According to Anton Valukas, the examiner appointed to investigate the collapse of Lehman Bros by a New York bankruptcy court, Lehman used Repo 105 to hide from creditors, markets, ratings agencies, regulators and even members of its own board quite how much it had borrowed relative to its capital.
Or to put it another way, the firm used Repo 105 to exaggerate its financial strength in 2008, which was when this really mattered because of widespread concerns about the robustness of many banks.
The ruse worked like this.
Lehman was highly and dangerously dependent on raising hundreds of billions of dollars of short-term finance every day, in what's known as the repo market.
This is a market used by US investment banks in which assets can be swapped for short-term loans.
But because the finance raised in this way has to be repaid within days, the assets - in an accounting sense - are never deemed in an accounting sense to have left the repo-ing banks' balance sheets.
Except that Lehman found a ruse to use the repo market to make it look as though the assets had been removed in a permanent way.
Apparently (and this is quite difficult to believe) the accounting rules allowed Lehman to report a reduction in assets if it exchanged those assets for funds at a conversion rate of 105 to 100: so if Lehman exchange assets with a value of $105 for loans at a value of $100, that $105 of assets could be removed from the balance sheet when reporting group financial results.
Astonishing.
Now according to Valukas, this ruse allowed Lehman to report that its assets were $38.6bn lower than was really the case at the end of the 2007 financial year. And the reduction increased to $49.1bn at the end of the first quarter of 2008 and $50.4bn by the middle of 2008.
Why did this matter?
Well, one of the most important measures of an investment bank's financial strength is its leverage ratio, or the ratio of its reported assets to its reported capital. The lower the ratio, the stronger a firm will appear to be: the bank will appear to have more capital relative to its loans and investments to absorb any losses on those loans and investments.
So by removing $50.4bn of assets from its reported balance sheet using Repo 105, Lehman reduced its reported leverage ratio from 13.9 to 12.1.
That may not sound a lot, but in the context of the fraught market conditions of 2008 - after Bear Stearns imploded - it could have been the difference between life and death for Lehman.
In particular, it was hugely dependent - as I've said - on raising short-term finance from the conventional repo market. And if its creditors in that market had known the true state of its leverage, they might have ceased lending to it even earlier than they did - which would have brought forward the date of Lehman's demise.
I'm slightly under the cosh now. But when time permits later in the day, I'll also look at why Valukas believes there is a case to claim damages from Lehman's chairman, Dick Fuld, three chief financial officers - Christopher O'Meara, Erin Callan and Ian Lowitt - and the firm's auditor, Ernst & Young.
For different reasons, he believes there may also be claims on JP Morgan and Citibank, for the way they demanded collateral from Lehman in its final days, and from Barclays, for the alleged improper transfer of certain assets to Barclays as part of its purchase of the rump of Lehman.
