Still a Pain:
The Cost of Prescription Drugs to
Seniors With the Medicare Drug Benefit

Dean Baker[1]

December 4, 2003

            After extensive debate, Congress finally passed a prescription drug benefit just before recessing for Thanksgiving. While the political implications of this bill are still being debated, the numbers clearly indicate that this bill will not end the problems of seniors who have to pay for prescription drugs. Even with the new benefit in place, seniors will be paying considerably more for prescription drugs when the bill takes effect in 2006 than they did when it first became a campaign issue in the 2000 election.

            In fact, under the new plan, seniors in the middle income quintile will pay an average of $1,650 a year in out-of-pocket expenses for prescription drugs in 2006. This figure is nearly 60 percent more than they paid in 2000, even after adjusting for inflation. Expenses are projected to continue to rise so that by 2013 middle-income seniors will be paying more than two and a half times as much for prescription drugs (adjusting for inflation) as they did in 2000.

            Table 1 shows projected spending for prescription drugs by income quintile for the years 2004 through 2013, compared with spending in 2000. These projections are derived from the Congressional Budget Office’s projections for average prescription drug spending, along with data from the Bureau of Labor Statistics’ consumer expenditure survey for the distribution of spending by income quintile. The numbers assume that every person buys the insurance plan provided by the new bill.

Table 1
Average Annual Spending by Elderly Households
on Prescription-Drugs by Income Quintile
(year 2000 dollars) 

Source: Author’s calculations and Baker and Schmitt, 2003: see appendix.

            The increase in spending also shows up as a sharply higher share of after-tax income being spent on prescription drugs. The share of after-tax income that will be spent on prescription drugs, for a household in the middle quintile, is projected to rise from 6.1 percent in 2000, to 9.0 percent when the insurance first comes into existence in 2006, to 13.7 percent in 2013 at the end of the current projection period. Table 2 shows projections of the share of after-tax income that will be spent on prescription drugs by income quintile, assuming that all elderly households purchase the insurance provided for by Congress.

Table 2

Shares of Elderly After-Tax Income Used for Prescription Drugs, by Quintile 

Source: Author’s calculations and Baker and Schmitt, 2003, see appendix.

                Table 2 shows that the share of after-tax income being spent on prescription drugs will rise sharply for all income quintiles over the period from 2000 to 2013, even with the prescription drug benefit in place. The share of after-tax income for households in the bottom quintile that is spent on prescription drugs is projected to rise from 9.4 percent in 2000, to 13.5 percent when the benefit first takes effect in 2006. It is projected to rise further to 20.8 percent by the end of the projection period in 2013.

             The data in table 2 clearly show that paying for prescription drugs will be an even bigger problem for most seniors in 2006, even with the new benefit in place, than it was back in the 2000 election campaign. The problem is projected to get even worse in future years. By the end of the projection period, the average household over age 65 will be spending more than twice as large a share of its income on prescription drugs as it spent in 2000. While the new program will clearly provide some assistance to seniors, it does not come close to solving the problem. Unless the cost of drugs is brought under control, it is difficult to see how either seniors themselves, or the federal government, will be able to afford the prescription drug bill of the elderly.

Appendix:

The projections in Tables 1 and 2 rely on the projections for drug expenditures and after-tax income in Baker and Schmitt, 2003 (“Growing Pain: The Expense of Drugs for the Elderly,” Washington, D.C.: Center for Economic and Policy Research [http://www.cepr.net/Growing_Pain_Issue_Brief.htm]). The appendix of this paper includes a detailed discussion of the construction of these projections. This paper uses the same projections, assuming that drug expenditures are reduced by the amount of the insurance payment (plus the premium). The calculations effectively assume that all households have the average level of drug costs for their income quintile, and that everyone buys the insurance premium. This may lead to somewhat of understatement of the benefits of the insurance, since the plan will cover 95 percent of costs for seniors with high drug expenditures.

These calculations assume that the availability of Medicare prescription drug insurance does not affect the willingness of private companies to provide insurance to retirees. If this benefit leads to lower coverage rates, then this would lead the projections to understate the increase in expenditures on prescription drugs. These projections also do not take account of any price effect of this drug benefit – the possibility that the additional spending may lead to more rapid price increases by drug companies, nor the possibility of increased usage as a result of the benefit – the possibility that poorer seniors may be more willing to buy some drugs if they have Medicare insurance that covers 75 percent of the cost. Both factors would also lead to higher projected levels of expenditures. 

Finally, these projections do not take account of the means-tested benefit that will provide large subsidies to low income households. In addition to an income test, eligibility for this benefit also requires that households meet a very low asset test – they must possess less than $6,000 in financial assets. Given this asset requirement, it is not possible to determine eligibility for this program based on the calculations in the earlier paper. Since this new means-tested benefit will largely replace existing Medicaid benefits (the effect of the Medicaid benefit is included in these projections), it is not clear that its omission leads to substantial errors in these projections.