As Prepared for Delivery
Thank you, Adam, I appreciate the opportunity to speak here today. I last gave a talk at PIIE a couple of years ago, while wearing my previous hat as an academic. One great thing that you immediately notice about PIIE’s experts and affiliates is that in addition to economic theory, they care about the practical details, institutional and otherwise, that can constitute the difference between a successful or unsuccessful policy. Now, as a policymaker at Treasury, I have a more visceral appreciation for this reality, including when it comes to cross-border lending and to cross-border borrowing. I thought, therefore, this would be a great place to discuss some practical ways to do both better.
Developing and emerging markets were in a very difficult spot at the start of the year. The COVID-19 pandemic led to a 2 percent fall in their collective output in 2020, the first overall contraction in the post-World War II period. Governments, appropriately, borrowed and spent to weather the shock and fight back against the disease. Their debt levels jumped from 54 percent of GDP before Covid to 64 percent by the end of last year.
And then, at this time of great vulnerability, Russia unleashed its brutal and unjustified war against Ukraine. The war has caused food and energy prices to jump and accelerated inflation around the world. As advanced economy central banks have raised rates, cross-border investment flows have retrenched. Net outflows across emerging markets in the first quarter of 2022 were the largest since the global financial crisis. The dollar, in real terms, has appreciated to levels not seen for several decades.
Together, these ingredients all but assure debt distress in a number of countries. We’ve all seen the recent images of social unrest in Sri Lanka. While the specter of a systemic sovereign debt crisis has not materialized, the difficult global conditions are amplifying domestic vulnerabilities, and economic stresses are appearing around the world. More, I fear, may come.
In the short-run turmoil following some episodes of distress, we should be ready to quickly deliver assistance, potentially taking steps such as repurposing existing international programs to fill urgent needs. But what can be done to help after that? And what can we do to reduce the likelihood of these events in the first place?
Today, I would like to discuss the logic behind working together to resolve unsustainable debts, why multilateral restructuring has in practice become more difficult, and how roadblocks in that process prevent countries from being able to take advantage of the international financial institutions that we’ve all worked so hard to build. I’ll suggest some concrete steps that creditors, borrowers, and the global community can take -- stopping some practices and ramping up others -- to achieve better outcomes now and into the future.
1. Dealing with Overhang
Let me start by noting that it’s a good thing that developing countries and emerging markets can borrow from abroad and we should continue to pursue policies that allow them to do so. These countries have less capital per worker than advanced economies and so have many investment opportunities offering high economic and social returns. Given their incomes will be higher in the future, it makes good sense for them to finance these opportunities with foreign capital. Not to mention the additional benefits carried by cross-border investment, such as positive technology transfers. These financial flows are also a good thing for the United States, offering our citizens and companies a chance to diversify risks, finance projects complementary to domestic production, and more generally deepen cultural and commercial ties.
Of course, macro shocks and other unpredictable developments can stress the sustainability of this borrowing. For many countries, such stresses can be resolved by pursuing the appropriate reforms, perhaps involving the International Monetary Fund (IMF) or others, without any need to restructure. However, for countries with debt loads that significantly exceed the value of what will likely be repaid – a condition referred to as debt overhang – we have a now-standard playbook.
The first step is to reduce the country’s external debts to make sure the borrower doesn’t forgo useful projects. After all, with debt overhang and no debt restructuring, some of the return to future investments will go toward repaying the existing creditors, leading to underinvestment. It is in the country’s interests and also in the collective interest of the creditors to avoid this outcome.
Given this alignment of interests, won’t the creditors on their own offer the efficient level of relief? Unfortunately, no. Together, the creditors would benefit from restoring sustainability and allowing the country to obtain new financing to pursue high-return activities. But individually, each creditor would rather be repaid in full and let the others take losses. Further, some creditors, focused on their own interest, may prefer deeper cuts to local spending than what a global social planner would suggest. Much of the international financial architecture developed over the past several decades aims to deal with these incentive and collective action problems. The Paris Club, for example, was formed to coordinate debt restructuring among bilateral official creditors. The London Club was formed to similarly organize commercial bank creditors.
Once a borrower makes a good faith effort to restructure its private debts and official bilateral creditors provide “financing assurances” – promises that they’ll sufficiently restructure what is owed to them -- we reach the second step, an agreed IMF program. The IMF then provides lending -- coupled with conditionality and expertise -- to help the country restore macroeconomic stability and catalyze fresh lending for high return investments and growth.
I do not mean to make it sound so easy, nor to imply that these elements are sufficient for a good outcome. It is only natural that there may be imperfections when countries work with the IMF to design and implement programs. And, of course, additional shocks and uncertainties often intervene. But having this playbook in place and ready to deploy has been good for the world.
Will this playbook work if we see a new wave of debt crises among developing and emerging market countries? Four major changes in the international lending landscape over the past decade have strained its efficacy. First, the debt burdens of developing and emerging market countries have risen considerably. Second, use of non-traditional arrangements, including collateralized borrowing, has proliferated. Third, private sector creditors have grown in importance. Fourth, while many official creditors have shifted their focus toward offering grants, non-traditional official creditors have increased their lending to developing and emerging markets. In particular, China has vastly expanded its portfolio of loans and trade credits and is now by far the largest bilateral official creditor in the world. All these elements have introduced new complexities to the needed coordination among creditors in debt restructurings. My remarks will focus on some steps that creditors, borrowers, and international financial institutions should take in response.
2. Recent Complications in Sovereign Debt Restructuring
One important change in the creditor landscape stems from how China restructures its bilateral debts. Chinese policy and commercial banks typically rely on limited cash flow treatments and do not write down large losses. Researchers have found that the majority of Chinese debt relief deals have come with significant delays and have not reduced the borrower country’s nominal debt burden. Instead, the deals involved lengthening maturities or grace periods, and in fewer cases, interest rate reduction or new financing. Only four cases since 2000 have reportedly involved haircuts on Chinese official debt. In some cases, such as the Republic of Congo in 2018, debt restructurings have even increased the net present value of China’s loans. As a result, the restructurings typically do not resolve the debt overhang and can stoke uncertainty about the need for repeated rescheduling.
Does the approach of any one country in this process matter all that much? In fact, China’s enormous scale as a lender means its participation is essential. Estimates of the total stock of outstanding Chinese official loans range widely from roughly $500 billion to $1 trillion, concentrated in low and middle-income countries. China became the world’s largest official creditor in 2017, surpassing the claims of the World Bank, IMF, and all Paris Club official creditors combined. A recent study estimates that as many as 44 countries now owe debt equivalent to more than 10% of their GDP to Chinese lenders after factoring in both on- and off-balance sheet liabilities. Failure to act on these debts could imply years of ongoing difficulties with the servicing of debts and with underinvestment and lower growth in low and middle income countries.
In November 2020, the G20 established the Common Framework to bring China and other non-Paris Club creditors into a multilateral effort to restructure the debts of low-income countries on a case-by-case basis. This helpful instance of global economic cooperation carried the hope that by bringing all creditors together, the Common Framework would result in needed debt treatments in a timely and orderly manner. However, that ambition has not yet materialized.
There have been three Common Framework cases to date. In the case of Chad, China delayed the formation of a creditor committee. In the case of Ethiopia, China delayed the formation of a creditor committee until it was clear that the IMF program would expire and discussions on a debt treatment would not take place because of the conflict. China did join the creditor committee for Zambia, but only after six months of delays following the staff level agreement with the IMF. Encouragingly, China announced in late July that it and the other official bilateral creditors would provide Zambia with debt treatments, and the IMF approved a three-year lending program for Zambia on August 31. But even if, as I hope, the situations ultimately improve in all three countries, these delays carry suffering and uncertainty, may discourage others from requesting needed treatments, and preclude the best outcomes. Successful efforts under the Common Framework have, alas, been uncommon.
Progress has also been slow on debt restructurings for middle-income countries outside of the Common Framework. Furthermore, China has engaged in unconventional practices that have allowed the IMF to move forward in several recent cases without obtaining standard financing assurances. For example, in 2020, China agreed to $2.4 billion in new lending as a financing assurance for Ecuador’s IMF program and to offset upcoming interest payments due to Chinese creditors, even as private creditors agreed to a $17.4 billion debt restructuring. However, rather than delivering the promised new financing in a matter of months, China only came to terms with Ecuador this past week. In Suriname, China and India have so far failed to provide financing assurances that the IMF considers specific and credible, leading to the IMF’s reliance on an unusual application of its policy to move forward with a program. In the case of Argentina, China is not restructuring debt service, while Paris Club creditors are likely to do so, and China has opted instead to promise net financing by offering new loans.
In many of these cases, China is not the only creditor holding back quick and effective implementation of the typical playbook. But across the international lending landscape, China’s lack of participation in coordinated debt relief is the most common and the most consequential.
3. What Can Creditors, including China, Do?
The immediate priority for creditors is to conclude the pending Common Framework cases as quickly as possible. We are also open to expanding the Common Framework to middle-income countries and should prioritize discussions of ideas, including those offered by new creditors, to boost the speed and predictability of the process.
Creditors should also focus on making their lending more transparent, since a lack of clear information on a borrower’s indebtedness makes lending to them riskier ex-ante and makes resolutions more difficult ex-post. G7 countries have largely met their commitment to publish their direct lending portfolios on a loan-by-loan basis, and at the U.S. Treasury, we’ve greatly improved our online foreign credit reporting system. G20 creditors should also follow the norms for financing terms and contractual clauses they endorsed in 2018.
A number of features common in China’s lending reduce transparency or differ from international norms in ways that exacerbate the coordination problem in multilateral restructurings. Chinese loan contracts often contain non-disclosure agreements. As a result, liabilities to China are systematically excluded from multilateral surveillance or restructuring efforts. One study found that all contracts with Chinese state-owned entities after 2014 contain strong confidentiality clauses that prevent the borrower from disclosing any terms unless required to do so by law. Chinese loan contracts also commonly feature novel clauses that allow Chinese lenders to cancel loans and demand immediate repayment in a wide variety of circumstances. Collateral is included in up to half of all Chinese loan contracts, and arrangements for repayment commonly involve escrow accounts.
All these elements limit a borrower’s ability to engage in standard multilateral restructuring processes and incentivize the borrower to cut side deals on more generous terms with the Chinese creditor. In fact, around three-fourths of all Chinese debt contracts contain clauses that expressly commit the borrower to exclude Chinese debt from Paris Club restructurings or from any comparable debt treatment. Countries that seek a Paris Club debt restructuring find themselves either having to violate the terms of their borrowing from China or to violate the principle of comparability of treatment.
Additional problems may stem from institutional fragmentation within China’s internal lending bureaucracy. In contrast to the typical practice among traditional official creditors, debt distress appears to be managed by the specific Chinese creditor, rather than by the Ministry of Finance (MOF) or an agency that is not the lender, like the People’s Bank of China. But Chinese policy banks and state-owned commercial banks do not report directly to the MOF or to the PBOC. As a result, the MOF and PBOC have often been unable to provide details on behalf of the creditor banks on a debt treatment for the purpose of providing financing assurances to the IMF. This institutional fragmentation can also keep China’s lenders from coordinating on an assessment of borrowers’ debt sustainability prior to extending loans.
Some of these institutional frictions may simply reflect the fact that China has grown so rapidly as a creditor. And they are only natural given China is so new to the restructuring process. That said, there are many ways China could quickly reduce these frictions.
In addition to improving its own tracking and transparency of all its foreign lending, China might consider implementing different financial structures, such as the creation of a sovereign asset management entity, or “bad bank,” as a way to allow the various creditor agencies to isolate distressed loans. Specialized management expertise can then focus on restructuring while the development banks and other lending institutions are freed up to return to their normal focus. With the right design features, such a structure might restore incentives toward resolving, rather than holding out, on troubled debts. A useful parallel may be made with reforms to the Chinese bankruptcy system. Amidst large increases in corporate debt and bankruptcies, China reformed its corporate bankruptcy code in 2007 and introduced specialized courts over the following decade. Research suggests these actions led to an increase in the scale and speed of resolutions and boosted productivity in China.
Changes in institutional structure or legal and management practices related to the restructuring of China’s external debts as part of the multilateral process would not only benefit low-income borrowers around the world but could also benefit China itself.
Private creditors could also do more to improve their own transparency. The OECD launched the Debt Transparency Initiative in 2021 to operationalize the Voluntary Principles for Debt Transparency that were helpfully put together by the Institute for International Finance (IIF), but few have participated so far. As with many of these suggestions, by reducing risk, information frictions, and coordination problems, such steps not only help borrowers but can also benefit the creditors themselves.
4. What Can Borrowers Do?
Borrowers can of course take their own steps to maximize the likelihood that, if they do end up in debt distress, they can obtain relief and restore stability and growth, including with an IMF program.
The antidote to non-disclosure by creditors is transparency by borrowers, and a number of countries have made remarkable progress on this dimension. Benin strengthened its level of debt disclosure by creating a debt portal and extending the coverage of its debt statistics. Regular trainings and interactions with the World Bank and IMF have improved capacity. Along with prudent macro-fiscal policies, Benin’s reputation in markets strengthened sufficiently that it was able to issue international debt, even in the middle of the pandemic. Burkina Faso, with the help of the Bank and Fund, last year issued its first Statistical Debt Bulletin that meets international best practices and includes detailed information on loan terms and conditions. It includes government direct debt, guarantees, and public-private partnership contracts. Madagascar in 2014 adopted a Law on Public Debt and Guarantees that introduces reporting requirements, including to their Parliament. As the sovereign debt expert Anna Gelpern has suggested, more countries should consider passing laws requiring their governments to, in essence, make their public debt public.
With a clearer picture of their own consolidated external liabilities, borrowers could implement measures that minimize the likelihood of individual ministries or state-owned enterprises from contracting unsustainable debts. Many low-income countries already have legal provisions that require authorization of all public sector borrowing by the central government, and more should consider doing so. Some, including West African countries such as Cote D’Ivoire and Gambia, have gone further by allowing their central government to impose debt limits on all public entities.
Borrowers should also look around, evaluate which types of arrangements produce positive long-term outcomes, and exercise caution when agreeing to unusual contract terms. If a borrower country has an abundance of labor that is qualified to work on infrastructure, is there a good reason to agree to only deploy foreign workers on financed projects requiring standard construction skills? Is there a good reason for payments to be made via an offshore escrow account that can be controlled by the creditor?
Borrowers might further enhance their sharing of information and best practices with technical-level regional dialogues and conferences, including presentations by experts in procurement, project audits, and debt management. For example, the Uruguay Chamber of Construction recently held such a conference, promoting international best practices in public contracting that brought together experts from across the world to share their experiences. Government officials, financial and legal experts, and construction and infrastructure firms shared their perspectives on how, at each stage of the procurement life cycle, decisions should not entirely reflect price, but also the quality and anticipated return of the projects and the social benefit brought to the local community.
5. What Should Multilaterals and the Rest of the World Do?
We should continue to support initiatives that foster sustainable lending practices and transparency, like the G20’s guidelines or the OECD’s and IIF’s debt transparency initiative. We should also continue to support initiatives that provide capacity building assistance to developing countries, like the World Bank Debt Management Facility.
We should build on recent successes such as the adoption of enhanced collective action clauses in bond contracts to minimize coordination problems and reduce disruption from holdout creditors. The G7 Private Sector Working Group (PSWG), led by the U.K., is making helpful progress in developing templates for majority voting provisions that serve a similar purpose in non-bonded debt.
We might also strive to create deeper markets for well-designed state-contingent securities, which could reduce the need for debt restructurings in the first place. Borrowing countries should issue whatever securities are best suited to their needs, but we should continue efforts to brainstorm, create, and stress-test contract templates, whether the ideas come from the private sector or from the official sector, such as the climate resilient debt instruments currently worked on by the PSWG.
We must minimize the chances that the IMF provides financing that could ultimately be used to repay select creditors. The IMF’s and WB’s debt sustainability analyses (DSA) constitute the core of efforts to coordinate and overcome debt overhang, modeling quantitatively the needed policies and, potentially, debt treatments. The expanding set of creditors and complexity of lending approaches may have made it more difficult, but the IMF must remain vigilant so that restructurings both meet the terms of the DSAs and are promised with financing assurances that hit the IMF’s standard of specific and credible, especially for creditors without a robust track record of meeting this standard. Details about financing assurances – their form, scale, provenance, etc. – should be more transparently reported and tracked in staff reports. All official bilateral creditors must be treated equally in these restructurings.
Finally, we should increase the scale of high-quality development financing, particularly through the multilateral development banks, by stretching their existing resources and using them to mobilize private capital. This would not only advance our development goals, but has the added benefit of providing developing countries with a stronger set of alternatives when considering loans with potentially onerous terms.
As I mentioned earlier, official bilateral creditors, led by China and France, agreed in late July to offer debt treatment for Zambia. This is a great step, but it is only the first step, toward finalizing technical details and actually delivering the relief. Now that the IMF program for Zambia has been approved, it is vital that creditors move expeditiously to hash out, and then transparently meet, the terms of the debt treatment.
And Zambia is not the only pressing case. Sri Lanka urgently needs a debt restructuring and, unfortunately, is not eligible for the Common Framework. On September 1, Sri Lanka reached a staff-level agreement with the IMF on a robust set of policies to restore economic stability. In order to bring this lending program forward, creditors now need to step up to negotiate a debt treatment that is line with this economic program.
We are in new territory, but the coming weeks offer a real opportunity for progress. Borrowers, lenders, and multilaterals all have a role to play. Swiftly concluding the first Common Framework case, and making clear progress on multilateral restructuring outside of the Framework, would be a big win not only for current and future debtors and their citizens, but also for all official creditors, whether traditional ones or new ones. We must build on recent experiences, apply lessons learned, and push ahead in these cases to prevent any depreciation of our global financial infrastructure.
 See Krugman, Paul, “Financing vs. Forgiving A Debt Overhang,” NBER Working Paper, 1988 and citations therein.
 Horn, Sebastian, Carmen Reinhart, Christoph Trebesch, “Hidden Defaults,” AEA Papers and Proceedings, May 2022.
 The Republic of Congo’s total indebtedness to China increased by 26% in net present value terms after the 2018 restructuring. Gardner, Alysha, Joyce Lin, Scott Morris and Brad Parks, “Bargaining with Beijing: A Tale of Two Borrowers,” Center for Global Development and AIDData, November 2020.
 Horn, Sebastian, Carmen Reinhart, Christoph Trebesch (2021), “China’s Overseas Lending.” Journal of International Economics, Volume 133.
 Nearly 70% of China’s overseas lending is now directed to state-owned companies, state-owned banks, special purpose vehicles, joint ventures, and private sector institutions, which are often unreported but present balance sheet risks when risks materialize. Malik, A., Parks, B., Russell, B., Lin, J., Walsh, K., Solomon, K., Zhang, S., Elston, T., and S. Goodman. (2021).” Banking on the Belt and Road: Insights from a new global dataset of 13,427 Chinese development projects.” Williamsburg, VA: AidData at William & Mary.
 The Chad case remains delayed as creditors cannot agree to the terms of the memorandum of understanding.
 Gelpern, Horn, Morris, Parks, and Trebesch, “How China Lends.”
 In the case of the United States, credit renegotiation is managed by State and Treasury regardless of which agency provides the credit. The budget costs of debt treatments are typically managed by Treasury.
 Li and Ponticelli, “Going Bankrupt in China." Review of Finance, 2022.