All the talk of bulging budget surpluses and an expanded Medicare obscure a bigger story that is overlooked by the press, excluded from political debate and ignored by Washington's think tanks. This is the slow conversion of the federal government into an agency for subsidizing America's retired. Consider: In 1950, Social Security was less than 2 percent of federal spending (there was no Medicare health insurance for the elderly). By 1998, Social Security, Medicare and other spending on the elderly exceeded a third of the budget. By 2020, the costs of the baby boom could push that toward 60 percent.

Is this good for us? Almost no one is asking.

It is hard to criticize big budget surpluses, which, by the White House's latest reckoning, could total $5.9 trillion over the next 15 years. Let us count the blessings. Even smaller surpluses would allow complete repayment of the publicly held federal debt (about $3.7 trillion). This would ease the costs of a retiring baby boom by erasing the 13 percent of spending that goes to interest payments. But the potential benefits of surpluses are so obvious that they obscure many possible dangers.

Five years ago, the people now predicting continuous surpluses predicted continuous deficits. Since February, the White House has raised its estimate of the 15-year surplus by an astonishing $1.1 trillion. There were small shifts in economic assumptions (for example, economic growth edged up a few tenths of a percent annually). All are defensible. Their huge effect on the projected surpluses underlines the flimsiness of the exercise.

The predictions are fictions not because the people making them are dishonest or incompetent, but because no one can see that far into the future. The main reasons the budget has swung into surplus are that the economy outperformed forecasts and that tax revenues exceeded estimates by about 2 percent of national income -- equal now to roughly $170 billion a year. These were pleasant surprises. If such surprises transformed the budget outlook in five years, what surprises (for good or ill) lurk over the next 15? No one knows.

The trouble is that President Clinton and congressional leaders are treating these projected surpluses as if they already exist: a treasure to be distributed. The effect of this seductive approach is to relieve pressure to face the cost of retiring baby boomers, which is one aspect of the future we can anticipate. In 1998 there were 9 million Americans between 35 and 54. We can crudely estimate their retirement benefits. The Clinton administration itself is projecting that these will account for almost 60 percent of federal spending in 2020. Sooner or later, the costs could overwhelm budget surpluses.

They also make the federal government a vast transfer system. The working population is taxed and its income bestowed on retirees. The justification has been the belief that most of the elderly are poor and feeble. But things have changed since the creation of Social Security (1935) and Medicare (1965). People save for retirement. The elderly are healthier. The National Institute on Aging reports a "dramatic . . . reduction in rates of disability among older persons."

These changes pose profound moral, social and political questions. How much should society tax younger (and sometimes poorer) people to subsidize older (and sometimes wealthier) people? Because the elderly's health has improved, should eligibility ages for Social Security and Medicare be raised? Should benefits for the wealthier elderly be reduced? Will the growing costs of federal retirement programs ultimately crowd out spending for other important national goals (defense, research, education)?

Clinton ignores these questions. His pleas to "save Social Security" and "save Medicare" mean: preserve all benefits. This amounts to feathering the retirement of the baby boom at the expense of its children. But given the voting power of the elderly, the political appeal is obvious. The proposal to extend Medicare coverage to prescription drugs is more of the same. In isolation, a strong case can be made for this, because about 35 percent of the elderly have no drug coverage under other insurance. But no case can be made for a drug benefit -- which, once established, might be expanded -- without a larger debate about curbing the growth of federal retirement programs.

And that won't happen soon. No effective voice speaks for a fundamentally different point of view. Republicans are terrified of being branded as anti-elderly (and Democrats pray that they will). The press focuses on political skirmishing and program details. Think tanks churn out technical studies on Social Security and Medicare, but from the Brookings Institution to the Heritage Foundation, they don't address the central issue of the government's evolution into a welfare agency for the retired.

So the looming debate on budget surpluses is not the debate the country needs. In some ways, the differences between the White House and the Republicans are smaller than they seem. Both would devote much of the surpluses to repaying federal debt. Good. But both propose big commitments for the rest -- and that's bad. The Republicans speak of tax cuts up to $1 trillion over a decade. Clinton proposes many big programs (these costs are for 15 years): $540 billion to finance "universal savings accounts" for all Americans; $543 billion to buy common stock for the Social Security system; and $156 billion for a "child and education" trust fund.

Some fail on their own demerits. At the moment, the economy doesn't need a tax cut. Having government buy stocks for Social Security is fraught with problems. But all these schemes share a larger defect. They make long-term commitments with imagined money. If budget surpluses don't materialize, the commitments would remain. The best outcome would be deadlock -- neither Clinton nor the Republicans would prevail. The surpluses would then quietly reduce the federal debt, which, for now, is their best use.