There’s quite a bit of uncertainty out there, even as equities are clearly predisposed to extending the rally and loath to throw in the towel on what, until late last week anyway, was widely described as a “melt-up”, if not a “blow-off top” (of course, neither of those terms are real – there is no set definition for either).
In a testament to just how wobbly things are amid virus jitters, Treasury futures volumes (which have been running high) jumped Wednesday morning in the US as headlines suggested Lufthansa and American Airlines are set to cancel some flights to China, where the Wuhan outbreak has killed more than 130 and sickened more than 6,000.
Looking briefly at the specifics, American will cancel flights from Los Angeles to Shanghai and Beijing, starting February 9 and lasting through March 27, according to CNBC. Service to China will continue from Dallas Fort-Worth, apparently.
The largest monthly drop in pending home sales since May of 2010 didn’t help. Equities turned negative and dollar-yen dipped. Although the knee-jerk moves were quickly faded, it speaks to the push-pull between a market that wants (demands?) more on the upside for risk assets, but recognizes the outside chance the virus will deep-six what was still only a nascent recovery in global growth.
Given the fanfare around Monday’s rout which brought the first one-day decline of 1% or more since October (bottom pane in the visual), it’s worth noting that the current blip is just that – a blip. It barely even registers (top pane).
That’s either good news, or bad news, depending on how you want to look at things. On one hand, it’s a testament to the market’s resilience. On the other hand, it suggests that Friday and Monday will not be enough to let the steam out of the screeching tea kettle, if you will.
“Pull backs of 3-5% in the S&P 500 have been typical every 2 to 3 months historically [and] the last such pull back occurred in early October”, Deutsche Bank reminds you, on the way to noting that “at over 3 1/2 months, the duration of the rally since then is already well above average (86th percentile)”.
Another two weeks without a 3% pullback and the current stretch will be in the top 10% by duration, the bank’s Parag Thatte cautions. And it’s not just the duration. The size of the rally (~15%) is “also larger than the historical average between such pullbacks (+10%)”.
Normally, pullbacks need a catalyst and yet, as regular readers (and anyone who’s been paying attention for the last half-decade) are acutely aware, downside price action is almost always made worse by stretched positioning, both from the discretionary crowd and the systematic community.
Deutsche reminds you that in January of 2018, “positioning for discretionary investors was extremely elevated along with that of systematic strategies”. Over the course of the last several weeks, there have been all manner of comparisons between the early 2020 melt-up and a similar surge post- the 2017 Trump tax cuts. Thatte goes on to write that “in contrast, near record positioning currently reflects systematic strategy allocations at maximum and discretionary positioning clearly overweight but below early 2018 levels”.
You can get as granular with this analysis as you like. Nomura’s Charlie McElligott, for example, provides extensive documentation, backtests, P/L, etc., and offers specific trigger levels vis-à-vis CTA de-risking across virtually all assets.
But from a kind of “roll it all up and give me the broad strokes” perspective, DB’s Thatte offers this:
Within systematic strategies, Vol Control funds, CTAs and Risk Parity funds are all at historical maximum equity allocations and the risk to their exposure is clearly to the downside, especially for Vol Control. However with volatility having been subdued for an unusually extended period, short lived spikes are unlikely to trigger selling. Discretionary investor positioning on the other hand is usually closely tied to growth and other fundamental indicators. Unlike in January 2018 when discretionary positioning was extreme in the context of very strong growth, it has now been rising since September even as growth remains subdued, and has completely disconnected from it.
And really, for those of you who don’t revel in tracking systematic exposure and otherwise trekking around in the quant weeds, that latter point is perhaps easier to digest.
Even most casual observers are aware that ISM manufacturing remains largely disconnected from everything other than bonds. Although the US services sector has held up better, the disconnect is evident there too. (And save me the “chart crime” accusations – it is what it is. It’s not like I’m plotting exotic data series here or something. These are just simple things presented “as is”.)
With that in mind, and in the context of the last excerpted passage from DB above, I’ll leave you with two final visuals that help drive the point home.