CS

http://pfr.sagepub.com/ Public Finance Review

http://pfr.sagepub.com/content/37/5/613 The online version of this article can be found at:

DOI: 10.1177/1091142109344585

2009 37: 613Public Finance Review Changhoon Jung, Chul-Young Roh and Younguck Kang

Cities Level of Capital Spending, Taxes, and Long-Term Debt in American

Longitudinal Effects of Impact Fees and Special Assessments on the

Published by:

http://www.sagepublications.com

can be found at:Public Finance ReviewAdditional services and information for

http://pfr.sagepub.com/cgi/alertsEmail Alerts:

http://pfr.sagepub.com/subscriptionsSubscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

http://pfr.sagepub.com/content/37/5/613.refs.htmlCitations:

What is This?

- Aug 31, 2009Version of Record >>

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

Longitudinal Effects of Impact Fees and Special Assessments on the Level of Capital Spending, Taxes, and Long-Term Debt in American Cities Changhoon Jung Auburn University

Chul-Young Roh East Tennessee State University

Younguck Kang KDI School of Public Policy and

Management

This article examines whether the use of impact fees and special assessments

affect the level of capital spending and two major own source revenues of

local capital spending (taxes and long-term debt) by analyzing a panel of

695 American cities with populations over 20,000 during the time period

of 1980–2000. Since impact fees and special assessments are heavily used

in a growing community and because it covers less than half the costs of new

development, the findings demonstrate that the private financing of public

infrastructure (impact fees and special assessments) increases the level of

local capital spending. It also leads to an increase in the level of long-term

debt use. Although it provides partial tax relief, it is not a strong substitute

for taxes. Thus, impact fees and special assessments are not a substitute for

local capital spending. It is rather a supplemental revenue source to fund local

capital infrastructure.

Keywords: special assessments; impact fees; capital spending

Authors’ Note: Please address correspondence to Changhoon Jung, Department of Political

Science, Auburn University, Auburn, AL; e-mail: [email protected].

Public Finance Review

Volume 37 Number 5

September 2009 613-636

# 2009 The Author(s)

10.1177/1091142109344585

http://pfr.sagepub.com

613

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

1. Introduction

In the wake of the taxpayer revolts and tax and expenditure limitation

(TEL) movements that have swept the nation since the late 1970s, a grow-

ing number of localities in many states have used impact fees and special

assessments to help pay for the existing, new, or expanded infrastructure

needed to serve communities. The impact fee is a one-time fee designed

to pay for the increased demand for public services generated by a new

development. Impact fees for different facilities are currently used in local

governments in twenty-eight states (Duncan Associates 2007), and it is

likely that more states will follow suit (Clarke and Evans 1999; Nelson and

Moody 2003; Jeong 2006; Duncan Associates 2007). Similarly, special

assessments have been frequently used as a tool to finance urban develop-

ment and redevelopment in many special assessment districts and localities.

Given that the areas served by the new facilities are required to pay the cost

of needed improvements, special assessments are similar in their function

and economic effects to fiscal impact fees (Yinger 1998; Bierhanzl and

Downing 2004).

With the growing use of impact fees in local governments over time,

many have conjectured that the impact fee may influence the level of local

capital spending and other local revenue sources that are used to fund local

infrastructure (e.g., property taxes, total taxes, long-term debt proceeds).

Conceptually, since both impact fees and special assessments are designed

to transfer a portion of the capital costs for the new and expanded infrastruc-

ture from the public (taxpayers) to the private sector (developers and

builders), these revenue sources could be considered alternative means of

private financing of the portion of the public infrastructure, especially if the

impact fees do not affect the rate of development in the community (Yinger

1998; Clarke and Evans 1999; Burge, Nelson, and Matthews 2007). Since

the infusion of revenue from impact fees could affect other revenue streams

used to finance local capital spending (e.g., property taxes, total taxes, and

long-term debt proceeds), the use of impact fees may substitute for other

local revenue sources that were used to fund local infrastructure. However,

if the impact fees cover only a portion of infrastructure costs for the new

development and lead to an increased rate of development in the commu-

nity, then impact fees may increase the level of local capital spending rather

than substituting for other revenue streams typically used to fund local

infrastructure. In this case, impact fees may serve as a supplemental revenue

source for local capital spending. In practice, impact fees usually finance

614 Public Finance Review

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

less than half of the costs for new development, leaving local tax payers to

cover the remaining portion of the costs of new development in the form of

higher taxes or higher long-term debt use (Burge, Nelson, and Matthews

2007; Duncan Associates 2007). Furthermore, since the use of impact fees

could affect the path (rate) of development and/or the desired level of public

services in the community, this would further complicate the effects of

impact fees on the level of local capital spending and the streams of other

revenue sources that are used to pay for local capital spending (Burge,

Nelson, and Matthews 2007).

This article examines the effect of impact fees and special assessments

on the level of local capital spending and two major own-source revenues

of local capital spending (taxes and long-term debt) in a panel of 695 Amer-

ican cities with populations over 20,000 during the time period of 1980–

2000. Given that no studies have empirically examined the relationship

between impact fees (and special assessments) and capital spending and

major sources of revenue of local capital spending (taxes and long-term

debt) by using a longitudinal national data set, this study provides important

insights into our understanding on the nature of the impact fees and special

assessments.

The next section discusses analytical framework and literature review.

This is followed by section 3, which describes data, research methodology,

and variables used in the study. Research findings are then presented and

discussed. The final section concludes the article.

2. Analytical Framework and Literature

2.1. Analytical Framework and Research Hypotheses

Traditionally, the proceeds of long-term debt, current taxes, and intergo-

vernmental grants coming from federal and state governments have been

three major revenue sources to fund local capital projects (expenditure),

although the relative importance and composition of these revenue sources

for capital spending has varied among jurisdictions, projects, and time.

However, with the taxpayer revolts and voter-imposed TEL movements

sweeping the nation, many local governments have also relied on nontax

revenue sources such as charges and miscellaneous revenues (e.g., impact

fees and special assessments) to address growing needs of local capital

spending.

Jung et al. / Longitudinal Effects of Impact Fees 615

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

Since most jurisdictions prohibit using the proceeds of long-term debt to

finance current operating expenditures, almost all long-term debt proceeds

are used to fund capital improvements. Thus, the proceeds of long-term debt

have been one of the most important revenue sources for most local capital

improvement projects (Leonard 2004). Two basic types of long-term debt

include general obligation (GO) bonds and revenue debt. GO bonds are

secured by the full faith and credit of the taxing powers of the issuing gov-

ernment, thus legally obligating the local governments to levy taxes on all

assessable property within its jurisdiction at whatever level is necessary to

meet the debt service payments on the bonds. The repayment of the princi-

pal and interests of the GO debt usually comes from an increase in local

property tax rates or other taxes (e.g., the sales tax). In general, this unlim-

ited GO debt is an appropriate financing vehicle for capital projects that

benefit the entire community (e.g., public schools). However, since local

governments must pass referenda to use GO debt, and the principal and

interests of GO debt is usually serviced by raising the property tax millage

rate or other local taxes (e.g., sales tax), the tax payer revolts and the TELs

that swept the country in late 1970s and 1980s has resulted in less use of GO

debt vis-à-vis revenue debt, as demonstrated in the analysis of 695 panel

sample cities in figure 1.

Unlike GO debt, the repayment of the principal and interests of revenue

debt (bonds) is secured by the revenues collected from a specific project

being financed. Since the use of revenue bonds does not require the passage

Figure 1

Relative Composition of Outstanding General Obligation (GO) and

Revenue Debts of the Sample Cities

616 Public Finance Review

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

of referenda, many local governments have turned to revenue bonds and

other forms of financing (charges and fees) schemes since the late 1970s

to fund the growing demands of local infrastructure arising from population

growth and local economic development purposes (Leonard 2004).

Other than the property taxes, other tax revenues (such as local sales tax)

are also used to finance long-term capital spending. Since the 1980s, some

states like Georgia, South Carolina, and Wisconsin have permitted their

localities to use local sales taxes to pay for local capital spending (Jung

2002). Nontax revenue sources such as charges and fees and miscellaneous

revenues (e.g., impact fees and special assessments) have also emerged as

important sources of financing for growth and economic development since

the 1970s. These revenue sources have been used to bypass political, legal,

and administrative limitations imposed on the use of taxes (especially prop-

erty tax) and long-term debt and to realize the benefit-received principle of

infrastructure financing (Downing and Frank 1983; Mullins and Joyce

1996; Bierhanzl and Downing 2004).

Nontax revenue sources frequently used to fund growth and develop-

ment include development fees, fiscal impact fees, special assessments, and

tax increment financing (TIF). Development fees are user charges that

cover public infrastructure within the new development. These fees are

imposed directly on the developer of the new property. Although similar

to development fees, fiscal impact fees are often levied to pay for additions

to public infrastructure. The costs of these additions, such as an expansion

to a water treatment plant, will be shared between the current residents and

the new developments. Similar to both development fees and fiscal impact

fees, special assessments are based on the benefits-received principle. Spe-

cial assessments serve to make sure that residents pay for additional ser-

vices, such as sidewalks or street lights, which are added to their area.

While impact fees are only used for new capital investment purposes, spe-

cial assessments could be used not only for new development but also for

operations and maintenance, and the collection is also often recurring.

TIF provides an additional method for financing new development. Basi-

cally, TIF allows for the subsidization of new infrastructure costs within the

TIF district by diverting some of the property tax collection within the TIF

away from general revenue directly to infrastructure improvements. Through

TIF, property taxes in new developments and surrounding areas are set up by

local governments to mimic user charges. As a result of the improvements to

the public sector infrastructure, tax assessments are structured around the

benefits that accumulate for property owners (Bierhanzl and Downing

2004). Although there are similarities between development fees, fiscal

Jung et al. / Longitudinal Effects of Impact Fees 617

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

impact fees, special assessments, and TIF, it should be noted that develop-

ment fees, fiscal impact fees, and special assessments are loosely defined

as taxation while the TIF is kind of a subsidy to pay for the infrastructure.

While the use of impact fees and special assessments has increased over

time, there is little empirical evidence as to how impact fees and special

assessments affect the level of local capital spending and other fiscal revenues

streams used to fund local capital spending. While one may predict that impact

fees and special assessments simply substitute for other revenues sources for

local capital expenditure like taxes and long-term debt, this would not be the

case if the use of impact fees and special assessments affects the path (rate) of

development and/or the desired level of public services in the community.

It should be noted that, depending on the nature and design of the impact

fees, the use of impact fees may slow down, speed up, or have no effect on

residential and other types of construction. Conceptually, the use of impact

fees and special assessments could result in one of three reactions in the

development path in the community, which will in turn affect the level of

capital spending and two major own-source revenues of local capital spend-

ing (Yinger 1998; Ihlanfeldt and Shaughnessy 2004; Burge, Nelson, and

Matthews 2007). First, if the impact fees and special assessments work as

direct excise taxes on construction, it may slow down the rate of develop-

ment in the community since the additional taxes increase the price of land,

which has a negative impact on the number of parcels of affordable and

available land for new development. Second, if the collection of impact fees

and special assessments provides offsetting benefits such as better quality

local infrastructure and increased housing prices, then the rate of develop-

ment in the community may remain constant. Third, if the use of impact

fees and special assessments brings more benefits than the costs of new

infrastructure financing, then the rate of development in the community

actually may increase. In this case, the provision of infrastructure financed

by impact fees and special assessments would facilitate further develop-

ment, which may cause further population growth and business attraction.

Numerous factors are involved in determining the rate of community

development, which makes the prediction of the influence of impact fees

and special assessments on the level of capital spending and own-source

revenues of capital spending quite complicated. Other factors being equal,

potential consequences of the imposition of impact fees and special assess-

ment on the level of capital spending and its components (taxes and long-

term debt) would be as follows.

Under the first scenario, where the imposition of impact fees and special

assessments results in a slowdown of development in the community, the use

618 Public Finance Review

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

of impact fees and special assessments may result in a decrease in the level of

local capital spending since there will be less development and this will lower

the overall need for local capital spending. As the level of local capital spend-

ing decreases, it may also lower the level of taxes and long-term debt use.

Under the second scenario, the imposition of impact fees and special

assessment results in a constant rate of development in the community. If the

level of service provision remains the same, then the levying of impact fees

and special assessments may have no effect on the level of local capital

spending since the collection of impact fees will now partially finance infra-

structure costs that were once fully financed with taxes and debt proceeds. In

this case, impact fees and special assessments may act as a dollar-for-dollar

substitute for taxes, since the bulk of the principal and interest of GO debt is

repaid with increases in property tax and other taxes. If impact fees and spe-

cial assessments provide a dollar-for-dollar tax reduction, then they may also

lower pressure for the further use of long-term (GO) debt issuance.

Under the third scenario, where the provision of infrastructure financed by

impact fees and special assessments facilitates further development, the

impact fees will result in additional local capital spending needs due to an

increasing rate of development in the community. Given that impact fees usu-

ally cover less than half the cost of infrastructure for new development, the

collection of impact fees and special assessments may increase the level of

local capital spending, since the local government must spend additional

money to cover remaining infrastructure costs for new development under

this scenario. With this increased capital spending, the impact fees may result

in higher taxes and additional long-term debt use. In this case, the impact fees

may or may not provide tax relief. If there is any tax relief, it may be only

partial tax relief. In this case, the impact fees may increase the level of newly

issued debt (or outstanding long-term debt) since local governments should

rely more on debt issue to fund the costs of capital expansion for further

development while providing some tax relief. It seems highly likely that this

scenario provides the best fit for predicting the influence of the impact fees

and special assessments on the level of capital spending and two major

own-source revenues of local capital spending or taxes and long-term debt.

2.2. Research Hypotheses

Based on the above analytical framework, we propose that the provision

of infrastructure financed with impact fees and special assessments would

facilitate further development, which will result in additional local capital

Jung et al. / Longitudinal Effects of Impact Fees 619

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

infrastructure spending needs. The collection of impact fees may or may not

provide partial tax relief, but since it does increase the rate of development,

it would also increase the level of long-term debt use for funding further

development in the community. Thus, we come up with the following

hypotheses. Hypothesis 1: The use of impact fees and special assessments increases the level

of local capital spending.

Hypothesis 2: The use of impact fees and special assessments provides partial tax

relief.

Hypothesis 3: The use of impact fees and special assessments leads to an increase

in the level of long-term debt.

2.3. Literature

Reflecting growing interest in the use and effects of impact fees and spe-

cial assessments, a sizable amount of research has addressed different

aspects of impact fees and special assessments. Although the economic

effects of special assessments are similar to those of impact fees, most of

the research focuses on impact fees in part due to their increasing use as rev-

enue and its relative newness compared to special assessment in local

finance. The 2007 National Impact Fee Survey by Duncan Associates

describes the recent use and pattern of impact fees in the nation (Duncan

Associates 2007). Compared to its 2003 survey, the survey reports growth

in the number of states and localities using impact fees, as well as the pur-

poses for which they are used. The survey also reports that impact fees are

widely used in the South and West, particularly Washington, Oregon, Cali-

fornia, Arizona, Colorado, and Florida. Impact fees are less frequently used

in the Northeast and Midwest.

Several articles report factors affecting the adoption of impact fees

(Frank and Downing 1988; Clarke and Evans 1999; Lawhon 2003; Jeong

2006). These works find that enabling state legislation, community growth,

fiscal health and stress, form of government, and administrative capacity

influence the likelihood of local adoption of impact fees. In general, impact

fees and special assessments are used in states where the state government

permits its local government to use the revenue. In these states, growing

communities and fiscally stressed cities are more likely to adopt impact

fees. Likewise, reformed cites (council-manager cities) and cities that have

higher administrative capacity are more likely to collect impact fees.

A growing body of literature also addresses the effects of impact fees on

several aspects of community growth and development. These include the

620 Public Finance Review

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

effects of impact fees on housing price and housing production and

economic development and job growths. Although Burge, Nelson, and

Matthews (2007) summarize major findings on the various topics, these

issues are still controversial and complex. This section briefly introduces

the main issues of these topics.

With respect to the effect of impact fees on housing price, the ‘‘tradi-

tional view’’ and the ‘‘new view’’ of the property tax make different predic-

tions concerning the short-run impact. The ‘‘traditional view’’ considers

impact fees to be an excise tax on the production of housing. This perspec-

tive assumes that the imposition of impact fees leads to an increase in hous-

ing price, which lowers profits for developers’ on the quantity of new

homes built (Snyder, Stegman, and Moreau 1986; Downing and McCaleb

1987; Huffman et al. 1988; Delaney and Smith 1989a, 1989b; Singell and

Lillydahl 1990; Altshuler and Gomez-Ibanez 1993).

Yinger (1998) and the following works by Nelson and Moody (2003),

Ihlanfeldt and Shaughnessy (2004), and Burge and Ihlanfeldt (2006a) chal-

lenge this view. The ‘‘new view’’ perceives impact fees not as an excise tax

but as benefit principle–based user fees and charges; that is, although the

impact fee imposes costs for developers in the permitting process, it also

brings several offsetting positive effects just as do other forms of benefit

principle–based charges and fees. This perspective says that impact fees

reflect the cost of providing the valued facilities needed to serve new devel-

opment, and thus they may offset property taxes that would otherwise have

been assessed. Put differently, improved infrastructure in the newly devel-

oped area, made possible by the imposition of impact fees, will eventually

increase the property value of the home in the area. This will also lower

future property tax rates in the area (Burge, Nelson, and Matthews 2007).

This mechanism needs a bit more explanation since it explains how

impact fees could affect the level of property taxes. If one assumes that a

local government relies almost exclusively on property taxes to finance new

facilities, and that there is no change in the quality of the facilities accom-

panying the move from property taxes to impact fees, then this view sug-

gests that prices for both new and existing homes will go up. However,

the increase in prices should equal the capitalized value of the property tax

savings that homeowners expect from the reduction in the tax rate. In this

case, the property tax rate declines because the imposition of the impact

fees shifts the costs of new infrastructure from existing property owners

to developers. For example, Ihlanfeldt and Shaughnessy (2004) demon-

strate that a U.S.$1 impact fee collection resulted in about U.S.$1.20 prop-

erty tax relief in Dade County, Florida. Thus, if appropriately structured, the

Jung et al. / Longitudinal Effects of Impact Fees 621

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

use of impact fees and special assessments could transfer a portion of the

burden of infrastructure costs from the local government (taxpayers) to the

private sector (builders and developers) because builders and developers

help pay for the infrastructure that provides access, utilities, and other ame-

nities critical to a development’s success (Yinger 1998; Clarke and Evans

1999; Ihlanfeldt and Shaughnessy 2004).

Concerning the effects of impact fees on housing production, the main

question is whether the imposition of impact fees will affect the production

of new affordable housing (Brueckner 1997; Skidmore and Peddle 1998;

Mayer and Somerville 2000; Burge and Ihlanfeldt 2006a, 2006b). Critics,

who perceive impact fees as excise taxes, argue that impact fees may nega-

tively affect the amount of buildable land and housing, and thus, the number

of units of affordable housing will go down. However, the opposite view,

which perceives impact fees as fees or charges, proposes that the use of

impact fees for infrastructure enhancement may actually increase the supply

of buildable land in the community rather than stifling housing production.

Since the units of housing production are determined both by supply and

demand in the housing market, the effects of impact fees on residential con-

struction rates are more complicated than their effects on prices. To date,

few studies have directly investigated the effects of impact fees on housing,

and Burge, Nelson, and Matthews (2007) concludes that the effects are

quite ambiguous.

With respect to the effects of impact fees on economic development, the

‘‘traditional view’’ of impact fees suggests that the imposition of impact

fees may stifle economic development and job growth. However, the ‘‘new

view’’ sees that better infrastructure provision with the impact fees collec-

tion may further promote economic development and job growth. Empirical

findings demonstrate that impact fees spent on infrastructure tend to stimu-

late economic development and job growth (Nelson and Moody 2003;

Jeong and Feiock 2006).

While a sizable amount of research has addressed various effects of

impact fees, to date only Clark and Evans (1999) have analyzed whether the

collection of impact fees results in increased local capital spending. They

analyzed the relationship between the use of impact fees and the level of

capital spending, by examining eighty-five American cities for a single

cross-section year, and they found that the use of impact fees lowered the

level of capital spending. Although the study is one of the first to look at

the relations between impact fees and capital spending, Clark and Evans

(1999) did not control for other revenue sources that could affect the level

of capital spending (e.g., long-term debt, taxes, intergovernmental grants,

622 Public Finance Review

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

and charges). Likewise, they did not consider the influence of impact fees

on the path of development in the community. Moreover, their study did not

examine how impact fees affect the major revenue sources for capital

spending (taxes and long-term debt). Since Clark and Evans (1999) exam-

ined only a single year data set for a limited sample, it is difficult to general-

ize from their study. Given that local capital investments are largely

nonrecurrent and require many years to be replaced (or repaid), it is neces-

sary to examine a large national sample over a long period of time to get a

better understanding of the effects of impact fees on the level of capital

expenditure and major revenue sources of capital expenditure. This study

fills the research void by analyzing the effects of impact fees and special

assessments on the level of capital spending and two major own-source rev-

enues of capital spending (taxes and long-term debt) for a panel of 605

American cities over a twenty-one-year period.

3. Research Methods

3.1. Data

This study used an annual data set of a panel of 695 American cities with

populations over 20,000 for the entire study period (1980–2000). In 1980,

there were more than 19,000 American cities, of which about 1,300 cities

reported having a population of over 20,000. Since we are interested in

building a panel sample, we selected cities whose demographic, socioeco-

nomic, financial, and other variables are available for the study period. This

reduced the size of the panel sample to 695 cities. It would have been pre-

ferable to examine more cities, especially cities with populations under

20,000 to better understand longitudinal local fiscal behavior. However,

given the practical difficulty of collecting relevant variables for a panel data

set over twenty-one years, this study limited the sample to cities with popu-

lations of over 20,000, especially because of the practical difficulty of col-

lecting demographic, socioeconomic, fiscal, institutional, and other

variables for smaller cities. It should be noted that many small cities did not

consecutively submit their financial and other data set information to the

Census Bureau during the study period.

With respect to impact fees and special assessments, the Census Bureau

did not report impact fees and special assessments separately in its annual

Local Finance series. Rather, impact fees and special assessments are

lumped together and reported as ‘‘special assessments’’ under the

Jung et al. / Longitudinal Effects of Impact Fees 623

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

‘‘miscellaneous revenue’’ section. If impact fees and special assessments

were separately reported, a separate estimation of the effect of each revenue

source would be possible. However, the aggregation of the impact fees and

special assessments makes it almost impossible to analyze the separate

effects of each. It should be noted that most local governments in most

states do not report impact fees and special assessments separately either.

Rather, they report it under ‘‘other revenues’’ or ‘‘miscellaneous revenue.’’

For example, the State of Georgia reports the impact fees and special assess-

ments of its local governments under ‘‘other revenues’’ in its annual Report

of Local Government Finance.

The Census Bureau defines special assessments as ‘‘compulsory contribu-

tions and reimbursements from owners of property who benefit from specific

public improvements; and impact fees to fund extension of water, sewer,

roads, and other infrastructure facilities in new developments.’’ Furthermore,

the Bureau goes on to elaborate that ‘‘these contributions and reimbursements

are designed to defray all or part of the costs of such improvements, either

directly or through payment of principal and interest on debt issued to finance

them’’ (www.census.gov/govs/www.class_ch7_misc.html). It should be

noted that special assessments cover a range of upgrades from general street

improvements such as sidewalks to utility improvements such as waterlines.

In general, these special assessments are assigned based on the assumed ben-

efits that would result from the improvements.

\The annual finance data set for the study was collected from the Census

Bureau’s Local Finance. Demographic and socioeconomic variables, such as

size of population, income, and age, were collected from the Census Bureau’s

decennial Population and Housing files and County and City Data Book. Forms

of municipal governments were collected from the International City/County

Management Association’s (ICMA) The Municipal Year Book series.

3.2. Research Model and Variables

The local finance literature indicates that the level of capital expenditure,

total taxes, and long-term debt are affected by demographic, socioeco-

nomic, fiscal, and institutional factors (Chicone and Walzer 1986; Poterba

1995; Jung 2001). Since we estimate the influence of impact fees and spe-

cial assessments on the level of capital expenditure, total taxes, and long-

term debt, we use the same sets of control variables to maintain consistency

and facilitate the comparison of estimates among models. Thus, the primary

model we use in the study can be written as

624 Public Finance Review

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

Level of capital expenditure, total taxes, long-term debt ¼ f (demo- graphic, socioeconomic, fiscal, institutional factors).

The three dependent variables are capital expenditure, total taxes, and

long-term debt. These, and all other fiscal variables, are expressed in per

capita values to mitigate possible heterogeneity problems among different

localities.

The primary independent variable is lagged per capita impact fees and

special assessment (Impact feest – 1). Since the current level of impact fees

collection could either increase or decrease current levels of taxes or long-

term debt, and vice versa, we use a lagged value of impact fees and special

assessments to alleviate potential reverse causality problems with the

dependent variables in the study.

To capture the demographic factors, we use three variables: population

size, population density, and percentage of population under eighteen. If

annual population statistics were available for the panel sample, we could

use the population growth rate to estimate the effects of population growth

on dependent variables. However, the annual population is not available,

and so we use the natural log of population size to indirectly capture the

effects of population growth on the dependent variables. Population density

is expressed as population per square mile. The percentage of population

under eighteen is frequently used to capture local service demands. Other

things being equal, if the percentage of population below eighteen

increases, this would place more pressure on local governments to increase

capital spending to provide additional infrastructure (e.g., public schools).

To capture socioeconomic factors in local finance, we use a per capita

income variable. Other things being equal, higher income leads to a higher

level of, or more varied, local services. To reflect the impact of institutional

factors on local finance, we use the form of municipal government. Profes-

sional cities under a council-manager form of governance are known to be

more active in developing and using new revenue sources for long-term

growth and economic development compared to the mayor-council or other

forms of cities (Mullins and Joyce 1996; Jeong 2006). For these reasons,

council-manager cities are more likely to use impact fees and special assess-

ments than other cities. Professional city management has also been shown

to affect the level of taxes, capital expenditures, and long-term debt

(Chicone and Walzer 1986; Poterba 1995; Jung 2001). To capture the like-

lihood of such innovation, we use a council-manager dummy variable,

coded 1 for council-manager form cities and 0 for others.

Since it is necessary to control for other internal and external sources of

revenue such as charges and fees (total charges), intergovernmental grants,

Jung et al. / Longitudinal Effects of Impact Fees 625

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

taxes, and long-term debt that could also affect the level of dependent vari-

ables, we use lagged values of these fiscal controls to alleviate the possibil-

ity of reverse causality and endogeneity problems (Wooldrige 2006). To

control for inflation, all the fiscal variables are converted into constant dol-

lars using the 1980 gross domestic product (GDP) implicit price deflator for

state and local government purchases as a base.

Table 1 presents variable definitions, sources of data set, and summary

statistics. As table 2 indicates, 181 cities collected some amount of impact

fees and special assessments consecutively during the entire study period,

while fifty-one cities never used this source.

3.3. Statistical Models

Since there are variations in the level of capital spending, taxes, and

long-term debt among sample cities over time, we use a fixed effects model,

which controls unit-specific fixed effects in the estimation (Wooldrige

2002). Likewise, since the year could also affect the dependent variables,

control for time effects must be considered. Hausman tests and joint signif-

icance tests of time effects for all the proposed models in the study show

that the sample data have both unit-specific and time-specific effects, and

we address these issues by using a two-way fixed effects model in the esti-

mation (Wooldrige 2002; Baltagi 2008).

4. Findings

Table 3 presents the results of the two-way fixed effects regression

model estimating the effects of impact fees and special assessments on the

level of local capital spending, property and total taxes, and long-term debt.

As pointed out above, since the primary own-source revenue for local cap-

ital spending comes from local taxes (such as property and sales taxes) and

the proceeds of long-term debt (both GO and revenue debt), an analysis of

the effect of impact fees and special assessments on taxes and long-term

debt provides a better understanding of how each component affects the

level of local capital spending. Since the property tax is the primary tax rev-

enue for most local governments as well as the primary revenue source to

repay the principal and interests of GO debt, we present separate effects

of the variable.

626 Public Finance Review

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

T a

b le

1

V a

r ia

b le

D e fi

n it

io n

a n

d S

u m

m a

r y

S ta

ti st

ic s

V a ri

a b le

s D

e fi

n it

io n

a n d

S o u rc

e o f

D a ta

M e a n

S ta

n d a rd

D e v ia

ti o n

M in

im u m

M a x im

u m

D e p e n d e n t

v a ri

a b le

s

C a p it

a l

e x p e n d it

u re

P e r

c a p it

a c a p it

a l

sp e n d in

g a

7 9 .0

3 8 1 .5

8 .6

1 ,6

7 2

C u rr

e n t

e x p e n d it

u re

P e r

c a p it

a c u rr

e n t

o p e ra

ti n g

sp e n d in

g a

4 0 5 .1

2 8 0 .7

1 7 .6

9 ,0

9 2

T o ta

l e x p e n d it

u re

P e r

c a p it

a to

ta l

e x p e n d it

u re

a 4 8 4 .2

3 1 8 .5

2 1 .3

9 ,1

4 8

P ro

p e rt

y ta

x P

e r

c a p it

a p ro

p e rt

y ta

x e sa

9 7 .5

1 0 5 .0

5 1 0 .3

1 ,0

3 8 .2

T o ta

l ta

x e s

P e r

c a p it

a to

ta l

ta x e sa

9 7 .5

1 0 5 .1

1 5 .3

1 ,0

3 8 .2

L o n g -t

e rm

d e b t

o u ts

ta n d in

g

P e r

c a p it

a c o m

b in

e d

g e n e ra

l o b li

g a ti

o n

(G O

)

a n d

re v e n u e

d e b t

o u ts

ta n d in

g a

5 3 1 .2

9 7 5 0 .4

0 2 9 ,5

8 0

G O

d e b t

o u ts

ta n d in

g P

e r

c a p it

a G

O d e b t

o u ts

ta n d in

g a

2 6 4 .6

2 9 9 .5

0 4 ,5

9 8 .5

R e v e n u e

d e b t

o u ts

ta n d in

g P

e r

c a p it

a re

v e n u e

d e b t

o u ts

ta n d in

g a

5 1 6 .8

9 8 8 .9

0 3 4 ,2

6 7 .9

L o n g -t

e rm

d e b t

is su

e d

P e r

c a p it

a c o m

b in

e d

G O

a n d

re v e n u e

d e b t

is su

e d

a 1 1 6 .4

2 7 5 .3

0 1 2 ,5

1 7 .5

G O

d e b t

is su

e d

P e r

c a p it

a G

O d e b t

is su

e d

a 9 2 .8

2 1 9 .6

0 1 2 ,3

7 9 .3

R e v e n u e

d e b t

is su

e P

e r

c a p it

a re

v e n u e

d e b t

is su

e d

a 2 3 .6

1 4 8 .3

0 1 0 ,1

1 2 .9

In d e p e n d e n t

v a ri

a b le

s

P o p u la

ti o n

(l n )

P o p u la

ti o n

(l n )a

4 .8

6 0 .3

5 4 .3

6 .9

D e n si

ty P

o p u la

ti o n

d e n si

ty p e r

sq u a re

m il

e b

3 ,7

0 0 .5

3 ,5

9 5 .9

4 6 .5

5 1 ,6

0 6

B e lo

w 1 8

P e rc

e n ta

g e

o f

p o p u la

ti o n

b e lo

w 1 8

y e a rs

o f

a g e

b 2 5 .5

4 .7

8 6 .2

4 8 .7

In c o m

e P

e r

c a p it

a p e rs

o n a l

in c o m

e b

1 2 ,4

7 0

6 ,2

5 0

7 ,8

1 9

6 5 ,5

0 7

C o u n c il

-m a n a g e r

If c o u n c il

-m a n a g e r

fo rm

o f

g o v e rn

m e n t

c o d e d

1 ,

o th

e rw

is e

0 c

0 .5

9 0 .4

9 0

1

Im p a c t

fe e

P e r

c a p it

a im

p a c t

fe e s

a n d

sp e c ia

l a ss

e ss

m e n ta

4 .1

2 1 1 .4

5 0

2 9 2

(c o n ti

n u e d )

627

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

T a

b le

1 .

(c o

n ti

n u

e d

)

V a ri

a b le

s D

e fi

n it

io n

a n d

S o u rc

e o f

D a ta

M e a n

S ta

n d a rd

D e v ia

ti o n

M in

im u m

M a x im

u m

G ra

n ts

P e r

c a p it

a in

te rg

o v e rn

m e n ta

l g ra

n ts

fr o m

st a te

a n d

lo c a l

g o v e rn

m e n ts

a

9 7 .5

1 1 1 .7

0 1 ,1

9 6

T o ta

l c h a rg

e s

P e r

c a p it

a to

ta l

n o n ta

x re

v e n u e s

e x c e p t

fo r

m is

c e ll

a n e o u s

re v e n u e sa

2 2 1 .1

2 2 3 .0

1 .1

2 1 1 ,1

8 9 .1

a S

e e

C e n su

s o f

L o c a l

G o v e rn

m e n t

F in

a n c e ,

1 9 8 0 – 2 0 0 0 .

b S

e e

C o u n ty

a n d

C it

y D

a ta

B o o k

1 9 8 2 ,

1 9 8 7 ,

1 9 9 2 ,

1 9 9 7 .

c S

e e

IC M

A ,

T h e

M u n ic

ip a l

Y e a r

B o o k

1 9 8 0 – 2 0 0 0 .

T a

b le

2

N u

m b

e r

o f

Y e a

r s

in W

h ic

h Im

p a

c t

F e e s

a n

d S

p e c ia

l A

ss e ss

m e n

ts W

e r e

C o

ll e c te

d in

th e

S a

m p

le C

it ie

s

N u m

b e r

o f

y e a rs

u se

d 2 1

2 0

1 9

1 8

1 7

1 6

1 5

1 4

1 3

1 2

1 1

1 0

9 8

7 6

5 4

3 2

1 0

N u m

b e r

o f

c it

ie s

u se

d 1 8 1

4 7

4 1

2 9

3 3

2 3

2 1

2 0

1 8

2 3

2 1

1 3

2 6

2 2

1 7

2 8

1 4

1 5

1 4

2 1

1 7

5 1

628

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

T a

b le

3

E ff

e c ts

o f

Im p

a c t

F e e s

a n

d S

p e c ia

l A

ss e ss

m e n

ts o

n C

a p

it a

l E

x p

e n

d it

u r e ,

T a

x e s,

a n

d L

o n

g -T

e r m

D e b

t— F

ix e d

E ff

e c ts

M o

d e l

V a ri

a b le

s C

a p it

a l

E x p e n d it

u re

P ro

p e rt

y T

a x

T o ta

l T

a x e s

L o n g -t

e rm

D e b t

O u ts

ta n d in

g

L o n g -t

e rm

D e b t

Is su

e d

P o p u la

ti o n

(l n )

– 3 3 .5

3 * * *

(7 .3

0 )

1 0 .0

6 * * *

(5 .4

7 )

4 .3

0 * *

(1 .9

7 )

– 1 4 6 .2

4 * *

(4 .1

1 )

– 8 9 .7

9 * * *

(4 .8

4 )

D e n si

ty – 0 .0

0 0 0 3

(0 .1

5 )

– 0 .0

0 0 0 7

(0 .7

6 )

– 0 .0

0 0 0 4

(0 .3

5 )

– 0 .0

0 1

(0 .6

2 )

– 0 .0

0 0 4

(0 .4

5 )

B e lo

w 1 8

– 0 .2

3 (1

.2 9 )

– 0 .1

9 * * *

(2 .6

0 )

– 0 .2

9 * * *

(3 .3

4 )

6 .2

3 * * *

(4 .3

9 )

1 .1

6 (1

.5 6 )

In c o m

e (p

e r

c a p it

a )

2 .4

8 e – 0 6

(0 .0

1 )

– 0 .0

0 0 2

* * *

(3 .2

3 )

– 0 .0

0 0 2

* *

(2 .1

9 )

0 .0

0 6

* * *

(4 .2

7 )

0 .0

0 2

* * *

(2 .8

7 )

C o u n c il

-m a n a g e r

– 2 .5

3 (0

.5 2 )

– 2 .7

2 (1

.3 8 )

– 2 .9

2 (1

.2 5 )

7 3 .7

8 *

(1 .9

3 )

1 7 .9

7 (0

.9 0 )

Im p a c t

fe e s t

– 1

0 .2

9 * * *

(4 .2

4 )

– 0 .0

6 * *

(2 .2

0 )

– 0 .1

2 * * *

(3 .6

1 )

0 .6

6 (1

.2 4 )

1 .2

3 * * *

(4 .4

6 )

G ra

n ts

t –

1 0 .2

2 * * *

(1 6 .5

1 )

0 .1

0 * * *

(2 0 .3

2 )

0 .0

8 * * *

(1 3 .4

8 )

0 .4

5 * * *

(4 .4

2 )

0 .1

7 * * *

(3 .1

3 )

T o ta

l ta

x e s t

– 1

0 .0

4 * *

(2 .4

0 )

0 .8

6 * * *

(6 .3

2 )

0 .1

6 * *

(2 .2

3 )

T o ta

l c h a rg

e s t

– 1

– 0 .0

2 * * *

(2 .6

7 )

0 .0

0 3

(1 .4

1 )

0 .0

0 6

* *

(2 .0

1 )

1 .2

7 * * *

(2 9 .2

0 )

– 0 .1

2 (0

.9 9 )

L o n g -t

e rm

d e b t

o u ts

ta n d in

g t

– 1

0 .0

2 * * *

(1 5 .6

8 )

0 .0

0 3

* * *

(6 .3

0 )

0 .0

0 3

* * *

(4 .8

0 )

C o n st

a n t

4 0 9 .3

6 – 3 7 .4

1 8 9 .2

2 1 8 1 8 .5

3 9 9 7 .7

3

N 1 3 ,9

0 0

1 3 ,9

0 0

1 3 ,9

0 0

1 3 ,9

0 0

1 3 ,9

0 0

R 2

.0 9

.1 1

.1 0

.1 0

.0 3

N o te

: T

h e

n u m

b e rs

in p a re

n th

e se

s a re

a b so

lu te

t v a lu

e s.

T im

e (y

e a r)

e ff

e c ts

a re

n o t

re p o rt

e d

h e re

. *

p <

.1 .

* *

p <

.0 5 .

* * *

p <

.0 1 .

629

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

The first column of table 3 presents estimates for the capital expenditure

model; the second column (the property tax model) and the third column

(the total taxes model) show the estimates for the property tax and for total

taxes, respectively. The fourth and fifth columns present estimates for out-

standing long-term debt and for long-term debt issued, respectively. To get

detailed effects of impact fees and special assessments on different types of

debt (GO and revenue) and expenditures (current and total expenditures),

we also present the results of these estimates in table 4.

The positive and statistically significant coefficient (0.29) of the impact

fees variable (Impact feest – 1) in the capital expenditure model in table 3

indicates that one dollar of impact fees and special assessments leads to

an increase of about twenty-nine cents in total capital spending. However,

a positive but insignificant coefficient of the impact fees and special assess-

ments (0.17) variable in the final column (the total expenditure model) in

table 4 indicates that impact fees and special assessments do not increase

the level of total government expenditure. Similarly, a negative and insig-

nificant coefficient of the impact fees and special assessments variable in

the fifth column of table 4 indicates that impact fees and special assess-

ments do not increase the level of current expenditure. Given that total

expenditure is composed of current operating plus capital expenditure, these

findings indicate that impact fees and special assessments do not lead to an

increase in the level of current operating or total expenditure, but they do

significantly increase the level of capital spending. Additionally, impact

fees and special assessments do not lead to a dollar-for-dollar increase in

capital spending since portions of the impact fees and special assessments

are used to provide tax relief or other revenue relief.

An examination of the relative magnitude of major revenue sources on

capital spending in the capital expenditure model in column 1 of table 3 pro-

vides an interesting finding. While a one dollar increase in impact fees and

special assessments revenue increases the level of local capital spending

about 29 cents, an increase of one dollar in federal and state grants brings

about a twenty-two-cent increase in capital spending. This is followed by total

taxes with a four cent increase and long-term debt with a two cent increase.

These results suggest that impact fees and special assessments could be an

effective and powerful tool to increase the level of local capital spending.

To estimate the effects of impact fees and special assessments on the

magnitude of tax relief, we present both a property tax and a total taxes

model for the reasons explained above. A negative and statistically signif-

icant coefficient (–.06) of the impact fees and special assessments variable

in the property tax model in the second column of table 3 indicates that a

630 Public Finance Review

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

T a

b le

4

E ff

e c ts

o f

Im p

a c t

F e e s

a n

d S

p e c ia

l A

ss e ss

m e n

ts o

n G

e n

e r a

l O

b li

g a

ti o

n (G

O )

a n

d R

e v

e n

u e

D e b

t a

n d

E x

p e n

d it

u r e s

V a ri

a b le

s

G O

D e b t

O u ts

ta n d in

g

R e v e n u e

D e b t

O u ts

ta n d in

g G

O D

e b t

Is su

e d

R e v e n u e

D e b t

Is su

e d

C u rr

e n t

E x p e n d it

u re

T o ta

l E

x p e n d it

u re

P o p u la

ti o n

(l n )

� 8 0 .8

8 * * *

(6 .0

5 )

� 6 5 .3

6 *

(1 .9

0 )

� 5 0 .6

7 * * *

(3 .3

6 )

� 3 9 .1

2 * * *

(3 .7

5 )

� 2 9 .1

3 * * *

(3 .9

2 )

� 6 2 .6

7 * * *

(6 .9

8 )

D e n si

ty 0 .0

0 0 4

(0 .5

9 )

� 0 .0

0 2

(0 .8

7 )

4 .5

3 e �

0 6

(0 .0

1 )

� 0 .0

0 0 4

(0 .8

1 )

� 0 .0

0 0 4

(0 .9

6 )

� 0 .0

0 0 4

(0 .8

7 )

B e lo

w 1 8

0 .8

8 (1

.6 5 )

5 .3

6 * * *

(3 .9

0 )

0 .4

0 (0

.6 6 )

0 .7

6 *

(1 .8

2 )

� 0 .0

7 (0

.2 0 )

� 0 .2

9 (0

.8 2 )

In c o m

e (p

e r

c a p it

a )

0 .0

0 3

* * *

(6 .4

0 )

0 .0

0 3

* (1

.9 3 )

0 .0

0 1

* *

(2 .1

0 )

0 .0

0 0 8

* *

(2 .0

6 )

� 0 .0

0 0 3

(1 .0

6 )

� 0 .0

0 0 3

(0 .8

7 )

C o u n c il

-m a n a g e r

� 5 .5

0 (0

.3 8 )

7 9 .2

9 * *

(2 .1

5 )

5 .5

1 (0

.3 4 )

1 2 .4

6 (1

.1 1 )

� 2 .8

2 (0

.3 5 )

� 5 .3

6 (0

.5 6 )

Im p a c t

fe e s t �

1 1 .4

1 * * *

(7 .1

3 )

� 0 .7

6 (1

.4 9 )

0 .5

6 * *

(2 .4

8 )

0 .6

7 * * *

(4 .3

4 )

� 0 .1

2 (1

.0 9 )

0 .1

7 (1

.2 7 )

G ra

n ts

t �

1 0 .2

7 * * *

(6 .8

4 )

0 .1

9 *

(1 .9

1 )

0 .1

1 * *

(2 .5

1 )

0 .0

6 * * *

(4 .3

4 )

0 .5

1 * * *

(2 4 .0

5 )

0 .7

3 * * *

(2 8 .3

5 )

T o ta

l ta

x e s t �

1 0 .3

0 * * *

(5 .7

9 )

0 .5

7 * * *

(4 .2

9 )

0 .0

8 (1

.3 7 )

0 .0

8 * *

(1 .9

8 )

0 .4

4 * * *

(1 5 .3

4 )

0 .4

8 * * *

(1 3 .9

2 )

T o ta

l c h a rg

e s t �

1 0 .0

3 *

(1 .8

6 )

1 .2

4 * * *

(2 9 .4

8 )

� 0 .0

8 * * *

(4 .4

0 )

� 0 .0

4 * * *

(2 .9

7 )

0 .1

1 * * *

(1 1 .2

1 )

0 .0

9 * * *

(7 .9

0 )

L o n g -t

e rm

d e b t

o u ts

ta n d in

g t �

1

0 .0

4 * * *

(2 2 .0

5 )

0 .0

6 * * *

(2 6 .2

7 )

C o n st

a n t

1 ,0

4 9 .0

6 7 6 9 .4

7 1 0 5 .6

9 4 0 4 .7

2 5 1 7 .0

3 9 2 6 .3

7

N 1 3 ,9

0 0

1 3 ,9

0 0

1 3 ,9

0 0

1 3 ,9

0 0

1 3 ,9

0 0

1 3 ,9

0 0

R 2

.0 5

.0 9

.0 2

.0 2

.1 9

.2 3

N o te

: T

h e

n u m

b e rs

in p a re

n th

e se

s a re

a b so

lu te

t v a lu

e s.

T im

e (y

e a r)

e ff

e c ts

a re

n o t

re p o rt

e d

h e re

. *

p <

.1 .

* *

p <

.0 5 .

* * *

p <

.0 1 .

631

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

one dollar impact fees and special assessment revenue collection provides

about 6 cents of property tax relief. Likewise, the negative coefficient of

impact fees and special assessments in the total taxes model (the third col-

umn of table 3) indicates that a one dollar impact fees and special assess-

ment collection provides about twelve cents total tax relief. These

findings suggest that impact fees and special assessments provide a small

amount of tax relief for both property and total taxes, which is far from

being a dollar-for-dollar tax relief as reported in Ihlanfeldt and Shaughnessy

(2004). Our finding confirms the new view’s perspective that the imposition

of impact fees shifts part of the property tax (and the total tax) burden from

existing property owners to builders and developers.

Since the influence of impact fees and special assessments on long-term

debt could be analyzed by examining both the outstanding and newly

issued debt, we present both the outstanding long-term debt and newly

issued long-term debt models in table 3. We are more interested in the

effects of impact fees and special assessments on additional debt use, so

we place more emphasis on the long-term debt issued model. However,

as several factors affect the amount, types, and timing of long-term debt

issue, we also examine both the outstanding (unpaid accumulated debt) and

newly issued debt to understand the influence of impact fees on the use of

long-term debt. In addition, since local governments could use both GO and

revenue debt to fund additional infrastructure costs, we present the break-

down of total outstanding and newly issued debt in table 4 to get a better

understanding of the relative influence of the impact fees on the use of these

debt instruments.

The long-term debt outstanding model and the long-term debt issued

model of columns 4 and 5 of table 3 present the effects of impact fees and

special assessments on the level of outstanding long-term debt and newly

issued long-term debt, respectively. Although the R 2

is very low, the posi-

tive and significant coefficient of the impact fees and special assessments

variable (1.23) in the long-term debt issued model indicates that a one dollar

increase in impact fees and special assessments leads to an increase of

U.S.$1.23 in new long-term debt issue. This suggests that the use of impact

fees will bring more development into the community, some of which is

financed with newly issued long-term debt. When the total newly issued

debt is broken down into the GO and revenue debt, the collection of

impact fees and special assessments brings about a fifty-six-cent increase

in GO debt while bringing about a sixty-seven-cent increase in revenue

debt, as seen in the third (GO debt issued model) and the fourth (revenue

632 Public Finance Review

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

debt issued model) columns of the table 4. This confirms the trend that

revenue debt is more heavily used to fund additional infrastructure

(Leonard 2004).

Unlike the newly issued debt, impact fees and special assessments in the

outstanding long-term debt model are not significant, as seen in the fourth

column of table 3. However, when the total outstanding debt is broken into

GO debt outstanding (first column) and revenue debt outstanding (second

column), as seen in the table 4, impact fees and special assessments signif-

icantly increase the level of outstanding GO debt, which they do not do for

revenue debt even in this model. The above empirical findings confirm all

three hypotheses proposed earlier.

Taken together, the above findings show that impact fees lead to an

increase in local capital spending. Thus, the use of private financing of pub-

lic infrastructure via impact fees and special assessments did not substitute

for municipal capital spending. While impact fees and special assessments

provide a small amount of tax relief, they do not provide a dollar-for-dollar

property or total tax relief. This suggests that impact fees and special assess-

ments are not strong substitutes for local taxes. Moreover, since impact fees

and special assessments lead to an increase in the level of long-term debt

issue, they do not remove the pressure to use long-term debt. Since one dol-

lar in impact fees and special assessments results in an increase of 29 cents

in capital spending and since the magnitude of long-term debt use

(U.S.$1.23) far outweighs that of total tax relief (12 cents), it could be inter-

preted that impact fees and special assessments are more of a supplement to

capital spending.

The implications of the empirical findings in the study suggest that sce-

nario 3 (impact fees speed up development in the community) is operative

in the panel sample cities; that is, it is highly likely that impact fees and spe-

cial assessments are heavily used in growing cities and that the use of

impact fees and special assessments promote further development. Since

impact fees and special assessments usually cover less than half of costs

of new development, their use results in more local capital spending, which

is financed with long-term debt issue and current taxes. In the process, the

use of private financing of public infrastructure does not provide a large

enough revenue source for the local government to offset or substitute for

municipal capital expenditure. This means that local governments must

make additional capital expenditures to subsidize the infrastructure financ-

ing for the new development and redevelopment in the growing areas. Thus,

the findings indicate that impact fees and special assessments do not

Jung et al. / Longitudinal Effects of Impact Fees 633

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

significantly substitute for local infrastructure spending. Rather, they serve

as supplemental revenue to local capital spending.

5. Conclusion

Our estimation results suggest that the use of impact fees and special

assessments accelerates the rate of development in the community. Since

typical impact fees cover less than half the costs of new development in

growing communities, the remainder is covered with increased debt issue

or taxes. These findings show that a one dollar collection of impact fees

and special assessments led to an increase in capital spending of about

twenty-nine cents. Thus, the use of private financing of public infrastructure

did not substitute for municipal capital spending. Although the impact fees

and special assessment provided a partial tax relief of about twelve cents,

they were not powerful tax relief measures. Our study also shows that

impact fees and special assessments led to an increase in the level of

long-term debt issue, rather than removing the pressure for the use of new

long-term debt. Thus, we conclude that impact fees and special assessments

supplement local capital spending rather than serving as a substitute for

local capital spending.

Although this study improved our understanding of the effects of impact

fees and special assessments on the level of local capital spending and

major revenues sources of local capital spending, our study has several lim-

itations. First, since most local governments do not report these impact fees

and special assessments separately to the Bureau of Census, we were not

able to analyze a separate effect of impact fees and special assessments.

Since a separate source of revenue for impact fees and special assessments

might be found only in the comprehensive annual financial reports of indi-

vidual cities, a future study could build on this study’s data set by using

individual cities’ financial reports. Second, in part due to the difficulty of

collecting necessary socioeconomic variables, this study could not analyze

cities with population below 20,000. To get a thorough understanding on

the topic, future studies could include cities with population below

20,000 in the analysis. Third, in general, the explanatory power of the mod-

els estimated is fairly low in part due to the employment of limited variables

within the national sample. Future studies could analyze single-state sam-

ples using more detailed and relevant variables that could affect the depen-

dent variables. Even so, our study is one of the first national analyses that

empirically examines whether the use of impact fees and special

634 Public Finance Review

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

assessments affects the level of local capital spending and major revenue

sources of local capital spending.

References

Altshuler, A. A., and J. A. Gomez-Ibanez. 1993. Regulation for revenue: The political econ-

omy of land use exactions. Washington, DC: The Brookings Institution.

Baltagi, Badi H. 2008. Econometrics analysis of panel data. 4th ed. West Sussex, UK: John

Wiley & Sons Ltd.

Bierhanzl, Edward J., and Paul B. Downing. 2004. User charges and special districts. In Man-

agement policies in local government finance, ed. J. Richard Aronson and Eli Schwartz,

315-51. 5th ed. Washington, DC: ICMA.

Brueckner, Jan. 1997. Infrastructure financing and urban development: The economics of

impact fees. Journal of Public Economics 63:383-407.

Burge, Gregory S., and Kieth R. Ihlanfeldt. 2006a. Impact fees and single-family home con-

struction. Journal of Urban Economics 60:284-306.

Burge, Gregory S., and Kieth R. Ihlanfeldt. 2006b. The effects of impact fees on multi-family

housing construction. Journal of Regional Science 46:5-23.

Burge, Gregory S., Arthur C. Nelson, and John Matthews. 2007. Effects of proportionate-share

impact fees. Housing Policy Debate 18:679-710.

Chicone, David L., and Norman Walzer. 1986. Factors affecting property tax reliance. Public

Choice 49:17-28.

Clarke, Wes, and Jennifer Evans. 1999. Development impact fees and the acquisition of infra-

structure. Journal of Urban Affairs 21:281-83.

Delaney, C. J., and Mark Smith. 1989a. Impact fees and the price of new housing: An empirical

study. American Real Estate and Urban Economics Association Journal 17:41-54.

Delaney, C. J., and Mark Smith. 1989b. Pricing implications of development exactions on

existing housing stock. Growth and Change 20:1-12.

Downing, Paul B., and James Frank. 1983. Recreational impact fees. National Tax Journal

36:477-90.

Downing, Paul B., and Thomas McCaleb. 1987. The economics of development exactions. In

Development exactions, ed. James Frank and Robert Rhodes, 42-69. Washington, DC:

Planners Press.

Duncan Associates. 2007. National Impact Fee Survey. Prepared by Clancy Mullen, Duncan

Associates, Austin, TX on August 22.

Frank, James E., and Paul B. Downing. 1988. Patterns of impact fee use. In Development

impact fees: Policy rationale, practice, theory, and issues, ed. Arthur C. Nelson, 3-21.

Chicago, IL. Planners Press.

Huffman, Forrest E., Arthur C. Nelson, Marc T. Smith, and Michael A. Stegman. 1988. Who

bears the burden of development impact fees? Journal of the American Planning Associ-

ation 54:49-55.

Ihlanfeldt, Kieht R., and Timothy M. Shaughnessy. 2004. An empirical investigation of the

effects of impact fees on housing and land markets. Regional Science and Urban Econom-

ics 34:639-61.

Jeong, Moon-Gi. 2006. Local choices for development impact fees. Urban Affairs Review

41:338-57.

Jung et al. / Longitudinal Effects of Impact Fees 635

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from

Jeong, Moon-Gi, and Richard Feiock. 2006. Impact fees, growth management, and develop-

ment: A contractual approach to local policy and governance. Urban Affairs Review

41:749-68.

Jung, Changhoon. 2001. Does the local option sales tax reduce the property tax burden? The

Georgia case. Public Budgeting and Finance 21:73-86.

Jung, Changhoon. 2002. The effect of local earmarking on capital spending: The Georgia

counties. State and Local Government Review 34:29-37.

Lawhon, Larry L. 2003. Development impact fee use by local governments. In 2003 municipal

year book, 27-31. Washington, DC: International City/County Management Association

(ICMA).

Leonard, Paul A. 2004. Debt management. In Management policies in local government

finance, ed. J. Richard Aronson and Eli Schwartz, 315-51. 5th ed. Washington, DC: ICMA.

Mayer, Christopher, and C. Tsuriel Somerville. 2000. Land use regulation and new construc-

tion. Regional Science and Urban Economics 30:639-62.

Mullins, Daniel R., and Philip G. Joyce. 1996. Tax and expenditure limitations and state and

local fiscal structure: An empirical assessment. Public Budgeting and Finance 16:75-101.

Nelson, Arthur C., and Mitch Moody. 2003. Paying for prosperity: Impact fees and job growth.

Washington, DC: The Brookings Institution Center on Urban and Metropolitan Policy.

Poterba, James M. 1995. Capital budgets, borrowing rules, and capital spending. Journal of

Public Economics 56:165-87.

Singell, L. D., and J. H. Lillydahl. 1990. Housing impact fees. Land Economics 66:82-92.

Skidmore, Mark, and Michael Peddle. 1998. Do development impact fees reduce the rate of

residential development? Growth and Change 29:383-400.

Snyder, Thomas, Michael Stegman, and David Moreau. 1986. Paying for growth: Using devel-

opment impact fees to finance infrastructure. Washington, DC: Urban Land Institute.

Wooldrige, Jeffrey M. 2002. Econometrics analysis of cross section and panel data.

Cambridge, MA: The MIT Press.

Wooldrige, Jeffrey M. 2006. Introductory econometrics. Mason, OH: Thomson

South-Western.

Yinger, John. 1998. The incidence of development fees and special assessments. National Tax

Journal 51:23-41.

Changhoon Jung, PhD, is an associate professor in the department of political science at

Auburn University. His research interests include public finance and public budgeting.

Chul-Young Roh, PhD, is an associate professor in the department of health services admin-

istration in the College of Public and Allied Health at East Tennessee State University. His

research interests include public finance and health care administration.

Younguck Kang, PhD, is an assistant professor at the KDI School of Public Policy and

Management in Korea. His research interests include public finance and public policy analysis.

For reprints and permissions queries, please visit SAGE’s Web site at http://www.sage-

pub.com/journalsPermissions.nav

636 Public Finance Review

at CA STATE UNIV SAN BERNARDINO on March 8, 2012pfr.sagepub.comDownloaded from