GLOBAL FINANCIAL ENGINEERING, INC.

Public Doctrine Series · The Engineered Long End — Paper II of V

The Two Balance Sheets

Why Washington Needs Tokyo to Raise Rates, and Fears What Happens When It Does

Document GFE-PUB-DOC-20260909-B · 9 September 2026 · Dr. Glen Brown, President & CEO, Global Financial Engineering, Inc. · Global Internal Governance Chamber

Educational doctrine, not investment advice. Published simultaneously on globalfinancialengineering.com and globalaccountancyinstitute.com.

Abstract

Paper I of this series closed on a question: who holds a sovereign's twenty-year promise at these prices, and at what premium. This paper answers the first half. Japan holds roughly 1.1 trillion dollars of United States Treasury securities, the largest foreign position in the world, and has held above one trillion for more than a decade. Over the same months in which the United States Treasury enlarged its long-dated buybacks to support the long end of its own curve, senior American officials pressed Tokyo to communicate a path toward fiscal sustainability and toward higher policy rates. The two policies point in opposite directions, and the mechanism connecting them is not political but arithmetic. A Japanese institution does not compare a Treasury yield with a Japanese government bond yield; it compares a Treasury yield net of the cost of hedging the currency with a domestic yield that carries no such cost. When domestic yields rise, that comparison moves against the foreign asset — and the largest single foreign buyer of American duration has less reason to buy. This paper sets out the arithmetic, quantifies what has already changed, and shows why a buyback programme and a request for Japanese rate rises are two halves of a single contradiction. It closes with an executed argument in which the institutions' own futures book held European duration short and held no yen position at all — and why that silence is itself reportable.

I. The Position, and Why Its Size Is the Whole Story

Japan is the single largest foreign holder of United States Treasury securities, with roughly 1.1 to 1.2 trillion dollars on the books through 2026 — a figure that has stayed above one trillion for over a decade. Against total foreign holdings of approximately 9.3 trillion, Japan alone accounts for something near thirteen percent of all foreign-held American government debt. Japan, the United Kingdom and China together hold about a third of the foreign total.

Concentration of that order changes the character of the relationship. When one country holds more than an eighth of all foreign Treasury exposure, its buying and selling decisions carry influence over the world's benchmark debt market that no policy statement can substitute for. And the direction has already turned: Japan's holdings declined roughly 3.3 percent year on year to June 2026 even as total foreign holdings rose 2.3 percent. The largest holder was trimming while the aggregate was growing.

The composition matters as much as the size. Japanese Treasury holdings are not principally a central-bank reserve position. They are held by life insurance companies, by pension funds including the government pension investment fund, and by banks — private institutions with domestic liabilities, regulatory capital requirements, and boards that answer for unrealised losses. They are not obliged to hold American paper. They hold it because, for thirty years, there was nothing at home that paid.

II. The Arithmetic That Governs the Flow

This is the section most commentary omits, and without it the entire subject is unreadable.

A Japanese life insurer has liabilities in yen. If it buys a dollar-denominated Treasury, it takes on a currency exposure it cannot prudently leave open, so it hedges — historically at ratios between fifty and sixty percent of foreign assets, using foreign-exchange swaps and outright forwards. The cost of that hedge is set, in essence, by the difference between short-term dollar and yen interest rates. The relevant return is therefore not the Treasury yield. It is the Treasury yield minus the hedge cost.

For three decades that calculation was trivially favourable, because Japanese short rates sat at or below zero and Japanese long yields offered almost nothing. Hedged Treasuries paid; domestic bonds did not; capital flowed outward, and in flowing outward it funded American asset prices. That flow was one of the largest and most persistent in global finance.

The arithmetic has now inverted. As the Bank of Japan normalised policy, short-term yen rates rose and the hedge became expensive; simultaneously long-dated Japanese government bonds began to pay a real yield for the first time in a generation. By 2025 the hedged return on long Treasuries had turned negative for a Japanese buyer against a domestic alternative, and by 2026 the ten-year Japanese government bond yield had reached about 3 percent — its highest since 1996 — with the thirty-year having touched levels never seen since the tenor was introduced.

The consequence follows without any need for intention. A Japanese institution that sells a low-coupon domestic bond and reinvests at 3 percent at home, in its own currency, with no hedge to roll and no basis risk, has done something simpler and safer than buying American duration and paying to protect it. The flow reverses not because anyone decided to punish anyone, but because the spreadsheet changed.

Scale gives the reversal its weight. The Bank for International Settlements has documented foreign-currency exposures at Japanese life insurers and other non-bank financial institutions running into the hundreds of billions of dollars in derivatives alone, with the cost of rolling hedges rising since 2022. A repricing of that hedge is not a marginal adjustment to one investor's portfolio; it is a change in the terms on which a very large pool of savings is willing to fund a foreign government.

III. The Contradiction, Stated Plainly

Now place the two policies side by side.

On one side, the United States Treasury enlarged its long-dated buyback programme through August and September 2026 — from two billion dollars per operation to at least four, then to six for the first operation — for the declared purpose of supporting liquidity at the long end of its own curve, against a backdrop of federal debt above forty trillion dollars and reduced foreign official purchases.

On the other, senior American officials pressed Tokyo publicly on fiscal sustainability and on the path of policy rates, and Japanese borrowing costs rose to thirty-year highs in the days that followed. Market pricing moved toward a further rate rise at the September meeting, with the Bank's own governor stating that a rise would be considered at every meeting.

Each policy is defensible in isolation. A stronger yen relieves imported inflation in Japan and eases a currency tension that has troubled both capitals. A larger buyback supports a market under strain. But taken together they form a loop. Higher Japanese policy rates raise the hedge cost and raise the domestic alternative; both effects reduce the appeal of hedged Treasuries to the largest foreign holder of them; reduced foreign demand at the long end is precisely the condition the buyback was enlarged to offset. Washington is asking Tokyo to do the thing that thins the bid, and then bidding for the paper itself.

The buyback, in this reading, is not merely a liquidity operation. It is the shim between two policies that are one problem — the dollar's terms of trade on one side, the Treasury's cost of funds on the other — and the shim is six billion dollars against a stock of forty trillion.

IV. What an Engineer Notices About This Structure

Three properties of the loop deserve naming, because they determine how it behaves rather than how it is described.

It is a positive-feedback loop, not a stable control system

A stable system corrects toward a set-point. This one amplifies. Higher Japanese rates reduce Japanese demand for Treasuries; reduced demand raises American long yields; higher American long yields widen the gap that hedging must cover and increase the pressure for further support operations; and the support operations, funded from bills, shorten the American maturity profile in a way that makes the Treasury more sensitive to its own front end. Nothing in that chain pushes back toward equilibrium. A system with no negative feedback is not managed by adjusting the input; it is managed by changing the structure.

The actors are private, and cannot be instructed

This is the property most often missed. Japan's Treasury position is not primarily a sovereign reserve that a government can be persuaded to hold. It sits with insurers and pension funds acting under domestic regulation, with asset-liability matching obligations and boards accountable for unrealised losses. Japan's financial regulator has already moved to review major life insurers' balance sheets for unrealised losses on bond holdings after a sharp move in domestic long yields. Those institutions will do what their liabilities and their regulators require. No amount of diplomacy changes the hedged-yield arithmetic on which they act.

The adjustment horizon is years, not quarters

Institutions of this kind plan on two- and three-year cycles, and their portfolios turn over slowly. That cuts both ways honestly. It means the reversal will not arrive as a single dramatic event, which is a genuine argument against alarm. It also means that once turned, the flow does not turn back quickly, and that the effects observable in 2026 are the leading edge of an adjustment already several years underway rather than its conclusion. Precedent supports caution in both directions: an earlier episode of improving domestic yields in Japan produced far less repatriation than expected.

V. What Would Falsify This Reading

Four observations would refute the account above, and all four are available in public data.

If Japanese holdings of Treasuries stabilise or rise over the coming quarters while Japanese policy rates continue upward, the hedged-return mechanism is weaker than argued, and habit, mandate or asset-liability structure dominates the arithmetic. If the cross-currency basis narrows so that hedge costs fall even as short rates rise, the arithmetic improves without any policy change and the loop loosens. If long-dated Treasury yields decline over the remaining buyback operations while Japanese yields also rise, the foreign-demand channel is not the binding constraint on American long rates. And if Japanese life insurers' published investment plans show renewed appetite for foreign bonds at current hedge costs, the institutions themselves will have contradicted the model — which is the strongest refutation available, since they are the actors in it.

The instruments to check are the monthly Treasury International Capital release, the semi-annual investment plans the large Japanese insurers publish, the cross-currency basis, and the Bank of Japan's own decisions. This paper's reading should be graded against them rather than against its own plausibility.

VI. The Executed Argument

Per the standing template of this series, the institutions' own book is reported retrospectively, in the past tense, as an exhibit and never as a template, with the paid and the unpaid given equal weight. The futures book is referred to as GATS7.

Doctrine restated

A book positioned for a global repricing of the term premium is positioned for it wherever the repricing occurs, and the currency in which it expresses that position is a matter of instrument availability rather than of conviction. The doctrine also holds a harder companion clause: where a book holds no position, it has no reading, and the absence must be reported rather than filled in.

Signature observed

Through August and into September 2026, GATS7 held a short position in FGBL_U and FGBL_Z (Euro-Bund futures, September and December) across sixteen legs, every one at or above breakeven and five locked in profit. It held long currency futures against the dollar — 6A_U (Australian dollar), 6B_U (British pound), 6C_U (Canadian dollar), 6E_U (euro), 6N_U (New Zealand dollar) — with 6S_U (Swiss franc) short. It held long equity-index futures in both the American and European complexes, which carried the book's entire floating deficit.

It held no position in 6J (Japanese yen futures) and none in ZN or ZB (United States Treasury note and bond futures). On the two instruments most directly implicated in this paper's argument, the book was silent.

Expression

The European duration short was the book's only rates expression, and it paid, because European duration was being sold alongside American duration in a global repricing that no single treasury authored. The dollar-negative currency complex was consistent with yields rising for term-premium reasons rather than growth reasons — the signature this series has identified as the regime's mark. The index longs paid the cost of the higher discount rate that both the buyback and the pressure on Tokyo were, in different ways, responses to.

Consequence

Realised: the book stood at a lifetime profit factor above five across one hundred and seventy-three closes with a maximum balance drawdown below one percent. Unrealised: the equity-index complex remained under water and bounded by constitutional stops, and its hedge coverage was found by measurement to be roughly a fifth of what doctrine required — a defect named and corrected in the same week it was found. On the yen and on American duration the book earned nothing and lost nothing, because it was not there.

Lesson banked

The most disciplined thing a reader can take from a book is where it is silent. GATS7's rates voice in this period was European; its yen voice did not exist. A commentator may write about the yen without holding it and lose nothing by being wrong. An institution that reports its own book must say plainly that on the central instrument of its own argument it held no position, and therefore that the argument in this paper is reasoning from public arithmetic rather than testimony from an executed trade. The distinction between the two is the whole difference between doctrine and opinion.

Paper III of The Engineered Long End — Yields Rising with a Falling Dollar: Reading a Term-Premium Regime from a Book That Has No Opinion — follows.

Attribution: Dr. Glen Brown / Global Internal Governance Chamber.