Backdating did harm investors.

Bill Sjostrom —  22 September 2006

Three Michigan B school profs have a new paper up on SSRN entitled “The Economic Impact of Backdating Executive Stock Options.� The paper adds some important data to the backdating debate. Specifically, the paper looked at 45 firms implicated in the backdating scandal and found that over a 21-day period surrounding the revelation of backdating, the average cumulative abnormal return of the stock of these firms was approximately negative 8%. It also found that the average market capitalization loss per firm during the period was $510 million. In light of these findings, I think it is now untenable to argue that backdating has caused little or no harm to investors. Yes, the monetary effects of backdating were timely disclosed and promptly incorporated into share price. However, as I noted in a comment to this post and as alluded to in the paper, the drop in price likely reflects reduced investor confidence in the firms’ management and internal controls exacerbated by the media frenzy and anticipated diversion of firm resources to deal with internal and external investigations, damage control, etc.

4 responses to Backdating did harm investors.


    color me skeptical. All of the information about option grants had been generally available for years. The footnotes to the financials provide all the information investors need as to the economic effects of options—no. outstanding, avg. weighted exercise prices, remaining term. The markets didn’t particularly care. What markets do care about is uncertainty and that is what the publicity and regulatory overreaction to the various practices provided. Potential changes in management and litigation costs will drive down stock prices. My guess is that if you looked at stock prices of the affected companies 90 days after the announcement, most of the price drop would have been recovered.


    It goes without saying that this study obviously contributes, like Walker’s, to our knowledge about what is going on with backdating: who is doing it, who isn’t, in what industries, etc. The star of the show, however, is obviously not this valuable descriptive analysis. The $500 million per firm gets the headline, for obvious reasons. A few quick thoughts/ reactions.

    1. $500 million per firm!!!!? Really? Context and common sense suggest to me that this estimate is implausible in terms of a loss to shareholder value *caused* by backdating alone. To their credit, the authors are careful to note that they are measuring the impact of the revelation — which is often associated with the sort of investigation/ litigation, etc. that Bill refers to in his post.

    Just for a reality check, multiplied by the 100 or so firms implicated in the backdating scandal, this paper suggests that the cost of backdating to shareholders was about the size of the GDP of Morocco. I’m not saying it isn’t possible, but I am saying that I would need a very solid identification of an isolated “backdating effect.”

    2. I’m not so convinced this event study can carry the weight of its punchline result if we read the result to mean that backdating caused $500 million per firm in shareholder losses (which I think is how it will be received). As Larry points out at his blog, atttributes all abnormal returns in the 21 day period surrounding the backdating as attributable to the announcement, i.e. the control group is the market as a whole. Why not control for within industry effects? This seems like an obvious way to add some information to the estimates — and is not too much labor added. As an aside, David Walker’s analysis of the semi-conductor industry is set up perfectly to run this sort of test, and would add some value to his analysis.

    3. Finally, and in any event, we are still left with the problem of disentangling whether the losses imposed are due to backdating itself or, as Bill suggests, “an anticipated diversion of firm resources to deal with internal and external investigations, damage control, etc.” I suspect that there are ways to tease out a true measure of the impact of backdating as to be distinguished from litigation/investigation/diversion of resources … but that is for another post.

Trackbacks and Pingbacks:

  1. Sunday links: de-levering LBOs « Abnormal Returns - September 24, 2006

    […] Bill Sjostrom at Truth on the Market points to a paper that shows the effect of the options backdating scandal on company stock prices. […]

  2. Ideoblog - September 23, 2006

    The harm from backdating?…

    I said yesterday that the key question about backdating is what happened to the affected firms’ shares when the backdating was disclosed or otherwise became known. I cited some anecdotal evidence that there was no such harm. Bill Sjostrom links…