Guide
The permanent portfolio, backtested with real funds
Harry Browne's permanent portfolio (25% each in stocks, long-term Treasuries, cash and gold) fell far less than a 60/40 but grew more slowly: from 18 November 2004 to 20 November 2026 it grew 6.9% a year against 7.9%, with a worst fall of 19.0% against 30.0%. Its worst year was not 2008 but 2022, when all four slices fell together.
What the permanent portfolio is
The permanent portfolio is a mix designed in the 1980s by the investment writer Harry Browne. It splits the money into four equal slices of 25%: stocks, long-term government bonds, cash (very short-term government bonds) and gold. Each slice is there for one kind of economy:
- Stocks for prosperity, when companies grow and markets rise.
- Long-term government bonds for deflation and recessions, when interest rates fall and these bonds rise in price.
- Gold for inflation and loss of confidence in money.
- Cash for tight money, when interest rates are high and other assets struggle.
The promise is not the highest return. It is that some slice should be holding up whatever happens, so the whole never falls very far. This guide tests that promise with real funds, step by step, so you can judge whether the trade suits you.
Permanent portfolio and 60/40 side by side
Permanent portfolio
- Stocks, long bonds, cash, gold
- A slice for each economy
- Smaller falls
- Slower growth
- Hurt when rates rise fast
60/40
- Stocks and bonds
- Built for growth with a cushion
- Deeper falls in stock crashes
- Faster growth
- Also hurt when rates rise fast
The follow-along example
We build the permanent portfolio from four funds you can buy, and run it against a classic stock and bond mix, each starting with $10,000, dividends and interest reinvested, from 18 November 2004 (the first day all the funds have prices) to 20 November 2026.
Follow along: the four portfolios
Portfolio What it holds Rule Permanent portfolio 25% VTI, 25% TLT, 25% SHY, 25% GLD Reset to a quarter each once a year Permanent portfolio, never rebalanced The same four, same start Left alone VTI 60 / IEF 40 60% US stocks, 40% intermediate Treasuries; our benchmark Reset once a year VTI 100% US stocks, for reference Held VTI holds almost every listed US company. TLT holds US Treasury bonds with twenty or more years to run. SHY holds Treasuries maturing within three years, the closest fund to cash. GLD holds gold bullion. IEF holds Treasuries maturing in seven to ten years. Rebalancing means selling what has grown and buying what has lagged to get back to the target weights. The benchmark is the yardstick: a 60/40 mix, 60% stocks and 40% bonds.
How we tested this
Step 1: Look at the worst fall first
The permanent portfolio exists to keep falls small, so start with its max drawdown: the largest drop from a high to a later low (max drawdown explained).
- Permanent portfolio worst fall 19.0% (9 November 2021 to 20 October 2022), longest time below a previous high 2.5 years
- Permanent portfolio, never rebalanced worst fall 20.9% (9 November 2021 to 20 October 2022), longest time below a previous high 3.7 years
- VTI worst fall 55.5% (9 October 2007 to 9 March 2009), longest time below a previous high 4.4 years
- VTI 60 / IEF 40 worst fall 30.0% (9 October 2007 to 9 March 2009), longest time below a previous high 2.4 years
Full period: Indices Lab backtest, total return, mixes rebalanced yearly, single funds held throughout, 18 November 2004 to 20 November 2026. Hypothetical results: past performance does not predict future returns.
Follow along: the deepest fall and the longest wait
Portfolio Max drawdown Longest time below a previous high Permanent portfolio 19.0% 2.5 years VTI 60 / IEF 40 30.0% 2.4 years VTI 55.5% 4.4 years The permanent portfolio's worst fall was far smaller than the 60/40's and a fraction of the all-stock fund's. The surprise is when it came: not in the 2008 crash, but from its late-2021 high through 2022.
So the first lesson: a mix built for every economy can still be caught by one it was not built for.
Step 2: Check each bad year on its own
A calendar-year view shows which slice did the work in each bad year, and when none did.
| 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 | 2023 | 2024 | 2025 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Permanent portfolio | +8.5% | +10.7% | +13.4% | +2.1% | +7.9% | +14.5% | +11.5% | +6.5% | −2.0% | +9.5% | −2.9% | +5.7% | +10.9% | −1.8% | +16.5% | +16.8% | +4.0% | −13.9% | +11.4% | +11.6% | +22.5% |
| Permanent portfolio, never rebalanced | +8.5% | +11.0% | +14.3% | +1.5% | +6.4% | +17.4% | +11.4% | +6.3% | −9.4% | +8.6% | −3.6% | +6.4% | +12.5% | −2.5% | +19.0% | +19.3% | +6.7% | −15.2% | +15.4% | +17.9% | +30.3% |
| VTI | +6.3% | +15.7% | +5.4% | −37.0% | +28.9% | +17.4% | +1.0% | +16.5% | +33.4% | +12.5% | +0.4% | +12.8% | +21.2% | −5.2% | +30.7% | +21.1% | +25.7% | −19.5% | +26.0% | +23.8% | +17.1% |
| VTI 60 / IEF 40 | +4.8% | +10.4% | +7.4% | −15.0% | +14.7% | +14.2% | +6.8% | +11.3% | +17.6% | +11.2% | +0.8% | +8.1% | +13.7% | −2.7% | +21.6% | +16.6% | +14.1% | −17.8% | +17.1% | +14.0% | +13.5% |
Full calendar years only. Full period: Indices Lab backtest, total return, mixes rebalanced yearly, single funds held throughout, 18 November 2004 to 20 November 2026. Hypothetical results: past performance does not predict future returns.
Follow along: the years that tested it
Year Permanent portfolio VTI 60 / IEF 40 VTI 2008 +2.1% −15.0% −37.0% 2013 −2.0% +17.6% +33.4% 2022 −13.9% −17.8% −19.5% In 2008 stocks crashed, but long Treasuries, cash and gold all rose, and the permanent portfolio ended the year up. In 2013 it was the other way round: stocks had a strong year, while gold and long Treasuries both fell, and the mix lost money while the 60/40 gained. In 2022 rates rose fast to fight inflation and all four slices fell in the same year, so nothing held up. It was the mix's worst calendar year of the test.
The lesson: the four slices protect each other only while they move differently. When rising rates hit stocks and bonds together, as in 2022, even four slices behave like one.
Step 3: Measure what the calm cost
Now the price of the smaller falls. Read growth as CAGR, the compound annual growth rate: the steady yearly rate that turns the start value into the end value (how CAGR works).
- Permanent portfolio $43,643
- Permanent portfolio, never rebalanced $48,620
- VTI $87,482
- VTI 60 / IEF 40 $53,601
Full period: Indices Lab backtest, total return, mixes rebalanced yearly, single funds held throughout, 18 November 2004 to 20 November 2026. Hypothetical results: past performance does not predict future returns.
Follow along: growth over the whole period
Portfolio CAGR Gap to 60/40, per year End value of $10,000 Permanent portfolio 6.9% −1.0 points $43,643 VTI 60 / IEF 40 7.9% $53,601 VTI 10.4% +2.4 points $87,482 About a point a year looks small. Compounded over two decades, it left the permanent portfolio with clearly less money at the end than the 60/40.
Step 4: Decide how you will rebalance
This is the trap most people miss. Left alone, a four-slice mix does not stay four slices: whatever grows fastest takes over. Compare the yearly reset with the same four funds never touched.
Follow along: reset once a year or leave it alone
Reset yearly Never rebalanced CAGR 6.9% 7.5% End value $43,643 $48,620 Volatility 7.5% 9.3% Max drawdown 19.0% 20.9% Longest time below a high 2.5 years 3.7 years 2013 −2.0% −9.4% The untouched version ended with more money, because stocks and gold grew and were never trimmed. But it stopped being a permanent portfolio: after gold's run to 2012, gold had grown to close to half of it, so the 2013 gold fall hit it much harder, and it then spent years below its old high. By the end, stocks and gold made up most of it.
Volatility is how much a portfolio swings up and down, as a yearly percentage (volatility explained). Rebalancing kept it low and kept the four-way balance that the whole idea depends on. Browne himself suggested resetting only when a slice drifts far from a quarter, rather than on a fixed date; either way, pick a rule before you start and keep to it.
Step 5: Check every start date
One start date can flatter or punish any mix. A rolling window test takes every possible 5-, 10- or 20-year stretch, one starting each month, and counts how often the permanent portfolio finished ahead of the 60/40. The chart shows every 10-year window.
Permanent portfolio finished ahead of VTI 60 / IEF 40 in 3.4% of 145 rolling 10-year windows (window ends by month). The last window Permanent portfolio won ended August 2015.
Full period: Indices Lab backtest, total return, mixes rebalanced yearly, single funds held throughout, 18 November 2004 to 20 November 2026. Hypothetical results: past performance does not predict future returns.
Follow along: how often the permanent portfolio won
Window Ahead of 60/40 Windows tested 5 years 23.4% 205 10 years 3.4% 145 20 years 0.0% 25 Calendar years 47.6% It won close to half of the calendar years but almost never a 10-year stretch. The few 10-year windows it won all started in 2004 or 2005 and ran through the whole 2008 crash; the last ended in August 2015.
A caution: the long windows overlap heavily, so treat them as one long result rather than many separate tests.
Step 6: Split the history in two
A long backtest can hide two different stories. Split it at the midpoint.
Follow along: each half on its own
Portfolio Growth per year, first half Second half Max drawdown, first half Second half Permanent portfolio 7.2% 6.7% 14.1% 19.0% VTI 60 / IEF 40 7.5% 8.4% 30.0% 22.1% In the first half, which holds the 2008 crash, the permanent portfolio grew almost as fast as the 60/40 with half its worst fall. In the second half it fell behind on growth, and its worst fall, in 2022, was close to the 60/40's.
Its own growth was steady in both halves; what changed was the 60/40, which grew faster once stocks led for a decade.
Step 7: Measure return for the swing
The best case for the permanent portfolio is a smooth ride. The Sharpe ratio divides return by volatility, and the Sortino ratio does the same counting only the falls, so higher means more return for each unit of swing (the risk ratios compared).
| Basket | Yearly return | Volatility | Sharpe | Worst drop | $10,000 became |
|---|---|---|---|---|---|
| Permanent portfolio | 6.9% | 7.5% | 0.93 | 19.0% | $43,643 |
| Permanent portfolio, never rebalanced | 7.5% | 9.3% | 0.82 | 20.9% | $48,620 |
| VTI | 10.4% | 19.0% | 0.61 | 55.5% | $87,482 |
| VTI 60 / IEF 40 | 7.9% | 10.4% | 0.79 | 30.0% | $53,601 |
Full period: Indices Lab backtest, total return, mixes rebalanced yearly, single funds held throughout, 18 November 2004 to 20 November 2026. Hypothetical results: past performance does not predict future returns. Sharpe uses a 0% risk-free rate.
Follow along: return for the risk taken
Portfolio Volatility Sharpe Sortino Permanent portfolio 7.5% 0.93 1.33 VTI 60 / IEF 40 10.4% 0.79 1.11 VTI 19.0% 0.61 0.86 The permanent portfolio swung the least and earned the most per unit of swing. A higher ratio still does not mean more money: both other portfolios ended ahead.
Step 8: Decide what you are aiming for
The permanent portfolio suits someone who values a small worst fall over the last point of growth, and who will rebalance even when it means buying the slice that has just done worst. Write down the largest fall you could sit through and the growth you need, then test it against the mixes you are considering.
Before you hold a permanent portfolio
- Write down the largest fall you could sit through
- Check 2008, 2013 and 2022 separately
- Choose a rebalancing rule and keep to it
- Test every start date, not one
- Know which funds fill each slice
- Compare it with the mix you would hold instead
What this example cannot show: the high-inflation 1970s; markets outside the US; other ways to build the slices (an intermediate bond fund instead of TLT behaves very differently); and money taken out every year in retirement.
Questions
What is the permanent portfolio?
A mix designed by Harry Browne that holds 25% each in stocks, long-term government bonds, cash and gold, reset to equal slices from time to time. Each slice is meant to hold up in a different economy: growth, recession, inflation and tight money.
What return has the permanent portfolio had?
Built from VTI, TLT, SHY and GLD and reset once a year, it grew 6.9% a year from 18 November 2004 to 20 November 2026, turning $10,000 into $43,643. Past returns are not a forecast, and gold and bond returns depend heavily on the period.
Permanent portfolio vs 60/40: which did better?
The 60/40 (VTI 60 / IEF 40) grew faster, 7.9% a year, and finished ahead in almost every 10-year window. The permanent portfolio had the smaller worst fall (19.0% against 30.0%) and the higher Sharpe ratio (0.93 against 0.79).
How did the permanent portfolio do in 2022?
It returned −13.9%, its worst calendar year in the test. Interest rates rose quickly, and stocks, long bonds, short bonds and gold all fell in the same year.
Which ETFs can build a permanent portfolio?
This test used VTI for stocks, TLT for long-term Treasuries, SHY for cash and GLD for gold. Other funds tracking the same four slices exist; a shorter bond fund in place of TLT changes the results a lot, so test the exact funds you would hold.
About this page. Figures on this page come from Indices Lab’s own data, powered by EODHD, and every number is recalculated after each market close; the text is reviewed by a person before it is published. Indices Lab is a research and educational tool: everything here is hypothetical and informational, not investment advice and not a recommendation to buy or sell any security. Past performance does not predict future results. See our methodology.