An exchange with a reader reveals that what he got from this post was not the possibly interesting theory that blatant fraud maybe harder to uncover than envelope pushing, but that I'm excusing the SEC.  I'm not.  It's early days yet, but I'm going to go ahead and assume that at least several people at the SEC deserve to be fired for their pitiful oversight.  Similarly, I think it was perfectly just to fire Howell Raines over Jayson Blair.

That outright fraud is hard to detect doesn't mean that it's okay to let it go undetected for decades.  It's hard to know exactly what one could have known, in another's place; hindsight bias is damn-near overpowering.  But I'm going to go ahead and say that both the SEC should have known that a firm with a tiny audit firm and no outside custodian needed a very hard second look.

So it's not by way of excusing the SEC or the fund managers who didn't pick it up; not at all.  I'm just interested in the theory that it might actually be harder to detect outright fraud than "aggressive" accounting or financial structuring.  The aggressive have to work with the stubborn material of relaity, and there is a limit to how far they can bend it to their will.  The fabulist, on the other hand, can cut "reality" to suit.

Consider some of the most famous frauds of the last few years.  Compare Michael Bellesiles to Stephen Ambrose and Doris Kearns Goodwin, for example.  It was much, much harder to convince people of Bellesiles' guilt--in part, yes, because he was closer to the establishmen than Ambrose and Goodwin, but also in part because it was so hard to check his sins.  Ambrose and Goodwin had text that was easily compared with other texts.  Bellesiles had counts that only he had ever done of often imaginary archives located in out-of-the-way places.  Indeed, had he confined his fraud only to imaginary archives, he might never have been caught.

Or think of Jayson Blair, Stephen Glass, Jack Kelley; the latter two got away with their crimes for years.  As long as they faked their notes, there was no one to complain.  The longest and most successful fraud was committed by Jack Kelley, who was also the most total fabulist--but careful to make the "events" he wrote about occur in remote, hard-to-check locales, with sources and quotes that were just slightly too good, rather than outright unbelievable.

Think of Worldcom, with its outright fraud.  Some analysts could, and did, think there was something wrong about Enron--its off balance sheet activity was multiplying too fast.  Enron was brought down by those fears, which trashed its credit and thus destroyed its ability to trade. But Worldcom didn't get fancy with special purpose vehicles; they just flat out claimed that their operating expenses were actually investment, like you classifying your electric bill as an investment in the electric utility.  They created an entire reality with their estimates that internet usage was doubling every nine months.  They were brought low not by crack outside investigators, but because the holes in their cash flow became too big to hide, and their own internal auditors spotted the problem.

It seems to me that Madoff falls into that same category, and that he was so successful because, like Jack Kelley, he kept his fraud modest.  He didn't produce spectacular returns--he underperformed the market when it had a good year.  What he did was produce unbelievably steady returns.  His fund was just slightly too good to be true.  And because he was in a position to create, with his trading firm, a complete alternate reality for investors, it took a long, long time to catch him.

There were red flags; that doesn't mean that the fraud couldn't be detected.  It just meant that it was less likely to be detected before serious money was lost.  Which is probably a valuable guide to making this sort of thing less likely in the future.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.