What Monetary Sovereignty Cannot Do: Britain's Own Currency and the 27-Day Constraint

There is a test that decides, for some economists, whether a country’s government is genuinely free to spend. It is not debt-to-GDP. It is not the deficit. It is whether the state issues the currency its own obligations are denominated in.

If it does, the argument runs, the state can always pay. It cannot run out of pounds, because it makes the pounds. Bond markets cannot force it to default, because it never needs their money in the first place. On this view, the countries that suffer at the hands of creditors are the ones that borrowed in someone else’s currency, or handed their monetary policy to a central bank they do not control.

That test is a real insight, and it is mostly correct. It is also insufficient, and it is worth showing why - not with a foreign example, but with Britain in September 2022.

The claim, stated properly

The accounting claim is not controversial among people who have looked at it. A sovereign state with its own non-convertible currency and a floating exchange rate settles its spending by crediting bank accounts, and it taxes to withdraw purchasing power from circulation. It does not collect tax in order to have something to spend. That is a description of how the plumbing works, and it is not seriously disputed.

What follows from it is what really matters. If money is not the constraint, something else must be. The answer, in this framework, is real resources: inflation arises when total spending outruns the economy’s capacity to produce. The constraint is not the bank balance. It is the supply of things.

That is a better account of the world than “the government must budget like a household,” and it is already the position this site has taken. Where Does Money Come From? sets out what the framework actually says, and why the conventional story fails.

It also leads somewhere useful. The 2 percent inflation target most central banks organize their lives around has no real evidence base, as will be shown below. And Hyperinflation makes the point that a currency collapse is a collapse of institutional credibility, not an arithmetical consequence of printing.

This piece is about the step after that. If the constraint is real resources and institutional credibility, then a country with its own currency is not automatically protected from either. Britain has the strongest version of the freedom the test describes. It also has a recent, documented case of that freedom failing to protect an elected government from the bond market.

What happened in September 2022

On 23 September 2022, the UK chancellor announced a budget of unfunded tax cuts. Markets did not wait for the details to be modelled.

Gilt prices fell steeply, and yields and swap rates rose with them. The mechanism was not a general loss of confidence in the currency. It was specific and structural. UK defined-benefit pension funds held large positions in long-dated and index-linked gilts, managed through liability-driven investment (LDI) structures. Those structures are highly sensitive to changes in interest rates. As gilt prices fell, the funds faced collateral calls, and they met them by selling gilts. That selling pushed prices down further, which generated more collateral calls. Liquidity in long-dated gilts evaporated.

The Bank of England intervened to stop the spiral. The chancellor was dismissed on 14 October, and the prime minister resigned on 20 October - 27 days after the announcement.

This was not the bond market refusing to buy. The settlement system for gilts is the Bank of England itself, which is why a sovereign issuer cannot be forced into default through a buyers’ strike. That argument is correct, and Where Does Money Come From? sets it out at length.

What happened in 2022 was different in kind. The forced selling came from domestic institutions meeting collateral calls, and the constraint arrived as a collapse in market liquidity rather than a refusal to lend. The Bank could settle the trades. It could not stop the policy from being reversed. The cost was not default. It was the loss of control over a decision the government had every legal right to make.

Every element of the monetary-sovereignty test was satisfied. Sterling is the UK’s own currency. The UK’s debt is denominated in that currency. The Bank of England is the UK’s own central bank, and it sets the UK’s own interest rate. None of it mattered to the particular constraint that arrived.

The Bank’s own account, published as a staff working paper in 2023, describes the episode as a liquidity crisis in a specific market, driven by the derivative and repo positions of liability-driven investors. That paper is not a political document. It is the central bank explaining, in technical detail, how a solvent sovereign found itself with no choice.

Why “sovereignty” missed it

The test asks one question: who makes the money? The 2022 episode asks a different one: who holds the debt, and on what terms?

A state can be sovereign over its currency and still be exposed through its own institutions. UK pension funds were not foreigners seeking to punish Britain. They were domestic funds, doing what their mandates and their accounting rules required, with leverage built into the structure. Their forced selling was not a judgment about the government’s solvency. It was a mechanical consequence of how the funds were designed.

This is the point The ECB Hiked Into a Supply Shock makes from the other direction. When a central bank raises rates, the question worth asking is what the hike is for - the inflation, or the institution’s credibility. In 2022 the Bank of England raised rates into a domestic institutional fault line, and then had to intervene in a specific market to contain the damage.

The state had the money. It did not have the institutions that would have let it use the money on its own terms.

The 2 percent target, and where it came from

If the constraint is inflation, then the target used to judge failure is worth examining. It does not survive the examination.

The 2 percent figure has an origin, and it is documented. In 1988, New Zealand’s finance minister was asked on television whether he was satisfied with inflation having just fallen below 10 percent. He replied that he would ideally want a rate of between 0 and 1 percent. According to Don Brash, then governor of the Reserve Bank of New Zealand, the remark was entirely off the cuff.

Having said it, the central bank had to work out what the target should be. Brash’s team knew the literature on cost-of-living measures suggested an upward bias. They estimated that bias for New Zealand at around 0.75 percent, rounded it up to 1 percent, and set the upper bound of the target at 2 percent. One of Brash’s colleagues later described it, plainly: it was not ruthlessly scientific.

Canada adopted inflation targeting next, and chose the same 2 percent. Others followed. Brash’s own explanation, offered with a laugh, is that central bankers meet at Basel and talk about these things.

The number was even less settled where it mattered most. Analysis of Federal Open Market Committee transcripts shows that US officials overwhelmingly favored a target of 1.5 percent at first. The consensus moved to 2 percent only after 2008, and for a reason that had nothing to do with evidence about price stability: a higher number leaves more room to cut rates before hitting zero.

What the literature actually finds. The largest meta-analysis on the subject - 777 estimates drawn from 116 primary studies published between 1989 and 2026 - puts the optimal long-run inflation rate at about 0.6 percent per year. That is below the target most advanced economies use, and far below the 4 percent that serious economists have argued for on the grounds that it gives central banks more room to cut. The same paper is careful about its own limits: measurement error in published price indices could close that gap, widen it, or reverse it, and the structural literature cannot pin down the sign of the correction it needs.

What is defensible. A small positive rate has real arguments behind it, and they are worth understanding. The Boskin Commission estimated in 1996 that US consumer price measurement overstated inflation by around 1.1 percentage points a year, which means a measured rate of 2 percent may correspond to something closer to 1 percent in lived experience. And nominal wages are hard to cut, so a modest positive rate lets real wages adjust without nominal reductions.

Neither argument produces 2 percent. They produce “a small positive number.” The specific figure is a convention, in the same family as 10,000 steps a day and five portions of fruit and vegetables - invented precision attached to a defensible direction.

This is where the critics of orthodox policy are right, and it is worth saying so. A target with this history is not a scientific boundary. Mervyn King, who ran the Bank of England, called inflation-fixated colleagues “inflation nutters.” The people who treat 2 percent as a law of nature are not reading the evidence. There is not much evidence to read.

The part that does not follow

Everything above concedes a great deal to the monetary-sovereignty case. The diagnosis is largely right. The target is a convention. Interest rates are a blunt instrument, and raising them into a supply shock does not touch the supply.

None of it establishes the prescription.

The proposed restraint on inflation, in this framework, is taxation - withdrawing demand by taking money out of circulation, concentrated on higher incomes and accumulated wealth rather than imposed on all borrowers through the interest rate. For inflation caused by profiteering, the proposal is regulation, excess profits taxation and competition policy. For inflation caused by energy or commodity shocks, it is direct intervention: price controls and publicly provided alternatives.

Take those in turn, because they are not equivalent in what they can deliver.

Price controls have a track record, and it is not encouraging. A controlled price below the market-clearing level reduces the incentive to supply, and shortages follow. That is not a theoretical objection; it is what happened in the 1970s, in the same economies now being discussed.

Excess profits taxation raises revenue and is popular. It does not add a tonne of steel, a barrel of oil or a trained nurse to the economy’s capacity. It moves money. When the constraint is real resources, moving money does not relieve it.

Taxation to withdraw demand is the load-bearing proposal, and it has never been operated at scale. No government has run an inflation-control policy of this kind through a sustained inflationary episode. That does not make it wrong, but it means the case rests on a mechanism with no track record, being offered in place of a mechanism that has one - and the mechanism it replaces is the one that stopped inflation in Britain the last time it ran out of control.

And the diagnosis underneath it is not settled either. The claim that spending caused the recent inflation rests on the 2021-22 episode, and the people who modelled it do not agree with each other.

The San Francisco Fed’s estimate, published in October 2021, put the American Rescue Plan’s effect on inflation at about 0.3 percentage points in 2021 and a little over 0.2 in 2022 - a modest contribution, from a labor-market-tightness model. A Federal Reserve note the following July put pandemic fiscal stimulus at around 2.5 percentage points in the United States and 0.5 in the United Kingdom, from a cross-country comparison. The same institution, two methods, a fivefold gap.

The point is not that stimulus is irrelevant. It is that the size of the effect is genuinely contested, and a case resting on “demand-pull is rare” is resting on an estimate that is not settled.

And there is a structural problem underneath the whole prescription. Taxing to withdraw demand and spending to build capacity are opposite operations. In the aggregate they cancel. The plan is therefore not a claim about capacity at all. It is a claim about composition - that the same money, spent differently, would do more good.

That claim may be true. Childcare spending genuinely raises labor supply. Targeted infrastructure spending can relieve a specific bottleneck. But the capacity argument rests entirely on that leg, and it is the leg with the worst delivery record in British public life.

Where the capacity actually went

There is a version of this argument that does not need the tax mechanism at all, and it is stronger for it.

Brexit reduced the UK’s long-run productivity. The Office for Budget Responsibility’s own assessment is a reduction of around 4 percent, and it is the evidential basis for saying Brexit has damaged the economy rather than merely changed it. That is a real capacity reduction in the same economy whose capacity is supposed to be the binding constraint on spending. Nothing about monetary sovereignty prevents it.

A government with its own currency and a smaller economy to spend against is not more free. It is less. That is the framework’s own logic applied honestly, and it points somewhere the framework’s advocates resist: the constraint that matters most in Britain is not the printing press. It is the size and productivity of the economy the printing press operates in.

What the test should have asked

The monetary-sovereignty question is worth examining. It separates countries that can always settle their own debts from those that cannot, and that distinction explains a great deal - Argentina’s repeated crises and the eurozone’s treatment of Greece are not accidents of policy design.

But it is a necessary condition, not a sufficient one, and treating it as decisive produces the 2022 error in reverse: confidence that sovereignty confers insulation.

The better test adds two questions.

Who holds the debt, and what forces act on them? Britain’s answer in 2022 was domestic pension funds with rate-sensitive leverage. Sovereignty does not protect a state from its own institutions.

What is the capacity constraint, and what is being done about it? Here the framework asks the right question and points at the wrong answer. Raising taxes does not expand capacity. Nor does regulation. Nor do price controls, which reduce supply. Only investment does - and investment is the one thing the prescription treats as a detail.

The position that survives is uncomfortable for both sides. The orthodox case that a state must budget like a household is wrong, and the 2 percent target is a convention dressed as a science. But sovereignty is a permission, not a shield. It tells you what a state may do. It does not tell you what its institutions can survive, or what its economy can produce.

Ask both questions. Britain in 2022 failed the second one with every advantage the first could confer.

Related: Hyperinflation · Where Does Money Come From? · The ECB Hiked Into a Supply Shock · The 90% Tax Rate Myth