What are the Odds?

Not really sure how to write this… best to start at the beginning and see where it goes. Expect bad writing ahead. In fact give up now!

Had an interesting conversation on Friday on the  subject of dabbling in the stock market. It bought back all sorts of memories of conversation that i used to have during the tech bubble, in my silicon valley days. The general gist of this conversation was that venturing a small (essentially disposable) amount of money the stock market, and making small trades to profit on the intra-day movements of shares was a easy way to make money. The thesis being that it would be possible to expend a period of time learning enough about how the markets work, and profitably use that experience making small trades.

This bothered me… but i struggled (on the right side of several pints of Old Speckled Hen) to articulate why…

The obvious analogy is something like online poker. Surely if its possible to make money playing online poker it must be possible to do the same with stocks. It’s merely a matter of amassing enough information about a specific segment of the market to calculate odds as to whether a particular stock will rise or fall.

On the face of it, this doesn’t seem unreasonable (ignoring the work of people like Stiglitz on information asymmetry). After all, if the markets are reacting to events, and you are aware enough of events to make predictions of the actions of other entities in the market, you should be able to make predictions about individual stocks.

Certainly if you had adopted such a strategy over the long term, for many of the last, say, fifty years you would have profited nicely. Except that those fifty years have been punctuated by massive dislocations that could well have wiped out all your prior gains. This is the story of the turkey in Taleb’s The Black Swan, everyday it gets fed, and if it bases it’s expectation of being fed on it’s prior experience, it’s odds of getting fed improve everyday. This relationship holds through out the year… but the week before Christmas brings an event that is so unlikely that it has become not even worth considering.

Investing for the long term would appear to require less detailed information (but perhaps deeper understanding?), and the introduction of an element of time allows the use of strategies to spread and limit risk, diversifiction, stop-loss, periodically realising gains, etc. There are arguments against things like modelling risk and diversification, but for the investor with the guts to pull the plug and get out, the damage of extraordinary events can be limited. I think.

In the short term (day trading) things seem less clear cut. In essence you are relying on your ability to predict small movements in a market with so many players that it may as well be chaotic. If you aren’t dealing with the kinds of sums of money that can ‘move the market’ you are in essence gambling. There are no doubt situations where the degree to which it is possible to identify winners and losers is so obvious that it’s possible to profit – taking short positions in the banks earlier this year was probably one of them. Their situation was so dire, and the institutional holding that had to be liquidated so huge, that it was very hard to lose in a short bet. And yet, here we are three months later with the market up 30% off it’s lows, one of the biggest rallys in history. How you like to have been caught short when this started? How many were wiped out by margin calls on the way up, and how many will take huge loses on the way back down?

With the number of entities in the market it doesn’t seem likely to me that it’s possible to make any short term predictions with any kind of certainty. What looks like good news to one entity is, bad news to another. Events unrelated to any information available in the market, the machinations of entities with enough capital to move the market  (let alone individual stocks), etc. are completely unpredictable. All attempts to model the likelihood of such events can only really influence decisions that involve an element of time.

Which brings me back around to online poker. The odds of winning at poker, blackjack, etc are knowable. Given enough experience it’s possible to develop a sense of when certain risks are worth taking, and when they aren’t. This contrasts with the situation in the stock market which requires incredibly complex systems of hedging to minimise the risk, while remaining an unknowable problem.

The only way to limit risk (outside of adopting complex hedging strategies, not really accessible to the individual trader) is to play with money that you can completely afford to lose. If you’re taking small positions, and turning them over quickly, your profits are going to be eaten by fees and taxes. 

At this particular time, with the financial system in a state of disarray, and reality continually obfuscated and obscured, it seems a risky business and essentially random game of chance. If i had the time in inclination to spend time in front of the computer gambling, i think i’d stick to poker.

[As i expected this attempt at articulating my current thoughts on investing is a mess. Sorry. Maybe i’ll try and write something more structured another day.]



30 year bondBeen watching this graph recently – it has been behaving badly…

That’s the yield on the US Treasury 30 year Bond. You can see the yield being forced down 2%, to 2.5%, from November to the end of last year. Then it creeps back up to 3.5% over a period of several months, where it stabilises until mid-April. At which point it starts to slowly, relentlessly creep back up. All of which raises some interesting questions, like ‘what’s happening?!’, and ‘what does it all mean?!’.

That graph shows the interest rate (yield) that the government has to pay on it’s debt. The more worried that ‘the market’ is about the amount of money that the Treasury wants to sell, the higher this yield moves.

Which would be simple, but of course, things are never allowed to be simple. In order to have some control over the rate, the Federal Reserve (central bank) can print up some new money, and buy the debt from the Treasury.

Unfortunately this ruse doesn’t appear to be working, as you can see – the rate is moving up. How can this be? Surely the Federal Reserve is so powerful, in theory being able to print an unlimited amount of money, there should be no limit to the scope of it’s action.

The problem for the Fed is that the more money that they print, the more the damage the value of the currency. The more they damage the value of the currency, the higher they would be forced to push interest rates to protect it. Not doing so risks crashing the currency, which when we are talking about the dollar is a big deal. The world holds trillions of dollars in reserves, and would not take kindly to having it rapidly devalued.

Another option is to squeeze money out of the equity / stock markets. Money exiting the stock market, seeking safe haven in the bond markets would help stabilise the yield. Or at least that’s the theory… in  my mind it doesn’t really work, or to the extent that it does, is of marginal effect. Someone holding stock decides to sell it, and move the funds into treasuries. For this to happen a second party must be on the other side of the transaction, and buy the stock. Where is the extra money coming from?

It’s possible to argue that the extra money is cash that is sitting on the sidelines (Sideline Cash Myth) but this doesn’t seem to stand up to much examination.

It seems to me that this puts us back in our usual position: between a rock and hard place. On one hand the government needs to finance debt in order to attempt to stimulate the economy / compensate for all the money being destroyed as debts go bad, interest rates need to stay low to keep this debt affordable, and reduce financing costs. And on the other, the currency needs to be protected or the market will not be willing to buy the the additional debt due to low yield, and fears of default.

Therefore the governments of the west are attempting to keep the system ticking along in the hope that enough growth will return to allow the debts to be paid down. At this point in the cycle, the debts are so huge that, i believe, this is no longer possible, or is at best, extremely unlikely.

Keep an eye on that graph. Nothing radical is going to happen now, but i suggest this is how you’ll know who (the central banks or the markets) is “winning” the war.

As an aside, a lot of this talk is about the dollar, but that’s mostly because it the easiest set of data to work with. The same situation is playing out in the other western economies, and the various central banks are fighting the same battle. The interesting exception to this rule, as always it seems, is the Bank of Japan, which despite printing money like it’s going out of fashion, has kept the yield on the japanese long bond remarkably stable. Again, as usual, i don’t understand the mechanism by which this can be explained…