Governing Arizona: perspectives for the future / an encomonic public policy assessment

              Governing Arizona:
Perspectives for the Future
An Economic Public Policy Assessment
________________________________________________________________________
Understanding Arizona’s Fiscal History
A Review of Expenditures and Revenue in Arizona from 1980- 2010
The Fiscal Shape of the American States
Trends and Issues in State Budgeting in the 21st Century
A Report of Governing Arizona
Published by:
The Communications Institute
Funded by:
The Thomas R. Brown Foundations
This Governing Arizona Report is a draft version. The final version will include additional content from the Governing Arizona forum and participating scholars and experts. We appreciate the contributions from the authors of these reports and from the participants in the 2009 forum.
This research in the public interest was funded by the Thomas R. Brown Foundations as part of the Foundations‟ commitment to expanded use of economic analysis in public policy decision making. The Foundations‟ goal is to provide policy makers at all levels with the knowledge to make better decisions that benefit the public and the state of Arizona.
This study is part of the Governing Arizona Project, a program of The Communications Institute with funding from the Thomas R. Brown Foundations. The Institute administers the economics and public policy programs of the Thomas R. Brown Foundations, of which this study is a part.
Cover Design by Dean Swartz
Copyright 2009 The Communications Institute &
The Thomas R. Brown Foundations
All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means without permission in writing from The Communications Institute.
Published by The Communications Institute
Los Angeles, California
Mailing address:
18860 Nordhoff Street, Suite 204
Northridge, Ca 91324
818.349.5555
To order additional copies or for additional information
Contact the Institute by phone or email
research@ tci1. org or fax request to 818.279.0568
Preface
Arizona‟ s hallmark has always been a western „ can- do‟ spirit that allows individuals to pursue their aspirations and their dreams. As the state‟ s Centennial approaches, it is important to note that this outlook, along with an abundance of economic opportunities, has resulted in dramatic growth in Arizona over the past century and that population increase has placed increasing strain on the state‟ s aging infrastructure. Though low taxes and limited government have kept down the cost of living in Arizona, the state‟ s future economic strength will also be driven by a robust business environment dependent on an educated work force, sufficient public services, and an equitable tax system.
The nation‟ s financial downturn has significantly impacted Arizona by reducing tax revenue which, in turn, impacts public education, public safety, law enforcement, transportation, and other vital services traditionally provided by the government. It has also impact nearly every state and community in the nation. The recession has also impacted private businesses through slowed sales and has forced some companies to close their doors. It is critically important that leaders in the public and private sectors balance the appropriate role of government in providing needed services to the public while evaluating the potential need to realign the state‟ s fiscal and tax policies.
The Fiscal Shape of the American States- The first half of this report looks at trends in fiscal activity over the last 30 years for the 50 U. S. states. It assesses the most important factors that have driven government spending and the success of measures to bring spending under tighter control where it has become a serious problem. It also reviews which government sectors have increased spending the most as well as important trends in state and local revenues.
This section of the report is authored by Mathew D. McCubbins, one of the nation‟ s leading experts on state and local government finance. McCubbins holds a Ph. D. in economics from Caltech and is a Professor of Law at the University of California, San Diego and Co- Director of the USC – Caltech Center for the Study of Law and Politics. Ellen Moule, Graduate Student, University of California San Diego Department of Political Science, is a co- author of the study.
Understanding Arizona’s Fiscal History – The other key purpose of this report is to provide policy makers, leaders across the state, citizens, and the media with a solid, objective analysis of Arizona‟ s fiscal condition. This report is intended to put Arizona‟ s current state fiscal condition in a national and historical context over the past thirty years. The state‟ s population has doubled in that time and is now more than 6.5 million, and many leaders and citizens are new to Arizona. Yet knowledge of history can help us make better decisions for the future.
Historian David C. McCullough noted “ A nation that forgets its past can function no better than an individual with amnesia.” Konrad Adenauer, the Chancellor of Germany after the tragedy of World War II reminded us “ History is the sum total of the things that could have been avoided.” Leaders and citizens would be wise to heed these words and look at our history in order to better plan for our future. This report provides that essential perspective.
We are indebted to Alan Maguire for his diligent work in putting this report together. He spent endless hours researching printed reports and assembling and interpreting mountains of data. Alan began working on state fiscal issues more that 30 years ago working with numerous legislatures, governors, and leaders from both political parties, and few Arizonans have his perspective and expertise.
We believe that it is critically important for all Arizonans regardless of partisan affiliation to work together to solve our problems in the best interests of the people and not lose sight of who we serve. John F. Kennedy challenged us nearly a half century ago when he declared “ Let us not seek the Republican answer or the Democratic answer, but the right answer. Let us not seek to fix the blame for the past. Let us accept our own responsibility for the future.”
President Kennedy was correct and we believe the purpose of Governing Arizona is consistent with this challenge from the past as we build for the future. We believe this report makes a constructive contribution to this effort in providing the knowledge and perspective to intelligently tackle these difficult issues on behalf of the people.
Pete Rios, Ken Bennett
Co- Chair, Governing Arizona Co- Chair, Governing Arizona
Member of the Board of Supervisors Secretary of State
County of Pinal State of Arizona
Former President, Arizona State Senate Former President, Arizona State Senate
TABLE OF CONTENTS
Understanding Arizona’s Fiscal History
A Review of Expenditures and Revenue in Arizona from 1980- 2010 ……………………………..….…… 1
The Fiscal Shape of the American States
Trends and Issues in State Budgeting in the 21st Century ……………………………………………………… 21
Chart Details ………………………………………………………………………………………………..………. 52
References ………………………………………………………………………………………………………..……. 56 Understanding Arizona’s Fiscal History
______________________________________________________________________________________
A Review of Expenditures and Revenues in Arizona from1980- 2010
Alan Maguire
President
The Maguire Company
A Report of Governing Arizona
Published by:
The Communications Institute
Funded by:
The Thomas R. Brown Foundations
Understanding Arizona’s Fiscal History
EXECUTIVE SUMMARY
Much has changed in Arizona over the past thirty years. The sources of tax revenue to the state general fund have changed and the uses of those monies for operating budgets have changed as well. The changes in the general fund reflect hundreds, if not thousands, of separate policy decisions by Arizona’s state lawmakers. Looking back at these changes, now as the state faces unprecedented fiscal challenges, may provide a useful foundation for future discussions of the possible ways forward for the State.
In the thirty years since 1980, the allocation of state tax revenues in the general fund have changed significantly. The largest percentage change in state general fund revenues has been the virtual elimination of any property tax revenues. More importantly the shares attributable to the individual income tax and the sales tax have increased, while the share from the corporate income tax has declined.
Over the thirty year period since 1980, the largest percentage increase in operating spending has been in corrections followed by health and welfare and the largest percentage decline has been in university funding. Although corrections had a larger percentage change, the actual dollar amount shift in spending is much greater toward health and welfare.
INTRODUCTION
The State of Arizona is facing a time of unparalleled economic challenges. Over the last year, the State has been forced to revise and reduce authorized expenditures on multiple occasions as actual revenue collections have continued to fall farther and farther below forecast ( budgeted) levels. While the likely need for spending reductions was clearly foreseeable, the extent of the misalignment of authorized spending levels and realized revenues was less predictable.
The continuing debate surrounding these ongoing budget revisions has occasioned much discussion of Arizona‟ s overall fiscal situation. Many of these discussions have focused on what should be done to address the current circumstances and many have proposed approaches to prevent the re- occurrence of such events in the future. Advocates from all sectors of the political landscape had used the situation as justification for their preferred public policy “ solutions”. Some conservatives have argued that the solutions lie in lower government spending and a lower, more business friendly tax environment. Some liberals have argued that higher taxes and government programs intended to strengthen the state‟ s economy are the answers. Almost all of the discussions have referenced what Arizona “ ought to do” based on the successes of other states, or on national averages, or theoretical treatises.
Furthermore, the outlook for the state‟ s fiscal situation shows little signs of improving for at least several years. Consequently, the misalignment of authorized spending levels from the general fund and the expected future tax revenues is not a short- term phenomenon likely to naturally remedy itself.
The purpose of this review is to look primarily at what Arizona actually has done. This is a review of state general fund taxes and operating expenditures over the period from 1980 through the present. The state general fund is the principal recipient of state tax collections. Special taxes, like highway user taxes, flow to special funds like the highway user revenue fund, but the state‟ s general fund is the recipient of general state tax collections and the source from which most state supported and legislatively appropriated programs are funded. The state receives, distributes, and expends large amounts of federal money, but the focus of the annual state budget process is, and will continue to be, on the monies that come and go through the state general fund ( and consequently, most of the public discussion).
1 It is worth noting, although not directly relevant to this review, that Arizona‟ s current mismatch between authorized expenditures and tax collections is one among the largest, on a percentage basis, of any state. Furthermore, much of the spending from the general fund has been mandated by voter- approved initiatives and referenda and thereby largely beyond from legislative modification, making the challenges even greater.
The goal of this review is to provide some straight- forward information about how the operating expenditure and tax patterns of the state of Arizona, as reflected in its general fund, have changed since 1980. Looking at changes of the 30 year period may provide some useful perspective on the trends and patterns of the state‟ s tax and spending practices. The period is long enough to reveal substantial trends, stretches beyond cyclical economic patterns, bridges multiple gubernatorial administrations, and spans many multiple two year legislative terms. This historical perspective may provide a foundation for future discussions of the possible ways forward for the State.
Many things have happened in Arizona since 1980. The state‟ s population has grown from just over 2.7 million to just under an estimated 7 million. That growth represents an increase of over 4 million people – a more than doubling of the state‟ s population. If the conventional wisdom is true that says for every three people who come to Arizona two leave, that means that 12 million people have migrated into Arizona since 1980 and 8 million have moved on.
But also many things remain the same as thirty years ago. Phoenix remains the most populous city; Maricopa still has the largest county population. There are still 90 state legislators and still one Governor.
METHODOLOGY
This review was undertaken based on information from published materials, both printed and on- line, previously produced by the Joint Legislative Budget Committee of the Arizona Legislature and the Arizona Department of Revenue through mid- October 2009. The author is indebted to the staffs of these fine agencies for their assistance in locating a variety of materials, some of which are quite aged. Any errors in the review are the authors and should in no way reflect in these organizations.
Tax and expenditure information was assembled for each decennial year – 1980, 1990, 2000, and 2010. In each instance ( except 2010), the “ decennial year” figures for the purpose of this review were calculated using data for three years were averaged– the year prior, the decennial year, and the year following. For example, 1989, 1990, and 1991 were averaged to represent 1990. The use of the three year average reduces the impact of any anomalous one year amount or event. For 2010, only 2009 and 2010 were averaged.
Tax information was based on actual historical collections and estimated amounts for 2009 and 2010. Expenditure information was based on appropriated amounts, as reflected in the annual Appropriations Reports.
2 ARIZONA‟ S GENERAL FUND TAX REVENUES
In 1980, the largest source of tax revenues to the state general fund was the transactions privilege tax, commonly referred to as the sales tax, and the accompanying use tax. The retail sales tax rate was 4%. The sales tax contributed slightly less than one- half of all general fund tax revenues.
FY 1980 Average General Fund Tax RevenuesSales & Use Tax47% Property Tax6% Other13% Personal Income Tax25% Corporate Income Tax9% Sales - 47% Income - 34% Property - 6% Other - 13% Source: JLBC, TMC
Figure 1
Income taxes imposed on individual households and corporations represented slightly more than one- third of state general fund taxes in 1980.
Property taxes imposed by the state constituted approximately six percent of general fund tax collections and other taxes represented about thirteen percent of the taxes deposited into the state‟ s general fund.
NOTABLE TAX LAW CHANGES BETWEEN 1980 AND 1990
In 1978 and 1980, exemptions, deductions, and credits against the individual income tax were indexed to increase with inflation.
In 1980, food sold for home consumption was exempted from the state transaction privilege ( sales) tax.
In 1982, inflation indexing was expanded to adjust the income tax bracket for inflation.
3 In 1982, the corporate income tax was again conformed to changes in the federal income tax including provisions for accelerated depreciation.
In 1983, the sales tax rate was increased by 25%, such that the general retail rate increased from 4% to 5%. The increase was initialed intended to be a temporary increase.
In 1987, a “ windfall” subtraction from the individual income tax was established to offset higher Arizona income liabilities resulting from conformance to the federal tax code.
In 1987, “ enterprise zones” were established and corporate income tax credits were established for employees in such zones.
In 1988, the preferential corporate income tax treatment for capital gains was eliminated and the subtraction from income for certain dividends was restricted.
In 1989, the individual income tax “ windfall” subtraction was limited, as was the deductibility of consumer interest payments. The subtraction for public pension incomes was broadened and limited. In 1989, the tax credit for enterprise zone employees was restricted.
( Note: in addition to the tax changes listed here, and in the subsequent sections, there were dozens of smaller changes that also impacted tax collections in varying ways.)
*****
FY 1990 Average General Fund Tax RevenuesSales & Use Tax48% Property Tax4% Other10% Personal Income Tax32% Corporate Income Tax6% Sales - 48% Income - 38% Property - 4% Other - 10% Source: JLBC, TMC
Figure 2
4 By 1990, the share of sales taxes in the general fund had grown slightly, to 48% from 47% in 1980. Sales tax collections were the largest source of general fund taxes at roughly one- half and the retail sales tax rate had risen to 5%.
Income taxes in 1990 had expanded to 38% of general fund taxes from 34% in 1980, while property taxes had declined from a six percent share to only four percent. Similarly other taxes had dropped to ten percent from 13 percent in 1980.
NOTABLE TAX LAW CHANGES BETWEEN 1990 AND 2000
In 1990, the individual income tax rate structure was consolidated and altered. The standard deduction was increased and new, qualified tax exemptions were established. The “ windfall” subtraction was eliminated and the other additions and subtractions from gross income were modified. Inflation indexing of the exemptions, deductions, and tax brackets was eliminated.
In 1990, the corporate income tax brackets and tax rates were replaced with a single 9.3% rate. The subtractions from income for federal income and foreign taxes were eliminated.
In 1991, the deductibility of the medical expenses from the individual income tax was gradually increased over several years to full deductibility from the individual income tax.
In 1991, the double weighted sales factor was established for corporate income taxes of multi- state corporations.
In 1992, a corporate income tax credit for qualified research and development tax credits was established.
In 1992, the amount of the personal, dependent, and elderly individual income tax exemption were increased.
In 1993, the maximum allowable income tax credit for research and development expenses was increased.
In 1993, a subtraction from corporate income was established for alternative fuel vehicles and equipment.
In 1994, the corporate income tax rate was reduced to 9% and separately, income tax credits for pollution control equipment and related construction materials were established. Also, an income tax credit for certain corrective actions for an underground storage facility was established.
In 1994, the corporate income tax subtraction for alternative fuel vehicles and equipment was replaced by a tax credit.
In 1994 and 1995, individual income tax rates were reduced a combined total of approximately 20%. The standard deduction was increased and new, limited, personal and dependent tax credits were established. The share of income tax revenues shared with cities was increased.
5 In 1996, a corporate tax credit for increases in qualified employment in military re- use zones was established and a tax credit for certain qualifying retail business based on employment positions was established.
In 1997, the income tax credits for alternative fuel vehicles and equipment were extended.
In 1997, two K- 12 education individual income tax credits were established for public school extracurricular activities and private school tuition.
In 1998, individual income tax rates were reduced with the top marginal rate dropping to 5.17% from 5.6% and the certain personal exemption amounts were increased. The share of income tax revenues shared with cities was increased.
In 1998, a corporate income tax credit for coal used in electrical generation was established.
In 1998, the deductions and credits for alternative fuel vehicles and facilities were expanded.
In 1998, the corporate income tax rate was reduced to 8% from 9% and the sales “ throwback” rules were modified.
In 1998 and 1999, the family tax credit was increased and expanded and certain dependent exemptions were increased. Individual income tax rates were reduced by a combined total of approximately 2.5%.
In 1999, the income tax credits for alternative fuel vehicles and stations were changed to a percentage of cost.
In 1999, the corporate income tax rate was reduced to 7% from 8% and some corporate tax credits were eliminated.
FY 2000 Average General Fund Tax RevenuesProperty Tax1% Other6% Corporate Income Tax8% Personal Income Tax35% Sales & Use Tax50% Sales - 50% Income - 43% Property - 1% Other - 6% Source: JLBC, TMC
Figure 3
6 By 2000, the share of sales taxes in the state general fund had increased 50%, from 48% in 1990.
Income taxes had grown to 43%, up from 38% ten years earlier. The share of personal income taxes had increased to 35% from 32%, and the share of corporate income taxes had also grown to 8% from 6%. Over the decade from 1990 to 2000, property dropped to only one percent from four percent and other taxes declined to six percent from ten percent.
NOTABLE TAX LAW CHANGES BETWEEN 2000 and 2010
In 2000, the state transaction privilege ( sales) tax rate was increased to 5.6% from 5%, but the increased amounts are not deposited into the state general fund, but rather dedicated and distributed for a variety of education related activities.
In 2000, the limits on the amount of the two K- 12 education individual income tax credits were increased.
In 2000, the tax credits for the purchase of alternative fuel vehicles and related systems were expanded. Later, a moratorium was imposed on the income tax credits and ultimately, the tax credit program was eliminated.
In 2000, a corporate income tax credit for employee technology skill training was established.
In 2000, the voters approved an increase in the state sales tax rate ( to 5.6% from 5% for retail purchases), for education related purposes, but the revenues from the increase are not deposited into the general fund.
In 2001, the corporate income tax enterprise zone program was extended.
In 2001, the individual income tax standard deduction was increased.
In 2002, the deduction from corporate income tax purposes for certain stock dividends was eliminated.
In 2004, the income tax credits for corrective action related to underground storage tanks was eliminated and the enterprise zone tax credits for employment were lengthened.
In 2005, income tax credits were established for motion picture production in the state.
In 2005, an optional enhanced sales factor formula for computing corporate income taxes was established.
In 2005, the termination dates and renewal dates for military reuse zones were extended.
In 2005, the individual income tax standard deduction was inflation indexed.
In 2006, a corporate income tax credit for contributions to school tuition organizations was established with an annual cap.
In 2006 and 2007, individual income tax rates were reduced by a combined total of 20%.
7
In 2007, the motion picture production tax incentive program was modified and credit limits for each production were increased.
In 2008, the amount of the research and development tax credit against individual and corporate income taxes was expanded through 2018. A minimum distribution amount was established for the amount of income taxes shared with cities.
*****
FY 2010 Average General Fund Tax RevenuesProperty Tax0% Other7% Sales & Use Tax56% Personal Income Tax30% Corporate Income Tax7% Sales - 56% Income - 37% Property - 0% Other - 7% Source: JLBC, TMC
Figure 4
In the decade between 2000 and 2010, sales taxes as a share of general fund tax revenues rose to 56% from 50%. Total income taxes, personal and corporate, declined to 37% from 43%. Personal income taxes as a share of the general fund tax revenues declined to 30% from 35% and the share of corporate income taxes remained also most the same. Other taxes contributed 7% and property taxes were negligible.
It is important to note that between 2000 and 2010 general fund tax revenues experienced significant volatility. Tax revenues grew rapidly in the early years of the decade and then fell precipitously in the most recent fiscal years. However, the 2010 allocations ( actual the average of FYs 2009 and 2010) represent the current situation of the state general fund. The significant volatility might argue for a different, more normalized comparison period but for the fact that 2010 is the current fiscal year and the point in time of this review. It also remains to be determined in the recent tax revenue patterns will be an aberration or the “ new normal”.
8 ARIZONA‟ S GENERAL FUND OPERATING SPENDING
In 1980, the largest single category of the general fund operating spending was state support of the K- 12 education system representing 44% all of spending. Overall education spending, K- 12 plus universities, totaled 63% of general fund operating spending. ( While there is other education- related spending, including for community colleges, this review focuses on the K- 12 and university portions of education spending.) Health and welfare spending – Health Services and Economic Security – was 16% of general fund spending. Corrections spending was 5% and general government ( and all other) totaled 17%.
FY 1980 Average General Fund OperatingK- 1244% DHS5% DES11% Corrections5% Gen Gvt / Other17% AHCCCS0% Universities18% Education - 63% H & W - 16% Corrections - 5% Gen. Gvt / Other - 17% Source: JLBC, TMC
Figure 5
NOTABLE OPERATING SPENDING CHANGES BETWEEN 1980 AND 1990
( Note: in addition to the changes listed here, and in the subsequent sections, there were dozens of statutory changes that also impacted spending in varying ways Only the most fiscally significant have been listed here.)
In 1980, the state adopted a new K- 12 educational finance formula that established the general fund as the balancing source for local school district budgets. The new formula substantially equalized per pupil operating spending across districts in the state and substantially equalized school district operating property tax rates across the state. Consequently, the general fund‟ s obligation to K- 12 funding became a function of overall authorized K- 12 budgets and the amount of revenue generated by the equalized school district operating tax rate, or “ qualifying tax rate”.
In 1981, the state created the Arizona Health Care Cost Containment System ( AHCCCS). The new agency was the state approach to joining the federal Medicaid program.
*****
9 FY 1990 Average General Fund OperatingK- 1237% DHS5% DES9% Corrections8% Gen Gvt / Other15% Universities16% AHCCCS10% Education - 53% H & W - 24% Corrections - 8% Gen. Gvt / Other - 15% Source: JLBC, TMC
Figure 6
By 1990, K- 12 education as a share of total general fund spending had declined to 37% from 44% and all combined education spending was 53% down from 63%. Over the same period, health and welfare spending increased to 24% from 16%. Corrections spending increased to 8% from 5%, while general government / other spending decreased slightly to 15% from 17%.
NOTABLE OPERATING SPENDING CHANGES BETWEEN 1990 AND 2000
In 1996, voters approved a ballot initiative expanding health and welfare spending to be supported by revenues from the state lottery.
Through the 1990s, litigation regarding the level and nature of state funding of behavioral health services significantly expanded the general fund operating spending for these services.
Through the later 1990s, the state general fund provided an increasing share of the costs of providing long- term care for qualified older residents.
Through much of the 1990s, litigation continued regarding the reliance of K- 12 school districts on local property taxes to finance school building construction. The resolution of the litigation ultimately involved the creation of the state school facilities board and transference of much of the costs associated with the construction and maintenance of K- 12 school facilities to the general fund.
*****
10 FY 2000 Average General Fund OperatingK- 1240% DHS4% DES7% Corrections9% Gen Gvt / Other20% AHCCCS8% Universities12% Education - 52% H & W - 20% Corrections - 9% Gen. Gvt / Other - 20% Source: JLBC, TMC
Figure 7
In 2000, K- 12 spending from the general fund increased to 40% from the 1990 level of 37%. Combined education spending remained largely unchanged from 1990 at 52%. University spending declined, again, to 12% from 16%. Health and welfare spending declined to 20% of total general fund operating spending from 24%. Correction spending increased slightly to 9% of the general fund, while general government / other increased to 20% from 15%.
NOTABLE OPERATING SPENDING CHANGES BETWEEN 2000 AND 2010
In 2000, voters approved a referendum to increase education spending to be supported by an increase in the state sales tax. Expenditures resulting from this action are in addition to the amounts expended from the general fund.
In 2000, voters approved two initiatives that expanded health and welfare spending to be supported in part by monies to be received from the state‟ s share of the master tobacco tax: and while some of the spending and all revenue are outside the general fund, the proposition established program requirements that require general fund spending.
Through the early to mid 2000s, K- 12 spending was expanded to cover local school district costs associated with all day kindergarten.
*****
11 In 2010, K- 12 education is budgeted at 45% of the general fund, up from 40% in 2000 and combined education spending also grew to 55% from 52% in 2000. Health and welfare spending is budgeted to increase to 26% from 20%, while corrections‟ share increased slightly to 10%. General government is budgeted to decline to 9% from 20%.
FY 2010 Average General Fund OperatingK- 1245% DHS5% DES8% Corrections10% Gen Gvt / Other9% AHCCCS13% Universities10% Education - 55% H & W - 26% Corrections - 10% Gen. Gvt / Other - 9% Source: JLBC, TMC
Figure 8
It is important to note, as mentioned earlier, that between 2000 and 2010 general fund tax revenues experienced significant volatility. Tax revenues grew rapidly in the early years of the decade and then fell precipitously in the most recent fiscal years. The changes in tax revenues have in turn forced changes in general fund operating spending, especially in the most recent two fiscal years. However, the 2010 allocations ( actual the average of FYs 2009 and 2010) represent the current situation of the state general fund. The significant volatility might argue for a different, more normalized comparison period but for the fact that 2010 is the current fiscal year and the point in time of this review. It also remains to be determined in the recent tax revenue patterns and resulting operating spending patterns will be an aberration or the “ new normal”.
12 AN ALTERNATIVE HISTORY
The preceding charts illustrate the significant changes in the allocation of general fund tax revenues and operating spending over the past 30 years. These changes have occurred due to a vast number of individual policy decisions made by state lawmakers – legislators and governors – as well by voter initiatives.
Often these decisions are necessarily considered within the narrow time frames of contemporaneous events at the time of the decision. Examining the cumulative changes over a longer time frame reveals trends and patterns that are not as apparent in shorter time horizons.
CHANGES IN THE ALLOCATION OF GENERAL FUND TAX REVENUES
What would state general fund tax revenues in FY 2010 look like if the tax allocation made in 1980 ( technically the average of FYs 79, 80, and 81) and FY 1990 ( technically the average of FYs 89, 90, and 91) were applied to 2010?
Using the 1980 tax revenue allocation for the state general applied to the level of 2010 tax revenues, corporate income tax and property tax collections would be higher and individual income tax and sales tax collections would be lower.
In fact, using the 1980 allocation of state general fund tax revenues, property taxes would represent almost three- quarters as much tax revenue as corporate income taxes. Overall income tax revenues would be about 10% lower and be 3% lower share of general fund tax revenues. Personal income tax revenues would be lower, while corporate income tax revenues would be higher.
Over the thirty year period since 1980, the largest percentage change in state general fund revenues has been the virtual elimination of any property tax revenues. More importantly, the shares attributable to the individual income tax and the sales tax have increased, while the share from the corporate income tax has decreased.
Assuming there were no changes to total tax revenues for the general fund, the following amounts would have been collected in FY 2010, based on the percentage allocation in 1980.
Calculation of FY 2010 General Fund Tax Revenues
Based on General Fund Percentage Allocation in 1980
Revenue Category
1980%
2010 Revised Amount
Change from Actual 2010 Amount
% Change
Sales & Use
48%
$ 3,269,000,000
-$ 509,000,000
- 13%
Personal Income
25%
$ 1,659,000,000
-$ 362,000,000
- 18%
Corporate Income
9%
$ 575,000,000
$ 108,000,000
23%
Subtotal Income
33%
$ 2,234,000,000
-$ 254,000,000
- 10%
Property
6%
$ 400,000,000
$ 382,000,000
2168%
Other
13%
$ 855,000,000
$ 381,000,000
80%
Total GF Taxes
100%
$ 6,759,000,000
$ 0
0%
Source: TMC
Table 1
13 FY 1980 Average General Fund Tax RevenuesSales & Use Tax47% Property Tax6% Other13% Personal Income Tax25% Corporate Income Tax9% Sales - 47% Income - 34% Property - 6% Other - 13% Source: JLBC, TMC
FY 2010 Average General Fund Tax RevenuesProperty Tax0% Other7% Sales & Use Tax56% Personal Income Tax30% Corporate Income Tax7% Sales - 56% Income - 37% Property - 0% Other - 7% Source: JLBC, TMCFigure 9
*****
Again assuming there were no changes to total tax revenues for the general fund, the following amounts would have been collected in FY 2010, based on the percentage allocation in 1990.
Using the 1990 tax revenue allocation for the state general applied to the level of 2010 tax revenues, again, property tax collections would be significantly higher. Sales tax revenues would be 13% lower, declining to 48% of general fund tax revenues from 56%. Overall income tax revenues would be about roughly the same as a general fund tax source. Individual income tax collections would be higher and corporate income taxes lower.
Calculation of FY 2010 General Fund Tax Revenues
Based on General Fund Percentage Allocation in 1990
Revenue Category
1990%
2010 Revised Amount
Change from Actual 2010 Amount
% Change
0
Sales & Use
48%
$ 3,276,000,000
-$ 502,000,000
- 13%
Personal Income
32%
$ 2,140,000,000
$ 119,000,000
6%
Corporate Income
6%
$ 387,000,000
-$ 80,000,000
- 17%
Subtotal Income
37%
$ 2,527,000,000
$ 39,000,000
2%
Property
4%
$ 304,000,000
$ 286,000,000
1623%
Other
10%
$ 651,000,000
$ 177,000,000
37%
Total GF Taxes
100%
$ 6,759,000,000
$ 0
0%
Source: TMC
Table 2
14 Over the twenty year period since 1990, the largest percentage change in state general fund revenues was still the virtual elimination of any property tax revenues. The share attributable to the sales tax has increased as has the share from corporate income taxes, while the share from the individual income tax has declined.
FY 1990 Average General Fund Tax RevenuesSales & Use Tax48% Property Tax4% Other10% Personal Income Tax32% Corporate Income Tax6% Sales - 48% Income - 38% Property - 4% Other - 10% Source: JLBC, TMC
FY 2010 Average General Fund Tax RevenuesProperty Tax0% Other7% Sales & Use Tax56% Personal Income Tax30% Corporate Income Tax7% Sales - 56% Income - 37% Property - 0% Other - 7% Source: JLBC, TMC
Figure 10
*****
CHANGES IN THE ALLOCATION OF GENERAL FUND OPERATING SPENDING
What would state general fund operating spending in FY 2010 look like if the spending allocation made in 1980 ( technically the average of FYs 79, 80, and 81) and FY 1990 ( technically the average of FYs 89, 90, and 91) were applied to 2010?
Using the 1980 allocations and the 2010 overall spending levels, overall education spending would increase by over $ 700 million, or roughly 15%, or to a 63% share from a 55% share of general fund operating spending. More specifically, K- 12 funding would increase only slightly, but university funding would increase by over $ 700 million, an increase of over three- quarters, or to an 18% share from a 10% share of general fund operating spending.
On the other hand, health & welfare spending would decline over $ 900 million or over 40%, or to a 16% share from a 26% share of general fund operating spending. The decline in health and welfare spending occurs in large part because AHCCCS, the state‟ s version of Medicaid, was not enacted until later ( at the time, Arizona counties were responsible for providing most indigent health care), although there would be reductions in DHS and DES spending as well.
Correction spending would be lower as well by roughly one- half and general government / other spending would increase significantly.
*****
15 Assuming there were no changes to total spending from the general fund, the following amounts would have been appropriated in FY 2010, based on the percentage allocation in 1980.
Calculation of FY 2010 General Fund Operating Budgets
Based on General Fund Percentage Allocation in 1980
Expenditure Category
1980%
2010 Revised Amount
Change from Actual 2010 Amount
% Change
K- 12
45%
$ 4,076,000,000
$ 23,000,000
1%
Universities
18%
$ 1,653,000,000
$ 733,000,000
80%
Education Subtotal
63%
$ 5,729,000,000
$ 756,000,000
15%
AHCCCS
0%
$ 0
-$ 1,183,000,000
- 100%
DHS
5%
$ 432,000,000
-$ 64,000,000
- 13%
DES
11%
$ 974,000,000
$ 289,000,000
42%
Health & Welfare Subtotal
16%
$ 1,406,000,000
-$ 958,000,000
- 41%
Corrections
5%
$ 419,000,000
-$ 484,000,000
- 54%
Gen Gvt / Other
17%
$ 1,516,000,000
$ 685,000,000
82%
Total GF Operating
100%
$ 9,070,000,000
$ 0
0%
Source: TMC
Table 3
FY 1980 Average General Fund OperatingK- 1244% DHS5% DES11% Corrections5% Gen Gvt / Other17% AHCCCS0% Universities18% Education - 63% H & W - 16% Corrections - 5% Gen. Gvt / Other - 17% Source: JLBC, TMC
Figure 11
FY 2010 Average General Fund OperatingK- 1245% DHS5% DES8% Corrections10% Gen Gvt / Other9% AHCCCS13% Universities10% Education - 55% H & W - 26% Corrections - 10% Gen. Gvt / Other - 9% Source: JLBC, TMC
Over the thirty year period since 1980, the correction share of general fund operating spending had the largest percentage increase followed by health and welfare and the largest percentage decline has been in university funding. Although corrections had a larger percentage change, the actual dollar amount shift in spending is much greater toward health and welfare.
16 Again assuming there were no changes to total spending from the general fund, the following amounts would have been appropriated in FY 2010, based on the percentage allocation in 1990.
Calculation of FY 2010 General Fund Operating Budgets
Based on General Fund Percentage Allocation in 1990
Expenditure Category
1990%
2010 Revised Amount
Change from Actual 2010 Amount
% Change
0
K- 12
37%
$ 3,396,000,000
-$ 657,000,000
- 16%
Universities
16%
$ 1,446,000,000
$ 526,000,000
57%
Education Subtotal
53%
$ 4,842,000,000
-$ 131,000,000
- 3%
AHCCCS
10%
$ 876,000,000
-$ 307,000,000
- 26%
DHS
5%
$ 411,000,000
-$ 85,000,000
- 17%
DES
9%
$ 826,000,000
$ 141,000,000
21%
Health & Welfare Subtotal
23%
$ 2,113,000,000
-$ 251,000,000
- 11%
Corrections
8%
$ 718,000,000
-$ 185,000,000
- 20%
Gen Gvt / Other
15%
$ 1,397,000,000
$ 566,000,000
68%
Total GF Operating
100%
$ 9,070,000,000
$ 0
0%
Source: TMC
Table 4
FY 1990 Average General Fund OperatingK- 1237% DHS5% DES9% Corrections8% Gen Gvt / Other15% Universities16% AHCCCS10% Education - 53% H & W - 24% Corrections - 8% Gen. Gvt / Other - 15% Source: JLBC, TMC
Figure 12
FY 2010 Average General Fund OperatingK- 1245% DHS5% DES8% Corrections10% Gen Gvt / Other9% AHCCCS13% Universities10% Education - 55% H & W - 26% Corrections - 10% Gen. Gvt / Other - 9% Source: JLBC, TMC
17
Using the 1990 allocations and the 2010 overall spending levels, overall education would be almost unchanged. However, K- 12 funding would be lower by over $ 600 million or over 15%, to a 37% share from a 45% share of general fund operating spending, while university funding would increase by over $ 500 million or by over half, to a 16% share from a 10% share of general fund operating spending.
Health & welfare spending would decline over $ 200 million or over 10%, to a 24% share from a 26% share of general fund operating spending. Both AHCCCS and DHS would decline while DES spending would increase.
Correction spending would be lower as well by roughly by one- fifth, or about $ 180 million, to an 8% share from a 10% share of general fund operating spending and general government / other spending would increase significantly.
Over the twenty year period since 1990, the largest percentage increase has been in AHCCCS and the largest percentage decline has been in university funding. The combined shifts toward AHCCCS, DHS, corrections, and K- 12 have been offset by allocation from general government and universities.
*****
CONCLUSION
In the thirty years since 1980, the allocation of state tax revenues in the general fund has changed significantly. These changes are the result of a myriad of individual tax changes made by state lawmakers, as delineated in the prior pages. The largest percentage change in state general fund revenues has been the virtual elimination of any state property tax revenues. More importantly, the shares attributable to the individual income tax and the sales tax have increased, while the share from the corporate income tax has declined.
In other words, if the state general fund derived its tax collections in the same proportions as it did in 1980, personal income taxes would be lower, and sales taxes would be lower. On the other hand, corporate income taxes would be higher.
Over the thirty year period since 1980, the allocation of operating spending supported by the general fund has also changed significantly. These changes are the result of policy choices made by state lawmakers, voter initiatives, and judicial decisions, as delineated in the prior pages. The largest percentage increase in operating spending has been in corrections followed by health and welfare spending and the largest percentage decline has been in university funding. Although corrections had a larger percentage change, the actual dollar amount shift in spending is much greater toward health and welfare.
In other words, if the state spent its general fund for operating budgets in the same proportions as it did in 1980, support of health and welfare spending would be significantly lower, while the share allocated to the universities would be significantly higher.
Finally, the changes in the general fund tax revenues and operating spending since 1980 reflect hundreds, if not thousands, of separate policy decisions by Arizona‟ s state lawmakers, voters, and judges. Hopefully, looking back at these changes in a time of unprecedented fiscal challenges may provide a useful foundation for future discussions of the possible ways forward for the State.
18 General Fund Tax Revenue Allocations 1980 through 2010
FY 1980 Average General Fund Tax RevenuesSales & Use Tax47% Property Tax6% Other13% Personal Income Tax25% Corporate Income Tax9% Sales - 47% Income - 34% Property - 6% Other - 13% Source: JLBC, TMCFY 1990 Average General Fund Tax RevenuesSales & Use Tax48% Property Tax4% Other10% Personal Income Tax32% Corporate Income Tax6% Sales - 48% Income - 38% Property - 4% Other - 10% Source: JLBC, TMC
Figure 13
19
FY 2000 Average General Fund Tax RevenuesProperty Tax1% Other6% Corporate Income Tax8% Personal Income Tax35% Sales & Use Tax50% Sales - 50% Income - 43% Property - 1% Other - 6% Source: JLBC, TMCFY 2010 Average General Fund Tax RevenuesProperty Tax0% Other7% Sales & Use Tax56% Personal Income Tax30% Corporate Income Tax7% Sales - 56% Income - 37% Property - 0% Other - 7% Source: JLBC, TMC General Fund Operating Spending Allocations 1980 through 2010
FY 1980 Average General Fund OperatingK- 1244% DHS5% DES11% Corrections5% Gen Gvt / Other17% AHCCCS0% Universities18% Education - 63% H & W - 16% Corrections - 5% Gen. Gvt / Other - 17% Source: JLBC, TMC
FY 1990 Average General Fund OperatingK- 1237% DHS5% DES9% Corrections8% Gen Gvt / Other15% Universities16% AHCCCS10% Education - 53% H & W - 24% Corrections - 8% Gen. Gvt / Other - 15% Source: JLBC, TMCFigure 14
20
FY 2000 Average General Fund OperatingK- 1240% DHS4% DES7% Corrections9% Gen Gvt / Other20% AHCCCS8% Universities12% Education - 52% H & W - 20% Corrections - 9% Gen. Gvt / Other - 20% Source: JLBC, TMC
FY 2010 Average General Fund OperatingK- 1245% DHS5% DES8% Corrections10% Gen Gvt / Other9% AHCCCS13% Universities10% Education - 55% H & W - 26% Corrections - 10% Gen. Gvt / Other - 9% Source: JLBC, TMC
The Fiscal Shape of the American States
_______________________________________________________________________
Trends and Issues in State Budgeting in the 21st Century
Mathew D. McCubbins, Ph. D. Distinguished Professor and Chancellor's Associates Chair
Department of Political Science
University of California, San Diego
Co- Director of the Caltech USC Center for the Study of Law and Politics
University of Southern California Ellen Moule Doctoral Student
Department of Political Science
University of California, San Diego
A Report of Governing Arizona
Published by:
The Communications Institute
Funded by:
The Thomas R. Brown Foundations The Fiscal Shape of the American States
Trends and Issues in State Budgeting in the 21st Century
This report looks at trends in fiscal activity for the fifty U. S. states over the last thirty years. The report is comprised of three sections. In the first section, we examine trends in state and local expenditures. We find, as already been well documented elsewhere, that expenditures have grown as a percentage of personal income. This growth has favored state governments at the expense of local governments. We also report on the rising employee retirement liabilities, an expense often overlooked in political discussions about the size of government. In our second section, we review trends in state and local revenues. Dramatic changes in revenue streams have occurred over the last thirty years, causing property tax revenues to be supplanted by user fees and new fiscal innovations, often involving fiscal chicanery, with the end result that fiscal policy has become less transparent to voters. Finally, in the third section, we examine the origins of some of these trends. Specifically, we review how fiscal institutions, such as tax and expenditure limits, supermajority limits, balanced budget amendments, and rainy day funds, have affected fiscal outcomes in the American states. SECTION 1: STATE AND LOCAL GOVERNMENT EXPENDITURES
In this section we review trends in state and local government expenditures over the last thirty years. We first consider the growth rate in expenditures by state and local governments vis- à- vis the federal government. We will also show that Arizona‟ s state and local government growth rate is higher than national trendlines. We then turn to a discussion of the balance of expenditures between state and local governments and detail the rise of special districts, assessment districts and other new forms of local sub- government. Finally, we review the changes that have occurred in spending across functional categories for state and local governments. We make particular note of the rising costs of public pensions and retiree healthcare. All estimates provided herein are in constant dollars.
Per- capita Expenditures
Figure 1 presents per- capita direct general expenditures nationwide and in the state of Arizona. As shown in this figure, general expenditures per- capita have increased in constant dollars. Whereas, nationwide, state and local governments spent only $ 3,345 per person in 1977 ( a little over twenty one percent of the median income), they now spend over $ 7,000 per person ( over 31 percent of the median income). In the late 1980s through 1990, Arizona spent more per- capita than nationwide average for per- capita state and local spending. This trend ended in 1991. Currently, Arizona spends $ 1,306 less per person than the nationwide average.
21 Figure 1: State and Local Expenditures Per- Capita, Arizona and Nationwide
Expenditure Growth Rates: State, Local, and Federal Governments
With only a few exceptions, growth rates for all governments have been positive over the last thirty years, even after adjusting for inflation. Noticeably in Figure 2, expenditure growth rates have fluctuated cyclically with the economy. In the 1970s, growth of federal, state, and local expenditures followed very similar patterns: growth rates dropped in the early seventies, peaked in 1975, and then fell together with the oil crisis of the late seventies. Local governments experienced the slowest growth of all types of government from the mid seventies to mid eighties, experiencing something rarely seen in American government, negative growth rates between 1979 and 1981.
Figure 2: Expenditure Growth Rates of Federal, State, and Local Governments
22 From the mid eighties through the year 2000, however, both state and local government expenditure growth rates far exceeded that of the federal government. Whereas the federal government‟ s average growth rate was only 1.0% per year during the decade of the 1990s, state and local expenditures grew by more than three times that rate, averaging, 3.7% and 3.0% respectively. State and local growth rates were even higher in Arizona, where the average growth rate for the state government was 4.5% and local expenditure growth was 3.3%. Federal government expenditure growth rates changed with the election of the Bush administration and the events of September 11, 2001, where- after the federal government growth rate far exceeded that of state and local governments through at least 2005.
Figure 3: Expenditure Growth Rates of Federal, State, and Local Governments
Figure 3 compares Arizona‟ s state and local government expenditure growth rate to state and local governments nationwide. Not only did Arizona have a higher than average growth rate, but in fact it had the highest average growth rate of any state over the last thirty years, followed by Florida and Nevada. Each of these states had average growth rates of over 5% over the last thirty years. It should be noted, however that these states have also experienced substantial population growth during these decades. Accounting for population, Arizona actually has the 9th lowest per- capita expenditure growth rates among states. By, contrast North and South Carolina experienced the largest average growth in their per- capita state and local expenditures. Each of these states has experienced an average growth of over 3% in terms of per- capita expenditures adjusted for inflation over the last thirty years. North and South Carolina are also two of the top three states, accompanied by Ohio, with the largest growth rates relative to increases in state personal income in the last thirty years.
23 Who Controls? State Versus Local Governments
Total local government general expenditures have historically surpassed state general expenditures, though this trend appears to be ending. In 1970, 63% of all general expenditures were by local governments. Today this margin has decreased to 57%. Forty- four of the fifty US states have moved in the direction of more state spending in the last thirty years. This trend holds in the state of Arizona, where local governments accounted for 67% of all general expenditures in 1977 compared to 59% in 2006. It is important to note that these calculations exclude non- general expenditures, namely those for utilities, liquor stores, and, most importantly, employee retirement. Including these non- general expenditures would show that balance between local and state spending has shifted even more significantly towards state governments in the last thirty years.
The state with the largest reliance on local expenditures is Nevada, followed closely by California. Today 66% of all state and local general expenditures are spent by local governments in Nevada. By contrast, Hawaii is the state that most relies on state government expenditures. Today, 81% of all state and local expenditures are spent by the state government in Hawaii. Local Governments and the Rise of Special Districts and Special Assessments
Though local governments appear to be in decline relative to state governments, it is noteworthy that spending as reported by the Bureau of the Census for local government often hides spending by special districts, business improvement districts, maintenance assessment districts, among others, whose revenue and the associated uses are not always reported as such by local governments. There are four major categories of local governments: counties, municipalities and townships, school districts, and special districts. Whereas municipalities and townships appear to be on the decline as a proportion of total local government expenditures nationwide, expenditures by special districts and counties are on the rise. However, neither of these nationwide trends is evident in the state of Arizona
The mix of local governments utilized varies significantly across states. In general, Eastern states rely primarily on municipal and township governments. For the last thirty years, municipal and township expenditures have comprised approximately 60% of all local expenditures in Eastern states. This is a significant difference from all other regions of the country where municipal and township expenditures have historically comprised 20 to 30% of all local expenditures. In contrast, county governments are relatively more important in Southern states than in other regions of the country. On average, 30% of Southern state and local expenditures are by county governments. Arizona is typical of Western states, with the largest share of local expenditures coming from school districts, followed by municipalities.
Figure 4 displays shares of total local government spending in the U. S. by who controls the spending. Figure 5 provides the same graphic for the state of Arizona. As noted previously, the greatest change within local governments nationwide over the last thirty years is the decline of expenditures by municipalities and townships. Municipalities and townships comprised 36 percent of all local government expenditures in 1977 and only 32 percent of local government expenditures in 2006. This trend, however, does not hold for the state of Arizona. Over our time series, municipal spending has actually jumped in Arizona, from 25% to 33% of all local government spending.
24 Nationwide, we have also seen the ascendancy of special districts. According to the latest census data available, the number of special districts has increased from 25,987 in 1977 to 37,381 in 2007 ( these do not include assessment districts, business improvement districts and the like). Special district expenditures now make up 11.6% of all local general expenditures nationwide, up from 7.2% in 2002 and 5.3% in 1977. Special districts now serve many public service functions once provided by municipalities, such as fire protection, park maintenance, rubbish collection, lighting and so on.
This fiscal innovation has not yet been widely employed in the state of Arizona. Currently, as far as we can tell from compiled data sources, special districts comprise less than 2% of local government expenditures. Indeed, Arizona ranks 40th among the states in number of local governments. Arizona had only 645 local governments ( 301 special district governments) as of October 2007. In comparison, Washington, a state with a very similar population size, has 1,845 local governments and 1,229 special districts.
Figure 4: Share of Total Local Direct General Expenditures by type of Local Government Nationwide
Figure 5: Share of Total Local Direct General Expenditures by type of Local Government in Arizona
25 State and Local Spending: Trends by Spending Category
This section explores state and local spending trends by functional category. As we show in this section, not all categories of spending have grown at the same rate. Figure 6 shows several of the major trends of state and local government spending by functional category. Each category will be discussed in turn. In what follows we will compare these national trends to trends in the state of Arizona ( in Fig. 7)
Figure 6: Cumulative State and Local Government Expenditures by Functional Category
Figure 7: Cumulative State and Local Government Expenditures by Functional Category for the State of Arizona
26 Education
Figure 6 shows that the largest functional category of state and local spending, education, has experienced relatively less growth in the last thirty years compared to most other functional categories. Indeed, average elementary and secondary education expenditures are on the decline as a proportion of all state and local general expenditures. Whereas 1977 expenditures on elementary and secondary education comprised more than 26% of all general state and local expenditures, this figure has fallen to 22% nationally. Arizona has experienced a similar decline, though the magnitude of spending as a share of general revenues is far larger than nationwide estimates. Arizona spent 45% of its general expenditures on education in 1977 compared to only 33% today.
Declines in education as a proportion of all direct general expenditures are explained in part by a relative decline in the number of school age children as a fraction of the population. 1 Nationwide, the percentage of school- age children of the total population was 29% in 1970 compared to 20% today. Similarly, in Arizona, school- age children comprised 31% of the population in 1970 but only 21% today. The decrease in the proportional number of school age children explains, in part, the decreased spending on education as a proportion of general expenditures.
When education expenditures are calculated on a per- child basis, with school- age children defined as those aged 5- 19, a noticeable upturn in state and local expenditures on elementary and secondary education over the last thirty years becomes evident. The increase is strongest in the Eastern states, the region that has seen the largest relative decline in the population of school- age children. In 1977, the average Eastern state spent $ 4,000 ( 2006 dollars) per child on elementary and secondary education. In 2006, by contrast, this figure is now closer to $ 10,000 per child. Arizona‟ s gains in spending per- child is just as striking. Arizona spent $ 6,765 per child in 1977 compared to $ 9,061 per child in 2006.
Figure 8: Average State and Local Expenditures per Child on Elementary and Secondary Education, By Region
Often, political debates center on whether state and local governments spend too much or too little on education today. Some people argue that schools are fraught with waste and that spending and learning are unrelated. People on the other side of the issue bemoan overcrowded classrooms, crumbling buildings and out
1 A decline in the proportion of students in the population, however, does not explain the relative decrease in state and local expenditures for higher education. Expenditures on higher education have decreased from 9.5% to 7.7% of all general expenditures from 1977 to 2006. This latter trend is surprising given increases in university enrollment and tuition costs ( Callan 2002.)
27 of date textbooks. What is missing from this debate, however, is the biggest financial crisis in the education system, the looming costs of teacher pension and retiree health care. The question is not are we spending the right amount, but rather are we spending our money on the things related to education or has our education system just become another very large pension and healthcare plan?
While state programs cover most public school employee pensions, many school districts supplement these benefits with their own retirement plan. Currently, very little is known about these school district liabilities for pensions and retiree healthcare coverage. While several studies are currently underway, one by Pew and another by the National Center for Education Statistics, to date we have no reliable estimate of these costs nationwide. The best estimates we have, based on only a subsection of school districts, are not promising. The state of California, for example, has found that school districts across the state have a retiree healthcare liability of at least $ 15.9 billion. Most startlingly, this estimate accounts for only half of the 1,036 Californian school districts ( McNeil 2008). Los Angeles schools alone are estimated to have an unfunded health- care liability of $ 10 billion.
Given these estimates, and limited data availability, it is impossible to know the amount per pupil that is currently, and in the future will be, spent on retiree benefits nationwide by American school- districts. As noted by Michael Podgursky, a leading expert on teacher- retirement issues, " These benefits are sucking dollars from the classroom" ( McNeil 2008). Arizona‟ s outlook, in this respect, is rosier than the rest of the country. According to the latest census data, no Arizona school districts have independent public pension funds.
Highways
Another functional category noticeably in decline as a share of total general state and local spending is highways. In 1977, highway expenditures accounted for 8.5% of all state and local general expenditures nationwide. Today this percentage has fallen to 5.4%. The fall is even more dramatic considering that highways expenditures were once 20% of all state expenditures ( Garrett and Wagner 2004).
The decline in highway spending is most noticeable in Southern and Western states. In 1977, the average Western state spent more than twice as much on highways than all health and public safety related expenditures combined. Today, the average Western state spends less on highways than on either health or public safety individually. Arizona is a good example of this trend, 10% of state and local general expenditures were allocated to highways in 1977 compared to only 7% today.
There is substantial evidence that declining expenditures on highways has damaged public infrastructure. The 2006 report of the Federal Highway Administration of the US Department of Transportation reveals several substantial problems. For example, of the nearly 600,000 bridges across the US, the Department of Transportation classified 13.1%, over 70,000 bridges, as structurally deficient. This list included the I- 35W bridge in Minneapolis that collapsed in 2007, killing 13 people and injuring 145. An additional 13.9% of all bridges were classified as functionally obsolete. ( Federal Highway Administration 2006). Arizona‟ s bridges have so- far fared better in reviews by the Federal Highway Administration. Only 1.5% ( 33 total) bridges were classified as structurally deficient in the administration‟ s last report.
Roadways have received similarly low grades nationwide. In the same report, the Department of Transportation classified less than half ( 44.2%) of all travel on roads for which data are available occurred on pavements with " good" ride quality. Fifteen percent of travel occurred on surfaces that the department considered unacceptable. As noted in the report, “ Poor road surfaces impose costs on the traveling public in the form of increased wear and tear on vehicle suspensions and tires, delays associated with vehicles slowing to avoid potholes, and crashes resulting from unexpected changes in surface conditions.” Specifically, the Federal Highway Administration estimates that outdated and substandard road and bridge design, pavement conditions, and safety features are factors in 30% of all fatal highway accidents. ( Federal Highway Administration 2006).
28 The Department of Transportation has also identified congestion as a major problem facing today‟ s highways. The typical duration of congested travel conditions in urbanized areas is currently 6.6 hours per day, up from 5.9 hours per day in 1995. The department estimates that the average driver commuting during hours of peak traffic would be stuck in traffic for the equivalent of more than 45 hours per year. Congestion problems will be particularly problematic in the future if funding is kept at a level where only maintenance, as opposed to improvements, is feasible.
Similar congestion problems have been reported in Arizona. For example, the annual Urban Mobility report by the Texas Transportation Institute suggests that the percentage of travel during peak hours that occurs in congested conditions in Phoenix is currently 68%, up from 43% in 1982. This report also estimates that congestion costs a peak- hour commuter an additional $ 1,034 per year. The traffic problem is similar in Tucson. Only 29% of peak vehicle- mile travel was congested in 1982, compared to 61% now. The average Tucson peak- hour commuter spends an additional $ 923 per year as a result of traffic.
Welfare
Another major shift in state and local government expenditure over the last thirty years is in the category of welfare. Welfare herein is defined as any cash payment made directly to an individual contingent upon their need. This includes expenditures associated with federal programs such as supplemental security income ( SSI), temporary assistance for needy families ( TANF) ( previously Aid to Families with Dependent Children), and the Medical Assistance Program ( Medicaid). In 1977, welfare spending comprised 12.6% of all state and local general expenditures. This category of spending has risen as a proportion of all general expenditures to 17.5%. This growth has been even more substantial in Arizona. Arizona only spent 4.3% of general expenditures on welfare in 1977, compared to 17% today.
It is important to reconcile two diverging trends within the category of welfare spending. This divergence is between the growth of Medicaid and the decline of cash assistance programs. A study by the Institute for Federalism and Intergovernmental Relations puts these trends into perspective ( Marton and Wildasin 2007A). This report shows that annual Medicaid expenditures as a share of annual state total expenditures increased from 10% to 16% from 1988 to 2003. In contrast, annual cash welfare expenditures ( such as those from TANF or SSI) decreased from 1.9% of total state expenditures to 1.2% of total state expenditures in that same time frame.
Additional research suggests that the diverging trends are explained in part by differences in intergovernmental funding for these programs. Prior to the 1996 welfare reforms, both Medicaid and cash assistance programs were funded with open ended matching grants by the federal government. Since 1996, however, matching funding for cash assistance was replaced with a system of federal block grants. It is argued that this change in funding has prompted states to substitute cash assistance for medical assistance. For example, Marton and Wildasin ( 2007B) argue that, because TANF benefits are no longer subsidized by the federal government on the margin, states have an incentive to increase matched Medicaid spending instead of the unmatched TANF program. ( Marton and Wildasin 2007B). In addition to differences in funding sources, increasing Medicaid expenditures are also the result of increasing health costs, the topic to which we now turn.
29 Health and Hospitals
Despite the present rhetoric on the subject, according to census figures, state and local spending on health and hospital- related expenditures has remained relatively constant over the last thirty years. This is a surprising finding given increases in medical care costs. Spending on health and hospitals accounted for 8.4% of all state and local general expenditures across the U. S. in 1977 compared to 8.5% today according to the most recent estimates available. The census definition for expenditures on health and hospitals includes government expenditures on hospitals and expenditures for general health activities, categorical health activities and programs, health- related inspections, community health care programs, regulation of air and water quality, rabies and animal control, and ambulance and emergency medical services.
These no- growth estimates can be explained in part by the fact that the census estimates encapsulate only a small proportion of all health- related expenditures by state and local governments. Most notably, these estimates exclude Medicaid expenditures as well as health- related expenditures for state employees, public university students, and the incarcerated. From 1998 through 2003 the National Association for Budget Officers ( NASBO) provided a report that compiled all direct personal health expenditures. NASBO totaled the amount that state governments spent on everything related to public health. The growth of direct personal health expenditures in the five years of this study is significant. In 1998 health related expenditures comprised 27.1% of total state spending. In fiscal year 2003, this figure grew to 31.5% of state spending ( NASBO reports on public health expenditures 1998, 2003).
Even including all health related expenditures, state and local payments towards healthcare pale in comparison to the expanding payments by private individuals and the federal government. Figure 9 puts growth in government health and hospital spending into perspective. Figure 9 uses data from the United States Department of Health and Human Services to show the relative growth of health expenditures by the federal government, state and local governments, and non- government sources. Non- government health expenditure refers to all health- related expenditures by private health insurance companies as well as out- of- pocket expenses by individuals. As is immediately evident in Figure 9, the growth of state and local expenditures on health is small compared to private and federal expenditure growth.
Figure 9: Health Expenditures by Source Nationwide
30 The National Health Statistics Group has compiled state- by- state statistics on health- care related expenditures. Their measurement includes all spending for privately and publicly funded personal health care services and products ( hospital care, physician services, nursing home care, prescription drugs, etc.). Arizona stands out in their ranking. As of 2004, Arizona had the second lowest health- care expenses per- capita, outspending only the state of Utah. They estimate that Arizonans spend $ 4,103 per person on healthcare expenditures. In comparison, public and private health care spending in Massachusetts totals $ 6,683.
Public Safety
Another important functional category of state and local expenditures is public safety. As shown in Figure 10, spending for public safety is on the rise. Police funding grew most rapidly in the mid- 70s through the 1990s. Declining crime rates in the nineties seem to have led to a plateau in the new millennium. These trends are very similar in the state of Arizona ( Figure 11)
Figure 10: Total State and Local Expenditures on Public Safety Nationwide
Figure 11: Total State and Local Expenditures on Public Safety in Arizona
31
By contrast, expenditures on corrections have continued to rise. Increased expenditures on corrections are not surprising given the increase of the prison population. A recent report by the Pew Center on the States shows that the prison population in the US has tripled since 1987. For the first time in history more than one in every 100 adults is now confined in American jails or prisons ( Pew 2008). In addition, the aging of the prison population is increasing the amount of health- related expenditures in prisons. A report by the Council of State Governments showed that from 1998 to 2001 health- care spending in state prisons grew 10% annually. As of 2003, Arizona spends $ 82.6 million on health care for its prison population ( NASBO 2003).
Arizona‟ s spending on public safety is higher than the national average. For example, nationwide, $ 379 is spent on police and fire protection per- capita, compared to $ 406 per- capita in the state of Arizona. Similarly, $ 210 per- capita is spent on corrections nationwide, compared to $ 232 in Arizona. The higher expenditures in Arizona may in part be explained by employee retirement pension guarantees. In Arizona, public safety employees retire under the auspices of the Public Safety Personnel Retirement System. This system had 82.1% of its liabilities funded in 2007, but only 68.9% in 2007. Unfunded liabilities will cause expenditures to continue to rise in the category of public safety.
Employee Retirement
The major category of spending missing ( it is missing from both the general revenue and general expenditure columns) is spending on employee retirement. Pension obligations are at the heart of many discussions on state finances and many government officials and academics alike view these obligations as an impending crisis. Figure 12 shows the growth rates of employee benefit expenditures and general expenditures in constant dollars. As is immediately evident from Figure 12, employee retirement expenditures are growing at rates much higher than the general budget. Pension expenditure growth is regularly above 5%, which is realistically still too low to cover future liabilities ( and the assumptions made in retirement plans project future rates of return that are unrealistically optimistic, implying that more reasonable assumptions would cause the deficits to mushroom beyond comprehension). Growth in expenditures on employee retirement benefits has been even higher in the state of Arizona. In the last thirty years, the average growth rate in this state is 9.8%.
Figure 12: Smoothed Growth Rates of Employee Retirement Benefit Payments
Versus Direct General Expenditures
32 One of the most worrisome issues today regarding future employee retirement spending concerns “ Other Post- Employment Benefits”, namely health- care coverage for retirees. It was only recently, 2006, that new rules from the Government Accounting Standards Board ( GSAB) went into effect calling for state and local governments to report future healthcare liabilities ( Hyde 2008). In its most recent report, California estimated that post- employment benefit liabilities over the next thirty years would total $ 118 Billion ( California Public Post- Retirement Benefit Commission 2007).
Arizona‟ s state public employee retirement system has been lauded for its high degree of funding. In the last actuarial evaluation, in June of 2008, it was reported that the retirement‟ s system future liabilities were currently funded at 102%. Despite this high level of funding, other research on Arizona has suggested that “ Aggregate statistics can mask the fiscal weakness of individual plans because assets in well- funded plans are not transferable to underfunded plans” ( Lahey and Anenson 2007).
Summary and Discussion
This section has reviewed the major trends in state and local government expenditures. We have shown that local governments have historically spent more than state governments, though the tides seem to be turning. We have also shown that different types of local governments have different expenditure trends. While municipal and township expenditures are on the decline, there has been a noticeable upswing in special district expenditures. The creation of new districts, most likely at the expense of municipal departments, largely explains this trend. Recent research has suggested that these districts have significant policy and democratic deficiencies and should be the subject of future research ( Burns 1994; Berry 2009; Kogan and McCubbins 2009). Arizona has not partaken in this national trend, and instead is increasing reliance on municipalities.
Finally, we have also reviewed shifts in the functional categories of state and local spending. We have shown that highways and cash assistance programs have declined in priority, whereas expenditures on corrections and Medicaid are rising. Additional health- related expenditures are also on the rise, but growth in this category has not yet rivaled the growth seen in federal government and personal expenditures. Finally, payments for retirement benefits, particularly heath insurance for retirees, loom as a significant expense for state and local governments in the future. This is particularly true for school districts, whose future liabilities have not yet been accurately estimated for most states.
SECTION 2: STATE AND LOCAL REVENUES AND DEBT
This section reviews trends in state and local revenue collection and borrowing practices. In a comparative perspective, state and local revenue collection is quite substantial. State governments in the US are more reliant on own source revenue than any other sub- national governments in the world, with the sole exception of Canadian provinces.
33 We start this section by reviewing relative levels of state, local, and federal revenues as a proportion of personal income. We show that state governments collect more revenue than local governments if employee retirement contributions are included in total revenues. We then look at trends in sources of revenue in state and local governments. We also show that there has been a sea change in how state and local governments raise revenues. Whereas property taxes were once the primary source of revenue, state and local governments now rely more on charges and fees for services, essentially privatizing public services. We look at this trend in detail, examining the categories of services for which charges and fees are collected. We argue that new fiscal practices are less transparent than their predecessors and are often imposed through undemocratic practices. This leads us to agree with Berry ( 2009) that a Tiebout sorting of residents, where people choose among different locations within a state on the basis of the package of taxes and benefits delivered by the government, does not arise ( Tiebout 1956). Rather the stacking of dozens of overlapping governments with a mixed bag of taxes, charges, fees, and with a mixture of services, not directly tied to those revenues, makes the Tiebout sorting a difficult, if not impossible, puzzle. Consequently, the distribution of taxes and services by the many layers of government in any locale may be far from optimal.
Finally, we conclude by analyzing trends in state and local borrowing. We show that state and local governments are more likely to borrow using non- guaranteed debt, or engage in tricks to hide debt by, for example, leasing government infrastructure instead of purchasing it outright ( or worse, by selling government infrastructure and then leasing it back, thus having the tax- payers unwittingly pay for the same infrastructure twice, while new infrastructure projects go wanting) or by “ borrowing” from pensions or retiree healthcare funds by under- funding them ( in part by assuming unrealistic future rates of return for the funds).
State, Local, and Federal Revenues as a Proportion of Personal Income
Figure 13 shows federal, state, and local total revenues as a percentage of personal income. This proportion is a useful gauge of the total size of government. Since the mid- seventies, federal revenues have ranged from 19% to 24% of personal income. In contrast, local revenues have ranged from 6.8% to 8.5% as a percentage of personal income nationwide and, by comparison, 12.8% to 15.7% in the state of Arizona. State government revenues have ranged from 8.5% to 12.4% nationwide and 11.1% to 14.6% in the state of Arizona.
The recent, noticeable dip in state government revenues is primarily the result of decreased revenues from investments on employee retirement contributions. Whereas the states received $ 225 Billion ( 2006 Dollars) from earnings on investment for employee retirement contributions in the year 2000, those accounts actually lost over $ 70 Billion in 2002 due to the dot com market crash. Arizona‟ s losses in 2002 were mainly by local governments. While the state of the Arizona had positive returns, local governments lost $ 53 million in insurance trust revenues.
Figure 13: Federal, State, and Local Revenues as a Proportion of Personal Income 34 Putting aside the effect of investments, there are several other take- away points evident from Figure 13. First, federal revenues have been more volatile than state and local revenues over this series. The Reagan tax cuts of the 1980s noticeably decreased revenues as a proportion of personal income, with a rise only occurring with the start of the Clinton administration in 1993. In contrast, both state and local governments have shown mostly monotonic increases. State government revenue comprised 9.6% of personal income at the beginning of the series, compared to 12.4% at the end of the series. The rise of local government revenues has been less substantial, but nonetheless positive, from 7.8% to 8.5%.
Figure 14: Components of State and Local General Revenue
Figure 14 decomposes state and local government revenues by source category nationwide. The composition of state and local general revenues has undergone significant changes in the last fifty years. In 1963, approximately 32% of all general revenues where collected via property taxes, 23% by sales taxes, 7% from income taxes, 14% from federal aid, and 23% from charges, fees, and miscellaneous sources. Over time, we have seen an increase in the importance of federal aid, charges and fees, and income taxes, and a decrease in the relative importance of property taxes and sales taxes. These national trends hold in the state of Arizona ( Figure 15).
Figure 15: Components of State and Local General Revenue in Arizona
35 Property Taxes
The decline in the use of property taxes is one of the most significant trends in state and local government finances of the past century. Prior to the 1930s, property taxes accounted for almost half of state government revenue. ( Brunori 2005). Compare this to the year 2005, when property taxes accounted for only 16% of general revenues. Analysts posit that the decline in property taxes is a consequence of tax revolts, tax incentives, tax exemptions, and school financing controversies that have occurred over the last thirty years ( Brunori p. 815). Several scholars have also hypothesized that Supreme Court rulings that enforce equalizing school finances is the ultimate source of these revolts. They speculate that these rulings have decreased public willingness to pay property taxes, since they no longer see the benefits in their own local schools. ( Fischel 1989, but also see Martin 2006).
Figure 16: Average State Components of General Revenue, By Region
Figure 16 displays the composition of general revenues using state averages categorized by region. The decline in property taxes is evident in every region except the South. The other three regions have compensated for falling revenues with increased charges, fees, and miscellaneous revenues. Charges and fees are now the leading source of revenue for most states, though the average Eastern state still relies more on property taxes than is typical in other regions of the country.
There is, however, wide variation in reliance on property taxes between states. The state with the highest reliance on property taxes is New Hampshire. In 2006, 34% of all New Hampshire general revenues were derived from property taxes. New Hampshire is followed by New Jersey and Connecticut, which collected 28% and 26% of their general revenues through property taxes respectively. The states with the lowest reliance on property taxes are New Mexico and Alabama, both with only 6% of general revenues arising from property taxes. Arizona falls somewhere in- between, with 15% percent of general revenues coming from property taxes in 2006 ( down from 26% percent in 1977).
36 Income Taxes
As both Figures 14 and 16 show, revenues derived from income taxes have increased over the last thirty years nationwide. Currently, 41 states collect personal income taxes. Of these states, eleven adopted the personal income tax within the last fifty years. The increased number of states collecting income taxes is the most obvious reason why income taxes now comprise a larger share of all state and local general revenues, most recently 14%. Currently, Wyoming, Washington, Texas, and Nevada collect no income taxes, including no corporate income taxes. The states with the highest proportional reliance on income taxes are Maryland, Massachusetts, and Oregon.
Arizona has had both personal and corporate income taxes since 1933. There has been only a modest increase in Arizona‟ s reliance on income taxes over the last thirty years. In 1977, 8.3% of general revenues came from income taxes, as compared to 11.2% in 2006. Generally, Arizona is characterized as having a lower income tax burden relative to the other 40 states that collect income taxes.
Most income taxes collected by states governments are personal income taxes, as opposed to corporate income taxes. Figure 17 shows personal income taxes as a proportion of total income taxes for both state governments and the federal government. Both types of governments have increased their reliance on personal income taxes relative to corporate income taxes over the last thirty years. For most of this series, with the significant exception of most of the 1980s, states governments relied on personal income taxes to a slightly larger extent than the federal government. This is reflective of the fact that states compete for businesses using incentives of lower corporate taxes. In 2005, 85% of all state income taxes were derived from the personal income tax. Arizona's reliance on personal income taxes has changed very little, at least since 1977. In both 1977 and 2006, 79% of Arizona's total income taxes collected were from the personal income tax.
Figure 17: Percentage of Income Tax from Personal Sources
37 Charges and Fees
We turn now to a discussion of changes that have occurred in the largest category of revenue composition, the “ other” category. The largest components of this " other" category are general charges and fees, licensing taxes, and miscellaneous revenues. In 2006, charges and fees accounted for more than half of this category, miscellaneous revenues approximately one- third, and licensing taxes approximately one- tenth. Bowler and Donovan ( 1995) have suggested that this switch to using charges and fees is an attempt to hide the true cost of government. Further, some public opinion polls have shown surprising support for user fees, at least in comparison to taxes ( Matsusaka 2004). We will focus first on the largest category, general charges and fees.
There is wide variation between states in terms of reliance on charges, fees, and miscellaneous revenue. Here Connecticut is noteworthy, as it has the lowest reliance on charges, fees, and miscellaneous revenues at 15% of general revenues. Indeed, the seven states with the lowest proportional reliance on charges and fees are all in the northeastern part of the United States. The state with the highest proportional reliance in charges and fees in 2006 was Alaska, with 44% of all general revenues coming from this category, primarily through charges to the oil industry. Indiana and South Carolina follow Alaska, both relying on charges, fees, and miscellaneous revenues for 33% of their general revenues. In Arizona, 21% of general revenues come from charges and fees ( up from 16% in 1977).
Figure 18 displays the increase in charges and fees for all state and local governments over the last thirty years. It is important to note that the charges and fees displayed in this figure are only general revenue charges and fees. Non- general revenue, such as most associated with utilities or liquor stores are excluded. In 2006, liquor store and utility revenue for all state and local governments was over $ 114 Billion.
Figure 18: Cumulative Representation of the components of general revenue charges and fees
38 Figure 18 presents the major categories of charges and fees in cumulative form. The category at the bottom on the chart, other education, represents the charges and fees collected by elementary and secondary schools, primarily through school lunch programs. There has been very little movement in this category over time. In contrast, charges associated with higher education ( tuition) have increased significantly over the last thirty years in real dollars. Revenue associated with higher education tuition now approaches $ 100 Billion dollars annually.
The category with the second largest increase in fee collection is state and local hospitals. This category includes charges from patients, private insurance companies, and public insurance programs to public hospitals as well as receipts of hospital cafeterias and gift shops. These charges have almost quadrupled over the last thirty years in terms of real dollars. The take- home point from Figure 18 is that most categories of charges of fees have increased over the last thirty years.
Most charges and fees in the state of Arizona are derived from education. Most significantly, tuition from institutions of higher education comprises 33% of all general charges and fees collected. Waste water disposal provides 12% of all charges and fees, along with an additional 8% from solid waste management. Hospitals collect 12% of all general charges and fees. Similar to the nationwide figures provided, these figures do not include non- general charges, such as those associated with utilities. Arizona collects fewer fees than national average most notably in the category of highways.
Though licensing fees comprise significantly less general revenue than charges and fees, there are changes within this category that are worth noting. Trends in state and local licensing taxes appear in Figure 19. The two largest components of licensing fees are corporate licensing and motor vehicle licensing. Each of these categories experienced growth over the last thirty years, but wane in comparison to the changes in the “ other” licensing category. The miscellaneous category of licensing is composed of various permits, such as those for fishing or hunting, marriage licenses, or impact fees paid in connection with issuance of building permits. From fiscal year 2004 to fiscal year 2005, this category almost doubled from $ 14 Billion to $ 28 Billion. This was an area of exceptional growth in Arizona, where other license taxes increased from $ 100 million to over $ 400 million.
Figure 19: State and Local License Taxes
39
The final category of “ other” revenues is a miscellaneous category. The census uses this category to measure revenues from special assessments, interest earnings other than employee retirement funds, private donations, royalties, sale of property, fines and forfeitures, lottery revenue, and other miscellaneous sources such as the tobacco lawsuit settlements. Again, these miscellaneous sources comprise approximately one- third of the total “ other” category in Figure 14.
Special assessments are a particularly intriguing new revenue source. A recent paper by Vladamir Kogan and Mathew McCubbins ( 2009) show that special assessments are vastly underreported to the Census and state governments. Special assessments bill property owners for public goods and services in proportion to the " special benefits" that they receive. While these charges appear on individual‟ s property tax bills, they are not technically property taxes because their amount is not tied to the value of an individual‟ s home. The authors‟ contention is that special assessments pose democratic problems since these fees are adopted by the vote of property owners ( who may not be citizens and indeed, may not even be people, in that land can be owned by corporations), not all citizens, and their administration is far from transparent. A review of a selection of property tax bills from the state of Arizona suggests that special assessments are not yet widely used in this state.
Sales Tax
Sales taxes have decreased as a proportion of total general revenues for state governments ( from 29% of all general revenues in 1977 to 21% today) but, in compensation, have risen in relative importance for local governments ( from 4% in 1977 to 6% today). This national trend has been even more noticeable in state of Arizona. State governments, in 1977, were once highly reliant on sales tax. In 1977, 41% of all general revenues came from this source. State reliance on sales taxes has declined since then, dropping to 29% of all state general revenues. In contrast, local governments have increased their reliance on sales tax. In 1977, 6.5% of all local government general revenues were collected from sales tax. Today, this proportion has risen to 11%. Today, Arizona local governments are substantially more reliant on local sales taxes than most states, ranking as the 11th highest.
Though sales tax represents only a small proportion of all local general revenues overall, recent research suggests that incentives to collect additional sales tax dollars have significantly shaped land- use choices by local governments. In a hypothesis known as the “ fiscalization of land- use planning”, scholars have suggested that local governments plan their cities with an eye to short- term sales tax windfalls, perhaps at the expense of long- term fiscal health.
Most research on the fiscalization of land- use planning has focused on the state of California. As we will discuss later in this report, researchers have made connections between the adoption of fiscal institutions – such as California‟ s limitation on property taxes, Proposition 13 – and changing fiscal practices. For example, Schwartz ( 1997) connects the adoption of Proposition13 to land- use practices that favor shopping centers, car dealerships, and large- scale discount retailers over smaller “ mom and pop” businesses. Lewis ( 2001) extends this research, in a survey of city managers, finding that “ the quest for retail development and sales tax revenues is a prime motivation for land- use decisions” ( Lewis 2001).
40 Debt
Borrowing is an important source of money for all governments, and most businesses and individuals. In the US, popular commentators such as the New York Times‟ David Brooks have bemoaned the “ culture of debt”. Figure 20 compares the relative outstanding debts of governments, individuals, and businesses using data from the Federal Reserve. As evident in the figure, debt from personal home mortgages has risen exponentially in the last thirty years. It is now the single largest source of indebtedness in this country. Coupled with debt from personal lines of credit, it becomes clear that US personal debt in enormous.
Figure 20: Debt Outstanding, By Sector
Surprisingly, the sector with the smallest amount of debt outstanding is state and local governments. That said, when debt outstanding is calculated as a proportion of total general revenues, the fiscal inequality above looks somewhat bleaker. In general, local governments borrow to a greater extant than state governments, as a proportion of their general revenues. Figure 21 shows the amount of long- term debt outstanding in all state and local governments as a proportion of general revenue. As noticeable from Figure 21, cumulatively local governments have owed more than a single year‟ s general revenue since 1986. In contrast, state governments owe just slightly more than half of their cumulative general revenue, a roughly constant trend over our time series. We have also plotted Arizona‟ s total state and local debt as a proportion of general revenue, which was strikingly high in the late eighties.
Clingermayer and Wood ( 1995) have shown that state government indebtedness is primarily a function of economic conditions. Fiscal strain affects both a state's need to borrow as well as its capacity to repay. In addition, politicians are also more likely to borrow when interest rates are low, or when borrowing could be self- beneficial ( such as before a competitive election). Clingermayer and Wood also show that Tax and Expenditure Limits are positively associated with indebtedness, a topic to which we will return in the next section.
41
Figure 21: State and Local Long Term Debt Outstanding as a Percentage of General Revenue
More noticeable than changes in the debt financing over the last thirty years are changes in the type of debt accrued. Figure 22 shows that there was a significant increase in the usage of non- guaranteed debt, that which is not backed by the full faith and credit of the government. The upswing in non- guaranteed debt is particularly noticeable through 1987. Thereafter, local government returned to greater general obligation borrowing whereas states continued to borrow without guarantee. Currently, more than 70% of state outstanding debt is of the non- guaranteed variety.
Figure 22: Proportion of Long Term Debt that is Non- Guaranteed
42 This picture of state and local debt is, if anything, rosier than reality. One important form of debt not included in this estimates is debt in the form of tax- exempt leases. Tax- exempt leasing has been called the " quiet revolution" of local financing ( California Debt Advisory Commission 1991 Report). Instead of falling into long- term debt on capital investments, localities can opt to lease instead of own. Tax- exempt leases are commonly used to finance both minor equipment needs ( fire- trucks, agency office supplies) but also major capital outlays ( school buildings, courthouses, and utility assets). Often, tax- exempt leases are set up so that locality can buy the property for a nominal fee, often $ 1, at the end of a contracted lease term.
The benefit of tax exempt leasing is that they avoid the technical classification of debt, and are therefore not restricted by constitutional debt limitations. Tax- exempt leases avoid classification as debt because they do not technically incur a long- term obligation on the government lessee. If the local government were simply to choose not to appropriate lease payments in a given fiscal year, the lease obligation becomes null and void. Further, if the property was somehow destroyed, the lease obligation is also nullified. Under such arrangements, future obligations of lease payments are not legally binding and are therefore not officially classified as debt. Despite this legal arrangement providing for cancellation, however, the cancellation of leases is very rare. One reason for this is that lease contracts often having non- substitution clauses. For instance, if a municipality wanted to cancel its lease of fire trucks before full payment, it would not be able to buy its own fire trucks
Although future lease payments are not reported as debt, the contract is very similar to borrowing with bonds. As noted by the California Debt Advisory Commission, " there should be no confusion as to the essence of the transaction: a government agency is borrowing funds from investors to finance the construction or acquisition of a capital asset. The funds borrowed through the issuance of tax- exempt lease obligations must be repaid in regular installments of principal and interest, just like the funds borrowed through the issuance of municipal bonds." That said, lease payments are unidentifiable by Census estimates because their payments are reported only as capital expenses. As such, no estimates exist nationwide on the true level of state and local debt. This lack of transparency, and evasion of debt limitation requirements in many states, is another example of the democratic deficiencies of many new fiscal practices.
These leasing arrangements are at the forefront of Arizona‟ s drive to cut its budget deficit. Recent reports suggest that Arizona is considering selling 32 public buildings with the intention of leasing them back. These properties include the state‟ s capitol building. In sum, the arrangement could gain the state $ 350 to $ 700 million to cover part of this year‟ s state budget deficit. Given long- term leasing costs, however, this quick- fix could end up costing taxpayers much more. 2
2 “ To cut deficit, Arizona may sell its Capitol” Christian Science Monitor September 8, 2009.
43 SECTION 3: FISCAL INSTITUTIONS
Fiscal institutions are structures, set in place by law or constitutional amendment, that govern the behavior of elected officials in making fiscal decisions. State and local governments are subject to far more fiscal restrictions than are officials in the federal government. For more than 100 years, statutory and constitutional requirements have attempted to limit or change the degree to which state and local governments can raise taxes, borrow, and spend money. In this section we will provide an overview of the most common fiscal institutions in the American states. Our discussion will include tax and expenditure limits, balanced budget requirements, rainy day funds, and supermajority limits. We will examine the extent to which these fiscal institutions have in fact bound state and local governments and discuss the difficulties associated with empirical assessment. Arizona has passed each of these fiscal institutions, so we will pay particular attention to this case.
Tax and Expenditure Limits
Tax and Expenditure Limits ( TEL) are a ubiquitous fiscal institution in the US states. A TEL is defined as any law, statutory or constitutional, that pegs taxing or spending to an explicit rule with the aim of slowing the growth of government. The most common TELs stipulate that revenue or expenditures cannot grow at rates greater than a fixed percentage, inflation, population growth, or personal income growth.
TELs have been most commonly passed through direct democracy. From 1970 to 2006, there were 44 passed and 94 proposed TELs. Adopting these measures was particularly common in the wake of California‟ s Proposition 13, a stringent property tax measure that limited taxes to 1% of assessed value, which in turn was limited to 2% growth per year. Indeed, there were 20 TEL proposals via direct democracy alone within two years of Proposition 13. To this day, limitations on property taxes continue to be the most popular form of TEL at the ballot box.
Estimation of the effectiveness of TELs has been highly inconclusive. Studies that have looked at a pooled time- series of the fifty states have shown both that TELs are effective ( Misiolek and Elder 1988, Elder 1992, Shadbegian 1998, Bails and Tieslau 2000, and New 2001) and that TELs are ineffective ( Mullins and Joyce 1996; Kousser, McCubbins Moule 2008). In part, these mixed findings may be a result of different dependent variables. Studies that have looked at the effect of TELs on state only spending have found some evidence of TEL effectiveness, while studies that look at both state and local spending combined find no supportive evidence. This divergence suggests that TELs are circumvented by moving money to different levels of government. Most strikingly, however, Kousser, McCubbins and Moule 2008 found no evidence of the effectiveness of spending or revenue limits when looking at single states over time, regardless of the choice of dependent variable. The sole exception to their finding was Colorado‟ s 1992 TABOR amendment.
In general, property tax limitations appear to be more effective than other types of TELs. Figures 23 and 24 sheds some light on this conclusion3. Figure 23 compares property taxes, as a proportion of personal income, in states with property tax limits to all other states. Note that this categorization is static even though states may have adopted their limits at any point during this time series. Similarly, Figure 24 compares the spending patterns of states with spending limits to all other states. As is clear from the two figures, property tax limits appear to have a far greater affect than spending limits.
3 These graphics, and those that follow, use a local polynomial smoothing technique to summarize fiscal outcomes for groupings of data.
44 At the beginning of the series, states that would eventually pass property tax limits collected more property taxes than the comparison group. These states experience a notable and sustained drop in property tax collections as more and more states adopt limits. In comparison, any dip in spending in states with spending limits appears temporary. Visual inspection does not suggest that spending limits substantially transformed expenditure patterns in these states.
Figure 23: Effect of Property tax limits on Property Taxes as a Percentage of Personal Income
Figure 24: Effect of Spending Limits on Direct General Expenditures as a Percentage of Personal Income
45 Strong evidence exists that suggests that TELs are circumvented with a variety of fiscal innovations, such as devolution, borrowing, or assessments of charges and fees. One of the most common findings is that TELs push governments to raise charges and fees instead of taxes. This was evident in California, a case examined by Kousser, McCubbins, and Rozga ( 2007). California passed a property tax limit in 1978 ( Proposition 13) and a spending limit in 1979 ( the Gann Limit). Kousser, McCubbins, and Rozga ( 2007) show that both limits contributed to increases in charges and fees. Not surprisingly, California collected additional charges and fees to make up for a drop in property taxes. More unexpected, however, was that the state‟ s spending limit further contributed to this trend. This is because California‟ s spending limit limited the appropriation of tax revenue to an index based on population and inflation growth. To circumvent this limit, all politicians had to do was collect more non- tax revenues, such as charges and fees, since these were not limited by the letter of the law. Today, California relies heavily on charges and fees for general revenue.
Arizona is another excellent example of TEL circumvention. Arizona has passed two constitutional TELs, a spending limit in 1978 and a property tax limit in 1980. Figures 25 and 26 show fiscal outcomes in the state of Arizona from 1970- 2006. The vertical lines in each of these figures represent the passage of the relevant TEL. As clear from Figure 25, Arizona‟ s property tax limit immediately reduced property taxes. This drop in property taxes, however, corresponds almost perfectly with a matching rise in charges and fees. Arizona‟ s spending limit ( Figure 26), in contrast, does not seem to have any affect. This limit is tied to personal income, which generally grows at a rate faster than government expenditures.
Figure 25: Arizona Property Taxes and Charges and Fees before and after adoption of a property tax limit
46 Figure 26: Arizona Direct General Expenditures before and after adoption of a spending limit
Balanced Budget Requirements
All states except Vermont have enacted balanced budget requirements. The details of these requirements, however, vary widely among the remaining 49 states. In some states, it suffices that the governor simply submits a balanced budget, without respect to the solvency of the budget that actually passes later in the budget process. In other states, the legislature must pass a balanced budget. Some states allow budget deficits for a single year as long as the shortfalls are corrected in the next fiscal year. The majority of states, however, 29 of the 49 with requirements, have more stringent rules that require passage of a balanced budget and do not allow carryover to the next year. ( Knight et al. 2003). Arizona is not among the majority of states in this regard, because its balanced budget requirement allows deficits to carryover.
There is a significant literature on the effectiveness of balanced budget requirements. Unfortunately, most requirements were passed before annual fiscal data was readily available for all fifty states. As a consequence, scholars primarily have to rely on cross- sectional variation, instead of variation that exists over time. This is problematic for estimation purposes because there may be state- specific, unobservable characteristics that lead a state to adopt one form of balanced budget requirement over another. Since data is not available to look at differences within a single state, before and after that state adopts the requirement, these unobservable differences cannot be accounted for empirically.
Most scholars testing the effectiveness of these limits categorize the requirements by stringency. Alesina and Bayoumi ( 1996) show that more stringent balanced budget requirements are associated with larger average surplus as well as with lower cyclical variability. Alt and Lowery ( 1994) and Poterba ( 1994) also find a significant affect of the more stringent forms of balanced budget requirements, but only when there is a unified government. Both studies show that states with unified governments had sharper reactions to budget deficits if they were unable to carry deficits into the next fiscal year. Primo ( 2006) suggest that the effectiveness of balanced budget requirements hinges on existence of politically independent courts. While these studies do show some positive correlation between balanced budgets requirements and fiscal solvency, the fact that the results are contingent on specific political arrangements suggests that it is not the limits in and of themselves that are effective.
Briffault ( 1996) similarly takes a more skeptical view of these requirements. He suggests that many state budgets are balanced on paper only, and others are balanced because of political and fiscal constraints having little to do with provisions mandating a balanced budget. Because no counterfactual exists, in other words we don't know what a state's budget deficits would look like if they didn't have a balanced budget requirements, it is difficult to sort out correlation from causation for this fiscal institution. We tend to agree with this analysis of this fiscal institution.
47 Rainy Day Funds
Rainy day funds, also known as budget stabilization funds, are other new fiscal innovations. Whereas only nine states had formal rainy day funds in the recession of the early 80s, 46 states were able to rely on such funds in the 2001 recession ( Wagner and Elder 2004). Today, only four states have no form of Rainy Day Fund: Arkansas, Illinois, Kansas, and Montana. Rainy Day Funds provide a way for states to put aside excess revenues so that these savings can be used when there are unexpected revenue shortfalls.
Legislation that enacts rainy day funds varies significantly by state. Some states require that a specific percentage of all general fund appropriations be deposited in the rainy day fund. Other states require that all surpluses be deposited, while still others only require deposits by appropriation. Methods for withdrawing funds also vary by state. Some states only allow withdrawal when revenue estimates fall below target. Other states have automatic appropriations to cover any budget deficits. Still other states have more specific rules allowing money to be used to cover natural disasters or state emergencies. A full description of the variety of limits in place can be seen in Wagner and Elder ( 2004). Moreover, Sobel and Holcombe ( 1996a), Knight and Levinson ( 1999), and Wagner ( 2004) find that states with rules that required deposits and specify withdrawal allowances both save more and face lower borrowing costs than states with weaker rules or states without rainy day funds. Arizona‟ s rainy day fund sets an explicit formula for fund deposits, but is more flexible about withdrawal since the requirement is only statutory, not constitutional. On several occasions the state legislature has been accused of “ raiding” the budget stabilization fund for non- recession related fiscal emergencies.
The other major criticism of Arizona‟ s budget stabilization fund is that it is capped at 7% of the previous year‟ s general fund, which many argue is not enough savings for a major recession. Sobel and Holcombe ( 1996) and Elder and Wagner ( 2004) calculate the optimal size of a rainy day fund. Sobel and Holcombe, analyzing the recession of the early nineties, conclude that states would have needed reserves equal to 30% of expenditures to stave off effects of the 1990- 1991 recession. Elder and Wagner's estimation, using different methodology and expanding the scope of the study, estimates that a rainy day fund equal to 5.1% of the budget would be sufficient to buffer against future recessions so that expenditures would not have to be cut and revenues would not have to be raised. Estimates for each state are calculated individually for more detailed estimations of the needs of specific states, some of which require significant more buffering than others.
Super Majority Limits
Supermajority limits are another important fiscal institution in the US states. Supermajority limits come in three varieties: supermajority limits on state legislatures to pass budgets, supermajority limits on state legislatures to raise taxes, and supermajority limits on the public to raise taxes. In general, supermajority limits are commonly passed in conjunction with a TEL.
A total of nine states have supermajority requirements for state legislatures to pass appropriation bills. Arkansas, California and Rhode Island all require a supermajority vote ( ranging from 2/ 3 in California and Rhode Island to 3/ 4 in Arkansas) in order to pass most appropriations bill. The other six states, Connecticut, Hawaii, Illinois, Maine, Mississippi and Nebraska, only require a supermajority under certain conditions. For example, Connecticut and Hawaii require supermajority passage only if the general fund ceiling limit ( mandated by each state's TEL) is surpassed. In Maine and Nebraska, supermajority votes are only required on emergency appropriations in the case that the state budget is not passed in time for the next fiscal year.
48 A total of 15 states require that state legislatures have supermajority support for any increase in taxes, including Arizona. Most of these limits, 12 of the 15, apply to all tax increases. All fifteen of these requirements were passed either by initiative or referendum. In general, a supermajority limit on legislatively- imposed tax increases is a relatively recent phenomenon. Of the 15 limits in place, 9 of them were enacted after 1990. An even newer frontier in supermajority limits is limits that prevent increases in charges and fees in addition to taxes. This proposals have been made in response to the contention that supermajority limits cause increases in charges and fees.
A third type of supermajority limit applies to a state‟ s voters. In California, for example, all local general obligation bond measures must pass the voters by a two- thirds margin. In 2001, this requirement was weakened for school bond measures, which now only require a 55% voter supermajority. Similarly, Oregon and Washington have at one time imposed supermajority requirements on local school levies. These forms of limits are highly contentious and are often repealed over time. For instance, Massachusetts voters once needed a supermajority to pass overrides of their local property tax limit, Proposition 2 1/ 2. This requirement was repealed within a year of the proposition's original passage.
Figure 27: Effect of Supermajority limits on State and Local Taxes as a Percentage of Personal Income
Figure 28: Effect of Supermajority Limits on Own Source General Revenue as a Proportion of Personal Income
49 Figures 27 and 28 show the effect of legislative supermajority limits to raise taxes, the most common form of supermajority limit. These figures group states into two groups, one group that never adopts supermajority limits and another group that adopts a supermajority limit to raise taxes at some point in the time series. As can be seen in the figures, supermajority limits appear to reduce both taxes and general own source revenues.
Both Figures suggest that these limits have a dramatic affect on fiscal outcomes. Unfortunately, true causal inference is marred by the fact that the states that adopt supermajority limits are also the ones that adopt many of the other fiscal institutions discussed herein. It may also be the case, much like TELs, that supermajority limits could reduce taxes while simultaneously increasing charges and fees, or they could lead to the fiscalization of land use. More research must be done to isolate the effect of this interesting fiscal institution.
Arizona adopted a legislative supermajority limit in 1992. This limit stipulates that the legislature can only raise taxes with a two- thirds supermajority vote. Figure 29 presents Arizona taxes before and after the adoption of this limit, which is represented by a vertical line in the year 1992. This figure does not provide evidence that adopting a supermajority limit dramatically changed Arizona tax burden.
Figure 29: Arizona Taxes Before and After Adoption of a Super Majority Limit
Summary
This section has reviewed the fiscal implications of TELs, rainy day funds, balanced budget amendments, and supermajority limits. With the exception of the strictest laws governing the creation of rainy day limits, we are skeptical about the efficacy of these fiscal institutions.
50 It is helpful to think of the consequences of fiscal institutions in terms of their primary, secondary, and tertiary effects. The primary effects of fiscal institutions should be that they accomplish their stated goals. In other words, TELs should cut the size of government, balanced budget requirements should lead to balanced budgets, rainy day funds should be accessed only in times of need, and supermajority limits should prevent tax increases. The research we have reviewed suggests that these primary effects are only achieved in a minority of cases.
The secondary effects of fiscal institutions are the unintended consequences. These are most clear in the case of TELs and supermajority limits. As a result of these institutions, taxes have been replaced with charges and fees, borrowing innovations, and the fiscalization of land use. Some of these shifts are evident in the data available by the Census bureau, but we suspect that under- reporting of assessments and non- reporting of leasing underestimates the secondary effects.
The tertiary effects of fiscal institutions are the consequences of the secondary effects, in other words the unintended consequences of the unintended consequences. We believe the tertiary effects are found in the democratic deficiencies of new fiscal innovations. New forms of financing that have been developed to avoid the bite of fiscal institutions, such as special assessments or leasing practices, are far less transparent than classical sources of revenue and debt. Administration practices in local governments receive very little monitoring and, in the case of assessment districts, are sometimes exempt from the state‟ s open- meeting and open- record laws.
SECTION 4: CONCLUSION
This report has analyzed the trends in fiscal activity for the fifty U. S. states over the last thirty years with comparisons drawn to the state of Arizona. We have provided both descriptive information about the expenditure and revenue practices of the fifty states and highlighted the most worrisome trends. We end this report with three conclusions.
( 1) First, in terms of state and local expenditures, the biggest priority is calculating the liabilities associated with pensions and retiree healthcare coverage. Currently, state and local expenditures on healthcare wane in comparison to the federal government and private expenses. If employee retirement healthcare liabilities are honored, this will certainly not be the case in the future.
( 2) Our second conclusion, on the topic of state and local revenues, is that new financing practices have transformed revenue sources. Today, state and local governments are heavily reliant on charges and fees for public services, suggesting what could be analogous to a privatization of government services. These new forms of revenues hide the true cost of government and are less transparent that previous sources, namely property taxes. We worry that these fiscal innovations have democratic deficiencies.
( 3) Finally, this report analyzed state and local fiscal institutions such as tax and expenditure limits, supermajority limits, balanced budget requirements, and rainy day funds. For the most part, we find precious little evidence that fiscal institutions are effective at slowing the growth of government. Instead, they appear to have led to the fiscal innovations discussed above.
51 FIGURES INFORMATION
Figure 1: State and Local Expenditures Per- Capita, Arizona and Nationwide
Data: State and Local Direct General Expenditures, State Total Personal Income
Suggested Citation: State & Local Government Finance Data Query System. http:// www. taxpolicycenter. org/ slf- dqs/ pages. cfm. The Urban Institute- Brookings Institution Tax Policy Center. Data from U. S. Census Bureau, Annual Survey of State and Local Government Finances, Government Finances, Volume 4, and Census of Governments ( 1977- 2006)
Figure 2: Expenditure Growth Rates of Federal, State, and Local Governments
Data
Federal: Data from the Congressional Budget Office, available at http:// www. cbo. gov/ ftpdocs/ 100xx/ doc10014/ Historicaltables09Jun09web. XLS
State: Total Expenditures, various editions of State and Local Government Finances ( Census)
Local: Total Expenditures various editions of State and Local Government Finances ( Census)
Figure 3: State and Local Expenditure Growth Rates, Arizona and Nationwide
Data: State and Local Direct General Expenditures ( E014)
Suggested Citation: State & Local Government Finance Data Query System. http:// www. taxpolicycenter. org/ slf- dqs/ pages. cfm. The Urban Institute- Brookings Institution Tax Policy Center. Data from U. S. Census Bureau, Annual Survey of State and Local Government Finances, Government Finances, Volume 4, and Census of Governments ( 1977- 2006)
Figure 4: Share of Total Local Direct General Expenditures by type of Local Government Nationwide
Data: County, School District, Special District, and ( sum) Municipal and Township Direct General Expenditures as a percentage of Local Direct General Expenditures
Suggested Citation: State & Local Government Finance Data Query System. http:// www. taxpolicycenter. org/ slf- dqs/ pages. cfm. The Urban Institute- Brookings Institution Tax Policy Center. Data from U. S. Census Bureau, Annual Survey of State and Local Government Finances, Government Finances, Volume 4, and Census of Governments ( 1977- 2003)
Figure 5: Share of Total Local Direct General Expenditures by type of Local Government in Arizona
Data: County, School District, Special District, and ( sum) Municipal and Township Direct General Expenditures as a percentage of Local Direct General Expenditures
Suggested Citation: State & Local Government Finance Data Query System. http:// www. taxpolicycenter. org/ slf- dqs/ pages. cfm. The Urban Institute- Brookings Institution Tax Policy Center. Data from U. S. Census Bureau, Annual Survey of State and Local Government Finances, Government Finances, Volume 4, and Census of Governments ( 1977- 2003)
Figure 6: Cumulative State and Local Government Expenditures by Functional Category
Data: State and Local
Education: E024 - Total Education Direct Expenditures
Welfare: E090 - Total Welfare Direct Expenditures
Health: E052 - Total health and Hospital Direct Expenditures
Highways: E065 Total Highways Direct Expenditures
Public Safety: E019 ( Police and Fire) + E021 ( Correction) Direct Expenditures
Other: Total Direct General Expenditures ( E014) minus all categories above
Suggested Citation: State & Local Government Finance Data Query System. http:// www. taxpolicycenter. org/ slf- dqs/ pages. cfm. The Urban Institute- Brookings Institution Tax Policy Center. Data from U. S. Census Bureau, Annual Survey of State and Local Government Finances, Government Finances, Volume 4, and Census of Governments ( 1977- 2006)
Figure 7: Cumulative State and Local Government Expenditures by Functional Category for the State of Arizona
Data: State and Local
Education: E024 - Total Education Direct Expenditures
Welfare: E090 - Total Welfare Direct Expenditures
Health: E052 - Total health and Hospital Direct Expenditures
Highways: E065 Total Highways Direct Expenditures
Public Safety: E019 ( Police and Fire) + E021 ( Correction) Direct Expenditures
Other: Total Direct General Expenditures ( E014) minus all categories above
Suggested Citation: State & Local Government Finance Data Query System. http:// www. taxpolicycenter. org/ slf- dqs/ pages. cfm. The Urban Institute- Brookings Institution Tax Policy Center. Data from U. S. Census Bureau, Ann