In his Corporate Governance Industry, Paul Rose focuses on the for-profit companies, including Glass Lewis and ISS, that have been established to develop corporate governance metrics, rate firms according to those metrics and, in some cases, advise companies how to do well in the ratings (henceforth referred to collectively as the CGI).
The paper’s first half describes what these firms do, putting their work in the context of the light state and federal (at least until SOX) regulation of internal corporate governance. The rest of the paper discusses potential problems and solutions. The problems include the obvious conflicts inherent in providing both ratings and advice; the dubious evidence on correlation between governance variables and firm performance; and the costs of applying inflexible criteria to diverse firms. The potential solutions, about which Rose is appropriately wary, include government regulation of the CGI.
I liked the paper's topic and approach, as indicated by the fact that I noted it on my blog even before becoming involved in this Forum. I found the paper’s descriptions useful and criticisms accurate, as far as they went. My comments will focus on what more I’d like to see in the paper. (Though the paper is plenty long enough, the author could make room by significantly cutting the descriptive part of the paper.)
First, what’s the problem? I don’t like how Wheaties taste, but I would defend to the death consumers’ right to buy them. Should we take the same approach to issuers’ and investors’ purchase of products from the CGI? One can’t criticize the CGI’s output like one would criticize, say, the SEC and Congress, whose edicts are forced on firms. So, assuming for the sake of argument that Rose’s criticisms of the CGI are apt, why would anybody buy their products?
Perhaps issuers buy faulty advice because the CGI effectively controls a sizeable slice of institutional investors’ votes. Directors like to be reelected. But then why are institutional investors voting per advice that we’re assuming is faulty, given that their portfolios could suffer and their customers would leave? Is it that they figure that, even if ISS is wrong, this won’t affect relative returns enough to drive away customers? On the other hand, if they don’t vote as ISS says, they open themselves to criticism. So, in effect, the investors are buying “criticism insurance.”
But then we should ask who would criticize institutional investors not following assumedly faulty advice? Well, among others, Gretchen Morgenson and other journalists of her ilk. Morgenson in particular, as I've discussed at length, has repeatedly advised mutual fund investors to sell funds that are not acting according to her CGI-influenced criteria. Investors probably won’t sell funds that are performing well even if their managers are voting “wrong,” but what if the fund isn’t performing well, for whatever reason, or the investor is choosing a fund in a very crowded field in the first instance? Surely a prominent NYT columnist achieves the sort of salience in investors’ minds that the behavioral finance people are telling us matters to investors’ decisions.
Let me emphasize that I’m not suggesting that Rose reach my particular, perhaps idiosyncratic, conclusions, but only that these are questions he might ask.
Second, assuming CGI advice does affect governance even if it’s faulty, I’d like to know more about the costs of this faulty advice. Rose alludes to these costs, but doesn’t elaborate. For example, he suggests that the CGI is inhibiting innovation or adaptation to specific circumstances. Some examples would help elucidate the theory. More importantly, I’d like to see testable hypotheses. For example, we might be able to estimate the costs of faulty CGI metrics by looking at what happens to the value of companies that follow CGI advice or where governance changes otherwise can be attributed to this advice.
Finally, I’d like more on the paper’s regulatory implications. The paper does a good job discussing the costs and benefits of regulating the CGI, but what about the paper’s implications for regulation of corporate governance generally? My own take would be that criticisms of CGI’s metrics demonstrate the folly of mandating particular criteria. Also, the existence of the CGI suggests, as Henry Butler and I discuss in our AEI book, that even if we like some of the stuff that SOX did, SOX wasn’t necessary.
Again, Rose doesn’t have to agree with these conclusions, but might ask the question. After all, he does spend some time in the paper talking about how the CGI fits with the existing matrix of corporate governance regulation. I’m suggesting that he connect the dots: what does the CGI tell us about what the matrix should be?
So, in general, the paper is a worthwhile contribution that could be developed into something even more consequential.
TrackBack URL for this entry:
Links to weblogs that reference Larry Ribstein on The Corporate Governance Industry: