How Markets Work
Plus: Norinchukin, US Mortgage Rates, Crypto Financial Crisis
Welcome to another issue of Net Interest, my newsletter on financial sector themes. This week, the subject is markets and why extreme moves are part of their fabric. For More Net Interest – this week on Norinchukin, US mortgage rates and the crypto banking crisis – sign up as a paid subscriber. Paid subscribers also get access to the fully searchable archive of over 100 issues.
Today’s edition of Net Interest is brought to you by Third Bridge.
Third Bridge Forum is the biggest archive of expert interviews in the world. Just last year over 16,000 investment professionals from 1,000 firms across private equity, public equity and credit downloaded over 500,000 interviews. The coverage is extensive – covering both public and private companies, in any sector, across all major geographies. I’ve seen it for myself – the insights Forum delivers are in-depth and unique. If you want to request a free trial visit thirdbridge.com/net.
How Markets Work
The first day of August 2007, I walked into my office and fired up my Bloomberg terminal as normal. I’d been at the hedge fund for a year now, working as part of a small team managing a fund focused exclusively on financial stocks. A few months earlier, the Financial Times had described our office as resembling a university library more than a trading floor, and it was certainly quiet that day – not least because several colleagues were away for the summer. We had just closed the books on an excellent month, “the Fund’s most satisfying since inception”. For the first time, most of our return came from the short side of our portfolio – stocks we were short were down between 10% and 20% amidst deteriorating conditions in the financial sector.
The following week, though, things would get weird. Stocks we were short began going up, and stocks we were long began going down. We came into the month with a net market exposure of just 15% and a bearish tilt, but stocks ceased behaving as we would have anticipated. One of our shorts was Countrywide Financial – a giant mortgage originator that wouldn’t survive the financial crisis; over three trading days at the beginning of August, it rallied by 16%.
We didn’t know it at the time, but we got caught up in what became known as the Quant Meltdown of 2007. On Monday 6 August, some fund somewhere began liquidating. People have since speculated that it may have been a fund operated by a commercial bank needing to raise cash quickly in response to margin calls on losing mortgage or credit related positions, but nobody really knows. The liquidation of that fund impacted market prices, and other funds began to take losses. Stop losses kicked in, deleveraging policies were invoked and losses spread. Renaissance Technologies, a hedge fund firm with one of the best track records in the industry, told investors that a key fund was down 8.7% by the end of the first full week in August. One fund managed by Goldman Sachs was down more than 30%.
The episode wasn’t due to some piece of fundamental newsflow – newsflow would eventually take a turn for the worse, but it would be two more months before the overall market peaked. In fact, market levels barely reflected the turmoil that was being inflicted underneath. Rather, losses stemmed from a liquidity spiral: too many funds heading for the exit at the same time. At the time, it was seen as a highly unusual occurrence. David Viniar, Chief Financial Officer of Goldman Sachs, remarked, “We were seeing things that were 25-standard deviation moves, several days in a row… There have been issues in some of the other quantitative spaces. But nothing like what we saw last week.”
Our fund gave back a lot of the gains we had secured in July but went on to finish the year strongly. Perhaps I should consider myself lucky for having been there to witness a 25-standard deviation move – these things come around only once every 1.3x10^135 years. The thing is, it wasn’t the last such high standard deviation move I would see; it wasn’t even the first.
A few years later, the Flash Crash happened. The Dow Jones Industrial Average plunged 600 points in the space of five minutes as I looked on from my home office. Some stocks such as Accenture dropped to a penny a share; others such as Apple rose to $100,000 a share.
A few years after that, the Swiss Franc jumped 19% against the Euro almost instantaneously. “I think it was something like a 20-plus standard deviation move,” said David Viniar’s successor as Goldman Sachs’ CFO.
And this very week, US Treasury yields experienced a four or five standard deviation move, depending on where you look on the yield curve.
One model of a market is as a simple mechanical equilibrium system. In this model there are two groups of investors: optimists who think the security is going up and pessimists who think it’s going down; the market price reflects an equilibrium at which both groups are happy. In effect, the market price represents a consensus view of all the expectations and information in the market.
In many situations that model works but it falls short in a number of ways. First, it doesn’t explain why there’s so much trading. Saturday Night Live once aired a sketch where the news anchor announces, “And today on the New York Stock Exchange, no shares changed hands. Everyone finally has what they want.” Not only does that never happen, it is seen as a sign of dysfunction when it does. Last Tuesday, Japan’s 10-year government bond didn’t change hands once on the open market, “a fresh sign of the country’s dysfunctional debt market,” reported Bloomberg.
Second, the model doesn’t account for the number of tail events I alone have seen in my professional career. Such events, and the realisation that nobody knows what causes many of them – the Flash Crash was as inexplicable as the Quant Meltdown – suggest that markets possess their own internal dynamics.
A better model of markets, understood by most practitioners and in the ascendancy among commentators, is of markets as an evolving ecosystem. There aren’t just two types of investor – there are multiple types. They include:
Fundamental investors who trade on … fundamentals;
Technical traders who look at past prices and trading volumes to determine strategies;
Liquidity traders who enter and exit the market because they are long or short liquidity; and
Market makers who make money taking the other side of other participants’ trades.
When interactions between these players are taken into account, a complex system emerges, more akin to a biological system than a mechanical one. Adding to the complexity is that each player operates under the guidance of two different functions – a cognitive one and a manipulative one. This is the basis of what George Soros calls reflexivity. Market participants observe the market and transact according to their mandate, whether it’s growth or relative value or some other style. That’s their cognitive function. But they also affect the market through their actions – that’s their manipulative function.
An extreme example is the blow up of hedge fund LTCM in 1998 (supposedly a 10-standard deviation event). Facing strains on its capital, the firm sent a letter to clients asking for more funding. Unfortunately for LTCM, this was like shouting ‘Fire!’ in a crowded auditorium. LTCM’s founder John Meriwether cast it like this:
“The hurricane is not more or less likely to hit because more hurricane insurance has been written. In the financial markets this is not true. The more people write financial insurance, the more likely it is that a disaster will happen, because the people who know you have sold the insurance can make it happen.”
The key insight of this model is that it’s the interactions between players and their environment that drive the behaviour of the market, rather than the actions of the players themselves. These interactions create feedback in the system that can lead to extreme market movements as market participants imitate each other to varying degrees.
In his 2004 book, The (Mis)Behaviour of Markets, mathematician Benoit Mandelbrot lists a number of features of markets that can only be explained by viewing them through this lens.
Markets are risky – extreme price swings are the norm.
Trouble runs in streaks – market turbulence tends to cluster.
Markets have a personality – prices are determined by endogenous effects peculiar to the inner workings of the markets themselves, rather than solely by the exogenous action of outside events.
Markets mislead – bubbles and crashes are inherent to markets; they are the inevitable consequence of the human need to find patterns in the patternless.
However empirical these features are, their underlying cause is frequently disregarded by market participants. In particular, there is a tendency to underestimate the role that liquidity traders play in the market. Perhaps that’s because while everyone is acquainted with the optimists and the pessimists that make up traditional investors, they are typically less familiar with shadowy liquidity traders. When was the last time you sat next to a liquidity trader at a dinner party?
Ironically, back in August 2007, we identified the importance of liquidity as an exogenous factor in our analysis. “The withdrawal of liquidity, though sporadic to date, has become evident across certain asset classes,” we wrote in our monthly letter to investors. We then listed connectedness, lack of transparency and leverage as reasons for remaining bearish on the financial sector. Yet we stopped short of factoring ourselves into the analysis – or more generally, hedge funds like ours that rely on liquidity to risk-manage portfolios. When liquidity recedes, market participants sell what they can rather than what they want, and that becomes a key source of contagion. After a point, liquidity withdrawal is rarely contained to one asset class.
If there’s one recurring theme of extreme market events, it’s the withdrawal of market liquidity. That’s what happened in August 2007 and it’s also one explanation for what underpinned the Flash Crash, when liquidity demand came in faster than supply. Since those episodes, liquidity has been abundant, supported by central bank policy. Yet as it gets withdrawn, various corners of the market are showing signs of strain. What the past has shown is that even if everyone sees it coming, interaction between market participants can lead to weird outcomes. Good luck out there.
So many good books on the broader theme of complexity economics. There’s The Origin of Wealth by Eric Beinhocker; Why Stock Markets Crash by Didier Sornette; Adaptive Markets by Andrew Lo; The End of Theory by Richard Bookstaber and The (Mis)Behaviour of Markets by Benoit Mandelbrot and Richard Hudson.