Closing the Global Crisis Finance Gap

Closing the Global Crisis Finance Gap Why and How the IMF Should Address Weaknesses in the Global Financial Safety Net EXECUTIVE SUMMARY Many low- and middle-income countries are struggling with deteriorating debt sustainability and external shocks, including the economic repercussions of the COVID-19 pandemic, Russia’s war in Ukraine and the tightening of monetary policy in advanced economies. Amid these developments, it is crucial to explore whether the world – and its most vulnerable countries – is well prepared to buffer balance of payments difficulties caused by systemic shocks. This policy brief explores the status of the Global Financial Safety Net (GFSN), the set of institu-tions and arrangements that provide external short-term finance to prevent or backstop a financial crisis, including the IMF, Regional Financial Arrangements (RFAs) and bilateral currency swaps. We demonstrate how structural inequalities in the GFSN could be reduced by reforming facilities and funding of the International Monetary Fund (IMF). Since the IMF is the element of the GFSN that is accessible for all countries, the IMF lending policy and lending capacity is key to ensuring equitable access to global crisis finance. By considering that high-income countries (HICs) have higher relative access to GFSN resources (in comparison to their gross domestic product (GDP)) through swaps and RFAs than low- and middle-income countries, the IMF could offset inequalities by increasing lending capacity to low-income countries (LICs) as well as middleincome countries (MICs). But for this to happen, IMF funding needs to expand. This year presents a unique opportunity for the IMF to expand its funding resources and increase its lending capacity to LICs and MICs. The IMF is undergoing its 16th General Review of Quotas, a process undertaken every five years where the IMF assesses whether member quotas are adequate to cover their balance of payments financing needs and the sufficiency of IMF funding. Subsequently, the IMF Board of Governors may decide to increase quotas, which are the primary and predictable source of funding to the IMF. This policy brief provides two possible estimations to assess how much additional funding the IMF would need to address the inequalities of the GFSN. First, we calculate the necessary amount to equalize access to crisis finance in the GFSN so that LICs and MICs have the same relative crisis finance volume as HICs compared to their GDP. Second, we calculate low- and middle-income coun-tries’ short-term gross external financing needs (GEFN) and the required crisis finance volume that would cover their GEFN. We find that IMF quotas of LICs and MICs would need to, at the very least, double to address the current structural gaps in the GFSN and at least triple to meet their GEFN in case of a systemic shock. Key findings: • IMF lending capacity should increase by $550 billion (or 127 percent of quotas) if it aims to lift access to external crisis finance for LICs and MICs in the GFSN to the same relative volume that is accessible for HICs. • $8 billion is needed to equalize relative GFSN access for LICs, $297 billion for lower middle-income countries (LMICs) and $244 billion for upper middle-income countries (UMICs), excluding China. • IMF lending capacity should increase by $1.16 trillion (or 267 percent of quotas) if it aims to cover the short-term GEFN of LICs and MICs. • $25 billion is needed for LICs, $433 billion for LMICs and $697 billion for UMICs, excluding China. Policy recommendations: A fit for purpose GFSN not only requires a higher lending capacity for low- and lower middle-income countries and a larger size in general, but also a fundamental change in IMF policies. To that end, we recommend that the IMF: 1. Increases the access limit of unconditional facilities for LICs and MICs, which are key to creating a level playing field in the GFSN across different country income groups and making the GFSN better prepared to deal with future systemic crises. LICs and LMICs could gain access to timely, unconditional, voluminous balance of payments finance similar to what had been given through the Rapid Financing Instrument (RFI) and the The Rapid Credit Facility (RCF) during the COVID-19 pandemic. A comprehensive revision of IMF facilities should consider the preference of LICs and MICs to respond to temporary illiquidity with unconditional IMF financing. These reforms are key to creating a level playing field in the GFSN across different country income groups and making the GFSN better prepared to deal with future systemic crises. 2. Considers the short-term GEFN of its member countries and the structural inequalities of the GFSN as a relevant baseline for the required lending capacity. If the IMF aimed to cover the GEFN of the most vulnerable countries through increasing quotas, they would need to increase by $1.16 trillion (or 267 percent of the current quota resources). IMF lending capacity should increase by $550 billion (or 127 percent of quotas) if it aims to lift access to external crisis finance for LICs and MICs in the GFSN to the same relative volume that is accessible for HICs. These estimates exclude possibly required increases of quotas related to GEFN of HICs. 3. Improves its funding resources to better support its operation. Increasing IMF resources through quotas, although preferred by some member states, is not the only option. The IMF Board may also consider increasing the New Arrangements to Borrow (NABs) and making it a perpetual (not renewable) source of funding for the IMF. As the odds of facing another systemic crisis like COVID-19 are not unlikely, the GFSN must be made fit for purpose, which means creating sufficient resources and adequate policies to support the most vulnerable countries. Moreover, the looming sovereign debt crisis in the Global South (Ramos et al. 2023) highlights the importance of addressing the deficiencies of the GFSN. The 16th General Review of Quotas, currently underway and slated to conclude in December 2023, is an opportunity to strengthen the IMF’s funding to a scale sufficient to cover the needs of countries that are underserved by the GFSN. INTRODUCTION The International Monetary Fund (IMF) is currently undergoing its 16th General Review of Quotas, a process undertaken every five years when the IMF assesses the adequacy of its quota-based funding resources. Quotas determine the IMF’s financial and governance structure as they are the basis for each member’s voting power, access limits, share of Special Drawing Rights (SDRs) allocation, as well as the maximum amount of financial resources that a member is obliged to provide to the IMF (IMF 2022a). Quota resources are also the main source of IMF funding with approximately $631 billion, followed by multilateral borrowing with $478 billion (New Arrangements to Borrow, NABs) and bilateral borrowing agreements (BBAs) accounting for $184 billion in 2022. In December 2023, following the 16th General Review of Quotas, the IMF Board of Governors may decide to increase quotas, which could concomitantly boost the IMF’s lending capacity. At the same time, many low- and middle-income countries (LICs and MICs) are struggling with deteriorating debt sustainability and external shocks, including the economic repercussions of the COVID-19 pandemic, Russia’s war in Ukraine and the tightening of monetary policy in developed countries. Since the COVID-19 pandemic, political and scholarly attention has increasingly focused on crisis finance and whether the world is well prepared to buffer balance of payments difficulties caused by systemic shocks (Woods 2022). In a context described by some as “polycrisis,” where shocks can be compounded and can emerge from several sources – such as climate, finance and health (Tooze 2022) – it is crucial to explore the suitability of the Global Financial Safety Net (GFSN)—the set of institutions and arrangements that provide external short-term finance to prevent or backstop a financial crisis. Until the end of the Bretton Woods era in the 1970s, the IMF was the only element of the GFSN. Regional Financial Arrangements (RFAs) were then established in several parts of the world, partly to substitute the IMF and partly to complement existing crisis finance sources. Recently, the GFSN has further diversified as central banks engage in the provision of bilateral currency swaps.1 Apart Literature on the GFSN also commonly refers to foreign exchange reserves as a fourth component. Here, we do not include foreign exchange reserves as a GFSN component for two reasons. First, standard economic literature shows that in all types of balance of payments crises in emerging markets and developing countries, the third-party actor must come from outside the country and provide timely, voluminous liquidity with smoothly decided conditionality for crisis prevention and backstop (Obstfeld 1996). Second, we assume that the crisis prevention and backstop capacity of foreign exchange reserves only hold when they are maintained; otherwise, they will be rapidly depleted. Empirical investigations into the role of foreign exchange reserves as crisis prevention and backstop show that the intervention of an external third-party actor is indispensable to backstop a financial crisis. In fact, regional funds and currency swaps dominate national foreign exchange reserve accumulation as a precautionary measure (Aizenman 2010; Aizenman, Jinjarak and Park 2011). 1 from the diversification of instruments, the GFSN’s lending capacity has also increased. Before the COVID-19 crisis, the GFSN’s lending capacity was around $3.5 trillion (Mühlich, Fritz and Kring 2021). Approximately $1 trillion is available at the IMF, $1 trillion is in regional liquidity pools and at least $1.5 trillion has been offered by central banks (Kring et al. 2022). However, this lending capacity is not equally available to all countries, and lower-income countries have substantially less access to the GFSN (Zucker-Marques, Mühlich and Fritz 2022). Since the IMF is the single largest component of the GFSN—both in terms of lending capacity and membership—the debate on the adequacy of the GFSN is often focused on the Fund’s lending capacity (see also IMF 2021a). The GFSN provides small lending capacity to LICs and MICs compared to the lending capacity it provides to high-income countries (HICs) as share of their respective gross domestic product (GDP). More specifically, while the combined access to the GFSN for HICs is about 7.8 percent of their GDP, LICs have access to 6.1 percent of their GDP in the GFSN, LMICs have 4.8 percent and UMICs have 5.2 percent. To rebalance the GFSN and make it fit for purpose, the IMF needs to extend access to unconditional facilities and strengthen its quota-based resources to ensure it offers appropriate lending and is sufficiently and reliably resourced to address potential future systemic crises. The IMF’s 16th General Review of Quotas has sparked a debate around the adequacy of the quota formula and ways to address the Fund’s resource requirements, as well as reform proposals aimed at boosting the representation of LICs and MICs. On the one hand, these proposals seek to improve access to crisis finance through quota formula reforms, while on the other hand, they aim to introduce redistributive elements into the quota system. As the IMF’s Managing Director Kristalina Georgieva said at the Group of 20 (G20) Finance Ministers meeting in February 2023, “[…] a successful quota review—which the IMF’s membership has committed to complete by December 2023—is critical for a strong Global Financial Safety Net” (IMF 2023b). At the same time, India as the current G20 president has made clear that quota reform needs to deliver on a more up-to-date quota formula and redistribution (Economic Times 2022). Similarly, the Intergovernmental Group of 24 (G24) has made numerous suggestions for creating a more balanced quota contribution (see, for example, G24 2023). Such measures would further contribute to giving LICs and MICs similar means of crisis finance as HICs in the GFSN, provided that the IMF also permanently opens access to unconditional lending facilities for LICs and MICs. This policy brief takes stock of the status of the GFSN and demonstrates how structural inequalities in the GFSN could be reduced by reforming IMF facilities and funding. We provide two possible estimations to assess how much additional funding the IMF would need to address the inequalities of the GFSN. First, we calculate the necessary amount to equalize access to crisis finance in the GFSN so that LICs and MICs have the same relative crisis finance volume as HICs. This would require an additional $550 billion ($8 billion to equalize relative GFSN access for LICs, $297 billion for lower middle-income countries (LMICs) and $244 billion for upper middle-income countries (UMICs)). Second, we calculate LICs’ and MICs’ short-term gross external financing needs (GEFN) and the required crisis finance volume that would cover their GEFN. The GEFN of LICs and MICs are almost twice as much as they could borrow from the GFSN in case of a financial crisis. In 2021, LICs’ GEFN were about 11.7 percent of their GDP, LMICs’ were 9.1 percent and UMICs’ were 12.5 percent. Since HICs can cover their GEFN with other GFSN elements—primarily bilateral central bank currency swaps—additional resources are needed to augment access to the GFSN for LICs and MICs. We find that covering the GEFN of LICs and MICs would require an additional $1.16 trillion ($25 billion for LICs, $433 billion for LMICs and $697 billion for UMICs). In sum, if the IMF aimed at positioning itself as a strong global element of the GFSN, quotas of LICs, LMICs and UMICs would need to, at the very least, double to address the current structural gaps in the GFSN and at least triple to meet their GEFN requirements in case of a systemic shock. Covering these resource requirements by an IMF quota increase would require all members to increase their contribution, including LICs and MICs. But according to the IMF Articles of Agreement,2 unless prescribed differently by the Board, when quotas are changed, 25 percent of payment are required to be in hard currency or SDRs and the remaining 75 percent in local currency. Hence, a quota increase that would cover these resource requirements could provide LICs and MICs with leverage for their external resources for crisis finance in case the IMF opens access limits to its unconditional lending facilities. We recommend that the IMF increases its lending capacity to cover the needs of LICs and MICs that do not have access to the same external crisis finance resources in the IMF, RFAs and swaps as HICs and consider three additional fundamental policy changes, including (i) increasing the access limit of unconditional facilities for LICs and MICs, which are key to creating a level playing field in the GFSN across different country income groups and making the GFSN better prepared to deal with future systemic crises3; (ii) considering the short-term gross external financing needs of its member countries and the structural inequalities of the GFSN as a relevant baseline for the required lending capacity and (iii) improving its funding resources to better support its operation. Although some policymakers have expressed preferences in increasing IMF funding through quotas (see Merling and Kring 2023), it is not the only option to support higher lending capacity to LICs and MICs. For instance, considering that some countries have voluminous crisis finance sources in the GFSN (like access to unlimited swaps from the US Federal Reserve (Fed)), the contribution of these countries to the IMF could be allocated for LIC and MIC lending purposes. Although Portugal, Greece, Ireland and Iceland accessed IMF lines after the 2008 global financial crisis, the largest economies in Europe have not accessed IMF lines for more than 40 years (the latest was the UK in the 1980s and Spain in 1978) (Ocampo 2017). Accounting for the little demand that advanced economies have for IMF resources, the Fund could reallocate resources to fund Poverty Reduction and Growth Trust members or provide unconditional lending to developing countries. Alternatively, advanced countries could increase and make permanent the NABs, which could give the IMF a reliable source of funding apart from being a direct support to LICs and MICs. Finally, the IMF could issue new SDRs and preferably distribute them in a manner that is independent from the quota systems but respects the GEFN of LICs and MICs. As of December 2022, 40 percent (32 countries) of LICs and LMICs used half or more of their accumulated SDR allocation. LICs and LMICs used at least one-third of their SDRs on average. This contrasts with the pattern of HICs, which are net buyers of SDRs and hold 102 percent of their allocation. In fact, only 5 percent of HICs used more than 50 percent of their SDRs. SDR use underscores the findings of Zucker-Marques, Mühlich and Fritz (2023), who find that countries in balance of payments distress due to external crisis events demand unconditional resources. Increasing IMF resources therefore needs to be combined with access to unconditional lending for LICs and MICs to be fully utilized. Article III, part III. The most recent temporary increase in access limits to conditional lending facilities that the IMF board introduced on March 6, 2023 (IMF 2023a) points to the right direction. During the next 12 months, the difference in the access that LICs and MICs have to the IMF will thereby diminish. More specifically, while the combined access to the GFSN for HICs is about 8.6 percent of their GDP, LICs have access to 7.7 percent of their GDP in the GFSN, LMICs 5.7 percent and UMICs 6.1 percent. Meanwhile, the unconditional access to IMF lending facilities that was the key element in response to the COVID-19 pandemic for LICs and MICs is not covered by this increase in access limits. Their elevated access limits will end by mid-2023. The gap between lending capacity provided in the GFSN and the short-term GEFN will diminish accordingly. 2 3 There are a few caveats. First, the calculations are based on retrospective data and do not account for the rapidly changing global economic conditions for LICs and MICs. Second, this policy brief does not undertake any ventures into different crisis scenarios but assumes a systemic shock that would hit all LICs and MICs at the same time and with the same magnitude. Third, we assume the IMF is the lender of last resort in the GFSN. In that sense, we estimate the funding needs for the case of IMF financing and do not account for possible additional crisis finance from RFAs or currency swaps. INEQUALITIES OF THE GFSN Structural Inequalities The GFSN is structurally unequal, as LICs and lower-middle-income countries (LMICs) have less access to resources to backstop temporary illiquidity (see Kring et al. 2022; Mühlich, Fritz and Kring 2021; Mühlich et al. 2020), which affects their resilience to effectively handle balance of payments problems. Figure 1 shows the access to the GFSN components by income group, demonstrating that the lower the income group, the lower the access to GFSN volumes (as weighted share of GDP). Figure 1: GFSN Lending Capacity by Income Groups in USD Million and as Weighted Share of GDP (%) 5,000,000 4.0% 4,500,000 3.5% 4,000,000 3.0% 3,500,000 2.5% 3,000,000 2,500,000 2.0% 2,000,000 1.5% 1,500,000 1.0% 1,000,000 0.5% 500,000 0.0% 2018 2019 2020 2021 Low income 2018 2019 2020 2021 Lower middle income 2018 2019 2020 2021 Upper middle income 2018 2019 2020 2021 High income IMF conditional IMF unconditional Limited Swaps (incl. China) Unlimited Swaps (excl. US) RFA GFSN/GDP % weighted by country GDP in global GDP Source: Authors’ own calculations based on the Global Financial Safety Net Tracker (Kring et al. 2022) and World Bank World Development Indicators. Notes: Left-hand scale: Lending capacity of the GFSN for three years by income group in USD millions. Right-hand scale: Lending capacity of the GFSN for three years as weighted share of GDP Unconditional IMF, includes cumulative access limits of the Rapid Financing Instrument, the Rapid Credit Facility, Flexible Credit Line, Precautionary and Liquidity Line and the Short-term Liquidity Line. Conditional IMF includes cumulative access limits of the Catastrophe Containment and Relief, the Stand-by Arrangement, the Extended Fund Facility and the Extended Credit Facility as of the respective year. Swaps are divided into limited and finite swaps (dark blue) and unlimited and infinite swaps (light blue). Currency swaps: Maximum amount of active currency swaps by country. Country income groups are based on 2020 World Bank classification. The inequality of the GFSN not only includes volume of resources but also quality of access and quantity of options. To analyze the inequalities in the GFSN more systematically, Zucker-Marques, Mühlich and Fritz (2023) developed a composite index that enables comparison of various GFSN instruments (IMF arrangements, RFAs and swaps) based on their volume and quality. The index also enables an assessment of the countries that hold a stronger position in the GFSN regarding their access to better more diverse instruments. Several economies, including eurozone members, Switzerland, Canada and Japan have access to the best short-term liquidity finance—swap agreements from the Fed—which are denominated in most liquid currency (the US dollar), are unlimited and permanent and come without policy conditionalities or market stigma. These economies also have access to several swap agreements from central banks around the globe, as well as IMF arrangements (Zucker-Marques, Mühlich and Fritz 2023). In contrast, the vast majority of LICs are excluded from almost all GFSN instruments except IMF lending. These countries had temporarily elevated access to unconditional IMF lending during the COVID-19 pandemic, as they lack access to any kind of swap, either from the Fed, the People’s Bank of China or from other central banks. In addition, most of the countries are not a member of a regional fund. This represents a twofold exclusion from the GFSN—geographically (nonexistence of a regional fund) and structurally (no access to bilateral central bank currency swaps)—that makes those LICs at the bottom of the GFSN hierarchy completely dependent on IMF conditional crisis finance (Zucker-Marques, Mühlich and Fritz 2023). Based on economic concepts of balance of payments crisis resolution (Obstfeld 1996), the index developed by Zucker-Marques, Mühlich and Fritz (2023) comprises criteria such as relative volume, currency of contract, conditionality, limitedness, predictability of funds and market stigma. Based on the data provided in the Global Financial Safety Net Tracker as of December 2021, Figure 2 shows Figure 2: The Composite GFSN Index: Aggregated Scoring by Country, 2021 7 50.96 points Russia 19.31 Canada 42.65 China 49.91 Japan 50.96 India 17.00 Brazil 17.00 Zambia 7.00 Australia 20.10 Source: Zucker-Marques, Mühlich and Fritz 2023. the accumulative index values per country, revealing that HICs in the North America and Europe have access to many GFSN instruments (as shown by Figure 1) as well as higher-quality options and a wider variety of them. On average, HICs in these regions reach an index of 35 points, while LICs and LMICs in sub-Saharan Africa, Latin America and Central Asia have less than half as many index points. A key finding of the GFSN index is the role of the IMF as an equalizer of the GFSN during the COVID19 crisis, in terms of providing better quality instruments to LICs and LMICs (Zucker-Marques, Mühlich and Fritz 2023). Although these country groups often lack access to better-quality and high-volume GFSN instruments, the IMF’s decision to increase access limits in the Rapid Financing Instrument (RFI) and the Rapid Credit Facility (RCF) proved instrumental in reshaping the geography of the GFSN during the pandemic. In essence, according to the composite GFSN index, while access to the GFSN remained highly unequal despite the IMF’s efforts, the reform of unconditional IMF facilities resulted in a less hierarchical distribution of the GFSN lending capacity (Zucker-Marques, Mühlich and Fritz 2023). IMF Pandemic Response: Increased Access to Unconditional IMF Facilities for LICs and MICs In response to the COVID-19 pandemic, the IMF has shown that it is feasible to expand the crisis prevention and backstop capacity of the most vulnerable countries. In 2020, it temporarily raised access levels to the RFI and the RCF (IMF 2022a, 2021b, 2021c). Normal access was lifted from 50 percent of a country’s quota per disbursement to up to 100 percent until the end of 2021. The IMF’s goal was to quickly allow easy access to liquidity for member countries that could not access unconditional lending through facilities that require prequalification, such as the Flexible Liquidity Line or the Precautionary Credit Line. As Mühlich, Fritz and Kring (2021) show, it was the reformed unconditional facilities that were in use by LICs, LMICs and UMICs during the COVID-19 pandemic, as well as by countries that had previously circumvented the use of IMF programs (see Mühlich and Fritz 2018). Due to the increase in IMF lending limits, we estimate that GFSN lending capacity for LICs has risen by 4.5 percentage points to reach 10.6 percent of GDP, while for LMICs, the lending capacity has increased by 2.5 percentage points, reaching 7.3 percent of GDP. For UMICs, it increased by 2.2 percentage points to reach 7.4 percent of GDP. HICs could access 10 percent of GDP during that time, an increase of 2.2 percentage points. The IMF’s response to the COVID-19 pandemic offers valuable lessons on how to improve the GFSN, which should be considered before returning to the previous access limits, as the IMF plans to do by mid-2023. Access to the IMF’s unconditional facilities is key for the most vulnerable countries: for 80 developing countries, the IMF is the only source of a short-term financing instrument as of December 2021 (Zucker-Marques, Mühlich and Fritz 2023). For most LICs and LMICs, the provision of a timely, voluminous and unconditional crisis finance depends entirely on the IMF (when these lines are available to them, like during the COVID-19 crisis). In contrast, HICs respond to temporary illiquidity by using other GFSN instruments like central bank currency swaps. For this reason, the probability of them requesting IMF resources is comparatively low, so the IMF could consider providing LICs and LMICs relatively larger access to unconditional facilities. The next section discusses two ways to estimate potential additional resources requirements of the IMF to increase the unconditional lending limits for LICs and MICs. ESTIMATING QUOTA INCREASES Increasing Quotas to Elevate IMF Lending Capacity for LICs and MICs If the IMF aims to position itself in the GFSN and address its weaknesses (IMF 2021a), then both GFSN inequalities and its overall size and lending capacity should be an issue for the Fund as the global element of the GFSN. Assessing the adequate size of the GFSN is not an easy undertaking, and it is an inherently uncertain exercise. In 2016, IMF staff suggested three possible approaches to such estimation: metric based (considering GDP, trade, GEFN, etc.), access based (considering demand for funds) and considering historical crisis and scenarios involving systemic shocks (IMF 2016, p. 20). Based on the IMF’s estimation, an additional 536 billion SDRs (equivalent to 112 percent of a quota increase) would be required to restore the ratio of quotas to relevant economic indicators to the average levels observed in previous reviews. However, the IMF (2021a) highlighted that even more resources would be needed in case it is willing to address the remaining weakness of the GFSN and the international monetary system. Bearing this in mind, we provide an analysis on the potential required increases in quotas for the IMF to act as an equalizer of the GFSN’s inequalities, which Mühlich et al. (2020), Mühlich, Fritz and Kring (2021) and Zucker-Marques, Mühlich and Fritz (2022) have identified is a crucial weakness of the GFSN. Our estimations are based on GFSN lending capacity as share of GDP and on GEFN. As mentioned, quotas are not the only source of IMF funding, as the Fund also relies on NABs from a group of members and institutions and BBAs. Although both are understood as complementary resources as of July 2022, combined, they have in fact been larger than quotas. This is particularly challenging since NABs and BBAs are temporary sources of financing whose renewal from 2023 or 2024 onward is pending (IMF 2022b). Therefore, we refer only to the required increase in quotas as stable and long-term funding for the IMF. As Figure 3 shows, not all resources collected by the IMF are fully translated into lending capacity. In the case of quotas, a key reason for this gap is the currency denomination of IMF quotas. As per IMF regulations, countries must fund at least 25 percent of their quota with “usable currencies,” which are currencies that compose the SDR basket. Figure 3: IMF Total Resources and Lending Capacity, in USD billion, as of June 2022 Lending Capacity 422 Resources 377 144 631 - 200 Total lending capacity $944 478 400 Quotas Source: IMF 2022b and IMF Exchange Rate Archives. Note: SDR per dollar exchange rate is 0.76 as per June 2022. 600 800 Multilateral Borrowing 184 1,000 Bilateral Borrowing 1,200 Total resources $1,293 1,400 The remaining 75 percent may be denominated in the member’s own currency, which may not be used to fund IMF lending. For countries that issue currencies with a high market liquidity like the US and Euro area countries, it means that 100 percent of their quota is paid in “usable currencies.” As a result, IMF effective lending capacity is not exactly the same as total resources (Morrison 1998). Estimating Required Lending Capacity and Quota Increases Q U OTA I N C R E A S E S TO L E VE L OUT GFSN L E NDI NG CAPACI T Y AMONG I NCOME GROUP S Another alternative to estimate the increase of IMF quotas is to compare LICs’ and MICs’ access volume in the GFSN to the access volume of HICs. To do this, we use the Global Financial Safety Net Tracker (Kring et al. 2022), GDP as of 2021 and World Bank country group classification as of 2021. As mentioned, the IMF announced returning to pre-pandemic access limits by the end of 2021 (IMF 2021c). Assuming their access to RFAs and exclusion from bilateral central bank currency swaps remained the same, after 2021, LICs find a lending capacity in the GFSN of 6.1 percent of GDP, LMICs of 4.8 percent, UMICs of 5.2 percent and HICs of 7.8 percent (Figure 4). Based on these data, assuming that access to, and the lending capacity of, all other GFSN elements remained the same, the sum of GFSN lending capacity for LICs would need to increase by about $8 billion to reach a level of GFSN/GDP comparable to the lending capacity of HICs which is 7.8 percent of GDP. By the same logic, the GFSN lending capacity for LMICs would need to increase by $297 billion and for UMICs excluding China by $244 billion. Together, this would add up to an additional lending capacity of up to $550 billion that would need to be mobilized in the GFSN to provide LICs, LMICs and UMICs (excluding China) with a similar lending capacity as HICs. As of February 2023, the IMF’s forward commitment capacity (FCC, the Fund’s quota-based lending capacity within the coming 12 months) accounts for $214 billion, which is less than half of the resources that Figure 4: GFSN Lending Capacity for LICs and MICs, Gap Compared to HICs, 2021, % of GDP Source: Authors’ calculations. Note: China is excluded from the figure. Lending capacity of the GFSN for three years by income group Unconditional IMF, includes cumulative access limits of the Rapid Financing Instrument, the Rapid Credit Facility, Flexible Credit Line, Precautionary and Liquidity Line and the Short-term Liquidity Line. Conditional IMF includes cumulative access limits of the Catastrophe Containment and Relief, the Stand-by Arrangement, the Extended Fund Facility and the Extended Credit Facility as of the respective year. Country income groups are based on 2020 World Bank classification. IMF access to unconditional facilities returned to 50 percent of quota at the end of 2021, while IMF access to conditional facilities returned to 145 percent of quota at the end of 2021. we estimate should be available for these income groups, in addition to existing lending capacity if a systemic shock occurs. Based on the perspective of GFSN coverage by income group, the IMF would need to increase the quotas of LICs, LMICs and UMICs by at least 127 percent if the new resources were to cover the potential needs of LICs and MICs. While the lending capacity of external providers of crisis finance is relevant to determining the resilience of countries to external shocks, mere accessibility of lending capacity does not show much about a country’s actual short-term financing needs. For this, we need to consider the GEFN for a country to sustain its economic activities in the short term. QUOTA INCREASES TO MEET THE GEFN OF THE MOST VULNERABLE COUNTRIES To approximate the unresolved question of what sufficient crisis finance would look like, we suggest making use of the GEFN, a measure used by market analysts as well as the IMF and the World Bank to assess a country’s liquidity needs and debt profile vulnerabilities (Bloomberg Intelligence 2018; IMF 2016, 2018). The GEFN of a given year is composed of a country’s current account deficit (or surplus), debt service of public and private borrowers (public and publicly guaranteed and private nonguaranteed) and the short-term debt stock (due within 12 months). In this sense, the GEFN measures a country’s dependence on external financing to sustain its economic activities during the year. We have compiled data from the World Bank International Debt Statistics (IDS) and the IMF’s balance of payments to estimate the GEFN of LICs and MICs.4 Figure 5 shows that in 2021, the GEFN for LICs accounted for 11.7 percent of their GDP, LMICs 9.1 percent and UMICs (excluding China) 12.6 percent. Figure 5 shows that for all income groups, the GEFN is higher than the lending capacity Figure 5: GEFN and GFSN Lending Capacity, LICs and MICs, 2021 Source: Authors’ elaboration based on IMF balance of payments, World Bank IDS, IMF finances and Global Financial Safety Net Tracker (Kring et al. 2022). Note: China is excluded from the figure. Lending capacity of the GFSN for three years by income group Unconditional IMF, includes cumulative access limits of the Rapid Financing Instrument, the Rapid Credit Facility, Flexible Credit Line, Precautionary and Liquidity Line and the Short-term Liquidity Line. Conditional IMF includes cumulative access limits of the Catastrophe Containment and Relief, the Stand-by Arrangement, the Extended Fund Facility and the Extended Credit Facility as of the respective year. Country income groups are based on 2020 World Bank classification. 4 Information on GEFN is not available for all countries. Of the 133 IMF members that are LICs or MICs, data are available for 95 countries. offered by the GFSN. Hence, GEFN is, on average, around 11 percent of the economy for each of these country income groups. In 2021, the GFSN lending capacity (i.e., a country’s annual access limit to IMF unconditional and conditional lending facilities; to RFA lending, if it is a member, and its active central bank currency swaps, if any, was elevated to 8.5 percent of GDP for LICs, LMICs and UMICs on average, not least due to the elevated access limits that the IMF had provided during the COVID19 pandemic. Yet, GFSN lending capacity decreased to 5.4 percent of GDP on average after the return to pre-pandemic access limits at the end of 2021 (see previous section). Under these circumstances, a tightening of global economic conditions—for example, through further rising interest rates in key currency economies and a corresponding reduction of market access for short-term refinancing for developing countries—could make it difficult for these countries to service their external obligations. If the IMF were to bridge the gap between the GEFN and the access to short-term emergency finance, it would need to increase the availability of finance to LICs in the amount of 5.6 percent of their GDP, to LMICs 4.3 percent and for UMICs (excluding China) 7.5 percent. Altogether, the GFSN lending capacity would need to increase by $1.16 trillion to cover the GEFN of the most vulnerable countries. Considering that total IMF quotas currently account for about $650 billion (477 billion SDR), which translates to a lending capacity from quotas of $431 billion (319 billion SDR), the IMF should increase its quotas by at least 267 percent if the new resources are to cover the potential GEFN of LICs and MICs. CONCLUSION For LICs and MICs, the GFSN is not fit for purpose if another external shock like the COVID-19 pandemic were to occur. As long as access to central bank currency swaps is not coordinated in the GFSN, and as long as access to RFAs remains geographically uneven, we recommend that the IMF take steps to rebalance the GFSN by bridging gaps in short-term crisis finance for the most vulnerable countries. Ahead of the 16th General Review of Quotas at the IMF, we therefore recommend that the IMF take the following actions: 1. Increase the access limit of unconditional facilities for LICs and MICs, which are key to creating a level playing field in the GFSN across different country income groups and making the GFSN better prepared to deal with future systemic crises. LICs and LMICs could gain access to timely, unconditional, voluminous balance of payments finance similar to what had been given through the RFI and the RCF during the COVID-19 pandemic. A comprehensive revision of IMF facilities should consider the preference of LICs and MICs to respond to temporary illiquidity with unconditional IMF financing. These reforms are key to creating a level playing field in the GFSN across different country income groups and making the GFSN better prepared to deal with future systemic crises. 2. Consider the short-term GEFN of its member countries and the structural inequalities of the GFSN as a relevant baseline for the required lending capacity. If the IMF aimed to cover the GEFN of the most vulnerable countries through increasing quotas, they would need to increase by $1.16 trillion (or 267 percent of the current quota resources). IMF lending capacity should increase by $550 billion (or 127 percent of quotas) if it aims to lift access to external crisis finance for LICs and MICs in the GFSN to the same relative volume that is accessible for HICs. These estimates exclude possibly required increases of quotas related to GEFN of HICs. 3. Improve its funding resources to better support its operation. Increasing IMF resources through quotas, although preferred by some member states, is not the only option. The IMF Board may also consider increasing the NAB and making it a perpetual (not renewable) source of funding for the IMF. Additionally, the IMF may consider increasing the issuance of SDRs, preferably without pegging it to the quota system but by considering the GEFN of its members, LICs and MICs, in particular. It is important to highlight that increasing IMF funding based on traditional conditional lending would not result in an improved crisis finance for LICs and MICs. The high use of SDRs by LICs and MICs, as well as empirical findings on the use of the GFSN during the COVID-19 pandemic (see Zucker-Marques, Mühlich and Fritz 2023), show that access to unconditional crisis finance is key, and LICs and MICs are not provided with such access to the same extent as HICs. The odds of facing another systemic crisis like the COVID-19 shock are not unlikely. For this reason, the GFSN must be made fit for purpose, which means having sufficient resources and adequate policies to support the most vulnerable nations. At the same time, it is important to assess access to, and the lending capacity of, each of the three components of the GFSN – the IMF, the RFAs and central bank currency swaps. For the GFSN to become fit for purpose, each of its components must be designed and reformed with a view of all its elements. Only this holistic perspective will show the external financing capacity accessibly by a country to increase its resilience to external shocks and financial crises. In this sense, we recommend coordinated crisis finance between the IMF, RFAs and central banks to make the GFSN fit for purpose, reducing the potentially costly influence of “liquidity diplomacy” (Mühlich et al. 2020; Mühlich, Fritz and Kring 2021) that makes unilateral use of power asymmetries and to bring GFSN resources to the most vulnerable countries.