Graham 50/50
Graham 50/50 (standard division for the defensive investor)
Where it stands today
Our calculation, 1926-07 – 2026-09
Growth of $10,000
Where it came from
Benjamin Graham (born Benjamin Grossbaum in London, 9 May 1894; died 21 September 1976) co-wrote 'Security Analysis' with David Dodd in 1934 and ran the Graham-Newman fund from 1936 with Jerome Newman. Security Analysis defined an investment as an operation promising safety of principal and a satisfactory return. It was aimed at professionals picking individual securities after the 1929 crash and the depression.
'The Intelligent Investor' first appeared in 1949 (Harper & Brothers) as a book for non-professionals. Graham revised it several times (printings dated 1954, 1959 and 1965 are cited in the copyright listing), the last time in 1971-72; the result was the Fourth Revised Edition (1973), with a preface and appendices by Warren Buffett. Jason Zweig added commentary after Graham's death, in the 2003 edition. Buffett called the book by far the best on investing ever written, and says he bought it in 1949 and that it changed his financial life.
In the book he splits investors into the defensive (passive) investor and the enterprising investor. For the defensive investor Graham sets a simple policy: divide funds between high-grade bonds and high-grade common stocks. As quoted by a secondary source, he says the investor should never have less than 25 % or more than 75 % of funds in common stocks, so bonds range inversely between 75 % and 25 %, and the 'standard division' is an equal one, 50-50.
His logic for moving within the band depends on the market's price level. He says tradition favours raising the stock share when 'bargain price' levels appear after a long bear market, and cutting the stock share below 50 % when the market 'has become dangerously high' by the investor's judgment. So the 25 % stock mix is the extreme for an expensive market and the 75 % mix is the extreme for a cheap one. Graham leaves this to the investor's judgment rather than a mechanical formula, and secondary summaries note that he does not tie the split to age.
Practical details attributed to Graham in secondary summaries (not verified against the book): the bond side should be high-grade, with U.S. government bonds favoured over corporates; the stock side a diversified list or fund shares; changes should be rare and rebalancing on fixed dates (summaries mention every six months). Treat these as unverified until the book text is checked.
Harry Markowitz is often connected to a 50/50 split, but not to Graham: in a late-1997 interview with Jason Zweig he said that in the mid-1950s, shortly after publishing 'Portfolio Selection' (1952), he split his TIAA-CREF contributions 50/50 between bonds and stocks to minimise future regret, instead of using his own optimisation. In a 2009 Q&A he said he no longer used 50/50 and spread money across asset classes. I found no source in which Markowitz refers to Graham's rule, so the commonly repeated link is not supported by what I could open.
What is in it
| High-grade common stocks (diversified) | 50% | Graham's standard division. |
| High-grade bonds | 50% |
How we model it
| US large companies (S&P 500) | 50% |
| Intermediate government bonds (10 y) | 50% |
For a euro investor
How to build it with European funds
| Part | Share | Example funds (ISIN) |
|---|---|---|
| US large companies (S&P 500) | 50% | iShares Core S&P 500 UCITS ETF USD (Acc) IE00B5BMR087Vanguard S&P 500 UCITS ETF (USD) Distributing IE00B3XXRP09 |
| Intermediate government bonds (10 y) | 50% | iShares USD Treasury Bond 7-10yr UCITS ETF (Dist) IE00B1FZS798 |
The idea
Protect against both a fall in stocks and inflation/a rise in prices by always holding both asset types; the defensive investor wants safety and little work. Equity share adjusted modestly to valuation.
How people buy it
Today: a total/large-cap U.S. index fund and a government or high-grade bond index fund (e.g. an intermediate Treasury fund), in the stated proportions. Graham himself had no index funds (he died in 1976, the first retail one was launched in August 1976).
Rebalancing as published
Keep to the chosen ratio and change it only on valuation grounds within 25-75 %; secondary summaries mention fixed rebalancing dates (e.g. every six months) - unverified.
Variants
- 25/75, 50/50, 75/25 are the three ends/centre of the Graham range
- Modern 60/40 and 80/20 index portfolios
- Valuation-based tilting (e.g. using CAPE) as a mechanical version of Graham's judgment
- Markowitz's personal 50/50 (TIAA-CREF), not connected to Graham in sources found
Criticism
- Graham's range is a judgment rule; he gives no formula for how far to move at a given valuation, so results depend on the investor's skill at judging 'dangerously high'.
- Advice was written before index funds and with 1930s-1970s bond markets in mind; the bond leg behaved differently in the inflation of the 1970s (general, unsourced).
- The stock/bond split is not tied to age or goals (secondary summaries).
Through the eras
| Era | Dates | Total | Per year | Worst drop | S&P 500 |
|---|---|---|---|---|---|
| Roaring Twenties boom | 1926-07 – 1929-08 | +73.2 % | +19.0 % | −2.8 % | +153.7 % |
| Wall Street crash and Great Contraction | 1929-09 – 1932-07 | −47.6 % | −19.8 % | −53.5 % | −78.2 % |
| New Deal recovery | 1932-08 – 1937-02 | +120.7 % | +18.8 % | −13.9 % | +268.7 % |
| Recession within the Depression | 1937-03 – 1938-03 | −25.1 % | −23.4 % | −25.1 % | −49.5 % |
| Rearmament and World War II | 1938-04 – 1945-09 | +96.8 % | +9.4 % | −14.0 % | +205.8 % |
| Postwar expansion | 1945-10 – 1965-12 | +396.1 % | +8.2 % | −12.4 % | +1212.9 % |
| Golden age of the 1950s | 1950-01 – 1959-12 | +160.8 % | +10.1 % | −7.6 % | +438.8 % |
| Go-go years and nifty-fifty | 1966-01 – 1968-12 | +19.7 % | +6.2 % | −8.7 % | +34.0 % |
| The Great Inflation (stagflation) | 1969-01 – 1982-07 | +101.7 % | +5.3 % | −24.2 % | +84.6 % |
| 1973-74 oil shock and bear market | 1973-01 – 1974-12 | −18.9 % | −10.0 % | −24.2 % | −41.7 % |
| Second oil shock and the Volcker disinflation | 1979-01 – 1982-08 | +43.1 % | +10.3 % | −9.8 % | +59.2 % |
| Great bull market in stocks and bonds | 1982-08 – 1999-12 | +1088.2 % | +15.3 % | −16.7 % | +2060.4 % |
| Black Monday | 1987-08 – 1987-12 | −11.5 % | — | −16.7 % | −21.8 % |
| Savings-and-loan crisis and 1990 recession | 1990-07 – 1991-03 | +8.8 % | — | −8.0 % | +7.4 % |
| 1994 bond market massacre | 1994-02 – 1994-12 | −5.8 % | — | −8.2 % | −3.2 % |
| Asian and Russian financial crises | 1997-07 – 1998-10 | +23.3 % | +17.0 % | −7.2 % | +22.4 % |
| Dot-com bubble and crash | 2000-03 – 2002-10 | −3.8 % | −1.4 % | −12.2 % | −37.2 % |
| Recovery and housing boom | 2002-11 – 2007-09 | +53.8 % | +9.2 % | −3.4 % | +97.0 % |
| Global financial crisis | 2007-10 – 2009-03 | −13.2 % | −9.0 % | −19.3 % | −44.7 % |
| The decade after the crisis | 2009-04 – 2019-12 | +179.3 % | +10.0 % | −6.5 % | +411.2 % |
| Fourth-quarter 2018 sell-off | 2018-09 – 2018-12 | −6.5 % | — | −6.5 % | −14.4 % |
| Covid crash and stimulus boom | 2020-02 – 2021-12 | +26.4 % | +13.0 % | −5.8 % | +53.6 % |
| Inflation shock and rate hikes | 2022-01 – 2022-12 | −18.1 % | −18.1 % | −20.8 % | −19.9 % |
Calendar years
Any N-year period
| Any N-year period | Worst | 10 % worse than | Median | 10 % better than | Best |
|---|---|---|---|---|---|
| 1 y (1192) | −35.7 % | −4.5 % | +9.3 % | +21.3 % | +64.5 % |
| 3 y (1168) | −20.6 % | +0.6 % | +8.0 % | +15.5 % | +25.8 % |
| 5 y (1144) | −6.7 % | +3.5 % | +7.8 % | +13.9 % | +23.2 % |
| 10 y (1084) | +0.6 % | +4.7 % | +7.8 % | +13.4 % | +16.4 % |
| 15 y (1024) | +2.5 % | +5.5 % | +8.1 % | +12.8 % | +15.6 % |
| 20 y (964) | +3.3 % | +6.0 % | +8.2 % | +12.1 % | +14.8 % |
| 30 y (844) | +5.7 % | +6.9 % | +8.6 % | +10.9 % | +11.9 % |
How it behaved in history
- 1929-1939. Graham and Dodd's Security Analysis (1934) came out after the 1929 crash; Graham's performance was about 20 % a year 1936-1956 against roughly 12.2 % for the market per Wikipedia (this concerns Graham-Newman stock picking, not the 50/50 policy). en.wikipedia.org
Published results (by others)
- Markowitz said he personally used a 50/50 stock-bond split in his TIAA-CREF account (regret minimisation); it is not a statement about Graham (mid-1950s practice, interview late 1997): 50/50 jasonzweig.com
- Markowitz noted a 50/50 mix left unattended in a bull market drifts toward 70/30 or 80/20 (as reported in a search summary of his 2009 Q&A) (2009): 50/50 to 70/30 or 80/20 jasonzweig.com
What we could not verify
- Exact wording of the 25-75 % rule is taken from a blog quoting the book; page/chapter (Chapter 4 in the usual numbering) not verified against the book.
- Whether the 25-75 % limits were already in the 1949 first edition is not confirmed (the 1973 text is the reference); the 1934 Security Analysis did not (as far as known) give such a rule - not verified.
- No published return series for Graham's 50/50 mix found.
- Markowitz did use 50/50 himself (Zweig interview, 1997); the often-heard claim that he cited Graham's 50/50 as the basis was not found in any source, so this site does not claim that link.
Sources
- www.retirementinvestingtoday.com
https://www.retirementinvestingtoday.com/2015/06/ - en.wikipedia.org
https://en.wikipedia.org/wiki/The_Intelligent_Investor - en.wikipedia.org
https://en.wikipedia.org/wiki/Benjamin_Graham - en.wikipedia.org
https://en.wikipedia.org/wiki/Security_Analysis_(book)
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.