Invest for what you keep
Subscribe

Choose · what to own · Decision 03 of 03 · Diversifiers

A real diversifier cuts the worst drawdown at five percent of the portfolio; an equity flavor deepens it

Five percent of a 60/40 moved into each candidate, 1985 to 2025; then bitcoin sized from one percent to twenty since 2015.

+2.3ptsShallower worst drawdown with 5% in trend following, and +0.34 points a year of return. Correlation to the 60/40: -0.02.
+1.9ptsShallower with 5% in gold, for −0.11 points a year of return. Correlation: 0.00.
−0.3ptsCommodities: no help with the drawdown and −0.14 points a year. A 150-year case on paper, not paid in the last 40.
−1.5ptsDeeper with 5% in large value stocks: an equity flavor is not a diversifier, whatever its label says.
Two bar panels, 1985 to 2025: what moving 5% of a 60/40 into each candidate did to annualized return and to the worst drawdown. Trend following +0.34 points of return and +2.3 points shallower drawdown; gold −0.11 and +1.9; commodities −0.14 and −0.3; large value and large growth add return and deepen the drawdown.
60/40 of Shiller stocks and 10-year Treasuries with 5% carved out pro rata into each candidate, rebalanced monthly, nominal, 1985 to 2025. Trend: AQR time-series momentum plus the one-month bill. Commodities: AQR equal-weight index. Gold: LBMA.
+1.3pts a yearReturn added by 2% bitcoin in a 60/40, rebalanced monthly, 2015 to 2026, for +0.7 points of drawdown.
36%What a 1% slice grew into when never rebalanced, taking the worst drawdown from 19% to 41%.
75%Bitcoin's own worst fall over the period, with 72% annual volatility and a 0.20 correlation to stocks.
Line of a 60/40 with 0, 1, 2, 3, 5, 10 and 20 percent bitcoin, rebalanced monthly, 2015 to 2026: annualized return against worst drawdown. From 8.7% a year and a 19% drawdown with none, to 10.0% and 19% at 2%, 15.2% and 23% at 10%.
60/40 with the labeled share in bitcoin, carved out pro rata and rebalanced monthly, 2015 to 2026. Month-end Coinbase prices via FRED (CBBTCUSD). Eleven years is a short record, and the card says so.
Verdict

Trend for the drawdown, gold for the regime, bitcoin at a size you will actually trim after it triples. Commodities on faith in the long record, or not at all. Anything that adds return and deepens the drawdown at 5% is equity in a costume.

Strategy cards

The 3 cards behind this decision

  1. Trend followingEvidence supports Trend earns its keep exactly when a 60/40 portfolio needs it most, and it charges for that insurance in the calm years between; that trade is worth making, not avoiding. 30–120 bps a year · section 1256 · Trend following made money in all five of the worst US equity years since 1985
  2. BitcoinEvidence mixed Bitcoin is a sizing and rebalancing question, not a belief question: one to two percent, rebalanced, has been a cheap option; any size you would not trim after it triples is too much. 12–150 bps a year · mixed · card only, read coming
  3. CommoditiesEvidence mixed Commodities are the diversifier that works over 150 years on paper and has not paid a 60/40 holder in the last 40: hold them for the inflation you fear, not the record you have seen. 25–95 bps a year · mixed · card only, read coming

A card states what a strategy is, its evidence with period and source, when it worked and failed, what access costs, its tax profile, capacity and behavior demand, and a verdict. Open cards carry no evidence yet and are published only when filled.

1 short read in this decision, about 4 minutes, below.

Start here1 read · 4 min

The few ideas everything else rests on.

  1. Trend following made money in all five of the worst US equity years since 1985

    AQR's diversified trend strategy was positive in every one of the five worst US equity years since 1985, averaging +21.6% while stocks fell 6 to 39%.

    4 min
Concepts used here
Section 1256 contractTerm structure
Concepts

Terms this topic relies on

All concepts
Term structure