Entitlements

His combined congressional/civil-service pension provided him initially with $1,560 a month, or $18,720 a year. He also received retirement pay, $55l a month, for service on active military duty and in the reserves, which he began getting in 1975. And in 1980, when he turned sixty-five, Keith drew a third government benefit, his monthly Social Security check.

Like other federal benefits, Keith's were fully "indexed" against inflation, which meant that they were increased once, and in some cases twice, a year, in step with changes in the Consumer Price Index (CPI). Because of these increases, Keith, whose highest salary as a congressman had been $42,500, found himself by 1982 taking in nearly $61,000 a year in federal pensions. His congressional pension had more than doubled since 1973, to $3,419 a month. His military pension had risen to $903 a month, and his Social Security payment was $691 a month. Another round of cost-of-living adjustments made this summer raised Keith's federal pensions yet again, to a total of more than $65,000.

And that was not the end of federal benefits for the Keith household. Keith's wife had retired from the federal government during a "reduction in force " or RIF, when she was forty-six years old. Since she had put in twenty-five years of federal service, she was, as a victim of a RIF, entitled to a pension, which started at $550 a month in 1976. By this year, it had nearly doubled, to some $1,000 a month. All told, then, the Keith family's federal pensions came to more than $77,000 as of last summer--and all of it was fully indexed to future increases in inflation.

To Hastings Keith, this seemed excessive. He wrote a letter to the secretary of the Treasury, Donald Regan, explaining why he wanted to turn over three checks, totalling $3,107, each representing a portion of his monthly allotments that he considered unwarranted.

Keith's hope had been to present his checks in a public ceremony, where he would register his complaint about the system that depleted the Treasury to finance windfalls such as his. The government shared little of his enthusiasm for the plan. Instead, he was granted an audience with an assistant secretary of the Treasury, who listened politely to his case, but did nothing.

His rebuff at the Treasury did not stop Hastings Keith, who has been telling his story to congressional committees and devoting his extra money to the effort to change the federal retirement system. But the government's lack of interest in his cause helps explain some of the current agonies over the federal budget and federal deficits.

Although the generosity of congressional pensions makes the sums in this example extreme, they illustrate a basic fact of modern government. Benefits that are indiscriminately awarded, and then ceaselessly increased, add more to the federal budget than do those more familiar evils "cost overrun" and "welfare fraud." Yet politicians of every camp have found it perilous to tackle, or even to acknowledge, this problem.

Since early this year, both friends and foes of the Reagan Administration have understood that the next four years might well produce the largest sustained budget deficits in American history. When the administration released its budget proposals for fiscal year 1983 last spring, it did what no other administration had ever done: for the years ahead it predicted a smooth, robust return to national prosperity yet it still foresaw budget deficits that would hover in the vicinity of $100 billion a year.

Last year's reductions in tax rates are part of the reason for the deficits, as are the administration's plans for a sustained military buildup. But while these two policies have been debated to death, neither the administration nor its critics have yet wrestled with the largest single force behind the deficits. Neither liberals nor conservatives have yet devised a confident approach to that portion of federal spending known as "entitlements."

Entitlements, sometimes called "payments to individuals," are technically defined as benefits for which people qualify automatically, by virtue of their age or income or occupation. Social Security is such an entitlement, by far the largest. So are medical programs such as Medicare and Medicaid, civil-service and military pensions, unemployment insurance and price-support payments for farmers, and (with certain technical quibbles over definitions) subsidized housing and food stamps.

From a budgetary point of view, the significance of such programs is that, at least in the short run their costs cannot be controlled. If the Congress appropriates $2 billion to build a dam or a highway, it can be confident that no more than $2 billion may legally be spent. But when it authorizes extended unemployment benefits, or a different reimbursement formula under Medicare, it can set no limit on the money that will ultimately flow from the Treasury, since the government is legally obligated to provide benefits to anyone who can prove that he is eligible.

The Reagan Administration took office committed to a fundamental redefinition of federal responsibilities, but it will almost certainly depart having made no dent in entitlements. From less than one third of federal spending in 1970, entitlements rose to represent nearly half (of a much larger total) in 1980. Even if the Reagan Administration should continue to be granted its every wish for more military spending, and even if Congress were to concur in all its proposed cuts in the rest of the budget, entitlements would still make up nearly half of the budget in 1983 and 1984.

The two national parties have virtually raced each other to distance themselves from this dangerous issue. Old-style liberals to "neo-liberals," the Democrats have attacked almost every other part of the administration's economic plan, without volunteering their ideas about this half of the budget. In its top-to-bottom re-examination of government spending, the administration exempted from Budget Director David Stockman's scrutiny not only the military but also the largest and fastest-growing entitlements.

The "entitlement problem" is often thought of as a "welfare problem." To a trivial extent, it is: most federal programs for the needy, including the classic welfare program, AFDC (Aid to Families with Dependent Children), do fall within the entitlement budget. But they make up a small part of the whole. Only a sixth of all the money spent on entitlements is for programs that are "means-tested," or aimed at the poor, and those programs are the slowest-growing part of the entitlements budget. For example, AFDC at $7 billion, costs one third as much as civil-service pensions, and it shrinks while the pensions grow. Rather than a "welfare problem," the growth of the entitlements is actually a "retirement problem."

October 1 marks the beginning of the government's fiscal year, and in fiscal year 1983, just begun, the federal government will, according to the latest congressional budget resolutions, spend some $770 billion for all its varied activities. Of that total, $362 billion will go toward entitlements; of the $362 billion, $264 billion will be for retirement programs. In other words, one out of every three dollars the federal government spends this year will be spent on pensions or on medical care for those over the age of sixty-five.

Social Security, of course, heads the list of retirement programs; it will cost about $170 billion in 1983. Medicare, the medical-care program for people over the age of sixty-five, is second largest, at $49 billion. Together, these two programs account for more than half of the entitlement budget and more than a quarter of all federal spending. Two other major retirement items are civil-service pensions, $21.2 billion in 1983, and military retirement pay, $16.2 billion. By way of comparison, the largest means-tested program, Medicaid (Medicare's counterpart for the needy) is expected to cost $18.5 billion in 1983, and food stamps will be about $12 billion.

Retirement programs not only dominate the federal budget, they also represent its largest area of growth. The basic reason for this growth is demographic: over the past generation, more people have been surviving to retirement age, and the birthrate has declined. In 1950, roughly one American out of every forty-three drew benefits from the main Social Security account, known as Old Age and Survivors Insurance. Now about one in seven receives payments (including those who draw from the newer Disability Insurance account). In 1950, every 100 working Americans had to provide, through their Social Security taxes, the support for six retired people. Now each worker's obligation is five times as great (thirty-one retirees per 100 workers). That ratio should remain steady for the next twenty-five years, while the children of the postwar baby boom remain in the workforce. But early in the next century, the ratio should shift again. Fifty years from now, each person in the office or on the assembly line may be responsible not only for his own livelihood but also for half the support of a retired countryman.

The average single man who retired in 1960 received six and a half times as much money in benefits as he ever paid in. (These comparisons are adjusted for inflation, and for the interest that the taxpayer could have been earning on his contributions.) The man who retired in 1970 got back more than three times as much as he contributed, and the man who retired in 1980 should eventually receive two and a quarter times as much money as he paid in. But by the time one of today's teenagers retires, in the year 2030, he will have paid more into the system than he will ever get back.

Moreover, while life-spans were increasing, the retirement age was being lowered. The average man who turned twenty in 1940 could expect to live six months past his sixty-eighth birthday. Since the retirement age was then sixty-five, he could expect to spend three and a half years receiving the retirement benefits toward which he had contributed during his forty-five years of work. In 1956, the Social Security retirement age for women was lowered to sixty-two; it was lowered for men in 1961. (Retiring at sixty-two instead of sixty-five means a maximum reduction of 20 percent in Social Security benefits, but most of those eligible choose to take their benefits as soon as possible.

The average man who turned twenty in 1977, could expect to live to the age of seventy-five, or thirteen years past his retirement age. Consider how this young man's expectations would differ from his older counterparts. The thirteen years he would spend in retirement would be nearly one third as long as his forty-two-year working career. Forty years ago, the average retirement was only one twelfth as long as the average working career.

Demographic trends, powerful as they are, do not fully explain the growth of the retirement programs; political maneuvering deserves much of the credit. The explosion in retirement benefits derives in large part from a reform gone awry.

The reform in question was an attempt to take the Politics out of federal benefits. In 1962, Congress undertook a reform of civil-service pensions. Until that time, Congress had to approve each pension increase, in rituals very much like its regular votes to raise the ceiling on the national debt. Good-government spokesmen argued that this sapped the Congress's energies and debased its deliberations, by enticing congressmen to play politics with pension adjustments. It would be far better, they said, to delegate adjustments to the automatic workings of a formula.

The reformers also recommended an improvement over the military's system of raising pensions whenever active-duty pay went up. In those days when federal salaries were considered to be "low," it was argued that the government might need to increase its pay selectively to attract the right talent; there was no logical reason to pass this on as a windfall to retirees. Pensions should rise only in accordance with the cost of living.

Starting in 1962, therefore, pensions for civil servants were to be adjusted once a year, in proportion to the rise in the CPI. Indeed, starting in 1969, the adjustment was one percent more than the CPI increase. This one percent "kicker" was supposed to make up for the lag between the rise in prices and the annual adjustment in the checks.

The other major step came in 1972, when Congress voted to index Social Security benefits to the CPI (but with no "kicker"). Until then, Social Security benefit levels had also been adjusted by the Congress. Switching to the index plan was, once again, supposed to shield the system from political demagoguery. The logic seemed so compelling that by the end of the 1970s, 30 percent of all federal spending was directly indexed to the CPI, and another 14 percent was indexed to other measures.

The architects of indexing failed to foresee what would happen next, and understandably so: through most of the fifties and sixties, inflation had averaged less than 2 percent a year. But at just the moment when so large a share of federal spending was being tied to the CPI, the historic American inflation of the 1970s began. As prices increased by 8, 10, 12 percent each year, exactly in pace rose federal payments.

No, not exactly in pace. In fact, the payments rose faster, because of peculiarities in the design of the indexing formulas. The kicker for federal pensions, for one thing, put federal retirees far ahead of the game, until it was removed in 1976. Civil-service retirement pay cost $2.8 billion in 1970; after ten years of sustained inflation, it reached $14.7 billion in 1980, an increase of 525 percent. Perversely, indexing meant that it could be more profitable to retire than to work. Eric Heimel wrote in the Journal of Contemporary Studies that a four star general who retired in 1971 would have seen his retirement pay increase, through indexing, to $64,000 in 1981, while a four-star general still on active duty in 1981 would be earning $54,000.

In principle, indexed increases did nothing more radical than enable federal pensioners to keep even with inflation. But in practice, their effect was quite different. During the inflation of the 1970s, few prices rose faster than those of energy and housing; but the increases in those prices, dramatic enough in reality, were exaggerated by the formulas used to calculate the Consumer Price Index. The costs of home ownership, which made up one quarter of the "market basket" of goods whose prices determined the CPI, were calculated in a way that made the index extremely sensitive to changes in the mortgage rate. For example, if the mortgage rate were to rise by one point in a month, from 10 percent to 11, that change would, all by itself, raise the CPI for that month by almost one full point, which could be "annualized" into an inflation rate of 12 percent. The Congressional Budget Office has estimated that this and other quirks in the CPI's treatment of housing costs exaggerated the real rise in consumer prices by as much as 6.1 points between 1978 and 1980, with each of those points triggering an extra $2 billion in federal spending each year. As for the cost of energy, the CPI market basket assumed that families would keep buying the same amounts of gas and oil and electricity, no matter how much prices might rise or fall. The real market basket of American consumers has of course changed markedly from that of 1972. The era of cheap fuel ended, and consumers cut back, but the CPI ignores such adjustments totally.

Beyond these technical defects in the index was a larger question of fairness, for "full indexation" led to very different fates for different segments of the American public. It created a class that had nothing to fear from the inflation that was the scourge of the rest of the populace. Those who retired from private companies were, if anything, more vulnerable to inflation than their working counterparts, since more than 90 percent of private pension plans offer no cost-of-living protection. (Hastings Keith's pension from the private firm where he spent twenty years started out at $160 a month and dropped to $100 when he began receiving Social Security) But because of the technical oddities of the CPI, those who received federal pensions and Social Security could come out ahead when prices rose. For one class of beneficiaries, indexation all but shattered the connection between the nation's productivity and their economic welfare.

There is one further engine of growth at work within the retirement accounts: medical insurance. In financial terms, the medical programs are potentially the most explosive of all; they also present the most intractable of the entitlement dilemmas.

Through the past decade, no major program run by the federal government has grown more quickly than medical care for the aged. Between 1970 and 1980, federal payments for medical entitlements, of which Medicare makes up three quarters (Medicaid accounts for nearly all of the rest), rose by 133 percent in real dollars (i.e., adjusted to remove the effects of inflation). Medicare now costs less than one third as much as Social Security, but it is increasing at a much faster rate. According to projections based on the 1981 Social Security Trustees' report, Medicare could, by the year 2006, surpass Social Security to become the largest single item in the federal budget.

That Medicare should be outstripping Social Security suggests that it is subject to pressures other than demographics, since the clientele of the two programs is almost the same. The pressures driving Medicare costs ever higher are built into the medical system.

American doctors, researchers, and hospitals have proven far more brilliant in combating the ravages of age than in figuring out how to do so at a controllable cost. This imbalance presents choices that are inescapably cruel: who shall be denied kidney dialysis if the cost of serving everyone is too high? Who shall pay the cost if no one is to be denied? These are precisely the sorts of choices that democratic governments find most awkward to make, and medical-entitlement plans reflect that uneasiness. Under Medicare and Medicaid, the choices finally do get made, but only in a backhanded fashion that does not pretend to be rational and does not succeed in being economical.

The choices fall to the doctors: if your physician authorizes a trip to the hospital and elaborate tests, Medicare will pick up the costs of the next sixty days (after you pay the first $260). In principle, the brake on expenditures is the cumulative effect of half-a-million doctors' prudent decisions. In practice, the limit is often set by local availability of hospital beds, of specialists, and of machines.

After years of bitter struggle, Congress enacted the Medicare program in 1966. Organized medicine, led by the American Medical Association, represented the principal opposition; as part of the political bargaining that ensured congressional passage, Medicare took on features that guaranteed that doctors and hospitals would not suffer financially under the plan. Doctors would be reimbursed according to a schedule of permissible fees--but the schedule would be based on rates set by the doctors themselves. Moreover, the doctors were free to bill their patients for supplements on top of the charge to Medicare. Hospitals, unlike doctors, agreed to accept Medicare's reimbursement as payment in full for a patient's bill. But that reimbursement would be based on the hospital's rendering of "reasonable costs" of care. Hospitalization insurance was automatic under Medicare. To get the extra coverage for doctors' bills, each person had to pay a premium, now $12.20 a month. Nearly everyone chooses the additional coverage, and so is effectively shielded from rising medical costs--which was, of course, one of the fundamental purposes of the plan.

The result of these arrangements was another of the "cost-plus" schemes so common in government contracting. No party to the transaction had both the incentive and the ability to economize on care. A hospital-building boom left new beds waiting to be filled. Filled they soon were, and the charges were passed on to Medicare or to private insurers. These "third-party reimbursers" watched their payments shoot through the roof--hospital charges have typically risen by 15 to 20 percent each year.

This general quandary for American medicine has created a particular challenge for Medicare, because expensive new medical techniques are most often used to treat older patients. Nearly a quarter of all Medicare payments are made in the last year of a recipient's life. While Medicare has been a resounding success in enabling retired Americans to meet their routine medical needs without financial terror, its budget is tremendously skewed by the costly last efforts to fend off death.

Political arguments about medical care have simmered down in the past half-dozen years, because of the near-universal assumption that the time has not been right to create another expensive new entitlement, in the form of a national health system. Yet the very difficulty of balancing our public accounts may soon refocus political attention on the basic questions of who gets what kind of medical care, and how we can pay for it.

As it drew up its proposals for the 1983 budget, the administration decided not to recommend any reductions in Social Security, and only minor adjustments in Medicare and federal pension plans. In so doing, it started out by declaring one third of the budget off-limits for spending cuts. Another quarter of the budget belonged to the military, and its share was supposed to increase to one third over the next few years. A further 15 percent of the 1983 budget was committed to repaying interest on the national debt.

Together, these three areas account for nearly three fourths of all federal spending without even counting the other entitlements. If they could not be cut, an administration pledged to frugality and embarrassed by mammoth deficits had little choice but to slash everything else. The administration proposed "zeroing out" most job-training programs cutting research-and-development funds, and taking other difficult steps. There was one final possibility for savings: the "other" entitlements, those aimed at the poor. This possibility the administration seized.

Between Jimmy Carter's budget for fiscal year 1981 and Ronald Reagan's proposals for 1983, the major pension programs increased by 20 percent and Medicare increased by 30 percent. But in that same period. AFDC was cut by 20 percent, the low-income energy-assistance program by 30 percent and a variety of nutrition programs, including food stamps, by 15 percent.

The unequal sharing of the sacrifice became one of the administration's central political problems: it led even conservative pillars such as Senator Robert Dole to complain about shoving the burden onto the poor. Even so, the cuts were not enough to solve the budget problems. No matter how deeply it might probe elsewhere, an administration that chose not to confront the biggest entitlements (to say nothing of the military) was left with the prospect of deficits so stupendous that they could double the national debt within five years.

In the middle sixties, it took the simultaneous pursuit of a war on poverty and a war in Vietnam to throw the federal budget into chronic deficit. Now it might require nothing more ambitious than meeting our routine obligations.

Can the entitlements explosion be contained? The ultimate answer demands a reconsideration of basic political premises that we have found comforting since the New Deal. Yet certain technical changes, primarily involving pensions for public employees, provide a place to start.

The single most important technical change is to begin dismantling preferential federal pensions and bring new federal employees into the Social Security system.

Federal employees now enjoy the nation's most generous pension plan. Most private pensions aim at replacing 20 to 25 percent of a worker's previous earnings. Social Security, reduced living expenses, and, yes, savings are supposed to make up the rest. A federal employee with thirty years service will typically receive 56 percent of the average salary for his last three years on the job, congressmen 80 percent. All of it is fully indexed, of course (although the Congress did, this summer, limit, indexing for retirees under the age of sixty-two, on the theory that most of them were holding other jobs). Federal retirees receive more in pensions than retirees from all private businesses combined.

The pension plans of private businesses usually require the worker and his firm to invest, in varying proportions, the resources that will pay dividends during retirement. Federal pension plans don't come close to doing so. The largest contributions come neither from the workers nor from the agencies that employ them but from general federal funds. Patrick Owens reported in Newsday that in 1980, federal workers and their agencies contributed a total of $3.6 billion toward their retirement funds. The Treasury contributed an extra $11.9 billion. The unfunded liabilities of the federal retirement systems--the commitments made to workers but backed by no investment or trust fund--now amount to some $1 trillion, roughly as much as the entire national debt. According to Hastings Keith, unfunded liabilities average $149,000 PER PERSON in the military, $173,000 in the civil service, and $530,000 in the foreign service.

The generosity of federal pensions is a legacy of the 1920s, when federal pay was poor. For the past twenty years, the federal government has embraced the principle of "comparability," offering salaries comparable to those in business. Why should the principle not extend to pensions as well?

"Integrating" federal employees into Social Security would ultimately mean more generous benefits for low-paid federal employees, especially those who do not work for the government long enough to earn pension rights, since they would be able to count their years in government service toward their Social Security benefits. At the other end of the scale, it would mean an end to the windfalls now available to "double-dippers": government employees who, having qualified for pensions while still in their late forties or fifties, take private jobs and become eligible for Social Security as well. Robert Myers, a former chief actuary for Social Security who now directs a presidential commission on Social Security, which will recommend reforms for the system this winter, contends that THREE QUARTERS of all retired government workers are either receiving a second pension or are working in a job that will qualify them to do so.

A few groups of federal employees have already set an example that might be widely emulated. The retirement plan for the Tennessee Valley Authority has since the 1950s been integrated with Social Security. Like most private pension plans, it does not count military service toward its pensions; it applies its disability standards more strictly than does the civil-service retirement system; and it discourages early retirement. Except for the provisions for early retirement, it offers benefits comparable to those of other civil servants; yet, unlike most other systems, it is actuarially sound. The Federal Reserve pension plan and a few others operate on similar principles.

The logic of comparability also suggests a change in military pensions. Career soldiers can now retire on half pay after twenty years of service, or three-quarters pay if they serve for thirty years. (Members of the military are also eligible for Social Security.)

The demands of military duty obviously differ from those of any other work. But the twenty-year career costs a great deal while adding very little to our national defense. Gradually converting the military to a thirty-year career might mean increasing active-duty pay, especially for the sergeants and petty officers who are now the services' scarcest resource. In the long run, this would probably cost less than adding to the mountain of unfunded pension liabilities; and by reducing many officers' sense that their military duty can be only a "first career," it might well enhance the services' integrity.

Yet all such reforms take us only so far. No matter how we might crack down on the retired colonels and GS-15s, it will not make much more difference in the entitlements budget than cracking down on welfare mothers would. Our national accounts will remain out of balance until we wrestle with the IOUs that middle-class America has issued to itself, through Social Security and Medicare. The two awesome challenges of the entitlements are to control medical costs and to re-examine our notion of who "needs" public help.

The budgets for Medicare and Medicaid, in their leaps to the sky, reflect the same pressures that are driving Blue Cross premiums and personal medical bills up and up. Tinkering with the government programs is the minor solution to this problem: the major one is changing incentives in the medical system. In forcing this issue to the Center of political attention, the entitlements mess may indirectly do some good.

One political camp, exemplified by Senator Edward Kennedy, contends that the government can never contain medical costs unless it controls the medical system. To that end, Kennedy has tirelessly pushed for a nationwide, largely publicly financed, centrally planned approach to health care. The opposite view, espoused by David Stockman in his days in the Congress, holds that rising medical costs reflect imperfections in the medical market. The answer, therefore, is to unleash market forces through more competition.

Of the two views, Stockman's seems more in harmony with the politics of the time. The Reagan Administration plans to unveil a new medical-reform scheme, emphasizing "market forces," about the time of the President's State of the Union address next January.

The logic of the market means encouraging the pre-paid medical plans known as health-maintenance organizations. It also means--though this may or may not be a feature of the President's plan--removing some of the padding that now protects the typical patient from his medical bills. Economists lament "third party" payment schemes, whether run by Blue Cross or by the U.S. government, because they blunt the instinct for frugality that is so powerful when one's own money is at stake. When the patient can pass the bill to someone else, why should he worry about unnecessary tests? Why should he wait two weeks for an appointment at the doctor's office when he can use the hospital emergency room as a clinic? He will ultimately pay the bill, of course, through higher taxes or insurance premiums; but the market functions poorly when its messages are so long delayed and so weak.

One government economist, who chooses to remain unnamed, has taken reliance on the market a radical step further. Medicare's biggest economic problem, the economist argues, is the last-resort measures taken before an elderly patient's death, which dwarf the costs of previous care. When patients and their families decide how hard the doctors should fight, and with what machines, finances do not color the deliberations, because Medicare covers nearly all hospital fees. The economist argues that when such a patient dies, his estate should be tithed to recover a share of his "Medicare profit"--the difference between the premiums he has put into the system and the benefits paid on his behalf. This, the argument continues, would repair the social contract between the generations. It would also build into individual decisions the same realities that the nation faces, as high-priced medical possibilities grow more rapidly than resources.

This may seem a ghoulish suggestion, as its author is aware, but when public leaders dismiss out of hand the possibility of questioning the rules, even though those rules lead straight to bankruptcy, are a few ghoulish thoughts so much more dangerous than the conventional wisdom? Whether or not this specific proposal is workable, the spirit behind it is constructive, because it demonstrates a willingness to think of new ways out of a predicament.

The main idea waiting to be spoken is that people have different claims on public support. Yes, there are cases in which the state bestows its benefits equally on all. The Children of rich and poor alike should be entitled to schooling at public expense. But to extend that logic to pensions and subsidies leads to commitments beyond our national means. The people who really NEED help need more than they are now getting. If they are to have it, other people must have less.

In the case of medical care, there are certain obligations that the state should assume on everyone's behalf. Any burn victim, to choose one example, should be entitled to the best care that is available, regardless of the cost. But there are other obligations that the nation cannot assume, simply because there would never be enough money to pay for them. Medical technology now offers answers to many of mankind's universal complaints. Corneal surgery can eliminate much of the need for eyeglasses. Joints can be rebuilt, arteries reamed, gums surgically lifted and the underlying bone scraped to remove plaque. We all "need" such services, but if all these innumerable repairs to all parts of all bodies were carried out, the cost would rival the gross national product.

How, then, can we decide where to draw the line? One way is to ask each patient to share more of the cost, giving him a stake in the decision about costly care. But if that is done without recognizing the differences in economic need, it will deny some people a more basic level of medical care.

A twenty-dollar physician's fee, or a $500 deductible for hospitalization, would mean quite different things to a retired couple with pensions and dividends of $30,000 and a widow surviving on $4,000 from Social Security. For the one, it would mean some sacrifice; for the other, denial of care. Yet our retirement plans, in their magnificent evenhandedness, treat the poor widow and the well-off couple the same. When we provide for some, we must provide for all. If we change the rules so as to balance the budget, we revoke the widow's right to treatment. But if we do not, we lose control of our financial future.

The original genius of Social Security was precisely that it did treat everyone the same. No one need feel humiliated by accepting its benefits, because it was not welfare. Everyone was included in the plan. As a political ideal, this is most attractive; but, in Social Security as in Medicare, it may simply have become too costly to sustain.

Consider two of the major proposals for bringing Social Security's commitments into line with its resources. One is to raise the retirement age, in recognition of medical improvements and demographic shifts. Raising the retirement age to sixty-eight by the end of this century would keep the average time in retirement one quarter as long as the average adult life-span--the same balance that has prevailed for the past few years. Another proposal is to hold down the automatic increases in Social Security and other benefits, perhaps by gearing them to the Consumer Price Index or the average increase in wages, whichever was less in a given year. If wages rose faster than prices, the buying power of the pension would be fully protected. If the reverse, people of all ages would share in the sacrifice of a less productive economy. (The CPI will, in any case, soon be revised to eliminate the housing-cost bias.)

But if changes like these were applied "fairly," with equal effect on all, the results would be unfair. Some people need the payments; others do not.

To speak of "the aged," or "Social Security recipients," as one homogeneous group no longer makes sense. As recently as twenty years ago, it did. Those above sixty-five were then an economically distinct group. Like black Americans or single women heading households, they were on the whole poorer than other people. The economic differences within the group were less important than the gap between them and the American norm. To aim a program at all retired people not only made political sense, in avoiding the taint of welfare; it also made economic sense, for it transferred money to people in need.

That has changed. Older Americans are now economically very much like their children's generation, and their grandchildrens'. Michael Hurd and John B. Shoven, of the National Bureau of Economic Research, have reported that the per capita income of people over sixty-five is now higher than for the population as a whole. In 1978, it was 121 percent of the national norm. Two thirds of all people over sixty-five own at least one home.

As the differences between the generations have diminished, the differences among older people have grown more acute. On average, those above sixty-five enjoy parity with those below; but the average conceals many retired people who are desperately poor. About a fifth of all retired couples have incomes above $24,000; but of the 5.8 million single women sixty-five or older, fully half have incomes below $5,000. As the world works, the wealthiest couples generally receive the highest Social Security benefits (because they earned more when working) but depend on them least. For the couples making more than $24,000, the comparatively large Social Security check represents about one sixth of their income. For the poorest elderly women, their smaller check is more than 80 percent of their total support.

Can it be fair to treat these people "equally," to hold the widow's $300 a month and the couple's $900 a month to an "even" 5 percent increase? Only a perverted sense of fairness is thereby honored; yet it is this kind of equity that our evenhanded entitlement policy now serves.

In practical terms, the solution might be to provide a full cost-of-living adjustment for only a certain portion of Social Security benefits--say, the first $500 a month. A more significant step might be to make Social Security benefits subject to federal income tax. The automatic workings of the IRS are administratively far simpler than any other means for concentrating scarce resources on people in need. Social Security is already subject to a tax of sorts--the "earnings test," which reduces benefits by $1 for each $2 a recipient earns above $6,000 per year. (The limit is $4,400 for those under the age of sixty-five.) But the test does not apply to people above the age of seventy-two, and next year the age will fall to seventy. More important, it applies only to EARNINGS, from jobs, and not to investment income or other pensions, which are the major source of income for the most affluent older people.

Yet to tax Social Security would be heresy: that attitude is nearly universal, its emotional and political power a legacy of the era when we could effortlessly afford to treat everyone "equally." As I have tried to suggest, that era has ended, but adjusting our assumptions takes time. The only political proposition more challenging than taxing Social Security would be carrying the logic to its next step, and asking why there should be extra exemptions on Form 1040 for everyone over the age of sixty-five. Since older Americans are, on the whole, economically even with everyone else, why are the exemptions necessary? And since they transfer money to the wealthiest members of the group (those who pay taxes) and do little or nothing for the poorer 50 percent, how can they be defended?

We are understandably reluctant to face seemingly mean-spirited questions such as these. Yet our reluctance helps explain why the entitlements have grown so large, and why we seem so powerless to control them. We are also hindered by a widespread misunderstanding of where most people stand on the economic pyramid, and who should therefore be entitled to help. A family whose income is in the mid-$30,000s is part of the richest 20 percent of all Americans. A family in the mid-$50,000s is in the upper 5 percent. Yet most such people would be horrified to think that they were anything but "middle class," entitled to public help in financing their homes or sending their children to school and to the rewards of a lifetime of work in the form of monthly Social Security checks. As individuals, they undoubtedly deserve help; but if all of us are entitled, where will the money come from?

The entitlements problem has forced such questions upon us. The cost of ignoring them is not simply the obvious--the automatic growth of federal spending--but also the subtle destruction of other public goals. This effect transcends ideology. It matters little whether you want to improve education for all children, or build more aircraft carriers, or lift the burden of taxation from the shoulders of America's entrepreneurs--all causes are in jeopardy as long as more and more of us are "entitled" to support from everyone else.