When Past Is Not Prologue: The Weakness of the Economic Evidence against Health Insurance Mergers

Abstract

The DOJ’s challenges of both Aetna/Humana and Cigna/Anthem mergers in 2016 are fundamentally rooted in a timeworn structural analysis: More consolidation in (where the market is a hotly-contested issue, of course) means less competition and higher premiums for consumers. Following traditional structural playbook, DOJ argues that mergers would result in presumptively anticompetitive levels of concentration, and that neither new entry not divestiture would suffice to introduce sufficient competition. It does not (in its pretrial brief, at least) consider other dynamics (including especially complex and evolving regulatory environment) that would constrain firms' ability to charge supracompetitive prices.The parties contend that things are a bit more complicated than government suggests, that government defines relevant incorrectly, and that there is no correlation between number of insurers in a given county and insurance premium pricing.The trials will, of course, feature expert economic evidence from both sides. But until we see that evidence, we are left to evaluate basic outlines of economic arguments based only on existing literature.A host of mergers' critics have determined that literature condemns mergers, based largely on a small set of papers purporting to demonstrate that an increase of premiums, without corresponding benefit, inexorably follows health insurance consolidation. In fact, virtually all of these critics base their claims on a 2012 case study of a 1999 merger (between Aetna and Prudential) by economists Leemore Dafny, Mark Duggan, and Subramaniam Ramanarayanan, Paying a on Your Premium? Consolidation in U.S. Health Insurance Industry, as well as associated testimony by Dafny, along with a small number of other papers by her (and a couple others). This paper challenges critics' claims, taking a close analysis of Paying a Premium paper as a jumping off point. While our analysis doesn’t necessarily undermine paper's limited, historical conclusions, it does counsel extreme caution for inferring study’s applicability to today's proposed mergers. That said, different markets and a changed regulatory environment alone aren’t only things suggesting that past is not prologue. When we delve into paper more closely we find even more significant limitations on paper's support for claims made in its name, and its relevance to current proposed mergers. In short: extrapolated, long-term, cumulative, average effects drawn from 17-year-old data may grab headlines, but they really don't tell us much of anything about likely effects of a particular merger today, or about effects of increased concentration in any particular product or geographic market.