In 1952 Harry Markowitz published an idea that would earn him a Nobel Prize: what matters is not the risk of an individual security, but its contribution to the risk of the WHOLE portfolio. A volatile asset can reduce overall risk if it moves out of step with the others.
The efficient frontier
For a given level of risk, there is a combination of assets that maximises the expected return. The set of these optimal combinations forms a curve: the efficient frontier. Any portfolio below it is sub-optimal — it takes as much risk for less return.
The central role of correlation
That is where the mechanism lies: combining two weakly correlated assets produces a portfolio whose risk is LOWER than the weighted average of the individual risks. Mathematically it is the only “free lunch” in finance, and the whole theory follows from it.
The limits, which you should know
The theory rests on FUTURE returns, volatilities and correlations, estimated from the past. Yet correlations spike during crashes: at the precise moment you need diversification, it weakens. The theory illuminates; it does not protect against everything.
What to take from it in practice
You do not need to optimise your portfolio mathematically. Keep the principle: judge each position by what it brings to the whole, not in isolation. Adding a tenth technology stock adds almost no diversification.
À retenir
- ✓What matters is the risk of the whole, not that of each security in isolation.
- ✓The efficient frontier = the best possible return for each level of risk.
- ✓Correlations rise during crises: diversification weakens exactly when you need it most.