Back in May, with the shock from February's VIX explosion still fresh in traders' minds, Goldman's Chief Markets Economist Charlie Himmelberg became the latest Wall Street strategist to admit the threat to the market posed by HFT. Picking up where our original warning from April 2009 left off, the Goldman strategist warned that HFTs – due to their inability to process nuanced fundamental information - may trigger surprisingly large drops in liquidity that exacerbate price declines, and result in flash crashes, something we have observed previously on countless occasions.
Himmelberg highlighted the growing market share of HFT and algorithmic trading across all markets, and warned that the growing lack of traditional, human market-makers has made the market increasingly fragile.
He is, of course, correct as active traders will attest, if nothing else then by the collapse in market liquidity around critical, market-moving events when HFTs strategically “pull out” from the market, making price swings especially sharp and resulting in a spike in volatility as shown in the schematic below.
As we discussed in greater detail back in April, the relentless, and increasingly commoditized ascent of HFTs, as well as the change to market structure and topology in a post-Reg NMS world, prompted Himmelberg to conclude that we live in a world where the biggest threat is not market leverage, but periods of sudden, unexpected and acute losses of liquidity. Or, as he put it, “liquidity is the new leverage.” This is how he explained it:
That analogy is meant to invoke the potential unrecognized problems or imbalances that build up over the course of long expansions. Financial leverage was obviously the imbalance that built up during the pre-crisis period, but that has been contained in the current cycle. In this cycle, there have been dramatic shifts in the way that secondary markets source liquidity, but this market structure has not yet been stress-tested by a recession or major market event. I therefore see a risk that markets are paying too little attention to liquidity risk, much as they previously paid too little attention to the risks posed by excess leverage.
Furthermore, the fact that for the past decade global capital markets and risk assets have been constantly prodded higher courtesy of central bank liquidity injections, has exposed them to increasing instability not only at the micro level but at the macro: while virtually all HFTs – and roughly a third of all active asset managers (who were simply too young during the global financial crisis) have never encountered a market crash, the big test will be what happens, and how the market would react during the next crisis in which there is no “big picture” central bank intervention, to make “buying the dip” at the micro, HFT level, the correct response, even though so far aggressively purchasing the “crash” has been the correct response every single time…
… as volatility always inevitably tumbled, making selling vol one of the preferred “carry” strategies for numerous investors classes (ultimately leading to the historic VIX explosion of February 5 which blew up several of the most popular retail vol-selling strategies such as inverse VIX ETNs in a matter of minutes).
But what happens if one day the market crashes, HFTs, quants, algos, CTAs and various other program traders pile on by selling or merely shorting into the plunge and the market does not rebound?
That was the main concern voiced by Brian Levine, Goldman’s co-head of Global Equities Trading, who back in June in an internal Goldman interview said that while most facets of the current marketplace are not new per se, there is one aspect of that the current broken market structure that does keep him up at night. As he admits in the interview, “what’s more worrisome to me is a real flash crash, which I define as a situation when the market “breaks.”
Indeed, the market breaking is surely high on the list of every trader’s worst nightmares, and reminds us of what we predicted several years ago, namely that when the “big one” finally hits for whatever reason, there won’t be a 20%, 30%, 40% or more drop in seconds. The market will simply be halted indefinitely (see “How the market is like SYNC which was halted indefinitely”).
This is how Levine described his trading nightmare, the one in which the crash is not a “flash” and the market simply breaks:
The data is wrong, everything trades at dislocated prices relative to the NBBO, and everyone—justifiably—widens their spreads. That happens almost every time there’s volatility, largely because message traffic increases dramatically. This is due to the fact that the opportunity set is greater and there’s no economic disincentive for sending messages to the market, so more electronic orders come in. This slows the system, widening spreads and generating price dislocations, which triggers even more orders and compounds the delays—a predicament that is only further exacerbated by the fragmentation of the equity markets. As this happens, stocks may trade outside of the NBBO briefly in millisecond or microsecond increments, constituting what I consider a genuine flash crash. All of this becomes a negative feedback loop that causes more volatility.
Interestingly, if you define a flash crash by the percentage of executions that took place outside the NBBO, one of the largest ones occurred in 2008 after the first TARP bill failed, according to internal analysis we did a few years ago. And the market didn’t snap back, with the SPX closing down 10% on the day and on its lows. I think that may have been why there wasn’t talk of a “flash crash” afterward, but clearly the market structurally failed pretty badly that day, too. This suggests to me that, in a situation with actual bad news, the current US market structure may not be able to handle it, and there could be a downward spiral.
In other words, there will come a day “with actual bad news” when the selling onslaught is so broad, not even BTFD HFTs will be able to resist the sudden avalanche of selling. That’s the day when the increasingly fragile market, one in which “liquidity is the new leverage” will officially break and stocks will “trade outside of the NBBO constituting a genuine flash crash” in a “negative feedback loop that causes more volatility.” A selloff from which there will be no “snap back.”
We bring all this up now because one of the main culprits who have been identified for the relentless selling in recent weeks has been the all-encompassing category of "algo traders", which for the sake of simplicity (if not necessarily accuracy) includes quants, HFTs, CTAs and trend-following strats, risk-parity and vol-targeting funds, passive investors and numerous other signal and factor-based market participants. In fact, earlier this week none other than Treasury Secretary Steven Mnuchin blamed HFTs for exploding market volatility (granted, it was a thinly veiled attempt to deflect attention from both the ongoing chaos at the White House and the Fed's ongoing financial tightening).
In this vein, on Friday another major bank joined Goldman as listing algo traders as the biggest latent threat to the market, when Deutsche Bank's chief international economist Torsten Slok not only said the nearly 20% drop in the S&P can be blamed in part on algos but - in his annual list of 30 top risks to the markets - said that the biggest threat is that "algo-driven, risk-parity fire sale in equities and credit continues."
While the catalog is not meant to be ranked in order of risk severity, Slok told MarketWatch that the risks at the top are of more immediate concerns, and are those which Slok's clients have consulted with him most often.
What is most surprising, is that Slok put uncontrolled algo-selling above such conventional concerns as a global economic slowdown, a spike in the dollar, a semi-failed Treasury auction, a spike in US yields and/or inflation, quantitative tightening, Italy's fiscal situation, a housing market crash, a Chinese crash, and so on (understandably, a bank run on Deutsche Bank was missing from the list of 30).
So why is Slok so scared of algos?
"I have a Ph.D. in economics. I am the chief international economist at Deustche Bank Securities, and I spend all day thinking about the explanations of why the stock market goes up and down," he told MarketWatch. "Usually there’s a good explanation, be it from earnings revisions, new economic data or surveys of sentiment."
"The very unique thing about markets since October is that we have on all three fronts seen little change whatsoever,” he said. “If you look at all these data, it is just really strong, and it can’t justify the decline we’ve seen in the stock market."
So, in picking up where Mnuchin left off, Slok merely assigns the "incomprehensible" selloff on those who often trade without a clear reason or logic.
"If you can’t explain why the market is moving, you should be suspicious of the moves the market makes," Sløk said, adding that the rise of algorithmic trading and rules-based investment strategies like risk-parity investment funds can likely take the blame for the outsized moves in the market we’ve seen in recent weeks.
As a result of the dominance of algo trading, Slok argues that momentum has emerged as the most important force in markets, something we have claimed for year. However, one key reason why trading has become so complicated for most, and certainly the algos, is that there is currently virtually no momentum in the market - with the MTUM ETF which tracks momentum stocks having its worst month and quarter since its 2013 inception - results in making any attempt to piggyback on the market a money-losing trade.
What does that mean for traders? According to Slok, investors, or rather their computers, "are selling because stocks are falling, not because of changes to the fundamentals." Indeed, the more stocks fall, the more selling emerges; just this past Thursday we noted that according to Nomura, for the first time in three years, CTAs had just flipped net short, suggesting that any further selling would only result in more shorting - and selling.
Beyond the risks posed by algorithmic trading, the theme that ties many of the risks together on the list, Sløk said, is the difficulty investors are having quantifying them. The slowdown of the Chinese economy is one example, as are rising trade tensions between the U.S. and China. In both of these situations, investors have little history to turn to help them understand how these forces will act upon equity markets.
"The nature of the challenges facing the market, therefore, is itself a type of risk, as their unquantifiability discourages investors from buying risk assets, because they don’t know how to hedge them. We’re in an unusually difficult situation,” Sløk said. "The fact that there are so many topics on this list, and that they are so diverse, tells you something important about the market."
Indeed it does, but it also tells us that when one is trying to explain away the bursting of the biggest ever bubble created by central banks, it increasingly appears that "algos" (whatever that means) will be thrown under the bus, because we are hardly the only ones who note that it is only when the market is dropping that Wall Street starts blaming "algorithms." When it's smooth sailing higher - as has been the case for much of the past decade - not a single trader could be bothered with the threat that algos pose.
Still, despite some latent criticisms, we wouldn't discount Slok's list; in fact one looks at his "30 risks to the market in 2018" which he published a year ago, shows that many of his "hypotheticals" emerged as credible threats to risk assets.
Incidentally, while Slok certainly raises some critical risks for the coming year, we don't think algo trading is the biggest one: after all, all that would take for momentum to reverse violently to the upside will be for the Fed and other central banks to throw in the towel on tightening and resume easing, in the form of either more NIRP or outright QE. In fact, in light of the ongoing escalation in tensions between Trump and Powell, one can argue that it is only a matter of time before Powell is fired - especially if the market slides deeper into a bear market - and is replaced with some uber-dove.
Which is why in our opinion, the biggest risks listed by Slok are the last two:
- Fed and ECB re-start QE and risky assets dont rally, and
- Monetary and fiscal policy are out of ammunition and the world experiences a Minsky moment.
We find those two to be especially critical because as BofA's Michael Hartnett noted last week, we may already be experiencing the traces of the first. As Hartnett wrote last Friday, markets are starting to "lose the plot" as it is rare for combo of such capitulation out of risk, capitulation into US dollar & Fed dovishness not to spark a rally, with the BofA CIO writing that "the only reason it would not is fear of policy impotence." In other words, fear that the central banks no longer have control.
If that is indeed the case, then Slok is correct: a global, and long-overdue Minsky moment will be the biggest risk for 2019.