I've been reading and rereading the full text of Warren Buffett's annual address to Berkshire Hathaway's shareholders, the one where he describes derivatives as "financial weapons of mass destruction". It is one of the most chilling pieces of financial analysis I have ever seen.
Suddenly, all the other concerns of business journalists seem trivial. What does it matter how Gordon Brown balances the books next month? Who cares which bidder finally gets Safeway?
If Buffett is right (look up the 2002 report at www.berkshirehathaway.com), a "time bomb", to use his words, hangs over us all. It is capable - my interpretation - of crippling the banking system of the west, threatening billions of savers and ushering in an economic ice age.
Two weeks on, his warnings are already fading. The banks have, naturally, pooh-poohed his apocalyptic words as over-the-top. Yet on matters financial, Buffett has a horrible habit of being more right, more often than anyone else. That is how he has accumulated the second-biggest personal fortune in the world (only Bill Gates is richer). His worry is derivatives - financial contracts routinely entered into by the world's banks and corporations. These are in effect bets - bets on any number of financial outcomes, from the level of the FTSE 100 to the price of crude oil or the difference between two countries' interest rates.
Some of these bets are legitimate. A British exporter, for example, might want to hedge against the risk of the pound strengthening and could use a derivative contract to do so. Some are entirely bogus. Enron used energy derivatives to flatter its profit-and-loss account.
There are two crucial points about derivatives. In every derivative contract (or bet) there has to be a counter-party, someone on the other side of the bet. For the bet to pay off, the outcome (which may not be for ten or 20 years) not only has to be positive, but the counter-party has to be solvent to pay up.
The other point is that valuing these bets in the balance sheet is often more art than science. While the value of some derivatives is easily established because markets are made in them and prices constantly quoted, there is no market in many of the more esoteric bets. When a banker's bonus or a chief executive's reputation depends on the most favourable view being taken, how much credence would you put on the numbers?
Derivatives are a zero-sum game. For every winner, there has to be a loser. But in the real world - and this is the point so tellingly argued by Buffett - the sums are so complex and the accounting rules so vague that both parties to a derivatives contract are able to book a notional profit on the same bet.
Eventually, however, every bet matures. Then the losers have to cough up real money. The financial world is festooned with bets. It only takes one bank to get into difficulties for other, seemingly innocent banks to be infected. Central bankers call it systemic risk. The rest of us call it the domino effect.
Buffett ruefully admits to problems in closing down his own derivatives business, inherited as part of an insurance company acquisition. Like hell, he says, the derivatives business is easy to enter and almost impossible to exit.
Despite the collapse in 1998 of Long Term Capital Management - whose Nobel Prize-winning economists and mathematicians thought they had invented a risk-free money machine - the derivatives business has continued to expand unchecked. The latest estimate for the face value of all outstanding derivatives is $130trn - or four times the world's entire GDP. HSBC alone has outstanding derivatives with a face value of $535bn - roughly a third of Britain's GDP.
Yes, I know that the face value of these contracts often does not reflect the money at risk, and that the banks employ vast armies of clever PhDs to get the sums right.
The trouble is that banks are not like other businesses. They enjoy a privilege not conferred on metal-bashers or greengrocers or anyone else - the right to take deposits. Those deposits are in effect underwritten by the taxpayer, who has to cough up if a bank fails. Should banking regulators be allowing deposit-takers to indulge in such huge and complex bets? I think not.
Yet regulators have stood by while derivatives multiply exponentially and all they can do now is demand greater capital strength and better disclosure. But even this much is proving hard: French and German banks are fiercely resisting changes to international accounting rules that would give a clearer picture of the bets they are taking.
Do I really believe that a major bank could fold, toppling others and triggering a worldwide depositor panic? I suppose it is unlikely. But rereading Buffett, I must admit to a tingle of relief that I owe my bank more than it owes me.
Patrick Hosking is deputy City editor of the London Evening Standard
