Geoff Gannon June 3, 2018

Anything Times Zero

To Focused Compounding members:
Since I read “Fortune’s Formula” and “A Man for All Markets” this past week, let’s talk blackjack. In blackjack, the player has an advantage over the casino if he’s counting cards. A card counter can bet nearly the minimum when he suspects the rest of the deck has cards unfavorable to him in higher proportion than a fresh deck and he can bet nearly the maximum when he suspects the rest of the deck has favorable cards in a higher proportion than a fresh deck. Applying this to stocks, let’s say you’re convinced Wells Fargo is a safe bank and Bank of the Ozarks is a risky bank. You have $10,000 to invest in bank stocks. These are the only two bank stocks you know anything about. You have one question: what happens if instead of taking your $10,000 and putting $5,000 into Bank of the Ozarks and $5,000 into Wells Fargo you instead put $1,000 in Bank of the Ozarks and $9,000 in Wells Fargo. You still start off with $10,000 worth of bank shares, but now you are acting like a card counter – betting nearly the maximum when you think the rest of a deck (Wells Fargo’s future) is favorable and betting nearly the minimum when you think the rest of deck (Bank of the Ozarks’ future) is unfavorable. How important is your decision to split your money 10/90 in favor of Wells Fargo?

Let’s say the chance of Wells Fargo stock going to zero in any one year is 0.5% and the chance of Bank of the Ozarks going to zero in any one year is 5%. Over a single year, a bigger annual upside – especially in the form of a quicker catalyst – can make up for a stock being 10 times riskier. Stocks are volatile. And any extra chance of a 50% pop in the stock’s price this year could overcome a 4.5% difference in the rate of catastrophe. So, if you frame your own investment lifetime as lasting only a single year – the math says it’s perfectly fine to bet as much on Bank of the Ozarks as on Wells Fargo. Catastrophic failure is not a big deal over one year. And you’ve promised yourself you’ll only play one hand. You’ll buy both Wells and Bank of the Ozarks today and sell twelve months from today no matter what. Whatever result you get won’t compound. That makes failure cheap. And if there’s some upside catalyst you see for Bank of the Ozarks this year – that catalyst could overcome the 4.5% greater chance of catastrophe this year. But, that’s framing the choice as a one-year bet. Buffett has owned Wells for 27 years. So, let’s ask: what is the difference between a 99.5% annual survival rate and a 95% annual survival rate if you’re committed to letting each bet ride for the full 30 years? Now failure isn’t cheap. It’s expensive, because it kills compounding. If you keep letting the risky bet ride, you will – sometime after five years – have survived a 1-in-4 chance of catastrophe, after 10 years you’ve dodged a 4-in-10 chance, after 15 years you’ve survived a 50/50 chance of disaster, after 20 years a 2-in-3 chance of disaster, and after 30 years a 4-in-5 chance of total failure. Meanwhile, in Wells Fargo you’ve dodged a 1-in-6 chance of failure after the full 30 years. If the surviving stock had been compounding at even just 6.2% a year it is now worth 6 times what you originally paid. Let’s look at the 50/50 split approach. If you held for 30 years, your two $5,000 bets would result in an 80% chance of having zero dollars in the first bucket (Bank of the Ozarks) and an 85% chance of having $30,000 in the second bucket (Wells Fargo). A 10/90 split would – after 30 years – result in you having an 80% chance of having zero dollars in the first bucket (Bank of the Ozarks) and an 85% chance of you having $54,000 in the second bucket. The tiny bet / giant bet approach works when holding for 30 years. Why? It’s not because the safe stock ever offers better one-year odds than the risky stock. It’s just that the safe stock offers the best chance of never killing compounding.