(iPoster) Public Opinion of Corporate Consolidation and Regulation of M&A

• In Event: Poster Session: Public Opinion 2

Abstract

Mergers and acquisitions (M&A) are among the most consequential forces that shape the structure of the modern American economy. Every year, US companies engage in M&A worth over two trillion dollars. These events quietly reshape markets and by extension, the everyday economic life of ordinary people. M&A and government responses to them can alter voters' evaluation of government performance. While the average voter might not consider concentration a core issue, they experience the consequences as consumers and workers.

Yet, the modern system for reviewing and regulating M&A is primarily a domain of elite discussion. The main institutions overseeing these economic activities, the Federal Trade Commission (FTC) and the Department of Justice’s (DOJ) Antitrust Division, have long operated at the margins of public attention. Since the 1990s, antitrust policy and enforcement have been widely regarded as a bi-partisan technocratic subject, where discussions are predominately among legal scholars and economists, and hardly engaged by the public. In the late 2010s, antitrust resurfaced as a prominent policy discussion, driven by regulators such as Lina Khan and Tim Wu.

Despite these elite discussions, far less is understood about how the public views corporate consolidation and the government's role in firms' M&A activities. Antitrust is a unique setting where not only we are able to understand how voters react to fundamentally important economic activities, but the issue itself is also relatively detached from partisan labels such that we can more cleanly understand how voters think markets should be regulated. Questions include what makes a transaction worth intervening, what makes a government intervention legitimate. In short, antitrust provides a policy space where partisan signals are weak and general base level information is low among voters.

In this paper, I show that average voters can hold reasonable and coherent evaluations of both corporate consolidation and antitrust enforcement, and employ familiarity as a heuristic; these evaluations are reflected in real world agency behaviors. From an original survey experiment, I find that respondents are more likely to want the government to stop large (in transactional value) mergers that occur within the same industry. Respondents also use familiarity - measured by revelations of well-recognized brands - as a heuristics to judge how potentially harmful a merger might be. For example, respondents would want the same level of government intervention between a $50 million merger with revealed brands and a $100 million merger if they don't know about the brands. Branding effects increase respondents' perception of firm market power and the sheer bigness of the firm, both of which are reasonable channels in considering the anticompetitiveness of the transactions.

Agency decisions to approve or stop a merger can change voters' evaluation of the agencies. Respondents increase their confidence in the FTC/DOJ when agency actions are congruent with their previously expressed preferences and falls when they are not, and have a baseline positive association between the government stopping a merger and higher confidence. Respondents are also able to update their opinions in the event of government failures. Learning that regulators failed to either secure a merger block in court (i.e. making a mistake when stopping a merger) or intervene in a harmful merger (i.e. making a mistake when approving a merger) significantly reduces one's confidence in the agencies, even among respondents who otherwise agreed with the initial decision, implying that the public distinguishes between incompetence and policy disagreement.

I also construct a novel dataset that links government and financial records in the context of Hart-Scott-Rodino reviews, and show that the survey-found voter preferences of government actions are generally reflected in the FTC/DOJ enforcement decisions. The agencies are much more likely to block large transactions - the top 1% in deal size - between firms from the same industry. Transactions where firms involved own recognizable brands attract more government scrutiny as well. At the industry level, those with more recognizable brands are more likely to get challenged. The effects of brands are more prominent among large (top 10 or 1% by transactional value) deals or horizontal (same-industry) mergers, which suggest that familiarity acts more like a shortcut when picking among the most risky mergers by economic factors. This result is especially meaningful since block attempts are exceedingly rare, occurring in around 0.6-3.4% of eligible filings each year. Notice that agencies, unlike the average citizens, have a lot more knowledge of the firms and the transactions. Whether the effects of familiarity should be taken as a signal of anticompetitiveness in addition to transaction observables or as suggestive evidence of pandering is worth further investigation.

Author

• Saikun Shi, Stanford University