Mark Spitznagel: tail hedging and portfolio crash protection

Mark Spitznagel's hypothetical reference allocation puts 3.33% of a portfolio in Universa's tail hedge and 96.67% in the S&P 500, the split Universa used to report its results in client letters in 2018 and 2020. The hedge seeks a large payoff in a sharp crash to offset losses on the equity allocation.

Universa reported $21.3 billion of regulatory assets under management in its Form ADV filed on 31 March 2026; the filing says that figure sums protection size or contributed capital, mostly protection size, so it is not cash invested in options.

Key takeaways

1. The hedge is a sleeve, not the portfolio. Universa's letters pair 3.33% in its tail strategy with 96.67% in the S&P 500, and judge the result on the whole mix.
2. The 3,612% headline applies to the sleeve's capital. In March 2020 the 3.33% mix was reported at +0.4% while the S&P 500 fell 12.4%.
3. Spitznagel's test is compound growth. In Safe Haven (2021), a hedge is only cost-effective if adding it raises the portfolio's compound annual growth rate over time.
4. AQR's researchers dispute the cost. Their 2021 paper found that out-of-the-money index puts gave negative returns in each decade of a 35-year sample to 2020 and preferred trend following on cost.
5. Copio adopts the whole-portfolio growth test. Our analytical standard is higher expected geometric growth after protection costs; a calmer return path alone does not meet it.

What is Mark Spitznagel's tail-hedging allocation?

It is a small allocation to deep out-of-the-money protection held alongside a large equity position, with the hedge judged by its effect on the total portfolio. Universa's March 2018 decennial letter measured a hypothetical portfolio of 3.33% in its tail hedge and 96.67% in the S&P 500 (copy of the letter on Scribd).

The letter said that, over its first ten years, the 3.33% allocation had added 2.6% to the compound annual growth rate of an S&P 500 portfolio, whose own rate was 9.7% (excerpt reposted on TalkMarkets). It described 3.33% as the approximate allocation size Universa recommends in practice, relative to a client's total equity exposure.

The April 2020 letter repeated the method, describing the 3.33% and 96.67% mix as the hypothetical portfolio it recommends to clients (Institutional Investor, 8 April 2020). A January 2023 letter used a 2% allocation instead, according to Bloomberg News (Bloomberg Línea, 3 April 2023).

Reported compound annual returns: March 2008 to February 2018, 12.3% for the 3.33% mix against 9.7% for the S&P 500; March 2008 to March 2020, 11.5% against 7.9%; 15 years to the January 2023 letter, 11.8% for a 2% mix against 9.6%.

Source: 2018 letter (Scribd copy) and TalkMarkets excerpt; Institutional Investor and ValueWalk on the 2020 letter; Bloomberg Línea on the 2023 letter; chart data (CSV). These are hypothetical portfolios built by Universa from its own net returns, and Bloomberg reports the 2023 calculation assumes quarterly rebalancing. No investor necessarily held these mixes, and client programmes are tailored.

Universa's chief operating officer Brandon Yarckin told finews.com in August 2025 that clients "often shift from a 60/40 portfolio to something like 70 percent equities plus a tail hedge" (finews.com, 25 August 2025).

What does Safe Haven argue?

Safe Haven: Investing for Financial Storms (Wiley, August 2021, 240 pages) argues that the only purpose of risk mitigation is to raise the rate at which wealth compounds (Wiley). Spitznagel's first two principles are that investing happens sequentially, with returns multiplying through time, and that the goal is to maximise realised wealth, equivalent to the compound annual growth rate (Chapter 1 excerpt, pp. 14-15).

From this he derives a testable rule: "If a strategy cost-effectively mitigates a portfolio's risk, then adding that strategy raises the portfolio's CAGR over time" (p. 20).

The geometric argument is set out in his January 2019 paper Amor Fati. The compound return is the exponential of the average of log returns, and because the logarithm curves downward it penalises large losses more than proportionally (Universa, Amor Fati, p. 3). A strategy can therefore raise the arithmetic average and still lower compound growth if it adds deep losses.

The book's framework discusses Daniel Bernoulli's treatment of risk and the St Petersburg merchant, who insures a ship cargo even though the premium exceeds the expected loss (book index; finews.com).

Spitznagel is explicit about the limits. He writes that what he does "is not to be attempted by nonprofessionals" and that the book is "not a 'how-to' book" (excerpt, p. 4).

Did Universa make 3,612% in March 2020?

A Universa client letter reported a net return of 3,612% in March 2020 for its Black Swan Protection Protocol, "based on required invested capital at the start of the year", and 4,144% year to date (Yahoo Finance, 9 April 2020). That denominator is the capital needed to hold the hedge positions, not the size of the portfolio being protected.

March 2020: hedge sleeve +3,612% on required capital; 3.33% Universa mix +0.4%; 3.33% tail-index mix -11.4%; S&P 500 -12.4%. Source: Yahoo Finance, 9 April 2020; Institutional Investor, 8 April 2020. The portfolio figures are hypothetical mixes calculated by Universa itself, for one month. They do not show the premiums paid in the preceding years.

Critics focused on that basis. In 2023 Boaz Weinstein of Saba Capital and Derek Kaufman, formerly of Citadel, accused Universa of presenting returns selectively, and Cliff Asness of AQR shared Weinstein's complaint, Bloomberg reported (Bloomberg Línea). Bloomberg also reported that the January 2023 letter likened a 2% allocation to an annuity paying 114% a year, while another letter put that figure at 79%. Yarckin replied that Universa could not comment on returns and that what it does is raise a portfolio's rate of return by reducing its systematic risk.

How big is Universa?

Universa's Form ADV, filed with the SEC on 31 March 2026, reports regulatory assets under management of $21,349,131,206 across 45 accounts, of which $7.9 billion is attributable to non-US clients (Form ADV, Item 5.F).

Its notes state that this figure is "computed as the summation of either the protection size (the amount of equity market risk that the client seeks to protect) or the contributed capital" of each vehicle, "with protection size being the most common approach", as of 31 December 2025. For the Black Swan Protection Protocol, the minimum commitment of $150 million "refers to the protection size".

The headline size therefore mostly describes how much equity exposure clients protect, not cash committed to options. Bloomberg reported in 2023 that a regulatory filing defined Universa's then $16.4 billion of assets as the amount of equity market risk a client seeks to protect. Yarckin told finews.com that the firm is "not structured to deal directly with retail".

What is Nassim Taleb's role at Universa?

Universa calls Nassim Nicholas Taleb its Distinguished Scientist and says he and Spitznagel began formal tail hedging for client portfolios over twenty years ago (Universa, About us). They ran the hedge fund Empirica from 1999, before Spitznagel launched Universa in 2007 (Business Insider via Yahoo, 7 September 2026).

In Safe Haven, Spitznagel writes that Taleb "would later become Scientific Advisor at Universa"; Taleb wrote the book's foreword (Chapter 1 excerpt, p. 6; contents). Taleb's own barbell, 85-90% in Treasury bills and the rest in speculative bets, is examined in our Nassim Taleb barbell strategy study.

What do AQR and other critics say about tail hedging?

AQR's researchers argue that buying crash insurance has a high long-term cost. In a 2020 AQR paper, later published in The Journal of Portfolio Management, McQuinn, Thapar and Villalon wrote that "the tail insurance strategies with the largest wins in crash months are likely ones that in good times lose all or most capital allocated to them, perhaps many times over" (AQR). They argue that slow, protracted drawdowns do more damage to long-term wealth than short crashes.

A February 2021 AQR paper found that a strategy of buying 5% out-of-the-money one-month index puts lost money from January 1985 to March 2020, and that out-of-the-money puts gave negative returns in each decade of that sample despite 1987, 2008 and 2020 (Ilmanen, Thapar, Tummala and Villalon, Journal of Systematic Investing, pp. 2-3).

It reported that the CBOE Eurekahedge Tail Risk Index, a peer group of tail managers, returned around -2% a year since 2008 and around -8% a year in the 2010s (p. 4). The authors concluded that "the cost advantage favors Trend over Put" (p. 1). The AQR article in this series covers that position.

An April 2024 paper argues that "most of the protection that bonds provide in a 60/40 (or similar) portfolio comes simply from not owning equities" (Lagnado, Universa, p. 3). It also tested the leveraged 60/40 portfolio that Asness has defended: from 1926 to February 2024 it reports 10.2% a year for the S&P 500 and 10.7% for 60/40 with 62.5% leverage, a gap it says is not statistically significant (pp. 4-6). The same critique of leveraged bond exposure bears on Ray Dalio's risk parity approach.

Neither side's evidence covers every kind of fall. Bloomberg noted that during the 2022 rate-driven sell-off, volatility stayed too low for most tail strategies to pay off.

Our put option guide checks a published 1986-2018 benchmark in which an S&P 500 portfolio protected with monthly puts about 5% out of the money cut the maximum month-end drawdown from 50.95% to 38.92% but compounded at 6.64% a year, against 9.80% for the index, before transaction costs and taxes. Those puts were about 5% out of the money rather than deep out of the money, so the benchmark cannot settle the performance of Universa's private strategy.

When does tail hedging improve geometric growth?

Copio's research objective is maximum expected long-term geometric growth after costs, within the investor's spending needs and capacity for loss. A tail hedge earns its place by improving that whole-portfolio outcome. We share Spitznagel's focus on the hedge's portfolio effect: preserve productive exposure, limit damaging losses and reinvest realised protection gains. Our research thesis is that this can be a route to compounding through active downside protection; a smaller drawdown alone is insufficient.

Universa's reported March 2020 blend gained 0.4%, but that hypothetical monthly result cannot establish the best long-term allocation. The comparison needs the equity allocation, premiums, trading costs, tax treatment and reinvestment rule, across both crash and quiet periods. An overpriced or poorly matched hedge can reduce growth even when it pays during one sell-off. Our resilient portfolio allocation guide makes this net compounding test the standard for comparing protection with cash reserves and reduced exposure.

What can an individual investor learn from Mark Spitznagel?

Universa's funds are institutional, and he discourages amateurs from copying his trades. Questions for evaluating protection:

• Do I judge my protection by the compound return of the whole portfolio after its costs, or by how it performed in one bad month?
• When I read a hedge return, what is the denominator: premium, required capital or the protected portfolio?
• Is my defensive allocation a hedge, or simply less equity?
• Which falls does my protection cover: a sudden crash, a slow bear market or a 2022-style fall in stocks and bonds together?
• Could I keep paying for protection through years of small losses, and do I have a rule for reinvesting a payoff?

Our analysis of crash-protection costs and payoffs examines how premium budgets, renewal and reinvestment rules affect such a sleeve, and our study of past crashes shows how each asset class behaved.