Omnibus Budget Reform: Why? How?

Mr. Chairman and distinguished members of this committee: I very much appreciate the opportunity to appear today and contribute to deliberation about your effort to create a useful package of budget process reforms. Given the wide array of proposals that have been floating around Capitol Hill, bringing them together to create some order and hopefully some analysis is a fine idea.

I hope you appreciate the difficulty of analyzing materials that are not yet anything resembling a bill, that contradict each other in numerous ways, and that refer to a vast array of processes, ranging from congressional procedures to internal management of agencies. I think it is important that the participants today, reporters and the public recognize what kind of materials we are discussing. We have here a draft outline of a bill. It includes references to language that exists in a total of twenty (by my count) other bills that have been introduced in the House or Senate, drafted for either body, or passed with such language as amendments. It also refers to four different Office of Management and Budget Circulars, an array of materials produced by an organization called “Citizens for Budget Reform,” and in a number of cases says the necessary language still has to be drafted. I urge you to view this hearing as an opportunity to gather some general responses to the general trend of ideas here, but to plan on holding further hearings on a more coherently drafted and carefully pruned document once you have produced one.

At this stage, I believe it makes most sense to divide my comments into three parts. First, I will discuss how the materials here seem to define “what’s wrong” with the budget process. Second, I will point out some major contradictions among the measures proposed here. Last, I will describe what is right with current budgeting, what some of the proposals here ignore about how budgeting works, and therefore why I would advise against passing legislation that included such measures.

The Evils This Bill Proposes to Fix

I must emphasize here that I am looking for a logic in the texts that should not be associated, in all cases, with individual sponsors. Nevertheless, the following themes seem evident.

1) The budget must be balanced by cutting spending. The bill provides a wide variety of measures to balance the budget. The most direct is taken from H.R. 1131, and is in essence a new version of the Gramm-Rudman system of targets and sequesters that failed miserably in the 1980s. The difference between H.R. 1131 and Gramm-Rudman is that the former demands explicitly that the deficit be diminished solely by cutting spending. I believe the assumption that deficits should only be reduced by spending cuts also fits other proposals here, though it is not always so explicit.

Second, the proposals here assume that the budget is not balanced for a series of reasons. The most prominent are:

2A) The existence of entitlements. Entitlements are programs that promise benefits to persons or firms so long as they meet certain conditions, such as being unemployed after a period of work, or sick and age 65 or over. The purpose of entitlements is to protect people from certain risks. This bill proposes to eliminate all but the entitlements for Social Security and interest on the public debt. All other entitlement programs—Medicare, Medicaid, unemployment Insurance, Military pensions, civil service pensions, food stamps, and so on—would be turned into annually appropriated programs, at the mercy of a series of further procedures that are designed to discourage appropriations. In short, this bill says that in case of economic downturns or other bad luck, the risk to the budget is far more important than the consequences for individual citizens.

2B) “Baselines”. The bill would require that all formal comparisons within the budget process focus on nominal dollars, the difference between one year’s spending and the next year’s proposal. Baselines, in this context, means estimates of what it would cost to provide the same services as offered under current law next year as this. Because of inflation and population growth, the baseline is on average higher than the previous year’s spending. Clearly if the public knew that, it would be more likely to support higher spending; so the purpose of abolishing “baselines” is to focus attention on spending rather than what the spending buys.

2C) Too much attention to the details of budgeting. Budget processes exist to resolve a difference between preferences about a total (how much we spend) and the details (what we buy). Usually we want to buy more but spend less. Most systems go through a series of iterations: there are suggestions for spending, suggestions for a total, the details are cut to fit the total but not enough, the total may be raised slightly but not enough to include all that’s left, the details are cut again, and so on. The point here is that you don’t really know how much you want to spend until you know what you would have to give up; or what you want to buy until you know what it will add up to. But allowing the total to be affected by attention to the details is likely to lead to more spending than just setting the total arbitrarily. Therefore this bill tries to strengthen the current trend towards setting totals with as little attention to their programmatic consequences as possible.

That trend is already dominant in the appropriations process, for which all parties in Congress as well as the administration have proposed shrinking “caps” through 2002 and nobody has specified what programs would be cut to meet those targets. This bill follows that logic in the Gramm-Rudman-Redux proposals of HR1131, the caps in HR2599, and the transformation of Budget Resolutions into laws with binding caps by budget function each year before Congress considers specific legislation to meet those totals.

2D) “Pork”. This bill provides the President with an item veto so he can reduce spending to meet budget targets. But I should emphasize that at least it is limited to that purpose, so it is a far more moderate proposal than what Congress appears to be set to pass momentarily.

2E) Majority rule. Many provisions in these proposals assume that majorities cannot be trusted. It goes beyond the current requirements for 3/5 of the Senate to overturn a point of order against measures that would increase the deficit. Instead, some parts of the bill use extraordinary majorities of 3/5 in both chambers and some of 2/3 in both chambers. Sometimes these are applied not only against measures that would increase the deficit compared to current law but against legislation that fails to meet a target for deficit reduction.

Third, this bill seeks to respond to public and probably internal disgust with the failure to pass budget legislation on time. At the same time, it seeks to enforce budget controls by preventing action on some items before action on others. So, in general, it includes an array of actions about the budget schedule. The problems and responses include:

3A) Adopting budget-related legislation without previous agreement on an overall budget plan. This package includes a requirement that overall budget total laws be adopted each year before any other action.

3B) Too much time spent on budgeting. So this package includes a change from annual to biennial budget resolutions.

3C) Government shutdowns due to failure to enact appropriations. So this package includes procedures for automatic continuing resolutions at the previous year’s exact level of spending if appropriations have not been enacted. In the case of former entitlements that have been transformed into annually appropriated programs, this requires further provisions allowing agency heads or state governments to make all necessary cuts to account for the fact that normally one year’s spending will not provide the same services the next year.

3D) Government shutdowns due to failure to raise the debt ceiling. Actually the package is ambivalent about this. It provides a one-time debt-ceiling increase, but also includes measures to ensure that if the next increase is not passed, the Secretary of the Treasury will not be able to avoid a shutdown.

Fourth, this bill seeks to increase cooperation between the President and Congress, by making cooperation necessary early in the process. It defines the problem in two ways.

4A) Presidents making proposals that do not explain how to balance the budget. So it includes provisions to require presidents to make such proposals.

4B) Presidents not having to agree to Congressional budget targets early in the year. So the budget resolution would be converted into a law. Of course, that means if they cannot agree on a budget resolution they cannot do anything, since 2/3 majorities are required to do anything without a resolution, and 2/3 majorities could have overrided the president’s veto of a resolution.

The fifth and last category of presumptions is that the deficit could be reduced and government’s presumably negative effect on the economy limited by technical changes to government management to make it more efficient. The package therefore includes a huge hodgepodge of management reforms, ranging from separation of an Office of Management from an Office of Budget, to requiring many more forms of accounting, to drastic changes in regulatory procedures. We may group these into:

5A) The government runs on automatic pilot. So programs must sunset, there must be zero-based budgeting, and of course there should be no entitlements except the two that it would be suicide to eliminate, interest and Social Security.

5B) Citizens aren’t aware enough of the terrible deficit, so government must provide a wide array of documents to tell them how bad it is.

5C) The true costs of government are hidden. So government must do much more cost-benefit analysis, emphasize accrued long-term obligations, etc.

5D) The government does not pay enough attention in budgeting to performance. So this bill joins the recent round of efforts to cut spending by having “performance budgeting.”

Contradictions Within the Package

In general, the presumptions within this package are not contradictory. Elimination of entitlements, distrust of majorities, accounting changes, spending cuts, etc. all fit an ideology that says government is an inherently corrupt violation of the free market, to be limited in all ways.

Yet some legislators, and many members of the public, may not distrust government quite so much, while being attracted to parts of these proposals. And there remain a number of contradictions that will require choices by you in putting together the final package.

Some are rather subtle. If the public has more information, does it make sense to require 2/3 majorities? A relatively more informed and attentive public may be more likely to be incensed by being thwarted by supermajority requirements.

Others are much more direct. You can’t have both a binding annual budget resolution and a biennial resolution process. The deficit reduction targets with 3/5 requirements fit awkwardly with the new binding budget resolution system with 2/3 requirements. How does an automatic CR (to prevent gridlock) fit with measures to make the debt ceiling even more of a roadblock? There would be no contradiction if the budget were already balanced, but it is not. How does an automatic CR fit with the Armey provisions that forbid appropriations without reauthorizations, and require sunsetting of authorizations? Where does the HR822 outlay reduction commission fit with either of the three new budgeting schedules—the biennial process, the move of the fiscal year to match the calendar year, or the schedule provided with the binding resolution in HR2929? How can caps be based on the “most recently published current law baseline” (HR2599) if current law baselines have been abolished (HR2929)?

Finally, there is an overarching contradiction between the effort to constrain budget decisions to force deficit reduction, and the claimed interest in ensuring that government not be shut down by conflict. Budgeting involves fundamental conflicts over the size and purposes of government. The only way to ease decisions and make them more timely is to facilitate compromise—and even that may not be enough. Yet this bill greatly favors one side, requires supermajorities and creates a wide variety of opportunities for hostage-taking and brinksmanship. On balance, I have to expect that the goal of making the process more timely will be obliterated by the other measures in this package.

Major Difficulties With This Package

The most obvious objection to the non-bill before us is that it is so contradictory and incoherent. Once it were cleaned up, however, it would still represent some combination of the assumptions that I described in the first section, so I will direct my attention here to those basic assumptions.

First of all, the case for a balanced budget is far weaker than its proponents imagine, and the case for achieving it solely by spending cuts is even weaker. I do not expect most members of this committee to agree, of course. But I wish you would consider two points.

1A) Every responsible estimate of the effects of deficits on the economy finds that balancing the budget, though perhaps helpful, does not provide markedly positive returns. Thus in its update on the Economic and Budget Outlook in August of last year, CBO reported that the effect of enacting the policies in the Congressional Budget Resolution for FY 1996, which would balance the budget by 2002, would be to increase GDP by a tenth of a percent per year, and a total of half a percent by 2002. That was the payoff for the cuts in Medicare, Medicaid, discretionary spending and other programs that were proposed. It is very small—easily wiped out by all sorts of unpredictable economic fluctuations. Ultimately the real reason this Congress has had so much trouble convincing the American public to support those policies is that it does not have strong evidence that they are worth the cost.

1B) Current rhetoric emphasizes that total spending will increase because of “entitlements,” basically those that serve the elderly: Social Security, Medicare, and Medicaid. This is true. But the causes are crucial. When the population ages and a larger proportion is retired, one of three things MUST happen. Either the elderly will get less per capita. Or their share of the national wealth will increase due to government transfer programs. Or their share will increase due to earnings from investments, which means wages will be lower compared to the return to capital. The third option takes just as much from workers as the second. The difference is, the third is much more likely to favor people who had money to invest, and the second is much more secure for all citizens. We can argue about the proper balance of effects, but surely we should expect part of the result to be larger transfer payments to the larger number of elderly. We can argue about how much other programs could be cut from the current base, but at some point you hit bottom, and then the effects of an aging population must cause spending to increase. So policy that focuses only on spending either denies demography, or simply chooses that the people who rely most on Social Security and Medicare will lose most from the aging of the population. I doubt that is wise.

Second, focusing entirely on numbers and ignoring program effects is not good government. In particular:

2A) Most entitlements are entitlements for good reasons. Medicare that had to be funded by an annual vote each year would be a much worse guarantee of coverage. Imagine what would happen if the appropriations were delayed and the Secretary of HHS issued regulations that caused some person to die from being unable to get care! Remember also that some of the entitlements are specifically designed to respond immediately to economic downturns, as “automatic stabilizers.” Under the logic of this package, if unemployment increased, 2/3 majorities would be needed to pay extra benefits. That’s an awful idea.

2B) We can argue all we want about “baselines,” but the fact remains that how much it will cost to buy what we bought last year is a fundamental question that anybody, in any budget process, asks. Amid a raft of proposals to get agencies to direct their budgeting towards outputs, it is ludicrous to say that Congress should not ask how much it would cost to maintain the existing level of output.

Third, this bill is fundamentally antidemocratic, in two senses.

3A) The supermajorities in this bill bias policymaking dramatically, against what often will be majority will. They are justified, I suppose, by the assumption that the unbalanced budget is such a disaster that it must be prevented by any means. But the fact is, the deficit is no such disaster. If it were, the benefits of balance would be much greater. That does not mean there is no place for some systemic bias towards fiscal conservatism. Within a system that makes any action difficult, measures that provide further obstacles to policy changes that increase the deficit, as exist under current law, might be justified. But this package goes far beyond that point.

3B) The procedures in this bill are designed to inhibit rational debate on the budget. As I stated earlier, budgeting in any situation is a process by which preferences about details are made to confront preferences about totals, so that each can affect the other and differences be resolved. This bill would foreclose that process in many ways. At some points it sets into law arbitrary preferences about totals—with no attention, at this point, to what the effects would be. At other points it requires that, in any given year, totals be set before any serious debate about how to achieve them. At other points it encourages government shutdowns, or gives the advantage in shutdowns to people who want to cut entitlements. This is, at best, arrogant: based on a belief that if the public seems conflicted between balancing the budget and the cuts needed to achieve balance, the authors of this bill know how the public “really” wants to resolve the conflict. I think Democracy means making it possible for the public to decide for itself.

Also, some of the measures in this bill are simply unpractical. That is especially true of the government management proposals.

4A) Nobody has ever been able to make performance budgeting work, because the outputs of federal programs are exceedingly difficult to measure, and could not be compared to each other even if they were measured. Every decade or so somebody comes up with the same idea under a new name—Program Budgeting, PPBS, ZBB, MBO, “Reinvention”—and it’s always bogus.

4B) An Office of Federal Management will never do much good, for two reasons. First, if there are serious inefficiencies in government, they are almost always due to policy choices: whether to have a supercollider or space shuttle or third submarine; or to create a housing program for the poor in a way that involves private sector developers. And there are always plenty of people with (often plausible) arguments as to why these are not such bad choices. Second, no central agency can ever know more than a smidge about managing specific programs. Management requires detailed knowledge of the law, of policy environments, of practical difficulties, of human resources within an agency, and many other matters. It’s not a matter of the kind of simple rules that could be imposed by an office with one staffer for every $3 billion or $5billion or $10 billion of spending.

Last, and related especially to the concerns about democracy, some of the proposals come very close to abolishing the rule of law and substituting arbitrary allocation of federal funds. Most evidently,

5) The provisions for allocating funds for annually appropriated former-entitlements cede huge amounts of discretion. Some of it goes to the heads of federal agencies. The rest goes to state governments. But in either case, the votes of Congress and signature of the President will no longer determine how programs operate. Law about program benefits would no longer be law: it would be law-if-Congress- appropriates-enough-money. It is not at all clear from the available materials what would happen if the courts objected to these terms.


Some of the ideas within this text deserve further consideration. Many of the accounting presentations would be useful, so long as they do not replace the basic current documents and one does not expect revolutionary results. The item veto provisions are a far less extreme change in our basic institutions than the version that is expected to be enacted soon. There is always room for debate about whether specific programs should be entitlements, though I would probably disagree with the authors in most cases. Measures to encourage the President to submit more thorough budget proposals are a reasonable idea, though he should also be expected to say when and why he objects to a total mandated in advance. There is a case to be made for a fallback Continuing Resolution position, legislated in advance—especially if it applies only to the current discretionary programs.

Overall, however, much of the basic ideas in this package are bad budgeting. Ultimately they are united by two things. One is a notion of the merits of a balanced budget that is unsupported by the evidence. The other is a refusal to acknowledge that the public’s moderation on the issue could be appropriate, and deserves respect from politicians.

The public wants a balanced budget. It also wants a set of programs and level of taxes that would not create a balanced budget. This is not a flaw of democracy. It is normal. Contradictory desires are the reason for politics and government. We want peace in Bosnia but not to be the world’s policeman. We want to get rid of Castro but not to invade. We want to guarantee health care to everybody but not to pay the price. We want a clean environment but a decent business climate. So we compromise.

In the battle of the budget we could compromise on smaller deficits, rather than either the level that would result from current law or a truly balanced budget. This Congress has an opportunity to enact such a compromise. President Clinton’s proposals are not what the majority of Congress wants, but they are a far cry from his proposals a year ago, and would substantially reduce the deficit. If it really cares about both the deficit and democracy, this Congress will compromise on the substance, instead of messing around with the process.