Veryan Allen: Investing, trading or gambling?

15.1.08

What is the difference between investing, trading and gambling? The first two come down to a divide in the holding period continuum; microseconds to months is trading but years is investing. Superficially investing and gambling are quite similar; putting money at risk in the hope of making more money. But most investors would balk at the idea of being called a gambler.

Surely the difference is that investing is deploying capital when you have an edge while gambling is betting when you do not have an edge. To make consistent absolute returns it is necessary either to have an advantage or identify someone else with one. An edge does not eliminate the possibility of small, manageable losses but it does mean persistent and predictable performance.

By definition there is no edge in beta and it is not very reliable over less than multi-decade time frames. Short only equity seems to be working quite well so far in 2008 and stock indices in many major developed markets have erased most gains from last year. The S&P 500 is now around 1400 just as it was 12 months ago and in January 2000. Investors don't seem to have received much equity risk premium or been compensated for the volatility despite what the economics textbooks say but then stocks can't read. It could be worse; in Jan 1988 the Japanese Nikkei was at 24,000 and now, 20 years on, it is at 14,000. Prudence mandates acknowledging the possibility of an extended bear market and constructing a portfolio that can grow without the benefit of beta. It just snowed in Baghdad; unlikely things can and do happen. Buy a snowplow even if you have never seen snow.

There are reasons to be bullish of course. There always are. The "private equity put" and "Greenspan put" evaporated in 2007 to be replaced by the "Sovereign Wealth Fund put". Many economists are predicting a recession which, given their track record, means there is a good chance there won't be one. Many US financial institutions will report earnings soon and with new CEOs and new stock options the temptation to write down doubtful CDOs, SIVs and loans to very conservative levels and adopt a kitchen sink approach to disclosing bad news must be very high. Last quarter can be blamed on former management but not the next quarter. Ben Bernanke promising substantial rate cuts was clearly preparing the market for the bad news to come. Short sell the rumor, buy the fact?

Sometimes a stock, bond, commodity, currency or any other specific asset goes up and sometimes it goes down. Timing these moves is hard but a few can do it. The relationship between them opens up anomalies and inefficiencies that can also be exploited. Casinos and the CIA are now using something called NORA or Non-Obvious Relationship Awareness in their security work. Successful investing is very dependent on non-obvious relationship awareness between securities. It was the key to success in 2007 and will be more so in 2008. This is where many go wrong; looking at a single stock, pair of securities or one asset class when it is the ENTIRE interrelated puzzle that needs solving.

Assets cannot be looked at in isolation as they all have an effect on each other. Commodities move stocks, currencies impact bonds and vice versa. NORA showed how long biased credit strategies could hurt some of the more crowded "market neutral" equity trading strategies. Some central bankers think raising interest rates will curb inflation and lowering interest rates will help the economy. Not necessarily anymore as global capital flows and new, non-obvious relationships between assets and geographies may have changed the economic game. High rates in Iceland or New Zealand or low rates in Japan or Taiwan haven't had quite the effect that "theory" anticipated.

Strategies make money out of and between assets. But in implementing a strategy a fund must either have a wide protective moat of a talent-based barrier to entry or keep it secret. Many things in the public domain did NOT work last year but is that surprising? The Dogs of Dow, the January effect, the "Magic formula" of value investing are too well known to work anymore. Those arbs, among others, are gone. I hope for the sake of the long only crowd that the "First 5 days in January" effect is NOT predictive for 2008. However I would be pleasantly surprised if the Dow and Nikkei don't dip below 10,000 sometime this year.
　
I spent New Year in the bastion of statistical arbitrage, Las Vegas, the only city in the world named after a volatility metric. I am always long vega and long Vegas. The "usual" opinion on casinos is you can't beat the house just like conventional wisdom in finance is that you can't beat the market. In general that is indeed true. The sweat equity, concentration and aptitude required to perform such a difficult task on a consistent basis is rare. Difficult yes, impossible no. Like others I've taken the time to try to develop an edge in picking fund managers and picking securities.

As in financial markets there are advantages that can be developed in casino games to change the negative expectation of gambling to the positive expectation of investing. But it requires dedication and proprietary research. Losses may still occur since the edge applies over time. Many people are aware Blackjack can be beaten but disclosure of many of the techniques and changes in the rules have reduced that edge. The first time I visited Vegas I knew basic strategy and the probabilities almost as well as Ed Thorp and could memorize cards almost as well as Dustin Hoffman and I did reasonably well; nowadays I am content to break even.

Despite the increased sophistication and monitoring at casinos there are still professional backjack players making good money from innovating their strategy and developing their talent. Just like a hedge fund keeps refining and adapting its edge and finding new ones. Even roulette and dice games can be "beaten" where beaten means having a small probabilistic bias that overcomes the house's advantage; it just takes very high skill AND practice to do it. Poker is a game of luck over one deal but skill over many deals. When I look up at the sports book in a Nevada casino I see potential mispricings and arbitrages all over the board just like on a futures floor or page of stock price quotes; it just takes hard work and deep domain knowledge of the teams, players and horses to identify them.

Slot machines are interesting too. The house always has the edge but that does NOT mean they should ALL be avoided. The POSSIBILITY of enormous returns for a very low capital outlay provides a different value proposition so I did sometimes play slots with mega jackpots. "Experts" say that the odds of winning are so remote (1 in 100 million or so) as to make them a loser's game but that is because they don't understand the payoffs and probabilities correctly.

Just like a national or state lottery, the chance that the jackpot will be won is 1.00, ie a certainty. Someone WILL win it. If you don't play you have ZERO chance of winning. If you DO play you have an unlikely but NON-ZERO chance. Since any number divided by zero is infinity, that means the simple act of risking a few bucks increases the chance of winning by an infinite multiple! Therefore the optimal trading algorithm on a lottery or a Megabucks slot machine IS to play but only with small cash. Similar to buying far out of money options; even if most expire worthless, you only need a single high pay out.

On New Year's Eve I happened to put $20 into a Wheel of Fortune slot machine and won $1,000. Deducting fees of 5% and 50% that is a return of 2,400%! So now you know what the "highest" performing "hedge fund" was last year - the Nevada Slots Opportunities Fund. A stupid claim of course but sadly such dumb, unrepeatable luck has been used to market many a fund. That return was "pure alpha" of course as I had the "skill" to pick the right machine in the right casino at the right time. I have seen even sillier arguments in powerpoint presentations and at investment conferences.

Suppose I had then lent the $1,000 to someone who "promised" to pay back $2,000 if they won speculating on real estate. What if I assigned an overly optimistic default probability to this "trade" and launched the Enhanced Nevada Credit Fund on the back of this "amazing" mark-to-model yield. Sounds even dumber but that is what Merrill Lynch, Citigroup, Bear Stearns, Northern Rock, Sowood and Dillon Read among several others were effectively doing last year. Dress it up with the high finance of Klio and Norma subprime CDOs but basically it was all alchemy of loan finance and fictitious capital that was unlikely to be paid back. High yield only makes sense if it is higher than the risk.

Whenever I hear the argument for long term passive investing I wonder what temporal era is meant - geological or cosmological time? Over periods of relevance to living humans I'd rather invest in alpha than gamble on beta. Suppose in 2020 or 2030 major equity indices are LOWER than today? Lost year, lost decade? Gold may be above $900 today but remains far below its inflation-adjusted high set over 600 years ago. I'll take different strategies applied to assets over asset classes themselves every time.

Most investors cannot wait long enough for the beta bet to pay off and why should they when they can allocate to fund managers with the skill to generate reliable absolute returns from their edge? Investing and trading both have important roles to play in a portfolio but it is no place to gamble. The only trades to make and the only managers to pick are those with positive expectation and the odds in their favor.