Ten years ago today, in the months leading to the start of the global recession, the ‘Quant Quake’ of August 10, 2007 shook Wall Street. The Quant Quake saw some of the top quant funds in the US, including those belonging to AQR and Goldman Sachs, lose billions of dollars in a matter of weeks. Many of the funds at the time were highly correlated, and as one firm began selling its equity positions, that created a seemingly unstoppable domino effect.
At the time, few understood what caused these algorithms to falter. However, looking back, it’s clear that some of the inherent risks that caused billions in losses were overlooked.
While the global recession and mortgage market meltdown weren’t “expected,” the idea that drastic downturns in other parts of the market — such as the mortgage or credit markets — would have no impact on the stock market and the performance of quants in hindsight look pretty naïve. And with the success of quant strategies, one should have, or at least could have, expected a wide array of copy cat managers would come along trying to cash in.
But what’s changed since the Quant Crash of 2007? Have we found ways to protect the market from unexpected and extreme downturns driven, at least in part, by quants? Have funds or fund managers found ways to better understand their algorithms?
It doesn’t seem like it. Only three years after Quant Quake, another notable incident of huge market fluctuations driven by algorithms occurred: The Flash Crash of 2010, in which trillions of dollars were lost from the US equity markets, only to be mostly recovered, all within about 36 minutes.
Looking back over the past ten years, it’s clear that not only have we not solved the problems that created Quant Quake, but the financial markets are more vulnerable than ever to another fast, unexpected, quant-driven fluctuation. Here’s why.
The rise of automated trading in hedge funds
Since 2007, the usage of automated trading strategies has only increased with more than 75% of the US stock market now traded via algorithms. Quantitative strategies have become seen as the active manager’s way of generating higher returns to outperform the growing amount of capital moving into passive investing strategies. This has been seen in the growth of the number of quant funds over the years (from ~10% of the market in 2007 to ~17% in June 2017), as even more fundamental hedge funds are turning to quants or platforms that enable non-technical PMs to run their strategies or ideas through algorithms.
The likelihood of another sudden market crash led by automated trading is almost inevitable. As these markets become even more interconnected and more and more traders or funds are using automated trading strategies, the effects of such a crash will not be contained just within a small community of hedge fund managers as it was during the Quant Quake. Instead, it will be felt by all investors and funds, and throughout the broader market.
The shift to automated trading isn’t just happening within the hedge fund community. In March 2017, Blackrock announced a move to consolidate many of its actively managed mutual funds and rely more on stock picking algorithms, triggering other traditional asset managers to follow suit. The has created a more populous network that relies on quantitative strategies for returns.
Copycats continue, bringing amateurs along
If in 2007, part of the issue was that too many of the trading strategies were too similar, has this risk decreased at all? To the contrary, there are more copy cats today than there were in 2007. Furthermore, there are more investors that are not as quantitatively savvy using various platforms or solutions that allow them to enact quantitative strategies as if they were. This also means that there are more investors whose primary focus, or at least significant focus, may be other asset classes, inherently linking quantitative investing to the rest of the financial market.
Regulators come up empty
It took the SEC four months to issue its report on the cause of the Flash Crash in 2010, and it is still generally unknown which fund was the initial trigger for the Quant Quake in 2007. This issue of the ‘black box’ for quant algorithms has been, at least for now, accepted. So, when a large downturn occurs, these quant strategies begin losing money, and investors want to understand what went wrong, will we still be okay with not knowing what is driving the decisions of our algorithms?
Regulators have made some attempts to protect against large-scale market fluctuations which are exaggerated by these algorithms. However, initiatives such as circuit breakers are just superficial solutions. We saw this two years ago when markets went into free fall and circuit breakers were tripped more than 1,300 times and didn’t help share prices of major companies from dropping tens of percentage points in just minutes.
While regulation and more market oversight could help to lessen the risk of a bigger, more widespread Quant Quake, a key issue is transparency. As markets become more complex and interconnected, we need to make sure to understand how they are connected and how actions in one market may or may not impact another.
The equity market risks are now amplified
While easier said than done, some of the features that have defined the equity market since the Great Recession have only amplified flash crash risks. For most of the last decade, share buybacks have been the main source of capital inflow into the US equity market. On top of that, the quiet IPO market alongside an active M&A market have also led to a decrease in the number of publicly traded companies. The combination of these trends means that there are fewer shares outstanding in the market, which in turn means that even smaller buy and sell orders can lead to exaggerated swings.
To me, this market set up sounds familiar, and doesn’t sound as though we are far off from another Quant Quake if a significant downturn in the market occurs.
Why haven’t we learned? Are events like the Quant Quake now just inevitable? Or are we just not taking the necessary steps to mitigate the risks?
The obvious, but uncool, answer: regulation
As someone whose first real awakening to economics and finance was the collapse of Lehman Brothers, there are takeaways from the causes of the Great Recessions that can, and should, be useful in mitigating the potential impact of another Quant Quake.
Much of Wall Street lives under the assumption that less is more in terms of oversight. Ten years ago, the industry’s attitude was that it didn’t matter if anyone understood the quality of the mortgages packaged into CDOs. Now, it’s as if it no one cares that few understand how algorithms make trading decisions. As our markets and the assets in it become more complex and interconnected, there are fewer and fewer people involved, even those in the weeds making trades on a daily basis, that really understand.
An analogy from structural engineering world comes to mind. If a large crowd of people on a stadium balcony begins to jump in synch, no matter how strong the balcony is, at some point it will begin to bend and recoil. When this up and down motion inevitably becomes too extreme for the material with which the balcony was made to handle, it may collapse and fail.
The past decade’s market volatility is a similarly exaggerated movement. With the growing number of funds crowding into quantitative strategies, without the right oversight, it’s not too far off to believe that sooner or later, the structure that is the US equity market may too see swings that are too extreme, bringing it to the point of collapse. This is not to say this scenario is necessarily destiny, however, action is needed to mitigate these risks, decrease the likelihood of such a downturn, and further support the market.
Jillian Williams is an Investment Associate at Anthemis.
What’s Changed in the 10 Years Since Quant Quake? Not Much. was originally published in #hackingfinance on Medium, where people are continuing the conversation by highlighting and responding to this story.Like This Post