Since my previous post, the S&P 500 has appreciated +3.19% to $4,535.43, with the VIX and VVIX depreciating by -10.03% and -9.21% to 16.41 and 105.48, respectively. My last callout only truly referenced how participants were pricing butterfly securities, implied correlation, and general futures and options positioning in terms of notional contracts. Even then, it was apparent a decent portion of the market (particularly on the ES side) was extremely bearish, with the vast majority being bullish (as evidenced by a lower VIX/VVIX/vol contract positing, etc.). From peak to trough of the S&P500's worst week in several months, on 8/13/2021 to 8/19/2021, the VIX appreciated by over 40% from 15.45 to 21.67, the SPX falling -1.39% from 4468 to 4405.80, and the VVIX appreciating by over 10% from 117.81 to 130.41. As was stated in the previous post, the primary risk factor to be considered was an overreaction in implied volatility, which was clearly seen in this time frame.
One more quick thing. In my past posts, I have generally referenced the same instruments and analysis sources for my thoughts on the market. From now on, this blog will take the direction of analyzing different ideas such as: local volatility models, stochastic volatility models, replicating variance swaps and the VIX index, and other ideas I have been interested in pursing. As the primary purpose of this blog was to show the average investor that there is so much more to making a decision on buying or selling any security than a twitter tweet or some 1-factor technical indicator, I believe this purpose has been fulfilled. All resources that I have used have been free, and easily accessible to anyone. Hopefully I have inspired a some of you to consider more inputs into an investment decision. With performance evaluation and the future of this blog out of the way, let's get into this blog post.
Update on Market Positioning
I hate to sound like a broken record, but I am not going to elaborate on the definitions of Non-Commercial, Dealer/Intermediary, and Leveraged Funds as much as I have in the past since I have already covered these extensively. However, the quick definitions are as follows: Non-Commercial - a type of market participant that is not participating for explicit hedging purposes. Dealer/Intermediary - this includes participants like market-makers and banks that warehouse risk, and thus hedge off exposure once a deal has been made, hence why dealer/intermediary positioning is usually net short whatever contract is being analyzed, Leveraged Funds: a fund, primarily hedge funds/CTA's, Family Offices, that speculate and invest at their own discretion.
This graph shows the Non-Commercial positioning on VIX futures contracts. Clearly there exists a short bias, but the important thing to note is how extreme this short bias has accelerated/decelerated, and in what fashion relative to past trends. For example, Non-Commercial long, short, and spreading positioning has moved -14.83%, +9.35%, and -4.54%, respectively, from around the timeframe of my last post. In addition, the current 11-month Z-Score of Non-Commercial long - short positioning is 0.81, suggesting that these participants are obviously still net short, but not really to a significant manner relative to past history. Nothing too crazy to gather here. In terms Leveraged Funds and Dealer/Intermediary positioning on VIX futures, nothing too crazy has changed since my last blog post. Dealer/Intermediary long, short, and spreading positioning has done -2.51%, -3.46% and +14.70%, respectively, with Leveraged Funds -0.54%, -18.20%, and -14.36% for long, short, and spreading positioning, respectively. The current 11-month Z-Score for Dealer long - short positioning is 0.15, with that of Leveraged Funds long - short being 1.16. This shows no extreme positioning for implied volatility exposure by either party, and such not an immense amount of information can be gained from this. It is important to keep in mind that, especially in this particular case, dealers and intermediaries are not predominantly net short (at least in this current environment), since there is such a demand for yield in the carry trade of short vol. Because of this, Leveraged Funds actually increase net short vol exposures once implied volatility regimes revert back to normalcy (since vol is, in effect, a rate variable that displays auto-correlative properties, but I digress). This complex causes dealers/intermediaries to be more net long than in highly volatile environments.
This graphs gives a great visual representation of the net positioning of a variety of market participants on the S&P 500 E-Mini futures contracts. Again, now you may notice that, in this current market environment, dealers are predominantly net short investor of thus specific futures contract for the inverse of the reasoning given above. The variety of market participants within this chart seem to take up similar amount of the percent of total OI among ES futures contracts, with a slight thinning of control by the Asset Management group. Dealer/Intermediary net ES futures positioning has +11.68%, +6.63%, and +20.76% for long, short, and spreading positions, respectively, since 7/6/2021. Leveraged Funds have +14.91%, +11.90%, and +62.21% for long, short, and spreading positions, respectively, within the ES futures complex. The main takeaways here is that the 2 main market participants that I analyze for information purposes, the Dealers and Leveraged Funds (for reasons adequately explained in earlier posts) have increased their spreading positions drastically, with spreading positions of the two taking up nearly 10% of total open interest on ES futures contracts. The primary market participants, in terms of ES futures contracts, do not see a clear direction, and are thus increasing their neutral positions with no directional bias (spreading).
The primary purpose of this graphic is to show how bearish overall market positioning is on ES futures contracts, in particular the EW4U1 contract (I chose this simply because it had roughly 24 days left to expiration, no other technical reasoning). As you can clearly see, put options contracts on this particular future occupy about 84% of total OI positioning, with more than 98% of said puts being out-of-the-money, emphasizing the negative skew that is prevalent in equity markets. To reiterate this view, the Put to Call OI ratio is 5.18, with that of options volume being 3.84 on this contract.
Options-Implied Probability Distribution of Terminal Stock Prices
The above graphic is a great representation of what option markets are pricing in for the S&P 500 SPDR ETF, SPY, to end up by expiration (30 days). I took the price of butterfly securities (a popular metric for the second derivative of call price with respect to strike as I mentioned in my previous blog post). There are a variety of ways one could interpret this graph. First and foremost, options market participants are assigning 21.93% risk-neutral probability that the SPY ETF will end up between $455 and $465 by expiration. This is the single-most likely outcome as derived by ATM options prices. Now, I would like to clarify that this is not a risk-neutral probability density functions, as that would take more time and calculations, and similarly involves taking the integral of some higher-order polynomial to derive pseudo-implied probabilities. I go more in depth in my previous post behind the calculations of even this simple approximation. The market implies a 34.63% probability that the SPY ETF ends between 405 and 455 by expiration, and a 44.47% probability that the SPY ends between 455 and 495 by expiration. In essence, for 0 to +5% and 0 to -5% move in the ETF, the market implied probability is ~73.87% for a move between $453 and $475, and ~49.59% for the SPY to end up between $430 and $453 by expiration. Clearly, there exists overwhelming optimism for the SPY ETF to continue its robust performance. One thing I would like to point out is the degree of skewness of pseudo-implied probability distribution of the terminal stock price by expiration. This is due to the observed volatility skew prevalent in the options market, something I hope to touch on in more detail in future posts. The forward price of the SPY ETF as derived by at-the-money options prices is $456.02 by expiration (again, 30 days from now). This is to reiterate the degree of bullishness that is reflected within this ETF.
Price, Volume, and Volatility
The main takeaways is the continued bullishness of the S&P 500 SPDR ETF, SPY, with that of the log-performance of the VIX and VVIX index. As such, the implied vol-of-vol index recently broke trend, suggesting the potential entrance of a lower volatility on multiple fronts, on SPX derivatives, as well as VIX derivatives. With price appreciating more than 7%, and the vol landscapes down over -6% on both fronts, in addition to implied vol-of-vol breaking down, begets an interesting position. If you trust consensus options positioning, then this would be a clear signal to get short implied volatility and adhere to the carry trade. However, on the other hand, this could be a potential opportunity to add long-vol exposure on a cheap cost, with that of the VIX index, as well as the VVIX index testing the lower-bound of their respective regimes. The volume on the SPY ETF has accelerated 11% day-over-day, but has decelerated -14.57%, -16.15%, and -22.45% on a 5-day, 1-month, and 3-month timeframes, respectively. Meanwhile, price has risen +0.63%, +2.56%, and +7.67% on a 5-day, 1-month, and 3-month timeframe. Since vol has been declining while the underlying has risen, we can see that the correlation break between spot and vol that occurred a few months ago has solved itself, begetting the mention that a sharp rise in implied volatility would be the primary risk factor at play here. Thinning market breadth could spur a sharp spurt of implied volatility to the upside across the board. If one were to pursue a long-vol-oriented approach, the next of this blog post will detail specific contracts that might be reasonable.
Analyzing Contract Pricing
In this section, I will seek to find a few ES futures contracts that appear over/under priced on a relative fashion.
This graph depicts the front-month ATM implied volatility complex of ES futures contracts, typically less than 20 days to expiration on said contract. As one can clearly see, there was a reasonable spike due to the SPX and VIX movements. However, this front-month implied volatility is nearing its subsequent lows on a 1-month basis, suggesting a cheapening of implied volatility exposure on the S&P 500 derivatives complex.
This graph essentially depicts the degree of skewness in call and put options on ES futures contracts. As we can clearly see, deeper OTM puts are considerably more expensive than OTM calls, with the current difference between 15 delta and 25 delta calls implied volatility - puts implied volatility (a risk-reversal) being more than -8% and -4%, respectively, suggesting an expensive premium for downside protection. However, this premium is at relative lows compared to previous values on a 1-month basis, suggesting a decent time to partake in an outright downside hedge. Would I recommend this direct of an avenue? Possibly no, but for someone who is not as focused on volatility products in particular, this could be enough justification for a direct hedge.
This graph depicts how, overall, downside protection via put options on front-month ES futures contracts have gotten relatively more expensive on a 1-week basis, but still overall more cheap than a month ago, as depicted by the green line. Keep in mind, that the steeper the curve one plans to obtain, the more convex the derivative instrument that you buy, as the gamma of said options is "low", but the deltas of those volatility products accelerates in accordance to its gamma, providing the convex payout of options instruments. There are many other sensitivities, as well as higher-order sensitivities, but that is beyond the scope of this post.
This graphic depicts a smoothed-constant maturity of ES futures contracts, with the respective individually-listed ES futures contacts plotted with their respective at-the-money implied volatility. This is great for seeing a simple "relative value" play, looking where one contract might be more expensive, on a relative basis in terms of implied volatility, than another. If one were looking for a roughly 30-day exposure (for simplicity's sake), then the majority of ES futures contracts around that DTE should be fairly priced, relative to their other contracts. In addition, if I could also mention how ATM IV for these contracts are cheaper on a relative basis to that of the curve from 8/25/2021. In terms of a specific contract play, it is not too obvious of a decision, and this case would not be as critical as it has been in the past. Now, for the same idea of a graph, but the risk-reversal of said contracts rather than the at the money implied volatility.
This graph the smoothed, constant-maturity of the same ES futures contacts, but this time displaying the difference in implied volatilities between the 25 delta call and put. As one can clearly see, this graph gives us further information on which contract we would play, if seeking downside exposure in the form of a put contract. the ES future EYCU21 is extremely expensive on a relative basis, so really anything between E4AU21 and EW4V21 would be reasonable. There is no extreme skew prevalent within these contacts, and are below the smoothed line, as well as in the lower 25% of RR.
The purpose of this graphic is to show the historical 1-month realized volatilities of the constant maturities ES futures contracts. We can clearly see that approximately 50% of all observations for realized volatilities for each contact are predominantly between 8% and 9%, with some outliers at 7% and 11.25%. Looking at where historical volatility has been is extremely important in assessing the relative price of a volatility product. If historical volatility is at extreme lows, yet implied volatilities are extremely high, is there something that the option market is pricing in that we are unaware of, or is there an extreme dislocation in the market in regards to future volatility? These are some of the extremely important questions we must consider when buying these derivative products, delta 1 or not.
A quintessential part of buying or selling securities, especially in the volatility space, is knowing the relative pricing of implied and realized volatility to the past on a Z-Score basis. Even though the realized distributions of most stock returns exhibit leptokurtic properties (skinny peak, fat tails) as opposed to the gaussian distribution, it is still useful to use this estimate. The important line is that of the E-Mini S&P 500 ES futures contracts. A line to the left indicates a negative Z-Score, and vice versa to the right. The most noticeable feature is the green dot, which represents the ES_30 ATM implied volatility on a 1-year basis, which exhibits a Z-Score of -1.2939, representing how the current at the money implied volatility value for the ES_30 constant-maturity complex is relatively cheap compared to its history. The Z-Score of -1.2939 represents a -1.2939 standard deviation move away from the mean 1-year ATM implied volatility of 17.29%. Using a standard normal cumulative distribution function, we can calculate probabilities that the ES_30 ATM implied volatility will be below the current value by expiration (9.79% chance), that it will be between the current value and the 1-year average (40.21%), and that it will be above the current value (90.21%). To echo this fact, the 25 delta risk-reversal skew, on a 1-year basis, for the ES_30 constant maturity futures contract, has a Z-Score of 0.8195, showing that it is, relatively speaking, somewhat high (within 1 standard deviation of the mean, so nothing too crazy, but begets the same point).
This is a great graphical representation of the effects of time skew (a contango implied volatility term structure, hence how selling vol is generally a carry trade), as the at the money implied volatility for the ES futures contract expiring this month is 2.35% vol points higher, with the 20-day realized volatilities of both contracts considerably lower. The ES ATM implied volatility Z-Score of the EWU21 and EWV21 are -1.42 nd -1.25, respectively, suggesting a rather low point of ATM IV to historical standards, especially how 20-day HV is at a Z-Score of -0.55 for both contracts.
Conclusion
There is extremely bullish market positioning on the listed options space on the SPX and SPY, with different metrics implying higher prices. However, that on the ES space is extremely bearish. With the market at all-time highs, a potential dislocation in the pricing of implied volatility, as well as implied volatility-of-volatility, with thinning market breadth (volume) presents an opportunity to acquire insurance for a reasonable price. I did not go into the different avenues of executing said insurance, nor did I reference the specific pricing of events if they did occur and how that would effect our book. That is up to your discretion, I simply wanted to provide objective market information. To reiterate the introduction, this blog will make the transition to more of an educational standpoint, as my task of introducing fantastic, free market information has been carried out successfully. If anyone has any questions as to my sources, feel free to reach out. As always, get that green!
Comments
Post a Comment