Why it is vital to close the ‘hedging’ exemption in the Volcker rule: a simple explanation
In the wake of the JPMorgan Fail-Whale trading scandal, there is now renewed pressure on regulators not to give in to industry lobbying seeking exemptions to the Volcker Rule banning proprietary trading for Banks who want a government guarantee on their liabilities. So it is important to lay out the reasoning behind not allowing an exemption for so-called ‘hedging’ trades.
The short answer to bankers bleating about the need for some freedom to hedge their positions in the market is that hedging trades tend to blow up. The Fail-Whale trade was a beautiful illustration in point. The longer answer is something like the following.
Here is a rough analogy for what hedging trader extraordinaire Bruno Iksil was, allegedly, doing:
He arrived at the virtual roulette table we call the ‘derivatives market’ and saw that his bank had laid down a multi-billion dollar bet in chips on various black numbers (recall, a roulette table has 18 black numbers, 18 red numbers and a green Zero).
So being the prudent hedging trader he was, he ran the numbers and said, “the wise thing to do here is to lay down an equivalent multi-billion dollar bet on Red. That way, if the bets on black numbers fall through, we still save ourselves from losses by winning on the ‘Red’ bet”. Brilliant strategy.
But then he thought, “gosh, we could lose on Red too, so I’d better hedge that as well! I’ll put down a further bet on ‘Black’, just in case the Red bet falls through too”. So he puts down a yet further massive bet on Black. Wow, now the bank’s position was double hedged! He hedged the first huge bet on some black numbers with an equally large bet on Red, and hedged the latter with a further bet on Black.
The bank was now so well-hedged this could not possibly go wrong. This is the kind of prudence the Hedging exemption to the Volcker rule is meant to allow. See? What’s not to like?
Then the wheel starts spinning, and the ball settles on … suspense … the green Zero!
Golly, who could have foreseen that?
The bank loses a bundle, its share price tumbles, and poor Iksil shakes his head at his bad luck – despite his flawless “hedging” strategy, takes his million dollar bonus from previous trades, and walks across the street to the next hedge fund.
If this sounds incredibly stupid, that’s because it is. A more generous account will involve complex mathematical models about how winning on red is so well correlated with losing on black (“more than 9 out of ten times, it will work!!”), and how no hedge is perfect, and so on and so on. But this is basically what these “risk-mitigation” trades amount to.
We have seen it again and again.
Look at Deutsche Bank’s 3.4 billion dollar loss in 2009 (here, scroll down to Sales and Trading Revenue). It was attributable to the Bank “hedging” a long-position on US automotive sector bonds by buying CDS (a kind of tradable insurance policy) against losses on those positions. How could they lose money on a hedge? Because, once again, the hedge was wasn’t really a hedge. It was a bet, a bet that a third outcome – where they lose money on both the long and the short positions – wouldn’t come about. And they lost that bet. Because, as the bank euphemistically puts it, there was ‘widened basis risk’. In other words, they were betting on the shorts falling faster than the long positions rose. They weren’t hedging at all.
Look at Morgan Stanley’s 9 billion dollar loss in 2008 on a similar bet. They ‘hedged’ a short position on some risky derivatives with a long position on somewhat correlated derivatives, where a loss on the first would be offset by gains on the latter. Oops, the correlation broke down. Whocoodanode?!
And so the list goes on.
All these losses are attributable to the hedge being ‘imperfect’. That is, when they put on a ‘hedge’ trade, they are not actually buying an insurance policy against losses on some position they have on the books. They buy some cheaper insurance, on something roughly correlated with their pre-existing position. In our analogy, they are at that roulette table, gambling on the likelihood that the ball doesn’t land on the green Zero. It isn’t hedging, it’s gambling.
To take another analogy, it is like owning a house, and not buying insurance against wind-damage on your own house, but on your neighbor’s house, where the policy is much cheaper (since his house is more wind-resistant), and doing so in the hope that if your house suffers wind-damage, so will his. It is just irresponsible.
So why don’t they just do ‘perfect’ hedges, buying insurance directly against losses?
First, for the obvious reason that there is no money in that. Banks make their extraordinary profits by gambling, not by spending money offsetting risk. Second, and more justifiably, there often isn’t an insurance product that does directly offset losses on assets they have.
So, to the extent that banks should be allowed to hedge against losses on their assets, (1) these hedges should be directly correlated – i.e. it must be an insurance policy on that asset. (2) They should be held, as should the asset, until maturity (i.e. not put on the trading books, or held as Assets For Sale). And (3) the traders and managers charged with hedging should not be paid bonuses.
But, naturally, real risk mitigation is the kind of thing banks have no interest in. Where’s the money in that?