It seems to me Amihai Glazer deserves the credit here

I keep on seeing papers and notices of the notion that commonly owned firms — say commonly owned by diversified mutual funds — might collude rather than competing vigorously against each other.  After all, maximization of joint profit would seem to be the imperative, not firm-by-firm profit.

It is not so widely known who first came up with that idea, and it is my former colleague at UC Irvine, Amihai Glazer.  I know this because Ami and I had a co-authored paper on this topic, could it have been as early as the late 1980s?  (I don’t remember the year.)  And while it was genuinely joint work, the key idea came entirely from Ami, not from me.

We tried to publish the paper at several journals, but they all told us it was crazy.

I should also note Ami and I never quite agreed on what the paper meant.  I always viewed it as more of a theoretical curiosity.  I’m still not sure I understand Ami’s take, but in general he saw it as closer to a real world possibility or maybe stronger than that.  I also thought that even if joint ownership came about, forces akin to those discussed in the socialist calculation debate still would require something akin to firm-by-firm competition, rather than managed collusion (what about just picking managers with sluggish temperaments, thus leading to an intermediate solution?)  I don’t think that was Ami’s view back then at least.

I don’t know where my copy of the paper is, I hope Ami still has one.  In the meantime, I hope credit goes where credit is due, and that is to Amihai Glazer.


Here is what Azar, Schmalz, and Tecu (2018, Journal of Finance) give as the intellectual background for the idea in their lit review. I believe that the papers in the last paragraph, including Reynolds and Snapp (1986) and Bresnahan and Salop (1986), are most directly related:

>>We thus complement a long but mostly theoretical literature arguing that shareholders with diversified portfolios seek to maximize joint portfolio profits as opposed to individual firm profits, and as a result large-shareholder diversification can reduce competition in product markets, for example, Rotemberg (1984), Farrell (1985), Gordon (1990, 2003), Admati, Pfleiderer, and Zechner (1994), Hansen and Lott (1996), Rubin (2006), Margotta (2010), Azar (2012), and Azar (2017).

>>This literature has a rich background. Under imperfect competition, when shareholders hold more than one firm, they may disagree about the firm’s objective (see, for example, Hart (1979)). A theory of shareholder preference aggregation is therefore necessary. To that end, Azar (2012, 2017) develops models of oligopoly firm behavior in which competition for shareholder votes among potential managers leads firms to aggregate and internalize shareholder interests, including holdings in competitors.

>>Reynolds and Snapp (1986) extend classic oligopoly models to allow firms to hold shares in competitors. Bresnahan and Salop (1986) introduce a modified Herfindahl-Hirschman Index (MHHI) to quantify the competitive effects of horizontal joint ventures. We use O’Brien and Salop’s (2000) version of the MHHI to measure common ownership concentration.

It's also significant that A. Glazer does not list TC's cited paper in his online CV: Maybe it was just an idea 'in the air' that had multiple authors? Not uncommon in innovation and IMO should be acknowledged as helpful (I don't like 'first in time is first in right' laws, or 'first past the post wins all')

There's a 1990 paper by TC, AG and ?? (i forget)... Seems to have been cited a number of times... Of course, so apparently was his paper on ski lifts.

Indeed, several authors worked on this topic in the early 1980s, with informal statements of the idea already written in the late 1970s:

Don't most CEOs and other executives receive incentive compensation in the form of their own company's stock rather than a portfolio of industry competitors' stocks?

While that is true, the boards that determine their compensation could simply, by collective agreement, grant them incentive pay in the form of some market-cap weighted percentage of all of the other companies party to the agreement. Though maybe this runs afoul of anti-monopoly laws?

But the boards don't do that, which would seem like pretty strong evidence that common ownership by diversified mutual funds has not lead to collusion, at least not yet. Executive compensation does not incentivize collusion nor have boards tried to shape compensation in that direction. Though, I suppose recent proposals to de-emphasize incentive compensation, either in the name of reducing short-termism or inequality, could be viewed as closet attempts to encourage collusion.

That approach seems way to obvious and so the potential for facing anti-trust charges probably would not be implemented I would think. Rather than the transparent market weighted approach setting the performance goals and targets for each CEO such that a similar outcome is likely would be a more opaque solution.

Here is some empirical work to help inform this debate:

As for the economic theory behind, it's important to realize that the economic mechanism need not be collusion, but decreased incentives to compete. Are diversified mutual funds particularly aggressive at pushing managers to compete market share away from each other (compared to, say, a Richard Branson, or Elon Musk)?

There is an argument to be made for paying CEOs for their stock's outperformance over competitors (as some hedge fund managers are paid), not absolute performance, as that is mainly due to factors out of their control. The fact that this is not the case could mean that common stock owners don't want to incentivise CEOs to compete.

Isn't the competition already built in? It's not that they don't want to, but that they don't have to?

The original issue was whether common stock ownership leads companies to collude. By Josh's logic, but applying it to shareholders, even if shareholders owned only one company's stock, there could be problems unless those shareholders were shorting all competitors!! Indeed, it is possible that completely independently owned firms, with no overlapping ownership, might collude as a cartel. However, the temptation to "cheat" often breaks such cartels. Regardless, the issues there have nothing to do with common ownership.

The point is that it is the incentives facing the firm managers, not the non-managing shareholders, that matter. Even if the non-managing shareholders own stock in competing companies, the firm managers most often do not.

Let’s take your point as true for the sake of the argument - what are the incentives of the firm managers? In the case of concentrated common ownership, the firm managers may have a variety of a incentives that harm, or insufficiently incentives, competition in the way a non-concentrated common owner would.

Starting with just one incentive that has motivated corporate governance research for decades: the market for corporate control.

To that end, I forgot if it’s Schmaltz et al’s work or Florian et al’s work that demonstrates if the firm manager seeks to maximize their likelihood of reelection, then they will maximize the profits of their most influential shareholders.* Increasingly, as Coates, Gormley, & Bebchuk et al have all pointed out (among others), that aforementioned “most influential shareholder” is increasingly and most often an institutional investor (typically one of the big four) who is also a leading shareholder of a horizontal competitor.

So, for one example of the several mechanisms proposed, do you think firm managers act in their own self interest by maximizing their chances for re-election?

*this was done off a cold recollection of the literature. I ~believe~ in this model, profit rights of shareholders are set equal to their voting rights, which may or may not reflect reality.

'It is not so widely known who first came up with that idea'

Who knew that the word 'cartel' was first introduced at UC Irvine?

A 'group of similar companies who agree prices between them in order to increase profits and limit competition'

Admittedly, that is a UK definition, as the standard U.S. definition includes ' independent .' As if the pursuit of profit is somehow only left in the hands of individuals, without concern for anyone else profiting from making mutually beneficial arrangements involving pricing, a perspective that Adam Smith would have found laughable.

Seriously. I would say I came up with the idea, but JP Morgan beat me too it by more than a century.

Was anyone disappointed that Amihai Glazer turned out to be a white mail? I was hopping for a POC with a name like "Amihai"

meaning he would threaten to disclose good things about you if you didn't pay him to be quiet? ;-)

Amihai is a Hebrew name and, for what it's worth, many of those would NOT be disappointed that he is a "white" "mail," may not consider a Jewish male to be "white."

I'm sure that some commenters here found it their red letter day, actually.

Man, have to remember to turn on javascript when commenting after the last comment.

Maybe I should just male that suggestion to myself.

The mutual fund had its genesis in collusion, the collusion of the heirs of wealthy industrialists seeking a higher returner to support the growing legions of heirs. Well, collusion may be too strong of a term to describe the effort, but the goal, a higher rate of return, would be achieved only by increasing the rate of return attributable to the combination of the heirs' resources into what would be a common enterprise. Is Cowen opposed to the collusion he is describing among commonly-owned (through mutual funds) firms? I would assume that Cowen believes it's a feature not a bug, for Cowen is a friend of monopoly. Yes, the mutual fund and its enormous growth is in many ways tantamount to monopoly. Does it increase the rate of return? It seems to: why else have mutual funds become so large as to be a dominant force on Wall Street. But there's a downside: instability. It's a house of cards, so when one falls they all fall. The concentration of wealth among individuals coupled with the concentration of wealth through mutual funds exaggerates the rate of return on the way up (as asset prices are bid up) but also exaggerates the losses on the way down (as mutual selling creates a cascade).

Glazer is the co-author of Why Government Succeeds and Why it Fails. Cowen wrote a positive review (that is quoted on the Harvard University Press web page for the book). Another reviewer offers this: “Most economists have a list in their heads of the conditions under which markets fail. Most political scientists have a set of theories about why politicians may fail to represent notions of the ‘public interest.’ The authors here ask a different type of question, which is often overlooked: under what economic conditions are government policies likely to work?" I suppose many would answer: there are none. That's not very helpful. Unfortunately, government policies are more often than not adopted in times of crisis, so not much thought goes into whether the conditions are ripe for the policies to work. It's more of a "do something" or a "do nothing" rather than a well-thought out analysis of the problem, the options, and the circumstances. In any case, policing mutual funds for unlawful collusion is about as likely as the adoption of Senator Warren's wealth tax. Calling Professor Glazer, we need you.

To clarify, it wasn't one set of heirs but different sets of heirs of many different fortunes who combined their fortunes in search of a higher rate of return. History is both interesting and offers many lessons for those willing to take the time to learn.

Matt Levine loves the topic but I don't remember he has ever mentioned Mr. Glazer.

This looks like the paper:

Cowen, Tyler and Amihai Glazer. 1991. “Why Do Firms Compete: Shareholder Diversification and Joint Profit Maximization

I've worked with top execs at a wide variety of firms, and a large majority of them don't really have the temperament to enter those direct competitive battles that lead to strong competition: Many are conservative enough to not even try to create new markets: They'd rather compete by creating a new upmarket, higher profit vehicle than by becoming more efficient and lower prices. This is true even in cases where I consider the executive exceptionally competent: Their most aggressive moves to capture higher market share of existing markets were sly and relatively hard to detect. It's more likely to see an industry lower costs while keeping prices up, knowing that any change in pricing would be matched by the competition, than to see companies really taking prices down as low as they can go.

That's not to say that there aren't aggressive CEOs out there, or specific industries where margins are really razor thin and actual pricing wars happen (look at what happened to airlines for a while), it's just that everyone knows those situations rarely create big winners, and only the brightest and most aggressive, in a position where they have an efficiency advantage that they believe will be very difficult to meet by the competition, are going to be pulling the trigger. I'd argue that even if you installed Bezos as CEO in a pharma company or a bank, you would not see Amazon-like behavior.

Interesting, thanks

Comments for this post are closed