Asset Allocation6 分钟阅读

The Efficient Frontier

Why do putting everything into one stock and mixing several assets produce such different results? Here is a magic curve connecting the portfolios that 'extract the maximum return for the same risk.'

What is the efficient frontier?

The efficient frontier is a curve connecting the 'most efficient portfolios.' 'Efficient' here means this: among portfolios that bear the same risk (the degree to which the price swings), the combination with the highest return; and conversely, among those aiming for the same return, the combination with the lowest risk.

Drawn as a graph, the horizontal axis is risk (the standard deviation of returns, i.e., volatility) and the vertical axis is expected return. When you mix several assets in various proportions, countless portfolios scatter as points, and connecting only the optimal points among them—the ones for which 'nothing is better'—forms a curve that bulges toward the upper left. Being on this curve means it's a 'smart combination that can't be improved further,' while being below the curve means it's a 'combination taking a loss when it could earn more for the same risk.'

Being 'on the curve' doesn't mean it's unconditionally good. The upper-left of the curve is low-risk/low-return and the right is high-risk/high-return, so 'which point to choose' depends on each person's risk tolerance.

Who created it? (Markowitz and the Nobel Prize)

This concept was first formalized by Harry Markowitz in his 1952 paper 'Portfolio Selection' (Journal of Finance, March issue). At the time he made a paradigm shift of 'let's measure risk by the variation (variance) of returns,' and showed mathematically that mixing several assets can lower risk without giving up much return. This is the starting point of today's Modern Portfolio Theory (MPT).

For this achievement, Markowitz received the 1990 Nobel Prize in Economics. It was a joint award with William Sharpe and Merton Miller. From publishing the paper in 1952 to the Nobel Prize took about 38 years, showing just how far ahead of its time the idea was.

Sources: Wikipedia, Britannica Money, UBS Nobel Perspectives, and econlib.org cross-checked.

Why does 'mixing' reduce risk? (The power of diversification)

The key is correlation. When you mix assets that move in different directions, one holds up when another falls, so the overall swing shrinks. The remarkable thing is that it's not simply an 'average'—when correlation is low, risk can drop even below the average of each. This is exactly the part Markowitz proved mathematically.

Here, mixing in a risk-free asset (an almost certain return, like a deposit) makes the story even more interesting. When you draw a straight line from the risk-free rate that just touches (is tangent to) the curve, that touching point is called the 'tangency portfolio,' and this point is the combination with the greatest excess return per unit of risk. This 'reward-per-risk' metric is exactly the Sharpe Ratio.

What the theory misses (know the limits before using it)

The efficient frontier is powerful, but blind faith in it is dangerous. First, the theory assumes returns follow a normal distribution (a bell shape), but real markets have 'fat tails,' meaning extreme events like crashes and surges happen far more often than the theory suggests. The sharp declines of 2008 and 2020 are prime examples.

Second, to draw the curve you need to plug in each asset's expected return and correlation, but these values are mostly pulled from past data. The past doesn't guarantee the future, and even a slight change in the inputs greatly shakes the 'optimal' answer. Third, even the long-standard 60/40 (60% stocks, 40% bonds) portfolio relied on the premise that stocks and bonds move oppositely, and there is debate that as periods where the two moved together increased after the 2010s, the diversification effect weakened.

So it's safest to view the efficient frontier not as a 'tool for predicting the future' but as a 'framework for thinking about how to mix assets.' Actual drawdowns and drawdown durations can be far rougher than the theoretical curve.

常见问题

Q. Which point on the efficient frontier should I choose?

The answer differs for each person. The left is low in both risk and return, and the right is high in both. If sleeping soundly at night matters, you choose closer to the left; if you can withstand swings, closer to the right. Either way, being 'on the curve' means it's the best combination at that risk level. This is not a recommendation but a matter of choice based on each person's risk tolerance.

Q. If I know the efficient frontier, can I predict future returns?

No. This curve is only a 'theoretical optimal map' made from past data; it does not guarantee or predict future returns. Even a slight change in the expected returns/correlations you plug in shakes the optimal point, and the real market swings far more than the theory. It's best understood strictly as a framework for thinking when considering asset allocation.

Q. If I diversify, can I avoid losses entirely?

That's not the case. Diversification reduces the risk of concentration in a specific asset, but in a sharp-decline phase where the whole market collapses, several assets can fall together. In fact, even combinations like 60/40 have had periods where stocks and bonds fell together. Diversification is a tool to 'ease the swings,' not a guarantee of 'zero loss.'