Private Debt or Government Debt? Two Theses on the Cause of a Financial Crisis (Part 1 of 3)

• Two men, one pile of debt, and our own stake in the answer
• Richard Vague: the variable is the speed of private credit, not the level
• The balance sheet he turns upside down, and the deficit argument with it
• The rule he would write, and the two exceptions he cannot hide
• Jack Mallers: the funding cost of the sovereign, and the trap at the front end
• The carry trade, the three exits, and the only road left open
• The four collisions where the two systems genuinely disagree
• Where they are talking past each other, and the mechanism they share
• The tests that would settle it, and what comes next

Richard Vague says the danger is private credit growing too fast and the argument about government deficits is aimed at the wrong debt, while Jack Mallers says the danger is the funding cost of the sovereign itself — so this episode states both systems at full strength, names each man’s incentives, separates the four places they genuinely collide from the places they are talking at different layers, and reports the one mechanism they both affirm and read in opposite directions.

DeepDives · 2026 · An AI–human collaboration that weighs every source, mainstream or alternative, under strict rules of truth and logic.

§Two men, one pile of debt, and our own stake in the answer

Two men, one pile of debt, and a disagreement about which part of it is the dangerous part. Richard Vague says the danger is private credit growing too fast, and that the public argument about government deficits is pointed at the wrong debt. Jack Mallers says the danger is the funding cost of the sovereign itself, that the government has pushed its borrowing to the front end of the Treasury curve where it must be rolled over constantly, and that the arithmetic of that roll ends in a weaker dollar. Neither is a neutral observer, and that should be said at the top rather than buried at the end.

In this disagreement the incentives are as legible as the arguments, so let us be exact about who these two are. Richard Vague spent a career as a banker, co-founded credit-card issuers and an electricity supplier, and served as Pennsylvania’s Secretary of Banking and Securities. He has written a body of work on debt and crisis, and he now runs a research organisation that publishes the data his own case rests on. That means his name is on the dataset as well as on the thesis. That is not a scandal. It is a stake, and a stake is better named before the argument than discovered after it.

Jack Mallers is the founder and chief executive of a bitcoin payments company, and his forecast runs like this. The debt has floated to the front end of the curve. The average cost of the stock sits far below the cost of new money. The interest bill widens the deficit with every hike. And every road out of the problem ends in a weaker dollar, against which bitcoin and gold get repriced. If he is right, the asset his company is built around is the asset that wins.

Which means both men own assets that would gain if their own thesis turned out to be the true one. That is not a reason to dismiss them.

We should dispose of our own interest too, because we have one. This show has published a prior on this exact ground. The prior is that fiscal dominance plus reserve-currency status means the eventual adjustment lands on the currency rather than on an outright default, and that hard assets reprice against it. That is a house position, held in public, and it will be put on the table in the third part of this examination and hit as hard as we hit either guest. Not here, and not yet.

So set the shape of the three parts, so that nobody waits for a verdict that is not coming. This first part states each man’s system at full strength, in his own terms. It names the exact point where the two systems collide, separates that from the places where they are merely talking at different layers, and reports the one place where they converge on the same mechanism and draw opposite conclusions from it. The second part takes the claims to the record and checks them against the data. The third part does the synthesis, our own prior included. Nothing is decided today.

§Richard Vague: the variable is the speed of private credit, not the level

Vague’s central claim is not that debt is high. It is that the level of debt is the wrong variable and the speed of the change is the right one. In his telling, the major financial crises have been around excessive private debt, and excessive government debt has not created a financial crisis in a major western developed country. He puts the number at forty-three crises across the six largest economies over two hundred years, and he says the common factor in them is velocity, meaning the speed of change in lending rather than the level of it.

His own illustration does real work in the argument. One high-rise condominium in South Beach works. Twenty of them in three years does not. A market can absorb a slow accumulation of credit and be wrecked by a fast one, and the wreckage does not arrive looking like a government balance sheet.

Then there is the money-creation mechanism, which is the part of his case most people have not thought through. Take a ten-thousand-dollar loan. The bank books a loan asset on one side and a matching deposit on the other. The deposit is the bank’s liability to the borrower, and that deposit is new money. Deposits are the money supply. On his figures, US deposits went from five trillion dollars in 2000 to twenty-one trillion dollars in 2025. Roughly ninety per cent of that came from private banks lending to households and businesses, and roughly ten per cent from the Federal Reserve. The point of all of it is that money is made at the point of lending, and most of the lending in an economy like this one is private.

That mechanism is not Vague’s invention, and it is the strongest single item he has. The Bank of England’s Quarterly Bulletin for the first quarter of 2014 carried an article titled Money creation in the modern economy, written by McLeay, Radia and Thomas. It says the majority of money in the modern economy is created by commercial banks making loans, and that banks do not act as intermediaries lending out deposits placed with them. If you are looking for the load-bearing beam in Vague’s structure, it is there.

§The balance sheet he turns upside down, and the deficit argument with it

His picture of the balance sheet is where he turns the deficit argument on its head. His figure is that US household wealth is about one hundred and eighty trillion dollars, and that it dwarfs the government’s forty trillion. The top ten per cent of households own almost eighty per cent of the real estate and the stocks, and the bottom half own less than ten per cent. Relative to income, he says debt at the top ten per cent has risen by five to ten per cent, while debt for the lower part of the income spectrum has risen seventy-five per cent.

Those distributional numbers are not the part to bet the argument on, and it is worth being honest about why. The direction is consistent with the Federal Reserve’s own distributional accounts, and the direction is what his argument needs, but the precise pair was not reproduced when this was checked. His structural claim is that credit growth inflates assets owned disproportionately by the people who are already able to borrow, and that part does not depend on the exact gap.

Then there is his answer on the deficit itself, which is the claim that gets him the most argument. He says that no financial crisis in a developed economy since the Second World War has come from rising government debt, and he says he is not concerned about debt at one hundred, one hundred and twenty or one hundred and thirty per cent of gross domestic product. When the government spends, the money does not disappear. When the government runs a trillion-dollar deficit, households run a trillion-dollar surplus. In 1981 the government’s net worth was minus forty per cent of gross domestic product and household net worth was positive three hundred and fifty per cent. His summary of the whole system is two sentences. Assets equal liabilities. Income equals expense. And on the interest, he says that almost all of the interest the government pays goes into the checking accounts of households.

The buyer question is where a lot of people stop him. His answer is that the money supply is not depleted by the sale of government debt, because the proceeds are spent in the same instant, and so there will not be a shortage of buyers.

He says he sees no correlation between rising government debt and inflation, and if anything the opposite relationship. He also concedes that the forty trillion is probably sixty or seventy trillion once unfunded obligations are counted, though not one hundred or two hundred trillion. Comparisons with developing countries fail, in his account, because those countries borrow in a foreign currency, so a devaluation wrecks them. A developed country borrows in its own, which is the whole difference.

§The rule he would write, and the two exceptions he cannot hide

Vague is not only describing; he is prescribing, and the prescription is the part people skip. He proposes a warning rule. Growth in the debt-to-gross-domestic-product ratio of any category above fifteen per cent per year should compel regulators to intervene. Not a ceiling on the level of debt, but a speed limit on the growth of it, applied to every category, private as well as public, because speed is the variable that does the damage in his account.

He goes back to Thomas Willing, and to the Bank of North America chartered in 1781, and the Bank of the United States in 1791. He also reaches back to the Bank of England, chartered in 1694. The history is doing a job. It says that private credit has been the organising force in the economy since the beginning, and that the people who run it have a constituency of their own.

Then the exceptions, named before we move to Mallers, because the rule cannot be stated as absolute with the counterexamples left in a pocket. The euro-area crisis of 2010 to 2012 is the test case. Greece, Ireland, Portugal, Spain and Italy each faced a sovereign funding crisis, and Greece restructured its debt in 2012. That is an event Vague’s rule does not cover on its face. His own caveat is that those countries borrowed in a currency they did not issue, so on his reading the episode tests the caveat rather than the claim. The second exception is the pandemic. Federal debt as a share of gross domestic product went from one hundred and six point eight per cent in the first quarter of 2020 to one hundred and twenty-four point one per cent in the third quarter, and consumer inflation followed it up to about nine per cent in 2022. Debt exploded and inflation exploded. The decade after 2008 is the case he reaches for, where debt rose and inflation fell. So the relationship does not hold in either direction. That helps his wider point about there being no reliable causal link, and it breaks the specific generalisation he states.

§Jack Mallers: the funding cost of the sovereign, and the trap at the front end

Mallers begins where Vague ends, with the stock of government debt, but he is asking a different question about it. He is not asking whether the sovereign can pay. He is asking what it costs to fund, and what happens to the price at which the funding clears. That is the spine of his case.

The way he sets it up, the government has struggled to sell long duration, so its issuance has floated to the front end of the curve. That means the debt is refinanced constantly at market rates instead of being locked in for decades. The debt is therefore repriced every few months, and the repricing is the trap. He says the United States must refinance eight point three trillion dollars of debt this year, against under a trillion a year for half a century. On that specific figure no Treasury source states it, so treat eight point three trillion as his number and not ours.

The average cost of the existing stock of debt is far below the cost of new money. The weighted average interest rate on all marketable Treasury debt was three point four seven five per cent at the end of August in 2026. The ten-year was above five per cent and the thirty-year above five and a half. So every time a maturing bond is rolled into a new one, the government pays more on that slice, and the interest bill ratchets upward even if nothing new is borrowed. That is arithmetic, not opinion, and it is the strongest quantitative support he has in the whole episode.

And then the loop closes on itself. The doom loop is his name for it. The central bank hikes, so the interest bill rises. The interest bill widens the deficit. A wider deficit means more bonds issued. More supply at higher yields pushes yields higher. And the central bank faces pressure to hike again. He says the central bank raised rates the week before and the ten-year went from four point nine to five point two five per cent.

§The carry trade, the three exits, and the only road left open

The carry trade is the piece of his case that is hardest to see and does the most work. The rest of the world owes fourteen trillion dollars in dollar debt and owns twenty-two trillion dollars of US assets, and a little under half of those assets is US government debt. If the dollar strengthens, those obligations get heavier in local terms, and foreign holders are forced to sell US assets to meet them, and that selling pushes yields up and prices down. Which means the dollar cannot be allowed to strengthen much, because strengthening the dollar is what triggers the selling. It is the release valve, and it is the only one that arithmetic leaves open.

Then he lays out three exits, and he closes two of them himself. Raise rates, and the interest bill goes up, so it makes the problem worse. Cut spending, and the arithmetic bites back, because he runs the numbers on a five hundred billion dollar cut and says it takes debt-to-gross-domestic-product from one hundred and twenty-five to one hundred and thirty per cent, because the denominator, the economy, falls further than the numerator, the debt. So the only line left is the interest expense itself, and the only mechanism he can find is to cap yields, run the economy hot, and let the currency weaken. What matters here is the shape of it. On his own accounting, two of his three exits are closed, and the one that is left is the one that debases the currency.

On the outcome he is not hedging. Every road ends in a weaker dollar, is his line. He expects bitcoin and gold to be repriced against it, he says he has not been more bullish than he is now, and he warns of a possible flash crash on the way. He adds a distributional argument of his own, that more than a third of US household wealth is now held by people over seventy, which would mean a great deal of the wealth in the system will change hands within a generation. That specific share could not be confirmed, so it stands as his claim rather than as a fact.

§The four collisions where the two systems genuinely disagree

Put the two systems side by side, because up to here both have been treated generously and the collision should be named precisely.

The first is the causal variable. Vague says the variable that causes crises is the velocity of private credit growth, the speed of the change in private lending. Mallers says the variable that matters now is the sovereign funding cost, the price at which the government’s own debt clears. Those are not two answers to one question. They are answers to two different questions, and each man treats his as the one that decides the outcome.

The second is the direction of causation, which comes first, and that is the real fork. In Vague’s order, the private sector moves first. Credit builds in the private economy, the run-up outruns the borrowers, and the crisis arrives from below. The sovereign is a follower in that story, and a well-behaved one. In Mallers’ order, the sovereign moves first. The government issues, the market prices the funding, the interest bill feeds the deficit, and the private sector is the one that gets squeezed by the rate that comes out of the government’s need to borrow. Both orders are internally consistent, and they cannot both be the description of the same crisis, because they name different first movers.

The third is what breaks first. Vague is watching for a wave of private defaults, the way the subprime lending boom of 2007 ended in one. Mallers is watching for a failure at the front end of the Treasury market, a point where the constant refinancing of the shortest bills meets a bid that is not there, or is there only at a price the government cannot wear. One man is looking at the borrower’s balance sheet. The other is looking at the term structure of the sovereign’s own debt.

The fourth is the sharpest, and it is whether the buyer base can fail at all. That one is not a difference of forecast; it is a difference of ontology. Vague’s accounting says no. The proceeds of a bond sale are spent in the same instant, so the money supply is not depleted and the system does not run out of the means to buy its own debt. The buyer base cannot fail in that framework, because it is not a fixed pool that can be emptied. Mallers’ price says yes. He is not claiming the money is gone. He is claiming that the price at which the debt clears can become one the government cannot fund without limit. The buyer can be present and still be unwilling at that price, and in Mallers’ telling that is failure enough.

§Where they are talking past each other, and the mechanism they share

There is another half to this, because those four disagreements are not the whole picture, and some of what looks like conflict here is not conflict at all. Vague is describing an accounting identity. A sovereign that issues its own floating currency and borrows in it does not face a solvency wall the way a household or a firm does, because it cannot be forced into default on debt denominated in its own money. That is an identity, and it is not falsifiable by a price. Mallers is describing the price and the term structure at which that same sovereign’s stock of debt clears. Those are two different objects of study. A balance sheet can be solvent and still be expensive to roll. Vague can be right about the identity and Mallers can be right about the price, at the same time, without either of them having made an error.

Ask Vague whether the government can run out of money and he will tell you no, because it issues the money. Ask Mallers the same question and he will tell you that no is beside the point, because the government can pay a price for the money it issues that is ruinous without being insolvent. Neither of those answers contradicts the other. They are answers about a stock and answers about a flow, and the two of them are being read as if they are answers about the same thing.

Then the convergence, which is the sharpest single point in the whole trilogy and should be stated plainly. Both men, independently and for opposite reasons, say that the interest the government pays on its debt lands in the accounts of households, pension funds and banks, and stays inside the system. Vague offers it as the reason not to fear the debt. Mallers offers the identical mechanism as the reason the deficit is stimulative and inflationary. Same fact. Opposite conclusion.

Vague’s version goes back to the net-worth history and the checking accounts. When the government runs its trillion-dollar deficit, households run their trillion-dollar surplus, and almost all the interest the government pays goes into the checking accounts of households. So the interest is not lost to the system. It is recycled within it, which is why he thinks the fear is misplaced. Mallers takes the same payment stream and reads it the other way. Money that the government pays out as interest is money that households and funds and banks now hold, and it is money they can spend. The interest bill is not a drain on the economy, he says. It is an injection. It is stimulative. And a stimulus that arrives whether or not the economy needs it is, on his reading, an inflationary force that the government cannot easily switch off. Both men are describing one transfer. One calls the transfer a reason for calm. The other calls the same transfer a reason for alarm. And neither of them disputes the mechanism at all.

The dispute is about the consequence. That is what makes this pair worth three parts instead of one. They are not arguing about a fact. They are arguing about what the fact means.

§The tests that would settle it, and what comes next

There are three, and the first is Vague’s own. Watch private credit. If there is a run-up in private debt past his threshold before the next crisis begins, growth in the debt-to-gross-domestic-product ratio of that category above fifteen per cent a year, his rule gets a scalp and his causal story gets its check. If the next crisis instead arrives from the sovereign side, with private credit quiet, that is evidence for Mallers and against Vague.

Second, watch the sovereign funding. A badly covered long auction, or a genuine turn in foreign demand, is the event Mallers is predicting and Vague says cannot bite. The thirty-year auctioned in September was covered two point six one times at a high yield of five point three one per cent, the twenty-year cleared two point five seven times, and the ten-year two point seven one times. The claim that there is no market for long-dated debt is directionally about price and literally too strong. What would matter is a long auction that fails to clear, or foreign holdings that turn down for more than a month or two, and foreign holdings are already flat to falling while the deficit runs near two trillion a year, which points the direction Mallers is describing without yet proving the event he predicts.

Third is the referee: the path of inflation expectations and real yields. Mallers’ case implies that the market should be pricing a weaker dollar and higher inflation into long bonds. Right now it is pricing the opposite of a debasement. The ten-year breakeven is two point three four per cent, the five-year five-year forward breakeven is two point three six per cent, and the ten-year real yield is two point nine per cent, up from one point eight four per cent in October of 2025. If Mallers is right, that is the price that has to move, because a debasement story that leaves breakevens anchored in the low twos is a story the market has not bought yet. If those breakevens break higher while real yields stay high, that is the combination that would tell you the market has started to believe him. And watch the speed of the moves rather than only the level. Between late August and late September the seven-year auction cleared at four point five one two per cent and then at five point zero eight five per cent, which is fifty-seven basis points in a month. Moves at that speed are the kind Mallers means when he says the price can start breaking things.

There is the shape of it. Two coherent systems, four real collisions, a stretch where they are simply talking about different layers of the same object, and one mechanism they both affirm and read in opposite directions. No verdict today. The next part takes Vague’s claims to the record, then Mallers’ claims to the record, and checks each one against the numbers rather than the rhetoric. The third part pulls it together, puts our own prior next to theirs, and says what we think. Where the record is thinner than either man’s confidence, we will say so and say why.

One more thing before we go, and it matters for how you listen to the next two parts. The two of them agree on more than they would like to admit. They agree that the debt has grown to a scale that commands attention. They agree that the interest bill is real and growing. They agree that the mechanism by which the government pays that interest puts money into private hands and keeps it there. What they divide over is which of those agreements is the one that decides the outcome. Vague thinks the level of sovereign debt is a distraction from the real variable. Mallers thinks it is the variable. And on that, the record is the only thing that can adjudicate, which is exactly where we are going next. Take each man’s central claim, find the data that would count against it, and see which way the data actually runs. That is the work, and it starts next time.

This has been a deep dive into whether private debt or government debt is the thing to watch.

Sources: this examination names its sources inline — the two theses stated by Richard Vague and Jack Mallers, with their commercial interests noted; Vague’s forty-three-crisis audit and his research organisation’s published data; the Bank of England Quarterly Bulletin article Money creation in the modern economy by McLeay, Radia and Thomas; the Federal Reserve’s distributional accounts and household net-worth series; the euro-area sovereign funding crisis of 2010 to 2012 and Greece’s 2012 restructuring; the pandemic-era federal debt and consumer inflation series; Mallers’ refinancing figure, reported as his claim and not as a Treasury number; the weighted average interest rate on marketable Treasury debt, the Treasury auction results, the breakeven and real-yield series, and the foreign holdings data; and the market tape for the ten-year, the thirty-year and the policy rate. Every source is weighed with the same skepticism regardless of politics or business model. The audio version of this article, read by DeepDives, is available on Wavlake.