We Sold China the Keys to Our Supply Chain, Then Called It a Threat

April 24, 2026

supply-chain geopolitics rare-earths corporate-incentives manufacturing

The same CFOs who spent the 1990s applauding offshored rare earth processing as a “lean operations win” are now testifying before Congress about Chinese supply chain aggression. I find this genuinely hard to watch — not because the threat isn’t real, but because the framing is so dishonest. We didn’t stumble into rare earth dependency. We sprinted toward it, quarterly report in hand, and called it genius.

The Arbitrage Was the Point

Here’s what actually happened: refining rare earth elements is dirty, expensive, and technically complex. China offered to do it cheaper. Western companies, answerable to shareholders who care about margins and not much else, said yes. Again and again, for thirty years. The executives who structured those deals got bonuses. The shareholders got returns. The workers in those regions of China got the pollution. And everyone went home satisfied.

This wasn’t naivety. The engineers knew what they were building. The procurement teams knew what single-source dependency looked like. The CFOs knew that “lean” was a polite word for “we’ve eliminated all the buffer.” They did it anyway because the incentive structure rewarded the near-term gain and externalized the long-term risk onto governments, taxpayers, and future leadership teams who’d have to clean it up.

What we’re living through now isn’t a geopolitical surprise. It’s a balance sheet coming due.

The Quarterly Earnings Problem Is the Only Problem

I’ve been thinking about this through the lens of software engineering, where I spend most of my time. In systems design, we have a name for what the West did with rare earths: we optimized for throughput and eliminated redundancy. In the short term, that looks like efficiency. In the long term, it looks like a single point of failure sitting underneath your entire semiconductor industry.

The difference is that in software, when you create a single point of failure, you own the consequences pretty quickly. A production outage at 2am has a way of concentrating minds. But when a corporation creates a geopolitical single point of failure, the feedback loop is measured in decades. The people who made the decision are long gone — cashed out, retired, or elevated to boards where they make more of the same decisions at higher abstraction.

This is the core problem, and it has nothing to do with China. A rational adversary, seeing this structure, would exploit it. The question worth asking is why we handed them the lever.

The Rediscovery Theater

Now we’re watching what I’d call rediscovery theater. Congressional hearings. Executive orders about “critical minerals.” Breathless coverage of domestic rare earth projects that have been technically feasible for years but economically unattractive under normal market conditions. Companies suddenly very interested in Australian and Canadian deposits they walked away from in the 2000s because they couldn’t compete on price with subsidized Chinese processing.

The implicit theory seems to be that if we just build some mines and wave enough federal dollars around, we can onshore the supply chain and solve the problem. Maybe. But we haven’t actually reckoned with the mechanism that created it, which means we’ll recreate it somewhere else the moment the next cheap arbitrage appears. The quarterly earnings machine doesn’t care about strategic risk. It cares about this quarter.

In my experience building companies, the hardest thing isn’t identifying the long-term risk. It’s building a structure where people are actually accountable for it. Startups fail at this constantly — you fix the short-term problem in front of you and the long-term architecture slowly turns into rubble. Large corporations fail at it even more reliably, because the timescales are longer and the accountability is more diffuse.

What a Real Fix Would Require

A real fix wouldn’t be a mining subsidy. It would be a serious conversation about whether the fiduciary structure of public companies — legally obligated to optimize shareholder value on short time horizons — is compatible with maintaining strategic supply chains in critical sectors. That’s a harder conversation than building a mine in Nevada.

It would also mean being honest about tradeoffs. Domestically processed rare earths are going to cost more. The clean energy transition is going to cost more. Semiconductors might cost more. These aren’t arguments against doing it — they’re just the actual price of redundancy, which we’ve been avoiding paying for thirty years. The bill is real. Calling it “Chinese aggression” doesn’t change what’s on the invoice.

I don’t think we’re going to have that conversation. It’s easier to frame this as something done to us than something we chose. But I keep sitting with the image of those CFOs at those hearings, and I wonder if any of them, privately, recognize what they’re looking at: the compounded interest on decisions that made perfect sense at the time.

The question I can’t shake is whether there’s any institutional structure — corporate, governmental, or otherwise — that’s actually capable of holding strategic risk over the kind of timescales it operates on. Or whether we’re just going to keep being surprised by the invoices we wrote ourselves.