Two trends have dominated financial markets in recent decades. First, there is an arms race going on in instruments and trading algorithms. Math jocks are going braino-a-braino to devise increasingly sophisticated strategies, to the point where even specialists are unsure how to value portfolios. At the same time, there has been an escalation of leverage; unregulated investment funds are lending to the tune of 30 or more times their equity base. Maybe these developments are related.
The current argument is that over-the-top leveraging is the consequence of a regulatory breakdown, and there is certainly truth to the extent that the failure of oversight has enabled investment banks and hedge funds to do whatever they want — but why did they want to extend themselves so far?
Here is one possibility: the new math-intensive strategies are chasing tiny margins. The trading programs are designed to perceive opportunities for arbitrage a nanosecond before anyone else, taking advantage of the slightest misalignment of related prices. We have also witnessed ever more elaborate strategies involving complex tradeoffs between risk and return to create composite positions whose alpha is perceived to be a shade higher in relation to its beta.
The profit margins on these strategies are minuscule, but if everything goes as programmed, predictable. This means that they can be turned into respectable earnings only through intense leverage. By investing in positions at the rate of 3000% of capital, you turn 1% annualized margins (which correspond to much smaller margins on any given trade) into 30%. And the geniuses who devise these opaque instruments tell you that the risk is small relative to the return.
What the risk jockeys always seem to miss is that estimates of portfolio risk depend on covariances among the individual elements, and these in turn are determined by the structural properties of the system in question. And no one can know what they are, because the system is too complex, there are too few data points, and the structure keeps changing. So the models work, trade after trade, until there is an unforeseen systemic event, after which all hell breaks loose.
Then the high degree of leverage, which was necessary to make the strategy pay sufficient dividends, magnifies the risk instead of the return.
If this analysis is correct, it suggests that putting a ceiling on leverage may also slow down the drive toward unfathomable financial complexity. That would be a good thing, and not just for us dummies.