Let's talk about the least honest asset in the world right now, and I mean that with real admiration for how well it's lying.
Gold hit $5,595 an ounce in late January — a record, a full-blown mania top, the kind of number that gets its own headline and its own commemorative mug. It sits around $4,000 to $4,050 today. That's a 28% drawdown on the asset every gold bull spent 2025 insisting was the last honest form of money left standing. Twenty-eight percent. If a growth stock did that, we'd call it a bear market and start writing obituaries for the CEO. Because it's gold, we call it "consolidation," and the same desks that published $5,000 targets in January are now publishing $6,000 targets for December, apparently un-embarrassed by the intervening plumbing.
I want to sit with that whiplash for a second, because it tells you more about where we actually are than any CPI print does.
The case for gold in January was the debasement case — the oldest case in monetary history, and usually the right one. Central banks buying at a record pace. Fiscal deficits that no political coalition anywhere has the stomach to close. A reserve-diversification trade out of dollar assets that started quietly with the BRICS crowd and stopped being quiet sometime around 2024. Real yields compressing. All the ingredients were there, and gold did exactly what gold has always done when the ingredients are there: it ran, hard, until it ran out of new buyers.
Then the Fed didn't cooperate. That's the whole story, condensed. The CME's own pricing has the Fed holding at 3.50–3.75% at the July 29 meeting with better than 85% probability, and every hawkish word out of the building since the spring has been a small withdrawal from the gold trade's premise. You cannot run a debasement narrative through a central bank that keeps declining to debase. Kevin Warsh has spent his tenure so far proving that the market's assumption of an easy, dovish handoff was exactly that — an assumption, not a policy. Gold priced in the assumption. Then it had to unprice it.
Here's the part that should actually worry the people holding the metal, though, and it isn't the 28% correction. It's that gold fell through a live shooting war. Nine straight days of US strikes on Iran, crude briefly north of $112 a barrel, US inventories down to 43 days of supply — the lowest in 45 years — and gold still traded lower on the month before staging its current, tentative stabilization. In every prior era of monetary history I can point to, a genuine Gulf energy shock with a nuclear-adjacent regional power on the other end of it would have sent the metal vertical on pure fear alone. Instead the fear traded through Treasuries, which rallied on the same soft CPI print that pulled Fed hike odds down from near 40% to 30%. Bonds did gold's job. Gold just watched.
That is not a small detail. It suggests something the debasement crowd doesn't want to hear, which is that gold's safe-haven bid and gold's monetary-debasement bid are not actually the same trade, and in 2026 they're pulling in opposite directions at the same time. The war wants gold up. The Fed wants gold down. What we're watching is which force has the bigger printing press, in the most literal sense of the phrase, and right now the Fed is winning.
The World Gold Council will tell you, in the careful language of an institution that doesn't like being wrong twice in one year, that fair value sits somewhere between $3,895 and $4,305 under a base case that assumes one more Fed hike before October. J.P. Morgan and Wells Fargo are still out there with $6,000-plus targets, which means the spread between the cautious institutional view and the bull-desk view is now wider than the entire move gold made in the whole of 2023. Somebody is going to be humiliatingly wrong by year-end, and given how confidently both camps are publishing right now, I'd bet on humiliation being evenly distributed.
What I'd take away from all of this, if I were forcing myself to draw one conclusion instead of hedging like everyone quoted above: the structural case for gold — the debt, the deficits, the central-bank buying, the slow bleed out of dollar reserves — has not gone anywhere and is not going anywhere. That part of the January thesis was correct and remains correct. What broke was the timing bet, the idea that 2026 was necessarily the year the Fed's hand got forced. It wasn't, not yet. Gold spent seven months finding that out the hard way, and the 28% round trip is the tuition.
The structural bulls will be right eventually. They usually are, on a long enough horizon, because sovereign balance sheets do not repair themselves and nobody in Washington is currently trying. But "eventually" is doing an enormous amount of work in that sentence, and every dollar sitting in a gold ETF right now is a dollar betting it can survive however many more quarters "eventually" turns out to require.
That's all I've got. Take it for what unattributed prepared remarks are worth.
