Page 10 of 14

Appendix D: Sensitivity Analyses

This appendix discusses the possibilities that the central conclusions of this report are wrong.

PREDICTIONS OF STRUCTURAL DEFICITS

The prediction of widespread structural deficits in state and local government throughout the next decade is the cornerstone of the conclusions in this report. That prediction naturally depends on definitions of what constitutes current service spending, which are discussed in this section. Given the definitions that are common in state budgeting and similar projections, the prediction of structural deficits depends on demographic, economic, and federal budget assumptions.

Demographic Assumptions
The underlying demographic assumptions reflect relatively stable patterns of demographic change. One major variable in the population estimates is immigration. But as indicated below, over time, immigration has somewhat parallel impacts on spending and revenues.

Changes in longevity would have major impacts on total government spending. Most of the adverse impact, however, would be concentrated on the federal budget, which bears total responsibility for maintaining safety net income through the Supplemental Security Income program, for preventing poverty in retirement through Social Security, and for providing the bulk of medical care for the aged through Medicare. State and local impacts would be confined primarily to health care spending supplementing Medicare and, most important, to increased nursing home outlays shared by the federal government and the states through Medicaid.

Changes in birth rates would not have much impact on state and local spending during the eight-year projection period of the forecasts in this report. The primary impact of children on state and local budgets comes only when they reach school age, so there is about a five-year lead time between birth rate changes and spending impacts.

Economic Assumptions
The predictions assume a long-term rate of real growth in the U. S. economy, about 2.4% as measured by the total output of goods and services (Gross Domestic Product or GDP) adjusted by a price index known as the GDP price deflator. Economic growth in excess of the predicted rate is an unmitigated blessing for government finances. It adds to revenues because tax collections are closely related to total economic activity. But it does not add to spending pressures, which are based on such factors as the number of children in school. More rapid growth than previous predictions, more than all other factors combined, accounts for turning the federal deficit into a surplus and for the strong current fiscal position of state governments.

It is possible that national economic growth could continue to exceed what forecasters view as the long-term trend. The fiscal impacts would depend on how the growth came about. Total output can be considered as the product of two factors: (1) total worker hours, and (2) productivity or output per worker-hour. So long as growth resulted from unexpected gains in productivity, there would be unambiguous gains in private living standards and in government revenues. However, such gains would be unprecedented in the nation's economic history. They are increasingly hard to realize because a growing proportion of economic activity is taking form (e.g., beauty care, most health care, and many other forms of personal services) where productivity gains are harder to realize than in manufacturing and construction. If unexpected growth results from a larger percentage of the nation's population working or a lengthening of the work day or work week, the fiscal consequences would also be overwhelmingly favorable. However, output gains that result from immigration of the labor force would not be favorable, as the unanticipated population growth would soon lead to as many sources of additional spending as the added revenues.

To the author and many other observers, the downside risks to assumed economic growth rates seem more serious than the upside risks. Individual forecasters each have their own favorite causes of concern. The author's is the increasing dependence of growth in national output on money from abroad. Currently, Americans are able to spend about $900 per person (including children) per year more than the value of their production because capital inflows from other nations (the reciprocal of a negative balance of trade in goods and services) now equal that amount. Centuries of economic history suggest that other nations will not subsidize the consumption of one nation to that extent indefinitely.

While the projections assume an inflation rate, the projections of structural deficits are largely insensitive to assumed rates. Inflation has nearly equal impacts on spending and revenues.

Federal Budget Assumptions
Assumptions about the federal budget affect higher education planning in two ways. First, the assumptions affect presumed federal outlays for higher education made directly in the form of Pell grants and other programs. Second, the assumptions affect presumed state and local outlays for higher education through the impact of federal spending on the fiscal position of state and local governments. With the federal budget supporting nearly 25% of all state and local spending, that fiscal position is now highly sensitive to what happens to future federal aid.

The outlook for federal domestic discretionary spending -- the category that includes both direct federal support of higher education and most grants to state and local governments -- is mixed.

On the one hand, actions in 1998 suggest this report's assumptions in this area will be easily realized. These projections assume that federal aid to state and local government will simply remain a constant percentage of personal income. The projections of federal surplus -- which governed the congressional budget allocations in 1998 and the many decisions in Congress to increase aid programs -- suggest that federal officials have both the desire to spend more on domestic discretionary programs and the fiscal capabilities to do so.

On the other hand, the spending caps of the 1997 budget agreement remain in place and both the President and the congressional leaders have indicated that these agreements will not be violated. These caps were not seriously restrictive in 1998, as they were designed to avoid pre-election budget trauma for members of Congress. But they bite heavily in 1999, with consideration of the fy 2000 budget. The caps will require that domestic discretionary outlays be reduced by about $28 billion from the fy 1999 levels. Appropriations needed to maintain federal domestic discretionary spending as a constant share of personal income would need to account for this $28 billion reduction. Similar problems will arise as federal officials consider budgets for fy 2001 and subsequent years.

The conflicting perspectives on future federal spending for domestic discretionary programs create choices that will provide the basis for debates on the federal budget in 1999 and the years beyond. Those debates could result in federal spending in support of state and local spending (and direct federal support of higher education) that could vary substantially, up or down, from the growth assumed in this paper.

REMEDIES FOR STRUCTURAL DEFICITS

Running Unbalanced Budgets
The financial picture that this report paints for higher education presumes that state and local governments will continue to maintain balanced budgets, as they are required to do by many constitutional provisions and statutes. These laws and the traditions behind them withstood the intense fiscal challenges of the Great Depression (circa 1929 to 1938) and appear likely to withstand the lesser challenges predicted by the projections in this report.

Because of balanced budget requirements, state officials will not allow a structural deficit of the magnitude described in the projections to occur. Instead, in the states where these deficits occur, state officials will deal incrementally with the mismatch between revenues from current taxes and the spending needed to maintain current services -- that is, they will respond to projected deficits as they become apparent.

Tax Increases
Although the text of the report does not predict this, the discussion of fiscal pressures on higher education suggests that most of the adjustments will be made by reductions in spending below the level required to maintain current services.

In fact, many adjustments are likely to be made by increasing taxes. Increases are particularly likely in the taxes with the least responsiveness to economic growth -- particularly the per-unit taxes on motor fuels, tobacco, and alcoholic beverages. For various reasons, increases in these taxes are not as politically formidable as increases in general taxes. Increases are also more likely in states combining rapid growth and inelastic tax systems, with Texas providing a good example. Reflecting public attitudes toward growth-oriented spending, the willingness of elected officials in those states to raise taxes is substantially greater than is found on average throughout the nation.

For the past two decades, state and local taxes have taken an almost constant percentage of personal income. This means, on balance and over the business cycle, tax increases have predominated over tax cuts enough to overcome the less-than-unit elasticity of state and local tax systems. That pattern may well continue, thereby reducing somewhat the impact of structural deficits on spending in general and higher education spending in particular.

Spending Cuts
The text of the report and Appendix B suggest that all possibilities for providing increases at less-than-current service level requirements have major policy consequences and thus major political consequences. Particularly over the short term, this is not entirely true. In many but not all state and local activities, small structural deficits can be handled by spending constraints with no immediate policy consequences.

Current service budgets presume that major ratios of governmental inputs to governmental outputs will remain constant. Examples of such ratios are faculty members to students in higher education, teachers to students in public schools, caseworkers to clients in social services, police officers and firefighters to population in local governments, and guards to prisoners in corrections. Such ratios can be allowed to inflate somewhat by holding spending increases to slightly below the levels that would be required to maintain them. Of course, there are consequences in each field paralleling the consequences most readers would understand for such ratio changes in public higher education.

Current service budgets also presume that the relationship between the compensation of public sector employees and the compensation of private sector employees will remain constant. As a result, the projections presume the same increase in real (inflation-adjusted) compensation of all workers, which, by definition, is roughly equal to the annual improvements in productivity assumed in the economic projections.

Given the assumptions of this report, this means university faculty members, public school teachers, and other public workers would be seeing annual average increases in purchasing power of about 1.5% a year. In the long run, providing smaller increases in compensation for public sector workers creates no apparent problems. For example, annual pay increases matching inflation would be readily defensible for many elected officials and would cut 1.5% a year from the wages and salary component of state and local outlays. The wage and salary component of state and local spending is so large that an inflation-matching compensation salary policy would, by itself, eliminate (on a national average basis) the structural deficit predicted in this report.

Such a policy is untenable in the long run because the public sector must compete for workers, particularly at the entry level, with the private sector.

Spending Shifts Out of the Public Sector
Nearly all discussions of government spending and taxes presume the same accounting conventions used to prepare and present budgets and fiscal statistics. These exclude those extractions of funds from the public that are "off-budget" in the jurisdiction(s) under consideration, and they often exclude the spending financed by those revenues. Major examples are public university tuition and fees, charges by state toll road authorities, and charges levied for services of public institutions with functions that parallel those of the private sector, including providing hospital services, recreation, electric power, and natural gas.

The projections in this report presume that such charges will finance the same proportion of total spending in each function that they currently finance. For example, if a particular state's higher education institutions are defraying 33% of their spending through tuition at the beginning of the projection period, then they are presumed to defray 33% at the end as well. This means that the projections presume, for example, tuition increases per student equal to increases in spending per student.

In times of fiscal adversity, such as those forecast for state and local governments in the early years of the next decade, state and local officials often find many ways to shift spending from financing by tax collections to financing by user charges. One example is the tendency to increase tuition to offset the impacts of appropriation freezes on institutions of public higher education.

Achieving Public Purposes by Requiring Private Spending
Government budgets and financial reports are imperfect indicators of the impact of government on the freedom of citizens to control the uses of their resources. By mandating private sector spending, governments can achieve the same objectives as are achieved by public sector spending and taxes. For example, supporting the consumption of low-income individuals is currently accomplished by tax-financed spending for cash welfare payments and subsidized health care.

It is also accomplished by subsidized low rates for phone service. Those rates, often called "lifeline" service, do not fully compensate telecommuni-cations providers for their costs of services. The offsetting extra revenue is collected from all other customers. While the payment of extra costs by the other customers is a government-mandated reduction in consumer purchasing power, it does not count as a tax. While the subsidies to low-income households are government-directed spending, they do not count as spending.

There are constantly growing examples of such devices. Two that have drawn public attention recently are: (1) the extraction of impact outlays from housing and commercial property developers for such purposes as building public streets and roads and constructing public schools, and (2) the funding of a multi-billion dollar federal program to provide telecommunications access to schools through charges built into telephone rates.

MAJOR REALIGNMENTS IN THINKING ABOUT GOVERNMENT ACTIVITIES

The analyses in the text and appendices of this report presume that the attitudes of the public and of elected officials toward government spending and taxes and toward major public programs will remain relatively constant over the projection period. The projection methodology simply applies these constant attitudes to changed circumstances dictated by the economic, demographic, and fiscal environments predicted during the projection period. As summarized at the end of Chapter Two, "With attitudes and decision-making procedures unchanged, and with the impacts of strong and weak state fiscal circumstances on higher education spending well known, forecasting state approaches to higher education spending becomes simply a matter of forecasting the state fiscal circumstances likely to prevail in the next decade."

The conclusions in this report are thus sensitive to any major changes in attitudes. Several such changes are possible.

Role of Government
One possible change is a major revision in thinking about the role of government and thus its size, scope, and cost. One possibility is that more people will believe that government does too little, much improving the outlook for higher taxes. On the other hand, another possibility is that more people will believe that government does too much, creating an environment for reductions in taxes and spending.

Redefining Baseline Spending
There are areas in which changes in attitudes and resulting changes in policy could substantially redefine baseline spending. Accepting a public obligation to finance fundamental health care would be one such change, with enormous fiscal implications. So would redefinition of the scope of free public education to encompass the education of three and four-year olds and/or grades 13 and 14. Likewise, certain activities now viewed as part of baselines could be defined as outside the traditional responsibilities of government. This has been happening with one major outlay in federal budgets (housing) and one in state budgets (recreation). As these examples suggest, movement in either direction tends to be gradual, almost glacial. As a result, such changes are not built into the assumptions used in this report.

DOWNLOAD | PREVIOUS | NEXT

© 1999 The National Center for Public Policy and Higher Education