As access to vaccines continues to hamper developing countries’ response to the COVID-19 pandemic, many of these countries also face significant public debt burdens. The Debt Service Suspension Initiative (DSSI), intended to ease pandemic-related burdens on low and middle income countries, expires at the end of 2021. The DSSI saved 43 countries $5.7 billion in total, paling in comparison to their overall debt levels.  Last month, the Group of 20 (G-20) central bank heads and finance ministers offered a broader plan, the Common Framework, for restructuring debt to official creditors. They also noted the importance of involving private sector creditors in debt relief and restructuring.

While permissive conditions in international capital markets have allowed many developing country governments to borrow – from bond investors and commercial banks, as well as from other governments – easy money doesn’t persist indefinitely.  Indeed, sovereign default accelerated in 2020, as did sovereign credit ratings downgrades. Ecuador, Lebanon, and Zambia were among the ranks of defaulting governments.

Last month, the risk premium for Sri Lanka’s outstanding bonds has increased dramatically, reflecting investors’ concerns that the government would not be able to repay at least $2.5 billion in dollar-denominated debt coming due in the next year.  Anticipating a $1 billion bond repayment due on July 27, Sri Lanka’s government introduced new controls on investment outflows. These controls were intended to keep foreign currency in the country, as the maturing debt is denominated in dollars. Sri Lanka did, in fact, meet its July 27 payment deadline, but many observers anticipate default on the remaining foreign currency-denominated obligations.

As Sri Lanka contemplates asking the International Monetary Fund for assistance and the G-20 continues to discuss debt burdens, it is worth considering not only how much governments borrow, but also the ways in which they do so. When governments borrow by issuing bonds, as over 130 governments have done since 1990, they do so with varying terms. Bonds vary in their yield (the interest rate), their time maturity and their currency of denomination, as well as in their governing laws and collective action provisions. Governments and investors make tradeoffs across these terms: short-term borrowing, for instance, usually comes with a lower interest rate, but requires the government to more frequently refinance or repay its debt.

These choices over terms matter: borrowing in foreign currencies tends to be cheaper, but it requires governments to generate foreign exchange – as Sri Lanka must do at present. Two decades ago, it seemed that developing countries had little choice but to borrow in foreign currencies: international investors did not view their currencies as sufficiently stable. So investors were unwilling to purchase domestic currency-denominated debt. This “original sin” problem meant that governments frequently needed to generate foreign currency to  repay their debts.

In a newly-published article [ungated until the end of September], we document the shift away from original sin: among non-OECD countries, over 90 percent of the volume of internationally-listed sovereign bonds issued by developing countries in the mid-2010s were denominated in domestic currencies. While the most credit-constrained countries may still borrow in foreign currencies (and may access credit mostly from official, rather than private, sources), many global South countries are no longer tainted with “original sin.” Macroeconomic fundamentals, such as the level of income per capita and the level of debt, explain only part of the variation in the issuance of domestic versus foreign currency bonds.

Left governments, all else equal, prefer the autonomy that comes with borrowing in their own currencies

Rather, politics within countries also affect borrowing choices: we find that left-leaning governments are – compared to the right-leaning counterparts – significantly more likely to borrow in domestic currencies. We suggest that this is evidence of persistent partisan differences: left governments, all else equal, prefer the autonomy that comes with borrowing in their own currencies. Right-leaning governments, by contrast, may welcome the tying of hands (of the current government as well as it successors) created by foreign currency denomination. We also establish that domestic currency issuance is easier for left governments without a history of currency and inflation crises, as well as for those with more autonomous central banks or with fixed exchange rate regimes.

These choices of currency denomination can make a big difference: when governments find themselves needing to refinance or repay existing debt, those with domestically-denominated debt will not need to generate foreign exchange in order to repay. Moreover, borrowing in domestic currency means that exchange rate crises – which may result from country-specific problems or from contagion – will not necessarily result in sovereign debt crises.

Developing country governments… are perhaps less constrained by the dictates of international capital markets

At the same time, there may well be tradeoffs: issuing in domestic currency is associated with shorter debt maturities as well as higher interest rates. Governments that want to gain some insulation from market pressures, via domestic currency or longer-term borrowing, likely will need to pay higher borrowing costs – meaning fewer funds available for other purposes.

The broader lesson here is that developing country governments often have choices when it comes how to access credit; they are perhaps less constrained by the dictates of international capital markets than they once were. And their sovereign borrowing choices often reflect not only macroeconomic profiles, but also governments’ attention to their domestic political circumstances. Some governments may seek funds from commercial banks, rather than from bond issues, because they want to avoid public disclosure of their fiscal activity. Or they may be drawn to official credit from China, because even though it is less concessional, it also comes with fewer economic performance requirements than credit from the World Bank.

Yet when global or regional financial conditions shift and countries find themselves in debt distress, these choices matter for the nature, speed and likelihood of crisis resolution. Private sector creditors may be less willing to participate in debt restructuring than their official sector counterparts. Among official sector creditors, we may observe a divide between traditional “Paris Club” lenders and new entrants such as China. And, on the debtor side, governments may privilege some creditors over others – making each set of creditors worry, especially in the absence of a global sovereign debt restructuring mechanism or its analog, about securing the right to repayment.

Hence, when we talk about debt, we need to talk not only about how much governments borrow, but also from which creditors, and on what terms. Choices about borrowing in foreign or domestic currencies, at long or short maturities, or from bilateral versus official creditors, have important implications for how exposed governments find themselves to debt crises and to external market pressures.