A common theme among value or fundamental investors is that stocks should have a “margin of safety”, so as not to lose all their value, go bankrupt or be restructured, when they encounter difficulties. This is solid common sense, yet has no obvious rational reason to be true.
Like any quality, or defect, an investment might have, it’s possible to discount that advantage or defect, and market forces should tend to. In a perfectly efficient market every investment returns the same in risk adjusted terms.
In a simple example, which portfolio would one choose between a basket of stocks who have a binary outcome of their value tripling or going to zero, with balanced odds (half fail and half succeed) and a portfolio of stocks where half stay flat, and half grow by 20%; or equivalently are distributed within that range. For a portfolio large enough we get returns (expected value):
- risky portfolio: 50% (1.5 = 3 * 0.5 + 0 * 0.5)
- sedate portfolio: 10% (1.1 = 1.2 * 0.5 + 1.0 * 0.5)
Anyone should pick the risky portfolio. Through diversification, we can re-introduce the “margin of safety” to the portfolio as a whole, however risky the individual items. All that matters is the pricing relative to expected value of each investment, as long as the risk is individual (idiosyncratic and not correlated).
Of course in practice it seems to turn out — this is an expression of the low volatility anomaly — that the price of risky investment seems to be higher than the rational expected value. Hence the “margin of safety” rule of thumb being usually a good one to follow.
Good rule, wrong story, I would say. It is very much like folk wisdom often is.
It just captures the low volatility anomaly with an inaccurately simplistic story, which could be detrimental if low volatility became expensive, or when risky investments with little margin of safety are genuinely, if unusually, good value in probabilistic terms.