Prophetic and reasoned: hedge funds and financial complexity

OK, one more post on the book I mentioned last time.  The conclusion of this 2007 book has one of the best paragraphs I’ve read on the financial system – with the benefit of 20/20 hindsight, of course (emphasis mine):

Market crises are not born from nature. They are not transmitted by economic or natural catastrophe. The machinery of the market itself can take a small event and distort it. The more closely we try to follow the ideal, thereby adding complexity and more tightly coupling the actions of the market, the more frequently crises will occur. Attempts at that point to add safety features, to layer on regulations and safeguards, will only add to the complexity of the system and make the accidents more frequent. And when blowups happen in the future I can guarantee that the focus will be directed improperly: not at the issues of market design but at hedge funds where the events are observed. They will be implicated for the simple reason that they are engulfing more and more of the risk-taking landscape. The perception of hedge funds being what it is, they will take the blame and become subject to increased regulation. But blaming hedge funds is a little bit like The Simpsons episode in which a meteorite hits Springfield and the townspeople gather, shouting, “Let’s burn down the observatory so this never happens again!” True, the hedge funds are the institutions that have the appetite for the risk; but there is nothing inherent in hedge funds, nothing that they represent as a unified set, that makes them the singular cause of anything.

Thoughts:

• A lot of financial alchemy transforms one type of risk (e.g. liquidity risk) into another type of risk (e.g. correlation risk).  An example from 2007 was the proliferation of RMBS CDOs and their synthetic cousins: they took a simple, illiquid asset (a mortgage) and created a liquid asset with a sting in its tail (the CDO).  The latter risk wasn’t obvious until the horribleness of 2008 occurred.
• The fallout from the credit crunch/financial crisis did indeed focus on hedge funds, both the investor-funded versions and the bank-funded prop desks.  But this was misguided, in my opinion: the gains made by Paulson and others were massive, but were only a symptom of underlying issues with the US housing market.  Remember: once these liquid RMBS CDOs were widespread, and default correlation assumed constant, we believed there was no longer need for mortgage underwriting scrutiny.
• Regulators have had a field day with adding considerably more safety checks/switches since the crisis; the uncertainties around Dodd-Frank and several other pieces of legislation (e.g. how will they ever implement and police so many new rules?) may lead us back to crisis.  For example, we now have banks which have cut lending to shore up capital positions; ‘shadow banking’ to fill the gap in credit; hedge funds specifically designed to arbitrage the capital treatments of different sources of bank capital; etc.  How will it end?

In sum: in many situations, more rules may not equal better risk management.  Probably good to know for trading, as well as life in general.

What cockroaches can teach traders: reading list

A well-defended cockroach. Source: Google images.

The latest on my reading list is A Demon of Our Own Design: Markets, Hedge Funds, and the Perils of Financial Innovation by Richard Bookstaber.  It’s a bit dated, but right on the mark for what was to come in 2008.

The author’s experiences with risk management in a world of increasing financial complexity is the main theme of the book.  Now that I’m nearing the end, I’ve found a tasty morsel of advice for systematic traders – or, indeed, just about anyone.  Seek simplicity and robustness in system design/management.  The author uses biological systems for his metaphors; in this case, the very simple defence mechanism of the cockroach results in lots of false positives (i.e. the roach is too conservative), but is therefore well equipped to handle the unknown unknowns of the world.  His counterexample is a fish in Lake Victoria: very specialised for its environment – but gets wiped out when a foreign species is introduced.  Though the fish was perfectly designed for its environment – optimal in a lot of ways – it was ill-prepared for external shocks.

This reminds me of several folks’ ideas of systematic trading: why can’t we just program the algorithm to take all the (in hindsight) obvious precautions to avoid losses?  Well, aside from the observation that many of the precautions actually cost us in the long-run, a hyper-specific trading system is basically useless going forward.  The markets continue to make fools of us all; therefore we need to choose systems which have more robust (read: simplistic) trading and risk management.  Not only will that probably keep us profitable as markets change, they will probably be better equipped to survive external shocks.

As an example (but not meant as a plug): trend-following strategies are beautiful to me.  They’re extremely simplistic trading rules, with built-in risk management.  The latter is usually pretty blunt (e.g. stop-losses, or just signal reversals), but means good risk control in a variety of cases.  That’s probably why trend-followers have been around for decades, quietly churning decent – though maybe not show-stopping – performance through crises and more normal times.