On April 20, 1983, Republican and Democratic congressional leaders joined President Ronald Reagan at the White House to sign landmark legislation strengthening Social Security. They were addressing a genuine crisis. Beginning in the 1970s, a lethal combination of high inflation and high unemployment had sent benefit payments soaring as payroll tax revenues dwindled.

"Just a few months ago," Reagan noted, "there was legitimate alarm that Social Security would soon run out of money. On both sides of the political aisle, there were dark suspicions that opponents from the other party were more interested in playing politics than in solving the problem." But then, he said, a bipartisan commission found a solution, political leaders of both parties "set aside their passions," and a compromise was reached.

If only that were true today. Now, with the baby boom generation nearing retirement, Social Security needs another adjustment. The mechanics of fixing the program aren't complicated, but the politics of fixing it are. The problem isn't so much insolvency as it is ideology. I served on that 1983 commission, representing then-House Speaker Thomas P. "Tip" O'Neill, the Democrat from Massachusetts, and I know that repairing Social Security is well within our grasp -- all we need is the political will.

Much has been made recently of Reagan's innate optimism, but even the late president would be hard put to bring Democrats and Republicans together today in the "spirit of compassion and commitment" that he invoked 21 years ago. Back then, Reagan reaffirmed "for all time our nation's ironclad commitment to Social Security" and said the bill "struck the best possible balance between the taxes we pay and the benefits paid back." By contrast, the present occupant of the White House has committed himself, if reelected, to begin dismantling the Social Security system by introducing private savings accounts to be financed, in part, by siphoning off some of the payroll tax revenue now committed to Social Security.

Meanwhile, on Capitol Hill there is no discernible interest among Republican leaders in the kind of bipartisan spirit whose essence, as Reagan noted, "is to give up a little in order to get a lot." And the Democrats' strategy is to say "leave the system alone" and let Republicans make political mistakes by proposing changes.

So we appear headed for another standoff. But it doesn't have to happen.

First, let's consider the nature of today's challenge. There's a high probability that Social Security faces a long-term imbalance of revenues and expenses, which, if left unaddressed, could eventually lead to insolvency. But even the experts aren't sure whether the shortfall will be minor (and relatively easy to close) or not so minor.

Indeed they can't be sure. For long-range planning, Social Security's trustees make 75-year forecasts. This is a highly uncertain exercise. Think about a 75-year forecast made in 1929. No matter how astute, it couldn't have foreseen the Depression and World War II, either of which would have skewed any forecast beyond recognition. We need to keep that cautionary tale in mind as we consider the forecasts that drive the Social Security debate today. They're planning aids, not prophecies cast in stone.

The reports released this year by Social Security's trustees and by the Congressional Budget Office (CBO) provide further evidence of this uncertainty. The estimators are nonpartisan professionals using their best judgment. But CBO anticipates a long-term deficit about half as large as the one foreseen by the trustees.

Does it matter which forecast is closer to the mark? Not really. Even the trustees anticipate an eventual shortfall of relatively small size: about 2 percent of payrolls. Reagan and O'Neill could have worked with a gap of that size. We can, too.

Most of the financing for Social Security comes from deductions from workers' earnings and taxes on employers' payrolls. Closing the financing gap means raising tax revenues or lowering benefits in one form or another. But that doesn't mean we have to raise payroll tax rates across the board; nor do we have to do radical surgery that would alter the nature of the program. All it requires are changes that should be made in any event. If these changes produce more deficit reduction than needed, I would urge their adoption anyway and a cut in the payroll tax rate.

I want to propose four such changes:

* Raise the tax ceiling to cover more of what high-income earners make. Right now, workers and employers don't pay Social Security taxes on any individual earnings over $87,900 a year -- limited contributions for limited benefits. That ceiling is adjusted every year. In 1983, Congress set a ceiling that would subject 90 percent of covered earnings to the Social Security tax. Currently, however, only about 85 percent of covered earnings are being taxed (and credited to workers' accounts for purposes of calculating benefits). This is a major reason why Social Security faces a long-range deficit.

This isn't the result of any policy change, but rather the accidental byproduct of the fact that over the past 20 years incomes of those above the ceiling have risen much faster than the incomes of those below. To gradually restore coverage to 90 percent, we should boost the cutoff point by 2 percent more each year than would occur under current law, which automatically raises the ceiling by the amount that average wages increase. By 2036, the goal would be reached.

The contributions of the 6 percent of earners who are paid more than the cutoff amount would go up. They would receive higher benefits, but the system would gain because benefits are weighted toward lower income earners.

This change would reduce the financing gap anticipated by the trustees by an estimated 0.61 percent of payroll. (Remember, we only need 1 or 2 percent of payroll to bring the system into balance.)

* Fix the cost of living adjustment. Social Security benefits are adjusted annually to maintain purchasing power. In determining the amount of the cost of living adjustment (COLA), Social Security uses a consumer price index (CPI) developed many years ago by the Bureau of Labor Statistics. That index measures changes in the price of a representative basket of goods and services. The Bureau, however, has developed a more accurate CPI that takes into account the effect of consumers substituting one type of purchase for another. (If the price of beef goes up, maybe you'll buy chicken.) Adopting this more accurate CPI would produce slightly lower COLA increases.

This would reduce the financing gap by another 0.3 percent of payroll -- bringing us to almost half the 2 percent goal.

* Dedicate a limited estate tax to Social Security. Under present law, the estate tax is being phased out; by 2010, only estates valued at $3.5 million or more will be taxed. If the tax is kept in place at the 2010 level (rather than being completely eliminated, as President Bush proposes), the revenues, if earmarked for Social Security, would reduce the Social Security financing gap by another 0.6 percent of payroll.

This would mark a departure from traditional Social Security financing. And with the nation facing record-breaking deficits, we can't count on general tax revenues to cover the Social Security shortfall. The only credible long-term solution is the addition of a dedicated tax.

You could earmark a lot of things, but this one makes sense. Social Security is saddled with what economists Peter Diamond and Peter Orszag have dubbed a "legacy debt" from paying substantial benefits to the first generations of covered workers, even though they and their employers had made very limited contributions. Paying those benefits was sound social policy; it helped lift millions of the elderly out of poverty, without raising payroll taxes to impossible levels or waiting for decades to make meaningful payments.

But now the legacy debt forms a big part of the projected 75-year deficit. There is no good reason why future earners should have to pay the entire legacy debt.

If the CBO's forecast is correct, these three changes would produce a surplus for Social Security. If the Social Security trustees' more pessimistic estimates are closer, we would still need to close a gap of about 0.5 percent of payroll.

* Create a failsafe mechanism. We can meet this residual gap, if it exists, by instituting a flexible payroll tax rate that would take effect later. Even before taking effect, it would be adjusted as forecasts change. This "balancing rate," as I call it, would quantify the shortfall and represent a commitment to filling it. This failsafe provision would be activated automatically if Congress neglected to adjust revenues and expenses to changes in the estimates.

I propose that the balancing rate become effective the year the trust fund would otherwise start to decline. To illustrate: After making the first three changes proposed above, the trust fund would continue to grow until about 2055. To avoid drawing it down at that point and to fund a full 75-year period from now, the program would need to schedule a balancing rate increase, effective in that year, of roughly 0.9 percentage points each on employers and employees. Of course, if the CBO estimate turned out to be accurate, the rate would never have to be increased -- in fact it could be reduced.

In the absence of the failsafe mechanism, the remaining gap in Social Security could be financed for the next 75 years by increasing the contribution rate today by 0.2 percentage points each on employees and employers. But if the program actually were to start collecting this amount now, it might collect more than necessary, overcharging today's workers. My approach -- making prudent adjustments now and providing for a balancing rate some 50 years from now -- reduces the likelihood of over financing while at the same time guarding against underfunding.

This strategy has the advantage of ending false alarms about Social Security's finances. As things stand, the trustees' annual release of their latest 75-year estimates becomes a sky-is-falling media circus. The middle-range estimates are reported as if they were meant to be exact and immutable rather than as one of many possibilities. What is really the trustees' tentative expectation of a possible distant shortfall is transformed into a looming "crisis," prompting millions of Americans to believe, wrongly, that Social Security is doomed unless it gets a radical overhaul.

Social Security doesn't require radical "reforms" because it hasn't failed. The system that has served so many so well for so long just needs some timely maintenance work. Let's get started.

Robert Ball was the commissioner of Social Security from 1962 to 1973 and chief Democratic negotiator on the 1983 commission whose recommendations led to the last major changes in the program. He is author of "Insuring the Essentials" (Century Foundation).