Montrose Journal Winter 06
SECOND-GUESSING THE MARKETS: DO WE UNDERSTAND RISK ANY BETTER THAN WE DID? -DAVID BLAKE, FINANCIAL ANALYST AND COMMENTATOR
There really was a Man Who Broke The Bank at Monte Carlo. Charles Bell did it 12 times in one session. He died penniless. We know about him because François Blanc, the owner of the Casino, made a fuss about it, draping the roulette table with a black shroud while new money was brought in to replenish the croupier's stock of chips, the "bank" of the song. He died rich but nobody wrote a song about him. In finance as in gambling the attention goes to those who make the big coup or suffer the big loss, but the smart money goes for a small but regular percentage of the gross.
What is actually happening in a Casino? The player who bets on 9 is taking the risk of losing some money and the owner is taking the risk of losing 35 times as much. (M. Blanc's great technological breakthrough was to put a zero on the wheel, so there is only a 1 in 356 chance of being forced to pay out. In financial markets, institutions try to give themselves the same kind of edge by charging more for what they sell you than that they buy.)
But unlike the casino, some players do seem able not just to break the bank on a few occasions but to do better than average over long periods of time.Warren Buffet is 76 and he still does so by buying firms he thinks are worth more than the market price. The Harvard and Yale endowment funds get average returns well into double digits the past 20 years. And the question of why and how this happens poses a huge problem for the conventional theory of markets, the Efficient Market Theory.
Essentially, this says that the roulette wheel is not rigged. There are so many players out there in the financial markets that collectively they manage to make sure that all the assets are priced to take account of all the publicly available information in the world. If something looks cheap, there's a reason. So even if the lunch seems free, you'll probably end up having to go without your dinner to pay for it.
Some people might be able to cheat by using inside information, but they usually get caught when their success draws attention. Some people might have a run of good luck; but over time they get caught out. Note that this does not mean that all prices today will be the same tomorrow. The small country with only a few scattered farmers living off international aid might strike oil and its currency will move up.
But it does mean that if Russian dollar bonds yield 12% and the US government bonds give you only 6%, the market knows there is a good reason for that. And that reason is Risk.
Russia defaulted a few years ago, so who knows when they might do it again? Brazil devalued by over 70% between 1998 and 2002, so maybe earning 15%a year on a Real bank account in Rio isn't such a good idea after all. The pharmaceutical company whose shares seem so inviting might rely on a drug which is just about to be exposed as having dreadful side-effects. Sometimes there's no obvious explanation of what makes an asset "risky", but the fact that people have to be paid extra to hold it shows that there must be a risk there which we haven't thought of.
Average it out over time and we'll see that the extra risk takes its toll on your returns, dragging them down (to/by) the market average. The standard tool of investment analysis, the Capital Asset Pricing Model (CAPM) starts out by looking at how much money you can get risk free from government bonds and then asks whether the extra risk of other things is worth taking. The theory got the authors a Nobel Prize in economics.
It's only a small step from saying that the market embodies all the available information to saying that your portfolio therefore should be the same as the market. In this view of the world, Risk isn't some abstract danger out there which we have to avoid at all costs. It's a tradable good which people buy and sell depending on their circumstances. And that's where one of the problems of the whole theory comes in.
Suppose you are Finance Director of a large firm with a lot of sales in India.What you most definitely do NOT want is to have to include a note in your annual report saying that your firm lost $ 400m because the Rupee got devalued before you had brought the money home. So you go into financial markets and buy protection, getting rid of the risk of an embarrassing loss by buying options to sell Rupees at a fixed exchange rate. Over a 10 year period the firm may pay out far more buying this protection than it saves when the devaluation finally occurs. The people who sold it to you make money, just as insurance companies make money regularly over hundreds of years selling fire insurance. Yet if the markets were really efficient this couldn't happen.
Foreign exchange markets, the most liquid and therefore potentially the most efficient in the world are one classic example of where it's possible to identify strategies that consistently make money, even after accounting for the mishaps along the way, by holding currencies with high interest rates which suffer occasional devaluations.
Yet most people focus not on that but on the very rare cases where someone like George Soros makes a killing. It's understandable. No one likes losing money. Funds which go up 15% and then go down 10% leave their clients 5% better off but they don't get thanked for it. Instead, they get pilloried for taking risky decisions which lost money. That's irrational, but people are irrational.
There are lots of other examples where strategies have worked over very long periods of time which appear to contravene Efficient Market Theory. Now it's been my long experience with economists that when something happens which contradicts their theory, they don't change the theory. They start trying to prove that what you say is happening isn't really happening at all. So those people who claim to have found examples where the EMT doesn't work because even allowing for risk it's possible to do better than average are told that they have just got the numbers wrong. That is, somehow or other, they have mispriced Risk and underestimated how big the danger is in making an investment. Money invested in shares does much better over time than money in bonds? You must be wrong about that. All that is happening is the market, in its infinite wisdom, senses that equities are so risky that it demands a better return in anticipation of the day the market crashes. And even if buying equities has made more money than bonds right through the stock market crashes as well as the booms, that just shows the next crash might be even bigger.
Now there are some very obvious poster children for the theory that these investments which seem too good to be true aren't true. Think of three big crashes we've have had in the past 10 years, Nick Leeson at Barings ($800m), LTCM ($4bn) and Amaranth ($8bn.) All of them started out with someone who thought they had found a way to make steady money by trading things where they thought the market priced assets wrongly. All of them ended up taking enormous positions because that was the way to make enormous amounts of money. And all of them went bust. But these examples have other things in common which made them particular. At sensibly run firms, however sure you are that the market is wrong you don't bet the firm. You remember Keynes' old rule that "Markets can stay irrational longer than you can stay solvent."
Even more crucial, you make sure that you don't end up being so big in the market that everyone knows that if prices go down you will get wiped out. A good rule of thumb is that if the only reason prices go up is that you are buying, you might find it uncomfortable if anyone starts trying to force prices down. That's what seems to have happened to Amaranth, which got so enthusiastic about buying natural gas contracts that it was enormously exposed to other traders who were able to drive prices down by selling.
That's an area where we can try to make well calculated decisions. But a lot of time, probably too much time, is spent worrying about a quite different kind of risk, the event which is highly unlikely to happen but would have enormous impact if it did.
For example, some time next year the scientists at CERN in Geneva will switch on their latest and biggest toy. It's a 17 mile long Large Hadron Collider which does exactly what it says on the label. It's large and it sends Hadrons (sub-atomic particles) round in circles and then makes them collide with each other.When they do, we'll find out the answer to some basic questions about the nature of matter. If we're still here.
Because one of the things that the LHC will do is probably make minute Black Holes. And as anyone with a GCSE in Star Trek can tell you, Black Holes have the ability to suck in all the matter around them, gobbling up complete stars in an instant so that they vanish. So a small planet like Earth would be little more than finger food.
The scientists assure us we shouldn't worry. There is, according to a physicist at Brown University only a "minuscule risk" that the LHC will actually end up destroying the planet. And if we want further reassurance Oxford's Future of Humanity Institute has collaborated with MIT to put a number on it. There is, they say at worst only a 1,000,000,000,000 to 1 chance that the guys from Geneva will wipe us all out in any given year from their experiment. That is a risk which they describe as "reassuringly small."
Or to take another example: in Berlin, the Deutsche Oper recently called off a production of Idomeneo because there was "an incalculable risk" that it might provoke a terrorist incident. Incalculable maybe, but presumably greater than zero.
In the US, the Administration pursues what is known as the "One Per cent Doctrine." If there is a One Per Cent chance that a threat is real the USA has to respond as if it is 100%certain that it is true.
These three examples come from outside the world of finance but they illustrate a problem which financial markets face every day. How do you cope with the risk of a very unlikely event which will wipe you out if it happened. It's the financial equivalent of a Black Hole, devouring all your wealth or possibly in the extreme devouring the whole financial system.
Traditional risk control systems look at the past and then say, for example, that 99%of the time the assets you have bought will keep at least 98%of their value. It's a very strong bet that you won't lose more than 2%. But what about that 1%of the time? And what about the 0.0000001%of the time that a huge natural disaster sweeps everything away? Interest in this issue has grown a lot since 9/11/01 because that sensitised us to the idea of disaster coming out of a clear blue sky (literally.)
These very rare events are called Black Swans and no one really knows how to handle them. By their nature we have no warning they are going to happen. It feels worrying to feel that we are not protected against them. Yet if you start protecting yourself against everything the universe could throw against you, you'll end up with no assets to protect. For most people most of the time, the best thing to do is to sell Black Swan insurance and hope that no one will be around to collect if it all goes wrong.
One final point to suggest that in spite of all the risks, it is possible to beat the odds. Charles Wells may have been the most famous man to break the bank at Monte Carlo but he wasn't the first. Joseph Jagger did it in 1873, but he did it the hard way, paying people to watch all 6 roulette wheels until they found one where a slight imperfection meant 9 numbers came up more often than the others. He made $5m in today's money, walked out with it when the casino changed the wheel, quit his job at the cotton mill in Yorkshire and lived a life of luxury.