Saturday, December 14, 2019

The Votes of Retail Shareholders

Alon Brav, Matthew Cain, and Jonathan Zytnick have a fascinating new paper analyzing the voting behavior of retail shareholders (and I linked to it once before but I'm pretty sure since then it's been updated with a lot of new data).  Bottom line: They got access to the votes cast by retail shareholders from 2015 and 2017 and made a lot of interesting findings, including:

Retail votes matter. Collectively, they have as much influence on outcomes as the Big Three (Vanguard, BlackRock, and State Street).

Expressed as a proportion of the shareholder base, retail shareholders hold a higher percentage of small firms than large ones, and their participation in voting is higher in small firms.

Retail shareholders are more sensitive to management performance than the Big Three; they are more likely to turn out, and more likely to vote against management, when companies have underperformed.  The Big Three, by contrast, are less sensitive to performance in terms of voting behavior.

Retail shareholder voting has an observable cost/benefit component.  Retail shareholders vote more often when their economic stake is greater; when management has underperformed; in controversial votes; and when they live in a zip code less associated with labor income (suggesting more time to devote to their portfolio).

Retail shareholders with higher stakes in the subject firm are more likely to vote in favor of management proposals and against shareholder proposals as compared to institutions; by contrast, retail shareholders with lower stakes are less likely to vote at all, but when they do vote, they’re more likely support shareholder proposals – particularly for the largest firms

So what are the implications of all of this?

Well, first, I’m struck by how everyone from the Progressives (Adolf Berle, William Riley) to modern thinkers like Einer Elhauge have assumed that the separation of ownership and control would lead shareholders to be less concerned about corporate social responsibility – the theory being that as shareholders feel less responsible for corporate behavior, they’ll shed morality in favor of wealth maximization.  Yet, at least according to this data, smaller stakes and thus greater separation leads shareholders to greater support for social responsibility (i.e., shareholder proposals, many of which trend along these lines). Which makes its own sense: shareholders with smaller stakes may identify more as laborers, community members, etc than as shareholders, and feel less of an economic hit when companies sacrifice profits to benefit these other groups.  Plus, these results may not mean the general principle is wrong, exactly; the argument was always that dispersed shareholders were absentee landlords, and Brav, Cain, and Zytnick find precisely that, in that lower stakes are correlated with lesser involvement in governance.  Still, these findings are some counterweight to the assumption that dispersion breeds lack of social conscience.

Beyond that, we can ask what these findings suggest for those who would seek to enhance retail shareholder voice.  For example, some have argued in favor of technological tweaks that would make it easier for retail shareholders to vote.  What would the effects be? Well, it depends on who you think these nonvoters are.  Perhaps, like the retail voters in the study, they’d turn out to be more attentive to corporate performance, less supportive of underperforming managers, and less supportive to shareholder proposals, than the current crop of voters – but again, as Brav, Cain, and Zytnick suggest, the infrequent voters are not the same as the frequent voters, and technological fixes may boost infrequent voters specifically.  Thus, changes that enhance retail participation may in fact result in greater support for ESG initiatives.

And then there’s the question of requiring mutual funds to pass through votes to beneficial owners.  That’s been proposed by a number of academics, including Caleb Griffin, Sean Griffith (for certain types of votes), Jennifer Taub, and Dorothy Lund, and has – as I understand it – even been suggested even by Bernie Sanders, who wants to ban asset managers from voting workers’ retirement fund shares unless they are following instructions (though it’s unclear to me whether he envisions straight pass through or more like a separate workers’ organization that sets voting policy).  And here we have the same question: We might imagine the silent retail shareholders would become more like the retail shareholders who vote today, in which case they’d oppose ESG proposals and would be more hawkish on management performance as compared to institutional investors, or they might be more like the infrequent/lower stakes retail voters of today, who support ESG but barely bother to vote at all.  Or they might be something entirely different: as Jill Fisch, Annamaria Lusardi, and Andrea Hasler note, investors whose only market exposure is a 401(k) plan are far less financially sophisticated than other investors.  Presumably these are mutual fund investors whose voices would be promoted by a pass-through regime, and they might have particularly idiosyncratic preferences, if they have preferences at all.

We might also ask whether the retail voters from 2015 to 2017 are reflective of the retail voters of 2020, or 2021.  As I previously pointed out, new technologies may encourage greater retail ownership, and these new traders may have entirely different preferences, especially if – as some data suggests – they treat stock trading as more of a leisure pastime than an investment opportunity.

But my big takeaway is this: I’ve previously argued that we should have more disclosure of the identity of voting shareholders, and, in particular, votes of high-vote shares and insiders should be distinguished from the votes of low-vote shares and unaffiliated investors.  Brav, Cain, and Zytnick have convinced me that retail shareholders have a distinct point of view, as well, and – to the extent consistent with these voters’ privacy – their votes should be disclosed separately, as well.

Ann Lipton | Permalink


Post a comment