THE 5% WORLD

For almost fifteen years, governments in advanced economies learned a very dangerous lesson.

Debt was cheap.

Sometimes extraordinarily cheap.

Central banks bought sovereign bonds by the trillions. Interest rates hovered around zero. In parts of Europe and Japan, governments could actually borrow at negative yields. Fiscal expansion became easier to justify because the apparent cost of carrying additional debt seemed almost trivial.

That world is disappearing.

Not because investors have suddenly stopped buying government bonds.

Quite the opposite.

They are still buying them.

They simply want to be paid much more for doing so.

And that distinction may define the next phase of the global debt problem.

The United States is approaching a 5% 10-year Treasury yield. Britain is already above 5% across important parts of its curve. Japan’s 10-year government bond has touched 3% for the first time in three decades. Meanwhile, governments and corporations are expected to raise an extraordinary $29 trillion from bond markets in 2026.

The bond market is not broken.

It may be working perfectly.

And that could be the problem.

1. Britain: almost 6% — and investors are still lining up

The United Kingdom offers perhaps the cleanest demonstration of the new regime.

On September 10, the British 30-year gilt yield reached 5.925%, its highest level since 1998.

The 20-year reached 5.890%.

The 10-year gilt climbed to 5.348%, a level last seen in 2007.

These are extraordinary borrowing costs for a developed sovereign that issues debt in its own currency.

But something even more interesting is happening.

There is no buyers’ strike.

At the latest £5 billion auction of May 2030 gilts, investors submitted approximately £16.2 billion of bids.

The bonds were sold.

The market functioned.

The government obtained its money.

But the average yield was 4.786%, the highest for that maturity range since October 2023.

Only two days earlier, Britain had sold long-dated debt at the highest yield recorded in a gilt auction or syndication since the Debt Management Office was established in 1998.

This leads to a distinction that is becoming central to understanding sovereign debt:

availability of capital is not the same thing as affordability of capital.

Britain can borrow.

The question is what Britain must sacrifice elsewhere in the budget to pay for that borrowing.

And this is where the mathematics becomes political.

2. The world’s $2 trillion interest bill

The UK is not an isolated case.

Across OECD economies, governments paid more than $2 trillion in interest in 2025, equivalent to approximately 3% of GDP.

Britain, France and the United States now spend more servicing public debt than they spend on defence.

Meanwhile, OECD governments are expected to borrow approximately $18 trillion in 2026.

Outstanding sovereign bond debt across OECD countries has reached approximately $61 trillion, up from $55 trillion in 2024.

But perhaps the most important number is 78%.

According to the OECD, approximately 78% of OECD government borrowing in 2026 will not finance new bridges, hospitals, defence systems or AI infrastructure.

It will refinance existing debt.

That changes the problem completely.

The world’s governments are increasingly refinancing debt created in the era of cheap money in the era of expensive money.

And the transition is far from complete.

One-third of OECD fixed-rate sovereign debt outstanding at the end of 2025 matures between 2026 and 2028.

More importantly, 10-year market yields have recently been roughly two percentage points above the yields on the bonds being replaced.

Imagine refinancing a mortgage from 2% to 4%.

Now imagine doing it with trillions.

That is roughly the mechanism governments are entering.

3. America: $40 trillion meets the bond market

The United States is where the experiment becomes truly global.

Federal debt has crossed $40 trillion.

At the same time, the 10-year Treasury yield reached approximately 4.85% on September 10, its highest level since 2023.

That matters because the Treasury yield is not simply America’s borrowing cost.

It is arguably the most important price in global finance.

Treasuries sit underneath mortgage rates, corporate credit, equity valuations, emerging-market capital flows, bank balance sheets and countless financial models.

When the supposedly risk-free rate moves toward 5%, virtually every other asset must compete with it.

But Washington now faces a fascinating contradiction.

The government needs enormous amounts of financing.

Yet the economic conditions that would naturally produce much lower yields are precisely the conditions it does not currently have: very low inflation, weak nominal growth and abundant excess savings.

Instead, the United States has large fiscal deficits, resilient economic activity, enormous investment requirements and renewed inflation pressure.

And so the market is asking for more.

4. The $6 billion experiment

The Treasury has now increased its long-duration bond buyback operations to as much as $6 billion per operation, triple the previously planned amount.

The objective is primarily to improve market liquidity and alleviate pressure in less-liquid older securities.

The reaction was revealing.

The 10-year yield went up, briefly reaching approximately 4.85%.

This does not mean Treasury buybacks are useless.

It means they solve a different problem.

A buyback can improve market functioning.

It can change the composition of securities available.

It can reduce specific liquidity distortions.

What it cannot do is eliminate the government’s underlying financing requirement.

Goldman Sachs makes essentially this distinction: changing the maturity composition of borrowing does not change how much the government ultimately needs to borrow.

Or put more simply:

Debt management is not fiscal policy.

You can rearrange the maturity profile.

You cannot financially engineer away the deficit.

5. And the market is not panicking

This is where I disagree with some of the more dramatic commentary currently circulating.

Calling this a “Treasury crisis” is premature.

The rise in yields has largely remained orderly.

Volatility has not demonstrated the sort of systemic dysfunction associated with a genuine market breakdown.

And investors continue to buy US government securities.

This distinction matters enormously.

If investors were refusing Treasury debt, Washington would have an immediate financing crisis.

Instead, something subtler is happening.

Investors are effectively saying:

We will finance you — at the market price.

That price is increasingly uncomfortable.

And perhaps that is more consequential over the long term than a temporary panic.

6. The great refinancing trap

For years, economists debating public debt often focused on the relationship between:

r = interest rate

and

g = economic growth.

If nominal economic growth exceeds the effective interest rate on government debt, high debt ratios can remain surprisingly manageable.

But this logic contains an important lag.

Governments do not refinance their entire debt stock every morning.

The average interest cost adjusts gradually as old securities mature and are replaced.

That protected governments when rates initially increased.

The debt stock still contained enormous quantities of bonds issued at 0%, 1% or 2%.

But the protection gradually expires.

The OECD estimates that around one fifth of fixed-rate sovereign bonds outstanding in 2025 were already issued at yields above 4%, the highest proportion since 2015.

Every refinancing cycle therefore replaces a little more of the cheap-money world with the new one.

This is why today’s 5% yield matters even if tomorrow’s budget does not immediately collapse.

The interest bill compounds slowly.

And then politically.

7. Inflation has stopped being the government’s friend

There is another paradox.

Moderate inflation can help highly indebted governments.

It raises nominal GDP and erodes the real value of fixed-rate debt.

Historically, inflation has therefore sometimes functioned as a silent form of debt reduction.

But that mechanism has limits.

The OECD estimates that in 2026 inflation should reduce the OECD aggregate debt-to-GDP ratio by about 2.4 percentage points.

Higher interest payments, however, are projected to add approximately 2.5 percentage points.

For the first time in this cycle, the interest-rate effect essentially overwhelms the inflation benefit.

That is a remarkable turning point.

The inflation tax is no longer necessarily winning the race against the interest bill.

8. Then oil returned

And now the problem has become even more complicated.

Oil has moved back above $100 amid renewed Middle East disruption.

That has revived inflation concerns across the developed world.

The ECB has consequently returned to tightening, with markets and economists expecting the policy rate to move from 2.25% to 2.50% amid renewed energy-driven inflation pressure.

Britain faces the same dilemma.

The Bank of England meets on September 17. Markets have dramatically increased expectations for another increase later in the year as energy costs threaten the inflation outlook.

And that creates perhaps the defining contradiction of the current macroeconomic environment:

Governments desperately want cheaper money precisely when inflation makes cheaper money harder for central banks to deliver.

9. This is where fiscal dominance enters the story

Fiscal dominance is often described too dramatically.

People imagine a finance minister telephoning a central-bank governor and ordering him to print money.

Modern fiscal dominance is unlikely to begin that way.

It can be much quieter.

Imagine a government with rapidly increasing debt-service costs.

Higher rates increase the interest bill.

The higher interest bill increases the deficit.

The larger deficit requires more borrowing.

More borrowing increases bond supply.

Higher supply can increase the term premium investors demand.

Higher yields then increase future debt-service costs.

The sequence becomes:

higher yields → higher interest expense → larger deficits → more issuance → higher yields.

This is the sovereign debt doom loop that investors increasingly discuss.

At some point the central bank faces an uncomfortable political environment.

Inflation says:

keep rates high.

Fiscal arithmetic says:

bring them down.

The central bank may remain legally independent.

But independence becomes increasingly expensive to exercise.

That is the version of fiscal dominance worth watching.

10. Quantitative tightening makes the problem harder

There is another structural change that receives too little attention.

For years, central banks themselves were enormous buyers of government bonds.

QE created a buyer that was essentially insensitive to price.

The ECB did not buy Italian bonds because it found an attractive risk-adjusted return.

The Federal Reserve did not conduct QE because Treasury yields offered a compelling investment opportunity.

These were monetary-policy operations.

Now balance sheets are shrinking.

The OECD notes that central banks remain major domestic holders of government debt, but quantitative tightening means markets increasingly depend upon price-sensitive investors, including households, foreign investors and hedge funds.

This is enormously important.

The question is not:

Who will buy the debt?

Someone almost always will.

The question is:

At what yield?

A central bank can be relatively insensitive to price.

A pension fund can be somewhat sensitive.

A hedge fund is extremely sensitive.

Changing the marginal buyer changes the price governments must pay.

11. Governments are responding by shortening maturity

Governments have already found one way to make the immediate numbers look less painful.

Borrow shorter.

Long-term yields are expensive.

Shorter maturities can initially be cheaper.

According to the OECD, the proportion of sovereign issuance with maturities longer than ten years fell in 2025 to its lowest level since 2009.

Treasury bills accounted for roughly 48% of total sovereign borrowing, close to a record, and are expected to remain around that level in 2026.

This lowers the immediate interest bill.

But it creates another problem.

Refinancing risk.

Shorter maturity means governments must return to markets more frequently.

In effect, sovereigns are exchanging:

interest-rate risk today

for

rollover risk tomorrow.

That is not necessarily irrational.

But it is not a free lunch either.

12. Japan has joined the 5% world’s logic — even at 3%

Japan is the most extraordinary case because the absolute numbers look different while the mechanism is the same.

Japan’s 10-year government bond yield reached 3% on September 1, the first time it had done so since September 1996.

Three percent hardly looks frightening next to British gilts approaching 6%.

But Japan has spent decades building its fiscal and financial structure around exceptionally low interest rates.

Its public debt burden is enormous.

Therefore a movement from near-zero yields toward 3% represents a profound regime change.

It also creates a global consequence.

Japanese investors historically searched abroad for yield because domestic government bonds offered almost nothing.

If JGBs now offer meaningful returns, Japanese insurers, pension funds and banks have less incentive to send capital abroad.

The global significance is not that Japan suddenly sells every Treasury it owns.

It does not need to.

The marginal flow merely needs to change.

One of the world’s largest pools of savings may gradually become less available to finance everybody else’s deficits.

13. Then comes AI

And governments are not competing only with one another.

They are increasingly competing with corporations.

The OECD estimates governments and companies will collectively borrow $29 trillion from bond markets in 2026, 17% more than in 2024 and roughly twice the amount of a decade ago.

And the AI revolution is transforming the corporate side of this equation.

The OECD estimates that nine major AI-related companies could issue around $1.2 trillion of corporate bonds between 2026 and 2030 to finance investment.

Think about the capital requirements converging simultaneously:

AI data centres.

Semiconductors.

Electricity generation.

Grid infrastructure.

Defence.

Energy transition.

Government deficits.

Debt refinancing.

All of them compete, directly or indirectly, for capital.

For fifteen years we discussed secular stagnation and excess savings.

The coming decade may instead be defined by something closer to a war for capital.

If so, expecting interest rates simply to return to their post-2008 averages may prove one of today’s biggest analytical mistakes.

14. The US-China paradox

And then there is China.

On September 10, the US 10-year Treasury yielded approximately 4.85%.

China’s equivalent government bond yielded only 1.68%.

The difference reached a record 3.17 percentage points.

At first glance, that seems absurd.

The United States has the world’s reserve currency, deeper capital markets and stronger nominal growth.

Yet Washington pays almost three times Beijing’s benchmark nominal borrowing rate.

The explanation is not that markets suddenly believe China is safer than America.

China faces almost the opposite macroeconomic problem.

Weak domestic demand.

Deflationary pressure.

Weak credit demand.

Abundant domestic savings.

The United States faces stronger nominal demand, persistent inflation pressure and enormous fiscal financing requirements.

Bond yields reflect those differences.

And this exposes one of the great ironies of today’s geopolitical competition:

The West is trying to out-invest China precisely while the cost of capital in much of the West is rising dramatically above China’s.

That deserves much more attention than it receives.

15. So is a sovereign debt crisis coming?

Not necessarily.

That conclusion is too easy.

Britain can finance itself.

America can finance itself.

Japan can finance itself.

European governments continue issuing enormous quantities of debt.

The OECD explicitly notes that sovereign bond markets have continued functioning effectively and have absorbed record issuance volumes.

That is why I would resist the fashionable language of an imminent global bond-market collapse.

The more interesting scenario is slower.

Governments continue borrowing.

Investors continue buying.

But the price gradually changes the political economy.

More tax revenue goes toward interest.

Less remains for infrastructure.

Or defence.

Or pensions.

Or healthcare.

Governments increase taxes.

Or reduce expenditure.

Or tolerate more inflation.

Or pressure central banks.

None of those choices requires a spectacular financial crisis.

16. The real sovereign risk may be political

This is ultimately where the debt debate becomes more interesting than debt-to-GDP ratios.

Governments can mathematically stabilize debt.

Raise taxes.

Reduce spending.

Reform pensions.

Increase retirement ages.

Cut subsidies.

Reduce welfare growth.

Increase productivity.

The problem is that voters may reject many of those solutions.

The Financial Times recently captured this political constraint well: investors increasingly question whether governments are willing — or electorally able — to make the difficult fiscal decisions required to stabilize debt.

That changes the risk calculation.

The fundamental sovereign risk of the next decade may not be inability to pay.

It may be inability to agree politically on who should pay.

Taxpayers?

Pensioners?

Public employees?

Bondholders through inflation?

Savers through financial repression?

Future generations through more debt?

There is no painless answer.

17. Financial repression may return

If governments cannot grow out of their debt and voters reject austerity, history offers another possibility.

Financial repression.

That can include policies encouraging or requiring domestic financial institutions to hold government bonds, regulations that create captive demand, negative real interest rates, capital controls or monetary policies that indirectly suppress sovereign borrowing costs.

The FT identifies financial repression as one potential route governments could use to escape increasingly difficult debt dynamics.

And this is where fiscal dominance becomes particularly relevant.

The political temptation eventually becomes obvious:

If 5% is fiscally painful, make sure the market cannot demand 5%.

But suppressing nominal yields while inflation remains elevated simply transfers the cost.

From governments.

To savers.

The debt has not disappeared.

Someone else is paying for it.

18. Welcome to the 5% world

The defining feature of the new regime is therefore not simply that interest rates are higher.

It is that high debt and high rates now coexist.

That combination was largely absent from the post-2008 world.

Before 2008, borrowing costs could be high, but public debt was generally much lower.

After 2008, debt exploded, but central banks suppressed borrowing costs.

Today we increasingly have both.

Record debt.

Record issuance.

Higher real yields.

Reduced central-bank buying.

Greater refinancing requirements.

Massive AI investment.

Defence spending.

Energy infrastructure.

And renewed inflation risk.

The OECD’s numbers summarize the transformation almost perfectly:

$61 trillion of outstanding OECD sovereign bond debt.

$18 trillion of expected OECD government borrowing in 2026.

78% of borrowing required merely to refinance existing debt.

And approximately $29 trillion of combined government and corporate borrowing expected this year.

This is not primarily a story about governments running out of money.

Sovereign issuers rarely “run out” of money in the conventional sense.

It is about the price of money.

Conclusion — The bond market isn’t broken. That’s the problem.

For years, policymakers behaved as though low interest rates were almost a permanent characteristic of advanced economies.

They were not.

They were a regime.

And regimes change.

Today investors are still willing to finance Britain near 6%.

They are still willing to finance America near 5%.

They are still buying Japanese government debt around 3%.

Government bond auctions continue to clear.

Liquidity remains available.

There is no generalized sovereign buyers’ strike.

And therefore the most interesting conclusion is almost the opposite of the popular one.

The bond market isn’t broken.

It is repricing the privilege of being financed.

Governments can choose how much they spend.

They can choose how much they tax.

They can choose how much debt they issue.

But unless central banks ultimately intervene, they cannot permanently choose the price at which private capital finances them.

And that brings us to the question I believe may become one of the defining macroeconomic questions of the next decade:

What happens when the interest rate required to control inflation becomes higher than the interest rate governments can politically afford?

We may discover that fiscal dominance does not begin with a central bank losing its legal independence.

It begins when maintaining that independence becomes increasingly expensive.

Welcome to the 5% world.

The money is still there.

It just isn’t cheap anymore.

Sources: OECD Global Debt Report 2026 and sovereign borrowing outlook; Financial Times analysis of the global sovereign interest bill and US-China yield divergence; Reuters reporting on UK gilt markets and current monetary-policy developments; US Treasury/bond-market reporting on the expanded buyback programme.