Whenever we get a technical selloff like what unfolded on October 10, when systematic flows and option gamma hedging precipitated a sharp drawdown, there's a rush in certain bearish circles to quantify the assumed impact of further programmatic de-risking. More simply: There's a demonstrable tendency for folks to extrapolate as to what might happen if momentum signals turn more negative, if key downside strikes are hit and if volatility stays elevated.
That's understandable. While technical drawdowns and flash crashes are a fixture of modern markets, and while the wider investment community is certainly more attuned to those risks than they were, say, three years ago, days like October 10 and February 5 have retained their ability to "shock and awe" because, frankly, most people still haven't wholly accepted just how deeply embedded and consequential these strategies are in markets.
People fear things they don't understand, and "fear sells" (as they say), so there's a cottage industry in the financial media (and the financial blogosphere such that it exists) for postulating worst-case scenarios following technical selloffs. You saw that last month with belabored attempts to suggest that selling from slower-moving, vol.-targeting funds would invariably overwhelm discretionary buying and fundamental flows following the October 10 momentum unwind.
The difference between analysts and the financial media/blogosphere is that analysts also talk about how systematic flows drive the price action on the way up - systematic re-risking, if you will. For instance, following the February rout, JPMorgan's Marko Kolanovic suggested that re-risking from systematic strategies would help support the market going forward. On May 1, he wrote the following:
Systematic investors thoroughly de-risked by selling ~$300bn of equities this year. If volatility can stay contained and the market can make modest gains, systematic investors would re-risk. For instance, three-month price momentum is more likely to turn positive in early May (three months since the February crash), which could result in buying of up to ~$50bn by trend followers. Volatility-sensitive investors would be slower to re-risk at ~$10bn per week, but this could accelerate if volatility declines more substantially. In addition, buybacks are expected to pick up activity after individual companies announce earnings, and may add up to ~$20bn of inflows per week at its peak. We often hear from clients ‘there is no one to buy equities’ and the above summary argues that is not correct.
U.S. equities would go on to log a better than 2% gain in May on their way to a five-month winning streak before the bottom finally fell out last month.
On Tuesday morning, at roughly 10:15 AM ET, Kolanovic was out with a new note, and in it, he suggested that October's "rolling bear market" could morph into a "rolling squeeze higher" into year-end. His rationale revolved partially around re-risking from systematic investors whose exposure was cut markedly during last month's selloff.
“Systematic investors are also near the bottom of their exposure – volatility targeting strategies’ equity holdings are similar to February lows, and many CTAs are outright short or out of equities”, Marko wrote in his Tuesday missive, before noting that “a ~10% decline from the peak and markets turning negative for the year triggered all kind of institutional stops, driving the selloff deeper.”
I don't want to dive too deeply into the weeds (those interested can read more here), but suffice to say Marko thinks that as long as volatility declines into year-end, systematic investors could re-risk to the tune of ~$100 billion.
The point is, if you want to flag systematic deleveraging on the way down, you also need to account for its impact on the way back up, lest you should end up in that unfortunate camp of bloggers who scream from the rooftops about the risk of further systematic deleveraging following technical drawdowns only to go completely silent when the washout occasioned by those drawdowns creates cleaner positioning, setting the stage for systematic re-risking and a technical bullish flow going forward.
Ok, so in that same vein, one of the interesting things about October was that once the systematic selling ran its course (or if "ran its course" isn't quite accurate, then we can say "once the October 10 rout helped clear the deck"), the market was unable to sustain a rally. Why? Well, here's Marko with the quick explanation:
Technical selling from the first part of the month (~$150bn) abated, only to be replaced by hedge fund de-risking and a rout in Tech stocks. The global HF equity beta dropped from ~95th percentile in September to ~15th percentile now (over the past 5 years), one of the largest and fastest declines on record.
As regular readers are undoubtedly aware, that's something I talked about over and over again in these pages (and even more frequently over on my site) last month. October was characterized by wild swings for consensus hedge fund longs in Growth and Tech (QQQ). The Long/Short crowd's exposure plunged in the first couple of weeks of October as a steepening in the curve catalyzed a rotation into Value (IVE) versus Growth (IVW), and that rotation set the stage for the Momentum unwind that played out the following week when, on October 10, the iShares Edge MSCI USA Momentum Factor ETF (MTUM) had its worst day on record.
Not to put too fine a point on it, but I warned you about that five days in advance in "Jobs Report, Tech Woes, And Curve Steepeners: Thoughts On Friday". Here's a look back at just how outsized the moves were with respect to the swings between Growth and Value last month (notable days are annotated):
In a testament to the rotation, the iShares Russell 1000 Value ETF (IWD) saw the largest monthly inflow of the year in October.
That ETF outperformed its Growth counterpart by the widest monthly margin since January 2008 last month.
Generally speaking, that kind of thing is bad news for hedge funds. This isn't a perfect representation, but it gets the point across:
That's just the HFRX Equity Hedge index versus the S&P (SPY) and what you want to note is that hedge funds started to massively underperform from July, following Facebook's (NASDAQ:FB) post-earnings plunge. They (hedge funds) were underexposed as the market made new highs in late August, leading to a frantic grab for exposure (i.e., they needed to catch up) which helped push benchmarks to new highs in September. Tragically, that "grab" (which left the Long/Short crowd with near-record exposure) came at just the wrong time. They were forced to massively de-risk as crowded longs were shellacked in October.
On Monday, that dynamic hit a kind of hopeless nadir following the negative tariff headlines. Specifically, Monday was the fifth worst one-day drawdown for the Long/Short crowd of the year, according to Nomura's Charlie McElligott, who has a number of models that track performance of various strategies. Here are some amusing factoids about that from Charlie's Friday note:
Monday showed to be the largest underperformance in the top 10 ‘most crowded’ longs vs the SPX since 2010 on forced-deleveraging.
Both “gross-” and “net-” exposures @ ~ 2 year lows, with 5Y delta-adjusted “gross” just 25th %ile and 5y delta-adjusted “net” at just 9th %ile.
Well, when it rains it pours, because on Tuesday and Wednesday, U.S. stocks staged their best two-day run since February. Then on Thursday, ostensibly positive news on the trade front (Trump's tweet) catalyzed a massive risk rally in Asia.
As noted above, the systematic crowd was just waiting to re-risk following the October washout and on top of that, macro funds were sitting on some dry powder after going into last month on the right side of the trade (more on that latter point here). And so, CTAs jumped on the move higher, with Nomura's Trend CTA model showing S&P positioning rising to +60% Long by Thursday from just +31% Long two days previous. As for macro funds, McElligott writes that the surge higher for equities developed after the macro crowd took profits on downside hedges which, Charlie notes, "meant yards of equities delta to buy over 48 hours and [a] quick pivot into outright upside expressions through early December."
That, clearly, is a disaster for the Long/Short hedge fund crowd. Simply put (and McElligott says exactly this), they "literally low-ticked" their exposure on Monday at the worst possible time. Here's a handy visual:
And here's S&P futures from the Monday lows through the Friday highs:
That, McElligott wrote on Friday, is "soul-crushing" for equities funds who had "massively de-grossed/de-netted/de-beta’d their portfolio risk over the past month’s market calamity." Again, they "literally low-ticked" it.
So what did they do? Well, that's the subject of some debate. The common sense, quick take on this situation is that those funds were hamstrung in their ability to catch up after October. That kind of abysmal performance leaves you unable to re-risk - you've got no P/L to play with and investors are probably calling up and asking for explanations. This is something Marko Kolanovic alluded to in his Tuesday note cited above. To wit:
Investors should keep this risk in mind – namely that an October ‘rolling bear market’ turns into a ‘rolling squeeze higher’ into year-end. This would cause further underperformance of active managers relative to broad indices.
For his part, Nomura's McElligott thinks the Long/Short crowd was in part responsible for this week's rally. They are, Charlie contends, running a "massive synthetic short-gamma in the equities space." With options expensive, those fundamental investors might have been forced to panic-grab S&P futs and ETFs in a hapless effort to not get left even further behind as the market rallies. Here's an excerpt from a Thursday note:
U.S. Equities / SPX continue to trade like there’s an implicit & incremental source of “short gamma” in the market: that “short gamma” source would likely be contained within the “fundamental” universe (both HF and MF), who are not actually short option delta per se, BUT are increasingly likely to act as “forced buyers” the higher we travel. They are de facto “getting shorter” the higher the market moves in their absence — thus in my mind, they are “dynamically hedging” with spastic trading in Spooz (as opposed to options due to their “richness” right now).
Needless to say, that is hilarious. I mean, it's the exact opposite of "hilarious" if you're one of those funds or, worse, an investor in those funds, but what I want to drive home to readers here, is that this kind of thing is brought to you by modern market structure. This is the interplay between: 1) systematic strats (CTAs and vol.-targeting funds re-risking after their exposure dropped in October), 2) "smart"-beta (factor-based investing helping to turbo-charge wild swings from one style to another), and 3) high position concentration/ overcrowding by hedge funds who have been forced to pile into Tech/Growth over the years as Howard Marks' "perpetual motion machine" dynamic pushed those names (and the benchmarks they drive) inexorably higher, leaving active management scrambling to justify higher fees by simply trying to out-benchmark the benchmarks.
This is what's going on behind the scenes and as you can see, it is a tangled mess, which is complicated immeasurably by the myriad geopolitical narratives that bombard the market on a daily basis.
The upshot going forward is that there seems to be some general agreement around the idea that between systematic re-risking (assuming equity volatility moves lower), buybacks and rebalancing by fixed-weight managers (i.e., reallocating to equities after the drawdown), there's scope for a squeeze higher into year-end.
But that's just the narrative from the folks who focus intently on the dynamics described above. There are of course three wild cards: 1) the midterm elections in the U.S., 2) the evolution of the trade war, with the main question being whether or not the Trump administration will move ahead with tariffs on the remainder of Chinese imports, and 3) the Fed, which appears to be countenancing the drawdown in equities as a way of transmitting the equivalent of a 25 bps rate hike through the financial conditions channel.
On that latter point, it would take a roughly 15% drawdown in the S&P from the highs to achieve the financial conditions equivalent of a 25 bps rate hike. If we get that, along with a December hike, the market would likely to start to anticipate the end of the hiking cycle sometime in mid-2019. The only question at that point is whether the Fed will have already tightened the economy into a slowdown.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.