rethinking property tax incentives for business

7/27/2019 Rethinking Property Tax Incentives for Business

1/80

RethinkingProperty Tax Incentives

for Business

Policy Focus Report Lincoln Institute of Land Policy

D a p h n e a . K e n y o n , a D a m h . L a n g L e y, a n D B e t h a n y p. pa q u i n


2/80

Rethinking Property Tax Incentives or BusinessDaphne A. Kenyon, Adam H. Langley, and Bethany P. Paquin

Polic Focus Report Series

The policy ocus report series is published by the Lincoln Institute o Land Policy to address

timely public policy issues relating to land use, land markets, and property taxation. Each report

is designed to bridge the gap between theory and practice by combining research ndings, case

studies, and contributions rom scholars in a variety o academic disciplines and rom proes-

sional practitioners, local ocials, and citizens in diverse communities.

About This Report

State and local governments across the United States use several types o property tax

incentives or business, including property tax abatement programs, rm-specic property tax

incentives, tax increment nancing, enterprise zones, and industrial development bonds com-

bined with property tax exemptions. The escalating use o property tax incentives over the last

50 years has resulted in local governments spending billions o dollars with little evidence o

economic benets.

This report provides an overview o use o property tax incentives or business and oers

several recommendations. State and local governments should consider orgoing these oten

wasteul incentive programs in avor o other, more cost-eective policies, such as customized

job training, labor market intermediaries, and the provision o business services. I ending

property tax incentives is not easible, state governments should consider a range o policy

options, such as placing limits on their use, requiring approval by all aected governments,

improving transparency and accountability, and ending state reimbursement or local property

taxes orgone because o incentives. Local governments can avoid some o the pitalls o busi-

ness property tax incentives by setting objective criteria or the types o projects eligible or

incentives, targeting incentives to mobile rms that export goods or services out o the region,

limiting total spending on incentives, opening the process or decision making on incentives,

and orging regional cooperative agreements.

Copyright 2012 by Lincoln Institute o Land Policy

All rights reserved.

113 Brattle Street

Cambridge, MA 02138-3400 USA

Phone: 617-661-3016 or 800-526-3873

Fax: 617-661-7235 or 800-526-3944

Email: [email protected]

Web: www.lincolninst.edu

ISBN 978-1-55844-233-7

Policy Focus Report/Code PF030


3/80

K e n y o n , L a n g L e y & P a q u i n R e t h i n k i n g P R o P e R t y t a x i n c e n t i v e s 1

. . . . . . . . . . . . . . . .

Contents

2 Eecutie Summar

4 Chapter 1: Oerie o Propert

Ta Incenties or Business

5 Increased Use o Property

Tax Incentives

7 Economic Development Goals

10 Obstacles to Achieving Development

Goals

12 Pitalls with Discretionary Property

Tax Incentives

13 Chapter 2: Propert Taes on Business

13 Why Businesses Pay Property Taxes

15 Policies Aecting the Property Tax

Burden on Business

19 Eective Tax Rates on Business

Property

21 Summary

22 Chapter 3: The Impact o Propert

Taes on Firm Location Decisions

22 The Site Location Process

23 The Eect o Input Cost Dierences24 Economic Theory

26 Empirical Evidence

29 Summary

30 Chapter 4: Tpes o Propert Ta

Incenties or Business

30 Property Tax Abatement Programs

33 Firm-Specic Property Tax Incentives

34 Tax Increment Financing

38 Enterprise Zones

41 Industrial Development Bonds

Combined with Property TaxExemption

44 Widespread Use o Incentives

44 Summary

45 Chapter 5: Tools or Assessing

the Eectieness o Propert

Ta Incenties

46 Transparency

46 Impact o Incentives on Firm

Location Decisions

49 Benet-Cost Framework or

Evaluating Incentives

51 Economic and Fiscal Impact

Analyses

51 Summary

52 Chapter 6: Policies to Reduce

Reliance on Propert Ta Incenties

52 Reduction o Interlocal Competition

53 Tax Reorm

55 Nontax Alternatives

57 Summary

58 Chapter 7: Findings and

Recommendations

59 Alternatives to Incentives

59 State Options or Reorming

Tax Incentives60 Local Options or Reorming

Tax Incentives

62 Reerences

66 Appendi Tables

74 Appendi Notes

75 Acnoledgments

76 About the Authors

76 About the Lincoln Institute

o Land Polic


4/80

2 P o l i c y f o c u s R e P o R t L i n c o L n i n s t i t u t e o f L a n d P o L i c y

. . . . . . . . . . . . . . . . . .

Executive Summary

1 percent o total costs or the U.S. manu-

acturing sector. Second, tax breaks are

sometimes given to businesses that would

have chosen the same location even without

the incentives. When this happens, property

tax incentives merely deplete the tax base

without promoting economic development.

Third, widespread use o incentives within

a metropolitan area reduces their eective-

ness, because when rms can obtain similar

tax breaks in most jurisdictions, incentives

are less likely to aect business location

decisions.

This report reviews ve types o property

tax incentives and examines their character-

istics, costs, and eectiveness.

The use o property tax incentives

or business by local governments

throughout the United States has

escalated over the last 50 years.

While there is little evidence that these

tax incentives are an eective instrument

to promote economic development, they

cost state and local governments $5 to

$10 billion each year in orgone revenue.

Three major obstacles can impede the

success o property tax incentives as an eco-

nomic development tool. First, incentives

are unlikely to have a signicant impact

on a rms protability since property taxes

are a small part o the total costs or most

businessesaveraging much less than


5/80


. . . . . . . . . . . . . . . .

Thebestevidenceonpropertytaxabate-

ment programs indicates they are eective

initially or the rst jurisdictions that use such

incentives, but once they prolierate across ametropolitan area they no longer promote

economic growth.

Evidenceontheimpactof taxincrement

nance on economic activity is more mixed,

but this mechanism may be overused and

nance less benecial projects when one local

government is able to divert revenue rom

another local government without its approv-

al, such as a city diverting a school districts

revenue.

Enterprisezones,whichtypicallyinclude

property tax incentives as part o a larger

incentive package and are usually targeted

to distressed areas, have limited eectiveness.

Verylittleinformationisavailableregarding

either rm-specic property tax incentives or

property tax exemptions in connection with

issuance o industrial development bonds.

Despite a generally poor record in promoting

economic development, incentives can be help-

ul in some cases. When these incentives attract

new businesses to a jurisdiction they can in-

crease income or employment, expand the tax

base,andrevitalizedistressedurbanareas.Ina

best case scenario, attracting a large acility can

increase worker productivity and draw related

rms to the area, creating a positive eedback

loop. This report oers recommendations to

improve the odds o achieving these economic

development goals.

Alternatives to tax incentives should be

considered by policy makers seeking more cost-

effectiveapproaches,suchascustomizedjob

training, labor market intermediaries, and busi-ness support services. State and local govern-

ments also can pursue a policy o broad-based

taxes with low tax rates or adopt split-rate prop-

erty taxation with lower taxes on buildings than

land.

State policy makers are in a good position

to increase the eectiveness o property tax in-

centives since they control how local govern-

ments use them. For example, states can restrict

the use o incentives to certain geographic areas

or certain types o acilities; publish inormation

on the use o property tax incentives; conduct

studies on their eectiveness; and reduce de-

structive local tax competition by not reimburs-

ing local governments or revenue they orgo

when they award property tax incentives.

Local government ofcials can make

wiser use o property tax incentives or business

and avoid such incentives when their costs ex-

ceed their benets. Localities should set clear

criteria or the types o projects eligible or in-

centives; limit tax breaks to mobile acilities that

export goods or services out o the region; in-

volve tax administrators and other stakeholders

in decisions to grant incentives; cooperate on

economic development with other jurisdictions

in the area; and be clear rom the outset that

not all businesses that ask or an incentive will

receive one.


6/80


. . . . . . . . . . . . . . . . . .

C h a p t e r 1

Overview o Property Tax Incentivesor Business

The United States is emerging rom

the worst economic downturn since

the Great Depression. The country

must create jobs to tackle a major

unemployment problem, while also address-

ing signicant scal challenges at all levels

o government. Many local governments

have attempted to deal with these dual

challenges by using property tax incentives

or business, hoping they can spur economic

development and expand their tax base.

But whether tax breaks can achieve

these goals or not is an open question at best.

Some leaders believe that incentives can

be an eective tie-breaker that governments

can use to tip business location decisions

in their avor. For example, the vice president

o marketing or the Chattanooga Area

Chamber o Commerce argues:

Businesses look at a lot o actors in

deciding where to locate. But i they think

they can get the labor, transportation,

and their other needs in more than one

community, then they are going to look

at the incentives to decide where to go.

(Chattanooga Times Free Press2010)

Yet many economists and policy analysts

who have studied tax incentives argue that


7/80


. . . . . . . . . . . . . . . .

F i g u r e 1 . 1

Increasing Use o Propert Ta Incenties

they are oten given to rms that would

have chosen the same location regardless o

tax breaks, in which case they are a costly

tool with no signicant eect on economicdevelopment.

Tax incentives have the potential to

achieve a variety o economic development

goals, but overuse and poorly designed pro-

grams can leave localities with smaller tax

bases and no improvement in their local

economies. The dramatic growth in their

use over the past 30 to 40 years and the long-

term scal challenges acing many state and

local governments suggest that policy makers

need to rethink how they are using incentives.

This report oers recommendations or

howtoincreasetheoddsof realizingdevel-

opment goals with property tax incentives

whileminimizingthecommonpitfalls.

INCREASED USE OF

P ROP ERTy TAx INCENT IvES

Like many other economic development

tools, the use o property tax incentives has

grown dramatically in recent decades, with

the most rapid growth occurring in the

1970s and 1980s (gure 1.1). There are

several reasons or this growth. At the root

is the increased mobility o business over

recent decades. Transportation and com-munications costs have declined dramati-

cally, supply chain management has im-

proved, and previously closed economies

have opened up in Asia and other areas.

As a result, rms are more sensitive to costs

that vary by location, such as labor and taxes,

and increased competition means that busi-

nesses ignoring these cost dierences may

risk bankruptcy (Davidson 2012). With

greater mobility, the potential or incentives

to alter rm location decisions has grown.

Competition to attract a smaller number

o industrial acilities has placed pressure

on state and local government ocials to use

all the tools at their disposal, including prop-

erty tax incentives. Figure 1.2 shows that

over the past three decades, the value o U.S.

manuacturing output has been stagnant,

growing only 4 percent since its 1978 peak

compared to 89 percent growth or the econ-

omy as a whole. Manuacturing employment

has declined 41 percent over this period.

15

81

31

2422

33

38 3735

4042

37

49

0

10

20

30

40

50

Property Tax Abatements Tax Increment Financing Enterprise Zones

NumberofState

sAllowingIncentives

64 79 91 05 10 70 80 90 00 10 81 85 90 10

Note: Property tax abatements are stand-alone programs that are not part o broader economic development programs.

Sources: Appendix Tables A.1, A.2, and A.3; Kerth and Baxandall (2011); U.S. Department o Housing and Urban Development

(1991); Wassmer (2009, 223224).


8/80


. . . . . . . . . . . . . . . . . .

Central cities have borne the negative

eects o these economic changes most

heavily, which has led some states to adopt

enterprisezones,taxincrementnancing,

and other types o geographically targeted

incentives meant to help distressed areas.Among the 100 largest cities in 1960, 44

had lower populations by 2010, which is

particularly striking since over this period

the U.S. population grew 72 percent and

many central cities annexed large amounts

o land (Gibson 1998; U.S. Census Bureau

2012).Incontrast,thepercentageof Amer-

icans living in the suburbs grew steadily

rom 15 percent in 1940 to 45 percent in

1980, and reached 50 percent in 2000

(Hobbs and Stoops 2002).Tax incentives are politically appealing

to local ocials. Because their cost is less

transparent and they are not subject to an-

nual appropriations, tax expenditures can

be more attractive than direct expenditures

on economic development, even i the eect

on tax rates and the ability to und other

services is similar. Policy makers also may

F i g u r e 1 . 2

Actiit in the U.S. Manuacturing Sector, 19502010

Sources: U.S. Bureau o Economic Analysis (2006; 2011).

10

12

14

16

18

20

22

600

800

1,000

1,200

1,400

1,600

1,800

2,000

1950 1960 1970 1980 1990 2000 2010

Employment(millions)

RealValueAdded($2010,

billions) Value Added Employment

argue that they are not really orgoing tax

revenues because without the incentives the

rm would have located elsewhere and thus

paid no taxes to the jurisdiction. However,

this is not always the case (box 1.1). Since

attracting large acilities is a highly visiblesign o success, local ocials may ace

considerable pressure to oer incentives.

A sel-perpetuating cycle can also drive

up the use o tax incentives over time. Their

use in one locality puts pressure on neighbor-

ing jurisdictions to oer incentives as well.

Localities may eel they have no choice but

to oer incentives i tax breaks are actively

used in surrounding jurisdictions; instead

o using incentives to gain an advantage to

attract rms, they are used just to remainon a level playing eld with their neighbors.

Some evidence also indicates that once

a municipality starts using property tax in-

centives it is unlikely to stop oering them

(Sands and Reese 2012). Oering tax breaks

to one rm makes it more likely that other

rms considering locating or expanding in

that jurisdiction will also lobby or incen-


9/80


. . . . . . . . . . . . . . . .

tives. This sel-perpetuating cycle means

that tax incentives can move rom being the

exception to the norm, and will be expected

by all rms rather than serve as a targetedtax break.

ECONOM IC DEvEL OP M ENT

GOAL S

Local governments use property tax incen-

tives to pursue a variety o economic devel-

opment goals. Policy makers must set clear

goals, think hard about the methods by

which tax incentives can help achieve those

goals, and consider obstacles that could

prevent success (table 1.1).

Increase Income or Employment

Business acilities that export goods or ser-

vices to national or international markets

provide an important economic base or a

local government or metropolitan area.

These acilities include manuacturing

plants, corporate headquarters, R&D cen-

ters, warehouses, back-oce support, and

services or people living outside the region,

such as nance and insurance.

Such rms increase an areas aggregate

income in direct and indirect ways. The

B o x 1 . 1

Do Ta Incenties Reall Tip Firm Location Decisions?

Perhaps the greatest dilemma or policy makers considering in-

centives is the limited inormation about the true importance o

property taxes in an individual rms location decision. Firms consider

dozens o actors during site selection, but government ocials rarely

know which actors are most important. They may eel compelled to

oer tax incentives since it is one o the ew location actors they

can infuence directly.

Policy makers may think that tax cuts and incentive oers are decisive,

but this assumption is oten wrong. When businesses lobby or tax

breaks, they have a clear motive to exaggerate the importance o in-

centives, because otherwise they are unlikely to receive any breaks.In act, evidence shows that in some cases businesses negotiate

or tax incentives ater they have already chosen a location (Fisher

2007, 65).

T a B l e 1 . 1

Propert Ta Incenties and Economic Deelopment Goals

Goal Goal Ma be Reached i Incenties: Goal Ma Not be Reached i Incenties:

Increase

Income or

Emploment

Attractfacilitiesthatexportgoodsorservicesout

o the area

Promoteindustryclustersthatincrease

productivity in the area

Havelittleimpactbecausepropertytaxesaccountfor

a small share o total business costs

Createjobsthatlargelygotoin-migrantsorcommuters

Createjobsthatarelow-wageorpart-time

Requiregovernmenttoeffectivelypickwinners

Improe

Fiscal Health

Obtainpartialpropertytaxesfromrmsthatwould

have located elsewhere without tax breaks

Attractsupplierspayingfulltaxesbyprovidingtax

breaks or anchor rms

Obtainothertaxesorfeesfromthermthat

oset orgone property taxes

Aregiventormsthatwouldchoosethesame

location even without tax breaks

Aregiventofacilitiesthatrequirecostlyinfrastructure

investments by the jurisdiction

Extendforalongertimeperiodthanthelifespan

o recipient plants

Promote

Urban

Reitalization

Redirectbusinessinvestmentwithinametroarea

to distressed areas

Offsetlowerbusinesscostsinwealthierareas

Havelittleimpactonrelativetaxburdensdue

to widespread use o tax breaks

Areutilizedaggressivelybywealthyareas

Requireverylargetaxbreaksperjobcreated

to attract investment to distressed areas

rm spends money directly on its payroll,

inputs rom local suppliers, and services

rom local businesses. The indirect eects

occur when these workers and companies

then spend a large share o their incomes

on locally provided goods and services, and

those rms and their workers in turn spend

this money at other local establishments.

This chain o events is oten measured by


10/80


. . . . . . . . . . . . . . . . . .

a multiplier, which is the ratio o the total

increase in income, employment, or output

across the local economy divided by the

initial direct increase (Morgan 2010).Using tax incentives to attract these types

o acilities may increase a localitys per capita

income and employment rate, although the

latter eect is less likely given the high rate

o U.S. labor mobility. Conversely, provid-

ing tax incentives or retail establishments,

housing developments, and other businesses

serving the local population is extremely

unlikely to increase income or employment.

The local population can only support so

many o these businesses, and expansion

by one rm will likely displace sales or

competitors.

Inaddition,attractingalargefacility

may increase the productivity o other rms

in the area and the wages o their workers.

Aninitialclusterof rmsspecializedinone

industry can create a positive eedback loop:

workers with industry-specic skills will

move to the area, which will increase the

number o other similar rms in the area,

and in turn the concentration o rms

supplying inputs. Meanwhile, the sharingo knowledge among workers and rms

will increase productivity and the rate o

innovation, leading to increased wages or

workers in the industry, which will draw more

skilled employees, rms, and suppliers.

Greenstone, Hornbeck, and Moretti

(2010) provide evidence o how attracting

one large acility can generate these types

o productivity spillovers, sometimes known

as agglomeration economies. For 47 large

manuacturing plant openings, the authors

compare economic trends or the winner

county and one or two loser counties that

were runner-ups. Beore the plant openings,

winning and losing counties had similar

trends in productivity and other economic

variables. Five years ater the opening, pro-

ductivity at existing plants in the winning

counties had grown 12 percent more than


11/80


. . . . . . . . . . . . . . . .

in the losing counties, and wage growth was

also signicantly higher. Although this study

did not have data on incentive oers, i they

had played a decisive role in attracting largeplants then these spillover eects on produc-

tivity and wages could justiy the cost o

the incentives.

Improve Fiscal Health

A common goal or individual municipalities

and counties using property tax incentives

is to improve scal health, which occurs i

revenue growth attributable to incentives

exceeds growth in public service costs

related to the business expansion.

If thermtrulywouldnothavechosen

the locality without the incentive, then local

ocials can conclude that some property

tax revenue is better than none. This con-

clusion makes sense i the rm pays partial

property taxes on the acility, or i the rm

will pay ull taxes in the uture once a time-

limitedincentiveexpires.Intheory,ajuris-

dictioncanmaximizerevenuebynegotiating

taxes down to the level at which the rm

just slightly preers that location to alterna-

tive sites, and maintain its scal health by

lowering taxes to the point at which they

equal the cost o providing public services

to the rm (Glaeser 2001).

Oering incentives or one rm could

also boost tax revenues i that acility attracts

other suppliers who would pay ull taxes,

or i it increases the property tax base in

other ways. Greenstone and Moretti (2004)

ound that attracting a large acility increased

property values in winning counties by 6.6

to 10.2 percent relative to runner-up coun-

ties over the course o six years.

Other taxes or ees paid by a rm could

also oset revenue losses rom property tax

incentives.Inparticular,whileincentivizing

retail acilities may be unnecessary i they

are tied to specic sites with high market

exposure, attracting large retail stores can

substantially increase sales tax revenues or

the locality, which is especially important

in states with property tax limits.

However, or counties, municipalities,and towns combined, property taxes raise

about 2.5 times more revenue than sales

taxes.In2007,propertytaxesaccounted

or 36.9 percent o own-source revenues

or these local governments, while sales

taxes accounted or 14.3 percent. Sales taxes

exceeded property taxes in only ten states

(State and Local Government Finance

Data Query System 2012).

Promote Urban Revitalization

Redirecting business investment within a

metropolitan region to areas with high un-

employment or declining populations is a

justiable policy goal. Areas with declining

populationstendtohaveunderutilizedin-

rastructure, so business investment in these

areas is less likely to require costly new in-

rastructure to provide services or a new

acility than areas with growing populations.

Inaddition,thesocialbenetsfromnew

jobs may be greater in these areas, because

a larger proportion o people without jobs

has been involuntarily unemployed or

long periods o time (Bartik 2005); workers

with prolonged periods o unemployment

suer rom an erosion o job skills that

hurts long-term earnings (Bartik 2010);

and inner-city residents may have diculty

obtaining jobs in wealthier suburbs due

to limited knowledge about opportunities,

diculties commuting, or discrimination

(Anderson and Wassmer 2000).

As described in chapter 3, property tax

incentives are much more likely to sway a

rms choice o a specic site within a given

metropolitan area than to alter its broader

choicebetweendifferentregions.If incen-

tives are oered primarily in poorer areas

and center cities, they can help oset the

act that the costs o business may be higher


12/80


. . . . . . . . . . . . . . . . . .

in these areas or a variety o reasons, in-

cluding higher property taxes, lower quality

public services, higher crime or land prices,

andtheneedtoredevelopbrownelds.In-centives can be considered a compensating

dierential to make these areas more com-

petitive with suburban areas that would

otherwise be more protable locations or

many new acilities.

OBSTACLES TO ACHIEv ING

DEvEL OP M ENT GOAL S

Achieving these economic development

goals with property tax incentives depends

on a wide range o actors and is ar rom

guaranteed (box 1.2). Three general obsta-

cles apply to all three goals: property taxes

are a small part o total costs or most rms;

tax breaks are sometimes given to businesses

that would have chosen the same location

even without incentives; and widespread

use o incentives reduces their eectiveness.

Specic obstacles relate to the goal o

increasing income or employment with tax

incentives. First, most new jobs created by

business investment will go to in-migrantsor commuters instead o existing residents,

because people move to areas with strong

economic growth. For example, an analysis

o 18 studies by Bartik (1993) ound that

between 60 and 90 percent o jobs created

by employment programs go to in-migrants

or unintended beneciaries, while a study

byBlanchardandKatz(1992)suggeststhat

in the long run all newly created jobs will

be taken by in-migrants.

Incomegrowthorpovertyreduction

may be more realistic goals than increasing

the employment rate, but these benets

depend on the characteristics o new jobs,

such as the wage level and percent o ull-

time workers. Relying on selective incen-

tives to improve the economy requires local

governments to pick winners by strategic-

ally oering incentives and identiying key

rms and local sectors that can sustain

competitiveness.

Achieving the goal o improved scal

health by oering tax incentives also depends

on several actors. Most important is the cost

o new inrastructure and expanded public

services, which depends on the current use

o existing inrastructure. Because o these

costs, projects that require new inrastructure

are unlikely to improve scal health in the

shortrun(AltshulerandGmez-Ibez

1993)

Another issue is that expecting a rm to

pay ull taxes in the uture once an incentive

has expired is oten unrealistic. Based on

several studies, Fisher (2007) has estimated

that the median manuacturing plant is

open or approximately 8 to10 years. Since

the duration or property tax abatements

exceeds 10 years in about two-thirds o pro-

grams (Dalehite, Mikesell, and Zorn 2005),

a majority o acilities may have closed


13/80

K e n y o n , L a n g L e y & P a q u i n R e t h i n k i n g P R o P e R t y ta x i n c e n t i v e s 11

. . . . . . . . . . . . . . . .

B o x 1 . 2

A Cautionar Tale in Michigan

In 2009, the State o Michigan oered

over 35 business tax incentive pro-

grams (Anderson, Rosaen, and Doe 2009).

The most expansive o these is the In-

dustrial Facilities Property Tax Abatement

program (Act 198). Crated in 1974, Act

198 provides geographically targeted

property tax abatements or the creation,

expansion, renovation, or addition o in-

dustrial property (Sands and Reese

2012; CRC 2007). Practically any local

government may establish an industrial

development or plant rehabilitation dis-

trict. Once a district is established, any

qualiying business wishing to develop

within the district can apply or an ex-

emption certicate subject to local and

state approval and conditional upon job

retention and creation. Instead o pay-

ing property taxes, certied businesses pay a substitute

tax equal to 50 percent o the property tax or new acilities

and equal to the property tax on the unimproved value o

renovations or rehabilitations (Mikesell and Dalehite 2002;

Signicant Features o the Proper ty Tax 2012).

Sands and Reese (2012) report that between 1974 and

2005 the program abated $77.4 billion in real and person-

al property, with an average o 600 exemption certicates is-

sued each year since 1980. The cost to local governments

in lost revenue between 1990 and 2005 was roughly $84

per person per year. In 2008, industrial property abated

by this program accounted or 20.5 percent o the total

industrial tax base (Anderson, Bolema, and Rosaen 2010).

Despite their widespread use, the impact o the Act 198

abatements is unclear. Over the 19902005 period, busi-

nesses receiving abatements reported they would create

234,000 new manuacturing jobs and retain 728,000

manuacturing jobs that otherwise would have been lost.

Yet the number o jobs reported is not the same as the

number o jobs actually attributable to the abatements,

because the promised jobs do not always materialize and

many that do would have been created even without the

abatements. In act, in some industries the number o jobs

reportedly created or retained through abatements actually

exceeds the total number o all jobs in those industries.

More generally, Michigan lost a slightly higher percentage

o manuacturing jobs than the country as a whole over

the time period. Although manuacturing job losses may

have been even greater in the absence o abatements, the

abatements were not eective in preventing substantial

job losses (Sands and Reese 2012).

Evidence shows the abatements have not eectively tar-

geted incentives to distressed areas or central cities. Among

communities that awarded abatements between 1998

and 2000, distressed areas were no more likely to award

them than fourishing communities, but spent more per

job retained or created than wealthier areas. Furthermore,

suburbs award abatements at a higher rate per capita than

central cities. The suburbs report more jobs per capita as

a result o incentives and had higher investment per capita.

Abatements may promote sprawl to the extent that new

investment spurred by the abatements is more likely to

occur outside o central cities (Reese and Sands 2006).


14/80


. . . . . . . . . . . . . . . . . .

beore they ever paid the ull tax rate. For

these reasons, some studies have ound that

greater reliance on property tax incentives

increases scal stress or local governments(Mullen 1990).

Finally,promotingurbanrevitalization

with property tax incentives depends on

theirgreaterutilizationindistressedareas

than wealthier communities; i both types

o areas use incentives aggressively, then

relative tax burdens may change little. How-

ever, in practice, economic development in-

centives do not appear to be notably more

common in low-income areas (Peters and

Fisher 2004). There is also evidence that

tax incentives are more cost eective in areas

with high incomes and low unemployment,

and thus their use could actually widen

economic disparities between high- and

low-income areas (Goss and Phillips 2001;

Sands and Reese 2012). While the social

benets o creating jobs with tax incentives

may be greater in areas with high unemploy-

ment, the costs could be even greater i

it takes substantially larger tax breaks to

induce business investment in these areas.

P ITFALLS wITH

D ISCRET IONARy P ROP ERTy

TAx INCENT IvES

Discretionary tax incentives, which are dis-

tinct rom as-o-right incentives given to all

rms meeting certain criteria, have other

pitalls. Selective use o incentives raises

majorconcernsabouthorizontalequityand

the distribution o taxes, because granting

tax breaks to some mobile businesses likely

means that long-standing local businesses

or homeowners will pay more. This type

o system is likely to be viewed as unairby many taxpayers.

Decisions to grant discretionary tax

incentives are sometimes not transparent

or are made in ad hoc ways without clear

economic justication. This process may

be unduly infuenced by political consider-

ations, with incentives granted to well-

connected rms or campaign contributors.

For example, Felix and Hines (2010) ound

that communities in states with more cor-

rupt political cultures were more likely

to oer incentives.

A related concern is that politicians may

grant incentives regardless o the economic

rationale. Politicians can grant incentives

and claim that they played an instrumental

role in attracting a new acility to the com-

munity, even i a rm may have located

there without incentives. Wolman and

Spitzley(1996)ndevidenceof thistype

o credit-claiming among elected ocials.

Conversely, i politicians decide not to

oer incentives, they could be blamed i

the rm chooses to locate elsewhere. A nal

consideration is that negotiation over tax

incentives signicantly increases the cost

o property tax administration or the local

government.Itisalsoeconomicallyinef-

cient or rms to spend time and money

lobbying or tax breaks instead o ocus-

ing on improving their business.


15/80


. . . . . . . . . . . . . . . .

C h a p t e r 2

Property Taxes on Business

Business property includes nonresi-

dential, income-producing prop-

erty such as commercial, industrial,

arm, mineral, railroad, or public

utility properties (Cornia 1995). This report

ocuses primarily on commercial and indus-

trial property.

wHy BUSINESSES PAy

PROPERTy TAxES

Inordertoputpropertytaxincentivesinthe

proper context, it is important to consider

the reasons or requiring businesses to pay

property taxes.

To und services received. State and

local governments provide a wide array o

services that benet business activity, in-

cluding a small proportion that directly and

solely benet businesses, such as economic

development support. Other types o state

and local government expenditures, such

as on the court system, transportation, and

public saety, provide critical benets or

bothbusinessesandhouseholds.Education

is the single largest expenditure o state and

local government. Although education pro-

vides direct benets to individuals, it also

benets businesses by increasing the pro-

ductivity o their employees.

Oakland and Testa (1996) examine several

rationales or state and local taxation o busi-

ness, concluding that the primary basis or


16/80


. . . . . . . . . . . . . . . . . .

taxing businesses is to recover the cost o

government services provided to them. The

authors urther argue that taxing businesses

in accordance with benets provided isboth air and ecient.

To generate revenue or local gov-

ernments. Although popular discussion o

property taxes tends to ocus on those paid

by homeowners, the assessed value o busi-

ness property is an important part o the

taxbase.In1986,themostrecentyearthat

the U.S. Census collected data on assessed

property values, 39 percent o the property

tax base could be attributed to businesses.

This included commercial properties (16

percent), industrial properties (6 percent),

arms (7 percent), and personal property

(10 percent). The latter can be classied as

business property since most states no longer

tax household personal property. Residential

property accounted or 55 percent o the

tax base, split between single-amily houses

(48 percent) and multiamily properties

(7percent).Vacantlotsaccountedforthe

remaining 6 percent (U.S. Department

o Commerce 1989).

More recent data can be obtained

at the state level, but not all states report

assessed values by property type, and the

states that do report may not divide the

property tax base into the same categories.

Thirty-one states report some division o

their property tax base by property type

(table 2.1). For these states on average,

nearly 60 percent o the property tax base

was residential, 22 percent was commer-

cial and/or industrial, and 19 percent was

categorizedasother,whichincluded

various types, such as personal property

and vacant land.

Revenue rom business property taxes

also constitutes a substantial proportion o

all property tax revenue collected. Accord-

ing to Phillips et al. (2011), in FY2009 busi-

nesses contributed $247 billion in property

T a B l e 2 . 1

Propert Ta Base b Propert Tpe in 31 States, 2009

State Residential

Commercial

and/or Industrial OtherU.S. Aerage 59.8% 21.6% 18.6%

Alaska 59.7 22.4 17.9

Colorado 46.2 25.7 28.1

Delaware 71.0 29.0 0.0

District o Columbia 58.4 40.9 0.7

Florida 74.2 17.1 8.8

Hawaii 68.6 24.8 6.7

Idaho 69.7 22.7 7.5

Illinois 65.0 32.5 2.5

Indiana 49.3 29.0 21.7Iowa 44.5 30.2 25.3

Kansas 51.9 25.8 22.3

Kentucky 65.8 24.3 9.9

Maryland 80.2 18.0 1.8

Massachusetts 83.0 14.5 2.5

Michigan 69.4 20.1 10.5

Minnesota 66.9 13.3 19.8

Missouri 53.7 21.4 24.9

Montana 47.1 13.6 39.3

New Hampshire 79.6 16.3 4.1

New Jersey 62.5 15.2 22.3

North Carolina 65.0 15.8 19.2

North Dakota 43.3 23.3 33.5

Ohio 69.3 20.8 10.0

Oregon 52.5 19.0 28.5

South Dakota 39.2 24.1 36.6

Tennessee 55.5 27.1 17.4

Texas 52.3 20.3 27.4

Utah 47.2 19.4 33.4

Vermont 60.9 16.5 22.6

Washington 75.4 16.6 8.0

Wisconsin 72.1 20.3 7.6

Wyoming 15.2 10.5 74.3

Notes: The other 19 states do not repor t divisions o their tax base into classes or residential

andcommercialand/orindustrialproperties.Statesdenitionsofotherpropertyvarywidely.

Source: Signicant Features o the Property Tax (2012).


17/80


. . . . . . . . . . . . . . . .

tax revenue to state and local governments,

constituting 58 percent o all property taxes

raised and 40 percent o all state and local

taxes paid by business (gure 2.1). These

estimates include multiamily housing,

although many other researchers would

not include it as business property. Business

property taxes have been quite stable overthe past two decades, although they did

jumpsignicantlyin2009and2010.It

is likely that business properties will help

shore up total property tax revenues in

coming years, as the dramatic all in hous-

ing values weighs down residential tax

payments.

To add progressivity to the state-

local tax system. For those concerned

with state and local government use o

regressive taxes, such as reliance on the

general sales tax, levying property taxes on

businesses can be a way to add a progres-

sive element to the total state-local tax sys-

tem. The property tax, particularly the part

o the property tax levied on businesses, is

oten conceived as a tax on capital. Owner-

ship o capital is proportionately greater

or higher-income households, so any tax

on capital places a higher tax burden

on high-income households than on

low- and moderate-income households.

P OL IC IES AFFECT ING THE

PROPERTy TAx BURDEN ON

BUS INESS

Property tax incentives or business canonly be understood ully within the context

o other major policies aecting the prop-

erty tax burden on business.

State Constitutions

Although they vary enormously and have

evolved over time, state constitutions together

with case law set the ramework that guides

legislative action regarding business property

taxes. The most important constitutional

provisions are the uniormity clauses includ-

ed in 39 state constitutions, which require

property taxation at a uniorm rate within

a jurisdiction, although they are subject to

important qualications that vary by state

(Coe 2009).

One example o a uniormity clause is

Alabamas constitutional requirement that

all taxable property shall be orever taxed

F i g u r e 2 . 1

Propert Taes on Business, Fiscal years 19902010

Sources: Cline et al. (2011); State & Local Government Finance Data Query System (2012); U.S. Bureau o Economic Analysis (2011).

20

30

40

50

60

1.50

1.75

2.00

2.25

2.50

1990 1995 2000 2005 2010

PercentofTotalTaxes

(lines)

PercentofPrivate

SectorOutput(bars)

Percent of Total Property Taxes

Percent of Total State/Local Business Taxes

Percent of

Private Sector

Ouput


18/80


. . . . . . . . . . . . . . . . . .

atthesamerate(ArtXI,sec.217(b)).

Arizonasconstitutionalrequirementthat

all taxes shall be uniorm upon the same

class o property provides an importantclue to the practical application o most

suchclauses(Art.IX,sec.1).

Although these clauses ostensibly require

all property to be taxed at a uniorm rate,

in reality most allow dierential taxation

between dierent classes o property at the

same time that they require uniorm taxa-

tion within a given class. But even that re-

quirement is subject to the exceptions that

arise rom property tax exemptions, which

areallowedinmoststates.Itisimportantto

realizethatcourtsareinclinedtogivestate

legislatures extensive leeway in their power

to tax, as long as it does not violate any

explicit provision o the state constitution

(Coe 2009, 131).

State constitutions also commonly

address the issue o exemptions, but they

range rom strictly limiting the state legis-

latures discretion in granting exemptions

(e.g.,Arizona),toallowingthelegislature

broadlatitude,asinIdaho,whoseconstitu-

tion states, the legislature may allow such

exemptions rom taxation rom time to time

asseemnecessaryandjust(Art.VII,sec.5)

(Coe 2009, 150151). The Florida consti-

tution addresses the issue o property tax

exemptions or the purposes o economic

development: Any county or municipality

may . . . grant community and economic

development ad valorem tax exemptions to

new businesses and expansions o existing

businesses.(Art.VII,sec.3(c)).

A recent legal case challenging tax

incentives or business went all the way

to the U.S. Supreme Court (box 2.1).

Classifcation or Split Roll

Classication or split roll taxation is a policy,

either constitutional or statutory, that applies

dierent eective tax rates to dierent classes

B o x 2 . 1

Cuno Supreme Court Case

In 1998 the City o Toledo, Ohio and two local school

districts oered DaimlerChrysler, Inc., a $280 million

tax incentive package to expand operations within the

city. The company estimated that the $1.2 billion devel-

opment o a new Jeep manuacturing acility would cre-

ate thousands o new jobs. The tax incentive package

included a 10-year, 100 percent property tax exemption

and a state ranchise tax credit (DaimlerChrysler Corp

v. Cuno [2006]).

Led by Toledo resident Charlotte Cuno, a group o nine

Ohio taxpayers and some area businesses led a law-suit against DaimlerChrysler, the State o Ohio, and the

City o Toledo charging that the tax incentive package

violated the U.S. Commerce Clause and the Ohio Equal

Protection Clause (Carty 2006). The case was led in

state court, but DaimlerChrysler moved the case to

ederal court where the U.S. District Court ruled that

the incentives violated neither the U.S. nor Ohio clause

and dismissed the case (Lunder 2005).

On appeal, in 2005 the U.S. Court o Appeals or the

Sixth Circuit armed the U.S. District Courts ruling

upholding the property tax exemption, but reversed its

ruling on the ranchise tax credit, maintaining that the

credit ran aoul o the U.S. Commerce Clause. In March

2006, the U.S. Supreme Court reviewed the lower court

decisions and dismissed the case, ruling that the plain-

tis had no standing to challenge the credit or the

state ranchise tax.

Summing up the unanimous ruling, Supreme Court Chie

JusticeJohnRobertswrote,Indeedbecausestate

budgets requently contain an array o tax and spending

provisions, any number o which may be challenged ona variety o bases, aording state taxpayers standing to

press such challenges simply because their tax burden

gives them an interest in the state treasury would inter-

pose the ederal courts as virtually continuing monitors

o the wisdom and soundness o state scal adminis-

tration, contrary to the more modest role Article III

envisionsforthefederalcourts(DaimlerChrysler

Corp v. Cuno [2006]).


19/80


. . . . . . . . . . . . . . . .

of property.Effectivetaxratesarecomputed

by dividing total tax liability by total prop-

erty value. Comparison o eective instead

o statutory tax rates is particularly impor-

tant when comparing one jurisdiction that

assesses property at market value with an-

other jurisdiction that assesses property at

some raction o market value. For example,

a jurisdiction can levy an eective property

tax rate o 1 percent either by assessing

property at 100 percent o market value and

employing a statutory tax rate o 1 percent,

or by assessing property at 50 percent o mar-

ket value and employing a 2 percent tax rate.

States that employ classication typically

use it to apply higher tax rates to commer-

cial, industrial, and other business property

than to residential property. Classication

can be accomplished in two ways: statutory

tax rates can vary by class, or the ratio o

assessed value to market value can vary by

class. As an example o the latter, Alabama

applies a uniorm statutory tax rate to all

types o property, but assesses utility prop-

erty at 30 percent o market value; commer-

cial and industrial property at 20 percent;

and residential property at 10 percent (Sig-

nicant Features o the Property Tax 2012).

Twenty-six states plus the District o

Columbia employ some orm o property

tax classication and Caliornia policy mak-

ers have been considering adopting a split-

roll property tax in order to increase property

taxes on businesses relative to residential

property (Lee and Wheaton 2010; Sherin

2009). Many state constitutions address the

issue o classication. Some, like Floridas,

prohibit classication, but others give the

state great leeway in creating a classication

system. Some place limits on classication,


20/80


. . . . . . . . . . . . . . . . . .

such as the Massachusetts constitution,

which limits the number o permissible

classes to our (Coe 2009).

Certain states, including Connecticut,Illinois,Massachusetts,NewYorkand

RhodeIsland,allowlocalgovernments

some discretion in adopting or adjusting

property tax classication. Others, such as

Colorado, have adopted a system termed

dynamic classication in which eective

tax rates or each property class are changed

over time in order to maintain a specic re-

lationship between the share o the property

tax paid by residential properties and other

properties (Bell and Brunori 2011).

Assessment Practices

Although classication systems are generally

used to impose greater eective tax rates on

business than residential properties, a state

can accomplish the same thing as a de acto

rather than a de jure policy. Some states

even had long-standing policies o assessing

business properties at a greater proportion

o market value than residential properties

beore enacting legislation establishing clas-sication systems to codiy such practice.

Another way in which business properties

can be systematically taxed dierently rom

residential property is by using a dierent

appraisal methodology. O the three stan-

dard methodssales, income, and cost

the sales method is most oten used or resi-

dential properties and least oten or business

properties. Although each methodology

should in theory lead to the same valuation,

in practice they may dier. One concern is

that the cost method might systematically

undervalue properties, which would tend

to lead assessors who employ that method

to undervalue business properties relative

to residential properties (Cornia 1995).

Personal Property Taxes

Inconsideringpropertytaxesonbusiness,

it is important to include personal property

as well as real property, which consists o

land, improvements to land, and buildings.

Personal property includes machinery and

equipment, inventories, and xtures such as

urniture or oce equipment, and is typically

taxed only when owned by a business. Per-

sonalpropertyischaracterizedbyitsmobil-

ity, whereas real property is immovable (Almy,

Dornfest,andKenyon2008).Inpartbecause

o this greater relative mobility, the case or

taxing business personal property is weaker

than that or taxing business real property.

For example, a business could easily move

inventories rom a high-tax to a low-tax

jurisdictioninordertominimizetaxliability.

Over time, personal property has become

a smaller part o the U.S. property tax base,

as most household personal property and

later some business personal property was

removed rom the tax base. Personal prop-

erty as a share o the local property tax


21/80


. . . . . . . . . . . . . . . .

F i g u r e 2 . 2

Eectie Propert Ta Rates or Urban Commercial Propert ($1 million alue), 2010

base was 17 percent in 1956, 13 percent

in 1971, and 10 percent in 1986 (Mikesell1995).In1961,fourstatesexemptedper-

sonal property rom taxation; by 2011,

12 states had exempted personal property

(Mikesell1995;Thompson/ReutersRIA

2012).Inthepast12years,8statesreduced

their reliance on personal property taxes,

including raising exemption levels and

eliminating personal property taxes on

inventories (Drenkard 2012).

Ohio and Michigan recently reduced

taxation o business personal property aspart o their tax reorm initiatives. Ohio

adopted a new commercial activity tax,

exempted new tangible personal property

rom taxation, and enacted a ve-year

phase-out o taxes on existing personal

property. Michigan replaced its Single

Business Tax with a new business tax struc-

ture at the same time that it signicantly

reduced personal property taxes or both

commercial and industrial taxpayers(NeubigandCline2008).

EFFECT IvE TAx RATES ON

BUS INESS P ROP ERTy

The most comprehensive measure o

eective tax rates is the one calculated or

the largest city in each state by the Minnesota

Taxpayers Association (MTA), which esti-

mates eective tax rates or commercial,

industrial, and homestead properties (Min-

nesota Taxpayers Association 2011). TheMTA takes a number o actors into account,

such as dierences in assessment practices;

exemptions, credits, or reunds that apply

to a majority o taxpayers; tax rates or all

state and local governments that serve a

city; and tax classication when it is used.

Figure 2.2 shows that eective property

tax rates or urban commercial properties

2.50% to 4.01%

1.96% to 2.49%

1.40% to 1.95%

0.65% to 1.39%

Rate for Largest

City in Each State

Note: In most cases property

tax structures are uniorm acrossstates, with the exception o

Illinois and New York. This map

illustrates the eective tax rate

or Aurora, Illinois (2.39%). The

rate or Chicago is 1.79%

Source: Minnesota Taxpayers

Association (2011, 21).


22/80


. . . . . . . . . . . . . . . . . .

2.00% to 3.15%

1.50% to 1.99%

1.20% to 1.49%

0.44% to 1.19%

Rate for Largest

City in Each State

F i g u r e 2 . 3

Eectie Propert Ta Rates or Urban Industrial Propert (50% Personal Propert, $1 million alue), 2010

with a $1 million market value range be-

tween 0.7 percent in Cheyenne, Wyoming,

and 4 percent or Detroit, Michigan. Theserates are highest in the Midwest and Middle

Atlantic states and lowest in the West. They

vary or many reasons, including reliance

on other local revenue sources (e.g., sales

tax and user ees), property values, and the

level o local government spending.

Because some states tax personal prop-

erty and others do not, estimates o eective

tax rates or industrial property depend on

the proportion o the total property value

that is personal property. Figure 2.3 shows

eective tax rates or urban industrial prop-

erty valued at $1 million, assuming that hal

of thevalueispersonalproperty.Effective

tax rates or industrial property are some-

what lower than or commercial property,

ranging rom 0.4 percent in Wilmington,

Delaware, to 3.2 percent in Columbia,

South Carolina. These rates are highest

in the Midwest and South and lowest

in the West.

Inthemajorityof citiestheeffective tax rates or commercial properties exceed

those or homesteads. Figure 2.4 shows the

ratio o commercial to homestead eective

property tax rates or the largest city in each

of 25states.Effectivetaxratesoncommer-

cial properties exceed rates on homestead

properties in 20 o them. A ew cities, such

asBaltimore,Maryland,andVirginiaBeach,

Virginia,haveahighertaxonhomestead

properties than on commercial properties.

The MTA has tracked the ratio o eec-

tive tax rates o commercial versus home-

stead property since 1998, when that ratio

was 1.76, indicating that on average across

the country commercial properties were

taxed about 76 percent higher than home-

stead properties. That ratio declined until

2002, then rose through 2008, and has

declinedslightlysincethen.In2010the

Note: In most cases property

tax structures are uniormacross states with the exception

o Illinois and New York. This

map illustrates the eective tax

rate or Aurora, Illinois (1.44%).

The rate or Chicago is 1.18%.

Source: Minnesota Taxpayers

Association (2011, 23).


23/80


. . . . . . . . . . . . . . . .

F i g u r e 2 . 4

Ratio o Commercial to Homestead Eectie PropertTa Rates, 2010

Notes: Figure shows the largest city in the 25 most populous states, with the

exception o Illinois and New York, which show the second largest city. The U.S.

average is or the largest city in each state, with New York City excluded.

Source: Minnesota Taxpayers Association (2011, 14).

ratio was 1.72 (Minnesota Taxpayers

Association 2011).

Whether or not eective tax rates or

industrial properties exceed those or home-steads depends on the split o industrial

property between personal and real prop-

erty.In2010,thenationwideaverageof

eective property tax rates on median value

homes across the United States was 1.34

percent. This ell short o the 1.43 percent

eective tax rate or urban industrial prop-

erty valued at $1 million, assuming that 50

percent o the total property value was per-

sonal property. However, it would exceed

the 1.3 percent rate i personal property

was assumed to account or 60 percent o

total property value (Minnesota Taxpayers

Association 2011).

Itisimportanttonotethateffectivetax

rates measure the initial incidence o prop-

erty taxes, but other studies explore nal

incidence, a more complicated concept that

takes into account the act that the ultimate

burden o taxation always alls on persons.

That is, depending upon actors such as

whether a business serves a local or national

market, the nal incidence o business taxes

will all on business owners, workers, or

consumers.

SUM M ARy

Requiring businesses to pay property taxes

is based on three rationales: businesses ben-

et rom local government services; business

property tax payments are an important

revenue source or local governments; and

business property tax payments add a pro-

gressive element to the state-local tax sys-

tem. This chapter surveyed policies other

than property tax incentives or business

that serve to either increase the property

tax burden on business (e.g., classication

or split-roll systems) or decrease the burden

(e.g., phasing out personal property taxes).

Effectivetaxratesoncommercialand

industrial property vary enormously across

the United States or a variety o reasons.

Inthelargestcityinmoststatestheeffective

tax rates on commercial property exceed

those or homeowners, but in some states

the reverse is true.

0 1 2

VA: Virginia Beach

MD: Baltimore

NC: Charlotte

NJ: Newark

WA: Seattle

CA: Los Angeles

WI: Milwaukee

IL: Aurora

TX: Houston

MI: Detroit

OH: Columbus

GA: Atlanta

FL: Jacksonville

PA: Philadelphia

TN: MemphisU.S. Average

NY: Buffalo

MO: Kansas City

AL: Birmingham

LA: New Orleans

MN: Minneapolis

AZ: Phoenix

IN: Indianapolis

SC: Columbia

CO: Denver

MA: Boston

3

Higher tax rates on

commercial properties

Equal tax rates

Higher tax rates on

homestead properties


24/80


. . . . . . . . . . . . . . . . . .

C h a p t e r 3

The Impact o Property Taxeson Firm Location Decisions

The site location process used by

many businesses, as well as eco-

nomic theory and empirical studies,

suggests that the impact o proper-

ty taxes on rm location decisions depends

on the type o acility and the geographic

area under consideration.

THE S ITE L OCAT ION P ROCESS

With over 36,000 jurisdictions in the United

States and a much larger number o poten-

tial sites, rms could not possibly evaluate

all sites across the many location criteria

that are typically considered in such deci-

sions.Instead,thesitelocationprocessnor-

mally occurs in several stages during which

rms systematically narrow the geographic

area under consideration and compare

with increasing detail the competing

locations. The importance o property

tax dierentials in rm location decisions

varies with the stage o site selection.

A two-stage process is one way to think

about site selection, in which a rm rst

chooses a metropolitan area and then a

specic site within that region. Property

taxes are relatively unimportant in choosing

a metropolitan area since tax dierences

have a much smaller impact on prots than

dierences in costs or labor, transportation,

energy, and rent or occupancy. However,

since eective property tax rates can vary

signicantly within a metropolitan area,

dierences in property taxes can be a

deciding actor when selecting a single site

within an area. At this stage, state taxes


25/80


. . . . . . . . . . . . . . . .

and regulations as well as energy costs are

oten constant; labor cost dierences are

small because o the ease o intrametropoli-

tan commuting; and there is little variationin proximity to suppliers or consumers.

Ady (1997) describes a more detailed

three-stage process or acility location,

developed by Fantus Consulting, which

isnowusedwidelybysiteselectors.Inthe

rst stage, the search is narrowed to a broad

region, several states, or several counties

based primarily on wage dierentials, trans-

portation variables (or manuacturing), and

project-specic essentials such as access to

port acilities, right-to-work laws, or prox-

imity to an engineering school. Taxes will

be considered, but only at a high level to

eliminate clearly uncompetitive states.

Inthesecondstage,3to5communities

will be chosen out o a list o as ew as 15

to 20 or as many as 50 to 100. The ocus is

on modeling operating costs or the specic

project in each community. According to a

database o rms using Deloitte & Touche/

Fantus Consulting or site selection during

19921997, Ady (1997) reports that total

operating costs or a typical manuacturing

acility can be estimated with the ollowing

weights or ve categories o input costs:

labor (36 percent), transportation (35), utili-

ties (17), occupancy (8), and taxes (4). Again,

taxes are relatively unimportant at this stage

because they account or a small part o

geographically variable costs.

But in the third stage, when choosing a

specic site, rms examine actual properties

that can meet their needs, and then all taxes

and incentives are compared in detail and

the quality o public services is measured

careully. As Ady (1997, 80) says, The only

case where taxes alone could sway a loca-

tion decision is a company relocation in a

relatively autonomous geographic area,

such as a city or metropolitan area.

THE EFFECT OF INP UT

COST D IFFERENCES

The importance o dierences in each cost

actor will depend on each actors shareo total costs or the rm and the extent o

variation across states, regions, or jurisdic-

tions. Large variations will have little eect

on rm location decisions i a cost actor

accounts or a small share o total costs, while

actors accounting or a large share will be

unimportant i there is little variation across

competing regions.

When a manuacturing rm chooses

a region in which to locate its acility, its

decision is typically driven by proximity tosuppliers and consumers (and the transpor-

tation costs to reach them) and the wages,

skills, and availability o local workers. That

is because three-quarters o costs or the

average manuacturing rm are inputs pur-

chased rom suppliers, with labor account-

ingformostof theremainingcosts.Infact,

gure 3.1 shows that the manuacturing

F i g u r e 3 . 1

Input Costs as a Share o Total Costs or the ManuacturingSector, 20042009

Note: See Appendix Notes or an explanation o the calculations.

Sources: Phillips et al. (2011); U.S. Bureau o Economic Analysis (2011).

21.8%

2.7%

0.8%0.3%

0%

5%

10%

15%

20%

25%

Labor Energy State/LocalTaxes

Property

Taxes


26/80


. . . . . . . . . . . . . . . . . .

F i g u r e 3 . 2

Impact o Relocation to Dierent States on Total Costs or an Aerage Manuacturing Facilit

Note: See Appendix Notes or an explanation o the calculations.

Sources: Moodys Analytics, Inc. (2011); Phillips et al. (2011); U.S. Bureau o Economic Analysis (2011); U.S. Energy Inormation Administration (2011).

sector spends nearly 75 times more on

labor (21.8 percent) than on property taxes

(0.3 percent).

Eventhoughtaxesvarymoreacrossstates than do labor costs, dierences in

labor costs are still much more likely to

drive rm relocations. Figure 3.2 estimates

the eect on an average manuacturing

acility o relocating rom a high-cost state

to a low-cost state, using actual data on

state input costs and the share o total costs

or each input. Moving rom the state with

the th highest state-local taxes to the state

with the th lowest will reduce business

costs by 0.4 percent on average whereas

moving rom the state with the th high-

est labor costs to the th lowest will save

almost 9 times as much (3.1 percent).

The importance o taxes is much greater

when a rm chooses a specic site within a

metropolitan area. At this stage, dierences

in the cost o labor, energy, state taxes,

and transportation are normally small, but

property taxes oten vary more across indi-

vidual jurisdictions within a given region

than they do across states. For example,

among 103 Massachusetts municipalities

in the Boston metropolitan area that havenotzonedoutindustry,effectiveproperty

tax rates on industrial properties ranged

rom 1.13 percent in the municipality with

the tenth lowest rates to 2.82 percent in the

municipality with the tenth highest rates.

This means that a rms total operating

costs could be reduced by 0.5 percent by

locating in the low-tax municipality instead

of thehigh-taxone(seeAppendixNotes

or calculations).

ECONOM IC THEORy

Dierences in eective property tax rates

on new investment will aect a jurisdictions

ability to attract mobile capital investment

in direct and indirect ways. Above-average

property taxes on business will directly

reduce business investment, because higher

property taxes decrease the rate o return

oninvestment.AccordingtotheNewView

0.1%

0.2%

0.7%

2.1%

0.2%

3.1%

0.0% 0.5% 1.0% 1.5% 2.0% 2.5% 3.0% 3.5%

Property Taxes

State/Local Taxes

Energy

Labor

1.4%

0.4% Change in Total Costs from Moving:From: State with 5th highest costs for input

To: State with 5th lowest costs for input

From: State with 12th highest costs for input

To: State with 12th lowest costs for input


27/80


. . . . . . . . . . . . . . . .

of thepropertytaxrstputforthbyMiesz-

kowski (1972), the average business property

tax rate constitutes a prots tax that reduces

the rate o return on business property na-tionally. Property taxes above the national

average will reduce business activity, land

prices, wages, and the employment rate.

However, higher property taxes are oten

associated with higher-quality public services

forbusinessandwillbecapitalizedinto

lower land values. These indirect eects will

tend to increase business investment. Higher-

quality police and re protection, highways,

inrastructure, utilities, and education all

aect rm location decisions (Ady 1997).

Public services aect rm costs and produc-

tivity, such as the impact o police protection

on insurance rates and the impact o edu-

cation on labor productivity.

If propertytaxdifferentialswerecom-

pletely oset by dierences in the quality o

public services, then property taxes would

be benet taxes and have no eect on

rm location decisions. Oates and Schwab

(1991) have argued that under perect com-

petition all local government taxes would

be benet taxes, because jurisdictions would

bid against each other to attract mobile

businesses up to the point where the cost

o providing public services to a rm would

exactly equal the amount it pays in taxes.

Inthiscase,thelevelof publicservices

provided would be economically ecient,

although there would be no scope or

redistribution at the local level.

Some research has cast doubt on the

property tax being a benet tax or business.

According to Oakland and Testa (1998), in

1995 businesses paid twice as much in state

and local taxes as the cost o public services

they received. However, these estimates de-

pend on assumptions about how the benets

o public goods are shared between house-

holds and business. For example, i 25 per-

cent o public education spending is count-

ed as a benet or business, then the esti-

mated ratio o taxes-to-benets drops

rom 2.06 to 1.31.

Inaddition,propertytaxesarecapital-

izedintolandvaluesthatis,forotherwise

identical properties with similar location

advantages and public services, the one with

higher property taxes will have a lower land

value,whichequalizestotalexpensesover

the lie o the property. However, Yinger

et al. (1988) ound that property tax dier-

entialsarenotfullycapitalizedintoprop-

erty values.

Thus despite some caveats, the balance

o evidence supports the basic intuition that

rm location decisions are responsive to

dierences in property taxes. However, the

net eect o a property tax cut on business


28/80


. . . . . . . . . . . . . . . . . .

investment in a jurisdiction is much smal-

ler than the direct eect, and depends on

whether the tax cut is nanced by reducing

public services or business and the extentthat land values increase in response to

a tax cut.

Inaddition,whilelowerpropertytaxes

should increase business investment, the

eect on employment is less clear, because

lower property taxes reduce the cost o

machinery and equipment relative to labor.

Job growth induced by greater business

investment (i.e., scale eect) could be out-

weighed by job losses due to substituting

machinery or labor (i.e., substitution eect).

Finally, the eect o property taxes on the

location decision or a specic acility can

be signicantly dierent rom the average

eect or all rms.

EM P IR ICAL Ev IDENCE

There are three common approaches

or estimating the eect o taxes on local

economies: surveys, regression analysis,

and representative rm models. Surveys

ask business decision makers about the role

o taxes and incentives in their acility loca-

tion decisions. While surveys can be infuen-

tial, they are unreliable since those surveyedhave an incentive to exaggerate the eect o

taxes and incentives on their decisions as a

way to lobby or preerred policies. Regres-

sion analysis and representative rm models

are more reliable because they look at a

rms actual decisions and take into account

many o the other local actors that aect

protability to determine the true impor-

tance o taxes and incentives.

An examination o studies done between

1990 and 2011 suggests that the best litera-

ture reviews on this issue are still Bartik (1991)

andWasylenko(1997),whosummarize

the results o roughly 90 studies that used

regression analysis to estimate the eect o

state and local taxes on economic activity.

Table 3.1 describes the methodology used

in these studies, including the measures o

economic activity, which include employment,

rm births and relocations, investment, and

income. One key result is that dierences


29/80


. . . . . . . . . . . . . . . .

T a B l e 3 . 1

Impact o State and Local Taes on Economic Actiit:A Summar o Empirical Eidence

Ta Dierences Across Regions Ta Dierences within a Region

Impact o Ta Cuts

on Economic Actiit

Increase in Economic Activity rom

10% Cut in Total State & Local Taxes

Increase in Economic Activity rom

10% Cut in Local Property Taxes

Median Estimate 2% to 3% 16% to 20%

Most Likely Range 1% to 6% 10% to 30%

Ta Reenue Change Per

Job Created b Ta Cuts

Annual Recurring Change

in Total State and Local Tax Revenue

Annual Recurring Change

in Local Property Tax Revenue

Median Estimate $17,337 $1,035

Most Likely Range $52,011 to -$3,853 $0 to $1,553

Methodolog These studies measure the long-runrelationship between dierences in

taxes or entire regions (states or metro

areas) and dierences in employment,

rm births and relocations, investment,

income, and gross product or these

regions.

These studies measure the long-runrelationship between dierences in

property taxes or individual jurisdic-

tions within a region (typically a metro

area) and dierences in employment

and rm births and relocations or

these jurisdictions.

Note: See Appendix Notes or details and calculations.

Sources: Bartik (1991; 2005).

in taxes within a given region have a ve to

ten times greater impact on economic activ-

ity than dierences in taxes across regions.

This key distinction refects the two-stage site

selection process described earlier. However,

while a 10 percent cut in property taxes or

one jurisdiction is associated with a 16 to

20 percent increase in economic activity,

the site location process suggests this eect

islargelyazero-sumgamefortheregionas

a whole, because the increase in economic

activity in one jurisdiction is oset by de-

creases in other jurisdictions (Wassmer 2009).

The across-region results in table 3.1 also

rule out the possibility o across-the-board

tax cuts generating enough new economic

activity to actually increase tax revenues at

theregionallevel.Infact,themedianesti-

mate suggests that creating one job through

tax cuts would require a recurring annual

loss in state and local tax revenue o $17,337.

Again, the story is quite dierent or indi-

vidual jurisdictions. The within-region

results suggest that decreased revenues rom

lower tax rates could be more than oset

by increased revenues rom new economic

activity, so that lower property tax rates

could increase revenues by $1,035 per year

or each job created.

To obtain reliable estimates o the eect

o taxes on economic activity, it is crucial

to measure the quality o public services

accurately. Otherwise the eect o higher-

quality public services (expected to increase

economic activity) can be incorrectly attrib-

uted to the eect o higher taxes (expected

to decrease activity), which will underesti-

mate the eect o taxes.

For example, Phillips and Goss (1995)

nd that studies that control or public ser-

vices estimate the eect o tax dierences

betweenregionstobetwicethesizefound

by studies that do not. The average eect o

a 10 percent cut in state and local taxes is a

4.48 percent increase in regional economic

activity in studies with a public service


30/80


. . . . . . . . . . . . . . . . . .

control variable, and 2.16 percent in studies

that do not account or dierences in public

services. The 4.48 percent estimate could be

interpreted as the eect o tax cuts holdingpublic services constant, while the 2.16 per-

cent estimate is the combined eect o tax

cuts and accompanying reductions in public

services.

Regression analyses must overcome a

host o other econometric problems and

measurement issues to obtain reliable esti-

mates o the eect o taxes (Wasylenko 1997).

Because o these signicant challenges, re-

gression studies may have inexplicably large

variations across industries, statistically in-

signicant coecient estimates, a very wide

range o elasticity estimates, or be dicult

to replicate using data rom dierent years

(McGuire 2003). One possible reason or

these wide variations is suggested by Ady

(1997), who is skeptical o trying to reach

general estimates o the eect o taxes on

economic activity because the importance

o taxes varies so much across industries,

the stage o site selection, and even individ-

ual rms in the same sector. For example,property taxes place a higher burden on

capital-intensive industries, such as most

manuacturing rms, than on labor-inten-

sive industries, such as many service-sector

rms (gure 3.3).

An alternative methodology is the repre-

sentative rm approach, which combines

models built to accurately refect the nan-

cial statements o typical rms with detailed

inormation about state and local tax provi-

sions. Starting with the same pre-tax prot

rate or each city or state, researchers calcu-

late the marginal ater-tax prot rate or

new investment projects in each location

or specic industries. These studies allow

or a much more complete picture o the

tax system, including the treatment o

depreciation, tax credits, exemptions,

F i g u r e 3 . 3

Capital Inestment per Job or Selected Industries, 2010

Note: Based on an analysis o 6,500 large mobile business investments in 2010 worth $137 billion.

Source: Ernst & Young LLP (2011, 11).

0 100 200 300 400 500

Business Support Services

Financial and Real Estate Services

Health Care Insurance

Professional Services

Wholesale and Retail Trade

Transport, Storage, and Logistics

Motor Vehicle and Parts Manufacturing

Food Manufacturing

Machinery and Fabricated Metal Manuf.

Semiconductor and Electronic Manuf.

Investment per Job ($ Thousands)

MostCapital-

Intensive

MostLabor-

Intensive


31/80


. . . . . . . . . . . . . . . .

and apportionment ormulas; and how

these eatures interact with rms ederal tax

payments, geographic distribution o sales

and existing acilities, and asset types. Theseactors oten have a larger impact on rms

prots than do statutory tax rates.

Fisher and Peters (1998) look at 16

manuacturing sectors in 112 cities and nd

small dierences in eective marginal tax

rates or most cities, although there are sig-

nicant dierences between the highest-

and lowest-tax cities. Similarly, Papke (1995)

nds that tax dierences have very little

eect on ater-tax rates o return among six

states in the Great Lakes region. However,

Papke (1987) nds signicant tax dierences

across states, and her analysis suggests that

higher eective tax rates do reduce capital

investmentinastate.Inaddition,thisre-

search shows how ederal deductibility o

state and local taxes signicantly reduces

the eect o property tax dierentials on

rms prots.

SUM M ARy

Research suggests that taxes play a role in

explaining dierences in economic activity

betwee

rethinking property tax incentives for business

Documents