Federal budget reforms: why they aren't working and what we can do about it
V. V. Chari
Published July 1, 1991 | July 1991 issue
Why is the federal budget still such a mess? The Gramm-Rudman-Hollings reforms of last fall are the third attempt to change the budget process, and rumblings out of Washington already strongly suggest the fall reforms are not long for this world. Meanwhile, the budget deficit continues to grow (it is projected to be $300 billion for 1991), and so does the bill for our debt that we will pass on to our children. A gargantuan budget deficit is not our only problem. There is widespread agreement that we are not spending enough to maintain our infrastructure of roads, airports, sewer systems and so on. Federal capital spending has declined to roughly 2.2 percent in the late 1980s of our gross national product from 4.4 percent in the early 1960s. While the deficit grew, and capital spending fell, a variety of problems like the S&L crisis were not dealt with promptly and must now be handled at substantially greater cost.
There is a common theme to these problems, and reforms of the budget process will be futile until this theme is recognized and confronted. These problems are symptoms of an inherent, and persistent, tendency of our political system to encourage an excessively short-term view in policy making. This bias against the future is systemic, rather than a problem that arises because policymakers care less about the future than voters.
The bias arises because most policy decisions involve difficult tradeoffs between current benefits or costs and future costs or benefits. It is difficult enough for policymakers to make judgments about such decisions. It is even more difficult for voters who do not have all the information available to policy makers to judge whether the policy makers have made wise choices. If voters did have all the information, they would boot out policy makers who made unwise choices and retain those who made better decisions. But acquiring information about all the choices open to policy makers is costly. Each voter has an incentive to let other voters acquire the information needed to monitor elected officials. The result is that most voters end up with less information than policy makers. How then do they tell good policy makers from bad ones? Quite rationally each voter judges policy makers by the tangible outcomes of the chosen policies. Frequently, this means weighing current consequences of decisions, which are tangible, much more heavily than future costs or benefits which are much more difficult to assess accurately.
What is a rational elected official who is concerned with reelection to do when confronted with these incentives? It should not be surprising that policy makers tend to adopt decisions which yield tangible benefits and minimize current costs even if such decisions prove very costly in the future. The system is stuck in an equilibrium which neither voters nor most policy makers particularly like. These considerations are not purely theoretical. We could have handled the S&L crisis at far less cost in the mid-1980s. There is widespread agreement that if nuclear armaments plants had been maintained better at a relatively modest cost over the last 20 years, we would not now be facing substantially higher expenditures on such plants. The recent and ongoing controversy over the closing of defense bases illustrates the difficulty we all face in trading off certain current costs against uncertain future benefits.
It is tempting to throw up one's hands and declare the problem unsolvable, but fortunately we do not have to do that. The first step is to recognize that this bias against the future is not unique to governments. Publicly traded corporations face exactly the same problem. Shareholders do not have the incentives to acquire information and managers have incentives to shortchange the future. How do corporations handle this problem? They adopt accounting systems which record, albeit imperfectly, the future costs of current decisions and they adopt rules constraining the actions of managers. For example, the accounting systems require accrual accounting: that is, expenses are recorded when incurred, not when the payments are made. The federal government operates, by contrast, on a cash basis. Under the government's accounting system, the deficit falls in the current fiscal year when salaries are shifted from late September when the fiscal year ends to early October of the next fiscal year. Under Gramm-Rudman-Hollings the government actually resorted to this practice to meet deficit targets.
Corporations, as well as state and local governments, also routinely separate capital and operating budgets. This sensible practice spreads the cost of capital projects over the time that such projects yield services. Expenditures that yield only current benefits, however, appear in the operating budget. The federal government's cash budget obscures the essential difference between the two types of expenditures. When combined with the policy making bias against this future, it makes capital spending highly susceptible to the knife of budget cuts. Reductions in capital expenditures result in a loss of future capital services and, given the bias, seem more attractive than reductions in current services.
The arguments for changing the accounting system to incorporate accrual accounting and separating the capital and operating budgets are obvious and compelling.
But they are not enough. Because the system is so inherently biased against the future, it is also essential to adopt a rule requiring that the operating budget be balanced while allowing the capital budget to be debt financed. Allowing deficits on the operating budget encourages policy makers to use debt finance as a way of delivering benefits to current generations and imposing costs on future generations. Removing this option is essential if we are to move away from the equilibrium in which we are stuck. It is entirely appropriate that the beneficiaries of programs pay for them. From this perspective, capital projects should be paid for by their beneficiaries which include future generations, but current expenditures should be paid for by current generations. How do we enforce this rule? Exactly as under current practices, which involve sequesters if targets are not met.
One problem with these rules is that balancing the operating budget may require sharp changes in revenues or expenditures during wars or recessions. A simple remedy is available. The federal government should establish rainy day funds. These funds would be separate from the operating budget and would be built up in good times and drawn in bad times.
I can anticipate the reactions to these proposals. The government would move operating expenditures to the capital budget, the rainy day funds would never be built up, and how do we get from here to there. But my proposals should not be compared to some hypothetical, unachievable ideal. Rather, they should be compared with the current system and with competing proposals. Under the current system, effectively everything is on the capital budget since all expenditures can be financed with debt. To those who would argue that the rainy day funds would never be built up, I point to the state of Minnesota, which has done a creditable job of building up such funds in the 80s and resisting raids on them recently. How do we get from where we are now to my proposals? The reforms of the budget process last fall are a useful starting point. The problem with them was that they loaded all the pain of deficit reduction at the end of a five-year horizon. Any credible attempt to reach a balanced budget must be front-loaded to a substantially greater extent. We will know that we have started on that road when we hear screams of anguish from every lobbyist in Washington.
Will these modest, but focused, proposals work or do we need a draconian change like a constitutional amendment? I believe that matters are not yet so dire that our problems cannot be solved legislatively. What has been lacking is not the will to act but the way to solve problems. In good part this is because reform proposals have not been based on a clear understanding of the underlying problem. No set of reforms will solve all our problems but the budget process has gone wrong for too long. We do not have to give up or make wholesale changes to get it right.