One interesting subject at the intersection of law, finance, and human behavior is the impact of looming litigation and enforcement actions on the affected company’s stock price. This post is going to be a bit of a think piece, as truly fleshing out some of the concepts here would require not just more research but industry expertise and empirical modeling.
With that caveat, a core assumption of Wall Street traders, financial economists, and the investing public holds that new information, including new information on legal risk, will generally move the market if it’s material. Materiality in this sense is basically circular; it’s something that moves the market. That’s a pretty high bar. However, securities law and accounting rules (GAAP, which applies to public and nonpublic companies alike) treat materiality in the disclosure context much more liberally. Where the chance of a litigation loss is “probable” or even “reasonably possible”—terms that mean somewhere between 10-70% chance of a loss—it usually must be disclosed.
The gap between the market sense of “materiality” and the legal and accounting sense is critical. As this Perkins Coie note discusses, the SEC “has increasingly taken the position that it is not enough that a possible loss or range of loss cannot be determined ‘with precision and confidence,’ and has indicated that it may ask companies to provide support for an assertion that an estimate cannot be made, particularly as litigation progresses.” So the possibility of a litigation loss must be disclosed even if its probability and magnitude can’t really be estimated yet and many of the underlying facts aren’t yet known.
The drafters of the legal and accounting rules may not have intended a significant gap between their definition of materiality and the economic definition, but I suspect the gap was known. One policy goal behind a rule that promotes early disclosure of relatively improbable risks is to make it hard for companies to conceal the risk of losses, and thus to promote better cost internalization and risk management. That’s a crucial objective, but it is pregnant with tradeoffs. Accuracy is probably one downside, since the disclosure obligation kicks in before much is known about the probability or magnitude of the risk. It could be that early disclosure tends to exaggerate the market response, because the risks are simply impossible to price at the early stage they are disclosed.
Share price movements of public companies following high-profile scandals provide some opportunity to measure the ability of the market to price disclosed-yet-unresolved legal risks. General Motors, for example, is currently in the throes of an unfolding legal and regulatory scandal. The company announced a major recall February 7, 2014 and has announced nearly 50 more since, involving a total of 17 million cars in the US and a few million more abroad so far (here’s a timeline of events).
GM’s problems have a lot of legally and politically salient features beyond those of most corporate legal scandals. Grim deaths of totally blameless consumers, grieving families, a simple fix that would have cost pennies per car to implement. And GM’s new CEO, Mary Barra, has admitted that the company knew about these problems for a decade. Criminal and civil investigations are ongoing, and Congress is investigating.
Of course, at at time like this, the impact on the company’s share price is not the main concern, but looking at it might provide an interesting, rough-and-ready measure of the market’s perception of this newly-disclosed legal risk. The cost of the mechanical fix is currently expected to come in around $700 million, but the legal and reputational damage is harder to gauge. Is the market doing a decent job of pricing them, or are traders overreacting to dramatic events? And what if any role are the securities laws and accounting rules playing here?
Let’s try to assess the reaction first. As depicted below, between February 7 (the announcement of the recall) and yesterday’s close, GM shares (blue) significantly underperformed those of its chief US rival, Ford (in red), as well as the broader S&P 500 index (green). GM is essentially flat (+0.5%), whereas Ford is up 12.5% and the S&P is up 8.9%. This could be a recall effect.
But did the market go too far in punishing GM? In March, some financial journalists suggested that GM’s recalls might be unjustifiably depressing share prices and this week TheStreet rated GM stock a buy. Does this lend support to that view?
It’s not entirely clear (and the reasons why underscore how hard it is to get market predictions right). However, if current trends continue, the market may have seriously overestimated GM’s legal risk in April relative to the facts known at that time.
The case of GM involves a watershed event in the form of a series of damaging disclosures made in a short, discrete window of time, followed by a rapid and massive drop in share prices, followed in turn by a strong rally.
Adding more competitors makes for a richer comparison. In this second chart, GM (blue) is underperforming rival Daimler, but outperforming both Honda and Toyota, with Honda down over 4% in the same window: All these stocks are performing poorly in this period, except Daimler. It could be that GM’s weak share price is attributable to the weakness of the auto sector in general relative to the S&P. Even so, I think this window may be somewhat obscuring the share-price impact of the unfolding recall. At the risk of stumbling into empirical territory, I think it’s clear here that selecting the right time window is important.
I have three observations in that connection:
(1) In the first chart above, note how the dropoff for GM does not begin in earnest until about a month after the first announcement (it begins on March 10) and worsens for another month after that (until April 11). Two additional charts, depicting shorter windows, appear to give us some evidence that the market overreacted and then corrected itself.
(a) Here is GM (blue) versus Ford, Daimler, Honda, and Toyota during the period of GM’s most severe decline, March 10-April 11, 2014—GM is down 14% while performance for the other companies comes in between -8% and +4%. A bad month for GM shareholders.
(b) The story since April 11 is basically the opposite—GM (still blue) has now recovered virtually all of the losses it sustained during its month of pain. This might suggest the market came to the conclusion that GM’s legal, reputational, and other recall-related risks were not as great as originally assumed.
(2) Looking at a longer window reveals that something is weighing on GM shares this calendar year. My instinct is that these historic, immensely embarrassing recalls are playing a role, though of course other factors may be at play too. Here are the same five big carmakers, YTD—GM is doing very badly, with returns around 20-25% below two of its competitors in this five-and-a-half-month window:
(One interesting question is why Honda, down about 15% YTD, is doing even worse than GM (-11% YTD), but since the Japanese carmaker hasn’t had any legal disasters, I assume something more conventional is going on there; the unexpectedly strong Yen has probably played a role.)
(3) One safe conclusion here is that the recalls are continuing to weigh on GM’s stock price. But perhaps the more interesting thing to watch is whether the worst is over for GM investors. If the post-April 11 trend continues, that suggests those who bought GM stock at the bottom may have been better at judging GM’s legal exposure than those who sold in the preceding month. Here’s GM v. the S&P since April 11—GM has beaten the market by 75% (+13.69% versus +7.78%):
Necessarily, any conclusions we draw from this are not only tentative, but depend somewhat on the question we’re asking.
If the question is the obvious human one of whether it would have been wise to buy GM stock on April 11, that’s hard. I won’t venture investment advice (which I am obviously unqualified to do), but I think it’s important to underscore some of the downside risks here even if the stock ends up doing really well from now on. Myriad uncertainties exist today as they did on April 11, including what will happen if more deaths are revealed, a court determines GM cannot (or the company itself decides it will not) use a bankruptcy shield to reduce its liability, individual plaintiffs start receiving enormous jury verdicts, or C-suite officers are criminally charged. This is all is on top of the “ordinary” idiosyncratic risks of investing in individual companies, like the possibility that for unrelated reasons consumers start preferring other brands. On a risk-adjusted basis, therefore, it’s entirely possible GM has not been an especially good investment since April 11. And, as I suggest above, even if you made the right directional bet on GM on a given date relative to its cohort of large publicly-held automakers, it is possible that that cohort will underperform the market as a whole. It is hard to isolate the price effects of legal risk.
The question of the impact of securities law and GAAP here is much harder to discern. Crucially, these rules do not appear to have been observed here, given that GM knew about these serious problems for years and failed to disclose them. So arguably the rules didn’t have any effect on share price, until the risk of litigation losses was so material that any reasonable rule would require disclosure. GM is probably not the best company to use in trying to size up the impact of those rules on share prices. But if GM shares continue their current post-April 11 trajectory, the episode may provide an illustration of the market’s more general difficulty in pricing legal risk.
Disclosure: I have no financial stake in how this shakes out as I do not hold any GM stock, apart from index funds. However, I intruded on a few friends’ weekends on Saturday, April 5 with emails proclaiming that GM shares had dropped really far, guys, so my ego has a little at stake.
NOTE: updated 9/7/14 to add discussion of securities and accounting rules. No updates to GM stock performance or recall issues.