What creditors lose when a government defaults

The median restructuring writes off nothing and still costs creditors 37.0%

165 of the 329 restructurings cut the face value of the debt by exactly zero, which is why the median face-value reduction is 0.0% while the median present-value haircut is 37.0%. The present-value haircut compares the discounted value of the new instruments with the discounted value of the old claim, so it prices maturity extension and coupon reduction, both of which a face-value measure records as nothing. It exceeds the face-value reduction in 267 of 329 episodes. 9 restructurings carry a negative present-value haircut, meaning the new instruments were worth more at the exit yield than the claim they replaced, and 4 carry a negative face-value reduction, meaning principal rose. Neither column is clamped at zero here.

Source: Meyer, Reinhart & Trebesch (2022), Sovereign Bonds since Waterloo | Cruces & Trebesch (2013), Sovereign Defaults: The Price of Haircuts | Asonuma & Trebesch (2016), Sovereign Debt Restructurings: Preemptive or Post-Default Both series cover the same 329 episodes; the bars are two measures of the same deals, not two samples. Methodology

The two-century record, 1823 to 2024

Every restructuring in the file, on the year it concluded, sized by the debt it treated. The record is not evenly filled: 1950-1979 holds 19 restructurings across 17 countries, while 1980-1999 holds 142 across 61. The era table below carries the coverage.

Source: Meyer, Reinhart & Trebesch (2022), Sovereign Bonds since Waterloo | Cruces & Trebesch (2013), Sovereign Defaults: The Price of Haircuts | Asonuma & Trebesch (2016), Sovereign Debt Restructurings: Preemptive or Post-Default Methodology

The eras are not one dataset. The 136 episodes sourced to Meyer, Reinhart and Trebesch run 1823 to 1993; the 174 sourced to Cruces and Trebesch run 1940 to 2013. Nothing before 1940 comes from the Cruces-Trebesch construction, and the exit yield that construction discounts at is carried only in the Cruces-Trebesch file, for 166 of its 187 episodes. The long-run file has no discount-rate column at all. The heading over the haircut column is the same from 1823 to 2024; the reconstruction behind it is not.

The biggest losses

Source: Meyer, Reinhart & Trebesch (2022), Sovereign Bonds since Waterloo | Cruces & Trebesch (2013), Sovereign Defaults: The Price of Haircuts | Asonuma & Trebesch (2016), Sovereign Debt Restructurings: Preemptive or Post-Default PV = present-value haircut, Face = face-value reduction. Both are properties of the same deal and are never combined. Methodology

Two haircuts for the same deal, and the discount rate decides

The Cruces-Trebesch file carries 187 restructurings from 1978 to 2013 with two haircuts for each one. The present-value haircut discounts both the old claim and the new instruments at the exit yield. The market haircut compares the new instruments with the face value of the old claim, leaving the old claim undiscounted. They differ on 92 of the 187 episodes, and where they differ the market haircut is the larger in 78 of 92 cases, by up to 22.2 percentage points. The medians are 32.5% and 40.3%. The exit yield doing the discounting is not a technicality: it ranges from 7.9% to 41.2% across the 166 episodes that have one, with a median of 14.8%. Averaging the two measures together would be meaningless, so this page never does.

Source: Cruces & Trebesch, haircut dataset (2014 update) 21 of the 187 episodes carry no exit yield. Methodology

Where the two measures diverge most

Source: Cruces & Trebesch, haircut dataset (2014 update) Methodology

And how long it takes

A default spell is the file's grouping of a country's restructurings into one episode of default. There are 206 of them across 93 countries. The median lasts 8 years, the mean 11.3, and the longest 64. 21 spells opened and closed inside a single year; 64 needed more than one restructuring, and the worst needed 7. In 2000-2024 the median spell ran 3 years against 11 in 1823-1899.

The endpoints are close to, but not exactly, the obvious reading. A spell's end year is the year of its last restructuring in 203 of the 206 spells, and its start year is the earliest default year of those restructurings in 205 of 206. Duration counts both endpoints in all 206.

Source: Meyer, Reinhart & Trebesch (2022), Sovereign Bonds since Waterloo | Cruces & Trebesch (2013), Sovereign Defaults: The Price of Haircuts | Asonuma & Trebesch (2016), Sovereign Debt Restructurings: Preemptive or Post-Default The spells table covers 93 of the 94 countries with a restructuring. 1 spell id referenced by an episode has no row in the spells table: ALB_1991-1995 (Albania). Methodology

China cut face value 4 times in 55

China's own restructurings of its overseas loans are recorded separately, with no haircut attached: 55 deals in 29 countries between 2000 and 2020, of which 49 are typed “debt rescheduling only”, 4 cut face value, and 2 have no type recorded. 10 of the 55 cut the interest rate. The file records no present-value haircut for any of them, so the loss China took cannot be compared with the numbers above. What can be said is how often principal was reduced: 4 times in 55.

Source: Horn, Reinhart & Trebesch, China's overseas lending and debt restructurings 17 of the 29 countries in China's restructuring record also have a restructuring in the haircut file. The interest-rate flag is set or absent, never zero, so "not recorded" is not the same as "no". Methodology

What is missing: the defaults that have not ended

Source: Bank of Canada-Bank of England Sovereign Default Database (CRAG) Bank of Canada terms (attribution). Amounts are the stock of debt in default across all creditor classes, including domestic arrears, so they are not comparable with the debt-treated column above. Methodology

Every country with a restructuring

94 countries have at least one concluded restructuring in the file and get a page. A country with no episode gets no page.

Source: Meyer, Reinhart & Trebesch (2022), Sovereign Bonds since Waterloo | Cruces & Trebesch (2013), Sovereign Defaults: The Price of Haircuts | Asonuma & Trebesch (2016), Sovereign Debt Restructurings: Preemptive or Post-Default Country label is the name attached to the country's most recent episode, so Serbia's 19th- and 20th-century episodes are filed under the label of its 2004 deal. Methodology