Policy Research Working Paper 8794 Debt in Low-Income Countries Evolution, Implications, and Remedies Sebastian Essl Sinem Kilic Celik Patrick Kirby Andre Proite Macroeconomics, Trade and Investment Global Practice March 2019 Policy Research Working Paper 8794 Abstract Debt vulnerabilities in low-income countries have increased reserves, and undiversified exports. A proactive effort to substantially in recent years. Since 2013, median govern- identify and reduce debt-related vulnerabilities is a prior- ment debt has risen by about 20 percentage points of gross ity for many low-income countries. Policy makers should domestic product and increasingly comes from non-conces- focus on mobilizing domestic resources, improving debt sional and private sources. As a result, in most low-income transparency, and strengthening debt management prac- countries, interest payments are absorbing an increasing tices. These efforts should be complemented by measures proportion of government revenues. The majority of low-in- to strengthen fiscal frameworks, improve the efficiency of come countries would be hard hit by a sudden weakening public expenditures and public investment management, in trade or global financial conditions given high levels and develop domestic financial systems. of external debt, lack of fiscal space, low foreign currency This paper is a product of the Macroeconomics, Trade and Investment Global Practice. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/research. The authors may be contacted at sessl@worldbank.org, skiliccelik@worldbank.org, pkirby@worldbank.org, and aproite1@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Debt in Low-Income Countries: Evolution, Implications, and Remedies1 Sebastian Essl Sinem Kilic Celik Patrick Kirby Andre Proite JEL Classification: [E62, H63, F34, F35, E44, E43] Keywords: Debt, low-income countries, debt vulnerabilities, non-concessional debt, debt management, financial markets, interest rates. 1 Essl (Global Macro and Debt Analytics Group of the Macro, Trade, and Investment Global Practice, World Bank; sessl@worldbank.org), Kilic Celik (Prospects Group of the Macro, Trade, and Investment Global Practice, World Bank; skiliccelik@worldbank.org), Kirby (Prospects Group of the Macro, Trade, and Investment Global Practice, World Bank; pkirby@worldbank.org); Proite (Global Macro and Debt Analytics Group of the Macro, Trade, and Investment Global Practice, World Bank; aproite1@worldbank.org. We thank Carlos Arteta, Ajai Chopra, Doerte Doemeland, Gerard Kambou, Sergiy Kasyanenko, Wee Chian Koh, Ayhan Kose, Franziska Ohnsorge, Rudi Steinbach, and Franz Ulrich Ruch for valuable contributions, comments and suggestions. Ishita Dugar, Brent Harrison, and Claudia Marchini provided excellent research assistance. The findings, interpretations and conclusions expressed in this paper are entirely those of the authors and should not be attributed to the World Bank, its Executive Directors, or the countries they represent. I. Introduction In recent years, many low-income countries (LICs) have gained access to additional sources of finance, including private and non-Paris Club creditors.2 While this has enabled these countries to fund important development needs, it has also led to higher levels of public debt. The increasing share of market-based debt exposes many LICs to interest rate, and refinancing risks. These trends take place as the external environment is becoming more challenging and borrowing costs are expected to rise around the world, as described in World Bank 2019. This means that, in the event of an abrupt deterioration in market conditions, some LICs may struggle to refinance debts from foreign sources and are at risk of capital flow reversals and dislocating currency depreciations. In this context, it is important for LICs to develop their domestic financial systems, strengthen capacity for domestic resource mobilization, improve macro-fiscal frameworks, and improve their resilience to shocks through the sound management of public debt and investment. Against this backdrop, this essay addresses the following questions:  What are the key characteristics of the recent rise in LIC debt?  How does rising debt relate to other LIC vulnerabilities?  How can better debt management help reduce LIC vulnerabilities?  How can complementary policy measures reduce LIC vulnerabilities? II. Key characteristics of the recent rise in LIC debt A recent sharp rise. Debt relief under the Heavily Indebted Poor Countries (HIPC) initiative and the Multilateral Debt Relief Initiative (MDRI) helped to reduce public debt among LICs from a median debt-to-GDP ratio of close to 100 percent in the early 2000s to a median of just over 30 percent in 2013.3 This downward trend reversed sharply thereafter, with the median debt ratio rising to above 50 percent by 2017 (Figure 1). The increase was large relative to other EMDEs, whose median debt rose by less than 11 percentage points of GDP from 2013 to 2017, compared to 20 percentage points for LICs. It was also broad-based: debt ratios rose in almost 90 percent of LICs, and a third experienced debt increases of more than 20 percentage points. The key role of fiscal deficits. Primary fiscal deficits had largely been closed among LICs by 2006, but widened steadily following the global financial crisis, especially among commodity exporters suffering from falling commodity prices. Rising deficits may also be the result of LICs’ increased ability to borrow as a result of HIPC and MDRI debt relief (Bayraktar and Fofack 2011; 2 LICs refers to countries meeting the World Bank Group’s definition of countries with per capita gross national income below $995 per year in 2017. This group includes 33 countries (Annex). 3 Most LICs—27 of 33—benefited from one or both of the HIPC and MDRI programs. 2 Figure 1: LIC government finances (A) Gross government debt (B) Gross government debt by LIC category Percent of GDP Percent of GDP 140 Median 60 All LICs 120 Interquartile range Commodity exporters 100 Others 80 40 60 40 20 0 20 2010 2011 2012 2013 2014 2015 2016 2017 2000 2002 2004 2006 2008 2010 2012 2014 2016 (C) Primary fiscal balance (D) LICs with largest increase in government debt Percent of GDP Percent of GDP 4 LIC median 150 Mozambique Zimbabwe Commodity exporters Sierra Leone Gambia, The 125 2 Others Benin 100 0 75 -2 50 -4 25 -6 0 2000 2002 2004 2006 2008 2010 2012 2014 2016 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Source: International Monetary Fund, World Bank. 2017 A. Dashed blue lines denote the interquartile range, while the solid blue line is the median. A.B.C. The sample includes 30 low-income countries, of which 2 are oil exporters, 8 metals exporters, and the remaining 20 are non-resource-intensive. It excludes Somalia, South Sudan, and Syria due to data restrictions. A.B. Figure shows median gross government debt in percent of GDP. B.C.D. LICs= Low-income countries. 3 Marcelino and Hakobyan 2014). The primary balance of most LICs has been negative since the mid-2000s, and all but five (of 31 with available data) LICs had primary deficits in 2017, with a third carrying a primary deficit exceeding 3 percent of GDP. Uses of borrowed funds. A rising debt burden is typically less of a reason for concern if it is used to finance investment that raises a country’s potential output, and therefore its ability to repay loans in the future (World Bank 2017). In some LICs, growing deficits reflected a push to finance public investment, as suggested by the doubling of median LIC public investment as a share of GDP from 3 percent in 2000 to 6 percent in 2015. This was the experience of Guinea-Bissau, Madagascar, Mali, and Nepal, where wider fiscal deficits were matched by higher public investment (IMF 2018a). These countries form a minority, however, as a substantial part of LIC borrowing has been used to finance a rise in current consumption. In resource-intensive countries in Sub-Saharan Africa, for example, the bulk of increased spending enabled by a rise in commodity prices went to public sector wages (World Bank 2018a). Some borrowing may also have been redirected toward the accumulation of private assets stored abroad.4 Dependence on external debt. Given their typically small local creditor base, a significant share of LIC borrowing comes from abroad and is denominated in foreign currencies. The resulting currency mismatch poses a challenge to LICs, as a depreciating currency can lead to a rise in the domestic value of the country’s debt burden and interest payments. This challenge is more severe in countries with a significant share of external debt priced at market rates, and less so for countries benefiting from the low interest rates on concessional debt. The median LIC carries external debt, including both public and private debt, equivalent to 28 percent of GDP and almost half of total debt. Median external debt as a share of GDP has risen about 3 percentage points since 2012, with several important outliers. Commercial debt issuances have contributed to especially sharp rises in external debt burdens in Mozambique and Tajikistan. In Uganda, external debt as a share of GDP has more than doubled since 2012, to more than 40 percent of GDP in 2017. The maturity composition of LIC external debt has remained broadly stable—short-term debt remained moderate at 5 percent of total external debt in 2016. Shift toward non-traditional creditors. The composition of public debt has shifted over the last decade, becoming increasingly non-concessional as LICs have increased their reliance on financing from non-traditional sources (Figure 2). The median share of non-concessional debt in public debt rose to 55 percent in 2016 (the latest year for which data are available), an increase of nearly 8 percentage points since 2013, and 15 percentage points compared with a decade earlier. Commercial creditors have become an important source of credit for some countries (World Bank and IMF 2018a). Ethiopia, Mozambique, Rwanda, Senegal, Tajikistan, and Tanzania have all 4 Ndikumana and Boyce (2011) find that for every dollar in external loans to Sub-Saharan Africa, capital outflows increased by roughly 60 cents in the same year. 4 Figure 2: Public debt in LICs (A) Change in creditor composition of public (B) Share of non-concessional debt and publicly guaranteed external debt, 2007-16 Non-Paris Percent of government debt Club bilateral 80 Median 70 Commercial Interquartile range 60 50 Bonds 40 30 Multilateral 20 Paris Club 10 bilateral 0 2000 2002 2004 2006 2008 2010 2012 2014 2016 -6 -4 -2 0 2 4 Percent of GDP (C) Interest payments (D) Share of LICs in debt distress or at high risk of distress Percent of revenue Percent of countries 12 Median 40 High In debt distress 10 Interquartile range 35 30 8 25 6 20 4 15 2 10 0 5 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 0 2013 2014 2015 2016 2017 2018 Source: International Monetary Fund, World Bank. A. GDP-weighted average across 32 low-income countries. “Bilateral” includes public and publicly guaranteed (PPG) loans from governments and their agencies (including central banks), loans from autonomous bodies, and direct loans from official export credit agencies. “Multilateral” includes PPG loans and credits from the World Bank, regional development banks, and other multilateral and intergovernmental agencies. It excludes loans from funds administered by an international organization on behalf of a single donor government. “Bonds” include PPG bonds that are either publicly issued or privately placed. “Commercial” includes PPG debt from commercial bank loans from private banks and other private financial institutions, as well as export and supplier credits. B.C. Dashed blue lines denote the interquartile range, while solid blue line is the median. Includes 30 low-income countries and excludes Somalia, South Sudan, and Syria due to data restrictions. D. Figure shows the percent of low-income countries eligible to access the IMF’s concessional lending facilities that are either at high risk of, or in, debt distress. A country is considered to be in debt distress if it is experiencing difficulties in servicing its debt, as evidenced, for example, by the existence of arrears, ongoing or impending debt restructuring, or if there are indications that a future debt distress event is probable. The sample includes 30 low-income countries. 5 issued commercial public debt since 2010, generally denominated in U.S. dollars.5 Non-Paris Club creditors, notably China, have also become a more important source of financing over the past decade, especially in Sub-Saharan Africa (World Bank 2015a). In 2016, non-Paris Club debt accounted for more than a fifth of the median LIC’s external debt, and about 13 percent of their public debt (World Bank 2018b). Major recipients of lending from non-Paris Club creditors include the Democratic Republic of Congo, Ethiopia, Mozambique, Tanzania, Uganda, and Zimbabwe (Atkins et al. 2017; Eom, Brautigam and Benabdallah 2018). Lending arrangements for commercial and non-Paris Club debt are often not public, and they can be complex and varied (World Bank and IMF 2018b). Some of this debt is collateralized, which could reduce budget flexibility by earmarking revenues, could weaken the creditor’s incentive to assess the borrower’s debt sustainability, and (if large) could increase funding costs from other creditors who may reassess the probability of being repaid. Moreover, increased exposure to non- Paris Club and commercial creditors may pose coordination challenges for debt resolutions in the future, making the consequences of debt distress even more disruptive, especially if debt is collateralized (World Bank and IMF 2018c). Rising cost of debt service. As debt loads have grown and become less concessional, interest payments have absorbed a growing share of government revenues. Among LICs, the median interest payments-to-revenue ratio rose to over 5 percent in 2017, up from just over 3 percent in 2013. The increase in the ratio was due to rapidly rising interest payments, with median interest payments among LICs having grown by over 128 percent versus 31 percent growth in government revenues. Drivers of rising debt. Countries with the fastest rise in debt were often fragile and affected by a combination of conflict, weak governance, or commodity-dependence (World Bank 2018c). In The Gambia, government debt increased from nearly 60 percent of GDP in 2013 to an estimated 88 percent in 2017, with interest payments absorbing 42 percent of revenue. The rise in debt was a result of loose fiscal policy, bailouts of state-owned enterprises, and widespread misman- agement by the previous government prior to a transition to democracy in early 2017 (IMF 2018b). In Mozambique, the government debt-to-GDP ratio has increased by close to 50 percentage points since 2013, reaching an estimated 102 percent in 2018, with interest payments rising from 2.6 percent of revenues to 16.5 percent over the same period. The deterioration was underpinned by rising deficits as fiscal policy remained loose amid lower commodity prices and subdued growth, and was aggravated by the inclusion of previously undisclosed external commercial debt in 2016 5 Of 11 LIC debt issuances since 2010, all were denominated in U.S. dollars, with the exception of one of Senegal’s two issuances in 2018, which was euro-denominated. 6 (IMF 2018c). The country is in debt distress, and several payments to external borrowers have been missed. Zimbabwe is also classified as being in debt distress. Over the last five years, government debt has risen substantially from just over 48 percent of GDP in 2013 to an estimated 82 percent in 2017. Persistently large fiscal deficits have partly been the result of an elevated public wage bill, which absorbed 90 percent of revenues in 2017 (IMF 2017). In addition, revenues remain subdued amid weak growth and structural rigidities, while transfers to the agricultural sector have kept non-wage expenditure elevated. Moreover, the deficits have partly been financed through an overdraft facility at the Reserve Bank of Zimbabwe that, given insufficient reserves, has led to money creation and exacerbated foreign-currency shortages. Risk of debt distress. Higher levels of public debt, much of it external, and an increased reliance on commercial loans make many LICs vulnerable to currency, interest rate, and refinancing risks (Devarajan 2018; Gill and Karakülah 2018a,b).6 LIC vulnerabilities are reflected by the fact that almost all LICs have the lowest or second lowest grade in the OECD’s country credit risk classification.7 Because of rising arrears or the need for debt restructuring, 11 LICs were assessed as being in debt distress or at a high risk of debt distress as of November 2018, compared to only six in 2015.8 For LICs assessed at low or moderate risk of debt distress, safety margins have eroded. III. Other LIC vulnerabilities Private debt. Due to shallow domestic capital markets and limited access to international finance, the median LIC has total private debt equivalent to only 18 percent of GDP, significantly less than the 41 percent ratio for the median non-LIC EMDE (Figure 3).9 Nonetheless, LIC private sector debt has been on a steady upward trend since 2005, rising by almost 8 percentage points. Excess 6 Separately, some countries such as The Gambia are vulnerable to rollover risk because of heavy reliance on short- term domestic debt (IMF 2018d). 7 There is one exception: The credit rating for Senegal has improved recently in the OECD credit risk classification, improving from 6 to 5 in a 0-7 rating system, with a higher number indicating higher credit risk (OECD 2018). 8 A country is considered to be in debt distress if it is experiencing difficulties in servicing its debt, as evidenced, for example, by the existence of arrears, ongoing or impending debt restructuring, or if there are indications that a future debt distress event is probable. LICs in debt distress are The Gambia, Mozambique, South Sudan, and Zimbabwe. LICs at high risk of debt distress are Afghanistan, Burundi, Central African Republic, Chad, Ethiopia, Haiti, and Tajikistan. There is a total of 30 LICs that have a debt sustainability analysis (DSA) available under the Joint World Bank / IMF debt sustainability framework (DSF). 9 Private sector debt refers to the sum of commercial banks’ and other financial corporations’ claims on the non- financial private sector, in percent of GDP. 7 Figure 3: Risks to LIC debt sustainability (A) Private sector debt (B) Median export concentration Percent of GDP Non-LIC EMDEs 0.5 50 Export concentration index LICs 0.4 40 30 0.3 20 0.2 10 0.1 0 0.0 Non-LIC EMDEs LICs 2000 2002 2004 2006 2008 2010 2012 2014 2016 (C) Growth (D) Global commodity demand growth Percent 2010-15 average Percent 8 12 5 Baseline projection 6 10 4 2010-16 4 8 3 2 6 2 0 4 1 -2 2 0 Energy Agriculture Metals -4 0 2017 2018 2019 2020 2017 2018 2019 2020 2017 2018 2019 2020 2017 2018 2019 2020 LICs Oil-exporting Metals- Non-resource- LICs exporting LICs intensive LICs Source: BP Statistical Review, Haver Analytics, United Nations, United States Department of Agriculture, World Bank, World Bureau of Metals Statistics. A. Domestic credit to the non-financial private sector provided by commercial banks and, if data are available, by other financial corporations. Median debt, based on 148 EMDEs and 29 LICs. A.B. Non-LIC EMDEs= Emerging market and developing economies excluding LICs; LICs= Low-income countries. B. Orange lines indicate interquartile ranges of Herfindahl-Hirschmann concentration index, which measures the degree of product concentration, with values closer to 100 indicating a country’s exports are highly concentrated in a few products. D. To ensure comparability, 2010-16 is model-predicted commodity demand growth. 8 private debt can sometimes be transformed into public debt, either directly through bailouts or indirectly through countercyclical government spending in response to private deleveraging, suggesting that the line between public and private debt can blur (Mbaye, Badia and Chae 2018). Growth subject to downside risks. Growth in LICs is expected to remain resilient, supporting their ability to service debt, but risks are tilted to the downside. LIC growth is expected to average 5.6 percent in 2018 and accelerate to just over 6 percent in 2019-20, supported by rising agricultural output and continued infrastructure investment (World Bank 2019). However, over the next decade, weaker growth in major emerging markets may slow global demand for metals, which dampens growth prospects for LICs that depend on metals for government and export revenues (World Bank 2018c). Downside risks to this outlook predominate and include the possibility of a faster-than-expected slowdown among major trading partners (including China, a major commodity consumer); a renewed plunge in commodity prices; a deterioration in international financial conditions; and the possibility of natural disasters, conflict, or severe weather events. Elevated debt, lower investment growth, increased risks. Rising levels of non-concessional public debt, often at variable rates, make some LICs susceptible to a sudden increase in borrowing costs, especially when they have substantial refinancing needs in coming years or have borrowed in foreign currencies. As advanced economies continue to withdraw monetary policy accommodation, new debt issuances and debt rollovers may become more expensive, resulting in rising LIC debt service costs that could weaken investment and lower medium-term growth (World Bank 2015b, 2016, and 2017). Fiscal consolidation, while often necessary, can also dampen growth in the short term. In the absence of sufficient lending made available at concessional terms, there is a risk that high public debt will lead to higher interest rates, crowding out private investment and slowing growth.10 Similarly, rising interest payments to domestic creditors may encourage policy makers to engage in financial repression—using administrative or other means to channel domestic savings toward the purchase of public debt—which can dampen private sector investment and limit the development of domestic financial markets (Fry 1997). Substantial current account deficits. Almost all LICs carry persistent, substantial current account deficits, with an estimated median of 6.8 percent of GDP in 2017 (Figure 4). Forty percent of LICs had current account deficits that widened by at least 3 percentage points of GDP over the last decade. Among metals exporters, rising deficits reflected the pickup in import-intensive mining investment, while in non-resource-intensive countries it often reflected high public investment. 10 Bevan (2012) argues that although evidence in the literature for the crowding out effect on investment in LICs is weak, it may be more important where financial depth is low. 9 Figure 4: External positions in LICs (A) Current account balance (B) Current account funding Percent of GDP Ratio of current account deficit 0 2.0 All LICs -2 Commodity exporters -4 1.5 Others -6 1.0 -8 -10 0.5 -12 -14 0.0 -16 LICs Metal exporting LICs 2004 2006 2008 2010 2012 2014 2016 Foreign direct investment Official development assistance (C) Foreign direct investment (D) Median foreign reserves Standard deviation of FDI flows Months of import coverage 6 6 Median for non-LIC EMDEs 5 5 4 4 3 3 2 2 1 1 0 LICs Commodity Commodity Metal exporting 0 exporting LICs importing LICs LICs LIDCs LICs Source: International Monetary Fund, World Bank. A. LICs= Low-income countries. Figure shows median current account balance in percent of GDP. The sample represents a total of 30 low-income countries, of which 2 are oil exporters, 8 are metals exporters, and the remaining 20 are non-resource-intensive. It excludes Somalia, South Sudan, and Syria due to data restrictions. B. The sample represents a total of 21 low-income countries, including 6 metals-exporting LICs, with current account deficits in 2016. C. EMDEs= Emerging market and developing economies. Standard deviation represents the median standard deviation of foreign direct investment in percent of GDP from 2000 to 2017. D. LIDCs=Low-income developing countries; LICs=Low-income countries. See Annex for details. Orange lines indicate interquartile range. Data is as of the last reported year, mostly 2016. 10 Countries relying on capital inflows to finance a large and persistent current account can be more vulnerable to currency crises, as weaker investor confidence can result in a slowdown in capital inflows, leading to higher borrowing costs, downward currency pressures, difficulties in rolling over debt, and possible macroeconomic and financial market stress (Roubini and Wachtel 1999). Current account deficits in LICs, however, are typically financed by capital inflows from development assistance, remittances, foreign lending, and foreign direct investment. The stable, long-term and often concessional nature of this financing mitigates some of the risks usually associated with large current account deficits. Foreign direct investment and development assistance flows were generally more than adequate to finance LICs’ current account deficits—the median LIC received inflows of these types 1.6 times as large as its current account deficit. In more than half of LICs, development assistance alone was greater than the total current account deficit. Median FDI inflows were equal to about half the current account, except for metals exporters where it was considerably more. FDI flows to LICs, however, are particularly sensitive to fluctuations in global growth and liquidity (Burger and Ianchovichina 2017). Among these countries, commodity exporters, particularly metals exporters, are particularly vulnerable to sudden swings in FDI flows that accompany changes in the external environment—FDI flows are more than twice as volatile in metal-exporting LICs than in other EMDEs. While external vulnerabilities can be mitigated by a strong foreign reserve position, more than 40 percent of LICs with available data have reserves close to or below three months of imports. IV. Role of better debt management Goal of sound debt management. In most LICs, government debt is the largest domestic financial portfolio, and debt management operations can be substantial relative to public spending and economic activity. A sound macro-fiscal policy framework requires that public debt is sustainable and can be serviced under a wide range of circumstances at reasonable costs. While ex ante the level of debt is mainly determined by fiscal policy, ex post the composition of debt can play an important role in safeguarding debt sustainability. Effective debt management plays a critical role in funding the government’s financing needs in a timely fashion, helping ensure low debt servicing costs at an acceptable degree of risk, and supporting the development of domestic securities markets. In addition, debt management can help minimize fiscal risks stemming from contingent liabilities, such as guarantees or on-lending to state-owned enterprises or through public-private partnerships, through effective monitoring and reporting. The benefits of sound debt management are fourfold:  Lowers debt servicing costs. In many LICs, debt service payments absorb a significant share of public revenues (notably in Burundi, the Central African Republic, and Chad), 11 reflecting a combination of low revenue bases, sizable debt loads, and a shift toward non- concessional terms. Effective debt management can help avoid excessive debt service costs by increasing awareness of the financial options available, enabling countries to borrow at competitive costs with a prudent degree of risk.  Supports financial sector development. More developed local-currency bond markets can promote economic stability by reducing the reliance on external debt, facilitating the implementation of counter-cyclical fiscal policies, and enhancing resilience to sudden reversals of capital flows. Public debt instruments can serve as a benchmark for pricing of private sector debt instruments. Local-currency bond markets can enable diversification from bank financing and provide a savings vehicle for a variety of investors to support growth (World Bank and IMF 2014).  Reduces economic volatility. Effective debt management can reduce economic volatility by selecting debt instruments that help insulate the government balance sheet from uncertainties. Both currency and interest rate shocks can be mitigated in this fashion, making a country less susceptible to contagion and financial risks, and supporting cheaper and more stable funding for the private sector.  Enhances public sector transparency and medium-term planning. A key element of sound public debt management is the public and comprehensive reporting of government debt, which improves the capacity of policy makers and the broader public to assess the fiscal position and appropriately weigh public balance sheet risks alongside spending and revenue priorities. Evolution of debt management in LICs. Despite some improvements, debt management in LICs still suffers from substantial deficiencies. Weaknesses in debt transparency, notably in monitoring and reporting, are pervasive. Medium-term debt strategies are becoming more common but have shortcomings in quality and implementation. Capacity and institutional set-up are often lagging.  Debt transparency. Better compilation and monitoring of public debt and guarantees are needed to ensure that risks are detected before they materialize (World Bank 2007). Recent examples of hidden debt and discrepancies among debt statistics point to continued low debt recording capacity, weak legal frameworks, and governance challenges. Debt Management Performance Assessments (DeMPA) suggest that, of the 17 LICs with available data, minimum requirements in debt recording are met by only eight, and monitoring guarantees are met by only four. Due to shortcomings in accuracy, timeliness, coverage and completeness of debt records, only four of these 17 countries met the minimum requirements for debt reporting and evaluation (Figure 5). Only a third of the 59 countries eligible for International Development Association borrowing report private sector external debt statistics (World Bank and IMF 2018d). 12 Figure 5: LIC policy frameworks (A) Countries meeting DeMPA minimum (B) Countries meeting DeMPA minimum requirements, select categories requirements, select categories Number of countries Number of countries 12 Most recent DeMPA 12 Most recent DeMPA 10 1st DeMPA 10 1st DeMPA 8 8 6 6 4 4 2 2 0 0 framework Managerial and Reporting Coordination Coordination DMS Audit LGLD Records DA and DS and CBM and BCP Borrowing Borrowing Domestic SD, SC Structure External Evaluation Debt with MP with FP CFF Legal (C) Central bank transparency index (D) Exchange rate regimes Index Number of LICs 6 LICs Other EMDEs 35 Flexible Pegged 5 30 4 25 20 3 15 2 10 1 5 0 0 1998 2000 2002 2004 2006 2008 2010 2012 2014 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 Source: Bloomberg, Debt Management Performance Assessments (DeMPA), Dincer and Eichengreen (2014), 2015 International Monetary Fund, Shambaugh (2004), World Bank. A.B. BCP=Business Continuity Planning; CBM=Cash Balance Management; CFF=Cash Flow Forecasting; DA=Debt Administration; DMS=Debt Management Strategy; DS=Data Security; FP=Fiscal Policy; GLD=Loan Guarantees, On lending Derivatives; MP=Monetary Policy; SD=Segregation of Duties; SC=Staff Capacity. Sample covers 17 low-income countries. C. Unweighted averages. The range of the index is from 0-15, 0=least transparent and 15=most transparent. D. De facto exchange rate regime from the Exchange Rate Regime Classification of Shambaugh (2004) is used to determine whether a country has a pegged or flexible exchange rate. The original classification has four categories: “1” reflects no fluctuation at all, “2” indicates movements within 1 percent bands, “3” indicates movements within 2 percent bands, and “4” indicates a one-time devaluation with 0 change in the remaining 11 months of the year. Shambaugh (2004) assesses these movements against relevant base currencies. The constructed dummy variable indicating a pegged exchange rate regime was defined to equal 1 for countries classified as 1, 2, 3, or 4. A value of 0 is assigned to flexible exchange rates—i.e., exchange rates that routinely fluctuate outside a 2 percent band. Based on 31 LICs. 13  Debt management strategies. A growing number of countries are producing medium-term debt management strategies. However, their quality varies significantly, and implementation  Broader issues. Some of the most pressing challenges include insufficient legal frame- works, weak capacity, lack of coordination between fiscal and monetary policy, inefficient management of cash and fiscal risks, and poor audit and risk control procedures. V. Complementary policy measures Domestic resource mobilization. Among LICs, there is considerable scope to enhance tax revenues and reduce the need to rely on debt financing (Baum et al. 2017). In the median LIC, government revenues accounted for only 19 percent of GDP in 2017, well below 28 percent of GDP in the median non-LIC EMDE, reflecting the prevalence of informal activity (Chapter 3). This highlights the need to broaden tax bases, especially for higher-income households, in a way that minimizes economic distortions and that carefully manages trade-offs between efficiency and equity (World Bank 2018d). Unexpected revenue windfalls from sudden improvements in a country’s terms of trade can be set aside to reduce fiscal deficits and debt. Improving spending efficiency. LICs have significant infrastructure needs that require debt financing. However, debt sustainability concerns associated with the financing of infrastructure may be lessened if these expenditures are accompanied by stronger long-term growth and better macro-fiscal, budgeting, and financing frameworks. There may also be room to cut unproductive spending (often subsidies) in order to allow for more growth-enhancing or better-targeted programs.11 Debt used to finance projects that generate a revenue stream is less likely to be unsustainable. There is also often considerable scope to improve the efficiency of investment spending by improving the institutions and procedures governing project appraisal, procurement, and monitoring. By one estimate, a country moving from the lowest quartile to the highest quartile in the efficiency of public investment could double the impact of that investment on growth (IMF 2015). Development of local financial markets. Reliance on external funding means that there is often a currency mismatch in LIC borrowing and revenues, leaving countries vulnerable to swings in the value of the currency. The development of local currency bond markets can help mitigate this risk, though they are often a relatively high-cost option. These markets require a functional money market, primary and secondary markets, a diverse base of investors, a stable regulatory system which includes reliable custody and settlement systems, and a significant improvement in debt management capacity. Sound macroeconomic policy and financial sector stability are also critical, as is transparent and effective communication by the government. Alongside improved debt 11 Credible and well-designed institutional arrangements—such as fiscal rules, stabilization funds, and medium-term expenditure frameworks—can help build fiscal space, improve the management of revenue windfalls, and strengthen policy outcomes (Huidrom, Kose, and Ohnsorge 2016). 14 management, growing local financial markets can help countries graduate from concessional lending by mitigating some of the costs and risks associated with non-concessional debt. Better data collection. Transparency about balance sheets is a pre-requisite for sound debt management. Among other gaps, there is often limited data on contingent liabilities (especially those arising from state-owned enterprises and public-private partnerships) and the assets held by LIC governments. These data limitations are especially acute for debt issued by commercial and non-Paris Club creditors. Improving data collection practices for LIC debt would help policy makers make informed and appropriate borrowing decisions and allow the public to hold the government accountable for its fiscal management (World Bank and IMF 2018d). Monetary policy and exchange rate regimes. More resilient monetary policy frameworks and foreign reserve buffers can help mitigate the impact of terms-of-trade and other shocks, including on the fiscal position (Adler, Magud, and Werner 2017). More LICs could join the growing number of EMDEs where improvements in the monetary policy regime have reduced inflation and, where appropriate, allow greater exchange rate flexibility to absorb shocks. Rigorous and transparent lending standards. Creditors also have a role to play in containing debt vulnerabilities. The Addis Ababa Action Agenda calls for debtors and creditors to work together to prevent and resolve unsustainable debt situations. Creditors can aim for good practice in lending, drawing on principles for sustainable lending such as those being championed by the G20 countries (G20 2018). VI. Conclusion In recent years, a broad-based rise in borrowing has increased public debt vulnerabilities in LICs. The composition of debt has also shifted, as many LICs have increased their exposure to non-Paris Club creditors and market-based debt, which may pose coordination challenges for any future debt resolution. While increased access to market funding has provided LICs with opportunities to address development needs, it has also exposed some countries to currency, interest rate, and refinancing risks. The number of LICs at high risk of debt distress or in debt distress has increased significantly, and safety margins in many LICs currently assessed at low or moderate risks of debt distress have eroded. External gross financing needs are likely to rise further as current account deficits widen and large international bonds fall due. By increasing the effectiveness of resource mobilization, public spending, and debt management—supported by better data collection—LICs can reduce the probability of costly defaults, enhance debt transparency, support sustainable financial sector development, and reduce economic volatility. 15 ANNEX. Comparison of LIDCs and LICs This paper discusses LICs following the World Bank Group definition of countries with per capita gross national income below $995 per year. This group includes 33 countries. It differs from other reports (such as IMF 2018a and World Bank and IMF 2018c), which include additional middle- income countries following the IMF definition of low-income developing countries (LIDCs). The term “LIDC” refers to countries with low per capita gross national income and comparatively weak socioeconomic indicators. List of countries in LIDCs and LICs  Low-Income Developing Low-Income Countries Low-Income Developing Low-Income Countries Countries (LIDCs) (LICs) Countries (LIDCs) (LICs) 1 Afghanistan 1 Afghanistan 31 Malawi 17 Malawi 2 Bangladesh       32 Mali 18 Mali   3 Benin 2 Benin 33 Mauritania 4 Bhutan       34 Moldova         5 Burkina Faso 3 Burkina Faso 35 Mozambique 19 Mozambique 6 Burundi 4 Burundi 36 Myanmar       7 Cambodia       37 Nepal 20 Nepal   8 Cameroon       38 Nicaragua         9 Central African Republic 5 Central African Republic 39 Niger 21 Niger 10 Chad 6 Chad 40 Nigeria       11 Comoros 7 Comoros 41 Papua New Guinea       12 Congo, Dem. Rep. of 8 Congo, Dem. Rep. of 42 Rwanda 22 Rwanda 13 Congo, Republic of       43 São Tomé and Príncipe         14 Côte d'Ivoire       44 Senegal 23 Senegal   15 Djibouti       45 Sierra Leone 24 Sierra Leone   16 Eritrea 9 Eritrea 46 Solomon Islands       17 Ethiopia 10 Ethiopia 47 Somalia 25 Somalia 18 Gambia, The 11 Gambia, The 48 South Sudan 26 South Sudan 19 Ghana             27 Syrian Arab Republic   20 Guinea 12 Guinea 49 Sudan       21 Guinea-Bissau 13 Guinea-Bissau 50 Tajikistan 28 Tajikistan 22 Haiti 14 Haiti 51 Tanzania, United Rep. 29 Tanzania, United Rep. 23 Honduras       52 Timor-Leste         24 Kenya       53 Togo 30 Togo   25 Kiribati       54 Uganda 31 Uganda   26 Kyrgyz Republic       55 Uzbekistan         27 Lao P.D.R.       56 Vietnam         28 Lesotho       57 Yemen, Rep. 32 Yemen, Rep.   29 Liberia 15 Liberia 58 Zambia       30 Madagascar 16 Madagascar 59 Zimbabwe 33 Zimbabwe (continues in the next column)  Sources: IMF and World Bank.     16 References Adler, G., N. E. Magud, and A. Werner. 2017. “Terms-of-Trade Cycles and External Adjustment.” Washington, DC: Working Paper 17/29, International Monetary Fund, Washington DC. Atkins, L., D. Brautigam, Y. Chen, and J. Hwang. 2017. "China-Africa Economic Bulletin #1: Challenges of and Opportunities from the Commodity Price Slump", China Africa Research Initiative, Johns Hopkins University School of Advanced International Studies, Washington DC. Baum, A., A. Hodge, A. Mineshima, M. M. Badia, and R. Tapsoba. 2017. “Can They Do It All? Fiscal Space in Low-Income Countries.” IMF Working Paper 17/110, International Monetary Fund, Washington, DC. Bayraktar, N., and H. Fofack. 2011. "Capital Accumulation in Sub-Saharan Africa: Income-Group and Sector Differences." Journal of African Economies 20 (4): 531-561. Bevan, D. 2012. “Aid, Financial Policy, Climate Change, and Growth.” Working Paper, 2012/77, United Nations University, World Institute for Development Economic Research, New York. Burger, M. J., and E. I. Ianchovichina. 2017. "Surges and Stops in Greenfield and M&A FDI Flows to Developing Countries: Analysis by Mode of Entry." Review of World Economics 153 (2): 411- 432. Devarajan, S. 2018. “Low-Income Country Debt: Déjà Vu All Over Again?” Washington, DC: Future Development, Brookings Institution. Dincer, N. and B. Eichengreen. 2014. “Central Bank Transparency and Independence: Updates and New Measures.” International Journal of Central Banking 10 (1): 189-253. Eom, J., D. Brautigam, and L. Benabdallah. 2018. “The Path Ahead: The 7th Forum on China- Africa Cooperation.” China-Africa Research Initiative Briefing Paper 1, Johns Hopkins School of Advanced International Studies, Washington, DC. Fry, M. J. 1997. "In Favour of Financial Liberalisation." The Economic Journal 107 (442): 754- 770. G20 (Group of 20). 2018. “G20 Leaders’ Declaration Building Consensus for Fair and Sustainable Development.” G20 Meeting, Buenos Aires, Argentina. Gill, I., and K. Karakülah. 2018a. “Is Africa Headed for Another Debt Crisis?” Duke Center for International Development, Durham NC. ———. 2018b. “Sounding the Alarm on Africa's Debt.” Washington, DC: Future Development, Brookings Institution. Huidrom, R., M. A. Kose, and F. L. Ohnsorge. 2016. “Challenges of Fiscal Policy in Emerging and Developing Economies.” Policy Research Paper 7725, World Bank, Washington, DC. 17 IMF (International Monetary Fund). 2015. Making Public Investment More Efficient. Washington, DC: International Monetary Fund. ———. 2017. Zimbabwe. Staff Report for the 2017 Article IV Consultation. Washington, DC: International Monetary Fund. ———. 2018a. Macroeconomic Developments and Prospects in Low-Income Developing Countries. Washington, DC: International Monetary Fund. ———. 2018b. The Gambia 2017 Article IV Consultation—Staff Report. Washington, DC: International Monetary Fund. ———. 2018c. Mozambique 2017 Article IV Consultation—Staff Report. Washington, DC: International Monetary Fund. ———. 2018d. IMF Fiscal Monitor: Capitalizing on Good Times, April 2018. Washington, DC: International Monetary Fund. Marcelino, S. and M.I. Hakobyan. 2014. “Does Lower Debt Buy Higher Growth? The Impact of Debt Relief Initiatives on Growth.” IMF Working Paper 14/230, International Monetary Fund, Washington, DC. Mbaye, S., M.M. Badia, and K. Chae. 2018. “Bailing Out the People? When Private Debt Becomes Public.” Working Paper 18/141, International Monetary Fund, Washington, DC. Ndikumana, L., and J. Boyce. 2011. “Capital Flight from Sub-Saharan Africa: Linkages with External Borrowing and Policy Options.” International Review of Applied Economics 25 (2): 149- 170. OECD (Organisation for Economic Co-operation and Development). 2018. Country Risk Classifications of the Participants to the Arrangement on Officially Supported Export Credits. Paris: Organisation for Economic Co-operation and Development. Roubini, N., and P. Wachtel. 1999. "Current-Account Sustainability in Transition Economies." In Balance of Payments, Exchange Rates, and Competitiveness in Transition Economies, edited by M. I. Blejer and M. Skreb, 19-93. Boston, MA: Springer. Shambaugh, J. C. 2004. “The Effect of Fixed Exchange Rates on Monetary Policy.” The Quarterly Journal of Economics 119 (1): 301-352. World Bank. 2007. Managing Public Debt: From Diagnostics to Reform Implementation. Washington, DC: World Bank. ———. 2015a. Global Economic Prospects: The Global Economy in Transition. Washington, DC: World Bank. ———. 2015b. Global Economic Prospects: Having Fiscal Space and Using It. Washington, DC: World Bank. 18 ———. 2016. Global Economic Prospects: Spillovers Amid Robust Growth. Washington, DC: World Bank. ———. 2017. Global Economic Prospects: Weak Investment in Uncertain Times. Washington, DC: World Bank. ———. 2018a. Africa’s Pulse, No. 17, April 2018. Washington, DC: World Bank. ———. 2018b. International Debt Statistics 2019. Washington, DC: World Bank. ———. 2018c. “Are Poor Countries Headed Towards Another External Debt Crisis?” Internal note. ———. 2018d. “Approach Paper: Public Finance for Development Evaluation.” Internal note. ———. 2019. Global Economic Prospects: Darkening Skies. Washington, DC: World Bank. World Bank and IMF (World Bank and International Monetary Fund). 2014. Revised Guidelines for Public Debt Management. Washington, DC: World Bank and International Monetary Fund. ———. 2018a. Debt Vulnerabilities in IDA Countries. Washington DC: World Bank and International Monetary Fund. ———. 2018b. G-20 Note: Improving Public Debt Recording, Monitoring and Reporting Capacity in Low and Lower Middle-Income Countries: Proposed Reforms. Washington, DC: World Bank and International Monetary Fund. ———. 2018c. Debt Vulnerabilities in Emerging and Low-Income Economies. October 2018 Meeting of the Development Committee. Washington DC: World Bank and International Monetary Fund. ———. 2018d. G-20 Note: Strengthening Public Debt Transparency: The Role of the IMF and the World Bank. Washington, DC: World Bank and International Monetary Fund. 19