Foreign-currency debt is where emerging-market government crises most often break, because a government cannot print another country’s money. But borrowing in your own currency is not a guarantee of safety: Russia defaulted on its own-currency debt in 1998. And euro-area countries behave partly like borrowers in a foreign currency, because they cannot print euros themselves.
Why it matters: The currency a debt is owed in tells you a lot about where it can break.
Summary: Foreign-currency debt is where emerging-market sovereign crises most often break, because a government cannot create another country’s money. But own-currency defaults do happen, Russia 1998 being the best-known case, and euro-area debt behaves partly like foreign-currency debt.
- Reinhart and Rogoff counted 68 overt domestic defaults since 1800; a Bank of Canada and Bank of England study found 32 local-currency defaulters in 1960 to 2019.
- Euro-area governments borrow in a currency they do not control; the ECB’s 2012 OMT backstop calmed that risk.
- Restructurings run through collective action clauses, the Paris Club, the G20 Common Framework and local-law exchanges.
- Face-value cuts understate losses: a 37% face cut can be a 53% present-value loss at a 9% exit yield.
- Zambia took about 3.5 years from default to its bond deal; Ghana’s Eurobond exchange cut face value by 37%.
Volume I established, through Sri Lanka’s 2022 default, and Capital Markets Part 10 extended through the 1997 Asian Financial Crisis, that foreign-currency debt is where emerging-market sovereign crises most often break. The mechanical reason belongs here: a government that borrows in its own currency can, in the most extreme circumstance, always technically create the money needed to service that debt (however inflationary a genuine last resort), while a government borrowing in a foreign currency has no such option — it must earn or acquire that specific foreign currency through exports, foreign investment, or foreign reserves, and simply cannot create more of another country’s money by decree. This is the single structural reason emerging market sovereign crises concentrate so heavily around foreign-currency-denominated debt specifically, rather than local-currency obligations.
That is a tendency, not a law. Governments do default on debt in their own currency, because printing money to pay is a choice with costs, and sometimes the costs of inflation, a currency collapse or a banking crash look worse than a default. Carmen Reinhart and Kenneth Rogoff counted 68 overt defaults on domestic debt since 1800, against about 250 on external debt, and called domestic defaults “often hidden” (NBER Working Paper 13946, 2008). A Bank of Canada and Bank of England study found 32 sovereigns that defaulted on local-currency debt between 1960 and 2019 and concluded that such defaults “are more common than is often supposed.” The best-known case is Russia in August 1998: on August 17 the government abandoned its defense of the ruble, defaulted on its ruble-denominated Treasury bills (GKOs) and bonds and forced their restructuring, and declared a 90-day moratorium on commercial external debt payments. It could have printed rubles; it chose not to.
The euro area is the other exception, in the opposite direction. Greek or Italian government bonds are denominated in euros, legally their own currency, but no national government controls the European Central Bank. Paul De Grauwe put it plainly in 2011: “Members of a monetary union issue debt in a currency over which they have no control.” Euro debt therefore behaves partly like foreign-currency debt, which is why, by De Grauwe’s figures, Spain paid about 2 percentage points more than the United Kingdom in early 2011 despite lower debt. The pressure eased only when the ECB stepped in as a conditional lender of last resort: Mario Draghi said on July 26, 2012, that “the ECB is ready to do whatever it takes to preserve the euro,” and on September 6 the ECB announced Outright Monetary Transactions, purchases of one- to three-year government bonds with “no ex ante quantitative limits,” available only to countries under a European rescue-fund program.
The rule “own-currency debt is safer” holds most of the time. These are the situations in which it bends.
| Situation | What changes | Why |
|---|---|---|
| Euro-area member | Euro debt behaves partly like foreign-currency debt | The national government cannot direct the ECB to create euros; only a conditional backstop such as OMT closes the gap |
| Very short local-currency debt | A rollover crisis is possible even in the home currency | Russia’s 1998 GKOs had to be refinanced constantly; when buyers vanished, the government chose default over unlimited printing |
| Local debt held mainly by domestic banks | Default or a haircut becomes a banking crisis | The doom loop of Section 3.5: Who Holds Sovereign Debt, and Why It Matters; Sri Lanka excluded banks from its 2023 domestic exchange |
| Local debt held heavily by foreigners | Currency and bond sell-offs feed each other | Foreign holders sell both the bond and the currency, so yields and the exchange rate move against the government at once |
| Dollarized economy | All debt is effectively foreign-currency debt | A country that uses the US dollar as its currency has no central bank able to create it |
| Statutory debt limit | Default risk becomes political, not economic | A reserve-currency government can be legally barred from borrowing to pay bills it can afford; Fitch cited “repeated debt-limit political standoffs” in its 2023 US downgrade |
When no combination of primary surplus, growth and refinancing can stabilize the debt (Section 3.7: Debt Sustainability Analysis — r versus g), the government has to restructure: change the terms of its debt by cutting principal, lowering coupons or extending maturities, ideally before it misses a payment rather than after. The machinery below is what makes that possible without a court.
There is no bankruptcy court for countries. A restructuring is a negotiation with each creditor group, and the machinery exists mainly to stop a minority from blocking the deal.
- Bondholders and collective action clauses (CACs). A CAC lets a qualified majority of bondholders change the payment terms for all holders of a bond. Older clauses voted bond by bond, so a fund could buy a blocking stake in one issue. The model clauses adopted from 2014 add single-limb aggregated voting, in which a 75% majority across all series combined binds every series, and rewrite the pari passu clause to rule out the New York courts’ equal-payment reading from the Argentina litigation.
- Holdouts. Creditors who refuse a deal and sue for full payment. After Argentina’s 2001 default, holdouts blocked payments on a restructuring approved by 93% of bondholders until 2016, when Argentina paid holdouts more than $6 billion, funded by a $16.5 billion bond issue. Greece’s 2012 holdouts were paid in full, about €6.5 billion according to the IMF.
- Official bilateral creditors: the Paris Club. An informal group of creditor governments dating from 1956, when Argentina first met its public creditors in Paris. It decides by consensus, requires an IMF program, and applies comparability of treatment: the debtor may not give other creditors terms more favorable to them than the Club’s.
- The G20 Common Framework. Endorsed on November 13, 2020, to bring non-Paris Club lenders, above all China, to the same table for low-income countries. It has been slow: Zambia applied in February 2021 and closed its bond deal in June 2024.
- Domestic debt. Restructured under local law, sometimes with banks carved out to avoid the doom loop (Section 3.5: Who Holds Sovereign Debt, and Why It Matters).
A face-value cut understates the loss. Suppose each $1,000 of an old bond is exchanged for $630 of a new 10-year bond paying 5% a year, a 37% face cut. Discounted at a 9% exit yield, the new bond is worth $468.28, so the loss in present value is 1 − 0.468 = 53.2%; at a 12% exit yield it is worth $380.83, a 61.9% loss. That gap is why Greece’s 2012 deal, with a face-value decline of about 52%, is estimated to have cost creditors 59–65% in present value.
| Country | Default | What was restructured | Outcome |
|---|---|---|---|
| Greece | 2012 exchange | Private-sector bonds; collective action clauses written retroactively into €177.3 billion of Greek-law bonds by the Greek Bondholder Act | €199.2 billion exchanged, 96.9% of eligible principal; face value cut about €107 billion (52%); present-value loss 59–65% |
| Zambia | November 2020, a missed $42.5 million bond payment | $6.3 billion of official bilateral loans (agreed June 2023, nearly 40% present-value reduction) and $3 billion of Eurobonds (June 2024), under the Common Framework | About 3.5 years from default to the bond deal |
| Sri Lanka | April 12, 2022, suspension of external debt service | Domestic debt (2023, banks excluded); about $12.5 billion of international bonds, including macro-linked bonds whose payments are tied to economic performance | Bond exchange with participation approaching 98%, settled December 2024; IMF program of SDR 2.286 billion (about $3 billion) approved March 2023 |
| Ghana | December 19, 2022, suspension of most external debt payments | Domestic debt exchange (about 85% participation, February 2023) and $13 billion of Eurobonds (October 2024) | Eurobond face value cut 37%, about $5 billion; over 98% consent; $4.3 billion of debt-service savings during the IMF program |
Two patterns stand out. Each began with an IMF program, because official creditors require one and its debt sustainability analysis sets how much relief is needed. And each took years: Sri Lanka’s IMF program projected central-government gross financing needs falling from 34.5% of GDP in 2022 to 15.9% by 2026, more than twice the 15% emerging-market line in Section 3.3: The Debt-to-GDP Ratio — What It Actually Measures (and Doesn’t) at the start.
Classify every sovereign bond by who controls the currency it pays in, not by its label. If the issuer’s own central bank creates the currency, the main risks are inflation and depreciation; default is possible but rare, and most likely when the debt is very short or held by fragile domestic banks. If the issuer cannot create the currency (foreign-currency debt, euro-area members without an ECB backstop, dollarized economies), analyze it as foreign-currency debt: reserves against foreign-currency debt falling due within a year, gross financing needs and the external balance decide the outcome. If foreign-currency debt is already in distress, value it by the present value of a likely exchange bond, not by the face-value cut. Ignore the label “local currency” as a safety signal for a foreign investor, whose return also depends on the exchange rate.
Treating a high-yield local-currency government bond as safe because “the government can always print.” For a dollar-based investor, printing is exactly the risk. A bond yielding 10% while the currency falls 30% returns 1.10 × 0.70 − 1 = −23% in dollars, with no default at all. Against a 4.5% dollar alternative, the break-even is a currency fall of only 1 − 1.045 ÷ 1.10 = 5.0% in a year. Avoid it by comparing the yield gap with plausible depreciation, using the country’s inflation gap, reserves and current-account balance, and by hedging or sizing the currency exposure separately from the credit exposure.
Can a country default on debt in its own currency?
Yes. Russia defaulted on its ruble Treasury bills and bonds in August 1998, and a Bank of Canada and Bank of England study counted 32 sovereigns that defaulted on local-currency debt between 1960 and 2019. A government that controls its currency can avoid default by printing, but it may judge the resulting inflation or banking damage worse than a restructuring.
What happens when a country defaults on its debt?
It usually loses market access, agrees to an IMF program and negotiates new terms separately with bondholders, official bilateral creditors and domestic holders. Bondholders receive new bonds worth less than the old ones; Ghana’s 2024 Eurobond deal cut face value by 37%. The process typically takes years: Zambia needed about three and a half.
What is a collective action clause?
A bond term that lets a qualified majority of bondholders change the payment terms for everyone, so a small minority cannot block a restructuring. Under the model clauses adopted from 2014, a 75% majority across all bond series voting together binds every series.
The fiscal rule. The Fiscal Responsibility and Budget Management (FRBM) Act, 2003 sets India’s fiscal anchors. The FRBM Review Committee chaired by N. K. Singh (report January 2017) recommended a general government debt ceiling of 60% of GDP, 40% for the central government and 20% for the states. The Union Budget 2026-27, presented on February 1, 2026, estimated the central government’s fiscal deficit at 4.4% of GDP for 2025-26 and 4.3% for 2026-27, and anchors the path on the debt ratio: central government debt of 55.6% of GDP in 2026-27, down from 56.1%, on a path to 50 ± 1% by 2030-31. That is Section 3.7: Debt Sustainability Analysis — r versus g‘s arithmetic in practice: with nominal growth above the interest rate, a falling deficit brings the ratio down. Gross market borrowing for 2026-27 is budgeted at ₹17.2 lakh crore and net borrowing at ₹11.7 lakh crore; the ₹5.5 lakh crore gap between them is mostly maturing debt being refinanced, the gross-financing-needs idea of Section 3.3: The Debt-to-GDP Ratio — What It Actually Measures (and Doesn’t).
The tax base. The Goods and Services Tax, introduced on July 1, 2017, replaced a web of central and state indirect taxes with one tax shared between the central government and the states and set by the GST Council. Its 56th meeting (September 3, 2025) collapsed the four main slabs into a standard rate of 18% and a merit rate of 5%, with a 40% rate for a few demerit goods, effective September 22, 2025. Because GST revenue is shared, a rate cut is a fiscal decision for every state at once.
State Development Loans and the RBI. India’s states borrow in their own right by issuing State Development Loans (SDLs), and under Article 293(3) of the Constitution a state still owing money to the central government needs its consent to borrow. Unlike the US, where the Treasury is the debt manager (Section 3.2: The Sovereign Debt Issuance Process), India’s debt management office sits inside the central bank: the Reserve Bank of India manages the central government’s debt under the RBI Act and, by agreement under Section 21A, the debt of all 28 states and of the union territories of Jammu and Kashmir and Puducherry. It auctions government securities (G-secs) and SDLs on its E-Kuber platform, yield-based for new issues and price-based for reopenings, with settlement on T+1 and primary dealers underwriting. Running monetary policy and debt management in one institution is the setup in which the fiscal-dominance warning signs of Section 3.1 deserve the closest watch.
Index inclusion and credit rating. On September 21, 2023, J.P. Morgan announced it would add Indian government bonds to its GBI-EM Global Diversified index. Inclusion ran from June 28, 2024, to March 31, 2025, at about 1 percentage point of index weight a month, up to a 10% cap, covering 23 Fully Accessible Route (FAR) bonds with a notional value of about $330 billion; Goldman Sachs estimated about $30 billion of passive inflows as a one-off adjustment. That widens the creditor base, and per Section 3.5 it also adds the most mobile holder type. S&P raised India’s long-term rating from BBB− to BBB on August 14, 2025, citing growth and the government’s commitment to fiscal consolidation, which moved India one notch above the investment-grade line of Section 3.6: Sovereign Credit Ratings.
Because a government cannot create another country’s money, emerging-market sovereign crises most often break on foreign-currency debt, but own-currency defaults such as Russia’s in 1998 do occur, and euro-area debt behaves partly like foreign-currency debt. When debt cannot be stabilized it is restructured through collective action clauses, the Paris Club and the G20 Common Framework, usually alongside an IMF program, as Zambia, Sri Lanka and Ghana did between 2023 and 2024.
Six questions on this chapter. Decide on your answer first, then click “Reveal Answer.”
1. In August 1998, what did Russia default on and force into restructuring?
- Its loans from the International Monetary Fund
- Its dollar Eurobonds held by foreign investors
- Its euro-denominated loans from European banks
- Its ruble-denominated Treasury bills and bonds
Reveal Answer
Answer: D. On August 17, 1998, Russia defaulted on domestic ruble debt (GKOs and OFZs), an own-currency default.
2. Each $1,000 of an old bond is exchanged for $700 of a new 10-year bond paying 6% a year. At a 10% exit yield the new bond is worth $527.95. What is the present-value loss?
- About 47%
- About 37%
- About 30%
- About 53%
Reveal Answer
Answer: A. PV loss = 1 − $527.95 ÷ $1,000 ≈ 47%, well above the 30% face-value cut.
3. Under the 2014 single-limb aggregated collective action clause, what binds all bondholders to a restructuring?
- A unanimous vote of all holders of every bond series
- A 75% majority across all bond series voting together
- A 50% majority within each bond series voting separately
- A 75% majority within each bond series voting separately
Reveal Answer
Answer: B. Aggregated voting stops a holdout from blocking the deal by buying a blocking stake in a single series.
4. A dollar-based investor buys a local-currency government bond yielding 12%; over the year the currency falls 20% against the dollar. What is the approximate dollar return?
- About +12.0%
- About −8.0%
- About −10.4%
- About −20.0%
Reveal Answer
Answer: C. Dollar return = 1.12 × 0.80 − 1 = −10.4%, with no default by the issuer.
5. Worked problem: Foreign-currency debt is 30% of GDP and the currency falls 30% against the dollar. What is the debt ratio afterward?
Reveal Answer
Answer: Ratio = 30% ÷ (1 − 0.30) = 42.86% of GDP.
6. Worked problem: Debt is 40% of GDP and the currency falls 25%. What is the ratio?
Reveal Answer
Answer: 40% ÷ 0.75 = 53.3% of GDP, with no change in borrowing.
- IMF Fiscal Monitor and WEO databases (DataMapper) — Gross and net debt for Japan, US, Ireland, Russia
- RBI, FAQs on Government Securities Market
- ECB, Draghi speech (July 26, 2012)
- ECB, Technical features of Outright Monetary Transactions (September 6, 2012)
- IMF Blog, Acting Collectively: A Better Way to Restructure Government Debt (November 2014)
- IMF, Press Release 23/79 on Sri Lanka (March 2023)