By: Eric de Bodt, Jean-Gabriel Cousin, Micah S. Officer & Richard Roll (CLS Blue Sky Blog)
A growing literature highlights the important effect of economic and political policies on mergers and acquisitions (M&A). M&A often involves major issues of corporate investment and resource allocation, and so inefficient interference in the M&A market can have significant and long-lasting economic impact. In a new paper, we investigate whether antitrust enforcement by the Department of Justice (DOJ) or the Federal Trade Commission (FTC) has the substantive and lasting effect of deterring U.S. mergers and acquisitions.
The DOJ and the FTC follow strict procedures for regulatory interventions, as described in the Horizontal Merger Guidelines[1], updated in 2010. These interventions may have substantial direct and indirect effects on future activity in the M&A market. Specifically, current antitrust regulation in a particular industry may directly discourage other potential acquirers in that industry from announcing deals for fear that they too will be subjected to regulatory scrutiny. Furthermore, merging parties have private information inaccessible to regulators, and seemingly substantial political bias in the regulatory process can contribute to unpredictable outcomes and enforcement uncertainty.
For example, regulators discouraged the Sprint/T-Mobile merger under the Obama administration in 2014 but then allowed it under the Trump administration just four years later. This uncertainty may indirectly discourage future deals as industry participants choose to defer, delay, or even terminate them until this policy uncertainty lifts. We refer to these potential effects of regulatory enforcement on future M&A activity as deterrence effects. Only a few, in some ways contradictory, papers directly test whether these effects exist and are significant (e.g., Eckbo (1992) and Clougherty and Seldeslachts (2013)).
To ascertain whether a deterrence effect exists in the U.S. M&A market, we follow the conditional probability approach used in Clougherty and Seldeslachts (2013) to study the probability that firms will be subject to future acquisition attempts in an “industry” currently subject to an antitrust-related enforcement action. The definition of industry plays a critical role here, and we build on the text-based similarity scores from Hoberg and Phillips (2010) to define industry clusters. Our main tests use the 10 nearest-neighbors cluster based on the target firm to define the “industry” that is potentially affected.
These clusters capture the 10 firms most similar to a proposed target based on the firm’s own description of its product market. Neighbor firms in the product market are likely to be the most relevant competitors for antitrust regulators because, by definition, they produce products similar to those of, and compete for the same customers as, the target firm (which is the subject of a current regulatory enforcement action). Another benefit of this H&P product market-based approach over traditional industry classifications (such as SIC or NAICS codes) is that similarity scores are recomputed each year, tracking the transformation of firms and industries, while SIC and NAICS codes are rarely updated at the firm level.
Our results are based on a sample of 6,285 M&A deals. In this sample, we observe substantially fewer future deals for firms in industry (i.e., 10 nearest-neighbors) clusters that have recently been subject to antitrust-related enforcement around mergers and acquisitions. Furthermore, the relatively few such deals that we do observe following regulatory enforcement in an industry are, on average, significantly smaller (measured using deal value) than future deals that we observe in industries not subject to antitrust enforcement, consistent with a desire to avoid DOJ or FTC investigations…
