Actually, Sovereigns Do Go Broke
In a vivid recent example, the government of Greece, in its 2012 debt restructuring, paid private holders of its defaulted debt 25 cents on the dollar, so these creditors suffered a 75 percent loss from par value. Greek government debt was at the vortex of Europe’s 21st century sovereign debt crisis. Various governments of Greece have defaulted seven times on their debt, which has been in default approximately half the time since the 1820s.
With such a record, how soon would the lenders be back this time?
Pretty soon, as usual. Defaults were the past; new loans proclaim a belief in the future. Thus in July, the print edition of the Financial Times informed us, “Greek debt snapped up as investors seek higher yields”! That’s a headline that would not have been predicted a few years ago—except by students of financial history who have observed the repeating cycles of sovereign borrowing, default, and new borrowing.
“Greece has seen vigorous demand for its latest bond sale,” read the Financial Times article. “The Mediterranean country received orders of more than €13 billion for the seven-year bonds, well above the €2.5 billion on offer.” And the “higher yield”? A not very impressive 1.9 percent. The recently again-defaulting Greek government has succeeded in borrowing at the same interest rate as the United States government was at the same time for the same tenor. Of course the currencies are different, but this is nonetheless remarkable.
Note the common but inaccurate figure of speech used in the article. It talks about the country borrowing, when it is in fact the government of the country that borrows. That these two are not the same is an important credit consideration. Governments can be overthrown and disappear, while the country goes on. Governments can and do default on their debt with historical regularity.
Breaking the Faith
Notorious in this respect is the government of Argentina, which has “broken good faith with its creditors on eight occasions since it declared independence from Spain in 1816,” as James Grant reminds us. That is a default on average about once every 25 years. Obviously the lenders reappeared each time—in 2017, they bought Argentine government bonds with a maturity of 100 years. That is long enough on average to cover four defaults. In August 2019, the Argentine government announced it would seek to restructure its debt once again, and its 100-year bonds at the end of the month were quoted at 41 cents on the dollar.
In contrast to this, an optimistic columnist for Barron’s pronounced in that same August that sovereign bonds “have minimal to no credit risk because they are backstopped by their governments.” This financially uneducated statement is reminiscent of the notorious Walter Wriston line that “countries don’t go bankrupt.” Wriston, then prominent in banking as the innovative chairman of Citicorp, was defending the credit expansion that would shortly lead to the disastrous sovereign debt collapse of the 1980s. While sovereign governments indeed do not go into bankruptcy proceedings, they nevertheless do often default on their debt.
The great philosopher, economist and historian, David Hume, famously argued two and a half centuries ago, “Contracting debt will almost infallibly be abused, in every government.”
Max Winkler shared a realistic appreciation of the risk involved, as he was writing during the sovereign debt collapse of the 1930s. His instructive and entertaining book, Foreign Bonds: An Autopsy (1933), provides a simple but convincing explanation for the recurring defaults. Considering “politicians in the borrowing countries, from Abyssinia to Zanzibar,” Winkler memorably observed:
The position they occupy or the office they hold is ephemeral. Their philosophy of life is carpe diem. . . . Tomorrow they may be swept out of office. Today they can live only by yielding to the multiple undertaking of expenditures, proposed by themselves and their temporary adherents. . . . In order to enjoy the present they cheerfully mortgage the future, and in order to win the favor of the voter they . . . exceed the taxable possibilities of the country.
This sounds familiar indeed. We only need to update Winkler’s A to Z country names—we could make it “from Argentina to Zimbabwe.” Otherwise, the logic of the politicians’ behavior he describes is perpetual. It applies not only to national governments but to the governments of their component states, cities, and territories, like over-indebted Illinois and Chicago; New York City, which went broke in 1975; and Puerto Rico, now in the midst of a giant debt restructuring, among many others.
The observation fits the governments of advanced, as well as emerging, economies. This political pattern includes the expanding debt of the United States government, although it has not defaulted since 1971. In that year, it reneged on its Bretton Woods agreement to pay in gold. The U.S. government also defaulted on its gold bonds in 1933. Then Congress declared that paying these bonds as their terms explicitly provided had become “against public policy.”
The always insightful Chris DeMuth, writing in The American Interest and pondering the long-term trend of rising U.S. government debt, proposes that we have seen “the emergence of a new budget norm.” This is “the borrowed benefits norm.” “Voters and public officials,” he writes, have forged “a new political compact: for the government to pay out benefits considerably in excess of what it collects in taxes, and to borrow the difference.” He points out that the benefits “are mainly present consumption and are not going to generate returns to pay off the borrowed funds. Borrowing for consumption leads to immoderation now, immiseration down the road.”
This scholarly language captures the same behavior Winkler described in more popular terms in 1933.
A Habit of Default that Few Seem to Have Noticed
How frequent are defaults on sovereign debt? In their modern financial classic, This Time Is Different (2009), Carmen Reinhart and Kenneth Rogoff counted 250 government defaults on their external debt between 1800 and 2006, or 12 sovereign defaults per decade on average (of course, there have been more since 2006). In addition, they found 70 defaults on domestic public debt over that period.
A study by the Bank of Canada finds that, since 1960, 145 governments “have defaulted on their obligations—well over half the current universe of 214 sovereigns.” That is on average 24 defaulting governments per decade.
The study considers “a long-held view among some market participants . . . that governments rarely default on local currency sovereign debt [since] governments can service such obligations by printing money.” It points out that “high inflation can be a form of de facto default on local currency debt.” Holders of U.S. Treasury bonds found that out in the Great Inflation of the 1970s, when the bonds became called “certificates of confiscation.” But not counting the inflation argument, the Bank of Canada still finds 31 sovereigns with local currency defaults between 1960 and 2017. “Sovereign defaults on local currency debt are more common than is sometimes supposed,” it concludes.
The Wikipedia “List of sovereign debt crises,” relying heavily on Reinhart and Rogoff, shows 298 sovereign defaults by the governments of 88 countries between 1557 and 2015.
“The regularity of default by countries on their sovereign debt” is how Richard Brown and Timothy Bulman begin their study of the Paris Club and the London Club. These are organizations of governmental and private creditors, respectively, to negotiate with over-indebted governments. The first Paris Club debt rescheduling was in 1956 for Argentina; the London Club’s first was in 1976 for Zaire. (A to Z again.) The clubs have been busy since then. “Reschedulings increased dramatically from 1978 onwards,” Brown and Bulman observed in 2006. The current webpage of the Paris Club reports that in total it has made 433 debt agreements with the governments of 90 debtor countries.
The cycle of sovereign borrowing, default, and new borrowing has a long and continuing history. “Defaults will not be eliminated,” Winkler wrote in 1933. He further predicted that “debts will be scaled down and nations will start anew,” and that “all will at last be forgotten. New loans will once again be offered, and bought as eagerly as ever.” He was entirely right about that, and now we observe once again “Greek debt snapped up.”
How far back in time do government defaults go? Over 2,300 years in Greece. As Sidney Homer, in A History of Interest Rates, tells us: “In 377-373 B.C., thirteen [Greek] states borrowed from the temple at Delos, and only two proved completely faithful; in all, four-fifths of the money was never repaid.”
Shall we expect the fundamental behavior of politicians and governments to change?
Image credit: Route66