Common Ownership May Reduce the Entry of Cheaper Generic Drugs

In new research, Martin Schmalz and Jin Xie examine how shareholder preferences influence the United States pharmaceutical industry. They find that generic drug manufacturers sometimes harm their firms’ own value when doing so benefits shareholder portfolios, who frequently have stakes in competing brand-name firms.

A basic and convenient assumption of economics is that public corporations maximize their own value, irrespective of shareholders’ preferences. The Fisher separation theorem proves that the assumption that a firm maximizes its market value is warranted under the condition that the firm is not influential enough to affect market prices and must accept prevailing prices, known as a price taker. But, when the firm is not a price taker, and therefore has more market power in supracompetitive product markets, the prediction of the theorem may not hold because firms might aim to maximize the value of their most influential shareholders’ portfolios. For example, if influential shareholders of one firm also have large ownership stakes in the rivals of that firm, these shareholders could financially benefit from making decisions that harm company value in one firm for the benefit of its rivals. 

In our research, we document that some pharmaceutical firms in the United States predictably make decisions that systematically destroy their own value while benefiting the portfolio value of their largest shareholders when those shareholders own stakes in rival companies. These results not only contradict the prediction of the Fisher separation theorem, but also suggest that antitrust laws should extend their focus from merely reviewing mergers and acquisitions to closely scrutinizing purely passive investments in competitors. 

Institutional background

Pursuant to the Hatch-Waxman Act of 1984, generic manufacturers are allowed to file Paragraph IV certifications with the Food and Drug Administration (FDA) to enter a drug market before patents covering a branded product expire. A brand incumbent typically responds by filing infringement lawsuits against generic entrants, often to protect what it believes to be its intellectual property rights, but sometimes to simply delay the entry of a generic competitor. The two parties then either enter into a settlement agreement or go to trial with the likelihood that the court might dismiss the case in favor of the generic upstart. 

The FDA won’t approve more than one Paragraph IV application for the same drug until after the first generic manufacturer markets its generic version of a branded drug for 180 days. This gives generic manufacturers an incentive to enter the market and earn duopoly profits alongside the brand-name incumbent. A common tactic to maintain high drug prices is for brand firms to enter “pay-for-delay” settlements with this first generic entrant. Typically, the brand incumbent can pay the first generic to forfeit the 180-day exclusivity period by not launching its product before the automatic forfeiture deadline or delaying marketing, which precludes other generic firms from entering the market without actually offering a competitive product. These deals harm consumers. 

Our study investigates whether potential generic entrants, whose largest shareholders hold stakes in the brand incumbent, post “fake entries” to insulate the incumbent from competition. A fake entry occurs when a generic challenges a patent only to later surrender its profitable entry opportunity by entering into a settlement agreement that helps the incumbent maintain its monopoly power. 

Wealth transfer

What are the common ownership links between generic and branded firms in the U.S. pharmaceutical industry? For example, on average, when the same shareholder holds both generic and brand companies, which is likely around 22% of the time in our sample, the largest generic shareholder (top-one shareholder of the generic firm) owns 10.3% of shares of the generic firm and 4.3% of the brand firm. The top-five shareholders of generic manufacturers hold 28.5% of generic equity and 20% of brand equity. Because brand firms are much larger and their products are more profitable, these large generic shareholders earn much more of their revenue from brand profits than generic profits.

This common-ownership linkage is illustrated by the Paragraph IV litigation that occurred between Barr Laboratories (generic) and Merck (brand) in the last quarter of 2009. In this case, Merch was trying to limit the generic availability of its branded drug, Temodar, which is used to treat certain types of brain tumors. Table 1 lists Barr’s top ten shareholders and their corresponding stakes in Merck. Capital Research and Management Company, Barr’s largest shareholder, held 13.2% of the generic firm and 8% of the brand. The second- and fifth-largest shareholders, Wellington and BlackRock, held 3.3% and 2.5% of Barr, and 4.3% and 10% of Merck, respectively. Collectively, these top ten shareholders owned 33.5% of Barr and 28.3% of Merck. 

Table 1. Barr Laboratories vs. Merck 2009Q4 Paragraph IV litigation

Our findings suggest common-ownership links are associated with opposing stock-price reactions to a settlement between commonly owned first-to-file generics and the brand. In our study, we created a “profit weight” measurement to differentiate low and high common ownership between brand and generic firms. The higher the profit weight on the brand, the more the generic shareholders care about profits made by the brand incumbent than profits made by the generic firm itself. Across all cases we studied, for the brand, public asset managers with both high and low common ownership links between the brand and generic saw an increase in daily abnormal stock market returns following the event, with the high-common-ownership group outperforming the low group by an additional 1%. For the first generic, stocks in the low-common-ownership group experienced positive abnormal returns surrounding the settlement, whereas those in the high-common-ownership group experienced negative returns. On average, brands benefit when they share more investors with generics, while generics are worse off.

Are the behaviors of these common-owned “first-to-file” generics rational? We calculate the gains and losses for major generic shareholders around the settlement announcements that coincide with negative generic returns. In generic-brand pairs with high common ownership, the top-three generic shareholders collectively lose over $400 million, while the top-three brand shareholders gain more than $1.8 billion. This pattern, however, does not hold for pairs with low common ownership. 

Fake entry

Given that generic shareholders profit from settlements on the brand side, it is natural to ask whether common ownership—specifically, the degree to which common owners benefit from wealth transfers from generic entrants to brand incumbents—predicts fake entry. Indeed, we observe a strong positive correlation between common ownership and the likelihood of a firm being a first-to-file generic, indicating that economic incentives explain these decisions. In this case the costly actions of filing the first Paragraph IV application, including running risk assessments and lab tests, finding expert analyses, and conducting thorough research, are outweighed by the financial benefits of a “pay-for-delay” settlement. Among potential generic entrants whose largest shareholders derive more of their portfolio value from brand profits, the probability of being the first generic is 28.2%. In contrast, for potential entrants whose largest shareholders do not hold large stakes in the brand, the probability is only 16.7%.

We also find that, conditional on a generic firm filing a Paragraph IV challenge and the brand suing in response, generic firms whose large shareholders have a greater financial interest in brand profits settle with the brand incumbent 48.9% of the time. For generic manufacturers whose major shareholders have a lower interest in brand profits, the settlement rate is only 38.1%. Notably, the first generic whose largest shareholders also hold significant brand equity is 13.1 percentage points more likely to settle than other first generics. 

Our findings suggest generic firms’ largest shareholders, who hold significant stakes in the brand, may sacrifice the generic’s value for the benefit of the rival. For example, a brand might propose a settlement that is unattractive for the generic firm but appealing to its shareholders if they have large stakes in the brand. Considering the counterfactual clarifies the intuition behind this “fake entry” strategy: a generic whose shareholders have no stakes in the brand would be less willing to settle in ways that transfer wealth to the brand. Such a generic would be more likely to compete aggressively, potentially benefiting consumers but harming shareholders as a group. If such a generic settled, it would likely do so under terms forcing the brand to share rents. 

One remaining question is why and how have generic entrants and brand incumbents, who are suspected of engaging in anticompetitive practices, managed to escape antitrust scrutiny? There are compelling arguments that the threat of litigation is not substantial enough to deter settlement agreements. First, within the legal framework of settlements, risk-averse litigants are permitted to resolve disputes through negotiation, and hence, reverse payments, such as in the case for pay-for-delay, are not necessarily anticompetitive, especially when risk aversion and asymmetric information shape decision-making processes. In fact, when the Federal Trade Commission challenged reverse payment settlements, some courts ruled that as long as the generic drug’s entry date preceded the patent’s expiration, the settlement was within the “scope of the patent” and therefore beyond the reach of the antitrust laws. Second, specialized expertise in pharmaceutical patent law is scarce. Due to staff shortages and budget limits, the FTC only focuses on the most egregious pay-for-delay cases.

Conclusion

The conceptual importance of the findings is that they collectively reject the prediction of the Fisher separation theorem, on which much of corporate finance theory is built, and which holds that firms maximize their own value irrespective of shareholder interests. We hypothesize that the failure of the prediction is due to the firm not being a price taker—one of the key assumptions of the theorem. If the findings were to be held more generally, re-examining standard questions on corporate finance while relaxing the assumption of own-firm value maximization may be worthwhile. Whether the empirical facts documented in the laboratory of generic entry enjoy broader support is also an interesting area for future antitrust research.

Author Disclosure: The author reports no conflicts of interest. You can read our disclosure policy here.

Articles represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty.

Subscribe here for ProMarket’s weekly newsletter, Special Interest, to stay up to date on ProMarket’s coverage of the political economy and other content from the Stigler Center.