Yet the research is shockingly clear.
“Using pharmaceutical industry data, we show that acquired drug projects are less likely to be
developed when they overlap with the acquirer's existing product portfolio, especially
when the acquirer's market power is large due to weak competition or distant patent
expiration,” the paper finds.
“Conservative estimates indicate 5.3% to 7.4% of acquisitions in our sample are killer acquisitions. These acquisitions disproportionately occur just below thresholds for antitrust scrutiny.”
“Patients suffer because there are fewer drugs and the drugs that are developed and brought to market are sold at higher prices,” argues Jeff Skinner, Executive Director of the Institute of Innovation and Entrepreneurship (IIE).
These findings are the result of exhaustive and pioneering research. To begin with, the authors created an economic model that models the circumstances under which a monopolist (or oligopolistic set) of competitor(s) would have an economic incentive to ‘acquire-to-kill’ (AtK) an invention. They then consider a number of parameters including:
- The remaining patent (monopoly) time remining on the existing (incumbent) patents (the incentive to AtK increasing with such patent life)
- The market overlap of the acquirer with the startup project (the AtK incentive increasing with such overlap)
- The product development overlap (e.g. a similar chemistry) resulting in synergies that potentially reduce cost, and thus (perhaps) AtK incentive.
“The AtK strategy can be profit-maximising for an incumbent and that the incentive is at its highest when the market overlap is high and the product development overlap is low,” says Skinner.
“Colleen also models the situation where the existing industry is an oligopoly. For example, a small number of existing firms dominate the market between them. She finds that the same incentives exist, although the AtK strategy is of less benefit to any one incumbent.
“She then proceeds to find out whether such behaviour exists in the real world. To do this she needs comprehensive data on project-level outcomes of both acquired and non-acquired projects, overlaps between acquirer and acquired (ie target) firms and market and technological competition. Such data exists for pharmaceutical project development and is thus a great ‘laboratory’ for testing her model.”
The authors used the Pharmaproject database, covering about 16,000 projects initiated by some 4,600 firms between 1989 and 2010 and acquisition data from a wide range of credible financial databases, thoroughly ‘cleaned’ to prevent duplicate acquisition events.
Cunningham and co-authors write in the paper: “For each drug in our database, we can identify whether it went through any acquisition event during its development life cycle and if it did, the acquirer, the timing of acquisition, and development activity in the years pre- and post-acquisition.”
The parameters of the earlier model (product and development overlap) are constructed for each acquisition, and used for the empirical tests, comparing post-acquisition development of drugs acquired by incumbents with overlapping projects, to otherwise similar projects acquired by non-overlapping incumbents, to otherwise similar non-acquired projects.
Analysing strategic intent
To supplement their main analysis, the authors compare the behaviour of firms making acquisitions at just below and just above the antitrust review thresholds. They find that the eventual product launch rate is much lower (1.8% versus 9.1%) and the discontinuation rate is much higher (94.6% versus 83.3%) for below-threshold acquisitions compared to those right above the threshold.
Skinner says: “This provides supporting evidence of strategic behaviour by acquiring firms that is consistent with killer acquisition motives.”
He adds: “Moreover, an alternative hypothesis that firms quite legitimately acquire entrepreneurial firms for human capital doesn’t hold water. Only 22% of pre-acquisition inventors move to the acquirer post-acquisition, while 78% move to other firms.
“So, what’s the impact of the ‘acquire to kill’ incentive? Well, first, Colleen estimates that without the incentive, achieved, perhaps, by lowering the antitrust threshold, the total number of viable drug projects would increase by 4.3%, or around 13 projects a year.