Global Public Debt Heads Toward 100% of GDP: Treasuries Are Still a Safe Haven, but Taiwan Can No Longer Hold Blindly

The IMF's "100%" comes with three necessary qualifications.

First, it is global public debt as a share of global GDP, not a simple average across all countries. Large economies and high-debt countries carry more weight, and individual countries' positions vary widely.

Second, it is a forecast for 2029. The IMF estimates that global public debt was close to 94% of GDP in 2025, and projects it forward to 100% based on assumptions about fiscal deficits, growth, interest rates and policy.^[1] War, the business cycle, inflation, tax revenue and policy changes could all revise that path.

Third, the debt ratio is not, on its own, a default trigger. Whether a government can sustain its debt also depends on interest costs, nominal growth, tax-raising capacity, debt maturity, currency denomination, investor composition, the monetary regime and policy credibility.

At the same 100% ratio, a country that finances itself long-term in its own currency with a stable investor base faces an entirely different risk from one that depends on short-term foreign-currency debt and has weak tax capacity.

The real function of the 100% figure is to remind the world that fiscal space is narrowing, and that when the next shock arrives, governments may not have as many low-cost options as they did in the last crisis.

The rise in global public debt is no longer merely a temporary result of a recession.

The IMF points out that spending on social needs, defense and strategic autonomy is rising, and so is the interest burden; global public-debt interest expenditure has risen from about 2% of GDP to close to 3% over four years.^[1] The high interest rates of the post-pandemic period are gradually working their way into governments' average financing costs: as long-term debt issued cheaply in the past matures, it has to be refinanced at higher rates.

This adjustment is slow, but it is cumulative.

A central bank cutting rates does not mean a government's interest bill falls immediately. As long as long-term yields remain elevated, or the market demands a higher term premium, new issuance and the refinancing of maturing debt will push up costs year after year. Rising interest expenditure squeezes other parts of the budget, and governments may respond by raising taxes, cutting spending, continuing to borrow, or letting inflation and financial repression absorb part of the adjustment.

Wars in the Middle East, energy, climate and demographic spending are all making it harder to simply "wait for the cycle to turn before dealing with it." The IMF's assessment is that much of this pressure has shifted from cyclical to structural; if governments merely wait for growth to bring the debt ratio down on its own, their options will only narrow.^[10]

Government bonds carry two kinds of "safety" that are often conflated.

The first is credit safety: whether the issuer pays interest and principal on time. The second is price safety: whether, when a holder needs to sell, the market price is close to the purchase cost.

A long-dated government bond with good credit can still fall sharply in price when market interest rates rise. The reason is that once newly issued bonds offer higher rates, older bonds must trade at a lower price for investors to earn a comparable return. Buy-and-hold investors who do not need to cash out early can keep collecting interest; but those who need to sell, post collateral, mark to market, or face redemptions feel the pressure immediately.

This is why sovereign debt does not need to default to trigger a financial tightening.

A decline in the value of banks' fixed-rate assets can constrain lending; funds facing redemptions may sell bonds for cash; and leveraged investors facing margin calls may unwind positions at the same time. Price movements pass through balance sheets and turn into greater difficulty for households and businesses in obtaining credit.

U.S. Treasuries remain one of the world's most important benchmark assets. Market size, trading depth, the dollar's international standing, the rule of law and financial infrastructure all make them difficult for any single alternative to quickly replace.

But being a core asset does not mean risk does not need to be managed.

Holding short-term Treasury bills, 10-year notes and 30-year bonds carries different price sensitivity; holding a position with one's own capital versus with short-term repo financing carries different liquidity risk; and an investor earning dollar income faces a different exchange-rate risk from one carrying New Taiwan dollar liabilities.

So "are Treasuries safe or not" is an oversimplified question. Four better questions are:

How is the issuer's credit? How long is the holding period? Is the funding source stable? Is the investor carrying a currency mismatch?

Change the answer to any one of these, and the meaning of "safety" changes with it.

Research published by the U.S. Federal Reserve in June 2026 estimated that large hedge funds' total U.S. Treasury exposure reached $4 trillion as of September 2025, with $2.4 trillion long and $1.6 trillion short; cash-futures basis trades stood at roughly $830 billion.^[3] The Fed also explicitly cautioned that these figures are estimated from reporting data, not an exact tally of individual trades.

The logic of a basis trade is to buy relatively cheap cash Treasuries and sell relatively expensive futures, waiting for the spread to converge. Each individual spread is small, so investors typically use high leverage and short-term financing to amplify the return.

In normal times, these trades can supply market liquidity. Under stress, the problem lies in funding maturity.

If price volatility raises margin requirements, or short-term funding providers tighten their terms, funds may be forced to sell cash bonds, cover futures positions, or cut other holdings. When many institutions deleverage at the same time, a spread that was originally very small can widen quickly, and market liquidity can thin out just when it is needed most.

The BIS treats this entanglement of high public debt with non-bank leverage as a new challenge facing central banks: a central bank may need to intervene to restore market function, yet not want that intervention to be understood as backstopping fiscal risk or leveraged trades.^[2]

The U.S. Treasury estimated in May 2026 that it would need to borrow a net $671 billion in privately held marketable debt in the July-to-September quarter, assuming an end-of-quarter cash balance of $950 billion.^[5] Quarterly borrowing estimates move with revenue, spending and cash management, and cannot be read directly as an annual deficit; but they still show that the market needs to keep absorbing a large volume of new supply.

When government bonds offer a higher yield, every other asset has to answer anew the question of "why is it worth taking on more risk."

Corporate-bond spreads may need to widen, the discount rate used in equity valuations may rise, and mortgage and corporate-loan rates may find it harder to return to ultra-low levels. Even if central banks cut their policy rates, the supply of long-term government bonds, inflation expectations and the term premium can still keep long-end rates elevated.

This is the aggregate force of the sovereign-bond market: it is not one corner of the financial markets, but the pricing floor beneath nearly every long-term asset.

Taiwan's own direct fiscal position differs from that of the world's high-debt countries.

Ministry of Finance data show that the central government's debt ratio has fallen in recent years; citing Standard & Poor's, the Ministry of Finance states that all levels of Taiwan's government combined are estimated to owe about 22.4% of GDP by the end of 2026.^[8] Figures for the central government, all levels of government, and public debt use different accounting bases and should not be mixed together, but overall Taiwan is not a major source of this round of rising global public debt.

Taiwan's exposure lies mainly in financial and private-sector balance sheets.

Taiwan's foreign assets stood at $3.267 trillion at the end of 2025; in the first quarter of 2026, the banking sector added to its holdings of foreign bonds, one reason residents' outward securities investment increased.^[6][7] The central bank's Financial Stability Report also notes that life insurers' profitability and capital-adequacy ratios have declined, and that they face higher market risk.^[9]

So global sovereign-debt risk does not need to pass through a government default before it reaches Taiwan. It can arrive via the valuation of overseas bonds, the dollar exchange rate, hedging costs, bank capital, insurance liabilities and corporate financing costs.

Taiwan's low government debt is a buffer; it is not a firewall against the financial markets.

The first is duration risk. The longer an asset's maturity, the more sensitive its price is when interest rates rise. Life insurers' liabilities are also long-dated, but the cash flows of assets and liabilities do not necessarily match exactly.

The second is exchange-rate risk. Holding assets in dollars while carrying liabilities in New Taiwan dollars requires hedging, and hedging costs move with the interest-rate differential between the two currencies, the foreign-exchange market, and the supply and demand for hedging instruments.

The third is credit and spread risk. Not all overseas bonds are sovereign debt; corporate bonds, financial bonds and structured products can all see their spreads widen at the same time when the economy weakens.

The fourth is liquidity risk. Assets that could in principle be held to maturity may still need to be sold at an unfavorable price if policyholders surrender their policies, collateral is called, or regulatory capital comes under pressure.

Viewed separately, all four risks may sit within their limits; combined, they can point simultaneously toward the same need for cash. This is why a stress test cannot change only one variable at a time.

Most Taiwanese SMEs do not directly hold large amounts of U.S. Treasuries, yet the effects still reach them.

If banks turn more conservative because of bond valuations, capital or liquidity, lending terms may tighten; changes in dollar interest rates and exchange rates affect imports, exports and foreign-currency loans; and once large companies face higher bond-issuance costs, they may push their funding pressure onto payment terms and supplier prices.

The companies most vulnerable are not necessarily those with the most debt, but those with the most concentrated refinancing. A company with modest total debt but a large loan maturing all in the same quarter, and no backup credit line, can be forced to accept a high interest rate right when the market is at its worst.

Another risk is floating-rate exposure. Companies tend to assume that once policy rates fall, their interest burden will ease quickly; but actual loan pricing also includes banks' funding costs, credit spreads and maturity. If global long-term bond yields stay elevated, the improvement in financing conditions may come more slowly than expected.

The first is a refinancing calendar. List every loan maturing over the next 24 months, its renewal terms, collateral and decision lead time. Do not look only at the annual total.

The second is an interest-rate/exchange-rate matrix. Test at minimum the combination of rising interest rates, currency appreciation or depreciation, and declining revenue; also calculate, currency by currency, whether foreign-currency income can naturally hedge foreign-currency liabilities.

The third is a liquidity waterfall. List, in order, cash, available credit lines, assets that can be quickly liquidated, deferrable investment, and shareholder funding; note who must approve each layer and how quickly it can be accessed.

The purpose of these three tables is not to predict whether the 10-year Treasury yield will rise or fall tomorrow, but to confirm that the company still has options if the market temporarily closes, rates stay high for a while, or banks delay extending credit.

When sovereign-debt markets turn volatile, central banks face a dilemma.

If they do not intervene at all, market dysfunction can be amplified through collateral, banks and the supply of credit; if they intervene every time prices fall, they may encourage governments and investors to take on more risk, creating an expectation that "the central bank will bail us out anyway."^[2]

Policy objectives therefore need to be kept separate.

Prices that simply reflect new fiscal and inflation information do not necessarily need to be stopped; it is when the market cannot trade, bid-ask spreads become disorderly, or the financing chain breaks that market function is genuinely at stake. The tools used to address this should also be distinguished from the stance of monetary policy, and paired with supervision of leverage, liquidity and non-bank institutions.

For Taiwan, the key is for different agencies to look at the full balance sheet together. The central bank watches markets and the exchange rate, the Financial Supervisory Commission watches financial institutions, and the Ministry of Finance watches government financing; if all three judge stability only within their own remit, they may miss how the same wave of interest-rate and exchange-rate movement transmits across balance sheets.

Global public debt heading toward 100% of GDP is not a countdown timer. It is a backdrop of pressure: governments need to borrow more, interest costs are rising gradually, the investor base is becoming more complex, and non-bank leverage makes market volatility easier to accelerate.

U.S. Treasuries remain a core global asset. What genuinely needs to be abandoned is not Treasuries themselves, but the illusion that "as long as the credit rating is high, price, maturity, currency and liquidity don't need managing."

Taiwan has lower government debt and a huge net external asset position, which gives it a better starting point than many economies. But these advantages only become a real buffer when the balance sheet can withstand the combined pressure of interest rates, exchange rates and liquidity.

A safe haven has never meant a place with no waves.

It should mean this: when the waves come in, the ship's maturity profile, leverage, cash and mooring lines are enough to keep it inside the harbor.