If you invest $10,000 and end up with $20,000 ten years later, your return is 100%. Or is it 7.2% per year? Or 8.4%? Or 6.1%? The answer depends on which return metric you use—and choosing the wrong one can lead to deeply misleading conclusions. A stock that "averaged 12% per year" might have actually delivered 9% in your account. A real estate investment that "doubled in value" might have underperformed a savings account once you account for the timing of your cash flows. Understanding the differences between simple return, CAGR, IRR, and time-weighted return is essential for evaluating investments honestly.
This guide covers each major return metric, when to use it, why average returns mislead, and the most important return concepts for retirement planning—real versus nominal returns, sequence of returns risk, and volatility drag.
Simple return: the basic calculation
Simple return (also called total return or holding period return) measures the total gain or loss on an investment over the entire period, expressed as a percentage of the initial investment.
Simple Return = (Ending Value − Beginning Value) / Beginning Value
If you invested $10,000 and now have $15,000, your simple return is ($15,000 − $10,000) / $10,000 = 50%. Simple, intuitive, and limited—because it ignores time.
The limitation
A 50% return over 1 year is spectacular. A 50% return over 20 years is terrible. Simple return doesn't distinguish between them. To compare investments held for different periods, you need an annualized measure.
CAGR: compound annual growth rate
CAGR answers the question: "What constant annual return would have produced this final value?" It assumes the investment grew smoothly at the same rate every year—even though actual returns fluctuated.
CAGR = (Ending Value / Beginning Value)^(1 / Years) − 1
If $10,000 grew to $15,000 over 5 years:
CAGR = (15,000 / 10,000)^(1/5) − 1 = 1.5^0.2 − 1 = 0.0845 = 8.45%
The investment didn't actually return 8.45% each year—it might have returned +20%, −10%, +5%, +15%, +12%—but CAGR tells you the equivalent steady rate. This is the metric to use when comparing investments held for different time periods.
When CAGR misleads
CAGR has two limitations. First, it assumes you invested a single lump sum at the start. If you contributed additional money over time (like a 401(k)), CAGR overstates your actual return because it credits all growth to the initial amount. Second, CAGR hides volatility—an investment that returned 8.45% annually without ever losing money is far more valuable than one that averaged 8.45% but lost 40% in a single year.
IRR: internal rate of return
IRR is the metric that handles real-world investing—multiple contributions and withdrawals at different times. It's the discount rate that makes the net present value of all cash flows equal to zero.
For an investment where you put in $10,000 in year 1, another $5,000 in year 2, withdraw $3,000 in year 4, and end up with $18,000 in year 5, IRR is the annualized return that makes the math work.
How IRR is calculated
IRR can't be solved algebraically—you need iterative numerical methods (which is what Excel's =IRR() and =XIRR() functions, and Google Sheets' equivalent, do). The XIRR function handles irregular cash flow dates, which is what most real-world investing looks like.
When to use IRR
IRR (specifically XIRR) is the correct metric for:
- 401(k) and IRA accounts with regular contributions
- Real estate with down payment, ongoing costs, and sale proceeds
- Any investment where you added or withdrew money over time
- Comparing your actual portfolio performance to a benchmark
For a 401(k) where you contributed $500/month for 10 years and ended with $90,000 (total contributions: $60,000), your simple return would be 50%. But your XIRR—the actual annualized return you earned—might be 6.5% or 8.2% depending on when markets rose and fell during your contribution period.
Time-weighted vs money-weighted returns
These are the two formal methodologies institutional investors use:
Time-weighted return (TWR)
Measures the performance of the investment itself, independent of when money was added or withdrawn. Each period between cash flows is calculated separately and geometrically linked. TWR is what mutual funds report—it tells you how the fund manager performed regardless of investor cash flow timing.
Money-weighted return (MWR)
Same as IRR—measures the actual return you earned, weighted by how much money was invested at each point in time. Your MWR can be higher or lower than the fund's TWR depending on when you bought and sold.
Why the difference matters
If you invested $10,000 in a fund at the start of a year, the fund rose 50%, you added $100,000 at the peak, and then the fund fell 33%—you'd have lost money overall, even though the fund's TWR was 0% (50% × 0.67 = 1.0). Your MWR would be deeply negative because most of your money was invested at the worst time.
This is the investor performance gap: investors systematically underperform the funds they invest in, because they buy high and sell low. DALBAR's annual study consistently shows investors earning 2–4 percentage points less per year than the funds they hold.
Total return: dividends matter
Total return includes both price appreciation and dividends/interest. A stock that goes from $100 to $110 paid a $3 dividend has a total return of 13%, not 10%. Always compare total returns, not price returns.
This matters enormously for dividend-paying stocks and funds. From 1930 to 2023, the S&P 500's price return was about 5.5% annually, but total return (including reinvested dividends) was about 10.2%. Dividends contributed nearly half of long-term compound growth.
Dividend reinvestment
Reinvesting dividends (via a DRIP—Dividend Reinvestment Plan) automatically buys more shares with each dividend payment, compounding growth. Without reinvestment, you lose the compounding effect on dividends. Over 30 years, the difference between reinvested and non-reinvested dividends can be hundreds of thousands of dollars on a moderate portfolio.
Real vs nominal returns
Nominal return is what you see quoted—10% means your account went up 10% in dollars. Real return is nominal minus inflation—what your money actually gained in purchasing power.
Historically, U.S. stocks have returned about 10% nominally and 7% really (after 3% average inflation). Bonds have returned about 5% nominally and 2% really. Cash has returned about 3% nominally and 0% really—just keeping pace with inflation.
The compounding effect of inflation
Over 30 years, 3% annual inflation erodes 60% of purchasing power. $1 million in 30 years has the buying power of about $412,000 today. When planning for retirement, always think in real terms—a $1 million target in today's dollars is a $2.4 million target in 30 years' nominal dollars.
The Fisher equation
The precise relationship: (1 + nominal) = (1 + real) × (1 + inflation). For small rates, this approximates to nominal ≈ real + inflation. But for large rates or high inflation, the precise formula matters.
Why average returns mislead: volatility drag
"The market averages 10% per year" is a misleading statement because arithmetic mean (simple average) and geometric mean (compound rate) diverge in the presence of volatility.
The math
Suppose an investment returns +50% in year 1 and −50% in year 2. Arithmetic average: 0%. But $100 becomes $150 (year 1), then $75 (year 2). You lost 25% over two years, despite an "average" return of 0%.
The geometric mean (CAGR) is √(1.5 × 0.5) − 1 = √0.75 − 1 = −13.4% per year. That's the actual return you experienced.
Volatility drag
The gap between arithmetic mean and geometric mean is called volatility drag. The approximate formula:
Geometric Mean ≈ Arithmetic Mean − (Variance / 2)
For U.S. stocks with 10% arithmetic return and 16% standard deviation (variance = 0.0256): geometric mean ≈ 10% − 1.28% = 8.72%. The volatility "drag" is about 1.3 percentage points per year.
This is why diversified portfolios outperform concentrated ones at the same expected return—diversification reduces volatility, which reduces drag, which increases compound return. A 60/40 portfolio might have a lower arithmetic return than 100% stocks but only slightly lower geometric return, because it's less volatile.
Benchmark comparisons
Always compare your returns to an appropriate benchmark. A 7% return is great if the benchmark returned 4%; it's poor if the benchmark returned 12%.
Common benchmarks
- S&P 500: large U.S. stocks. Use for comparison of any U.S. equity portfolio.
- MSCI World or ACWI: global equities. Use for international portfolios.
- Bloomberg U.S. Aggregate Bond Index: investment-grade bonds. Use for bond portfolios.
- 60/40 portfolio (60% stocks, 40% bonds): balanced portfolio benchmark.
- Custom benchmark: weight multiple indexes to match your asset allocation.
The right benchmark matches your asset allocation. Comparing a conservative 40/60 portfolio to the S&P 500 is meaningless—you're taking less risk, so you should expect lower returns.
Risk-adjusted returns
For comparing investments with different risk levels, use risk-adjusted metrics:
- Sharpe ratio: (return − risk-free rate) / standard deviation. Higher is better.
- Sortino ratio: like Sharpe but only penalizes downside volatility.
- Treynor ratio: return per unit of beta (market risk).
These let you compare a low-risk bond fund returning 5% with a high-risk stock fund returning 10% on an apples-to-apples basis.
Sequence of returns risk: critical for retirees
For accumulators (those still saving), the order of yearly returns doesn't matter—only the average. A portfolio that returns −30%, +40%, +5% over three years ends up in the same place as one returning +5%, +30%, −20%.
For retirees, sequence matters enormously. Withdrawals amplify losses: if you retire with $1 million and withdraw $40,000/year, a 30% market drop in year 1 means you're withdrawing from $700,000, not $1 million. The portfolio may never recover even if later returns are strong.
The numbers
Consider two retirements, each starting with $1 million, withdrawing $40,000/year adjusted for inflation, over 30 years. Both have an average return of 7%—but different sequences:
- Good sequence: strong returns early, weak returns late. Portfolio survives 30+ years.
- Bad sequence: weak returns early (e.g., 2000–2002 or 2007–2009), strong returns late. Portfolio depletes in 15–20 years.
The difference is dramatic. Retiring in 2000 (just before the dot-com crash) versus 2002 (after the crash) produces vastly different outcomes, even with identical long-term averages.
Mitigation strategies
- Hold 2–3 years of expenses in cash or short-term bonds, so you don't have to sell stocks during a downturn.
- Reduce withdrawal rate during market downturns—4% becomes 3% or 2% when portfolio drops significantly.
- Use a bond tent: increase bond allocation in the years just before and after retirement, then gradually shift back to stocks.
- Consider annuitizing part of your portfolio to lock in lifetime income and reduce sequence risk on the remainder.
- Be flexible on retirement timing: working 1–2 more years through a market crash can dramatically improve outcomes.
Practical application
When evaluating any investment, ask which return metric is being quoted and whether it's appropriate for your situation:
- Mutual fund performance numbers are typically TWR (CAGR for specific periods). Use these to compare funds.
- Your own portfolio performance should be calculated as XIRR (money-weighted). This tells you what you actually earned.
- For long-term planning, use real (inflation-adjusted) returns. A 7% real return assumption is reasonable for a stock-heavy portfolio; 4–5% real for a balanced portfolio.
- For retirement income planning, model sequence risk—test your plan against historical bad sequences like 1966, 1973, 2000, and 2007 retirement cohorts.
Common return calculation mistakes
- Using arithmetic average instead of geometric. "Average return of 12%" usually means arithmetic mean; the actual compound return was lower.
- Ignoring dividends. Price return understates total return by 1–3 percentage points per year for stock funds.
- Comparing nominal to real returns. A 7% nominal return is worse than a 5% real return at 3% inflation.
- Using CAGR for portfolios with contributions. CAGR overstates returns when you added money over time. Use XIRR instead.
- Cherry-picking time periods. "10% per year since 2010" is true but includes one of the strongest bull markets in history. Look at 20+ year periods.
- Comparing to the wrong benchmark. A diversified portfolio shouldn't be judged against the S&P 500 alone.
- Ignoring taxes and fees. A 10% gross return becomes 7–8% after a 1% expense ratio and 20% tax drag.
Putting it into practice
The honest way to evaluate investment performance: calculate your XIRR using actual cash flow dates, compare to a benchmark that matches your asset allocation, look at multi-year periods (5–10 years minimum), think in real terms, and account for fees and taxes. Single-year returns are noise; long-term compound growth is signal. The investor who consistently captures 7–8% real returns over 30 years will outperform the one chasing 15% nominal returns in any given year.
To calculate your portfolio's CAGR, IRR, and total return with real-world cash flows, try our Investment Return Calculator. Enter your contributions, withdrawals, and ending value with dates, and it computes both money-weighted (XIRR) and time-weighted returns—plus benchmarks for comparison.