Golden Butterfly
A passive five-asset portfolio at 20% each, derived from the Permanent Portfolio with a second equity fund (small cap value) to tilt it toward prosperous periods.
What the Golden Butterfly is
The Golden Butterfly is a static portfolio of five equally weighted assets, 20% each. Its author, Tyler, of the Portfolio Charts website, presented it in April 2016. According to his explanation, it starts from the Permanent Portfolio and adds a second equity fund, of small value companies (small cap value), to tilt it toward prosperous environments. Keller also uses it as a benchmark in his LAA paper, with the same five assets at 20%.
Quick facts
- Author
- Tyler (Portfolio Charts)
- Year
- 2016
- Origin
- Community
- Access
- Public and free
- Type
- Passive
- Assets
- 5
- Frequency
- Annual
- Difficulty
- Low
Who it is for
Anyone who wants a diversified passive portfolio with few assets and a single review a year, and who is interested in a variant of the Permanent Portfolio with more equities.
Who it is not for
Anyone who does not want to hold gold or small cap value, or who expects the small cap value premium to repeat as in the past.
When it fits
For anyone who wants a static five-asset portfolio that needs no signals, and does not mind holding 20% in gold. For studying how a Permanent Portfolio changes when part of the cash is replaced by value and small cap equities, it is a good reference.
Current signal
Golden Butterfly
Calculated with the closes of September 30, 2026.
No portfolio changes from last month.
Rules, step by step
| Asset (US ETF) | What it is | Weight |
|---|---|---|
| SPY | US stocks, large companies (S&P 500) | 20% |
| IWN | US small value companies (Russell 2000 Value) | 20% |
| TLT | Long-term Treasury bonds | 20% |
| SHY | Short-term Treasury bonds (1-3 years) | 20% |
| GLD | Gold | 20% |
There are no signals. You buy the five at equal weight and rebalance to return to 20%. In what we have consulted the author does not fix a rebalancing frequency, so the site's convention is annual, at the end of December. For small cap value, the ETF can be IWN, VBR or IJS: they differ in index and available history. In the backtest we use IWN.
Backtest and out-of-sample period
With real ETFs from Jan 2005 to Sep 2026, with annual rebalancing and a cost of 0.1% per trade, the portfolio returns 7.5% a year, with a maximum monthly drawdown of -17.8%, volatility of 8.1% and a Sharpe of 0.78. The worst year was 2022 (-13.8%) and the best, 2025 (+20.6%). It turns over 7% a year on average. The S&P 500 bought and held, measured from Mar 1993, gives 10.8% a year with a maximum drawdown of 51%; the start dates differ, so that comparison is not clean.
Since Apr 2016, the date of the article with the portfolio's theory, it returns 7.7% a year with a maximum drawdown of -17.8%. It is the only stretch in which the rules were already published.
The author's figures
Portfolio Charts publishes the portfolio with US data from 1970 to 2025, in real terms, that is, adjusted for inflation, with annual data. Its real volatility is 8.3% and its real maximum drawdown, 18.0%. The longest drawdown it records lasted five years. Those figures are not comparable with those above, which are nominal and monthly, but they have one advantage: they cover decades in which ETFs did not exist.
7.7% a year · max drawdown -17.8% · S&P 500 over the same period: 15.1%
An "out-of-sample" result: the period after the rule was published, which could not have been used to tune the rules.
Growth of $100 (logarithmic scale) since , with 0.1% costs per trade. Data: US-listed ETFs, total return (dividends reinvested), before taxes.
Drawdowns from peak
| Strategy | CAGR | Volatility | Max drawdown | Sharpe |
|---|
Year by year
| Year | Golden Butterfly | S&P 500 | 60/40 |
|---|---|---|---|
| 2026 (to Sep 2026) | +3.6% | +12.7% | +5.8% |
| 2025 | +20.6% | +17.7% | +13.8% |
| 2024 | +11.0% | +24.9% | +14.7% |
| 2023 | +12.1% | +26.2% | +17.2% |
| 2022 | -13.8% | -18.2% | -17.0% |
| 2021 | +9.4% | +28.7% | +15.9% |
| 2020 | +13.8% | +18.3% | +15.0% |
| 2019 | +17.7% | +31.2% | +21.9% |
| 2018 | -3.9% | -4.6% | -2.4% |
| 2017 | +10.3% | +21.7% | +14.0% |
| 2016 | +10.8% | +12.0% | +7.6% |
| 2015 | -3.7% | +1.2% | +1.3% |
| 2014 | +8.6% | +13.5% | +11.7% |
| 2013 | +5.0% | +32.3% | +16.9% |
| 2012 | +8.7% | +16.0% | +11.1% |
| 2011 | +8.2% | +1.9% | +7.4% |
| 2010 | +16.1% | +15.1% | +12.8% |
| 2009 | +9.9% | +26.4% | +13.1% |
| 2008 | -4.0% | -36.8% | -14.9% |
| 2007 | +8.6% | +5.1% | +7.2% |
| 2006 | +13.3% | +15.8% | +10.5% |
| 2005 | +7.3% | +4.8% | +3.9% |
The first year may be partial. Returns are net of trading costs.
Taxes on rotating
The results above are before taxes. This portfolio is rebalanced once a year: historically it turns over only 7% of the portfolio a year on average, so it generates few sales.
In most countries each sale of an ETF in a taxable account realises a gain or loss that may be taxed, so rotating monthly is best done inside a tax-advantaged account (such as an IRA or 401(k) in the US, or an ISA or SIPP in the UK) where one is available. Rules vary by country and by account: check yours, or ask a tax adviser. See our methodology for how the results are calculated.
Criticisms and limits
The split was designed with the history already known and the author himself presents it as an illustrative portfolio. The small cap value premium has been weaker in recent decades. And the author's figures are real (inflation-adjusted) and annual, so they cannot be compared directly with those of our engine.
- It was designed looking at the past. The split came out of the already-known history, and the author himself presents it as an illustrative portfolio.
- The small cap value premium is the bet. It has been weaker in recent decades, and the portfolio depends on it more than the Permanent Portfolio does.
- Figures that are not comparable. The author's are real and annual; ours are nominal and monthly, and start in Jan 2005.
- ETF to be decided. IWN, VBR and IJS do not track the same index, and the result changes a little depending on which is chosen.
How to run it
- Map each ETF to the fund you actually hold, and check its fees and dividend policy before buying.
- Set an annual rebalancing date.
- Work out the orders with the rebalancing calculator.
Similar strategies
Frequently asked questions
What weights does the Golden Butterfly have?
20% in each of five assets: US stocks (SPY), small cap value (IWN), long Treasury bonds (TLT), short Treasury bonds (SHY) and gold (GLD).
How does it differ from the Permanent Portfolio?
The [Permanent Portfolio](/en/strategies/cartera-permanente/) splits 25% between stocks, long bonds, gold and cash. The Golden Butterfly splits 20% between five assets: it adds small cap value and holds short bonds in place of cash.
How often is it rebalanced?
The author does not fix a frequency in the source we consulted. We use once a year, at the end of December, as the site's convention.
Why do the author's figures not match those on this page?
Because his are real (after inflation), annual and from 1970. Those of our engine are nominal, monthly and use ETFs from Jan 2005.
Is this investment advice?
No. It is educational information about a published portfolio. Backtests and past performance do not guarantee future results, and you can lose money.
Sources and review date
- Portfolio Charts, "Golden Butterfly Portfolio" · accessed October 10, 2026
- Portfolio Charts, "The Theory Behind the Golden Butterfly" (18 April 2016) · accessed October 10, 2026
- Portfolio Charts, "Methodology" · accessed October 10, 2026
- Keller, "Lethargic Asset Allocation (LAA)", SSRN 3498092 (uses it as a benchmark) · accessed October 10, 2026
Page last reviewed on October 11, 2026; figures and signal are recalculated at every monthly close (latest: September 30, 2026). How we work: editorial policy and methodology.
Educational and informational content. It is not investment advice or a personal recommendation. The results shown are historical simulations (backtests) using US-listed ETFs; past performance does not guarantee future results and you may lose some or all of the capital invested. Real returns will depend on your costs, taxes, currency and execution timing.