Ernst R. Berndt, Iain M. Cockburn, Douglas L. Cocks, Arnold M. Epstein, Zvi Griliches
January 1, 1998
Recently controversy has surrounded the issue of whether Social Security payments to the elderly should continue to be adjusted automatically according to changes in the Consumer Price Index (CPI). One issue in the public policy debate concerns whether price inflation is different for the elderly, particularly because the official Bureau of Labor Statistics price indexes for medical care have been growing more rapidly than the overall CPI, and medical care expenditures constitute a larger proportion of the elderlys budget than of the youngs.Using annual IMS data from 1990 to 1996, we examine empirically whether elderly-nonelderly price inflation differentials exist for prescription pharmaceuticals. We assess prices for prescription drugs destined for ultimate use by the elderly versus the nonelderly at three points in the distribution chain: initial sales from manufacturers, intermediate purchases by retail pharmacies, and final sales from retail pharmacies to patients or payors. We find that at the initial point in the distribution chain, no differences in price inflation exist for the aggregate of drugs destined for use by the elderly versus those for the nonelderly. At the intermediate sell-in point to pharmacy distribution, we examine antibiotics (ABs), antidepressants (ADs), and calcium channel blockers (CCBs). For ABs, since 1992 price inflation has been somewhat greater for the elderly than for the young, reflecting in part the elderlys more intensive use of newer branded products having fewer side effects, adverse drug interactions and more convenient dosingattributes of particular importance to the elderly. For ADs, price inoation is considerably less for the elderly than for the young, due in large part to the elderlys greater use of older generic products. For CCBs, elderly-nonelderly differentials are negligible. None of these differentials adjust for variations in quality.At the final retail sell-out point, we examine only ADs. We find that because retailers obtain larger gross margins on generic than on branded products, and because the elderly are disproportionately large users of generic ADs, the elderly-nonelderly price inflation differential benefiting the elderly at the intermediate point is reduced considerably at final sale.