Tax Policy – Designing a State and Local Government Relief Package

Key Findings

  • The COVID-19 pandemic and the attendant economic contraction will wreak havoc on state and local tax revenues, with projections of a 15-20 percent decline in state revenues.

  • Lawmakers and governmental associations have called for between $300 billion and $1 trillion in state and local aid.

  • As much as $535 billion of the nearly $3 trillion the federal government has already appropriated flows through to states and localities, though only a portion is flexible funding.

  • In designing a state and local government relief package, federal lawmakers must determine (1) how to allocate the funding, (2) how much to appropriate, (3) how much flexibility to grant state and local governments in spending it, (4) whether to provide all funding immediately or spread it out over months or years, and (5) whether to repurpose existing appropriations.

  • Allocation could be by population, economic or budgetary conditions, or coronavirus cases. Population-based measures, while imperfect, avoid distorting state incentive structures and the substantial lag associated with economic data.

  • All levels of government should participate in the fiscal response, with states expected to revise budgets in light of the crisis.

  • If state aid is intended to provide several years of assistance, staggered disbursements can help ensure that states do not postpone difficult but necessary decisions.

  • Lawmakers could provide additional flexibility in drawing down the $150 billion Coronavirus Relief Fund.

  • If lawmakers wish to craft a state relief package, they should act expeditiously but not hastily, as their decisions will cast a long shadow on state financial systems.


As states brace for sharply lower tax revenues, there are increasing calls for federal assistance to close state budget gaps. The federal government will spend trillions of dollars responding to the economic contraction, with at least hundreds of billions going to state governments in some form. Flexible aid is a matter of controversy—both how much (if any) there should be and how it is allocated.

The National Conference of State Legislatures has requested $300 billion in flexible state and local aid. The National Governors Association (NGA) has requested $500 billion in state aid,[1] a figure echoed in legislation introduced by Senators Bob Menendez (D-NJ) and Bill Cassidy (R-LA), under which states and localities would share $500 billion.[2] House Speaker Nancy Pelosi (D-CA) has indicated that she will pursue an approach under which states and localities have separate pools of funding worth about $500 billion each, yielding a $1 trillion aid package.[3] All proposals dramatically outstrip the $143.6 billion in flexible aid provided to state and local governments during the Great Recession, a figure which includes provisions already mirrored in the Families First Coronavirus Response Act (FFCRA) and the Coronavirus Aid, Relief, and Economic Security (CARES) Act.[4]

This paper reviews the aid made available to states thus far, attempts to situate state aid proposals in the broader revenue and budget context, and offers a few considerations for the design of any additional program of federal aid to backfill state revenue losses.

Scope of State Revenue Losses

Projecting state revenue losses during an economic contraction is always challenging, but in the present crisis it is an exercise in epidemiology as well as economics. Much depends on how long the health crisis lasts and how soon certain sectors of the economy can reengage. It is not simply a matter of lost months of economic activity. As time passes, more businesses will fail, meaning fewer jobs to which those currently furloughed or laid off can return. Supply chains will disintegrate, making recovery more difficult. Capital will be destroyed, making reinvestment less affordable.

Forecasters do not know when the pandemic will abate; at best they might have some sense of when policymakers in a given state will attempt a phased reopening. How steep state revenue losses are and how long they will continue will substantially depend on the answers to these questions. Still, to even begin to discuss what sort of additional federal aid to states might be appropriate, it is necessary to advance some preliminary estimates.

The states have thus far been optimistic on the whole. Of 15 states which have forecast revenue declines against their prior FY 2021 baseline, Arizona, Arkansas, Hawaii, Kentucky, Minnesota, Pennsylvania, South Carolina, and Virginia anticipated revenue losses of 1 to 9 percent, and only Alaska, Michigan, New Mexico, New York, and Oklahoma have officially projected losses of 15 percent or more. In some cases, moreover, those figures form the high end of ranges, such as a range of 4-16 percent in Michigan or 9-17 percent in New York.[5] At times, however, it seems as if states are anticipating the losses they feel prepared to cover, rather than preparing for the losses they should be anticipating.

Moody’s, the ratings service, projects that FY 2021 state revenues could be 18.5 percent lower than prior projections, and 14.3 percent lower than states actually raised in the last fiscal year.[6] The NCSL, in its request to Congress, projected revenue losses of 15-20 percent.[7] Local governments tend to rely on a more stable tax mix than states, so slightly better outcomes seem likely. Taken together, state and local revenue declines of about 15 percent form a reasonable assumption, subject to revision as time passes.

States’ enacted general funds, in aggregate, were $913.2 billion in FY 2020.[8] Anticipated state tax revenue (for general funds and the own-source share of non-general funds) was $1.10 trillion, and local governments likely generated another $760 billion in tax revenue, for a combined estimated $1.86 trillion in state and local tax revenue.[9] A $1 trillion package, therefore, would replace about 54 percent of state and local tax revenues for a year—three to four times higher than estimated FY 2021 losses.

Undoubtedly, however, state and local tax revenue will take years to recover—as was the experience of the Great Recession and prior periods of economic decline as well. Moody’s projects that state tax collections between FYs 2020 and 2024 will be $601 billion less than what they would have been had the budgeted FY 2020 revenues remained constant throughout, and $461 billion less than static FY 2019 levels. Local tax revenues should fall less precipitously, since localities rely heavily on property taxes, which are significantly more stable. If, however, we assume that local government tax revenues decline at about two-thirds the rate that state revenues do, their five-year losses would represent another $164 billion, bringing the total state and local tax revenue losses—compared to a FY 2020 baseline—to $760 billion.

Greater losses are of course possible, and this assumes flat spending, not a continuation of the spending trajectory prior to the crisis, but it is notable that a current proposal from congressional leadership would actually provide state and local governments with more revenue than they are expected to lose over the next five years. In the past, those governments have been expected to bear some of the loss themselves. Any federal assistance plan must, moreover, be situated within the context of the provisions already implemented for state aid—inadequate, perhaps, but not insignificant.

Aid Currently Available to States

About $500 billion of the nearly $3 trillion the federal government has already appropriated in response to the COVID-19 pandemic flows through to states and localities, in addition to access to hundreds of billions of dollars in loans. A more precise number is difficult to establish, because it requires judgment calls on just what constitutes aid to states and localities: for instance, when the federal government enhances funding for a jointly administered federal-state program but all the benefit immediately flows through to the beneficiaries, regardless of the state’s policy choices, or when some portion of the funding could also be accessed by nonprofits.

A sizable amount of the funding is intended to facilitate responses to the COVID-19 pandemic and is unavailable for general governmental purposes, but in many cases, enhanced funding for specific state programs can provide states with flexibility to reduce their own contribution to that expenditure, freeing up state revenues for other purposes. This is the case, for instance, with $30.9 billion in education stabilization funds; the money must be spent on education, but it can supplant existing state education funding, allowing states to reduce their own contributions and reallocate those dollars elsewhere.

Broadly speaking, funding for states and localities under the CARES Act and other recent legislation and federal governmental actions includes up to $274 billion in emergency supplemental appropriations benefiting states and localities (depending on what is counted), $150 billion in the Coronavirus Relief Fund to cover both health and economic responses to the crisis, $50 billion in Stafford Act disaster relief made available by emergency declaration, $50 billion in increased Medicaid funding, and $11 billion in funding for state and local COVID-19 testing. These appropriations, taken together, total $535 billion. State and local governments may also apply for economic stabilization loans from a pool of $454 billion available to businesses and governments.

Emergency Supplemental Appropriations

The $274 billion in emergency supplemental appropriations for states and localities is the most fraught of these figures.[10] That number comes from the U.S. Senate Appropriations Committee, though observers may differ in exactly how much of the $339.9 billion in emergency appropriations to count as flowing to states and localities, particularly if the intention is to assess these governments’ abilities to cope with a loss of revenue. The $100 billion in new reimbursements to hospitals and health care providers, for instance, is vitally important, and a cost that might otherwise fall to state governments—but it also represents an expense that would not have existed absent the COVID-19 pandemic and therefore was never part of the budgets on which states are now falling short. An appropriation of $45 billion for disaster relief, similarly, is crucial in alleviating a cost states might otherwise incur, but does not address any revenue shortfalls.

On the other hand, states may be able to curtail some of their own hospital spending due to the federal assistance, freeing up those dollars. And not only the Education Stabilization Fund ($30.9 billion), but also the Transit Infrastructure Grants ($25 billion) and other supplemental appropriations, providing additional flexibility for states. Much of this funding comes in the form of new or expanded block grants. The table below lists several of the more significant emergency funding provisions benefiting states and localities.

Table 1. Emergency Supplemental Appropriations to States and Localities Under the CARES Act
Supplemental Appropriation Amount
Hospital and Health Care Provider Reimbursements $100.0 billion
FEMA Disaster Relief Fund $45.0 billion
Education Stabilization Fund $30.9 billion
Transit Infrastructure Grants $25.0 billion
Strategic National Stockpile Funding $17.0 billion
Airport Improvement Program Funding $10.0 billion
Community Development Block Grants $5.0 billion
Homeless Assistance Grants $4.0 billion
Child Care and Development Block Grants $3.5 billion
State Emergency Preparedness Grants* $2.5 billion
Economic Development Administration Funding $1.5 billion
Tenant-Based Rental Assistance $1.3 billion
Community Services Block Grants $1.0 billion
Project-Based Rental Assistance $1.0 billion
Other Appropriations (per Senate allocations) $26.3 billion

* Includes $1 billion authorized under the Families First Act, not the CARES Act.
Source: Senate Appropriations Committee.

Coronavirus Relief Fund

The Coronavirus Relief Fund (CRF), under which states and localities receive $150 billion to cover expenses made in response to the pandemic, is by far the best known of the state aid provisions, and represents the single largest transfer directly to state government coffers. Under the CRF, state governments receive $110 billion, eligible local governments receive $29 billion, tribal lands $8 billion, and U.S. territories and the federal district $3 billion. Allocations are largely by population, though all states are provided a minimum of $1.25 billion in funding.

Only local governments with populations of 500,000 or greater are eligible to receive direct aid, and in such cases, they receive 45 percent of the aid associated with their population, while the state retains 55 percent. The state government receives all aid associated with the population share outside these eligible local jurisdictions. The District of Columbia’s share, calculated separately from state and local shares, is $495 million.

Table 2. State and Local Shares Under the Coronavirus Relief Fund
State Total Allocation State Share Local Share
Alabama $1,901,376,504 $1,786,453,683 $114,922,821
Alaska $1,250,000,000 $1,250,000,000 $0
Arizona $2,822,569,714 $1,857,099,389 $965,470,325
Arkansas $1,250,000,000 $1,250,000,000 $0
California $15,322,206,367 $9,526,137,621 $5,796,068,746
Colorado $2,233,145,460 $1,673,950,246 $559,195,214
Connecticut $1,382,561,117 $913,181,962 $469,379,155
Delaware $1,250,000,000 $927,233,331 $322,766,669
Florida $8,328,721,940 $5,856,159,554 $2,472,562,386
Georgia $4,117,266,353 $3,503,081,996 $614,184,357
Hawaii $1,250,000,000 $862,823,979 $387,176,021
Idaho $1,250,000,000 $1,250,000,000 $0
Illinois $4,913,928,948 $3,519,156,999 $1,394,771,949
Indiana $2,610,646,554 $2,442,324,311 $168,322,243
Iowa $1,250,000,000 $1,250,000,000 $0
Kansas $1,250,000,000 $1,034,052,049 $215,947,951
Kentucky $1,732,491,935 $1,598,690,705 $133,801,230
Louisiana $1,802,727,754 $1,802,727,754 $0
Maine $1,250,000,000 $1,250,000,000 $0
Maryland $2,344,417,741 $1,653,367,701 $691,050,040
Massachusetts $2,672,802,119 $1,655,720,328 $1,017,081,791
Michigan $3,872,742,971 $3,080,874,821 $791,868,150
Minnesota $2,186,958,839 $1,870,033,167 $316,925,672
Mississippi $1,250,000,000 $1,250,000,000 $0
Missouri $2,379,996,144 $2,083,827,229 $296,168,915
Montana $1,250,000,000 $1,250,000,000 $0
Nebraska $1,250,000,000 $1,083,865,742 $166,134,258
Nevada $1,250,000,000 $836,051,100 $413,948,900
New Hampshire $1,250,000,000 $1,250,000,000 $0
New Jersey $3,444,370,826 $2,393,995,126 $1,050,375,700
New Mexico $1,250,000,000 $1,067,817,494 $182,182,506
New York $7,543,778,952 $5,219,021,994 $2,324,756,958
North Carolina $4,067,110,763 $3,585,606,805 $481,503,958
North Dakota $1,250,000,000 $1,250,000,000 $0
Ohio $4,532,845,506 $3,757,703,970 $775,141,536
Oklahoma $1,534,449,890 $1,281,597,801 $252,852,089
Oregon $1,635,570,763 $1,388,745,825 $246,824,938
Pennsylvania $4,964,406,011 $3,935,406,029 $1,028,999,982
Rhode Island $1,250,000,000 $910,740,118 $339,259,882
South Carolina $1,996,588,712 $1,905,229,176 $91,359,536
South Dakota $1,250,000,000 $1,250,000,000 $0
Tennessee $2,648,244,148 $2,363,576,014 $284,668,134
Texas $11,244,137,605 $8,038,804,878 $3,205,332,727
Utah $1,250,000,000 $936,475,736 $313,524,264
Vermont $1,250,000,000 $1,250,000,000 $0
Virginia $3,309,937,372 $3,109,689,845 $200,247,527
Washington $2,952,933,375 $2,167,209,642 $785,723,733
West Virginia $1,250,000,000 $1,250,000,000 $0
Wisconsin $2,257,846,523 $1,997,414,905 $260,431,618
Wyoming $1,250,000,000 $1,250,000,000 $0

Source: CARES Act; Tax Foundation calculations.

Use of payments from the fund is limited to covering costs that are necessary expenditures incurred between March and December 2020 due to the COVID-19 pandemic and which were not accounted for in the budget most recently approved as of the enactment of the CARES Act (March 27). According to guidance from the Treasury Department, the funding cannot be used to close revenue gaps, but it can be expended on both direct health responses and economic relief, and can be applied to payroll expenses for public sector workers “whose services are substantially dedicated to mitigating or responding to” the COVID-19 crisis, including public safety, public health, health care, human services, and similar employees.[11] State and local governments could, for instance, provide a financial relief package for individuals or businesses affected by the crisis, paid for out of the fund.

Governments do not need to make separate applications to the federal government for each expenditure; Treasury transfers the funding to state and local governments, which may then spend it for authorized purposes. A cost meets the requirement of not being accounted for in the budget if either of the following conditions is met: (1) the cost cannot lawfully be funded using a line item, allotment, or allocation within that budget, or (2) the cost is for a substantially different use from any expected uses of funds from such a line item, allotment, or allocation.[12]

Although Treasury has construed eligible uses broadly, the language of the CARES Act precludes using the CRF as flexible funding to cover general government expenditures or patch budget holes. Many states may find it difficult to spend their full allocation on direct responses to the pandemic, which could lead to poorly targeted spending or unutilized funds at the same time that states struggle to balance their budgets. This possibility has led some to propose that states be granted additional flexibility in how they spend their share of the $150 billion.

Other Funding and Financial Assistance

In March, an emergency declaration was made under the Stafford Act, freeing up $50 billion in aid administered by the Federal Emergency Management Administration (FEMA). This money will be provided to state and local governments to facilitate their response to the virus.[13]

The Families First Coronavirus Response Act (FFCRA) provided a 6.2 percentage-point Federal Medicaid Assistance Percentage (FMAP) increase, meaning that the federal government will temporarily take on a greater share of the joint federal-state cost of Medicaid. This is designed both to assist states should Medicaid enrollment increase during the crisis, and to reduce the state’s share so that those state dollars can be put to other purposes.

The Coronavirus Economic Stabilization Act, part of the broader CARES Act, authorizes $454 billion in loans, loan guarantees, and other investments to support eligible businesses, states, or municipalities related to losses incurred as result of the coronavirus outbreak, which can help governments weather the downturn.

Finally, several other aspects of the CARES Act, while not directly funding states, do provide a budgetary boost. The Federal Pandemic Unemployment Compensation program, which provides an additional $600 a week to those receiving unemployment benefits, and the Pandemic Unemployment Assistance program, which expands eligibility for unemployment compensation and fully federal funds coverage to newly eligible claimants, provides income which is taxable in most states. Similarly, the Paycheck Protection Program helps stave off layoffs, maintaining both individual and business tax collections at higher levels than would be expected absent the assistance.

Designing a State Aid Package

As state and local revenue losses come into focus, Congress is likely to authorize additional funding, provide flexibility in the utilization of already appropriated funding, or both, to help governments weather the downturn. The decision to provide such aid, however, creates a new set of choices, including, but not limited to: (1) how to allocate the funding, (2) how much to appropriate, (3) how much flexibility to grant state and local governments in spending it, (4) whether to provide all funding immediately or spread it out over months or years, and (5) whether to repurpose existing appropriations.

Allocation Method

It is possible to imagine many different ways to allocate a new round of aid to states, each with unique advantages as well as shortcomings. The most neutral distribution, consistent with the allocation of aid under the Coronavirus Relief Fund, would be to apportion it on the basis of population, with or without the gross-up for smaller states that created so much consternation in CRF. Alternatively, a measure of the economic impact of the crisis could be used, such as the decline in gross state product or the employment-population ratio. Some have also proposed fiscal measures, like the size of a state’s budget shortfall or its revenue losses against projections. Finally, money could be disbursed in line with the virus’s spread.

One obvious advantage of a population-based measure is its timely administrability, especially contrasted with economic formulae. State populations are already known; the most recent estimates are dated July 2019, and changes since then are modest. Some (largely temporary) movement notwithstanding, the COVID-19 pandemic would not have an appreciable impact on state populations. By contrast, any reliable measure of the impact of the current crisis on state’s economies or revenues will necessarily lag by several months in a situation where even a week can produce marked shifts.

Most states’ April sales tax receipts—representing the first month of a stark decline in consumption in much of the country—will only come in at the end of May. Declines in income tax withholding are beginning to be felt, but only barely. Painful revenue losses are coming, but for now, states have few objective ways to measure those losses, and the data will always be months behind. Similarly, the advance estimate of real Gross Domestic Product (GDP) for April or onward will not be available until the end of July, and state-level statistics will likely be released in October. The most recent state data release, in early April, was for the fourth quarter of 2019.

Using a public health measure like number of confirmed COVID-19 cases is also plausible, and the prevalence of the virus might serve as a rough proxy for the intensity and length of the economic dislocation producing state revenue losses (business closures, stricter social distancing orders, and public wariness). It is also likely to induce states to change testing regimens to unlock access to additional funds, which could be beneficial if it incentivized greater testing in line with the recommendations of health officials, but could also see states try to enhance funding outcomes by revising definitions of a likely coronavirus case or changing testing eligibility in line with the design of the funding formula rather than with best practices. If, for instance, funding flowed based on either the proportion or total number of positive tests, states would be incentivized to disproportionately test (proportion) or at least unduly prioritize (total count) those highly likely to have the virus, neglecting many others.

Each approach has other potential drawbacks too. A pure population count, while arguably the fairest approach, does not take regional impacts of the virus into account. It is also indifferent to the components of a state’s funding mix—income taxes, sales taxes, severance taxes, hospitality taxes, gross receipts taxes, and the like—even though different tax regimes and the composition of each state’s economy will contribute to the severity of the downturn.

Whether this is virtue or vice is not entirely straightforward. There is understandable, and often justifiable, resistance to rewarding states for favoring highly volatile revenue streams, imposing taxes on too narrow a base, or inadequately preparing for an economic contraction. This is not merely a matter of resentment by states that grew government more slowly during good years to better prepare for a crisis. It also reflects a concern about moral hazard, should the lesson of the current crisis be that states need make no provision during a period of growth, and in fact may be foolish for doing so, as all differences between them and their less prudent peers will be wiped out by the governmental response. At the same time, however, states’ needs will also vary in ways that have nothing to do with their level of preparedness or the design of their tax code.

Methodologies based on state economies or revenue shortfalls, in addition to timing issues, can create perverse incentives or reward state policy choices in ways that inherently politicize the process or create opportunities for gaming. Some proposals, for instance, would have state funding conditioned on how much actual collections diverge from initial projections. But projections are not reality, and some states were far more optimistic in their projections than others. It is far from clear why a state should receive greater financial assistance from the federal government not because its needs are greater but because its revenue forecasts were less realistic.

States vary, moreover, in the size of their budgets and the taxes used to pay for them. Even assuming no differences in relative decline in revenue, a state that spends more per capita could receive more relief, not because service provision is more expensive there or needs are greater but rather because the state chooses to spend more than its peers.

This is one of the signal features of fiscal federalism: each state can decide on its own (within certain constraints) what the appropriate level of spending is, and how to raise the revenues to pay for it. But the system starts to break down if the costs of greater government, or the tax savings of fewer services, are wiped out by federal transfers. Similarly, it would seem inappropriate and certainly inequitable were the crisis used as a justification for the federal government to bail out the pension funds of states which have made no real effort to maintain solvency, or to otherwise save states from the consequences of poor choices during an extended period of economic growth, while many of their peers acted with greater circumspection.

Using either fiscal conditions or health measures, moreover, adds an additional wrinkle: snapshot dates. Population is estimated annually, and while there are, of course, modest inflows and outflows at all times, these are small enough to be considered irrelevant for purposes of a funding package. There is simply no question of North Dakota overtaking North Carolina, or New York slipping behind Maryland. By contrast, economic or health measures can change rapidly; an outbreak in one state, or successful suppression in another, can easily allow them to swap places, and some state budgets will undoubtedly prove less resilient than others.

If a measure other than population is chosen, therefore, the choice of snapshot date (or dates) for allocations suddenly becomes vitally important. Distributions based on conditions in early May could make very little sense by June, to say nothing of later in the year—particularly if there is a second wave of the coronavirus outbreak, with entirely different hotspots. Would funding itself come in waves, responding to changing conditions? How much of a lag would be tolerated?

Sometimes, in the face of extraordinary complexity, the best answer is simplicity. A population-based measure is imperfect, but it is likely the best we have: the fastest and fairest method of allocating relief, and the approach least likely to skew incentives or undermine sound policy decisions. In line with the CRF, a population-weighted approach should be favored, either as the exclusive or at least primary method of allocation.

Amount of Relief

Based on Moody’s estimates, FY 2021 state tax revenues stand to be about $154 billion lower than they were in FY 2020, and $33 billion lower than Moody’s anticipates in FY 2020, a year the ratings services estimates will itself close 11 percent lower than FY 2019.[14] Neglecting inflation, the two-year revenue decline for states is an estimated $274 billion, or—put another way—$330 billion lower than two years at originally projected FY 2020 revenue levels. And a scaled estimate puts total state and local losses for FYs 2020 and 2021 slightly above $400 billion, or just over $480 billion lower than two years at anticipated (but unrealized) FY 2020 levels.

The following table shows anticipated revenue losses for FYs 2020 and 2021 against both actual revenues from FY 2019 and the amounts states projected at the start of their current FY 2020 budgets.

Table 3. Anticipated Revenue Losses Against FY 2019 Actual and FY 2020 Projected Baselines
  FY 2019 Actual Revenues   FY 2020 Preliminary Budget
Years State Local Total   State Local Total
FY 2020 $120.5 $56.9 $177.4   $148.5 $68.3 $216.8
FY 2021 $153.5 $72.4 $225.9   $181.5 $83.5 $265.0
Both Years $274.0 $129.3 $403.3   $330.0 $151.8 $481.8

Sources: Moody’s Investor Service; U.S. Census Bureau; Tax Foundation calculations.

States entered FY 2020 with about $77.4 billion in revenue stabilization (“rainy day”) funds[15] and another $15.8 billion in unappropriated balances, bringing states’ total balances at the outset of the current fiscal year to $93.2 billion.[16] This represents about 28 percent of the amount states would need to cover to operate for both the mostly completed FY 2020 and soon to begin FY 2021 at budgeted FY 2020 levels.

States should not, however, continue to operate at those levels. State general fund spending declined by nearly 13 percent in real terms between FYs 2007 and 2010, dropping 0.6 percent in FY 2008 (reflecting growth in the first half of the year followed by declines at year’s end), 5.9 percent in FY 2009, and a further 6.7 percent in FY 2010. Cuts along a similar trajectory would save $65 billion through FY 2021 and a further $116 billion in FY 2022.

It may not be desirable for states to cut spending as deeply as they did during the Great Recession, but it is impractical and likely disadvantageous for them to operate as if nothing has changed. The federal government’s ability to deficit spend confers advantages that argue in favor of a federal role in stabilizing state revenues, but it is reasonable for states to make some adjustment to their own spending during an economic downturn, and to draw upon their reserves.

Local governments, though their losses may not be as steep (in most, but not all, jurisdictions), tend to have far fewer reserves and less financial flexibility—including, in many cases, less to cut, given that municipalities have a greater emphasis than state governments in delivering core services. Any aid package will undoubtedly account for the financial challenges faced by localities as well and should reflect this reality.

Revenues, moreover, will take time to recover to prior levels. Using Moody’s projections as a starting point, state and local tax collections between FY 2020 and FY 2024 could come in $760 billion below a FY 2020 baseline, almost two-thirds of which will be experienced by the end of FY 2021. Some of this should come from the federal government in the form of aid and loans, and some from states in the form of adjustments to both expenditures and revenues. Policymakers must grapple with the vexing question of just how much of the burden should be assumed by the federal government, when, and in what form.

Despite the severity of the downturn during the Great Recession, lawmakers only provided $143.6 billion in general fiscal relief, and about $290.4 billion in total, covering no more than 40 percent of state budget gaps.[17] Congress may choose to be substantially more generous this time, and not just proportional to greater revenue losses, but amounts should be chosen with care, with the expectation of all levels of government participating in the solution, rather than the federal government facilitating a business as usual approach for subnational governments or even subsidizing a surge in new expenditures that will prove difficult to sustain once federal aid is withdrawn.

Timing of Aid

If an aid package is intended to help see state and local governments through several years of lower revenues, policymakers should consider whether to stagger disbursements rather than provide all resources upfront, which might enable states to postpone important fiscal policy decisions by overutilizing funding in the first year. If the federal government provides funding intended to see the states through the next three or four years, states should not be in a position to make further application for additional funding unless conditions worsen beyond projections. It makes more sense to provide multiple years’ worth of relief in installments than to provide it to states all at once, with the potential that political pressures will lead to immediate overreliance. It may also make sense to provide for FYs 2020 and FY 2021 at present, and revisit additional years of assistance later, rather than attempting to project the economic impact of a pandemic for years into the future.

There are other timing considerations as well. Most states have adopted their budgets for the fiscal year beginning July 1 (FY 2021), and many of these budgets incorporate only limited adjustments from trajectories established before the crisis began. Federal lawmakers are likely to want to act in time for states to have access to additional funding to close out FY 2020 or at least for the entirety of FY 2021, though ideally in a way that does not relieve them of any responsibility to make their own adjustments or encourage them to postpone important decisions.


Flexibility is a relative concept, as the experience of Great Recession-era relief demonstrates. In its maximalist form, flexibility involves a direct grant of funding to be used for any valid governmental purpose, or for offsetting revenue losses. By this definition, the federal government did not provide any flexible funding to states to help them cope with the revenue privations of the Great Recession. But because money is fungible, another form of flexibility is associated with funding for existing programs that allows the states’ own contributions to be repurposed to meet other needs.

To this end, the State Fiscal Stabilization Fund within the American Recovery and Reinvestment Act (ARRA) of 2009, under which $53.6 billion was distributed to states by the U.S. Department of Education,[18] is generally regarded as flexible funding. The money had to be used to cover state education expenditures, but states were free to drawn down their own funding of education commensurately, essentially turning the program into an injection of funding for general purposes. Most also regard the FMAP increase, termed the State Fiscal Stabilization Fund, a form of flexible aid, because $90 billion in additional Medicaid funding provided by the federal government freed up an equivalent amount of state dollars for other purposes.

Notably, the Families First Act has already increased FMAP by 6.2 percent, worth $50 billion, with another $30.9 billion in an Education Stabilization Fund that is being used flexibly much in line with the State Fiscal Stabilization Fund during the Great Recession. Other block grant increases, like the $25 billion for transit infrastructure, could also provide governments with additional flexibility. In other words, flexible funding during the COVID-19 crisis already rivals total flexible funding during the Great Recession, though its designations are less clear.

By contrast, the CRF is not flexible, inasmuch as it can only be spent on expenditures directly arising from the COVID-19 pandemic and states’ economic and public health responses to the crisis. While certain ongoing personnel costs can be covered with the funds, provided those workers are engaged in coronavirus response, it is not like the old stabilization funds or even the new block grant increases in enabling states flexibility in covering their existing budgets.

Flexibility should be favored in most cases, as it allows states to meet their most urgent needs, rather than use a one-size-fits-all approach that may overfund some programs while states struggle to meet obligations in other areas. Flexible funding also has less risk of becoming permanent than does enhanced federal spending on specific programs. Still, it is important to recognize that much of the funding already provided under the CARES Act, while not directly available to backfill revenue losses, is nonetheless fairly fungible, freeing up states’ existing revenues to meet other needs.

Given states’ complaints about being unable to use a significant portion of their funding under the $150 billion CRF,[19] a first step for Congress might be repurposing the CRF, permitting the states to use the remainder of their allotted amount for general, or at least a broader range of, expenditures. Block granting support for existing joint federal-state programs may also provide states with flexibility that could enable important cost savings.

Another potential way for the federal government to support states and localities with a fair amount of flexibility would be the restoration of Build America Bonds, part of the ARRA response to the Great Recession. State and local governments could issue these debt securities to cover capital expenditures, with the federal government subsidizing 35 percent of their interest payments, either through refundable tax credits for the bondholder or by direct payments to the issuers. This lowers the cost of financing for state and local governments, allowing them to issue more bonds and, where necessary, to offer more generous terms, ensuring that capital projects could continue to be financed, and allowing existing revenues to be used for other vital purposes.[20]

Unintended Consequences

No sizable transfer from federal to state governments can be created without some undesirable side effects, even without reference to the burgeoning national debt necessary to support such a transfer. Policymakers should, however, be aware of these negative ramifications and seek to minimize them where possible.

If transfers are so generous as to allow states to increase spending during the crisis, or are structured in such a way that a state could injudiciously run through a multiyear appropriation in a single year, they may only postpone, not attenuate, states’ fiscal crises. Spending decisions made while receiving generous federal assistance would either have to be scaled back afterward, which can be difficult, or supported by higher taxes, which may be inadvisable. Whatever role the federal government may have in aiding states in the present crisis, federal lawmakers should avoid injecting themselves into a broader debate about the size of state governments.

There is also a risk of an undesirable level of added entanglement of state and federal budgets. States already administer a range of fully or partially federally funded programs, but our system of fiscal federalism encourages states to make their own decisions on many spending priorities and the funding necessary to underwrite them. What works for Massachusetts may not work for Montana, and vice versa. As the federal government takes on a greater role in financing state budgets, there is a danger that states will come to rely on this aid even after the crisis has ended, making it difficult for the federal government to eliminate the new transfers—and that any lingering aid will also involve greater federal influence over state policymaking. Taking this concern seriously favors flexible over dedicated funding in most cases, as enhancements of dedicated funding may be more difficult to withdraw, though it also entails caution regarding both the size and timing of fiscal relief.

Some states, moreover, were fiscally imprudent well before the crisis. Illinois has a nearly $7 billion backlog of unpaid bills,[21] and six states—New Jersey (34 percent), Illinois (39 percent), Indiana (42 percent), Colorado (43 percent), Kentucky (46 percent), and Connecticut (47 percent)—have pension systems that are less than 50 percent funded.[22] An aid package that bails out mismanaged pension funds or otherwise saves states from the consequences of imbalanced systems not only rewards poor fiscal choices but also delays potentially vital state reforms. It seems appropriate for an aid package to respond to the COVID-19 pandemic, and not to be structured in such a way as to disincentivize needed reforms. This would argue for a population-based allocation method, or at least one that is not based on budget gaps or measures of a state’s fiscal health.

The choices made about the size, allocation, timing, and flexibility of relief will have meaningful consequences, and can interact with each other in complex ways. The following table summarizes some of the key implications of each major option.

Table 4. Key Considerations in Designing a State Relief Package
Design Choice Considerations

Allocation Method

Choice of allocation method interacts with every other choice federal lawmakers face on the size, timing, and flexibility of a relief package


Favors timely administration, does not skew state incentives or reward fiscal imprudence, fails to capture disparities in the virus’s impact

Economic or Fiscal

Tied more directly to specific needs, but long lag in relevant economic data, consistency issues, discriminates based on size of government, can reward poor planning, and dissuades states from necessary reforms

Coronavirus Cases

Proxy for need independent of state’s budgetary choices, but has snapshot date issues, neglects costs in states that have kept case counts low at significant economic cost

Amount of Relief

Inadequate levels could force states to make economically harmful tax and budgetary decisions that slow the recovery, but excessively generous provision staves off necessary reform and introduces moral hazard

Timing of Aid

Multiple years of funding in a single installment could delay difficult decisions, leading to a sharper contraction in the end


Greater flexibility will allow federal funding to flow to states’ most urgent needs and is less likely to create long-term dependency


States face mounting revenue shortfalls amid considerable uncertainty. State and local government officials are no stranger to recessions, but the current crisis defies easy comparison. We can only guess at the long-term fiscal ramifications of the crisis, though it is certain that the economic toll—for individuals and governments alike—will be substantial.

There may well be a role for the federal government in providing additional assistance to states as they grapple with shortfalls, helping them to stave off painful spending cuts or counterproductive tax increases, but in designing any such aid, federal lawmakers should be careful not to eliminate incentives for state and local governments to prepare for the next downturn or to induce them to undertake imprudent expenditures. Although imperfect, allocation by population likely represents the most equitable, and certainly the most straightforward, approach to distributing aid to states.

If lawmakers intend to craft a relief package for states, they should act expeditiously but not hastily. The lesson of coronavirus response legislation to date is surely that hastily drafted provisions can have unintended consequences, and that imprecision can be the cause of great confusion and consternation. The decisions they make could reverberate in state capitols for years and cast a long shadow on state financial systems. Designing a state aid package merits careful deliberation; it cannot simply be a matter of picking a number.

[1] National Governors Association, “National Governors Association Outlines Need For ‘Additional And Immediate’ Fiscal Assistance To States,” Apr. 11, 2020,

[2] Daniel Flatley, “Senators Propose $500 Billion State and Local Government Rescue,” Bloomberg Quint, Apr. 20, 2020,

[3] Erik Wasson, Billy House, and Laura Litvan, “Pelosi Says States and Cities Seek $1 Trillion in Next Stimulus,” Bloomberg, Apr. 30, 2020,

[4] Congressional Budget Office, Economic Effects of the American Recovery and Reinvestment Act of 2009, Mar. 2, 2009, (archived copy).

[5] Center on Budget and Policy Priorities, “States Grappling with Hit to Tax Collections,” May 4, 2020,

[6] Moody’s Investors Service, “Revenue Recovery from Coronavirus Hit Will Lag GDP Revival, Prolonging Budget Woes,” Apr. 24, 2020.

[7] National Conference of State Legislatures, “Re: Flexible Stimulus Funds for States to Stabilize Economy,” Apr. 16, 2020 (letter to Congress).

[8] National Association of State Budget Officers, “Fiscal Survey of States,” Fall 2019, 8,

[9] Id.; Moody’s Investors Service, “Revenue Recovery from Coronavirus Hit Will Lag GDP Revival, Prolonging Budget Woes”; U.S. Census Bureau; author’s calculations.

[10] Senate Appropriations Committee, “$340 Billion Surge in Emergency Funding to Combat Coronavirus Outbreak,” April 2020,

[11] Department of the Treasury, “Coronavirus Relief Fund Guidance for State, Territorial, Local, and Tribal Governments,” Apr. 22, 2020,

[12] Jared Walczak, “New Guidance on State Aid Under the CARES Act,” Tax Foundation, Apr. 23, 2020,

[13] National Conference of State Legislatures, “President Trump Declares State of Emergency for COVID-19,” Mar. 25, 2020,

[14] Moody’s Investors Service, “Revenue Recovery from Coronavirus Hit Will Lag GDP Revival, Prolonging Budget Woes.”

[15] Jared Walczak and Janelle Cammenga, “State Rainy Day Funds and the COVID-19 Crisis,” Tax Foundation, Apr. 7, 2020,

[16] National Association of State Budget Officers, “The Fiscal Survey of States,” 75.

[17] The Henry J. Kaiser Family Foundation, “Update: State Budgets in Recession and Recovery,” October 2011, 5,

[18] U.S. Government Accountability Office, “Recovery Act: Funds Continue to Provide Fiscal Relief to States and Localities, While Accountability and Reporting Challenges Need to Be Fully Addressed,” GAO-09-1016, September 2009, 39-40,

[19] See Pat Ralph, “CARES Act Restrictions Prevent New Jersey from Utilizing Full Benefits, Gov. Murphy Claims,” Philly Voice, Apr. 23, 2020,

[20] Andrew Ang, Vineer Bhansali, and Yuhang Xing, “Build America Bonds,” The Journal of Fixed Income 20:1 (Summer 2010),

[21] Illinois State Comptroller, “Bill Backlog,” May 6, 2020,

[22] Data-Z, “Pension Database,” Truth in Accounting, 2018 data,

Source: Tax Policy – Designing a State and Local Government Relief Package