The Hour Between Dog and Wolf

by John Coates

  You have now put on a spread trade. And look at the finances of your small hedge fund. You bought 100 apples at 40¢, so you pay out $40; but you also sold 100 oranges for 60¢, so receive $60, and you can use that to pay for the apples. The arbitrage position has been put on with no capital required (although in the financial markets you are required to post a small amount of margin or collateral), meaning that you could in principle amass enormous positions and eventually dominate your local fruit market. This simple game of leverage permitted the hedge fund Long Term Capital to leverage their capital into such large positions that when they went wrong they wiped out the company’s capital and threatened to destroy the entire financial system.

  Your trade is tiny in comparison to those of even the smallest hedge fund. Nonetheless, over the next few days the fruit market goes on a rollercoaster ride: rumours one day of a good harvest cause prices to drop 25 per cent; rumours the next day of a strike among migrant fruit pickers cause them to spike 50 per cent. No matter – apples and oranges go up and down together, so what you lose on one leg, as it is called, of your hedged position you make on the other. At one point, however, the price spread did exceed 20¢, widening to 25¢, with apples going down in price and oranges going up, and this gave you the jitters. There is no scarier predicament in the financial markets than a spread trade in which the two securities become uncorrelated and start moving in opposite directions, something that unfortunately tends to happen during crises. Hedged positions are billed as market neutral, but the term is something of a misnomer, because they carry substantial risks.

  But this week in the fruit market the correlations hold strong, and after a few days the old price relationship of a 10¢ difference reasserts itself, with apples at 30¢ and oranges at 40¢. Now you decide to unwind your trade. You sell the 100 apples you bought at 40¢ at the new price of 30¢, losing 10¢, for a total loss of $10; you buy the 100 oranges you sold at 60¢ at the new price of 40¢, making 20¢, for a total profit of $20; you give back the 100 oranges you borrowed and pay the stall owner $5. Summing up, you have made $5 on your spread trade, for little risk (or so the thinking goes) and minimal use of capital.

  If you understand this simple example of shorting and hedging and spread trades, then you have in your hands the basic principles of high finance. Spread trades are the staple trading strategy of both banks and hedge funds. Within the banks it is not just the arb desk that employs them, for flow traders do too, on the Treasury, mortgage, corporate bond desks. Flow traders facilitate customer trades, but they often use these customer flows as a means of building up large spread trades. For example, if Martin had believed that the ten-year Treasuries he sold to DuPont were expensive relative to, say, two-year Treasuries or bond futures, he could have bought either of these securities instead and established a large spread trade. Flow trading desks were originally intended mainly to serve clients, but over time the arbitrage positions they held grew exponentially in size and P&L, until during the housing bubble years they dwarfed the client trades.

  To make the jump from the fruit market into high finance, however, you have to stretch your imagination to consider spread trades between, say, Treasury bonds and mortgage-backed bonds, or between Treasuries and stocks, or between German and Greek bonds, or gold and silver, even the amount of rainfall in California relative to Kansas. In addition, you have to consider numbers that verge on the incredible – aggregate positions running into the trillions and profits into the hundred millions, even billions. Your small spread trade made $5, but if you were to scale up your position to a size comparable to the big hedge funds, and bought, say, $4 billion-worth of apples, then your profit would have amounted to $500 million, a decent week in the fruit market.

  * * *

  And this is precisely what Scott has been doing this year, setting up spread trades, mostly between the stock and bond markets. He believes strongly that stocks are undervalued, so he has been buying them, and that bonds, especially corporates, are overvalued, so he has been shorting them. He has met with unprecedented success at this form of arbitrage, and can boast a P&L year to date of almost $17 million, an exceptionally strong showing given that it is early in the year. Scott does the mental arithmetic, calculating his year-end P&L if he keeps making money at this rate – maybe $40 million – and warms to the idea of a $5 million bonus at year end.

  But Scott’s success has led him to throw caution, the most valuable trait in an arb trader, to the wind. He is now convinced that all his spread trades are in essence nothing more than a cumbersome way of buying the stock market. He is one of the many traders who has taken the Fed’s message as a green light for a market rally, so recently, in addition to the spread trades he maintains as his core strategy, he has spent more and more time doing nothing more sophisticated than buying stocks and waiting for them to go up. So far he has displayed an almost occult ability in calling market direction, so his boss Stefan has let him continue.

  Specifically, Scott bought futures contracts on the S&P 500 index, contracts called e-minis. His initial position, put on after the Fed announcement, amounted to 2,000 contracts, meaning that for a 1 per cent move in the stock market he makes or loses about $1.3 million, a reasonable-sized trade, but nothing too dangerous. Scott’s strategy has worked well: the stock market has rallied almost 100 points, making him $10 million.

  Cooler minds would conclude that the prudent thing to do now is unwind his core position, take some profit off the table, or at the very least scale down its size. But that does not seem to happen during bull markets. Scott is certainly not thinking that way. Far from it. He can barely contain his excitement. For Scott has seen hove into view a rare trade, the big one, his great white whale. Those who deal in the lore of the market refer to this opportunity as ‘the retirement trade’, the one that will make so much money they will never have to work again. The retirement trade, it should be noted, is something of a myth; not because such opportunities do not exist – they do – but because very few bankers, despite their professed dream of spending the rest of their lives on a golf course, actually want to leave this game. They would miss the excitement, the feeling that they are at the centre of the world, which in a sense they are. Most would find it difficult to replicate the dopamine rush they get from trading.

  Stocks may have rallied a quick 15 per cent this spring, taking their dividend yield down below 2 per cent, but, Scott argues, just look at the macro-economic backdrop! Asia has only just started to grow, as have Africa, the Middle East, Latin America and Eastern Europe (the Russian default of 1998 quickly fading from memory). At no time in history have so many geographic regions been brought within the ambit of the world markets. This is George W. Bush’s New World Order, and in such a world old valuation tools cannot be applied. Scott believes the market is on the cusp of being revalued.

  Something similar happened back in the early 1980s. At that time, the price–earnings ratio of the S&P index amounted to little more than 9 – a low number compared to its long-term average of 15 – largely because many investors had lived through the Crash of 1929 and the Great Depression, and retained a deep-seated fear of stocks. But as this generation aged and lost their clout in the market, dark memories of financial crisis that had haunted the economy for fifty years began to fade. After the 1970s fewer and fewer people remembered the hard times of the thirties, and took to spending and investing with the same blithe spirit as their grandparents in the 1920s. Stocks proceeded to rally for the next 25 years, taking their price–earning ratios up to a peak of 44 in 1999. The aftermath of the Tech Crash of 2001 has, Scott believes, presented a similar opportunity. Doubts and fears linger among survivors of that crash. But not in Scott. He decides against reducing his core position, and instead builds on it.

  THE WINNER EFFECT

  The euphoria, overconfidence and heightened appetite for risk that grip traders during a bull market may result from a phenomenon known in biology as ‘the winner effect’. I first heard of this m
odel during the dot.com years, while listening to a lecture at Rockefeller, and I thought, as I have explained previously, it was the most compelling model of irrational exuberance I had encountered. It was this model that led me back to research, to see if the winner effect exists in the markets.

  How does the winner effect work among animals? Does it exist in humans? Can it explain the heightened risk-taking and manic behaviour I had witnessed on Wall Street during the dot.com bubble?

  Biologists studying animals in the field had noticed that an animal winning a fight or a competition for turf was more likely to win its next fight. This phenomenon had been observed in a large number of species. Such a finding raised the possibility that the mere act of winning contributes to further wins. But before biologists could draw such a conclusion they had to consider a number of alternative explanations. For example, maybe an animal keeps winning simply because it is physically larger than its rivals. To rule out possibilities such as this, biologists constructed controlled experiments in which they pitted animals that were equally matched in size, or rather that were equally matched in what is called ‘resource holding potential’, in other words the total physical resources – muscular, metabolic, cardiovascular – an animal can draw on in an all-out fight. They also controlled for motivation, because a small, hungry animal eating a carcass can successfully chase off a larger, well-fed animal. Yet even when animals were evenly matched for size (or resources) and motivation, a pure winner effect nonetheless emerged.

  Research on the winner effect began with this statistical finding, but it lacked an explanation. How can winning contribute to further wins? Some scientists argued that a victory imparts information to an animal about its own resources and abilities relative to those of rival animals, and this knowledge permits it to choose fights it can win. Others thought that winning may leave physical traces, such as pheromones and other chemicals, on an animal which broadcast its recent victory, and these can deter subsequent opponents from escalating an encounter. But perhaps the most persuasive account is one that highlights the role of testosterone in these competitions.

  When two male animals face off they experience a pronounced rise in their testosterone levels. Testosterone’s allotted role in male bodies is to prepare them for precisely these sorts of confrontations. Hence the anabolic effects on muscle mass and haemoglobin. Testosterone has also been found to quicken reactions, to sharpen a type of visual skill known as visuo-motor scanning, and to enhance another visual ability known as camouflage breaking. Just as important as the physical preparation is the hormone’s tendency to increase an animal’s persistence and fearlessness. After all, there would be little point in equipping an animal with a greatly augmented fighting capacity if it were not willing to use it.

  Testosterone thus prepares an animal for a competition, but it is what happens next that drives the winner effect. After the fight is over the winning animal emerges with even higher levels of testosterone, the loser with lower levels. These hormonal signals, it has been argued, make sense: if you have just lost a fight, you had better retire into the bush and nurse your wounds; while if you have won you can expect an increased number of challenges to your newly elevated status in the social hierarchy. Among animals these effects can be large. In one study it was reported that in a competition for rank among recently introduced rhesus monkeys, the winner emerged with a tenfold increase in testosterone, while the loser experienced a drop to one tenth of base levels, and these new levels for both winner and loser persisted for several weeks. So powerful is this effect that in some species dominant males will intervene in a fight, apparently protecting the weaker male, but with the intention of depriving the likely winner, and potential future rival, of the benefits of the winner effect.

  In many species, male animals display a dazzling ornamentation – bright colours, fancy wattles and combs, lush nuptial plumage – and ostentatious behaviour, strutting and threatening. Some lizards, for example, when preparing for a fight, bob their heads and pump themselves up and down on their forelegs, much like a triumphant athlete punching the air. But the effects of winning and losing can alter these animals’ physical appearance. In defeat they can go into rapid decline, their bravado dimmed, some even see their colours wash away, their testicles and brains shrink, and they lapse into a lethargic, even a depressed, state.

  Life for the winner is more glorious. It enters the next round of competition with already elevated testosterone levels, and this androgenic priming gives it an edge that increases its chances of winning yet again. Through this process an animal can be drawn into a positive-feedback loop, in which victory leads to raised testosterone levels which in turn leads to further victory.

  Does the winner effect occur in humans? The question is controversial. Many social scientists have denied the existence of winning streaks or what are called ‘hot hands’ in sports, claiming that athletes and fans who believe otherwise are subject to an illusion. But I think the winner effect does exist among humans. First, let us look at the stats. My colleague Lionel Page and I have done just that. To do so we followed as closely as possible the protocol used by biologists when studying animals in the wild. Lionel searched for athletes who were equally matched for physical resources and motivation, and then tested to see if winning on its own contributed to further wins. He came up with what I think is the most rigorous test of the winner effect in sport. Lionel managed to find a database of 623,000 professional tennis matches; from this sample he considered only those between players who were one seed apart; then within this subset he further narrowed his search to matches that went to a tie break in the first set; and even further, to tie breaks decided by the narrowest possible margin, two points. In short, he looked at tennis players who were as closely matched as possible in rank and in playing form on the day of the match. Studying only these close matches, he found that the winner of the first set had a 60 per cent chance of winning the second one, and, since the matches considered were the best of three sets, the match. In sport, winning contributes to further wins.

  Does a testosterone feedback loop power the winner effect in humans? The pre-competition surge in testosterone has been documented in a number of sports such as tennis, wrestling and ice hockey, as well as in less physical competitions, such as chess, even medical exams. Winning athletes experience a post-game spike in testosterone, suggesting that a positive-feedback loop is indeed the physiological substrate to winning and losing streaks. Incidentally, these testosterone-driven sporting victories appear to be more common when an athlete is on home turf, the so-called home advantage. Athletes on a winning streak may thus have a very different body chemistry than those on a losing streak. In all these experiments, with both animals and humans, the winners experienced a self-reinforcing upward spiral of testosterone.

  It is not surprising, therefore, that sports scientists spend a great deal of time figuring out how to raise testosterone levels in their athletes – legally, of course. We are in the curious situation of harbouring within our bodies chemicals that can unlock our full potential – kindling our competitive spirit, focusing our attention, granting access to our metabolic resources, raising us to a state of flow – yet we cannot readily access them. How frustrating! We hold the keys to victory within us, but usually we cannot find them. We would love to self-administer these drugs, but it cannot be done by a mere act of will; you cannot simply say, ‘I want lots of testosterone now!’ or ‘I want to play my best, now!’ It just does not work. Instead we must engage in all sorts of occult rituals and physical exercises before our bodies will even consider our request for more power. The situation is comparable to having money in the bank but being unable to touch it, unless, say, you perform some ceremonial dance, at which point the gates protecting your fortune swing open.

  What rites need to be performed before our bodies will grant us access to our own physiological wealth? Sports scientists probably know more than anyone about what needs doing. Their techniques include alte
ring the balance between aerobic to anaerobic exercises, between fun and gruelling sessions, their timing and length, weights lifted – in general the greater the weight, the shorter the sessions, the larger the anabolic result – diet, amount of sleep, and so on, until their athletes achieve just the right levels of testosterone. A previous victory, as we have seen, is also a potent way of unlocking our testosterone. In fact, it has been found among ice hockey players that replaying a video of a previous win can increase their testosterone levels and thus their chances of winning the upcoming game.

  Another influence on testosterone levels is a rivalry: Ayrton Senna and Alain Prost in Formula One racing; Muhammad Ali and Joe Frazier in boxing; John McEnroe and Björn Borg in tennis. Often felt as a thorn in an athlete’s side, the rival may bring out the best in him. After Borg retired from tennis, McEnroe confessed that there was ‘this void, and I always felt it was up to me in a sense to manufacture my own intensity thereafter’. That is something many athletes do before a game: manufacture intensity, get the juices flowing. Often, without understanding the physiology involved, they try to artificially provoke the effects of challenge, even victory, before a game starts. American football players, for example, pound locker doors, and boxers strut when entering the ring and stare down their opponents. Even people in business, before and during negotiations, can prepare themselves by striking what are called ‘power poses’ – feet widely placed, chest out, arms crossed; or feet up on desks, hands behind head: basically poses that take up more room – and these too can raise testosterone levels. Soldiers often perform similar rituals before battle, to get the hormones flowing and to invoke Ares, god of war. In this process, music can play an important role. Napoleon complained that the barbaric music of the Cossacks drove them to such a fury that they were able to wipe out the cream of his army; and General Nikolai Linevich later said that for the Russian army, music was a ‘divine dynamite’.