Jack Mallers on Europe: The Record So Far (Part 2 of 3)

• The examination this part compares against
• Where the operator and the documents agree
• The foreign seller, the long end, and one checkable recap
• The fair test: what aged well
• What did not age well, and the one item that ran the other way
• The price the market has not paid
• The two items left unresolved
• The long bond, repriced rather than abandoned
• Where the ground position refines us
• The holder with a name, and the French inventory
• The through-line: not the destination but the route

Part one graded his older claims one by one against the outside evidence; this part sets that verdict beside his new episode and asks where the record has caught up with his forecasts, and where it has not.

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

§The examination this part compares against

Start with what the earlier work carried forward, because it is the spine of this part. The last time this show took Jack Mallers’ debt-and-dollar thesis seriously, it set his account of the sovereign funding problem against a private-credit writer, Richard Vague, and graded the claims one by one. Vague argues that the crisis risk sits in private credit rather than in government debt, and the earlier trilogy put the two accounts side by side on those terms. Both men were told the same thing before either was graded, which is that the ledger would not grade the person rather than the claim. Each item came out Supported, Unresolved or Contradicted, and where a guest was right, the script said so plainly and moved on.

The central unconfirmed item was his figure of eight point three trillion dollars of United States debt needing to be refinanced within a year. The trilogy could not find a Treasury source stating that number, and it told listeners twice to treat it as his figure rather than ours, and to hold the mechanism rather than the digit. It is the same place this part leaves it.

What held up was narrower and stronger. 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, which is the average coupon cost of the whole stock. The trilogy called it arithmetic rather than opinion, and the strongest quantitative thing the sovereign case had. It is still the number every other claim in this part has to beat.

So the earlier verdict was not a dismissal, and it is worth being precise about the shape of it. The roll, the mechanism by which a maturing bond is replaced at a higher rate, was supported. The trillion-dollar interest bill was supported. The forty-trillion-dollar headline was supported as a gross figure. The carry-trade mechanism was unresolved, the doom loop was unresolved, and the debasement price was contradicted by market prices on the day of recording.

Which brings us to the new episode, which is called Europe Is Screwed. His new claims arrive with checked numbers attached, and not all of them land where his older ones did. Where the older work and the new work agree by independent routes, that agreement is the evidence under examination. This part is a comparison, not a re-grading.

§Where the operator and the documents agree

Start with the convergence, because it is the most useful thing in the episode and the least flashy. There is a set of items where a working operator reading the market in real time and a series that triangulates documents reached the same conclusion by methods that share almost nothing. When two independent methods converge, both conclusions get stronger, and it does not matter that one of them was spoken on a live stream.

The first is the cost of the stock against the cost of new money. His older account had the debt priced as a pile of individual loans, each issued at an older and lower rate and each needing to be refinanced near today’s rates. Our own reading of the current data has the same geometry. The weighted average rate on the marketable stock was three point four seven five per cent at the end of August, while the marginal ten-year sat at five point two four and the marginal thirty-year at five point five six.

That is the roll, and it is the load-bearing arithmetic in both versions. Every refinancing happens at a rate far above the rate it replaces, so the average cost of the stock drifts up toward the marginal cost even if no new money is borrowed. That is not a forecast about politics or incentives. It is subtraction, and both the operator and the document series arrived at it from opposite directions.

The second convergence is the interest bill widening the deficit. He says interest expense is about a trillion dollars a year and the total deficit about two trillion this year and above two trillion next. Our check of the Treasury’s own interest series has interest on public issues at nine hundred and eighty-one point eight billion dollars in fiscal 2026 through the end of August, with fiscal 2025 at nine hundred and seventy billion. The bill is crossing a trillion dollars rather than sitting comfortably above it, and both readings see it climbing.

§The foreign seller, the long end, and one checkable recap

The third convergence is the foreign holder as the marginal seller. His account has the world owing fourteen trillion in dollar liabilities while owning twenty-two trillion dollars of United States assets, a little under half of that government debt. Our own read of the holdings data is narrower but points the same way. Foreign holdings stood at nine trillion two hundred and sixty-seven billion dollars in December of 2025, against nine trillion three hundred and forty-nine billion the month before, which is flat to falling while the deficit runs near two trillion dollars a year. The mechanism he supplies is the practical one, which is that a foreign holder sells when it needs money itself.

The fourth convergence is the direction of the debasement trade, and this is where the two methods agree most cleanly and the market agrees least. He reads a sovereign funding problem unfolding toward a weaker dollar, and both his account and ours expect the pressure to show up first in prices at the long end of the curve. The long end is where the number moves when the buyer goes quiet, and it has moved.

His new episode’s own recap numbers check out against outside sources. The ten-year yield reached five point three zero four per cent on the thirtieth of September 2026, past the intraday peak of 2007 and the highest since May of 2002, on Reuters reporting. The thirty-year mortgage was seven point two eight per cent in early October. Total public debt first printed above forty trillion dollars on the eighteenth of August, and sat at forty trillion and thirty-three billion dollars on the twentieth.

He reads the market in real time and he runs a balance sheet inside it. We triangulate documents, which means the Reuters wire, the Banque de France, the Council on Foreign Relations, the European Central Bank and the International Monetary Fund, and those are different instruments pointed at the same object. Where they agree, the reading is stronger for the disagreement in method.

§The fair test: what aged well

Now the cheapest fair test available, which is to put his older numbers beside the current data and read the record. It is simply what happened next to what he said then.

Begin with what aged well, because more of it did than the register of the coverage suggests. On the direction of yields he was right and early. His older account had the ten-year above five per cent and the thirty-year above five and a half, and the current reading is a ten-year that reached five point three zero four per cent on the thirtieth of September, the highest since 2002. The direction he described is the direction the market took.

On the mortgage rate he was also right. His account had the thirty-year mortgage above seven per cent, and the check has it at seven point two eight per cent in early October. That number is not a small claim, because the mortgage rate is what transmits the long end of the curve into households, and it is the one the Treasury Secretary is least comfortable defending. He had the transmission right.

On the deficit trajectory the split is worth keeping sharp. He says the deficit will work from two trillion dollars a year to three trillion even with no new spending and no further hikes, on the argument that refinancing at today’s rates does the work. The Congressional Budget Office expects about one point nine trillion dollars for 2026, so the deficit level itself checks out. The escalation is his projection rather than a published forecast and should be heard as his, while the mechanism underneath it is the one both methods share.

On the warning about foreign holders he aged well as well, though the event has not arrived. His older line was that the buyer for the debt stock had stopped buying, and our check found foreign holdings flat to falling while the deficit ran near two trillion dollars a year. That points the direction he is describing without yet being the break he expects.

What aged well also includes the simplest item, which is the size of the problem. The forty-trillion-dollar headline is supported as a gross figure. Treasury’s own count on the twenty-eighth of September 2026 put total public debt outstanding at forty trillion, one hundred and two billion dollars to the nearest billion, and gross debt at about one hundred and twenty-three per cent of gross domestic product.

§What did not age well, and the one item that ran the other way

Now the items that did not age well, and there is a specific reason to name each one rather than wave at the lot of them. The largest is his figure of eight point three trillion dollars of United States debt needing to be refinanced within a year. The earlier trilogy could not find a Treasury source that states it, and the new episode does not supply one. It stays exactly where it was left, which is his figure rather than ours, unconfirmed rather than wrong, with the mechanism held and the digit not.

A second item is unconfirmed in the same way. His claim was that Asia runs a one point eight trillion dollar trade surplus and sends only about one hundred billion dollars of it into United States long-term bonds, which is his evidence for the marginal buyer moving domestic. The stock data point the same way, but the flow figure could not be reproduced. So the direction is sound and the number is his.

A third is his projection that United States interest reaches four and a quarter per cent of gross domestic product in 2027, against under one per cent for Japan. His figure sits above recent official net-interest projections, and the basis could not be confirmed. It belongs with the other unverified items rather than in the column of things we can check.

A fourth is the demographic claim. He says more than a third of United States household wealth is now held by people over seventy, which would mean a great deal of the wealth in the system changes hands within a generation. That specific share was not checkable in the earlier work and should be repeated as his claim rather than as a fact.

Then there is the item that did not merely fail to confirm, but ran the other way, and this is the clearest single correction in the whole record. He says no country or empire has survived debt above one hundred and thirty per cent of gross domestic product. Japan’s gross debt is about two hundred and four to two hundred and seven per cent of gross domestic product, the highest among advanced economies, and Japan has not defaulted. Singapore’s gross figure is about one hundred and seventy-two per cent, against about one hundred and twenty-six per cent for the United States, on International Monetary Fund figures from April of 2026.

That is a level claim, and levels are the easiest thing in this argument to check. The historical record does not support the threshold he states. The honest way to hold it is that high debt is a real drag and a real risk, and that the specific sentence, that nothing above one hundred and thirty per cent survives, is contradicted by two of the world’s most indebted advanced economies.

§The price the market has not paid

Then the debasement price, which the earlier examination graded as contradicted and which the new data has not rescued. If the market were pricing the erosion of the dollar’s purchasing power, real yields would be falling and breakevens rising, because a breakeven is the inflation the market expects over the life of a bond. In fact the ten-year breakeven was two point three four per cent, the five-year five-year forward breakeven was two point three six per cent, and the ten-year real yield was two point nine per cent, up from one point eight four per cent in October of 2025.

Two caveats were kept on that item and both still apply. Prices can be wrong, and expectations are not outcomes. What the figures establish is narrower and firmer than a prediction, which is that the debasement trade, as a price in the bond market today, has not been bought.

§The two items left unresolved

Two items remain unresolved rather than contradicted, and the distinction matters. The first is the doom loop, the sequence in which the central bank hikes, the interest bill rises, the deficit widens, more bonds are issued, and yields rise again. The earlier examination called it a forecast rather than a description, because the record shows the policy rate is the lagging variable rather than the leading one. The federal funds effective rate was three point eight eight per cent on the twenty-eighth of September 2026, up from three point six three per cent in the middle of that month, and a year earlier it was four point three three per cent.

So the loop’s steps are coherent and its timing is not established, which is why it stays unresolved rather than moving to either column. The mechanism describes how the system transmits stress; the prediction is that the transmission happens on the next turn of the screw. The mechanism we can carry. The next turn is a question.

The second unresolved item is the carry-trade conclusion, and the earlier work drew a sharp line through the middle of it. The mechanism is real, which is that the rest of the world owes dollars and owns United States assets, so a stronger dollar is what triggers the selling. But the conclusion drawn from it, that the dollar cannot be allowed to strengthen much, runs into a problem the trilogy flagged. It says the system can be broken by a stronger dollar while the forecast on the table is a weaker one.

Unresolved is the fair grade, and the wall it runs into is worth naming once more. Nothing in the mechanism requires the United States to be unable to fund itself, and the claim that it does is a separate step that the evidence does not take. A structure in which foreign holders are price-sensitive and the dollar’s moves change their behaviour is a real structure. What it delivers on a schedule is not visible inside it.

§The long bond, repriced rather than abandoned

There is one further item from the old ledger that the current data makes sharper, and it is the claim that the government cannot sell long-dated debt. The earlier grade was directionally supported but literally too strong, because the thirty-year bond auctioned on the tenth of September 2026 cleared a bid-to-cover of two point six one times, which is the ratio of bids received to the amount of debt offered. The twenty-year on the fifteenth cleared two point five seven times, and the ten-year on the ninth cleared two point seven one times. The auctions cleared.

Of the two claims inside that sentence, the price claim is the one that matters, and it is the one that has held. The thirty-year yield was five point five six per cent, the highest since 2004, and the ten-year at five point two four was the highest since 2007. A market can clear with no shortage of buyers and still charge more for the privilege. The buyer did not vanish. The buyer repriced.

And the speed of the repricing is the part that cuts in his favour. The seven-year auctioned at four point five one two per cent on the twenty-seventh of August 2026 and at five point zero eight five per cent on the twenty-fourth of September. That is fifty-seven basis points of repricing in a month on one maturity, and a basis point is a hundredth of a percentage point. A sovereign that is refinanced continuously has to find the buyer again every few months, and the price it pays to do so is being reset that fast.

Read the record as a whole rather than item by item. The direction of yields, the mortgage rate, the deficit level, the size of the debt, the trillion-dollar interest bill, the roll and the warning about foreign holders all aged well, while the eight point three trillion refinancing figure, the Asian surplus flow, the interest-as-a-share-of-output projection and the demographic share failed to confirm. The debt threshold, the debasement price and the carry-trade conclusion ran the wrong way or ran ahead of the evidence. That is a mixed record with a clear shape, and the shape is that he was early and roughly right on direction, and precise but unverified on magnitude.

§Where the ground position refines us

Which brings the examination to the part where he refines us, because a ground position gives an operator things that documents do not. We triangulate the aggregate. We cannot see the order book, the phone calls, or what a Treasury Secretary intends. He sells into that market and runs a balance sheet inside it, and that is the case for listening closely even where the ledger is thin.

The first thing he refines is the mechanics of a currency intervention and how a Treasury Secretary signals a trade to the market. On the thirty-first of July 2026 the United States sold euros from its international reserves and bought yen alongside Japan, after the yen fell to a forty-year low, which the Council on Foreign Relations and Fortune both report. He reads the euro purchase as a message rather than a currency choice, and the message is the part documents struggle to price.

The signalling he describes is documented, which is why the reading deserves weight. Reuters photographed the Treasury Secretary’s notepad reading buy Japanese yen, five to ten billion, at Camp David on the thirty-first of July, and the range that appears in his account is drawn from that photograph, with the actual amount undisclosed. Then on the twentieth of August he said on CNBC that people have bad information, that he has asymmetric information, and asked what he knows that the market does not. His reading of that language is that it is a solicitation rather than a confession.

The second thing he refines is the decision to fund a yen purchase with euros rather than dollars. The choice of euros is the detail the documents confirm, and the inference he draws from it is his own. Selling dollars to buy yen would strengthen the yen and weaken the dollar, which is not the signal he thinks the Treasury wanted to send. Selling euros strengthens the yen and leaves the dollar where it was, and he reads the deferral as a statement about which currency is being asked to absorb the pain.

His account of that operation is more careful than the press version, and it is worth saying so. He says the intervention was the first time the United States bought yen to support Japan’s currency since 1998, and the Council on Foreign Relations’ Brad Setser, who worked on currency policy at the Treasury, agrees. The looser framing in parts of the press, which counts any intervention for Japan, includes 2011, when the direction was the opposite one. He has the distinction right, and it is the press that blurred it.

§The holder with a name, and the French inventory

The third thing he refines is the practical floor level of what a holder does when a currency moves against it. His line is that foreign holders are the sellers of first resort when they need money themselves. Where we see an aggregate fall in foreign holdings, he sees a named holder with a position to defend and a currency to repair, and that is a step closer to the hand on the sell button.

The fourth thing he refines is the detail that Japanese investors are the most overweight holders of French bonds. They held an estimated twenty-three trillion yen, about one hundred and forty-five billion dollars, of French bonds as of July of 2026, the most overweight position in the euro area relative to global bond benchmarks, and about six point six per cent of Japan’s overseas bond holdings, on Bloomberg calculations reported by CTOL and Equiti. His claim that Japan is buying French debt is not a smear. It is an inventory, and the inventory is documented.

That detail is what turns a general worry into a named channel.

There is a fifth thing, and it is a concession rather than a refinement. He says he has no proof and calls the motive speculation.

Two of his claims that sit near this beat are checkable and check out, which is worth stating because it is the kind of thing that gets lost in a critique. Non-residents held sixty-one point two per cent of France’s long-term debt securities at the end of March 2026 and bought one hundred and thirty-two billion euros of French debt securities in the first quarter, on Banque de France figures. His one sentence, that around sixty per cent of French government debt is owned by foreigners, is verified.

And his number on the France-to-Germany spread is in the right place, which is the same grade the older ledger kept giving. He puts it at one hundred and thirty-five basis points, and Reuters on the first of October had the premium at a fourteen-year high of roughly one hundred and thirty-one basis points, with other measures that week reading about one hundred and forty to one hundred and fifty-two. The BIS chart note had it above one hundred and forty, the widest since the 2012 crisis. His figure is a shade high and directionally correct.

Where his ground position is hardest to match is the euro’s fall against the yen, and there he understated the move rather than overstated it. He says the euro is down four per cent against the yen since the intervention, and the European Central Bank’s own reference rates have the euro at one hundred and eighty-six point nine nine yen on the thirtieth of July and one hundred and seventy-eight point one five on the sixth of October, which is a fall of four point seven three per cent. Against the dollar over the same window the euro fell only one point eight per cent. The euro-specific move he described is real and slightly larger than he said.

§The through-line: not the destination but the route

Now the through-line, and it is the thing that ties the earlier examination to this one. The trilogy against Vague concluded that every road out of the debt problem ends in a weaker dollar, because the stock was issued at old rates, refinances into new ones, and the arithmetic of funding a sovereign that borrows in its own currency and cannot raise enough revenue points at the currency. That conclusion is where the old work landed, and it is where the new episode begins.

What the new episode adds is not the destination but the route. If every road ends in a weaker dollar, the useful question is which road gets taken first, and by whom, and on what trigger. His answer is a sequence rather than a prediction. Phase one is the Fed tightening and the dollar strengthening while the weakest balance sheet cracks, and he puts France first. Phase two is the rescue, in which something breaks and someone is bailed out and money is created and the dollar turns lower.

That sequencing is the actual claim, and it is a different claim from the one the older trilogy graded. The older work graded whether the destination was where the roads led. This episode is an argument about the order of travel and about who is asked to absorb the loss on the way. A destination claim and a route claim fail in different ways, and only the route claim can be wrong about the calendar while being right about the direction.

Which is exactly where the case becomes hard to check. The direction can be verified against prices and holdings, and it has been. The route runs through motive, through a Treasury Secretary’s intent, through the order in which sovereigns are sold, and none of that sits in a document a series can reach. He can see the intent from a position inside the market. We can see the consequences, and the consequences arrive only after the intent has acted.

Part three takes up where his case breaks, where it cannot be checked from documents, and what a listener would watch to settle it, because between a direction that has arrived and a route that has not, the third part is the one that has to be rebuilt on dates.

This has been a deep dive into what the record shows so far about Jack Mallers’ debt-and-dollar thesis on Europe.

Sources: this examination names its sources inline — Jack Mallers on his show and in the episode Europe Is Screwed, and Richard Vague, both weighed as claims rather than as persons; the Treasury’s published debt, interest and auction data; the Treasury International Capital holdings series; the Reuters wire; the Banque de France; the Council on Foreign Relations, including the work of Brad Setser; the European Central Bank’s reference rates; the International Monetary Fund; the Congressional Budget Office; the Bank for International Settlements; Bloomberg, as reported by CTOL and Equiti; and the market prices for breakeven inflation, real yields and the federal funds effective 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.