Capital Ideas

Peter Bernstein

4 ideas

1. Bachelier's ignored 1900 random walk thesis

   In 1900 Louis Bachelier, a French doctoral student, modeled Paris bond and option prices as a random walk. He argued that the expected gain of a speculator is zero because prices already reflect all available information. His thesis got only a mediocre grade and was ignored for more than fifty years, until economists rediscovered it in the 1950s. The mathematics of efficient markets and option pricing had existed decades before anyone was ready to use it.

2. Diversification prices risk, not individual securities

   Markowitz showed that a security's riskiness depends on how it covaries with the rest of the portfolio, not on its own volatility in isolation. So an investor should choose portfolios on an efficient frontier, the set that maximizes expected return for each level of variance. This turned 'risk' from a vague feeling into a measurable quantity that can be traded off against return.

3. Only undiversifiable risk earns a premium

   Sharpe's CAPM argues that specific risk can be eliminated for free through diversification, so the market will not pay anyone for bearing it. Expected returns are therefore set only by beta, a security's sensitivity to movements in the overall market. Stocks that are volatile but uncorrelated with the market should earn no more than the risk-free rate plus their small beta-driven premium.

4. Price changes as unforecastable news arrivals

   Fama's efficient-market view treats every price change as a response to new information. By definition, new information is unpredictable, so if competing investors quickly incorporate what is known, future price moves cannot be forecast from past prices or public data. Seen this way, a professional manager's persistent failure to beat an index shows the market is working, not that managers are incompetent.

Save and mark ideas in the app