Saturday, January 16, 2016
There has long been a debate about whether corporations should be forced to disclose non-financial information about their operations, particularly information pertaining to social responsibility. For example, as Marcia Narine has repeatedly discussed, Dodd-Frank’s “conflict minerals” disclosure requirement may be not only ineffective to pressure companies into making ethical purchasing decisions, but may even be counterproductive, by causing companies to pull out of the Congo entirely (thus devastating the regional economy) rather than endure the expense of ensuring that their purchases do not indirectly finance armed groups.
Further to this issue, Hans B. Christensen, Eric Floyd, Lisa Yao Liu, and Mark Maffett have recently released a paper studying the effects of Dodd-Frank’s requirement that mining companies disclose information about their compliance with the Federal Mine Safety & Health Act of 1977. They find that after mine-owning companies became subject to Dodd-Frank’s disclosure requirements, they demonstrated a marked decrease in safety violations and injuries, counterbalanced by a decrease in productivity (apparently because they are spending more time on safety compliance). They also find that mines that disclose an “imminent danger order” from regulators post-Dodd Frank not only experience an immediate stock price reaction (especially the first time such an order is received, as compared to subsequent orders), but also experience a change in investor base – specifically, mutual funds (and particularly mutual funds that focus on “socially responsible investment”) divest their holdings.
I find these results fascinating for several reasons.
First – as the authors point out – mine safety information has always been publicly available, but hard to decipher. It’s located on the website of the Mine Safety and Health Administration in a searchable database. The Christensen et al. study is a nice demonstration of the principle that piecemeal, hard-to-decipher information has less of an impact than information compiled in SEC filings. Which, by the way, was exactly what the court held in In re Massey Energy Co. Sec. Litig., 883 F. Supp. 2d 597 (S.D. W. Va. 2012). There, plaintiff-stockholders of Massey Energy argued that the company misled investors about its safety profile. The defendants argued that information about their safety record was publicly available in the MSHA database. The court rejected this “truth on the market” argument, recognizing that difficult-to-parse mine-level data might not offset more prominent misstatements announced to the investing public. This new study validates the court’s reasoning in Massey.
Second, the paper seems to document a real, substantive effect of disclosure on companies' behavior. "Name and shame" apparently does work, at least in this context.
Third, the paper adds to the literature regarding the effects of investor “taste” on asset prices– representing a challenge to simpler models that imagine that asset pricing is purely a function of expected returns. Of course, it is possible that investors treat mine safety data as relevant to returns, to the extent that the higher levels of production from unsafe mines comes at the cost of a higher risk of an expensive mine disaster. But the fact that socially-responsible funds react more strongly to mine-safety disclosures than other funds suggests that taste is part of the story. (Caveat: I am confused as to how the authors selected their mutual funds for the study; if they compared socially responsible funds to all mutual funds, including index funds, the greater reaction of socially-responsible funds may be traceable to the fact that the socially-responsible funds are more likely/able to respond to news about individual companies).
In any event, none of this is to say that disclosure is the best way to regulate corporate behavior – direct command and control, or economic incentives, may be preferable for a lot of reasons, including the fact that disclosure only impacts public companies – but, well, it’s not nothing.