[EAI Special Commentary] U.S. Debt, Dollar Hegemony, and a Fracturing Financial Order : What It Means for Korea

Ⅰ. Is U.S. Federal Debt Sustainable?

The U.S.-China rivalry is commonly pictured as a contest in which one side wins and the other loses. The financial order, at least, tells a different story. The United States cannot sustain its debt system without inflows of foreign capital, and China cannot expand its own payment system without the infrastructure that the United States and Europe established. Neither can fully replace the other, and neither can fully disengage. This mutual entanglement is what defines the international financial order today.

Against that backdrop, the debate over whether U.S. debt is sustainable has intensified, and it begins with sheer scale. Gross federal debt crossed $38 trillion in October 2025, roughly 123 percent of GDP. Even by the narrower measure that markets watch most closely, debt held by the public, it reached 99.8 percent of GDP at the end of fiscal 2025, the highest level since the years immediately following the Second World War. A peacetime economy now carries a wartime debt ratio, and this is what ignited the debate.

Yet sustainability is determined not by the size of the debt but by the relationship between two rates: the real interest rate the government pays (r) and the economy's nominal growth rate (g). When g exceeds r, growth absorbs the debt of its own accord; when r exceeds g, the debt swells even if the government takes no action. For nearly sixty years the United States remained in the safe zone, with g above r. That cushion has grown thin. Growth now stands at roughly 5.6 percent and the interest rate at roughly 4.6 percent, a gap of barely a percentage point, and it is narrowing rapidly as newly issued Treasuries reprice at higher yields.

The result manifests as a surge in interest costs. Net interest reached about 3.1 percent of GDP in fiscal 2025, approaching or already exceeding the previous record of 3.2 percent set in 1991. The decisive difference lies in the direction. The 1991 peak arrived once and subsided, while the present trajectory only climbs. The Congressional Budget Office (CBO) projects net interest at 4.6 percent of GDP by 2036 and above 6 percent by the mid-2050s, at which point roughly 40 percent of all federal revenue would be devoted to interest alone. The implication is clear: a rising interest bill is a structural burden that no administration can reduce in the near term.

That this fiscal strain has not yet detonated is due to the inflow of foreign capital. The United States absorbs some $1.2 trillion of it a year, and its net international investment position stood near –$27.5 trillion at the end of 2025, rivaling the federal debt itself. The sole condition on which the U.S. financial system operates is that this capital continues to arrive. Should the flow halt or reverse, in what economists term a "sudden stop," a U.S.-born financial crisis would become unavoidable. This is precisely the mechanism that struck Korea in 1997.

The debate thus divides in two. One camp holds that Washington must act preemptively to reduce the debt and its interest costs; the other holds that reserve-currency status will sustain the system without any such intervention. Both, however, have misidentified the question. The issue is not whether the United States goes bankrupt. One of the largest problems the world economy will confront over the next decade is to whom a non-bankrupt America transfers the bill.

Ⅱ. Will the U.S. Government Go Bankrupt?

It will not, at least not over its debt in any meaningful sense. Concerns about the debt are usually expressed as fears of default, but that fear rests on the wrong question. The reason is straightforward: the United States is a reserve-currency issuer that can create its own money to service what it owes. A further route is available. The Federal Reserve can purchase more Treasuries, a drift into what economists call fiscal dominance, in which monetary policy is subordinated to fiscal need. That drift has already begun: in December 2025 the Fed ended the quantitative tightening it had conducted since June 2022.

To gauge where the Fed may be headed, one might look to Japan. The Bank of Japan holds more than half of all Japanese government bonds; the government's debt has effectively been transferred onto the central bank's books. The Fed holds roughly 10 percent of Treasuries today, a figure that rose above 20 percent during the pandemic before receding. Measured against the path Japan has traveled, considerable room remains. It is no exaggeration to say that the next decade of U.S. monetary policy will turn on how quickly, and how extensively, the Fed buys Treasuries. Bankruptcy, in short, is not a genuine prospect.

Yet the Fed's purchases do not make the fiscal crisis disappear. The crisis is not erased; it changes form and is transferred as a cost onto particular groups. The channels all converge on a single mechanism: the United States dissolving its debt burden through inflation, which is to say the gradual erosion of the real value of the dollar. This single process imposes a burden on three groups in three distinct ways.

First, American citizens bear it through prices. Financing debt by expanding the money supply raises prices and erodes real wages and purchasing power, and much of that burden falls on the middle class.

Second, foreign holders bear it through the quiet erosion of real value. Foreign governments and institutions hold some $9.5 trillion in U.S. assets. So long as inflation persists, the nominal value of those assets and the nominal coupon on the Treasuries remain unchanged, yet their real, inflation-adjusted worth is quietly diminished. The Bank of Korea, which holds roughly $360 billion in U.S. securities, some 88 to 90 percent of its foreign reserves, is fully exposed to this erosion. The loss occurs with no default and no entry on any ledger. This is the silent loss.

Third, countries begin to withdraw their confidence in the dollar system. Having observed the United States offload its debt burden in this manner, central banks diversify their reserves to reduce dependence on the dollar and turn toward alternative payment infrastructure. That transition cannot occur without disorder in the international financial order.

This threefold transfer aligns with the strategy of the second Trump administration: to weaken the dollar and strengthen export competitiveness, and to ease the debt and interest burden through inflation. The burden ultimately falls on foreign holders. There is no nominal default, yet the danger is being amplified as the financing base of the Treasury market shifts.

Ⅲ. Why the Market Has Grown More Dangerous?

If bankruptcy is not the genuine risk, the real danger lies in who purchases the debt. Over the past decade or so, the financing base of the Treasury market has changed quietly but fundamentally. The share of Treasuries held by foreign governments has fallen from about 50 percent in 2008 to roughly 30 percent today. The privilege the United States long enjoyed, official foreign demand formed without regard to the debt's fiscal risk, is drying up.

A closer examination of that decline reveals three countries moving for three different reasons. Japan, still the largest holder at about $1.2 trillion, would sell if it could but cannot: it already holds so much that a large-scale sale would first destroy the value of its own portfolio, a form of forced bandwagoning. The United Kingdom's holdings rose more than fivefold, from $160 billion in 2013 to $863 billion, but not by London's design; the figure is the statistical footprint of global hedge funds and pension funds domiciled there. China reduced its holdings over the same period from $1.32 trillion to $684 billion, an unmistakable and deliberate act of balancing. The same asset is held by three countries under three logics.

What matters is who filled the vacuum. The answer is hedge funds registered in the Cayman Islands. Between 2022 and 2024 they absorbed roughly 37 percent of newly issued Treasury notes and bonds, and they are now the largest foreign holder, ahead of China, Japan, and the United Kingdom. Their holdings reached about $1.85 trillion at the end of 2024, an increase of $1 trillion in two years.

These funds now hold roughly 8 to 10 percent of the Treasury market, a share on the scale of a national central bank but opposite in character. Their Treasuries are mostly acquired with borrowed money; the largest funds are leveraged up to eighteen times their own capital. The difficulty is that this structure collapses at the smallest shock. When markets turn anxious and Treasury prices decline even slightly, lenders value the same bonds less generously and demand additional cash. A fund operating on leverage, lacking that cash, has no choice but to sell the very Treasuries it pledged.

This is where the real danger begins. Because dozens of funds run nearly identical trades, when one sells Treasuries the price falls further, the collateral behind every other fund is marked down, and they too must sell. Selling begets selling. March 2020 was precisely this: as the COVID shock forced these trades to unwind simultaneously, basis traders unloaded roughly $100 billion in Treasuries and froze the U.S. Treasury market, the deepest in the world, for several days. The Fed was compelled to intervene and purchase Treasuries.

This leverage is now double what it was then. Not every shock, to be sure, triggers such an unwinding; when the turmoil over Trump's reciprocal tariffs struck in April 2025, the Fed's liquidity backstops held and no significant damage followed. The fact remains, however, that leverage accumulated to a record high constitutes a permanent fragility in the market. The market long regarded as the safest may instead become an amplifier that magnifies a crisis.

Even so, the dollar system will not collapse easily. The paradox is that there is no alternative destination. Capital departing the U.S. market has nowhere ready to go. The eurozone is constrained by structural stagnation and the limits of fiscal union; Japan's market is too small; and China does not fully open its capital account, so that capital which enters cannot readily exit.

In relative terms, the U.S. market remains the deepest, most transparent, and most liquid available, and that relative advantage produces the crisis paradox. The deeper the systemic stress, the more capital crowds into dollar assets as a haven, and the firmer dollar hegemony becomes. The evidence is that in the fifteen years since the U.S.-born crisis of 2008, that hegemony has grown stronger rather than weaker.

History likewise suggests a gradual clock. Even after sterling departed the gold standard the British Empire had built, in 1931, it took until the mid-1950s—more than two decades and a world war later—for the pound to fully relinquish its hegemonic role. To imagine the dollar being displaced on a five- or ten-year cycle misreads that record.

What truly warrants concern is not American bankruptcy but the manner in which the tools used to avert it corrode confidence in the financial system, so that U.S.-born financial shocks grow larger in amplitude and shorter in cycle. With no alternative financial order yet mature, this uneasy coexistence may endure for a long time. The question, then, is what sustains that coexistence.

Ⅳ. Hegemony Lies Not in the Currency but in the Infrastructure

The debt swells exponentially, foreign governments retreat, and the market grows fragile, yet the dollar system still does not fall. To locate the source of that durability, one must identify the true foundation of dollar power. Contrary to common assumption, it is not the currency. It is the three-layer payment infrastructure behind the currency.

That infrastructure may be understood as a three-story building. At the top is SWIFT, the messaging network that links some 11,500 financial institutions worldwide and conveys payment instructions. In the middle are CHIPS and Fedwire, the clearing systems through which funds actually move, handling roughly $6 trillion a day. At the bottom are accounts at the Federal Reserve, where every dollar transaction ultimately settles. The freezing of some $300 billion in Russian central-bank assets in 2022 was executed at precisely this bottom layer.

The three layers are complements that cannot substitute for one another, and circumventing only one accomplishes nothing. China's renminbi settlement system, CIPS, is the decisive case. It succeeded in bypassing the clearing layer, yet about 80 percent of its transactions are still messaged over SWIFT. This is why U.S. secondary sanctions retained their force even when Russia used CIPS: sanctions operate not at the stage where a transaction is processed but at the stage where it becomes visible, and an alternative that cannot replace the messaging layer cannot conceal that visibility.

Here a pattern emerges that appears, at first, contradictory. As America's debt troubles surfaced, central banks reduced their dollar assets; the dollar's share of reserves declined steadily from 71 percent in 1999 to 57 percent in 2025. Judged by that alone, the world appears to be moving away from the dollar. Yet the currency actually used to settle international payments tells the opposite story: the dollar's share of settlement rose from 33 percent in 2012 to about 50 percent in 2025. The dollars being stockpiled are diminishing, even as the dollars actually being exchanged are increasing.

The discrepancy arises because "stockpiling" and "using" are entirely different matters. How large a dollar reserve a central bank holds is a political choice. It diversifies in order to distance itself from the risk of U.S. sanctions and asset freezes, and it may do so tomorrow if it wishes. Which currency a bank or firm uses to settle a trade or move funds, however, is not a choice but a matter of infrastructure. With no system yet able to replace the dollar, day-to-day practice has no option but to use it. Thus, at the very moment the world seeks to withdraw from the dollar politically, it binds itself to the dollar more tightly in technical terms.

That paradox reveals where dollar power truly resides. Even as debt fears lead states to stockpile fewer dollars, working dependence on the dollar only increases so long as no alternative infrastructure exists. Hegemony lies not in the currency but in the infrastructure that processes it. A decline in dollar holdings does not shake the hegemony; only a weakening grip on the infrastructure that clears dollars can do so. The question, therefore, is whether alternative infrastructure has reached the point of genuinely threatening that grip.

Ⅴ. The Alternative Order: Time, Friction, and the Paradox Between

Alternative infrastructure has evolved over the past decade from partial workaround to full bypass. China's renminbi settlement system grew roughly forty-fourfold, from 4 trillion yuan in 2016 to about 175 trillion in 2024, its membership expanding from 19 direct participants at launch to more than 190 direct and 1,500 indirect across 124 countries. As noted, however, it bypasses only the clearing layer; 80 percent of its messaging still runs over SWIFT.

The effort to overcome that limit is under way in digital currency. A digital currency issued directly by a central bank is termed a central bank digital currency, and China's digital yuan is the leading example. By the end of 2025 it had processed some 3.4 billion transactions worth about 16.7 trillion yuan ($2.4 trillion), the largest such experiment in the world. Most of it, however, is domestic, and its reclassification as a deposit liability in January 2026 indicates that its standing remains unsettled.

The decisive case is mBridge, a settlement network shared by several central banks. Its five members are China, Hong Kong, Thailand, the United Arab Emirates, and Saudi Arabia, which joined formally in June 2024. They settle through their own messaging, multilateral netting, and participating central banks' ledgers, touching neither SWIFT nor dollar clearing. It is the first integrated alternative to bypass all three layers at once. Cumulative transactions reached about $55.5 billion across more than 4,000 payments by November 2025, roughly 2,500 times the $22 million of the 2022 pilot, with about 95 percent settled in digital yuan.

Two developments increased the network's weight. First, the Bank for International Settlements (BIS), which had coordinated the project as a neutral party, withdrew on October 31, 2024. The stated reason was that the project had reached self-sustaining maturity, but the timing fell a week after President Putin proposed a similar alternative network at the BRICS summit in Kazan. Governance thereby passed from a neutral multilateral body to the participating central banks, China and the UAE above all.

Second, the BIS turned its resources to the Western camp's counterpart, Project Agorá. Involving seven Western central banks and more than forty commercial banks, that parallel project entered testing in January 2026. It takes up the same problem of cross-border settlement, but under Western governance and, crucially, in a way that preserves the existing correspondent-banking system rather than replacing it. The international payment system, alike in form yet institutionally divergent, has begun to split into two.

The scale of that split, however, remains small. mBridge's cumulative $55.5 billion is negligible against the roughly $6 trillion cleared each day through the U.S. clearing systems, CHIPS and Fedwire. Moreover, while 137 countries and currency unions, about 98 percent of world GDP, are experimenting with digital currencies, only a handful of small economies have actually issued one for general use. The foundation is being laid; functional replacement remains distant.

In short, constructing an alternative order requires time and friction, and during that transition the paradox may hold: with no alternative at hand, dollar hegemony may grow stronger still. The first year of the second Trump administration has been precisely such a phase. Over the longer term, however, the need to hedge against a dollar decline does not disappear, and alternative infrastructure is gradually developing the capacity to accommodate it.

Ⅵ. Conclusion: Implications for Korea

This uneasy coexistence weighs on Korea with particular severity. As noted, the U.S. securities held by the Bank of Korea constitute the bulk of its foreign reserves and already expose it to the silent loss. To this is added the $350 billion investment package finalized in October 2025. The package deepens that exposure, for as funds shift from short-term Treasuries into high-risk, twenty-year projects, the capacity to defend against inflation diminishes.

The exposure does not end there. Korea is exposed on several fronts at once: to the dual pressure of the U.S. and Chinese markets in trade; to the absence of any hedge in payment infrastructure, having joined none of the China-led alternative networks; and to a negotiating structure in which security, currency, and trade are bundled together.

The issue is not only direction but speed. While Saudi Arabia, the UAE, India, Indonesia, and Malaysia maintain their security ties to the United States even as they distribute risk across alternative infrastructure, Korea is accelerating its integration into the U.S. bloc, not only in security but in economics and finance. Set against a world moving to distribute its risk, Korea's trajectory runs in nearly the opposite direction.

Yet one must not rush to a hasty conclusion. Dollar hegemony will not fade easily. As with the British pound, a transfer of primacy takes decades, and the deeper a crisis runs, the more capital floods into the dollar as a haven, firming its position. The alternative order is unfinished, and filling that void will itself require a long time. To presume collapse and withdraw precipitously would be reckless.

The fact that the dollar will not fade, however, is no ground for complacency. The United States will not go bankrupt, yet the very policies that prevent bankruptcy breed chronic instability. The more Treasuries the Fed absorbs, the more the cost is transferred, quietly, through inflation. This instability is not a passing episode but a structural constant, one likely to persist for a decade and more. The dollar stands firm, and yet the waves it generates do not subside; they are more likely to grow rougher. That paradox is the challenge Korea confronts.

Korea's task, then, is not to choose one side but to calibrate direction and speed with care. This is not the moment to withdraw precipitously, nor to deepen the alignment complacently on the assumption that no default will come. A short-term posture that reacts to every movement in the exchange rate and every shift in U.S.-China relations cannot achieve that balance. What is required is a national vision and strategy for traversing a decade and more in which instability is the constant, one that maintains a footing in the dollar order while keeping open a door to the alternatives, so as not to be captive to its risks, neither hasty nor complacent.■

References

Arslanalp, Serkan, Barry Eichengreen, and Chima Simpson-Bell. 2022. "The Stealth Erosion of Dollar Dominance and the Rise of Nontraditional Reserve Currencies." Journal of International Economics, 138.

Atlantic Council. 2026. "Central Bank Digital Currency Tracker." Accessed July 21, 2026.https://www.atlanticcouncil.org/cbdctracker/.

Dalio, Ray. 2025. How Countries Go Broke: The Big Cycle. New York: Avid Reader Press.

Kang, Myungkoo. 2026. "Testing the Alternative: mBridge, the Hengli Episode, and the Empirical Viability of a Bifurcated Financial Order." Working Paper, Baruch College, CUNY.

Miran, Stephen. 2024. "A User's Guide to Restructuring the Global Trading System." Hudson Bay Capital.

Rogoff, Kenneth. 2025. Our Dollar, Your Problem: An Insider's View of Seven Turbulent Decades of Global Finance, and the Road Ahead. New Haven: Yale University Press.

■ Myungkoo Kang_is a Professor of Political Science at City University of New York.

■ Edited by: Jaehyun Im_EAI Research Associate
    For Inquiries: 02 2277 1683 (ext. 209) | jhim@eai.or.kr