Safe Haven + The Dao of Capital

Mark Spitznagel · 2013–2021

Thesis

The investor who wins over time is the one who accepts small immediate losses for a later advantage, recognizes when central banks have distorted markets, and protects against crashes cheaply, because a portfolio's compound return depends far more on avoiding large losses than on its average return.

The main ideas

Shi, positional advantage, beats li, immediate gain. The best investments take an indirect, roundabout path.
Nature's model is the conifer, which grows slowly in poor ground and wins after the forest fire.
Production, and investing, gain by going roundabout: building capital first and harvesting later.
Central banks distort markets. A ratio of market value to net worth shows when the distortion is extreme.
What matters is the geometric, compound return, not the arithmetic average. Large losses destroy the compound return.
A good safe haven raises the compound return of the whole portfolio. Most popular havens do not; cheap crash insurance does.

Shi and li

In The Dao of Capital (2013), Mark Spitznagel, founder of the hedge fund Universa Investments, borrows two terms from Chinese strategy. Li is the direct, immediate advantage: the quick gain, the obvious move. Shi is positional advantage: accepting a weaker position now in order to be strongly placed later. His examples come from the crossbow, from the board game weiqi, known in the West as go, and from Clausewitz's indirect strategy in war.

Spitznagel draws on the Laozi for the attitude he wants: one gains by losing and loses by gaining. He takes from it the idea of wu wei, non-coercive action, glossed in the book as not doing: waiting until the moment for action has come instead of forcing it.

The mechanism is time. A strategy that sacrifices immediate results to build position compounds into a larger advantage than one that grabs every available gain, but only for an investor patient enough to accept looking wrong in the meantime.

Implication: be willing to accept small, steady losses or underperformance if they buy a position that pays off decisively later. Judge such a strategy over a full cycle, not a year.

The conifer

Spitznagel's model from nature is the conifer. In a mature forest, the fast-growing broadleaf trees dominate the fertile ground, and conifers are pushed to the poor, rocky margins, where they grow slowly. But when wildfire sweeps through, the broadleaf trees burn, and the conifers, which survive or reseed quickly, repopulate the cleared land. The fire reallocates the forest's resources to the patient trees.

He applies the image to markets. In long booms, aggressive investors who take every immediate gain flourish; patient investors who hold cash or protection look foolish. Crashes are the market's wildfires, and they transfer resources to those who were positioned for them.

The mechanism is that resources flow, after a disruption, to whoever is positioned to use them. The patient strategy pays not continuously but at the moments of disruption.

Implication: position yourself to be among those with resources when others are forced to sell, accepting slower growth in the meantime.

The roundabout path

The Austrian economist Eugen von Böhm-Bawerk called it Produktionsumweg, roundabout production: instead of producing goods directly, invest first in tools and capital that make later production far more efficient. Spitznagel adds the time preference of the Austrian school, the idea that how much people discount the future shapes how much roundabout investment a society undertakes.

His business example is Henry Ford, whom he calls the roundabout entrepreneur: Ford reinvested heavily in the means of production, accepting lower immediate returns, to build a far more productive enterprise.

He traces the same logic through the Austrian school more broadly, from Carl Menger to Ludwig von Mises, and through Frédéric Bastiat's distinction between the seen and the unseen effects of economic decisions.

Implication: favor investments, in businesses and in your own capabilities, that build productive capacity over those that maximize this year's output.

Distortion

Spitznagel argues that central banks, by holding interest rates artificially low, distort the signals that would otherwise guide roundabout investment, encouraging speculation and pushing asset prices above their natural level.

To measure the distortion, he borrows from forestry. The nineteenth-century forester Martin Faustmann calculated whether land was worth planting by comparing its expected value from timber with its replacement value; above one, invest; below one, do not. Spitznagel applies the same ratio to the stock market as a whole, comparing the market value of corporate equity with corporate net worth. He calls it the Misesian Stationarity index, and reviewers have compared it with Tobin's q. A value near one indicates an undistorted market; a high value indicates distortion and a market vulnerable to a crash.

When the index is high, his strategy is tail hedging, protection against crashes, which he calls Austrian Investing I and describes as really central bank hedging.

Implication: when market values run far above the underlying net worth of the businesses, increase protection rather than exposure.

Arithmetic and geometric returns

Safe Haven (2021) restates the argument in the language of mathematics. Investors usually think in arithmetic averages: the average return across possible outcomes. But an investor experiences returns in sequence, compounding over time, and the relevant measure is the geometric average, which measures the growth of capital through time.

The two diverge because large losses hurt compound returns more than equal gains help them. Spitznagel calls the difference a volatility tax. His illustration is a dice game. On four faces you gain 5 percent, on one you gain 50 percent and on one you lose 50 percent. The arithmetic average is a gain of about 3.3 percent a roll; but a player who keeps rolling, staking everything each time, steadily loses money, because the large loss compounds against all the gains. He calls the danger the logarithmic falls: avoiding the plunge matters more than collecting more gains.

The mechanism, related to what Nassim Taleb, who wrote the foreword, calls non-ergodicity, is that an individual living through one sequence of outcomes does not receive the average of all possible outcomes.

Implication: judge every investment by its effect on the compound growth of your whole portfolio, and give large losses far more weight than their probability alone suggests.

Safe havens

A safe haven, in Spitznagel's definition, is valuable only if it is cost-effective: it must raise the compound return of the portfolio, not merely reduce its volatility. He sorts havens into three kinds: stores of value, such as gold and cash; alpha havens, assets expected to outperform in a crisis; and insurance, which pays off sharply in a crash.

Testing them, he finds that most popular havens fail. Treasury bonds protect, but cost too much in lost returns to raise the portfolio's compound growth; gold is found wanting. Diversification across risky assets fails because correlations rise in crashes. Only the insurance kind, a small allocation to protection with explosive payoffs in a crash, comes out as cost-effective, because it allows the rest of the portfolio to take more risk safely.

The mechanism is that a small, cheap position that pays off many times over in a crash removes the large loss that drags down compound returns.

Implication: test any hedge by asking whether the whole portfolio grows faster with it than without it, after its cost, not whether it feels safe.

Limits

Other researchers find tail hedging too costly. The investment firm AQR, in papers such as "Chasing Your Own Tail (Risk)," argued that systematically buying crash protection costs more over time than it saves, and that other strategies protect more cheaply; Spitznagel's supporters reply that AQR tested different options.
The strategy's best-known results are disputed. Universa's reported gains in March 2020 were measured on the capital the hedge required, not on assets under management, and the investors Boaz Weinstein and Derek Kaufman accused the firm of presenting its figures selectively.
The distortion signal has been early. Spitznagel has warned repeatedly of an impending crash since The Dao of Capital was published in 2013, and markets rose for long periods in between.

The Black Swan + Antifragile + Skin in the Game + The Bed of Procrustes — the theory of rare events, fragility and ruin on which this strategy rests.
LEAPS — the long-dated options used in hedging strategies of this kind.
Devil Take the Hindmost — the history of the distortions and crashes the strategy is built to survive.