Betting on the Odds
This post was first made on the ValueAsia website here. It was published in Chinese here.
A friend told me he spent much of last year re-reading the fifty-plus year canon of Warren Buffett’s annual letters and found it full of contradiction. Buffett counsels against focusing on macroeconomics but in the early 2000s, made a multi-billion dollar wager that the US dollar would depreciate. He scorned the gold bugs for adoring a shiny but unproductive metal, yet was happy to speculate in silver. He warned of the catastrophic danger of derivatives but frequently bought and sold them himself. He has told us to bet on the horse and never the jockey, yet regularly lavishes praise on the managers of his investments in highly competitive, often commoditised industries. And while he admonishes deep research into companies with ‘wide moats’ like See’s Candies and Coca-Cola, he gleefully purchased a basket of ‘no moat’ South Korean stocks after an afternoon’s reading.
How can we reconcile Buffett’s words with his actions? Is he contradicting himself? My take is that he is not. A narrow interpretation of Buffett’s experience would probably conclude that one should only buy businesses with a ‘wide moat’. But a broader interpretation might be that Buffett invests in anything where he can determine that the probability of success is overwhelmingly in his favour.
What do I mean by that? First, we need to begin thinking in terms of contingencies. A typical stock pitch begins with a description of the company and closes with a target price: “Company XYZ’s shares are trading at $50 but I think they’re worth $100”. Let’s assume that the analysis and reasoning behind this conclusion are reasonable, and that the shares could be worth $100. A more nuanced listener, however, would recognise that this is just one contingency of many; the shares could just as well be worth more, or less. What matters is the range and distribution of those contingencies, and where on that range we believe the current price implies we are sitting. A more practical conclusion for the pitch therefore might simply state that there is an overwhelming probability that the shares are worth more than they are today; or conversely, that there is a negligible probability that they are worth less. Analysts used to calculating valuation out to three decimal places might be shocked by this suggestion. But as Buffett’s teacher Benjamin Graham once said, “I don’t need scales to tell you that a 300lbs man is overweight”.
Second, we need to define the situations where we have a reasonable chance of determining that range and distribution of contingencies. The way Graham stacked the odds in his favour was to invest in a diversified portfolio of ‘cigar butts’: debt free stocks trading at less than their immediate liquidation value. With virtually zero probability that these businesses could be worth less, the upside would take care of itself. But Graham’s approach has its drawbacks: a) such extreme valuations are rare; b) the opportunities are typically small in size and so hard to scale as a portfolio grows; c) they needed to be regularly bought and sold, thus incurring capital gains taxes (in America); and d) without a catalyst, the stocks could languish for extended periods of time.
As Berkshire Hathaway grew and at his partner Charlie Munger’s behest, Buffett modified Graham’s approach to look for larger, long term investments. But he never ceased to ask for a wide margin of safety. Rather, his insight was to seek companies with ‘wide moats’ within his ‘circle of competence’. The former led him to situations where the future could be reasonably determined, while the latter ensured that it could be reasonably determined by him. Moreover, in terms of future contingencies, ‘wide moat’ businesses by definition should have a narrower range and a distribution skewed towards more positive outcomes.
Legendary horse handicapper Steven Crist said that winners don’t bet on the horse, they bet on the odds. And if we reconceive Buffett’s investment style from ‘wonderful companies’ to ‘wonderful odds’, then I think his actions across all those different investments remain consistent with his words. Those that look only for the wonderful business miss the point: Buffett’s were good investments not simply because they were wonderful businesses. They were good investments because they were reasonably predictable businesses whose future prospects were mispriced. Seen in this light, I think the best way to define margin of safety is to demand odds overwhelmingly in one’s favour. As Buffett said, "I don't try to jump over 7-foot hurdles. I look for 1-foot hurdles that I can step over".