Abstract
Emerging markets and developing economies are currently facing major challenges from global shocks, including a slowdown in global growth; food and energy price increases; decline in risk appetite of international investors; unsustainable debts in low-income countries; and ongoing climate risks. National policies have not sufficed to meet these challenges. Efforts at the national level must be complemented by changes in the global economic and financial architecture designed to make the world a safer place. In this article, we focus on the financial aspects of such reforms. The financial agenda as we see it has seven key elements: (i) reform of central bank swap lines, (ii) reform of IMF-contingent credit lines, (iii) SDR reallocation, (iv) reform of credit rating agencies, (v) creation of currency hedging instruments, (vi) inclusion of climate-resilient debt clauses in new debt instruments and (vii) steps to streamline the debt restructuring process. We detail this agenda and urge the G20 members to implement the recommended measures.
Keywords
1. Introduction
Recent events, from the COVID-19 pandemic to Russia’s invasion of Ukraine and now the global slowdown, are reminders that small open economies, and even not-so-small open economies, do not entirely control their own fate. Even when they deploy their entire arsenal of economic policy tools, they are not able to fully insulate their economies and residents from global shocks.
Those shocks have been coming fast and furious. First is the slowdown in global growth, driven in 2022 by China and prospectively in 2023 by the United States and Europe, which, even if they avoid recession, are certain to grow more slowly. This will mean weak external demand for emerging markets, many of which depend heavily on exports. Second is the terms-of-trade shock, given the unusual situation that food and energy prices have been going up even while global growth goes down. This negative shock disproportionately impacts food and energy importers. Third is the decline in risk appetite on the part of international investors as economic and financial conditions become more volatile and returns on safe assets rise, driven by sharp increases in the policy interest rates of advanced-country central banks. This shift manifests itself in the curtailment of portfolio capital flows into emerging markets and in some cases in portfolio outflows from emerging and developing countries (EMDCs). Thus, India saw a cumulative portfolio capital outflow of more than US $30 billion in the 12 months ending in October 2022. 1
Historically, this toxic mix has resulted in recession and, more than once, financial crises. Thus, it is revealing that official and private forecasters anticipate that emerging markets and developing countries will continue to grow in 2022–2023 despite this unfavourable global backdrop. IMF forecasts are representative: the Fund sees EMDCs growing by 3.9 per cent in 2022 and 4.0 in 2023. 2 Some will say that the other shoe has yet to drop and that more downward revisions of growth forecasts will yet follow. Dozens of low-income countries face severe debt-servicing difficulties. But that emerging-market economies as a class have avoided outright recession and financial crisis testifies to the progress they have made in strengthening their policy frameworks and institutions.
The particulars of that progress are well known. Emerging markets have in many cases brought down formerly high rates of inflation, often through the adoption of inflation targeting as a monetary framework, together with measures strengthening the independence of their central banks. They have turned to greater exchange-rate flexibility to facilitate adjustment. They have accumulated foreign exchange reserves to permit intervention when the exchange rate is buffeted by shocks (although a share of that reserve cushion has now been worked down as a result of recent intervention). They have strengthened fiscal rules and institutions and maintained public-debt-to-GDP ratios that on average are less than half of those of advanced economies. They have succeeded in issuing a sharply higher share of government debt in local currencies. 3 They have embraced macro-prudential policies. And monetary, fiscal and regulatory authors have done a better job of communicating their intentions and actions to financial markets and other stakeholders.
Although this progress at the national level continues, progress at the national level alone is not enough. It is past-due time to implement changes in the global economic and financial architecture designed to make the world a safer place for emerging markets and developing countries.
In this article, we focus on the financial aspect. We do so in part because we have little to say about reforms that might limit the risk and incidence of global recessions and terms-of-trade shocks. In terms of limiting global recessions, advanced countries could avoid excessive fiscal stimulus that causes inflationary pressures to develop. Their central banks could avoid falling behind the curve so that, to catch up, they are forced to jack up interest rates with a vengeance. 4 They could limit their dependence on unreliable sources of energy. They could engage in a modicum of fiscal consolidation once recovery is secure. In terms of food- and energy-price shocks, major economies could diversify their sources of supply, invest in self-sufficiency and sustainability and cooperate with one another, importantly in Europe but also more broadly. They could avoid international conflicts that threaten major supply disruptions.
The other reason we focus on the financial dimension is that important aspects of this agenda remain unaddressed. Our goal in this article is to remind the G20 of the details of this agenda and to urge members to get on with it.
2. The Agenda
The financial agenda as we see it has seven key elements: reform of central bank swap lines; reform of IMF-contingent credit lines; SDR reallocation; reform of credit rating agencies; creation of new hedging instruments for economies with currency mismatches; inclusion of climate-resilient debt clauses in the new debt raised; and steps to streamline the debt restructuring process.
2.1 Generalise Central Bank Swap Lines
Bilateral swap lines have proliferated in recent years (see Figure 1). Central bank swap lines generally, and the dollar swaps provided by the Federal Reserve in particular, have been highly effective in calming financial markets during periods of volatility. Bahaj and Reis (2022) examine deviations from covered interest parity as a measure of financial stringency and stress. They find that such deviations are smaller for currencies issued by central banks with access to Federal Reserve swap lines than for comparable currencies issued by central banks lacking such access. Further, they find a positive impact on the condition of financial institutions in the recipient countries. 5 Although there now exists a broad-based network of some 170 bilateral swap lines worldwide (Perks et al., 2021), the Bank for International Settlements (BIS) concludes that the Fed’s dollar swaps are especially important and powerful, given the dominance of the US dollar in cross-border financial transactions and the magnitude of cross-border claims on banks operating in the United States (BIS, 2020).
The Federal Reserve has acknowledged the utility of dollar swaps and made five of its temporary swap lines permanent, thereby making it easier and quicker for the recipients to tap them. But it has provided these facilities selectively, and it has not been transparent about its criteria for deciding who gets access. Other central banks similarly provide swaps of the currencies they issue to a limited number of partners. The European Central Bank maintains a euro swap line with Poland, where euro-denominated mortgage obligations are prevalent; it has tendered a total of 28 such agreements since the global financial crisis. The People’s Bank of China (PBOC) has entered into 41 bilateral currency swap agreements, arguably with the goal of encouraging trade settlement in local currencies (notably its own) rather than dealing with financial distress (Tran, 2022). 6 Still other central banks and governments provide ad hoc loans of foreign currency to top up their partner’s reserves; Saudi Arabia’s deposits with the Egyptian and Pakistani central banks are examples (The Economist, 2022). Finally, there exist regional swap agreements such as the Chiang Mai Initiative Multilateralization, under which central banks may obtain dollar or local currency swaps from partner central banks participating in the arrangement—though it is notable that this facility has not been drawn on in its two-plus decades of existence. More detail on these arrangements is provided in Appendix A.
Emerging and developing members of the G20 are all recipients of swaps of one form or another, but most smaller EMDCs are not. The G20 should, therefore, encourage central banks to broaden their networks of currency swaps. The Federal Reserve can afford to extend swaps to additional central banks without balance sheet risk to itself (if these additional central banks have other assets that can act as collateral). Other central banks with partners that do business in the former’s currency can similarly take steps to provide swaps more widely. Temporary swaps can be made permanent. Central banks with ample dollar reserves can make these available to partners as in the case of Saudi Arabia in Egypt and Pakistan, though it would be highly desirable for such arrangements to be formalised where they are ad hoc, and for the terms to be transparent. The Reserve Bank of India might negotiate similar arrangements in South Asia, while the South African Reserve Bank could do likewise in southern Africa. Such arrangements would go part way towards filling the holes in the global financial safety net.
There have also been more ambitious proposals (e.g., Gallagher & Gao, 2021) for routing central bank swap lines through a multilateral organisation such as the IMF, perhaps transferring to the Fund the power to decide who on the receiving side qualifies. However, central banks issuing hard currencies would be reluctant to cede these prerogatives. Governments and not central banks are IMF members.
Fortunately, there exists an alternative that can, in principle, fulfil the same functions, namely, the IMF’s contingent credit lines, to which we now turn.
2.2 Reform IMF-contingent Credit Lines
The introduction of IMF-contingent credit lines was stimulated by the observation that holding foreign reserves is costly (Rodrik, 2006) and that even ample reserves may not suffice to insulate countries from global shocks beyond their control.
To remind the reader, the Fund now has a trio of contingent lines. Table 1 compares their features. The Flexible Credit Line (FCL) introduced in 2009 was intended to encourage countries to seek IMF assistance before experiencing a full-blown crisis. Pre-qualified countries can draw on an FCL at any time within the period covered, without a cap of amounts, for a renewable period of an initial one or two years. But only countries with ‘very strong fundamentals’ (in the words of the IMF’s website) can qualify. 7 And countries must apply.
Comparison of the Key Features of the Short-term Liquidity Line (SLL), Flexible Credit Line (FCL) and Precautionary Liquidity Line (PLL)
The Precautionary and Liquidity Line (PLL) was added subsequently to address the liquidity needs of countries with sound economic fundamentals but with ‘some remaining vulnerabilities’ that prevent them from qualifying for an FCL. Durations are shorter, while amounts are capped at 250 per cent of the quota for the first year and 500 per cent of the quota for the entire arrangement, reflecting the existence of these vulnerabilities. Qualified countries are subject to ‘focused ex post conditionality’ designed to eliminate those remaining vulnerabilities and undergo biannual reviews by the IMF Executive Board.
Finally, the Short-term Liquidity Line (SLL), established in 2020 in response to the COVID-19 pandemic, is designed to be drawn by countries with very strong policies but facing temporary adverse capital account conditions. The qualification criteria are similar to those for the FCL, but the periods are shorter, the amounts are limited to 145 per cent of the quota, and the fees are lower if the facility is used purely on a precautionary basis. The fees convert to those applying to the FCL if the line is actually drawn.
In their first decade, only five countries signed up for an FCL or PLL: Mexico, Poland, Colombia, Macedonia and Morocco. Peru, Chile and Panama joined in 2020–2021. Of these eight, only Macedonia, Morocco and Colombia have actually drawn. Table 2 provides the details. Table 3 shows that resources potentially made available are large in proportion to the reserves that these countries hold. So, failure to apply is a paradox. Policymakers in countries with strong policies may not see the need. Or they may fear that applying sends an adverse signal to the markets. Application to the PLL may be further discouraged by the need to subject the country to IMF conditionality, regular staff monitoring and periodic Executive Board oversight.
IMF Credit Lines Approved and Availed of Country
aNumber of renewal indicates the number of times the agreement was renewed.
bTerm 1, 0.5 indicates that the agreement was first signed for 1 year, and later renewed for 6 months.
cThe agreement was first signed for 2 years and then renewed twice for 2-year and 1-year terms, respectively.
Country Reserves vs IMF Credit Lines
Lisi (2022) concludes that negotiating an FCL or PLL generally leads to a reduction in sovereign spreads and a smaller increase in spreads in the event of an adverse shock such as the COVID-19 pandemic. 8 He finds no negative impact on the spreads of countries actually drawing on these lines. The problem for analysis is that these findings are based on a very small sample, as just described, and that countries applying for contingent lines are not randomly selected, as Essers and Ide (2019) show. In our view, the case for contingent IMF lines remains to be made empirically, though the underlying analytical case remains strong, insofar as foreign currency resources have insurance value but warehousing reserves is costly.
The G20 should, therefore, endorse measures to enhance their role. Advanced countries could apply for contingent lines as a way of weakening the adverse signalling effect that deters emerging-market governments. More ambitiously, the IMF could prequalify countries rather than requiring them to apply. It could include in the Article IV report whether a country qualifies or not and the amount of the line. The charges attached to initial qualification could be eliminated entirely. Lines could disburse automatically when there is an ‘EM sell-off’ identified by the IMF staff and verified by the Executive Board.
Acceptance of a larger role for the IMF in global financial management, through the extension of contingent credit lines and provision of other forms of finance, rests on the legitimacy of the institution in the eyes of its members. As shown by the quota shares in Figure 2, the imbalance between the votes and voices of advanced and emerging G20 members is growing, not shrinking. Continued quota reform should, therefore, be an integral element of the G20 agenda.
2.3 Reallocate SDRs
The historic decision of IMF members to authorise a new $650 billion allocation of Special Drawing Rights (SDRs) in response to the COVID-19 economic crisis was supposed to be accompanied by reallocation of those SDR resources from high-income countries that do not need them to low-income countries in balance of payments and fiscal distress. Yet more than a year later, there has been little such reallocation. An unprecedented number of low-income countries have drawn on their own SDR allocations in the interim. But the bulk of the 2021 allocation remains immobilised in the hands of high-income countries that are the majority recipients.
To facilitate that reallocation process, the IMF agreed to create (and operationalised in October 2022) the ‘Resilient and Sustainability Trust’, or RST, which builds on the earlier Poverty-Reduction and Growth Trust (PRGT). Where the PRGT pools SDR-related donor funds (SDRs which can be swapped for currencies or equivalent amounts in those currencies) for lending to low-income countries, mainly for balance of payments needs, the RST is designed to address the needs of both low- and middle-income countries with longer-term funding needs, including those related to climate change and pandemic readiness. SDRs are lent rather than donated to lessen bureaucratic constraints. RSF arrangements have a 20-year maturity and a 10.5-year grace period during which it is not required to repay the principal. Concessional interest rates are in line with those on the PRGT. This effectively removes pre-existing obstacles to SDR reallocation.
Borrowing from the RST, however, requires a government to request an IMF programme. Although this can include a non-financing ‘IMF-supported programme’, this requirement can still act as a deterrent. In addition, the RST is initially capped at 150 per cent of the quota or SDR 1 billion, whichever is smaller, which translates into a sum significantly smaller than the SDR allocation received by high-income countries in 2021. The IMF foresees mobilising only $42 billion of SDRs for reallocation. As of October 2022, only six members had signed agreements to lend their SDRs, for a total of $20 billion. Staff-level agreements had been reached with only three countries (Barbados, Costa Rica and Rwanda), with negotiations still underway with a handful of others (IMF, 2022).
This is progress, but relative to ambitions attached to the 2021 SDR allocation the RST remains underpowered. The 150 per cent of the quota cap can be lifted. Conditions attached to the associated staff-monitored programmes can be further simplified and streamlined. 9 The G20 can resolve that additional advanced-country governments beyond the pioneering six should contribute to the trust.
2.4 Reform Rating Agencies
Sovereign rating changes (especially downgrades) have a substantial (negative) impact on financial conditions in emerging markets (see, e.g., Kraeussi, 2003). Stability can be threatened by so-called cliff effects, when sovereigns are downgraded from investment to non-investment grade, forcing institutional investors to liquidate their positions in response to regulatory requires or their own mandates. Ratings are procyclical, causing them to amplify economic and financial cycles: upgrades facilitate and encourage overborrowing during upswings, and during downswings precipitate financial crises (Ferri et al., 1999; Griffith-Jones & Kraemer, 2021). Fear of downgrades may also be part of the explanation why countries have been reluctant to participate in the G20’s Debt Service Suspension Initiative and Common Framework (United Nations, 2022).
These problems are likely to grow more severe as sovereign bonds become more complex, with the addition of contingencies related to inter alia climate-change- and public-health-related risks (see below). Policymakers in emerging markets will want to know how credit ratings will be affected by the addition of these clauses, and officials will want to know on what basis rating agencies are gauging the resulting risks. In contrast to government agencies, rating agencies do not publish formal debt sustainability analyses underlying their judgements. They do not provide fan charts surrounding their central cases; they do not provide scenario analyses. They do not provide model-based ratings and then explain how judgmental factors, including political judgements, cause them to modify model outputs.
The aforementioned are all best practices that could be required by regulators. 10 The cliff-edge problem can be addressed by regulatory reform at the national level. National authorities could modify regulations that require financial institutions to abruptly liquidate claims on a country when it is downgraded or to abruptly add significant amounts of additional risk capital. A more gradual approach to adjusting risk weights could be substituted. So too could rules that mandate gradual adjustments of portfolio shares in response to incremental changes in overlapping tiers of ratings. The mandates of financial institutions could be altered to require them to maintain an average rating for their entire portfolio, not to restrict each and every holding in the portfolio to a certain grade. The G20 could help by establishing a committee to identify best practices for risk weighting and regulation at the national level, keeping these concerns in mind.
In addition, more systematic and regular dialogue between rating agencies and government officials in more countries can reduce misunderstanding on both sides. More extensive dialogue between rating agencies and multilateral institutions could prevent the former from inferring that participating in the programmes of the latter is a sign of economic and financial weakness.
Commercial credit ratings are only as good as the data used as inputs, and data on the external debt of governments (and its composition) are imperfect and incomplete. Greater accuracy and transparency of debt statistics are also issues when a sovereign debtor and its diverse creditors meet to negotiate debt restructuring and agree on comparability of treatment (see below). It is equally an issue for the rating agencies, which can be reduced to making guesses in the absence of hard information. Improving debt data is an obvious way of modestly enhancing the rating agencies’ performance.
Complaints from emerging markets that credit ratings are arbitrary and unfair and that investors have a tendency to react strongly (and possibly over-react) are fuelled by rating agencies’ lack of transparency and by their reluctance to acknowledge uncertainty surrounding their judgements. Rating agencies publish the ‘building blocks’ of their methodologies, but only partially (without detailing the ‘qualitative overlay’) and only in general terms. 11 Even if the methodologies of the Big Three rating agencies (Standard & Poor’s, Moody’s and Fitch) are broadly similar, their ratings can differ markedly.
As we show in Appendix B, commercial credit ratings for emerging markets depend heavily on indicators of the quality of institutions and governance, in addition to the data on debt alluded to in the previous paragraph and other macroeconomic indicators. When one looks across the G20 advanced and emerging markets, EMs receive lower ratings even after controlling for a comprehensive set of available debt and macroeconomic indicators. In other words, when one adds a dummy variable for EM status, its coefficient is strongly negative (indicating a lower rating) and highly significant, even after controlling for observed macroeconomic indicators. But when one adds institutional and governance indicators (such as the World Bank’s ‘Worldwide Governance’ and ‘Doing Business’ measures), this differential disappears. Policy makers in emerging markets argue that the weight attached to such indicators is arbitrary and opaque and that the measures in question are of dubious quality. Emerging markets with no history of debt default nonetheless receive lower ratings than their observed debt loads and macroeconomic performance would otherwise lead one to expect.
Addressing these concerns requires more than just improving the debt and macroeconomic statistics used in rating-agency exercises. It requires efforts on the part of multilaterals and others to improve the quality of the institutional/governance measures they produce, while being more transparent about how they produce them. It requires more transparency on the part of rating agencies on exactly how they use the resulting measures.
2.5 Create Hedging Instruments
Many low-income and not a few middle-income countries continue to have no choice but to borrow in foreign currencies (Eichengreen et al., 2022). This exposes them to financial risk and economic dislocation from exchange-rate volatility. Periods like 2022 when the US dollar rose sharply, making it more difficult for such countries to service and repay their dollar-denominated debts, illustrate the point.
Readily available hedging instruments at the relevant maturities and affordable costs would help to mitigate these dangers. Private markets in such hedging instruments exist for only a small number of emerging economies, such as Chile, with relatively well-developed financial markets (see Alfaro et al., 2021). Developing such markets for additional countries and currencies would be a significant step towards reducing financial fragility, although this goal remains out of reach for many emerging markets and most frontier economies.
Entities such as Currency Exchange Fund NV, or TCX, show how such a market could be structured. TCX was established 15 years ago by a group of national governments (German, Dutch, British and Swiss), development finance institutions or DFIs (such as KfW from Germany and JBIC from Japan), and additional donors to provide swaps and forward contracts in emerging-market currencies.
TCX operates mainly in connection with development finance. When a bank or other entity in a developing country borrows from a multilateral development bank or other DFI, that DFI is not in a position to provide a loan in the borrower’s local currency because of the risk to its own balance sheet. TCX provides the DFI with a hedging instrument in the relevant currency and tenor, enabling it to extend a domestic-currency loan. 12 The original maturity of these instruments averages three to five years. Interest rates on those local currency hedges are above interest rates on dollars, adjusted for differential inflation. In other words, the borrower pays a premium for hedging its currency risk but reaps benefits in terms of stability and risk reduction. This pricing model has been sufficient for TCX to break even or even earn a modest positive return in most years. In the absence of other hedging instruments, there is no secondary market to which TCX can transfer its exposures (although on occasion, the Fund has been able to sell portions of its portfolio to other interested parties, typically investment banks). Currency diversification mitigates some of the residual balance sheet risk, but not all. Capital, therefore, has to be sufficient to enable the Fund to meet its commitments in the event of balance sheet losses. Unfortunately, TCX’s capitalisation is limited to $1.1 billion US (as of end-2020). This limited capital limits the hedges that it can provide. As a result, the Fund has a balance sheet of derivatives of only some $5 billion US 13 TCX to date has hedged loans mainly to micro-borrowers and small- and medium-sized enterprises, although its loans for infrastructure and sustainable energy projects have been growing.
TCX’s capital is minute by the scale of the problem. Only four G20 governments are shareholders in the Fund. 14 Most of the hedges provided by TCX go to relatively small private borrowers in frontier markets, not to frontier and emerging-market governments exposed to currency mismatches. A G20 agreement to provide the funding needed for TCX or an equivalent entity at the IFC or elsewhere to scale up significantly would help to address the currency mismatch problem that creates financial fragility. This demonstration effect may then attract commercial banks and other private financial entities to provide currency hedges for this growing market.
2.6 Insert Climate-resilient Debt Clauses into Debt Contracts
Climate change poses special risks to developing countries, especially those that are low-lying and lack the financial resources to invest heavily in resiliency. A climate-change–related disaster can also turn into a financial disaster insofar as such countries find themselves unable to service their debts and see their capital-market access curtailed.
Financial market participants, with support from the international policy community, have made progress in standardising a variety of other contingent clauses and inserting them into debt contracts, thereby beginning to address this incomplete-markets problem. The market in privately-issued catastrophe bonds allows insurers to reinsure against losses from earthquakes, hurricanes and related natural disasters. Last September, Barbados issued the world’s first government bond with a clause allowing payments to be suspended in the event of another global pandemic. In addition, it issued a dollar-denominated global bond in conjunction with a restructuring that included provisions for payments to bondholders to be delayed for up to two years in the event of a specified natural disaster (earthquake, tropical cyclone and excessive rainfall).
The decision of Fitch Ratings to assign a B rating to Barbados’ disaster bond suggests the existence of a market for such issues. The market will be deeper and more liquid, however, and any adverse signal (issuing such a bond may be taken as a sign that a government anticipates lacking the capacity to cope, financially, with the shock) will be less if a broad set of countries, including advanced-country governments, issue such bonds. The G20 countries should be open to including such clauses in their own bilateral, regional and multilateral lending to climate-sensitive low-income countries. They could use regulation to persuade and incentivise private creditors to do likewise. They might subsidise interest premium for such contingent lending through multilateral institutions, in the same manner as some advanced countries have done for catastrophe/pandemic bonds issued by the World Bank.
It is useful that a private sector working group convened by the UK government (PSWG, 2022), including investment banks, legal experts, academic experts and multilateral financial institutions, have agreed on a template or term sheet for such bonds. A standard template will make for a more homogenous, liquid market. It will reduce the transaction cost of issuance. The G20 should encourage and endorse this initiative.
2.7 Create a More Efficient Mechanism for Restructuring Debts
According to the World Bank, as many as 60 per cent of all low-income countries are in or at high risk of debt distress (Ahmed & Brown 2022). (We provide more detail on the financial situation of these countries in Appendix C.) ‘The Common Framework for Debt Treatments’ agreed by the G20 in November 2020 is intended to expedite necessary debt restructurings. This framework was designed to give the Chinese government a seat at the table alongside existing Paris Club bilateral lenders and to ensure that private creditors would provide comparable relief. Yet more than two years later, only three countries, Chad, Ethiopia and Zambia, have applied for relief through the Common Framework. Only one, Chad, has completed the process and actually obtained relief.
The heads of the World Bank and IMF have suggested that distressed debtors seeking relief under the Common Framework should receive statutory protection from asset seizures by national courts when suspending debt service payments. This will help to relieve immediate debt distress and encourage more countries to apply for relief under the framework. But that protection needs to be implemented by creditor-country governments through legislation or executive order. The G20 can adopt a resolution to this effect. More countries can also be encouraged to apply if the Common Framework is extended from low- to middle-income countries such as, for example, Sri Lanka. The IMF can also speed the process by offering impartial, blunt appraisals of exactly how much relief is necessary.
Beyond the immediate need to fix the Common Framework, there is the need to address the increasingly diverse and fragmented nature of the creditor base, which heightens free-rider problems and complicates debt restructuring. To this end, new creditors such as China and India should be admitted as official members of the Paris Club. 15 The Common Framework is supposed to be an alternative to the Paris Club, but it operates on an ad hoc basis and lacks the precedents and secretariat of its long-standing counterpart.
In addition, the G20 should support ongoing efforts to streamline the restructuring process through the adoption of collective representative and collective action clauses in sovereign bond contracts. Most new debt issues by emerging markets and developing countries now include collective action clauses (though legacy debt does not). Two-limb voting clauses have now been used in the recent Ecuador and Argentina restructurings, although the simpler single-limb voting provision in some recent issues has not yet been used (IMF, 2020). However, other instruments such as newly-issued syndicated loans and foreign-law-governed sub-sovereign bonds still do not include CACs; these should be added. Legislation adopted at the national level should make this a requirement.
In addition, more creditor countries can adopt ‘anti-vulture fund’ legislation, along the lines of acts adopted by the United Kingdom, Belgium and France. Doing so will prevent private creditors from holding up renegotiation by rushing to the courthouse (Gill & Buchheit, 2022).
Finally, efforts to restructure problem debts tend to be stymied by less than complete information on who owes what to whom. Not all official creditors provide comprehensive information on their loans, which complicates efforts to agree on burden-sharing (World Bank, 2021). Nor is the problem limited to official creditors. In 2021, the OECD launched a ‘Debt Transparency Initiative’ encouraging private creditors to provide more complete information on their loans and investments (OECD, 2021). Few private creditors have participated so far, however (Neiman, 2022). The G20 governments can make this a regulatory requirement.
3. Conclusion
The G20 finance track has no shortage of problems to address. Few of these problems have simple solutions, but it should still be possible to make progress on them during the Indian G20 presidency. In this article, we have suggested six specific areas where concrete progress is possible. There is no reason why the G20 cannot address these issues simultaneously.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The authors received no financial support for the research, authorship and/or publication of this article.
Global Safety Nets
Economies have been strengthening global financial safety nets through bilateral swap lines (BSLs) and the regional finance agreements (RFAs). Both arrangements have proliferated since the global financial crisis of 2008–2009. In this appendix, we first focus on the bilateral swap lines, and then briefly summarise the developments regarding the RFAs. 16
While BSLs existed prior to the global financial crisis, they gained notice in 2008–2009, when the US Federal Reserve Board offered them to five advanced economies and four emerging market economies. Since then, these have proliferated rapidly. Between 2007 and 2020, 91 bilateral swap lines have been signed around the world; involving 21 countries as their issuers and 43 countries as the recipients.
In 2008, the Federal Reserve Board offered swap lines to the Bank of England, European Central Bank, Bank of Japan, Swiss National Bank and Bank of Canada. It then extended BSLs to nine other countries: Australia, Denmark, South Korea, New Zealand, Norway, Singapore, Sweden, Brazil and Mexico. These additional BSLs expired in 2010 but were reintroduced in response to the COVID-19 pandemic. During the global financial crisis, the ECB also initiated swap lines to Latvia, Hungary, Poland, Denmark and Sweden.
Subsequently, China led the expansion of the global BSL network. China signed six BSLs in 2009, including with Argentina, Indonesia and Malaysia, and rapidly expanded them to 30 countries at end-2019. China is currently the leading provider of these lines by number, followed by the United States (Figure A1). While the United States limits its agreements to advanced economies and select emerging markets, Chinese swap lines are more diverse and have a wider reach. Japan and South Korea have also offered swap lines to multiple countries (Figure A2).
While bilateral swap lines were originally intended to address temporary foreign currency needs, they are now being used to advance a variety of other objectives too. These include: increasing the use of local currency of the issuing country in bilateral trade; serving geopolitical ends; and financing more persistent BOP deficits. In many cases, these swap lines appear to be regarded merely as a form of insurance and are not actually drawn. For example, India has never withdrawn from its ongoing swap line with Japan.
Many advanced economies and emerging markets have been issuing the swap lines and/or have received at least one (see Figure A3 and Table A1). Advanced economies have traditionally offered swap lines to each other (39 of 57 such lines). But nearly a quarter of the swap lines have been signed among the emerging and developing economies, the so-called south–south agreements.
China has, in particular, offered swap lines to emerging markets and developing economies. China’s clients include 9 countries among the G20 countries and more than 20 countries outside the G20.
The use of bilateral swap lines is quite prevalent within the G20 countries (Table A2). Of the 91 swap lines that have existed during 2007–2020, 45 have been between two G20 countries; and in all of the remaining, at least one G20 country has been a counter party.
The global bilateral swap line network is estimated to have been worth US$ 1.9 trillion at end-2020 (Table A3), dominated by the Fed’s permanent standing bilateral swap line network among advanced economies (estimated at US$ 610 billion), and the network of BSLs between Asian countries (estimated at US$ 500 billion).
Alongside bilateral swap agreements, countries have also developed a number of regional financial agreements (RFAs). These include the BRICS Contingent Reserve Arrangement (CRA), Chiang Mai Initiative (CMI), European Stability Mechanism (ESM) and Fondo Latinamericano de Reservas (FLAR).
The various RFAs were created to pool the resources of the respective countries in order to provide initial unconditional financing during the situations of temporary liquidity needs of foreign exchange and to supplement the eventual financing available from the IMF. 19 Some of these regional arrangements also reflected the discomfort and perceived unreliability of the IMF to step in during country-specific or region-wide BOP crises. To some extent, the CRA also came about due to the limited and slow pace of reforms to the IMF voting structure as seen by the emerging economies. 20
As Medhora (2017) points out, FLAR and the CMI were established in response to dissatisfaction with the IMF’s handling of various Latin American debt crises and the 1997–1998 Asian financial crisis.
Determinants of Credit Ratings
In this appendix, we analyse the correlates of credit ratings for the G20 countries.
We follow Griffith-Jones and Kraemer (2021), who observe: ‘Rating agencies derive their ratings applying published methodologies. While the methodologies, as well as the ratings, differ between the three agencies, the main building blocks are the same. They consist of an analysis of: (i) institutional and governance quality; (ii) economic growth and resilience; (iii) public finances; (iv) external accounts; and (v) monetary flexibility’, after which they apply a ‘qualitative’ overlay. The credit committee can revise the indicative scores in either direction based on their subjective assessment. Thus, the final rating outcome results from a combination of ‘objective quantitative and subjective qualitative factors’.
We regress the numerical credit ratings of the G20 countries for 2019 on the following variables: GDP growth, the fiscal deficit (as a percentage of GDP), public debt (as a percentage of GDP), the current account deficit (as a percentage of GDP) and inflation. We also include a dummy which takes a value of 1 for emerging markets and 0 for advanced economies. In additional regressions, we also include (log) per capita GDP.
We use the credit ratings assigned to the G20 countries by the three largest international credit rating agencies: Moody’s, S&P Global and Fitch. We calculate the average annual ratings for each country in two steps. First, we use the concordance provided by Mohapatra et al. (2018) to convert alphabetical ratings into numerical ratings that are comparable across agencies. Those numerical ratings run from 1 to 21. The lowest rating (coded by the Moody’s, S&P and Fitch as Ca, C and C, respectively) is given the value 1; and the highest rating (coded by Moody’s, S&P and Fitch as Aaa, AAA and AAA, respectively) is given the value 21. Second, we take the average of the monthly ratings for each country. 22
Average ratings differ significantly between advanced economics and emerging markets. While the average rating of an AE is 19, the emerging-market average is 7.5 points lower (11.6, viz., barely above the junk grade, which starts at 11). The low average rating for emerging markets is affected by Argentina’s 3.6 rating, but even when Argentina is excluded the emerging-market average is 12.6, viz., just 1.5 points above junk. Hence, it may take just one downgrade for an emerging market to fall from the investment grade to the junk grade.
Regression results are in Table B1. Most coefficients have their expected signs. Faster-growing countries have higher ratings. Countries with a higher ratio of public debt to GDP get a lower credit rating; while countries with higher inflation also have lower ratings.
But even after controlling for these determinants of credit ratings, the advanced economy differential persists: the emerging-market dummy still has a negative and significant coefficient. Only when we also include per capita income, which of course differs systematically between advanced and emerging markets, does that coefficient decline in size and become less significant.
For robustness, we dropped Argentina and repeated the regressions for 2021. The results are very similar.
Figures B1–B5 juxtapose sovereign ratings against current account deficits, fiscal deficits, public debt, inflation and GDP growth. The figures convey the same impression as Table B1. They indicate that the advanced economies have higher ratings than emerging markets, despite the fact that many of them have higher fiscal deficits, public debts, inflation and current account deficits and lower GDP growth.
We now add governance indicators to the analysis. 23 The dataset includes government effectiveness, or perceptions of the quality of public services, civil service and its degree of independence from political pressures, the implementation of policies and their quality, and the government’s commitment to these policies; regulatory quality, or perceptions of the government’s ability to implement sound policies and regulations that permit and promote development of the private sector; political stability and the absence of violence, or perceptions of the likelihood of politically motivated violence, and political instability (including terrorism); and control of corruption, or perceptions of the extent to which private gains motivate the exercising of public power and also the degree to which states are captured by elite and private interests.
Each of these indicators ranges from −2.5 to 2.5, with higher values corresponding to stronger/better governance. We also calculate an average governance indicator, by taking a simple average of these four indicators.
Table B2 reports the results when we include the average value of the governance indicator. The results are similar to before, but now the governance variable enters positively and significantly, while the EM dummy no longer differs from zero.
Debts of Low-income Countries
In 2005, the IMF and World Bank jointly developed a framework to conduct debt sustainability assessments (DSAs) of low-income countries (LICs). 24 The latest assessment of the 70 LICs, published in November 2022, identified 10 of these countries as in debt distress, 27 at high risk of distress, 25 at moderate risk of debt distress and 7 at low risk of debt distress. 25
Using the IMF’s Fiscal Monitor and World Bank’s International Debt Statistics database, we calculate the average total government debt and government debt raised externally, both expressed as a percentage of GDP, across countries at different levels of debt distress. Table C1 shows that the debt burden of countries in distress was roughly 90 per cent of GDP in 2019, and further increased by almost 10 percentage points of GDP by 2021. The debt level of countries at high risk of debt distress was 48 per cent of GDP in 2019, but increased by 10 percentage points of GDP, to an average of 58 per cent by 2021.
The table also shows that countries in debt distress have raised 40 per cent of their total public debt externally; countries at high risk of distress have raised a larger fraction of debt externally; countries at moderate risk have raised more than 60 per cent of their debt externally; and countries at low risk have raised half of their debt externally.
We decompose the externally raised general government debt stock into bilateral debt (which we further disaggregate into debt owed to the G20 countries and to the non-G20 countries, and to countries in the G20 that are members of the Paris Club and those that are not); debt owed to the multilateral agencies; and debt raised from private creditors. In 2019, external public debt raised by LICs totalled $340 billion. Of this, about half of the debt was owed to the multilateral institutions, 35 per cent to bilateral creditors and the rest to the private sector.
A large proportion of this bilateral debt has been extended by the G20 countries (amounting to $102 billion in 2019, which further increased to $123 billion in 2021). China has been by far the largest creditor, accounting for half of all the bilateral debt accruing to the G20 countries (see Figures C1 and C2). In all, more than half of all bilateral debt is extended by the G20 countries not in the Paris Club (see Table C4).
Acknowledgements
The authors thank Prof Sanket Mohapatra for sharing the data on credit ratings, and Ayesha Ahmed, Aakansha Atal, S. Priyadarshini and Sakshi Rathee for valuable research assistance.