Jeffrey Wernick
Buried in the Senate’s version of the defense authorization bill is a provision that changes the relationship between the American state and private industry. Section 1051 would give the Pentagon’s Office of Strategic Capital explicit authority to buy ownership stakes in private companies, and it would create a new account in the Treasury to pay for them. The government would no longer merely fund, contract with, and regulate private firms. It would own them.
The text is written to sound contained. The equity power is limited to critical minerals, materials, chemicals, and batteries. Stakes are capped at forty percent, the shares are non-voting, no single investment can exceed five hundred million dollars, and an oversight board is attached. Supporters point to these limits and call the concern overblown.
The categories are not contained. “Critical materials” reaches semiconductors and advanced materials. “Critical minerals” reaches mining, refining, and processing across dozens of elements. “Chemicals” reaches industrial chemistry, energy, and pharmaceuticals, since drug supply chains have already been declared a national-security matter. “Batteries” is not a product. It is the entire energy-storage and electrification economy, which means electric vehicles, grid storage, and much of the power systems inside consumer electronics. Take the four words at their plain meaning and, before anyone stretches a thing, they already touch mining, chemicals, pharmaceuticals, semiconductors, energy, autos, and defense.
Then there is the stretch, and Trump has shown how far he will take a national-security clause. He has used it to justify tariffs on steel, on aluminum, on cars. His administration has taken equity in more than twenty companies already, through a patchwork of agencies and legal theories, including ten percent of Intel. Applied to its outer limit, the term “critical” can be made to cover most of the industrial base. Under a broad but defensible reading, the companies eligible for a government stake under this provision represent something on the order of a fifth to a quarter of the national economy. Under the maximal reading Trump has repeatedly attempted, the arguable perimeter approaches a third. The dollars per deal are capped. The reach is not. And caps in first-generation legislation are the part that gets raised later.
The limits do not matter.
The architecture does.
Congress is not tolerating a one-time emergency. When the government took stakes in banks and carmakers in 2008, those were temporary rescues from bankruptcy, done with congressional approval and an exit in view. This is different in kind. It builds a permanent, chartered pathway for the executive branch to acquire ownership of private companies as ordinary policy, with a standing fund to finance it. The categories are broad today. The machinery is broader, and machinery, once built, is used. The Senate bill does not stop the equity spree already underway. It ratifies it, giving the practice a single statutory home and a legal basis, which is to say Congress looked at Trump’s power grab in progress and chose to write it into law rather than end it.
Eisenhower warned against exactly this in 1961, and not against a strong defense or against the existence of arms makers. It was against the fusion of the two, against what he called the acquisition of unwarranted influence by the military-industrial complex, and against the day when public policy might become the captive of the interests it was meant to govern. His remedy was distance. Government had to hold the arms industry at arm’s length, because only at arm’s length could it buy honestly, regulate fairly, and judge impartially.
This bill abolishes the distance. Consider what the government becomes to a company it owns. It is the company’s largest customer. It is the regulator of the company’s market. It is the awarder of the company’s contracts. And now it is the company’s shareholder, with a direct financial interest in the company’s rising value. When the same hand buys the product, writes the rules, picks the winners, and profits from the stock, impartial judgment is not weakened. It is impossible. The government cannot fairly decide whether a competitor should win a contract instead of the firm the government owns, because the government profits when its own firm wins. Northrop and Anduril make rocket motors too. The government has no stake in them. It has a billion dollars riding on L3Harris. Eisenhower warned that the complex would gain undue influence over the state. This goes further. It dissolves the boundary between them.
Then the word that is supposed to make all of this safe. Non-voting shares are offered as the proof that the government will not control these companies. It is the most cynical word in the bill, because it surrenders the one power that would make the control visible and keeps every power that produces the control. A private minority shareholder with non-voting stock genuinely is passive, because a private shareholder has no power over a company except the vote. Strip the vote and you strip their only lever. But the government’s power over a company it owns was never the vote. Its power is the contract it can withhold, the regulation it can tighten, the license it can pull, the investigation it can open. Non-voting shares disclaim the one form of influence the government least needs and retain every form it most has.
No one has to vote. No one has to issue a directive. The company holds an investor that is also its regulator and its largest buyer, and the company knows the value of its own shares, shares the government now profits from, depends on staying in that investor’s favor. So the firm anticipates. It aligns itself with Trump’s preferences before it is asked, because displeasing the shareholder who is also the regulator is suicide, and not one vote has to be cast for the company to understand that. The absence of a formal vote does not make the control absent. It makes it deniable. “We never directed them” becomes literally true and entirely meaningless, because you never have to direct a company that is busy directing itself toward whatever it thinks you want. Non-voting shares do not remove the government’s influence. They convert overt control into anticipatory obedience, and let everyone claim the control is not there because no vote was taken. The government takes the upside and disclaims the responsibility, profits from the company and answers for nothing, a silent partner who also happens to be the sheriff, the customer, and the tax collector, holding silent-partner shares precisely so that when the company does exactly what the sheriff wanted, everyone can say the sheriff never said a word.
The defense of the bill collapses on its own terms. Its supporters make two claims at once. The stakes are small and capped, so do not worry about control. And the stakes are valuable, so this helps strategic companies. Both cannot be true in the way they need. If the money is genuinely small and immaterial, then a company large enough to matter to national security can raise it in private markets without difficulty, and private capital comes without the disclosure, the political scrutiny, and the entanglement of having the state as an owner. So why would any such company take the government’s money instead? A rational firm chooses the worse capital over the better capital only if the money is not the point. Something else is attached. And if the money is not immaterial, if it is a large enough inducement to change a company’s behavior, then it is large enough to give the government the leverage the “non-voting, capped, limited” structure was sworn to prevent. There is no version where the stake is both too small to worry about and worth taking on its own terms.
The only coherent reading is that the money is a pretext and the privilege is the product. The equity buys preferential treatment from a customer who now profits when you win, forbearance from a regulator who is now your shareholder, and protection from competitors who have no stake. It buys access to the one office that controls the fund, and a place on the inside of it rather than the outside. This is why companies that plainly do not need the money have taken it. Intel is not capital-starved. A firm taking a government stake it does not financially need is telling you, by its own choice, that the stake purchases something other than capital. The equity is not the investment. It is the membership fee. The “return to the taxpayer” is the cover story that makes the membership fee look like an investment. The cap does not make it safer. The cap proves the money was never the mechanism. The privilege was.
If that sounds like a hypothesis, the record already supplies the instance. A rare-earth startup linked to the President’s son received a six hundred and twenty million dollar Defense Department loan in November of 2025, and a member of Congress raised the obvious concern that the President’s family may have profited. And when the Senate Armed Services Committee had before it an amendment that would have barred the investment office from taking stakes in companies where top executive-branch officials or their immediate family hold a significant ownership interest, the committee killed it. Congress was handed the chance to close the self-dealing channel before it opened, and declined. The perverse incentive is not a prediction about where this leads. It is a decision that has already been made, on the record, to leave the door open.
Madison saw the deeper danger, and it is not about defense contractors at all. It is about the delegation. He built the separation of powers on a single insight, that the accumulation of powers in the same hands is the very definition of tyranny, and that the sharpest tool for keeping those hands apart was the purse. Congress appropriates. The executive spends what Congress specifies, on whom Congress specifies. The legislature was never to hand the executive both the money and the discretion to choose its targets, because that combination, the power to fund joined to the power to decide who is funded, is the accumulation Madison spent his life trying to prevent. This bill fuses them. It gives Trump a fund and the authority to decide which companies qualify, under a term elastic enough to reach a quarter of the economy. Congress has not appropriated for a purpose. It has delegated the purpose.
Here is the mechanism that makes this worse than an ordinary expansion of the state. Under the old arrangement, a company that wanted government money had to work the system. It lobbied members of Congress, cultivated committees, spread its plants across many districts so that many representatives had a stake in its survival. The process was corrupt, but it was diffuse. Power was spread across hundreds of members, which meant no single person could deliver the prize, and the diffusion was itself a kind of protection.
This bill collapses that. It puts the checkbook in one office, under one branch, at the discretion of one administration. A company that wants an equity infusion no longer needs to lobby four hundred and thirty-five members of the House and a hundred senators. It needs to call one person. It needs to please the man who controls the account. And a company that displeases him learns what it is to be on the outside of a fund it cannot reach. The lobbying does not stop. It simply discovers that influence now has a single address. One checkbook. One call. This is how a diffuse corruption becomes a concentrated one.
That is the machinery of a cartel. Not a cartel in the sense of men in a room conspiring, but in the sense that matters. A system that dispenses privilege to those positioned to receive it, that rewards proximity to power over performance in the market, that draws favored firms into dependence on the state and makes them captive to it. And every incentive in it points the same way. The company’s incentive is to please the shareholder-regulator preemptively. The executive’s incentive is to use the stake as leverage while denying leverage exists. Congress’s incentive is to delegate rather than choose and be blamed. And the competitor’s incentive, the firm the state does not yet own, is to seek a stake of its own, because the only defense against a rival with a government shareholder is to acquire one too. So the arrangement spreads, not by mandate but by the logic of the incentive, until an entire sector is pulled in. Nobody has to intend the cartel. Each party pursuing its own advantage builds it, and “non-voting” ensures that at every step everyone can say they did nothing.
The bipartisan support should be the alarm, not the comfort. The same move is arriving from the other side, in a senator’s proposal for the government to take half of every major artificial-intelligence company. Republican defense hawks and democratic socialists have converged on the identical instrument, the state taking equity in strategic industry, because the logic of it flatters everyone. To one side it is national security. To the other it is economic justice. To both it is the state acquiring ownership of the private economy, and its appeal across the spectrum is not evidence that it is safe. It is evidence of how the boundary erodes, with everyone nodding.
There is a pattern in how Trump acquires power. When he seizes it directly, the courts sometimes stop him. They have trimmed his tariffs. So when the court will not give him the power, he goes to Congress to be handed it. A power struck down by a judge is one thing. A power granted by statute wears the legitimacy of law and is far harder to challenge. This is the search for power not subject to review, and Congress is obliging it, because its members find their advancement in loyalty to Trump rather than in the power of their own seats. Madison built the system on the assumption that ambition would counter ambition. Here the legislature hands its ambition to the executive. It is not being overpowered. It is volunteering.
Strip it to what it is. A defense bill would give one branch a standing fund to buy ownership of private companies, let that branch define which companies qualify under a term elastic enough to reach a quarter of the economy, and route the whole apparatus through a single office answerable to a single man. It would keep every power that lets the state control a company and surrender only the vote, the one power that would make the control visible. It would take a corruption that was at least diffuse and make it concentrated. It would take a boundary Eisenhower said was essential and erase it. It would take the purse Madison called the people’s most complete weapon and hand it to the executive. And it would do all of this with limits on the paper and a compliant Congress writing the grant, so that on the record, everyone consented.
They will call it the arsenal of freedom. It is a checkbook, held by one hand, with one number to call.
Editor’s Note: At a moment when the once vaunted model of responsible journalism is overwhelmingly the play thing of self-serving billionaires and their corporate scribes, alternatives of integrity are desperately needed, and ScheerPost is one of them. Please support our independent journalism by contributing to our online donation platform, Network for Good, or send a check to our new PO Box. We can’t thank you enough, and promise to keep bringing you this kind of vital news.
You can also make a donation to our PayPal or subscribe to our Patreon.