United States General Accounting Office Report to the Chairman, Committee on GAO Finance, U.S. Senate !’ ._ t ApriMl90 COMMUNITIES IN r FISCAL DISTRESS State Grant Targeting Provides Limited Help Human Resources Division B-236433 April 13,199O The Honorable Lloyd Bentsen Chairman, Committee on Finance United States Senate Dear Mr. Chairman: This report responds to your request that we assess whether states are meeting the needs of fiscally distressed communities, particularly in light of the expiration of federal revenue- sharing in 1986. We identified numerous state general fiscal assistance programs available to fiscally distressed communities. In this report, we (1) examine the extent to which these programs reduce differences in tax burdens among general purpose local governments in 48 states and (2) compare the success of these state programs with that of the old federal revenue-sharing program in reducing differences in tax burdens between distressed and better-off communities. Copies of this report are being sent to other interested congressional committees and members, as well as to national associations representing state and local governments. We also will make copies available to other interested parties upon request. Please contact me at (202) 275-1655 if you or your staff have any questions concerning this report. Other major contributors to it are listed in appendix XIV. Sincerely yours, Linda G. Morra Director, Intergovernmental and Management Issues ExecutiveSummery The most direct means of reducing such disparities is by targeting state and federal grant funds to fiscally distressed communities, thereby reducing their tax burdens compared with “better-off” communities. State and federal grant programs have the inherent flexibility to serve such policy ends. Unrestricted general fiscal assistance grants are better suited than special purpose grants to achieving disparity reduction goals. Special purpose aid is aimed at satisfying specific public service needs that may not correspond to where fiscal disparities are greatest. This report focuses on how general fiscal assistance grants reduce dis- parities among general purpose local governments. In performing its analysis, GAOexamined data for 1986 (the last year of the federal reve- nue-sharing program), conducted a SO-state telephone survey, and inter- viewed state officials in 11 states. Because of data limitations, GAO did not separately analyze the disparity-reducing effects of services pro- vided directly by state governments, or local education and special pur- pose aid to local governments. State and federal grant programs reduce local financing burdens. How- Results in Brief ever, grants from both have decreased as a share of local revenues. Between 1977 and 1987, state and federal aid dropped from 40 to 31 percent of local revenues. Reductions in state aid accounted for 2 points of the 9 percentage point drop; an overall reduction in federal aid accounted for the rest. As a result, local governments have had to finance an increasing share of their expenditures from local resources. At the same time, differences in per capita incomes widened between poorer and more affluent counties. As the ability to bear tax burdens is directly related to income, this suggests that fiscal disparities between poorer and more affluent communities have increased. In GAO’Sanalysis of 1985 data, disparities between fiscally distressed communities, such as Starr County, Texas, and better-off communities existed in all states. But the extent of disparities differed substantially across states. States provided $10.9 billion in general purpose fiscal assistance to local governments in 1985” and federal revenue-sharing added another $4.6 ‘Weusedfiscalyear1985databecause this wasthe latestlocalgovernment tax dataavailablefor our analysis.Analysisof morecurrentdataonstateprogramshouldyieldresultssimilarto those presentedhere,asstatesinfrequentlychangethe formulasusedto distributethis aid. Page3 GAO/~90-69 StatesHelpCommunitiesin maxi DMXTSB Executivesummary Federal Revenue-Sharing The amount of disparity reduction achieved depends on both how much More Targeted to funding is provided and the extent to which it is targeted to fiscally dis- tressed local governments. Even though it had less than half the funding Distressed Communities of most state programs, the federal revenue-sharing program produced a greater reduction. In 31 of the 48 states analyzed, federal revenue- sharing reduced disparities more than did state programs because it was more targeted to distressed communities (see p. 42). GAOis making no recommendations. Recommendations GAOdid not solicit agency comments. Agency Comments Page6 StatesHelpC!ommunitie~ GAO/‘HRb9O89 in &cal DI~trees contents Appendix V: General Fiscal Assistance Programs in 66 Minnesota Appendix VI: General Fiscal Assistance Programs in New 69 Jersey Appendix VII: General Fiscal Assistance Programs in New 76 York Appendix VIII: General Fiscal Assistance Programs in 81 Rhode Island Appendix IX: General Fiscal Assistance Programs in 82 Tennessee Appendix X: General Fiscal Assistance Programs in Texas 86 Appendix XI: General Fiscal Assistance Programs in 87 Vermont Appendix XII: General Fiscal Assistance Program in 88 Wisconsin Appendix XIII: State and Federal Intergovernmental Aid 91 to General Purpose Local Governments, by State (Fiscal Year 1985) Appendix XIV: Major Contributors to This Report 93 Tables Table 2.1: Share of Public Services Delivered by State 23 Governments, by State (Fiscal Year 1985) Table 2.2: Share of Local Public Services Financed by 26 State and Federal Grants, by State (Fiscal Year 1985) Table 2.3: States Ranked by Extent of Local Fiscal 32 Disparities, Assuming No State and Federal General Fiscal Assistance (Fiscal Year 1985) Table 3.1: State and Federal General Fiscal Assistance as 36 a Percentage of Local Revenues, by State (Fiscal Year 1985) Table 3.2: Reduction in Fiscal Disparities Attributable to 39 Combined State and Federal General Fiscal Assistance, by State (Fiscal Year 1985) Table 3.3: Reduction in Fiscal Disparities in 16 States 42 With the Widest Disparities (Ranked by Per Capita Funding) (Fiscal Year 1985) Table 3.4: Reduction in Fiscal Disparities Attributable to 43 State and Federal General Fiscal Assistance, by State (Fiscal Year 1985) Table I. 1: Standard Deviation of Local Tax Burdens Per 62 Dollar of Expenditures, Before and After Receipt of General Fiscal Assistance, by State (Fiscal Year 1985) Pagr 7 GAO/llELNXM9StatesHelpCommunitiesin Pkd Distress Page9 GAO/HRD9O89 StatesHelpCanmdties in Fkd Distress chapter 1 introduction Figure 1.1: Distdbutlon 01 Local Qovernment Revenue Sources (1977-87) PWCWII 1977 1660 1661 1662 1963 1964 1966 1666 1667 Years L-l Federal Grants State Grants m own-sourca Revenue Source U S. Bureau of the Census, Governmental Finances, table 29 for 1985-1987. table 23 for 1980 and 1982-84, and table 24 for 1977 and 1981 increased because of the relative declines in the amount of state and federal aid to localities. Fiscal disparities-that is, differing abilities to finance comparable local Local Fiscal public services-are an inherent result of using a decentralized Disparities: A Matter approach to providing public services. Such differences can lead to wide of Public Concern variations in (1) the level of public services among localities with com- parable tax burdens or (2) tax burdens among localities providing com- parable service levels. Differences in tax burdens relative to services received raise equity con- cerns that ultimately must be answered through the political process: Page11 GAO/HRLWM9StatesHelpC!ommunities in PiscalDistress chapter1 Intmduction this $4.6 billion program exacerbated disparities. At that time, we esti- mated that the expiration of federal revenue-sharing could increase fis- cal disparities among local governments on average by 10 to 15 percent.3 Several members of the Congress have proposed a less expensive and Objectives, Scope,and more targeted replacement for the general revenue-sharing program. To Methodology help assess the need for such a program, the Chairman of the Senate Finance Committee asked us to examine the extent to which state and federal assistance programs enable local governments to meet their pub- lic service needs with comparable local tax burdens. To do so, we developed the following objectives: (1) to assess the rela- tive extent of local government fiscal disparities within each state, (2) to identify state policies and strategies that affect the magnitude of these disparities, and (3) to assess the extent to which state and federal gen- eral fiscal assistance programs alleviated them in 1985, the year before the expiration of federal revenue-sharing. The uses of general fiscal assistance grants are unrestricted; local offi- cials can use them to finance any of the services they provide. Because of their unrestricted nature, general fiscal assistance programs easily can be designed to reduce the gap between fiscally distressed and better- off communities. Consequently, the programs represent a pool of resources that could be used for this purpose. Some states have designed their general fiscal assistance aid to reduce fiscal disparities but many have not. Our analysis assesses the extent to which general fiscal assis- tance programs reduce the gap between fiscally distressed and better- off communities, whether or not they were intended to do so. While our study analyzes the disparity reduction achieved on a state-by-state basis, it does not assess reductions in disparities among local govern- ments across states. We conducted our review between October 1987 and March 1989 in accordance with generally accepted government auditing standards. To develop information on applicable policy issues and economic theory, we did a literature search and reviewed GAO’Spast work in this area. 30urreport,LocalGovements:TargetingGeneralFiscalAssistanceReduces FiscalDisparities (GAO/HRLMG-1131, examinedfti disparitiesamonggeneralpurposelocalgovernmentsandthe impactof federalrevenue-sharing onreducingthosedisparities. Page13 GAO/HBD9989StatesHelpCommunitiesin FiscalDistress Chapter1 Introduction on state programs should yield results similar to those presented in this report because states infrequently change the formulas used to dis- tribute this aid. l used per capita income, averaged over the years 198084, as a proxy for local revenue-raising capacity. Income is a comprehensive measure of residents’ “ability-to-pay,” widely used by analysts when measuring tax burdens. We did not use the legal tax base from which local govern- ments directly raise their revenues, because residents’ personal incomes better measure ability-to-pay taxes than their property values. l used the 1984 population as a proxy for public service needs among local governments. Data required for more precise measures of public service needs in every state does not exist. While many factors deter- mine public service needs, most are associated with population size. . did not attempt to reflect the varying costs of providing similar public services in different localities. No consistent unit-cost data applicable to local governments across the states exists. Identifying State To identify state policies and strategies to reduce fiscal disparities Strategies among local governments, we conducted a 50-state telephone survey. In each state, we contacted senior staff members of the executive and legis- lative branches to obtain a broad overview of how each state deals with local fiscal disparities. State program descriptions and discussions of state law in this report are based on information obtained from these interviews and explanatory materials provided by these officials. We also interviewed officials from the U.S. Bureau of the Census, National Association of State Budget Officers, and the Advisory Commission on Intergovernmental Relations. In addition, to develop an in-depth understanding of their policies for addressing fiscal disparities, we visited 11 states: California, Kansas, Massachusetts, Minnesota, New Jersey, New York, Rhode Island, Ten- nessee, Texas, Vermont, and Wisconsin. In selecting them, we balanced differences in geographic location, population, and approaches to pro- viding state general fiscal assistance. Our field work enabled us to relate the results of our statistical analyses of disparities to specific state strategies and policies that alleviate them (see apps. II through XII). Page16 GAO/HlZD9069States,HelpCommunities in FiscalDistis chapter 1 Introduction Figure 1.3: States Selected by GAO for Fieldwork M Minnesota I \ Wisconsw n _I h /A’ c iv T’L - New York Vermont ” . -’ - Massachusetts Rhode Island New Jersey California KaflSaS Tennessee Fieldwork States Page17 GAO/HRD-9949 StatesHelp Communitiesin &cd Distress chapter 2 PlscnlDisparities:Their Nature,Sources, andExtent Two basic factors separate better-off communities from those that are The Nature of Fiscal fiscally disadvantaged: Disparities 1. The tax burdens borne by local residents to finance public services and 2. The level and quality of services these taxes finance. Better-off communities can finance relatively high levels of public ser- vices with relatively low tax burdens. But disadvantaged communities must bear relatively high tax burdens that can finance only relatively low levels of public services. Two New Jersey cities, Woodbine and Alpine, illustrate an extreme example of fiscal disparities (see fig. 2.1). In 1983, Woodbine had the lowest per capita income ($5,013) of all New Jersey communities and Alpine the highest ($39,004). With a tax burden equal to 1.7 percent of its residents’ per capita income, Woodbine raised $83 per resident in 1985. In comparison, Alpine enjoyed a more favorable fiscal situation. With a tax burden equal to 0.6 percent of its per capita income, it raised $232 per resident. Thus, with a tax burden about one-third of Wood- bine’s, Alpine raised almost three times as much revenue. Page19 GAO/HUB9989StatesHelpCommunitiesin PiscalDSsh-esa Chapter2 FiscalDisparities:TheirNature,sources, and Extent statewide average can provide services at a lower effective tax rate than relatively poor communities1 . The unit cost of providing public services. For example, the starting sal- ary of a police officer in San Francisco, California was just over $29,000 in 1987 compared with about $16,000 for a similar position in Glen County, California. Even allowing for its higher per capita income, San Francisco must bear a higher per capita tax burden for each policeman it hires per 1,000 residents than Glen County does. l The level of public service needs among communities, resulting from dif- ferences in geographic or socioeconomic conditions. For example, High- land Park and Metuchen are two New Jersey communities with almost equal populations. Yet, Metuchen has 132 miles of streets to maintain, five times more than Highland Park. In 1985, Metuchen’s expenditures for streets were $435,000 compared to $279,000 for Highland Park. Local governments have limited ability to alter socioeconomic character- istics that contribute to the rise of fiscal disparities, especially over the short run. The value of tax bases, which are the sources of revenues to pay for public services, depends largely on a community’s economic con- dition and its prospects for employment and business opportunities. Through economic development strategies, communities with relatively low tax bases may augment their taxable resources. But even when suc- cessful, these developments require years to reach fruition. Similarly, unit costs of services, such as wages and salaries or office space and land are largely influenced by remunerations available in the private sector. Also, a community faced with relatively high public safety and welfare service needs is ill-suited to alter the underlying causes of long- term societal problems such as crime and poverty. Therefore, local gov- ernments that are fiscally disadvantaged may require state and federal government assistance in meeting their public service needs. States differ significantly in the extent to which they centralize service Centralizing Service delivery. On average, state governments delivered just over half of all Delivery Affects Local noneducation public services in 1985, but there was significant varia- Disparities tion. Some states, such as Vermont and Alaska, provided over three- quarters of such services to their residents, while in Florida and Nevada, the state provided less than 40 percent. ‘Theuseof percapitaincomemeasures mayoverstatethetrue tax burden.Somecommunities are ableto shift a substantialshareof their localtax burdento nonresidents by “exporting”taxes.For example,Stratton,Vermont,is a majorski resort.In 1987,townresidentsownedonly 2 percentof the taxablerealestatein thetown Page21 GAO/‘HRD9O-69 StatesHelpCommunitiesin FiscalDistress Chapter 2 Fiscal Disparities: Their Nature, Sources, and Extent Table 2.1: Share of Public Services Delivered by State Governments, by Percent of public Percent of public State (Fiscal Year 1965) services delivered services delivered State by state State by state Vermont 78.6 Nebraska 55.3 Alaska ___- - ~_.__ 754 Idaho 55.2 Rhode Island _ -.__- 74.7 Missouri 55.2 Marne _~ -~ --.__72.2 Oklahoma 54.9 North Dakota 71.3 Michigan 54.3 West Virginra 66.3 .-__--- New Jersey 54.2 Connectrcut __---~-~ -- 66.1 Wyoming 52.6 Massachusetts 65.8 ~-____.- Mississippr 52.0 South Dakota 65.3 Kentucky __-.- --~-~ - ~. 63 3 U.S. Average’ 51.3 Oregon __-. 62 4 South Carolina 61.6 Tennessee 50.6 Maryland -__ 61 4 Wisconsin 49.6 PennsylvanIa 60.0 Iowa 48.3 New Hamoshrre 59 8 Ohio 47.6 Washington __~~~ ~~ ~~~ 59 1 Texas 46.0 Montana 59.1 Minnesota 45.6 Alabama -.__-- __-- 59.0 ~__~-__ Kansas - 44.0 Arkansas 58.9 Indiana 43.5 New Mexico 50.3 Colorado 42.2 Utah 57 7 California 41 2 lllrnois 57 2 New York 41 .I Loursrana 56.0 Florida 38.6 Georgra 55.9 Arizona 38.6 North .___ Carolrna .~ ~__.55.8 Nevada 38.0 Virginia 55 5 Source US. Bureau of the Census. Governmental Fmances, 1984-85. aWe calculated the U S average by welghtlng each state’s share of state-delwered services by its respective population we Grants from states and the federal government can be classified into two groups: 1. Categorical grants:’ support specific program activities ranging from very narrowly defined functions, such as medical care for low-income aidwhoseuseis restrictedto specificprogramareas,regardless ‘This mcludesall intergovernmental howbroad.Thisdefinitionwouldencompass blockgrants Page 23 GAO/HRLMO-S9 States Help Commuxdtie in Fiscal Distress Chapter2 FiscalDisparities:Their Nature,Sources, andExtent Table 2.2: Share of Local Public Services Financed by State and Federal Grants, State Percent State Percent by State (Fiscal Year 1985) Oklahoma 64 4 Oregon 34.3 South Carolina 61 .l Alabama 32.6 Mississrppi 60.7 West Virginia 30.0 Arizona .~ 56.4 Pennsylvania 29.9 Wisconsm 57 0 Virginia 29.7 California ___~ ~-~~ 56.2 Washinoton 29.0 Arkansas 48.2 Connecticut 27.9 ~-~ ldaho- - 47.8 ____-__ Kentucky 23.8 Michigan 47.8 New York 23.7 Nevada 46.7 ~__-- Vermont 23.7 New Mexico 46 1 Rhode island 23.0 Minnesota 43.8 Texas 22.1 Louislana- 43.3 Montana 21 .a Indiana 42.3 Alaska 19.9 Ohio ~-___ 42.0 New Hamoshire la.4 Iowa 39.8 Maryland 16.3 Missouri 38.6 - North Carolina 15.8 North Dakota 38.4 ~__~-___ Flonda 15.2 Tennessee 38.3 Kansas 11.7 Nebraska 37.5 Maine 10.4 Wyommg 37.3 Utah 8.6 Massachusetts -.___ 36.9 Colorado a.5 South Dakota -~___ ~-- 36.0 ___~_ Georgia 2.8 New Jersey 35.6 _____-- __~ Illinois 35.4 U.S. Averaae 34.8 Source: U S Census Bureau, “Revenue-Sharing Allocation File for Entitlement Period 17” and “Tax and Intergovernmental Aid File for Frscal Year 1984/R” (computer-based files). Local public services exclude public educatron. and grants exclude state and federal ard-to-education. Half of Tennessee’s highway aid is distributed equally among the coun- ties, 25 percent is distributed by land area, and the remaining 25 percent by population. Federal categorical aid usually is allocated by formulas, most often baaed on program costs or such usage factors as population or potential caseloads. In fact, few state or federal categorical programs allocate funds according to residents’ taxpaying ability. Page26 GAO/‘HRD9O89 StatesHelpCommunitiesin PiscalDistress Chapter 2 Fiscal Disparities:Their Nahm, Sow, and Extent share of the state’s total population, the other 35 percent according to their percentage of the state’s total assessed tangible property. The county government receives half of the county area’s allotment and the remainder is divided among its municipalities according to their share of the county’s population. At the federal level, general revenue-sharing was the major general fis- cal assistance program that used a tax-baaed targeting method to dis- tribute aid. This program allocated funds on the basis of the tax burdens, per capita incomes, and populations of local governments. The way local geographical boundaries are set and the extent of local Other State Policies restrictions on revenue-raising can increase or decrease the extent of Affecting the Extent local fiscal disparities. of Local Disparities Boundaries Can Affect Establishing geographic boundaries for local governments creates com- munities with differing fiscal capacities if fiscal needs and resources are Disparities not evenly distributed. Although it is theoretically possible to draw com- munity boundaries so each community has an equal ability to raise reve- nue, in practice this has not been done. Giving local governments the ability to adjust their political boundaries can affect the extent of fiscal disparities, as illustrated by the situation in Texas, Texas home rule cities” can unilaterally annex adjacent unincorporated areas As the population of adjacent areas increases, home rule cities have the option of annexation-with or without the consent of the neighboring communities. Thus, annexation allows central cities to expand their tax bases while providing outlying areas with municipal services such as water and sewer. Such policies long have been credited with lessening the deterioration of central city tax bases caused by out- migration in Texas. While this policy has been successful in preventing the erosion of some central cities’ tax bases, it has not completely solved the problem of local fiscal disparities. For example, many poor unincorporated areas in the state-particularly near the border-have grown in population but ‘%e stateconstitutionof Texasallowsanycity with a populationover6,000to adopta homerule charter.Homerule citieshavethepowerto doanythingtheywishthat is notspecificallyprohibited by statelaw.Ofthe 1,121municipalitiesin Texas,217arehomerule cities. Page27 GAO/HRTMO89 StatesHelpCommunitiesin FiscalIKstfese Chapter2 Fiacd Disparities:Their Nature,So-, andExtent average resident tax burden per $100 of expenditures. Tax burdens are local taxes as a percent of a county’s average personal income. The dispersion in tax burdens for Florida counties is shown in figure 2.2. The largest burden was in Union County, where residents pay taxes equal to 1.94 percent of their income for each $100 of public services they receive. In contrast, the lightest burden was in Palm Beach County, where residents pay taxes equal to 0.69 percent of their income for each $100 of public services they receive. Thus, the heaviest burden was nearly three times the lightest. Page29 GAO/IilfD90-99StatesHelpCommmitiesin FiscalDistress Chapter 2 pisal ~H~parities: Their Nature, Sources, and Extent Figure 2.3: Dispersion of County Tax Burden Per $100 Dollars of Public Services in 99 Iowa Counties (Fiscal Year 1985) 2.0 Tar Burden Per $100 Dollars 01 Services 1.9 1.8 1.7 1.6 1.5 1.4 w Decatur 1.3 wm m 1.2 m n .C I n w 8 n l wm m 1.1 l W. w m mm m m B I n n I I mm li . n 1.0 0.9 l n+ n u - m n n mm m mrnrn l +m mm. mmm mm l m l .+ m 'mmm n ,mmmm I mm Q H 0.8 n U-Polk 0.7 0.6 Note The 99 counties are arrayed alphabetlcally on the honzontal axis Source US. Bureau of the Census, ‘Revenue-Sharmg Allocatton File for Entitlement Period 17” and ‘Tax and Intergovernmental Ald iYe for Fwal Year 19@4/85” (computer-based files). For the 48 states in our analysis, we constructed an index of local fiscal disparities. The state in which local disparities were equal to the national average was assigned an index value of 100. Table 2.3 uses this index to rank the states. It shows that the potential range of fiscal dis- parities, in the absence of general fiscal assistance, is substantial. Ken- tucky’s fiscal disparities would have been three times greater than Nevada’s” - I ‘Wemeasured localdisparitiesby thestandarddeviationin local tax burdens per dollarof services. Thestandarddeviationis a statisticalmeasure of dispersionthat provides an empirical methodfor measuring disparities.Seeapp I for a discussion of why wechose thii measure of dispersion. Page 31 GAO/HRW3@69 States Help Gxnmunitie~ in Fbcal Distress Chapter2 FiscalDisparities:Their Nature,Sources, andExtent State Index no. States with relatively small disparities: ____ Connecticut 07a Minnesota --~______ 86 Ohio 85 Nebraska El California 80 Washington 80 Pennsylvania 76 Montana 74 Rhode Island _____ 72 North Dakota 72 Kansas 70 Indiana 67 Oregon ____ - 61 Wyomlnq 60 Iowa 54 Nevada 53 Note: See app I for the actual standard dewations and the methodology used to calculate them aA state wth dlspanties equal to the natlonal average would have an Index number of 100 We rounded the Index numbers and grouped the states Into thirds to develop the three categories. New York’s Index number was 87.1, while Connecticut’s was 87 0 Source U.S. Bureau of the Census. “Revenue-Sharing Allocahon File for Entitlement Period 17” and “Tax and Intergovernmental Ald File for Fiscal Year 19&I/85” (computer-based files). The extent of potential disparities-large or small-is not systemati- cally correlated with state population or geographic sizes. Some low- population states such as Wyoming and Nevada would have a relatively small range of disparities, while others such as Alaska and South Dakota would have a wide range. Populous states such as Florida and Texas would have large disparities, while California and Ohio would not. Of the four geographically largest states, Alaska and Texas have large disparities while California and Montana do not. However, there is a regional pattern. Potential disparities are most prominent in the Southwest (see fig. 2.4). They are less evident in the western and north- ern Plains states. Despite these regional trends, neighboring states can vary widely in the range of disparities, e.g., North Dakota and South Dakota, New Jersey and Pennsylvania, and Utah and Nevada. Page33 GAO/HRD-906g StatesHelp Communitiesin FiscalDi&ress ChaDter 3 Federal Revenue-SharingReducedLocailFiscal Disparities More Than Did Most State Programs In 1985, state general fiscal assistance to communities totaled $10.9 bil- lion and federal revenue-sharing added another $4.6 billion. Together, these programs comprised about 12 percent of total local revenues and reduced fiscal disparities among local governments by 18 percent. Ana- lyzed separately, state programs reduced disparities by about 9 percent and federal revenue-sharing by about 11 percent.’ Despite substantially more funding, most state programs reduced disparities less than did fed- eral revenue-sharing. If they were to target existing aid more specifi- cally to fiscally distressed localities, many states could further reduce disparities without additional funding. General purpose fiscal assistance, both state and federal, is an impor- General Fiscal tant source of revenue for local governments. States provided $10.9 bil- Assistance: An lion, or $47 per person, in fiscal year 1985, and the federal government Important Local allocated $4.6 billion, or $19 per person. In 1985, state and federal gen- eral fiscal assistance comprised 12.3 percent of local government reve- Revenue Source nues (see table 3.1). In 10 states, it accounted for more than 20 percent of revenues; in 17, less than 10 percent. Table 3.1: State and Federal General Fiscal Assistance as a Percentage of Percent of local revenues Local Revenues, by State (Fiscal Year State Combined State Federal 1985) Nevada 38.5 35.9 2.6 New Jersey 38.2 35.3 2.9 Wisconsin 32.8 29.2 3.7 New Mexico 31 9 27.2 4.7 Mississippi 31.1 22.8 8.3 South Carolina 27.5 17.6 10.0 Wyoming 26.7 22.3 4.3 Massachusetts -26.1 22.4 3.7 Arizona -~~ 25.3 22.3 3.0 Minnesota 22.7 19.6 3.1 Mtchigan 19.9 15.7 4.2 West Virgbnla 186 4.0 14.6 Arkansas 18.4 9.0 9.5 North Dakota 17.8 12.8 5.0 Florida 17.5 14.6 3.0 Idaho 16.5 8.9 7.7 lndlana 16 1 11.7 4.4 (continued) ‘Theoveralldisparityreductionfor generalfiscalassistance programsis lessthanthedisparity reductionachievedby the state andfederalprogramsseparatelybecause, in somestates,stateand federalaidoffseteachother Page 36 GAO/HIuT9069 States HelpCk~nununitiea in FiscalDistress chapter3 FederalRevenue-SharingReduced Local FiscalDisparitiesMoreThanDid Most state Frognuns local revenue in 33 states. Within these states, state aid ranged from a high of almost 36 percent of local revenues (in Nevada) to a low of 4.5 percent (in New York). In the 15 states where federal revenue-sharing was greater than state aid, federal aid never exceeded 15 percent of local revenues. Funding levels for state general fiscal assistance programs varied con- siderably by state for several reasons. In some cases, a high degree of funding was related to a state’s concern about local public service needs. For example, states with large general fiscal assistance programs such as Minnesota and Wisconsin have programs aimed in part to reduce dis- parities among their local governments. In other cases, it was an explicit political choice to have small general assistance programs. For example, Kentucky, Texas, and Utah provided very little aid, in part because they have state constitutional provisions that prohibit general fiscal assis- tance grants to local governments. And in other states, officials expressed the view that ameliorating fiscal disparities was not a state government responsibility, nor was it seen as an important policy issue. Looking at disparity reduction from a national perspective, states with Larger or Better- the widest disparities (identified in table 2.3) would need to have larger Targeted Aid or more targeted general fiscal assistance programs compared with Programs Would states where disparities are smaller.Z Red&e Fiscal Five of the 16 states with the widest disparities, when measured with- Disparities out considering state and federal general fiscal assistance, had large pro- grams. Funding, expressed as a percent of local revenues, substantially exceeded the 8.7 percent national average. They were: New Jersey (35.3 percent), New Mexico (27.2 percent), Arizona (22.3 percent), Florida (14.6 percent), and Alaska (12.6 percent). If these relatively highly funded aid programs were more targeted to fiscally distressed local gov- ernments, substantial disparity reductions could be achieved. At the other extreme, general assistance provided 1 percent or less of local revenues in 7 of the 16 states (Colorado, Oklahoma, Texas, Ver- mont, Utah, Kentucky, and Missouri). These states cannot reduce dis- parities by much, even with highly targeted programs, because of relatively low funding. ‘In general,increasingthefundingof existingstategeneralfiscalassistanceprogramswouldreduce fiscaldisparities.However.if theincreasedfundingweredistributedamonglocalgovernments according to eachlocalgovernment’s shareof statetaxescollected(return-to-originassistance), dis- paritieswouldbeunaffected Page37 GAO/HRLMO49 StatesHelpCommunitiesin FiscalDistrem clmpter 3 FederalRevenueSharingReducedLoeel FiscalDisparitiesMoreThanDid Most state programs average-income counties, After general fiscal assistance aid, Starr County’s fiscal disadvantage falls from $37.21 to $29.36, a Xl-percent reduction. Similarly, the income of the average resident in Robertson County was about 20 percent below Wheeler County’s average income and his/her fiscal disadvantage without general fiscal assistance would be $7.46. After general fiscal assistance aid, however, this disadvantage falls to $7.31, a 2-percent reduction. For local governments in all 254 county areas in Texas, general fiscal assistance aid was distributed among county areas in a way that reduced these fiscal disadvantages by 15.5 percent, on average, as shown in table 3.2. Table 3.2: Reduction in Fiscal Disparities Attributable to Combined State and Disparity reduction Federal General Fiscal Assistance, by As a ercent of State (Fiscal Year 1985) U.0 . average Per capita State Percent (U.S.=1 W) general aid Nevada .--___ 54.1 302 $227 South Carolina 39 7 219 57 Arkansas _~-. 36.1 ___. 199 38 West Virainia 34.4 190 29 Maine 32.6 180 48 Arizona _ ~~ _._~ ~~~ 32.5 180 146 Louisiana 28.7 159 47 Florida --~~~ 27.5 152 90 Tennessee 26.4 146 40 Alabama -___ 26.1 144 37 Rhode island -~ -.______ 24.4 ____. 135 52 Iowa ____ ~~~ .~ 24.1 133 63 Mississippr 23.8 131 86 New Jersev 23.2 128 133 Minnesota 23.1 128 155 South Dakota 22.7 -- 125 - 44 Michigan _- 22.5 124 100 Illinois 22.1 122 52 New t-lampshrre ---_____. ~~ 21 2 117 41 Nebraska 21.1 117 49 Idaho 20.7 114 41 New Mexico 20.6 113 141 Alaska -.____-~ 20.4 112 264 Vermont 18.7 103 23 Massachusetts ~~- .____- ____ 18.5 102 147 Georgia ____- 16.4 102 22 (continued) Page39 GAO/HRIHO89StatesHelpCommunitiesin FiscalDistress Chapter3 FederalRevenu&haring ReducedLocal FiscalDisparitiesMoreThanDid Most state programs In reducing fiscal disparities, the amount of general fiscal assistance aid Targeting More is less important than its targeting. The 10 states with the largest per Important Than capita assistance, as listed in table 3.2, reduced disparities about 25 per- Funding in Reducing cent, on average. However, this was not significantly different than the 10 states with the least per capita assistance, which reduced disparities Fiscal Disparities by about 14 percent, on average. While the states with high per capita amounts of aid achieved 11 percent more disparity reduction, they aver- aged almost eight times more assistance than that provided by states with low amounts of per capita aid.” Vermont and Georgia reduced disparities more than the national aver- age with per capita assistance aid at about one-third the national aver- age. These states could have accomplished such reductions only by highly targeting their aid programs to their most fiscally distressed gov- ernments. Such success demonstrates that improved targeting of state aid programs can meaningfully reduce local disparities, even with rela- tively few dollars. Among the 16 states with the widest disparities shown in table 2.3,9 Disparity Reduction had state programs that reduced disparities more than the national Mixed in States With average (see table 3.3). The programs in four states (Alaska, Arizona, Widest Disparities New Jersey, and New Mexico) reduced disparities more than the national average because of their comparatively high funding levels. General assistance grants in these states equaled or exceeded $120 per capita-over two-and-one half times the national average. Florida achieved the largest reduction (21.8 percent) through a combination of above-average funding ($75 per capita) and targeting to disadvantaged communities. The four other states with above-average reductions (South Dakota, Maine, Louisiana, and Tennessee) provided below- average funding, ranging between $25 and $30 per capita, but had better-than-average targeting to disadvantaged localities. “Thecorrelationbetweenall states’percentage disparityreductionsandtheir percapitaassistant wasa low .32.Thissignifiesa weakstatisticalrelationshipbetweenfundinglevelsandthesizeof disparityreductions. Page41 GAO/HRD-SO.69 StatesHelpCommunitiesb FiscalDistress chapter 3 Federal Revenu&harlng Reduced Lacal Fiscal Dlsparlties More Than Did Most state Proglam.9 sharing in 17 states, only Arkansas had a program whose targeting effectiveness was superior to that of federal revenue-sharing. In a majority of the 31 states, federal revenue-sharing did better even though the per capita amount of its assistance was less (see table 3.4). Tebte 3.4: Reduction in Fiscal Disparities Attrtbutable to State and Federal General State program Federal revenue-sharing Ftscal Assistance, by State (Fiscal Year Disparity Disparity Per capita 1985) reduction Per capita reduction grant States (percent) grant (percent) (percent) United States 8.0 $47 11.0 $19 Federal revenue-sharing superior: West Virginia __~~.. (2.4) 6 37.3 23 Mississiaoi, 1.4 63 24.1 23 Maine 17.5 27 22.2 21 Alabama -- 5.5 18 21.6 19 Idaho 4.8 22 18.8 19 North Carolina (0.5) 30 170 20 Vermont 2.5 2 16.9 21 Kentucky --(0.3) 2 15.9 20 Georaia 3.8 3 15.5 19 Rhode island 11.8 31 15.2 21 Texas 0.6 3 15.0 15 Michigan 12.2 79 13.7 21 New Mexico 12.4 120 13.3 21 Wisconsin 6.6 175 13.0 22 North Dakota 0.7 47 12.8 18 Utah 1.3 1 12.4 22 New Hampshire 11.9 29 12.3 12 Pennsvlvania 0.9 3 12.1 18 Missouri __~~~~ ~---. 1.1 1 11.0 16 Wyoming 9.5 159 10.9 31 Oklahoma 3.3 3 10.2 17 Maryland 5.6 33 10.0 20 Kansas __.~ 5.1 19 9.0 16 Ohro 8.1 36 8.8 18 (continued) Page 43 GAO/lllUh9C59 States Help Communities in Fiscal DWresa Chapter 3 Federal RevenueSharing Reduced Local FiscalDisparitiesMore ThanDid Most state programs in Kentucky, North Carolina, and Montana also were primarily return- to-origin programs. The main reasons that state general fiscal assistance programs generally did not reduce local fiscal disparities as effectively as did federal reve- nue-sharing were: 1. Most state programs were not designed to achieve disparity reduction as was intended by the federal revenue-sharing program. This is espe- cially true of state programs that distribute funds on a return-to-origin or per capita basis. 2. General fiscal assistance programs may be poorly designed uninten- tionally and thus not achieve the fiscal disparity reduction objectives desired by state policymakers. The expiration of federal revenue-sharing in October 1986 ended the State Reaction to Loss annual flow of $4.6 billion to local general purpose govermnents. The of Federal Revenue- most common state response was to do nothing. Other reactions included Sharing Minimal creating new general fiscal assistance programs, increasing local taxing authority, and better targeting of funds. We identified 13 states, as of 1987, that had responded to the loss of general revenue-sharing. Their responses varied. For example, Rhode Island and Delaware created new general assistance programs. Massachusetts increased funding for its major general fiscal assistance program. Connecticut officials cited the termination of federal revenue-sharing as a contributing factor in the expansion of existing general assistance programs and creation of a new one. And New Jersey expanded one of its aid programs for its municipalities. Other states, such as Illinois and North Carolina, responded by increas- ing the taxing authority of local governments. This left the decision to replace revenue-sharing funds to the individual communities. However, this approach will have little effect on fiscal disparities because it does not increase the revenue-raising capacity of distressed communities compared with those that are better-off. Another approach was to increase the targeting of funds to distressed communities, as Rhode Island did. In 1987, Rhode Island had seven sep arate general fiscal assistance programs, each with its own distribution formula. In 1988, the state eliminated these programs and created a new Page 46 GAO/HRD99+9 States Help Ciunmunities in Fiscal Mstress page 47 Appendix I Definition and Measurement of Local Fiscal Disparities to bear to finance a “foundation” service level. One such foundation level might be the state-wide average of local government expenditures, or perhaps 25 percent or some other percentage of the state-wide aver- age. A high tax burden needed to finance a foundation service level would indicate a poor fiscal condition; a relatively low tax burden, a good fiscal condition. Fiscal disparities then would be defined as differ- ences in these tax burdens compared with the state-wide average of all such tax burdens. We chose the power-equalizing criterion because: 1. Unlike the foundation approach, an analyst need not make a value judgment about what should be the foundation service level a state’s general assistance program guarantees local governments. 2. The power-equalizing criterion explained the distribution of state gen- eral assistance grants better than the foundation criterion in 9 of the 11 states included in our field visits. If fiscal disparities are defined as tax burden differences per dollar of Measuring the services provided across communities, a summary statistic is needed to Reduction in Fiscal express these differences. From various statistical measures of disper- Disparities sion, including the range, interquartile range, and coefficient of varia- tion, we chose the standard deviation. Because the standard deviation averages differences between each observation and the average value, it represents an absolute measure of dispersion. This was more appropriate for our analysis because small absolute differences in tax burdens are of no policy significance even if relative differences are large. If grants finance a large share of local ser- vices, local tax burdens will be small, and small tax burden differences need not concern us. The standard deviation will measure the average size of these differences. In contrast, the coefficient of variation expresses the standard deviation as a percent of the average value. It would indicate large relative differ- ences in local tax burdens even though absolute differences were small. This would lead to the conclusion that disparities were large even though they were of little policy significance. The range and interquar- tile range were rejected as summary statistics because they use only two observations in measuring the dispersion in tax burdens per dollar of public service benefits. Page 49 GAO/HRD90.69 States Help Communities in Fiscal Distress Appendix I Deflnltion and Measurement of Local Fiscal Disparities most influential observation deleted. After deleting the next most influ- ential observation, we continued this process until the estimated dispar- ity reduction stabilized (i.e., did not change substantially when successive observations were deleted). When analyzing state general assistance, we deleted observations only if doing so resulted in showing a larger disparity reduction for the state program. Thus, our analysis tends to overstate the disparity reduction provided by state general assistance. We did this to show the state pro- gram in the most favorable light. In the case of federal revenue-sharing, we did the opposite, deleting influential observations only if they reduced the amount of disparity reduction. This resulted in under- estimating the disparity reduction achieved by the federal program and ensured that our conclusion regarding the superior targeting of the fed- eral program is a conservative one. After completing our sensitivity analysis, we deleted no observations in 35 of the 48 states included in our analysis and no more than four in any 1 state. Thus, relatively few observations had to be deleted. The results of our analysis are shown in table 1.1. The standard devia- tion in tax burdens per dollar of services in the baseline (i.e., ult/e] = a[l/y]) is shown in the first column of numbers. Column 2, which expresses each state’s standard deviation as a percent of the U.S. aver- age, represents the extent of fiscal disparities in each state compared with the national average. These figures were reported in table 2.3 (ch. 2). The third column shows the standard deviation after the receipt of general fiscal assistance aid. The fourth column, the percentage dis- parity reduction (i.e., the percentage difference in standard deviations with and without general assistance aid) that were reported in column 1 of table 3.2 Page 61 GAO/-9989 States Help Commnnitie~ in Fiwal %trese Appendix I Def¶nition and Measurement of Local Fiscal Disparities Disparity Index reduction State Before (U.S.=lOO) After (percent) South Dakota 0.196 148 0.152 22.7 Tennessee .~- ~- 0.257 127 0.189 26.4 Texas 0.259 128 0.219 15.5 Utah 0.228 113 0.197 13.6 Vermont ~____ 0.272 -- 134 0.221 18.7 ~.. Virginia 0.225 .- 111 0.198 12.2 Ghmaton 0.162 80 0.141 12.7 West Virginia 0.216 -107 0.142 34.4 Wisconsin 0.179 -. 88 0.151 15.9 Wyoming 0.122 - 60 0.102 16.6 Source: U.S Bureau of the Census, “Revenue-Shanng Allocation File for Entitlement Period 17” and “Tax and Intergovernmental File for Fwal Year 1984/85” (computer-based flies). Another possible approach was to analyze the reduction in fiscal dispar- An Alternative ities provided by state and federal general fiscal assistance grants net of Approach Not Taken the state and federal taxes used to finance them. We did not do this because we could not identify the specific taxes used to finance these programs and obtain the data on a county-by-county basis. This “net fiscal incidence” analysis would produce different results, depending on the progressivity of state taxes. For example, if the inci- dence of state taxes used to finance state programs is more regressive than those used to finance the federal program, our methodology would understate the disparity reduction of the federal program compared with state programs. If the reverse is true, that is, state taxes are more progressive than federal taxes, our analysis would understate the dis- parity reduction of state programs compared with the federal program. Similarly, if the progressivity of state taxes is greater in one state than another, this could alter the ranking of states in table 3.2, where they are ranked by how much their general assistance programs reduced disparities. Page 53 GAO/HRDB@99 States Help Communities in Fiscal Distress Appendix II General Piscal Aesi.9tance Progrllma in caufomia Funding Level $76.6 million for state fiscal year 1986-87. Revenue Source 30 percent of the 10 cent tax on each package of cigarettes. Key Allocation Factors Local sales taxes and population. Formula Return to origin and per capita needs. The revenues first are split between counties and cities according to their share of the local sales tax. The funds allocated to counties are based solely on their share of the local sales tax. Of the funds allocated to cities, 50 percent is distrib- uted on the basis of population and the remaining 50 percent on the basis of each city’s share of the local sales tax. Mobile Home and Commercial Coach License Fee Program Objective To prevent revenue losses to local governments and school districts due to the state’s administrative takeover of this program from the counties. Funding Level $13.6 million for state fiscal year 1986-87. Revenue Source An annual license fee on mobile homes and coaches equal to 2 percent of their market value. Key Allocation Factors Market value of mobile homes and coaches. Formula Return to origin. Funds are distributed to cities, counties, and school dis- tricts according to the location of the mobile home or commercial coach being taxed. Taxes on vehicles located within a city are split evenly between the city, county, and school district. If the vehicle is located Page 56 GAO/HRD9949 States Help Commnnlties in Fiscal Distress Appendix U General Fiscal Assistance Pmgmw in califomia Funding Level $338.9 million for state fiscal year 1986-87. Revenue Source State appropriation. -. Key Allocation Factors Not applicable. Formula Return to origin. Localities are fully reimbursed for the property tax revenues lost by the exemption of a homeowner’s first $7,000 of assessed valuation. Open Space Subventions Program Objective To partially compensate local governments for the property tax revenue lost by assessing land on the value of its limited use (i.e., open space or agricultural) rather than full market value. Funding Level $14.9 million in state fiscal year 1986-87. -- Revenue Source State appropriation. Key Allocation Factors Land acreage. Formula Return to origin. Localities receive funds based on a partial reimburse- ment for revenue lost under this program as follows: Cities 1. Prime agricultural land in cities with populations over 25,000 at $8.OO/acre. 2. Prime agricultural land in cities with populations between 15,000 and 25,000 at $5.00/acre. Page 67 GAO/HUD-99-69 States Help cOmmunities in Fiscal Distress Appendix II General Fkcal AFd3tance Programs in callfonlia 2. $28.1 million on a proportional basis to the cost-of-living adjustment to which counties would have been entitled for 1987-88 for certain health programs if one had been provided. 3. $27.4 million on a basis proportional to the cost-of-living adjustment counties would have received in 1987-88 for the Medically Indigent Ser- vices, Community Services Block Grant, Aid to Families with Dependant Children (Am), Foster Care, and the Community Mental Health Program. 4. $27.4 million according to a number of factors relating to county costs for state-mandated programs. For the second year of the program, funds are distributed as follows: The state pays a portion of a county’s nonfederal share of costs for spec- ified mandated programs (AFLXJ,Food Stamp Administration, Commu- nity Health Services) that exceeds the percentage of the county’s expenditures of general revenues for those programs in fiscal year 1980-81. page 59 GAO/HRlMO89 States Help Cmnmunltiea in Fiscal Mstress Appendix Ill General Fiscal Assistance k%xQmms inxanslw Funding Level $19.6 million for state fiscal year 1987. Revenue Source State appropriation equal to 3.5 percent of the state sales and use tax credited to the State General Fund during the preceding calendar year. Key Allocation Factors Population and assessed valuation of real and personal property. Formula Return to origin/per capita needs, as follows: 1. Funds first are distributed to county areas. Of this distribution, 65 percent is based on the county’s share of the state’s total population and 35 percent on the county’s share of the total assessed valuation of real and personal property in the state. 2. The county government receives half of the funds allocated to the county. The remaining 50 percent is divided among all cities within the county according to population. Private Club and Drinking Establishment Liquor Tax Program Objective To share with local governments the tax revenue collected under this program. Each local government must allocate one-third of the proceeds to its general fund, one-third to its parks and recreation fund, and one- third to a substance abuse fund. Funding Level $7.6 million for state fiscal year 1987. Revenue Source 70 percent of the state’s 10 percent tax on alcoholic drinks served by clubs, caterers, and drinking establishments. Page 61 GAO/lIRD9069 States Help Communities in Pisal Disiress General Fiscal Assistance Programs in Massachusetts Additional Assistance Program Objectives To equalize fiscal disparities between local governments, maintain the ability of communities to provide essential services, and provide prop erty tax relief. Funding Level $714.7 million for state fiscal year 1987 (July 1 to June 30). Revenue Source State appropriation. Key Allocation Factors Needs and fiscal capacity Formula Tax base equalizing. Only increases in funding are allocated each year by the formula below. Every community’s base level of funding is guar- anteed to equal what it received the year before. Aid increases must be at least 50 percent of the increase received the year before, but cannot be more than 50 percent greater than that increase. Beginning in fiscal year 1988, Boston’s aid is set by the legislature and not subject to the formula. The basic formula is: Need = cost of local services - local revenue capacity. Massachusetts develops a cost index for every community within the state. Starting with a statewide average cost of providing services, the state adjusts each community’s cost up or down according to eight cost factors: 1. Weighted full-time student population. Students with special needs, and bilingual, vocational, and AFDCstudents receive extra weight in the formula. 2. Population density. 3. Manufacturing employment. Page 63 GAO/HlD9989 States Help Communities in Fiscal Diatrees Lottery Distribution Program Objective To equalize fiscal disparities between local governments, maintain the ability of communities to provide essential services, and provide prop- erty tax relief. Funding Level $195.0 million for state fiscal year 1987. Revenue Source Funded by net receipts from the state lottery. Key Allocation Factors Property tax revenues and population. Formula Tax base equalizing. The entitlement for each community is determined by its population and property tax revenues as follows: Population x 10 x revenues, where revenues is the statewide average per capita equalized property tax levy expressed as a percent of each community’s equalized revenues. This entitlement is adjusted by a coefficient to bring allocations calcu- lated for each community in line with funds available: Coefficient = funds available for distribution/ah entitlements statewide. Page 65 GAO/IlltD9669 States Help Commnnitiw in Fhcal Distress Homestead Credit Program Objective To facilitate the ownership of family homes. Funding Level $598.0 million for fiscal year 1987. Revenue Source State appropriation. Key Allocation Factors Market value of homesteads. Formula Return to origin. Local governments and school districts are reimbursed by crediting 54 percent of the property tax payment on the first $68,000 of market value for each homestead within their jurisdiction. The maxi- mum credit per homestead was capped at $700 for fiscal year 1987. Taconite Homestead Credit Program Objective To compensate homeowners in areas where taconite (low-grade iron ore) production companies pay production taxes to the state as opposed to local property taxes, from which they are exempt. Funding Level $11.2 million for state fiscal year 1987. Revenue Source State appropriation of t,aconite production taxes. Key Allocation Factors Local property taxes Page 67 GAO/HUD-SW39 States Help Communities in Fiscal Distress Appendix VI General Fiscal Assistance Programs in New Jersey Public Utilities Franchise and Gross Receipts Tax Program Objective To compensate local governments for the state preemption of taxation of public utility property and to share with them the franchise fees paid by public utilities. Funding Level $685 million for state fiscal year 1986. RevenueSource State appropriation based on state taxes assessed on public utility operations. Key Allocation Factors Value of utility property. Formula Return to origin. Proceeds of this tax are distributed to municipalities according to their share of the public utility’s scheduled property within their jurisdiction. Maximum aid is limited to $700 per capita or 75 per- cent of the aid received in 1979, whichever is greater. Business Personal Property Tax Replacement Revenue Program Objective To compensate municipalities for the elimination of the local property tax on business personal property. Funding Level $158.7 million for state fiscal year 1986. Page 69 GAO/HEB90.69 States Help Communities in Fkcal Distress Appendix VI General Fiscal Assistance Progrm in New Jersey Bank Corporation Business Tax Distribution Program Objective To return to municipalities 25 percent of the taxes collected under this program. Funding Level $16.2 million for state fiscal year 1986. Revenue Source State appropriation equal to 25 percent of the bank corporation business taxes collected. Key Allocation Factors Bank deposits Formula Return to origin. Municipalities receive 25 percent of bank corporation business taxes according to their share of total in-state deposits held by banks in offices within their jurisdiction. Financial Business Tax Distribution Program Objective To return to municipalities 25 percent of the taxes collected under this program. Funding Level $1.6 million for fiscal year 1986. Revenue Source State appropriation equal to 25 percent of the financial business taxes collected. Page 71 GAO,‘HlDgO89 States Help cOmmdtiea in lXsc.4 Distress Appendix VI General Fiscal Assistance Programs in New Jersey Revenue Source State appropriation from proceeds of the state income tax. Key Allocation Factors Population. Formula Per capita needs. To be eligible, a municipality must have an effective property tax rate exceeding $1 per $100 of valuation. Funds are allo- cated to all eligible communities on a per capita basis. Municipal Purposes Tax Aid Program Objective To provide property tax relief. Funding Level $30.0 million for state fiscal year 1986. Revenue Source State appropriation funded by unapportioned proceeds from the public utility franchise and gross receipts tax. Key Allocation Factors Population and equalized assessed valuation. Formula Per capita needs/tax base equalizing. Municipalities may qualify for funds from either of two separate allocations: 1. 23/27ths of the total allocation is distributed to eligible municipali- ties; 50 percent is based on population and 50 percent on the extent to which their per capita equalized assessed valuation is less than the state average. To be eligible for funds from this distribution, a municipality must have had a property tax rate equal to or greater than the state average for the previous year and its per capita equalized assessed valu- ation must be less than 90 percent of the state average. Page 73 GAO/HRD-99.69 States Help Chnmunities in Fiscal Distress Appendix VI GenenllFisealAsslstan~ProgralMhl New Jersey Distressed Municipalities Program Objective To aid distressed municipalities (enacted starting fiscal year 1987). Funding Level $17.5 million for fiscal year 1987. Revenue Source State appropriation. Key Allocation Factors Need and fiscal capacity. Formula Per capita needs. To be eligible, cities must qualify as distressed under such criteria as high property tax rates, low equalized property values, AFDCpopulation, and population. The Department of Community Affairs distributes the funds to eligible cities as it sees fit. There is no specified allocation formula. Page 75 GAO/HlKMS49 States Help Communities in Nacd Distress AppendixvII General Fiscal Assistance Pmgram in NewYork New York City There is no special formula. New York City which encompasses five counties, receives both the county and city allotment. State-Local Revenue- Sharing Program- Special City, Town, and Village Aid Component Objective To provide special unrestricted aid to all general purpose governments in the state except counties and New York City. Funding Level $96.4 million in fiscal year 1987-88. Revenue Source State appropriation. Key Allocation Factor Population, fiscal capacity, and land area. Formula Per capita needs/tax base equalizing. The allocation for towns and vil- lages is baaed on land area, 1979 local revenues, 1980 Census popula- tion, and 1980 full valuation of taxable real property. The allocation for 53 cities uses the 1970 Census population and 1979 full value real prop- erty tax rates. The remaining eight cities receive specific amounts as cited in the legislation. State-Local Revenue- Sharing Program- Excess Aid Component Objective To provide revenue for the general purposes of local governments. Page77 Appcmdix VU General Fiscal Assistance Program in New York New York City The city receives 40 percent of the total allocation. Cities, Towns, and Villages The locality’s share (49 percent of the total) is based on its percentage of the local revenues of all localities. Emergency Financial Assistance to Eligible Municipalities Objective To provide emergency financial aid to eligible localities in the state. Eli- gible local governments are Erie County, Buffalo, Niagara Falls, Yonkers, Rochester, and Syracuse. Funding Level $36.2 million in fiscal year 1987-88. Revenue Source State appropriation. Key Allocation Factor Fiscal capacity Formula Per capita needs. Eligible localities receive allocations as specified in appropriation legislation. Emergency Financial Aid to Certain Cities Objective To provide financial assistance to all cities with populations above 100,000 and below l,OOO,OOOthat have constitutional tax limits and/or large amounts of tax-exempt property. Funding Level $28 million in fiscal year 1987-88. Page 79 GAO/llElM989 States Help Conununities in Firscal Distmes Appendix VIII General Fiscal Assistance Programs in Rhode Islmd Note: We have not described the seven general fiscal assistance pro- grams that existed in Rhode Island for fiscal year 1987. Instead, we have described the new state general revenue-sharing program that replaced those seven programs in fiscal year 1988. State General Revenue-Sharing Objective To provide state aid to local governments. Funding Level $37.1 million in fiscal year 1987-88. RevenueSource State appropriation for state fiscal year 1987-88 equal to 6.1 percent of the state’s combined sales and income tax receipts. Future year alloca- tions are scheduled to increase by 5.5 percent. Key Allocation Factor Population, tax effort, and per capita income. ?ormula The state used the federal revenue-sharing formula as the basis for the new program’s distribution mechanism. The amount each municipality received was based on: Municipality’s population x tax effort x income f total state’s popula- tion x tax effort x income Regardless of the allocations generated by this formula, each municipal- ity was guaranteed at least 10 percent more aid than it received from the seven programs in fiscal year 1986-87. Page 81 GAO/HED&M9 States Help Communities in Pimd Mstmss Formula Return to origin. Revenues are distributed to the city in which a tax- payer resides. If a taxpayer resides outside the corporate limits of any city, the funds revert to the county of residence. Mixed Drink Tax Program Objective To share tax revenue with cities and counties. Funding Level $10.0 million for state fiscal year 1986-87. Revenue Source 50 percent of the 15percent tax on mixed drinks. Key Allocation Factors Property taxes paid. Formula Return to origin. Half of the revenues earmarked for local distribution are allocated between cities and counties according to their share of the county property tax for schools. The remaining half is returned to the city where the tax is collected. If the tax is collected in an unincorpo- rated area, it goes to the county. Alcoholic Beverage Excise Tax Objective To share tax revenue with counties. Funding Level $5.9 million for state fiscal year 1986-87. Page 83 GAO/HltlHJM9 States Help c0mmunitie.a in Piad Dl~trew AppendixIX General F%3cal ARaietanfe PrograIn inTennessee Funding Level $4.2 million for state fiscal year 1986-87. Revenue Source 97 percent of the severance tax on coal is earmarked for counties. One- third of the severance tax on crude oil and natural gas production is earmarked for counties. Key Allocation Factors Severance tax collections. Formula Return to origin. Funds are distributed to counties where the tax was collected. TennesseeValley Authority In Lieu of Taxes Program Objective To compensate local governments for the difference between actual tax losses resulting from the exemption of property owned by Tennessee Valley Authority (TVA) and what local governments receive directly from the TVA. Funding Level $39.8 million for state fiscal year 1986-87. Revenue Source Portion of the money received by the state from the WA. Key Allocation Factors Assessed value of TVA property. Formula Return to origin. The state determines what tax revenue local govern- ments would have received if TVA property was taxable. Subtracting what counties and cities receive directly from TVA, the state pays the difference. Page86 Appendix XI General Fiscal Assisti3ncePrograms in Vermont Vermont had no single general fiscal assistance program exceeding $1 million in fiscal year 1987. However, it has developed the Property Tax Rebate Program to ensure that individual residents are not overburdened by high property tax rates. The primary local revenue source used to finance local services is the property tax. The community imposes the necessary property tax rate to raise the revenue to pay for its services. Vermont refunds to residents the property tax paid that exceeds a set percentage of their household income. The percentage var- ies with income as follows: l Under $4,000-3.5 percent. . $4,000-7,999-4.0 percent. l $8,000-l 1,999-4.5 percent. l $12,000 and over-5.0 percent. The program is open to all Vermont residents, including both homeown- ers and renters. For renters, the state considers 24 percent of the rent paid as property taxes. In 1988, the state rebated almost $13 million to Vermont residents for property taxes paid in 1987. Of this amount, about $11 million went to households with incomes less than $20,000. Vermont’s program ensures that no homeowner or renter pays property taxes that exceed 5 percent of household income. A community may impose a high tax rate, but the state will, in effect, pay that portion of the tax rate that exceeds the applicable percentage of a resident’s income. In addition, it provides property tax relief only to residents. Nonresidents (who own 20 percent of the market value of the property in the state), do not benefit from this program because the state restricts eligibility to full-time Vermont residents. Page 87 GAO/HlDWf39 States Help Communities in Fiscal Distress Appendix XII General Fiscal Assistance Program in Wisconsin Counties receive aid under the same formula except that only 85 percent of the 3-year average of local purpose revenues is used, as opposed to 100 percent. 3. Utilities. The annual payment of this component is designed to com- pensate local governments for the cost incurred for providing services to public utilities whose property is not subject to local taxation. The amount of the payment is determined by the location of the public util- ity property. If the property is located within a town, the town receives a payment equal to a tax rate of $3 per $1,000 of book value and the county payment is equal to a tax rate of $6 per $1,000 of book value. If the property is located within a city or village, the county’s tax rate is reduced to $3 per $1,000 of book value, while the city or village receives a payment equal to a tax rate of $6 per $1,000 of book value. The maxi- mum payment is limited to $300 per capita for a municipality and $100 per capita for a county. In 1987, $13.6 million was distributed under this component. 4. Minimum/maximum payment, designed to ensure stability in the pro- gram by preventing wide fluctuation of local aid payments in any 1 year. The minimum payment in any year may not be less than 95 per- cent of the previous year. These minimum payments are funded by establishing a maximum growth limit for each year. For 1987, this was 4 percent over what was received the previous year. In 1987, $14.3 mil- lion was redistributed under this component. State Property Tax Credit Program Objective To provide property tax relief. Funding Level $146.7 million for the general government tax credit for state fiscal year 1987. Revenue Source State appropriation. Page 89 GAO/HRD9O49 States Help Communities in Fiscal Distress Appendix XIII State and Federal Intergovernmental Aid to General Purpose Lo& Governments, by State (Fiscal Year 1985) Dollars in millions Categorical aid General assistance State Total aid Dollars Percent Dollars Percent United States $53.307 $37.999 71 $15.308 29 Colorado 595 521 88 73 12 -...- Pennsylvania I ,778 1,519 85 259 15 Virginia 880 751 85 129 15 New York 10,325 8,760 a 1,564 15 Washington 794 655 83 139 18 Californra 10,622 8,651 81 1,971 19 Missouri 446 361 a 85 19 Oklahoma 317 251 79 67 21 Kentucky 371 289 78 82 22 ..- Georgia 558 432 77 126 23 Utah 168 129 77 38 23 Maryland 985 755 77 231 23 Ohio _~ 2,477 1,896 77 580 23 Oregon 502 383 76 120 24 Montana 91 69 76 22 24 Vermont 46 34 74 12 26 Alaska 518 379 73 139 27 Texas 926 640 69 286 31 North Dakota 144 99 69 45 31 Iowa 570 387 68 183 32 Indiana 963 649 67 314 33 Kansas 245 160 65 86 35 Arkansas 253 164 65 90 35 Loursiana 591 380 64 211 36 Connecticut 371 235 63 136 37 Idaho 112 71 63 41 37 Alabama 404 256 63 149 37 Illinois 1,627 1,027 63 600 37 North Carolina 825 520 63 305 37 Michigan 2,441 1,533 63 908 37 Nebraska 209 131 62 79 38 Minnesota 1,671 1,026 61 645 39 Florida 2,640 1,554 61 986 39 Tennessee 578 353 61 225 39 Rhode Island _~ 107 58 54 50 46 New Hamoshire 86 46 54 40 46 (continued) Page 91 GAO/HRD8989 States Help Communities in Pisd Distress &endix XIV Major Contributors to This Report John M. Kamensky, Assistant Director, (202) 275-0653 Human Resources Jerry C. Fastrup, Senior Economist Division, Washington, Charles C. Tuck, Assignment Manager D.C. Gail C. Harris, Evaluator Virginia T. Douglas, Reports Analyst Frank E. Putallaz, Evaluator-in-Charge New York Regional RoJeanne Liu, Evaluator Office Robert R. Poetta, Evaluator Hector M. Castillo, Evaluator San Francisco Regional Office (118820) Page 93 GAO/‘HRD!40-69 States Help Communities in Fiscal D&tress Appendix XIII State and Federal Intergove-ntal Ald to General Purpme Local Gw-ents, by State (Fiscal Year lSS6) Categorical aid General assistance State Total aid Dollars Percent Dollars Percent Arizona 946 500 53 445 47 South Dakota 64 33 52 31 46 Wyoming 197 101 51 97 49 Wisconsin 1,666 930 50 936 50 Maine 107 51 48 56 52 West Virginia 105 49 46 57 54 Mississippi 383 158 41 225 59 South Carolina 319 130 41 189 59 New Mexico -. 332 132 40 200 60 Massachusetts 1,361 520 38 853 62 Nevada 266 59 22 207 70 New Jersev 1.202 206 17 996 63 Note: Total aid column does not necessarily sum to aid component columns due to rounding of categori- cal and general fiscal assrstance amounts. Excludes Delaware and Hawaii. Does not include aid received by county units of government in Connectcut. Maine, Massachusetts, New Hampshire, New Jersey, Rhode Island. and Vermont, where our units of analysis are municipalities and towns Source: U.S. Bureau of the Census, “Revenue-Sharing AllocatIon File for Entitlement Period 17” and “Tax and Intergovernmental Aid File for Fiscal Year 1984/W (computer-based files). Page 92 GAO/HRDM89 States Help Communities in F&al Distress Appendix XII General Fiscal A.eslstance Progrm in wlscon.9ln Key Allocation Factors Property tax levy. Formula Return to origin. This program is not open to counties. Municipalities share in the distribution according to their share of the 3-year average of the state’s total property tax levy for general government purposes. This distribution is varied by a minimum/maximum funding adjustment. No municipality may receive less than 90 percent of what it received the year before. The maximum growth in a municipality’s allocation is determined by the amount of aid left when the minimum payments have been made. Page 90 GAO/HEBW9 States Help Commnnitles in Pisal Distress Appendix XII General l?isd Assistance Program in Wisconsin Shared Revenue Program Objective To (1) provide property tax relief, (2) equalize the fiscal capacity of local governments, and (3) compensate localities for utility properties not subject to local taxation. Funding Level $779.4 for state fiscal year 1986-87. Revenue Source State appropriation. Key Allocation Factors Population and assessed valuation. Formula Per capita needs/tax base equalizing. This program distributes funds based on four allocation components: 1. Population. This per capita payment ensures that every municipality in the state will receive a payment. Counties receive no payment under this component. The amount received for state fiscal year 1986-87 was about $30.00 per cap&a; $142.7 million was distributed in 1987. 2. Aidable revenues, based on the fiscal capacity of local governments; $623.1 million was distributed under this component in 1987. The formula for municipalities is: Aid = 3-year average of local purpose revenues x tax base weight. (Local purpose revenues = local property tax levies plus certain other local revenues.) Tax base weight = 1 - equalized property value per capita/standard- ized valuation per capita. The standardized valuation per capita is some- what like a state-guaranteed tax base. It is set such that funds available for distribution under this component exactly match aid entitlements. For 1987, this was set at $32,800 per capita. Page 88 GAO/HRD9069 States Help Communities in Fiscal Distress Annendix X General F’iscailAssistance Programs in Texas Mixed Beverage Tax Program Objective To share with local governments the taxes collected by the state under this program. Funding Level $44.2 million for state fiscal year 1987. Revenue Source 25 percent of the state’s 12-percent tax on the serving of mixed drinks is earmarked for counties and cities. Key Allocation Factors Mixed drink sales. Formula Return to origin. Of the tax collected under this program, 12.5 percent is distributed to counties on the basis of where the tax revenue was gener- ated. Another 12.5 percent is distributed to cities on the basis of where the tax was generated. Page 86 GAO/lflllMW9 States Help Commdtie~ in I’%cal Diahmn Revenue Source 17.5 percent of the state’s excise tax on alcohol is earmarked for counties. Key Allocation Factors Population and land area. Formula Per capita needs. 75 percent of the funds is distributed to counties according to population. The remaining 25 percent is distributed to counties according to their share of the state’s square mileage. Beer Excise Tax Programs Objective To share tax revenue with cities and counties. Funding $2.6 million for state fiscal year 1986-87. Revenue Source 10.05 percent of the state’s beer excise tax is earmarked for counties and 10.05 percent for cities. Key Allocation Factors Population. Formula Per capita needs. The county share is divided equally among each county. The city share is distributed based on population. Severance Tax Programs Objective To share with counties the taxes collected under this program. Page 84 GAO/HltD9O89 States Help Communities in Fiscal Distress Appendix IX General Fiscal Assistance Programs in Tennessee Mixed Beverage Tax Program Objective To share tax revenue with cities. Funding Level $92.7 million for state fiscal year 1986-87. Revenue Source 4.6 percent of the state’s general sales tax is distributed to cities less a portion for the University of Tennessee. Key Allocation Factors Population. Formula Per capita needs. Distribution is based strictly on population. Individual Income Tax Program Objective To share tax revenue with cities and counties. Funding Level $23.1 million for state fiscal year 1986-87. Revenue Source 3/8ths of the state’s tax on individual’s dividend and interest income. Key Allocations Factors Income taxes paid by local residents. Page 82 GAO,TiBD9O89 State Help Communities In Fiscal Distrese Appendix W General Fiscal Assiitance Program in New York Revenue Source State appropriation. Key Allocation Factor Eligibility determined by population, fiscal capacity, and value of tax- exempt property. Formula Return to origin/per capita needs. Eligible localities receive allocations based on their tax losses sustained due to tax-exempt property. The exception is Albany, which receives $2 million. Page 80 GAO/HRD9O-O9 States Help Commmitie~ in Fiscal Distress Appendix VII General Fiscal Assistance Prognun in New York Funding Level $65.7 million in fiscal year 1987-88. Revenue Source State appropriation. Key Allocation Factor Population, fiscal capacity, and land area. Formula Aid is 36.67 percent of the difference between the base year (1984-86) and the projected year when comparing per capita revenue-sharing aid summed with Special City, Town, Village Aid paid to all eligible munici- palities. No excess aid is given when the difference is zero or less. State-Local Revenue- Sharing Program- Needs-BasedAid Objective To provide revenue for the general purposes of local governments. Funding Level $70.0 million in fiscal year 1987-88. Revenue Source State appropriation. Key Allocation Factor Unemployment rates. Formula Per capita needs, as follows: Counties The county allocation (11 percent of total) is split into two separate parts: One is based on a county’s share of the unemployed population of all the counties, the second on a county’s share of all social services reimbursements received by all the counties. Page 78 GAO/ISRJ%W9 State Help Communities in Fiscal Distress Appendix VII General F’iscalAssistance Program in New York State-Local Revenue- Sharing Per Capita Aid Program Objective To provide revenue for general purposes of local government. Funding Level $800.7 million for fiscal year 1987-88. Revenue Source State appropriation. Key Allocation Factors Population. Formula Per capita needs/tax base equalizing. Fifty percent of the funds are dis- tributed among all cities in existence as of April 1, 1968, on the basis of population. The remaining 50 percent of the funds is distributed as follows: Towns $3.55 per capita. Counties $.65 per capita when the average of per capita full value assessment and personal income is $19,637 or more. An additional $.05 per capita for each $245 this average falls below $19,637. Cities $8.60 per capita when the average of per capita full value assessment and personal income is $19,637 or more. An additional $.05 per capita for each $245 this average falls below $19,637. Villages $3.60 per capita when the average of per capita full value assessment and personal income is $19,637 or more. An additional $.05 per capita for each $245 this average falls below $19,637. Population of Towns Residing $2.05 per capita when the average of per capita full value assessment Outside Villages and personal income is $19,637 or more. An additional $.05 per capita for each $245 this average falls below $19,637. Page 76 GAO/HRD9669 States Help Communities in PiscaI Distress Appendix VI General Fiscal Assistance Programa in New Jersey 2. If a municipality is not eligible for funds under that allocation, it can receive funds from the second allocation, which distributes the remain- ing 4/27ths of the funding. To be eligible, a community must have had a tax rate (primarily, the property tax rate) in excess of 50 percent of the state average and a per capita equalized assessed valuation less than twice the state average. Funds then are distributed to eligible municipal- ities on the same basis as under the first allocation. Municipal (Urban) Aid Program Objective To provide assistance to distressed communities. Funding Level $40.1 million for fiscal year 1986. Revenue Source State appropriation. Key Allocation Factors Population, needs, and assessed valuation. Formula Per capita needs/tax base equalizing. Municipalities must qualify to receive funding under this program. Eligibility criteria include minimum population, high tax rate, low assessed valuation, and minimum number of AFN children. Funds are distributed among qualifying municipalities with 60 percent based on their share of AFDC children. Distribution of the remaining 40 percent is based on population, tax rate, and tax base data. Page 74 GAO/HlUM9-69 States Help Communities in Fiscal Distress Appendix VI General Flcal.4s8istance Program8 in New Jersey Key Allocation Factors Financial business taxes. Formula Return to origin. Municipalities receive 25 percent of the total financial business tax generated within their jurisdiction. Payment in Lieu of Taxes, State Property Program Objective To compensate for local services provided state-owned property. Funding Level $14.1 million for fiscal year 1986. Revenue Source State appropriation. Key Allocation Factors Assessed valuation of state property. . . ,. = -L-L.. Appendix Vl General Flsd Al3dst..¶nce Progralll.¶ in New Jersey Revenue Source State appropriation. Key Allocation Factors Prior taxes on business personal property. Formula Return to origin. Payment to municipalities is based on what annual tax revenue they generated from this tax prior to its repeal in 1966, supple- mented by additional state payments. Funding has been capped since calendar year 1977. Insurance Premiums Tax Distribution Program Objective To return to local governments the taxes collected under this program. Funding Level $20.2 million for state fiscal year 1986. Revenue Source State appropriation funded by the state tax on insurance premiums. Key Allocation Factors Prior taxes on insurance premiums. Formula Return to origin. Distribution to counties and municipalities equals the amount received under former program increased by the annual per- centage increase of all tax revenues under this program. Page 70 GAO/HBD9089 States Help Communities in Piscd Di8tmm Appendix V General Fiexd Assistance Prorpam~ in Minnesota Formula Return to origin. After the homestead credit is applied to local tax bills, local governments are reimbursed for 57 percent of the remaining prop- erty tax rate up to a maximum credit of $465 or 66 percent up to a maximum of $520. The use of 57 or 66 percent is based on such criteria as the value of iron ore produced and the proximity of the homestead to the mines. Fiscal Disparities Program Objective Allow all communities in the Minneapolis-St. Paul metropolitan area to share in the benefits of the area’s economic growth. Funding Level $1.511 million for fiscal year 1987. Revenue Source 40 percent of growth occurring in each metropolitan county’s commer- cial/industrial property tax base. Key Allocation Factors Population and equalized value of local property. Formula Tax base equalizing, calculated as follows: Distribution index = city’s population x 2 x average fiscal capacity/ city’s fiscal capacity Each city’s fiscal capacity is its equalized market value per capita. Page68 GAO/HED4O49 States Help Communities in Fiscal Distress Appendix V General Fiscal Assistaxe FVograms in Minnesota Local Government Aid Program Objective To reduce the level of local property tax rates and to address disparities between jurisdictions in tax effort and tax capacity. Funding Level $323.7 million for calendar year 1987. Revenue Source State appropriation. Key Allocation Factors Fiscal capacity and population. Formula Tax base equalizing. There are separate allocation formulas for cities, towns, and counties: 1. For cities, the fiscal need is estimated by averaging the sum of the property tax levy and local government aid for the past 3 years. The city’s fiscal capacity is then estimated by multiplying its property tax rate by its adjusted assessed valuation and the result subtracted from the city’s fiscal need. This is the city’s preliminary aid figure. Sub- tracting the aid received the previous year establishes the increase nec- essary to meet the city’s fiscal need completely. This figure is adjusted by the appropriation limit set by the state. This figure is compared with maximum aid figures determined by per capita aid amounts. The lesser of the two figures becomes the aid received by the city less the state’s costs for developing demographic data. 2. For towns to receive aid, they must have a property tax rate equal to or more than $1 per $1,000 of assessed valuation. They then receive a 4-percent increase over the greater of 60 percent of all aid received in 1983 or 100 percent of the local government aid received in the previous year. This is the aid received less the state’s costs for developing demo- graphic data. 3. Counties receive a 4-percent increase over the aid received in fiscal year 1986 less the state’s costs for developing demographic data. Page 66 GAO/HRBW9 States Help Communities in Pisal Distress Appendix IV General Fiscal Asdtsnce Pro@vuns in Massachusetts 4. Service and trade employment. 5. Road mileage. 6. Pre-1940 housing stock. 7. Population below the poverty level. 8. Service level. This variable reflects the different service levels between communities. It is designed to recognize that in small, rural communities, residents privately arrange services such as water and sewer that are provided by governments in larger communities. The state then develops the revenue capacity of each community according to five factors. These are: 1. Property tax capacity. 2. Motor vehicle excise collections. 3. State aid from the previous year. 4. Local reserve. This can be a surplus or shortfall. It equals (net free cash + overlay surplus) - (reserve cushion equal to 2.5 percent of spending or $100,000, whichever is greatest). 5. Hotel/motel tax capacity. The need of each community is determined by subtracting revenue capacity from the state’s cost calculation. Communities with negative need have their need set at zero. Since the increase in state aid will not meet the total statewide need determined, need is reduced by the per- centage of the statewide need that the increase does meet. This results in the covered need for each community. It is adjusted to ensure that no community receives 50 percent more or less than last year’s increase in aid. Page 64 GAO/HRIMW39 States Help Communities in Fiscal Distress Key Allocation Factors Tax revenues from alcoholic drink sales. Formula Return to origin. Proceeds are distributed to the source of the revenues. Counties receive only the tax collected in unincorporated areas. Cities receive the tax collected within their jurisdictions. SeveranceTax Program Objective To share with local governments the severance tax on crude oil, gas, and coal collected under this program. Funding Level $4.4 million for fiscal year 1987. Revenue Source 7 percent of the revenue raised by the state’s severance tax. Key Allocation Factors Severance tax revenues. Formula Return to origin. Counties receive funds equal to 7 percent of the state severance tax collected within their jurisdiction. Counties retain 50 per- cent of these funds, with the remaining 50 percent going to school dis- tricts within their boundaries. Page 62 Appendix III General F’iscailAssistanee Progmms in Kansas Local Ad Valorem Tax Reduction Fund Program Objective To reduce property tax levies. Funding Level $26.9 million for state fiscal year 1987. Revenue Source State appropriation equal to 4.5 percent of the state sales and use tax credited to the State General Fund during the preceding calendar year. Key Allocation Factors Population and assessed valuation of real and personal property. Formula Return to origin/per capita needs, as follows: 1. Funds first are distributed to county areas. Of this distribution, 65 percent is based on the county’s share of the state’s total population, the remaining 35 percent on the county’s share of the total assessed valua- tion of real and personal property in the state. 2. Each county’s allocation is divided among all property tax-levying subdivisions except school districts according to their share of the total property tax levy in the prior year. The county government is included in this distribution. - County-City Revenue- Sharing Fund Program Objective To compensate local governments for the elimination of their participa- tion in the cigarette, liquor enforcement, and domestic insurance compa- nies privilege taxes. Page 60 GAO/HRD.9089 States HelpCommunities in FiscalMstresa Appendix n General Fleeal Assistance ProgralM in California 3. Prime agricultural land in all other cities at l.OO/acre. 4. All other open space at $.4O/acre. counties 1. Prime agricultural land within 3 miles of cities with populations over 25,000 at $8.00/acre. 2. Prime agricultural land within 3 miles of cities with populations between 15,000 and 25,000 at $S.OO/acre. 3. All other prime agricultural land at $l.OO/acre 4. All other open space at $.4O/acre. County Revenue Stabilization Program Objective To help compensate counties for the rising cost of state-mandated pro- grams (enacted for fiscal year 1987-88). Funding Level First-year allocation was $110.3 million for state fiscal year 1987-88. Subsequent allocation is estimated to be $15.3 million for state fiscal year 1988-89. Revenue Source State appropriation. Key Allocation Factors Population and needs ( 1987-88), needs (1988-89). Formula Per capita needs. For the first year of the program, funds were distrib- uted as follows: 1. $27.4 million to counties according to population. Page 68 GAO/IUUWO59 States HelpCommunities in FiscalDistress outside a city, the funds are split evenly between the county and school district. Special Supplemental Subventions Program Objective To compensate cities for the revenue loss experienced due to the repeal of the tax on personal property. Funding Level $56.9 million for state fiscal year 1986-87. Revenue Source State appropriation -~ Key Allocation Factors Personal property tax collections. Formula Return to origin. Funds are distributed to cities according to their share of the revenue loss sustained due to the repeal of the tax on personal property. In 1984-85, cities received 50 percent of the aid received in 1983-84. This percentage has declined by 10 percent each year; funding is to be terminated by 1989-90. - Homeowners’ Property Tax Relief Program Objective To compensate local governments for the revenues lost due to the home- owner’s exemption equal to the first $7,000 of assessed valuation. Page 56 GAO/HRIMM9 States Help Communities in Fiscal Distress Appendix II General Fiscal Assistance Programs in California Motor Vehicle License Fee Program Objective To prevent revenue losses to local governments due to the state’s assumption of the personal property tax on motor vehicles. Funding Level $1,547 million for state fiscal year 1986-87. Revenue Source An annual license fee on motor vehicles equal to 2 percent of their mar- ket value. Key Allocation Factors Population. Formula Per capita needs. Program revenues are allocated as follows: 1.40.625 percent to counties on the basis of population. 2. 40.625 percent to cities on the basis of population. 3. 18.75 percent, primarily to counties according to their population and share of revenues from a tax on business inventories prior to its repeal. Cities that did not levy a property tax prior to Proposition 131 received a portion of the 18.75 percent. This distribution to cities was repealed beginning in fiscal year 1988-89. Cigarette Tax Program Objective To offset revenue losses by local governments when the state preempted the right to tax cigarettes. ‘Proposition13wasthe voterinitiativethat limit&dlocalpropertytaxesto 1 percentof Bgqegged valuation. Page64 GAO/HELM089StatesHelpC~nu~~unlti~in Fksl Dbtrea.8 Appendix I Defhition and Measurement of Local Fiscal Disparities Table 1.1:Standard Deviation of Local Tax Burdens Per Dollar of Expenditures, Disparity Before and After Receipt of General index reduction Fiscal Assistance, by &ate (Fiscal Year State Before (U.S.=lOO) After (percent) 1985) Alabama 0.196 94 0.145 26.1 Alaska 0.270 133 0.215 20.3 Arizona 0.183 116 0.124 32.3 Arkansas 0.204 101 0.130 36.1 California 0.163 80 0.137 15.8 Colorado 0 230 113 ..-0.208 9.5 Connecticut 0.176 87 0.150 15.1 Florida 0.295 146 0.214 27.5 Georgia 0.206 102 0.168 18.4 idaho 0 191 96 0.152 20.7 lllrnois 0 195 96 0.152 22.1 lndrana 0.137 67 0.121 11.5 Iowa 0.109 54 0 083 241 Kansas 0 142 70 0 123 13.2 Kentucky 0.322 159 0.271 15.7 Louislana 0.268 132 0 191 28.7 Maine 0.311 153 0.209 32.6 Maryland 0.168 93 0 161 14.6 Massachusetts 0.206 - 102 0 168 18.5 Mrchiaar ” 0.193 95 0.150 22.5 Minnesota 0.175 86 0 135 23.1 Mrssissipp 0.208 105 0 158 23.8 Mrssourr 0.243 120 .0214 12.0 Montana 0.149 74 0.141 5.3 Nebraska 0.164 81 0.130 21.1 Nevada 0.107 53 0.048 54.7 New Hampshire 0.215 106 0.170 21.2 New Jersey 0.249 -- 123 0.191 23.2 New Mexico 6.265 156 0.211 20 4 New York 0.177 87 0.164 7.0 North Carolrna 0.194 96 0.166 14.7 North Dakota 0.995 72 -- 0.091 8.1 Ohio 0.171 85 0.145 15.6 Oklahoma 0.251 124 -0.219 12.9 Oregon 0 123 61 0.104 15.7 Pennsylvania 0.154 76 0 134 127 Rhode Island 0 151 72 0.114 24.3 South Carolina 0.212 106 0.128 39.7 (continued) Page 62 GAO/HRD-99-69 States Help Communities in Fiscal Distress Appendix I Deflnltion and Measurement of Local Fiscal Disparities To measure the effect of general fiscal assistance grants on fiscal dispar- ities, we compared the dispersion in local tax burdens per dollar of ser- vice benefits with and without the receipt of general assistance grants. If tax burdens are represented by t (per capita taxes paid by residents, r, as a percent of residents per capita income, y) and public service bene- fits by e (per capita expenditures), the tax burden per dollar of services can be expressed as the t/e. The standard deviation in t/e ratios is repre- sented by u(t/e). The percent reduction in fiscal disparities is simply the percent change in this standard deviation with and without general fis- cal assistance grants. To calculate the reduction in fiscal disparities, we first defined a base- line t/e ratio by assuming that local governments finance all services from local revenue sources. Using this assumption, per capita own- source revenues are set equal to per capita expenditures (i.e., r = e, where r is per capita local revenues). The baseline ratio t/e is then equal to [(r/y)/r], or l/y. In the absence of general assistance aid, this result implies that the baseline measure of fiscal disparities is 41/y). Fiscal disparity after the receipt of grants is represented by a(t/e), where “t,” the tax burden, is defined as [(r-g>/y], where g is the per cap- ita grant whose effect is being assessed. If the distribution of grants is disparity-reducing, then by definition the dispersion in t/e ratios would be reduced compared to the baseline case. In fact, if grants were targeted to completely eliminate disparities in fiscal condition, postgrant t/e ratios would be completely equalized and u(t/e> would be equal to zero. We examined data on local tax burdens and per capita expenditures in Sensitivity Analysis each of the 48 states included in our analysis. In some instances, the Performed receipt of general assistance aid dramatically changed the t/e ratio for one or a few individual counties within a state. Because the standard deviation can be heavily influenced by extreme values, we did a sensi- tivity analysis, redoing the analysis with and without potentially influ- ential observations. In performing the sensitivity analysis, we ranked all observations by the change in their t/e ratio with and without grants. We then calculated the percent reduction in a(t/e) compared with the baseline, a( l/y), with the Page 60 GAO/HRB9O49 States Help Communities in l+ca~ D%J-WS Appendix I Definition and Measurement of Local Fiscal Disparities In this appendix, we discuss in more detail our approach to measuring the reduction in local fiscal disparities. We elaborate on the conceptual and methodological issues related to measuring fiscal disparities and the effectiveness of general assistance aid in reducing them. After present- ing two definitions of fiscal disparities, we explain why we selected one. Additionally, we explain why we chose the standard deviation statistic to measure how much disparities are reduced by state and federal gen- eral assistance aid. Finally, we describe the sensitivity analysis we per- formed to assure that any bias in our analysis is in the direction of showing superior state targeting. For this report, we use the so-called “power-equalizing” criterion to Fiscal Disparities define fiscal disparities among general purpose local governments. In Defined the objectives, scope, and methodology section of chapter 1, we first defined the fiscal condition of local governments within each county as the ratio of residents’ local tax burdens to their public service expendi- tures. Tax burden was defined as local taxes paid, expressed as a per- centage of residents’ personal income, and public service expenditures were used to approximate the public service benefits provided by a local government. By our definition, local governments with relatively high tax burdens per dollar of expenditures have a relatively poor fiscal condition. Con- versely, low tax burdens per dollar of expenditures denote a relatively good fiscal condition. For our analysis, we define fiscal disparities as differences in fiscal condition among local governments. In addition, we aggregated local government expenditures up to the county level.’ Thus, measurement of disparities was baaed on county averages. In this report, we refer to governments with a poor fiscal condition as “fiscally distressed” and governments in good fiscal condition as “better-off” communities. This criterion-equalization of average tax burdens per dollar of local government expenditures-also is known in the public finance litera- ture as a power-equalizing or a percentage-equalizing program. It is one of two criteria commonly used in designing grant programs aimed at reducing fiscal disparities. An alternative definition of fiscal disparity-used in foundation grant programs-is the difference in tax burdens local residents would have ‘Seefn.7,ch.1 Page 48 GAO/HBD99-09 States Help Communities in Fiend Distress Chapter3 FederalBevenueShariugReducedLocal FiscalDisparitiesMoreThanDid Most state Pmgrztms general fiscal assistance program, in part to help localities offset their loss of federal revenue-sharing funds. The state decided to use the federal revenue-sharing distribution formula to allocate all funds in its new program. In 1986, before the change in formula, Rhode Island’s general assistance programs reduced disparities by 11.8 percent. Federal revenue-sharing, being more targeted to disadvantaged local governments, contributed a 15.2-percent reduction (see table 3.4). By adopting the revenue-sharing formula in 1988 to allocate all its general assistance aid, we estimate that Rhode Island will reduce disparities by 25.1 percent. Thus, Rhode Island, by replacing its old formula with the more targeted federal formula, will substantially reduce disparities. The degree to which general fiscal assistance aid is targeted to fiscally Conclusion distressed communities is the factor most responsible for the disparity reductions achieved by state and federal programs. In 1986, state pro- grams provided more than twice the funding provided by federal reve- nue-sharing. However, they did not reduce disparities as much as the federal program because they were not as targeted to distressed commu- nities as was the federal program, nor in most cases were they intended to be. If states were to target more of this aid to distressed local govern- ments, they could further reduce the differences in tax burdens between distressed and better-off communities without additional funding. Page46 GAO/HRD46439 StatesHelpCommunitleain FincalDistress Chapter3 FederalRevenueSharingReducedLocal Fiscal DlsparltiesMoreThanDid Most state Programs State program Federal revenue-sharing Disparity Disparity Per capita reduction Per capita reduction States (percent) grant (percent) (per!Ei; Colorado 0.9 6 0.7 17 Indiana 3.9 42 8.6 16 Oregon 8.5 24 8.6 20 Washington 5.0 15 7.9 17 Virginia 4.7 5 7.8 17 Montana (0 3) 4 6.1 23 New York 23 62 5.0 26 State fiscal assistance superior: Nevada 52 0 212 10.0 16 South Carolma 27.2 37 23.3 21 Arkansas 25.6 19 15.9 20 Florida 21.8 75 9.1 15 Arizona 20.6 129 8.6 17 New Jersev 18.5 123 -9.9 10 South Dakota 18.0 25 7.1 20 Louisiana 17.4 30 13.6 Ii Alaska 16.6 223 5.6 41 Minnesota 16.6 134 9.2 21 Tennessee 16.2 30 13.4 18 Iowa 16.1 44 10.8 19 lllinors 15.2 34 8.9 18 Massachusetts 12.4 126 8.5 21 Nebraska 12.0 30 11.3 19 Californra 11.1 57 5.9 20 Connecticut 6.5 27 7.5 16 Note: States rn bold letters are the 16 states where local fiscal disparrtres are relatively large before receipt of grants (see table 2 3) Source: U S Bureau of the Census. “Revenue-Shanng Allocation Frle for Entrtlement Period 17” and “Tax and Intergovernmental Ard Frle for Frscal Year 1984/&J” (computer-based files) In four states (West Virginia, North Carolina, Kentucky, and Montana), general assistance programs marginally worsened disparities because they returned revenues to the originating jurisdictions, As discussed ear- lier, return-to-origin programs provide funding to localities in proportion to the size of their tax bases. In effect, this targets funds to better-off communities as opposed to disadvantaged ones. For example, in West Virginia, the state’s primary general fiscal assistance program distrib- uted coal severance taxes to local governments in counties where the coal production took place. Similarly, general fiscal assistance programs Page44 GAO/fIUD9O69StatesHelpCommunitiesin FiscalDistress Chapter 3 FederalRevenueSharingReducedLocal F&cd Disparttks More Than Did Most state Programs Table 3.3: Reduction in Fiscal Disparities in 16 States With the Widest Disparities Percent (pgagad by Per Capita Funding) (Fiscal Year disparity Per capita State ~~~~-..____ reduction grant U.S. Average 8.8 $47 Alaska 16.6 223 Arizona 20.6 129 New Jersey 18.5 123 New Mexico 12.4 120 Florida 21.8 75 Louisiana 17.4 30 Tennessee 16.2 30 Mame 17.5 27 South Dakota 18.0 25 Colorado 0.9 - Oklahoma ___~ _____~ 3.3 Texas 0.6 3 Kentucky (0.3) 2 Vermont 2.5 2 Missouri 1.1 1 Utah 1.3 1 Source: U S. Bureau of the Census, “Revenue-Sharing Allocation File for Entitlement Penod 17” and “Tax and Intergovernmental Ard File for Fiscal Year 1964/1985” (computer-based files) All of the remaining seven states whose general assistance programs had little effect on disparities provided relatively small amounts of assistance ($6 per capita or less). Overall, federal revenue-sharing was better targeted to disadvantaged Better-Targeted governments than state assistance. Despite funding levels almost two- Federal Aid Reduced and-one half times the size of federal revenue-sharing, state general Disparities More Than assistance grants reduced fiscal disparities less than the federal pro- gram. Nationally, state programs reduced disparities by 8.8 percent, the State Aid federal program 11 percent.” In 31 of the 48 states, federal revenue-sharing was more effective in reducing local disparities because it was more targeted. While state gen- eral assistance programs reduced disparities more than federal revenue- “To measurethereductionin fiscaldisparitiesprovidedby stategeneralfiscalassistanceandfederal revenue-sharingprograms,wecalculatedthestandarddeviationsin localgovernments’ tax burdens perdollarof servicesbeforeandafterreceivingaid.A decrease in standarddeviationsthenwas expressedas a percentage reduction. Chapter3 FederalRevenueSharingReducedLocal Fiscal DisparitiesMoreThanDid Most state FTcJgrluns Disparity reduction As a cercent of U.S. average Per capita State Percent (U.S.=lOO) general aid U.S. Average 18.1 100 85 Wyoming 16.7 92 189 Wisconsin 15.9 68 196 Californta 15.8 87 77 Kentucky 15.7 87 22 Oreaon -i5.7 86 45 Ohio - 15.6 86 54 Texas - .~~ 15.5 86 18 Connecticut 15.1 83 43 North Carolma 14.7 81 49 Maryland 14 6 81 53 Utah 136 75 23 Kansas i3.2 73 35 Oklahoma 12.9 71 20 Washington 12.7 70 32 Pennsylvania 12.7 70 22 Vtrgtnia 12.2 67 23 Missouri 12.0 66 17 Indiana 11 5 64 ~- 57 Colorado 95 52 23 North Dakota 8.1 45 65 New York - 7.0 39 88 Montana 53 29 27 Note States tn bold letters are the 16 states where local fiscal dtsparitres are relatwely large before the recetpt of grants (see table 2.3) Source. U S. Bureau of the Census, “Revenue-Shanng Allocatton File for Entitlement Period 17” and “Tax and Intergovernmental Atd File for Frscal Year 1984/85” (computer-based ftles). The national average reduction in fiscal disparities was 18.1 percent. Texas was slightly below the average at 15.5 percent. A few states did much better or worse. Disparities were reduced by more than twice the national average in three states: Nevada (54.7 percent), South Carolina (39.7 percent), and Arkansas (36.1 percent). However, they were reduced by less than half the national average in three others: Montana (5.3 percent), New York (7.0 percent), and North Dakota (8.1 percent). Page40 GAO/HRB9@69 StatesHelpCommunitiesin FiscalDistress Chapter3 FederaIBevenueSharingRedwedLocal FiscalDbparkIeaMoreThanMd Most state Programa The remaining four states were near the national average in terms of funding. To offset local disparities as much as states with large pro- grams, they would need to target a larger proportion of their aid to fis- cally distressed governments. In 1985, combined state and federal general fiscal assistance reduced General Fiscal local disparities among general purpose local governments by 18.1 per- Assistance Reduced cent (see table 3.2).3 Much of this reduction was due to targeting, not the Disparities amount of aid provided. The reductions varied widely across states- ranging from as much as three times the national average (Nevada, 54.7 percent) to as low as one-third of the average (Montana, 5.3 percent). A comparison of two counties in Texas, Starr and Wheeler, illustrates how the percent reduction in disparities, shown in table 3.2, can be interpreted. That state’s 15.5~percent reduction represents the average reduction in tax burdens borne by the average resident in each county compared with the average-income county. In Starr County, the average resident’s income in 1984 was the lowest in the state at $3,704. This compared with Wheeler County’s personal income per resident of $9,916, which was close to the state average income of $9,913.4 Before general fiscal assistance, expenditures for public services were $136 per resident in Starr County compared with $353 in Wheeler County. Yet, the tax burden borne by the average Starr County resident was slightly more than twice that of the average Wheeler County resident, when expressed on a per-dollar-of-service basis. If a resident of Starr County earning the average income and bearing the typical tax burden, moved to Wheeler County, he or she would pay $37.21 less in taxes for every $100 worth of public services. This differ- ence in tax burdens represents the fiscal disadvantage of the average- income Starr County resident compared with the average-income resi- dent in Wheeler County. State and federal general fiscal assistance aid received by Texas local governments reduced the fiscal disadvantage of low- compared with “Wemeasured disparitiesby thestandarddeviationsin localgovernmenta’ tax burdensperdollarof services.Wecalculatedthestandarddeviationwith andwithouttotal generalfiial assistance aid.In everystate,thestandarddeviationwassmallerafteraccountingfor generalfiscalassistance. The decreasewasthenexpressed asa percentage reduction. ‘In our analysis,Wheeler’s tax burdenperdollar’sworthof servicesbecomes the standardwith whichothercountyareatax burdensarecompared. Page38 GAO/HBDSO49 StatesHelpCommnnitiesin Eisfal Mstnarr chapter 3 FederalRevenueSharingReducedLocal FiscalDlspiulties MoreThanDid Most state ~grams Percent of local revenues State Combined State Federal Maine 16.1 9.1 7.0 Alaska 14.9 12.6 2.3 North Carolina 14.4 a.7 57 Tennessee 14.2 9.0 5.2 Iowa 13.5 9.6 4.0 Nebraska - 13.3 8.1 5.1 Rhode island 13.2 7.9 5.4 South Dakota - 13.0 7.2 57 New Hampshire 12.9 9.1 3.8 Alabama 12.7 6.2 6.5 U.S. Average 12.3 a.7 3.5 Illinois 12.2 8.0 4.3 Louisiana 12.1 7.7 4.5 Ohto ~~~__- - 11.3 7.6 3.7 Oregon 10.7 5.9 4.9 Californta ~~~~-.- 9.8 7.3 2.5 Connecticut 9.0 5.7 3.3 Vermont 8.8 0.6 8.2 Kentuckv 9.3 0.9 8.4 Montana ~~__. 9.0 1.3 7.7 Marvland A -.______-..- ~~~ 8.9 5.6 3.4 Kansas 8.7 4.8 3.9 Washmgton 73 3.4 3.8 Utah 7.1 0.5 6.6 Oklahoma. 6.6 10 5.6 New York 6.3 4.5 1.9 Georgta 5.9 0.8 5.1 Pennsvlvania 5.6 0.9 4.7 Texas 5.2 0.9 44 Virginta ___- 5.1 1.2 33 Missouri 5.0 0.3 4.7 Colorado 4.0 1.0 3% Note. States in bold letters are the 16 states rn whtch local ftscal disparities were relattvely large before allowing for the effect of general ftscal assistance grants (see table 2.3) Also, figures do not add due to rounding Source. U S Bureau of the Census, “Revenue-Sharing Allocation Ftle for Entttlement Period 17” and “Tax and Intergovernmental Atd File for Fiscal Year 19&t/&35”(computer-based files). Nationwide, state aid was more than twice that of federal revenue- sharing when expressed as a percentage of local revenues-g.7 com- pared with 3.5 percent. State aid exceeded revenue sharing as a share of Page36 GAO/HRD-90.69 StatesHelp Communitiesin FiscalDistress Chapter 2 Fiscal Disparities:Their Nature,Sources, and Extent Figure 2.4: Extent of Potential Fiscal Disparities Within Each State (1985) Relativelv tame diwarities Near average disparities Relatively small disparities Source: U.S. Bureau of the Census, “Revenue-Sharing Allocation File for Entitlement Perlod 17” and “Tax and Intergovernmental Aid File for Fiscal Year 1084/85” (computer-based files). Page34 GAO/HRD.9O-69 State Help Communitiesin PiscalDistress Chapter2 PiscalDisparities:Their Nature,Sources, and Extent Table 2.3: States Ranked by Extent of Local Fiscal Disparities, Assuming No State Index no. State and Federal General Fiscal States with relatively large disparities: Assistance (Fiscal Year 1985) 159 Kentuckv New Mexico 156 Maine 153 South Dakota 148 Flonda 146 Vermont 134 Alaska 133 Loursrana 132 Texas 128 Tennessee 127 Oklahoma 124 New Jersey 123 Mrssourr 120 Arrzona 116 Colorado 113 Utah 113 States with near average disparities Virginia 111 West Vrrarnra 107 South Carolina 106 New Hampshire 106 Mississippr 105 Massachusetts - - 102 Georgia 102 Arkansas 101 U.S. Averaae 100 Idaho 96 Illinois 96 North Carolrna 96 Michigan - 95 Alabama 94 Maryland 93 Wisconsin 88 New York 87 (continued) Page32 GAO/HRD-99.69 StatesHelpCkmmunitiesin F’imalDistress chapter 2 Fiscal Dlspmlties: Their Nature, Sonrces, and Extent Figure 2.2: Dispersion of County Tax Burdens Per $100 Dollars of Public Services in 67 Florida Counties (Fiscal Year 198!3) 2.0 Tax Burden Per $100 Dollars of Services Union Co 1.9 n d 1.6 1.7 n I n 1.6 n n m n 1.5 q n n n 1.4 n n n n mm n 1.3 n mm n n n n n n 1.2 l 1.1 n n mm n I n n n I 1.0 mm n n n n n 0.9 n n n n n n n 0.9 n n m n Palm Beach Co. 0.7 n d n 0.6 Counties Averagestate tax burden Note. The 67 counties are arrayed alphabetically on the horizontal axis. Source U.S. Bureau of the Census. “Revenue-Sharing Allocation File for Entitlement Period 17” and “Tax and Intergovernmental Ald File for Fiscal Year 1994/W (computer-based files) In contrast, the dispersion in tax burdens among Iowa counties is rela- tively small, with county tax burdens bunched more closely around the state average, as shown in figure 2.3. The tax burden per $100 of ser- vices was 0.7 percent in Polk County compared with 1.34 percent in Decatur County. Without outside assistance, the dispersion in local tax burdens in Florida would be about two and one-half times greater than among Iowa’s counties. Page 30 GAO/HlD9989 States Help Communltles in FIscal Mstresa chapter 2 FiscalDisparities:Their Nature,Sources, and Extent have not been annexed by adjacent cities. There is no economic incentive for municipalities to annex communities with weak tax bases, especially when more resources would have to be spent on services in these areas than would be gained in additional tax revenues. Unilateral or even negotiated annexation is not legal or feasible in all states. In New Jersey, for example, there are no unincorporated areas. For a local government to expand, it would have to be through consoli- dation with another local government. Revenue Restriction May Restricting local revenue-raising authority may increase the level of fis- Increase Disparities cal disparities, especially if local governments with low incomes and high public service costs are prevented from exporting taxes.g For exam- ple, New York City imposes a tax on all income earned within the city. This allows the city to pass some of its tax burden to nonresidents. Yet, most major cities lack the authority to levy an earnings tax on non- residents who work within their jurisdictions and presumably use many of the city’s public services. For example, in 18 of the 20 largest metro- politan areas, the central city had less tax-raising ability than the sur- rounding suburban communities in 1986.“’ As of 1988, slightly over half of these 18 cities were unable to compensate for this disadvantage by shifting some of their tax burden to nonresidents by means of a local income or payroll tax. If localities had to finance all services without outside aid, sizable fiscal Extent of Local Fiscal disparities would result and differ substantially among states. For Disparities example, in Iowa tax burdens per dollar of services received in the most distressed counties were about 70 percent greater than in the “best-off” counties. In Florida, tax burdens in Union County were nearly three times greater than in Palm Beach County. To estimate the potential extent of local fiscal disparities within each state, we initially assumed that local governments delivered and financed all of the public services now being provided exclusively from own-source revenues. Using this assumption, we calculated a county’s “Tax limitationinitiativessuchasProposition13in CaliforniaandPropositionZ-1/2in Massachusetts donot widenfiscaldisparities,becauseall localgovemments areaffectedequally. “‘SeeLocalGovernments: Tar etin GeneralFiscalAssistance Reduecs FiscalDisparities (GAOmD-86-113,July 19&p. !I). Page28 GAO/HRINO89StateaHelp Communitiesin PiscalDistress Chapter2 FiscalDisparities:Their Nature,Sources, andExtent State general fiscal assistance programs use various methods for target- ing aid,” e.g.: 1. Return to origin, which does little to reduce local disparities. Essen- tially, the state collects revenues from a specific tax base and returns all or a portion of those taxes to local governments according to the share of the tax base lying within its borders7 Tennessee, for example, taxes individuals’ dividend and interest income. Of the total amount collected, the state keeps five-eighths and returns three-eighths to the municipal- ity or county in which the individual resides. The grant component of this tax does not reduce disparities because more funding is provided to localities whose tax bases generate the most revenues. 2. Needs only, in which the state distributes funds on the basis of popu- lation or other indicators of social needs, such as poverty or unemploy- ment. By reducing the local burden of financing these needs, programs contribute to reduced local fiscal disparities. For example, in fiscal year 1986, New Jersey’s state revenue-sharing program distributed this aid to municipalities according to each jurisdiction’s share of state popula- tion (see app. VI). 3. Tax base targeting, in which fund distribution is baaed on the level of the local tax burden. It is more efficient in reducing disparities than either of the other two approaches because it can reduce the gap between better- and worse-off communities with less funding. In 1987, Massachusetts’ Additional Assistance Program used local governments’ fiscal capacities and data reflecting their differing levels of social needs to allocate funds. After setting a level of expenditures subject to aid, based on indicators of each local government’s social needs, the state estimates the amount of revenues that the local government should be able to raise locally. Then t,he state’s Additional Assistance Program partially fills the gap between the estimated levels of expenditures and revenues. Several states have created general fiscal assistance programs that use a combination of these three allocation methods to distribute funds. For example, Kansas’s County-City Revenue Sharing Program used needs and return-to-origin factors to allocate its aid. The program formula allocated 65 percent of the money among county areas according to their “Seeapp II-XII for descriptions of specificstateprogramandformulasin the 11stateswevisited 7Thisincludesonly thereturnof state-unposed taxes,not thereturnof taxeslocallyimposedbut collectedby the state An rmmpleof thelatter wouldbea localoptionsalestax. Page26 GAO/HRD9689StatesHelpCommunitiesin FiscalDistress chapter 2 Fiscal Disparities: Their Nature, Sources, and Extent pregnant mothers, to more broadly defined activities, such as commu- nity services. The $38 billion in such aid comprised 71 percent of state and federal grant funding in 1985. 2. General purpose fiscal assistance is unrestricted aid that may be used to fund whatever public services local governments deem necessary. Therefore, it is most easily designed to offset fiscal disparities. Nearly all states provided general fiscal assistance, which totaled $10.9 billion in 1985. The major federal program providing general fiscal assistance was the general revenue-sharing program, which allocated $4.6 billion among 39,000 local governments in 1985, the year before it was terminated. Combined, state and federal grants form a significant revenue source for financing locally provided services. Nationwide, they exceeded $53 bil- lion in 19854 and represented 34.8 percent of local revenues (see table 2.2). Grant financing of locally delivered services tended to be only slightly greater in decentralized states than in centralized states.” However, there were exceptions among decentralized states. In California, Ari- zona, and Nevada, the proportion of grants substantially exceeded the national average, but in most other large decentralized states, such as New York and Texas, the proportion was well below the national average. In addition to the level of grant funding, how such funding is targeted Grant Targeting Can also affects the extent of local fiscal disparities. Even when grants ReduceLocal finance a relatively small share of local services, disparities can be sub- Disparities stantially reduced if the funds are targeted to fiscally disadvantaged communities. In practice, categorical grants tend to be distributed according to pro- gram needs, not the ability of local taxpayers to finance services from local resources. Tennessee’s local highway aid program typifies this. It uses highway funds to repair and maintain about 12,000 miles of high- ways. The state finances about 75 percent of total costs, county govern- ments 25 percent. “Seeapp.XIII for a state-by-state breakdownof this figure. “A weaknegativestatisticalrelationshipexistsbetweenthedegreeof servicecentralization(table 2.1)andthedegreeof financecentralization(table2.2).Thecorrelationcwfficientis -0.20. Page24 GAO/HRLS9049 States Help Communities in Fiscal Distress chapter 2 Piseal Dlsparlties: Their Nature, Sources, and Extent Centralizing the delivery of public services at the state level has advan- tages and disadvantages. It can reduce the size of local fiscal disparities by reducing the reliance on unevenly distributed local tax bases to finance services.2 The major disadvantage is that state-provided services are likely to reflect the average preferences of all state residents. Conse- quently, the level of services provided in each community is less likely to be responsive to local preferences. Decentralized service delivery, on the other hand, allows local governments to respond to such differences. Nationwide, we estimate state governments deliver over half of all noneducation public services provided to state residents (see table 2.1). For Vermont, the most centralized state, the state government provided 79 percent of all noneducation public services in 1985. The most decen- tralized state is Nevada, in which the state government directly pro- vided 38 percent of all such public services. The range of service centralization shown in table 2.1 reveals a pattern among small and large states. States with small populations and land area tend to be centralized and provide services at the state level. This includes all six New England states which, except for Massachusetts and Connecticut, have relatively small populations. Decentralized states tend to be geographically large (Nevada and Arizona), more populous (New York and Florida), or both (California and Texas). About 11 percent of the U.S. population lives in the 12 most centralized states and about 44 percent in the 12 most decentralized states. Another means of reducing local fiscal disparities is through the provi- Level of Grant sion of grants used to finance local services. Decentralized states, where Financing Another local governments deliver a larger proportion of services, provided Factor grants that were only slightly larger than those provided in centralized states. Consequently, decentralized states are unlikely to offset local dis- parities to the same extent as more centralized states. 2Fiscaldisparitiescanbenarrowedby servicecentralizationif statesdeliverservicesonthebasisof relativecommunityneedsandtheincidenceof statetaxeswith respectto personalincomeis more progressivethanthe incidenceof localgovernment taxes. Page 22 GAO/HEN089 States Help Commnnitiea in Fiscal Dhtrecrs Chapter2 PiecalDlsparlties:Their Nature,So-, and Extent VMP 3 I. llhmtratinn , -. - -. . . . . . - - . . - . . -. . nt Fiacnl Disaaritv -. .___. ___r ____, Between - __.. -- Alaine and Woodbine . . r ..- -..- ..~ ~~~ Eorouahe. ~. ” , New Jersev (1983) .I I $0 $25,000 $50,000 Income per capita 0% .5 1 .o 1.5 2.0 2.5 3.0% Tax burden $0 $75 $150 $225 $300 Revenue per capita m Alpine Woodbine Fiscal disparities arise in part from communities’ differing socioeco- SocioeconomicSources nomic characteristics, such as: of Fiscal Disparities l The ability of communities to bear local tax burdens. As the Woodbine/ Alpine example shows, communities with per capita incomes above the Page20 GAO/Hl-#w)BSStatesHelpCamunities in Pisd Dishess F’iscalDisparities: Their Nature, Sources, and Extent Fiscal disparities are differences in communities’ abilities to provide comparable levels of public services with comparable tax burdens on local residents. They arise from differing socio-economic factors, such as (1) the financial ability of local residents to bear tax burdens, (2) the unit cost of providing services, and (3) social conditions that affect pub- lic service needs, including high crime rates or poverty concentrations. To a large extent, the fiscal disparities arising from such socioeconomic differences are beyond the ability of local governments to control. As a consequence, these governments seek state and federal assistance so they can provide basic public services while keeping local tax rates com- parable to those of neighboring communities. Such aid can ameliorate the fiscal disparities among localities, depending on the degree to which it is targeted to those with relatively high tax burdens per dollar of ser- vices received. How the responsibility for delivering public services is divided between the state and its local governments is another factor affecting the degree of fiscal disparity. States differ significantly in the extent to which they centralize service delivery. Nationwide, we estimate, state governments delivered over half of all noneducation public services provided to state residents in 1985. Significant variation existed among states. In some, such as Vermont and Alaska, the state provided over three-quarters of such services to its residents, while in others, such as Florida and Nevada, the state provided less than 40 percent. If localities had to finance all the services they provide locally without state or federal grants, sizable fiscal disparities would result and their degree would differ substantially among states. For example, local fiscal disparities would be relatively small in Iowa, where they would be about half the national average. The tax burden per dollar of public services in Iowa’s most distressed county would be about 70 percent greater than in the “best-off” county. In contrast, fiscal disparities would be relatively large in Kentucky, New Mexico, and Florida. In Florida, for example, the tax burden per dollar of services would be nearly three times greater in the most fiscally distressed counties than in the “best-off.” Other factors that influence the magnitude of local fiscal disparities, dis- cussed but not analyzed in this report, are the setting of local govern- ment boundaries and state restrictions on the ability of local governments to shift taxes to nonresidents. Page 16 GA0/HRD9069 States Help Communities in F-is& Distress chapter 1 Introduction Assessing How Much State To measure the effects of state and federal general fiscal assistance aid in reducing local fiscal disparities, we calculated per capita spending on and Federal General local public services first without and then with the receipt of general Assistance Reduces fiscal assistance aid. We then compared the standard deviation in local Disparities tax burdens per dollar of public service benefits before and after the receipt of aid. The percent decrease in the standard deviation measured the disparity reduction attributable to general fiscal assistance aid.R It represents the average reduction in tax burden differences between fis- cally distressed and “better-off” communities. We then applied this method to state and federal aid separately in order to compare how much each reduced disparities. In performing our analysis, we analyzed only the distribution of state and federal general fiscal assistance grants. We did not consider other factors that could reduce disparities among local governments. Among these are special purpose state and federal aid programs and services provided directly by state governments to local communities. A more comprehensive analysis also would take into account who pays the state and federal taxes used to finance the grant programs. For a more com- plete discussion of this issue, see p. 53. *.!keapp.I for a morecompletediscussion of issuesrelatedto thedefiition of fiscaldisparitiesand measurement methodology. Page 16 GAO/HFtBSM9 States Help C!~mmunities in Pimd Distress - Chapter1 Introduction Assessing the Relative As an indicator of a local government’s fiscal condition, we used the ratio of its average effective tax rate to its per capita expenditures4 Extent of Fiscal Local effective tax rates serve as an indicator of the tax burden borne Disparities Within a State by local residents, and per capita expenditures indicate the public ser- vice benefits they receive. Next, we defined local fiscal disparities as differences in the fiscal con- ditions of local governments. We chose the standard deviation in local fiscal conditions as a summary measure of fiscal disparities in each state.5 To identify the states with the largest disparities, we ranked the 48 states6 in our analysis by the degree to which fiscal disparities were larger or smaller than the national average of all states. To overcome the unavailability of some data as well as certain other analytical problems, we l limited our review to general purpose governments. Only they were eli- gible to receive federal revenue-sharing, and this study was intended to help the Congress assess the possible need for a replacement program. Consequently, school and special districts (which provide a single ser- vice) were excluded. . totaled all economic, fiscal, and demographic data to the county area level within each state. This simplified the analysis, gave us comparable units of analysis,? and eliminated the need to account for the varying structure of local government service responsibilities in each state. By averaging out differences among individual governmental units within each county, this approach tends to understate disparities and yield con- servative estimates of the extent of fiscal disparities in a state. used fiscal year 1985 state and local government tax and grant data. We l did so because we wanted to compare the targeting of state and federal general fiscal assistance aid and 1985 was the last year the federal pro- gram provided aid to local governments. Analysis of more current data 4Effectivetax rates,whichdiffer fromstatutorytax rates,aremeasuredasthepercentage of the averageresidentpersonalincomein a countyareathat all localtaxesrepresent. %eeapp.I for a morecompletediscussion of issuesrelatedto thedefinitionof fiscaldisparitiesand our measurement methodology “WeexcludedDelawareandHawaiibecause theyhavetoofewlocalunitsof government to obtain statisticallymeaningfulresults 7Forselectedstates,however,weusedmunicipalitiesasourunit of analysisbecause eitherthere weretoofewcountiesto eerveasunitsof observationsor thecountygovernments hadlimitedgov- ernmentalresponsibilities. ThesestateswereConnecticut, Maine,Massachusetts, NewHampshire, NewJersey,RhodeIsland.andVermont. Page14 GAO/HRD99-69 StatesHelpC~mmnnitieain FiscalDistress Figure 1.2: Share of U.S. Population Living in Low-, Mlddle-, and High-Income Counties (1977-87) Pmcml of U.S. PapUlatkxl El High ~r-vxme (greater than 120% of U.S. average) Middle income (between 80% and 120% of U.S. average) Low income (less than 80% of U.S. average) Source: GAO calculations, based on U.S. Department of Commerce, Bureau of Economic Analysis Data. 1. Should taxpayers in poor communities bear higher tax burdens than rich communities to finance their basic public service needs? 2. Should disparities serve as an incentive for residents and businesses to move to communities with plentiful public services and low taxes? State officials assign differing degrees of significance to these equity concerns. Our survey of state officials and state aid programs shows that state perceptions of local fiscal disparities cover a wide range. Some states regard them as an important public concern that merits remedial action, while others see them as outside their scope of responsibility. In part, federal concern about local fiscal disparities prompted the crea- tion of the federal general revenue-sharing program in 1972. Some mem- bers of the Congress have expressed concern that the 1986 expiration of Page12 GAO/HlHMO89StatesHelp C~mmunibles in FiscalDistress Chapter 1 Introduction A major benefit of our federal system of government is its ability to pro- vide for a wide range of public services desired by its citizens. The fed- eral government provides services that address national needs, such as defense and regulation of interstate commerce. Typically, services from state governments benefit citizens across the entire state, such as the construction and maintenance of highways. At the local level, services are tailored to reflect the needs and preferences of local residents. The nearly 39,000 general purpose local governments’ in the United States include 3,042 county, 19,200 municipal, and 16,691 township governments.” The fact that they delivered 46 percent of all public ser- vices offered to state residents in 1985 underscores the important role of these governments. Local governments have shouldered an increasing share of the cost of public services provided at the local level. In the decade since 1977, the local share of general revenues raised at the local level increased 16 per- cent, from 60 to 69 percent, as shown in figure 1 .l Federal aid as a percentage of total local revenues fell almost three-fifths, from 12 to 5 percent. The expiration of federal revenue-sharing in 1986 accounts for about 44 percent of this decrease. Over this same time period, the state share of local revenues fell slightly, from 28 to 26 percent. The ability of local taxpayers in low-income counties to shoulder a larger share of service costs has deteriorated compared with more afflu- ent areas. Since 1977, the income gap between poorer and more affluent communities has widened. The number of residents living in counties with incomes more than 20 percent below the national average has risen from 16 to 19 percent of the U.S. population (see fig. 1.2). This 19-percent increase implies that income growth for most low-income counties lagged behind the national average. In contrast, incomes in the most affluent counties have increased faster than the national average. Their share of the 1T.S.population has risen from 11 to 16 percent, a 45 percent increase, As the ability to bear tax burdens is directly related to income, this sug- gests that fiscal disparities between poorer and more affluent areas have increased. At the same time, local financing responsibilities have ‘Generalpurposelocalgovernmentsincludecounties,municipalities, andtowns.Thisexcludesboth schooldistrictsandspecialpwpasedistrictsfor reasonsdiscussed onpg. 14. %urce: U.S. Bureau of theCrnsus,1987Censusof Governments, Vol.1,Governmental Organization, table3, p, 3 Page10 GAO/HRD-90.69 StatesHelpCommunitiesin FiscalDisti-ess - Figures Figure 1.1: Distribution of Local Government Revenue 11 Sources (197787) Figure 1.2: Share of U.S. Population Living in Low-, 12 Middle-, and High-Income Counties (1977-87) Figure 1.3: States Selected by GAO for Fieldwork 17 Figure 2.1: Illustration of Fiscal Disparity Between Alpine 20 and Woodbine Boroughs, New Jersey (1983) Figure 2.2: Dispersion of County Tax Burdens Per $100 30 Dollars of Public Services in 67 Florida Counties (Fiscal Year 1985) Figure 2.3: Dispersion of County Tax Burden Per $100 31 Dollars of Public Services in 99 Iowa Counties (Fiscal Year 1985) Figure 2.4: Extent of Potential Fiscal Disparities Within 34 Each State (1985) Abbreviations AFDC Aid to Families with Dependent Children 8-f fiscal year GAO General Accounting Office TVA Tennessee Valley Authority Page8 GAO/HRD90~9StatesHelpCommunitiesin FiscalDistress Contents Executive Summary 2 Chapter 1 10 Introduction Local Fiscal Disparities: A Matter of Public Concern Objectives, Scope, and Methodology 11 13 Chapter 2 18 Fiscal Disparities: The Nature of Fiscal Disparities Socioeconomic Sources of Fiscal Disparities 19 20 Their Nature, Sources, Centralizing Service Delivery Affects Local Disparities 21 and Extent Level of Grant Financing Another Factor 22 Grant Targeting Can Reduce Local Disparities 24 Other State Policies Affecting the Extent of Local 27 Disparities Extent of Local Fiscal Disparities 28 Chapter 3 35 Federal Revenue- General Fiscal Assistance: An Important Local Revenue Source 35 Sharing Reduced Local Larger or Better-Targeted Aid Programs Would Reduce 37 Fiscal Disparities More Fiscal Disparities Than Did Most State General Fiscal Assistance Reduced Disparities 38 Targeting More Important Than Funding in Reducing 41 Programs Fiscal Disparities Disparity Reduction Mixed in States With Widest 41 Disparities Better-Targeted Federal Aid Reduced Disparities More 42 Than State Aid State Reaction to Loss of Federal Revenue-Sharing 45 Minimal Conclusion 46 Appendixes Appendix I: Definition and Measurement of Local Fiscal 48 Disparities Appendix II: General Fiscal Assistance Programs in 54 California Appendix III: General Fiscal Assistance Programs in 60 Kansas Appendix IV: General Fiscal Assistance Programs in 63 Massachusetts Page6 GAO/HUD.9989 StatesHelpCommuniUea in FiscalDistress Executivesumm&uy billion. Combined, this aid reduced disparities by approximately 18 per- cent. But when separately analyzed, federal revenue-sharing was targeted more to distressed communities than was state aid. As a conse- quence, although the federal program had less than half the funding of state general fiscal assistance, it reduced disparities more than did most state programs. Using existing levels and sources of funding, local fiscal disparities could be reduced further if states targeted more of their aid to fiscally distressed communities. GAO’s Analysis Extent of Disparities Absent state and federal grants, fiscal disparities would be large in most states. For example, there would have been a 3-to-1 disparity in tax bur- dens between Florida’s best-off and worst-off counties. In Union County, residents would have had to pay taxes equal to 1.9 percent of their income for each $100 of public services they received, while in Palm Beach County residents would have had to pay taxes equal to just 0.7 percent of their income for each $100 of services (see pp. 28-29). In contrast, fiscal disparities would be smallest among Iowa counties. The tax burden per $100 of expenditures for services would be 70 per- cent higher in the most disadvantaged county compared with the best- off county. This 1.7~to-1 tax burden disparity was half the national average (see p. 29). General Fiscal Aid Has Nationally in 1985, state general purpose fiscal assistance and federal Little Effect revenue-sharing reduced tax burden disparities between fiscally dis- tressed and better-off counties by 18 percent. Results varied widely by state, ranging from a 55-percent reduction in Nevada to 5.3 percent in Montana.3 When analyzed separately, the federal program reduced dis- parities on average by 11 percent and state programs by almost 9 per- cent (see p. 33)” “GAOexcludedDelawareandHawaiibecause theycontainedtoofewlocalitiesfor statistical analysis. ‘The overalldisparityreductionfor generalfiscalassistance programsis lessthanthedisparity reductionachievedseparately by thestateandfederalprogramsbecause stateandfederalaidoffset eachotherin somestates. Page4 GAO/‘HRD9649 States Help Chnmtmiti~ in F-iscal Diswess Executive Summary Between 1977 and 1987, direct federal aid to counties, cities, and town- Purpose ships declined by about three-fifths, to 5.2 percent of local revenues. Although the cuts occurred in a number of programs, the 1986 expira- tion of the $4.6 billion general revenue-sharing program accounted for 40 percent of the reductions and affected nearly 39,000 local governments. Concerned that the funding cuts may have had a serious effect on fis- cally distressed communities, the Chairman of the Senate Finance Com- mittee asked GAO to examine (1) how well state governments were meeting these communities’ fiscal needs and (2) how the loss of revenue sharing affected such communities.l GAO'S primary objectives were to . identify states with wide differences in local tax burdens (fiscal dispari- ties) and l discuss the extent to which state general purpose fiscal assistance pro- grams reduce disparities between the fiscally distressed and better-off communities. Fiscally distressed communities are those in which residents bear sub- Background stantially higher tax burdens in order to obtain levels of public services comparable to better-off communities. Such tax burden differences are referred to as “fiscal disparities.” They arise from differing socio- economic factors, such as (1) the financial ability of local residents to bear tax burdens; (2) the unit cost of providing services; and (3) social conditions, including high crime rates or poverty concentrations, that affect public service needs. Some policymakers are concerned about fiscal disparities, seeing large differences in tax burdens as inequitable. The difference between Starr and Wheeler counties in Texas provides an example of the inequity. In 1984, the average personal income of Starr County was the lowest in the state while that of Wheeler County residents was equal to the state average. If a Starr County resident earning the county’s average income and bearing the typical tax burden were to have moved to Wheeler County, he would pay $37.21 less in taxes for each $100 of public services. to localfmcalneeds;in a separate ‘This reportfocusesonstateresponsiveness study, GAO is examin- ingtheeffectsof thelossof federalaid.
Communities in Fiscal Distress: State Grant Targeting Provides Limited Help
Published by the Government Accountability Office on 1990-04-13.
Below is a raw (and likely hideous) rendition of the original report. (PDF)