Last month, the impact of coronavirus caused stock markets to plummet. This led to a huge spike in volatility, with the VIX Index hitting a high of 82.69 on 16 March, benefiting macro strategies (according to the latest Q1 performance figures) as well as statistical arbitrage strategies like New Jersey-based Blueshift Asset Management.
Last month, the impact of coronavirus caused stock markets to plummet. This led to a huge spike in volatility, with the VIX Index hitting a high of 82.69 on 16 March, benefiting macro strategies (according to the latest Q1 performance figures) as well as statistical arbitrage strategies like New Jersey-based Blueshift Asset Management.
Blueshift runs a series of statistical arbitrage market neutral strategies and pays careful attention to the balance of liquidity in the market; the aim being to identify when to provide liquidity to generate alpha.
According to Blueshift’s CEO and CIO, Mani Mahjouri, over recent weeks all the indicators show that liquidity is as bad as it’s ever been. “In early March liquidity was worse than it was in Q4 2018,” says Mahjouri.
“The way we think about it is liquidity is really a commodity. There is a supply/demand for it and right now, demand is at record levels and supply is also at a record low level. So if you have the stomach to step in and provide liquidity, now is the time to do it because you earn outsized premiums.”
Knowing how to spot risk and understand the complex relationship between market volatility and liquidity in times of extreme stress is uniquely challenging but applying a scientific methodology to its trading models has helped Blueshift navigate the first few months of 2020 with minimal turbulence. Understanding the balance of liquidity in recent weeks has very much fallen into Blueshift’s wheelhouse.
Liquidity barometer is poor right now
“It comes down to taking a very granular approach to understanding market microstructure,” says Mahjouri. “We started as a group of 10 or 11 individuals trading an AUM of approximately USD10 million 10 years ago with algorithms that were trading around 1 per cent of all US stocks. Over the years, we’ve built on top of this and for us, understanding the fabric of liquidity is particularly important.”
He adds: “Every year the market gets a little bit more efficient. In that process, pockets of inefficiency get created. The overall barometer of liquidity is that it’s very poor right now, and it is continuing to get worse. This is causing severe moves in the market; perhaps more severe than they need to be. We think about the degree of severity relative to the degree of liquidity in the marketplace. There is a causal relationship. Understanding that is the key to unlocking a lot of alpha.”
Caution remains high across the hedge fund space at present. The risks are very high to getting caught on the wrong side of trades and while liquidity remains patchy, at best, the current trading environment is likely to remain choppy for some time to come.
Mahjouri does not think that stat arb is really substantially less risky right now.
“The market is crowded, liquidity is poor and there’s a realisation that anyone can put a vector into an optimiser and come out with a result that says risks are minimal,” he says. “What matters today is, ‘What are your risks in a stressed scenario?’ Typically, what ends up happening is there’s a correlation between risk and directionality in the market because so much of the hedge fund community is invested in beta. There’s an understandable tendency to reduce risk in such situations.”
Quant funds have risen to prominence in recent years. This should be welcomed because the way they approach risk is in complete contrast to discretionary managers. A discretionary manager might provide one key reason for buying a stock; by contrast, a quant manager might offer one out of a hundred possible reasons.
The point is, there are two ways you can slice the pie: one is to have a high conviction and run a concentrated portfolio, which absorbs risk at one end of the spectrum; and the other is to have a lower level conviction and run a diversified portfolio.
“The signals that are a discretionary manager builds require higher conviction because they can only make a limited number of bets. We, on the other hand, approach risk from a different level to try to get a different payout. It’s a different aspect of the risk spectrum that we look to get exposure to,” states Mahjouri.
Prior to establishing Blueshift in 2017, Mahjouri was the Chief Investment Officer at Tradeworx and before that, he was the 12th person to join Cliff Asness at AQR. Over the course of his trading career, Mahjouri has been through two significant bear markets. The last time market volatility was so high – following the global financial crash – he met his wife. Could another positive omen be around the corner? Time well tell.
One thing that is certain: the team of PhDs, many of whom are astrophysicists, that Mahjouri has assembled will be heavily debating market microstructure right now, and where the opportunities might lie as markets continue to whipsaw.
“It’s good to challenge each other’s views,” he says. “Too often, egos get in the way, especially in smaller funds. One of the nice things about taking a scientific approach is that there’s a clear objective and end result and part of our recruiting process is to find people who are willing to say you’re wrong.”
Faster signal strategy
Blueshift takes its name from astrophysics; a reference to the Doppler Effect, which describes an object in space moving closer towards another object. If radio waves are shifted into the ultraviolet part of the electromagnetic spectrum, they are said to be ‘blueshifted’, or shifted toward higher frequencies.
“We have a lot of astrophysicists in the team so we wanted something that reflected our background. We have our roots in HFT and we still operate high turnover strategies. To the extent that a typical institutional portfolio is a slow signal, adding us is like introducing a blueshift into the portfolio,” explains Mahjouri.
Applying a scientific method to trading means that like other systematic strategies, Blueshift takes an unbiased view, using whatever statistical techniques are relevant to the task at hand. Although not opposed to machine learning, equally the team is not looking to offload portfolio management to the machine.
At Blueshift, data is its lifeblood
“We look through a huge amount of it,” comments Mahjouri. “The thing about data is there’s a novelty factor to it so if nobody has it there is a value component but there’s also a descriptive component to it; some data is useful in certain scenarios and regimes and not in others. There’s no ‘one size fits all’.
“The key to success (to using data) in today’s markets is understanding that complex relationship. At Blueshift, it’s about our ability to comprehend information and extract information from our surroundings; both of those things are in exponential growth mode thanks to technology.”
