The recent coverage of JP Morgan’s credit derivative ‘whale’ trade — supposedly a market hedge rather than directional position for the bank — has seen many financial pundits wonder how it could be that JP Morgan has a short position on any of the underlying names in the credit against which it is supposedly selling protection?
The index is made up of 121 North American companies, all of which were investment grade at the time the index started trading in the fall of 2007.
Some commenters have found this hard to rationalise because they believe it makes more sense for the bank to be long that underlying exposure rather than short. In which case, how can selling protection be considered a hedge?
But that’s to take hedging at its most simple. And things, as we have learnt, are never simple.
For more on what JP Morgan’s whale trade could really be hedging it’s worth checking out John Carney’s coverage over at CNBC’s NetNet. He has a good theory about how the trade could actually be a hedge for inflation risk rather than anything else.
Hedging is not what it seems anymore.
To be fair, hedging has not been about buying an exact equivalent position to the one you’re exposed to for a long while now. (Or ever.)
So what counts as a hedge in this day and age? And how important is “smart hedging” to the profit model of a modern bank or trading entity?
First, here are what we consider to be three key attributes of a “smart hedge” (the sort that can make banks and trading houses heaps of money).
- That it correlates with your exposure.
- That it is cheaper to run than your underlying exposure.
- That it is liquid.
What is the main risk associated with such a hedge?
If any of these three factors change with respect to your original position, you experience what is known as basis risk. This basis risk has to be managed independently.
If things are going awry because the instrument you are using to hedge is no longer correlating, this shouldn’t be a problem if the hedge is a liquid enough to exit.
If the hedge becomes too expensive — again, not a problem providing it’s still liquid enough to exit.
If the hedge becomes illiquid, however– things only remain okay for as long as the hedge is still correlating and stays cheap.
If not, you may have a problem.
Conclusion? Liquidity is by far the most important attribute of a “smart hedge”. It’s also the reason why really exotic hedges (such as Turkish lira offsetting oil exposure) are popping up ever more frequently. It’s the liquidity, correlation and price that traders are looking at, not whether the hedge is an exact match substance wise.
The mathematics of exotic hedging, meanwhile, are complicated. We for one cannot say that we understand it. Though, it’s fairly clear that a lot of the progress is down to improved correlation modelling, which can help institutions manage residual exposures in the public market more cheaply, but also to offset (or match) exposures internally in a far more creative manner.
Very simplistically speaking, imagine something like this:
“We’ve got 500 buy orders for Turkish lira coming in from our FX desk. As it happens we’ve got the equivalent sell orders coming in in oil too. There’s a correlation here. We won’t need to hedge our lira or our oil exposure nearly as much.”
And for a more concrete example here’s a paper what was recently highlighted by Risk (though be warned, unless the term ‘quant’ or ‘rocket scientist’ appears in your job title, this will very likely hurt your head. What’s more, we are in no position to explain it ourselves) :
Hedging methods are divided into single-period and multiperiod forms. After reviewing some well-known hedging algorithms, two new procedures called the Dickey-Fuller optimal (DFO) method and the minimax subset correlation (MMSC) method are introduced. The former is a multiperiod, cointegration- based hedging method that estimates the holdings that are most likely to deliver a hedging error with no unit root.
The latter is a single-period method that studies the geometry of the hedging errors and estimates a hedging vector such that subsets of its components are as orthogonal as possible to the error. We test each method for stability and robustness of the derived hedged portfolio. Results indicate that the DFO method produces estimates that are similar to those for the error correction method, but more stable.
Likewise, MMSC estimates are similar to principal component analysis, but more stable. A generalized Box-Tiao canonical decomposition (BTCD) method, which is of the multiperiod class, is proposed. The BTCD estimates are also very stable, and cannot be related to any of the aforementioned methodologies. Finally, it is found that all three advanced hedging methods (MMSC, BTCD and DFO) perform well.
So, is there genuine reason to be worried? Is the world of hedging getting far too smart for its own good?
And more to the point, should we really be reassured when a bank says “don’t worry, this position is just a hedge”.
We are definitely not informed enough to make that judgment. But we’d be interested to hear your thoughts.
One thing we do know… trading houses and broker dealers don’t really like to discuss the principals behind their hedging strategies. This, of course, is understandable given that hedging is really where the bulk of their proprietary edge is found nowadays.
Banks as volume gobbling monsters – FT Alphaville
Stardate April 13, CIO sector, JP Morgan reporting VaR – FT Alphaville
JP Morgan’s giant unwitting catalyst trade – FT Alphaville
You Say “Voldemort” Like That’s A Bad Thing – Dealbreaker