Why Do Countries Repay Their Debts?

The study of international government finance has long been plagued by a seemingly irreconcilable paradox. In theory, a sovereign debt contract is little more than an ownership title expressing a claim on part of the state's future tax revenues. But since the counterparty in question is by definition always a sovereign power — whose actions have legal force within a given territory and whose privileges include the ultimate authority to ordain and rescind laws — this immediately raises the question of how such contractual claims are enforced in practice. In the absence of a world government or a higher legal authority capable of enforcing cross-border debt contracts and compelling national governments to live up to the letter of their prior commitments, a country that is either unwilling or unable to meet its financial obligations should in theory be able to default on its external debts without repercussions. Seeing that the servicing of interest on international loans effectively constitutes a wealth transfer from the debtor to its foreign creditors, and assuming that government representatives always act rationally in defending the national interest, we would expect distressed borrowers in particular to suspend foreign debt servicing much more frequently than they actually do.

We have already seen how, at the extreme, the logic prescribed by the assumptions of neoclassical economics would therefore mean that a self-interested government should try to borrow as much as possible before repudiating its accumulated external debts in total. If all sovereign borrowers displayed such opportunistic behavior in face of their international obligations, rational lenders would assess the risks and simply refuse to disburse further credit, causing global capital markets to collapse and external debt to disappear altogether. Yet this is clearly not what happens. If anything, the opposite is the case: global capital markets have been thriving ever since the 1970s, and heavily indebted peripheral countries generally do repay their debts, forking over hundreds of billions of dollars in interest payments to their rich-country creditors every year (see figure 1.1), even if they have to go through great pains to remain current on their foreign obligations. As the renowned sovereign debt lawyer Lee Buchheit puts it, "conventional wisdom is that sovereigns will rarely, if ever, default on their external debts in circumstances where it is clear that they have the capacity to pay." Economists at the IMF recognize that distressed borrowers appear to prefer avoiding default "even if that implies running down reserves, shortening the maturity of the debt, and ceding part of their economic policy sovereignty to multilateral institutions." If, for whatever reason, a government finds itself unable to repay its debts in full or on time, it will generally prefer to negotiate an orderly settlement with its creditors over a unilateral suspension of payments. In practice, it therefore remains extremely rare for countries to simply stop paying, let alone to repudiate their obligations outright.

In the economics literature, this striking puzzle at the heart of international lending has long been known as the "enforcement problem" of cross-border debt contracts: clearly there is some kind of cross-border enforcement at work, but the precise mechanism through which it operates is not immediately observable, and economists still do not understand how exactly it works. Apparently most governments do consider default to be costly, preferring to impose painful austerity measures on their own citizens instead. But why is this so? Over the past three decades, this question has inspired a number of hypotheses about the precise costs of default and the exact enforcement mechanisms of debtor compliance. In the resultant debates, four explanations emerged, centering on (1) the borrower's long-term reputation; (2) legal and trade sanctions; (3) democratic institutions; and (4) spillover costs, respectively. Yet despite the important advances that have been made by a new wave of sovereign debt scholarship in recent years, the economics literature is still at pains to answer the most basic question of all: how can external sovereign debt even exist in the first place?

In this chapter, I will briefly discuss the four conventional explanations of debtor compliance found in the economics literature and assess their validity in light of the available evidence. Since economists were the first to identify the puzzle at the heart of this book, it is important to take a closer look at why their abstract theoretical models have so far struggled to account for real-world outcomes. Given the somewhat specialist nature of these debates, the lay reader may want to skip this theoretical discussion and delve straight into the politics and history of sovereign debt in subsequent chapters.

Reputation: The Threat of Long-Term Market Exclusion

The sovereign debt puzzle outlined in the introduction really became an issue of interest for economists only in the late 1970s and early 1980s. The wave of sovereign debt defaults during the Great Depression had caused global capital markets to freeze up for the next forty years, and it was not until after the emergence of the Eurodollar markets in the 1960s and the eventual collapse of the Bretton Woods regime in 1973 that capital really began to flow across borders again. This in turn led to renewed scholarly interest in international lending. The first paper to systematically identify the fundamental paradox at the heart of the sovereign debt puzzle was the aforementioned theoretical contribution by Eaton and Gersovitz, who argued that countries ultimately honor their foreign obligations because they are concerned about their long-term reputation as borrowers. In this reputational model, governments borrow in order to smooth out consumption in the event of unforeseen shocks on the economy, giving them an inherent incentive to repay in order to retain access to international capital markets. "Should the country refuse to repay," Eaton and Gersovitz write, "we assume that it faces an embargo on future loans by private lenders and that this embargo is permanent." In short, countries honor their debts because repudiating them would leave them "forever unable to use international borrowing to smooth absorption across periods of varying income." In a somewhat less restrictive formulation, governments may still be able to access foreign credit following a default, but face significantly higher borrowing costs than their more compliant counterparts, as lenders assess the country's past repayment record and become more skeptical about its capacity to repay in the future, charging a higher risk premium to reflect these lingering fears of potential nonpayment. Schematically, we can represent the causal mechanism at the heart of this account as follows:

[FIGURE OMITTED]

From a quick glance at this visualization, it immediately becomes apparent that the reputation hypothesis hinges on two key assumptions: first, that lenders can and do act rationally and monolithically in their refusal to extend further credit in the event of a default; and second that they actively discriminate between sovereign borrowers on the basis of their past repayment records and demand a higher risk premium from borrowers with a history of default. When it comes to the three case studies undertaken as part of this research project, neither of these assumptions appears to hold up against the empirical evidence.

On the first point, it turns out that the countries that defaulted in the 1930s did not borrow systematically less in the 1970s, nor did they borrow on terms different from nondefaulters. While Latin American borrowers faced severely restricted credit access in the decades following the defaults of the 1930s, the effect was just as strong for a compliant borrower like Argentina, which did not default, as it was for the other countries that did. In fact, when international lending was resumed in the 1970s, there was no noticeable difference in borrowing costs between past defaulters and nondefaulters. Lenders were eager to let bygones be bygones; what mattered was not the historical repayment record of individual sovereign borrowers, but the immediate prospect of easy profits. The evidence from the 1980s therefore points in the direction of a relatively myopic investor attitude towards risk assessment, which was perhaps most blatantly expressed in the statement by Citibank CEO Walter Wriston, just before the crisis broke out, that "countries don't go bust." Angel Gurría, the current secretary general of the OECD who served as Mexico's director of public credit during the 1980s, recounts that "the banks were hot to get in" ahead of the crisis of 1982. "They showed no foresight. They didn't do any credit analysis. It was wild. ... We just issued promissory notes. We were selling them like hotcakes." For foreign creditors, "the prospect of default seemed too extraordinary to consciously consider."

This pattern of investor myopia was repeated in the lead-up to the Argentine default. If investors were really driven by past repayment records in their lending decisions, they should have remembered the country's reputation as a "debt-intolerant" serial defaulter. Instead, by the mid-1990s, Wall Street had embraced Argentina — by far the most recalcitrant debtor of the 1980s — as an investor favorite and a poster child for the Washington Consensus. "Every time we finished a meeting, the orders would come," the country's deputy secretary of finance Miguel Kiguel recalled. "People were desperate to buy Argentina."

Subsequent research on the emerging market borrowing of the 1990s has confirmed that past defaulters again did not face higher risk premiums than those with an unblemished record. As for the consequences of Argentina's default, the reputation hypothesis would lead us to expect a complete cutoff from all sources of foreign financing after December 2001. But while the government did lose access to international capital markets, it was still able to raise significant amounts of foreign credit through domestic bond auctions. After the debt restructuring of 2005, demand for Argentine bonds was so strong that the riesgo país — the risk premium charged by investors compared to U.S. Treasury bills — converged with Brazil's, which did not default on its debts. In fact, as soon as the restructuring was completed, the riesgo país that had plagued Argentina throughout its crisis returned to the levels previously experienced at the peak of the financial euphoria in 1997. More recently, in 2017, the right-wing Macri government re-entered international capital markets by issuing $2.75 billion worth in unprecedented 100-year bonds immediately after settling the country's long-standing legal dispute with U.S. vulture funds, raising widespread concerns that international investors might once again be overlooking Argentina's patchy repayment record.

The story is no different for Greece. Instead of taking into account the country's long-standing reputation as a "serial defaulter," which suspended payments in every single prewar lending cycle and spent roughly half of its existence as an independent nation in a state of default, financial markets provided the Greek government with almost the exact same borrowing costs as Germany as late as 2008, its risk spread only rising above 1 percent following the collapse of Lehman Brothers and the bankruptcy of Dubai. It was not until the announcement by Prime Minister George Papandreou in late 2009 that the previous government had been cooking the books, and that Greece's real deficit and debt load were much higher than previously thought, that the Greek-German risk spread began to widen dramatically. This appears to indicate that lenders were never really driven by Greece's long-term repayment record; rather, they based their investment decisions on short-term risk assessments. For years, until the global financial crisis of 2008–2009, investors appeared to reason that all debt in the Eurozone carried more or less the same default risk. As a result, Greece, with its perceived high growth potential, developed into an investor favorite, just as Mexico and Argentina had before it, and began to attract large amounts of credit in spite of its history of default — which is exactly the opposite of what the reputation hypothesis would lead us to expect.

It is perhaps not surprising, then, that subsequent research has cast significant doubt on the original reputation hypothesis. In a later study, published a decade and a half after his seminal contribution with Gersovitz, Jonathan Eaton was forced to acknowledge that the evidence to support his earlier theoretical paper was "ambiguous" at best. While a number of scholars have since sought to resuscitate the reputational framework by relaxing some of its most stringent assumptions, even proponents of this line of argument now admit that empirical support for this explanation is mixed and that most research of the past three decades finds the default premiums in sovereign credit markets to be negligible. To the extent that governments do pay higher interest rates after a default, most scholars would agree that these costs tend to be short-lived, ranging from a few months to about two years.

Many of these subsequent empirical refutations were already foreshadowed in an influential theoretical paper by Jeremy Bulow and Kenneth Rogoff published at the end of the 1980s, in which the authors pointed out that countries actually have other ways of insuring themselves against adverse shocks on the economy. Instead of repaying to retain access to international capital markets, a self-interested borrower could simply repudiate its obligations and invest the money it saves on interest payments in foreign capital markets, which — assuming sufficiently high returns — would provide a more profitable cushion for bad times. As a result, the reputational mechanism, hinging entirely on the debtors' need to retain credit access for future consumption smoothing, simply collapses. To have any effect at all, Reinhart and Rogoff conclude, "the reputation approach therefore requires some discipline."

Sanctions: Asset Seizures, Trade Embargoes, and Gunboat Diplomacy

Building on Bulow and Rogoff 's contribution, a second body of literature has proposed direct punishment as the main enforcement mechanism of debtor compliance. By imposing or threatening to impose sanctions, scholars in this tradition argue, private lenders and creditor states can directly coerce recalcitrant debtors to repay. Such creditor sanctions could take either of two forms: the seizure of a debtor's assets abroad, or the imposition of a trade embargo. Subsequent work has also highlighted the historical importance of so-called supersanctions, where external financial control, military coercion and the threat of outright occupation served as the principal enforcement mechanisms. What these different types of sanctions have in common is that they all seek, through some form of direct creditor action, to "raise the cost of default sufficiently high to make repaying the foreign obligations in the self-interest of the sovereign debtor." Like the reputation hypothesis, the sanctions approach therefore remains firmly within the boundaries of neoclassical cost-benefit analysis and rational choice theory. "In this sense," Reinhart and Rogoff note, "the reputation and legal approaches are not so different."

In the legal approach, schematically represented in figure 1.3, sovereign borrowers will do almost anything to avoid default because of the danger of lawsuits inflicting further damage on an already strained national economy. Since historically most emerging market and developing country debt has been denominated in foreign currencies and contracted in other legal jurisdictions, debtors have generally been liable to the laws of the country where the debt was issued (although this is now slowly changing, as emerging markets in particular have begun to issue more debt domestically and in their own currencies). The legal sanctions hypothesis holds that there is therefore no way for the borrower to protect itself from aggressive litigation pursued by creditors inside the issuing country. Selective default, discriminating between domestic and foreign creditors, is generally not an option either, especially in the case of securitized bond finance, where secondary markets add a veil of anonymity to bond holdings and where a suspension of payments is likely to trigger cross-default clauses. The notorious example of the U.S. vulture fund Elliot Associates buying up Peru's greatly depreciated postdefault bonds far below par, and then suing the government for its refusal to repay the face value, is often cited as an exemplary case of aggressive litigation bearing fruit for the persistent speculator. After Elliot succeeded in attaching some of Peru's foreign assets, including a Brady bonds payment that was to be channeled through a Brussels-based clearing house, Peru found itself forced to settle and repay part of its defaulted debt.