Mr. Chairman and Members of the Forum:
You have asked us to testify about the effects of former Senator and now presidential candidate Robert Dole's budget proposals upon the operations of Medicare.
I hope the members of this forum and the public recognize that in discussing this subject we face some unique difficulties. It is hard enough to predict the budgetary effects of various measures, or to analyze health care policy. But in this case the measures have been explained less clearly than one might hope. And we are also talking directly about politics. The effect of former Senator Dole's budget proposals on Medicare depends not just on what is in the proposals now, but on how you think a President Dole would respond if some of the assumptions behind those proposals, as is likely, turn out to be false.
I agreed to testify because I do have some ability to analyze all these matters. I have published books on both budgeting and health care policy. I am a student of Congress. But I think it's very important to realize that, even more than is usual in congressional hearings, we are talking today about probabilities, about risks, about priorities. We are not talking about certainties.
For reasons that I will explain, I think that discussion is appropriate. We are about to have a presidential election, and elections, I believe, are also about risks and priorities, not certainties.
But I hope you realize that, while we have no choice but to speculate, the job is made difficult by both the uncertainty of the world and the uncertainty created by some aspects of former Senator Dole's explanation of his economic proposals.
Why the subject of this forum is a fair question
We are talking about a proposal, and the proposal states that, "Medicare, Social Security, and Defense programs are "off-the-table. If Medicare is off the table, how can there be any threat to it?
The first answer is, in the Dole plan a large amount of Medicare cuts are "off the table" only in the sense that they have been hidden under the table. His economic plan assumes, as a starting point, the $158 billion in Medicare reductions over six years that were ordered in the Congressional Budget Resolution for Fiscal Year 1997.
Relative to standard projections of the program's costs if it were not changed, this is about a ten percent cut over the six year period, and a sixteen percent cut in the last year, 2002. People can argue about whether this plan is a good or bad idea, but it seems silly to pretend it doesn't exist.
The second reason to discuss Medicare and the Dole campaign's announced "Plan for Economic Growth" is that, in politics as in the rest of life, what people say they want is not the same as how they will make choices. We value family and work, but often have to violate our standards on one for the sake of the other. We love all our children, but when they fight we may have to favor one or the other. In politics, contradictions are especially common, because the life of a country is especially complex. So the voters have to look closely at both major candidates' plans, in order to distinguish not just promises, which may not all be kept, but risks and priorities.
Let me give an example. In 1980, President Reagan promised to do three things: balance the budget, cut taxes dramatically, and build up the military. He would have kept all three promises, if he could. But when his voodoo economics didn't work out, he had to choose. He decided the military and his tax cut were more important than his balanced budget. Voters might have wanted to think in advance about what choice he would make.
Now former Senator Dole promises to balance the budget, cut taxes, and not cut Medicare spending (other than the $158 billion he doesn't want to mention). It seems only fair to look at Mr. Dole's long record in national government and think about how he would choose.
I want to emphasize, as a citizen and student of politics, that I think voters should be looking at the priorities of both major party candidates. I think both President Clinton and former Senator Dole have long records as people who naturally cannot deliver or get everything they want, but have pretty clear priorities. You may not want this mentioned, but President Clinton campaigned for both deficit reduction and a middle-class tax cut. When he came into office, he needed to do more deficit reduction, so the tax cut could not be afforded. He was criticized for breaking a "promise." But President Clinton still followed the basic priorities identified in his campaign. The burden of the 1993 deficit reduction act was placed overwhelmingly on taxpayers in the top five percent of the country, and the bill lowered taxes on those working people who were struggling most.
President Clinton has also made clear, through the course of his administration, that he wants both to reduce the deficit and protect Medicare. He has said, in his own budget proposals, that a certain amount of change in Medicare is acceptable in pursuit of the balanced budget. But, by vetoing the Republicans' 1995 plan, President Clinton also drew a line, where protecting Medicare became more important than balancing the budget.
Former Senator Dole was on the other side of that line. That is one part of the evidence about his priorities, and so what the risks to Medicare would be if he were elected. Nobody can know for sure. But there is certainly a very high risk that, if forced to choose among giving up the tax cut, abandoning budget balance, or cutting more out of Medicare, President Dole and a Republican Congress would choose to cut more out of Medicare—eventually.
* * *
That is why I think it is fair to ask whether former Senator Dole's budget plan would threaten Medicare, even though he has declared explicitly that it would not. To answer the question we have to make three kinds of judgments: about the budget, about politics, and about Medicare.
On the budget, we and the American public have to judge whether the Republican candidate's budget proposals "add up." Is the Dole/Kemp campaign correct in claiming that a balanced budget and its tax cut can be obtained even without further Medicare cuts?
On politics, we have to ask what choices a President Dole and a Republican Congress would make if, indeed, they had to make the tradeoffs that they would prefer to pretend are unnecessary.
Then on Medicare, there are two questions. First, can cuts of the likely size be made without threatening the program's performance in significant ways? Second, is a Dole administration likely to make cuts of any given amount in a way that best protects the program, compared to what a President Clinton would do?
The Budget Numbers
The budget numbers in the Dole plan raise two questions. Are the economic projections in the Dole plan believable? If so, are the spending cuts in the plan in order to balance the budget and pay for the tax cut bearable for the country, never mind for Medicare?
On September 5 you gave a distinguished panel of experts an opportunity to discuss this part of the question. Many analysts have tried to make sense of the Dole plan's rather unspecific numbers. I do not want to get into the details of the economics because they are bigger experts on that subject than I am. So I only want to remind you of some of the reasons for skepticism today.
First, presidential candidate Dole's economic forecast presumes that his policies would have a feedback effect on revenues, that would reduce the loss of income from the tax cut, more than is likely. It is possible to believe, as your panel testified, that there would be some small feedback effect. But the estimate in the Dole plan is excessive. Indeed, back when he was a Senator, all those weeks ago, Dole himself built a record of sensible distrust of those theories. The assumption that these policies would markedly change individuals' levels of investment or work is unjustified. Candidate Dole's economic advisers have tried to set their deviation from standard understandings at a level that is not totally and absolutely absurd. But it is likely that President Dole would need to cut spending by more than he has admitted as a candidate. Or sacrifice some of his tax cut or budget balance.
Second, former Senator Dole's forecast includes a built-in contradiction: he assumes that revenues will be higher than CBO's current projections under current policies, yet that changing policies will increase revenues even more. Not only is the extra baseline revenue questionable, but if it did exist it would refute much of his claim that lower rates will change behavior in a desirable way.
Third, it is obviously difficult to guess what cuts a Dole administration and Dole Congress would make in order to achieve the plan's spending targets. The cuts not only are described very generally but in some cases are labeled in what can only be called a disingenuous manner: the "10% cut in non-defense administrative costs," for example. Richard Kogan's report for the Center on Budget and Policy Priorities, however, seems a reasonable attempt to fit policies to the misleading labels. Former Senator Dole seems to be proposing a 37 percent cut in purchasing power for the entire nondefense discretionary budget between the current fiscal year and fiscal year 2002. Since those programs were cut substantially in the past year, this would be a 43% cut from their level in fiscal year 1995.
In other words, we are asked to assume that cuts similar to those made in 1995 will be made every year. Members of Congress will remember how difficult that was. But, of course, meeting the Dole plan's standards would be much harder than that. First, we already know those cuts won't be made this year, so the annual cuts would have to be worse in coming years. Second, each year's cuts get more painful. It really is like slicing into a body; the deeper you go, the more you hit organs and bone, not just muscle.
Imagine what it would be like if, in 2002, your personal purchasing power were only 63 percent of what it is now. That's a radical, radical change. Is it possible? Sure. People lose their jobs and suffer other financial disasters frequently. But we rarely call it desirable. And we certainly would not plan to reduce our personal consumption by that much without thinking through what we would give up, before deciding that it's a good idea.
The Dole campaign's potential cuts in domestic discretionary programs seem to be about $62 billion larger, over six years, than the cuts that were included in the reconciliation bill of 1995 that President Clinton vetoed. It is thus radical by even 1995's standards.
But does the extremism of the discretionary spending cuts mean that a President Dole, instead, would propose larger Medicare cuts to balance the budget?
That brings us to the political choices that would be made necessary by the flaws in the Dole plan as currently described.
Even after a decade studying and writing extensively about the battles of the budget, I cannot pretend to know for certain what President Dole would do about his own plan's dubious aspects once he took office. I can, however, offer a reasonable guess
Given the chance, politicians muddle through rather than confront contradictions in their preferences. Just like the rest of us.
The new president would face, remember, two basic problems with his campaign position. First, it seems unlikely that a responsible CBO director would accept all of the Dole plan's revenue and savings estimates. There are just too many issues, ranging from the underlying baseline to the supposed economic growth from policy change to the probably exaggerated profits from asset sales. Therefore a revised proposal would have to find some extra savings. There are too many uncertainties to be sure about the scale of the error. But it would not be surprising if CBO said the policy had to find an extra $75-100 billion in spending cuts over the six year period in order to still pay for the tax cut within a balanced budget. The second problem is the irrationality of the assumed discretionary domestic spending cuts.
The first question is whether President Dole would back off on either the tax cut or balanced budget goal. Either strikes me as very unlikely.
Look first at the Dole campaign's stated positions. It offers "six key points" in an economic plan. The first three are a balanced budget amendment, a balanced budget by the year 2002, and the tax cut. Protecting Medicare is clearly subordinated to the need to create growth from this economic policy.
Consider next the politics. Without a Republican President, Congress in 1995, under Senator Dole's leadership, passed a reconciliation bill with spending cuts about as large as would be required in 1997. It is hard to see why they could not do the same in 1997. If the Republicans win the Presidency in this election it is hard to imagine that Congress would become less conservative. Republicans would surely retain their majority, and be convinced they had a mandate for their policies.
Moreover, Senator Dole, one of the longtime leaders of the tax-cut skeptics among Republicans, would have switched sides. As President Dole, it would be extremely difficult for him to abandon the tax cut. He would be seen as selling out the majority of his own party, going back on his major campaign promise. And why? In order to reduce a projected, vague, unspecific set of cuts to discretionary domestic spending programs? In order to accept economic estimates made by the triumphant coalition's supposed "enemies"?
Absent a recession—meaning assuming that President Clinton's moderately good economy persists—it seems foolish to assume that President Dole would cripple his presidency by reversing field and failing to adopt the balanced budget plan and tax cut that he will almost surely have the votes in Congress to pass.
Would President Dole, however, pay for budget balance and tax cuts by proposing larger Medicare cuts than he has put under the table now, in order to reduce the planned cuts in domestic discretionary spending?
I'm afraid that I cannot agree with people who assume the discretionary cuts are so stupid that President Dole would immediately, in 1997, limit them in order to cut entitlements.
Our politicians have a pretty strong record of preferring prospective, unspecific discretionary cuts to actual, specific entitlement reductions. Discretionary "caps" were the largest spending reductions in the 1990, 1993, and 1995 deficit reduction legislation. Gramm-Rudman promised massive discretionary cuts and hardly any in entitlements. Some sponsors of the legislation hoped the threat of ridiculous discretionary cuts would force Congress to cut entitlements, but that theory did not come true in any significant way.
So, if he got very lucky on the economic estimates, it seems likely that in 1997, President Dole would not increase his planned Medicare cuts. The Republican president and Congress would enact the planned "caps" on discretionary appropriations, and postpone facing the consequences for a short while.
If he had to find another $75 billion to $100 billion, though, he might be tempted to increase Medicare's share of the pain beyond $158 billion. And it would be almost impossible for him not to seek further savings in Medicaid.
Sometime around 1999, President Dole might have to confront the facts of the discretionary caps. They would be turning into appropriations bills, for which he would be responsible. At that point he and a Republican Congress would have to decide whether to compensate for scaling back the discretionary spending cuts by raising taxes, abandoning the balanced budget, or cutting Medicare and other entitlements further.
Again, I do not think one can simply assume that they would cut Medicare. But that seems more likely than the alternatives.
When the crunch came in 1999 or so, the majority of the Republican party would be very unlikely to believe that the problem was their tax cuts. They don't seem capable of recognizing that the Reagan tax cuts did not balance the budget even after fifteen years; why should they learn after two?
Moreover, it is clear from how they voted in 1995 that a majority of the party believe Medicare should be cut more than former-Senator Dole is currently admitting to proposing. The pattern of revolutionaries is to presume that if results are disappointing they must forge further ahead, not retreat.
And they will be strongly supported, in any efforts to cut Medicare, by the dominant faction in the establishment press and academy that insists that Medicare's longterm prospects constitute a "crisis" that requires large program cuts. Even some people who have testified to this forum, and some of its member, appear to believe that. I don't, but as an analyst I have to recognize the political power of the forces that insist Medicare must be cut. Given the choice between enlisting that support for "controlling entitlements," or selling out their own most fervently-held positions on tax cuts and the balanced budget, it seems far more likely that a Republican government would take the political risks of slashing Medicare.
Perhaps most important, it is possible that the Dole administration would already have taken steps, as part of its 1997 legislation, that would make Medicare cuts politically easier in the future.
These steps are very likely to be bad policy. But they are at the heart of the Republican Medicare agenda.
How Dole Might Cut Medicare
It is commonly asserted that the difference between President Clinton and congressional Republicans on Medicare is quite small. After all, President Clinton has endorsed some measures to save money in Medicare. It is widely agreed that the Hospital Insurance trust fund must be put in better condition, though that is partly a technical matter. Under the present estimates of the economy and program trends, President Clinton's proposed Medicare savings are somewhere around three-quarters of Congress' stated goal.
But this is an instance where the policy difference is much, much bigger than the difference in the numbers. Let me try to explain why.
The Republicans have reduced their proposed cuts in Medicare, responding temporarily to public disapproval, in two ways. First, they dropped their policy on Part B premiums. Whatever you think of that issue, the premiums do not in fact shape how the program operates. Second, they took advantage of changes in underlying economic and technical estimates of program costs. In essence, since the program is estimated to grow less quickly now than at the beginning of 1995, they can claim to be cutting less in order to meet much the same targets as before.
I want to emphasize that that is perfectly legitimate. After all, President Clinton all along disagreed slightly with CBO's estimates of the budget trend on Medicare as well as the general economy.
But it does mean that the important question about the Republican policies on Medicare is not how much they claim to "save" but how they propose to meet their targets.
Neither former Senator Dole nor his former colleagues have done much to answer that question at the moment. The campaign documents, as I mentioned before, say only that the FY97 budget resolution's savings are part of Dole's plan—without even mentioning that that includes $158 billion in Medicare cuts. And within Congress, his former colleagues have tried very hard not to say how they would meet their own targets.
So again, we have to talk about risks and priorities. The best evidence about how the Republicans would try to meet their targets is what they have proposed before, way back in 1995, when they had to write legislation. They delayed doing so as long as possible, waiting until this time last year. But they did, finally, draft specific proposals to cut Medicare.
I found those proposals to be extremely dangerous to the program and many of its beneficiaries. The Republican approach had three core aspects.
First, they wanted to offer vouchers with which people could purchase a wide variety of forms of health payment coverage, rather than just stay in the existing main Medicare system. These included not only HMOs but other insurers and even combinations of catastrophic coverage with Medical Savings Accounts.
Second, they then set legal caps on Medicare spending. The voucher amounts, then, would in essence be set in advance, regardless of developments in actual costs.
Third, they made the vouchers voluntary, claiming that people who wanted to stay in traditional Medicare could do so. But they required that spending in that program also stay within the caps. If it did not, then payments for providers within traditional Medicare would be cut however much was necessary in order to meet the targets.
As with everything else, one cannot be certain how this approach would have worked. But the risks are substantial.
1) Vouchers require a market in which insurers compete to sell their products and individuals choose among them. But that means that both sides will try to improve their situations, the insurers by avoiding enrollees who are more likely to be sick, and the individuals by choosing among insurers' somewhat different packages in order to maximize their own well-being.
This, of course, is the risk-selection and adverse-selection problem with which we are familiar in the current Medicare program. It is why the current option for people to join HMOs in Medicare seems to increase program costs. But in the GOP version, the effects should be significantly stronger. First, there would be many more choices, including essentially unmanaged care and medical savings accounts. The incentives for medical savings accounts in particular would encourage adverse selection, with healthier people choosing the MSAs. But even inclusion of the less-managed-care plans should increase the effect of risk-selection on the Medicare system. The smaller insurers who do not extensively manage care tend already to emphasize risk selection in their underwriting—it is their main way to control costs.
There are many people, not all conservative, who would like to find some way to make vouchers work for Medicare. Notions of some sort of competition among plans within a guarantee of coverage are common around the world—I coedited a book on the subject. But nobody yet has solved the risk-adjustment problem. And, where such options exist, marketing tends to be controlled more strictly than seemed to be contemplated in the Republican 1995 proposals. You need a heavy structure of regulation to make such competition safe, whether for the whole population or just the elderly. Given the fervor with which the Republicans criticized the Clinton plan's effort to specify that regulation in 1993, and the absence of significant legislative proposals from the Republicans to date, I think there is reason to worry that, in implementing vouchers, the Republicans would create a situation in which the healthier members of the Medicare population would mainly be buying coverage and the less healthy members would be more likely to be stuck in the traditional fee-for-service plan.
2) There are many ways to reduce increases in medical spending. Most effective ways involve some sort of hard cap on particular categories of spending. In that sense, setting targets for vouchers in advance is not bad policy: we do something of the sort already in Part B of Medicare, for physician services costs, and other countries do so more strictly.
Yet you cannot control the input of money into the health care system without worrying about how that will affect the supply of services. I have stated many times that the trend of spending observed in the early 1990s was unnecessary and could be reduced. I believe that certainly, in isolation, the system could adjust to the trends contemplated by President Clinton's Medicare plan; it might even be able to adjust to the amount of restraint specified in the Republican budget resolution, though I'm less sure of that.
But a voucher approach presumes that more money will be taken out of health care than Medicare itself will save. Vouchers require marketing and plan development and care management and thus a series of extra expenses. So in order to save $158 billion through the kinds of measures President Clinton has suggested now, and that Congress and the President have agreed on in the past, you only have to take $158 billion from hospitals, doctors, and other providers. But in order to save $158 billion from a voucher system you need $158 billion plus the billions more in added costs of administration, marketing, and profits.
I think that's particularly important in the current market. If current cost trends in the private arena can be sustained, that is good news for Medicare in one way. It clearly would be easier for Medicare to reduce its fees compared to current law. The odds that any provider would stop selling to the program would be low because the difference between Medicare and average private fees still would not widen. Indeed, it has narrowed in the past year.
The bad news is, the combination of a private squeeze and a Medicare squeeze may threaten some urban hospitals and other essential providers. Managed care plans seem to want to selectively contract with hospitals that have lower costs. There is a real danger that a voucherized Medicare would aggravate that trend.
So there are at least four ways in which the Clinton Medicare proposals threaten the underlying supply of services less than a Republican voucher approach probably would. First, the admitted Republican target is about $40 billion more to start with. Second, they would need to take even more extra from providers in order to pay the extra costs of the voucher system. Third, voucher providers would themselves tend to divert business from the essential providers. And fourth, the Republican savings target might be more than $158 billion.
3) The Republican proposals in 1995 had a further, particularly perverse, twist. The authors had to admit that methods of managed care were unlikely to save enough to meet the targets. Therefore they needed a "backup" system of savings. This is exactly the situation that the Clinton administration faced in 1993, and, in spite of their criticisms of "regulation," the Republicans in 1995 adopted essentially the same fix. They set up a system of fee reductions to ensure that the necessary savings were realized.
But in the Clinton plan, at least the cuts would have been imposed only on plans whose costs were above the target. The GOP version focused, instead, on the remaining, old-style, fee-for-service Medicare. So, for example, if healthier people chose the voucher plans, raising the costs in traditional Medicare, fees in traditional Medicare would be cut!
Instead of rewarding efficiency and punishing inefficiency in medical care delivery, this system rewards risk selection. Providers of care to traditional Medicare patients would get whacked for no fault of their own. Trends of costs for comparable patients could go down, and the old system would still get whacked.
It is possible to imagine that such results would be prevented by a really good risk-adjustment scheme. But so far one can only imagine that, just as we can imagine a trip to other galaxies. We don't know how to do it. And I have to emphasize that a party which has made reducing regulation one of its major political goals is highly unlikely to provide the kind of stringent regulation of a voucher system that would be required to even begin to meet the selection problems.
4) Once vouchers are created, however, any of these effects would only occur slowly. By 1999 or 2000 they might have begun, but it will be possible to raise many doubts about how severe they will be. It will also be possible to suggest "fixes", such as new risk-adjustment schemes. Politically, new risk-adjustment schemes don't have to work; they just have to show that the government is trying to respond to the problem, and get it through the next election.
So I do not think the difficulties would force a Dole administration to abandon its vouchers. And I do think vouchers provide a real temptation for further cuts. After all, in a voucher system, it is the plans, not the government, that have to find a way to hit the targets. It will be very easy to claim that the marketplace could do better. All government budgeters love systems that allow them to blame somebody else for cost control. What vouchers would do is make Medicare more like discretionary spending. The government could set general targets and avoid specifying how they will be met, just as today all parties set longterm targets for discretionary domestic cuts without saying how they will be met. I think it is likely that, facing that temptation, in a situation where even Republican programs were gravely endangered by domestic spending caps, a President Dole would support lowering of voucher amounts.
Conclusion I have attempted in this paper to provide a responsible political and policy analysis of a topic on which it is impossible to be precise: the effects on Medicare of former Senator Dole's economic program.
We are talking only about probabilities. If the improbable happens, President Dole might only have to cut all programs by as much as he has already proposed in order to balance the budget. Then we would be talking about $158 billion in Medicare cuts. But it seems more likely that we would be talking about more, either because the deficit targets aren't hit or a Dole administration eventually goes back to Medicare in order to reduce the pain of discretionary domestic spending reductions.
Probabilities involve politics as well as programs. In a world in which people did not have records or commitments, one could imagine that a President Dole and a Republican Congress would cut defense more than planned if the budget deficit were larger than they expected, or even raise taxes. But that seems less likely, based on their records and how politics work, than that they would turn to Medicare for more savings.
The best predictors of politicians' priorities are what they have done before and how fervently they commit themselves to a goal. Even that is nowhere near perfect. George Bush promised "no new taxes." If there is a recession, even former Senator Dole is unlikely to balance the budget. Maybe a President Dole would go back on his tax cut promise.
But for now, I have to look at the numbers, at the campaign statements, and at what then-Senator Dole and his colleagues supported in 1995, and I have to conclude that the risks to Medicare from candidate Dole's economic program are substantial.