Skip to content
Menu
Finabase Learn

Funds · Costs and choosing a fund

How to read fund performance: standardized and after-tax returns

What SEC-standardized fund returns mean, how loads and taxes change them, why the end date moves a trailing return, and how investor returns and dead funds differ.

On this page
  1. The standardized return
  2. Load-adjusted returns
  3. After-tax returns
  4. Net asset value return and market price return
  5. Calendar-year and trailing returns
  6. The fund's return and the investor's return
  7. Survivorship bias
  8. A reading order
  9. Where it fails or misleads
  10. How Finabase does it
  11. Frequently asked questions
  12. Summary
  13. Check your understanding

A fund's page can show a dozen different returns for the same fund: one year, five years, ten years, year to date, at net asset value, at market price, with the sales charge, after taxes, each calendar year. They are all correct. They answer different questions, and the one a reader notices first is rarely the one that applies to them.

US rules standardize how some of these numbers are calculated, so that one fund's ten-year return can be set beside another's. This lesson explains what the standardized figures are, what loads and taxes do to them, why a trailing return depends heavily on the day it ends, and two ways a fund's record differs from what investors in funds actually earned.

The standardized return

Form N-1A is the SEC form on which US mutual funds and ETFs register, and its instructions define the performance figures a prospectus must carry (the version cited here is dated December 2023). Item 4 requires two things directly after the fee table:

  • A bar chart of the fund's total return for each of the last ten calendar years, or for its life if shorter, with its best and worst calendar quarter stated beneath.
  • A table of average annual total returns for the 1-, 5- and 10-calendar-year periods ended at the last year-end, beside the returns of an appropriate broad-based securities market index.

Rule 482 under the Securities Act applies the same idea to advertisements and fact sheets: if performance is shown, it must include average annual total return for one, five and ten years, current to the most recent calendar quarter.

Four features of that definition are worth stating.

  1. It is a compound rate. T is the single yearly rate that turns the starting amount into the ending amount. It is not the average of the yearly returns, which is always higher when returns vary (CAGR: compound annual growth rate, with the formula and its limits).
  2. It is a total return. Dividends and capital-gain distributions are assumed to be reinvested on the day they are paid (Total return).
  3. It is net of the fund's expenses. The expense ratio and the fund's trading costs are already out. Nothing more should be subtracted for them.
  4. It is net of the maximum sales load, in the table. The bar chart is different: Form N-1A says sales loads are not reflected in the bar chart and requires a statement that returns would be lower if they were.

The Vanguard S&P 500 ETF's summary prospectus of April 28, 2026 shows the form filled in. Its bar chart lists ten calendar years, from 11.93% in 2016 to 17.84% in 2025, with 2018 at −4.42% and 2022 at −18.15%. The best quarter was 20.54% (ended June 30, 2020) and the worst −19.63% (ended March 31, 2020): the fund's worst and best quarters of the decade were consecutive.

Load-adjusted returns

A front-end sales charge is taken from the money before it is invested. A 5.75% load means $942.50 of each $1,000 goes into the fund (Fund share classes and loads: Class A, C, I and what each one costs). The standardized table shows the return on the full $1,000.

The Growth Fund of America's summary prospectus of November 1, 2024 gives Class A returns with the maximum 5.75% sales charge for periods ended December 31, 2023:

Period With maximum sales charge Same return without the charge Effect of the load
1 year 29.31% 37.20% 7.89 points
10 years 11.19% a year 11.85% a year 0.66 points a year

The middle column is computed here, not printed in the prospectus: 1.2931 ÷ 0.9425 = 1.3720. The load is paid once, so over one year it removes nearly eight points and over ten years two thirds of a point a year. The fund's bar chart for 2023 shows the return without the load, which is why the chart and the table of the same document disagree for the same year.

When comparing a load fund with a no-load fund, use the load-adjusted figure if you would pay the load and the unadjusted one if you would not. Many investors do not: retirement plans and fee-based accounts generally buy at net asset value.

After-tax returns

The prospectus table carries three lines for one share class:

Line What is taken out
Return before taxes Nothing beyond the fund's costs and any load
Return after taxes on distributions Federal tax on every dividend and capital-gain distribution, in the year paid. Shares still held
Return after taxes on distributions and sale of shares The above, plus tax on the investor's own gain when all shares are sold at the end of the period

Form N-1A specifies the assumptions: the highest individual federal marginal income tax rates in force at the time of each distribution, no state or local tax, and a statement that the figures are not relevant to investors who hold shares through tax-deferred arrangements such as 401(k) plans or IRAs.

Fund, ten years ended Before taxes After taxes on distributions After taxes on distributions and sale
Vanguard S&P 500 ETF (VOO), Dec. 31, 2025 14.78% 14.32% 12.34%
The Growth Fund of America, Class F-2, Dec. 31, 2023 12.09% 10.14% 9.45%

Read the middle column as the cost of holding the fund in a taxable account. For the index ETF, whose distributions are essentially the index's dividends, tax on them cost 0.46 points a year. For the actively managed fund, which also distributed realized capital gains most years, it cost 1.95 points a year. The periods are two years apart, so the two are comparable in size and not to the decimal.

The third column is mostly about the investor, not the fund. It taxes the whole gain at the end, so a fund that has risen a great deal shows a large reduction. For a one-year period it can look odd: VOO's one-year return to December 31, 2025 was 17.84% before taxes and 10.77% after taxes on distributions and sale, because selling after one year's 17.84% gain makes the whole gain taxable at once.

Net asset value return and market price return

ETF documents show two returns. NAV return is the portfolio's result. Market price return is calculated from exchange closing prices and so includes any change in the premium or discount between the first and last day (Premiums and discounts in ETFs: why price and NAV disagree).

For large funds the two are close. VOO's one-year return to December 31, 2025 was 17.84% at NAV and 17.82% at market price, and the ten-year figures were both 14.78%. The iShares Core S&P 500 ETF's sheet for June 30, 2026 shows a wider one-year gap, 22.29% at NAV and 22.05% at market price, shrinking to 15.47% and 15.44% over ten years. A premium or discount of a few hundredths of a percent at either end is a noticeable part of one year's return and a negligible part of ten.

Judge the fund on NAV return against its benchmark. Use market price return to see what a shareholder trading on those exact dates received.

Calendar-year and trailing returns

A calendar-year return covers January to December of one year. A trailing return covers a period ending on the latest reporting date: the last 1, 3, 5 or 10 years.

Trailing returns compress a history into one number, and all of them end on the same day. That makes them sensitive to two months: the one that has just been added and the one that has just dropped out of the window.

Trailing five-year return, annualized, by the month it ends -5% 0% 5% 10% 15% 20% 25% SPY (S&P 500) IWM (Russell 2000) 2005 2010 2015 2020 2025
Each point is the annualized total return of the five years ending in that month. Computed by Finabase from monthly prices with distributions reinvested.
Five years ending SPY, annualized IWM, annualized
February 2014 22.85% 26.43%
March 2014 21.10% 24.36%
March 2020 6.65% −0.28%
December 2021 18.34% 11.90%
December 2022 9.31% 4.03%
February 2025 16.68% 9.36%
March 2025 18.48% 13.17%
October 2026 11.96% 5.44%

Three things in this table recur in every fund's record.

  • The same fund, a different answer each year. SPY's five-year return was 18.34% a year at the end of 2021 and 9.31% a year later. The fund did not change. One year of −18% entered the window and one good year left it.
  • Dropping a bad month raises the figure. SPY's five-year return rose from 16.68% in February 2025 to 18.48% in March 2025, in a month when the market itself did not rise by anything like that: the window had just lost March 2020. The largest such jump in the record is 5.08 points in October 2013, when October 2008's fall of 16.52% left the window.
  • The window can begin at a low or a high. The 22.85% to February 2014 starts at the bottom of the 2007–2009 decline. The −6.67% to February 2009 ends there. Neither describes what the fund typically does.

Calendar-year returns are the remedy. A row of ten separate years shows the bad years that a trailing average hides, and no single year is counted in several figures at once. Reading both, and noting the start date of any chart, takes a minute and removes most of the distortion.

The fund's return and the investor's return

Every figure so far is a time-weighted return: it follows one payment made at the start and left alone. It measures the fund. Investors do not behave that way. They add money, often after a strong run, and take it out, often after a fall. The return on the dollars actually invested is a dollar-weighted return, the internal rate of return of the cash flows.

A worked example with a real fund shows the difference. ARK Innovation ETF (ARKK) is an actively managed fund that rose 152.8% in 2020 and fell 23.4% in 2021 and 67.0% in 2022. From the end of December 2016 to the end of October 2026 its time-weighted total return was 16.03% a year, by Finabase's calculation from monthly prices and distributions.

  • An investor who put in $10,000 at the end of 2016 and nothing more had $43,143 at the end of October 2026: 16.03% a year, the fund's return.
  • An investor who did the same and then added another $10,000 at the end of 2020, after the best year, had $49,743. The second $10,000 was worth $6,600. The dollar-weighted return on the two payments together is 11.97% a year.

Same fund, same dates, four points a year apart, and the difference is entirely timing.

Across all funds the effect is smaller but consistent. Morningstar's annual study, Mind the Gap 2026, estimated that over the ten years ended December 31, 2025 the average dollar in US mutual funds and ETFs earned 8.7% a year while the funds' aggregate total return was 9.9%: a gap of about 1.2 points a year. For US stock funds the figures were 12.8% and 13.3%.

What funds returned and what the average dollar in them earned, percent a year Fund total return Investor (dollar-weighted) return 9.9% 8.7% All US funds and ETFs 13.3% 12.8% US stock funds Ten years ended Dec. 31, 2025. Source: Morningstar, Mind the Gap 2026 (Aug. 6, 2026).
Total return of funds and the dollar-weighted return of the money in them. Source: Morningstar, Mind the Gap 2026.

Survivorship bias

A list of funds available today contains only funds that still exist. Funds with poor records tend to be closed or merged into others, and their histories go with them. Any average taken over today's funds is therefore an average of survivors, and it flatters the group (Survivorship bias).

S&P Dow Jones Indices' SPIVA scorecard counts every fund that existed at the start of each period for this reason. Its Mid-Year 2025 edition, with data to June 30, 2025, reported:

Period to June 30, 2025 US stock funds still existing Large-cap funds still existing Large-cap funds that trailed the S&P 500
1 year 96.60% 97.42% 72.61%
5 years 85.19% 87.60% 86.91%
10 years 63.81% 65.52% 85.98%
20 years 37.29% 34.70% 91.03%

Of 2,373 US stock funds at the start of the twenty years, fewer than four in ten were still operating at the end. A screen of twenty-year records run today can only show those survivors.

The same bias applies inside one fund family's marketing. A company with thirty funds will advertise the handful with the best recent numbers. That is selection, not evidence about the next thirty.

A reading order

  1. Find the date the figures end on, and whether they are NAV or market price returns.
  2. Read the ten calendar years before the trailing averages. Note the worst year.
  3. Compare the fund's return with its benchmark's over the same period, not with other funds over different periods.
  4. If you would pay a sales charge, use the load-adjusted row.
  5. If the account is taxable, read "after taxes on distributions".
  6. Ask what your own return would have been, given when you would actually have bought.

Where it fails or misleads

  • Past performance does not predict. Rule 482 requires every advertisement to say that past performance does not guarantee future results and that current performance may be lower or higher. The SEC's own investor guidance says to look at more than past performance. The requirement exists because the evidence supports it.
  • Ten years can be one kind of market. The decade to 2025 contains two sharp falls inside a long rise. A fund that looks low-risk on that record may never have been tested.
  • The benchmark may be the wrong one. The rules require a broad-based market index, which may not resemble what the fund holds (Choosing a benchmark: what to compare a strategy against).
  • Inception-date returns are not comparable. "Since inception" for a fund launched at a market low and for one launched at a high measure different things.
  • A new share class can borrow an old record. The performance shown for a recently created class is often that of an older class of the same fund, adjusted or not for the difference in fees. The footnotes say which.

How Finabase does it

Finabase does not publish standardized, load-adjusted or after-tax returns. Those are the issuer's figures and are in the prospectus and fact sheet.

A fund's symbol page has its quote, chart and news, and the chart can be set to any range, which makes end-date sensitivity easy to see for yourself: move the start by a year and watch the percentage change. The comparison tab rebases several funds to percent change from the first bar in view and tabulates price returns from one week to two years. Those are price returns, with dividends left out, so they sit below a fund's total return by roughly its yield, and a mutual fund that pays a large year-end distribution shows a drop on that day that is not a loss. The Dividends & splits tab lists each distribution so the missing part can be seen. Portfolios record your own purchases and dividends, which is the starting point for your own, dollar-weighted, result.

Frequently asked questions

What is a standardized return?

It is the average annual total return calculated the way SEC Form N-1A prescribes: a compound yearly rate for 1, 5 and 10 years, with distributions reinvested, after fund expenses and the maximum sales charge. Every US fund must show it, so funds can be compared on one basis.

Why is the after-tax return lower than the pre-tax return?

It subtracts federal tax on the fund's distributions at the highest individual rates, and in the third line also the tax on selling the shares at the end. In a tax-deferred account the pre-tax figure is the relevant one.

What is the difference between NAV return and market return for an ETF?

NAV return measures the fund's portfolio. Market return measures the exchange price, so it also picks up any small premium or discount at the start and end. Over long periods the two are nearly the same for large funds.

Why did my return differ from the fund's published return?

The published figure assumes one investment at the start of the period, held to the end, with everything reinvested. If you bought at other times, added or withdrew money, paid a sales charge or tax, or did not reinvest, your result differs.

Are 10-year returns better evidence than 1-year returns?

They contain more information, but they are still one sample with one start and one end. Read them with the individual calendar years and the benchmark's return for the same period.

Summary

The standardized return is a compound annual total return for 1, 5 and 10 years, after expenses and the maximum sales load. A 5.75% load turned a 37.20% year into 29.31% and cost 0.66 points a year over ten. After-tax lines assume top federal rates: tax on distributions cost 0.46 points a year for an S&P 500 ETF and 1.95 for a large active fund. ETF market returns differ from NAV returns by the premium or discount at the two ends. Trailing returns hang on their first and last months: SPY's five-year figure has ranged from −6.67% to 22.85% a year. And the fund's return is not the investor's: timing cost the average fund dollar about 1.2 points a year over the decade to 2025, and fewer than four in ten US stock funds lasted twenty years.

Check your understanding

5 questions on this lesson. Each answer comes with the reason, and nothing is scored.

  1. A fund with a 5.75% front-end sales charge shows a one-year standardized return of 29.31%. What did the fund’s portfolio itself return that year?

    1. 23.56%
    2. 37.20%
    3. 35.06%
    Show the answer

    B. 37.20% Only $942.50 of each $1,000 was invested. $1,293.10 ÷ $942.50 = 1.3720, a 37.20% return. The load is divided out, not added: 29.31 + 5.75 is not the answer.

  2. A fund’s five-year trailing return jumps from 16.7% to 18.5% a year in a month when the market was roughly flat. What is the most likely reason?

    1. A very bad month from five years earlier has just dropped out of the window
    2. The fund cut its expense ratio
    3. The fund paid a large dividend
    Show the answer

    A. A very bad month from five years earlier has just dropped out of the window A trailing return covers a fixed-length window that moves forward each month. When a sharp fall leaves the window, the annualized figure rises even if nothing happened this month.

  3. A prospectus shows 14.78% before taxes and 14.32% after taxes on distributions. Who should use the 14.32% figure?

    1. Every shareholder
    2. An investor holding the fund in an IRA
    3. An investor in the highest federal bracket holding the fund in a taxable account
    Show the answer

    C. An investor in the highest federal bracket holding the fund in a taxable account The after-tax lines assume the highest individual federal rates and no state tax, and the form requires a statement that they are not relevant to tax-deferred accounts.

  4. A fund returned 16% a year over ten years. An investor put in $10,000 at the start and another $10,000 just after the fund’s best year, shortly before a large fall. How does their dollar-weighted return compare with 16%?

    1. It is the same, because the fund’s return is the fund’s return
    2. It is lower, because more of their money was invested for the bad years than for the good ones
    3. It is higher, because they invested more
    Show the answer

    B. It is lower, because more of their money was invested for the bad years than for the good ones The fund’s published return is time-weighted: one payment held throughout. A dollar-weighted return gives more weight to periods when more money was invested, so adding before a fall lowers it.

  5. Of the US stock funds that existed twenty years before June 30, 2025, about 37% were still operating at the end. Why does that matter when reading fund averages?

    1. It does not: closed funds returned their money
    2. Averages over today’s funds leave out the closed ones, which tended to have weaker records, so the group looks better than it was
    3. It means the surviving funds are certain to keep outperforming
    Show the answer

    B. Averages over today’s funds leave out the closed ones, which tended to have weaker records, so the group looks better than it was This is survivorship bias. A study that counts every fund alive at the start, as the SPIVA scorecard does, avoids it.

Sources and further reading

  1. SEC — Form N-1A (SEC 2052, 12/23): Item 4(b)(2) (bar chart and performance table, after-tax returns) and Item 26(b) (average annual total return formula)
  2. SEC — Rule 482 under the Securities Act (17 CFR 230.482): performance in fund advertisements
  3. SEC — Mutual Fund Investing: Look at More Than a Fund’s Past Performance
  4. Vanguard — Vanguard S&P 500 ETF summary prospectus, April 28, 2026 (annual returns, best and worst quarter, returns before and after taxes)
  5. Capital Group — The Growth Fund of America summary prospectus, November 1, 2024 (Class A with sales charge; Class F-2 before and after taxes)
  6. iShares — iShares Core S&P 500 ETF fact sheet as of June 30, 2026 (NAV and market price returns)
  7. Morningstar — Mind the Gap 2026 (dated August 5, 2026): investor returns and total returns, ten years ended Dec. 31, 2025
  8. S&P Dow Jones Indices — SPIVA U.S. Scorecard, Mid-Year 2025 (data to June 30, 2025): underperformance and survivorship
  9. Finabase chart data — monthly bars and distributions of SPY, IWM and ARKK to October 2026; returns computed by Finabase