An IRS Conspiracy? Not Likely ... Yet

Yesterday, I wrote about the news that the Internal Revenue Service could not recover two years' worth of e-mails from Lois Lerner's hard drive. Almost as soon as the post went up, I saw the National Review reporting that it wasn't just Lerner: Documents from six other people, including the former chief of staff to the commissioner, are also unrecoverable. I put in a query to the House Ways and Means committee, but all I heard back was that they, too, are waiting for more details.

A lot of people have asked me whether this makes me less likely to believe in an innocent explanation. The answer is that I don't know what to think, because I don't yet understand what we're talking about: a couple of corrupted Word documents, six more hard drive failures? I have queries into the IRS and the House Ways and Means Committee trying to clarify this question. Meanwhile, because a lot of you are asking, here are my provisional thoughts:

  1. I have heard from a number of federal employees who lost data in exactly the way Lois Lerner did.
  2. I have also heard from a federal employee who says all of her e-mails have to be archived, on a server. The agency that the latter person works for is, let's just say, not known as a leading light of federal competence.
  3. I am inherently suspicious of any suggestion of a conspiracy, particularly one involving civil-service employees. Not because I think especially highly of civil servants, but because conspiracies are hard to get together, and hard to keep together -- someone is likely to blab. Civil servants have a lot to lose by helping political appointees pursue their partisan agendas and no particular reason to be helpful. If it turns out that the IRS engaged in wrongful conduct, I will be inclined to credit complex sociology-of-organizations explanations (where everyone at the IRS shares an unspoken and perhaps unrecognized belief that people who campaign against taxation are clearly political ideologues, while people who campaign for more environmental spending are just swell, public-spirited folks trying to save the planet). It will take a lot to convince me that career employees at the IRS sat down and deliberately and knowingly engaged in illegal targeting of conservative groups for the purpose of helping Democrats win elections. It will take even more to convince me that they blatantly coordinated with the White House, which would obviously be a ticket to loss of job, pension and benefits, plus maybe a side trip to the pokey.
  4. It would be phenomenally stupid for Lerner and others, if they were engaged in such a knowing conspiracy, to use their work e-mail accounts to coordinate it, rather than just calling each other on the telephone or, in extremis, using secret outside e-mail accounts.
  5. Even if there was such a conspiracy, the odds that people who were not initially involved in it decided, after the fact, to help them cover it up seem pretty low to me. At least one tech person would have to be involved, and why should they attach themselves to someone else's felony? If there was a cover-up, the most likely story is a single person deciding to get rid of evidence after Representative David Camp sent a letter asking about targeting of conservative groups. The second-most-likely story is a small group of people decided to do the same, perhaps with the help of a friendly person in the IT department. The least likely -- in fact, frankly unbelievable -- possibility is that the IRS as an institution is actively destroying evidence to cover up wrongdoing.
  6. It is not particularly easy for a non-technically-savvy user to fake the problem the IRS documents say that Lois Lerner had. I could do it pretty easily, with pretty high confidence that I wouldn't get caught. But I worked in tech. The average person wouldn't know how far back the backups went, or how to create an unrecoverable issue with her hard drive. And anything she did to fake an accident would have involved some risk of getting caught -- and making everything much, much worse.
  7. That said, it's possible that Lerner, or one of her close friends or relatives, does have these sorts of technical skills. This is one place where a larger conspiracy helps: the more people involved, the more likely one of them is to have a dear one who could walk the team through the process of faking a catastrophic data loss.
  8. Because this problem does seem to be genuinely common in the federal government, Lerner might well have been aware that lost archives couldn't be easily recovered.
  9. If Lerner destroyed her hard drive, odds are nearly 100 percent that she Googled instructions for doing so from either her home or work computer. I've heard it suggested that her Google searches from June 2011 might be recoverable with a subpoena.
  10. It is easier to fake a problem if no one is looking for it. At the time of Lerner's hard drive failure, there would have been no reason to suspect anything except a normal accident. And it's unlikely that anyone would have taken particularly heroic measures to get data off the drive; those measures are expensive, and Lerner's e-mails to the tech people emphasize that she was looking to recover personal files, not work files that might be potentially of interest to investigators.
  11. Coincidences do happen. I once sat down next to a small conference in Italy and realized that the guy sitting next to me was my college roommate's best friend, with whom I had wandered around our campus one lovely spring night 20 years before. What are the odds? But there you are: However small the odds, it did happen.
  12. Groups do manage to get themselves together to do crazy dangerous things, even though this seems, well, crazy and dangerous. The fact that Watergate happened at all is, in retrospect, completely amazing: How did all these guys agree to commit felonies for such a trivial potential payoff? But they did. The tech guys at Bernie Madoff's firm did decide to help him cover up his Ponzi scheme. Groups of civil servants have gotten together for massive embezzlement schemes. So the fact that a conspiracy is really unlikely doesn't actually make it impossible.
  13. For the moment, I am going to assume that the additional six sets of lost documents are less interesting than they sound in a news release, because most things are less interesting than they sound in news releases. It is very easy to imagine that they could have lost a couple of documents apiece for six different employees during the course of normal business. To be clear, of course six tragic hard drive failures in a relatively short period of time would make it very hard to believe in a benign explanation. As Lady Bracknell said of parents, "To lose one may be regarded as a misfortune. To lose both looks like carelessness." To lose six looks like . . . something quite a bit worse than carelessness. But I reiterate that we don't yet know that this is what happened; we just know that the House Ways and Means Committee says that documents are missing for at least six more employees. How many? What kind? How were they lost? I'll need to know that before I render judgment.
  14. The IRS did not do itself any favors by sitting on this for months; it appears to have known about this problem as early as February. In the intervening months, e-mails at other agencies that could have been discovered may have been lost. As Naomi Goodno says, "If this discovery issue had arisen in federal court, the IRS would have likely been subject to monetary sanctions and possibly an adverse inference instruction. Shouldn't the IRS be held to these standards?" At this point, I think calls for an outside investigation are justified.
  15. I really wish someone at the IRS would return my e-mail and explain what happened.

I assume at some point we'll get more detail. When that happens, you can expect me to write more. Until then, that's all I've got: It remains possible that there was some sort of attempt to destroy evidence, but without more information, this is not necessarily the most likely possibility.

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.