The Deficit Dilemma

In February 1983, the Congressional Budget Office published its baseline budget projections for the following five years. The bad news looked like this:

CBO was projecting $200 billion deficits as far as the eye could see. "To a great extent," that office explained, the 1983 and 1984

deficits are attributable to the economic recession, which has reduced federal revenues and increased federal outlays for unemployment compensation and other income maintenance programs. But even as these cyclical crises wither as economic recovery proceeds, the proposed deficits remain at the high level of 5.6 percent of GNP throughout the 1984–88 period. This indicates a long-term mismatch between federal spending and taxing.^[2]

CBO projected a deficit in 1988 that would consume 3.6 percent more of GNP than had 1981's deficit, although unemployment in those years would be similar.

This "long-term mismatch" was what President Reagan had called the "structural deficit." This 3.6 percent difference consisted of a rise in outlays of one percent of GNP and a drop in revenues by 2.6 percent. By this standard, then, the major source of the deficit problem was reduced revenue, though there also was a lack of spending restraint. The long-term mismatch could be ascribed, by Democrats, to short-term policy errors—mostly the ERTA tax cut.

Any analysis was flawed, however, if it assumed that 1981's 20.9 percent of GNP was a reasonable base level for federal taxes. The Democrats

lost the 1980 election in part because they had let taxes rise to their highest level since 1944. In 1978, revenues had been 19.1 percent of GNP, but bracket creep, payroll tax increases, the windfall-profits tax on oil, and the effects (from 1977 to 1979) of a progressive income tax on rising real incomes had increased the tax burden. Because the corporate tax had been diminished by the 1978 depreciation amendments, most extra money came out of individuals' pockets.

To be defensible, the 1981 tax level had to be argued for, and the public had to be convinced. Neither happened. There is little reason to believe that the American public saw these higher taxes as a necessary price for increasing government services. Rather, this tax increase was an aberration during a twenty-year period in which federal taxes had remained remarkably steady. Not only the overall tax level but, specifically, individual income taxes had increased from around 8.5 percent of GNP (or less) before the Carter administration to 9.9 percent in 1981.^[3] Gregory B. Mills and John L. Palmer summarize the trend in federal government tax burdens (as a percentage of GNP), comparing averages over five-year periods to the 1981 level:

The brunt of these increases fell on lower-income taxpayers. Democratic efforts in 1981 to direct reductions to the lower brackets and to cut somewhat less than the president proposed reflected the fact that the 1981 tax rate was higher than most people or politicians, with the conspicuous exception of Jimmy Carter, desired.^[4]

If a normal level of revenues is 19 percent of GNP, not 20.9 percent, then the long-term budget problem was not simply an artifact of the Reagan tax cuts. We must, therefore, turn our attention to the outlay side, where the president kept pointing his finger. Table 9 highlights five crucial relationships.

First, in 1979—a pretty good year for the economy—outlays at 20.83 percent of GNP were 1.2 percent of GNP above actual revenues and 1.8 percent of GNP above the (more or less normal) revenue level. These are the figures from which President Carter, in his FY80 budget, began his drive for spending restraint.

Second, spending growth from 1979 to 1985 can be explained entirely by increases in defense, social security, medicare, and debt interest. The real growth from FY79 to FY83 in income security is entirely a product

of the bad economy; it disguises the significant reductions made in domestic discretionary programs and in means-tested entitlements during this period.^[5] This is only one of many instances in which hard political work was, if not exactly undermined, then at least reduced in value by events public officials could not control.

Third, from 1979 to 1983 social security rose from 4.3 percent to 5.2 percent of GNP. This increase stemmed from the recession's effect on GNP and from the Carter-era stagflation. The recession slowed production, but it slowed neither the growth in program recipients nor the delayed effects of their COLAs. Stagflation caused benefits, linked to prices, to grow faster than either the economy or fund contributions, which increased more slowly with wages. These combined effects threw the social security system into crisis. They also caused an increase in the social security share of GNP during a time when economic growth had been expected to exceed demographically driven program increases. Thus, in the mid-1970s stabilization of the social security share had been reversed by unexpected economic bad news. Medicare, meanwhile, grew mainly because of the tremendous cost inflation in the medical business, a trend with which Congress had been wrestling for years.^[6] Given underlying demographic trends (more old people), only an optimist could predict even a stabilization of this medicare increase.

Fourth, net interest costs increased drastically because the government had to borrow more as deficits increased; then borrowing and interest began to feed on themselves, generating higher deficits, more borrowing, and, it was argued, higher interest rates. The economy's troubles—inflation and then recession, creating first an insistence by lenders for higher rates and then a great need to borrow—were the major cause of the government's, if no one else's, interest rate problems. Because fiscal policy was rather tight from 1980 until mid-1982, it is fairer to say that the economy caused this part of government spending than that spending caused the economy's troubles.

Fifth, the defense buildup began under the Democrats; indeed, the 1981 outlays are almost entirely a product of 1980 (mostly Senate) action; not until 1983 did the Reagan budgets significantly increase outlays.^[7] Some increase certainly would have occurred without Ronald Reagan. Still, a substantial part of the long-term increase in the "structural deficit" was due to administration plans to raise defense spending to 7.7 percent of GNP by 1988.

These trends in the two biggest entitlements, in interest costs, and defense suggest that, given the behavior of the economy, any president and any Congress would have faced a serious deficit problem by 1983. They suggest also that only part of the trouble was avoidable. If, like many observers in 1983, we decide that a deficit of 1 to 2 percent of

GNP is actually quite livable—in short, that Carter's original FY81 budget was just fine—we will still have trouble explaining how to get there. Table 1 shows deficits of 5.6 percent of GNP. Where can we get 4 percent of GNP in budget savings? How can we keep taxes at 20.9 percent? That would be a big chunk, but then there might not be a recovery in 1983 or a president in office to support it. Have a slower defense buildup? It's worth a try, but, given first Afghanistan and then the backlog of budget authority from FY81–83 action, it would be very hard to get much below the 1983 figure—only saving half a point of GNP. Reduce interest costs? It's too late.

Cut the other stuff? This was done in 1981 for FY82 and then slightly by attrition (increases slightly below inflation, for example) in subsequent years. Those reductions actually began under Carter. But, as David Stockman explained at the beginning of 1984,

The Great Society programs … peaked out at a cost of $140 billion in 1984 dollars in 1979–80. That cost will be down to $110 billion this year, no matter what Congress does. We have not succeeded in challenging the premise of the Great Society, but the level has been adjusted downward by about 20 percent…. Aside from defense, interest, and social security, we have shrunk the government by 15 percent.^[8]

Stockman was willing to junk the whole kit and kaboodle of Great Society programs, but he did not have the votes.

One could cut medicare by changing the guarantees of treatment or putting the doctors on salary; cut defense by retrenching America's commitments (why should the United States defend Japanese and German oil supplies in the Persian Gulf or support the defense of rich European nations?); cut interest with legislation limiting rates to, say, 6 percent (the states had long had such usury laws). But we are talking here about drastic changes in national policy, none of which on their own would be enough to solve the deficit problem. Even if you tossed out $110 billion of Great Society programs—food stamps, the Economic Development Administration, medicaid, Amtrak—you would have roughly a $100 billion deficit in 1985. That is experiencing considerable trauma to reach a still enormous deficit.

The mismatch between taxing and spending stemmed, in the main, from much earlier decisions about social security, medicare, and other pension programs, combined with the economy's stagnation after 1973. Table 10 provides data on outlays by major categories of spending. From 1962 to 1979, nondefense spending increased its share of GNP from 9.9 to 15.9 percent, yet total spending rose only from 19.5 to 20.8 percent. Domestic growth was financed not by taxes or deficits but by a lower defense share. Almost all the government's share from economic growth

went into domestic programs, so the relative proportions of defense and domestic changed dramatically.

As we argued in previous chapters, and as the administration acknowledged in its FY84 budget, defense need not grow with the economy.^[9] Yet in 1979 both the public and elites felt the threat and thus requirements had increased. Therefore, defense was bound to retrieve some of its budget share; the free ride for domestic spending was over.

Most domestic increase came from payments to individuals, such as social security and medicare. These may be divided into direct payments, such as social security, and those administered though the states, such as AFDC and medicaid. Most poverty programs expanded sharply from 1967 to 1972 as the federal government removed restrictions on access and as medicaid took effect. After 1973, however, the government made few new commitments. AFDC shrunk in real terms, but food stamps picked up the slack. Growth of poverty programs, as shown in Table 9 ("income security") was essentially cyclical.

The growth of "other grants"—the panoply of Great Society efforts to aid and influence local governments: revenue sharing, CETA public service jobs, Urban Mass Transit, Title 1 education fund, and so on—stopped in 1978, under Carter, as the budget crunch began. Poverty programs and these grants were the major victims of the 1981 reconciliation; in early 1983 OMB still planned major cuts. As for other program spending—everything from NASA to the FBI, the census, water projects, and national parks—they had stopped growing as a share of GNP in 1968.

What grew and kept growing were the non-means-tested entitlements funded directly by the federal government: civil service and military pensions and, most of all, OASDHI (that is, social security and medicare). These grew, however, not because of post-1973 decisions but because of pre-1973 commitments. They grew mainly because the eligible population grew: people were living longer, and more and more of them were fully vested in the social security system. In 1962 the Old Age and Disability funds had 16.8 million beneficiaries; in 1972, 25.2 million; in 1982, 31.9 million.^[10] In the case of medicare, they grew because virtually nobody anticipated the costs of coverage; Congress began struggling with medical inflation soon after the program began. They grew, finally, because politicians kept benefits in line with rising prosperity but out of line with recipients' contributions; legislators, eager to please a very powerful group of voters, seemed confident that continuing economic growth would enable current workers to pay the bill. The last big increase in 1972 tells both sides of the story: Democrats and Republicans competed to woo the elderly; the resulting increase "was financed largely by a change in actuarial assumptions" that politicians had little reason not

Table continued on next page

Table continued from previous

to accept.^[11] There were only two difficulties: the economy did not behave as expected, and almost everybody was asking the wrong question.

The economy went into a "quiet depression"^[12] in 1973, and social security got into deep trouble. The social security funding question—Would that system's taxes fund that system's expenses?—was politically important. But throughout the program's history, it begged a second question: If social security taxes increased, would those be new taxes or replace old ones? (Tax aficionados speak of "fiscal cannibalism" where one tax eats up another.) One way, the tax burden went up; the other way, either other programs went down or deficits went up. Acting "responsibly" in funding OASDHI, the government had the payroll tax rise from 3.1 percent of GNP in 1962, to 4.7 percent in 1972, to 6.6 percent in 1982.^[13] It was already scheduled to go higher; the 1983 compromise would speed up these increases. Whenever the government propelled increasing taxes out into the future—as it did in the original 1935 act and most subsequent modifications—it avoided the choice as to how the increase would affect the rest of federal activity.

Thus, the repetition of what appears a truism—social security is self-financing—by Ronald Reagan, his congressional opponents, and spokesmen for the elderly, though true as far as it goes, leaves the mistaken impression that the biggest program by far would not affect the rest of government. Nonsense. To believe this is to believe that it doesn't matter how high any tax is as long as it supports a given activity. Individuals and businesses feel taxation from all sources, so a big increase here is bound to affect the possibilities of other increases elsewhere.

Here's the catch: maybe the government should not have chosen. No one could see the size of any future problems; perhaps they would be small, for economic growth heals many ills. Perhaps it was for people in the future to choose whether they wanted higher taxes or fewer programs, or even (gasp!) higher deficits. If politicians in 1962, say, had decided that other programs would have to be cut by 1.6 percent of GNP (the amount OASDHI would grow) by 1972, how on earth could they have done so? Okay, one might say: If you can't plan, don't make the commitment. Don't have a national retirement plan or any policies with long-term increasing costs. Toss out civil service pensions and medicare while you're at it. But that's not possible or desirable. People in real life make long-term commitments like buying a house, sometimes even with an adjustable-rate mortgage. They have children, plan to send them to college; they will figure out later what to sacrifice to that end. The government, with heavy public support, committed itself to social security. There is little evidence that people, on the whole, dispute either that commitment or its necessity. Some problems come with the territory.

Is there, then, no room for choice? Hardly, one can always exercise

prudence. The 1972 increases, for example, could have been smaller. Many provisions, such as early retirement, could be less generous. Here we want to emphasize only that both Carter and Reagan, along with politicians of all parties, got caught in a fiscal tidal wave, beginning in 1979, that had begun long before; much of its force was inevitable. Most of the increase in spending after 1979 constituted old commitments hitting the beach of a new economy. Huge deficits would have occurred even without Ronald Reagan—unless you think taxes were going well above the level that helped defeat Jimmy Carter.

What could the politicians have done, in 1980–1981, about the coming tidal wave? Could they even know it was coming? In the president's FY84 budget, published at the beginning of 1983, Stockman answered those questions. He argued that events from 1970 to 1981 produced a "profound disequilibrium in the inherited 1981 budget." He wanted to get domestic spending down to 1970's share, but that was nearly impossible. The administration's solution was the magic asterisk of unspecified savings in the Economic Recovery Program, to shrink the numerator, and "an immediate, rapid and sustained expansion of GNP" to increase the denominator of the outlays/GNP ratio.^[14] This presumed, he reported rather drily, "the transition from rising to falling inflation and from low real growth to rapid output expansion would occur immediately and simultaneously, and without intervening financial and economic disturbance."^[15] If everything had worked, spending would have stabilized after 1984 between 19 and 20 percent of GNP.

What went wrong? In 1983 Stockman blamed the economy. Later, in his book, he blames himself and his colleagues for misestimating the economy. Rosy Scenario, he writes, overestimated GNP, and thus the revenue base, by $2 trillion over FY82–86.^[16] They thereby hid, from themselves and from others, the deficit consequences of the tax cut. Stockman is partially right: they did rig the forecast; if it didn't fool Domenici, Baker, and other congressional leaders, it did make it harder for them to oppose the package. But Stockman was right in 1983 as well; the economy performed far worse than anyone had expected. For example, the excessive GNP projections in Rosy Scenario were no higher than those made by Carter's CEA in his last (January 1981) budget. Stockman predicted more growth, Carter, more inflation, creating similar nominal GNP.^[17] Neither side anticipated the revenue consequences of actually stopping inflation; no one was predicting the slump that occurred or its far greater than normal decrease in revenues.

There is something to be said for erring with the conventional wisdom, as opposed to making up a logically inconsistent theory with which to err. Stockman's mea culpa is not all wrong. But, as we saw, the whole Republican establishment confirmed Rosy Scenario, albeit in some cases

grudgingly. The Republicans could do so because, panicked about the state of the economy, no one knew quite what to do.

It is hard to argue that the first two installments of the tax cut, in 1981 and 1982, were excessive either in prospect (when they seemed smaller because people anticipated more inflation, thus increasing revenues) or in effect (for they responded to a much bigger recession than was anticipated). As for business tax cuts, Democratic tax leaders accepted the basic size of the administration's package, for Democratic economists were also pushing "capital formation." The original Hance-Conable compromise, on June 4, 1981, would have cut taxes less than the final bill. But a lot of the add-ons—concessions to business lobbyists on June 9, oil and other breaks in the auction—were reversed, albeit with difficulty, in TEFRA. Given what economists were saying in 1981, as well as the actual course of the economy, it is hard to argue that tax policy as of early 1983 could have been much different, with one glaring exception: the upcoming third year of the tax cut.

The most thorough analysis of the deficit's causes was made by Gregory Mills and John Palmer of the Urban Institute in September 1983; thus, it included the deficit reductions from TEFRA, the social security rescue, and the economic picture as understood by Congress in its FY84 budget resolution. Mills and Palmer also assumed that taxes would not have risen beyond the FY81 level, remaining at 20.8 percent of GNP. Table 11^[18] indicates (compare lines 5 and 8) that policy did not significantly affect the deficit until FY84 and after. Even in FY86, only $100 billion of a projected $237 billion deficit would be due to policy choices. Of that $100 billion, similar shares are accounted for by tax cuts and defense buildup (lines 9 and 12, offset by line 13).

In 1983, therefore, the politicians faced $200 billion deficits far out into the future. Over FY84–86, however, $174 billion, $149 billion, and $137 billion (line 2 minus line 8) of those deficits were not, in any plausible way, their fault. After 1980's experience with a $15 billion deficit, we suspect, those deficits alone would have been enough to create political chaos.

Surely something must be wrong with the analysis because "everybody knows" that Ronald Reagan was responsible for the huge size of the deficit. Surely his tax cuts and defense increases are the villains. Aside from encouraging doubters to reread what they have just read, this view is so prevalent and so strongly held that it deserves direct confrontation.

Was Reagan responsible for the deficit? We answer: mostly no, but partially yes. To say that the nation would have stood still for even higher taxes in 1981 and 1982 not only goes against historical experience but also implies that Democrats would not have cut taxes. The evidence is that

they would have, albeit not as much. The difference between the two parties is essentially the third year of the tax cut.

Although Reagan did not start the defense buildup, he pushed it farther and faster than a more fiscally prudent president might. That part of the deficit, by no means the largest part but a part nonetheless, is fairly attributable to him. With the benefit of hindsight, we can observe that after 1984 a build-down occurred, thus reducing the impact of defense spending on the long-term deficit. That deficit reduction, of course, was not his fault. Even the famous Professor Hindzeit, less fallible than the rest of us, has not had sufficient time to test whether the rise in defense spending was worthwhile or not large enough.

If not Reagan, who was responsible? A good half or more of the deficit may be attributed to the recession, which, as far as anyone knows, was either inevitable or precipitated by the Federal Reserve Bank under the aegis of Paul Volcker. Having perhaps taken off the monetary brakes too soon in 1980, when inflation surged again, Volcker may have kept them on too long in 1981 and 1982. Possibly another Fed chairman might have brought the economy in with a softer landing. Inflation and employment might have declined more slowly, leading to less individual suffering and considerably more revenue. Maybe. But citizens were so distraught over inflation that "just right"—had anyone been able to figure out then what that would be—might have been considered too little and too late by market and populace. Because Reagan supported Volcker and because both men claimed credit for stopping inflation, credit for the good face of deflation has to go with blame for its bad side in reducing revenues.

Should the ghost of commitments past, as well as present supporters of social security, like Senator Moynihan, Representative Pepper, and Speaker O'Neill, take part of the blame? After all, social security payments (and, therefore, taxes) could well have been lower, thereby leaving more room for other taxes. They, too, or others like them who had the authority in the early 1970s, erred on the side of optimism in expanding the clientele and the awards under social security. No doubt they will accept blame for the most popular program in America.

A reasonable conclusion would be that all who govern now and in the past half century share some blame for the deficit: some more than others, but few are guiltless. Overall, American economic performance since the end of the Second World War has been good. Nevertheless, the 1970s left a lot to be desired. Much of this slowdown may be due to international oil increases or vast numbers of new entrants into the labor market or emergent economic powers on the world scene or too high taxes or insufficient demand or too little saving, some of which government could affect and much of which it could not.^[19] Finger pointing

does not, in any event, tell us whether (and, if so, how much) we should worry about the deficit—our next subject—or what public officials might do about it (and how), the subject of the rest of the book.