“A stock that has dropped very sharply compared to the market becomes riskier at the lower price than it was at the higher price”
-Warren Buffett
Various forms of Buffett principles and wisdom were consolidated into a book called “The Essays of Warren Buffett”. On multiple occasions so far, I have found myself thinking “I knew he was going to say that.”
Possibly the key insight that sums up Buffett’s investing career is that he actually does what he says he’s going to do in every investing environment. Most of us either can’t stay out of our own way or don’t have the luxury of waiting through the long periods of pain associated with capturing the “value premium” [not a Buffett phrase but a fundamentally parallel quant term].
The quote above is an excerpt on Buffett’s thoughts about the beta component of the Modern Portfolio Framework for a stock’s “risk” relative to a benchmark.
If you only read our commentary sparingly, you would still know that valuation based solely on cheap financial-statement generated multiples is not our starting point on global macro asset allocation.
However, we do spend quite a bit of time observing that stock, sector, or style factor exposures price in more future risk after going down relative to an index or benchmark.
Back to the Global Macro Grind…
QUAD1 has packed a real punch in 2017. Although it’s far out, the most probable set-up from a Growth, Inflation, and Policy standpoint is Real Growth Accelerating through Q1 of 2018. This set-up of course is a slave to the data, and you don’t get multiple quarters of growth accelerating without some turbulence mixed in - the last month of macro has been driven by reflation’s rollover pausing which has caused a more hawkish ripple effect felt in currencies, interest rates, and yield-sensitive sectors.
An obvious question we get is how do you know the outperformance in your preferred exposures haven’t already front-run the continued QUAD1 set-up you’re expecting, since after all the Nasdaq 100 and Tech sector have only added +3-4% of the +21-22% YTD performance since the beginning of June? What’s the non-consensus short-term development within this trivial growth accelerating environment that could present a surprise?
We can write about our tested risk-management framework for contextualizing derivatives markets until we’re blue in the face, but the bottom line is that volatility factors provide a good peek at what consensus thinks about forward-looking asset class or sector dispersion.
This process is rooted in futures and options contract positioning, open interest, volume and implied volatility trends to name a few.
Looking at relative volatility expectations within U.S. markets, a ratio series of the Nasdaq 100 volatility index (VXN) relative to the S&P 500 volatility index (VIX) currently trades in the 99th percentile of historical observations based on a 10yr time window.
VXN was introduced as an index in February 2001, so if we pull back the curtain to inception, this VXN/VIX ratio currently trades in the 91st percentile. The index was created in front of the last, and possibly most violent, leg of the tech bubble aftermath from January to April of 2001 – this period is the only point in time that this VXN/VIX ratio was pushed higher.
Anxiety about large-cap growth is ongoing. Using volatility factors to manage the sizable FAANG weightings that drive things like NDX, XLK, or QQQ, which are traditionally good QUAD1 exposures, is an important part of our risk management process. It provides color and idiosyncratic idea generation. After all these options markets are deep, liquid, and everyone has to care.
This process is especially important as the big Hedgeye GIP Model QUAD driven moves occur in the early shift from one QUAD to another. This QUAD 1 shift is not early stage, so we use the head-fakes to manage the shorter-term risk within the QUAD shifting.
In short, concluding that the 937bps of outperformance in XLK over SPY or the 1011 bps of outperformance for the Nasdaq 100 index over the S&P 500 index is a “crowded” exposure set to mean-revert any second now does not seem to be the case…
- After the Nasdaq, QQQs, XLK put in the all-time lows in realized and implied volatility in March and April, volatility expectations have been widely divergent from the rest of the equity market.
- CFTC net futures and options positioning from the speculative crowd has been trimmed by 73K contracts since the beginning of May to net long 40K contracts in Nasdaq Minis currently. This translates to -1.4x and -0.7x on a TTM and 3Yr Z-score basis.
- The blow-out in relative implied volatility for the Nasdaq 100 or technology sector began in June, and this trend has not mean-reverted. Relative to the S&P 500 (SPY), the front month implied volatility spread in XLK is +1.4x and +2.3x extended on a TTM and 3Yr z-score basis. On average over the last 5 years, XLK implied volatility usually trades at a 2-point premium to SPY. Since June this spread has trended 4-6 points wide and is now on the low-end of that range.
Observing trends in implied volatility and volatility surfaces are directly related to the observation that a stock looks relatively more risky after it’s gone down, a concept Buffett ponders above. Things like volatility (standard deviation/variance) and skew are statistical descriptions of a distribution. And again, from a macro utility perspective, they offer insight into prevailing and evolving investor expectations and sentiment.
To use an example with the “implied volatility premium” factor on a 30D time window:
- Implied volatility traded at its largest DISCOUNT to realized volatility of the year two weeks ago at the all-time high (15-20% discount) in the Nasdaq 100 index
- After Monday’s -1.1% decline (-1.9% w/w), the implied volatility PREMIUM popped to 21%
- Now after the index has fully recouped that decline, the implied volatility PREMIUM has compressed to 12%
So after a move to the high-end of the risk range or fractal pattern, forward looking expectations for volatility are lowest? This is a behavioral disconnect, observed through the implied volatility premium, and one we use to identify opportunity.
Monday of this week marked yet another time when the Nasdaq 100 index broke and closed below its 50-day moving average. This was driven by a dumping of FAANG-like baskets. Pull up a chart of the Nasdaq 100, or any FAANG stock – the “get out” strategy using the 50-Day moving average has not worked well and the recent pullback was no exception. Managing winning exposures within a robust macro framework requires new age tools.
Our immediate-term Global Macro Risk Ranges (intermediate-term TREND views in brackets) are now:
UST 10yr Yield 2.20-2.35% (bullish)
SPX 2 (bullish)
RUT 1 (bullish)
NASDAQ 6 (bullish)
VIX 9.41-10.98 (bearish)
USD 92.00-93.45 (neutral)
EUR/USD 1.16-1.19 (neutral)
Oil (WTI) 49.08-52.86 (bullish)
Gold 1 (neutral)
Good luck out there today,
Ben Ryan
Macro Analyst
