Commenting this phenomenon among others, two weeks ago JPMorgan as Marko Kolanovic rightly stated that "liquidity has become to a large extent driven by market volatility " by reinforcing a negative feedback loop between volatility, liquidity and investor flows, he cites, as the most recent examples, the unprecedented drop in the depth of the futures market or "record of low liquidity"We discussed previously (shown in the graph below), the flash currency crash on January 2 or the stock market crash on December 26.
This topic was then taken up by Goldman, who broadened this analysis by taking into account the third, evoking the impious trinity of the market, namely the volume of trade, and its impact on the reflexive feedback loop between liquidity and volatility (the readers can re-read the complete structure of Goldman's market valuation to the next post).
Yet, despite repeated attempts to explain what has happened, investor confidence in the stock market has been seriously undermined by soaring volatility, falling liquidity and widespread accusations that algos ( or program trading) are behind recent events that, for the most part, have remained quite confusing, even for the most seasoned financial professionals.
Of course, none of this is new, as evidenced by the January 1988 announcements of none other than Shearson Lehman to those who may not have even been born on Black Monday.
So, in a follow-up attempt to explain the nuances of the recent violent market movements (in response to what seems to have been a lot of confusion among customers, most investors still not aware that in the new abnormal case , "liquidity is the new lever " as Goldman Sachs explained last JulyAs for JPM, Marko Kolanovic, director of JPM, has published over the months one of his most detailed and informative notes, describing in detail the feedback loop between volatility, liquidity and flow, a loop that is more important than ever. it simply qualifies as "market fragility" (something of Goldman). has also touched on several occasions, most recently in "Goldman: 2019 – Lower Liquidity, Higher Volume, More Fragile. ")
Kolanovic explains that "the interest in this topic comes from various clients – those who evaluate the impact of broad market flows and their portfolios (fund managers), those who position themselves for and anticipate these. flows (speculators), and those who evaluate the robustness of different systematic investments (investors). "
So, to provide greater clarity to clients, the Quant first demonstrates the link between volatility and liquidity, which it did several weeks ago, while demonstrating that market depth – the key measure of liquidity – has become much worse in the last decade. the banks took over.
Figure 1 illustrates the relationship between the depth of the S & P 500 futures market and the VIX. It can be noted that this relationship is very strong and non-linear (for example, market depth decreases exponentially with VIX). Since an increase in volatility often leads to systematic sales, this relationship is the key to understanding market fragility and extreme events. The second question was whether this relationship was always the same or if the situation worsened with time. To answer this question, we show the historical relationship between liquidity and VIX over time (Figure 2, sliding slippage slope between liquidity and VIX). It can be seen that the negative relationship between liquidity and VIX has worsened over the last decade. (Note that an exponential relationship can be approximated locally with a linear relationship and tracked over time).
Finally, we note that in periods of high volatility, the VIX is almost the only driver of market liquidity. Figure 3 shows the percentage change in liquidity that can be explained with the VIX as a function of time (R-square sliding). The higher the VIX, the more liquidity is generated by the VIX and recently, up to about 80% of the liquidity changes have been explained by the VIX. To conclude, we have shown that there is a negative relationship between volatility and liquidity, that this relationship is reinforced over time, and that it is particularly strong in times of high volatility.
After covering the two main rays of the modern market function (evil), namely liquidity and volatility and their constant interaction, Kolanovic shows that volatility also plays an important role in determining flows of various systematic strategies (It defines systematic flows to include derivative hedging flows, flows of passive and quantitative investment strategies, the insurance industry and programmatic market flows).
According to the JPM, and as operators have very clearly observed last December, an increase in volatility – fast or slow – generally leads to an increase in systematic sales, what happens in a tight liquidity environment, thus producing a disproportionate impact on the market and a greater overall fragility of the market – so it is not surprising that crash-flashes occur most often when the VIX is to the north Kolanovic also notes that during periods of high volatility / low liquidity "," not only systematic strategies, but also discretionary managers, sell, although they tend to sell more slowly and / or later in the future. during the mass sale episodes ".
To represent these observations in a practical way, Kolanovic examines the different examples of systematic flows, their impact on the market and their own performance, as well as the speculative activity related to these flows:
The most important of all systematic flows in terms of size and impact is the hedging of index options. Figure 4 illustrates the weighted delta interest of the S & P 500 Index positions, compared to an asset estimate of two other "gamma-short" strategies, CTA strategies / Trend Tracking Strategies. and volatility. It can be seen that the most important component of systematic flows comes from the coverage of options, but given the increasing tracking of trends and volatility targeting these components, it is impossible to track them. ;ignore.
During periods of heightened volatility, option hedge flows generally have a significant impact on the market as it takes a few days for speculators to establish positioning and hedging patterns and begin to anticipate these flows. .
In addition to index delta coverage, another systematic strategy with predictable flows according to the JPM is based on reverse exchange traded products: similar to the index option hedge, "these products constitute a short gamma (note that the ETF gamma inverted is usually much smaller than the gamma option). "
And since flows are "predictable," one of the defining features of this strategy is that, to the extent that leveraged and inverse ETFs are short gamma values, their rebalancing translates into predictable systematic flows by speculators, which negatively affects the performance of these products, creating a potentially destabilizing return dynamic. The best and most recent example is the disappearance of the product of inverse volatility XIV:
When volatility rose in February, the size of the rebalancing could not be digested by the market. Liquidity providers, experiencing significant rebalancing flows, did not want to intervene until the product self-destruct. Figure 5 shows the rebalancing flow in% of the daily average volume of the VIX as well as the VIX level at which the product "self-destructs". As soon as the VIX has "scored" this level, the volatility is quickly blurred.
Continuing on the list of strategists whose gamma is short, we then encounter strategies of trend tracking and targeting of volatility.
As Ray Dalio very well knows, volatility targeting can be applied to any portfolio (eg, 60/40, risk parity, factor portfolio, a core PM platform, etc.); moreover, the flight targeting is explicitly a short gamma in a medium inversion market. As Kolanovic explains, the reflexive strategy reduces risk when volatility increases and increases risk when volatility declines. This is a design (risk exposure ~ 1 / volatility) and, as a result, the flows from these strategies are closely related to the hedging of options. It also means that once the market starts to sell, most of these short and pro-cyclical gamma strategies will strengthen the downward movement and vice versa; it is therefore the points of inflection that tend to be the most violent.
In terms of impact on the market, Kolanovic notes that in his estimates, the notional amount of these strategies to approximately $ 300 billion in multi-asset portfolios, which is much smaller than the weighted weighted T4 delta interest of $ 750bn notional just for the S & P 500 index. Moreover, these strategies are selling over several days (as opposed to hedges). options that are sold during a day). Similar to the other strategies mentioned above, the short gamma exposure of CTAs is not explicit, but remains intuitive. as they sell when an asset price goes down, and buy when it goes up.
That said, systematic investors do not sell only VIX peaks. Long-short equity hedge funds are also selling in the VIX picks, but in a less programmed and certainly less aggressive way.
This is indicated by the sensitivity of funds beta to the VIX. For example, CTA's beta to VIX is approximately 4 times higher than the beta of long short equity HF to VIX (for example, -3.5% vs. -0.9%). Figure 6 shows the difference between the beta of short-term equity HFTs and CTAs and VIXs. As VIX increases, CTAs reduce beta (sales shares) faster than long-short equity hedge funds.
Kolanovic explains that, while his analysis does not "judge" the merits (or failures) of various systematic short gamma strategies often used for risk coverage or control, he notes that "These strategies require adjustments because of the changing market" (for example, feedback loop liquidity-flow-volatility, speculative flows, etc.). But above all, again, it is still the reflexive nature of these strategies, where the modification of flows becomes a signal that influences future changes in flows, creating a feedback loop that reinforces itself or, as the says Kolanovic, "if the systematic flows are large enough to have a market impact, they will affect their own performance via speculative flows and market repercussions."
Finally, to list the strategies that would be affected, the JPM suggests that some simple checks include:
- Strategies in which the asset base and flows are important in relation to market liquidity (that is to say the crowd).
- Strategies where flows are correlated with volatility. The reason is the volatility-liquidity-flow feedback loop, which makes the effective asset size look "larger" than it is. Strategies that are intrinsically short gamma are more affected.
- Strategies that are fully transparent and do not adapt along the way (for example, if a strategy is documented in a prospectus or a theoretical white paper, it will more likely invite "imitators" or speculative flows).
- In addition to the above, a poorly performing strategy may indicate clutter or damage caused by speculative flows or short gamma exposure.
The above explanation should provide an essential framework for analyzing market developments in times of increased volatility (and decreasing liquidity), but it will not be enough to satisfy a market where a whole generation of traders has not yet seen a bear market in its own right. why the next time a general sale will occur, expect that virtually no one understands anything of the above, and why the explanations provided by the market quants aimed at developing the different feedback loops between the Leading market representatives, namely liquidity, volatility and capital flows, will be quickly forgotten, and the most likely answer will be another episode of aggressive selling followed by demands asking the Fed to bail everyone out. As the events of December and January demonstrated, it could well be the best "strategy" in the foreseeable future.