ResourcesGuide

All Weather Portfolio: Allocation, Returns and the ALLW ETF

What Bridgewater’s All Weather strategy really is, how the popular 30/40/15/7.5/7.5 copy differs, what the 2025 ALLW ETF holds, and one consistent 54‑year test of illustrative stand‑in mixes for the Permanent Portfolio and Golden Butterfly, against 60/40.

Published Updated 10 min readHow we research

Key takeaways

  • Bridgewater’s All Weather is a leveraged risk-parity strategy built in 1996 for Ray Dalio’s trust: it puts equal risk on rising and falling growth and inflation. The 30/40/15/7.5/7.5 mix is a simplified, unleveraged version popularized in a 2014 book, not Bridgewater’s fund.
  • The ALLW ETF (launched Mar 5, 2025, 0.85% expense ratio) uses futures and swaps; on Oct 23, 2026 its notional exposures added up to 187% of the fund’s value (State Street).
  • In one consistent unleveraged test from 1972 to 2025, an illustrative 30% S&P 500 / 55% 10-year Treasuries / 15% gold mix standing in for the retail All Weather returned 8.7% a year with a worst year of −15.1% (2022); 60/40 returned 9.4% with −18.0%, and 100% S&P 500 11.1% with −36.6% (Exent calculation on NYU Stern data).
  • Of the illustrative stand‑in mixes, 25% each in S&P 500, 10-year Treasuries, T-bills and gold had the shallowest worst year (−8.3%), and 40/20/20/20 sat between, at 9.2% a year and −10.3%. These substitutions do not establish the historical ranking of the named strategies.
  • 2022 is the stress test for bond-heavy mixes: stocks and the 10-year Treasury both fell about 18%. Built with real ETFs, the retail All Weather would have lost about 18.8% that year.

Market risk reading, cross‑asset context and an events calendar: see Exent Market intel →

What is the All Weather portfolio?

It is two different things that share a name. Bridgewater Associates’ All Weather is an institutional strategy that “emerged in 1996” for Ray Dalio’s trust assets and puts equal risk on four economic environments, using leverage so that low-volatility assets such as bonds weigh as much as stocks (Bridgewater, The All Weather Story). The “All Weather portfolio” on most personal-finance sites is a five-fund, unleveraged mix that Tony Robbins published in 2014 as Dalio’s “All Seasons” portfolio: 30% stocks, 40% long-term Treasuries, 15% intermediate Treasuries, 7.5% gold and 7.5% commodities.

Scope: both versions, the ALLW ETF, and how the retail mix compares with three other “lazy” portfolios in one test. Out of scope: other risk-parity funds, leveraged backtests and tax treatment. Portfolios here are hypothetical illustrations, not advice.

How is Bridgewater’s All Weather different from the 30/40/15/7.5/7.5 portfolio?

The idea is the same; the engine is not. Side by side:

Bridgewater All WeatherRetail “All Seasons” copy
Origin1996, for Ray Dalio’s trust; later offered to institutionsTony Robbins, MONEY Master the Game (2014)
How weights are setEqual risk to each environment (risk parity)Fixed dollar weights: 30/40/15/7.5/7.5
LeverageYes: Bridgewater calls it “an implementation tool”None
BondsGlobal nominal and inflation-linked bondsUS long-term and intermediate Treasuries
InstrumentsFutures, swaps and cash securitiesFive ETFs or funds
RebalancingActively managed (ALLW follows a daily model)Usually once or twice a year

The difference matters most for bonds. Without leverage, the only way to give bonds as much risk as stocks is to hold a lot of them and choose long maturities, which is why the copy is 55% Treasuries and 40% long-dated. That made it sensitive to rising rates, as 2022 showed (below).

What are the four economic “seasons”?

Growth and inflation, each coming in above or below what markets expect. Bridgewater describes All Weather as “four different portfolios each with the same risk,” one for each case, so the strategy does not need to forecast which arrives. Its paper names nominal government bonds for a disinflationary slowdown, stocks for rising growth, and inflation-linked bonds and commodities for rising inflation, when stocks and nominal bonds do poorly. Our simplified map:

Which assets tend to help in which environment (simplified)
Growth surprise
  • Growth risingstocks, commodities
  • Growth fallingnominal government bonds
Inflation surprise
  • Inflation risinginflation-linked bonds, commodities, gold
  • Inflation fallingnominal bonds, stocks
Bridgewater’s paper describes the stocks, nominal bonds, inflation-linked bonds and commodities cases; the gold and falling-inflation entries are our simplification. Tendencies, not rules.

Cross-asset ratios are one way investors read which environment the market is pricing, for example copper against gold or stocks against bonds; see the copper/gold ratio history and cross-asset signals explained. They describe the backdrop; they do not forecast it, and All Weather is designed not to need a forecast.

What is the ALLW All Weather ETF?

The State Street Bridgewater All Weather ETF (ALLW) puts a Bridgewater-modelled All Weather allocation into one exchange-traded fund, and it is closer to the real strategy than to the five-fund copy. Facts from the State Street fund page:

ItemALLW
InceptionMar 5, 2025
Gross expense ratio0.85%
Assets under management (Oct 23, 2026)$1,835.54 million
ManagementActive; Bridgewater supplies a daily model portfolio, State Street implements it
Notional exposure, global nominal bonds70.34%
Notional exposure, global equities42.64%
Notional exposure, inflation-linked bonds41.68%
Notional exposure, commodities32.75%
Total notional exposure (sum)187.41% of fund value
Volatility target of the modelTypically 10%–12% a year

Exposures above 100% come from futures and swaps, which State Street says “may result in leverage”. Commodities are held through a Cayman Islands subsidiary that lacks some protections of a registered fund. With a history of about 20 months, ALLW has no record through a full cycle, and its 0.85% fee compares with 0.03% for Vanguard’s S&P 500 ETF (VOO, fact sheet Sep 30, 2026).

All Weather vs Permanent Portfolio vs Golden Butterfly vs 60/40: which held up best?

In this test the illustrative 25/25/25/25 mix (the Permanent Portfolio stand‑in) had the smallest losses, the 40/20/20/20 mix (the Golden Butterfly stand‑in) came within 0.2 points a year of 60/40’s return with a worst year a little over half as deep (−10.3% vs −18.0%), and 100% stocks had the highest return with by far the deepest losses. One consistent test: annual data from 1972 (gold traded freely from the early 1970s) to 2025, rebalanced every January 1, no leverage, no fees or taxes. Exent calculation on NYU Stern’s annual returns (Damodaran).

Portfolio (target weights)Return per yearAfter inflationWorst yearLosing years (of 54)Worst 10 years, per year$10,000 became
100% S&P 50011.1%6.9%−36.6% (2008)11−1.4% (1999–2008)$2,885,521
60/40: 60 stocks / 40 bonds9.4%5.3%−18.0% (2022)9+2.8% (1999–2008)$1,306,792
Illustrative mix: 40% S&P 500 / 20% 10-year Treasuries / 20% T-bills / 20% gold9.2%5.1%−10.3% (2022)9+4.5% (1999–2008)$1,151,908
Illustrative mix: 30% S&P 500 / 55% 10-year Treasuries / 15% gold8.7%4.6%−15.1% (2022)6+4.3% (2013–2022)$901,470
Illustrative mix: 25% each S&P 500, 10-year Treasuries, T-bills and gold8.5%4.4%−8.3% (2022)8+4.0% (2013–2022)$812,437
  • 100% S&P 500
    Return per year
    11.1%
    After inflation
    6.9%
    Worst year
    −36.6% (2008)
    Losing years (of 54)
    11
    Worst 10 years, per year
    −1.4% (1999–2008)
    $10,000 became
    $2,885,521
  • 60/40: 60 stocks / 40 bonds
    Return per year
    9.4%
    After inflation
    5.3%
    Worst year
    −18.0% (2022)
    Losing years (of 54)
    9
    Worst 10 years, per year
    +2.8% (1999–2008)
    $10,000 became
    $1,306,792
  • Illustrative mix: 40% S&P 500 / 20% 10-year Treasuries / 20% T-bills / 20% gold
    Return per year
    9.2%
    After inflation
    5.1%
    Worst year
    −10.3% (2022)
    Losing years (of 54)
    9
    Worst 10 years, per year
    +4.5% (1999–2008)
    $10,000 became
    $1,151,908
  • Illustrative mix: 30% S&P 500 / 55% 10-year Treasuries / 15% gold
    Return per year
    8.7%
    After inflation
    4.6%
    Worst year
    −15.1% (2022)
    Losing years (of 54)
    6
    Worst 10 years, per year
    +4.3% (2013–2022)
    $10,000 became
    $901,470
  • Illustrative mix: 25% each S&P 500, 10-year Treasuries, T-bills and gold
    Return per year
    8.5%
    After inflation
    4.4%
    Worst year
    −8.3% (2022)
    Losing years (of 54)
    8
    Worst 10 years, per year
    +4.0% (2013–2022)
    $10,000 became
    $812,437

Proxies, because one data set has to cover all five: every stock sleeve, including small-cap value, uses the S&P 500; long-term and intermediate Treasuries both use the 10-year Treasury bond; short-term Treasuries and cash use 3-month T-bills; and the 7.5% commodity sleeve of the All Weather copy is held in gold, since the data set has no commodity index. So the test understates the Golden Butterfly’s small-value tilt, understates long-bond swings (both up and down) and replaces commodities with gold. The definitions come from their popularizers: Harry Browne’s Permanent Portfolio (Rowland & Lawson, Wiley) and the Golden Butterfly (Portfolio Charts). Browne’s own rule rebalanced only when a sleeve drifted outside 15% to 35%; this test rebalances yearly. The 40/20/20/20 mix stands in for the Golden Butterfly, 30/55/15 for the retail All Weather copy and 25/25/25/25 for the Permanent Portfolio. These substitutions do not establish the historical ranking of the named strategies.

Year100% S&P 50060/4040/20/20/20 mix30/55/15 mix25/25/25/25 mix
1974 (inflation shock)−25.9%−14.7%+4.8%+3.2%+12.5%
2008 (financial crisis)−36.6%−13.9%−9.5%+0.7%−2.7%
2022 (rates up, stocks down)−18.0%−18.0%−10.3%−15.1%−8.3%
2025 (gold +66.2%)+17.7%+13.8%+22.7%+19.5%+24.0%
  • 1974 (inflation shock)
    100% S&P 500
    −25.9%
    60/40
    −14.7%
    40/20/20/20 mix
    +4.8%
    30/55/15 mix
    +3.2%
    25/25/25/25 mix
    +12.5%
  • 2008 (financial crisis)
    100% S&P 500
    −36.6%
    60/40
    −13.9%
    40/20/20/20 mix
    −9.5%
    30/55/15 mix
    +0.7%
    25/25/25/25 mix
    −2.7%
  • 2022 (rates up, stocks down)
    100% S&P 500
    −18.0%
    60/40
    −18.0%
    40/20/20/20 mix
    −10.3%
    30/55/15 mix
    −15.1%
    25/25/25/25 mix
    −8.3%
  • 2025 (gold +66.2%)
    100% S&P 500
    +17.7%
    60/40
    +13.8%
    40/20/20/20 mix
    +22.7%
    30/55/15 mix
    +19.5%
    25/25/25/25 mix
    +24.0%

Sensitivity: from 2000 to 2025 the order tightens (S&P 500 8.0% a year; 40/20/20/20 mix 7.2%; 60/40 6.9%; 30/55/15 and 25/25/25/25 mixes 6.8%), and dropping 2025, a particularly strong year for gold, trims the gold-heavy mixes by 0.2 to 0.3 points a year (S&P 500: 0.1). Long windows, not one hot year, are the fair comparison. For the two-fund baseline see is 60/40 dead? and for gold’s role how much gold belongs in a portfolio.

How did the All Weather portfolio do in 2022?

Badly for a portfolio sold as all-weather: built with real ETFs and held from the first trading day, the 30/40/15/7.5/7.5 mix lost about 18.8% in 2022, roughly as much as stocks. Calendar-2022 total returns from Exent market data; hypothetical $100,000 at year-start weights, no rebalancing during the year:

Sleeve (ETF used)Weight2022 returnDollar changeContribution
US stocks (SPY)30%−18.2%−$5,460−5.46 pts
Long-term Treasuries (TLT)40%−31.2%−$12,480−12.48 pts
Intermediate Treasuries (IEF)15%−15.2%−$2,280−2.28 pts
Gold (GLD)7.5%−0.8%−$60−0.06 pts
Commodities (DBC)7.5%+19.3%+$1,448+1.45 pts
Total100%−$18,832−18.8%
  • US stocks (SPY)
    Weight
    30%
    2022 return
    −18.2%
    Dollar change
    −$5,460
    Contribution
    −5.46 pts
  • Long-term Treasuries (TLT)
    Weight
    40%
    2022 return
    −31.2%
    Dollar change
    −$12,480
    Contribution
    −12.48 pts
  • Intermediate Treasuries (IEF)
    Weight
    15%
    2022 return
    −15.2%
    Dollar change
    −$2,280
    Contribution
    −2.28 pts
  • Gold (GLD)
    Weight
    7.5%
    2022 return
    −0.8%
    Dollar change
    −$60
    Contribution
    −0.06 pts
  • Commodities (DBC)
    Weight
    7.5%
    2022 return
    +19.3%
    Dollar change
    +$1,448
    Contribution
    +1.45 pts
  • Total
    Weight
    100%
    2022 return
    Dollar change
    −$18,832
    Contribution
    −18.8%

The arithmetic: each weight × its return, summed: 0.30 × −18.2 + 0.40 × −31.2 + 0.15 × −15.2 + 0.075 × −0.8 + 0.075 × 19.3 = −18.8%. Long Treasuries did most of the damage, because bond prices fall when yields rise and long bonds fall the most. Our 10-year-bond proxy above (−15.1%) is milder for the same reason. The 2022 year in full is in the 2022 inflation bear market.

Who is the All Weather portfolio wrong for?

Anyone who needs growth over decades and can sit through stock drawdowns, because the price of smoother years was a much smaller ending balance. Counterexample: in the test above, $10,000 in 1972 grew to about $901,000 in the illustrative 30/55/15 mix and about $2.89 million in the S&P 500, more than three times as much. The mix also fails in its own terms when stocks and bonds fall together, as in 2022, and it relies on long Treasuries keeping their role as a hedge. Those it can suit: investors close to spending the money, people who would otherwise sell in a crash, and anyone for whom a −15% year is survivable but −37% is not. Compare lower-risk portfolios and whether bonds still hedge stocks.

  • Limitations: annual data hide intra-year falls; the worst calendar year is not the worst peak-to-trough drop.
  • The comparison uses proxies (S&P 500 for small-cap value, 10-year bonds for long and intermediate Treasuries, gold for commodities) and ignores fees, taxes and costs.
  • Bridgewater’s strategy and ALLW use leverage and different assets, so this test does not describe their results.
  • One period (1972–2025) in one country; a different start date or a different bond environment can change the order.
  • Past results, hypothetical portfolios; not advice and not a forecast.

Add market context with Exent Market intel

Market intel provides market context, cross‑asset analysis, an events calendar and regular written notes. These can inform your review of an existing plan; they do not reliably predict crashes or guarantee better outcomes.

  1. A Risk ON, Neutral or Risk OFF reading of US market risk built from four pillars, with an email when the reading changes.
  2. Cross‑asset correlations across stocks, bonds, gold and commodities, plus market internals and market stretch.
  3. Regular written notes that read ratios such as copper vs gold and stocks vs bonds, an events calendar and 26 Market Intel 101 lessons.

Exent is our product. Educational content, not investment advice.

Frequently asked questions

The retail version from Tony Robbins’s 2014 book is 30% stocks, 40% long-term Treasuries, 15% intermediate Treasuries, 7.5% gold and 7.5% commodities. Bridgewater’s own All Weather fund uses risk parity and leverage, not those fixed weights.

In an unleveraged 1972–2025 test with yearly rebalancing (Exent calculation on NYU Stern data, with substitutions), the illustrative 30/55/15 mix returned 8.7% a year, 4.6% after inflation, with a worst year of −15.1% in 2022. These substitutions do not establish the historical ranking of the named strategies.

Yes. The State Street Bridgewater All Weather ETF (ALLW) launched on Mar 5, 2025 with a 0.85% gross expense ratio. It uses futures and swaps; its notional exposures summed to about 187% of fund value on Oct 23, 2026.

In the 1972–2025 test the illustrative 25/25/25/25 mix (Permanent stand‑in) had the shallower worst year, −8.3% vs −15.1% for the 30/55/15 mix, both in 2022, and a slightly lower return, 8.5% vs 8.7% a year. These substitutions do not establish the historical ranking of the named strategies.

In the same test the illustrative 40/20/20/20 mix returned 9.2% a year with a worst year of −10.3%, against 8.7% and −15.1% for the 30/55/15 mix. It uses the S&P 500 for the small-cap value sleeve, so these substitutions do not establish the historical ranking of the named strategies.

Because 55% sits in Treasuries and bonds fell with stocks as rates rose. With real ETFs, 40% in TLT (−31.2%) and 15% in IEF (−15.2%) made the mix lose about 18.8% in 2022.

Bridgewater’s All Weather is a risk-parity strategy: it balances risk, not dollars, across environments and uses leverage to do it. The fixed-weight retail copy only approximates that idea.

Not over the long run in our test: 8.7% a year against 11.1% for the S&P 500 from 1972 to 2025. It lost much less in bad stock years such as 2008 (+0.7% vs −36.6%).

Sources

  1. The All Weather Story — Bridgewater Associates, Jan 1, 2012
  2. State Street Bridgewater All Weather ETF (ALLW) fund page — State Street Investment Management, Oct 23, 2026
  3. Historical Returns on Stocks, Bonds and Bills: 1928–2025 (histretSP) — NYU Stern School of Business, Aswath Damodaran, Aug 24, 2026
  4. MONEY Master the Game: 7 Simple Steps to Financial Freedom (summary) — AFCPE, The Standard, Jan 1, 2015
  5. Vanguard S&P 500 ETF (VOO) fact sheet — Vanguard, Sep 30, 2026
  6. The Permanent Portfolio (Rowland and Lawson) — Wiley, Jan 1, 2012
  7. Golden Butterfly Portfolio (definition) — Portfolio Charts

Add market context to your plan.

Market intel provides a Risk ON / Neutral / Risk OFF reading from four pillars, cross-asset analysis, an events calendar and regular written notes. Context for reviewing a plan, not a crash forecast.

See what Market intel includesLimited offer67% off your first month

Educational content, not investment advice. Examples are illustrative; past performance does not guarantee future results.

DISCLAIMER: Exent AI is an informational research platform and does not provide investment advice, financial advice, or act as an investment adviser, broker-dealer, or fiduciary. All content, analytics, and model outputs are provided for educational and informational purposes only and should not be relied upon as a recommendation to buy, sell, or hold any financial instrument. Investing involves risk, including the possible loss of principal. Exent AI does not execute trades, hold customer funds, or provide brokerage services. Past performance and hypothetical results are not indicative of future results. Third-party data and AI-generated outputs may be incomplete or inaccurate.