Gold and commodities: do they belong in your portfolio?

Gold has a marketing department five thousand years old. It's pitched as an inflation hedge, a crisis hedge, and 'real money' — usually loudest right after it's already gone up. The truth is less romantic: gold is a volatile, zero-yield asset with a genuinely useful diversification property and a long history of decade-long dead zones. Both the goldbugs and the dismissers are half right.

What gold actually is (financially)

Gold produces nothing: no earnings, no interest, no dividends, no rent. Its long-run real return is close to zero-to-modest — it roughly holds purchasing power over centuries while stocks multiply it. What it offers instead is non-correlation: gold often rises when stocks fall, when real interest rates drop, or when confidence in institutions wobbles. It's less an investment than a form of insurance that sometimes pays.

The record, without the mythology

• Inflation hedge: unreliable on any horizon shorter than generations. Gold peaked in 1980 at ~$850 and didn't reclaim that price for 28 years — through plenty of inflation. In 2022's 9% inflation, gold was roughly flat.
• Crisis hedge: better. Gold rose in 2008, in 2020's crash, and during various geopolitical shocks. As portfolio insurance against stock-market catastrophe, it has a real (if inconsistent) record.
• Volatility: comparable to stocks — annual swings of 20–30% are normal. 'Safe haven' describes its correlation, not its stability.
• Broad commodities (oil, metals, grain futures): a stronger inflation-year hedge than gold — commodity indexes soared in 2022 — but with brutal long-run drag from storage/roll costs and multi-decade stretches of negative real returns.

If you want exposure, do it cleanly

1. Use a low-cost gold ETF (expense ratios around 0.10–0.25%) in a tax-advantaged account — the IRS taxes physical-gold-backed fund gains at the higher 'collectibles' rate (up to 28%) in taxable accounts.
2. Cap it: 5% of the portfolio is a reasonable insurance sleeve; 10% is the aggressive end of defensible. Beyond that you're making a macro bet, not diversifying.
3. Rebalance on schedule — the entire benefit comes from mechanically selling gold after crisis spikes and buying it after long fades. Unrebalanced gold is just dead weight with mood swings.
4. Skip physical coins and bars unless you value them for non-financial reasons: dealer markups of 3–8%, storage, insurance, and wide resale spreads eat the returns.
5. Broad commodity futures funds: optional, small (0–5%), and only if you understand roll yield. Most investors get sufficient inflation protection from stocks, TIPS, and I bonds instead.

The bottom line

Gold is neither salvation nor scam: it's low-expected-return portfolio insurance with a decent crash record, an overrated inflation record, and a real cost measured in decades of drag. A disciplined 0–10% sleeve, held cheaply and rebalanced coldly, is defensible; more than that is a belief system. If you skip it entirely and just hold stocks, bonds, and some TIPS, you're missing nothing essential — which is, itself, the most honest thing you can say about any asset.

The century-long scoreboard

Zoom all the way out and the asset classes sort themselves unambiguously. One dollar invested in US stocks in 1925, dividends reinvested, grew to something on the order of $15,000 by 2025 in nominal terms. The same dollar in gold grew to roughly $130 — barely ahead of inflation's $18 — and a dollar in broad commodities did worse still, because raw materials generate no earnings, pay no dividends, and cost money to store and roll. Gold's defenders are right about the exceptions: it shone in the 1970s stagflation (up over 1,000% for the decade while stocks went sideways), held up in 2008, and hit records above $2,600 in 2024-25 as central banks bought heavily. Those episodes are why a small allocation genuinely diversifies. But the long gaps between them — gold went from its 1980 peak to 2007 with a negative real return, a 27-year drought — are why the allocation must stay small. An asset that produces nothing can only rerate; it cannot compound.

If you do carry an allocation, implementation matters more than conviction. Use a low-cost bullion ETF such as GLDM or IAUM at roughly 0.1 to 0.18% per year rather than coins in a drawer (illiquid, wide dealer spreads, storage risk) or leveraged miners (which are stocks, not gold). Know the tax quirk: physical-gold ETFs are taxed as collectibles, with long-term gains capped at a 28% rate instead of 15-20%, which argues for holding them in an IRA. Cap the position at 5 to 10%, rebalance it annually like everything else — trimming after spikes like 2024-25's, adding after droughts — and judge it over decades by whether it smoothed the ride, not by whether it beat the stocks it was never going to beat.

Held that way, gold is a small insurance policy with a century of mixed receipts — never the engine, occasionally the shock absorber.

The honest closing test applies to every shiny asset, not just this one: if you would not rebalance into it after a decade of losses, you do not believe in it as diversification — you are chasing its recent chart, and the chart always changes hands at exactly the wrong moment.