In March 2000, after reviewing the results of a clinical study of the pain-reliever Vioxx, Merck’s research chief, Edward Scolnick, typed an email. In it he said that evidence of heart-related risks from Vioxx was “clearly there”. Merck did not mention Scolnick’s conclusion publicly, however, and it kept selling Vioxx for four more years.
Several months ago, after Merck withdrew Vioxx from the market and Scolnick’s email was revealed, the events seemed a clear case of fraud or negligence: Merck knew its drug was risky and didn’t say so. Plaintiffs’ lawyers pounced, and criminal and regulatory investigations began. Subsequent explanations, however, have shown that the case might not be clear.
In recent years, email has been heralded as a miraculous new detection tool, one that enables regulators to peer through public statements and see the “unvarnished truth” beneath. The impact of email on corporate reputations and pocketbooks has indeed been breathtaking, but not always because it reveals the truth. Sometimes, email doesn’t reveal anything more than the appearance or possibility of impropriety. This appearance alone, however, can create massive liability.
Warning beacon
Before going further, I should acknowledge the obvious: on this topic, I’m not just an armchair observer (please see below). I can’t discuss my own ghastly email experiences, unfortunately, but I hope they at least stand as a warning beacon. I should also draw a distinction between the role of emails in cases like Merck’s and their role in cases like the insurance bid-rigging scandals, in which emails were actually used to commit alleged crimes.
As the Merck case demonstrates, the meaning of any remark depends on its context: specifically, the surrounding circumstances, conversations, and events, which might or might not also have been preserved. If Scolnick’s email represented his professional opinion about Vioxx in March 2000, then Merck probably did commit fraud (it depends whether Scolnick’s conclusion was synonymous with Merck’s). If the email merely captured Scolnick’s first impression, however, it could be innocuous. The hurdle for expressing a reaction is far lower than that for expressing a formal professional opinion, and it is not fraud to think or say one thing and then conclude another.
According to Merck, Scolnick’s email represents the latter: his initial impression of the Vioxx study. Merck says Scolnick sent the email immediately after he received the results of the study and then, subsequently, after reviewing more data, he concluded (wrongly it turns out) that the drug was safe after all. In this version of events, Scolnick might have made a regrettable decision, but he didn’t commit fraud.
The sobering lesson for executives is that it doesn’t really matter what Scolnick’s email means. Even if Merck’s explanation is true, the damage has been tremendous. Merck’s stock has been scalped, the company has set aside $675 million in defence-cost reserves, and a recent “Vioxx litigation update” ran to more than 1,100 words. The case is more complex than the Scolnick email, but, so far, the email is the smoking gun.
The difference between email and a conversation is that, when a conversation ends, both the words and context disappear. Email exchanges often feel like conversations, encouraging us to be less formal than in, say, memos, but when the email exchanges end, only the context disappears.
This would be fine, except that the realities of our legal system and human psychology are such that, once lost, this context may always be open to debate. Memories fade and, over time, hindsight bias can make even true intentions hard to believe (now that the Vioxx risks are obvious, it is hard to imagine that they weren’t always so). In corporate litigation, moreover, defendants often do not get the benefit of the doubt. The risk/reward often favours settling, so subjective interpretations can be as damaging as reality.
Choose carefully
Executives have always had to choose words carefully, but the introduction of email has taken this prerogative to another level. If Merck’s explanation of Scolnick’s email is accurate, it is hard to see how he could have foreseen that, by expressing his impression of the Vioxx study, he was doing something risky or wrong.
The rules for safe email usage are clear: no humour, no irony, no hyperbole, no sarcasm, no debate, no errors, no reactions, no doubts, no blowing off steam – in other words, none of the stuff that makes up the bulk of human business interaction. Email etiquette, however, encourages us to just keep on typing.
What can we do to protect ourselves? Short of studying the above rules, not much. Email seems on the surface to be a remarkably cheap, convenient technology, but it can have a major hidden cost. Pressing the “send” button, unfortunately, may always bear a distant resemblance to Russian roulette.
