The Bond Thesis, Examined

• The basics
• The claims
• The agreements
• The 1951 story, running backwards
• The disagreements, examined
• The verdict

An essay by DeepDives — companion to the episode of the same name. The audio argues in full; this is the examination in writing. Plain language throughout: every term explained where it first appears.

A guest appeared on a bitcoin podcast in the first week of September and spent an hour explaining what he thinks is happening to the American government’s debt, which has now passed forty trillion dollars. His name is Nik Bhatia, and his views deserve a careful look, because he is the most qualified bond-market voice in the bitcoin world. He traded bonds before he became interested in bitcoin. He teaches finance at the University of Southern California, in a course about bonds themselves. His first book, Layered Money, is about the machinery underneath the financial system — the overnight lending market and the layers of borrowing that most commentary never touches. None of that makes him right. It makes his arguments worth testing against the historical record, which is what this essay does. And one thing up front, because honesty cuts both ways: Bhatia is not a neutral observer. His research firm sells its analysis, and he owns bitcoin. The author of this essay has published his own conclusions about where the debt ends, and holds the assets that fit those conclusions. Nobody here is unbiased, so we name it on every side and test the arguments anyway.

§The basics

A government raises money by selling IOUs called bonds: promises to repay the money later, with interest. The interest rate on those IOUs is set by the market — by the people willing to lend. Confident lenders accept low rates; worried lenders demand high ones. The market where these government IOUs trade is the bond market, and the American government’s IOUs, called Treasuries, are the most important pieces of paper in the world, because the global financial system treats them as the safest thing that exists. When you hear that a yield went up, it means the interest rate the government must pay went up, because lenders became less willing to hold its debt. A yield is the market’s opinion of the borrower, updated every second.

Two terms do most of the work in this story. The first is the repo market. Repo is short for repurchase agreement, and the repo market is the overnight lending market of the financial system: banks and funds lend each other cash for a day at a time, putting up government bonds as a deposit, the way a pawnshop holds collateral. Almost every financial machine runs on this overnight lending, so when the repo market runs short of cash, the plumbing of the whole system backs up. The second term is bills. Government debt comes in lengths. Bills are the shortest: IOUs repaid within a year, sometimes within weeks. Bonds, strictly, are the long IOUs, running ten, twenty, or thirty years. Short IOUs versus long IOUs turns out to be the hinge of the whole story.

§The claims

Bhatia’s position, from his September appearance and his written research, is a system of six claims.

First, the buyback program is not control of interest rates. In August the government doubled a program that buys back its own older bonds, right after the interest rate on thirty-year debt hit its highest level in nearly two decades. Much of the bitcoin world called this yield curve control — the government setting the interest rate on its own debt by force. Bhatia disagrees, on technical grounds. The program buys only the older bonds, the ones that are no longer freshly issued and heavily traded, and which are therefore hard to sell quickly. In market language, the newest bond is on the run and trades constantly; older ones are off the run and trade thinly. Bhatia argues the buyback exists to keep that hard-to-sell corner of the market working, not to control prices. Dollar for dollar, he says, it is small. Then he conceded the part that makes the distinction slippery: the government pays for the buybacks by selling new short-term IOUs, the bills, and using the cash to buy long bonds. Selling short and buying long pushes the two ends of the market closer together — the technical definition of flattening the curve. So even on his own description, the program quietly does some of what the control crowd claims. He also cited the famous investor Stanley Druckenmiller, who has written that the government cannot paper over the truth: how much debt there is, and how much new borrowing the market must absorb.

Second, the government will fund itself with short-term IOUs. His phrase: the plug is with bills. The long bonds have become so expensive that the government would rather borrow at the cheap, short end, constantly rolling over bills that come due in weeks and months. But there is a limit to how many bills the market can absorb, because they crowd out everything else: only so much cash exists to buy them, and only so much cash to lend overnight against them. At a certain point the flood of bills causes a cash shortage in the repo market, and a cash shortage there can be fixed by exactly one player: the Federal Reserve, America’s central bank, the institution that can create cash from nothing. Bhatia notes this is not hypothetical — at the end of 2025 the Fed bought bills, and stopped once the overnight market calmed. His near-term map is mechanical: borrow short, saturate the market, seize up the plumbing, and the central bank is forced back in as the buyer of the government’s short-term IOUs.

Third, interest rates can only fall through a recession. The government’s annual interest bill is now around a trillion dollars — more than its military budget — and Bhatia argues this bill forces every decision. Cut the interest bill in half and the government’s finances become tolerable. How do rates come down? You need a recession: people get scared, stop spending, hide their money, and rates fall; the central bank cuts in response; inflation cools enough that lenders accept small returns again. He was honest about the conditional: prices are not exploding but not falling, there is no recession, there is an investment boom, and jobs are plentiful. The falling-rates scenario is a map of how it must end, not a forecast of next year.

Fourth, the near-term path is a negotiated deal. If you cannot cut rates because inflation is too high, and cannot raise them because that would break the government’s finances, you negotiate. Bhatia expects a managed arrangement between the Treasury, the department that borrows, and the Federal Reserve. He points out that the Treasury Secretary and the incoming Fed chair have been seen traveling together, and argues they must be discussing how to handle this. He has written about the historical template: the famous 1951 agreement between the Treasury and the Fed — the subject of the sharpest disagreement in this essay, below.

Fifth, the crisis lands in Britain, France, and Japan first. These three, Bhatia says, are much more likely to see massive central-bank or government intervention in the next six to twelve months than the United States. France cannot print its own money. Germany, notably, he exempts. And he suspects the American players are comfortable with rates rising abroad, as long as somebody over there breaks first — that Europe’s central bank may be forced to reverse its own rate-raising because the cost of lending long-term is being pushed onto it from America.

Sixth, there is a price at which buyers return. When long-term rates get high enough, holding government debt becomes a paying business, and investors can borrow cheaply in the overnight market to buy more of it, multiplying their returns. The repricing itself creates its own buyers.

§The agreements

Tested against the record, Bhatia agrees with this series’ conclusions more than either side might expect.

He agrees on the cast of characters. This series has argued that the story of 2026 is three debtors with three relationships to the printing press: the United States, which owes the world and prints its own money; Japan, which owes itself and whose central bank owns half its government’s debt; and Europe’s shared currency, which contains both countries that print, chiefly Germany, and countries that gave up their currencies and cannot. Bhatia’s crisis map is that framework applied: the countries that cannot print break first, and the biggest such country in 2026 is not Greece, the crisis country of 2011, but France — Europe’s second-largest economy, running a deficit the markets have begun to punish, with no currency of its own and no central bank that answers to Paris. Britain is the country whose bond market already broke one government this decade, in 2022. The rule of history is that the ones without a printing press go first, and the country that prints the world’s reserve currency — the currency other nations hold as their official savings — goes last. And the suspicion that America might be happy to let someone else break first is not paranoia; it is the oldest trick of the reserve-currency issuer, which can export its inflation, its rates, and its pain to everyone else.

He agrees on the engine. The debt is no longer a tool; it is the master, and the master’s first demand is cheaper borrowing. The trillion-dollar interest bill eats the tax revenue and forces every choice. On this the two analyses run on the same fuel.

And he reaches for the same history: the 1951 agreement between the Treasury and the Federal Reserve, which this series also used as the anchor of its bond-market episode. Two independent analyses, one from the record and one from the market’s machinery, both reaching for the same seventy-five-year-old agreement as the key to the present. That convergence is itself evidence both sides are looking at the same animal.

§The 1951 story, running backwards

The sharpest point of the examination is what the 1951 agreement actually means, because Bhatia is reading the template backwards.

In 1951, the Federal Reserve was escaping the government. During the Second World War the Fed had been ordered to keep government borrowing cheap no matter what, to help pay for the war. By 1950 that arrangement had become a trap: the Fed was creating money to buy the government’s debt, prices were rising, and the central bank wanted its independence back. The 1951 agreement was the Fed’s declaration of freedom — the founding moment of central-bank independence in America. Now look at today’s cast. The Treasury Secretary is the one with the agenda: doubling the buybacks, managing the long end, in constant contact with the incoming Fed chair. A deal struck in 2026 would not be the central bank escaping the government. It would be the central bank formally accepting its place beneath the government — the 1951 story running backwards. Bhatia calls the coming arrangement a managed situation, and he is probably right that management is what they will attempt. But management is the word every suppressed market uses for control while the control still works. The wartime arrangement was management. Japan’s years of capped interest rates were management. Europe’s whatever-it-takes promise of 2012 was management. The difference between management and control is a matter of degree and time, and the record says the market eventually takes back control — harder the longer the management lasted.

§The disagreements, examined

On the buybacks, Bhatia’s technical point is fair, and it is the kind of point only someone who has traded these instruments makes. Calling the program full control of interest rates is not accurate: it buys old, hard-to-sell bonds, does not aim at a rate, and is small. The honest response is to accept the correction and add the complication. Intent is not nothing, and the history of every control regime starts exactly where Bhatia says this one is not yet. The wartime arrangement began as cheap war financing, and the control emerged from the commitment. Japan’s version began as managing a mountain of debt, and the cap emerged later. Every regime that ended up suppressing its bond market described itself at the start as helping the market work better — Bhatia’s exact description. The question is not whether the program controls rates today. It is whether a government buying its own debt, and doubling the purchases whenever rates rise, is still just helping the market the day the market really tests it. The August doubling was the first step on that ladder. The record says the steps continue until the market or the inflation breaks.

On the road to the end, Bhatia’s map improves this series’ own, and the credit should be given plainly. Earlier episodes described the end as inflation and quiet confiscation — the debt repaid in watered-down money — without spelling out the path. Bhatia spells it out: borrow short because the long end is too expensive, flood the market with bills, starve the overnight lending market of cash, and the Fed, the only player who can fix that, buys the bills. The short end is the door the money comes through. And buying short-term IOUs to fund a government that cannot sell long-term bonds is creating money to pay the government’s bills — entering through the back door, dressed as plumbing repair rather than printing. Every suppression regime described its own money creation this way. The 1940s called it the war effort. Japan called it stability. The Fed will call it fixing the plumbing. The names change; the mechanism does not. One warning from history: borrowing short because you cannot borrow long is what every struggling country did in the year before its crisis. The short-term fix is real. It is also, in the record, a countdown.

On falling rates, the disagreement narrows to a genuine fork in history. Bhatia’s sequence — recession, cuts, cooling prices, cheap refinancing, Japan after 1990 — has happened, and the examples are America after 1945 and Japan after its bubble. The other sequence, from the other half of the record, is the 1970s: inflation first, rates forced up, the bond market breaking, and only then the recession — and that cure was brutal. Which road is 2026 on? The uncomfortable part for the falling-rates camp is that Bhatia’s own description of the present contains the ingredients of the 1970s: an investment boom, plentiful jobs, prices that are not falling, a government borrowing two trillion dollars a year, and a central bank whose independence is being negotiated away in real time. That is the recipe for the road where the bond market forces the issue first — which is why thirty-year rates hit their highest level in nearly two decades this summer, and why the government had to double its buybacks in response. The steelman for his reading is also real: a trillion-dollar interest bill drains money from the economy every year, and a government that must constantly refinance cannot afford the high rates that would break inflation quickly. There is a real ceiling on rates. The honest verdict: his near-term politics are probably right, the negotiated path is probably what they will attempt, and his long-term story is a bet against the 1970s, in a situation that looks more like the 1970s than he concedes. Watch the inflation numbers: if prices start falling, his road opens; if they stay sticky while the boom runs, the bond market’s road opens.

On Britain, France, and Japan breaking first, he is probably right, with one correction. The ones who cannot print go first: that is the rule, because the printer of the reserve currency is always the last domino. France is the prisoner at a scale Europe has never had to rescue, and the argument that Europe’s central bank will eventually print for its prisoners is the argument that produced whatever-it-takes in 2012; the flaw in the union has been hidden behind Germany’s strength, not repaired. Britain has the freshest scar and the classic combination of a budget deficit and a trade deficit. Japan is the slow-motion case: its central bank spent decades buying its own government’s debt and is now slowly selling, while Japanese insurers sell foreign bonds to pay obligations at home. His exemption of Germany is right, and matches this series’ argument: Germany can print, and it has chosen to borrow half a trillion euros for defense — turning from Europe’s safe parking spot into a competitor for its savings, which is the more important story. The correction is the one history always applies to timing: the six-to-twelve-month window is the weakest part of the thesis, because intervention is a choice made by people, not a mechanical schedule. And the someone-else-breaks-first strategy has a failure mode: when the prisoners break, the panic does not stay in the prison. In 2011 the crisis in small Greece pushed the rate on much larger Italy’s debt toward seven percent within months. The wires carry the panic everywhere — as they did in 2011 and again in 2022, home included.

§The verdict

On the mechanics, Bhatia is right: the buyback program is help for the old, hard-to-sell bonds, not a target on interest rates, and the honest description of the alternative reading is that it reads the government’s intention rather than the program’s mechanics. On the near-term map, he has improved the record: the short-term borrowing, the cash shortage, and the Fed buying the bills is the most believable path for the money creation to enter, it has already happened once, and his marker — the overnight lending market — is the right one to watch. On the order in which countries break, he is probably right. On the destination there is no disagreement: the debt is resolved by creating money and quietly taxing the lenders through inflation, in some order, because the government has ruled out every other option. And on the two genuine differences — the meaning of the coming deal, and the order of the endgame — the examination says this. The deal, if it comes, is the 1951 story running backwards: the central bank accepting its capture rather than escaping it, the return of controlled interest rates in modern clothes. And the falling-rates road requires prices to stop rising, which his own description of the present does not show, while the ingredients he describes are the ingredients of the road where the bond market forces the issue first. He may be right that the negotiated path holds for the next year. The record says the negotiation is the prologue.

The way to decide between the two readings is specific markers. Watch the share of government debt issued as short-term IOUs: if it climbs toward the levels that caused the last two cash shortages, his machine is running. Watch the overnight lending market: when it strains, the Fed is coming back in. Watch whether the buybacks get bigger, and especially whether they ever start buying the fresh, heavily-traded bonds — the moment the helping-the-market story ends and the controlling-the-price story begins, whatever it is called. Watch the gap between what France pays to borrow and what Germany pays. Watch whether Japan’s central bank keeps shrinking its holdings of its own debt. And watch the inflation numbers, because they choose between the two roads. Whoever’s markers fire first wins the round.

Nik Bhatia is not a neutral observer, and neither is anyone publishing this conclusion. His firm sells the analysis his framework produces, and his audience owns bitcoin; the falling-rates ending is the ending his framework needs. The author of this essay has published the conclusion that the dollar’s debt ends in inflation and quiet confiscation, and holds the assets that survive it. The incentives on both sides of this family argument point the same way, which is exactly why the argument should be tested rather than enjoyed. The assets that survive both roads are the same assets, for the same reasons. The governments and the plumbing come and go. The things that cannot be printed remain.