Guest Post: The Professors Respond
A recent article by three academics raising the question whether corporate securities lawsuit defendants underperform financially after their case settles has generated significant commentary on this site. In this post, the professors respond to the commentary.
The article in question is a March 18, 2010 paper entitled "Lying and Getting Caught: An Empirical Study of the Effect of Securities Class Action Settlements on Targeted Firms" (here) by Cincinnati Law Professor Lynn Bai, Duke Law Professor James Cox, and Vanderbilt Law Professor Randall S. Thomas.
The article describes the professors’ research in which they sought to discover whether getting hit with a securities a lawsuit and then subsequently entering into a settlement "weakens the defendant firm so that from the point of view of well-received financial metrics the firm is permanently worse off as a consequence of the settlement."
My initial post about the article provoked an unusual amount of reader commentary, including a comment about the academic’s research and analysis posted by former plaintiffs’ securities attorney, Bill Lerach. With Mr. Lerach’s consent, I republished his comment as a separate guest post, here.
The professors have prepared and submitted a response to the various comments about their paper. Here is the professors’ response:
The comments seem to have concentrated on the possible alternative causation of the underperformance of defendant companies involved in securities class actions, i.e., these companies had financial problems prior to the lawsuit and it was likely that these pre-existing problems prompted the companies’ management to lie, and thus, it should not be surprising to see that these companies underperform their peers after settlement. We do not deny that this could be an alternative explanation to the underperformance that we have seen in the data, and indeed we have explicitly talked about this alternative explanation at a number of places in our paper. However, our study has also revealed empirical evidence that is inconsistent with this intuitive explanation, and we thought we should report such evidence in the paper so that people can think about them and perhaps follow up with more research.
First, we are not seeing deterioration (post-lawsuit and post-settlement compared to pre-class period) in the defendant firms’ sales numbers. This holds true both in terms of defendant firms’ absolute sales numbers and relative performance to their peers. We know that sales reflect the bottom line of the financial health of a company and the robust sales shown in the data are inconsistent with the story that these firms are deteriorating on their own independent of the lawsuit.
Second, we are seeing deterioration in liquidity in post-settlement period but not in the post-lawsuit-but-pre-settlement period. If the liquidity constraint is caused primarily by other factors such as banks’ withdrawal of credit as a result of revelation of fraud (as Bill Lerach suggested), why are we seeing significant constraints only in the post-settlement period?
Thirdly, although the average Altman Z-scores for defendant firms were lower in post-lawsuit and post-settlement periods compared to the pre-class period, the inferiority was more prominent in the post-settlement periods. The significantly lower Altman z-score in post-settlement periods seems consistent with the heightened liquidity constraint we observe in the post-settlement period. Again, we do not rule out the possibility that defendant firms are deteriorating on their own, but we want to point out that the data can also be consistently explained by an alternative hypothesis, i.e., lawsuit and settlement had an independently negative effect on the financial health of the defendant firms.
The comments are legitimate and we appreciate the interest that people have shown in this topic. We have certainly thought about the causation issue in our research, but we could not explain completely what we were seeing in the data by the hypothesis that defendant firms were destined to underperform even if they were not dragged into a lawsuit. We have dutifully reported what we saw in the data.
I would like to thank the professors for taking the time to prepare a thoughtful response and for their willingness to have the comments posted on this site. My thanks to all of the readers who have engaged in this dialog. Further comments are still very much welcome.
Isn't this simply a verification of the perception that settlement is evidence that a defendant business is acknowledging culpability for the allegations that were the basis for the commencement of the litigation, notwithstanding it's protestations to the contrary?
This residual effect, if any, is just the "Liar's Discount" getting baked in.
What would be interesting additional research is to look at the existing database and ascertain whether there is a difference in the longterm "effect" where as part of the settlement of the securities fraud class action, companies adopted MEANINGFUL corporate governance changes tailored to address the governance failures that permitted the fraud. And/or companies that on their own (or as part of the settlement) replaced almost all of their senior-most executives. It is hard to shed the Liar's Discount where you're going on with "business as usual."
Let me break down the professors' flaws one by one. First, they say "We know that sales reflect the bottom line of the financial health of a company and the robust sales shown in the data are inconsistent with the story that these firms are deteriorating on their own independent of the lawsuit." Not true. Sales are a "top line" metric and many many many book-cooking schemes focus on expenses, margin, net income and earnings. Improper capitalization of expenses is one of the go-to schemes that are easy to cover up and difficult to expose. Corporate fraudsters love making the argument that revenues were immaterially changed by the fraud, or that cash flows were not impacted--But the issue is income and earnings, both of which can be falsified without touching revenue. Thus the more appropriate metric to look at would be if the net income or earnings per share go down following litigation and settlement. Any data on that, professors?
Second, the professors state that "we are seeing deterioration in liquidity in post-settlement period but not in the post-lawsuit-but-pre-settlement period." You don't say! That's because settlements are mostly cash outlays, plus insurance (thus increasing the price of D&O thereafter), not to mention that internal and external investigations cost millions (which subsidizes accounting firms and law firms who have a disincentive to find fraud, but that's another story). Also, litigation costs money. That has nothing to do with securities cases. And with the PSLRA, the costs are minimal until plaintiffs can plead a real case with real facts, thereby making it less likely that a bogus suit will get by. Thus, most cases that survive are likely real cases with real frauds. Which means that the settlements are not just a factor of avoiding litigation, but also avoiding the exposure of their fraud. The liquidity comment has no merit.
Third, the money spent on internal investigations, firing executives (and retraining others), cleaning house, and establishing real internal controls is expensive--and it should be. But it does not mean that the mere filing of a securities suit results in heightened liquidity pressure. See Point #2 above.
In conclusion, the professors see the results they want to see. They, like most, have an innate bias against "greedy plaintiffs' lawyers" and securities actions in general. They have drunk the kool-aid provided by the rich corporate complex and its enablers (Congress; the Bush administration; parts of the Clinton administration) and will always find a way to attack the plaintiffs' bar. Nothing new, not surprising. This is why these comments are important, so these guys can be called to account for their obvious biases.