What Is the Efficient Frontier? Markowitz Portfolio Theory Explained

Search Fintrev

Search our guides, reviews and glossary.

The efficient frontier is the set of portfolios that offer the highest expected return for each level of risk, or equivalently, the lowest risk for each level of expected return. Plotted on a chart with risk on the horizontal axis and expected return on the vertical axis, these portfolios form an upward-sloping curve. Any portfolio below the curve is inefficient, because another mix of the same assets delivers more return for the same risk. Points above the curve cannot be reached with those assets.

The concept comes from Harry Markowitz, who introduced it in his 1952 paper “Portfolio Selection.” It became the foundation of modern portfolio theory (MPT), and Markowitz shared the 1990 Nobel Memorial Prize in Economic Sciences for the work.

How the efficient frontier works

Markowitz framed investing as a trade-off between two numbers: expected return and risk, with risk measured as the variance or standard deviation of returns. For this reason the method is also called mean-variance optimization (MVO), and the full curve is sometimes called the mean-variance frontier.

Building a frontier takes three sets of inputs for every asset:

1. Expected return: often estimated from historical returns and commonly expressed as annualized CAGR.
2. Standard deviation: usually annualized, as the measure of each asset’s volatility.
3. Correlations or covariances: how each pair of assets moves together.

An optimizer then searches across every possible set of portfolio weights. For each target return, it finds the weights that produce the lowest portfolio standard deviation. Connecting those minimum-risk points traces the curve.

The math for two assets

For a portfolio with weight w in asset A and (1 − w) in asset B:

• Expected return: E(Rp) = w × E(RA) + (1 − w) × E(RB)
• Variance: σp² = w²σA² + (1 − w)²σB² + 2w(1 − w)ρσAσB

Here σ is standard deviation and ρ is the correlation between A and B. Return is a simple weighted average, but risk is not. The final term shrinks as correlation falls, which is why combining imperfectly correlated assets lowers risk without lowering the weighted-average return. With many assets, the same logic is written in matrix form using a covariance matrix.

Interpreting the curve

The frontier bends because of diversification. As you move right along it, each extra unit of risk buys a smaller increase in expected return, a diminishing marginal return to risk.

The leftmost point of the curve is the global minimum variance portfolio, the lowest-risk mix available. The full curve, including the lower half that bends back below that point, is the minimum variance frontier. Only the upper half is the efficient frontier. Every portfolio on the lower branch has a counterpart directly above it with the same risk and a higher return.

Example of the efficient frontier

Hypothetical inputs, for illustration only:

Assume a correlation of 0.2 between them.

50% A / 50% B: Expected return is 8%. Variance is 0.25(0.01) + 0.25(0.04) + 2(0.25)(0.2)(0.10)(0.20) = 0.0145, so standard deviation is about 12.0%. That is well below the 15% weighted average of the two assets’ risk.

Minimum variance mix: The weight in A that minimizes risk is (σB² − ρσAσB) ÷ (σA² + σB² − 2ρσAσB) = 0.036 ÷ 0.042 ≈ 85.7%. This portfolio has an expected return of about 6.6% and a standard deviation of about 9.6%.

Holding 100% of Asset A, the “safer” asset, is therefore inefficient. Adding roughly 14% of the riskier Asset B lowers risk from 10% to 9.6% and raises expected return from 6% to 6.6%. Mixes with more of Asset B than that sit on the efficient frontier, and each additional slice of B adds return at the cost of more volatility.

How the efficient frontier is used

Choosing a risk level. The frontier does not name a single best portfolio for everyone. It narrows the choice to efficient options, and the investor picks a point based on how much volatility they can accept: risk-averse investors sit toward the left end, risk-tolerant investors toward the right. In theory, this is where the frontier meets the investor’s indifference curve, the set of risk/return combinations that person finds equally satisfying.

Finding the optimal risky portfolio. When a risk-free asset is added, the straight line drawn from the risk-free rate to the point where it just touches the frontier is the capital market line. That tangency point is the portfolio with the highest Sharpe ratio, meaning the most excess return per unit of risk. Investors combine it with the risk-free asset (lending), or borrow to hold more of it, rather than picking a different risky portfolio on the curve.

Asset allocation. Advisers and institutions run mean-variance optimization across asset classes such as stocks, bonds and cash, often using index funds or ETFs as building blocks. The output is a starting allocation, usually adjusted with constraints such as no short selling or maximum weights.

Limitations of the efficient frontier

Normally distributed returns. The model treats risk as standard deviation, which fits a normal distribution. Real returns are leptokurtic, or heavy-tailed: moves of more than three standard deviations from the mean happen more often than a normal curve predicts. Standard deviation therefore understates the risk of large losses.

Estimation error. The frontier is only as good as its inputs. Expected returns are especially hard to forecast, and small changes in them can swing optimal weights sharply. Optimizers also tend to overweight assets whose returns were overestimated. A frontier drawn from historical data shows which mixes were efficient. It does not improve future performance by itself.

Unstable correlations. Correlations change over time and often rise during market stress, which shrinks the diversification benefit when it is most needed.

Idealized investors and markets. Markowitz assumed investors are rational and risk-averse, cannot move prices on their own, and can borrow and lend without limit at the risk-free rate. Real markets include risk-seeking investors, large players whose trades affect prices, and borrowing that is limited and costs more than the risk-free rate. Taxes and transaction costs are also left out.

Single-period view. The classic model optimizes over one period and does not account for rebalancing, changing goals or cash flows over time.

Efficient frontier vs. related concepts

• Capital market line: a straight line that includes a risk-free asset. It lies above the efficient frontier everywhere except at the tangency point.
• “Frontier” in production economics: terms like the best-practice or cost-efficiency frontier describe firm productivity, not portfolio risk and return.
• Efficient market hypothesis: a separate theory about whether prices reflect available information. It is unrelated to how the efficient frontier is built.

The Bottom Line

The efficient frontier matters to investors, advisers and institutions setting asset allocations across stocks, bonds and cash. It narrows the choice to efficient mixes, and each investor then picks a point based on how much volatility they can accept. Because inputs are hard to estimate, correlations shift under stress and real returns have heavy tails, its output works best as a starting allocation adjusted with constraints.