← Jared Dillian hub  ·  Research hub  ·  Research library

Jared Dillian — "It's not because commodities are terrible. It's because commodities have negative carry"

The same five sleeves, argued from the inside out: each one chosen for its cost of carry, its correlation and its checkability rather than its expected return.
2026-SEP-06 · Talking Billions (Bogumil Baranowski, Blue Infinitas Capital) · Jared Dillian (The Daily Dirtnap / Jared Dillian Money) · 53:20 · ▶ Watch · transcript · actionable insights
One-line take: a second book interview two days before the 2026-SEP-08 Excess Returns one, and worth reading separately because this host takes him through the construction rather than through the pitch. The new material is almost all mechanical. Why gold rather than commodities: he actually tried commodity indices first and the returns were poor — "it's not because commodities are terrible, it's because commodities have negative carry because you have to pay for storage… what you're referring to is in a futures curve you're looking at contango." Gold's cost of carry is negligible and "gold mimics the commodity indices over time," so it is the commodity sleeve with the storage bill removed. Gold's correlation to stocks is "zero. And if you go back 25 years ago, it was actually negative." How to express real estate: home equity counts as the sleeve ("not ideal because it's one house in this idiosyncratic geographic area"), otherwise "you can simply buy a REIT ETF"; the REIT indices run back to 1972, and adding real estate raises the return and the Sharpe "by quite a bit." Risk of ruin, scaled by wealth: "wealthy people think about the risk of ruin and middle class people don't" — a $300m Powerball winner buys T-bills and lives on $13.5m rather than putting it in SPY, while the man with $400k in his 401(k) is trying to get to 800, 1.6, 3.2. The ETF love/hate: he traded ETFs "when there were only about 300 in existence," yet the open-end mutual fund's once-a-day NAV is a feature, not a defect — "the more information you're getting on price, the more it affects your decision making" — and he would implement the whole portfolio in open-end funds if a physical-gold one existed. Indexing is reflexive: the host's observer-affects-the-observed framing gets named — "I think the word you're looking for is reflexive" — with the top seven stocks at 35% of the index and the only large-cap managers beating it being the ones overweighting those same seven ("you're just a maniac"); he credits Michael Green with the real work. Different drawdown arithmetic: worst 12%, second worst 9% (2008), third/fourth/fifth about 1%, and the number the calendar year hides — the S&P's 57% peak-to-trough from summer 2007 to March 2009 against the −38% print for 2008. He also concedes 60/40 beats the Awesome Portfolio by about 40 basis points, and that 1929–32 was stocks −89% with bonds +15%. On the tape right now: volatility "about average" but "correlation is very low… it does have a lot in common with the dot-com bubble," immediately fenced with "I'm not saying that the market's going to crash." And the closing conviction is about publishing, not markets: "books succeed when they tell people things they already believe," so he expects this one to sell badly. Timestamps link into the video.
Reading notes. (1) An allocation and psychology episode, like 2026-SEP-08 — the table is deliberately tiny and it is not a duplicate of that one. Only three securities are named at all, and the one carrying an argued position is again BTC (exclude it) — but the reason is new here: he concedes crypto genuinely raised the portfolio's Sharpe in 2019 and excludes it anyway. (2) The asset-class views again have no vehicle attached — gold, cash, bonds, real estate and the index itself are argued for pages with no fund named, so they live in the talking points and the macro themes; no proxy ticker has been invented. The one exception is the real-estate sleeve, where he names the vehicle class ("a REIT ETF") but no specific fund, so there is still nothing to table. (3) The VOO row is a reconstruction from the auto-captions. The ASR renders it "VO" and "V.", but the sentence is "the difference between Vanguard's S&P 500 mutual fund and VOO," which identifies the security unambiguously; it is used only as the ETF half of a mutual-fund-vs-ETF contrast and carries no view on the fund. (4) Fiscal AI is the show's sponsor, read by the host at ~03:07–04:18 and ~52:22–52:43, and 01:27–03:07 is the show's compliance disclosure. None of it is content, none of it is a Dillian pick, and all of it is excluded below (Fiscal AI's founder Braden Dennis is a separate source in this hub). (5) Lehman, Peter Lynch's Magellan Fund, Vanguard, Harry Browne's permanent portfolio and the pod shops (Millennium, Exodus Point, Balyasny) are scenery or history and carry no investment view, so they are not rows. (6) The auto-captions mangled several names; they are corrected in transcript.txt and here (Bogumil Baranowski, Blue Infinitas Capital, The Daily Dirtnap, Harry Browne, Sharpe/Sortino, Nick Maggiulli, Michael Green, the Magellan Fund, Balyasny). (7) "Nozzles, Inc." is Dillian's invented example company.

1. Stocks & names mentioned

Remarks of 2026-SEP-06 on Talking Billions. Stance reflects how each name was framed in this conversation (not a price rating). Three securities are named and only one carries a position — the substance is asset allocation. Research: QT Qualtrim · SA Seeking Alpha · STK Stock Analysis.

TickerNameResearchViewWhat he saidAt
SPYSPDR S&P 500 ETF TrustQT · SA · STKNeutralNamed as the thing a rational rich person will not do, in his risk-of-ruin argument: "let's say you played the Powerball and you won the jackpot and you had 300 million… Would you take $300 million and put it all in SPY?… over time you would probably end up with billions of dollars, but nobody does that. What they do is they take it and put it in T-bills." No view on the fund itself — the point is that the same allocation reads as prudent or reckless depending on the size of the balance sheet behind it.20:17
VOOVanguard S&P 500 ETFQT · SA · STKNeutralThe ETF half of a deliberate mutual-fund-vs-ETF contrast, and the rare case of a professional preferring the older product: "the difference between Vanguard's S&P 500 mutual fund and VOO… the mutual fund, you're getting the NAV once per day. VOO, you can look at your phone every 5 seconds and see where it's trading throughout the day. And that's bad." No view on the fund — the objection is to continuous pricing as an input to behaviour.23:09
BTCBitcoinSTKNegativeExcluded again, but with the concession stated this time: challenged that crypto improved the portfolio, he agrees — "in 2019, if you included Bitcoin, it increased the Sharpe of the portfolio" — and leaves it out anyway. "If you had six asset classes… guess which one you're going to be staring at every day?… Even if it was only 2% of the portfolio… because it's so volatile." The disqualifier is the attention it takes, not the risk-adjusted return: "if you include something that's like an 80-vol in the portfolio, it's going to increase your stress even if it's a tiny part."42:23

"View" is Dillian's framing in this conversation (Positive / Neutral / Negative), not a price rating. The substance of the episode is not in this table: it is how each of the five sleeves — stocks, bonds, gold, cash and real estate, 20% each — gets chosen and expressed, and the deciding criteria are cost of carry, correlation and how often the asset makes you look at it. No fund or ticker is named for any sleeve. Research: QT Qualtrim · SA Seeking Alpha · STK Stock Analysis.

2. Talking points

0:00 The whole programme is a reaction to one number: 50% of his own net worth

5:13 The aha moment was a modelling session, not an insight

6:13 Nick Maggiulli's optimizer lands on the same answer, minus the cash

9:10 The advertised returns are real; almost nobody receives them

10:51 A 16-vol index holding the nation's savings — and only this nation's

11:56 Advisor alpha — Vanguard's own answer, and why he thinks it is not enough

12:52 The drawdown ladder — 12, 9, then about 1 — and why pod shops exist

14:45 The read on today's tape: average vol, very low correlation

16:02 The Magellan gap — a 29% fund whose average holder did far worse

16:42 The life hedge, and the admission that it does not exist

19:46 Risk of ruin is a wealth-level behaviour, not a knowledge gap

22:15 The ETF love/hate — once-a-day pricing is a feature

24:39 "The word you're looking for is reflexive" — indexing changes what it measures

27:16 "Instant diversification" was true — until everyone bought the same thing

28:29 60/40 is incomplete, not wrong — and it actually beats him by 40bp

30:18 Why gold and not a commodity index — negative carry and contango

32:00 How to actually hold real estate — your house counts, otherwise a REIT ETF

33:43 Asked which sleeve to drop, he refuses — and drops into Harry Browne

35:19 The cash sleeve — and a rare disclosure that he is not running it himself

37:51 Gold's correlation to stocks is zero — and the 20% objection comes from 80%-stock people

39:55 Rebalancing: not optimised, on purpose

41:41 Crypto raised the Sharpe — and is still excluded

43:34 Peak-to-trough, not calendar year: the S&P's real 2007–09 number is 57%

45:37 FOMO — the years you win are the years everyone else is being destroyed

47:46 The scenario that breaks all five — he cannot construct one

48:51 His most contrarian belief is about publishing: books sell when they confirm

3. In plain English

What each name is doing in his argument, in everyday language. Only one of the three is a position; the other two are exhibits inside a wider point, and the blocks say so.

BTC — Bitcoin Negative

Bitcoin is the largest cryptocurrency, and the question is whether it deserves a slice of a five-part portfolio. His answer is no — but the interesting part is the argument he concedes on the way.

The standard case for adding it is that it improved the portfolio's risk-adjusted return. The Sharpe ratio measures how much return you get for each unit of price swing, and Bitcoin's gains were large enough that even a small slice raised it. Dillian agrees outright: "in 2019, if you included Bitcoin, it increased the Sharpe of the portfolio." On the measure he uses to justify the entire book, crypto passed.

He excludes it anyway, because the thing he is really managing is not in the numbers — it is how often you look. "If you had six asset classes… guess which one you're going to be staring at every day? You're going to be staring at the Bitcoin. Even if it was only 2% of the portfolio." His shorthand for an asset that moves roughly 80% a year is "an 80-vol," and the objection is simply that "it's going to increase your stress even if it's a tiny part of the portfolio."

So this is a rare case of a strategist overriding his own optimiser on behavioural grounds. If you already own crypto, his position elsewhere is unchanged: count half of it as gold and half as stocks rather than selling.

SPY — SPDR S&P 500 ETF Trust Neutral

SPY is the oldest and most-traded fund that simply holds the 500 largest US companies. Here it is not being rated — it is standing in for "all-in on the stock market" inside an argument about who can afford that.

The test he poses: you win $300 million on the Powerball. Would you put it all in SPY? Mathematically that is the highest-expected-value answer — "over time you would probably end up with billions of dollars." And yet, he says, "nobody does that. What they do is they take it and put it in T-bills and they get 4.5% a year and they live off 13.5 million a year."

The term for what the winner is avoiding is risk of ruin — the chance of a loss so large that you cannot recover from it, however good the odds looked in advance. His claim is that attention to it tracks wealth rather than sophistication: "wealthy people think about the risk of ruin and middle class people don't." The person with $400,000 in a 401(k) is trying to double and double again; the person with $300 million is trying not to go backwards.

The practical takeaway is not about SPY at all. It is that the same holding is prudent or reckless depending on whether you would still be fine after it halved — and that most people never actually ask themselves the question.

VOO — Vanguard S&P 500 ETF Neutral

VOO is Vanguard's exchange-traded fund tracking the same S&P 500 index as its long-standing mutual fund. Same index, same manager, same fees to a rounding error — which is exactly why he uses the pair.

The difference is when you can see the price. A traditional open-end mutual fund is priced once, after the close, at its net asset value; you cannot watch it move. An ETF like VOO trades all day, so "you can look at your phone every 5 seconds and see where it's trading throughout the day."

Most people file that as an improvement. Dillian — who traded ETFs professionally when only about 300 existed — calls it a defect for a long-term holder: "that's bad. The more information you're getting on price, the more it affects your decision making and it causes you to do stupid things." Once-a-day pricing is a friction that stops you acting on noise, and he would build the whole five-sleeve portfolio out of open-end funds if he could. The only thing stopping him is that "there's no open-end mutual fund for physical gold."

No view is expressed on VOO as a fund. The point is that the wrapper you choose changes your behaviour even when the holdings are identical — a consideration that never appears in a fee comparison.


Summary & timestamps derived from the public Talking Billions episode on YouTube (auto-transcript, cleaned, in transcript.txt) for personal study. Not investment advice. © Talking Billions / Bogumil Baranowski / Jared Dillian for source material.