Cause 39 of the banking crisis
BBC business editor Robert Peston on how the banking crisis came about
Howard Davies, director of the London School of Economics and founding chairman of the Financial Services Authority, has just published a characteristically witty and pellucid analysis of the competing theories for who and what caused the worst global banking crisis since the 1930s, The Financial Crisis - Who is to Blame?.
He identifies 38 causes, from the efficient market hypothesis and lax regulation through to avarice and traders' testosterone. It's an old-fashioned primer, not a polemic. And - some would say - none the worse for that, given that there are a good few myths still doing the rounds about who or what was at fault.
Why did he write it? Perhaps it was therapy, a need to reassure himself that the debacle wasn't his fault. And he can take comfort that the explosion of banks' balance sheets - the period of the fastest growth of their dangerous lending relative to their capital resources - took place after he had handed back his watchdog's teeth and climbed his ivory tower.
As luck would have it, Tim Bush, a former Hermes fund manager and member of the Accounting Standards Board's Urgent Issues Task Force, would this morning argue that Davies's 38 causes should be 39.
He's tabling a paper for the taskforce to minute, as a prelude to a proposal for reform of the way that British banks account for old-fashioned, plain vanilla loans.
His is not the modish complaint that putting a market price on tradeable loans and investments - the mark-to-market requirement of fair value accounting - meant that banks over-valued their assets in the boom era, and under-valued them in the bust years.
There is a genuine debate to be had about the merits for banks of allocating their capital on the basis of asset prices determined by the irrational herdlike behaviour of investors (including the herdlike behaviour of banks themselves).
But Bush makes a different point. And although he has become something of a one-man, single issue, slightly obsessive campaigner, what he says is certainly worth considering.
In simple terms, what went wrong - Bush says - is that in the UK and Ireland from 2005 onwards banks stopped making any general provisions against the risk that their loans could go bad. In that sense, they stopped the long and tested practice of factoring into the cost of a loan the probability that it might not be repaid.
British and Irish banks stopped making these provisions for possible non-repayment of loans at both the level of published accounts and at the lower level of the operating units.
This was not their choice, Bush says. They were forced to do it by the way that the Accounting Standards Board implemented the international accounting standard IAS 39 as Financial Reporting Standard 26, or FRS 26.
Bush argues that the implementation of FRS26 magically made lending seem less risky and cheaper for British banks - so (guess what?) they did much more of it.
If you look at the terrifying speed at which Northern Rock increased the supply of 100% mortgages, or the extraordinary acceleration of lending to property companies by HBOS, there would seem to be some connection between the worst excesses of the UK's credit bubble and the cessation of any requirement to factor in the probability that some loans would go bad.
And there is a similar correlation between the accounting change and the timing of the explosion in property lending by Anglo Irish Bank and Allied Irish Banks.
That said, it doesn't seem wholly plausible that an accounting change could turn on a credit tap quite so fast - in that an accounting reform surely couldn't have persuaded long-serving bankers to simply forget their years of experience about the riskiness of lending.
On the other hand, it is striking that there was not quite such a lending binge in continental Europe, where the new international accounting standard was not implemented at the level of operating units, where lending decisions are actually made.
Given that the reckless increase in lending by HBOS and Northern Rock generated massive losses for both of them, which ultimately undermined their solvency, you might ask why the Accounting Standards Board thought it a good idea to abolish the requirement to make general provisions against possible non-repayment of loans.
Well, the ASB for some years has strived to reduce the element of subjective judgement in the production of published accounts. The ideal would be that accounts show the world as it is, not the world as managers would like to see it. So it seemed a good idea that banks should only report losses on loans as and when the borrower stops repaying - and not when the bank manager fears that they may stop repaying.
However, it turns out that accounting rules which deny the importance of managers' judgement may have the unfortunate effect of transforming them into credit-spewing automata.
Here's what gets Mr Bush really excited.
He thinks that FRS 26 may actually contravene the requirement of company law that banks operate in a prudent manner consistent with the protection of their depositors and creditors.
Which carries the intriguing and resonant implication that shareholders might be able to claim that they were gulled by FRS26 into believing that Britain's banks were made of bricks when they were in fact made of sand.
You can keep up with the latest from business editor Robert Peston by visiting his blog on the BBC News website.