In 2016 BofAML's global equity derivatives desk expects volatility to maintain its gradual upward trend, however, to continue to be punctuated with violent but short-lived shocks owing to poor liquidity, extreme positioning and a market still heavily manipulated by (and dependent on) the central bank put. Despite below-normal levels of volatility across asset classes, we are in uncharted waters in terms of a lack of stability:
- Markets are setting records in terms of jumping from calm to stressed & back,
- CB liquidity is tightening, making markets more accident prone, and
- Our indicator of cross-asset market fragility is near its highs
[5]
Asset managers are struggling, with the poorest hedge fund performance relative to the risk they are taking since 2008, despite overall market volatility being only 1/4th of 2008 levels. Their poor performance is better explained by the extreme levels of market fragility, which by our metric is at 80% of its 2008 highs (Chart above).
Unfortunately, we don’t see conditions improving and only becoming more acute as liquidity continues to deteriorate, asset valuations become increasingly stretched, and the Fed navigates the unwind of the greatest policy experiment in history.
* * *
Measuring fragility
Normally, there is a strong relationship between the size of a market shock, and the amount of volatility generated by the aftershocks before markets calm to normal levels. However, in a local tail event the relationship is very different with a tendency for markets to see extreme dislocations which reverse at record pace. This is precisely what we have seen over the last 18 months across asset classes (Chart below).
Illustrating this for US equity volatility specifically, we note the relationship between the peak height in the VIX to the total volatility generated during the shock before it retraced back to pre-spike levels. Since 2014, we have been in uncharted territory in terms of large shocks occurring with very little total damage done to markets. This allows average volatility to remain muted but creates challenging conditions for managing risk and generating returns.
The Fragility Indicator
Extending this concept of sharp dislocations to cross-asset volatility, our Fragility Indicator measures the frequency and extent to which volatilities and credit risks in our GFSI index are witnessing abnormally large shocks, or outlier events from the group. So for example on 24-Aug-15, S&P 500 implied volatility rose to extremes versus other risks in GFSI, as shown in Chart 9, which indicated high fragility.
Normally, the magnitude of this fragility indicator is directly proportional to the total level of volatility. However, we have recently entered uncharted territory in terms of dislocations rising to near 2008 levels, despite average volatility still below normal, as shown in the chart above.
As shown in the chart below, the market across asset classes has in the last 18 months generated a large number of extreme outlier events that in numerous cases have set multi-decade records.
[6]
It is these events that continue to drive our fragility indicator to such high levels. However, as these episodes have not resulted in significant risk-contagion (in part a result of the power of the central bank put), overall volatility levels have remained muted. In other words, we have recorded a significant number of local shocks that have not become global.
One of the key reasons we believe local shocks occur is that they are driven by technicals including crowded positioning and markets’ malfunctioning from a lack of trading liquidity. However, with little fundamental depth, they fade quickly, aided (in particular) since 2013 by central banks that have never been more sensitive to market risk.
A key sign of a technical shock is one in which risk is dominated by a single asset class, without a fundamental re-pricing of risk across markets. This was precisely what we witnessed in both October 2014 and August 2015, when equities recorded as much as five times the stress of any other asset class. Monitoring this gap is key to deciding whether to fade a shock or add hedges for a more prolonged event, as major risk-off events are never isolated to a single asset class.
