This piece in the Wall Street Journal reports on a recent article by David F. Benson, James C. Brau, James Cicon, Stephen P. Ferris regarding the language used in charters and bylaws of companies going public.  As described in the WSJ, they conclude that companies with shareholder-unfriendly provisions – such as, for example, staggered boards or supermajority voting – are inclined to “camouflage” this fact by using more obscure, harder-to-parse language.  And this effect is more pronounced for companies that can expect they won’t be caught – such as, companies with a smaller analyst following and fewer institutional investors.  They also find that companies that use camouflage reap benefits in the form of higher pricing.  I was intrigued by the description in the WSJ, and thought the findings might be a useful point of discussion in my Sec Reg class, so I tracked down the actual study.  But I found myself a bit confused by the evidence offered to support their conclusions.

[More under the cut]

Researchers have previously found that corporate managers use obscure, vague, and difficult-to-parse language when disclosing bad news. For example, Feng Li demonstrates that firms use harder-to-parse language when reporting lower earnings, and the effect is particularly pronounced when firm managers possess unexercised options.

In their new paper, Benson et al. explore the readability of charter and bylaw provisions. They conclude that the same effect applies – firms that limit shareholder power in their charters and bylaws are not anxious to advertise that fact, and use “obfuscation” – successfully – to hide it.

But I have several questions. First, they describe their key findings as including the fact that firms are more likely to use camouflage when they have “unfriendly” shareholder provisions (“We find the following key results: IPO firms that have less investor-friendly governance provisions employ more obfuscation in their charters and bylaws.”). But when I get to the body of the paper, I can’t find a single hypothesis related to shareholder-unfriendly provisions, or even a definition of what they believe that means (other than generalized examples of the concept).  The article has a lot of detail operationalizing the concept of “camouflage,” but nothing addressing the substantive content of the governance provisions in the sampled companies.  To put it bluntly, I see no evidence that the authors even studied camouflage as a function of the content of the corporate governance provisions.

Second, they appear to be inconsistent – even within the paper – about whether more or less scrutiny leads to camouflage. You can tell a story either way: more scrutiny means you need the camouflage more (because without it, people are more likely to notice the governance provisions), or more scrutiny means the camouflage will be less effective, so there’s no use bothering. They find that when there are more institutional investors, and more analyst coverage, there is less camouflage – comporting with the latter story – but they also find that when a firm goes public in a “cluster” that diverts attention from any particular firm, there is less camouflage – which comports with the former story.  Indeed, the paper itself offers contradictory stories within the space of a few sentences:

We argue that firms that IPO in the middle of a wave (i.e., when a lot of other firms are going public) have less need to camouflage. This could occur for at least two related reasons. First, it could be because in periods when there are many firms in an industry going public, the amount of attention or scrutiny that any particular firm receives is relatively limited, (i.e. the firm will blend in because it receives less particular attention). Second, it could be that as more firms in the industry use camouflage, it becomes seen as the norm (i.e., the practice no longer stands out because everyone is doing it.). Moreover, under this view, the perceived penalty for getting caught using camouflage may also be lower. Thus,we hypothesize that IPOs occurring amidst many other IPOs will be less likely to use camouflage than those going public during periods of few other IPOs…

If I’m understanding this correctly, their hypothesis changes mid-paragraph between the idea that there is less need for camouflage because of lower scrutiny, and the ease with which camouflage can be used without detection/penalty, again because of the lower scrutiny.  Indeed, they conclude by saying “Firms will be more likely to use camouflage during periods of greater IPO activity,” which – if I’m understanding – directly contradicts their earlier claim that “firms that IPO in the middle of a wave (i.e., when a lot of other firms are going public) have less need to camouflage.” They ultimately conclude, “Firms going public within a cluster … are more likely to be assessed by analysts or investors relative to the averages of the other issuing firms. This reduces both the need to camouflage and the potential gains from any camouflaging,” which does not comport with their earlier hypothesis that camouflage is easier to hide, and carries fewer penalties, in a cluster – perhaps reflecting a mid-study change in hypotheses?

Third, companies that go public with unfriendly provisions in their charters and bylaws frequently disclose those provisions in their prospectuses, identifying the provisions as a potential risk factor. For example, when the Carlyle Group tried to include an arbitration provision in its partnership agreements, it disclosed that fact in its prospectus, which led to a minor public uproar.  I randomly glanced at the S-1 of a recent company to hold an IPO, Houlihan Lokey, and it prominently identifies right in the risk factors that its charter and bylaws have “unfriendly” provisions such as supermajority voting and a staggered board. If these factors are identified clearly in the company prospectus, how much effect can camouflage in the charters and bylaws themselves really have?

Unless there’s something I’m missing – it’s not impossible, it’s Saturday morning and two Mardi Gras parades went thundering by my house last night – the authors have found a fascinating object of study, but they aren’t quite all the way there yet.