I’m posting here a (very) lightly edited transcription of a presentation I gave last month at the Stigler Center conference on antitrust held at the University of Chicago. The goal of the panel I was on was to consider merger policy in relation to labor and labor issues. If you prefer to watch or listen, the entire panel (indeed, the entire conference) is available here, and my presentation starts at about 2:43 or so.
As a sidenote, this was one of those instances where the occasion and context generated the content of the presentation nearly as much as previous writings did–which is why I’m posting the presentation, and also why I’ve edited it very lightly. Here it is.
I want to first note: we’ve been talking about mergers at this conference nearly to the exclusion of other antitrust issues. When we talk about whether antitrust is an appropriate tool to deal with labor market harms, we shouldn’t just be thinking about mergers. We should also be thinking about other antitrust rules, notably those that govern vertical restraints. The treatment of vertical restraints has been really the biggest change worked by the Chicago School Revolution … and that has direct impacts on labor. I won’t talk about this too much, but there was a great New York Times article yesterday by Lydia DePillis on franchising and its effects on both labor and smaller firms. The article doesn’t talk directly about antitrust, but ultimately it’s these permissive vertical restraints rules that make the characteristic control without responsibility of the franchising business model –which labor advocates have targeted for decades now—possible. Those rules make the organizational arrangement of franchising possible, and they are of course an antitrust issue. The article notes that both the NLRB and the FTC are now interested in thinking seriously about franchising. So I think it’s important that we don’t just think about mergers here but also keep in mind other antitrust rules. In addition to vertical restraints and merger policy, I also think the rules on horizontal coordination have impacts on labor in a way that I think will become more clear in just a minute.
So, labor and mergers. Now, I’m going to suggest that we think about labor and merger policy, not just from the perspective of what the effect is that a specific merger or a set of mergers has on market concentration, and the effect that has on wages—although that is certainly important and I think all of the empirical and statistical evidence that the last speaker mentioned is very relevant … I think we should also think about the merger rules we adopt in terms of their systemic effects on labor, on labor unions, on workers. And so here I also suggest a more general shift in our thinking about all of these antitrust rules that we’ve been discussing: that we think about them not just in terms of how we would apply a given set of rules in a given adjudication, though that’s important, but also in terms of how they function prophylactically to create, or not create, the sorts of markets we want to see.
So specifically with respect to mergers, what is the effect of a permissive merger policy versus a quite stringent merger policy on the competitive and business strategies that executives, that boards are looking at, and that they may even be compelled into when everyone else is also pursuing them? What competitive strategies are open to firm controllers given a particular set of legal rules? In the context of a not very permissive merger policy, what would happen to business strategy? I’m just going to ask the question as a thought experiment: if we largely take mergers off the table as a competitive or business strategy, we make them the exception and not the rule … what would happen? What would be the implications for markets in general and for labor in particular? I’m not going to answer those questions exhaustively, but I’m going to suggest that we think about those questions.
We’ve also been having a kind of meta-debate here about what sorts of rules, standards, and goals are appropriate to consider. Something we’ve heard several times is that broadening the type of antitrust goals that are considered (from economic efficiency, or narrowly-defined economic welfare) will be intractable for decision-makers, will result in a confused antitrust policy that doesn’t do anything well, and will vest too much discretion in judges. And to address that, I want to to try to get us to at least agree on the subject matter that we’re discussing, even if there’s disagreement about those other things, the specific rules and standards and even what type of rules and standards antitrust should adopt. And I think that part of the subject matter we’re discussing when we’re discussing antitrust is: what kind of markets do we want to have in our society, and given that law inevitably shapes markets, what sorts of rules and processes should the law adopt?
And I want to distinguish here between considerations that may not be appropriate for individual judges or decision-makers, versus considerations are not appropriate for the law or policy as a whole to be oriented toward. Now, for individual judges in individual adjudications, I’m the last one to suggest that we increase the discretion that our federal judges have. I’ve argued against that in print. I think we should absolutely adopt bright line rules that constrain that discretion. But we should distinguish between the considerations that are appropriate for individual decision makers or for that matter, ALJs to consider in deciding individual cases that arise under the antitrust laws, versus the considerations that are appropriate for us at this conference or as academics, as advocates, as researchers, as legislators, to consider when we are talking about designing our antitrust laws. Because those are conceptually two completely separate questions. And if we’re not open to considering those questions systemically, then I don’t know what the point of having a conference like this actually is, or what the point of academic, at least legal, academic research is. That’s our job. I was a lawyer for many years: we are bound by the rules when we’re practicing lawyers, but that is not what we’re doing here. So, we do not have to entrust a judge with considering some of these bigger impacts that market concentration and permissive merger policy may have on the polity and on society – but it’s an appropriate thing for us to consider when we think about what type of rules to adopt.
And here again, a permissive or not-permissive merger policy is one of these rules that we can think about in this prophylactic way. As a quick aside, I think we should also think about FTC rule-making this way. Empowering and actually just allowing the FTC to do its job of engaging in competition rule-making is another way, a very important way, of consciously creating these prophylactic rules. I think that the authority they have to do this is extremely clear, given the statements of the Supreme Court and the evidence from the legislative history. The reason I’m bringing that up here is that it illustrates the more general point: the considerations the FTC should take into account when doing competition rule-making are different from how much discretion we’re going to give to individual decision makers. Okay, so I think I’ve made that point.
So, let’s come back to the main topic. What’s a merger? A merger is a specific method of expanding the scope of a particular form of economic coordination that has been authorized by law. There is nothing natural or necessary about firms as a compulsory form of economic coordination or organization. All throughout history, people have been innovating and creating and trading in all sort of organizational forms and economic institutions … through craft guilds, through mercantile guilds. There’s nothing compulsory or natural about the specific form of the modern firm—shareholder-driven and organized in the particular way that it is, with workers largely divorced from decision-making. There’s nothing natural about that. Now, I’m not suggesting that we go back to the guild system. I’m asking us to take a step back and think about what a firm is and what we’re doing when we say that the presumption should be that, in essence, firms or rather those that control them have the right to combine into still-larger and more powerful firms.
And I want to note that the firm is specifically an antitrust exception. It is a suspension of competition. You can say that it’s crazy to think about not allowing firm-based coordination; how would we produce? Well, there’s actually all kinds of ways. At the simplest level, if you were truly trying to maximize competition, you could take the output of a firm and divide it – you keep all the operational integration, but you could just divide control rights over that output among everyone who works for that firm and let them individually price their share of the output. That would be a more competitive outcome, and would potentially replicate the level of competition that the modern business firm displaced. I think it’s fine that we don’t do that, but I think we need to notice what is actually happening here. I’m not the first one to have this insight: Ronald Coase, most famously, had this insight, in essence. He didn’t talk about it in terms of antitrust, but he did talk about the firm as a suspension of competition and market exchange.
So that is what a merger is: expanding this particular way of suspending competition. We need to just be very clear about what we are actually doing and how we treat that form differently from other suspensions of competition … including labor unions, cooperatives, and yes, even looser horizontal coordination in the form of things like trade associations. (I’ll come back to that point.)
Okay, so mergers are an expansion of the scope of what is already the biggest and strongest antitrust exemption we have: the firm. So that’s one thing I want us to notice about mergers. The second point I want to make is about “efficiencies.” When when we talk about efficiencies that may be realized by a given merger, I think we need to consider two things. Which of them, first of all, actually are efficiencies versus something else? And secondly, we need to consider whether and how any genuine efficiencies could potentially be achieved through forms of coordination other than the expansion of the firm.
Now, taking the second point first, this idea of thinking about economic coordination broadly is not just an academic point. What could we do right now in terms of other forms of coordination that could realize some of the putative efficiencies that specific mergers may accomplish? For instance, could some of those genuine efficiencies –just considering the genuine efficiencies for a moment— could they be achieved through, for instance, industry-wide standards in various situations? Could they be achieved by a trade association? We have trade associations that do marketing for various industries. You’ve all seen the milk ads, California raisins, right? Are there, is there coordination through those types of mechanisms that could actually preserve more independent decision-making throughout the economy? That, by the way, is one of the goals of antitrust under settled law. That’s still good law, as far as I know: that dispersed decision-making is one of the goals of antitrust law. So, should we be considering those alternative forms of coordination … and indeed, should we, in any new merger guidelines, be considering a safe harbor for some of these alternative forms of coordination, potentially through dispersed coordination, through trade associations or industry-wide standards that could achieve some of the genuine efficiencies without the consolidation, and specifically without the consolidation in the form of the shareholder-driven firm that mergers currently signify? So I think we need to put that on the table and consider it.
And on the first point about efficiencies: we also have to ask, how many of these putative efficiencies are actually intensified forms of extraction versus actual efficiencies? And here, it’s important to first get conceptually clear about what we mean by “efficiency.” Do we mean allocative efficiency? Now, obviously that’s undermined by a corporate merger. I should say, I don’t find that concept actually terribly useful myself, but to the extent that we are adopting it, obviously a merger decreases allocative efficiency. Everyone has actually agreed on that: Bork, Williamson agree with me. Though somehow, we seem to sometimes forget this simple point.
So, allocative efficiency certainly isn’t an efficiency that’s contributed by a merger. It has to be a production efficiency. Then the question becomes: when are there true production efficiencies that are contributed by a given merger? In some cases, what we call a productive efficiency may actually be a form of extraction. I’m not saying they all are; I think there are genuine efficiencies. But I think that sometimes, specifically when we’re looking at price as an index of production efficiency—which is obviously a shorthand we use all the time—we mistake extraction, or just a transfer of benefits from one group to another, for efficiency. In other words, often low prices are not an indication of production efficiencies; they’re an indication of extraction. And I don’t mean that in just a moralizing, pejorative way; I think this has to be acknowledged regardless of your ultimate normative views.
Let me break it down. What is a production efficiency? A technical efficiency means you get more output for the same input. It is technical innovation. We have seen so much such technical innovation in history. We live at a time that we’re benefiting from it tremendously. But consider the difference between a machine that allows two workers to produce more (at the same quality) with the same effort, versus a new institutional or organizational arrangement that pays those two workers less to produce the same amount—or has them put in more effort for the same amount. Those are not the same thing. The second thing simply is not a technical efficiency. I’m not taking a normative position on that, right now. You can, I suppose, have the position that that’s a good thing and that we should drive wages down to subsistence levels. I do not think that—I do not think many people would admit that they think that even if they do—but I want us right now to notice conceptually there is a difference.
And the same thing applies when a merger leads to greater bargaining power with respect to other firms that are input suppliers or distributors, or whatever—firms in adjacent markets that you’re bargaining with. If you’re able to—if you ultimately even get lower prices, which we know we often don’t because instead those savings go up to shareholders— but even when you do get lower prices, to what extent is it coming from superior bargaining power with respect to suppliers, distributors, other trading partners versus true productive efficiencies? Because obviously if you now have—well, we should look at it in individual cases, but oftentimes, you now have more bargaining power with respect to trading partners in adjacent markets if you’re now a much bigger firm and you have a much bigger market share. So, again, we need to distinguish between true technical efficiency and extraction, both on the labor side and on the smaller firm side.
And by the way, that’s not mainly because we just care about small firms and we romanticize them, like that’s a nostalgic ideal as it’s often characterized. Those small firms have workers themselves. So even the supposed wage premium of large firms—we need to look very carefully at that, because to what extent is that a wealth transfer from the workers of other firms to the workers of large firms (to the extent there even is a wage premium)? Notably, aside from current harms, you don’t even maintain that benefit if you just get rid of small firms to extract from.
Right, so these are all things that I think we need to just take a step back and think about when we think about merger policy and workers. I suspect I’m over my time, but I also really want to include in things that are not efficiencies, the direct transfer that permissive merger policy may entail at a systemic level from, essentially, productive arms of the firm to non-productive arms, and therefore ultimately from workers to shareholders and deal-makers. If we talk about it in this individual way that sort of seems like we’re pointing fingers, but if we consider it systemically, we see how everyone including boards and executives are actually constrained by this.
So I just want to very briefly quote from Josh Mason, an economist who I think that maybe antitrust folks should pay more attention to. So, many social scientists and researchers have been talking about share buybacks as a form of this type of transfer from workers and from production in general to the financial and household sectors. Josh has this nice essay thinking about M&A itself as this same type of transfer. And he says that “when acquisitions are paid in stock, the total volume of shares doesn’t change. But when they are paid in cash, it does. In the aggregate, when publicly traded company A pays $1 billion to acquire publicly traded company B, that is just a payment from the corporate sector to the household sector of $1 billion, just as if the corporation were buying back its own stock.” So again, we are transferring from producers to savers at a systemic level. And again, you can have different views on the utility of that, but that is exactly what has been critiqued with respect to share buybacks. And whatever your opinion about it, it certainly isn’t a productive efficiency!
Finally, I just want to very, very briefly make the point I alluded to in the introduction: if you think about mergers and acquisitions in terms of this sense of merger policy, in terms of this sense of prophylactic rules, then you need to ask: what are the competitive strategies and tools that are on the table for a CEO, for a board, given a particular set of such rules? And sometimes—I mean, if you have taught business associations, as I have, then you read all kinds of cases with your students where the board doesn’t want to do the deal … no one who’s actually running the company wants to do this deal. Their hands are tied by the pressure that this set of rules creates, or at least their hands are strongly guided, toward doing deals that very few people actually want. And this is especially stark when we consider their primary job is to be a steward and a manager of certain productive assets. That’s the point of the firm: making the best use of productive assets. We’ve decided, as a society, that giving this job over to managers and CEOs and boards is the way we want to do that. So if that’s the goal, we must ask, with respect to merger policy: Wouldn’t it be useful to have prophylactic rules that actually channel their activity toward doing precisely that, rather than being forced to do deals that often do not benefit anyone [or benefit very few people]?