The story of Chinese economic expansion is usually told as a mining story — Belt and Road, African resource extraction, port deals. That framing misses the more consequential half. China isn’t primarily buying mines. It’s buying smelters, refineries, and chemical processing facilities. It’s buying the midstream.

The distinction matters enormously. A mine produces ore. Ore requires processing before it becomes a usable industrial input. The country that controls the processing controls the supply chain, regardless of who owns the land title. China understood this twenty years ago and has been systematically acquiring midstream capacity across every critical mineral supply chain.

Craig Tindale’s copper example illustrates the mechanism precisely. Chinese copper smelters have been offering Chilean and Peruvian mines a processing bounty — paying $100 per tonne to smelt copper at a loss. South Korean copper refineries need $50-75 per tonne to operate profitably. They cannot compete with a state-capitalist actor absorbing losses as a cost of strategic positioning. South Korean refineries lose market share. Chinese smelters gain it. Over time the alternative processing capacity disappears and the dependency becomes structural.

This is not trade competition. It is deliberate industrial warfare conducted through commercial mechanisms, exactly as the 1999 unrestricted warfare doctrine prescribes. The weapon is a below-cost processing contract. The objective is permanent midstream control.

Chinese state banks finance this at sovereign cost of capital — effectively zero real return requirement — because the return is measured in geopolitical leverage, not financial yield. No Western private equity fund can match that financing structure. The only credible response is state capitalism meeting state capitalism — which is exactly what Hamilton prescribed two hundred years ago.

Robert Friedland has spent decades actually building mines and understands the physics of the business in a way that most analysts do not. When he talks about copper supply, it’s worth listening — not because he’s bullish on his own assets, which he always is, but because he has earned that right the hard way.

Craig Tindale referenced conversations with Friedland in his Financial Sense interview to make a specific and sobering point about copper supply math. Friedland has just brought a major new copper mine into production in the DRC — one of the largest new copper operations in the world. Tindale’s assessment: we would need five or six mines of equivalent size coming online every single year just to keep pace with projected copper demand through 2030.

We are not building five or six major copper mines per year. We are not building one. The global pipeline of copper projects in advanced development is a fraction of what the demand trajectory requires, and that pipeline faces the full gauntlet of permitting delays, ESG financing constraints, community opposition, geopolitical risk, and the fundamental physical reality that a copper mine takes roughly nineteen years from discovery to full production.

Friedland’s Congo mine is genuinely significant. It is also a single data point against a demand curve that looks like a wall. The hyperscale data centers, the EV fleet, the grid electrification, the defense manufacturing — all of it runs on copper, and the supply response has barely begun.

The investment case for copper is not complicated. It is supply constrained against demand that is structurally mandated. The question isn’t whether copper prices will reflect this constraint. They will. The question is timing — and the timing is being driven by physical realities, not financial models.

For thirty years we told our kids to stay out of the trades. Get a college degree. Work in an office. The dirty jobs — welding, machining, electrical work, process operations — those were for people who didn’t have options. That narrative is about to reverse violently, and the people who understand it early will be positioned very differently from those who figure it out late.

Craig Tindale made the point without sentiment: we are going to need an enormous number of blue collar workers, and we don’t have them. The Colorado School of Mines needs to double in size. Every industrial training program in the country is undersized for what’s coming. The skills to safely operate a zinc smelter, manage a sulfuric acid processing line, commission a copper refinery — these have been allowed to atrophy for a generation because we decided we didn’t need them. We need them now.

You cannot re-industrialize with white collar workers alone. The physical processes that underpin a functioning industrial economy require people who can operate and maintain physical equipment, troubleshoot process failures in real time, and apply the kind of embodied knowledge that doesn’t exist in a spreadsheet or an AI model. When a valve fails at 2 AM in a processing facility, you need someone who knows what that valve does, why it failed, and how to fix it without shutting down the entire line.

The wage implication is already playing out. Electricians, pipefitters, and industrial mechanics are commanding salaries that would have seemed implausible a decade ago. That trend has years to run. The most valuable workers in the re-industrializing economy will be the ones who can actually make things. That’s not a prediction. It’s already happening.

When Trump moved aggressively on Venezuela and positioned military assets near the Strait of Hormuz, most commentary focused on the obvious: oil, sanctions, regional power projection. That’s the surface reading. The deeper reading is about China’s energy vulnerability and the logic of conjoined-twin warfare.

China controls the midstream of Western critical mineral supply chains. That’s their leverage. But China has its own chokepoint: energy. The Chinese economy is massively dependent on oil imports, and the majority transit the Strait of Hormuz. China cannot secure its own energy supply lines militarily in the Persian Gulf.

Venezuela was a Chinese client state with significant oil reserves. Iranian oil flows to China in volume. If the U.S. controls both — through sanctions enforcement or military positioning — it holds a counter-lever against Chinese rare earth coercion. You restrict our gallium, we restrict your tankers. The logic is brutal and simple.

Craig Tindale frames this as a classic unrestricted warfare equilibrium: each side applies pressure at the other’s soft points to prevent the balance from tipping too far. It’s not about winning outright. It’s about maintaining enough mutual vulnerability that neither side pulls the trigger on full economic warfare. Conjoined twins trying to choke each other — neither can kill the other without dying themselves.

The investment implication: energy geopolitics and critical mineral geopolitics are no longer separate analysis tracks. They are the same track. The companies, commodities, and regions sitting at the intersection of Middle East energy, African critical minerals, and strategic shipping routes are not just commodity plays. They are positions on the board of the most consequential geopolitical game of the next twenty years.

On April 1, 2026, the Ninth Circuit handed California employers a meaningful win in O’Dell v. Aya Healthcare Services, Inc., No. 25-1528. The court reversed a Southern District of California ruling that had used a procedural doctrine—non-mutual offensive collateral estoppel—to invalidate arbitration agreements for more than 250 opt-in plaintiffs based on two prior arbitrator decisions finding the employer’s agreements unconscionable.

The decision matters because it closes off a tactic that could have allowed a small number of adverse arbitration outcomes to wipe out an employer’s entire arbitration program in a collective or class action. For California employers who have invested in arbitration agreements as a risk-management tool, the ruling reaffirms a foundational principle: each arbitration agreement stands or falls on its own terms.

Here are five practical takeaways for California employers.

1. Arbitration agreements must be evaluated individually, not in bulk.

The district court in O’Dell allowed two arbitrator rulings—each involving a different employee and a different arbitrator—to preclude Aya from enforcing hundreds of other arbitration agreements signed by opt-in plaintiffs. The Ninth Circuit rejected that approach, holding that the Federal Arbitration Act requires courts to enforce arbitration agreements according to their terms and that applying non-mutual offensive collateral estoppel in this context conflicts with the FAA’s core principle of consent.

For California employers facing collective or class actions—particularly under the California Labor Code —this means that a single adverse arbitration ruling against one employee does not automatically bind the employer as to every other employee. Each agreement is its own contract, and each employee consented to arbitration on an individualized basis.

2. Delegation clauses remain powerful.

The Aya arbitration agreements contained delegation clauses that sent questions of the agreement’s validity to the arbitrator rather than the court. That structure was central to what happened next: when the initial four plaintiffs challenged the agreements, arbitrators—not judges—decided the unconscionability questions.

California employers should consider having clear, express delegation clauses in their arbitration agreements. A well-drafted delegation clause keeps gateway issues (validity, enforceability, scope) in the arbitral forum and reduces the number of issues courts can use to deny a motion to compel. O’Dell is a reminder that the architecture of the agreement—not just the substance—matters.

3. Split arbitrator results do not doom the rest of the program.

In Aya’s case, the arbitrators divided two-to-two on whether the agreements were unconscionable. The district court gave preclusive effect only to the two decisions finding the agreements invalid—turning a split into a sweeping defeat. The Ninth Circuit correctly observed that this approach transformed individualized arbitrations into something like an unauthorized bellwether class action that the parties never agreed to.

The lesson for employers: inconsistent arbitrator outcomes are a fact of life in large workforces. O’Dell confirms that adverse decisions in a handful of arbitrations are not a basis for invalidating agreements with other employees. Employers can continue to enforce their agreements on an individual basis, even when some arbitrators reach unfavorable results.

4. The ruling does not cure substantively flawed arbitration agreements.

O’Dell is a procedural win. It does not insulate employers from the substantive unconscionability challenges that remain the bread-and-butter of plaintiffs’ attacks on arbitration in California. The Armendariz factors can still apply, and California courts continue to scrutinize fee-splitting provisions, venue clauses, limitations on remedies, discovery restrictions, and presentation issues—as recent cases like Fuentes v. Empire Nissan (arbitration agreement in nearly unreadable font) illustrate.

Two of the four arbitrators in O’Dell found Aya’s agreements unconscionable because of their fee and venue provisions. That is a reminder that even a structurally sound arbitration program can be undone by one-sided terms. Employers should not read O’Dell as a reason to delay a substantive review of their agreements.

5. California employers should use this as a reminder to audit their arbitration programs.

Now is a good time for California employers to conduct a focused audit of their arbitration agreements. Key items to review include:

• Consider implementing a delegation clause assigning gateway issues to the arbitrator.
• Fee allocation consistent with Armendariz—the employer bears the costs unique to arbitration.
• A reasonable, neutral venue that does not impose unfair burdens on the employee.
• Mutual remedies and mutual coverage (both sides bound to arbitrate).
• A severability or savings clause that allows the agreement to survive if a specific provision is found unenforceable.
• Presentation that passes a readability check—legible font, clear headings, separate signature, and no buried consent.
• A carve-out framework that properly addresses PAGA and sexual harassment claims consistent with current California and federal law.

The Ninth Circuit has reaffirmed that arbitration programs built on individualized consent will be enforced on an individualized basis. Employers who pair that protection with a substantively defensible agreement are in the strongest possible position heading into the remainder of 2026.

The post Five Takeaways for California Employers from the Ninth Circuit’s Arbitration Ruling in O’Dell v. Aya Healthcare Services appeared first on California Employment Law Report.

The electric vehicle revolution has a supply chain problem the auto industry’s PR departments prefer you not think about carefully. The batteries that make EVs possible require lithium, cobalt, nickel, and manganese in quantities that dwarf current Western production capacity — and the processing of those materials is overwhelmingly controlled by China.

Lithium is mined in Australia, Chile, and Argentina. But the processing — converting spodumene concentrate into battery-grade lithium hydroxide — is dominated by Chinese refiners. Cobalt comes primarily from the DRC, where Chinese companies have secured the majority of mining rights and process most of the output. High-grade battery nickel processing is again concentrated in Asia, with Chinese firms controlling significant capacity in Indonesia.

The pattern Craig Tindale identifies across critical minerals plays out identically in the battery supply chain. The mine is visible. The midstream processing facility is invisible to most investors and almost entirely foreign-controlled. Western automakers have announced ambitious EV targets, built gleaming gigafactories, and signed celebrity endorsement deals — and the battery cells trace their material inputs through a processing chain running through Beijing.

The domestic battery supply chain investments in Nevada, Georgia, and Ontario are real and necessary. But they are years behind schedule, over budget, and dependent on material inputs that must be imported in processed form while domestic processing capacity is built.

For investors, the EV battery story has two chapters. Chapter one — which we are living through now — is Chinese processing dominance. Chapter two is genuine diversification, arriving in the better part of a decade. Knowing which chapter you’re in matters enormously for how you value companies in this space.

Here is a pattern that should disturb every investor and policymaker who cares about American industrial revival: a company receives $150 million in DoD funding to build critical mineral processing capacity. It lists on a public exchange. Shortly after the funding announcement, it becomes a target of aggressive short selling. The stock collapses. The company can’t raise additional capital. The project stalls or dies.

Craig Tindale has documented this pattern across multiple DoD-funded industrial startups, and he names it plainly: unrestricted warfare operating inside the capital markets. You don’t need to blow up a factory if you can bankrupt the company building it. You don’t need to steal the technology if you can make the enterprise economically unviable before it scales.

The mechanism is elegant in its simplicity. Small-cap industrial companies are inherently vulnerable to short pressure. Their market caps are modest. Their investor bases are thin. Their revenues are pre-commercial while capital needs are large. A well-funded, coordinated short campaign can destroy a company’s ability to raise capital in six months — faster than physical sabotage and with complete legal deniability.

The question Tindale poses — and it’s the right question — is: where are these short sellers coming from? What is the source of their conviction on companies that have secured government backing and operate in strategically critical sectors?

I don’t deal in conspiracy theories. I deal in incentives and patterns. The incentive for a state actor to use capital markets as a weapon against industrial revival is obvious. The pattern is real and documented. The practical implication is clear: government funding alone is not sufficient to protect industrial startups. They need structural protection from capital market attack — and we don’t have it.

Copper mining has a chemistry problem nobody in the investment community talks about. You cannot mine copper at industrial scale without sulfuric acid. You cannot refine it. You cannot do heap leach extraction. Sulfuric acid is as essential to copper production as copper is to electrification — and the West’s capacity to produce it is constrained in ways that don’t show up in any copper price model.

Craig Tindale laid out the reagent dependency with the clarity of someone who has actually mapped the industrial inputs rather than just the headline metals. Sulfuric acid. Chlorine. Ammonia. These are the invisible chemicals that sit behind every critical mineral extraction process. Control them and you control the mine, regardless of who owns the land title.

The irony is almost literary. A significant portion of industrial sulfuric acid is produced as a byproduct of copper and zinc smelting — the same operations the West has been systematically closing for environmental reasons. Shut the smelter, lose the sulfuric acid. Now the copper mine that was supposed to reduce China dependency requires reagent imports to operate. The circular dependency is complete.

This is the mechanical thinking we’ve lost. We see a smelter as a pollution source. We don’t see it as a sulfuric acid production facility whose output is essential to three other industrial processes downstream. We optimize for one variable — local air quality — without modeling the systemic effects. The result is a set of industrial metabolisms quietly starving.

For investors, the reagent gap points toward an underappreciated category: domestic industrial chemical producers in sulfuric acid, ammonia, and specialty solvents. These aren’t glamorous. They don’t get covered at tech conferences. But in a world where the material economy reasserts itself, the company supplying the acid to the mine supplying the copper to the data center is not a commodity business. It’s infrastructure.

The narrative is appealing in its simplicity: China has become too risky, so we’ll move production to India. Apple is already making iPhones there. Problem solved. It isn’t solved. Not even close.

Craig Tindale dismantled this narrative with one observation that should be required reading for every supply chain consultant selling the India pivot story. The ferroalloys — specialty iron compounds used in the precision components inside an iPhone — come from China. Move the assembly to India, and you’ve moved a label. You haven’t moved a supply chain. The finished product still depends on Chinese-processed inputs at every level of the bill of materials that actually matters.

India’s industrial capacity constraints run deeper than ferroalloys. The country lacks the railroad density to move heavy industrial inputs efficiently. It lacks the electrical grid reliability that precision manufacturing requires. It lacks the trained engineering workforce at the scale needed to absorb even a fraction of the manufacturing volume currently processed in China. It lacks the chemical processing infrastructure for the reagents that advanced manufacturing requires.

India ran out of magnesium during a titanium production run. That is not the supply chain profile of a country ready to absorb Apple’s manufacturing operations, let alone the semiconductor, defense, and critical mineral processing that actually matters for national security.

India has real industrial ambitions and genuine strengths. But potential measured in decades is not a solution to supply chain vulnerability measured in months. The India pivot is a story that makes Western executives feel better about a problem they haven’t actually solved. The material reality hasn’t moved. Only the assembly line has.