Bengen for retirees
Bengen's retiree portfolio (7 asset classes, 55 % stocks)
Where it stands today
Our calculation, 1926-07 – 2026-09
Growth of $10,000
Where it came from
William Bengen (born 1947) studied aeronautics at MIT, ran a family soft-drink bottling business until 1987, became a Certified Financial Planner in 1990 and ran a fee-only practice in Southern California until he retired in 2013. In October 1994 he published 'Determining Withdrawal Rates Using Historical Data' in the Journal of Financial Planning. Using US data from 1926, he asked what first-year withdrawal rate, then adjusted for inflation, would never have run out of money in 30 years from any starting date. The answer, about 4 %, became known as the 4 % rule.
The 1994 portfolio was simple: 50 % large-cap US stocks and 50 % intermediate-term US Treasury bonds, rebalanced annually. A 2023 replication in the Journal of Financial Planning reports Bengen's 4.0 % for this mix on 1926-1992 data. Bengen recommended holding between 50 % and 75 % in stocks, as close to 75 % as possible. He later named the lowest sustainable rate 'SAFEMAX'.
In a 1997 Journal of Financial Planning paper he added small-cap stocks (30 % of the stock part according to the replication) and raised the figure to about 4.3 %; his 2006 book 'Conserving Client Portfolios During Retirement' (FPA Press) states a SAFEMAX of about 4.5 % for tax-free accounts and about 4.1 % for taxable ones. A 2018 interview describes the small-cap addition as lifting the rate from about 4.2 % to about 4.5 %, with the worst-case retiree being one who started in 1968.
In later work (articles in FA Magazine from 2024 and the interviews of 2025) Bengen uses seven asset classes: US large-cap, mid-cap, small-cap and micro-cap stocks, international stocks, intermediate-term US government bonds and Treasury bills. His 'standard' allocation is 55 % stocks, 40 % intermediate bonds and 5 % bills, with the five stock classes equally weighted at 11 % each (FA Magazine, May 2024, Table 2). With this mix he put the universal SAFEMAX for a 30-year retirement at 4.7 %, based on about 400 quarterly start dates from 1926 to 2024; only the retiree who started in October 1968 could not have withdrawn more.
Bengen stresses that the standard mix is a reference. In his own tests the optimal allocation for an individual start date was almost never 55/45 and the best mix often used only one or two stock classes (small and micro caps). The static 55/40/5 portfolio is the one used here because it is the stated baseline for the 4.7 % figure.
What is in it
| US large-cap stocks | 11% | FA Magazine May 2024, Table 2 'standard allocation' |
| US small-cap stocks | 11% | |
| US mid-cap stocks | 11% | |
| US micro-cap stocks | 11% | |
| International stocks | 11% | |
| Intermediate-term US government bonds | 40% | |
| US Treasury bills | 5% | kept constant at 5 % in all of his tests |
How we model it
| Intermediate government bonds (10 y) | 40% |
| US large companies (S&P 500) | 22% |
| US small companies | 22% |
| Developed markets outside the US | 11% |
| Cash (Treasury bills) | 5% |
For a euro investor
How to build it with European funds
| Part | Share | Example funds (ISIN) |
|---|---|---|
| Intermediate government bonds (10 y) | 40% | iShares USD Treasury Bond 7-10yr UCITS ETF (Dist) IE00B1FZS798 |
| US large companies (S&P 500) | 22% | iShares Core S&P 500 UCITS ETF USD (Acc) IE00B5BMR087Vanguard S&P 500 UCITS ETF (USD) Distributing IE00B3XXRP09 |
| US small companies | 22% | SPDR Russell 2000 U.S. Small Cap UCITS ETF IE00BJ38QD84iShares MSCI USA Small Cap ESG Enhanced CTB UCITS ETF USD (Acc) IE00B3VWM098 |
| Developed markets outside the US (closest match) | 11% | iShares Core MSCI Europe UCITS ETF EUR (Acc) IE00B4K48X80iShares Core MSCI Japan IMI UCITS ETF IE00B4L5YX21 |
| Cash (Treasury bills) | 5% | Xtrackers II EUR Overnight Rate Swap UCITS ETF 1C LU0290358497iShares USD Treasury Bond 0-1yr UCITS ETF (Acc) IE00BGSF1X88 |
The idea
A retiree withdrawing a rising amount each year needs growth to outpace inflation, so about 55 % in stocks is the balance Bengen finds best for the worst historical start dates; diversifying the stock part into small, mid, micro and foreign shares raises the lowest sustainable withdrawal rate from about 4.1 % to about 4.7 %. The 5 % in bills is a cash reserve.
How people buy it
Not a product. In practice: US large, mid, small and micro-cap index funds, an international index fund, an intermediate Treasury fund and a money-market or T-bill fund.
Rebalancing as published
The 1994 study rebalanced annually. For the 2020s work the source read does not state the rebalancing interval.
Variants
- 1994 form: 50 % US large-cap, 50 % intermediate Treasuries.
- 1997-2006 form: adds small caps (about 30 % of the stock part), 4.3 % to 4.5 % SAFEMAX.
- A 'test' mix of 63 % stocks and 37 % bonds with 15 % each in large and small caps and 32 % bonds (FA Magazine 2024, Table 2).
- Bengen's unconstrained 'optimal' average: about 91 % stocks (small about 42 %, micro about 21 %, international 14 %, large about 10 %), bonds about 4 %, bills 5 %: it is a result of optimising on past data, not a recommendation for use.
Criticism
- The 4.7 % result rests on US history since 1926 with a small number of independent 30-year periods; a replication on bond index data from 1973 finds different numbers (4.2 % for 50/50, up to 4.9 % with 75 % stocks).
- Micro-cap and small-cap series are hard to hold cheaply and the equal 11 % weights are a convention, not an optimum, as Bengen himself says.
- Gain from adding asset classes beyond small caps is small: about 0.2 percentage points in total according to a summary of his work.
- Bengen himself says the rule does not apply in today's expensive markets without care, and recommends no lower than 5.25 % to 5.5 % only for those starting retirement at high valuations.
Through the eras
| Era | Dates | Total | Per year | Worst drop | S&P 500 |
|---|---|---|---|---|---|
| Roaring Twenties boom | 1926-07 – 1929-08 | +67.7 % | +17.7 % | −4.1 % | +153.7 % |
| Wall Street crash and Great Contraction | 1929-09 – 1932-07 | −51.8 % | −22.1 % | −58.8 % | −78.2 % |
| New Deal recovery | 1932-08 – 1937-02 | +185.4 % | +25.7 % | −21.2 % | +268.7 % |
| Recession within the Depression | 1937-03 – 1938-03 | −32.3 % | −30.2 % | −32.3 % | −49.5 % |
| Rearmament and World War II | 1938-04 – 1945-09 | +144.2 % | +12.6 % | −14.2 % | +205.8 % |
| Postwar expansion | 1945-10 – 1965-12 | +469.8 % | +9.0 % | −16.8 % | +1212.9 % |
| Golden age of the 1950s | 1950-01 – 1959-12 | +188.6 % | +11.2 % | −8.8 % | +438.8 % |
| Go-go years and nifty-fifty | 1966-01 – 1968-12 | +40.5 % | +12.0 % | −10.6 % | +34.0 % |
| The Great Inflation (stagflation) | 1969-01 – 1982-07 | +129.2 % | +6.3 % | −27.2 % | +84.6 % |
| 1973-74 oil shock and bear market | 1973-01 – 1974-12 | −23.7 % | −12.7 % | −27.1 % | −41.7 % |
| Second oil shock and the Volcker disinflation | 1979-01 – 1982-08 | +54.8 % | +12.7 % | −10.9 % | +59.2 % |
| Great bull market in stocks and bonds | 1982-08 – 1999-12 | +934.8 % | +14.4 % | −19.0 % | +2060.4 % |
| Black Monday | 1987-08 – 1987-12 | −13.8 % | — | −19.0 % | −21.8 % |
| Savings-and-loan crisis and 1990 recession | 1990-07 – 1991-03 | +6.2 % | — | −11.0 % | +7.4 % |
| 1994 bond market massacre | 1994-02 – 1994-12 | −5.0 % | — | −6.7 % | −3.2 % |
| Asian and Russian financial crises | 1997-07 – 1998-10 | +14.3 % | +10.6 % | −9.5 % | +22.4 % |
| Dot-com bubble and crash | 2000-03 – 2002-10 | −5.3 % | −2.0 % | −9.3 % | −37.2 % |
| Recovery and housing boom | 2002-11 – 2007-09 | +74.9 % | +12.0 % | −3.7 % | +97.0 % |
| Global financial crisis | 2007-10 – 2009-03 | −19.2 % | −13.3 % | −25.1 % | −44.7 % |
| The decade after the crisis | 2009-04 – 2019-12 | +175.6 % | +9.9 % | −9.3 % | +411.2 % |
| Fourth-quarter 2018 sell-off | 2018-09 – 2018-12 | −9.3 % | — | −9.3 % | −14.4 % |
| Covid crash and stimulus boom | 2020-02 – 2021-12 | +26.9 % | +13.2 % | −9.4 % | +53.6 % |
| Inflation shock and rate hikes | 2022-01 – 2022-12 | −16.3 % | −16.3 % | −19.8 % | −19.9 % |
Calendar years
Any N-year period
| Any N-year period | Worst | 10 % worse than | Median | 10 % better than | Best |
|---|---|---|---|---|---|
| 1 y (1192) | −39.9 % | −5.9 % | +10.0 % | +24.3 % | +103.8 % |
| 3 y (1168) | −24.1 % | +0.8 % | +9.4 % | +16.4 % | +34.5 % |
| 5 y (1144) | −9.4 % | +3.8 % | +8.7 % | +14.4 % | +25.8 % |
| 10 y (1084) | +1.4 % | +5.8 % | +8.8 % | +13.1 % | +16.4 % |
| 15 y (1024) | +3.6 % | +6.6 % | +8.8 % | +13.0 % | +15.8 % |
| 20 y (964) | +4.8 % | +7.0 % | +9.1 % | +12.6 % | +14.3 % |
| 30 y (844) | +7.0 % | +7.8 % | +9.5 % | +11.2 % | +12.8 % |
How it behaved in history
- 1966-1982 (stagflation). The binding case for the 4 % rule is a retiree who started at the peak in October 1968: stocks and bonds lost real value together during the 1970s inflation. Bengen describes 4 % as a worst case based on this start. en.wikipedia.org
- 1973-1974. In an independent replication using bond index data from 1973, the worst cohort for the 50/50 mix is the 1973 cohort (retiring 1973, funds lasting to 2002). financialplanningassociation.org
- 1929-1932. The study starts in 1926, so Depression retirees are inside the tests, but the sources read name 1968 as the worst start; a reader comment mentions the long recovery after 1929 only anecdotally. www.aaii.com
- 2000-2002 and 2008. No era-specific results for the seven-class portfolio found in the sources read.
Published results (by others)
- 1994 SAFEMAX for 50/50 large-cap/intermediate Treasuries with annual rebalancing (1926-1992 data): about 4.0 % (later cited as 4.15 %) financialplanningassociation.org
- Universal SAFEMAX of the seven-class portfolio (55 % stocks, 40 % bonds, 5 % cash), 30-year horizon (400 quarterly start dates, 1926-2024): 4.7 %; average historical starting rate 7.0 %; only the October 1968 retiree was limited to 4.7 % www.aaii.com
- Average SAFEMAX of the standard 55/45 allocation vs unconstrained optimal, across retirees (1926-2020s data): 7.25 % vs 11.72 % digitalmagazine.fa-mag.com
- Adding US small caps raised SAFEMAX from 4.15 % to 4.5 %; the other classes added about 0.2 points (1926-2024): 4.15 % -> 4.5 % www.advisorperspectives.com
- Independent replication (bond index data) of the worst cohort (retirement start 1973): 4.2 % (50/50), 4.5 % (adding 30 % small), 4.7 % (adding mid), 4.9 % (75 % stocks) financialplanningassociation.org
- Simplified Richer Retirement portfolio (same weights as our mapping) with data since 1970 (1970-2025): 30-year safe withdrawal rate about 4.8 % portfoliocharts.com
What we could not verify
- The text of the 2006 book 'Conserving Client Portfolios During Retirement' was not available; its asset mix is described only through secondary sources (small-cap addition, 4.5 %/4.1 % SAFEMAX). The 2006 book is not the source of the weights used here.
- The 11 % equal-weight table is from the FA Magazine article read through a page of the digital magazine; Bengen's 2025 book 'A Richer Retirement' (release date reported as 5 August 2025) was not read, so the book's weights may differ.
- Whether the stock classes in the 2025 portfolio are exactly equal-weighted or only for the baseline test: Portfolio Charts says Bengen finally used an equal-weight collection; no other source confirmed this.
- The 1994 SAFEMAX is quoted as 4.0 %, 4.15 % and 4.2 % in different sources; the 1997 and 2006 asset splits are given only by secondary sources.
- Mapping of mid and micro caps to our large/small classes is an approximation.
- The Wikipedia page was used for the biography, the 1994 article and the 1968 start, as a secondary source.
Sources
- digitalmagazine.fa-mag.com
https://digitalmagazine.fa-mag.com/fa-magazine-may-2024/0433186001713813105/p19 - financialplanningassociation.org
https://financialplanningassociation.org/learning/publications/journal/NOV23-revisiting-william-bengens-safemax-portfolio-withdrawal-rate-OPEN - www.advisorperspectives.com
https://www.advisorperspectives.com/articles/2025/08/29/bill-bengen-boosts-the-4-rule-to-4-7 - www.aaii.com
https://www.aaii.com/journal/article/321922-is-47-the-new-safe-retirement-withdrawal-rate - portfoliocharts.com
https://portfoliocharts.com/2026/02/16/sail-to-a-good-life-with-the-richer-retirement-portfolio/ - en.wikipedia.org
https://en.wikipedia.org/wiki/William_Bengen - www.financialplanningassociation.org
https://www.financialplanningassociation.org/sites/default/files/2021-10/DEC97%20JFP%20Bengen%20PDF.pdf - www.financialplanningassociation.org
https://www.financialplanningassociation.org/sites/default/files/2021-10/AUG96%20-%20JFP%20Bengen%20PDF.pdf - retirementresearcher.com
https://retirementresearcher.com/?p=7650 - roberthuebscher.substack.com
https://roberthuebscher.substack.com/p/rethinking-retirement-bill-bengens - www.aaii.com
https://www.aaii.com/journal/article/insights-on-using-the-withdrawal-rule-from-its-creator
Calculated on historical data. Past results do not predict the future. Models before the 1990s use stand-ins for classes that had no index yet; fees, taxes and fund costs are not included. This is mathematics, not an investment recommendation.