Can we fix the budget deficit by cutting “other” spending?
This rather long post points out that recent budget-deficit-reduction plans tend to assume large amounts of savings from “other” spending – spending that does not include the most prominent and problematic programs of government. It presents numbers to provide a sense of the scale of this phenomenon, and then explains why simply assuming large savings from this nondescript “other” segment of the budget could prove quite dangerous.
The federal budget has clear trouble spots. In the long term, the projected growth of healthcare spending, primarily for Medicare and Medicaid, will far outstrip the growth of the GDP, and is therefore unsustainable. Revenues have been well south of their historical norm for several years. In fact, for the last three years, revenues have been more than 4 percent of GDP below the lowest level of any year in the last half century in which the budget has been balanced. (There were, admittedly, only five such years.) A deficit equal to that revenue shortfall, 4 percent of the GDP, is by itself enough to grow the public debt faster than the economy – which again is not sustainable. The Social Security system has not paid for itself since 2010, and it is not projected to do so ever again without significant changes of policy. The cost of the Middle East wars is a matter of controversy, but it has added at least $1 trillion, and possibly much more, to the public debt.
With all of these well defined and widely known problems, it might come as a surprise that the leading budget proposals from policymakers in Washington – President Obama’s budget, and House Budget Committee Chairman Paul Ryan’s budget resolution – achieve very large shares of their budget savings not in these problem areas, but rather from “other” – that is, not revenues, not Medicare, not Medicaid, not Social Security, and not defense (and of course not interest on the debt, which cannot be cut directly but rather only through savings elsewhere in the budget.) There is the old joke about the drunk who looks for his lost keys under the lamp post, not because he lost them there but because that is where there is light. But why should our nation seek a solution so far away from the problem?
There are some mitigating circumstances. For one thing, with the debt already too high and rising far too rapidly, we are managing for cash. Budget savings are where you find them. And in truth, although health care indisputably is the long-term problem, it equally clearly is not the near-term solution. No one has proposed changing the terms of Medicare for people currently on the rolls, and Chairman Ryan goes further and explicitly “grandfathers” (pardon the pun) all future Medicare enrollees who now are age 55 and over. Any enacted plan almost certainly would do likewise. And for that matter, fundamental change in the practice of medicine to reduce costs significantly will take years (a conversation for another time). So by all means, all policymakers should look for legitimate savings throughout the budget.
But arguably, that is not what is going on here. Rather, the nondescript “other” category of federal spending is by all indications being used as though it were a bottomless pit, from which essentially limitless amounts of savings can be extracted without consequences. And for every unspecified dollar of savings from “other,” policymakers search for one fewer dollar of the large amount of required savings from the more-visible parts of the budget – or even indulge in one dollar of truly unjustified and unaffordable additional tax cuts.
First, the numbers: The Congressional Budget Office (CBO) estimated spending under President Obama’s budget. Chairman Ryan presented estimates of his own resolution. At Chairman Ryan’s request and direction, CBO also extrapolated the budget resolution’s numbers into the more-distant future, using the Chairman’s own assumptions.
On that basis, Chairman Ryan’s budget resolution would cut “other” in nominal dollars by more than 9 percent in one year – from 2012 to 2013 – growing to more than 26 percent (relative to 2012) by 2017. Thereafter, spending would increase, reducing the nominal cut to almost 17 percent by 2022. But taking account of inflation, using the assumptions in the resolution, “other” spending would be cut by almost 11 percent in just one year (2013), and then by about one-third (again, relative to 2012) in 2018, and more than 30 percent for the rest of the ten-year budget window.
Another yardstick would be the spending on “other” as a percentage of the GDP. In most recent years, “other” spending has been measured somewhere in the 5 or 6 percent-of-GDP brackets. It went higher in economic downturns – for example, to 9.0 percent of GDP in 1976. It spiked to 8.9 percent of GDP in 2009 because of the financial crisis and the stimulus bill, but by last year, fiscal year 2011, it had subsided to 7.4 percent of GDP. Under the budget “baseline,” which accounts for the spending caps in the current law (notably including the impending “sequester”) but otherwise assumes that government costs increase in step with those in the private sector, it would fall to about 4.8 percent of GDP by 2022. Clearly, “other” spending is not the major source of our expected budgetary problems.
The Ryan budget would cut “other” spending even further than the baseline – to 3.7 percent of GDP by 2022. The previous low point in CBO’s historical data (which go back to 1972) was 1997, at 5.1 percent of GDP. So the Ryan budget would go below the previous historical low by more than one-fourth.
By CBO’s estimate, President Obama’s budget would reduce “other” spending in nominal dollars by 2.9 percent in 2013, and 5.0 percent in 2014 (relative to 2012); but spending would then increase such that by 2022 outlays would be 5.2 percent greater than in 2012. However, those increases would fall well short of the rate of inflation, and so by 2022 spending would be 11.5 percent lower than in 2012. As a percentage of GDP, “other” spending would fall from 2011’s 7.4 percent of GDP to 4.9 percent – slightly lower than the record low of 5.1 percent in 1997, but higher than the Ryan budget by about one-third.
Are such reductions feasible? And if they were imposed, what would be the consequences?
Perhaps the first question implied by those is, “What is ‘other’ spending?” We know what it is not. The parts that are excluded, Social Security, Medicare and so forth, are all of the largest, most-analyzed, and perhaps best-understood pieces of the federal government. By definition, the pieces that remain are smaller, and seem almost anonymous – until you call the roll, at which point a political smokescreen rises.
As of 2011, about 58 percent of “other” is the money that is appropriated each year for domestic and non-defense international activities. The rest is spending on “other” entitlements – that is, other than Social Security, Medicare and Medicaid. For this post, I will focus on the larger part of “other,” the annual appropriations. A following post will discuss the “other entitlement” programs, which are the subject of legislation recently passed by the House Budget Committee.
Everyone is opposed to government spending in the abstract. The question is what government spending to cut in reality. For example, there are plenty of attacks on “earmarks,” and “pork.” But then there is the question of precisely what it is that constitutes an “earmark.” One leading earmark opponent added to her opposition a “redefinition” of the term, because “Advocating for transportation projects for one’s district in my mind does not equate to an earmark…I don’t believe that building roads and bridges and interchanges should be considered an earmark.” The definition of earmarks is not a trivial concern. The Predator drone military aircraft was originally funded by an earmark; the Pentagon did not request it, and said that it did not want it. By the most common definition, all aid to Israel is an earmark. But much of the politics of budgeting does not rely on such fine points. Many vocal opponents of earmarks in theory continue to pursue targeted funding of projects in their districts in practice. When I worked in the White House Office of Management and Budget, we once systematically collected all of the letters requesting that particular projects be included in the President’s budget – and sent by vocal opponents of “earmarks.” There were many such letters. (We chose not to make those letters public to maintain comity in our relationships with the Congress.)
Upon hearing about non-defense international spending, the instant reaction is to cut “foreign aid.” That is 0.6 percent of the overall budget, and 1.6 percent of non-defense annually appropriated spending. The next cry is the inevitable proposal to cut “congressional salaries.” (There is a bill to withhold the salaries of Members of Congress if the Congress does not pass a budget.) The entire budget of the Congress – including the Capitol Police, the Library of Congress, the arboretum, and everything else – is 0.1 percent of the budget, and 0.4 percent of annual non-defense appropriations. Zeroing out both foreign aid and the entire Congress in 2011 would have reduced the $1.300 trillion budget deficit by $0.026 trillion (rounding up), or about 2.0 percent.
What are some of the more substantial pieces of the annual appropriations? How about health care for veterans, which costs $56.7 billion? The administration of Social Security and Medicare (which is not counted in the cost of the programs themselves), at $11.6 billion? Highway spending, at $60.2 billion (including mass transit; financed largely with the proceeds of the federal gasoline tax)? Maintenance of the national parks, costing $4.2 billion? The administration of justice – the courts, the FBI, etc. – for $54.3 billion? How much further should we go? Higher education student loans? Food safety inspections? Air traffic control? And can we ignore homeland security?
Here is where the smokescreen rises: Ask the advocate of cuts in “other” spending how much he or she will cut health care for veterans. The answer invariably is, “Oh no, I would never cut that! There is plenty of money in my reduced budget to provide full health care for veterans! I will cut something else!” But the math is unforgiving. Commit to cut all domestic and non-defense international appropriations by one-third – the deepest inflation-adjusted cut under the Ryan budget resolution. Then rule out cuts in veterans health care, and the cuts in everything else increase to 37 percent. Rule out highways (and it is far from clear that the federal government could keep collecting the gasoline tax if none of the proceeds were spent on highways), and the cuts on everything else increase to 41 percent. Rope off Social Security and Medicare administration, and the cuts elsewhere increase to 42 percent; the parks, the courts and the FBI, and the necessary cuts increase to 47 percent. In sum, there is enough money in a reduced budget to finance any one priority, but that cannot end the discussion. There may well not be enough to finance all of the priorities that most Americans accept, but do not think about when they hear the magic words, “federal government.”
This mental exercise recalls a presentation that former Representative Lee Hamilton (D-IN) once made to a group of his constituents. Upon being asked what the Congress was going to do about the serious budget problem of the late 1980s (in retrospect, far less serious than today’s), Rep. Hamilton responded that “Reducing the federal budget deficit is the second-highest priority of every Member of Congress.” He then waited for the joke to sink in, which it did: After every Member of Congress has taken care of his or her highest priority, there isn’t a prayer of success on the second (the deficit). So, for example, speak to a vocal opponent of government spending from a district with a relatively young population, and that Member of Congress might advocate cuts to “meals on wheels” for the elderly, but defend spending on higher education student loans and grants. Talk to another spending-opponent Member who represents a retirement haven, and he or she might decry education spending and advocate more funding for meals on wheels. In theory, those two Members of Congress are allies. They might even use the same very general talking points. But when it comes time actually to cut spending, their prescriptions are incompatible.
Planning for savings in “other” government is understandable, and in fact necessary. It is made up of an enormous number of relatively small programs. Few people know those parts of government in detail, or anything like it. It has become customary to assemble budget plans with assumed reductions in future annual appropriations – indeed, since the advent of tight caps on annual appropriations more than 20 years ago, most budgets themselves essentially assume that unspecified future savings can and will be found, without documentation or even much analysis. The Congress appropriates one year at a time, after all, so budget proposals are made on the same basis. However, assumptions of savings in “other” government should be made cautiously. (And it may be telling that the last House Republicans to sign on to Chairman Ryan’s budget resolution apparently were the members of the Appropriations Committee, and that much of the adverse reaction to President Obama’s budget from among the Democrats was about appropriations issues.)
Are there opportunities for savings? Sure. But don’t bet the fiscal stability of the United States, in a time of enormous financial stress, that those opportunities will be large. For perspective, where might those opportunities come from, and how substantial are they likely to be?
One generic argument is that a budget baseline for annual appropriations that assumes constant growth in costs at economy-wide rates does not account for productivity growth. If productivity increases, this argument goes, then government costs should decline relative to average prices. This argument probably holds for some parts of the government, but not for others. And it misses some important points.
Consider the administration of Social Security and Medicare. Most of the personnel in that system deal with the enrollment and everyday problems of beneficiaries. You can get those federal employees faster computers, and you can improve office organization. But a fundamental bottleneck in that system always will be the rate at which the elderly people who walk into the offices speak and comprehend.
This issue is reminiscent of “the Baumol problem,” identified by economist William Baumol some years ago. He noted that it is in the nature of government that it is left with responsibility for activities that are least readily amenable to productivity increases – activities that involve hand-to-hand interaction, like aiding Social Security beneficiaries or providing health care to veterans. His quintessential example was funding for the performing arts, where he noted that it would be counterproductive to impose upon a symphony orchestra a target to play Beethoven’s Fifth Symphony 3.2 percent faster each year. That is an extreme instance of the problem, and government should seek greater efficiency in Social Security (and symphony) administration, healthcare delivery, and everything else it does. But ignoring the barriers to productivity improvement in government can only limit success in reducing the deficit, by assuming at least some unattainable savings and therefore postponing achieving real ones.
And there is another fundamental issue. Because the Social Security caseloads will grow faster than the overall population over the next quarter-century or so, the needs of these systems will increase accordingly. Merely keeping funding for that part of government (and every other part that must operate in scale with the population) in step with costs, barring productivity increases, would not allow it to keep up with demands for services that are growing in absolute size.
Or consider the transportation-related systems, like highways and air traffic control. Over the long term we would expect traffic not to be constant, as the budget baseline implies, but rather to grow somewhat faster, like the growth rate of the population or the trend rate of growth of the GDP. (Working the other way would be efficiencies that would reduce the need to travel to accomplish work-related tasks.) And because the burden on those systems is caused as much by congestion as by the amount of the load per se, system needs could easily grow even faster. The difficulty of moving growing volumes of traffic within the same constrained geographic area could actually grow faster than the traffic volume itself. Thus, for some activities of government, GDP growth, like population growth, could require a higher level of government activity, which in turn could increase costs.
The future of demands on government could be argued back and forth for days, weeks, even fiscal years. But in making decisions that will have lasting or even irreversible implications – largely in the form of accumulation of debt – we should not assume that substantial as yet unidentified savings will be forthcoming. This is especially true if, on the basis of those assumed savings, equally irreversible commitments will be made toward tax cuts or increases in other spending (such as defense).
Large reductions to “other” spending are the budget’s outlay side’s analog to the old tax ditty repeated often by the late Senator Russell Long (D-LA), for many years chair of that body’s Finance Committee: “Don’t tax you, don’t tax me, tax the fellow behind the tree.” “Other” spending is like the anonymous, unseen “fellow behind the tree.” The fond hope was and is that we can make substantial reductions to the federal government’s budget deficit in ways that no one will notice. We can’t – especially now when those reductions, reasonably measured, must be the largest in the history of the Republic.