Does common ownership really matter?

We derive a measure that captures the extent to which overlapping ownership structures shift managers’ incentives to internalize externalities. A key feature of the measure is that it allows for the possibility that not all investors are attentive to whether a manager’s actions benefit the investor’s overall portfolio. Empirically, we show that potential drivers of ownership overlap, including mergers in the asset management industry and the growth of indexing, could in fact diminish managerial motives. Our findings illustrate the importance of accounting for investor inattention and cast doubt on the possibility that the growth of common ownership has had a significant impact on managerial incentives.

That is from a new NBER working paper by Erik P. Gilje, Todd Gormley, and Doron Y. Levit.


So the agent-principal problem has gotten worse due to passive indexes?

Bonus trivia: reading Andrew Lo et al's "Non-random Walk Down Wall Street", which makes a brave attempt to show that data mining proves there's no random walk. But the differences between random and not random are very tiny, hardly exploitable except to a high-frequency trading bot.

I think the takeaway is that the principal-agent problem isn't as bad as the principal-chief executive agent-vice agent-agent director-agent supervisor-agent problem. Overlapping ownership structures give a lot more ways for incentives to get misaligned.

Managers in large companies have always been conservative. Play it safe, don't take needless chances that could blow up on you (analogous to survival strategies among traders that Taleb talks about).

Conservatism/risk aversion has grown in recent years, but much more due to greater HR/legal input on ANYTHING companies do than this hypothesis IMO.

"We hold these truths to be self-evident that I didn't fight the British to make a country full of cucks"
- President George Washington

Not only indexing, but basing investment decisions entirely on data mining (and algorithms that identify patterns in trading and stock prices). Of course, investors are told to forget stock picking (you cant beat the market) and, instead, invest in index funds that track the market. But now hedge and other funds are replacing analysts with computer engineers and quants to pick stocks based on patterns detected via data mining. Who is the principal and who is the agent in this scenario? I know, there's a hedge fund manager in Virginia who has been using the data mining technique for years and his returns are amazing. But if everyone uses the same approach, how does that work? Indeed, if everyone invests solely through index funds, how does that work? I appreciate that "analysts" are not really analysts but soothsayers. And I understand that history often repeats (i.e., the "patterns" that are identified by data mining). But if it's true, then why do economists do such a poor job as soothsayers? Can't they data mine historical economic data for patterns? Why is it rational for the manager of a hedge fund to rely solely on historical data and algorithms (is he really a "manager"), but not economists? Is it that economists, or many of them anyway, already know the answer they are looking for without the bother of data mining?

It is a very interesting question. Marketers often do attribution tests for advertisers (Millward Brown), but the attribution market for advertisers is mostly by third parties. Why can't attribution be done by the "demand" side companies since they have the technology? Because it's a technological nightmare. You are measuring actual activity with digital activity, sometimes the two occur at the same time. Or, is it that there is a conflict of interest?

Matt Levine has written extensively and brilliantly about this issue; well worth reading, I think.

turns out krugman cant print money
for the socialist democratic country of finland
& blames demented white men?

could a whale swallow Krugman?
hey Krugman!
so how come finland isnt
controlling finlands
health care costs like you predicted in your "paper"
is it because we also are demented white man/men?

we still gonna need the 3.1 billion euros

it's because the ratios are off. How could printing in color possibly be as expensive as it is?

hey krugman
not actually angry, white,demented &or from finland
just peckish & dissapointed
so please
what exactly is going on in here

I know that economists find financial markets rich data sources for analysis. I just wonder if there is any sampling bias -- Wall Street just isn't really like Main Street and the populations seem motivationally/behaviorally different from one another in important ways.

The whole point of stock-based compensation was to align managers' incentives with shareholders', an implicit recognition that, without such compensation, managers' incentives were often misaligned with shareholders'. Index funds may have increased common ownership among *shareholders*, but *managers'* stock compensation has not taken the form of common ownership. Managers receive own-company stock, not a portfolio of industry stocks or index funds. That would seem to suggest that worries about common ownership are overblown.

Maybe that was the original intent but it's been perverted into a technique that allows management to loot the coffers of corporations. Situations like this one are hardly rare.

I never really understood this argument. Unless you are in a very small company, your individual efforts to increase the value of the company are returned to you at a very low ROR. Your incentives shouldn't be much different than as with regular compensation.

See "Common Ownership, Competition, and Top Management Incentives" (, which finds that (1) because the value of any individual company's stock mostly moves in line with that of the industry at-large, for stock-based compensation to actually overcome this principal-agency problem, such that Managers are incentivized to care only about own-firm profits, firms should use Relative-Performance Evaluation Compensation (e.g., how that company's stock rises/falls relative to the stock of other firms in that industry). Indeed, another economist won a Nobel for showing that Relative-Performance Evaluations are the most efficient form of compensating firm managers. Yet, the Anton et al paper, cited above, finds that the use of Relative-Performance Evaluation Compensation decreases as the rate of common ownership increases; that is, as common ownership increases, it becomes more likely that a firm-manager's compensation will be more based on industry profits than own-firm profits (relative to the manager of a firm without a common owner).

Take it for what you will - but that's where the evidence is today.

Equity is not the same thing as ownership! For example, both the Small Business Act for Europe and the U.S. Small Business Act (1953) indicate that their primary purposes are to assist small businesses in fostering competitive markets (e.g., by preventing large businesses from forming market oligopolies and monopolies) and to address market failures (e.g., the difficulties faced by small businesses in accessing capital).

'Does common ownership really matter?'

Since when did Prof. Cowen become a socialist?

First, mergers are often done when weak firms are exiting because of competitive pressures. The best buyer for your distressed assets is another similar firm.

Or a dominant firm finds it easier to buy a desirable asset (perhaps buying market share) rather than develop something themselves.

If I'm an index investor do I care why the firms merged if I previously owned both firms? Not really. In the first case, the two firms are fighting to remain viable in a competitive environment. If I am a forward-looking investor I should see that long term the industry will have a shakeout with only the most efficient firms surviving, In the second case, the acquiring firm is growing through the most efficient means possible, by buying assets and lowering the transactions costs of acquiring similar assets. Why else do it?

Of course, if the firms are merging to create greater market diversification then management has probably done something that I as an investor can do quicker and easier on my own. But the market knows this and will not reward mergers that don't add value. As an investor, I would lose if I owed one or both firms. What they are doing is not adding value to either firm.

I would expect many mergers to, in their term, internalize externalities (or soften competition) because the mergers are a soft exit from the marketplace and reduce price competition. They are doing it to survive. But that is consistent with investors goals. Look at the Standard Oil mergers under Rockefeller.

The largest investors may force the firms to make tough choices In part because voting with their feet can be difficult as their exit can cause a downward cascade in value. As an index investor, I assume that occurs frequently and don't need to get involved

Comments for this post are closed