Real Return Calculator
Compute real (inflation-adjusted) investment returns via the Fisher equation. See how much purchasing power actually grows vs nominal illusion.
Real Return Calculator
Future values are before tax. "Real" values are in today's money; "break-even" is the nominal balance needed just to keep pace with inflation; "inflation illusion" is the part of the nominal balance that only keeps pace with prices.
Year-by-year projection
Your return at other inflation rates
What US assets actually returned
Geometric annual averages, 1928–2025, from Damodaran's US returns (NYU Stern, Jan 2026), deflated by BLS CPI-U, which averaged 3.0% a year. History, not a forecast.
How to use the Real Return Calculator
Enter your investment + horizon
The principal you're starting with, and how many years it'll compound. Longer horizons make the difference between real and nominal returns much more dramatic.
Set the nominal return
This is the headline percentage, before tax. From 1928 to 2025 the S&P 500 returned 10.0% a year, 10-year Treasuries 4.5% and 3-month Treasury bills 3.4% (Damodaran's US data, geometric averages). Use a realistic forward-looking number, not last year's result.
Set the inflation rate
US consumer prices (CPI-U) rose 3.0% a year from December 1927 to December 2025, and 3.7% in the twelve months to September 2026. The US Federal Reserve aims for 2% over the longer run, measured by the PCE price index rather than CPI. Over 1995–2024 the World Bank's consumer-price figures average 1.7% a year for Singapore, 2.3% for Malaysia, 2.5% for Thailand, 4.4% for the Philippines, 5.6% for Vietnam and 8.3% for Indonesia; Singapore's CPI was 2.3% higher in August 2026 than a year earlier (SingStat). The results table shows your return at inflation two points either side of what you enter.
Read both Fisher exact + approximation
Fisher exact uses (1+nom)/(1+infl)−1. The approximation (nominal − inflation) is close but overstates the real return when both rates are positive. At 8% nominal + 3% inflation, exact = 4.85%, approx = 5%. The gap grows with the rates. If the return is taxed each year, enter your tax rate: tax falls on the whole nominal return, so 20% tax on 8% at 3% inflation leaves 3.30% real, not 3.88%.
Real return — the only return that matters for purchasing power
Nominal return is the percentage your investment grows in dollar terms. Real return is what's left after inflation erases part of that growth. Only real return tells you how much more you can buy at the end than at the start — which is what investing for the long term is actually about. A 10% nominal return in 8% inflation is 1.85% real. A 5% nominal return in 1% inflation is 3.96% real — more than twice the purchasing-power gain despite the lower headline number.
The Fisher equation
Irving Fisher set out the translation from a money rate of interest to a real one in The Theory of Interest (1930): (1 + real) = (1 + nominal) / (1 + inflation), solved for real = ((1 + nom) / (1 + infl)) − 1. The common shorthand "real = nominal − inflation" is an approximation that works well at low rates (under 5%) but drifts at higher rates. At 10% nominal + 8% inflation, the approximation gives 2%; Fisher exact gives 1.85% — small at first glance, but compounded over 30 years, it's about a 4½% difference in final purchasing power (1.02^30 = 1.81 against 1.0185^30 = 1.73). For ASEAN-stage investors where both rates run higher, always use the exact equation.
Subtracting inflation from a return is a shortcut; dividing by it is the answer. The gap between the two grows when both numbers are large.
The APAC investor's real-return reality
Inflation differs sharply across the region — World Bank consumer-price data for 1995–2024 average 1.7% a year in Singapore, 2.3% in Malaysia, 5.6% in Vietnam and 8.3% in Indonesia — so the same nominal market return means very different real returns. A 10% year in Jakarta at Indonesia's long-run inflation is 1.6% real; a 6% year in Singapore at Singapore's is 4.2% real. Vietnam's inflation reached 23.1% in 2008 (World Bank), which is exactly when a nominal figure misleads most. Across the region the message is the same: judge investments by real returns, not nominal — especially when inflation is volatile.
Why this matters for retirement
Retirement can last 25–40 years, and a spending plan in today's money has to be compared with returns in today's money. A portfolio earning 0% real (3% nominal at 3% inflation) loses purchasing power on every withdrawal. Even a healthy average real return is not a safe withdrawal rate: William Bengen's 1994 study, which produced the 4% rule, began from a planner who assumed a 60/40 portfolio's average real return of almost 5.1% and showed why the order of bad years, not the average, sets what can safely be drawn. A model that mixes nominal returns with today's-dollar spending overstates how far the money goes.
Fisher's equation, a century of US real returns, and when high yields were illusory
From 1928 to 2025 the S&P 500 returned 10.0% a year while US consumer prices rose 3.0% a year. Subtracting gives 7.0%; the Fisher equation gives the true 6.8% real.
The Fisher equation is exact: (1+real) = (1+nom) / (1+infl). The "real = nom − infl" shorthand is only accurate at low rates.
Three-month US Treasury bills returned 0.3% a year real from 1928 to 2025. Their job is preservation, not growth — on average they only just kept pace with inflation.
The 4% rule is a real-terms rule: Bengen (1994) tested a first-year withdrawal of 4% followed by inflation-adjusted withdrawals, and in no past case did it exhaust a portfolio in under 33 years.
US real returns 1928–2025 (Damodaran's annual returns, deflated by CPI-U): stocks ~6.8%, 10-year Treasuries ~1.5%, T-bills ~0.3% a year. That's the canonical case for equity over fixed income for long-horizon savers.
In 1980 the 10-year Treasury averaged 11.5% while CPI rose 13.5% — a negative real yield. In 1984 it averaged 12.4% with inflation at 4.3%, a real yield of about 7.8%: the same "high-yield" era was illusory one year and exceptional a few years later.
Emerging-market stocks can post exciting nominal returns in high-inflation years — deflate them by local inflation before comparing them with developed markets.
TIPS (Treasury Inflation-Protected Securities) pay a fixed rate of interest on a principal that is adjusted with the CPI, so the fixed rate works like a real yield — though US federal tax can fall on the principal adjustment as well as the interest.
Mortgage debt has a negative real cost when inflation exceeds your fixed mortgage rate — you pay back in cheaper future dollars. A 3% fixed mortgage in 5% inflation has a real cost of 1.03 ÷ 1.05 − 1 = −1.9%.
In 2022, US CPI inflation peaked at 9.1% year-on-year (June) — any savings account or bond paying less than that lost purchasing power even as its balance rose.
Frequently Asked Questions
-
Fisher exact: (1+nom)/(1+infl) − 1. Approximation: nom − infl. At low rates the two are nearly identical. At 5% nom + 3% infl, exact = 1.94%, approx = 2% — close. At 15% nom + 10% infl, exact = 4.55%, approx = 5% — meaningful difference. Always use exact for serious financial planning.
-
Because the nominal gain itself gets inflation-eroded. If you grow $100 to $110 (10% gain) but prices also rose 8%, the $110 only buys what $101.85 bought before. You gained ~1.85% in purchasing power, not the headline 10%. The exact formula captures this compounding interaction; the approximation ignores it.
-
For long-term planning (retirement, FIRE, college savings): yes, always real returns. They represent actual purchasing power. For short-term cash flow modelling (next 1-2 years): nominal is fine. The longer the horizon, the more critical real returns become — over 30 years, even 2% inflation cuts purchasing power by about 45% (1 ÷ 1.02^30 = 0.55).
-
The 4% rule comes from William Bengen's 1994 study: a first-year withdrawal of 4% of the portfolio, raised with inflation every year after, never exhausted a portfolio in under 33 years in the US history he tested. Because the withdrawal rises with prices, it is a real-terms rule. It sits well below a balanced portfolio's average real return — Bengen's own example put a 60/40 mix at almost 5.1% real — because the worst sequences of returns, not the averages, set the safe rate. A plan that projects nominal returns against today's-money spending overstates what can be withdrawn.
-
The honest answer is "we don't know." US consumer prices rose 3.0% a year on average from 1927 to 2025; the Federal Reserve targets 2% (on its PCE measure); and US CPI inflation reached 9.1% in the year to June 2022. Run your numbers at more than one assumption — the results table does two points either side — and accept that the future will surprise you in both directions.
-
Yes, when inflation exceeds nominal return. Cash in a savings account at 0.5% in 5% inflation has a −4.3% real return — losing purchasing power every year despite the nominal number going up. US Treasury bills returned 2.1% in 2022 while consumer prices rose 6.5% over the year — a negative real return, as through much of the low-rate 2010s. Negative real returns are why "safe" assets aren't safe in high-inflation regimes.
-
Yes. Real return math is dimensionless. Enter the nominal return in any currency, the inflation rate for that currency's country, and the formula gives correct real return. SGD + Singapore CPI, MYR + Malaysian CPI, JPY + Japanese CPI, etc.
-
Yes — the real after-tax return is the most honest number, and the order matters. Tax falls on the whole nominal return, including the part that only compensates for inflation, so take tax off the nominal return first and then apply Fisher: (1 + nominal × (1 − tax)) ÷ (1 + inflation) − 1. Applying the tax rate to the real return understates the tax: at 8% nominal, 3% inflation and 20% tax that shortcut gives 3.88%, the correct figure is 3.30%. Enter your tax rate in the optional box. For a tax-free account, leave it blank.
-
History gives the scale rather than a target. From 1928 to 2025 US stocks returned 6.8% a year real, 10-year Treasuries 1.5% and Treasury bills 0.3% (Damodaran's data deflated by CPI-U). Bengen's 1994 example put a 60/40 stock-bond mix at almost 5.1% real. A portfolio expecting much more than the stock-market figure is assuming something history did not deliver on average.
-
No. The calculation runs entirely in your browser; the numbers you type are not sent to us or anyone else.
Method & sources
How it computes
Fisher: real = (1 + nominal) ÷ (1 + inflation) − 1, shown beside the shortcut nominal − inflation. After tax (tax on each year's nominal return): (1 + nominal × (1 − tax)) ÷ (1 + inflation) − 1. Projection: nominal = principal × (1 + after-tax nominal)^years; real = that ÷ (1 + inflation)^years. History rows: Damodaran US returns 1928-2025, annualised and deflated by BLS CPI-U (RealReturnKernel).
What this tool implements
- Fisher relation between money and real interest (Fisher, The Theory of Interest, 1930, ch. II): the exact quotient is the answer and the subtraction is labelled as the approximation, which overstates the real return whenever both rates are positive.
- Tax is taken off the NOMINAL return before deflating, because tax falls on the part of the return that only offsets inflation too: 20% tax on 8% at 3% inflation leaves 3.30% real, not the 3.88% that taxing the real return gives.
- US history is computed, not typed: $100 at the start of 1928 grew to $1,157,598.95 (S&P 500), $7,752.88 (10-year Treasuries) and $2,578.30 (3-month bills) by end-2025 in Damodaran's data; CPI-U went from 17.3 (Dec 1927) to 324.054 (Dec 2025). That is 10.0%, 4.5% and 3.4% nominal, 3.0% inflation and 6.8%, 1.5% and 0.3% real a year.
- Dated figures on the page, each re-checked at source on 2026-10-17: GS10 averaged 11.46% in 1980 and 12.44% in 1984 against CPI-U annual-average inflation of 13.5% and 4.3%; CPI-U rose 9.1% in the year to June 2022 and 3.7% in the year to September 2026; World Bank CPI means 1995-2024 of 1.7% (Singapore), 2.3% (Malaysia), 2.5% (Thailand), 4.4% (Philippines), 5.6% (Vietnam, 1996-2024) and 8.3% (Indonesia), with Vietnam at 23.1% in 2008; Singapore CPI 2.3% higher in August 2026 than a year earlier (SingStat M213751).
- The verdict compares the reader's real return with cash held at 0% over the same inflation; a positive nominal return always beats it, so the page never calls one "worse than cash".
- Corrected 2026-10-17: blank years read as 10 and blank rates as 0; "1,000" read as 1; "worse than cash under the mattress" for any real return below −2%; unsourced asset-class scenario rows; no after-tax figure and FAQ advice that understated tax; the subtraction shortcut used in the intro (2%) and fact 01 (7%); −2% for a 3% mortgage in 5% inflation (−1.9%); the 4% rule misattributed and its FAQ inverted; unsourced VN-Index and PSEi claims; two unclosed divs.
Sources
- Damodaran A. Historical Returns on Stocks, Bonds and Bills: 1928-2025. NYU Stern (updated 5 January 2026). https://pages.stern.nyu.edu/~adamodar/New_Home_Page/datafile/hist…
- U.S. Bureau of Labor Statistics. CPI for All Urban Consumers (CPI-U), U.S. city average, all items, series CUUR0000SA0 (BLS Public Data API). https://api.bls.gov/publicAPI/v2/timeseries/data/CUUR0000SA0
- Federal Reserve Bank of St. Louis. Market yield on U.S. Treasury securities at 10-year constant maturity (GS10). https://fred.stlouisfed.org/data/GS10.txt
- World Bank. Inflation, consumer prices (annual %), FP.CPI.TOTL.ZG (updated 8 October 2026). https://api.worldbank.org/v2/country/SGP;MYS;VNM;IDN;PHL;THA/indi…
- Singapore Department of Statistics. Consumer Price Index (CPI), 2024 as base year, monthly (M213751). https://tablebuilder.singstat.gov.sg/api/table/tabledata/M213751
- Fisher I. The Theory of Interest. New York: Macmillan; 1930 (FRASER). https://fraser.stlouisfed.org/files/docs/publications/books/theor…
- Bengen WP. Determining Withdrawal Rates Using Historical Data. Journal of Financial Planning, October 1994 (FPA reprint, March 2004). https://www.financialplanningassociation.org/sites/default/files/…
- TreasuryDirect. Treasury Inflation-Protected Securities (TIPS). https://www.treasurydirect.gov/marketable-securities/tips/
What can make this go out of date
- Nothing is fetched at runtime. The history rows are constants from Damodaran's January 2026 file and BLS CPI-U; they need refreshing when Damodaran adds 2026 (expected January 2027), and the dated inflation hints when they age.
Abridged — the full review record for this tool runs longer than the list above.
Pick up where you left off
Stored only in this browser — never sent to our servers.