The Check and the Ceiling

On Wednesday night, at the Republican midterm convention in Dallas, the President promised every adult American citizen five thousand dollars. The condition attached was that Republicans hold both chambers of Congress in November. There are roughly 245 million adult citizens. The promise is therefore worth about $1.2 trillion.

That is a large number, and most of the coverage since has been about whether it is legal. Some of it has been about whether it is a bribe. Very little of it has been about the simpler question, which is whether a check like this does what a check like this is supposed to do.

It is worth starting with the ledger, because this is the fourth one.

Notice what the three have in common. Each one is a promise to hand back a surplus. And in each case the surplus has to exist first.

The money that went backwards

Here is the part almost nobody has looked at. The Treasury publishes its receipts every month, and customs duties get their own line. You can pull it yourself.

Through August, eleven months into fiscal 2026, net customs receipts stood at $167.3 billion. In April, seven months in, they stood at $188.6 billion.

Read that again. The year-to-date total went down. A running total is not supposed to be able to do that.

It can do it for one reason. The figure is net of refunds, and in February the Supreme Court ruled that the emergency statute the tariffs were built on did not authorize them. The Court of International Trade then ordered Customs to give roughly $165 billion back. That money started moving in May.

In June the government paid out more than twice what it took in. This is the revenue stream that three separate promises have now been drawn against.

Annualize the eleven months and fiscal 2026 lands somewhere near $182 billion. Against a $1.2 trillion promise, that is close to seven years of every dollar the country collects at the border, spent on nothing but this, to make the payment a single time.

And the authority behind even that $182 billion has been rebuilt twice since February. The emergency statute fell on the twentieth. Within hours a 10 percent tariff went up under Section 122 of the Trade Act, which by its own text expires after 150 days unless Congress acts. It duly expired on 24 July. The same minute, Section 301 duties of 10 to 12.5 percent took its place across 60 economies.

What a check is for

Now set the money aside, because the arithmetic is the smaller problem.

The Toilet Paper Theory starts from a distinction. A shortage can be real, meaning the thing does not exist in the quantity people need. Or it can be counterfeit, meaning the thing exists in plenty and people simply cannot reach it. From the empty shelf the two look identical. They are not, and they take opposite remedies.

A check is a monetary instrument. It does exactly one thing. It restores reach.

Which means a check is the correct medicine for a counterfeit shortage and no medicine at all for a real one. If the goods are on the shelf and the household has no money, money fixes it. If the household has money and the goods are not there, money raises the price and changes nothing else.

This is not a theoretical claim. We ran the experiment in 2020, and it worked. The shortage then was counterfeit almost everywhere. Factories were running or could restart. Incomes had been cut by an order to stay home. The checks restored reach to people whose reach had been taken, and they did what they were supposed to do.

So the instrument is not the problem. The question is what disease we have now.

Reading the gauge

This site keeps an index for precisely this question. The Credit Counterfeit Index subtracts a real bind from a monetary bind. When money is the thing that is scarce, it runs high. The tipping point is +1.0.

The most recent reading, for the third quarter of 2026, is −1.08. The monetary bind sits at −0.08, which is to say at nothing. Credit spreads are tight. Banks are not tightening standards much. The real policy rate is near a third of a percent. By every input the index uses, money is not scarce in America right now.

The real bind sits at 1.0, and it is the higher of the two.

That is the whole finding. The binding scarcity today is real, not monetary. And a check is a monetary instrument.

Look at what the other dispatches on this site have found about where those real constraints sit. The work week has held near 34.3 hours for twenty years, through booms and busts alike, because labor is the one input that cannot manufacture more of itself. Housing has 3.56 million vacant units and 745,652 people without one, which is a divisibility problem that cash does not solve. Munitions production sits at 36,000 shells a month against a goal of 100,000, and Congress already voted that money.

That last one is the closest parallel, and it is the one to hold onto. We have already run this experiment recently. Money was appropriated. Capacity did not appear. Appropriation is not capacity, and a dividend is not supply.

The circle in the tariff version

There is a further oddity in funding the check from customs duties specifically, and it is worth saying slowly.

A tariff, as the earlier dispatch argued, is the one counterfeit shortage in this whole framework that gets manufactured on purpose. The goods exist. The factories abroad are running. The ships can sail. A toll booth in between makes the goods unreachable, which is the textbook definition of a counterfeit shortage.

So the proposal is this. Erect a barrier to reach. Collect money at the barrier. Hand some of that money back so people can reach past the barrier you erected.

The circle does not close, and it cannot, because the trip is not free. Collection costs something. Distribution costs something. And the tariff destroys some purchases entirely, which is the finding from the earlier piece: a household at the bottom does not pay the tariff and complain, it stops buying, and a purchase that never happens generates no revenue to rebate. You cannot return more reach than you removed. The rebate is always smaller than the barrier.

On the legality, which is not what people think

The loudest objection since Wednesday has been that this is a bribe. Conditioning payments on an election result certainly feels like one. As a matter of law it is the weaker argument.

In 1982, in Brown v. Hartlage, the Supreme Court considered a candidate who promised voters a financial benefit and held that there was nothing unlawful about it, so long as the benefit was not offered as a quid pro quo to a particular voter. A promise of lower taxes gives every voter a financial reason to vote a certain way. That has never made it a bribe. A general promise made to the entire electorate is campaign speech, and it is protected.

The real barrier is duller and far more solid. It is the Appropriations Clause, in Article I, which says no money may be drawn from the Treasury except by an appropriation made by law. Treasury cannot send $1.2 trillion because a President announced it. Congress has to pass something first.

Which is where the shape of Wednesday’s promise becomes interesting, because the condition is the confession. The payment was made contingent on Republicans holding both chambers. That is not the structure of a thing a President can do alone. It is an acknowledgment, built into the sentence, that this requires an act of Congress and therefore the votes to pass one.

The two earlier promises died at exactly this wall. The DOGE dividend needed the savings to be real and needed an appropriation, and got neither. The tariff rebate needed the tariffs to be lawful and needed an appropriation, and lost the first in February.

When a check would be right

The honest version of this argument has to include the case where it works, because there is one, and it may not be far off.

If the gauge flips, the answer flips with it. Should the monetary bind rise while the real bind recedes, meaning credit seizes up and money climbs to the top of the ranking, then the shortage that follows is counterfeit by definition. Goods sit unsold. Households cannot reach them. In that world a direct payment is not merely defensible, it is the best instrument available, because it works faster than a rate cut and it reaches households that a rate cut never touches.

So the objection here is not to the tool. It is to using it now, and to spending it early. A remedy fired at the wrong quadrant is not neutral. It is a remedy you no longer have when the right quadrant arrives.

And there is a direction problem on top of the timing. Adding $1.2 trillion of demand to an economy where the real bind is the binding one pushes toward the upper right of the phase plane, where prices rise and quantities do not. That is the overheating quadrant. It is the one place the framework says you should not want to go, and it is the direction this policy points.

What would prove this wrong

Three things, and they are all checkable.

If a payment of this size were made and the CPI did not move much, the claim here about real constraint would be badly weakened, and the reading of the gauge with it. If net customs receipts recover past $250 billion on an annual basis once the refunds finish clearing, the funding objection shrinks a good deal, though it does not disappear. And if the Credit Counterfeit Index crosses toward +1.0 before any payment is authorized, then the disease will have changed, the medicine will have become correct, and this dispatch will have been written about a moment that passed.

Watch the index rather than the announcement. It has told the truth about the last three.

One caution about a number above. The 245 million figure for adult citizens is a Census estimate, and the Vice President has since suggested an income cap that would cut the total. A cap would reduce the $1.2 trillion, possibly by a lot. It would not touch the argument, because the argument is about which scarcity binds, and that does not depend on the size of the check. But the headline cost should be read as the version without a cap, since that is the version that was announced.