The Fat Tails of VaR

Conclusion: As if risk management wasn’t difficult enough, there is a strong case that the “institutionalization” of risk management may actually be increasing risk and market inflection points.

Not only have we seen heightened volatility in financial markets over the past couple of years, but we have also witnessed accelerating volatility.  On the first point, we’ve highlighted a chart of the VIX, which is a volatility measure on the SP500, going back three years, which shows the strong relationship between an extreme in Federal Reserve Policy and this increase in volatility.  In our view, extremely loose Fed policy has both shortened the business cycle, as well as increased price volatility.

The Fat Tails of VaR - 1

To the second point, in conjunction with this broad increase in price volatility, we have also witnessed a number of accelerations in sell offs in certain markets.  The most obvious example of this is the “Flash Crash” that occurred on May 6th, 2010 and resulted in a 900 point drop in the Dow Jones Industrial Index, a 9% decline and the second largest intraday decline in history.  Just as quick as it happened, the Dow regained most of its losses within minutes on May 6th.

The Fat Tails of VaR - 3

While the “fat finger” theory has been largely disproven, the explanation given in “Findings Regarding the Market Events of May 6, 2010”, a joint report from the SEC and CFTC, still seems somewhat inadequate.  According to the report, the “Flash Crash” was likely initiated by a large and unnamed mutual fund firm that initiated a massive sell order of 75,000 E-Mini contracts, which had a value of $4.1BN.  According to the report, the algorithm the trader used was set to “target an execution rate set to 9% of the trading volume calculated over the previous minute, but without regard to price or time.”  While in “normal” markets, this execution may have gone off without a hitch, the equity markets on May 6th were more volatile and less liquid than the typical trading day.  As a result, this sell order initiated a dramatic sell off as the broad market was overwhelmed with sell orders, which was then amplified by high frequency traders.

According to the joint report:

“The combined selling pressure from the Sell Algorithm, HFTs and other traders drove the price of the E-Mini down approximately 3% in just four minutes from the beginning of 2:41 p.m. through the end of 2:44 p.m. During this same time cross-market arbitrageurs who did buy the E-Mini, simultaneously sold equivalent amounts in the equities markets, driving the price of SPY [an exchange-traded fund which represents the S&P 500 index] also down approximately 3%.

Still lacking sufficient demand from fundamental buyers or cross-market arbitrageurs, HFTs began to quickly buy and then resell contracts to each other – generating a “hot-potato” volume effect as the same positions were rapidly passed back and forth. Between 2:45:13 and 2:45:27, HFTs traded over 27,000 contracts, which accounted for about 49 percent of the total trading volume, while buying only about 200 additional contracts net.”

Another aspect to the “Flash Crash” that has been discussed less, but is perhaps just as relevant structurally, is the evolution of institutional risk management and its impact on accelerating volatility in global markets.  Specifically, most large financial institutions that commit capital utilize a number of standard measures of risk to determine their exposure and risk.  The most common of these metrics is Value at Risk, or VaR.

Like many consensus risk management measures, VaR was developed after a risk management event.  In this case, it was the stock market crash of 1987 that led many large institutions to develop and institute this measure of risk to evaluate their exposures across trading desks, groups, and geographies.  While every financial firm likely has their own tailored definition, VaR is typically defined across a probability, time horizon, and portfolio.  By way of an example, a 3-day 97% VaR of $3MM implies that the given portfolio has a 3% probability of losing $3MM in a 3-day period.

Critics of VaR, and there are many, deride the risk metric because it typically assumes normal markets and limited trading within the probability period.  Further, it doesn’t account for the potential loss period effectively.  Typically, the loss distribution is a fat tail, so the 3% in the tail potentially has massive loss potential.  Our friend David Einhorn put it most succinctly when he said VaR is, “like an airbag that works all the time, except when you have a car accident.”  Now to be fair, some of this criticism has been taken to heart and institutions continually experiment with VaR and other risk management metrics to improve them.  One of the more recent derivations is LVaR, or Liquidity Value at Risk.

According to the Bank of International Settlements the key attribute of LVaR is, “the holding periods in the risk assessment are adjusted to account for market liquidity, in particular by the length of time required to unwind positions.”  The implication of this tightened risk management metric is that it creates a feedback loop of sorts.  As liquidity decreases, the trader’s ability to take risk, according to LVaR, decreases and the trader is then forced to decrease the size of the positions.  The potential snowball effect is that across the market, traders could be forced to sell positions into increasingly illiquid and one way markets as they are forced to reduce exposure.

The recent action in the commodity futures markets appears to suggest this type of selling is occurring.  In a chart below, our Energy Sector Head Lou Gagliardi, shows that the oil futures contract spread is at a 5-year high.  Under an LVaR analysis, as liquidity in the oil futures market decreases, the exposures available to many institutional traders may decrease, which creates a snow ball effect in terms of market activity.  In the scenario of an extreme long or extreme short position, market price accelerated volatility may be particularly pronounced.

The Fat Tails of VaR - 2

In fact, we have seen this in a number of commodity markets this month. In one day in early May, oil was down more than 8% and silver was down more than 20%.  As if risk management wasn’t difficult enough, there is a strong case that the “institutionalization” of risk management may actually be increasing risk and market inflection points. Be wary of the fat tails of VaR.

Daryl G. Jones

Managing Director