In recent years, and particularly since the Supreme Court’s decision in Citizens United v. FEC, 558 U.S. 310 (2010), there have been increasing calls for the SEC to require public companies to disclose their spending on political activities.

The situation is complex because while there may be many reasons for transparency on the subject, it is difficult to tie disclosure specifically to the needs of investors as investors.  Most political spending is likely undertaken by companies to benefit the firm itself – that is, in fact, precisely why people find it objectionable – and it is difficult to articulate why investors as investors (rather than, say, as employees or as citizens) should care about political spending any more than any other ordinary business decision for which we have no required disclosures.

The SEC has resisted increasingly loud calls that it regulate in this area, likely due to this precise problem.  In December, Congress passed a budget that actually forbade the SEC from using funds to regulate political spending in the following year, though that has not ended the matter for Democrats.  (Interestingly, I note that it was only after the budget had passed both Houses that there started to be any real press on the subject and the issue still didn’t get much press traction until after the budget was signed into law. I’m no political expert but if I had to guess, I’d say that the reason for the relative stealth was that the SEC supported the measure as a way of alleviating some of the political pressure on itself).

In any event, over the years, there have been a variety of attempts to show that political spending is/is not beneficial to investors, often with a view toward providing a solid foundation for SEC regulation.  One study, Corporate Lobbying and Fraud Detection by Frank Yu and Xiaoyun Yu, found that political activity – namely, lobbying – helps firms conceal fraud.  Based on class actions filed between 1998 and 2004, and class period length, they concluded that firms that engage in political lobbying managed to keep their fraud under wraps for longer periods than non-lobbying firms.  They also found that lobbying expenses increased during fraud periods.

Which is why I read with interest a new study by Matthew McCarten, Ivan Diaz-Rainey, and Helen Roberts that extends Yu and Yu’s original research.  According to the authors, the Sarbanes Oxley Act has significantly mitigated the impact of lobbying.  They find that post-SOX, lobbying is no longer correlated with longer class periods/fraud detection; they attribute the change, at least in part, to various SOX corporate governance provisions, such as the requirement that the CEO and CFO certify SEC filings, and increased whistleblower protection.

These results are intriguing, though I have some reservations.  The authors treat any action filed in 2005 or later as post-SOX (to account for delays in SOX’s implementation since its passage in 2002).  But this was also a period of great upheaval in securities class actions.  In 2005, the Supreme Court decided Dura Pharmaceuticals, Inc. v. Broudo, 544 U.S. 336, which introduced loss causation as a new and heavily litigated variable in class actions – and one that can dramatically affect the length of a class period.  Moreover, over the years, courts have become increasingly strict in their analysis of Section 10(b) pleadings: As Hillary Sale documented, requirements for pleading scienter have ratcheted up over the years (and I’d argue that the standards have only tightened since her paper was published; among other things, confidential sources have become de rigueur).  Cases like Stoneridge Inv. Partners, LLC v. Scientific-Atlanta, Inc., 552 U.S. 148 (2008) and Janus Capital Group, Inc. v. First Derivative Traders, 564 U.S. 135 (2011) – as well as their precursors in the circuits – have made it increasingly difficult to bring claims not only against secondary actors, but also against corporate executives who do not directly speak to the public.  The combined effect of these developments make me question whether class action data – especially class action data that purports to measure changes over time – is a reliable proxy for fraud.

That said, none of these developments has anything to do with lobbying, so theoretically, they may not affect McCarten et al.’s findings.   The paper is an interesting one, both for what it adds to the debate on disclosure of political spending, and also for its implications about SOX’s effectiveness (which was, at the time of its passage, famously described as “quack corporate governance”).