How David Bellairian Sizes Gold in a Portfolio: Obligations First, Percentages Second
I’m David Bellairian, founder and CEO of Glencore Associates, a private investment and asset management firm in Los Angeles. In an earlier essay I explained why the firm holds several asset classes at once and what job each one does. The question I get most often afterward is about gold: not whether to own it, but how much. This essay is the answer, and the reasoning behind it.
Start with the job, not the forecast
Most gold arguments are forecasts in disguise. Someone is bullish on inflation, or bearish on the dollar, or worried about a particular central bank, and gold becomes the expression of that view. I don’t size gold that way, because I don’t think I can forecast any of those things reliably, and a position built on a forecast has to be unwound when the forecast changes.
At Glencore Associates, gold has one job. It is the reserve that does not depend on a counterparty, a functioning market, a balance sheet, or anyone’s promise to pay. Everything else the firm owns, including the income real estate and the concentrated stock book, relies on some chain of institutions working the way they are supposed to. Gold in a vault relies on nothing. That is the whole thesis. Once the job is defined that narrowly, the sizing question gets much easier.
The sizing rule
I size gold by asking one question: how much reserve would I need to keep the rest of the portfolio intact through a bad stretch without being forced to sell anything at the wrong time?
That means thinking in terms of obligations, not percentages. Property has taxes, insurance, and repairs whether or not tenants pay on time. A concentrated stock book can get cut sharply in a drawdown, especially when the 7-8% sell rule is doing its job and positions are being closed one after another. Private holdings need patience, and patience is only possible if nothing else is screaming for cash. The reserve exists so that none of those pressures ever becomes a forced decision.
So I work backward. I estimate what the portfolio’s fixed obligations would be across a period of poor conditions, add a margin for the things I know I have not thought of, and hold gold against that number. The percentage that falls out of that exercise is the allocation. I don’t start with “ten percent” or “five percent” and then justify it. The percentage is a result, not an input.
Why I don’t rebalance gold like a stock
A lot of allocation advice treats gold as another line to rebalance quarterly: trim it when it runs, add when it falls. I don’t do that, for two reasons.
First, it quietly turns the reserve back into a trade. The moment you’re selling gold because it went up, you have reintroduced the forecast you were trying to remove. Second, the reserve is sized against obligations, and obligations don’t move with the gold price. If gold rallies, my obligations did not shrink; the reserve just got more comfortable. If gold falls, the obligations did not grow; the reserve got thinner and I may need to add. The only rebalancing signal I respect is a change in what the portfolio owes, not a change in what gold is worth.
In practice, that means gold at Glencore Associates is reviewed, not traded. I revisit the obligation estimate when something in the portfolio changes materially: a property is bought or sold, the stock book grows, a new business is added to the private book. Those events change the number. Headlines don’t.
Physical, not paper
I’ve said this before and I’ll keep saying it because it matters to the sizing. A gold fund is a claim on gold that is settled through the same financial system the reserve is meant to be independent of. In an ordinary year the distinction is academic. In the year the reserve is actually needed, it is the entire point. So the firm holds metal, in custody it controls, and I size it knowing that a bar cannot be lent out, rehypothecated, or gated at redemption.
That choice does come with costs: storage, insurance, and the discipline of verifying what you hold. I treat those the way I treat property insurance: a cost of owning the asset properly, not a reason to own it improperly.
What this is not
This is not a claim that gold will outperform. Over most long windows, equities have, and I expect they will again. Gold at Glencore Associates is not competing with the stock book. It is the thing that lets the stock book be aggressive and lets the private book be slow. A reserve that earns nothing but holds when everything else is falling is doing exactly what it was bought to do.
If you take one idea from this: decide what job gold has in your own structure before you decide how much to hold. If you can’t describe the job in one sentence, the percentage will never feel right, because it isn’t answering a real question.
This essay describes how one private firm allocates its own capital. It is not investment advice, and nothing here is an offer or solicitation. Glencore Associates does not manage capital for outside investors.
— David Bellairian, Founder & CEO, Glencore Associates | Founder, AIDiscover | davidbellairian.com
More from David Bellairian: How I Structure a Five-Asset-Class Portfolio (Medium) · Ray Dalio’s “Holy Grail” of Investing (Substack) · Business Book Recommendations (Glencore Associates) · Essays on Investing, AI, and Building (Kindle) · davidbellairian.com
Comments