Start with the honest answer, because it is the one a Treasury official gives. A war bond is still government debt. Whether you call it a war bond, a defence bond or a gilt, the state borrows the money and repays it with interest, and on the public accounts the label changes nothing. So if the case for war bonds rested on the accounting, it would fall over. It does not. It rests on three things the accounting never captures: who lends, what the money is visibly tied to, and whether Britain can still tell itself a story its own savers will put money behind.
The idea is back because John Healey put it there. On 20 July 2026, in one of his first moves as Prime Minister, Andy Burnham made Healey his Chancellor, moving the former Defence Secretary who had resigned over defence funding straight into the Treasury. Healey has floated a new category of government bond, a defence or war bond, to help push military spending toward 3 per cent of national income by 2030. He inherited a nervous market: the ten-year gilt yield stood at 5.04 per cent on the day, the highest in two months, with oil-driven inflation keeping rates high.
A debt is a debt, and that part is true
Robert Peston put the standard objection cleanly, and every Treasury official will make the same one. A debt is a debt. Whether you call it a war bond or a gilt, the government still borrows the money and still repays it with interest. Wrapping borrowing in a flag, he argued, is a gimmick, and the idea floated alongside it, that such bonds might escape inheritance tax, is a second gimmick stacked on the first. He is right about the arithmetic, and it is worth conceding that fully before disagreeing. Money is fungible. A pound raised through a defence bond frees a pound of general revenue to spend on anything else, so the hypothecation is mostly presentational. On the public accounts, a purpose bond and an ordinary gilt are the same liability with the same interest cost.
Who lends is the real prize
This is the part of the case I would defend hardest. Around a quarter of UK government debt was held by overseas investors by 2023, above the advanced-economy average of 18 per cent. The Office for Budget Responsibility's own warning is that foreign holders treat gilts as interchangeable assets and can switch out quickly and in size when the mood turns, while domestic holders tend to sit on the debt for structural reasons. Japan carried debts far larger than ours for decades at unusually low rates in large part because its own citizens held them for the long term. A bond built for British savers changes the composition of the lender base, not just the wording on the certificate. That is a real macro gain, and it is the opposite of cosmetic.
Tie the money to something people can see
You cannot adopt an abstraction. In the war, communities adopted a named ship in Warship Weeks or an aircraft in Wings for Victory, and the tangibility did the work. The modern equivalent of "growth" is not a slogan, it is this rail line, that grid connection, these homes. A purpose bond forces a named pipeline and a measurable output, because a bond sold against a project invites the public to check whether the project happened. Clarity here is not decoration. It is accountability, and it runs in exactly the direction a serious fiscal conservative should want, toward spending judged on what it delivers rather than what it intends.
The 450 billion pounds sitting still
British savers held roughly £450 billion in cash ISAs by the end of the 2025-26 tax year, part of an ISA pool passing £1 trillion, and they are putting far more into cash each year than into anything productive. That is patient money parked out of a lack of confidence rather than a lack of capacity. The precedent for moving it is not nostalgia. When the Treasury issued its first green gilt in September 2021 it raised £10 billion and the book was more than nine times oversubscribed. Go back further and the pattern holds under far greater strain: Defence Bonds launched in November 1939 under the "Lend to Defend" campaign, and between 1939 and 1946 savings bank deposits rose from £509 million to £1,982 million while the number of accounts more than doubled. People will lend to a purpose they can see.
That is also the answer to the mood Steph McGovern reached for on the same podcast, the sense that people feel hopeless and want to be part of something. A bond you can point to is a modest way of being part of something, and it is a better one than most of what politics currently offers.
Who wants what
None of this settles the question, because the design decides everything, and the factions pulling on that design want different outcomes. The Treasury and the Debt Management Office will resist anything that fragments a deep, liquid gilt market, since fragmentation can raise the cost of borrowing rather than lower it. The defence hawks around Healey want the money and will take the label that unlocks it. A growth-and-infrastructure camp wants patient capital for projects the market underfunds. And there is a patriotic, saver-facing politics, the terrain Reform UK has been working, that sees a national bond as a way to bind people to the country's balance sheet. Each wants a different animal wearing the same name.
What Healey has not said
Which is why the silences in Healey's position matter more than the slogan. He has not said whether the bond would be genuinely hypothecated or merely branded. He has not said whether it would be a retail product sold through NS&I to households or a wholesale instrument sold to the same institutions that buy gilts already. And he has not said the hardest thing, which is that done honestly this may cost the Exchequer more, not less, through tax relief or an above-market coupon.
Three ways it can go
Relabel a wholesale gilt and call it a war bond, and Peston is right, it is a gimmick, and a thinner market may even carry an illiquidity premium that makes it dearer than the gilt it replaced. Build a genuine retail bond through NS&I, tied to a named pipeline and modestly tax-advantaged, and it shifts the lender base toward patient domestic capital, gives the public something legible to back, and draws in parked cash, at a real fiscal cost and with real delivery risk. Or promise a rail line and a hundred thousand homes, tie a bond to them, and deliver neither, the way the last housing targets were missed, and it deepens the very hopelessness it was sold to cure. A bond attached to a project that never opens is worse than no bond at all.
The real deficit is a story
The structural read is the one that should guide the decision. Britain's competitiveness gap is, in large part, an infrastructure gap, and infrastructure needs patient capital and a pipeline the public can watch land. The war bond is not a funding trick. It is a delivery-and-accountability device that only works if the projects are real. That is also the case for going local with it. Burnham's instinct to devolve gives a Greater Manchester or a Teesside bond, tied to something a resident drives past, more force than any national abstraction, because the lender can see the thing they paid for.
So explore it, and refuse it in the same breath if it turns out to be only a flag on the same old borrowing. Britain's deepest deficit is not fiscal. It is the absence of a project its own citizens believe in enough to lend to, and no bond can manufacture that. It can only carry it once it exists.
Questions readers ask
Do war bonds or defence bonds raise new money for the government? Not in the way the label implies. Money is fungible, so a pound raised through a war bond frees a pound of general revenue to spend on anything else, and on the public accounts a purpose bond and an ordinary gilt are the same liability with the same interest cost. The case for war bonds does not rest on the accounting. It rests on who lends, what the money is visibly tied to, and whether the country can tell a story its own savers will back.
What is the difference between a war bond and a normal gilt? Legally and fiscally, very little. Both are UK government debt repaid with interest. The difference is who buys and why. A gilt is sold mostly to institutions, including foreign investors who held about a quarter of UK government debt by 2023 and can sell quickly under stress. A retail war or infrastructure bond aimed at British savers shifts the lender base toward domestic holders who tend to hold for the long term, which is a real macroeconomic gain even though the accounting is identical.
Could an infrastructure bond help fund UK projects like rail and housing? Potentially, if it is genuinely hypothecated to a named pipeline rather than only branded. Britain holds roughly 450 billion pounds in cash ISAs earning little, and the 2021 green gilt was more than nine times oversubscribed, so the appetite exists. The risk is cost. Done honestly it may pay a tax break or an above-market coupon, and a bond tied to projects that never open would deepen public distrust rather than build it.
Why do supporters say war bonds would build public confidence? Because a bond you can point to gives savers a visible stake in a specific project, the way wartime savers adopted a named ship or aircraft. Supporters argue that clarity and shared purpose matter when people feel the country lacks direction. The counter-argument, made by Robert Peston, is that branding borrowing as patriotic is a gimmick that does not change the debt. Both can be true: the label is worth exploring for the confidence it builds, and worth refusing if it is only a flag on the same borrowing.
Sources: Trading Economics, UK 10-year gilt yield (5.04 per cent, 20 July 2026); Office for Budget Responsibility, foreign-held debt (about 25 per cent by 2023, above the 18 per cent advanced-economy average); Lloyds Banking Group, ISA savings to pass 1 trillion pounds (2026); UK inaugural green gilt, 10 billion pounds, over nine times oversubscribed (September 2021); NS&I heritage, wartime Defence Bonds and National Savings; Bloomberg, Burnham names Healey Chancellor (20 July 2026); and The Rest Is Money, "Why Healey as Chancellor? And can he find the money?". All figures are official estimates and subject to revision.