Murray Weidenbaum

Murray Weidenbaum is the Mallinckrodt Distinguished University Professor at Washington University in St. Louis, where he also serves as honorary chairman of the Weidenbaum Center on the Economy, Government, and Public Policy. He served as chairman of the Council of Economic Advisers under President Ronald Reagan from 1981-82.

The rapid growth of government spending and deficit financing underscores the importance of carefully analyzing the huge array of numbers contained in budget documents. That onslaught of statistical information also reinforces the need to recognize the many budget games regularly played with the data.

To start off on a positive note, the situation has improved since the time when policymakers could choose among three competing concepts of the budget. Back then, they could select at different points in time the budget concept that generated the best statistical support for the policy positions they were advocating.

That situation prevailed in the administration of Lyndon Johnson. That time was a high point (or rather low point) in budget manipulation. Of course, no president since has been blameless. That point can be quickly supported by referring to the following 10 examples of budget games. I will illustrate the seriousness of these games and try to show how to guard against them.

Game 1. Changing the way the overall federal budget is measured, especially what categories of spending are included and excluded.

Game 2. Arbitrarily estimating the allocation of future government spending to any specific year.

Game 3. Taking advantage of spending categories that are effectively uncontrollable in the normal budget process.

Game 4. Omitting expenditures that can easily be added later on.

Game 5. Fudging the economic assumptions underlying the budget estimates.

Game 6. Using clever policy assumptions.

Game 7. Downplaying congressional tilting of budget assumptions.

Game 8. Not stating all the key assumptions.

Game 9. Relying on very technical ways to slow down or speed up revenues and spending.

Game 10. Setting up arbitrary budget categories.

Game 1. Changing the Way We Measure the Overall Budget

In an article published back in the Johnson presidency, I described a new presidential budget with the title, "Federal Spending -- Up, Down, or Sideways?" Back then, three different measures of the budget were used: (1) the administrative budget, which ignored trust funds such as social security, (2) the consolidated cash budget, which incorporated the trust funds, and (3) the NIA budget which put the budget on an economic basis by using the relevant components of the national income and product accounts (NIA or GDP).

Depending on which of those three budget concepts government officials used back then, they could, with technical accuracy, state that federal expenditures for the year ahead were rising, declining, or stable. That choice is no longer available. Since then, the "unified budget" has been adopted. It uses a variation of the consolidated-cash approach, but with important omissions (such as some of the bailout programs).

Game 2. Arbitrarily Allocating Spending to a Specific Year

A different budget game is still played: take advantage of the fact that calculating the surplus or rather deficit for the forthcoming budget year draws on an estimate of future spending (that is, expenditures). However, Congress does not act on expenditures. It acts on appropriations, which are the legal authorizations for federal agencies to spend money.

The game consists of overestimating expenditures in the current year and underestimating expenditures in the year ahead. This is done because the attention of policy makers and those who report on the new annual budget is on the year ahead, which is the budget year. Nobody pays attention to the current year.

Thus, the Office of Management and Budget (OMB) can show that the deficit is coming down, based on a phony, understated forecast of next year's expenditures. Of course, when the current year is over, it will turn out that the government did not spend money as fast as it said it would. That pushes up the spending estimate for the next year. But by then, that next year is the new current year and, once again, attention has shifted to the next budget year, and the game can be played again.

It is a game, because each appropriation can be spent over a period of several years, depending on the specific nature of congressional action. The extreme case is military procurement and construction. In budget lingo, these large categories are financed by "no-year money." The contrast with ordinary appropriations is very substantial. The standard appropriation must be "obligated"-in effect, committed by contracts-within a 12-month period; actual expenditures may occur over a 36-month period. In the military case, however, the appropriation is available for obligation for an indefinite period. The imagination can provide numerous possibilities for playing budget games under that situation.

Game 3. Taking Advantage of Uncontrollable Spending

For example, spending on veterans' pensions is subject to annual appropriations. However, when Congress passes an inadequate appropriation because it took OMB's low spending estimate seriously, there's no real penalty for the underestimation that incurred. You can count on Congress later in the year graciously passing a supplemental appropriation when the Department of Veterans Affairs runs out of appropriated funds.

A variation of this game is for OMB to play with the estimates of the various "permanent indefinite" appropriations. That perennially is a fertile field for fiscal manipulation. Those items do not appear in the formal appropriation bills at all. A few key examples help to get the point across: Social Security, Medicare, and interest on the public debt. The actual amount spent is determined by the detailed provisions of permanent laws.

The spending department has great leeway in submitting low-ball appropriation and expenditure estimates for these permanent, indefinite accounts. In each case, the actual appropriation for the year is automatically calculated when the money is actually spent.

Game 4. Omitting Categories of Spending Which Can Be Added Later On

Yet another budget game is played by the Pentagon, especially when a war is on (such as in Iraq or Afghanistan). The federal budget can assume that hostilities will end very early. When that does not happen, OMB can send up a supplemental defense appropriation request - but that action will occur after the public discussion of the new federal budget is over.

An interesting side effect is the opportunity for the administration or the Congress or both to add low priority civilian pork to the bill. How can Congress withhold approval - or the president veto - an appropriation to pay for the military forces currently engaged in combat?

Game 5. Fudging the Economic Assumptions

A special word of warning: If any budget issue sounds very technical or boring, pay special attention to it. That can be camouflage for something really important. For example, a vital part of any detailed analysis with lots of numbers - such as preparing a budget - is selecting the underlying assumptions.

Thus, the forecast of corporate profits is a key determinant of the amount of corporate tax collections shown in the budget. Treasury officials can come up with rosy estimates of the flow of corporate tax receipts just by using very optimistic assumptions of the future profitability of American companies.

Many other categories of revenues and expenditures are sensitive to economic conditions. The standard example is unemployment compensation. Less obvious but often far more important is the swing in personal tax collections. For example, anticipating a stock market boom will raise the estimate of capital gains, an important element of personal income and of personal income tax collections. Again, there is no penalty for making a bad guess, even if the overestimate is clearly intentional.

Game 6. Using Clever Policy Assumptions

These assumptions may be as important as the economic assumptions. Take the "patch" on the alternative minimum tax (the widely reviled AMT) that Congress has voted in recent years. The "patch" is a rough attempt to offset the effect of inflation on the amount of income subject to the AMT, which is in effect a second income tax. If the Treasury makes a positive assumption about the enactment of the patch, the federal revenue stream will be much lower than if they assume no passage of a "patch."

Numerous other legislative actions influence government revenues or expenditures. A few current examples help to make the point: Will the Bush tax cuts be extended again? Which ones?

Game 7. Downplaying Congressional Tilting of Budget Assumptions

A fundamental difference between the OMB numbers and the estimates of the Congressional Budget Office (CBO) relates to the key assumptions underlying them. Of course, each uses somewhat different economic assumptions, but usually these are not the major reason for their differences. The key CBO assumption-mandated by Congress-is to estimate the budget trends for future years on the basis of existing law.

That sounds very sensible - until you stop to think about it. That single inflexible assumption means that no expiring legislation will be extended and no new legislation will be enacted during the period being forecasted. Nor will any existing legislative authority be repealed or even modified. I cannot recall any fiscal year in modern times when the actual budget outcome met this rigid assumption. Certainly no presidential budget is ever prepared on that basis.

To minimize boredom, I will not go through the details of the differences between OMB and CBO budget numbers, on the one hand, and those of the Government Accountability Office (GAO) on the other. Let us just note that the variations in assumptions produce quite different long-range spending and deficit estimates.

Game 8. Not Stating All the Key Assumptions

More about the importance of assumptions. Sometimes the most crucial assumptions are unstated. For example, the congressional explanation of the new health reform law skips over the fact that the cost estimates shown cover only items included in the budget. The many costs that the law imposes directly on health care providers and insurers are carried at zero.

Thus, in the back of the published analysis of the new law, page after page shows zero costs for many items voted by Congress. It takes a close reading by people versed in arcane federal lingo to even notice that all those zero costs result from the fact that the title of the table indicates that it only covers direct expenditures. Nothing in the accompanying text alerts the reader to the significance of this technical point. In the normal course of events, numerous so-called indirect costs will be borne by the users of health care services - those voters are also called patients.

Game 9. Relying on Very Technical Ways to Speed Up or Slow Down Spending and Revenues

Another budget game played by the fiscal authorities is taking place right now. On occasion, Congress is reluctant to increase the statutory limit on the size of the federal debt - even though it has been very generous in voting the outlays and tax cuts which generate the need for additional deficit financing. One Washington cartoonist described the situation in terms of a large dog being cajoled into moving into a small doghouse while his master simultaneously is bringing him a generous supply of additional food.

Under these circumstances, Treasury can and has drawn down its cash balance in order to finance temporarily some especially urgent outlays. The government can also delay investing the cash it receives from the Social Security and Medicare payroll taxes. That increases the cash available for direct expenditures. At the margin, the Treasury can and has slowed down the issuance of checks to government contractors and to beneficiaries of other spending programs.

In a limited way, Treasury also can speed up the process of cashing the checks sent in by taxpayers. Paying clerks overtime to reduce the normal backlog of unprocessed checks is not economically efficient. However, in the short run, such action increases cash flow, enabling Congress to continue debating whether to increase the debt limit.

Game 10. Setting Up Arbitrary Budget Categories

Real gamesmanship is involved in developing arbitrary categories of budgetary presentation. On occasion, analyses of federal finance focus on nondefense spending or omit (or segregate) capital outlays. One extreme exercise of this sort once led to the gripe, back in my Budget Bureau days, that if you exclude from your analysis enough of the spending programs that are going up, the overall budget will look like it is going down.

A Suggestion

What can we do about these budget games? Here's a suggestion. For several years, the Brookings Institution and the American Enterprise Institute (AEI) published competing analyses of each new federal budget. Brookings started the activity, drawing on its substantial array of full-time scholars. Often that included Charlie Schultze, a fine economist and former budget director. For a while, I did a similar but less comprehensive analysis for AEI. Both studies were widely used. In any event, after a few years, I went on to other matters. Subsequently, so did Brookings.

I still believe that such independent studies of each new federal budget are useful. They could serve as a deterrent to the budget games so often played - or at least alert the public to their impact. The national interest is not when government officials devote time and effort to playing "spin the budget." *