Costs of the Tax Cut and of a Medicare Prescription Drug Benefit, 6/14/02

This issue brief compares the cost of last year's tax cuts — and especially the costs of the tax cuts provided to the top one percent of the population — with the costs of two possible prescription drug plans for the Medicare population. The first such plan would cover, on average, one-quarter of the prescription drug costs of the Medicare population. The second plan would be like the first except that it would cover, on average, half of the Medicare population's drug costs.

Such prescription drug plans would entail significant expense, and the plan covering half the cost of prescription drugs — at a cost of approximately $700 billion over ten years — would be twice as expensive as the plan covering one-quarter of drug costs. Yet the cost of the tax cut is substantially larger than the cost of either of these prescription drug proposals. When the tax cut is fully in effect, the cost of the tax cut just for the top one percent of the population would exceed the entire difference in cost between the two prescription drug proposals.⁽¹⁾

Decades ago, people with serious medical problems might be treated through surgery or through medicine that could cure them, if a cure existed. Nowadays, however, drugs can address some conditions that surgery cannot, can allow people to live for years or decades without surgery, or can forestall certain medical problems for years. But many of the new drugs are expensive and need to be taken indefinitely. Those drugs are especially important to people eligible for Medicare — the elderly and people with disabilities — who often cannot easily afford them. The Medicare program, however, does not cover the cost of outpatient prescription drugs, since they were a far less important part of medical care in 1965 when Medicare was enacted, and so were not included in the program.

This leaves current policy-makers with a conundrum. The more that modern medicine relies upon prescription drugs and the more necessary and effective those drugs become, the more costly it becomes to provide a prescription drug benefit. At the same time, the more necessary and effective that prescription drugs become, the more important it is to include prescription drug coverage as part of any health insurance package, public or private, including Medicare.

This analysis is not based on cost estimates of specific legislative proposals for prescription drug benefits. Rather, for illustrative purposes, the analysis considers two general Medicare prescription drug plans. The first plan would use federal resources to cover 25 percent of the costs of prescription drugs for the Medicare population, while the second would use federal resources to cover 50 percent of the costs of prescription drugs for this population. Once these plans are fully effective, the second plan would cost the federal government about twice as much as the first. Assuming these alternative plans are phased in over three or more years, the first plan might cost $350 billion over the ten-year period through 2012 and the second, $700 billion over the same period. The costs assumed in this analysis thus are roughly consistent with the cost ranges for prescription drug plans currently under discussion in Congress.

**The Cost of Two Possible Prescription Drug Benefits and Last Year's Tax Cut**
in billions of constant 2002 dollars
	if fully in effect in 2003		when fully in effect, in 2012		in 2020
	dollars	% of GDP	dollars	% of GDP	dollars	% of GDP
"25%" prescription drug benefit	22	0.2%	53	0.4%	92	0.6%
"50%" prescription drug benefit	44	0.4%	105	0.8%	184	1.2%
last year's tax cut*	187	1.8%	235	1.8%	273	1.8%
last year's tax cut for top 1%*	69	0.7%	86	0.7%	100	0.7%
NOTE: dollar figures are adjusted for inflation so that they are comparable over time. * Assuming the tax cut is made permanent

The above table shows the annual costs of these two general prescription drug plans under three scenarios: as if they were fully effective in 2003; in 2012, assuming the plans are fully in effect by then; and in 2020, again assuming the plans are fully effective. These costs are expressed in constant 2002 dollars, which means they are adjusted for inflation.

We estimated the costs of these two prescription drug plans by taking 25 percent and 50 percent, respectively, of the national costs of prescription drugs for the Medicare population. The Congressional Budget Office recently projected these national costs for each year through 2012, allowing us to calculate directly the costs of our two assumed plans in 2003 and 2012 if the plans were fully in effect in those years. (No plan could be implemented quickly enough to be fully effective in 2003; these figures are provided for illustrative purposes.)

To project the national costs of prescription drugs for the Medicare population beyond 2012, we relied on long-term projections made by the Medicare actuaries at the Department of Health and Human Services. The actuaries' projections incorporate the expected growth of the Medicare population and the general growth of medical prices. We based our projections of national prescription drug costs on CBO's estimate of those costs in 2012; we assumed that after 2012, national drug costs would grow somewhat faster than overall Medicare costs. Prescription drug costs have grown faster than Medicare costs in the past and CBO assumes they will continue to do so through 2012. Our assumptions lead to projections of national drug costs in years after 2012 that are generally consistent with public statements about such costs made by HHS officials. See the Appendix for a more complete discussion of our projections.

The table shows that the cost of either of the two prescription drug plans initially is relatively modest compared with other budget costs but grows to become more substantial as time passes. For instance, the cost of the plan covering 50 percent of prescription drug expenses for the Medicare population would (if fully effective in 2003) have about the same cost that year as the increase in defense funding proposed by President Bush for 2003 or the cost in that year of the recently enacted "stimulus" bill. By 2020, however, the cost of both of the prescription drug benefit proposals would be noticeably higher.

Can Congress restrain the cost growth of prescription drugs shown in the table? Congress could, of course, provide a smaller drug benefit. But the smaller the benefit, the more the Medicare population may face difficulties as the price of prescription drugs continues to rise. As discussed in the box on the next page, we as a nation cannot reasonably choose to have a substantial share of elderly or disabled Medicare beneficiaries forgo pharmaceuticals that are increasingly necessary to their health, and even to their remaining alive.

Drinking water and prescription drugs: an analogy

Clean, potable water is a necessity of life and health. Currently, most water is provided by local or regional governments at reasonable rates. If water becomes extremely expensive, however, we would have no choice but to continue providing it to all, even though this would require increased fees or taxes. Water is not like opera; if opera becomes so expensive that only the very well off can afford it, society can nonetheless continue to function. Water is not even like space exploration, which must be financed publicly, not privately, but could be abandoned if its costs became prohibitive.

In two respects, medicine is like water. To begin with, both are provided more efficiently through group purchasing.* Second, as medicine (or water) becomes more expensive, we cannot in good conscience accept that it will be available only to the well off. At the same time, we cannot wish away the growing costs.

* Water is provided more efficiently through group purchasing because the infrastructure needed to distribute water implies a natural monopoly. Prescription drugs are provided more efficiently through group purchasing for a different reason — medicine is best financed through public or private insurance pools. The costs of a serious illness can exceed the resources of many or most of the population, but a pool spreads these costs to make them more affordable for each participant, and each participant would rather be insured than not.

One conclusion is that, as part of a Medicare prescription drug benefit, the government ought to use its purchasing power to negotiate lower prices, foster competition to reduce drug prices, or reduce costs through other means. The Center for Policy Alternatives reports that "[t]he exact same name-brand medicine and dosage sold at retail to an uninsured person for $100 will, on average, be sold to an HMO or the Medicaid programs for $65, and to the U.S. Departments of Defense or Veterans Affairs for $46."⁽²⁾ The estimates provided in this brief analysis do not assume that the prescription drug plans would squeeze prescription drug prices. For that reason, the costs given here for the prescription drug plans may be somewhat overstated.

Note that, under either of the drug plans discussed in this analysis, beneficiaries would have to pay a sizeable portion of prescription drug costs out of pocket.

The table on page 2 shows the cost of last year's tax cut over the same three periods: in 2003 as though it were fully effective in that year; in 2012, when fully phased in; and in 2020, assuming all provisions of the tax cut are extended. The table compares the costs of the tax cut with the costs of the two assumed prescription drug benefits.

In making our estimates of the cost of the tax cut, we start with the estimates of its cost that CBO published this January. We add to those costs CBO's estimate of the cost of extending provisions in the tax cut law that are scheduled to expire after 2010, as well as CBO's estimate of the cost of extending provisions with expiration dates before 2010.

We also include the costs of a partial "fix" to the individual Alternative Minimum Tax. The AMT was originally designed to ensure that well-off taxpayers with multiple tax exemptions or deductions would still pay their fair share of taxes. Because of design flaws, the AMT will grow to hit tens of millions of middle-class taxpayers by the end of this decade; last year's tax cut exacerbated the situation. In our projection of the cost of the tax cut, we assume that Congress will address the AMT in such a way that last year's tax cut does not make the scope of the AMT grow more than it would have without the tax cut. We use the Joint Committee on Taxation's estimate of the costs of such a "hold harmless" policy. Last year's tax cut includes a temporary provision that has this "hold harmless" effect, but that provision expires after 2004.

Finally, we take the resulting estimates from CBO and the Joint Tax Committee for the costs of the tax cut in 2011 and extend them into subsequent years by assuming the cost of the tax cut will remain constant as a share of GDP after 2011.⁽³⁾ This is the standard approach taken by CBO, OMB, GAO, and others in projecting revenues over the long term. (In the case of the tax cut, this approach is more likely to understate than to overstate costs.) These costs are shown in the table both in 2002 dollars (i.e., as adjusted for inflation so the amounts are comparable over time) and as a percentage of GDP.

The table on page 2 also shows the costs of last year's tax cut for the top one percent of the U.S. population. In a recent paper that is the most rigorous and comprehensive analysis available of the effects of last year's tax cut, William Gale and Samara Potter of the Brookings Institution estimate that when the tax cut is fully effective, the tax cuts going to the top one percent of the population will constitute 36.7 percent of the annual costs of the legislation. Thus, the overall costs of the tax cut can be multiplied by 36.7 percent to derive an estimate of the amount going to the top one percent.⁽⁴⁾

One way to see what it means for the top one percent of the population to receive 36.7 percent of the tax cuts is as follows: if the United States had 100 residents, for every $10 in tax cuts going to the individual with the highest income, the remaining 99 persons would share a total of $27 in tax cuts and thus receive an average of about 27 cents each.

The average income for households in the top one percent exceeds $1 million per year. Cutting the top marginal tax rate from 39.6 percent to 35 percent, as the tax cut ultimately does, would be less costly if the incomes of the top one percent of the population were smaller than they actually are. (If people at the top had smaller incomes, the reduction in the top tax rate would provide them a tax cut equaling 4.6 percent of a smaller amount of money.) It is precisely because the incomes of those in the top one percent are so high that tax cuts for the well-off — provided through cuts in the top marginal tax rate, the estate tax, and other provisions of the tax code — provide these individuals with so much additional after-tax income and are so costly.

As can be seen, the difference in costs between the two prescription drug benefits examined here is less than the cost of the tax cut for the top one percent of the U.S. population. While both the tax cut and a prescription drug benefit are ultimately quite costly and a prescription drug benefit will generally grow in cost more rapidly than the tax cut, it is difficult to argue that the tax cut is modest in cost and easily affordable but a prescription drug benefit is not.

Moreover, it is fair to conclude that the more the tax cuts for the top one percent cost, the less they are needed by persons in that income group; as just explained, the tax cuts for the top one percent are very costly precisely because the top one percent is so well off. The extra income the tax cuts provide to the top one percent will not appreciably improve the well-being of these individuals. (The evidence also suggests that these tax cuts will not significantly improve economic growth. To the contrary, the Brookings study by Gale and Potter concludes the tax cuts are more likely to have a small negative effect on growth than a positive effect.⁽⁵⁾)

For prescription drugs, the opposite is the case. Prescription drugs are expected to grow more costly over time in part because of new scientific breakthroughs that will provide superior treatments, improve health, and prolong life, but at inevitably greater cost. To the extent this is the case, the more that the costs of a prescription drug benefit increase over time, the more those costs are likely to represent essential medicine. Furthermore, the more that prescription drug prices rise for any reason — not just because of medical breakthroughs — the less affordable they will become for Medicare beneficiaries with low and moderate incomes, and the more a prescription drug benefit will be needed to protect the health of a large number of elderly or disabled people.

In making the projections of prescription drug costs used in this analysis, we relied on the most recent projection that the Congressional Budget Office has issued of national spending on prescription drugs for the Medicare population over the next ten years (2003 through 2012). CBO does not project national expenditures on prescription drugs after 2012. We assume that in years after 2012, the cost of prescriptions drugs per Medicare beneficiary will grow almost two percent per year faster than the cost of Medicare physicians insurance (Medicare Part B) per Medicare beneficiary. Because projected Medicare Part B costs beyond 2012 are published by the actuaries of the Department of Health and Human Services, we can, in this manner, estimate prescription drug costs for years after 2012.

By tying our cost projections to these HHS projections, we take into account the coming retirement of the baby boom generation, the projected growth of the economy, and the projected increases in medical prices generally. As just mentioned, we assume that the costs of prescription drugs will grow faster than the cost of Medicare Part B; this has been the case for some time and is projected by CBO to be the case for the next ten years. The estimate that the per-person costs of prescription drugs will grow almost 2 percent per year faster than the per-person costs of Medicare Part B produces long-term estimates of the costs of prescription drugs for the Medicare population that are broadly consistent with the long-term estimates the Administration has used when discussing the long-term costs of a prescription drug benefit.

1. This analysis is not intended to suggest that if Congress were to scale back elements of last year's tax cut that have not yet taken effect, the resulting savings should be used exclusively to provide a more adequate Medicare prescription drug benefit. There are a number of competing needs that could be considered if such resources became available. To name just a few: there is a need to increase budgetary savings — to reduce debt — in anticipation of the baby boomers' retirement; there are approximately 39 million Americans with no health insurance; child poverty remains at higher levels in the United States than in Canada or most of western Europe; there is a particularly pressing need for funds to combat health epidemics and rampant and extraordinarily severe poverty in parts of the developing world; the shortage of affordable housing in the U.S. remains at near record levels; and some funds from the rest of the budget are virtually certain to be needed as part as part of any Social Security solvency proposal that could have a chance of passing.

2. Playing Fair: State Action to Lower Prescription Drug Prices, Center for Policy Alternatives, June 2000. Similarly, the Lewin Group concluded that if uninsured persons "had access to private market discounts, the medicines that need, on average, would cost 30% to 39% less." Price Discounting Practices for Pharmaceuticals in the United States, April 2000.

4. See William G. Gale and Samara R. Potter, "An Economic Evaluation of the Economic Growth and Tax Relief Reconciliation Act of 2001," National Tax Journal, forthcoming. Available on the Brookings Institution website at http://www.brookings.org/views/articles/gale/200203.htm The estimate of 36.7 percent for the share of the tax cut enjoyed by the top one percent of the population when the bill is fully effective is reinforced by a very similar estimate made by Citizens for Tax Justice.