Expected Return

Expected return is the probability-weighted average return of a future distribution of possible outcomes. It is a model estimate based on stated probabilities, forecasts, or asset-pricing assumptions, not the return an investment is guaranteed or most likely to earn.

Key Takeaways

• Expected return combines possible returns with their probabilities.
• It is forward-looking; historical average return is a backward-looking sample statistic.
• The expected return can differ from the most likely outcome and the median outcome.
• Portfolio expected return is the weighted average of component expected returns when weights and return definitions are consistent.
• Two investments can have the same expected return but very different volatility, downside, liquidity, or loss severity.
• Model inputs, fees, taxes, inflation, currency, and investment horizon should be explicit.
• A precise expected-return estimate is not a promise, target, recommendation, or prediction of the next period.

Expected Return Formula

For discrete possible outcomes:

where:

• (E(R)) is expected return
• (p_i) is the probability assigned to outcome (i)
• (R_i) is the return in outcome (i)
• the probabilities sum to 1

Expected return is a mean of a modeled probability distribution. It does not describe the spread, asymmetry, or sequence of outcomes.

Worked Example: Scenario Return

Assume a one-year investment model has three mutually exclusive outcomes:

The expected return is:

    flowchart LR
	  A["One-year investment"] --> B["Strong: 20% chance, 12% return"]
	  A --> C["Base: 50% chance, 6% return"]
	  A --> D["Weak: 30% chance, -4% return"]
	  B --> E["Probability-weighted expected return: 4.2%"]
	  C --> E
	  D --> E

The investment cannot earn 4.2% in any of the three modeled scenarios. The number is the distribution’s mean, not a fourth scenario.

Expected Return Is Not the Most Likely Return

Consider a separate hypothetical investment with:

• a 90% probability of a 0% return
• a 10% probability of a 100% return

Its expected return is 10%, but its most likely outcome is 0%. A large low-probability gain pulls the mean above the mode and median.

This distinction matters for options, distressed securities, start-ups, insurance-like payoffs, and other skewed distributions. Reporting only the mean can hide the probability and size of loss.

Expected Return for a Portfolio

For portfolio weights (w_j) and component expected returns (E(R_j)):

Assume a portfolio has:

• 60% in Asset A with expected return of 7%
• 40% in Asset B with expected return of 3%

Then:

Correlation does not change this weighted-average expected-return calculation. Correlation and covariance do affect portfolio variance and diversification. The portfolio variance calculation therefore needs more than component expected returns.

The weights should represent the same point in time and sum to one for a fully invested long-only portfolio. Leverage, short positions, derivatives, and changing weights require additional care.

How Expected Return Is Estimated

No single method reveals the true future return.

Historical data can inform an expectation, but the analyst should explain why the sample, market regime, currency, and measurement basis remain relevant.

Expected Return Under CAPM

The Capital Asset Pricing Model estimates expected or required return from a risk-free rate, beta, and market risk premium:

CAPM provides a disciplined benchmark, but its market portfolio is unobservable and its rate, beta, and premium inputs are estimated. The output is conditional on that model; it is not an independent fact about the security.

Other factor models may include size, value, momentum, term, credit, or other exposures. More factors do not automatically make an estimate accurate if exposures and premiums are unstable or selected after seeing the results.

Expected Return Versus Required and Realized Return

Expected return and required return may be set equal in an equilibrium valuation model, but they answer different questions. An investor’s forecast can be above or below the return required for a decision.

Arithmetic Expectation Versus Compound Growth

The one-period expected return is an arithmetic probability-weighted mean. It should not be compounded mechanically as if the same deterministic return occurs every year.

Under uncertainty:

• the expected ending wealth depends on the full return process
• the expected compound growth rate is affected by volatility and sequence
• arithmetic mean return and geometric mean return differ
• changing probabilities and correlations can alter multi-period outcomes

An expected annual return of 6% does not guarantee that $100 becomes $106 after one year or $112.36 after two. Those are expected-value or deterministic projections only under additional assumptions.

Risk Cannot Be Read From the Mean Alone

Two modeled investments can have the same expected return:

The table does not claim Investment A exists without credit, inflation, liquidity, or reinvestment risk. It illustrates why expected return should be reviewed with standard deviation, downside scenarios, drawdown, liquidity, leverage, and loss severity.

Nominal, Real, Gross, Net, and After-Tax Expectations

An expected-return estimate is incomplete unless its measurement basis is stated.

• nominal expected return includes expected inflation
• real expected return measures expected purchasing-power growth
• gross expected return precedes specified fees and expenses
• net expected return deducts the costs defined by the methodology
• pre-tax and after-tax expectations depend on different cash flows and assumptions
• local-currency and base-currency returns can differ because of exchange rates

Do not subtract fees, inflation, and taxes from one estimate unless the underlying rates and timing are compatible. Compounding relationships are more accurate than simple subtraction when rates are material.

Estimation and Review Checklist

Before relying on expected return, document:

1. Investment, portfolio, currency, and horizon.
2. Return definition, including income and reinvestment.
3. Gross/net, nominal/real, and pre-tax/after-tax basis.
4. Scenario outcomes and evidence supporting each probability.
5. Historical sample dates and treatment of outliers or missing data.
6. Model, factor exposures, premiums, and valuation date.
7. Correlations and path assumptions for multi-period projections.
8. Downside, stress, and liquidity analysis.
9. Sensitivity to probabilities, returns, and fees.
10. Difference between the expected return and the decision’s required return.

Sources and Trust Checks

• OpenStax’s Probability Distributions explains probability-weighted expected values and portfolio weights.
• Investor.gov’s Performance Claims bulletin distinguishes hypothetical projections, backtests, and actual historical performance.
• Investor.gov’s Risk and Return explains that potential return and risk are related and market returns are not guaranteed.
• NYU Stern’s risk and return materials cover expected returns, risk premiums, beta, and asset-pricing assumptions.

Common Mistakes and Limitations

• Treating expected return as the most likely result.
• Assigning probabilities that do not sum to one.
• Mixing nominal scenario returns with real discount rates.
• Treating a historical average as a forecast without adjustment or explanation.
• Ignoring skewness, tail loss, liquidity, leverage, and path dependence.
• Comparing gross expected return with net historical performance.
• Adding expected returns across assets without portfolio weights.
• Assuming a one-period mean compounds unchanged over many periods.
• Using false precision for uncertain probabilities and premiums.
• Increasing the expected return to justify a preferred investment or valuation.

Related Terms

• Mean Return: Arithmetic or geometric average of a defined return sample.
• Total Return: Completed-period performance from price change and income.
• Standard Deviation: Dispersion measure that adds risk context to an expected return.
• Portfolio Variance: Portfolio risk measure incorporating variances, weights, and covariances.
• Capital Asset Pricing Model: Single-factor expected- or required-return model.
• Required Rate of Return: Decision threshold distinct from a forecast expectation.

FAQs

Educational Use

This article provides general financial education. Expected-return estimates are uncertain and do not constitute a guarantee, forecast of a specific outcome, or personalized investment, tax, legal, or portfolio advice.