For other versions of this document, see http://wikileaks.org/wiki/CRS-RL31936 ------------------------------------------------------------------------------ ¢ ¢ Prepared for Members and Committees of Congress ¢ ¢ This report provides background and analysis of the general revenue sharing program (GRS) as authorized in the State and Local Fiscal Assistance Act of 1972 (P.L. 92-512, the 1972 Act). The GRS program was extended three times before finally expiring on September 30, 1986. Over the almost 15-year life of the GRS program (1972 through 1986), more than $83 billion was transferred from the federal government to state and local governments. From 1972 to 1980, states received approximately one-third of the grants and local governments received two-thirds. State governments were excluded from GRS beginning in the 1981 fiscal year (FY). In 2003, policymakers suggested using the original GRS program as a model for a new, short- term, GRS program. The FY2004 budget resolution contained a proposal (H.Con.Res. 95, Sec. 605) expressing a sense of the Senate that $30 billion should be set aside over the next 18 months for state fiscal relief. Congress ultimately approved $20 billion in aid to states; $10 billion through Medicaid and $10 billion distributed by population. By comparison, in 1972, the federal government authorized $8.3 billion ($42.1 billion in 2008 dollars) for the first 18 months of the original GRS program. More recently, the recession that began in 2008 has prompted similar proposals. The rationale behind GRS in 1972 cannot be traced to a single political or economic objective, such as economic stimulus. The turbulent economic and political environment that characterized the 1960s and 1970s led proponents and opponents of GRS to modify their political and economic arguments as that environment changed. Generally, GRS could be implemented to (1) initiate intergovernmental fiscal reallocation; (2) address state and local government liquidity crises; and (3) synchronize federal and state-local fiscal policy. A revised GRS program intended to help close state budget deficits (estimated to be $31.0 billion for the remainder of FY2009 and estimated to be $64.7 billion for FY2010) has been advocated based on the last two objectives. The budget crisis facing state and local governments in 2009 has generated renewed concern at the state and local level. A GRS program designed as a countercyclical initiative would encounter two primary implementation issues: fiscal policy time lags and variability in the state response to GRS grants. In addition, as with all fiscal policy, the overall size of the additional federal spending is critical to the impact of the fiscal stimulus. This report provides general background and analysis and does not track current legislation. It will not be updated. ¢ Introduction ..................................................................................................................................... 1 Background on General Revenue Sharing....................................................................................... 1 Amount...................................................................................................................................... 1 Allocation Formula ................................................................................................................... 3 Economic Rationale for GRS Grants .............................................................................................. 6 Fiscal Reallocation .................................................................................................................... 7 State and Local Government Liquidity Problems ..................................................................... 7 Federal and State-Local Fiscal Policy Synchronization............................................................ 8 Analysis of GRS for Economic Stimulus in 2009........................................................................... 9 Magnitude of Anticipated Pro-Cyclical State Action ................................................................ 9 Implementation Issues............................................................................................................... 9 Fiscal Policy Time Lags.................................................................................................... 10 State Budget Options ........................................................................................................ 10 Table 1. GRS Transfers Made Through the State and Local Fiscal Assistance Act of 1972 and Subsequent Extensions .......................................................................................................... 2 Table 2. Amount Allocated to Each State under Three-factor GRS Formula with a Hypothetical $40 billion or $20 billion Appropriation................................................................. 5 Table 3. State Strategies to Eliminate FY2003 Budget Gaps .........................................................11 Table A-1. Change in Real GNP and Real Wages, 1980: Q2 to 1983:Q4 ..................................... 16 ¡ Appendix. A Brief History and Analysis of Prior GRS Legislation .............................................. 13 Author Contact Information .......................................................................................................... 17 ¢ This report provides a brief history and analysis of general revenue sharing (GRS). GRS is commonly defined as a program of federal transfers to state and local governments that does not impose specific or categorical spending requirements on the recipient government. The United States implemented a GRS program in 1972 that expired on September 30, 1986. Congress looked to the bygone GRS program once before as an option designed to address the fiscal year 2003 (FY2003) and FY2004 state budget shortfalls ($21.5 billion and $72.2 billion, respectively).1 Some observers have suggested that a revenue sharing program that provided states with grants to forestall spending cuts and tax increases in 2009 may deter pro-cyclical2 actions by states and produce national fiscal stimulus. The budget gaps for is estimated to be $31.0 billion for the remainder of FY2009 and for FY2010 it is estimated to be $64.7 billion.3 An examination of the GRS program that existed from 1972 to 1986 could provide some historical perspective if policy makers were to consider a revised GRS program in 2009. The first section provides a brief overview of GRS as authorized by the State and Local Fiscal Assistance Act of 1972 (P.L. 92-512, the 1972 Act) and the three extensions.4 The second section analyzes the economic rationale for GRS. The third section analyzes GRS in the context of its possible use for stimulus of the nation's economy in 2009 including estimated distribution to the states based on the original GRS formula. The Appendix provides a more detailed legislative history of the GRS program created by the 1972 Act and its three extensions. General revenue sharing (GRS) is typically defined as unconditional federal grants to state and local governments. These grants are intended to provide state and local governments with spending flexibility. The total grant amount is fixed annually, sometimes called "closed-ended," and allocated to the recipient governments by formula. GRS has not been explicitly identified as a primary tool to provide counter-cyclical assistance. The GRS program created by the 1972 Act exemplifies how a GRS program can work. Over the almost 15-year life of the GRS program (1972 through 1986), over $83 billion was transferred from the federal government to state and local governments. To achieve a comparable 1 Aggregate state deficit data are from the National Conference of State Legislatures, State Budget Update: April 2003, pp. 1-2. 2 The business cycle has peaks and troughs. Fiscal policy and monetary policy are used together to attenuate the size of those peaks and troughs to stabilize the economy. These actions are counter-cyclical. In contrast, pro-cyclical fiscal and monetary actions magnify the peaks and troughs, thus destabilizing the economy. 3 State cumulative deficit data, not including Puerto Rico, are from the National Conference of State Legislatures, State Budget Update:November 2008, p. 6. 4 For a more detailed description of the 1972 Act, see U.S. Congress, Joint Committee on Internal Revenue Taxation, General Explanation of the State and Local Fiscal Assistance Act and the Federal-State Tax Collection Act of 1972, committee print, 92nd Cong., February 12, 1973, (Washington: GPO, 1973). .snoitaluclac SRC dna elbat ni detic swal cilbuP :ecruoS 15.914,313$ 2.151,38$ 6891 ,03 rebmetpeS ot 2791 ,1 yraunaJ latoT dnarG 13.912,92$ 1.007,31$ 6891 ,03 rebmetpeS ot 3891 ,1 rebotcO latoT 77.937,9$ 7.665,4$ 6891 ,03 rebmetpeS ot 5891 ,1 rebotcO 71 doireP 77.937,9$ 7.665,4$ 5891 ,03 rebmetpeS ot 4891 ,1 rebotcO 61 doireP 77.937,9$ 7.665,4$ 4891 ,03 rebmetpeS ot 3891 ,1 rebotcO 51 doireP )581-89 .L.P( ylnO stnemnrevoG lacoL rof noisnetxE 3891 94.813,53$ 1.007,31$ 3891 ,03 rebmetpeS ot 0891 ,1 rebotcO latoT 38.277,11$ 7.665,4$ 3891 ,03 rebmetpeS ot 2891 ,1 rebotcO 41 doireP 38.277,11$ 7.665,4$ 2891 ,03 rebmetpeS ot 1891 ,1 rebotcO 31 doireP 38.277,11$ 7.665,4$ 1891 ,03 rebmetpeS ot 0891 ,1 rebotcO 21 doireP )406-69 .L.P( ylnO stnemnrevoG lacoL rof noisnetxE 0891 21.143,59$ 0.835,52$ 0891 ,03 rebmetpeS ot 7791 ,1 yraunaJ latoT 31.375,52$ 0.058,6$ 0891 ,03 rebmetpeS ot 9791 ,1 rebotcO 11 doireP 31.375,52$ 0.058,6$ 9791 ,03 rebmetpeS ot 8791 ,1 rebotcO 01 doireP 31.375,52$ 0.058,6$ 8791 ,03 rebmetpeS ot 7791 ,1 rebotcO 9 doireP 27.126,81$ 0.889,4$ 7791 ,03 rebmetpeS ot 7791 ,1 yraunaJ 8 doireP )884-49 .L.P( noisnetxE 6791 95.045,351$ 0.312,03$ 6791 ,13 rebmeceD ot 2791 ,1 yraunaJ latoT 44.798,61$ 0.523,3$ 6791 ,13 rebmeceD ot 6791 ,1 yluJ 7 doireP 13.072,23$ 0.053,6$ 6791 ,03 enuJ ot 5791 ,1 yluJ 6 doireP 10.805,13$ 0.002,6$ 5791 ,03 enuJ ot 4791 ,1 yluJ 5 doireP 27.547,03$ 0.050,6$ 4791 ,03 enuJ ot 3791 ,1 yluJ 4 doireP 38.481,51$ 0.889,2$ 3791 ,03 enuJ ot 3791 ,1 .naJ 3 doireP 41.764,31$ 0.056,2$ a 2791 ,13 rebmeceD ot 2791 ,1 yluJ 2 doireP 41.764,31$ 0.056,2$ a 2791 ,03 enuJ ot 2791 ,1 yraunaJ 1 doireP )215-29 .L.P( tcA 2791 lanigirO )tnerruc $( )lanimon $( 8002 2791 a setaD doireP tnemeltitnE )snoillim $ ni( b deltitnE tnuomA snoisnetxE tneuqesbuS dna 2791 fo tcA ecnatsissA lacsiF lacoL dna etatS eht hguorhT edaM srefsnarT SRG . 1 elbaT a $1 commitment in 1972 would be equivalent to a $5.08 commitment in 2008. thought of as the relative value of a commitment made in the past in current dollars. For example, nominal dollars and adjusted to 2008 dollars). The estimates provided in Table 1 for 2008 can be periods for the GRS grants (as provided for in the 1972 Act and subsequent extensions, both in distributed over the next 15 years. Table 1 provides detailed information on the 17 entitlement magnitude of assistance today, approximately $313 billion (in 2008 dollars) would need to be ¢ ¢ eno sdoirep rof edam erew stnemyap evitcepsorter ,suht ,2791 fo rebotcO ni wal otni dengis saw tca ehT .a .ylevitcepser ,3791 yraunaJ dna 2791 rebmeceD ni ,owt dna noitalsigel eht hcihw ni raey eht rof PDG eht no desab detaluclac saw srallod 8002 rof tnemtsujda ehT .b .dessap saw stnemeltitne eht gnizirohtua The payment periods in the 1972 Act were designed to roughly follow the budget calendars of state and local governments. The grants in subsequent extensions tracked the federal budget calendar.5 Note that after FY1980, only local governments, not states were entitled to GRS grants. GRS allocations were determined by a formula that used a combination of the following variables: tax effort, population, and per capita income.6 Generally, the greater the tax effort and population, the larger the grant. In contrast, the higher the per capita personal income, the smaller the grant. More specifically, section 106 of the GRS legislation stipulates that under the three-part formula, each state shall receive: an amount which bears the same ratio to the amount appropriated under that section for that period as the amount allocable to that State under subsection (b) bears to the sum of the amounts allocable to all States under subsection (b) The three-factor formula can be summarized symbolically: pop st × GTEF st × RIF st State "i" Share of GRS = A × 51 us i i i ( pop i × GTEFi × RIFi st st st ) i =1 where: Aus = total appropriation, popist = population of state "i", perincist = total personal income of state "i", 5 Unlike the federal government, most state and local government fiscal years begin July 1 and end on June 30. The fiscal year begins in July for 46 states, October for two states (AL and MI), April for one (NY), and September for one (TX). Thirty states use an annual budget cycle, while the other 20 use a biennial cycle. 6 An alternative, yet similar, five-variable formula (the "House" formula) was also used for determining the initial state share. The five variable formula included the three mentioned variables plus the state's urbanized area and income tax collections. The state chose the formula that produced the largest grant. Tax effort is a measure of taxes as a fraction of ability to pay. Two states with the same ability to pay and the same amount of taxes collected would receive equal tax effort scores. If a state raised more from the same ability to pay, it would receive a higher tax effort score. ¢ perinc us us = pop RIFi st , or state "i" relative income factor, and perincist pop ist taxesist GTEFi st = perinc st , or state "i" general tax effort factor. i The two ratios in the formula, the relative income factor (RIF) and the general tax effort factor (GTEF), were intended to adjust the state allocations based on the state's "ability-to-pay" and tax structure. The RIF for a state is the pre capita income for the U.S. divided by the per capita income of the state. If the state's RIF is greater than one, then it is considered relatively low income. Analogously, a RIF less than one indicates a state has relatively high income. In the three-part GRS formula, the higher a state's RIF, the greater the share of revenue. The GTEF was considered important for GRS because it created a disincentive for states to reduce taxes and rely more on the federal government for revenue over time. The GTEF is total state tax collections as a share of state personal income. In the GRS formula, the larger the GTEF component, the greater the share of revenue. Under the original GRS, the first step in the allocation procedure was to calculate each state's share based on the three variable formula. After each state's share was determined, one-third of the total amount was allocated to the state government and two-thirds to local general purpose governments within the state.7 The two-thirds portion was then distributed to each geographically defined county (parish) area within the state using the same three variable formula used to determine the state share. Each government within the county area then received an amount equal to the ratio of taxes it collected to total taxes collected by all general purpose governments in the county.8 The allocation formula was criticized for generating inequitable treatment of local governments. Generally, the arguments arose from "similar governments within a state receiv[ing] different revenue sharing payments, primarily because of their geographic location."9 According to a GAO report, "These inequities are created primarily by tiering allocation procedures whereby revenue sharing funds are first allocated to county geographic areas."10 7 The automatic state GRS allocation was discontinued after FY1980. 8 All tax calculations were adjusted to exclude taxes collected exclusively for schools. 9 More detail on this critique of the old allocation scheme can be found in the following: U.S. General Accounting Office, Changes in Revenue Sharing Formula Would Eliminate Payment Inequities; Improve Targeting Among Local Governments, GAO Report GGD-80-69 (Washington: June 10, 1980). 10 GAO, Changes in Revenue Sharing Formula Would Eliminate Payment Inequities; Improve Targeting Among Local Governments, p. ii. %2.0 73 976,840,94 57 953,790,89 erihspmaH weN %8.0 16 438,745,551 121 966,590,113 adaveN %6.0 86 601,955,021 631 112,811,142 aksarbeN %4.0 68 971,209,18 171 853,408,361 anatnoM %8.1 16 282,584,953 221 465,079,817 iruossiM %5.1 401 385,057,403 902 761,105,906 ippississiM %1.2 97 153,748,014 851 207,496,128 atosenniM %9.3 87 224,382,487 651 548,665,865,1 nagihciM %7.1 15 090,168,133 301 081,227,366 sttesuhcassaM %3.1 84 221,550,962 69 342,011,835 dnalyraM %6.0 19 522,334,021 381 944,668,042 eniaM %6.1 77 460,491,923 351 821,883,856 anaisiuoL %0.2 59 154,888,404 191 309,677,908 ykcutneK %0.1 27 844,768,002 541 698,437,104 sasnaK %0.1 96 767,042,702 931 535,184,414 awoI %5.2 87 211,115,794 751 422,220,599 anaidnI %4.3 35 762,596,386 601 335,093,763,1 sionillI %7.0 19 026,664,631 281 042,339,272 ohadI %6.0 101 469,784,921 202 829,579,852 iiawaH %2.3 86 752,013,746 631 315,026,492,1 aigroeG %7.4 25 013,521,449 301 026,052,888,1 adirolF %2.0 48 858,952,94 761 617,915,89 aibmuloC fo tcirtsiD %4.0 18 029,532,07 261 048,174,041 erawaleD %8.0 74 627,918,461 49 254,936,923 tucitcennoC %1.1 44 073,302,212 78 147,604,424 odaroloC %9.21 07 731,719,375,2 141 472,438,741,5 ainrofilaC %6.1 211 815,524,613 322 630,158,236 sasnakrA %2.2 07 986,781,444 041 873,573,888 anozirA %4.0 321 226,437,38 542 442,964,761 aksalA %6.1 17 926,916,723 241 852,932,556 amabalA %0.001 66$ 000,000,000,02$ 331$ 000,000,000,04$ setatS detinU latoT atipaC detairporppA atipaC detairporppA fo erahS reP tnuomA reP tnuomA lacitehtopyH lacitehtopyH noitairporppA noillib 02$ ro noillib 04$ lacitehtopyH a htiw alumroF SRG rotcaf-eerhT rednu etatS hcaE ot detacollA tnuomA .2 elbaT available. Only states are eligible in the example provided in Table 2. billion and $20 billion using data for 2007, the latest year where data for the full years is Table 2 below employs the three-part formula to allocate a hypothetical appropriation of $40 ¢ local government liquidity crises, and to synchronize federal and state-local fiscal policy. rationales behind GRS: to initiate an intergovermental fiscal reallocation, to address state and economic objective. This section of the report summarizes three frequently mentioned economic Because of this turbulence, the rationale behind GRS cannot be traced to a single political or political and economic arguments depending on the current political and economic conditions. environment changed dramatically. Thus, the proponents and opponents of GRS modified their of the 1972 Act and subsequent extensions, general economic conditions and the political From the time the active debate surrounding GRS began in the 1960s, through eventual passage .atad 7002 gnisu SRC yb delipmoc elbaT :ecruoS %2.0 86 105,794,53 631 200,599,07 gnimoyW %1.2 67 111,766,724 351 322,433,558 nisnocsiW %1.1 611 137,263,012 232 264,527,024 ainigriV tseW %0.2 36 827,189,604 621 654,369,318 notgnihsaW %1.2 55 603,599,524 011 116,099,158 ainigriV %4.0 411 049,477,07 822 978,945,141 tnomreV %3.1 59 130,782,252 191 260,475,405 hatU %7.5 84 545,177,141,1 69 090,345,382,2 saxeT %0.2 46 117,360,593 821 324,721,097 eessenneT %2.0 84 447,203,83 69 784,506,67 atokaD htuoS %7.1 97 060,285,943 951 911,461,996 aniloraC htuoS %3.0 46 721,620,86 921 452,250,631 dnalsI edohR %0.4 46 978,349,697 821 957,788,395,1 ainavlysnneP %2.1 56 574,967,442 131 059,835,984 nogerO %4.1 87 858,083,382 751 717,167,665 amohalkO %0.4 07 875,830,808 141 551,770,616,1 oihO %3.0 48 547,294,35 761 984,589,601 atokaD htroN %9.3 58 108,529,377 171 106,158,745,1 aniloraC htroN %6.5 85 763,479,421,1 711 337,849,942,2 kroY weN %1.1 901 806,055,512 912 512,101,134 ocixeM weN %3.2 45 255,646,564 701 401,392,139 yesreJ weN latoT atipaC detairporppA atipaC detairporppA fo erahS reP tnuomA reP tnuomA lacitehtopyH lacitehtopyH ¢ ¢ Fiscal reallocation has two components. Generally, under a GRS program, state and local tax regimes are partly replaced by the federal tax regime. Also, the federal spending objectives are replaced, in part, by state and local spending priorities. Proponents of reallocation cite the more "progressive," and thus desirable, structure of federal taxes.11 However, an assessment of the merits of a more progressive tax structure require subjective claims of what is "fair" taxation. Even if there is agreement that a more progressive structure is needed for fairness, it is unclear that GRS on the relatively small scale of the previously implemented program could achieve that objective. GRS would also shift government spending decisions for the grant amount from the federal government to state and local governments. The rationale for such a shift can be traced to the assertion that state and local governments are better able to understand and satisfy the preferences of their residents. A reallocation through GRS could also address the "assignment" issue. The assignment issue arises when the revenue productivity of a government does not match the spending requirements for the public services assigned to that level of government. Although these observations may be true for some publicly provided goods and services, it is not clear that nationally, the net gain in spending efficiency alone would justify a GRS program. And, the small relative size of a GRS program relative to overall tax collections would limit any gains in government spending efficiency. The arguments for and against fiscal reallocation are subjective because they rely on measuring fairness. Some would argue that a more progressive tax system is patently unfair, while others would argue that a tax system that redistributes income is more equitable and desirable. Fiscal reallocation would change the structure of government fiscal relationships, but analysis of the degree to which it does and the desirability of such a shift are beyond the scope of this report. ¢ State, and more specifically, local governments, often face fiscal liquidity problems that arise from revenues that fluctuate more dramatically with the business cycle than do expenditures. As the economy slows, revenue falls more sharply than expenditures, creating a budget deficit. Governments without sufficient reserves are then compelled to reduce expenditures or raise taxes to balance their budgets. State and local governments cannot use debt to close deficits because of state constitutional or statutory restrictions requiring a balanced budget. In contrast, the federal government can issue more debt when expenditures exceed revenue. A countercyclical GRS program could help alleviate these relatively short-term liquidity problems for states. Opponents of federal assistance to state and local governments during economic slowdowns suggest that poor state-local fiscal management creates deficit problems. State and local governments could "save" surplus revenue during economic expansions to then use when the economy contracts and revenue falls. If the rise and fall of revenue is symmetric, then the revenue 11 Most research has found that state and local tax regimes are generally more regressive because they rely much more heavily on sales and property taxes than on income taxes. The sales tax is viewed as a very regressive tax whereas the property tax has been cast as mildly regressive. However, some research has found that with a different set of assumptions, the property tax could be mildly progressive. ¢ saved should be sufficient to cover revenue shortfalls when the economy slows. However, research has shown that state government budgets are generally asymmetric over the business cycle.12 State and local governments tend to save less during expansions for a variety of reasons. Political pressure from voters to reduce taxes when large budget surpluses accrue is a commonly cited reason. ¢ ¢£ This objective is related to the liquidity objective discussed above. However, the rationale for a long-term GRS program designed for economic stabilization is somewhat different than a one- time grant to remedy a temporary fiscal imbalance. The federal government will typically employ monetary and fiscal policy to help stabilize consumption patterns and the price level as the economy cycles between periods of growth and recession. Generally, stimulative fiscal policy is implemented through tax reductions or increased government spending. In theory, tax reductions and/or increased government spending stimulates the demand for goods and services. The increased demand for goods and services then leads to economic expansion and recovery. This fiscal policy counters the economic downturn and is thus termed countercyclical fiscal policy. However, state and local governments may mitigate countercyclical federal fiscal policy if they are forced to raise taxes and reduce expenditures during recessions. Such a "pro-cyclical" state and local government response could undermine any federal fiscal stimulus. During economic downturns, this rationale played a more prominent role for proponents of general revenue sharing. While debating the 1976 extension, Senator Muskie offered the following rationale for GRS: we at the Federal level are trying to speed up economic recovery by cutting taxes, [while] state and local governments are being forced to raise their own taxes, thus delaying the impact of the Federal effort.13 The economic situation in the early to mid 1970s, about the time of initial passage of GRS, may seem similar to today's economic situation. However, the 1973-1975 recession was much deeper and longer and coincided with a sharp oil supply shock that the current downturn has not experienced.14 Nevertheless, the debate surrounding countercyclical aid to the states today is reminiscent of the 1975-1976 debate.15 12 Bent E. Sorensen and Oved Yosha, "Is State Fiscal Policy Asymmetric Over the Business Cycle?," Federal Reserve Bank of Kansas City Economic Review, Third Quarter, 2001, pp. 43-64. 13 Edmund Muskie, "Revenue Sharing and Countercyclical Assistance," in General Revenue Sharing and Decentralization, Walter F. Schefer, editor (Norman, OK: University of Oklahoma Press, May, 1976), p. 72. 14 For more on the relative size of U.S. recessions, see CRS Report RL31237, The 2001 Economic Recession: How Long, How Deep, and How Different From the Past?, by Marc Labonte and Gail E. Makinen. 15 Congress did enact two relatively small countercyclical assistance programs. P.L. 94-369 included an authorized maximum amount of $1.375 billion for countercyclical assistance over five quarters, beginning July 1, 1976. The funds would be released to state and local governments provided certain national economic thresholds were crossed. P.L. 95- 30 contained an extension of the countercyclical aid program, authorizing a maximum of $1 billion for FY1977 and $2.25 billion for FY1978. No federal funds were spent under either authorization. ¢ ¢ This section analyzes how GRS might affect the economy if implemented in 2009. The first subsection describes the potential size of GRS compared to current state deficits. The second section analyzes implementation issues that may arise if a new GRS program were authorized, including a discussion of how states might use new federal grants. ¢ The principal question is: "Will the supposed pro-cyclical state actions in the absence of federal assistance dampen the effect of federal fiscal policy?" From a national economic perspective, closing the remaining state FY2009 budget gaps with revenue sharing would likely have little if any effect on the national economy. The National Conference of State Legislatures reported that the remaining FY2009 gap for 38 states of $31.0 billion (as of November 2008) is approximately 0.22% of the U.S. GDP of $14.4 trillion, hardly enough to effectuate a stimulative response.16 The same NGA study, however, notes projected shortfalls of $64.7 billion for FY2010. The budget gaps for FY2009 are after closing a $40.3 billion budget shortfall before enacting the FY2009 budget. A one-time GRS type grant to states that closed the estimated FY2009 fiscal imbalance of $31 billion and forestalled anticipated state spending cuts and tax increases for FY2010 of $64.7 billion could provide significant fiscal stimulus. This assumes other federal spending would not be reduced and the states spent the federal grants immediately. The degree of stimulus would be tempered by the net spending response of the recipient government. Research has generally shown that for every $1 lump sum transfer, only a portion is translated into new spending.17 For example, assume a state has planned spending of $100 to be paid with own source tax revenue of $100. Under this leakage theory, a $10 transfer from the federal government would not lead to $110 of spending. Instead, the state may lower own-source tax revenue $5 and use half the federal grant to cover the tax reduction. The result would be an increase in government spending of $5, not the full $10 transferred. The above discussion assumed that federal spending would flow seamlessly from the federal government through states to the designated spending program. Two factors may result in a drag on this flow. First, state government administration may increase the lag time and second, each state would use the grant for budget priorities of varying stimulative effect. Following is a brief analysis of these two important implementation factors. 16 State cumulative deficit data, not including Puerto Rico, are from the National Conference of State Legislatures, State Budget Update:November 2008, p. 6. The GDP data are from U.S. Department of Commerce, Bureau of Economic Analysis, Table 1.1, Gross Domestic Product: Third Quarter 2008. 17 Edward M. Gramlich and Harvey Galper, "State and Local Fiscal Behavior and Federal Grant Policy," Brookings Papers on Economic Activity, vol. 1, 1973, p. 15. Gramlich and Galper concluded that between $0.25 and $0.43 of each $1 of unconditional federal transfer became new spending. The remaining $0.57 to $0.75 leaked from the spending stimulus. ¢ ¢ Time lags in implementation are the primary impediment to effective fiscal stimulus.18 Generally, the objective of fiscal policy during a recession is to boost aggregate demand and generate short term economic stimulus. However, if the stimulus comes too late, the increased spending may occur when the economy has already begun to revive and is approaching full employment. In that case, the stimulus becomes pro-cyclical and possibly inflationary. Policy makers should therefore use fiscal stimulus with caution because of the potential for mistimed action. GRS grants may be subject to two time lags, thus increasing the potential for mistimed fiscal policy. The first occurs at the federal level where policy makers must identify the need for stimulus then agree upon the size of the stimulus. Once the need and size are determined, Congress must then agree upon a grant allocation scheme that satisfies the competing goals of equity among jurisdictions and optimal stimulus. For example, suppose the grant allocation formula includes a component that provides greater assistance to states with greater need.19 If so, states that may have been more fiscally responsible would receive less, possibly violating the fairness criterion. However, from a broader macroeconomic perspective, aid that prevents more layoffs and state government budget cuts would seem to deliver greater short-term stimulus. Determining the structure of the allocation scheme could generate considerable debate, possibly delaying initial implementation efforts. The second time lag occurs at the state level. Federal grants that arrive before June 30, 2009, might avert some of the pro-cyclical state actions (e.g., budget cuts and tax increases) for many states. If the grants arrive too late for FY2009, state budget officials could simply add this revenue to the operating budget for FY2010 and perhaps avoid implementing tax increases and spending cuts that would otherwise begin on July 1, 2009. What could states do with unconditional revenue sharing grants? Generally, states have four options for federal grants (listed in order of stimulative response): · increase government spending, · reduce taxes (or rescind past tax increases), · reduce debt (or not issue more debt), and/or · contribute to a rainy day fund (or not draw down a rainy day fund). Increased spending would be the most stimulative in the short run, because the grant is immediately injected into the economy. This option for the states would include retaining state 18 For more on the effectiveness of fiscal policy, see CRS Report RL30839, Tax Cuts, the Business Cycle, and Economic Growth: A Macroeconomic Analysis, by Marc Labonte and Gail E. Makinen. 19 Note that the 1972 Act GRS allocation scheme included per capita income and "tax effort." Jurisdictions with greater tax effort received a larger share. Jurisdictions with lower relative per capita income, one potential measure of need, also received a larger share. ¢ employees who would have been furloughed, maintaining current operations that would have been reduced, and not scaling back social programs such as education and healthcare. Theoretically, this fiscal stimulus works best when government spending is quickly multiplied through the economy.20 This means that each dollar of the federal transfer payment stimulates the economy the most if the entire dollar is spent by the recipient and then spent again. The degree of stimulative effect of avoided state actions, such as not furloughing workers, depends on this "multiplier effect." Thus, to achieve the greatest stimulus, the most contractionary state actions should be the first avoided. The National Conference of State Legislatures (NCSL) asked budget officials from all states to categorize their spending strategies to reduce or eliminate budget gaps remaining for FY2009. Changes in taxes are difficult to implement in the middle of a budget year and are not included. Table 3below lists the strategies identified by NCSL and the number of states that proposed implementing those strategies for 2009. For FY2010, several state and local governments are likely going to increase taxes to help close budget gaps.21 The spending option for states that would produce the most relative stimulus for each dollar of spending would be to avoid net job losses (e.g., layoffs, furloughs, and, to a degree, early retirement and hiring freezes). To see why this is true, consider what would happen if net job losses occurred. First, layoffs reduce aggregate demand because when workers are laid off, their income would fall steeply until they find new jobs, causing their consumption to fall. (Even though all of the federal spending is not entirely multiplied through the economy because of employment taxes and income taxes, the stimulative action is relatively effective because the federal government is essentially "paying" the state employees.) Second, since government services are included in GDP, measured economic activity would be directly reduced as long as resources (workers) lay idle. In an environment of rising unemployment, it is unlikely that all of these resources would quickly be put back to use through market adjustment. If GRS prevented net job losses, these negative effects on the economy could be avoided. .3 elbaT spaG tegduB 3002YF etanimilE ot seigetartS etatS ygetartS ro gnisoporP setatS fo rebmuN ygetartS eht gnitnemelpmI ezeerF gniriH 32 stuC egatnecreP draoB-eht-ssorcA 02 snaB levarT 61 sdnuF rehtO 41 sdnuF yaD yniaR esU 11 sffoyaL eeyolpmE 01 stcejorP latipaC yaleD 01 ezeerF yralaS 5 20 See CRS Report RL30839, Tax Cuts, the Business Cycle, and Economic Growth: A Macroeconomic Analysis, cited earlier. 21 For example, see Laura Mahoney, "California Governor Vetoes $18 billion In Cuts, Tax Increases OK'd by Democrats," Daily Tax Report, January 8, 2009, p. H-1. ¢ ygetartS ro gnisoporP setatS fo rebmuN ygetartS eht gnitnemelpmI tnemeriteR ylraE 2 rebmevoN ,11 dna ,01 ,9 ,8 selbaT ",etadpU tegduB etatS" ,serutalsigeL etatS fo ecnerefnoC lanoitaN :ecruoS .8002 The saving behavior of potentially separated employees would likely enhance the stimulative effect of avoiding job losses. (However, avoiding induced early retirement may provide less stimulus than avoiding furloughs and lay-offs.) If the employees are early in their careers and/or are in low skill positions--likely candidates for furloughs or lay-offs--it is likely that their incomes are lower than the median for state employees. Research has shown that low income workers save a smaller portion of their income than high income workers.22 Thus, preventing the employment separation of low income workers should provide more relative stimulus than the alternative of not offering early retirement. Across-the-board cuts would affect a variety of spending programs that do not easily conform to one succinct appraisal. The stimulative effect of avoiding across-the-board cuts would vary from state to state based on the state's spending pattern. Aid to local governments also falls into an uncertain category because of differing intergovernmental transfers across states. The stimulative effect of avoiding cuts in local aid would be positive, though the magnitude is uncertain. ¡ Generally, tax cuts are less stimulative than direct spending increases, because individuals are likely to save some of their tax cut. Analogously, a rescinded or avoided tax increase would also be less stimulative than spending increases because taxpayers would likely save some portion of the reduced tax payment. ¢ ¢ Debt reduction and contributing to a rainy day fund would offer little stimulus because such action would be equivalent to an increase in public saving. In the short run, increased public saving does not stimulate the economy. If the federal grants were used to avoid tapping into tobacco revenue, the saving effect would be similar to contributing to a rainy day fund. The combined effect of the various potential responses of state and local governments to federal grants is difficult to quantify a priori. Nevertheless, one could confidently assert that $1 of federal grants would not lead to a corresponding $1 increase in fiscal stimulus. While some state and local governments may spend all the federal grants and not change pre-grant taxing and spending priorities, some portions of the GRS grants would likely be used for non-stimulative purposes such as substituting for previously planned spending or tax increases. 22 Julie-Anne Cronin, "U.S. Treasury Distributional Analysis Methodology," U.S. Department of Treasury, Office of Tax Analysis Paper 85, Sept. 1999, Table 6, p. 16. ¢ ¡ ¢ ¢ The GRS grants authorized by the State and Local Fiscal Assistance Act of 1972 (the 1972 Act) were essentially unconditional. A trust fund was established and annual appropriations were dedicated to the trust fund. Even though the grants were identified at the time as general revenue sharing, the legislation did include a list of "priority expenditures" for which the shared revenue sent to local governments could be used. (The grants to states were unconditional.) GRS grants could be used by local governments for the following acceptable operating expenditures: (1) public safety; (2) environmental protection; (3) public transportation; (4) health; (5) recreation; (6) libraries; (7) social services for the poor or aged; and (8) financial administration. "Ordinary and necessary capital expenditures" were also allowed.23 The grants could not be used for education. Note that the priority expenditure list was discontinued by the 1976 extension. In addition to the priority expenditure list, the 1972 Act also disallowed the use of GRS for matching federal grants. That restriction was also dropped in the 1976 extension. Congress believed GRS was necessary for a variety of reasons. The most prominent reason at the time was the perceived need for reallocation of government responsibilities arising from the changing citizen demands for government services (fiscal reallocation as cited earlier). The congressional sentiment behind the 1972 Act that created general revenue sharing is summarized well in the following passage from the Senate report accompanying the 1972 Act: Today, it is the States, and even more especially the local governments, which bear the brunt of our more difficult domestic problems. The need for public services has increased manyfold and their costs are soaring. At the same time, State and local governments are having considerable difficulty in raising the revenue necessary to meet these costs.24 The Nixon Administration seemed to have a similar perspective. When President Nixon signed the legislation, the President remarked that the GRS program would "place responsibility for local functions under local control and provide local governments with the authority and resources they need to serve their communities effectively."25 23 Section 103 of the "State and Local Fiscal Assistance Act of 1972." State governments usually maintain an operating budget and a capital budget. Generally, debt cannot be issued for the operating budget. 24 U.S. Congress, Senate Conference Report, report to accompany H.R. 14370, S.Rept. 92-1050, 92nd Cong. (Washington: GPO, 1972). 25 This quote is cited in the following: Graham W. Watt, "The Goals and Objectives of General Revenue Sharing," The Annals of the American Academy of Political and Social Science, vol. 419, May 1975. Mr. Watt was Director of the Office of Revenue Sharing, U.S. Department of the Treasury. ¢ However, the shift in the demand for and provision of government services was not the only justification for GRS. Observers at the time cited these additional reasons for implementing a revenue sharing program:26 · to stabilize or reduce state and local taxes, particularly the property tax; · to decentralize government; · to equalize fiscal conditions between rich and poor states and localities; and · to alter the nation's overall tax system by placing greater reliance on income taxation (predominantly federal) as opposed to property and sales taxation. Counteracting cyclical economic problems, such as state and local budget deficits induced by a slowing economy, was not explicitly mentioned as justification for GRS in the 1972 Act. However, when the debate began in 1974 on extending GRS beyond 1976, the countercyclical potential of revenue sharing apparently became important to policymakers. The counter cyclical arguments were likely initiated by the relatively severe recession that lasted from November 1973 through March 1975.27 ¡ The State and Local Fiscal Assistance Act of 1976 extended the GRS program through FY1980 with minor modifications. In the Senate report accompanying the legislation, Congress identified the following two reasons for the extension: (1) "Rapidly rising services costs coupled with sluggish declining tax bases has meant that State and local governments have had to raise tax rates and/or cut services," and (2) "A chronic problem State and local governments face is that the demand for public services is more elastic than the availability of revenues to finance them."28 The Senate report suggested that the extension of the GRS program "not only serves to help solve the fiscal problems of individual state and local governments, but also serves to stabilize the economy." The 1976 extension also eliminated the priority expenditure categories for local governments and the prohibition on states from using the grants for federal matching grants. Policymakers recognized the fungibility of local revenues which initiated the elimination of the spending restrictions. Although the fiscal stimulus features were mentioned during the debate surrounding extension, the ultimate purpose of revenue sharing was characterized as a long-term restructuring of the intergovernmental transfers. The desire to use revenue sharing as a countercyclical fiscal policy tool was not directly addressed in the 1976 extension. However, the reference to revenue sharing's ability to "stabilize" 26 Richard P. Nathan, Allen D. Manvel, and Susannah E. Calkins, Monitoring Revenue Sharing (Washington, D.C.: The Brookings Institution, 1975), p. 6. 27 For more, see Edmund Muskie, "Revenue Sharing and Countercyclical Assistance," in General Revenue Sharing and Decentralization, Walter F. Schefer, editor (Norman, OK: University of Oklahoma Press, May, 1976), p. 67-74. 28 U.S. Congress, Senate Finance Committee, report to accompany H.R. 13367, S.Rept. 94-1207, 94th Cong., Sept. 3, 1976. (Washington: GPO, 1976), p. 4. ¢ the economy may have arisen due in part to the countercyclical merits of GRS as suggested during the debate leading up to the extension.29 The total size of the extension, $25.5 billion, was approximately 2.5% of total state and local own-source tax revenue collected over the FY1977 to FY1980 period. Nationally, the transfer averaged 0.29% of national gross domestic product (GDP) annually over the four-year period. ¡ The State and Local Fiscal Assistance Act Amendments of 1980 (P.L. 96-604) extended the general revenue sharing program through September 30, 1983, but only for local governments.30 According to the House report accompanying act, the state share was eliminated as a means of helping to balance the Federal budget. The Committee believes that State governments are better able to adjust to the discontinuance of revenue sharing allocations than local governments.31 Until the 1980 Act, approximately one-third of the GRS grants had been allocated to the states. The 1980 Act reduced the GRS grants by one-third--from $6.850 billion to $4.567 billion--and only local governments received the grants (see Table 1). In addition to continuing GRS for local governments, the 1980 Act also authorized the creation of a "countercyclical assistance program" to be triggered by national economic downturns. The purpose of the program was to provide assistance to state and local governments during recessions. To achieve this, the program authorized $1 billion for each of the fiscal years, 1981, 1982, and 1983, subject to the trigger mechanism described in the House report accompanying the legislation: funding would be triggered when the national economy has experienced two consecutive quarterly declines in both real gross national product and real wages and salaries [emphasis added] (that is, corrected for inflation). Once a recession has been confirmed by these declines, funds would be provided for each recession quarter in relation to the severity of the recession. The program would be funded at a rate of $10 million for each one-tenth percentage point decline in real wages and salaries measured from the pre-recession base-- the average of the real wages and salaries for the two quarters preceding the decline. The amount of money allocated in any one quarter would be limited to $300 million. After setting aside 1% of the funds for Puerto Rico, Guam, American Samoa, and the Virgin Islands, the remaining funds would then be split evenly between state governments and "county 29 Around the time the extension was passed, Congress did enact two relatively small countercyclical assistance programs. P.L. 94-369 included an authorized maximum amount of $1.375 billion for countercyclical assistance over five quarters, beginning July 1, 1976. The funds would be released to state and local governments provided certain national economic thresholds were crossed. P.L. 95-30 contained an extension of the countercyclical aid program, authorizing a maximum of $1 billion for FY1977 and $2.25 billion for FY1978. No federal funds were spent under either authorization. 30 The 1980 legislation did provide for GRS grants for states if the state reduced other categorical federal grants-in-aid by an amount equal to the GRS grant. Essentially, states had the option of changing categorical aid into general assistance. 31 U.S. Congress, House Government Operations Committee, Report to accompany H.R. 7112, H.Rept. 96-1277, 96th Cong., Sept. 4, 1980. (Washington: GPO, 1980), p. 6. ¢ areas." The relative size of payments to states and county areas would have been based on the severity of the economic downturn in that area. The state portion would be adjusted by the state's tax effort. The greater the effort, the greater the grant. Apparently, the trigger threshold was never crossed. No grants were provided under the countercyclical fiscal assistance program. Table A-1 below reports the quarterly change in the real wage and real GNP for the second quarter of 1980 through the third quarter of 1983. The time periods reported in Table A-1 are the three federal fiscal years for which funding was authorized plus the two quarters before the first fiscal year of authorization. Note that for the 14- quarter time frame reported below, there were never two consecutive quarters where both the real GNP and real wage declined from the previous quarter. 4Q:3891 ot 2Q :0891 ,segaW laeR dna PNG laeR ni egnahC .1-A elbaT ) . s r a e y l a c si f l a r e d e f k r a m s e n i l l a t n o z i r o h d l o B ( doireP PNG laeR egaW laeR 2Q :0891 %11.2- %03.1- 3Q :0891 %32.0- %00.1 4Q :0891 %35.1 %03.0 1Q :1891 %10.2 %02.1- 2Q :1891 %57.0- %02.0- 3Q :1891 %42.1 %01.1- 4Q :1891 %50.1- %02.0 1Q :2891 %37.1- %01.0- 2Q :2891 %55.0 %09.0- 3Q :2891 %76.0- %01.0 4Q :2891 %30.0- %07.1 1Q :3891 %41.1 %05.0- 2Q :3891 %04.2 %02.1 3Q :3891 %87.1 %05.0 cimonocE fo uaeruB ,ecremmoC fo tnemtrapeD .S.U eht morf atad ylretrauq no desab snoitaluclac SRC :ecruoS .scitsitatS robaL fo uaeruB ,robaL fo tnemtrapeD .S.U eht dna sisylanA The 1980 Act is significant because the act discontinued revenue sharing for the states and formally introduced the concept of providing countercyclical fiscal assistance through federal grants to state and local governments as part of GRS legislation. Ultimately, the countercyclical assistance program was never funded and thus no countercyclical fiscal assistance was provided. Local governments generated $593.8 billion of own source revenue over the three fiscal years covered by the 1980 Act. GRS provided $13.7 billion in grants to local governments-- approximately 2.3% of total own-source revenue. The grants to local governments probably had little effect on the national economy given they represented 0.14% of U.S. GDP over the three- year time frame. The $1 billion for each of 1981, 1982, and 1983 for countercyclical aid, authorized but never spent, would have produced a negligible effect on the economy, even if fully realized. ¢ ¡ The final installment of the GRS program was signed into law on November 30, 1983, as the Local Government Fiscal Amendments of 1983 (P.L. 98-185). As with the 1980 Act, only local governments received grants. The 1983 extension was intended to stabilize the fiscal condition of local governments. The conference report accompanying the legislation stated that the tendency of State and Local governments to rely on relatively inelastic revenue sources, such as local property taxes, has limited their flexibility in responding to fiscal problems. To assist local governments in meeting the needs of their communities in a time of fiscal stringency, the Committee amendment extends the general revenue sharing program for three years.32 The final extension provided the same amount for local governments as did the 1980 Act ($13.7 billion) in three equal annual installments of $4.567 billion. This amount was equal to the amount received by local governments from 1977 through 1980. The countercyclical aid program was not extended. The GRS program ended September 30, 1986. Steven Maguire Specialist in Public Finance smaguire@crs.loc.gov, 7-7841 32 U.S. Congress, Senate Finance Committee, report to accompany S. 1426, S.Rept. 98-189, 98th Cong., July 20, 1983. (Washington: GPO, 1983), p. 2. ------------------------------------------------------------------------------ For other versions of this document, see http://wikileaks.org/wiki/CRS-RL31936