Failing To Identify The New Rx, Replacement Rx Split

5. FAILING TO IDENTIFY THE NEW Rx / REPLACEMENT Rx SPLIT HRP WINDFALL PROFIT REPORT: Our report’s main analysis assumed that the split between new and old prescriptions was 60/40. This assumed a program that meaningfully addressed the great suffering that many Medicare beneficiaries endure because they cannot afford their medications. We also provided a contingency table (Exhibit 3) for those who assumed a different split. Thus, while focusing our discussion on the estimate that 60 percent of the new federal spending would go to fill new prescriptions, we showed the effect on the rise in profits as that share of the federal subsidy ranges from 40 percent to 80 percent. The contingency table indicated that with new prescriptions getting 40 percent of the new federal spending, drug makers’ new profits would be an estimated $80 billion, rather than $139 billion, other things equal; with 80 percent going to new prescriptions, the estimated new profits rise to $198 billion over eight years. PRI/PwC: The PRI/PwC critique noted the assumption in our main analysis that 60 percent of the subsidy will constitute new spending, but did not mention the contingency table or the range of estimates that it offered. The critique focuses, as discussed elsewhere, on downplaying the potential rise in volume, but does not make explicit its own conclusion, if any, about the likely split between new and replacement prescriptions. When the critique presents its unexplained estimate that insurance coverage will rise from 58 to 70 percent of Medicare patients’ drug spending (discussed above), it states only that “some of the new coverage will replace coverage that Medicare enrollees already had.” (p. 4). REPLY: It is important to address this issue explicitly because the calculated rise in drug makers’ profits, as we noted (p. 8), is fairly sensitive to this split between new and replacement prescriptions. That is because the new volume means new revenues and profits, whereas drug makers may receive less revenue and profit on replacement sales than they received on those sales in the past. Whether and how much the drug makers’ revenues and profits decline on the replacement sales depends, as discussed elsewhere, on the extent to which drug makers give Medicare plans average prices that are lower (because discounts/rebates are greater) than those drugs sold for in the past. The new/replacement prescription split depends greatly on the extent of new demand under the new coverage—which we discuss under price-elasticity and elsewhere. While this split has strong implications for drug makers’ profits (the focus of our report), it also has important implications for the value of the Medicare drug benefit program. Much replacement spending does have public benefit. Many Medicare patients today are obtaining all the medications they need but they or others who pay for those drugs (state Medicaid programs or employers, for example) are struggling under the cost. Replacement spending, however, would only help by easing the cost burden for medications that people were already able to buy—whether through existing coverage or out-of-pocket payments. But much of the urgent desire for a Medicare drug benefit stems from the concern that many patients are unable to fill the prescriptions they need—or fill them half as often as they should. Will the new benefit address any substantial share of this need? Compared to our analysis, the PRI/PwC critique assumes that a much smaller share of the new money will go for increased volume, and a much bigger share will take over the cost of paying for existing drugs. This clearly means that PRI and PwC imply the law will do much less to bring new medical care to patients than we expected. But the PRI/PwC critique doesn’t make this implication explicit. They never offer an alternative estimate from their calculations of the split between new and replacement prescription. Proponents of the law should view our analysis as more optimistic than that offered by PRI/PwC. PRI/PwC suggest that the law will mean relatively little rise in new volume, while we suggest that the $400 billion—despite the law’s limited benefits, high patient cost-sharing, and high drug prices—will enable many patients to fill prescriptions that otherwise would go unfilled. 6. HIDING CRUCIAL CALCULATIONS HRP WINDFALL PROFIT REPORT: All of the calculations in our report are transparent and are explained fully. PRI/PwC: The PRI/PwC critique announces that it relies on “transparent assumptions and accepted research.” A central element of their critique is the projected rise in purchases of prescription drugs induced by the extent of new drug coverage. They state that they estimate a rise in the share of prescription drug use that would be covered by insurance (see top of p. 4 and note ix) to 70 percent from a current 58 percent. But the analysis itself is not disclosed. The PRI/PwC critique simply and unhelpfully claims in endnote ix that it rests on separate calculations for four or more types of current insurance and three aspects of the new law, including low-income subsidies and others. REPLY: Some of the PRI/PwC assumptions may be transparent, but most of their calculations are not. The critique’s main source of “accepted research” is the 30-year-old RAND Health Insurance Experiment, whose relevance to price-elasticity of demand under a new Medicare drug benefit is highly questionable at best, as we documented earlier. The critique’s two main failures to make transparent assumptions are (1) its presentation of the unsubstantiated CBO estimate of price discounts well in excess of those achieved by bigger buyers today, and (2) the asserted rise from 58 percent to 70 percent in the share of Medicare beneficiaries’ medications covered by insurance, a claim which rests on unreported calculations. On the latter point, the core of their analysis, their work is cloaked in secrecy. Undisclosed calculations—a “black box,” the workings of which are not disclosed— underlie the crucial estimates of the rise in use, which they assert will be substantially smaller than we suggest. While saying these estimates reflect calculations taking into account four or more types of current insurance and several aspects of the new law, they present no evidence on those calculations, providing no way to evaluate the specific elements therein. PRI/PwC’s critique takes places at two extremes. At one extreme, it offers a low priceelasticity of demand (relying on 30-year-old studies of a different benefit, a different population, and a different pharmaceutical climate) and a rise in the share of drugs covered by insurance from 58 percent to 70 percent (neither the numbers nor the methods of calculating them is documented). At the other extreme, as discussed elsewhere, the PRI/PwC critique asserts that drug makers’ revenues after implementing the new benefit will face between a 1.0 percent drop in total revenue and a 3.2 percent rise. But it does not show how these estimates were derived. Their claim that industry revenues might fall under the law (despite the $400 billion in new federal spending) is blandly attributed to conversations with anonymous sources in the pharmaceutical industry (PRI/PwC, note 4). The PRI/PwC critique fails in these separate activities. It also fails to present an integrated, connected analysis of changes in drug makers’ revenue, to tie together the individual pieces. As a result, according to the PRI/PwC critique, the $400 billion in new federal spending seems to simply to float away into the ether.