Some of you may remember the quant crash of August 2007.  As Andrew Lo wrote in his initial diagnosis[1], “losses were initiated by the rapid unwinding of one or more sizeable quantitative equity market-neutral portfolios…which then put pressure on a broader set of long-short and long-only portfolios.”

Fast forward nearly 12 years, and there was a strange phenomenon occurring in the market in May 2019 which didn’t get as much attention.  It was slightly more hidden, and appeared within the subset of stocks that had relatively high market betas.  Basically, there was a tremendously exaggerated (we believe) difference in performance between the previous winners and the previous losers.

This time it was in favour of quant-driven strategies, and before commenting further, we should say that - over time - we actually do expect there to be momentum in some share prices, and it should surprise no one (at least no one that thinks about momentum or factors) that there are typically annual excess returns to be had from being long a basket of winners and short a basket of losers.

But in May, it was next level.

And it wasn’t just May. Using components of the Eurostoxx 600, and now that we have the benefit of hindsight, we see that the high-beta “losers” initiated their dreadful underperformance vs. the high-beta “winners” on April 18, 2019, and continued underperforming for exactly two months (which also is interesting, but probably random) until the close on June 17, 2019.

And how well did the top-decile previous winner 15-stock portfolio selected from a top-quartile 150-stock high-beta European universe do during this period?  Well, that 15-stock basket that had already been up 50% on average over the previous 11 months generated another 3.5% positive returns.[2] 

Okay, so that in isolation isn’t a big deal. 

But how about the losers?  Well, the 15 worst performing stocks over the previous 11 months, which were already down 37%, lost another 16.6%.

So this “market-neutral” long-short portfolio during those two months made 20.1%.

And if we expand the portfolio size out to the top quartile winners and losers of the top quartile of beta (so baskets of 37 names on both sides), the market-neutral L/S portfolio still generated 15.2% excess returns.   

Irrationality 1.png

So this extreme momentum was still a thing even in a more diversified, ~75 stock long-short, zero-beta portfolio.  Now who would go out and buy a basket of stocks that already went up an average of 50% and sell a basket of stocks that were already down 37%, and execute that strategy with the gentle touch of a bull in a china shop?

This guy. 

Irrationality 2.png

The dumb robot. 

Actually, probably a gang of dumb robots.  We’ll call them “the machines”.

Well, it was the machines at first. But we contend that trend-followers (mostly also machines) soon joined in, and then as the tail started wagging the dog even more violently, some active fundamental managers may have become nervous, thinking they were missing something (or perhaps were stopped out by their risk managers).

This all mixed together to serve up a fiery cocktail for those that had exposures to downtrodden stocks of businesses they believed were successfully restructuring or fundamentally improving, but where Mr. Market wasn’t yet paying attention (you know, businesses that we and others like us like to own).

And before you ask if this all wasn’t just something particular to this dataset, or perhaps specific to Europe, here is a similarly built L/S portfolio based on the S&P 500, during the exact same period. 

Irrationality 3.png

So it happened in the US as well.

This “buy the risky winners and sell the risky losers” thing was a global phenomenon.[3] 

We don’t think this was some reward for bearing risk.  We don’t think it was a fundamental panic.  We don’t think there was idiosyncratic news driving this behaviour.  We don’t think this was actively irrational. 

It was passively irrational.[4]


[1] What Happened To The Quants In August 2007, Khandani and Lo, https://web.mit.edu/Alo/www/Papers/august07.pdf

[2] Yes, we know this isn’t precise because we are using ex-post betas that include the sample period, not ex-ante betas from the beginning.  And yes we know this isn’t precise because we haven’t measured to see if there were other factors driving returns, like growth or quality or even sector exposures.  For our purposes here, we don’t need to be overly precise.   We just need the gist.

[3] Yes, this is just sampling Europe and the US, but these companies are multinational, with exposure to suppliers and consumers globally.

[4] I realize that the preferred term for quant strategies is “systematically active” because they deviate actively from a benchmark.  I am using the term “passive” to indicate that a decision wasn’t made about any of the individual stocks in the momentum baskets.  They were in the baskets, academically, and thus bought or sold without regard to company specifics (other than poor prior share price momentum), and no active human decision was made to include or exclude them from a portfolio.



The views and opinions expressed in this post are those of the post’s author and do not necessarily reflect the views of Albert Bridge Capital, or its affiliates. This post has been provided solely for information purposes and does not constitute an offer or solicitation of an offer or any advice or recommendation to purchase any securities or other financial instruments and may not be construed as such. The author makes no representations as to the accuracy or completeness of any information in this post or found by following any link in this post