Japan discovers the price of money | Sovereign debt gets complicated | The Daily Brief #489

ELI5/TLDR

Japan spent thirty years in a world where money was free and prices never moved, so people just kept their savings in cash and it worked fine. Now inflation is back, interest rates are finally rising, and the whole economy has to relearn that money costs something — which breaks a lot of arrangements built on the assumption that it didn’t. The second half is about how lending to poor countries has quietly become a tangled mess of secret deals, hidden collateral, and rival creditors, so when one of them goes bust, nobody can even agree who’s owed what before they start fixing it.

The Full Story

Japan’s thirty-year sleep

Start with one strange number. At the end of March 2025, Japanese households held more than half their financial wealth in plain cash and bank deposits. For comparison, in the US cash is about 11.5% of household assets and over 40% sits in stocks. Even Indian households — in a much poorer economy — hold more equity than the Japanese.

Everywhere else, holding cash is slow self-harm. Inflation eats it. Money sitting still loses purchasing power, so you’re forced to put it into something productive just to stand still. Japan learned the opposite lesson, and learned it brutally.

If you invested in the Japanese stock markets in 1989, you wouldn’t see a single yen in returns until 25 years later in 2024.

In the late 1980s Japanese stocks and land became a national mania. Then the bubble burst, and the floor gave way. Asset prices collapsed, banks were buried in bad loans, and companies stopped investing. An economy that had genuinely scared the United States simply went limp — and stayed limp for three decades. Every time a little optimism flickered, something stamped it out: the 2008 financial crisis, the 2011 Tohoku earthquake.

In that world, cash was the rational choice. It paid nothing, but it lost nothing either, and nothing else could even promise that. Hold it long enough and the rational choice hardens into culture.

Why free money didn’t work

Here’s the counterintuitive part. Japan made money as cheap as money can get, and still nobody borrowed it.

By 1999 interest rates were zero. In the 2000s the central bank started quantitative easing — printing money to buy up assets. In 2016 it went to negative rates, literally charging people to leave money idle. It even started buying 10-year government bonds without limit to pin down long-term borrowing costs. Nothing took.

The economist Richard Koo has a name for what was happening: a balance sheet recession. After a bubble bursts and everyone is drowning in debt, companies stop behaving like profit-seekers. For a decade, every yen of Japanese corporate cash flow went to paying down debt and repairing balance sheets. Borrowing became taboo. Offer a Japanese firm free money to expand and it would politely decline, because its only goal was to get clean.

Debt, meanwhile, became taboo. Companies would not borrow to expand their operations, even if you offered them that money for free.

That’s the trick of it. Cheap money only helps if someone wants to borrow. If everyone is paying off debt at once, you can cut rates to zero and nothing moves.

And it fed itself. No investment meant no jobs — an “employment ice age.” People who kept jobs took pay cuts. Scared, everyone delayed spending. Companies cut prices to sell anything at all, and Japan slid into deflation: falling prices. Think of it like a snake eating its own tail. Low demand → no investment → flat wages → weak demand → low prices, round and round, until everyone simply expected prices and wages to stay frozen forever and planned their lives around it.

Where all that frozen money went

A frozen economy still has savers, and they had to put the money somewhere. Japanese banks couldn’t earn anything at home, so they lent for 30 or 40 years out, or went abroad. The one borrower with a big appetite was the Japanese government, which could borrow staggering amounts, for very long terms, at almost no interest.

By the end of fiscal 2024 that pile was 1,171 trillion yen of government bonds — about two and a half times the country’s entire GDP — at an average interest rate near 0.83%. After all the QE, the central bank itself held nearly half of it.

Outside Japan, the frozen yen created a famous money machine: the carry trade. Borrow yen at near-zero, sell it for dollars or rupees or pesos, park that in something with a real yield, and pocket the difference. The only real danger was the yen suddenly rising before you repaid your loan — which occasionally happened and sent a jolt through global markets. But mostly it was free money.

According to an August 2024 assessment, the scale of these carry trades was more than 40 trillion yen or over rupees 23 lakh cr.

By some estimates roughly a quarter of that flowed into India as foreign investor inflows — though these trades are deliberately opaque, so nobody knows precisely.

The thaw

Then the world changed. Japan imports almost all its food and fuel, and when Russia invaded Ukraine in 2022, both got expensive overnight. Painful for households — but it cracked the ice. With an ageing population and scarce workers, employees finally had leverage. In 2024 Japanese workers won an average 5.1% raise, the biggest in 33 years. Prices rose. And crucially, people began expecting prices to rise and acting accordingly — the psychology finally flipped.

So the Bank of Japan started dismantling its emergency machinery. Negative rates ended in March 2024. Rates went to 0.25% in July 2024, 0.5% in January 2025, and kept climbing until this year, when for the first time since 1995 they touched 1%.

Now sitting here in India, that number probably looks like nothing. But to a country that had learned to expect free money, it was a sea change.

The bill for normal

A return to normal money breaks things that were built for abnormal money. Banks could finally earn a decent lending margin — but the old low-yield bonds on their books are now worth less, because nobody wants a 0.5% bond when new ones pay 1.5%. That markdown has dented the capital of smaller banks. Insurers face the same squeeze and have gone shy on ultra-long bonds.

That matters because banks and insurers were exactly who the government leaned on to fund its enormous borrowing. Now it’s getting expensive. Interest costs on general bonds are climbing from about 8 trillion yen (FY2024) to 10 (FY2025) to 13 (FY2026). Across all instruments, debt servicing now eats 31.3 trillion yen — a quarter of the entire budget — and that will only grow as cheap old bonds roll off.

Meanwhile the Bank of Japan, holding half the bonds, plans to offload 350–370 trillion yen of them by March 2030. One twist: normally stock markets hate rate hikes, but Japan’s market read them as a signal of real growth ahead, and the Tokyo index roughly doubled in three years. And the carry trade has probably shrunk meaningfully — possibly a small piece of India’s recent foreign outflows.

The old world is dead. But who knows what the new one will look like.

The second story: sovereign debt gets tangled

The video’s second half shifts to how governments borrow — and what happens when they can’t pay.

It used to be simple. Few types of creditor, clear legal order. When a country went broke, everyone knew the queue: the IMF at the top, then rich-country governments (the “Paris Club”) and commercial banks. Creditors coordinated, did the maths on a transparent pile of debt, and worked it out. Painful, but legible.

That clarity is gone. Over the past decade, emerging-market debt has sprouted exotic new instruments, and that changes everything about a default. When Zambia defaulted in 2020 it took over four years to reach even a partial fix — not because anyone cheated, but because nobody could agree who was owed what, who ranked where, and who should eat the losses. Ethiopia has been stuck in restructuring for years.

The invisible pecking order

On paper, apart from the IMF, most government lenders rank equally. In practice a steep, mostly-hidden hierarchy has formed:

IMF at the top, protected by convention as lender of last resort.
Big multilateral banks (World Bank, African Development Bank), whose “preferred creditor status” means they rarely take losses as long as they keep lending.
Commercial lenders backed by a multilateral guarantee — if things go wrong, the guarantee pays them.
Collateralized lenders, secured against specific assets or future revenues.
Regional development banks, weaker protection.
Unsecured creditors at the very bottom, absorbing everyone else’s safety.

The problem isn’t just that this ladder exists — it’s that it’s mostly unwritten and getting more rigid. When a guaranteed commercial lender activates its guarantee mid-restructuring, its ordinary claim magically becomes a preferred multilateral claim, and everyone below sees their recovery shrink. Worse, creditors actively jostle to climb. China’s state policy banks — now the largest bilateral lenders to poor countries — often refuse to be classified as either “official” or “commercial” and demand secret loan terms. In Zambia, where China was the dominant lender, that very ambiguity drove years of delay.

Hidden collateral

Then there’s what’s pledged against the debt. Angola is the cleanest example: after its civil war ended in 2002, it rebuilt on Chinese loans repaid not in cash but in future oil shipments. By 2011 it had borrowed over $20 billion against its oil — most of it invisible to outsiders.

A landmark study… examined 52 resource-backed loans signed between 2004 and 2018 worth a combined $164 billion… Somehow only one of the 52 contracts was made public.

When World Bank researchers tried to match the African loans against their own records, they could find only half — some had been quietly booked as “advance payments to suppliers” rather than debt at all. And collateral is contagious: if pledging a revenue stream reliably bumps you up the queue, every new lender demands the same, and whoever lent first and unsecured gets shoved to the back. Collateralized debt is now 60% of external public debt in South Sudan, 54% in Guinea, 32% in Chad.

Angola is trying to escape by replacing Chinese oil-loans with Western private capital — but that’s pricier and creates fresh tangles. In 2025 it borrowed $1 billion from JP Morgan and posted $1.93 billion of new bonds as collateral. Pay up and the collateral returns; default and JP Morgan seizes it, leaping past every earlier creditor. Less Chinese, but no less complicated.

When the machinery jams

The G20 “Common Framework” of 2020 was meant to fix all this by pulling China and other newcomers into one orderly process. Four countries applied — Chad, Ethiopia, Ghana, Zambia — and the results have been grim. Zambia took over four years from default to a first bilateral deal. Ethiopia, five years and counting, with bondholders threatening to sue.

The reason: before anyone can negotiate haircuts, they have to establish who’s owed what and where they sit — and that basic fact is no longer knowable. In Zambia, China insisted bondholders were getting a sweeter deal than bilateral lenders; the bondholders disagreed; Zambia was stuck refereeing a fight between its own creditors over whose losses “counted” as comparable.

Even the clever fixes backfire. Value recovery instruments pay creditors more if the economy beats forecasts — sensible in theory, but fiendish to value. Greece’s 2012 restructuring (the largest in history) handed creditors GDP-linked warrants; when Greece recovered and tried to buy them back, creditors argued they were worth far more, and it ended in years of London litigation that Greece only recently won.

India is now a player here too — co-chairing Sri Lanka’s creditor committee with Japan and France, and signing the Zambia deal as a lender. As its lending grows, it’ll meet the same classification headaches.

The closing thought is a tidy historical arc: in the 1980s the hard part of a debt crisis was splitting losses among banks; in the 2000s it was neutralizing holdouts who refused the deal; in the 2020s the hard part comes before any of that — just figuring out which debts are official, private, guaranteed, collateralized, domestic, external, preferred, or contingent in the first place.

Key Takeaways

A balance sheet recession (Richard Koo) defeats cheap money. When everyone is paying down debt simultaneously, zero or negative rates do nothing, because the bottleneck is the absence of willing borrowers, not the price of credit. This is the core reason Japan’s stimulus failed for decades.
Japanese households hold over 50% of wealth in cash; the US holds ~11.5% and 40%+ in equities. Three decades of deflation made cash the only asset that reliably held value, and that rational choice fossilized into culture.
Japan’s 1989 stock peak wasn’t recovered until 2024 — a 25-year wait for zero return. This is the lived experience behind Japanese risk-aversion.
The yen carry trade ran to 40+ trillion yen (~₹23 lakh crore), with possibly a quarter flowing into India as foreign inflows. As Japanese rates rise (hitting 1% in 2025, first time since 1995), this trade unwinds — a plausible contributor to Indian foreign outflows.
Rising rates create losers among bond holders even as they help lending margins. Old low-coupon bonds get marked down when new bonds pay more, denting bank and insurer capital — the same institutions a government relies on to fund its borrowing.
Japan’s debt servicing is now a quarter of its budget (31.3 trillion yen) and rising, on a government bond stock 2.5x GDP.
Sovereign debt has a hidden, unwritten creditor hierarchy (IMF → multilaterals → guaranteed commercial → collateralized → regional banks → unsecured), and lenders actively maneuver to climb it mid-crisis.
Collateral is contagious. Once one lender secures a revenue stream to jump the queue, every new lender demands the same, pushing unsecured/early lenders to the back. Resource-backed loans are now the majority of external debt in several African states.
Opacity is the real enemy, not novelty. Of 52 resource-backed loans worth $164 billion (2004–2018), only one contract was public; some debt was disguised as “advance payments.” The useful distinction is transparent and contingent vs opaque and subordinating — not old vs new instruments.
The G20 Common Framework has been very slow (Zambia 4+ years, Ethiopia 5+) because you can’t negotiate haircuts until you know who’s owed what — and that’s now genuinely unknowable up front.

Claude’s Take

This is one of those Daily Brief episodes that quietly over-delivers. Both stories are doing the same trick: taking a thing everyone treats as plumbing — the price of money, the order in which debts get repaid — and showing that when the plumbing changes, the whole house shifts.

The Japan story is the stronger of the two, mostly because the balance-sheet-recession framing is genuinely clarifying. The puzzle “why did free money do nothing for thirty years” is a real puzzle, and “because nobody wanted to borrow it” is the kind of answer that reorganizes your intuitions. The episode is careful with its caveats too — it repeatedly flags that carry-trade numbers are opaque and estimates are soft, which is the right instinct.

The sovereign-debt half is solid and well-sourced (Lazard, World Bank, the Natural Resource Governance Institute study), if a touch listy. The strongest insight is the closing historical arc — that the hard part of a debt crisis has migrated from splitting losses, to defeating holdouts, to merely establishing the facts. That’s a real shift in the world, not just a complaint about complexity. The China angle is handled with appropriate restraint: it’s named as the dominant new force without tipping into editorializing.

Where it’s thin: it never quite quantifies how big a deal the carry-trade unwind is for India specifically, and the Japan story soft-pedals the scariest implication — a government with debt at 2.5x GDP facing genuinely rising interest costs is not obviously a happy ending, and “who knows what the new world looks like” is doing a lot of work to avoid saying so. But for a daily news segment, this is high-signal, honest about its uncertainty, and teaches durable mental models. An 8.

Japan discovers the price of money | Sovereign debt gets complicated | The Daily Brief #489 | Markets by Zerodha