a growing
tain shortcomings, in addition to those noted previously, community are indicative of changes in service scope
from a fiscal perspective: or quality, rather than attributable to higher public em-
� Most of the studies do not relate differences in ployee wages or increased service consumption at-
population density to the consumption of services
other than education. 12. Kain [391 states that any public sector savings associated
with higher density are likely to be more than offset by in-
� Differences in the long-term capital costs for high- creased construction costs. Private construction outlays, show-
density concentrated development and for scat- ing differences in cost per square foot as density varies are
tered development which allows future filling of shown in [43, 62]. Other data indicate higher construction cost
for facilities sponsored by the public sector in areas of high
density. For further discussion of population density as a factor
11. The Real Estate Research Corporation is presently in fiscal impact analysis, see Chapter 4 of this report.
undertaking a study which is applying its findings to selected 13. This research, at the University of Florida, was under
communities. the direction of Harry Merritt.

12 Fiscal Impacts of Land Development

tributed to new residents. In undertaking an examina- more than 10,000 residents, the satisfaction with serv-
tion of changes in the cost of public services, differ- ices does not parallel higher outlays.
ences in outlays due to shifts in service scope and qual- Several studies-reviewed in a recent article [56]
ity need to be distinguished from shifts due to growth. which surveyed resident satisfaction with police serv-
Three approaches can be used to measure differences ices-find that smaller police departments with lower
in the level of services over time: changes in input, per capita expenditures are considered to be better by
changes in output, and changes in the quality of output residents than those in larger suburban or central city
as perceived by the recipients of the service .14 jurisdictions where the per capita outlays are greater.
Input measures include changes in police per 1,000 This finding is consistent with the New Jersey data
residents, pupil-teacher ratios, or other personnel which indicate that higher per capita outlays for police
changes on a per capita or per household basis. For are not linked to greater service satisfaction.
example, it could be argued that the higher ratio of A survey of service satisfaction among residents of
police personnel in a growing community or a smaller the Washington, D.C. metropolitan area indicates a
pupil-teacher ratio represents an increase in the scope similar pattern: Fairfax County, Virginia, which spends
of services. This approach, however, does not provide substantially less per pupil compared to inner suburbs
insight into the relationship between personnel assigned and the central city, ranks highest among area jurisdic-
and differences in the quality of service, since additional tions for school services.15
personnel may be required to maintain existing service These results, while tentative, tend to support the
levels rather than to expand or improve services. premise that higher or lower expenditures for many
Output measures include such variables as the pro- services do not reflect perceived differences in quality
portion of crimes that are solved, number of fires from the viewpoint of the resident.", What is the ex-
under control within a certain time interval, average dol- planation for this? In the case of police services, for
lar damage per fire, attendance at parks per 1,000 example, more police personnel (and thus higher spend-
residents or achievement scores in schools. ing) per capita are likely to be in response to higher
An additional and very useful measure of service crime rates rather than to demands for improved serv-
quality, particularly from the perspective of elected ices that would bring greater public satisfaction. In con-
officials, is the satisfaction of residents. These perceived trast, higher outlays for libraries or parks tend to reflect
service qualities, as reflected in user satisfaction, may service improvements if wages remain constant. It
not be consistent with changes in other output meas- should be noted, however, that the surveys cited do
ures. For example, classroom size may have been re- not reflect the views of residents over time. It may be
duced and test scores increased over a five-year period, that as taxes increase, consumers anticipate more in
but citizen satisfaction with schools could be lower than services. The low service ratings in cities may also be
indicated in an earlier survey. expression of the general frustrations of residing in
One growth study includes ratings of all major local large urban centers rather than specific dissatisfaction
services grouped by size and growth rate of communi- with particular public services. Therefore, surveys need
ties [58]. On the basis of interview data, the study to be combined with quantitative output measures to
found that consumer satisfaction is highest in moder- estimate actual changes in service scope as well as per-
ately growing communities ranging in population from ceptions of changes in quality.
10,000 to 25,000. While educational services were Other limitations of the methods reviewed are that
ranked highest in small communities, no service was cumulative, scale, secondary, and distributional impacts
rated comparatively high in large communities. The usually are not incorporated in the analysis. In view
study. found no relationship between per capita ex- of the complexity of these issues, the considerable re-
penditure and service ratings with the exception of sources required to consider these effects and the rela-
libraries, where higher outlays did correlate positively tively short history of fiscal impact studies, one should
with greater consumer satisfaction. These results show not be overly critical of these limitations. (These issues
that in New Jersey, resident satisfaction appeared to be are discussed in Chapter 4.)
virtually independent of expenditure levels. Although 15. The Washington Post, April 3, 1973. Comparative ex-
per capita expenditures are higher in communities with penditure and related data for schools within this metropolitan
area are shown in [341.
14. A list of measures to meet specified objectives and data 16. That is not to say that in a particular jurisdiction satis-
collection procedures for each measure for major public serv- faction would not increase as a result of higher service ex-
ices can be found in a recent Urban Institute publication [86]. penditures.

Cost-Revenue Analysis-State of the Art 13

111. WHY FINDINGS DIFFER

A recent article in American City compared two impact is merely one of the criteria, albeit a popular
frequently referenced fiscal impact studies [63].1 The one, which should be applied in evaluating proposed
author notes that one study on the impact of growth land use changes. An examination of the various studies
points to "a forty percent surplus of revenues over ex- also indicates that many factors other than motivation
penditures" while a similar study showed that a large of sponsors also account for differences in conclusions
development would leave the local jurisdiction "more regarding the fiscal impact of new development.
than $100,000 in the red." He figuratively throws up Two analysts, given the same data on a proposed
his hands and concludes, development, may differ on the magnitude of the surplus
"Herein lies the problem. We don't really have or deficit, and they may even reach opposite conclu-
the facts-pro or con." sions as to whether a surplus or deficit is created. These
conflicts are attributable, in part, to the set of initial
The problem cited has been the subject of consider- assumptions, such as the relationship between property
able debate at public hearings which consider rezoning values, household income, and the level of anticipated
cases or other land use changes. taxes. This set of assumptions applied is influenced by
One reason for inconsistent findings was discussed in analysts' professional background and their familiarity
connection with study perspectives, namely, the objec- with the literature on the subject. As to expenditures,
tive of the sponsor of the study. Of the many develop- the allocation of costs among land uses and between
ment impact studies reviewed by this writer sponsored new developments and the rest of the community can
by developers, only two residential projects were shown result in considerable variations in estimates.
not to produce an excess of revenue to local govern- However, given the same data sources, the same
ments .2 This would lead to the conclusion that essen- initial assumptions, and the same allocation methods,
tially all such nonsubsidized developments produce an the fiscal impacts of similar developments in the same
excess of revenue to local governments. Studies spon- or different communities are still likely to vary.
sored by the public sector and nonprofit organizations The subsequent sections discuss how initial assump-
show mixed results. The American City writer was thus tions, allocation procedures, state-local fiscal structures,
expressing the frustration of many who are confused and intra-community locations affect study results.
as to the "real" impact of new development.
It is apparent, after examining the many cost-revenue DIFFERENCES IN REVENUE ASSUMPTIONS
studies, that not all residential developments produce Most fiscal impact studies sponsored by developers
a surplus, nor is it necessary that they should. The fiscal assume, for the purpose of calculating property taxes
1. The two studies noted are [51 and [8]. and revenues, that the official asses sed-t o-m arket value
2. One of the two studies was implemented by a university. ratio will be applied to properties in. the new develop-
An agreement was reached that the findings would be pub- ment. Further, it is assumed that this ratio will - be
lished irrespective of the outcome [131. maintained for an extended time period once the devel-

opment is completed. In fact, due to inflationary pres- taxpayers, as when utilities or other facilities are
sures, infrequent reassessments, and other factors, there initially built oversize to meet later needs.
tends to be a gap between the official and actual An example of differences in allocating costs to new
assessed-to-market value ratio. For example, the state developments is the previously cited study of eleven
of California requires a uniform assessment-maTket projects in California [8]. The analysts in the original
value ratio of 25 percent. That is, a dwelling with a study did not allocate the capital costs of services
market value of $40,000 is to be assessed at $10,000. such as libraries_ parks, and recreation facilities to
In fact, however, the median assessment-to-market some projects. The basis for this decision was that
value ratio in California during 1972 was only 20 per- a small increment of population will not require the
cent [83]. Thus, on the average, a $40,000 housing construction of new facilities. This reasoning, however,
unit would be assessed at $8,000, and would pay 20 is only valid if the projects considered represent the
percent less in property taxes than predicted by assum- final development to take place in a community. More
ing the 25 percent ratio. typically, a project is one of a future stream of devel-
Such differences between effective and official tax opments which will require an expansion of libraries
rates were the basis for some of the adjustments made and other facilities to maintain existing service levels.
in a study that recalculated costs and revenues allocated Therefore, a share of this projected cost should be al-
to eleven residential development projects in California located to these projects based on the anticipated level
[14]. The original study indicated that fiscal surpluses of facility usage by new residents.
would be shown for ten of the eleven developments In some cases, the costs of such outlays as education
[8]. After these adjustments, which increased costs and are prorated to residential development on the basis of
reduced revenues, the analyst concluded that fiscal defi- the residential share of the property base [71]. For in-
cits would occur for eight and that reduced revenue stance, if in a given community residential property
flows would result from the other three. forms 70 percent of the base, only 70 percent of edu-
Similarly, adjusting the assessed-to-market value cation outlays are allocated to residences, the balance to
ratio from the official 40 percent to the historical 35 industrial and commercial property. This approach ob-
percent reduced the projected surplus of a proposed viously reduces the public sector costs associated with
large development substantially [65]. residential development and makes such development
Other initial revenue assumptions that are open to seem to be more attractive than other land uses. How-
considerable question are that the percent of disposable ever, this approach appears questionable. Since com-
income subject to the sales tax is independent of house- mercial and industrial developments are not direct causes
hold income level and that the sales tax is a function
of real property values'[9]. of the demand for educational, social, recreational,
or library services, the costs of these services (unless
specifically designated to serve commercial and indus-
CHOICE OF ALLOCATION APPROACH trial facilities) should be allocated only to households.3
The methods for determining which local services Similarly, to avoid double counting, revenues which
should be allocated to new developments, and the share accrue as a result of new households, such as higher
to be allocated, can have a substantial impact on esti- sales tax receipts, should not be allocated to new shop-
mated expenditures associated with new projects. ping centers.
How to allocate development-related revenues and There is no commonly accepted methodology for al-
expenditures involve decisions by the analysts, and location of costs for services jointly used by residential
these decisions can lead to a range of differences in and commercial developments, such as transportation,
study findings. Some of the major allocation decisions public safety, and general government. Ideally, such
include the following: costs should be allocated on the basis of actual cost
incurred. In practice, however, it is difficult to allocate
� Determining which expenditures should be allo- the costs of general government services or highways on
cated only to the new development, jointly to the the basis of the demand for services from residential
new development and the balance of the com- and business sectors. As in the previously cited case of
munity, or only to the balance of the community.
� Distinguishing between costs and revenues as- 3. A critique [421 of an earlier Palo Alto study [40] indi-
cates that education costs should have been allocated to resi-
signed to residential and nonresidential (com- dential developments on the basis of its share of the property
mercial and industrial) land uses. base. This approach implies, incorrectly, that most new com-
mercial and industrial development is linked directly to the
� Distinguishing between costs to present and future population base of a community.

16 Fiscal Impacts of Land Development

Prince George's County, expenditures for jointly used cost as a function of location 'and structure, it is unlikely
services are often allocated to residential and commer- that this method of allocation approximates closely the
cial development in proportion to the share each rep- cost of fire protection for a particular new facility.
resents of the property base. This technique has merit Despite these and other limitations, it is recommended,
for services linked to the value of property, such as may in the absence of more intensive analysis, that service
be the case for fire protection. However, value of the consumption be used to allocate the cost of public
property is usually not linked directly to the amount of safety among land uses.
service demand generated by a new enterprise. The Cross-sectional data indicate that communities with
use of this approach may also confuse the concept of considerable commercial and industrial facilities have
allocating actual costs incurred with the concept of public safety and transportation outlays substantially
allocating costs on the basis of perceived benefits. above the level found in predominantly residential
An owner of an expensive housing unit may derive jurisdictions. This difference is not fully accounted for
more benefit as a result of police and fire protection by differences in property value attributed to commer-
compared to an owner of an inexpensive housing unit cial and industrial structures. It may be, therefore, that
located in the same neighborhood. However, unless the the use of allocation proxies such as property values
level of protection differs, the cost of providing the underestimate the actual cost of providing services to
service is likely to be the same. All studies reviewed the business sector. Higher costs may reflect differences
rely on the allocation principle of costs incurred rather in population characteristics of residents in communi-
than the value of service provided as perceived by the ties with large commercial and industrial facilities com-
recipient .4 pared to those found in primarily residential communi-
An alternative approach is 'to allocate jointly in- ties.
curred costs, such as for police and fire protection, on New capital facilities are often designed with -over-
the basis of directly assignable services provided to capacity, considering present demand, to take advan-
each type of land use. In the case of fire protection, tage of scale economies in anticipation of future growth.
one study assumes that 50 percent of the cost is pre- The decision to build a facility larger than necessary to
cipitated by actual fires (direct cost), while the other meet current demand, essentially "locks in" another de-
50 percent is allocated to reserve capacity [38]. The cision-to develop the area serviced by the new facility
allocation of the former to different types of land use, in order that the burden of paying for the facility is not
i.e., residential, commercial, and industrial structures, fully borne by present residents. In the long run, if the
can be determined from fire department records. It is expected development takes place, this should reduce
then assumed, unless more detailed data are available, the unit service cost to both present and future resi-
that the cost of servicing a fire is independent of struc- dents.6 However, present residents have to meet the
ture type.@` Based on this premise, if 30 percent of all total cost of the facility, most likely in the form of
fires occurred in commercial facilities, 30 percent of the debt service payments, until the facility is fully utilized.
cost precipitated by actual fires is allocated to these What share of the cost should be allocated to future
facilities. To allocate this aggregate cost among specific users?
facilities, the cost is divided by the total number of (1) One concept is to allocate to future users only
workers employed in commercial facilities. Using this the incremental cost, while allocating to the present
cost per employee value, if a proposed development is users what would have been their unit cost in the ab-
anticipated to employ 50 workers, and the per em- sence of scale economics (the unit cost as if the facility
ployee cost estimated at $50, the annual fire fighting had been designed only to meet present demand). Us-
cost allocated to the development would be $2,500. ing this concept, the scale benefits accrue only to new
The cost related to service capacity can be distributed users.
according to the market value of the physical struc- (2) An alternative approach would be to charge
tures. This approach has some merit, since the alloca- new users the average unit cost once the facility
tion among land uses is related, at least in part, to is fully used. An allocation based on these two altema-
service consumption. However, because of variations in tives would result, in the short run, in having present
residents pay for part of the benefit which will accrue
4. For a thorough examination of benefit concepts associated to future residents.
with public services, see: [33].
5. In fact, structure type, fire protection devices on the 6. Presumably, engineering calculations incorporating pres-
premises, type of material stored, and the location of a struc- ent value concepts demonstrated, given an anticipated rate of
ture are among factors which can contribute to differences in development, the long-range savings prior to. approval for
the cost of fire protection. construction.

Why Findings Differ 17

(3) A third alternative would be to charge new INTRA-COMMUNITY LOCATION
development, when it takes place, the incremental cost In addition to variation in the fiscal structures of
plus the additional outlay imposed on present users many states, the impact of a development can vary ac-
until full utilization is achieved. This alternative appears cording to its location within a community. Most new
the most equitable, although delays in the anticipated developments, particularly those having low density, are
rate of development can result in allocated unit cost built on the periphery of developed areas, thus requir-
to the new users which exceeds the unit cost to present ing new schools, utilities, and other capital expendi-
users. tures. Frequently, location is a more important factor
than the density of development in precipitating capital
STATE-LOCAL FISCAL STRUCTURE costs [64]. A study of Fairfax County, Virginia L27]
Residential developments with identical physical and illustrates the impact of alternative locations within the
resident characteristics in two contiguous states can county on the cost of capital facilities.
have sharply different local fiscal effects. For example, It should not, however, be assumed that the ex-
the state of Maryland pays for the construction of new istence of unused public facility capacity will neces-
schools from state revenues. On an annual cost basis, sarily spur development in an area. Locational decisions
the per pupil expenditure for a new school is about that are uneconomic from a public viewpoint may be
$200. The state payment of this large capital outlay rational from the perspective of private developers,
in Maryland reduces the local expenditures associated since their cost of unimproved land tends to be lower
with growth. In Virginia, school facilities are financed as the distance from employment centers and transpor-
from local revenue. In most other states, school facili- tation corridors is increased. Therefore, the public cost
ties are either paid from locally derived revenue or of incentives to private developers has to be compared
jointly by the state and the local school district. As to public savings which could result by encouraging in-
another example, North Carolina and Delaware pay vestment to areas with unused capacity.
about 80 percent of school district operating costs, Current expenditures as well as capital outlays can
while in New Hampshire education is almost totally a also differ in various locations within a community. As
local outlay. Similarly, the state-local funding responsi- noted in the Prince George's County study [27], there
bility for other services, particularly transportation, are major intrajurisdictional differences in per house-
varies among states. These differences in state-local hold outlays for police, fire, education, health, and
responsibility for financing public services have a sub- welfare attributable to both differences in household
stantial effect on the fiscal impact of new residents. In characteristics and, for some services, location.
one case, the burdens of new growth may fall heavily
on local taxpayers, in another, the state may pick up SPATIAL SCOPE OF ANALYSIS
the tab, and in another the costs may be shared by The spatial limits of fiscal impact analyses depend
state and local governments. primarily on the perspective of the sponsor.' A local
Sharp differences are also found among the local jurisdiction is likely to be concerned only with revenues
revenue structures of certain states. In many states, it can collect and costs of services the community pro-
such as New Jersey, the property tax is the dominant vides. A county government will seek to examine the
local source of revenue. By contrast, the property tax fiscal impact of a new development on all taxing dis-
accounts for only 8 percent of the local tax base in tricts within its boundaries, such as schools, flood con-
Boulder, Colorado. Since the effective tax rate as a trol, and fire protection. A regional agency will con-
percent of income varies on the basis of each commu- sider multicounty fiscal interactions.
nity's revenue package, the revenue flow from house- In some cases, what is a surplus to a local govern-
holds of similar income status can be expected to vary ment may result in a deficit to the county or region.
substantially among jurisdictions. Conversely, it has been shown that a development
Overlapping taxing districts can also affect fiscal which creates a deficit at the county level can produce
flows. For example, one proposed development in a surplus to the state [76].
Contra Costa County, California would pay taxes to
sixteen tax districts [11]. The development may create 7. See, for example, differences in the geographic scope
a surplus to the cemetery district (few early deaths are considered relevant to estimate fiscal impact by the same
projected!) but a deficit to the school district. organization in two locations, (42, 45].

18 Fiscal Impacts of Land Development

IV. FISCAL IMPACT ANALYSIS
MAJOR ISSUES

Most fiscal impact studies examined limit their scope of anticipated revenue from households and firms. Most
of analysis to direct cost-revenue effects of new devel- fiscal studies assume that the cost of providing an ad-
opment. A number of broader, longer-term aspects ditional unit of service is equal to the average cost. This
have been excluded from most of these studies. To a section will discuss factors which can affect the incre-
large extent, this reflects the undeveloped state of the mental cost of providing services to new residents and
art of fiscal impact analysis. However, these are major firms.
issues and it should serve a useful purpose to note in- For purposes of this report, scale economies or dis-
formation that is presently available about them as well economies will be defined as changes in the unit cost
as the gaps in present knowledge. The discussion that of providing services as the number of service units
follows, therefore, is addressed to the following ques- expands, holding service quality constant.' This broad
tions: definition incorporates changes in service cost as popu-
� What is the impact of new development on the lation expands that are attributable to many factors,
incremental cost of providing services and how some of which will be identified in this section. In some
will revenues be affected? cases, these are offsetting factors. For example, fewer
� How do service preferences of new residents in- internal roads may be needed if a new development will
concentrate new housing units in highrise apartments
fluence the tax and service structure in a growing rather than building detached housing units which may
community? be typical of older developments. In this case, the in-
� What distributional effects are likely to accom- cremental cost of roads for the additional housing units
pany various forms of new growth? will be below the average community cost. However, if
� What secondary or indirect effects of development adding a large number of units, regardless of their con-
need to be considered? figuration, requires greater per capita outlays for major
� How does the time horizon influence study re- roads in the community or region, the aggregate fiscal
sults? Should both short-term and long-term ef- impact may be neutral. If large communities have to pay
fects be analyzed? higher wages to their municipal workers, as compared
to smaller jurisdictions, because equivalent housing and
� What fiscal interactions exist among local com- other components of the cost of living are more ex-
munities and other jurisdictions? pensive, this will be considered a diseconomy. While
housing is generally a private cost, if it contributes to
THE FISCAL EFFECTS OF CHANGES higher wages for municipal employees, part of this
IN POPULATION AND DENSITY 1. The discussion in this section deals only with services pro-
An expanded scope of fiscal impact analysis should vided by most local or county governments. Specialized serv7
ices, such as museums, graduate schools, and comprehensive
consider the effects of cumulative development on the medical facilities require a very large population to take advan-
cost of providing additional services and on the level tage of scale economies.

private cost is shifted to the public sector. Similarly, ment may not accurately represent its impact on the
congestion is primarily a private cost, but delays in total community.
providing public services caused by congestion are like- Scale effects are difficult to identify when considering
ly to increase the cost of these services. an individual development. They may become crucial,
Higher or lower incremental costs due to changes in however, when evaluating the cumulative impact of
efficiency should not be confused with changes in de- many individual developments.
mand. The number of children attending public schools The scale effects associated with new population
may increase above the community housing unit aver- should be considered in terms of their impact on capital
age because new developments are comprised primarily outlay, operating expenditures, and revenues. To de-
of large, low-density housing. This increase, in itself, is termine scale effects over time, however, one should
not a scale diseconomy, since the number of units consider capital and operating costs concurrently.
serviced has changed from the previous average.
Changes in per pupil expenditures attributable to in- Capital Outlays
creased enrollment would be considered scale economies Additional development can affect the unit cost of
or diseconomies. constructing public facilities by (1) increasing the
The importance of considering scale effects can be ability to take advantage of scale economies in plant
illustrated by the following example: construction, (2) increasing the population and geo-
A community with a balanced budget is considering graphic size of the urbanized area, and (3) changing
the impact of two new developments which will add density within the area of development. These three
3,000 students to the 19,000 presently in public effects need to be considered independently.
schools. Data show that once a total enrollment of Capital facilities provided by the public as well as
20,000 students is reached, the unit cost per pupil the private sector can have a theoretically optimum
for servicing each additional 1,000 students increases size, at which point they can obtain a minimum unit
an additional 2 percent, holding service levels con- cost. For example, a particular tertiary treatment plant
stant. Let us suppose that the new developments, may require a population of about 30,000 to operate
given existing tax rates, raise sufficient revenue to at its most efficient point. If a community with 25,000
meet the cost of education and other public services, residents requires that such a facility be built to meet
including the higher cost due to scale diseconomies service standards, the addition of 5,000 persons would
imposed by new residents. A cost-revenue impact reduce the unit cost to each user. Optimum plant sizes
analysis of the development would show a neutral fis- for school buildings have also been suggested [38].
cal flow (revenues balancing expenditures). However, However, the optimum size tends to vary over time as
since the unit cost for all students, including original a result of technological and other changes. In de-
residents, would be raised, a -fiscal deficit would be termining the appropriate size of schools and other
shown by the balance of the community. Increasing facilities, both initial outlays and operating costs over
taxes to a level which would offset the aggregate the economic life of the facility have to be computed.
deficit would result in a fiscal surplus from the new In addition, private costs (such as the distance to drive
residents to the balance of the community, although a personal vehicle to obtain a public service) should be
the original residents (as well as the new ones) had considered.
their tax rates increased to meet the deficit caused A growing community can take advantage of scale
by the new residents.2 economics through comprehensive planning geared to
Given economics of scale, the opposite result would optimum facility size and by controlling the location,
be shown: original residents would benefit by lower density, and pattern of development to the degree pos-
tax rates since the increased number of residents serv- sible. Of course, optimum facility size needs to be
iced reduces the average unit cost. balanced against the many other fiscal and nonfiscal
The example illustrates two points: all residents- considerations that should be incorporated in land use
new and old---can be affected by economies and dis- plans and controls. For example, if a community builds
economies of scale-, and, if scale effects are present, a facilities larger than short-term demand dictates to take
fiscal flow study confined only to a proposed develop- advantage of scale economies, the decision to approve
additional development has been implicity taken.3 IM-
2. While this example is hypothetical, education finance data pact analysis would, however, remain a useful instru-
show per pupil costs rising as enrollment expands beyond a ment if alternative development projects are under con-
threshold value, due primarily to higher salaries in large
districts. 3. Capital cost allocation is briefly discussed in Appendix A.

20 Fiscal Impacts of Land Development

sideration for the same site or if the negative fiscal flow capita cost of primary roads increases by 50 percent as
from a particular development being considered is total population rises from 50,000 to 250,000 [44].
greater than the incremental cost of not having capital Per capita annual capital outlays for most functions,
facilities fully utilized. as well as gross debt outstanding, as shown in Table 1,
Although there has been some discussion about po- rise as urban population increases.5 Although these
tential scale economies in growing areas, empirical data data do not isolate the many variables which influence
on the subject remains limited .4 capital expenditures, it would appear that public sector
A second factor which can influence per capita out- savings which may accrue from scale economies in large
lays for capital facilities is the increase in the popula- urban areas are more than offset by other factors, such
tion and geographic size of the urbanized area as a re- as the necessity to construct and subsidize the operation
sult of new development, independently of scale econo- of mass transit systems. I
mies in plant construction. As population expands, is the The third factor one needs to consider is the impact
incremental cost of additional capital facilities above or of development patterns, particularly density, on capi-
below average cost? Data on this subject is again lim- tal outlays. Utilities such as water, sewerage, natural
ited, since studies examined deal primarily with the gas and electricity collect or distribute their product by
cost of facilities necessary for relatively small total a system of pipelines and cables. For example, the
populations. However, one study, based on both simu- length of sewerage trunk lines, and thus the initial
lated models and empirical data, indicates that the per capital outlay per dwelling unit, can be reduced by in-
4. A study of new communities in Great Britain concludes creasing the density of residential housing. Many of the
that scale economies in the construction of public facilities 5. The cross-sectional data in Table I show average costs
such as water and sewerage are exhausted when population which reflect historical development and population change
reaches a size of about 50,000 [44]. For an analysis of scale patterns, notAthe incremental cost of services to a new house-
economies in sewerage treatment plants, see [381. hold.

Table 1. PER CAPITA OUTLAYS FOR SELECTED MUNICIPAL FUNCTIONS, 1970-1971

City Population Sizes
(as of 1970)
Less Than 50,000- 100,000- 200,000- 300,000- 500,000- 1,000,000
FUNCTIONS 50,000 99,999 199,999 299,999 499,999 999,999 or more

CURRENT EXPENDITURES a
Police Protection $ 16 $ 21 $ 24 $ 26 $ 28 $ 41 $ 55
Fire Protection 9 17 20 20 20 21 23
Sewerage Operation 5 5 5 6 8 8 4
Sanitation 6 8 9 13 10 12 19
Parks and Recreation 6 12 14 16 20 18 13
Libraries 2 4 5 2 4 6 6
General Government b 8 9 9 11 9 17 16
TOTAL 52 76 86 94 99 123 136

CAPITAL OUTLAYS
All Functions 29 41 52 64 66 71 83
Water Supply 5 6 5 7 6 7 8
Sewerage 6 7 10 10 10 10 10
GROSS DEBT
OUTSTANDINGe 234 277 330 427 423 457 813
DENSITY (Per Square Mile) d n/a n/a 4,681 5,261 5,697 6,853 12,750

SOURCE: U.S. Bureau of the Census, City Government Finances in 1970-1971, Washington, D.C., 1972.
a. The data shown reflect total outlays per capita. The values are not adjusted for differences, if any, in the scope of services or
other factors.
b. Financial administration and general control.
c. Per capita.
d. Average population densities for each city size category. For 100,000-199,999, based on random sample; for other categories,
includes all cities.

Fiscal Impact Analysis-Major Issues 21

reports noted in Chapter 2 provide comparative data, a patrolman or half a squad car. Once added, how-
based on simulated conditions, on the cost of capital ever, there may be overcapacity until demand expands.
facilities as housing density is changed, with the total Although "lumpiness" of a service is a problem in
number of housing units held constant [36, 37, 38, 43]. evaluating the fiscal impact of some developments, the
However, since this report is aimed primarily at public major limitation of the typical cost-revenue analysis
costs and revenues viewed from the local perspective, which allocates current average per unit service costs
distinctions between private and public outlays should to both small and large developments, is that unit costs
be noted. Utilities such as water and waste treatment are not constant as a community grows. As a result,
usually charge user fees aimed at recovering the capital the collective impact of development on expenditures
and operating costs associated with the level of use. is likely to differ from projections based on current per
Systems within a development, such as internal roads capita average costs.
and sewer lines are, in most cases, constructed by the A considerable number of studies have examined the
developer. As a result, most savings associated with relationship between per capita service costs and pop-
increased density shown in these studies are private ulation size or density. The pattern, according to the
rather than public savings.6 more recent studies, is that unit costs of most services
The cost of purchasing additional land for public rise as population or density increases beyond a thresh-
facilities is also likely to rise as the available supply old level.'
within an urban area decreases. This has been shown A frequently found pattern is for per capita costs to
in a study which compares site acquisition costs for be reduced as a community grows to a population of
schools in jurisdictions with varying populations and between 10,000 and 50,000. In communities smaller
densities [38]. The report shows that large, high than 50,000 the incremental cost of providing services
density urban centers have higher land acquisition and is often below average cost, indicating that development,
construction costs (on a per acre and per square foot from a fiscal viewpoint, should be encouraged.8
basis, respectively) compared to lower density cities For communities with greater populations, as shown
and suburbs. Prudent advance acquisitions of sites also in Table 1, per capita outlays for labor-intensive services
may result in long-term savings. Since the cost of land usually funded from local revenue sources rise.9 How-
has been rising faster than the cost of construction, ever, one should be careful not to attribute higher costs
delaying site purchase to await scale economies as- primarily to greater population size. Population density,
sociated with larger facilities may offset operating as shown in Table 1, increases as population rises.
economies. Whether concentrated growth to take This is not surprising, since most cities cannot annex
advantage of scale economies or scattered growth within contiguous incorporated jurisdictions. Large cities
urbanized areas results in lower per capita outlays with comparable population size and similar median
depends on the utilization rate of the existing in- income, but lower densities, as in California, have sub-
frastructure [46]. stantially lower per capita outlays for most local serv-
ices. The per capita cost of roads, however, is more a
Operating Outlays function of population size than density.
Changes in operating outlays resulting from small Adding a new residential development to an urban
developments are difficult to isolate. Considered by area with a population of 50,000 is, from a transporta-
itself, a development of fifty or so housing units in a tion perspective, less expensive than adding the same
large community may have no visible impact on public development to an urban area of one million.
safety and other noneducational operating outlays. On the basis of his examination of service costs in
This is the result of the "lumpiness" of a service. That New Jersey, Sternleib concludes, "municipal officials
is, only the collective effect of a number of small de- in moderately growing communities looking to reduce
velopments are likely to trigger additional manpower per capita costs may try to hold population at the
or equipment. It is obviously impossible to add half 10,000 to 25,000 mark since it is at this population
6. A clear distinction between public and private costs is 7. For a review of some of the literature on this subject,
often not possible. For example, matching federal grants are see: [34]; also: [47, 48, 49, 50, 51, 57, 58].
frequently provided for the construction of water sewerage 8. Since communities of 50,000 residents or less are grouped
treatment plants, while the local government share is met by in Table 1, these scale economies are not evident in this tabu-
user charges. In most areas, roads internal to a project are lation.
built by the developer, while the maintenance of these roads is 9. Costs have also been rising faster in larger communities.
a public responsibility. In some cases, user charges do not Between 1965 and 1972 per capita current expenditures in the
reflect the full incremental cost of extending services to an largest cities have risen 140 percent, in cities with 50,000 to
area, shifting the cost to other users in the community. 100,000 inhabitants, only 85 percent [89].

22 Fiscal Impacts of Land Development

level that per capita costs are lowest" [58]. reliance on the state police in larger communities.
Most studies examine costs of specific services rather Increased per capita costs for fire protection in
than aggregate expenditure levels. For example, there higher density areas are due primarily to higher wages
is considerable evidence that small, isolated school dis- and more sophisticated equipment as well as more
tricts have higher per pupil outlays for equivalent school building fires and alarms per capita in communities with
services compared to somewhat larger school districts. highrise or intensive commercial-industrial develop-
However, beyond a minimum school district size, the ment.13 The need for fire stations in newly developed
reduced cost of transporting students in comparison to low-density residential areas is due primarily-to the
smaller districts is offset by higher per pupil instructional areas' distance from existing service, rather than the
outlays.10 Concurrently, a reduction in school enroll- number of new dwelling units. In very low density areas,
ment, which occurs in areas of the nation which have per capita public sector outlays for fire protection tend
outmigration (particularly in large cities and rural to be low, but the private cost, in terms of the proba-
areas) also leads to higher per pupil costs since plant bility of a total loss, is high and fire insurance rates
operation and maintenance costs remain unless a reflect this fact. As with other services, wages paid to
facility is closed."' firemen are likely to be an important reason for varied
In the case of expenditures for police services, almost per capita outlays between large and small jurisdictions.
all studies show that as population and/or density in- Despite the impact of location and building type on the
crease, per capita outlays rise once a threshold of about cost of fire protection services generally, a number of
10,000 residents is reached. This rise is attributable communities utilize a fixed ratio of firemen per 1,000
to more police manpower per capita, and, to a lesser residents. 14
degree, to higher wages and more sophisticated equip- Health and welfare costs are not directly affected by
ment utilized in larger communities. Data collected changes in scale associated with the collective impact
annually by the Department of Justice, which show more of development, except for higher wages for public
police per capita and higher crime rates as the size of a employees. Since new residents of nonsubsidized new
community increases, support these studies [92]. In housing tend to have incomes above the community
contrast, fiscal impact studies of new developments median or mean, per capita costs are likely to be
show that the demand for police services, as indicated reduced in areas of rapid growth. In the case of sub-
by crime rates, is at or below the community average.12 sidized housing, social service costs may increase; how-
It would appear, therefore, that if the incremental cost ever, these costs are essentially income transfers from
of providing police services to expanding areas is below one group of households to another. As these costs re-
the community average, the per capita communitywide flect broader income redistribution objectives, state and
costs should be reduced as the population grows. A federal agencies now provide the major revenue source
number of explanations can be offered for this incon- for these programs.
sistency: The new population demands more sophisti- Economies of scale are not shown in the case of
cated police services, or inmigrants to housing in older larger jurisdictions, even for capital intensive services
areas vacated as a result of housing expansion on the such as sewerage treatment and water supply. 15 Per capita
periphery have a higher crime incidence. Perhaps more operating costs for sewerage treatment plants, based on
important, however, is that crime-related services national data, increase for cities with up to one million
account for only a small share of total police calls. inhabitants. The nation's largest cities, however, have
Police services such as traffic control and protection of the lowest operating costs for sewerage treatment.
commercial facilities extend beyond the boundaries of The causes for the relationship between population
the new development. As a result, examining police size, density, and cost of services are not fully under-
demand only within a development is an inadequate stood. Factors which have been' cited include con-
measure of additional services which are required as a gestion and inefficiencies associated with large govern-
result of growth. Finally, the state-local provision of mental organizations.", There is more substitution of
police services shifts as a community grows, with less public for previously private or quasi-public services
10. This is attributable primarily to higher salaries for such as fire protection and recreation as density in-
equivalent education and experience.
11. This problem is present even in spatially large com- 13. For discussion of this subject, see: [591.
munities with net inmigration, such as Fairfax County, Vir- 14. In Santa Clara County, communities maintain a ratio
ginia, or Montgomery County, Maryland, due to the con- of one fireman per 1,000 residents. See: [80].
tinuing decline in the birth rate as well as to the new growth on 15. In New Jersey and Ohio, sewerage service costs are
the periphery of urbanized areas. higher for cities with over 250,000 persons. See: [511.
12. See, for example: [9, 24, 27]. 16. For a discussion of some of these factors, see: [53].

Fiscal Impact Analysis-Major Issues 23

creases. An important factor is the higher public anticipated revenue flows from each individual develop-
service wage structure found in larger communities, ment.
reflecting both higher living costs and stronger union Higher revenues to the public sector which can result
bargaining power and, in some cases, adverse working from greater economic activity, particularly commercial
conditions. These increased wages are not usually off- activity, in large urban centers may partially offset higher
set by rises in productivity, in contrast to much of per capita public services costs. For example, higher
the private sector, since most public service functions, wages for public employees may be representative of
with the exception of utilities and road maintenance, are higher income for most households residing in large
labor-intensive and do not enjoy many of the produc- urban centers. This, in turn, would reflect an increased
tivity gains associated with improved technology. ability to pay for public services by larger as com-
Two factors, in addition to wage differentials, appear pared to smaller jurisdictions. A greater share of the
dominant: the change in the socioeconomic characteris- total service cost can also be shifted to business firms.
tics of the population and more intensive nonresidential In California, median family income declines
land uses in urban areas with high population densities. slightly as the population of cities expands [411. These
The net outmigration of middle-income households from California data include only cities which had an in-
high density cities creates a greater demand for public crease in population during the last decade. Income
services, which requires more personnel per capita-the data from cities across the nation, shown in Table 2,
major cause for higher per capita operating outlays. indicate a mixed pattern. Both the largest and
Intensive commercial development in large metropolitan smallest cities for which data were obtained have per
areas results in private rather than public scale econo.- capita and family incomes above the level of other
mies, and requires higher capital outlays. cities.
The costs of local services, as a share of income, vary
Revenues from 2.1 percent for cities in the 50,000 to 100,000
The collective effect of development on income and group to 4.0 percent in the largest cities.19 However,
property values-and thus on potential revenues-has since the commercial share of all property is greater in
not been thoroughly examined by researchers. As a the large cities, part of the service cost is shifted from
result, only general observations can be made. Per households to business firms.
capita personal income rises more rapidly in growing Commercial and industrial property as a share of all
communities than in other jurisdictions. The dominant property increases from 29 percent in cities with pop-
factor in this pattern is that the new residents have ulation in the 100,000-200,000 category to 40.4 per-
permanent incomes which tend to be higher than cent in cities with over one million inhabitants. These
for the base population." However, the expansion data suggest that large cities, particularly those with
of general economic activity, higher sales, and ad- high population densities, do attract more commercial
ditional employment opportunities for existing resi- property than smaller communities, including most
dents are also likely to increase the personal income of suburbs of central cities. This reflects one of the
many residents. Since prices tend to be higher in large economic benefits of concentrated population. If one
urban areas, wages in both the private and public sector assumes that the percentage of residential property is a
reflect this increased cost of living. These effects tend crude proxy for the share of local taxes paid by house-
to reflect a long-term pattern of growth, and are un- holds, only about two-thirds of the service costs in cities
likely to be the result of one development. Land values' shown in Table 2 are borne by their residents. Adjusting
and thus property tax revenues, can also be expected to per capita outlays to reflect only the household share
rise in areas of rapid growth reflecting more intensive reduces somewhat the gap in the percentage, of income
use plus anticipation of further development."' allocated for selected local services as population size
Given the premise that there is a rise in both personal varies.
income and property value due to cumulative growth Impact of Outmigration
patterns, local revenues can be expected to increase
more than would be shown by simply estimating the The available data strongly suggest that large juris-
dictions, particularly those with high population densi-
17. Demographic and socioeconomic characteristics of in- 19. Service costs shown in Table I are for all cities, includ-
migrants to major urban areas are shown in [901. ing those which have declining populations. Large cities with
18. For example, a study of land sales in Minnesota [67] indi- declining populations have higher per capita outlays than cities
cates that the rise in sales price was faster in metropolitan which are gaining residents. However, the pattern of service
counties of rapidly growing areas compared to slower growing costs as a percentage of income rising as a function of city size
regions. appears to hold. See: [75].

24 Fiscal Impacts of Land Development

ties, have high per capita outlays for many services. Second, some research suggests that in the short run
In part, this is attributable to some of the factors rapidly growing communities have reduced per capita
identified which are independent of population char- costs following a spurt in population size because of
acteristics. In part, higher per capita costs reflect the initial excess capacity in public facilities or, more
outmigration of middle income households from cen- typically, a lag in the budgetary process. These results
tral cities and many inner suburbs during the last two should not be confused with longer-term trends.
decades, which does not significantly reduce the demand
for noneducational services, particularly social services SHIFTS IN PREFERENCES FOR SERVICES
by the remaining population. Reduced population size: AND TAXES
in this instance, does not lower per capita outlays, while Preferences of new residents for "tax packages" and
the cost of providing services to new development aimed services may differ sharply from those of the base
at middle income households should be substantially population. The importance of these preferences is
below the average for all families. Conversely, the suggested in the following illustrative example:
cost of providing services to similar middle-income A new development is planned in a pre-
households moving to a large, rapidly expanding suburb dominantly rural county on the fringe of a metro-
is likely to be at or above the community average. politan area. The new residents will consist pri-
Notes of Caution marily of young households-former central city
residents-with incomes above those of the base
There are two notes of caution in examining scale population. These new households will form a
effects. First, time series data showing community voting majority in their new jurisdiction. On the
expenditure patterns should be used with care. The familiar assumption of many fiscal impact studies
mere fact that per capita service costs in constant dollars that the new residents are comparable to the old
are rising over time in a growing community may not and that the previous per capita service costs will
necessarily be linked to the rate of growth but may persist, revenues and costs associated with the new
be due, rather, to an overall rise in public service costs. development are projected to be in balance.
Cross-sectional data which compare expenditure pat- When the new development is completed,
terns in slower growing jurisdictions are one basis for however, neither revenues nor expenditure pat-
determining whether such changes were typical of other terns behave as projected. New residents vote
communities, independent of their growth rate. to construct a better high school and to provide

Table 2. OUTLAYS FOR MUNICIPAL SERVICES AS A PERCENTAGE OF INCOME, 1969-1970

City Population Sizes
(as of 1970)
Less than 50,000- 100,000- 200,000- 300,000- 500,000. 1,000,000
50,000 99,999 199,999 299,999 499,999 999,999 or more

Outlays for selected local
functions (per capita) $52 $ 71 $ 81 $ 96 $ 98 $ 116 $ 138
Mean per capita income:
All cities n/a 3,309 3,318 " 2,984 3,165 3,222 3,451
Cities with increasing
population n/a n/a 3,385 " 3,050 3,241 3,260 3,677
Outlays for local functions
as percent of income e n/a 2.1% 2.4% 3.2% 3.1% 3.6% 4.0%

SOURCES: U.S. Bureau of the Census, City Government Finances in 1969-70, Washington, D.C., 1971; and U.S. Bureau of the
Census, General Social and Economic Characteristics, State reports, Washington, D.C. 1972.
NOTES:
a. Combination of total current expenditures (as defined in Table 1) and annual debt payments in 1969-70.
b. Based on random sample of fifty cities. Remaining data on these lines are for all cities in their respective size categories.
c. All cities. Note rise in current expenditures between 1969-70 (Table 2) and 1970-71 (Table 1). Increases amount to 12.4
percent for largest cities, 10.6 percent for smallest cities. Per capita metropolitan personal income increased by 6.3 percent between
1969 and 1970.

Fiscal Impact Analysis-Major Issues 25

extras, such as additional college preparatory There are some data, however, which show that growth
courses, that the older residents did not desire. results in more rapid rise in local taxes than the rise in
The newcomers want to improve access roads to income attributable to new residents [78], causing in-
employment centers where most of them work. creased tax burdens. 23
Collectively, in short, the new residents in this hypo- Shift in Revenue Sources
thetical case have public service preferences whose The property tax is the dominant source of revenue
costs far exceed the level previously provided to the in rural and small urbanized areas. As the demand for
base population. The new residents are willing to public services increases as a result of growth, revenue
increase their tax contribution but impose the higher sources tend to be diversified. Although property tax
rates on all residents to cover the cost of these outlays. rates are likely to increase, the relative importance of
Under these conditions, the new development fiscally
"pays its own way." However, the old residents are this revenue source declines as additional taxes and
implicitly subsidizing the new residents .20 fees-sales, income, utility, and business levies-are
added. These additional taxes reflect (1) preferences by
Shift in Service Demand new residents for a broader tax base as property tax
The hypothetical example above is based on obser- rates increase, (2) the ability of a larger jurisdiction to
vations of communities undergoing urbanization. Prop- efficiently administer new taxes, and (3) the increased
erty taxes in rapidly growing semirural communities of ability of expanding communities to shift part of their
Virginia have risen more quickly than in similar com- added tax burden to nonresidents.
munities with stable populations, primarily to finance If the people who lived in an area before the period
the expansion of locally provided public services. A of new growth do not want the new services and the
similar pattern has been shown in semirural Michigan higher taxes, and if they hold sufficient political power,
counties [84]. of course they can then block the bond issues or tax
The likely cause for these shifts in preferences in the increases.
urban fringe are the characteristics of new residents. Importance of Preference Shifts to Jurisdictions
They are typically younger, with more scbool-age chil- Communities which anticipate that new development
dren, of higher social rank (in terms of occupation), may bring inmigrants whose characteristics deviate
and with expectation of more lifetime income than the from the existing population base should pay attention
long-time residents of the suburban areas .21 As in- to preference shifts in assessing fiscal impact .24 Al-
comes rise, demand often grows for "luxury type" though studies are limited, the pattern which emerges
services-parks, recreation, libraries, and special school indicates that as the characteristics of residents shift
programs. This phenomenon may reflect, in part, the from rural to urban, higher local taxes as a percentage
federal. and state , income tax regulations that enable of income result. This is attributable to higher demand
higher income taxpayers to deduct their local tax costs for public services and higher wages for municipal
of these services, thus obtaining them at less cost than workers. A community which anticipates such change
if they purchased similar services privately.22 should examine the fiscal pattern of other communi-
Empirical data on expenditure shifts in growth areas ties in the region during the time span of rapid urbani-
attributable to preferences of new residents are some- zation as a means of estimating the likely impact of
what limited. There is considerable evidence that per new population on service demand and revenue needs.
capita outlays increase as population grows, but it is
not easy to distinguish between higher outlays due to DISTRIBUTIONAL EFFECTS
diseconornies of scale, discussed previously, and those
due to different preferences among new residents. Distributional effects associated with new develop-
ment which may have direct fiscal consequences are
20. Technically, the marginal cost to the older residents of generally overlooked in cost-revenue studies because of
the new service would be greater than the marginal benefit their complexity and the premise that they are outside
they perceive. Concurrently, the marginal benefit to new resi- the scope of these efforts.
dents may be higher than their cost. This section discusses three types of distributional
21. Inmigrants to central cities and some mature suburbs
may have different characteristics than the ones described here.
22. Taxes imposed by local and state governments are de- 23. For additional discussion of factors which increase pub-
ductible expenditures for federal income tax purposes on the lic costs more rapidly than revenues in an area of rapid growth,
questionable premise that there is no direct linkage between see: [601.
such taxes paid to benefits received. 24. For an example of such a development, see: [121.

26 Fiscal Impacts of Land Development

effects-the shift in service demand from one area to If the community restricts large detached housing units,
another within a community or region as a result of a there will be a spillover of demand to other juris-
new development; fiscal benefits from the standpoint of dictions within the region. If all suburban communi-
an efficient distribution of new households within a ties in the region restrict such housing, some demand
region; and the income distribution effect of changes will spill over to jurisdictions on the periphery of the
in the local tax structure or allocation of funds for local region, with some shift to the central city.
services. Some larger households, as the cost of larger units
rises, will occupy the new high-density housing, while
Changes in Housing Mix others will compete for the existing detached housing
One study of a new community, following the pattern stock within the community. These new households
of similar efforts, assumes correctly that the addition of are likely to have demographic profiles which differ
high-density housing with few bedrooms per unit will from those presently occupying either high-density or
attract households with fewer children than detached detached housing units. Thus, the number of children
housing developments.25 The study, however, further attending public schools after these readjustments will
assumes that larger families (those seeking three or differ from the level assumed based on existing patterns.
four bedrooms) would not seek lower-density housing This, in turn, would require that fiscal calculations be
in the balance of the county or even the state within adjusted to reflect the change in the number of school
which the development is located [36]. Based on this children per household.
assumption, it is concluded that the demand for local The point is that the number of families who will
and state services, particularly education, would be utilize public schools and most public facilities in a
sharpjy curtailed, resulting in a fiscal surplus from the region or state will not be significantly reduced by shifts
proposed development to the community and state. in the mix of housing. In spatially large jurisdictions
It is more likely, however, that aggregate demand ior most redistribution of service demand will be local.
schools and other services has not been appreciably Smaller communities, however, can affect service de-
reduced as a result of the development, but merely mand by altering their mix of housing.27
shifted or distributed to another location within the Efficient Location of Households
community, region, or state. If this alternative hypoth-
esis is valid, the approach of such studies needs to be The availability of public facilities should be one of
reexamined. the major considerations in determining the most effi-
Interurban migration is motivated by many factors, cient location of households from a regional planning
with improved employment opportunities (higher viewpoint. For example, a mature suburb with consider-
wages, less unemployment, or both) a major considera- able outmigration may be encouraging the construction
tion.26 It is doubtful if any significant migration to most of high-rise apartments, although substantial open space
regions is encouraged by a strong preference to live in remains for lower density development.28 Such a com-
a particular development or even community within a munity is faced with the problem of closing elementary
metropolitan area. The construction of smaller, less ex- schools in excellent physical condition due to the decline
pensive, higher-density units may encourage some in its school-age population. An adjacent community, be-
households who would otherwise not have moved to the cause of an influx of households with school-age chil-
community to locate in new units. However, for other dren, requires a large school construction program.
potential households, the higher density construction will From a regional perspective, this is an inefficient loca-
not meet their perceived housing needs and thus will re- tion of new households with children. Household loca-
sult in a shift of families with children to larger units tion is emerging as a major issue as large operating sub-
(new or old) in other parts of the jurisdiction or region. sidies are foreseen for mass transit, including the BART
system in San Francisco and Metro rail system in the
25. Note that most highrise developments, with the excep- Washington, D.C. metropolitan region. Utilization of
tion of very expensive condominiums, average no more than
two bedrooms per unit. Detached housing units tend to have 27. Inner suburbs of Washington, D.C., such as Arlington
three or four bedrooms. One study considers a typical high- and Alexandria, Virginia, which have experienced sharp in-
rise unit to have 900 square feet, a typical detached unit creases in the number of highrise apartment units, have rapidly
1,600 square feet [43]. declining school enrollment.
26. In addition to moves to take advantage of employment 28. High land values, in anticipation of intensive develop-
opportunities, households migrate to areas because of climate ment, are an important factor contributing to this pattern.
and scenery, for access to urban amenities, and to seek pri- There are other factors which encourage this trend. Thus, the
vacy. For a discussion of other nonemployment factors that pattern may emerge without the encouragement of local
influence interregional migration, see: L74]. officials.

Fiscal Impact Analysis-Major Issues 27

mass transit increases if intensive development is while the elderly are financially worse off.
located close to transit stations, which may reduce Constraining new housing, and thus population, in
operating deficits. However, there are insufficient data areas which are perceived as highly desirable will in-
to indicate the fiscal impact of household locations on crease average household income (in the absence of
operating deficits of mass transit systems. policies to subsidize units) since inmigration will be
The fiscal implications of alternative location strate- limited to more affluent families who can afford to com-
gies for new development require additional examina- pete for the existing housing stock. For example, Boul-
tion. A tentative step in this direction has been taken by der, Colorado estimates that new family income would
a community which estimated differences in capital out- increase from $18,259 to $19,591 if the population could
lays given alternative locations of new households [28]. be limited to 122,000 residents in 1990 rather than a
Unfortunately, data on regional approaches are limited higher level [24]. However, as noted earlier, the re-
since most communities are unwilling to surrender any gional demand for education and other public services is
of their land use controls to areawide authorities. unlikely to be substantially diminished by the growth
policies of one jurisdiction. Rather, the demand is
Income Distribution shifted to nearby communities.
A surplus from new development can either reduce
taxes or provide additional revenue to expand services. SECONDARYIMPACTS
Since most residents pay local taxes and consume some Most of the cost-revenue studies completed to date
local services, the beneficiary group tends to be large. focus on the primary fiscal effects of new development.
Since deficits from new development result in higher While it is recognized, for example, that a residential
taxes or fewer services, they similarly affect a large development creates a secondary demand for com-
group. If new development creates a surplus to the mercial facilities to provide goods and services for the
state government, all its residents may benefit. How- new residents, fiscal implications of such secondary
ever, the magnitude of surpluses or deficits has to be effects are usually not included as part of cost-revenue
substantial to result in any measurable impact. Com- analyses.
muters can also benefit if surpluses are used to improve From a community perspective, the following sec-
services, such as transportation, consumed by this ondary growth effects of new development which have
group. Both disposable income and income "in kind" fiscal implications require consideration:
can thus be affected by new development. 9 Changes in the value of surrounding land and
A secondary but important result of fiscal surpluses structures.
is that local property values increase, reflecting lower
taxes or more services which accrue to residents of the 9 Increased demand for goods and services resulting
property. This benefits present property owners, but from residential development.
increases the initial capital outlay necessary to purchase 9 Increased demand for housing and services as a
housing by potential inmigrants and others. result of industrial development.
Independent of surpluses or deficits, the preferenc 'es e Increased demand for housing and services as a
of residents in new developments, as noted earlier, can result of commercial development.
cause a change in either the tax structure or in the al- * Increased activity on the periphery of new de-
location of funds for services. Either shift is likely velopment.
to affect the income distribution of residents. For
example, a shift from a utility tax to an income tax is The secondary effects listed can have a major fiscal
likely to result in a less regressive (or more progressive) impact on a local jurisdiction. For example, if the
tax structure. If the aggregate level of taxes is un- value of open land increases near the new development,
changed, lower-income households would have a slightly property tax receipts increase without a concurrent in-
higher after-tax income than previoUSly.29 If residents crease in local expenditures. The geographic size of a
of new developments cause a shift of expenditures community and the intensity of the development and
from housing and hospital care for the elderly to ex- its location within the jurisdiction determine, to a con-
panding services in public schools, the middle- and siderable extent, to what degree secondary effects spill
higher-income new residents may have lower outlays for over to nearby jurisdictions.
otherwise privately financed after-school activities, Changes in Value of Surrounding Property
29. There are other, more significant nonfiscal impacts of
new development on income. These include higher employ- One approach for estimating the impact of a recent
ment rates and higher wages. development on land values is to compare before and

28 Fiscal Impacts of Land Development

after sales and/or assessment data for similar parcels housing. New residential development has not, how-
of land, taking into account inflation and other factors. ever, resulted in expanding commercial facilities signif-
Frequently, a professional appraiser or realtor active icantly within some New Jersey communities. Xsurvey
in the land market can estimate whether a new develop- of PUD residents in the state has shown that 80 percent
ment enhanced or reduced the value of surrounding do most of their shopping at regional shopping malls,
land and improvements, based on his or her knowledge which are frequently at a much greater distance from
of land and housing transactions. In addition, cross- their homes than the older "downtown" areas or smal-
section analysis has also been applied to estimate the ler suburban shopping centers [73]. This pattern may be
likely impact of new developments on nearby residential typical of communities with small land areas. A com-
property [851. This research shows that, in one large prehensive methodology to estimate the level of addi-
metropolitan area, the construction of apartments re- tional commercial activity likely to be generated by new
duced the value of detached housing nearby while housing is discussed as part of an impact analysis study
the construction of detached housing increased the [2]. A simpler approach is to aggregate total revenue
value of the existing predominantly single-family hous- which accrues to a community from commercial de-
ing stock. velopment, subtract the direct cost of providing services
While land values contiguous to new developments to commercial facilities, and divide the balance by the
can be expected in most cases to rise in anticipation number of housing units to obtain net commercial
of further development, particularly intensive develop- revenue per household. This approach provides a rough
ment, this may be offset by a proportionate reduction measure of the fiscal importance of commercial facili-
in relative land values elsewhere in the community or ties in the community based on historical residential
region if total demand for housing and commercial and commercial development patterns.
facilities is unaffected. In this case, increases in land
value would represent, a locational shift rather than Increased Demand for Housing and Services to
a net change. The validity of this assertion remains Households as a Result of Added Industrial
to be tested empirically. Development
Increased Demand for Goods and Services Most communities are anxious to attract 'new industry
Resulting from Residential Development in anticipation of fiscal surpluses. It is assumed, in
most cases correctly, that the cost of providing services
It is feasible to estimate the level of additional such as police and fire protection to a new facility is
demand created for goods and services if the house- less than local tax revenues which are likely to
hold income of new residents is known. These income accrue to the municipality. Frequently ignored is the
data, in turn, can be converted into estimates of in- fiscal impact caused by industrial development when
creases in the square footage of commercial space re- labor force requirements are not met from the existing
quired, and resulting additional net property and sales labor pool within commuting distance of its location.31
tax revenues. @ In making such estimates, it is necessary The resulting level of inmigration attracted into the
to consider what proportion of the new demand is community with new facilities in comparison to other
likely to be met by business enterprises within the com- areas within commuting distance is a function of housing
munity, compared to the share of demand "leaking out" at various price levels, land use controls, and accessi-
to other jurisdictions. In addition, already present bility of the new development to nearby communities.32
excess commercial space, built in anticipation of future A study of new industry in Kentucky shows that in
growth, should be taken into account. The proportion six of eight cases examined, the net fiscal impact, due
of additional sales volume not accounted for by higher to the inmigration of new households with school-age
disposable income of old and new residents or by likely children, was negative [16]. The communities, located
purchases from residents of other jurisdictions represents in rural areas, would not have attracted new house-
a shift in the location of new sales from another part of holds in the absence of the industries locating there.
the community rather than a net increase.30 A similar study of rural industrial development in
It is likely that some convenience goods retail stores
and gas stations, if zoning permits, will locate near new 31. The proportion of employees commuting among juris-
dictions in metropolitan areas can be obtained from the 1970
Census of Population. For an application of these data as part
30. In most communities, newer shopping centers reduce of an economic impact analysis, see: [21].
the volume of business in older shopping areas. While this 32. Given high vacancy rates, new housing demand may
shift has been common from central cities to suburbs, it also not be created, but service demand by households will never-
takes place within jurisdictions. theless increase.

Fiscal Impact Analysis-Major Issues 29

Oklahoma concludes that three new facilities caused a these transfers represent only a fraction of total retail
deficit to local government, nine showed a fiscal surplus employment. Substantial inmigration and additional
[22]. However, both deficits and surpluses tended to be housing demand as a direct result of new retail stores
small. is therefore unlikely.
An older study which examined new industry in sub- Office building tenants range from high salary pro-
urban areas shows generally favorable fiscal effects of fessional groups to firms employing large numbers of
these developments [18]. The magnitude of positive moderate wage clerical personnel. The number of
results was due to relatively few households moving to inmigrants depends on factors noted previously. The
the communities as a result of industry relocation. This proportion of commuters and residents by income, oc-
study concludes: "In order to improve their finances, cupation, and sex is shown in a study of highrise office
suburban municipalities should seek out and encourage buildings in San Francisco [211. The study concludes
expansion of industries which not only have a high that the type of employment influences the proportion
value of assessment per employee but also have em- of commuters to the city.
ployees who are not apt to relocate their residences in
the face of an industrial move." Increased Activity on the Periphery
The effects of various industrial sectors on a major of Development
local outlay-public schools-have been examined Major land use changes, such as PUDs and new
through the use of a model. The application of the communities, inevitably create additional development
model to a metropolitan area shows that expansion of contiguous to their location. This is part of the agglom-
five types of industries would create a fiscal deficit but eration effect noted in many urban studies. Even if un-
that expansion of ten other types would generate a improved land areas within the boundaries of the new
surplus to school districts [ 17]. development are sufficiently large to absorb future
What primary and secondary fiscal impact can Com- demand-as in Disney World, Florida and Reston,
munities anticipate as a result of new industrial develop- Virginia-development on the periphery nevertheless
ment? On the basis of examining studies referenced in appears to take place. This is usually due to controls
this section, the results are somewhat mixed. One of enforced by the owners and/or residents to preserve the
these studies, after noting that considerable private specific character of the development. A shopping
economic benefits -(higher wages and salaries) accrue center, for example, may exclude "undesirable" dis-
from industrial development, states that "disillusionment count stores, gas stations, motels, and fast-food stands
awaited communities seeking industry merely to expand which then try to locate nearby. This spillover
their tax base to support schools and municipal services" economic activity may include not only retail outlets but
[22]. Fiscal benefits from new industry in rural areas also certain types of housing not available within a PUD
appear minor, while frequently additional costs exceed or new community. There may be a substantial time
revenues, resulting in increased tax burdens for all resi- lag of two to four years between the initial development
dents. In the case of industry relocating within metro- and development on its periphery. In any event, the
politan areas, the fiscal results are likely to be more fiscal impacts need to be computed.
positive, since the likelihood of inmigration is reduced.
Increased Demand for Housing and Services Public Sector Capital Investments
Resulting from Commercial Development The previous discussion focused on private sector
The secondary effects resulting from the addition of effects which have fiscal implications. However, large
such commercial structures as shopping centers and new developments usually trigger a chain of local, state,
office buildings have not been the focus of fiscal impact and federal capital outlays. The largest outlays are
studies. However, studies of household location pat- for transportation facilities, particularly roads, and for
terns provide some insight into additional demand for utilities.33 These outlays, which generally follow growth
housing likely to be created. In many areas, new stimulated by private investments, create their own
shopping center employees relocate from older business momentum. For example, highway construction in
centers which are experiencing declining sales. Be- urban areas follows traffic levels which the existing
cause of comparatively low wages for sales personnel, transportation network cannot absorb. Once the public
and a high proportion of secondary employees (workers
other than heads of households), workers tend to be 33. The major planning tools available to influence the
pattern and location of growth remain the placement of water
local residents rather than commuters. While national and sewer lines, usually under local control, and the location
retai.1 chains may transfer some executive personnel, of major roads, usually a joint local-state decision.

30 Fiscal Impacts of Land Development

construction is initiated, additional private development In addition, rental apartment units are generally not
is encouraged, particularly near highway intersections maintained as well as owner-occupied property, which
and access roads. The addition of new rapid transit has the effect of reducing their value. For these and
facilities can redistribute growth to sites near transit related reasons, the initial difference in assessed value
stations. Such redistribution, however, is not likely to between single-family homes and apartments is likely
have a significant aggregate regional effect. to increase over time, a factor which needs to be incor-
porated into a fiscal analysis.
TIME HORIZON Investment property, including most apartments, is
In most cost-revenue studies, the levels of revenues usually not assessed on the basis of its market value, as
and expenditures are estimated through the year when is owner-occupied property. Gross rents or net cash
the last phase of construction is to be completed, or for flow, in the case of residential property, and gross sales
an arbitrary period such as ten years. Usually, the or value of product manufactured, in the case of com-
base data at the time the analysis is undertaken (for mercial and industrial property, are frequently used as
the development and community) are extrapolated for the basis for tax assessment purposes. These can vary
the period selected .31 substantially over the long run. Since assessment
This type of analysis has severe limitations, as practices for business property vary widely among
revenue and expenditure levels are likely to shift over communities, few generalizations are possible when
time according to the type of development and the long-term revenue estimates are attempted. In general,
historical assessment patterns need to be examined for
characteristics of population it attracts. However, since various land uses to project future revenue flows.
the time span selected for fiscal impact analysis is as An additional factor involved in the long-run projec-
much a political as an economic decision, and political tion of revenues from real property is the frequency of
pressures tend to reflect short-term effects, there is reassessment. It appears that many assessors adjust
more interest in next year's tax rates than in tax levels taxes to reflect higher rent receipts more frequently than
for the following decade.35 Nevertheless, specific kinds they respond to the appreciation of single-family units.
of long-term shifts that are likely to affect revenue and This may be a result of less concern with the reaction
expenditure estimates should be taken into account. of renters who are not conscious of the linkage between
Revenues taxes and rent payment levels, and, politically, less con-
cern with apartment dwellers who are not as likely to
Property appreciation (and depreciation in a less vote in local elections as people residing in owner-
inflationary economy) tends to be a long-run effect. occupied units.
There is considerable evidence that the rate of change in Income-related revenue sources, although not directly
property value-which clearly has an impact on the lo- linked to real property, may also be imputed from shifts
cal tax base-varies by type of land use. For example, in property value. If housing units are physically de-
single-unit detached homes tend to appreciate more teriorating, the income of their residents is likely to drop
rapidly in value than multiple unit structures.36 This relative to the income of people living in otherwise
pattern is due, at least in part, to initial differences in the similar but well-maintained housing. Concurrently, if
value of land in proportion to the total property value. a new development becomes socially desirable, the in-
Since land appreciates at a much more rapid rate than come of its residents (in constant dollars) will rise
improvements (buildings are likely, in fact, to depreci- over time.
ate in constant dollars), this is a significant factor in ex-
plaining why detached housing units appreciate more Expenditures
rapidly than townhouses and multi-unit dwellings [771. Future expenditure patterns are strongly influenced
by shifts in the demographic characteristics of residents.
34. See, for example: [9]. Although data on this issue are limited, it has been shown,
35. Long-term projections can take into account demo- for example [271, that in Prince George's County,
graphic changes of the population, but are subject to greater Maryland the number of children attending public
uncertainties. These uncertainties include technological changes,
life style changes, and shifts in intergovernmental respon- school per unit from apartments increased over time as
sibility for public services. the proportion of apartment units to all units increased
36. Land in suburban areas has been appreciating at 15 [27]. A study of New Jersey suggests that the number of
percent or more per annum in the last decade, or more school children from townhouses increased by 60 per-
rapidly than housing values. However, land is a greater share
of residential real property in densely populated Washing- cent between the initial date of occupancy of new units
ton, D.C. than in surrounding low-density suburban areas. and "maturity" of the units [73]. In contrast, a study

Fiscal Impact Analysis-Major Issues 31

of apartments over a ten-year time period in New Jersey for larger families. Costly services to this elderly house-
[581 indicates that the number of school children per hold, specifically medical and nursing care, are provided
unit has not varied.37 by the federal government. Thus, this household may
Detached housing units seem to follow a more con- again provide a fiscal surplus to the community.
sistent pattern, with more school-age children at the If this household resided over an extended time
time of initial occupancy. As a moderate income resi- period in the same community, with or without intra-
dential development matures, the number of high school jurisdictional moves, there would be little need to be con-
students compared to elementary school students will cerned if it yielded a fiscal deficit in a particular year.
increase .311 Finally, long-term projections of school en- However, there is a tendency within metropolitan areas
rollment are hazardous, since the birth rate is con- for households to locate in certain communities at
tinuing to decline. If this trend continues, there will be particular stages of their life cycle.40 This pattern,
a major readjustment in outlays for primary and second- strongly influenced by the availability of housing that is
ary education compared to other services. aimed at particular demographic markets, makes it
Per capita costs of local services such as public safety difficult to apply the life-cycle approach in estimating
.and general government seem to be rising faster over the impact of new development, except from a regional
time (in constant dollar terms) than private sector perspective.
services. This may be due to the lack of increase in As a result of complexities in projecting costs and
public sector productivity.39 As a result, wage in- revenues in the public sector, further efforts are required
creases, which have been rising rapidly, particularly in to develop knowledge about long-term benefits and
large cities, generally are not offset by higher output losses to a jurisdiction from a development. Until
levels, adding to the difficulty of projecting future costs. more data are collected on those long-term fiscal effects
Capital costs have also accelerated more rapidly than in growing communities, projections over an extended
operating costs in recent years. A community should time horizon are of only limited value.
examine the rate of inflation for construction and an-
ticipated increases in salaries for its municipal employees FISCAL INTERACTIONS AMONG JURISDICTIONS
in projecting long-term fiscal effects of new develop- Since local communities are not closed economies,
ment. many activities of individuals, firms and public agencies
in adjacent jurisdictions affect the fiscal impact of a
Life Cycle of Household new development on a local government. The type and
If service demands are viewed from the perspective size of the development will have a bearing on the
of a household, these vary sharply over its life cycle. As geographic extent and intensity of these spillover effects.
a young household without school children, the family For example, additional sales taxes from a new shopping
is likely to produce a small fiscal surplus to the com- center depend on the consumption level of shoppers
munity. During the period when the family has one or from other areas as well as on the increased propensity
more children in public schools, it is likely to receive of local residents to shop within a jurisdiction .41 Con-
more in services than it pays in taxes. After these currently, development activities in nearby communities
children complete their public education, which often have a fiscal spill-in effect. Frequently, a community is
coincides with the peak earning years of the family, the affected more by developments beyond its jurisdictional
same household will produce a substantial surplus to control than by those within its boundaries, A large in-
the community. dustrial facility located in a nearby community can
When this household becomes "senior citizens," its accelerate demand for new housing or services, such as
income is reduced. However, property taxes paid by the transportation. Conversely, the fiscal impact of non-
household tend to remain high (unless exemptions are residential development within the jurisdiction depends,
provided) since it may remain in a housing unit designed to a considerable degree, on the housing opportunities
for new employees outside the community. This ex-
plains why certain communities find it desirable to
37. Since the birth rate was reduced over the time span of
the analysis, there was an increase in school children com- 40. The segmented housing market is most visible in Southern
pared to detached housing. California where there is housing for singles, young married
38. This results in higher unit costs since per pupil outlays families with young children, mature adults, and for the elderly
for high school students are about one-third above those for ("leisure" developments).
elementary pupils. However, total enrollment will be reduced. 41. In a closed economy, a new shopping center could result
39. For a theoretical discussion of cost increases in the in little net change in tax receipts-only a shift by consumers
private sector, see: [47]. from older shopping areas to the new center.

32 Fiscal Impacts of Land Development

attract new industry, but not necessarily new house- development in certain areas over others. In most states,
holds, since they believe the latter may adversely affect low-density rural areas benefit more from highway aid
the fiscal surplus from new business enterprises. than do growing urban centers.
Within a metropolitan area, or within suburban com- Fiscal interactions among neighboring jurisdictions
munities which have open space remaining, a consider- and various levels of government should be incorporated
able share of new growth may merely represent a shift into all fiscal impact studies, with special attention to the
of population from the central city to its suburbs or impact of new development on state aid and federal
from older to newer suburban areas. Thus, what is fis- revenue sharing.
cally beneficial (or detrimental) to a community may
not be necessarily beneficial (or detrimental) from a FACTORS WHICH LIMIT THE SCOPE OF
regional perspective. As noted earlier, one study has FISCAL IMPACT STUDIES
shown that a development caused a fiscal deficit at the Why do so few studies consider the broader issues of
local level but a surplus at the state level [5]. As the size fiscal impact?
of geographic areas under study expands, spillover and Factors which may account for the scant attention
spill-in effects are reduced. As a result, fiscal impact, these issues have received include the following:
viewed from the state or metropolitan perspective, is less
sensitive to spillover and spill-in effects than is an Lack of Conceptual Framework
analysis that focuses narrowly on local effects. The impact analyses reviewed are ad hoc, with little
A new development can cause shifts not only in local attempt to formulate a comprehensive framework show-
revenue and expenditure flows, but also in state revenues ing interdependencies among various effects or to draw
and outlays to the local community. Therefore, in order on relevant social science disciplines to provide a theo-
to determine the net fiscal impact of a development on a retical construct.
community, both local and state fiscal flows affected by
growth need to be computed. For example, in New Insufficient Analyses of Empirical Data
Jersey, there is theoretically an equilibrium in state-local Facts about scale economies, preference shifts, and
education funding: state education aid drops as a re- so forth have not been gathered or studied with enough
sult of increases in per pupil property value, but local care to demonstrate precisely how these important
taxes derived from increased property values are sup- aspects affect costs and revenues. While differences in
posed to counteract the loss; Thus, a school district the tax structure and expenditure patterns among com-
should be fiscally neither better nor worse off by shifts munities have been examined and reported in the public
in per pupil property value as a result of new develop- finance literature, the results have not been widely in-
ment. However, in actuality, local governments in the corporated into cost-revenue studies.
state have been receiving only half of the state equaliza-
tion aid. School districts taxing their own properties are Shortages of Resources
therefore better off than those receiving aid. This en- To undertake cross-sectional studies and household
hances the attractiveness of high property value develop- surveys requires considerable resources. Most sponsors
ments to local school districts. Although most states are unwilling or unable to provide the level of funding
have some form of equalization based on differences in necessary to undertake these large-scale efforts.
property value, the situation in New Jersey is typical-
state aid does not suffice to put districts with low per Sponsor Objectives
pupil property value on a par with wealthier districts. Frequently, the interest of study sponsors is limited
Consequently, lower-income housing is discouraged by to short-term cost-revenue effects. At worst, there may
present state distributional formulas for educational aid. be concern that the results of a more comprehensive
State distributional formulas for highway construc- analysis would not be in the self-interest of the sponsors;
tion, mass transit, and other services tend to encourage at best, the sponsor's perspective is simply too narrow.

Fiscal Impact Analysis-Major Issues 33

V. IMPLEMENTATION AND USE OF
FISCAL IMPACT ANALYSIS

The reader of this report should not conclude, on the These communities can compare the characteristics
basis of the many limitations of fiscal studies reviewed, of newer and older residents. In most cases, it is found
that estimating the effects of new development in a par- that residents of new detached housing units have more
ticular community is not a feasible task. school-age children, have higher incomes and are some-
what younger than the community average. These kinds
GUIDELINES FOR INITIATING STUDIES of data, in turn, can be used to estimate the likely fiscal
impact of these households. These general data need to
The sections that follow briefly outline general guide- be followed up by more intensive surveys of new de-
lines for implementing a fiscal impact study, discuss fac- velopments with differing housing mixes. For a sample
tors to be considered in the selection of developments of these households, the consumption level for public
for analysis, and note some nontechnical issues. Ap- services other than schools should be ascertained and
pendix A discusses procedures for undertaking a fiscal compared to service usage levels in older areas of the
impact analysis in more detail. community. It can be assumed that residents of pro-
General Approach posed developments will have service consumption pat-
terns similar to those of residents in newer developments
To implement cost-revenue studies of residential surveyed, if housing characteristics do not differ sig-
developments in communities which do not have de- nificantly. The public cost of capital facilities associated
tailed data on the cost of new housing, it is first nec- with new development, since it can vary substantially
essary to establish a communitywide data base on by intracommunity location, requires direct estimates
revenue sources, capital costs, and operating expenditures from local agencies.
for major land use categories. Data on the socioeco- It can usually be assumed initially that the operating
nomic and demographic characteristics of the present cost of providing a unit of service (one student attend-
population by housing type and length of residence ing school, one police department service call) for new
are also needed. Part of the data on residents is residents will be similar to the average cost for all pres-
available from the 1970 Census of Population. How- ent residents. However, the quantity of units (number of
ever, since these data quickly become obsolete, they school children, or fire calls) utilized per household
need to be updated on the basis of local surveys. Two should be based on data from similar developments. If
large suburban counties have recently completed such other analyses undertaken in the community demonstrate
surveys. Montgomery County, Maryland undertook such significant scale economies or diseconomies as the num-
an update in 1974 by the use of a mail survey [65]. ber of units serviced increases, the incremental unit cost
Similarly, Fairfax County, Virginia utilized the state- should be adjusted to reflect these differences. As was
required school census of households as a means to noted in Chapter 4 (pages 19-33) of this report, there
obtain socioeconomic and demographic data linked to is considerable evidence of scale diseconomies for many
housing type and length of residence in the county. services as the size and density of the community in-

creases. The available data suggest, however, that pop- ment. Cross-sectional analysis provides insight into over-
ulation density (or changes in population characteristics all revenue and expenditure patterns among communi-
associated with density) appears to be a more important ties which have some common attributes, but it cannot
factor in higher per capita costs than population size. reflect the many variables which influence the expendi-
Therefore, scale effects should be considered if new ture patterns in a given community. Data from such
developments are projected to significantly change the analyses can be useful when considering broad alterna-
population density as well as the total population of tive growth strategies rather than for estimating the
the community. impact of an individual development. Time-series data
To implement studies of commercial and industrial are also of limited value when projecting the cost of
developments, it is first necessary to determine how such individual new developments.
property is assessed for tax purposes. In addition, the Econometric models are poteniially powerful tools
likely level of other business taxes should be identified. in communitywide cost-revenue analyses. However,
Public services consumed by these developments, further research and testing of existing techniques is
such as public safety and utilities, can be estimated required. Staff requirements limit the utility of this
from local agency records. approach for most local governments.
To estimate secondary effects of industrial and resi- When to Implement an Analysis of
dential employment requires data on the likely number Individual Developments
of residents, commuters, and imnigrants to the com-
munity which can be directly linked to new facilities. The availability of an adequate data base and access
In most cases, it is necessary to survey facilities already to trained personnel are necessary prerequisites for
constructed to project the likely residence of new em- undertaking fiscal analyses. However, since a growing
ployees, their characteristics, and the level of inmigra- urban area can be expected to have many proposed land
tion to the community. use changes, it would be inefficient to undertake a com-
Since all new development is likely to affect the plete-in contrast to a routine-analysis of each pro-
amount of revenue received from the state and federal posed development. The political importance of the
governments, this factor should be considered when fiscal issue will determine, in part, the level of effort.
evaluating both residential and nonresidential develop- The following other factors should be considered in
ment. selecting developments for fiscal impact evaluation:
Routine estimates of fiscal flows associated with The degree to which a new development differs,
small developments which are similar to existing de- either in the type of housing, density, unit value,
velopments can be obtained from expenditure and or likely population characteristics from the exist-
revenue data for these existing land uses and housing ing housing stock. A small detached-unit develop-
types. While these estimates are likely to differ some- ment in a predominantly residential community
what from actual fiscal flow, the degree of these devia- is unlikely to create a fiscal problem. A change
tions is probably not significant. in population characteristics, however, can shift
This discussion has placed emphasis on micro-level service demand from the existing pattern and thus
data on households and business firms in the commu- affect aggregate expenditures.
nity, rather than reliance on estimates from studies of 9 The number of additional units as a percentage
other communities, aggregate data, or the application of of total housing in the community. If a proposed
community standards to project service costs. Estimates development is expected to include, say, 15 or
by local officials of likely service demand from new de- 20 percent or more of the total residential hous-
velopments should be used to compare their estimates ing stock, it could result in fiscal changes even
with data obtained from the analysis of household and
firm characteristics, rather than as the principal source if the type of housing does not deviate from the
of information. If major deviations in estimates are community average. A development which sharp-
evident, the differences should be discussed with local ly increases the population of a community is
personnel to determine if factors causing estimates of likely to create scale economies and diseconomies
projected demand to differ can be identified. This which require examination.
process can be useful to both the analyst and the The fiscal effects of new industrial and commercial
official. development, if the household income of the pro-
The macro-level techniques discussed in this report jected labor force will differ substantially from
are more useful in examining the aggregate effects of the community average and if inmigration is
growth rather than the impact of an individual develop- likely.

36 Fiscal Impacts of Land Development

If a community initiates an analysis of the environ- has control over the proposal being considered but not
mental or social impacts of a proposed development, at the most likely alternative, the approach should differ.
least a rudimentary analysis of the fiscal impacts should For example, a new shopping center proposal may
also be included. The latter would help provide a more depend on the grant of a rezoning from moderate
comprehensive view of the effects of the development density residential to commercial land use. If rezoning
being considered. is not granted, one can anticipate that townhouses,
A large community with an established data base allowed under present zoning, will be built. In this sit-
should periodically sample smaller new developments uation, it is advisable that the fiscal and other impacts
to determine whether characteristics of new housing, of townhouses be compared with the shopping center
its residents, and the consumption of public services impacts.
deviates from previously found patterns. If such devia- -Local government should, in some cases, consider the
tions are shown, it may require that existing relationships fiscal and other effects of alternative development pat-
(such as house value to income, school children per terns, particularly at the area plan level. Many com-
bedroom) be adjusted. munities have adopted policies to provide a broad spec-
Nontechnical Issues trum of housing to meet the needs of present and
future residents. To meet this objective, it is useful to
A number of nontechnical questions frequently arise determine the pattern of growth which can provide
when a community undertakes an impact evaluation. housing that reflects the preferences and ability to
Should incremental one-by-one analysis be undertaken? pay of various population groups and that can make
Are the skills for such an analysis available or obtain-. such housing available at the lowest possible public and
able? Should proposal alternatives be considered? private costs.
-Urban planners state, with considerable justification, -Local officials, planners and interested citizens should
that evaluating the impact of individual projects is not also be aware of issues associated with fiscal impact
a subsfitute for land use planning. We recognize, how- studies, such as scale and income distribution effects of
ever, that local government is legally required, in many development which have been discussed in this report.
instances, to reach one-by-one decisions on individual
parcels of land. The techniques described in this report PRESENT USES AND POTENTIAL MISUSES
and in referenced studies can provide the data base nec-
essary to evaluate most fiscal implications associated From a policy perspective, the importance of cost-
with alternative comprehensive plans. The fiscal impact revenue or fiscal impact studies lies not in their
of individual residential developments within the bound- methodological niceties or theoretical symmetry, but
aries of a comprehensive plan can also be evaluated rather in the ways they are used or misused in the land
from this data base. use decision process.
-A frequently observed problem is that while planning It would be unrealistic to expect local officials to
departments have the responsibility for presenting accept study findings about fiscal effects out of context,
factual data for land use decisions, the line agencies without reference to other development effects. For ex-
have more technical expertise and access to the nec- ample, the kinds of neighborhoods and new development
essary information. The staff weakness of most planning that present residents want to encourage may be given
departments is particularly evident in economic analysis. more weight than purely fiscal considerations. In New
Ideally, a member of the planning staff should be an Jersey, many communities restrict or severely limit the
economist with a background in public finance and ur- number of apartments that can be built, although most
ban economics. In all but very large departments one local officials recognize that a favorable fiscal impact
person can be responsible for economic analysis. A less would result from new multifamily dwellings, particu-
effective alternative is to establish links with a nearby larly highrise apartments. They limit such development
university or private firm for periodic assistance. because of the opposition of residents, most of whom
-Should an impact evaluation estimate the effect of the live in single-family units and wish to maintain a homo-
development being proposed in comparison to no land geneous community.'
use change, or to the most likely alternative land use if
the proposal is not accepted? If local government has 1. The views of residents and community leaders on per-
the option to reject or modify both the proposal under ceived fiscal and social effects of apartments are discussed in
consideration and the most likely alternative, then the [19]. The importance attached by both residents and the courts
to maintaining a homogeneous community is reflected in the
anticipated results of no change in land use should be recent Supreme Court ruling on the constitutionality of a
used as a comparison. However, if local government single-family dwelling ordinance in Belle Terre, New York.

Implementation and Use of Fiscal Impact Analysis 37

To elected officials, one crucial fact about potential severe action. It is more likely that one of the following
residents who are being excluded is that they cannot options will be considered:
vote. * The development may be accepted as is. One
Officials at other times may use fiscal findings as a premise for following this option is that the deficit
kind of shield if there is opposition to a particular type will be offset by surpluses from other develop-
of housing and the type of people who are likely to ments. Or it may be accepted on the basis that
occupy it: if a study can project a possible fiscal deficit, nonfiscal social benefits outweigh the fiscal
this economic factor rather than the more fundamental deficits.
exclusionary sentiment will be cited as a cause for
rejection of the development. 0 The developer may be required to modify the
The local officials and planners who review fiscal proposal by changes in the housing mix or non-
impact studies must first evaluate the results from the housing land uses. For example, if a study shows
community's rather than the study sponsor's perspective. that three-bedroom apartments (in contrast to
Then, if negative fiscal effects are projected, they must one-bedroom units) cause a deficit, he could re-
determine whether there are other development options duce the number of such units. Alternatively, non-
that could offset such effects. residential facilities can be added or expanded.
The new development options facing local officials e A payment may be required from the developer
vary considerably, depending on the current stability to the community for units projected to cause
and strength of the community's treasury. The options a deficit. Such payments can be used to defray
clearly are related also to the jurisdiction's present sur- capital outlays linked to the development. For
plus or deficit of capital facilities. Communities that have operating outlays likely to cause a deficit, services
been forced to raise tax rates because of a sharp rise in could be priced in the form of user charges which
school enrollment, for example, are anxious to expand reflect the incremental cost of the service. This
their tax base by attracting fiscally beneficial com- approach has been suggested as a means of dis-
mercial and industrial development. Similarly, com- couraging growth in previously undeveloped ur-
munities with considerable outmigration, and thus with ban fringe areas [23].
underutilized public facilities, are also eager to attract
development; stagnating communities, as compared to * The proposed development may be required to
growing communities, are more likely to show a fiscal change its location to an area which has under-
surplus from additions of new households. utilized public facilities to reduce capital outlays.
In affluent communities and those anxious to main- * The level of public services may be reduced to
tain what they perceive to be an advantageous life style, maintain the existing tax structure and rates.
new development may be rejected even though its fiscal * The local tax structure may be revised in a way
benefits can be demonstrated. Even in these communi-
ties, prodevelopment arguments are more persuasive if that can turn public deficits from new develop-
accompanied by a projection of fiscal surplus. However, ment into a.public surplus.
when new development poses what residents consider to The position taken by the community regarding
be a social threat, certain forms of development are these alternatives depends on many variables, including
frequently restricted by the use of zoning or other its concern over social issues, its wealth, the need for
means. housing for its residents, and overall economic stability.
If a community anticipates future commercial or
Alternative Options for Dealing with Expectations industrial developments, it can accept deficit-causing
of Fiscal Deficits housing on the premise that the surplus from non-
Assuming that fiscal factors are of considerable residential development will offset the residential
importance to most communities, what options are deficit. The provision of subsidized housing can be
available if studies indicate a deficit from new develop- justified on its social rather than fiscal merits. A fiscal
ment? One alternative, of course, is to reject the appli- deficit can also be justified if the new economic activity
cation outright .2 This, even if legally permissible, is a will substantially decrease local unemployment. Thus,
residents may be willing to increase their tax contribution
to maintain present levels of service or subsidize new
2. In some communities, no reason has to be cited for re- firms if there are private sector benefits which contribute
jecting an application to rezone. Thus, fiscal deficit could be
the determining factor without an explicit statement to that significantly to the general welfare..
effect. The second option, such as changing the housing

38 Fiscal Impacts of Land Development

mix, is likely to reduce housing opportunities in the would not be passed on to consumers .4 If part of the
development for moderate-income families, particularly cost is absorbed by the consumer, moderate-income
larger families, and this has adverse social effects. A households will be adversely affected.
less restrictive approach is to modify the originally pro- The fourth option-a change in location for the
posed land use mix to include more commercial and in- development-can represent a substantial saving in
dustrial development, offsetting the negative fiscal flow capital outlay [28]. However, the option is unlikely to
from residential units. Such arrangements are part of be tenable because of differences in land costs and
the approval process for PUDs (planned unit develop- housing demand by location, the difficulties of assem-
ments) in New Jersey [10]. A similar approach is now bling large tracts of land in more developed parts of the
required by St. Charles County, Maryland for the ex- community, and the loss the developer may incur on the
pansion of St. Charles' new community [6 '6]. The long- land already purchased.
term fiscal and the other effects of these policies have The next alternative-reducing the level of public
not been evaluated. However, the experience in New services-,does not appear politically acceptable in most
Jersey points to the difficulty of attracting nonresiden- communities. The general response (with some notable
tial development to PUDs. exceptions) has been to increase tax rates rather than
The third option-requiring payments to the com- to reduce public services.
munity by the developer-is a practice being adopted Surprisingly little authoritative work has been done
by a number of jurisdictions. The legality of such re- on the reform of property taxes, commuter taxes, and
quirements has been questioned. On the basis of some local income taxes as these relate to development. The
court decisions it appears that such charges can be im- Minneapolis-St. Paul metropolitan area has adopted a
posed if the benefits from the use of those charges accrue system of tax base sharing to help spread the benefits
directly to the new residents. For example,, if funds are of nonresidential growth more equally among the
to be used for the construction of a new eleme 'ntary various jurisdictions and to minimize the pressures for
school attended by the children of the new residents, exclusionary and fiscal zoning. Several important studies
the charge can be imposed .3 underway should also shed more light on this option.
The effects of these charges are difficult to estimate, Other options for constraining or limiting develop-
although it is thought that the developer passes on at ment have been challenged in court. Policies which im-
least part of the cost directly to the purchaser or renter. pose a total limit on population or unreasonably delay
To the extent that the required charges would lead to a approval for new development may be open to challenge
reduction in the cost of land to the developer, the charge on constitutional grounds. A discussion of these legal
issues is, however, beyond the scope of this study.

3. There may be differences in court decisions among 4. The value of the land to the initial owner is reduced
states. However, the practice of requiring developers to deed since his net income is lower as a result of the cha rge. Sub-
some of their land for highway rights-of-way, schools, and sequent owners who purchase the property do so at a lower
parks, for example, is widely accepted. price and are not burdened with the charge.

Implementation and Use of Fiscal Impact Analysis 39

VI. SUMMARY- STUDY FINDINGS

EXPANDING USES OF FISCAL IMPACT STUDIES can cover projected public capital and operating outlays.
The major use of the fiscal impact studies that were Fiscal impact studies which examine service pricing
reviewed has been to help local officials determine alternatives associated with new development are at-
whether there was likely to be a surplus or deficit to tracting considerable interest. Some of these studies
their community as a result of new development. are aimed at determining whether present pricing
A second use of these studies has been to help mechanisms result in subsidies to new development
planners and policy makers formulate broad community- from older areas of a community.'
wide growth strategies. A number of other potentially
valuable uses are also emerging. These include, for METHODOLOGIES-STATE OF THE ART
example, comprehensive land use planning and capital Techniques applied in studies reviewed range from
improvement planning. simple, one-dimensional methods to complex econo-
Cost-revenue studies are being undertaken to examine metric models. The determination of which technique to
the fiscal impact of communitywide alternative develop- select depends on study objectives and available re-
ment patterns. Previous studies were concerned pri- sources. However, since there are only limited data
marily with the impact of density on capital outlays. available based on retrospective analysis, the reliability
However, communities are now undertaking studies of techniques reviewed have not yet been adequately
which will estimate the aggregate number of inmigrants assessed.
for each alternative plan along with their socioeconomic Given the limited state of knowledge, the most effec-
and demographic characteristics. Given these population tive approach is to estimate, directly by the use of
data and the anticipated level of nonresidential develop- surveys or indirectly from secondary sources, the likely
ment, revenues, capital outlays, and operatin 'g outlays demographic and income characteristics of new residents
can be estimated. The results will indicate which by type of housing. These data can be applied to esti-
development pattern is most acceptable from a fiscal mate both revenues expected to accrue and anticipated
perspective. This information, combined with projected demand for public services.
nonfiscal effects, will provide local decision makers with To'examine the fiscal impact of alternative growth
implications of alternative growth strategies. patterns or development in general, it is useful to apply
A more modest use of fiscal impact studies is re- time series data from the same community or cross-
lated to capital improvement programs. The size, loca- sectional data from similar jurisdictions. It is essential
tion, and timing of new facilities is linked to the level of that likely incremental costs for new households be de-
anticipated demand. Concurrently, the timing and loca-
tion of private development is based, to a considerable
degree, on the availability of public facilities. Public and 1. The National Science Foundation is presently funding an
private planning should be inte rated in such a manner effort by the Center for Urban and Regional Studies at Vir-
g ginia Polytechnic institute to examine service pricing alterna-
that the revenue flow-estimated from fiscal studies- tives in growing suburbs.

termined. The emphasis should be on empirical data many inmigrants. However, new shopping centers fre-
from similar development rather than on simulated quently reflect a shift in the location of retail activity
values. rather than increased aggregate sales, offsetting part of
the surplus. Office buildings, particularly those utilized
WHY FINDINGS MAY BE INCONSISTENT by professional groups, are likely to provide more
A major factor which explains inconsistencies in revenues than the cost of services consumed.
results from one study to another is the objective of the Industrial developments are found to have a mixed
sponsor, which affects the scope of the analysis, the effect when secondary impacts, particularly inmigration,
inital assumptions and, in some cases, the techniques are considered. Capital intensive facilities with high-
selected. The conclusions of studies sponsored by pri- salaried employees provide more revenue than the cost
vate developers frequently differ from those sponsored of services, while labor intensive facilities with wages
by nonprofit organizations or the public sector. close to or below the community average cause a deficit.
Differences in allocation of service costs among land The above discussion focuses on individual develop-
uses and between new development and the balance of ments. At the aggregate, community level, a balanced
the community can also affect findings. budget implies that residential developments collectively
Yet, even if the same initial assumptions are made will produce a deficit. This deficit is offset by nonresi-
and the same techniques are applied to similar develop- dential land uses-industrial, commercial, or agri-
ments in different locations, results may vary between cultural. Surpluses or deficits from any one develop-
locations because of variations in state-local fiscal struc- ment are absorbed among all property by adjustments
tures. Intracommunity location of new development, be- to the tax rate or changes in outlays for public services.
cause of differences in infrastructure required and other
causes, also affects public outlays and thus the fiscal
flow associated with a particular development. FISCAL IMPACT ANALYSIS-LOOKING
TO THE FUTURE
FISCAL IMPACT OF DIFFERENT Growth theories and analytic techniques have not
DEVELOPMENT TYPES been sufficiently' developed to incorporate consideration
Most new detached housing developments, particu- of some issues which can have a major long-term fiscal
larly during the 1960s, produced more revenues in effect.
comparison with the existing housing stock. Dur- Effects of Increasing Population Size and Density
ing this decade, the average square footage of new
housing, and thus the selling price in constant dollars, There is substantial evidence that there are dis-
increased substantially. Many of the new residents in economies of scale for many services provided by the
suburbs, where most new development has taken place, public sector when the size and density of a community
were higher-income outmigrants from central cities or reaches a threshold level which is at some point below
migrants with above-average income from other metro- 100,000 residents. Private benefits in larger areas prob-
politan areas. Their property and other tax payments ably offset increased public costs, but the percentage
exceeded that of the base population. Concurrently, of income allocated for public services is higher in large
their consumption of services, particularly education, urban areas compared to smaller communities. How-.
also exceeded the community average. Thus, both rev- ever, a reduction in population and density due to out-
enues and expenditures from new detached housing have migration can create even greater diseconornies. The
been above the level of older housing. cumulative fiscal effects of many one-by-one decisions
Garden apartments frequently create a surplus be- have not been sufficiently evaluated to reach conclusions
cause of the low number of school-age children com- regarding long-term fiscal impacts.
pared to those in detached housing. In communities
where these apartments create a fiscal deficit, it is usually Tax and Service Preferences
smaller than the deficit from single-family units. High- The preferences for both tax and service "packages"
rise privately-sponsored apartments and condominiums change as a community grows from a rural to urban
are usually aimed at young professionals and the stage. This is linked to changes in population charac-
elderly. Since these groups are unlikely to have school- teristics of new residents and to physical changes, such
age children, surpluses are consistently shown. as higher density, which make it more efficient to have
Commercial developments, such as shopping centers, the public sector provide additional services. There-
create a fiscal surplus since they are unlikely to attract fore, fiscal data gathered prior to those changes is un-

42 Fiscal Impacts of Land Development

likely to be appropriate for estimating current costs significantly to this issue of both local and national
and revenues. interest.
Spatial and Income Distribution Effects Redevelopment Projects
The impact of residential development whose mix of Almost all fiscal impact studies to date involve
housing differs from the community pattern does not communities growing in population and expanding in
significantly change the aggregate regional demand for general economic activity or large developments on the
services, since regional migration patterns are not sig- periphery of urban centers. This is a reflection of the
nificantly affected. Thus, service demand is redistributed concern by residents of such communities that rapid
rather than altered within the region or, in some cases, expansion may have adverse effects. It also reflects
within the jurisdiction. The income distribution of resi- growth pressures as a result of outmigration from central
dents can be affected by shifts in tax and service prefer- cities and inner suburbs, as well as interregional migra-
ences. tion.
Central cities, however, also have the capacity to
Secondary Impacts absorb a considerable amount of growth by redevelop-
, In addition to direct effects of development, a variety ment, rehabilitation of existing structures, and other
of secondary impacts or chain reactions need to be con- mechanisms. Since, as has been suggested, the fiscal
sidered. The linkage between the primary impact of new impact of a given household varies by its location within
industry,, the subsequent demand for additional hous- an urban area, the extent of the comparative fiscal
ing, and the need to expand public services for new benefit which results from a household locating in a
residents are part of this process. rehabilitated structure in a central city rather than in a
new unit in a suburban area should be explored. The
AREAS FOR FURTHER STUDY cost-revenue approaches will require some modification
Relatively little attention has been given to fiscal in cases where urban redevelopment is to be financed
implications of certain specific kinds of development, or by both private and public funds.
to the fiscal effects of development from a central city Retrospective Evaluation of New
and regional perspective. Such topics that require further Communities, PUDs, and Large Developments
study are discussed briefly below. A number of new communities, PUDs, and large de-
velopments for which initial fiscal impact studies had
Industrial Developments been undertaken have now been completed for a num-
Despite large-scale efforts on the part of many ber of years. These communities offer a potentially
communities and states to attract new industry, the large data base that could be used in studies to deter-
fiscal implications of adding industrial facilities have mine whether prior predictions of impact were realistic.
not been thoroughly investigated. Specifically, the link- The data also could be used in developing a model
ages between new industrial facilities and changes in for future evaluations.
income, demand for new housing, the costs of providing These developments would be examined at the micro
services to industry and additional households, and the level to estimate changes in the characteristics of the
additional revenues generated, particularly in metropoli- occupants by type of housing unit over time. For exam-
tan areas, require additional explanation. A comparision ple, one would examine a detached housing develop-
between the fiscal impact of attracting new industry on ment built in 1968, estimate (from mortgage company
central cities and on their suburbs would also be par- records or fiscal studies) the demographic and socio-
ticularly useful. economic characteristics of its original residents and the
original property value, and compare this information to
Pattern of Development the characteristics of present residents and current
A number of studies referenced in this paper have property values. Similarly, the change in composition
noted the fiscal impact of alternative development pat- and property value of commercial development centers
terns. However, these studies are restricted by assump- over time would be examined.
tions which may not be valid. While it is often assumed The examination of changes in household charac-
that there are differences in both capital and operating teristics within one or more large developments should
expenditures related to a particular pattern of develop- be parallel with a review of changes in public serv-
ment, little empirical data have been collected to dem- ice outlays, public employee personnel, and service
onstrate the validity of the simulated expenditure levels output measures for selected functions during the same
used. More comprehensive studies could contribute time period.

Summary-Study Findings 43

Regional Studies munities adopt various explicit measures to control
A fiscal impact study of selected metropolitan regions growth, in addition to implicit controls by zoning, it
should be aimed at determining differences in revenues has become more and more important to understand
and public services among communities of different the fiscal implications of such policies on the surround-
sizes and growth rates. The foundation for such an ing region.
effort already exists in previous research. The emphasis Policy Effects of Fiscal Impact Studies
of such a study would be the choice of communities by
residents, the fiscal implications of these decisions on Although considerable sums of money are allocated
communities, and the most efficient location of new by both the public and private sector for fiscal impact
households within the community or region from the studies, it is not known what influence these studies
public service cost perspective. have on land use decisions. A follow-up of selected
Another useful kind of regional study involves de- publicly and privately sponsored studies undertaken to
termining the fiscal impact of constraining development aid in the land use decision process would be useful to
in one community on surrounding jurisdictions. As com- determine their effect, if any, on that decision process.

44 Fiscal Impacts of Land Development

BIBLIOGRAPHY

This is a partially annotated bibliography of works 1973. General computer simulation model with
dealing with the fiscal impact of land development. It is fiscal submodel. Sponsored by HUD.
divided into the following categories: 4. Gale, Dennis E., The Municipal Impact Evalua-
A. General methodologies tion System, American Society of Planning Of-
B. Case studies of individual developments ficials, Report No. 294, September 1973. Com-
C. Similar developments in several locations puter-assisted procedure to examine revenue and
cost data. Sponsored by ASPO for local govern-
D. Communitywide studies ment use.
E. Regional studies 5. Muller, Thomas and Dawson, Grace, Fiscal Im-
F. Fiscal impact of alternative development patterns pact of Residential and Commercial Develop-
G. Studies examining demand for and cost of public ment, The Urban Institute, 1972. General
services methodology with application to a PUD in Vir-
H. Other referenced studies ginia. Ford Foundation funding.
1. Other bibliographies 6. Northern Virginia Planning District Commission,
An Approach to Financial Benefit-Cost Analysis,
The items in the bibliography are numbered and are 1973. Outline of a general methodology, with
referred to in the text, with appropriate numbers in fiscal data for Northern Virginia, for use by local
brackets (in contrast to the superscript numbers which governments.
refer to footnotes). 7. Stuart, Darwin G., and Teska, Robert B., "Who
A. GENERAL METHODOLOGIES Pays for What?" Urban Land, March 1971. Dis-
cussion of factors which influence fiscal impact of
1. Applied Decision Systems, Inc., The MSHDA residential and nonresidential development.
Community Impact Model, 1972. Part of com- Sponsored by the Urban Land Institute.
puter simulation system for Michigan State Hous-
ing Development Authority. Sponsored by HUD. B. CASE STUDIES OF INDIVIDUAL
2. Connecticut Development Group, Inc., "Cost DEVELOPMENTS (INCLUDING PUDs AND
Revenue Impact Analysis for Residential De- NEW COMMUNITIES)
velopments," September 1974. Detailed step-by- 8. Ashley Economic Services, The Fiscal Impact of
step procedures for local government to follow in Urban Growth-The California Experience, 1973.
estimating fiscal impact. Sponsored by HUD. Collection of fourteen fiscal impact studies, in-
3. Decision Sciences Corporation, Advanced New cluding residential and nonresidential develop-
Community Simulation Systems (NUCOMS), ment. Sponsored by California Builders Council.

9. Booz-Allen Hamilton, The Economic Impact of ban Communities," National Tax Journal, June
Reston on Fairfax County, Virginia, 1973. 1963. Fiscal impact of industry relocating within
Projected costs and revenues associated with the metropolitan area.
expanding new town of Reston. Study sponsored 19. New Jersey County and Municipal Government
by developers of community. Commission, "Planned Unit Developments in
10. Burchell, Robert W., Planned Unit Develop- New Jersey: A Social and Political Investment,"
ment-New Communities American Style, Cen- revised draft, June 1974. Fiscal, social and po-
ter for Urban Policy Research, New Brunswick, litical issues associated with PUDs in New Jer-
Rutgers University, 1972. General text on PUDs, sey. Sponsored by HUD.
including retrospective case study.
20. Peat, Marwick, Mitchell, and Company, "Impact
11. Contra Costa County Planning Department, Analysis of Past Rosslyn and Crystal City De-
Economic Supplement to Application 1840-7 En- velopment," draft final report, March 1974.
vironmental Impact Report, 1974. Fiscal impact Retrospective fiscal impact study of two inten-
of proposed large residential development, with sively developed areas in Arlington County, Vir-
emphasis on capital outlays. ginia. Sponsored by county government.
12. Economic Research Associated, Lynden Farms- 21. San Francisco Planning and Urban Renewal As-
Orange County, Virginia, 1973. Fiscal impact sociation, Impact of Intensive High Rise Develop-
study of second-home condominium development ment in San Francisco, preliminary drafts, 1973.
in rural county. Sponsored by developers. Analysis of fiscal, other economic, and environ-
13. Loewenstein, Louis K., and Walter, David W., mental impacts of intensive development in San
Municipal CostlRevenue Analysis for Planned Francisco business district. Sponsored by HUD.
Unit Developments, Institute of Urban and Re- 22. Shaffer, Ron and Tweeten, Luther, "Measuring
gional Development, University of California, Net Economic Changes from Rural Industrial
Berkeley, 1973. Privately sponsored study of Development, Oklahoma," Land Economics,
existing PUD which does not consider capital August 1974. Economic impact, including fiscal
outlays. effects of new industry in small communities.
14. Syrek, Daniel B., The Fiscal Impact of California Sponsored by federal Economic Development
Residential Growth-Analysis and Supporting Administration.
Tabulations, University of California, Davis,
1974. Critique of methodology applied to studies D. COMMUNITYWIDE STUDIES
of fiscal impact in California. Sponsored by Na- 23. Alesch, Daniel J., and Levine, Robert A., Growth
tional Science Foundation.
in Son Jose: A Summary Policy Statement, The
15. Wainwright and Ramsey, Inc., Harbor Bay Isle Rand Corporation, May 1973. Sponsored by Na-
Development-City of Alameda, 1973. Privately tional Science Foundation.
sponsored case study of proposed residential de-
velopment. 24. Ashley Economic Services, Cost Revenue Anal-
ysis of New Housing Development in the City
C. SIMILAR DEVELOPMENTS IN SEVERAL of San Diego, 1972. Cost estimates of new de-
LOCATIONS velopment by use of local budgets. Sponsored by
group representing developers.
16. Garrison, Charles E., "New Industry in Small 25. Boulder Area Growth Study. Commission,
Towns: The Impact on Local Government," Na- Boulder Area Growth Study (Six Volumes),
tional Tax Journal, December 1971. Fiscal im- December 1973. For Boulder City and County.
pact of new industrial facilities in small Kentucky Sponsored by local government, implemented by
jurisdictions. consultants.
17. Hirsch, Werner Z., "Fiscal Impact of Indus- 26. County of San Diego, Joint City-County Eco-
trialization on Local Schools," Review of Eco- nomic Analysis Project, 1973. Framework for
nomics and Statistics, May 1964. computer model to study impact of growth on
18. Loewenstein, Louis K., "The Impact of New In- individual developments and collective impact
dustry on Revenues and Expenditures of Subur- for area. Sponsored by local government.

46 Fiscal Impacts of Land Development

27. Doxiadis Urban Systems, Inc., Fiscal and Land F. FISCAL IMPACT OF ALTERNATIVE
Use Analysis of Prince George's County, 1970. DEVELOPMENT PATTERNS
Fiscal characteristics of county by type of land 36. Barton-Aschman Associates, St. Charles Com-
use, case study. Sponsored by local government. munities Impact Evaluation Study, draft, 1973.
28. Fairfax County, Fairfax County Five Year Coun- Case study projecting fiscal and other impacts of
ty-Wide Development Program, Volume 1, Pres- a new community on local and state government,
entation of Alternatives, August 1972. Differ- for Tri-County Council of Southern Maryland.
ences in capital outlays given alternative intra- Sponsored by HUD.
county location of new growth. Sponsored and 37. Howard County Planning Commission, Howard
implemented by local government. County 1985 General Plan, Report No. 1, 1967.
29. Fairfax County, Fairfax County Five Year Coun- Impact of alternative development patterns on
ty-Wide Development Program, Volume 11, Fi- service costs, particularly capital outlays. Spon-
nancial Plan, August 1972. Projected fiscal flows sored by county to evaluate impact of the new
for county based on alternative levels of popula- town of Columbia, Maryland.
tion growth. Sponsored and implemented by local 38. Isard, Walter and Coughlin, Robert, Municipal
government. Costs and Revenues, Federal Reserve Bank of
30. Fairfax County, Fairfax County Five Year De- Boston, Chandler & Davis Publishing Company,
velopment Program, Volume III, Evaluation of 1957. Estimated differences in service costs if
Standards and Criticism, Fairfax County, August residential density varies.
1972. Review of standards for local services. 39. Kain, John T., "Urban Form and the Cost of
Sponsored and implemented by local government. Urban Services," Program on Regional and Urban
31. Planning Department, City of Ann Arbor, The Economics, Discussion Paper No. 6, Cambridge,
Ann Arbor Growth Study, February 1973. Fiscal Harvard University, 1967. Review of earlier work
and other impact of growth in city. Sponsored and on the subject.
implemented by local government. 40. Livingston and Blayney, Open Space vs. Develop-
32. Ranck, Edward L., Owensboro, Kentucky Fiscal ment--City of Palo Alto, 1971. Comparison of
and Land Use Management Program, Lexington, various land use alternatives, including no de-
University of Louisville, March 1973. General velopment for a large tract of land. Sponsored by
approach applied to city of Owensboro. local government.
41. -, Santa Rosa Optimum Growth Study,
E. REGIONAL STUDIES 1973. Impact of two rates of population growth
and alternative development patterns on commu-
33. Greene, Kenneth V., Neenan, William B., and nity. Sponsored by local government.
Scott, Claudia D., Fiscal Interactions in a Metro- 42. Real Estate Research Corporation, "Analysis of
politan Area, Lexington, Mass., Lexington Books, Palo Alto Foothill Environmental Design Study,"
1974. Fiscal flows among communities in Wash- in Growth Cost-Revenue Studies, Associated
ington, D.C. SMSA. Sponsored by The Urban Home Builders of Greater Eastbay, 1972. Cri-
Institute and National Science Foundation. tique of the Palo Alto Study (see 40 above),
34. Muller, Thomas, Fiscal Issues in Metropolitan stating limitations of approach. Sponsored by As-
Growth, Metropolitan Washington Council of sociated Home Builders of San Francisco.
Governments, 1973. Various regional fiscal is- 43. -, Environmental and Economic Eflects
sues, including methods to offset negative impact of Alternative Development Patterns, Part 11, De-
of growth and service level differentials. Spon- tailed Cost Analysis, 1973. Fiscal, economic, and
sored by Washington COG. environmental impact of alternative growth pat-
35. Williams, Oliver, et al., Suburban Diflerences and terns. Sponsored by HUD and Council on En-
Metropolitan Policies, Philadelphia, the Uni- vironmental Quality. (Also published under title:
versity of Pennsylvania Press, 1971. Cross-sec- The Costs of Sprawl, Washington, D.C., U.S.
tional analysis of communities in large region, Government Printing Office, 1974.)
with emphasis on service cost and land use dif- 44. Stone, P. A., The Structure, Size and Costs of
ferentials among communities. Urban Settlements, Cambridge University Press,

Bibliography 47

1973. Public and private cost differences if total Police Departments: Too Small and Too Many?"
population or pattern of development is changed Masotti, Louis H., and Hadde, Jeffrey, eds., in
for new 'communities in Great Britain. The Urbanization of Suburbs, Vol. 7, Beverly
45. The Greater Hartford Process, Options for Fu- Hills, Sage Publications, 1973.
ture Development of East Farms, 1973. Fiscal 57. Shapiro, Harvey, "Economies of Scale and Local
impact of alternative development patterns for Government Finance," Land Economics, May
large tracts of land in semirural area. Sponsored 1963.
by community jointly with landowners. Part of 58, Sternleib, George, et al., Housing Development
study implemented by Real Estate Research Cor- and Municipal Costs, Center for Urban Policy
poration. Research, New Brunswick, Rutgers University,
46. Wheaton, William L., and Schussheim, Morton 1973.
J., The Cost of Municipal Services in Residential
Areas, Washington, D.C., U.S. Department of H. OTHER REFERENCED STUDIES
Commerce, 1955. Cost differentials if location or 59. Bahl, Roy, Metropolitan City Expenditures-A
pattern of residential development is varied. Corn- Comparative Analysis, Lexington,. University of
prebensive analysis of costs, without considering Kentucky Press, 1969.
revenues. Sponsored by Department of Com- 60. Breese, Gerald, et al., The Impact of Large In-
merce, implemented at universities. stallations on Nearby Areas-Accelerated Urban
G. STUDIES EXAMINING DEMAND FOR AND Growth, Beverly Hills, Sage Publications, 1965.
COST OF PUBLIC SERVICES 61. Boise Cascade Corporation, Importance to
Montgomery County of Selected Economic Ac-
47. Baumol, William J., "Macroeconomies of Un- tivities, 1970. Fiscal impact of alternative private
balanced Growth: The Anatomy of Urban firms and public facilities, including service con-
Crisis," American Economic Review, June 1967. sumption as a function of income and housing
48. Bergstrom, T. 0., and Goodman, R. P., "Private type. Sponsored by Montgomery County, Mary-
Demands for Public Goods," The American land Economic Development Commission.
Economic Review, June 1973. 62. Burchell, Robert W., and Hughes, James W.,
49. Bradford, P. F., et al., "The Rising Costs of Lo- "Financial Aspects of PUDs," The Appraisal
cal Public Services," National Tax Journal, June Journal; July 1974.
1969. 63. Chamberlain, Gary M., "Local Revenues vs.
50. Brazer, Harvey, City Expenditures in the United Land Use," American City, May 1974.
States, New York, National Bureau of Economic 64. DeLeeuw, Frank and Struyk, Raymond J., The
Research, 1959. Urban Institute Housing Model: Second Year
51. Gabler, L. R., "Economies and Diseconornies of Report, Washington, D.C., The Urban Institute,
Scale in the Urban Public Sector," Land Eco- 1973.
nomics, November 1969. 65. "Economic Impact of Reston on Fairfax County
52. Hanke, Steve H., "Pricing Urban Water," Public Government," memorandum to County Executive
Prices for Public Products, Mushkin, Selma, ed., from Fairfax County Office of Research and Sta-
Washington, The Urban Institute, 1972. tistics, Fairfax County, June 1973.
53. Hirsch, Werner Z., Urban Economic Analysis, 66. Educational Facilities Laboratories, Schools for
New York, McGraw-Hill, 1973. New Towns, 1973.
54. -, "Expenditure Implications of Metro- 67. Einde, K., and Raup, P., Rural Real Estate
politan Growth and Consolidation," Review of Market 1971, University of Minnesota, April
Economics and Statistics, August 1959. 1972.
55. -, "Supply of Public Services," Issues 68. Fairfax County Office of Comprehensive Plan-
in Urban Economics, Perloff, H., and Wings, ning, Interim Manual for Standard Evaluation
L., eds., Washington, Resources for the Future, Methodologies, Fairfax County, Virginia, 1974.
1968. 69. Washington Post, "Households Up 20 Percent
56. Ostrom, Elinor and Parks, Roger B., "Suburban in Montgomery Area," October 5, 1974.

48 Fiscal Impacts of Land Development

70. Levin, Betsy, Muller, Tom and Sandoval, Cora- in two cities, examining impact of growth pat-
zon, The High Cost of Education in Cities, Wash- terns on these changes.
ington, D.C., The Urban Institute, 1973. 81. "School Census in Fairfax Protested," Washing-
71. Leonard, William T., "Synopsis of Urban Land ton Post, August 5, 1974.
Institute Study," in Growth Revenue Studies, 82. Scott, Claudia D., Forecasting Local Govern-
Associated Home Builders of the Greater Bay ment Spending, Washington, D.C., The Urban
Region, 1973. Institute, 1972. Econometric model to predict
72. Montgomery County Planning Board, Fiscal Im- local expenditures applied to New Haven, Con-
pact Analysis, Stage I-Germantown Master necticut. Research sponsored by the city, pub-
Plan, Final Report, 1974. Sponsored and imple- lished by The Urban Institute.
mented by planning staff. 83. Schaenman, Philip S., and Muller, Thomas,
73. Mallack, Allen, written comments for Urban Measuring Impacts of Land Development: An
Institute-Housing and Urban Development De- Initial Approach, Washington, D.C. The Urban
partment Cost-Revenue Conference, January Institute, 1974. Analysis and data collection pro-
1974. cedures to estimate economic, environmental and
74. Morrison, Peter A., Population and the Shape of social impacts. Sponsored by HUD.
Urban Growth-Implications for Public Policy, 84. Sokolow, Alvin D., Governmental Response to
Santa Monica, Rand Corporation, 1972. Urbanization, Agricultural Report No. 132, U.S.
75. Muller, Thomas, "Fiscal Differences Related to Department of Agriculture, Washington, D.C.,
Urban Size and Changes in Population," Western 1965. Impact of urbanization on land values and
Regional Science Association, February 1975. local services.
76. - and Dawson, Grace, Impact of A nnex- 85. Stull, William J., "An Essay on Externalities,
ation on City Finances: A Case Study in Rich- Property Values and Urban Zoning," doctoral
mond, Virginia. The Urban Institute, Washington, dissertation, Department of Economics, Massa-
D.C., 1973. Detailed analysis of costs and rev- chusetts Institute of Technology, October, 1971
enues in a central city and a suburban area an- (unpublished). An empirical assessment of the
nexed to a city. effects of alternative land use on housing values
77. - "Implicit Grants to Property Owners in suburban communities, with critique of litera-
at the Local Level," in Redistribution to the Rich ture.
and the Poor, Boulding, Kenneth E., and Pfatt, 86. ne Urban Institute and International City Man-
Martin, eds., Belmont, Wadsworth Publishing agement Association, Measuring the Eflectiveness
Company, 1972. Impact of assessment practices of Basic Municipal Services, Washington, D.C.,
and property tax exemptions on income dis- 1974.
tribution in suburban Maryland counties. 87. Thompson, Wilbur R., A Preface to Urban Eco-
78 Netzer, Dick, "Financing Suburban Develop- nomics, Resources for the Future, Baltimore,
ment," Studies of the Nassau-Sujffolk Planning John Hopkins Press, 1965.
Region, Economic Research Bureau, Stony Brook, 88. U.S. Bureau of the Census, 1972 Census of
State University of New York, 1969. Changes in Governments, Vol. 2, Part 2, Taxable Property
public outlays as a result of rapid suburbani- Values, Washington, D.C., 1973.
zation 89. -, City Government Finances in 1964-65,
79. New Jersey County and Local Government Study 1969-70, 1970-71, and 1971-72, Washington,
Commission, Multifamily Housing and Municipal D.C., 1966, 1971, 1972, and 1973, respectively.
Services, Trenton, 1973. Study of multifamily
housing fiscal flows, characteristics of residents, 90. -, Mobility Between Metropolitan Areas,
attitudes of local officials and residents in a Washington, D.C., March 1973.
cross-section of New Jersey communities. Spon- 91. -, General Social and Economic Char-
sored by HUD. acteristics, United States Summary, Washington,
80. Santa Clara County Planning Department, D.C., June 1972.
Municipal Cost Revenue Analysis, San Jose, 92. U.S. Department of Justice, Crime in the United
1973. Changes in expenditures for public services States, Uniform Crime Reports, (annual issues).

Bibliography 49

1. OTHER BIBLIOGRAPHIES: revenue analysis. In addition, economic effects
93. Kain, John T., "Urban Form and the Cost of of annexation, residential, commercial, and in-
dustrial land uses and urban redevelopment are
Services," Program on Regional and Urban referenced.
Economics, Discussion Paper No. 6, Cambridge, 95. Real Estate Research Corporation, The Costs of
Harvard University, 1967. Review of earlier work Sprawl: Li Iterature Review and Bibliography,
on alternative development patterns. Washington, U.S. Government Printing Office,
94. Mace, Ruth L., "Costing Urban Development and 1974. A comprehensive bibliography on resi-
Redevelopment," Institute of Government, dential and commercial costs of development.
Chapel Hill, University of North Carolina, 1963. Separate sections on major public services and
Extensive bibliography of cost-benefit, cost- environmental effects. Many references annotated.

50 Fiscal Impacts of Land Development

APPENDIX A. MEASURING FISCAL IMPACT
A GENERAL APPROACH

Appendix A outlines a general approach to obtaining lation of the development, the public services allocated
data on net changes in government fiscal flow as a result to the development, and so forth.
of new development. Part I is extracted from the Urban Some of the major direct fiscal impacts of commercial
Institute report, Measuring Impacts of Land Develop- and industrial development are discussed, but not the
ment [83]. Part II provides brief comments on other secondary fiscal effects, such as those resulting from the
reports which illustrate various approaches to cost- inmigration, commuting, and shopping they stimulate.
revenue analysis. A detailed case study illustrating the procedures for
estimating fiscal impact has been developed by Muller
1. ESTIMATING NET CHANGES IN GOVERNMENT and Dawson elsewhere [51.*
FISCAL FLOW
Revenue Estimates
A new development's fiscal impact on local govern- Local revenues can be grouped into four categories:
ment-the net change in public revenues less operating 1 ) revenues associated with real property wealth-the
expenditures and (annualized) capital expenditures- largest source in most jurisdictions; (2) revenues as-
depends to a considerable extent on whether the gov- sociated with income and level of consumption, which
ernment will maintain or change its level and quality of are comprised primarily of local income, sales, and
services to the new development and to the rest of the
community after the development is completed. Concur- utility taxes; (3) per capita, per pupil, or other per
rently, the level of service to be provided is likely to 91population unit" revenues, which are derived from
depend to some extent on the estimated fiscal impacts. either a per capita tax, or redistribution from higher
That is, the community chooses a level of service based levels of government; and (4) miscellaneous revenues,
in part on its perception of what it can afford. To further which include fees, user charges, fines, licenses, and
complicate matters, maintaining the same expenditures minor items.
per capita is not necessarily synonymous with main- It is useful to identify separately the revenues from
taining the same quality of service, since the demands business enterprises and revenues from households;
for services and the costs of supplying them may change and the latter should be further classified as occupants
faster or slower than the rate of residental or business of single family, townhouse, and apartment units.
population growth. Revenues Related to Real Property
The methodology discussed here for assessing fiscal Real property is usually taxed by local governments.
impacts is based on the assumption that current service In general, the same tax rate applies to both residential
quality, tax structure, and tax rates are to be maintained. and nonresidential property.'
The discussion focuses on evaluating proposed resi-
dential development. * Numbers in brackets refer to material listed in the Bibli-
Retrospective analysis would use similar techniques ography preceding this appendix.
1. There are exceptions, as in Minnesota, where industrial
but would have much better estimates for the socio- and commercial property are taxed at a higher rate than resi-
economic and demographic characteristics of the popu- dential property.

Property tax revenues are computed by multiplying Revenues Related to Income
the tax rate by the assessed value of property. In most Revenues generated by new development may be
communities, the assessed property value is a per- directly related to income of residents as with local
centage of market or full property value. For example, income taxes. Or they may be indirectly related via con-
in California, assessments are based on 25 percent of sumption as with personal property taxes and local sales
market value. Thus, a $50,000 housing unit should be taxes. A number of communities impose utility taxes
assessed at $12,500. In most cases, however, there is a related to income insofar as higher income households
difference between the "official" and actual current have larger housing units and more appliances, and thus
market value, due to a time lag in updating assess- consume more energy and water. Excise taxes on
ments in an inflationary economy and to other factors. specific goods also relate to consumption patterns.
The true "effective tax rate," which should be the basis To estimate these income-related taxes, it is neces-
for estimating additional revenue from new real prop- sary to estimate the expected household income of new
erty, can be computed by dividing the current market residents, which may be derived from the relationships,
value of similar property in the community (estimated between property values and income. These relation-
from recent real estate sales) by tax payments from the ships, in turn, can be determined from census data and
property. The effective tax rate, with few exceptions, is consumer surveys.2
below the official rate. The average effective assessment If monthly rent payments for proposed apartment
ratio in California during 1971, as shown in the 1972 units have been set, income estimates can be derived by
Census of Government, was 20 percent of market value, assuming rent payments to be a specified share of
not the 25 percent ratio required by state legislation. income.3 The share of income allocated for housing
Thus an official or nominal tax rate of $15 per $100 of varies somewhat by location, age, and size of the house-
assessed value, for instance, would amount to a 3 hold, and by type of housing. Annual rental payments
percent effective tax rate, not 3.75 percent as one also may be estimated as representing, on the average,
might assume from use of the official assessment ratio. between one-seventh to one-ninth of the value of the
The estimated market value of land and structures housing unit.
is usually provided by the developer. It can be compared A more direct method for estimating income of new
to values of similar property to determine if it reason- residents is to examine applications to mortgage institu-
ably reflects the local market. The property tax revenues tions, developers, and apartment managers. However,
may then be estimated by multiplying the estimated access to these data is extremely difficult because of con-
market value of the new real property by the effective fidentiality.
tax rate, deducting for exemptions such as homestead, Sales and excise tax recipts can be estimated-given
old age-low income, or veteran status. Real property data on income-from various surveys on expenditures
taxes from the current (before development) land use by income class, household size, region, and metro-
on the development site should be computed and sub- politan area .4
tracted from the estimated revenue accruing from the
proposed development to yield the net change in real Income taxes can be estimated directly by application
property taxes. This is too often neglected in fiscal of appropriate rates to taxable income and size of house-
analysis. Likewise, if the people or businesses displaced hold. In a few states, local governments can impose a
by the development leave the jurisdiction, estimates of tax on income *of residents, or a tax on payrolls, based
other tax revenues lost-and expenditures reduced- on place of employment. Many cities in Pennsylvania
may be needed. and Ohio tax income earned in the community. In
For income-producing property, such as a large Maryland, all counties levy an income tax on residents
apartment building, property taxes might not be based
on the value of the building, but rather on gross or net 2. For a discussion of the demand for housing as a function
of income, see F. deLeeuw, "The Demand for Housing: A
income. This assessment approach tends to result in Review of Cross Section Evidence," The Review of Economics
higher revenues compared to taxes on the value of the and Statistics, Vol. 53, February 1971, pp. 1-10. He found that
building, unless many units are not occupied. the value of the new owner-occupied housing was generally
Changes in property tax revenue may also result if between 1.7 and 2.4 times annual income.
3. A number of government publications discuss these pro-
new development induces changes in land values else- portions. For example, see U.S. Department of Labor, Three
where in the community. Although such estimates are Standards of Living for Urban Families, Bulletin No. 1570-5,
very difficult to quantify with much confidence, to the Washington, D.C., 1969.
4. U.S. Department of Labor, Bureau of Labor Statistics,
extent they can be approximated the associated revenues Survey of Consumer Expenditures, Report No. 237-88. Wash-
should be accounted for. ington, D.C., 1965.

52 Fiscal Impacts of Land Development

of their jurisdiction. In some states, these local income basis of recent per capita receipts from these sources
or payroll taxes are not permitted. by the jurisdiction.
Personal property subject to taxation varies widely. Operating Expenditure Estimates
The most common items subject to this tax are auto-
mobiles and, to a lesser extent, major household goods. The importance and scope of local public services
Their value can be estimated by their relation to in- for which expenditures must be estimated can differ
come.5 In the case of automobiles, it is necessary to sharply among and within states. For example, water
ascertain the base used for estimating value (wholesale and sewage utilities and roads and highways are main-
price, loan value, or market price) and the effective tax tained by many localities but not others. Health and
rate. Businesses sometimes must pay personal property welfare often are not city responsibilities, and tend to
taxes based on machinery and inventory; these can be be small portions of some county budgets. However,
roughly estimated if the type of proposed industry is they are major expenditure items in cities such as New
known. York or Detroit and in many counties.
Utility taxes are frequently levied as a percentage of The allocation techniques discussed here assume that
utility bills. Estimates of average bills can be based on current local government personnel (teachers, main-
utility company data for various types of residences, tenance crews) generally are fully occupied. Thus, a new
such as large single-family dwellings, smaller single- development that creates additional demand for their
family dwellings, and apartments. services would, in the absence of additional resources,
reduce the quality of services. The allocation techniques
Per Capita Revenues estimate the cost of maintaining the existing scope and
Local governments in some states administer a per quality of services. The attempt to determine average
capita or "head" tax on all adults. More frequently, costs or additional costs for each service, if they can be
local government is the recipient of state or county determined, will be very useful for determining the
revenues distributed on the basis of the number of resi- impacts of new development. It is recognized, however,
dents or the number of students. For example, profits that existing personel may be underutilized because of
from the alcoholic beverage sales by the state are inefficiencies, anticipation of future demand, or other
distributed to local jurisdictions in Virginia based on reasons, in which cases judgmental adjustments in
population, while sales tax receipts are distributed on the operating expenditure estimates would have to be made.
basis of school-age residents. Federal revenue sharing Local operating expenditures can be grouped into
for local jurisdictions, as presently legislated, also uses those incurred in supplying services used (1 ) primarily
population as one criterion. (The other criteria are per by households, such as education, libraries, health and
capita income and tax effort; as income in a com- welfare, and recreation, and those used (2) by both
munity rises relative to other jurisdictions, revenue business enterprises and households, such as fire and
sharing funds are reduced; an increase in relative tax police, utilities, general government, and transportation.
effort leads to an increase in the federal funds.) Esti-
mates of all per capita taxes should be based on the Household-Relaled Expenditu rerr--Edu cation
expected change in local population or school enroll- In most local jurisdictions, public education is the
ment, applying whatever formulas are used for com- largest outlay, as high as 80 percent of operating ex-
puting such taxes. penditures in suburban areas of states in which the
state governments do not absorb the major share of
User Charges, Service Fees, Miscellaneous Revenues these burdens. Therefore, the factor which usually
User charges for utility services, other revenues from determines whether a residential development will result
public utility operations, and fees for public safety, in a fiscal surplus or liability is the projected incre-
recreational, and other services also can provide sub- mental expenditure for public education.
stantial revenue to local government [761. Such user The two most important factors which determine
charges, fees, and fines initially should be allocated school enrollment and therefore education expenditures
between business firms and households. The revenues are the type of housing and number of bedrooms per
accruing from households can be approximated on the housing unit. A number of studies show how to estimate
enrollment on the basis of these two factors [27, 58].
5. For data on the relationship between automobile value With few exceptions, detached single-family housing
and personal income, see the most current issue, Bureau of
the Census, Consumer Buying Indicators, Series P-65, Wash- units, particularly those with four or more bedrooms,
ington, D.C. and garden apartments with three bedrooms, have the

Appendix A. Measuring Fiscal Impact-A General Approach 53

most school-age children per unit. New detached ally differ from those of the base population.6 The type
housing units typically have more children tharf do older and unit value of housing can be used to estimate resi-
units. The fewest children per unit are found in highrise dent characteristics, especially income distribution.
luxury apartments and condominiums, one-bedroom Some of the complexities to be taken into account
garden apartments, and two-bedroom townhouses. for improved estimates of noneducational expenditures
In addition to housing type, racial and ethnic char- are discussed in the sections that follow. Cost situations
acteristics, which are related to children per family and vary too widely from community to community to
reliance on parochial schools, also influence public permit more than a general approach here.
school enrollment. Income, which is related to housing Health and welfare. Most health and welfare services
type, affects both the demand for higher quality educa- are directly linked to income. These expenditures tend
tional services and the reliance on private schools and to be concentrated in older, lower-income areas of a
thus affects public school population and budget. jurisdiction. Since the income of residents in new non-
The distribution of students among grade levels is subsidized housing can be expected to be considerably
frequently also a function of housing type. Apartment above the level that would qualify for welfare and
residents tend to have proportionately more children in health services for the indigent, the demand and thus the
elementary grades, where per pupil costs are up to one- incremental cost of social services is likely to be low.
third lower than in higher grades. The use of average per To assess the impact of new residents on social ser-
student expenditure throughout the school district, vices, the proportion of new households whose income or
without reference to these differences in grade level age is at a level which qualifies for social services should
distribution for each housing type, is likely to be mis- be estimated. In addition, the cost of health services
leading. available to all residents, regardless of income, should be
Statistics on children per unit considering the various computed. As population increases, it also may be
factors just cited can be developed from school attend- assumed that the unit cost of social services will rise
ance records for the community or similar communities, somewhat, because the cost of living and wages generally
if they are not already available from the school board. are higher in larger communities than in smaller ones.
The estimated number of new students per grade times Recreation and libraries. The demand for certain
the average cost per student in each grade yields the recreation and library facilities may increase with the
total estimated educational expenditure. inmigration of higher income residents. Concurrently,
In communities which support junior colleges and new developments may provide their private facilities,
other post-high school education, the impact of new reducing the pressure on public services. Some com-
developments on these facilities also needs to be esti- munities plan for a fixed quantity of recreation facilities
mated. Enrollment in such institutions is a function of and open space per capita regardless of private facilities.
household demographic characteristics and income. Preferably, the additional demand for recreational
In most states the level of state aid for public educa- facilities as a function of age, income, and location of
tion is based, at least in part, on per pupil property residents should be used to estimate new recreation oper-
values. Thus, a proposed commercial, industrial, or ating expenditures, modified by special circumstances
expensive residential development will increase the per and characteristics of local policy. In the absence of
pupil property base, decreasing the per pupil state con- such data, or, if the policy is equal allocation per capita,
tribution in the future. the current average cost per household should be al-
Household- Related Expenditures-Noneducational Services located to the new development.
One simplistic approach to estimating additional non- Services Utilized Jointly by Households and Businesses
educational expenditures associated with new house- Most local services are utilized by both households
holds is to assume that the cost per new resident will and business enterprises. As in the case of the house-
equal the average cost of these services per existing hold-related services, it is preferable to base cost esti-
resident. This easy computation is based on the premise mates on actual service additions that can be attributed
that (1 ) demand is independent of socioeconomic and to the new development. Where circumstances do not
demographic characteristics, or that (2) population allow this, estimates may be based on past expenditures
characteristics of new residents are similar to the base
population. It also implies that the unit cost of delivering 6. Characteristics of inmigrants to large urban areas can
Numerous studies suggest that these assumptions are be estimated from Bureau of Census Mobility of Metropolitan
questionable [34]. Characteristics of inmigrants gener- Areas, PC(2)-2c, Washington, D.C., 1973.

54 Fiscal Impacts of Land Development

per household, per business employee, or per $1,000 estimate the incremental expenditure. The actual alloca-
property value. tion should reflect the experience of similar communi-
As a first step for estimating these unit costs, it is ties in the state which have been growing rapidly in com-
useful to identify past expenditures for each sector- parison to those where growth has been small. This
households and business .7 In some communities, busi- method of estimating the future cost of services has
ness enterprises are concentrated in areas with few been applied to a number of communities [58].
residential structures, and the services devoted to them Fire services. The need for additional fire service
may be readily identified. For example, a fire company expenditures is determined by the accessibility of new
may serve primarily a central business district, so that developments to existing fire stations, the current de-
all or a large share of that cost can be allocated to mand level at those stations, and the types of pro-
business. posed structures.
In expanding areas, new commercial and perhaps The frequency of fires per housing unit in new
industrial property may be in fairly close proximity to residential developments, based on empirical data, is
housing, making it difficult to identify the actual re- usually below community averages. However, low
sources supplied for each. Several allocation schemes density development can require more fire stations per
have been devised. The most commonly utilized method housing unit to offset the longer travel times when
is to allocate expenditures for jointly used services- housing is spread out. And despite locational differ-
particularly public safety-to busines and residences in ences, certain communities maintain a fixed relationship
proportion to their relative property value. An alterna- between firemen and population."
tive is to rely on the number of employees in business The suggested approach is to allocate incremental
enterprises, as a proportion of total employees and resi- operating outlays for fire services on the basis of addi-
dents, for the allocation to business [38]. Both of these tional manpower required. If no added personnel are
approaches, however, tend to reflect benefit received needed, one can estimate the anticipated number of ad-
rather than cost incurred. Where demand data are ditional fire calls as a proportion of the total number of
available, these may be used as the basis of allocation. calls for the fire station nearest the development. This
For example, the number of fire calls associated with would indicate the share of the fire station operating
business versus the total calls could be used for al- cost to be allocated,to the new development.
locating fire services. The proportion of trips generated
by residences versus businesses could be used for allo- Police services. Per capita police outlays, once some
cating local transportation services. minimum population base is reached, increase as the
Some services are aimed directly at the business size of a city increases. It is not known to what extent
sector, such as the testing and sealing of scales. Their this is attributable to changes in the level and types of
costs should be fully allocated to business, even though police service provided, socioeconomic characteristics,
some benefits may accrue to the residents. population density, or other factors. The major factor
Once historical costs are allocated between business appears to be the higher level of crime per capita.
and residences, unit costs can be computed and used for New developments characterized in the main by low
estimating expenditures for new development. Some fur- density housing are likely to have low crime rates.9
ther comments on estimating costs of the major jointly Insofar as crime rates reflect direct demand for police
used services follow. services, the use of a crime index as a proxy for demand
General government. It is difficult to allocate most is likely to show that the incremental cost in new
general government services to a specific development. developments is below the average cost of providing
For small developments, general expenditures can be service. Thus using average costs might seem biased.
estimated on a per capita basis. However, police protection extends to roads, shopping
However, as the community grows, per capita ex- areas, and other facilities where residents shop and
penditures for general services tend to increase. A wider work. In addition, only a small share of total police
scope of services is offered, and more highly trained and calls are directly linked to crimes. Thus average costs
paid professionals are hired. For large-scale develop- may not be as poor a proxy as one might think at first.
ments, using past per capita costs may thus under- Another approach is to base costs on the estimated
additional manpower allocated to the new area, adding
7. A frequent mistake in cost-revenue analysis is to com- a proportional share of central administrative and re-
pute per capita (resident) costs by dividing total costs (for
businesses and residents) by the number of residents, rather 8. In communities within Santa Clara County, for example,
than dividing just the resident-related part of the costs by the a ratio of one fireman per 1000 residents is maintained [801.
number of residents. 9. See, for example: [24 and 27].

Appendix A. Measuring Fiscal Impact-A General Approach 55

lated overhead expenditures. Some communities apply allocated to it. For example, for a school the number of
a standard, such as 1.6 uniformed police per 1,000 pupil-years of education required by the development as
residents. This implicitly assumes that demand for a percent of the total pupil-years provided by the school
police services is independent of.new population cbar- over its expected life could be used to allocate the
acteristics or the housing mix. It also deals with the annualized capital costs of the school to the develop-
development alone, not reflecting its contribution to ment. This gives some of the benefit from the expected
the higher per capita costs associated with larger economy of scale to the new development. The balance
communities. of the annualized cost, until the facility is fully utilized,
The preferred but somewhat more difficult approach is shared by the total community. However, if earlier
is to estimate the additional manpower likely to be construction of the new facility is required because of
added, based on past experience with similar develop- the specific development, the analysis should take ac-
ments, if any, and discussions with police officials, so count of local funds requiring earlier outlay and the
that the latest policies can be reflected. To the costs likely costs of construction at different times, including
of manpower necessary to serve the new development anticipated interest costs on bonds that would be bor-
would be added expected increases in general costs due rowed by the local government for the project.10
to the community's increased population, based on ex- If the new development uses available space in
perience in other like communities. existing facilities, some would allocate only the short-
term incremental cost, some the long-term incremental
Capital Expenditure Estimates cost, and some the average cost. Which to use depends
Three major tasks are involved in estimating the costs on the viewpoint and purpose of the analysis. The short-
of public capital improvements associated with new term incremental cost (which may be zero) reflects the
development. out-of-pocket additional expense for the facility. The
� The allocation of facility costs between the long-term incremental cost reflects the cost attributed
existing community and the new development. to new development over the long run, including econ-
omies or diseconomies of scale they create. The average
� Choice of the lifetime and interest rates to be cost concept assumes each user bears an equal burden.
used in annualizing costs of new plant. For either case-whether the new development uses
� The timing of the new investment. old or new facilities-two separate capital expenditure
computations might be made, one emphasizing causation
Cost Allocation of costs, the other what the community will have to
Capital expenditures associated with new develop- spend. The first would indicate the relative burden on
ment can be divided into two categories. First, facilities services from the new development, the second the
linked directly with the development, such as new changes in fiscal outlays that would be needed.
schools, sewer lines, fire stations, and other new facili- In some cases, a new development triggers a new
ties to be utilized primarily by the new development. capital investment that will be used by all of the com-
Costs of these facilities can be allocated largely or wholly munity, and that will raise the per capita cost to the
to the new development. Second, facilities constructed or community for a service. An example is a tertiary sew-
expanded as part of a capital improvement program age treatment plant required to keep water pollution
which will be shared by existing as well as new residents below some limit. In most communities the practice is
or enterprises in the jurisdiction. Such facilties could in- to distribute the cost of such facitties equally among all
clude junior colleges, new sewage or water treatment users. The cost for fiscal impact analysis purposes might
plants, and health care centers. They are generally not be allocated to the entire community, but in some cases
triggered by a single development, unless it is very large. the costs might be allocated to the new develoment-
The costs-of the second category pose difficult, classic it depends on the reasons for adding the plant and
allocation questions involving consideration of scale whether overall service quality improves or remains the
economies and the optimum size and timing of new same.
plant construction. The approaches are widely argued Facilities fully utilized prior to new development,
and a full discussion cannot be included here. Only a such as public schools, should not be considered as part
few suggestions must suffice. of the capital cost attributable to the new development.
If a new facility is part of a capital improvement plan
and is initially underutilized in expectation of future 10. In an inflationary economy, it is frequently advantageous
to initiate construction in anticipation of future demands,
growth, only the share of the total cost needed to meet since annual debt payments are fixed while the tax base is
the demands created by the new development should be expanding.

56 Fiscal Impacts of Land Development

Annualizing Costs interest charges on a bond could be added to the
There are three means of paying for major capital amortized per annum economic life of the project. Thus,
if bonds for a project had a 30-year repayment period,
projects: (1 ) general revenues from current tax receipts,
(2) general obligation or revenue bonds, and (3) cur- but the useful economic life of the project was 40 years,
rent revenue combined with general obligation or the annual capital costs could be reduced, although the
difference would not be substantial. Technological and
revenue bonds."
Whether capital expenditures, which provide current other changes may result in a shorter useful economic
as well as future benefits, should be paid from current life for a capital investment than intially projected. In
revenues or over an extended time period, involves addition, elements of a particular project are likely to
issues of equity, since the composition of the popula- have differing economic life spans. As a result, basing
tion using the capital improvements undergoes change estimates on the anticipated economic life of a project
during the useful life of the investment.12 Most large has practical limitations. Whatever method of annualiza-
communities-for both fiscal and political reasons- tion is used should be clearly stated, since it is often the
tend to borrow funds, particularly for school facilities. basis for criticism in comparing fiscal flows.
The method of financing chosen by a jurisdiction Timing of Investment
influences the short-term and long-term costs of capital Unless existing facilities are underutilized, increases
investment. In a s 'low-growing jurisdiction, a substantial in population or business enterprise expansion require
portion of capital needs can be met from current rev- immediate new public and private sector capital invest-
enues, on a "pay as you go" basis. Capital expenditures ment. The alternative would be a reduction in the level
for such services as publicly owned utilities are usually and quality of existing services, such as double school
self-financing from user charges through revenue bonds, sessions, increased traffic congestion, or overcrowded
and thus impose no direct burden on the public sector recreational facilities.
fiscal structure. However, major capital costs, for In areas of rapid growth, public infrastructure invest-
services not funded by user charges, particularly in areas ments frequently lag behind population increases be-
of rapid growth, cannot be financed from current funds. cause of public sector fiscal limitations, such as legal
Therefore, general obligation bonds are issued for a limits on borrowing; lag in revenues from new develop-
selected payback period. ment; the initial diseconomies of scale (i.e., under-
The bond repayment periods selected by communi- utilization) associated with new facilities; or because
ties generally vary between 20 and 30 years. A sug- of inadequate planning. As a result, there is often a
gested approach for determining per annum cost in- short-term degradation in the provision of public
volves calculating the straight line amortization of services.
capital over the useful life of the investment and adding
the interest on the average balance outstanding. The 11. ADDITIONAL REFERENCES
interest rate selected when computing annualized capital
expenditures should reflect the bond market at the time It is recognized that the approach outlined in Part
of the analysis. In 1973, interest rates for communities I of this Appendix A provides insufficient information,
with a high bond rating fluctuated between 4.7 and 5.5 by itself, for undertaking a fiscal impact analysis. It is
percent. (These percentages are considerably below the therefore suggested that the interested reader review
private market rate, since the interest on these bonds is one or more of the case studies and general method-
not subject to federal income taxation.) ologies briefly noted below for additional information
An alternative approach for computing the capital on data sources, allocation methods, and computation
cost would be to estimate the useful economic life of the procedures.
investment (excluding the value of the land), inde- Only reports which focus on evaluation of individual
pendent of the bond repayment period. Presumably, developments are noted. Readers interested in methods
if the actual economic life could be estimated, the annual for estimating communitywide growth effects or of
determining the fiscal impact of alternative develop-.
11. Revenue bonds are those bonds secured with income ment patterns should refer to the Bibliography for ref-
received by a jurisdiction from the earnings of a revenue- erences to reports on these subjects.
producing enterprise, such as waterworks. General obligation
bonds are secured by an unconditional pledge of a jurisdiction's The Fiscal Impact of Residential and Commercial
credit, including its taxing power. Development: A Case Study (1972)
12. For a theoretical discussion of public debt issues, see
R. A. Musgrave, The Theory of Public Finance, New York, This case study [51 applies the general methodology
McGraw Hill, 1959. outlined in Part I of this appendix to a proposed mod-

Appendix A. Measuring Fiscal Impact-A General Approach 57

erate-scale PUD in a semi-rural county of Virginia. The The report provides very detailed tables and sched-
report discusses allocation procedures for capital ex- ules to be completed by those undertaking an analysis.
penditures in considerable detail and an appendix fo- Step by step guidelines accompany each table and
cuses on estimating the fiscal impact of proposed de- schedule. A section of the report provides a method-
velopments on state government. ology for estimating the level of commercial develop-
Population characteristics, particularly the number ment likely to be attracted as a result of residential
of school-age children, and household size by type of growth.
housing are estimated from a survey of similar develop- One limitation of this study is its reliance on the
ments in the community. Household size and income expected service demand based on the judgment of
are among factors suggested to project service demand. local personnel. The methodology is not applied to a
specific development.
One limitation of the report is that since water and
sewage services in the case study community are pro- Computer-Assisted Approaches
vided by an agency which is not part of the county gov-
ernment, the fiscal implications, if any, of new develop- A number of recent reports discuss computer-assisted
ment on the cost of these services are not considered. procedures to implement cost-revenue analyses. These
The report does not discuss minor public services which reports focus on the use of computers as aids in pro-
some large urban communities may provide. jecting costs and revenues by the use of standardized
procedures and data formats.
Municipal Cost-Revenue Analysis for Planned A well-written, concise description of such a com-
Unit Developments (1973) puter-assisted system is available [4]. However, the re-
This fiscal impact study [13] allocates costs and port provides only the most general guidance on how
revenues associated with a moderate-scale PUD in to obtain and allocate the necessary revenues and ex-
Northern California, where multiple taxing districts penditures. Another computer-assisted model discusses
complicate the evaluation process. Since early phases of cost curves for various categories of municipal services
the development were completed at the time the analysis based on changing population levels [5], Some guide-
was undertaken, household characteristics were obtained lines for estimating service costs are given, but in-
from an examination of mort age applications and rental sufficient information is presented to undertake an
agreements, with tax data 9derived from assessment analysis without reference to additional data sources.
Another computer-assisted approach to fiscal impact
records. It is assumed that per capita expenditures for analyses is aimed at assessing the fiscal impact of
the new developments are the same as the community publicly-financed housing developments in Michigan
average. The study aims at an objective evaluation of towns and cities [1]. It uses a model to develop five-
revenues and costs associated with the development. year forecasts of the demographic, physical and eco-
However, the study has two significant limitations: nomic c
capital expenditures are not included, thereby under- haracteristics of cities. These data are used
to estimate costs and revenues linked to new public and
estimating the overall public cost of new development; private housing.
and some of the state funds are redistributed on a Computer-assisted approaches have the advantage of
per capita basis, which differs from the actual dis- quickly calculating costs and revenues. However, none
tribution formulas. of the general models referenced incorporate all the
Cost-Revenue Impact Analysis for Residential major variables which influence revenues and expendi-
Developments (1974) tures within and among communities. The cost and com-
plexity of developing and maintaining a more general
The approach of this report [2] is to utilize the 1970 model, applicable to most communities in the nation,
Census of Population as the major data source on house- are probably too great to be justified. The development
hold characteristics. Estimates for public service demand of more sophisticated computer models, given the limi-
are to be derived from local government department tations of existing research, may in any case, be pre-
heads. mature.

58 Fiscal Impacts of Land Development

APPENDIX 1B. ATTENDEES, FISCAL COST-REVENUE
ANALYSIS CONFERENCE, 1974

The study reflected in this report benefited con- John P. Fowler Community Development
siderabIy from the Conference on Fiscal Cost-Revenue Department,
Analysis, held at The Urban Institute in Washington, San Diego, California
D.C. on January 23 and 24, 1974. The conference was Dennis E. Ciale Marcou O'Leary and Associates,
cosponsored by HUD. The attendees opened many Inc., Washington, D. C.
avenues of thought pursued in these pages, but re- Harvey 0alper The Urban Institute,
sponsibility for the content of the report remains solely Washington, D. C.
with the author.
Worth Bateman The Urban Institute, Robert W. Glover Livingston and Blayney,
Washington, D. C. San Francisco, California
Beverly Beidler Alexandria City Council, Virginia Robert Goldman National Science Foundation,
Melvin Bergheint Alexandria City Council, Virginia Washington, D. C.
and National League of Harry P. Hatry The Urban Institute,
Cities/ U.S. Conference of Mayors Washington, D. C.
Ed B. Brandt Decision Sciences Corporation, Harvey Herr Office of Research and Statistics,
Jenkintown, Pennsylvania Fairfax County, Virginia
Debra Brett Real Estate Research Corporation, Duane Himz Planning Department,
Chicago, Illinois Madison, Wisconsin
G. B. Chapman Office of Planning Coordination, James Hoben U.S. Department of Housing and
Metropolitan Washington Council Urban Development,
of Government, Washington, D. C.
Washington, D.C. Kenneth B. KenneyU.S. Department of Housing and
Wyndham Clarke U.S. Department of Housing and Urban Development,
Urban Development, Washington, D. C.
Washington, D. C. Alan Mallach County and Municipal Study
Grace Dawson The Urban Institute, Commission,
Washington, D. C. Trenton, New Jersey.
John F. Downs Jr. Nation Capital Planning Richard Martin Department of Economics,
Commission, Prince George's University of Hartford,
County, Maryland Connecticut
Bruce Drenning Barton-Aschman Associates, David Mosena. American Society of Planning
Washington, D. C. Officials, Chicago, Illinois

Thomas Muller The Urban Institute, Samuel J. Stein Research and Statistics,
Washington, D. C. State of New York,
George F. PetersonThe Urban Institute, Albany, New York
Washington, D. C. David Talbott Planning Office,
Harold Phipps Northern Virginia Planning District Falls Church, Virginia
Commission, Fairfax, Virginia Jack Underhill U.S. Department of Housing and
Urban Development,
Dale Price National Capital Parks and Washington, D. C.
Planning Commission, Joseph S. Wholey Arlington County Board,
Montgomery County, Maryland
Arlington, Virginia, and
Randall Chum Office of Comprehensive Planning, The Urban Institute
Fairfax County, Virginia George Wjnzcr U.S. Department of Housing and
Philip Schaenman The Urban Institute, Urban Development,
Washington, D.C. Washington, D. C.
Lee G. Shelton Michigan State Housing J. Ward Wright Office of Urban Services,
Development Authority, National League of Cities,
Lansing, Michigan Washington, D. C.

60 Fiscal Impacts of Land Development

THE URBAN INSTITUTE BOARD OF TRUSTEES

WILLIAM SCRANTON, Chairman
Scranton, Pa.
CHARLES L SCHULTZE, Vice Chairman
Senior Fellow, Brookings Institution, Washington, D.C.
JEAN FAIRFAX, Director, Division of Legal Information and
Community Service, NAACP Legal Defense and
Educational Fund, Inc., New York, N.Y.
EUGENE G. FUBINI, Consultant, Arlington, Va.
WILLIAM GORHAM, President, The Urban Institute,
Washington, D.C.
KATHARINE GRAHAM, Chairman of the Board,
Washington Post Co., Washington, D.C.
ROBERT V. HANSBERGER, Chairman and Chief Executive,
Futura Industries Corporation, Boise, Idaho
VERNON E. JORDAN, Executive Director, National Urban
League, Inc., New York, N.Y.
RICHARD LLEWELYN-DAVIES, President, Llewelyn-Davies
Associates, New York, N.Y., and London, England
BAYLESS A. MANNING, President, Council on Foreign
Relations, New York, N.Y.
STANLEY MARCUS, President, Neiman-Marcus, Dallas, Tex.

ROBERTS. McNAMARA, President, International
Bank for Reconstruction and I Development, Washington, D.C.
ARJAY MILLER, Dean, Graduate School of Business,
Stanford University, Stanford, Calif.
J. IRWIN MILLER, Chairman, Cummins Engine Co., Inc.,
Columbus, Ind.
JOHN D. ROCKEFELLER IV, President, West Virginia
Wesleyan College, Buckhannon, W. Va.
WILLIAM D. RUCKELSHAUS, Ruckelshaus, Beveridge &
Fairbanks, Washington, D.C.
HERBERT SCARF, Professor of Economics, Yale University,
New Haven, Conn.
FRANKLIN A. THOMAS, President, Bedford-Stuyvesant
Restoration Corp., New York, N.Y.
CYRUS R. VANCE, Partner, Simpson, Thacher & Bartlett,
New York, N.Y.
JOHN G. VENEMAN, President, Veneman Associates,
San Francisco, Calif.
JAMES VORENBERG, Professor, School of Law, Harvard
University, Cambridge, Mass.

@3 "666800002 5454