The short version - TLDR:

Copper is at record highs - around $14,000 a tonne (≈ $6.35/lb). The world is structurally short of it. The mining industry's response? Cut investment in new supply. That single fact tells you more about money in 2026 than any Fed meeting.

Start with the number that ends the debate everyone's having. Per Scotiabank's latest published research titled “The Copper Market Is Still Tighter Than You Think” - (53 pages I read so you don't have to)- copper consumption needs to grow just 0.9% this year to keep the market balanced, against a 2.9% average since 1960. Demand could run at a third of trend, a frankly recessionary scenario, and the market still tips short.

So forget the demand brawl - China, EVs, AI. The bears were right about all of it: China disappointed, EVs stalled, the "energy transition" underdelivered. Copper ripped anyway. This was never a demand story. It's a supply story - and the supply side has broken in a way that takes a decade to fix, because the people who could fix it have decided not to.

Below: why record prices aren't buying a single new mine, why the skeptics might be right anyway (and why I don't think they are), what it's worth, and how the smart money is actually playing it.

Start with the thing that breaks the textbook. Every econ course teaches the same loop: scarcity lifts prices, high prices summon supply, the shortage heals. So with copper at records and a shortage in plain sight, capital should be flooding into new mines. Instead, it's walking out the door.

Growth capex is set to average $7.4 billion a year in 2026–28, down 49% from $14.7 billion in 2023–25 - and only ~$6.7 billion a year is planned for 2026–30, versus $17 billion through the 2010s. Investment in new copper is being halved into the teeth of a shortage. Why? Four reasons. The first three are mechanical; the fourth is the one that matters.

It costs a fortune. Grades are falling, the easy ore is gone, and what's left often has no water - Chilean miners now desalinate seawater and pump it thousands of metres uphill. Scotia puts a new mine's capital intensity at ~$23,300 per tonne of annual capacity - about $2.3 billion for a mid-sized project - and lifted its long-term incentive price to $5.00/lb (from $4.50). At $4.50 copper, a typical project earns ~12%, under the hurdle. You have to believe copper stays high for a decade to sign that cheque.

It takes forever. Twenty years from discovery to metal in the developed world, closer to 29 in the US - a commitment that outlives the CEO, the comp plan, and several price cycles.

The jurisdictions are landmines. Exhibit A: Cobre Panama - a finished, world-class mine switched off by the government and left idle, what Scotia calls essentially the world's only material source of spare capacity, dark by political decree. When a government can vaporize a built mine overnight, who's racing to break ground on the next one?

And the capital itself changed - the real story. In a sound-money world the textbook works: high prices, high returns, patient capital, new supply. We don't live there. We live in a financialized, fiat world where the executives who overbuilt at the last peak got their heads handed to them, and the survivors learned the lesson the market beat into them - ambition gets punished, "discipline" gets rewarded. Shareholders want buybacks. CEOs are paid on this year's returns, not on cathode shipping in 2045.

I've spent my career on the deal side of this business, so let me translate the boardroom math: a twenty-year, two-billion-dollar bet in a country that might nationalize it is the last thing a rational team wants on its plate. Far easier - and far better rewarded - to hand the cash back, or just buy someone else's pounds.

The price is screaming BUILD. The incentives are answering: return capital.

That gap is the capital strike - and here's the part the sell-side won't put in a deck: it isn't irrational. It's the rational response to soft money. When your unit of account is being debased and policy turns capricious - tariffs that flip overnight, governments that switch off finished mines - the sane time-horizon for capital collapses. A twenty-year, illiquid bet on a real asset looks reckless next to a buyback you can book this quarter. Cheap, abundant money doesn't fund long-cycle supply; it chases liquid financial returns. So debasement doesn't just dilute the dollar - it raises everyone's discount rate on patient, real-world capital and starves the very investment needed to dig real things out of the ground.

Here's the contrast that frames the whole capital strike. Since the start of 2020, the U.S. money supply has expanded from about $15.5 trillion to $22.6 trillion - roughly 45% more dollars, conjured with a keystroke. Over the same stretch, global copper mine supply grew about 11% - the hard way, through grades, permits, and capital.

You can print dollars. You cannot print copper.

That asymmetry doesn't pump the copper price directly - copper isn't a monetary metal; its price is set by physical supply and demand. What it does is quieter and more consequential: when expanding the money supply is effortless and expanding the metal supply is a twenty-year slog, capital rationally chases the easy game - financial returns, buybacks - over the hard one. The deficit is the downstream result.

It's also durable, not a spike. The relief everyone's banking on - the 2027 wave of restarts (Cobre Panama, Collahuasi, Grasberg, Kamoa) - is real but temporary:

Scotia sees 2027 as peak supply growth for the decade, turning negative by 2030. Nothing's behind it, because nobody funded it - and even if every board reversed course tomorrow, see reason two: the metal wouldn't arrive until the 2040s.

We've seen this movie. Oil producers slashed capital from 2015 to 2020 - every cut applauded as "discipline" - and gifted us the 2022 spike. The cure for low prices is low prices. The cure for high prices is supposed to be high prices. This time someone unplugged the second half of the sentence.