Some of my favorite bloggers (Ribstein, Josh Wright and Geoff Manne (aka TOTM) are rather harshly attacking Matt Bodie's recent post about options backdating. Matt believes that options backdating allowed executives to extract more pay than they would have otherwise. TOTM argues that prices (incl both stock market and labor market prices) are set by a mostly efficient market, so talk about "pay norms" is irrelevant and counterproductive.
Hmm. What would have happened if firms hadn't backdated? Would executives have simply left for competitors where they would get paid their market rate? Certainly not in all cases -- labor markets are sticky and lumpy. It's not costless to switch jobs, and a new job will have unique duties that may or may not match up with your current job. Options backdating wasn't just about recruiting or retaining talent. Sometimes it was about executives manipulating documents -- departing from shareholder-approved plans -- to maximize the value of their compensation.
I think Matt should be praised, not vilified, for raising the issue of norms. Still, TOTM is right that we need more precision for "norms" to have explanatory power. So let's talk about it and work towards precision, and I'm happy to refrain from calling anyone greedy.
Yes, market forces explain a huge chunk of executive pay. But other things (like institutional practices, path dependence, regulation) also matter. Moreover, market forces are better at explaining the amount of pay than the form of pay. What, then, are the implications of options backdating? Is it merely a technical footfault? A victimless crime? Or is it something we should be concerned about?
I'll be exploring options backdating in a new tax paper; what follows are my preliminary thoughts. Civil comments and feedback would be appreciated, on or off line.
Culture of Noncompliance. Options backdating matters not just because proper disclosure might affect the form and substance of executive pay (as Matt and those other zany folks like Christopher Cox and Jeffrey Gordon would argue) but also because the manipulation may reflect a culture of noncompliance. A company that is willing to violate one set of rules (however silly those rules may be) is more likely to violate other, not-so-silly rules. Options backdating is a perfect example of overly-aggressive regulatory gamesmanship. It not only violates accounting rules, but, in many cases, tax law. Say what you will about accounting fraud, but can one really call tax fraud a victimless activity? Am I guilty of fetishization of tax law because I expect companies to pay taxes?
And I don't think it's crazy to think that "technical" violation of rules matter. Lots of rules are ticky tack. Most of securities, ERISA, 40 Act, and corporate finance law, taken piecemeal, is ticky tack. But companies should comply with the law, and when they don't it's the sign of a potential problem.
The Reverse Willie Sutton Fallacy. Ribstein and TOTM are willing to dismiss the stealing-from-shareholders argument because an efficient market ought to be able to look through the accounting treatment. This is the reverse-Willie-Sutton defense: it can't be stealing if there isn't any money in the bank. (This reminds me of the immortal words of H.I. McDonnough (Nicolas Cage) in Raising Arizona, "Nawww, honey, it ain't armed robbery if the gun ain't loaded.")
The empirical evidence here is mixed (look for an upcoming paper from BU taxprof David Walker). Most accounting experts believe, contrary to what Ribstein and TOTM suggest, that accounting does matter, at least to some firms. The inflated earnings from options backdating, for example, allowed some firms to avoid violating debt covenants; it allowed managers with earnings-based compensation bonuses already in place to get higher bonuses. There was a little money in the bank, and some executives took what they could when they thought no one was looking.
Tax Fraud. Regardless of where you come out on accounting compliance and the stealing-from-shareholders debate, I fail to see a reasonable argument for dismissing the tax consequences of options backdating as a victimless activity.
[A brief tax primer: Section 162(m) requires that a corporation may take a deduction for compensation in excess of $1 million, with respect to covered employees, only if the compensation is performance-based. Companies normally take a deduction for the spread between the exercise price and strike price of an option. But options are performance-based -- and the deduction allowable -- only if "the amount of compensation attributable to the options ... would be based solely on an increase in the corporation's stock price" (Committee Report; see also the 162 regs). In the money options don't count, since the stock price could decline and yet still provide compensation to the employee. The Committee Report also notes that the maximum number of options that an executive may receive during a specified period must be predetermined.
In simple terms, then, the rule is that in-the-money or backdated options are not deductible with respect to covered employees. (I should note that if in-the-money or restricted stock grants are subject to performance-based vesting rules, it's possible that the deduction is okay. Vesting based solely on continued employment is not enough. See 1.162-27(e)(2)) Additionally, to qualify as performance-based, the material terms have to be determined by a compensation committee of outside directors and disclosed to shareholders prior to payment. (162(m)(4)(C)).]
In the typical case, then, it appears that companies that were backdating options were also violating the tax law. This is stealing from the government. (Or, if you prefer, stealing from you and me by depleting the public fisc.) In the best case, companies may have forgone the deductions in order to comply with the tax law, in which case they are guilty of bad tax planning, but not tax fraud. (From a tax perspective, in these cases it would have been better to provide more at-the-money options rather than fewer in-the-money options.)
Wherever you come out on the accounting debate, I fail to see how you can dismiss the scandal, as Manne does, by saying "There's just no harm in the practice." Regulatory compliance matters. Should all these executives go to jail? I doubt it. But ignoring the problem would be irresponsible.
The next interesting question in my mind, then, is whether, as I assumed above, companies that violate some rules are more likely to violate other rules. Common sense suggests that they are. Cf. Enron, WorldCom. On the other hand, companies like Google and Apple have violated some technical rules (remember the Playboy interview by the Google founders?) without big skeletons lurking in the closet. So what's going on?
Cultural Differences. My tentative theory is one of cultural differences. Some companies rely regularly on outside counsel for advice. These sorts of companies (Google, Apple, Microsoft) might make regulatory mistakes, but when they do, they fix them. They work within the legal system. They seek the advice of counsel. They conduct internal investigations. Other companies (Enron, Comverse) disregard the advice of counsel, sift through gatekeepers until they find the one with the answer they want, distort the information they provide to counsel, or simply act behind their backs. I suspect that options backdating, like tax shelters and other forms of regulatory noncompliance, cannot normally be accomplished with the assistance of informed and reputable counsel.
It's worth noting that, as with Enron, the tax fraud here was collateral fallout from the accounting gamesmanship. Any tax lawyer would have spotted the issue. So what happened? We're not talking about complicated corporate tax shelters that have some slim chance of success in the courts and are worth a lot of money to the company. We're talking about petty noncompliance, picking up nickels in front of bulldozers just because you think you deserve that extra nickel. As an outsider, I don't care who owns the nickel. It's the lack of judgment that troubles me.
What causes the cultural differences? That's a harder question. There's some anecdotal suggestion that tech companies were more likely to be cavalier. (On the other hand, they may have just had more to gain given the volatility of stock prices for these stocks; backdating was worth more to them.) The practices may have spread through networks of directors serving on comp committees, or the networks of executives demanding it. This is something I'll be looking into more carefully in the next few weeks as I work on this paper.
TrackBack URL for this entry:
Links to weblogs that reference Options Backdating and Compliance Norms (Nawww, Honey, It Ain't Armed Robbery if the Gun Ain't Loaded):
1. Posted by Mike Guttentag on September 4, 2006 @ 14:21 | Permalink
I would add one refinement that is particularly ironic for the TOTM/Ribstein folks.
Part of the Manne/Ribstein/Wright argument appears to be that requiring the disclosure of the particular date on which the option strike price is set provides little or no benefit to the firm. This is an issue that Matt addressed by talking about norms, but it is an issue that can also be addressed by looking at the history of disclosure requirements. Manne/Ribstein/Wright point to the fact that share prices either reflect information about executives’ option strike prices, in which case it is irrelevant what the timing of the option dating is, or, as Ribstein argues, share prices don’t reflect this information in which case the firm’s shareholders may be better off. Arguments about share price accuracy with respect to requiring disclosure of these types of transactions are beside the point (particularly since the amounts involved are unlikely to be material).
The scholarship on disclosure requirements suggests a different rationale for these particular disclosure rules. It is customarily argued that there are two distinct reasons that firms adopt disclosure rules – accuracy enhancement and agency cost reduction, and that these two justifications for disclosure rules suggest that different kinds of information should be required to be disclosed (a distinction that I reject as overly simplistic, but that is another matter). Paul Mahoney in his 1995 Chicago Law Review article, Mandatory Disclosure as a Solution to Agency Problems, shows that at least from a historical perspective reducing agency costs is a significant factor in determining why firm disclosures are required and what types of information firms are required to disclose. Mahoney’s describes how disclosing information about transactions between the firm and its agents, which he calls agency information, has a long history. Presumably, investors who are not aware of transactions between the firm and its agents would need to discount security prices to reflect the extent to which the firm’s agents would be expected to enter into transactions that benefit the agents more than the firm. By credibly committing to disclose any transactions between the firm and its agents, the agents are providing a means to assure investors that funds will not be inappropriately transferred out of the firm. With the disclosure of these transactions, investors would be put on notice and provided information that facilitates heightened scrutiny. Further, we would assume the firm’s owners have some way to take action if dissatisfied with the transactions disclosed, and so the agents would be discouraged from entering into these types of transactions. The value of the firm increases.
Why does this add to the irony of the TOTM/Ribstein argument? A rule that requires the disclosure of the timing and amount of options was probably adopted to achieve these agency cost reducing benefits. If this is correct then this rule, while imposed by regulatory intervention, is not significantly different from a disclosure rule that the firm would choose to commit to ex ante. The culprit here is not overreaching government regulation, but the original investors in the firm who choose to adopt a set of disclosure rules to maximize the value of the firm.
2. Posted by Jake on September 4, 2006 @ 14:45 | Permalink
Backdating a document is immoral. The extent to which the law should make such immorality criminally punishable is debatable, and the current debate among TOTM and opposite views is valuable.
But it is wrong to dismiss option backdating as immaterial. Minor wrongful conduct, if not recognized and combatted, fosters greater wrongs. Consider the well documented story of the NYPD's shift under Giuliani toward a policy of more vigorous investigation and prosecution of supposed "minor" and "victimless" crimes, which ultimately led to an overall decrease in the NYC crime rate, more serious crimes included.
I wonder whether executives who backdate their own stock options brag about it to their mothers. ("Mom, you'll never believe what I pulled off on the stockholders today!")
3. Posted by Bill Sjostrom on September 5, 2006 @ 6:56 | Permalink
Another tax issue here is recipients of backdated options improperly claiming incentive stock option treatment under IRC 422.
4. Posted by Beth Young on September 5, 2006 @ 7:50 | Permalink
Having worked with/among institutional investors, at least on the voting (as opposed to investment) side, I find the TOTM arguments to be out of touch with the way shareholders approach option compensation. If shareholders had known that companies were granting in-the-money options, there would have been substantial resistance to approving any plans put up for a shareholder vote. ISS and other proxy advisors, as well as the big institutions that vote on plans, began to scrutinize "shareholder-unfriendly" practices such as repricing, in-the-money options and reload features pretty closely in the mid to late 1990s, as dilution soared and the amounts realized by executives did too. There's just no way the voters would have thought the way Manne and others at TOTM posit. Perhaps portfolio/fund managers would, but at most institutions those folks have very limited input into voting decisions on equity plans. So disclosure of option backdating would have significantly impaired the ability of companies--perhaps all companies until the extent of the practice became known--to get any more shares into equity plans. In the end, investors would have pressed even more companies to prohibit the practice in plan documents.
The other thing I think is being overlooked is that option grants have informational value for investors. Executives who are consistently receiving in-the-money rather than at-the-money or out-of-the-money options, would generally be viewed by executives as having a bearish outlook on the stock. That might not actually be true, but nonetheless would be the perception. Misleading investors about the exercise price of options interferes with this mechanism.
5. Posted by Jeff Lipshaw on September 5, 2006 @ 17:23 | Permalink
I have posted on this ad nauseum over at PrawfsBlawg, so I won't repeat it. I was persuaded by Holman Jenkins' description of the reasons for the backdating, and that strikes me still as the most plausible: managements were (and I assume still are) HIGHLY motivated by even small reported earnings impacts, even if pure theory would say it doesn't make any difference. (My guess is that is a statement either about the absence of perfect information, or the perception that there is not perfect information.) If you are not expensing, options are a way of paying compensation without showing a hit to earnings. So to agree with Larry Ribstein, it's not an issue of greedy CEOs trying to gild the lily. If you can accomplish the recruiting result (which benefits the shareholders) in a way that doesn't cause you to miss your guidance number or your whisper number, it's not irrational to do it.
I'm not convinced there's even an agency cost issue here, because we are not talking, if Jenkins is right, about motivations or interests as between agents and principals that are out of alignment. His point is that you could be backdating the options because that's what you think you should be doing in the best interest of the company and its shareholders.
The real problem to me, as I've said before, is something akin to the comment about the reduction of minor crime and its impact on major crime. If the rules allow you to do X without even a technical lie, no problem. But when the rules cause you an earnings hit that you can avoid with a technical lie, how do you then distinguish or regulate other technical lies within the organization? This is not an area of gray standards (e.g. the imposition of insider trading windows, which are all over the lot, or the "reasonable people can differ" judgment calls you make within GAAP accounting all the time). This is a rule, the effect of which you avoided by deliberately falsifying a record (a mere technical lie, someone called it), notwithstanding its materiality. If you falsify a document on emissions data, knowing that the legal limit is essentially arbitrary, and the only impact of being .001 above the limit as opposed to being .001 below the limit is the EPA investigation you are going to provoke, with no ultimate benefit to the company, the shareholders or the environment, should you be morally, legally or ethically absolved from the technical lie on the grounds that it was an arbitrary rule to begin with? I find that a troubling rationale.
6. Posted by Elizabeth Brown on September 6, 2006 @ 17:05 | Permalink
The New York Times has an article in today's paper that discusses a new article on the cost to shareholders of options backdating. See http://www.nytimes.com/2006/09/06/business/06options.html?_r=1&dlbk&oref=slogin
The New York Times article notes:
"Three researchers at the University of Michigan estimated that backdating stock options between 2000 and 2004 helped sweeten the average executive’s pay by more than 1.25 percent, or about $600,000. But the fallout from the recent options investigations has caused those executives’ companies to fall in market value by an average of 8 percent, or $500 million each.
'For about $600,000 a year to the executives, shareholders are being put at risk to the tune of $500 million,' the study concludes.
The working paper, expected be made public later this week and to be published in The Michigan Law Review next year, appears to be the first to put dollar figures on the costs and benefits of backdating. It analyzed thousands of stock option grants at 48 companies that announced they were under investigation as of the end of June, and measured the maximum gains for those executives if their options were backdated over a 90-day period as well as the drop in value at those 48 companies in the 10 days before and after news of a backdating inquiry was released."