A Framework for Hard Choices
Choosing Among Options to Address New York’s State and Local Fiscal Stress
One of the many distressing consequences of the Covid-19 pandemic is serious fiscal problems for state and local governments. In New York, the epicenter of the pandemic, the fiscal stress for its major public entities – the State of New York (the State), the City of New York (the City) and the Metropolitan Transportation Authority (MTA) – is acute. This policy brief identifies and assesses options available to governments to deal with their fiscal problems. A companion report released today demonstrates how the State can adopt a balanced package of actions that at this stage best addresses the needs of New Yorkers, and a future CBC report will apply the framework to the City.
The public health measures necessary to curb the spread of the pandemic have required a near shutdown of economic activity, leading to a sharp economic recession. This, in turn, has reduced the state and local tax base and deprived governments of anticipated revenues. Since state and local governments typically are required to adopt balanced budgets, revenue shortfalls require hard choices about how to bring revenues and expenditures in line with each other. In addition, the pandemic has generated unanticipated expenses for public health, health care, and other essential services. While many of the added expenses will be reimbursed by federal aid programs, the State and local entities face immediate cash flow problems and some expenses may not qualify for federal aid.
The current situation has some parallels and some important differences with the prior recessions of 2001–2002 and 2008–2009 and the fiscal stress stretching beyond those years. Those periods included major unanticipated revenue shortfalls, increased spending for safety net programs such as Medicaid, and significant federal aid to states and localities. The current situation continues to evolve, but it is likely to be a period of deeper and more rapid decline than previous recessions, and the time required for recovery is difficult to discern. Although the federal government has already taken major actions to help cope with economic issues, the full nature and scale of the federal response including aid to state and local governments is still emerging. Accordingly, the experience of previous recessions may be instructive, but the lessons should be interpreted cautiously in light of the pandemic’s unique impacts and the current political context.
The Options and Their Implications
A wide range of options is available to balance revenues and expenditures. State and local leaders should choose among them based on their impacts, with the aim of minimizing harm, particularly to New Yorkers most in need, and maximizing benefit for the greatest number of New Yorkers. Two other implications should be given strong weight in these decisions. First is the impact on the long-run attractiveness of the jurisdiction as a business location and place of residence. When the pandemic has passed, the actions taken should have had minimal impact on the factors that have made New York a desirable and economically competitive location. Second is the impact on future generations of New Yorkers. The public finance principle of “intergenerational equity” is a sound guideline; future generations should not be obliged, usually through long-term borrowing, to pay for services for today’s residents. Moreover, the debt service for such borrowing constrains future resources for services and makes future budget decisions harder. Wise public officials will give careful consideration to the implications of their decisions for their heirs.
With these considerations in mind, five broad options (with some variation) are assessed: use of reserves, expenditure reductions, tax increases, asset sales, and borrowing.
Use of Reserves and Unallocated Resources
Governments often have available resources in the form of reserve and unallocated funds that take various forms. “Rainy day” funds are explicitly set aside for periods of economic decline or natural disasters. Other reserves are included in annual budgets, typically as a “general reserve” for unanticipated events or budget flexibility. Elected officials also tend to view some unused funds set aside for special purposes as available reserves; notably many City leaders perceive funds set aside in the Retiree Health Benefits Trust (RHBT) as equivalent to a rainy day fund, and State leaders have viewed cash balances held by various public authorities as available to “sweep,” that is, transfer to the State’s operating budget. Sizable funds received by the State under special settlements from financial and other institutions over the last decade also can be viewed as reserves to the extent they remain unallocated or unused.
The implications of using the different forms of reserves vary. Rainy day funds and general reserves can be used in difficult times with little or no adverse impact. While it is appropriate to use the funds, disagreements can arise over the scale or pace of use of these reserves. Those concerned with long-term consequences generally favor more limited use of reserves in order to keep some resources available in case difficult times continue or new problems emerge. For example, the Citizens Budget Commission (CBC) has recommended using no more than two-thirds of any rainy day fund in the first year. It should be noted that reserves, once used, do not provide ongoing budget relief and serve as a bridge to a period when either revenue strength resumes or spending reductions substitute.
Use of other funds in an emergency generates a conflict between the immediate budgetary need and the funds’ designated purpose. For example, while the RHBT only contains a small fraction of the resources necessary to pay future benefits, drawing on those funds for budget relief makes funding those benefits more challenging in the long run. Similarly, drawing on the funds of public authorities such as the New York Power Authority or the New York State Energy Research Authority means less will be available for those entities’ missions or its customers may have to pay more in the future for services such as electric power or energy conservation.
Even in the best of times government leaders should improve the efficiency of public service delivery, thereby reducing expenditures or getting more value from resources. In harder times these efforts should be more substantial and extended to other forms of expenditure reduction. These expenditure cuts can take multiple forms, and the implications of these alternative forms of reductions vary.
Reducing expenditures by delivering services more efficiently is the most beneficial because consumers can receive the same or better level and quality of service, but at a lower cost. Such efficiencies can be achieved through enhanced use of technology, the reform of outdated work practices and processes sometimes perpetuated through organizational inertia, or more effective management and supervision. Some efficiency improvements can be implemented by managerial prerogative while others may require modifications to collective bargaining agreements.
Improving efficiency can be challenging, and using data-driven performance management can provide insight. In addition, change can be spurred by managerial incentives and “gain-sharing” arrangements. “Gain-sharing” may be useful where changes require labor-management agreement to change work rules; they can often achieve significant savings even if requiring some added compensation for affected staff.
However, efficiencies may negatively affect workers who may no longer be necessary or may earn less in overtime. Workforce reduction through attrition has the fewest negative consequences on employees; however, important limitations to this approach are that the employees who leave may not be well matched with the targeted efficiency and implementation is gradual. Buyouts—paying employees an incentive to retire or leave a job—also can prune the workforce and achieve short-run savings; however, buyouts have often increased long-term pension costs.
Efficiency improvements preserve services while saving resources, but expenditures also can be reduced by decreasing the volume or quality of services. Such reductions have obvious negative consequences– service recipients suffer, and public employees suffer because the reductions are achieved by reducing staff or reducing staff hours. However, service reductions can be focused on programs with limited benefits or that are lower priority.
Any necessary staff reductions can be implemented through attrition or layoffs. Attrition causes less direct harm to current workers, but limits managerial control over which organizational units are immediately affected. Layoffs can better target the service to be reduced, but directly affect people who may want to remain employed, are typically concentrated among the most recently hired workers, and can cause disruptions in agency operations due to “bumping” rules that allow those with more seniority to select assignments that become vacant. In addition to the direct impacts on consumers and staff, service reductions have a risk of lowering the relative attractiveness of a jurisdiction because the availability of high-quality services is one element of an area’s competitive advantage.
In some cases staff reductions may not mean fewer services because the demand for services has fallen and the ongoing staffing supports unused capacity. This situation is particularly relevant in the current pandemic. Some functions are staffed for service levels currently not required because nonessential activities are closed and some essential services are being delivered in new ways. Examples include building inspectors, driver license testers at the Department of Motor Vehicles, and some janitorial staff at closed public schools. City and State managers are reluctant to lay off these workers because the decline in demand for service is seen as temporary and the workers may be needed in the future; in addition, there are adverse impacts for the individuals and the economy.
Furloughs are a way to adapt to these situations. Employee work obligations and pay can be suspended for a temporary period with a commitment to return to full employment at the end of the emergency. However, the lost income will have a significant detrimental impact on the affected workers. Alternatively, another option for this situation may be partly paid furloughs, during which workers can receive income equivalent to most or all of their base salary (and continue to be covered by employment-based health insurance) through a combination of Unemployment Insurance Benefits (UIB) and partial pay from their employer. Such arrangements are available in European countries and could be adapted in New York through agreement with unions and adjustment to current UIB eligibility constraints combined with partial pay from the City or State. These arrangements can protect worker income and health insurance while providing fiscal savings.
Lower Employee Compensation
Expenditures can be reduced by lowering planned increases to employee compensation. This can take the form of a pay “freeze” under which workers do not receive scheduled contractual pay raises or more drastic steps such as lowered wages or “payless paydays” in which some time is uncompensated. The fringe benefit portion of compensation also can be reduced by requiring new contributions from workers and retirees for their health insurance.
The negative implications of this approach are concentrated among the affected employees who suffer reduced income. However, this may be justified when public sector fringe benefits exceed many private sector counterparts and wages for selected occupations are higher in the public sector (although the opposite is true for other occupations). To the extent state or local government employees’ compensation is brought into line with that of the private sector in the long run, these adjustments also have the positive impact of enhancing the jurisdictions’ attractiveness by helping to keep taxes at competitive levels.
Reduced Capital Investments
Like many jurisdictions, the State and the City have distinct operating and capital budgets. The two are connected in two ways: (1) The operating budget may allocate some revenue for capital investments, an arrangement known as “pay-as-you-go” or “paygo” capital; and (2) The capital investments may be financed through long-term borrowing with the debt service (interest and principal payments on long-term bonds) for the bonds paid through the operating budget.
Operating budget savings can be achieved in the near term by switching paygo financing to long-term borrowing (although total costs increase due to the interest expense). Operating budget savings also can be achieved by reducing the amount of capital investment financed by borrowing. In this case the short-run savings are less than the amount of investment foregone because the annual debt service removed from the operating budget is only a fraction of the total cost of the capital investment. Nonetheless, it can be a significant operating budget reduction; at current interest rates for 30-year bonds each $100 million of capital spending avoided saves about $6.5 million annually in debt service.
An important risk of this measure is neglecting needed repairs and renovations of essential infrastructure, which can jeopardize effective service delivery and reduce the competitiveness of the jurisdiction. This drawback can be minimized by making cuts to capital projects that are not for essential infrastructure (including projects characterized as “political pork”) or projects that are otherwise desirable expansions or enhancements that can safely be postponed.
Operating expenditures can be reduced by shifting some current obligations to another level of government or to the nonprofit sector. This does not lower the total cost of a service, but can yield savings to a particular governmental unit. At the State level the shifting is typically from the State to local jurisdictions by cutting aid to local governments that continue to provide the service, or by mandating some financing of current State obligations by local governments. At the local level the shifting may be from the locality to the nonprofit sector, typically by reducing payments for services without reducing required performance quality or amount of service delivered. Sometimes cuts to nonprofit organizations lead to efficiencies on their part; however, they also can result in lower quality and greater need for philanthropic support.
The implications of cost shifting are mixed. To the extent aid cuts lead to efficiencies by local governments and nonprofits, the consequences are positive. When the cost shift is to jurisdictions that should more appropriately finance the service, the consequence ultimately also is positive, though in the short run implications may vary. For example, wealthy school districts that have high levels of service can choose either to offset State aid cuts by increasing their taxes or better align their systems’ services with districts with closer to average spending. In contrast, shifting costs in a manner that inappropriately burdens local jurisdictions for services that should be financed from the broadest possible tax base is inequitable and should be avoided. For example, CBC’s studies of Medicaid financing (most recently in 2018) have demonstrated that New York’s large financing role for localities is highly inequitable and should be reversed rather than accelerated.
The implications of shifting costs from government to nonprofits are uncertain and potentially problematic. To the extent philanthropy cannot offset cuts in a time difficult for fund raising, essential services for the indigent may suffer in availability or quality or both.
Increasing tax rates to generate new revenue has the benefit of providing resources to support essential, and perhaps desired, services. The negative implication is the harm it may cause to the jurisdiction’s competitiveness and the cost it imposes on those who pay the new tax.
Tax increases can be designed to lessen the negative impacts. Raising tax rates temporarily for an emergency period moderates the impact on competitiveness. Current and potential residents and businesses may be more tolerant of a tax burden that is temporary in a period of stress than increases that become a more permanent feature of the jurisdiction. However, in New York the credibility of a “temporary” tax increase is widely questioned because such increases in prior recessions have not been fully reversed as times improved.
The impact of a tax on competitiveness should take into account specific circumstances of the jurisdiction at a point and over time. The burden of tax increases on individual incomes in a period of economic stress can be moderated by targeting the rate increases in a highly progressive way. Increases in the highest brackets of the personal income tax have been used to achieve such targeting. In New York this strategy has limitations because top marginal rates are already high relative to most other jurisdictions; the impact on tax bills relative to other jurisdictions has been further exacerbated by the federal cap on the deduction of state and local taxes. Competitiveness can be harmed because high income households are typically mobile and could avoid the tax (and, therefore, lower total tax collections) by relocating. As an alternative to this progressive targeting, it may be preferable to have any tax increase be relatively small for a broad swath of the population on grounds that the hardship of a catastrophe should be widely shared.
The immediate nature of economic stress also can affect the impact of tax increases. For example, potential widespread rent non-payment or delinquencies during this pandemic puts stress on multifamily and commercial property owners, making a significant increase in property tax rates burdensome.
Governments own assets that are physical, like buildings, and financial, such as tax liens. These assets can be sold to raise one-time resources to support operations. The negative implication of this approach is that the entity’s “balance sheet” value is lowered. The more fiscally prudent policy is to use proceeds from assets sales to create equivalent value on the balance sheet; that is, use the proceeds for capital investments or keep them as reserves.
However, exceptions to the fiscally prudent practice can be appropriate in some emergency situations. Long-term harm is minimized if the assets sold are excess capacity or obsolete facilities, or if the assets could be effectively managed and financed by the private sector. Examples include the sale of closed prisons or of recreational facilities like golf courses that could be operated privately. Still, like other non-recurring resources, this option serves as a bridge to a period when either revenue strength resumes or spending reductions substitute.
Borrowing for Operations
Borrowing can be used to raise money for operations as well as for capital projects. In this context an important distinction is between short-term borrowing, usually for less than one year, and long-term borrowing, which extends for more than one year.
Short-term borrowing is used to deal with cash flow problems; when the timing of expenditures and revenue collections are not well matched an entity may have to make cash disbursements before the revenue that supports those expenses is collected. If the revenue is realistically expected to be collected within a year (and used to support an expense within the same year), it is practical and prudent to borrow for a few months in order to be able to make the scheduled payments.
Formal short-term borrowing usually takes the form of tax anticipation notes (TANs) for expected tax collections or revenue anticipation notes (RANs) for expected intergovernmental aid. The only negative implication of this borrowing is the interest cost it carries. If the notes are paid on time, no longer-term negative impacts arise, and it constructively resolves cash flow problems. Greater negative impacts are created if such short-term borrowing is not fully repaid and is repeatedly “rolled over” or converted to long-term debt. In these cases the negative impacts are the same as those associated with long-term borrowing for operations discussed below.
Another form of short-term “borrowing” is the delay of payments due for which sufficient cash is not available. That is, governments can simply not pay their bills on time. In some cases, vendors receive interest on delayed payments. When no such requirement is present in a contract, government essentially receives an involuntary, interest-free loan from the affected vendor. While the government benefits from the lack of interest cost, the delay imposes a burden on the firm expecting the payment. A longer-run consequence of this practice likely is that vendors charge higher prices in anticipation of such payment delays.
A form of payment delay that extends beyond a single year is the deferral of government employer contributions to employee pension funds. These contributions are intended to be made annually based on calculations of what is needed to keep the pension funds actuarially sound. Governments may reduce the needed contribution in one or more years and pledge to make up the shortfall by increasing the contribution in future years. This is a form of borrowing with no negative implications for the affected employees, who still receive full pension benefits, but with substantial cost for future taxpayers who will have to make higher future contributions and pay interest on the deferred amounts because they must make up the expected return the pension fund would have earned if it had the money.
Long-term borrowing for operations is generally considered inappropriate because it violates the principle of intergenerational equity by requiring future taxpayers to pay for the services consumed by current residents. The practice also harms a jurisdiction’s competitiveness by imposing costs that have no benefit to future businesses and residents. The City and State already suffer to some extent from this practice in the form of debt issued on behalf of the City during its 1970s fiscal crisis, which is still being paid off, and long-term debt issued on behalf of the State in the 1990s to eliminate its notorious “spring borrowing” habit.
In addition to new borrowing for operating purposes, debt can be used to achieve savings during an emergency period by restructuring existing debt. As previously noted, debt service is part of operating budgets. Restructuring can take two forms. Debt can be “stretched,” meaning the term of the bond is extended. This lowers the annual debt service. Debt service also can be “backloaded,” meaning more of the principal and interest is paid in later years and substantially less in the earlier years.
Such restructuring has two clear negative implications. First, the total cost of the borrowing increases because interest costs are greater due to the longer period or the backloading. Second, the new arrangement violates the principle of intergenerational equity. Sound public finance generally requires that capital investments be financed with bonds of a duration equal to the useful life of the asset purchased and that the debt service be equal in each of the years reflecting the equal use of the asset by each set of future residents. Extending the bond beyond a period of useful life or making debt service unequal among cohorts of residents is inequitable.
Despite their harmful impacts, new borrowing for operations and debt restructuring are sometimes justified by “once in a generation” emergencies such as the terrorist attacks of September 11, 2001 and, perhaps ultimately, the current pandemic. Such borrowing should be only used a “last resort” measure after other appropriate actions are exhausted— and federal aid proves insufficient— and should have conditions to protect fiscal prudence in the future. For example, such borrowing by the State could be conditioned on a requirement that it balance its budget in accord with Generally Accepted Accounting Principles, a condition imposed on the City when it indulged in borrowing for operations in the 1970s. Such conditions can be part of bond covenants, not necessarily a constitutional amendment or state statute.