Blog: Demons, Butterflies and Margin of safety
If the answer to the first part of the question is obvious, we should revisit history. The notion that future is not predictable is a rather recent phenomenon. Newton theorized the laws of motion and universal gravitation in the late 17th century and post that, the motion of planets; eclipses and appearances of comets could be predicted years in advance with total precision. The French physicist PS Laplace manifested a super-intelligent being (now known as Laplace’s demon), who was aware of the motion of every particle in the universe. “If this intellect were vast enough to submit the data to analysis… for such an intellect, nothing would be uncertain, and the future, just like the past, would be present before its eyes,” he said. As if, the future was totally determinable, and we just needed enough information to decipher it.
However, Newton’s equations would explain things in a ‘fixed point attraction’ world (say, like planet earth revolving around the sun). But, introduce a third body (say, a moon – two forces of gravity now) and the equations didn’t work. It wasn’t clear for almost 200 years, until Henri Poincare suggested (in what later became the Chaos theory) that there is no fixed solution to the three-body problem.
Chaos came into the limelight in the 1960s when Ed Lorenz simulated the atmosphere on his computer using twelve variables and twelve equations. One day, he halted the simulation mid-way and manually entered the outcomes for the run to continue. That gave a totally different outcome compared to the previous runs. Ed later realized that he had entered the numbers rounded to three decimals, whereas the computer memory was working with six decimals. A change of ‘one part in a thousand’ led to totally different outcomes. This is better known as the ‘Butterfly effect’, i.e., a butterfly flutters its wings in a rainforest in Brazil, which can lead to a tornado in California. Essentially, the future was now not only unknown, but for some events, a very tiny change in initial conditions led to a dramatically different outcome.
To mitigate this conundrum, ‘margin of safety’ (MOS) investing was popularized by investors Benjamin Graham and Warren Buffet. In this principle, you purchased securities only when their market price is significantly below their intrinsic value. Intrinsic value can be arrived at by several methods, but all essentially involve forecasting the future. If you pay lower compared to a business’ intrinsic value, the downside to your investment will be limited even if the intrinsic value turns out to be lower.
The past few years, however, have not been kind to this theory (considered to be part of the ‘value investment’ framework). It has underperformed the ‘Growth at any price’ framework by a wide margin.
Consider this example. Over the past decade, while an Indian FMCG company’s share price is up 9X, its net income has grown only 4X. It now trades at a price-to-earnings ratio of c74x last year’s adjusted earnings. In order to justify its current price, a reverse discounted cash flow model has to assume that: (a) its free cash flows (FCF) will post 20% CAGR for the first decade; and (b) the terminal value is calculated using cost of equity of 10% & perpetual growth rate of 5%.
These are extraordinarily high estimates for any business to meet. It implies that the company’s FCF will jump 10x by 2040. Additionally, this is already built into the stock price now! For the stock to deliver returns higher than its assumed cost of equity, it will have to positively surprise the Street.
In addition, the outcome is sensitive to small changes in assumptions: For example, if we lower explicit forecast FCF growth to 10% (vs. 20%) and terminal growth to 3% (vs. 5%), the valuation of the business falls by 62%. The stock, nevertheless, continues to outperform even with little MOS in buying this business. This has happened largely because there is: (a) a scarcity of quality large investible businesses; and (b) inflow of money in the insurance & mutual fund industry continues to remain high. More money chases the same good quality asset.
While financialization of savings is on the rise, which continues to get channeled into mutual funds, and in turn, they keep buying businesses with little margin of safety, it is difficult to ‘predict’ if the value framework will ever make a comeback. Nevertheless, what is not sustainable, will find a way to not sustain; if it has not in the past, I believe it will not in the future as well. When, and not if, is the broader question.