“I consider everything a loss because of the surpassing worth of knowing Christ Jesus my Lord.”
-the Apostle Paul via Philippians 3:8
As a spiritual millennial, I can’t really find enough time in the day between taking and posting selfies, liking other people’s selfies, taking and posting pictures of my meals and liking other people’s pictures of their meals to find time to go to church. As such, one of the things I like to do to start each day is consume a daily devotional on my smartphone.
That verse comes from today’s ODB devotional titled, “The Ultimate Good” in which we are reminded that the best path forward in life isn’t just one filled with checking the boxes of good deeds, but rather one that is also grounded in the love that comes with an intimate knowledge of Jesus Christ.
My sincere apologies for the non-Christians among you; I certainly don’t mean to offend or impose. There is something to be said about the message, however, considering that even my Bloomberg terminal got in on the memo with their Quote of the Day today:
“To be nameless in worthy deeds exceeds an infamous history.”
-Sir Thomas Browne
Back to the Global Macro Grind…
In other news, domestic economic growth continues to accelerate on a trending basis off its 2Q16 lows – as most recently confirmed by this morning’s Jobs Report (we’ll have a note out on that shortly). This much was also confirmed yesterday with the advent of the JUN ISM Non-Manufacturing PMI, the MAY Trade data and weekly Initial Jobless Claims data, which rounded out the month of JUN. Specifically:
- Largely driven by a +2.8pt increase in New Orders to 60.5, the ISM Non-Manufacturing PMI accelerated to 57.4 in JUN vs. 56.9 in MAY. Excluding February’s 57.6 print, it’s the highest reading since AUG ’15. That’s important to note because this monthly indicator is currently the 7th highest-rated factor in our dynamically re-weighted, 30-factor predictive tracking algorithm that we’ve developed to forecast U.S. GDP growth intra quarter.
- With respect to the MAY Trade data: Export growth accelerated to +5.4% YoY from +5.1% in APR; Import growth decelerated to +6.6% YoY from +8.4% in APR; and growth in the Trade Balance decelerated to +12.0% YoY from +23.8% in APR. Each of these factors contributed a positive revision to our previous 2Q17E GDP estimate due to the fact that the latter two indicators are inversely correlated with Real GDP growth – which itself is no surprise give our persistent current account deficits.
- Initial Jobless Claims ticked up to 248k from 244k in the week prior, but the trend remains one that is hovering at historic troughs – especially on a population-adjusted basis. More importantly, the YoY rate of change of this data set is extremely informative as an indicator of real-time GDP growth trends. Specifically, the quarterly average YoY growth rate of Initial Jobless Claims carries a correlation coefficient of -0.86 to the YoY growth rate of Real GDP on a trailing 10Y basis, currently making it the 6th highest-weighted factor in the model.
Recall that our model itself is one that is non-linear in nature and Bayesian at its core. While it’s true that we use linear relationships to dynamically re-weight each factor in the algorithm, each factor’s weighted contribution to the forecasted marginal rate of change in YoY Real GDP growth – as well as the forecasted marginal rate of change itself – does not follow a linear beta + alpha relationship.
Rather, it employs a percentile-matching technique to account for the fact that economic growth could be gapping down (or up), insomuch as it could be slowly trending higher (or lower), or stochastically chopping around with no trending slope at all. Said simply, we want to avoid using linear relationships to forecast non-linear outcomes.
Bayesian with a large side of non-linearity – that’s Benoit Mandelbrot and Daniel Kahneman 101. And while my matriculation narrowly missed the former’s retirement at Yale (insomuch as I missed the latter altogether by eschewing Princeton), I have been blessed to receive their gospel in the form of their seminal literary works – The (Mis)Behavior of Markets and Thinking, Fast and Slow – which we firmly believe to be the “co-Bibles” of modern finance.
If those works are Biblical in nature to us, then Mr. Market himself is the Apostle Paul, someone who’s dedicated his life to preaching the gospel of the future to the masses – all we have to do is listen and be willing to repent.
One of the ways in which we’ve amplified our ability to do both is via analyzing proprietary volatility-based indicators – a process my colleague Ben Ryan captains for our team. Options markets are not deterministic, but rather, inherently probabilistic and possess the power to signal very worthwhile indications about consensus expectations of the future if you have a quantitatively-oriented process to extract said clues.
With respect to our favorite factor exposure on the long side across all of global macro (i.e. large-cap U.S. Tech stocks) he had some interesting callouts this morning that help strengthen our conviction that the ongoing -5.3% correction in the QQQ’s is just that – a correction and not the start of a new trend lower:
- The 6-month term structure on the QQQ ETF is now in backwardation, which means that front-month implied volatility is higher than implied volatility on a contract that expires in six months. Historically, this has provided a powerful buy-the-dip signal for this exposure. Refer to the Chart of the Day below for more details.
- In addition to that signal, the spread between the 6-month term structure of the QQQ’s and the SPY is as historically wide as it gets – which means that investors are as bearish on the prospective immediate-term relative performance of large-cap technology stocks as they have ever been. As the vast majority of our volatility indicators implicitly suggest, we want to be fading consensus leans in the futures and options markets.
- The aforementioned signal is confirmed by the fact that the spread between 30-day implied volatility between the QQQ’s and the SPY is a whopping ~3 standard deviations above its historical mean at the current 7pts wide.
- Open interest in QQQ puts has recently spiked to the highest level since 2008, which implies large investors are using the options markets to hedge for further downside – en masse, after the move.
- The 30-day implied volatility spread (to 30-day realized volatility) on the QQQ’s is lowly +8.3% vs. +33.7% for the SPY. While at first glance that might seem like relative complacency, it’s important for investors to consider the fact that, at 15.7%, realized volatility on the QQQ’s is in the 38th percentile of historical readings vs. only in the 6th percentile for the SPY at 7.5%. Adjusting for relative levels of volatility, that would imply buy-side consensus now expects a massive rotation out of large-cap tech stocks – again, after the move.
Fade consensus or chase consensus? You tell me which one is the better long-term risk management strategy…
All told, we’ll use this subtle reminder that the prices of the assets you are long don’t have to go up every day, week, or month to reiterate our conviction in remaining long of the #Quad1 playbook with respect to the domestic equity and credit markets.
Keep your head on a swivel,