MONGOLIA ECONOMIC UPDATE Macroeconomic Stability and Credit Growth Global financial crisis Socioeconomic reforms Co mm od ity Pri ces h owt c Gr omi on Ec Bank Credit Growth January 2020 MONGOLIA ECONOMIC UPDATE Macroeconomic Stability and Credit Growth January 2020 Contents Acknowledgments....................................................................................................................... 7 Abbreviations............................................................................................................................... 8 EXECUTIVE SUMMARY......................................................................................................................... 9 I. PERFORMANCE AND PROSPECTS......................................................................................... 11 A. Recent Economic Developments.................................................................................................................12 A1. Real sector: Economic growth is slowing as mining production moderates....................... 12 A2. Inflation has decelerated in recent months, reflecting the impact of slowing credit growth..........................................................................................................................14 A3. The labor market: A strong recovery but limited progress on reducing the gender wage gap and poverty.....................................................................................................15 A4. Fiscal developments: Impressive performance continued in 2019......................................... 17 A5. The balance-of-payments situation has improved, but external buffers remain weak.................................................................................................................................................23 A6. Monetary policy: Monetary conditions have gradually tightened, but risks to financial sector sustainability remain high.................................................................................. 27 B. Outlook and Risks..............................................................................................................................................31 II.UNDERSTANDING MONGOLIA’S BANKING SECTOR CREDIT DEVELOPMENT ................ 37 A. Context...................................................................................................................................................................38 B. Recent developments in bank credit..........................................................................................................39 B1. Current status of banking sector credit..............................................................................................39 B2. Main drivers of credit development....................................................................................................42 B3. Growing importance of non-bank financial institutions.............................................................. 43 C. Sectoral allocation.............................................................................................................................................43 C1. Household sector........................................................................................................................................43 C2. Private sector................................................................................................................................................44 C3. Credit concentration..................................................................................................................................45 D. Credit cycle and business cycle....................................................................................................................47 E. Policy recommendations ................................................................................................................................48 Annex: Estimating the drivers of bank credit in Mongolia .................................................. 50 |3| BOXES Box I.1. A surge in recurrent spending is explained by growth in the pension and wage bill.......20 Box I.2. Proposed amendments to the VAT are likely to be detrimental to revenue mobilization...................................................................................................................................21 Box I.3. Summary of the 2020 budget.................................................................................................................22 Box I.4. Exchange rate development and FX intervention in Mongolia..................................................26 Box I.5. Asset Management Company – An effective solution to the banking sector problems?..........................................................................................................................................29 Box I.6. The macro-fiscal implications of writing off loans of pensioners.............................................34 Box I.7. Global and regional outlook and risks.................................................................................................35 Box II.1. Credit booms in Mongolia......................................................................................................................47 FIGURES Figure I.1. Growth having peaked in early 2019, is starting to slow........................................................12 Figure I.2. Investment and consumption remain the key drivers of growth.........................................12 Figure I.3. Government consumption growth has significantly increased............................................13 Figure I.4. Robust household income growth has propelled private consumption growth...........13 Figure I.5. Production of key commodities have risen, except copper...................................................14 Figure I.6. The services sector has remained a key contributor to non-mining output...................14 Figure I.7. Inflation has moderated in recent months...................................................................................15 Figure I.8. …with slowing credit growth partly contributing to this moderation................................15 Figure I.9. The unemployment rate has declined consistently since 2016............................................16 Figure I.10. The labor market is shifting away from agriculture................................................................16 Figure I.11. The gender gap in labor force participation is large and widening.................................17 Figure I.12. After stagnating for years, public sector wages are starting to increase.......................17 Figure I.13. Rural poverty declined, while urban poverty stagnated in 2018.......................................17 Figure I.14. Fiscal balance (percent of GDP) has improved in recent years….......................................18 Figure I.15. …resulting in steady fall in government debt (percent of GDP) since 2017..................18 Figure I.16. CIT and social security contributions have been the main contributors to rising revenue..................................................................................................................................19 Figure I.17. Total expenditure has been growing steadily in recent years.............................................19 Figure I.18. A surge in recurrent spending mainly explained by the pension/welfare and wage bill....................................................................................................................................................20 Figure I.19. Implementation of the 2017 legal changes explains the increase in pensions/ welfare spending.............................................................................................................20 Figure I.20. Mongolia’s VAT statutory rate has been low by international comparison...................21 Figure I.21. VAT return filers (number of businesses during 2016–19)...................................................22 Figure I.22. Revenue collection and refunds have substantially increased...........................................22 Figure I.23. Macroprudential measures have significantly contained growth of imports...............24 Figure I.24. The balance-of-payments situation has improved in 2019.................................................24 Figure I.25. Gross international reserves increased........................................................................................25 Figure I.26. The U.S. dollar strengthened globally in 2019..........................................................................25 Figure I.27. The REER has appreciated relative to major trade partners................................................25 Figure I.28. Recent foreign exchange market interventions in historical perspective.......................26 |4| Mongolia economic update Figure I.29. FX loans have consistently declined since September 2018...............................................26 Figure I.30. Monetary policy was tightened in late 2018…..........................................................................27 Figure I.31. The macroprudential measures have contained credit and import growth.................27 Figure I.32. Interbank rates have increased moderately in tandem with the policy rate….............28 Figure I.33. Average lending rates have gradually declined amid relatively robust funding conditions...............................................................................................................................28 Figure I.34. The deceleration in broad money was driven by slowing credit …..................................28 Figure I.35. … and decelerating local currency deposits...............................................................................28 Figure I.36. Bank asset quality remains weak…................................................................................................29 Figure I.37. Banking system appears liquid…....................................................................................................30 Figure I.38. Rising currency mismatch in the banking system...................................................................30 Figure I.39. Government debt-to-GDP ratio is projected to fall further but remains higher than many selected peers.................................................................................................................32 Figure I.40. The government is expected to pay a higher amount of debt due on external bonds during 2021–24.......................................................................................................................32 Figure I.41. Real GDP growth (percent)...............................................................................................................35 Figure I.42. World commodity price forecast (Index=nominal U.S. dollars, 2016=100)..................35 Figure I.43. Regional growth....................................................................................................................................36 Figure II.1. Mongolia ranks well in financial sector development.............................................................38 Figure II.2. Mongolia’s credit-to-GDP ratio is comparable to that for lower middle-income countries…................................................................................................................................................40 Figure II.3. …relative to inflation, it has been higher than the credit to GDP ratio for some regional comparators..........................................................................................................................40 Figure II.4. Credit growth in Mongolia is high relative to peers................................................................40 Figure II.5. Mongolia has experienced several episodes of high credit growth..................................40 Figure II.6. Most of the credits were channeled to individuals..................................................................41 Figure II.7. Corporate credit expansion was mostly beneficial to domestic private investment..41 Figure II.8. A sharp deceleration of credit growth pushed the credit-to-GDP ratio below its long-term trend.....................................................................................................................................41 Figure II.9. The quality of corporate loans has significantly deteriorated..............................................41 Figure II.10. Provisioning remains below average for many key sectors...............................................42 Figure II.11. Provision of individual loans increased......................................................................................42 Figure II.12. Growing importance of non-bank financial institutions (NBFIs)......................................43 Figure II.13. A large part of outstanding individual credit is for mortgage and salary loans......................44 Figure II.14. Salary loans have been a key driver of individual credit growth since 2016...........................44 Figure II.15. Corporate credit is concentrated in the mining and mining-related sectors..............44 Figure II.16. Corporate credit growth has been dominated by trade and mining activities since early 2018....................................................................................................................................44 Figure II.17. Credit dollarization has been declining since mid-2018…..................................................45 Figure II.18. …for both corporate and individual borrowers.......................................................................45 Figure II.19. Loan and deposit concentration by allocation and client...................................................46 Figure II.20. Credit growth and economic performance...............................................................................47 Figure II.21. Credit cycle and business cycle.....................................................................................................47 Figure A.1. Impulse response of total credit to selected shocks...............................................................50 Figure A.2. Main drivers of credit growth (Variance Decomposition of VAR model)........................51 TABLES Table I.1. Summary of the 2020 approved budget.........................................................................................23 Table I.2. Key macroeconomic indicators............................................................................................................33 Table II.1. Mongolia: Policy Options to Manage Volatile Credit Growth................................................49 |6| Mongolia economic update Acknowledgments This edition of the Mongolia Economic Update was prepared by Jean-Pascal Nganou and Davaadalai Batsuuri, with contributions from Fang Guo, Ikuko Uochi, and Undral Batmunkh. The team is grateful to the Research and Statistics Department of the Bank Mongolia for an outstanding collaboration on the preparation of the special topic of this Economic Update. The team would like to thank Aaditya Mattoo, Ergys Islamaj, Ekaterine T. Vashakmadze, Katia D’Hulster, Liang Wang, Tungalag Chuluun, and Vera Kehayova for their helpful comments. The team also appreciates the constructive comments received from the Revenue Division of Ministry of Finance. The report was prepared under the overall guidance of Martin Raiser (Country Director), Hassan Zaman (Regional Director), Deepak Mishra (Practice Manager), and Andrei Mikhnev (Country Manager). The analysis in this report is based on statistics spanning from the 2000s to December 31, 2019. The team is grateful to Indra Baatarkhuu (Communications Associate) for her advice on the dissemination of the report, and to Diane Stamm (Consultant) for editing the report. Angar Enkhtur (Program Assistant) provided outstanding operational support. |7| Macroeconomic Stability and Credit Growth mongolia – GOVERNMENT Fiscal Year January 1 – December 31 Currency Equivalents (Exchange Rate Effective as of December 31, 2019) Currency Unit = Tugrug (MNT) US$1.00 = MNT 2734.33 Abbreviations AMC Asset Management Company AQR Asset Quality Review BoM Bank of Mongolia CIT Corporate Income Tax EMDEs Emerging Market and Developing Economies FATF Financial Action Task Force FDI Foreign Direct Investment FX Foreign Exchange GDP Gross Domestic Product H1 First half of year H2 Second half of year LLP Loan Loss Provisioning MNT Mongolian Tugrug MoF Ministry of Finance NEER Nominal Effective Exchange Rate NPLs Nonperforming loans NSO National Statistics Office OT Oyu Tolgoi PBOC People’s Bank of China PIT Personal Income Tax Q1 First quarter of the year Q3 Third quarter of the year Q4 Fourth quarter of the year REER Real Effective Exchange Rate VAT Value-Added Tax y/y Year-over-year |8| Mongolia economic update EXECUTIVE SUMMARY While Mongolia’s economy is starting to slow from its brisk pace of recent years, the medium- term growth outlook remains firmly positive. The strong economic recovery since 2017 has been underpinned by impressive fiscal outcomes, better coordination of macroeconomic policy, favorable global commodity prices, and strong recovery in private investment. Real GDP growth strengthened to 7.2 percent in 2018 from 5.3 percent in 2017. Following the gradual decline in commodity prices combined with a reduction in the quality of key mineral exports, growth moderated to 6.3 percent in the first nine months of 2019 and is expected to slow to 5.8 percent for the year as a whole. Medium-term growth prospects remain positive (averaging 5.6 percent in 2020–21), largely supported by robust growth in private consumption and continued steady investment in mining and manufacturing. This forecast is predicated on the government’s continued commitment to fiscal consolidation to contain public debt in the medium term by improving revenue mobilization and controlling public spending. There are significant downside risks to the growth outlook emanating from heightened uncertainties in the global and domestic environments. Domestically, risks are related to the possible relaxation of the government’s commitment to reforms on the eve of the general elections in 2020, thereby affecting market sentiment and the flow of foreign direct investment. Although the authorities have maintained fiscal discipline and tight credit policies, the recent decisions to roll back the increase in the social security contribution rates and to write off pension-backed loans, are seen as notable setback to progress on economic reforms. The lack of such commitment has been responsible for exacerbating fiscal and financial vulnerabilities in past elections. Globally, uncertainty surrounding the impact of the recent trade deal between China and the United States of America, especially China’s energy imports from non-US sources, poses considerable risk to our base case outlook.1 In addition, given the strong economic linkages between China and Mongolia, our growth outlook is likely to be adversely affected if the novel coronavirus outbreak is not quickly contained. Weather-related shocks (e.g., drought/flooding, harsh winter) and resumption of non-trade barriers at the border with China could affect Mongolia’s coal exports. Mongolia’s policy buffers, while stronger than prior to the last downturn in 2016, are not yet sufficient to fully mitigate the potential impact of negative shocks. The banking sector remains fragile and the failure to complete the banking recapitalization program could delay official financial support. Given that Mongolia remains on the gray list of the Financial Action Task Force (FATF), limited progress on anti-money laundering issues would pose additional risks.2 1 The deal signed in January 2020 includes Chinese commitments for the purchase of U.S.A energy exports. It is not clear whether this would affect Chinese energy imports from other sources. This could affect growth prospects for Mongolia negatively. At the same time, a renewed flare up of international trade tensions could depress global growth and exert downward pressure on commodity prices. 2 The FATF is an independent inter-governmental body, which develops and promotes policies to protect the Global Financial System from terror funding and money laundering activities. Mongolia was put in the FATF’s gray list on Octo- ber 18th, 2019 following limited progress on addressing anti-money laundering issues. Incidentally, ING group recently decided to exit Mongolia and stop serving as foreign currency clearing agent for the Mongolian banks. Moreover, with slow progress on FATF issues, there is risk that an already limited corresponding relation with banks in advanced econ- omies is likely to be further eroded with adverse effects on trade and investment. |9| Macroeconomic Stability and Credit Growth Strengthening fiscal buffers through continued fiscal consolidation and building up reserves by limiting excessive foreign exchange interventions should remain two important near-term policy priorities of the government. In addition, improving the business environment and strengthening financial sector stability, especially with regard to credit policy and soundness of the banking sector, will contribute to greater macroeconomic stability, job creation, and poverty reduction. The accelerated progress on the U.S.A-Mongolia trade agreement following the July 2019 visit of the Mongolian president to the U.S.A. could also contribute positively to the outlook | 10 | Mongolia economic update I. PERFORMANCE AND PROSPECTS | 11 | Macroeconomic Stability and Credit Growth A.Recent Economic Developments A1. Real sector: Economic growth is slowing as mining production moderates Having maintained a brisk pace of growth since early 2017, the Mongolian economy showed signs of gradual deceleration in the second half of 2019. Real gross domestic product (GDP) growth decelerated to 6.3 percent year-on-year (y/y) in the first nine months of 2019 from 7.3 percent in the first half (H1) of the year. With lower gold and copper grades in Oyu Tolgoi’s (OT’s) production, mining sector output contracted by 1.4 percent (y/y) in the third quarter after an average growth of 13 percent in the first two quarters. In addition, the closure of OT’s copper concentrator factory for maintenance in September also contributed to the slowdown in growth. However, coal production and exports volumes have remained robust despite gradually declining prices. Growth in the non-mining sector slowed down to 6 percent over the first nine months of 2019, from 8.1 percent in the same period of 2018. On the demand side, investment continued to grow rapidly (24.5 percent, y/y) in the first nine months of 2019, mainly driven by strong foreign direct investment (FDI) inflows (US$1.3 billion) (figures I.1 and I.2). Private consumption growth remained strong as well (7.3 percent during January–September 2019 from 6.4 percent a year ago) amid improved labor market conditions and increased wages. Figure I.1. Growth having peaked in early 2019, is Figure I.2. Investment and consumption remain the starting to slow key drivers of growth Mining Non-mining GDP at market prices Household Consumption Public Consumption Gross Capital Formation Net Exports 10% GDP 8% 20% 6% 15% 4% 10% 2% 5% 0% 0% -5% -2% -10% -4% -15% I I-III I I-III I I-III I I-III I I-III I I-III I I-II I-III I-IV I I-II I-III I-IV I I-II I-III I-IV I I-II I-III 2014 2015 2016 2017 2018 2019 2016 2017 2018 2019 Sources: NSO; World Bank staff estimates. Investment remains the largest contributor to growth. Gross capital formation expanded by 24.5 percent in the first three quarters of 2019—a deceleration from 35.7 percent a year ago. FDI remained robust during January–September 2019 (close to 10 percent of GDP) despite a slight decline from 11 percent in the first nine months of 2018. Mongolia’s economic growth remains heavily dependent on investment, which is largely funded by volatile FDI flows. In addition, public sector investment grew by 57 percent (y/y) during January–September 2019, reflecting increased budget investment in schools and hospitals and improved customs capacity (including laboratories, border checkpoints, and control and management centers). Moreover, accelerated progress toward completion of investment projects that have been under implementation for several years also helped explain these developments in public investment. | 12 | Mongolia economic update Final consumption grew rapidly in the first nine months of 2019 supported by robust growth in household income and increased government spending. Final consumption grew by 7.7 percent (y/y), relative to 5.8 percent in the same period in 2018, owing to a surge in government consumption—fueled by increased wages of public sector employees—and sustained expansion of private consumption (7.3 percent from 6.4 percent) (figure I.3). Meanwhile, real household income continued to grow rapidly (8.6 percent, y/y) in September 2019 (figure I.4).3 Negative net exports has slightly narrowed compared to a year ago due to an improved trade balance in 2019. Figure I.3. Government consumption growth has Figure I.4. Robust household income growth has significantly increased propelled private consumption growth Private Consumption Government Consumption 15% 40% Real Private Consumption (y/y): LHS 14.9% 14.0% Real Household Income (y/y): RHS 30% 10% 10.5% 9.6% 20% 6.9% 7.3% 5% 6.5% 5.2% 10% 3.2% 0% 0% -1.3% -5% -10% -6.3% -6.4% -10% -20% I-III I-III I-III I-III I-III I-III I-III I-III I-III I-III I-III I-III I-III I-III 2014 2015 2016 2017 2018 2019 2012 2013 2014 2015 2016 2017 2018 2019 Sources: NSO; World Bank staff estimates. Mineral production remained robust in 2019, though the pace of expansion moderated. The performance of the coal sector—which accounts for 48.2 percent of total mineral exports— remained robust in 2019, mainly driven by strong demand from China and high contractual prices (figure I.5).4 Meanwhile, iron ore production was up by 37.3 percent in the same period following a sharp increase in prices. However, copper production weakened, especially in the second half of the year—a contraction of 3.7 percent (y/y) in 2019—reflecting the lower quality of copper from OT and declining prices in global markets.5 Moreover, despite rising gold prices, gold production fell by 21 percent in 2019, which can be mainly attributed to the government’s decision to raise the gold royalty in March 2019.6 3 Rising public wages and recovery in income from livestock activities have supported household income growth. 4 In 2018, coal production and export grew by 6.2 percent and 8.6 percent, respectively, to reach historically high levels (that is, 50 million tons and 36 million tons, respectively). 5 The OT managment indicated that copper production contracted by 8 percent in 2019. The gold production also contracted by 15.2 percent in 2019. The contraction in copper and gold production is expected to continue in 2020, mainly due to lower grade ore. The shutdown of the concentration factory of the OT mine for maintenance reasons in September also contributed to a weaker performance. In 2021, higher grade ore and a significant increase in gold production are expected. 6 On March 3, 2019, the Parliament ended the subsidy on gold sales to the Bank of Mongolia (BoM). As a result, the royalty tax on gold sales increased from 2.5 percent to 5 percent. | 13 | Macroeconomic Stability and Credit Growth Figure I.5. Production of key commodities have Figure I.6. The services sector has remained a key risen, except copper contributor to non-mining output Growth change in 2019 (y/y) Non-mining real GDP growth by sectoral contribution (%) 40% 38% Agriculture Services Net Tax Other non-mining industries 30% Production Volume Non-mining GDP growth (%) Export Volume 10 20% 8 13% 6 10% 8% 4 6% 2 2% 2% 0 0% -2 -4% -2% -4 -10% Q1-15 Q2-15 Q3-15 Q4-15 Q1-16 Q2-16 Q3-16 Q4-16 Q1-17 Q2-17 Q3-17 Q4-17 Q1-18 Q2-18 Q3-18 Q4-18 Q1-19 Q2-19 Q3-19 Copper Coal Oil Iron Ore Sources: NSO; World Bank staff estimates. Note: Copper and gold accounted for over 30 percent of total export proceeds in 2018. Non-mining output growth has decelerated but remains robust, supported by the services sector. Non-mining GDP grew by 6 percent (y/y) in the first nine months of 2019, slightly below the 8.1 percent achieved during the same period last year (figure I.6). Agriculture sector growth accelerated to 4.1 percent during January–September 2019, from 2.1 percent in the previous year, reflecting relatively favorable weather conditions, lower loss of livestock, and strong survival rates. The services sector, mostly the transportation and trade sectors, continue to post robust growth (above 6 percent) owing to their strong links with the mining sector.7 The construction sector has also recovered, growing by 4.7 percent (y/y) in the same period from 0.3 percent in the first three quarters of 2018. The latter is explained by the continued development of OT’s underground project, increased government capital expenditures, and recovery in residential buildings with increased funding to the government’s housing mortgage program since April 2019. Manufacturing production decelerated to 2.3 percent (y/y) in the same period, from 10.1 percent in 2018, mainly due to a slowdown in food and textile productions, which both account for nearly 70 percent of the manufacturing production.8 A2. Inflation has decelerated in recent months, reflecting the impact of slowing credit growth National Consumer Price Index inflation has fallen sharply in recent months. Declining meat prices, decelerating credit growth, and some base effects from the previous year were the main factors for the declining inflation rate, including in Ulaanbaatar (figure I.7). Food inflation moderated to 8.3 percent (y/y) in December 2019 from 19 percent in September following a seasonal decline in meat prices.9 Mainly reflecting the implementation of a series of macroprudential measures by 7 Transportation services accelerated to 15.1 percent (y/y) in Q3 2019, from 8 percent in the first half of 2019. Traditionally, transportation activities tend to accelerate in the third quarter of the year, mainly due to acceleration of activities in the construction and services sectors under favorable weather conditions. 8 A slowdown in the food industry has mainly been explained by a moderation in meat processing compared to the peak of 2018 due to the government’s efforts to limit meat exports somewhat following the rapid increase in the retail price of meat in the domestic market during most of 2019. Moreover, China imposed several standards requirements on imports of meat products from Mongolia, which also contributed to a moderation in meat exports. 9 In September 2019, inflation (9 percent, y/y) was mainly driven by meat prices (over 40 percent, y/y) following accelerated meat exports. | 14 | Mongolia economic update the Bank of Mongolia (BoM), credit growth slowed to below 5 percent (y/y) in December from a peak of 26.5 percent in December 2018. On the supply side, the impact of the government’s ban on transportation and use of raw coal to improve Ulaanbaatar’s air quality, which triggered a massive hike in hardwood fuel inflation last year, dissipated in 2019.10 The price of hardwood fuel (coal and wood) declined by 8.1 percent (y/y) in December 2019 relative to its peak of 65 percent (y/y) in December 2018. This follows the introduction of subsidized/affordable refined coal briquettes to accompany the ban as well as some base effects. In addition, retail fuel prices also declined, by 1.9 percent (y/y) in December compared to 24.6 percent increase in the same period last year. Figure I.7. Inflation has moderated in recent Figure I.8. …with slowing credit growth partly months... contributing to this moderation Consumer price inflation (y/y, %): Ulaanbaatar CPI inflation, Credit growth (y/y, %) 12% Meat Price Inflation Non-Meat Food Inflation (y/y) 25% Credit growth: LHS 15% 10% Oil Core Inflation (y/y):RHS 13% 20% Others CPI Inflation (y/y): RHS 11% 8% Headline 15% 9% 6% 10% 7% 4% 5% 5% 2% 3% 0% 1% 0% -5% -1% Dec-15 Dec-16 Dec-17 Dec-18 Dec-19 Apr-16 Apr-17 Apr-18 Apr-19 Aug-16 Aug-17 Aug-18 Aug-19 -2% Oct-15 Oct-16 Oct-17 Oct-18 Oct-19 Jan-16 Jan-17 Jan-18 Jan-19 Jul-15 Jul-16 Jul-17 Jul-18 Jul-19 Apr-15 Apr-16 Apr-17 Apr-18 Apr-19 Sources: NSO Bulletin; World Bank staff estimates. Note: CPI = Consumer Price Index; LHS = left-hand side; RHS = right-hand side. The core inflation rate has also moderated following a sharp deceleration in credit growth. Supported by a combination of factors such as strong domestic demand, pass through of fuel and food prices, and exchange rate depreciation, the core inflation—defined as the change in prices of consumer items excluding meat, milk, and vegetables—was a key driver of high inflation in 2018. However, it has moderated considerably this year. A robust recovery in private consumption (6.6 percent, y/y) and high credit growth (26 percent) weighed on core inflation in 2018. However, after reaching 8.6 percent in September 2019, core inflation moderated to 4.2 percent in December 2019, reflecting the impact of credit growth deceleration (figure I.8). A3. The labor market: A strong recovery but limited progress on reducing the gender wage gap and poverty Labor market conditions continued to improve in 2018, with the unemployment rate steadily declining. Largely driven by an expansion of employment in the industrial sector, the national unemployment rate fell to 7.8 percent in 2018, down from 8.8 percent in 2017 and 10 percent in 2016 (figure I.9). The mining, manufacturing, and construction sectors were the largest contributors to employment growth in 2018, as the number of people employed increased by over 20,000 between 2017 and 2018. The services sector also added nearly 17,000 jobs during this period. However, despite an increasing share of employment in the industrial sector, the majority of the employed population remains engaged in low-end services and agriculture. 10 This policy went into effect in May 2019. | 15 | Macroeconomic Stability and Credit Growth The structural transformation of the economy has continued, though both agriculture and services have lost labor to the industrial sector. The employment share of agriculture declined over the past three years from 34 percent in 2016 to 26 percent in Q3 2019. (figure I.10) The employment share of the industrial sector, including mining, manufacturing, and construction, has been on the rise over the same period, reaching nearly 23 percent in Q3 2019 from 11 percent in 2016. The services sector, including wholesale and retail businesses, absorbed many workers during 2012–16 who migrated from agriculture-related activities, with its job share rising to 58 percent from 48 percent. However, the share of the services sector declined during 2018–19, reflecting a shift to the industrial sector from services, following a strong performance of the mining sector. The employment share of the mining sector increased to about 5 percent in Q3 2019 from 3.7 percent in 2016, reflecting increased activities in the mining sector over the past two years. In total, around 15,000 new jobs were created in mining, an increase of 36 percent. Figure I.9. The unemployment rate has declined Figure I.10. The labor market is shifting away from consistently since 2016 agriculture 11.6 Percentage share of employment by sector (%) 9.9 10 65 Services Agriculture Industry 8.8 55 8.2 7.9 7.9 7.7 7.5 7.8 45 35 25 15 5 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Jan-12 Jan-13 Jan-14 Jan-15 Jan-16 Jan-17 Jan-18 Jan-19 Sources: NSO; World Bank staff estimates. Despite improved labor market conditions, the gender wage gap remains wide and the female labor force participation rate is at a historic low. In 2018, the nominal monthly average wage and salary was MNT 903,000 for women and MNT 1,100,000 for men—a 22 percent gender wage gap. This gap is partly explained by the fact that many women are employed in informal or low-end service activities, while an increasing number of men have obtained jobs in better- paid sectors (such as mining, manufacturing, and construction). Indeed, the male labor force participation rate reached a historically high level (69.5 percent) in 2018, while female labor force participation, which has consistently declined since 2006, hit a historically low level (53.4 percent) in 2018 (figure I.11). To reduce gender disparities in Mongolia’s labor market, a recent qualitative study suggested a package of policy reforms including promoting a mandatory gender nondiscrimination policy in hiring, ensuring access to finance and training for female-operated micro-businesses, and expanding the quantity and quality of early childcare services (Schmillen and Weimann-Sandig 2018). | 16 | Mongolia economic update Figure I.11. The gender gap in labor force Figure I.12. After stagnating for years, public sector participation is large and widening wages are starting to increase Quarterly labor force participation rate (%) Cumulative real wage growth since 2010 (percent) 75. Total Male Female 90% National Average 80% Private Sector 70. 70% Public Sector 60% 65. 50% 40% 60. 30% 20% 55. 10% 0% 2019 Sep 2010 2011 2012 2013 2014 2015 2016 2017 2018 50. 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 Sources: NSO; World Bank staff estimates. 2017 After years of stagnation, public sector wages have increased. The authorities have steadily increased public sector wages since late 2018 (an 8 percent increase in both Q4 2018 and January 2019) to help address wage stagnation that began in 2012. The real average wage of the public and private sector grew sharply—by 33 percent—during 2010–12 (figure I.12). Wages in the private sector rose by another 32 percent during 2013–18, while the public employees saw their real wages contract by about 5 percent during the same period. However, as of September 2019, public employees have seen a 10 percent real wage growth (y/y) following two sequential wage increases by the government. Figure I.13. Rural poverty declined, while urban Poverty has declined in rural areas but remained poverty stagnated in 2018 unchanged in urban areas, resulting in little 50 49.0 National Average Urban Rural progress in lowering the national poverty rate. 45 38.8 40 Based on the 2018 Household Socio-Economic 35 33.2 35.4 34.9 Percent of total population 30.8 Survey (HSES), the 2018 national poverty headcount 29.6 30 27.4 28.4 26.4 27.1 27.2 rate was estimated at 28.4 percent, representing a 25 23.3 21.6 18.8 1.2 percentage point reduction from 2016. During 20 15 the same period, the rural poverty rate declined by 10 4.1 percentage points, while the urban poverty rate 5 was not statistically different from 2016 (figure I.13). 0 2010 2012 2014 2016 2018 While the incidence of poverty remains high in rural Sources: NSO; World Bank staff estimates. areas, with two-thirds of the total population of Mongolia living in urban cities, poverty has become more concentrated in urban areas. The share of the poor population in urban areas increased from 56.3 percent in 2012 to 63.5 percent in 2018, with more than 40 percent living in Ulaanbaatar in 2018. A4. Fiscal developments: Impressive performance continued in 2019 With three consecutive years of primary budget surplus, government debt has been steadily declining since 2016. The fiscal balance has improved significantly, from a deficit of 15.3 percent of GDP in 2016 to a surplus of 2.6 percent of GDP in 2018 and 1.4 percent in 2019 (figure I.14).11 This remarlable fiscal turnaround is explained by massive revenue overperformance (strong performance in the mining sector and considerable revenue mobilization efforts by the government through improved tax administration), and a commitment to spending control (for example, 11 Development Bank of Mongolia spending is excluded from the fiscal balance for 2016–19. It is monitored separately. | 17 | Macroeconomic Stability and Credit Growth streamlining of the wage bill through a hiring freeze, and rationalization of underperforming capital spending).12,13,14 In contrast to the practices prior to 2017, the government has avoided any budget amendments in 2018 and 2019 keeping spending in check, (see box I.1 for more details). Substantial improvements in the primary balance and use of the fiscal surplus to buy back expensive domestic debt resulted in a sharp reduction in government debt between 2016 and 2019 (figure I.15). Declining interest payments reflecting falling government debt and improved terms contributed to the fiscal consolidation in recent years.15 Figure I.14. Fiscal balance (percent of GDP) has improved Figure I.15. …resulting in steady fall in in recent years… government debt (percent of GDP) since 2017 Overall balance: RHS Revenue: LHS Expenditures: LHS 20 40 88 15 85 35 10 73 30 68 5 58 60 25 2.6 1.4 49 20 0 43 15 -5 -4.1 -3.8 -5.5 10 -10 -9.1 -9.2 -10.4 5 -15 -15.3 0 -20 2011 2012 2013 2014 2015 2016 2017 2018 2019e 2012 2013 2014 2015 2016 2017 2018 2019e Source: MoF; World Bank staff estimates. Note: LHS = left-hand side; RHS = right-hand side. In 2019, the fiscal balance was in surplus for the second consecutive year amid revenue overperformance. The budget execution of 2019 recorded a surplus mainly due to robust revenue performance, although the under-execution of the capital budget (capex) also continued. The overall balance reached a surplus of 1.4 percent of GDP in 2019 following a surplus of 2.6 percent in 2018. The primary balance also exhibited a surplus of 3.7 percent of GDP after 5.8 percent in 2018. Revenue exceeded targets by over MNT 721 billion (6.4 percent relative to target) in 2019. Strong revenue collection in 2019 was supported by a broad-based revenue stream (corporate income tax [CIT], including thanks to robust mining exports, personal income tax [PIT], and value- added tax [VAT]), and more than compensated a surge in current budget spending after two years of tight spending controls. The Parliament approved a slightly expansionary 2020 budget while committing to reduce public debt further. The approved 2020 budget assumes that government revenue will grow by 12 Improved value-added tax (VAT) administration through digital technologies, increased social security contributions, and increased excise taxes on alcohol and tobacco were key drivers of growth in non-mineral revenues. 13 With technical assistance from the World Bank, the Minister of Finance issued Order #295 in December 2018 approving regulation of and methodologies on evaluation, prioritization, and selection of publicly financed projects. The approved methodology is expected to help the government assess its public investment projects to be financed by state budget investment. The new regulation will be used starting with the 2020 budget. 14 However, after no wage increases for five years, the government sequentially increased public wages starting in late 2018, as a response to pressing labor union demands. 15 Interest payments quadrupled from about 1 percent of GDP in 2012 to over 4 percent in 2017 due to a sharp rise in government or government guaranteed debt. However, implementation since mid-2017 of the government’s debt strategy, which emphasized repaying expensive domestic debt, replacing short-term external commercial loans by concessional borrowing, resulted in a gradual decline in interest payments. Consequently, the gap between the primary budget deficit and the overall budget deficit has been narrowing. | 18 | Mongolia economic update 16.6 percent (reaching 31.7 percent of GDP), mainly through improved CIT collection and better VAT collection due to upgraded facilities at the customs/border posts (figure I.16)16. As explained below, these revenue projections may be overoptimistic in light of recent developments. On the spending side, the 2020 budget projects an increase of 19.7 percent (reaching 34.1 percent of GDP) relative to the 2019 planned budget (figure I.17). The major sources of the spending increase are social welfare and pensions (inflation adjustment- and pension fund-related expenses including implementation of the legal amendments introduced in 2017 but postponed in the 2019 budget), capital expenditures (mainly ongoing investment projects), the wage bill, and spending on goods and services (including the cost of the election and air pollution reduction). Public debt is expected to decline further (for more details on the 2020 budget, see boxes I.2 and I.3). Figure I.16. CIT and social security contributions have Figure I.17. Total expenditure has been growing been the main contributors to rising revenue steadily in recent years Total Revenue (% of GDP) 39.8 Total Expenditure (% of GDP) 40 6.8 7 2017 2018 2019 budget 2020 approved 6.6 6.5 2016 5.9 6.0 35 32.8 34.1 6 2018 5.1 4.9 5.1 4.8 29.0 30 27.7 5 4.5 24.6 2019 budget 3.8 4.0 22.9 23.3 4 3.6 3.5 3.5 25 3.2 2020 approved % of GDP 3 2.4 2.5 2.2 2.2 20 2 15 9.9 9.2 9.4 1 10 5.2 0 5 2.2 Direct PIT CIT SSC VAT 0.9 0.4 0.1 Mineral 0 Revenues Total Exp Recurrent Capex Net Lending Sources: MoF; World Bank staff estimates. Note: The total spending increase from the previous Note: Figures for the 2020 approved budget are before year was 3.8 percentage points of GDP in 2019 and 1.3 the Presidential veto of the SSC gradual increase. percentage points in 2020. CIT = Corporate income Tax; PIT = Personal income tax; SSC = Social security contribution; VAT = Value added tax. 16 The revisions of the tax law approved by the Parliament in March 2019, became effective in January 2020. It aims to broaden the tax base through simplification of filing and rationalizing the number of taxes paid by Small and Medium Enterprises (SMEs). | 19 | Macroeconomic Stability and Credit Growth Box I.1. A surge in recurrent spending is explained by growth in the pension and wage bill The 2020 budget exhibits a substantial increase in the recurrent spending relative to capital expenditure. The primary drivers of recurrent spending increases are pension/welfare-related expenditures including the implementation of the 2017 amendments to the pension insurance legislation (32 percent of the total increase), the wage bill (23 percent), and spending on reducing air pollution, environmental protection, elections and Olympic games (13 percent) (figure I.18). Specifically, the implementation of the 2017 legal changes would increase the spending on pensions and welfare by about 0.3 percent of GDP (figure I.19).17 Figure I.18. A surge in recurrent spending mainly Figure I.19. Implementation of the 2017 legal changes explained by the pension/welfare and wage bill explains the increase in pensions/ welfare spending Sources: MoF (the 2020 budget document); World Note: Implementation of 2017 legal changes on pensions Bank staff estimates. was initially postponed for a year in 2019 budget. The revenue mobilization plan in the 2020 approved budget may be overoptimistic. While assumptions on commodity prices are broadly consistent with the projections of international institutions including the International Monetary Fund (IMF) and the World Bank, assumptions on export volumes appear to be overly optimistic. In particular, the revenue estimates assume that coal exports will reach 42 million tons this year, a 14.7 percent increase over the 2019 outturn of about 36.5 million tons. The latter is likely to be optimistic given weak prospects in the coal market and the expected growth slowdown in China. 17 The authorities may also have over-estimated the impact of the new tax law package on revenues. The approved budget proposes to collect MNT 1.8 trillion for the CIT (an almost 50 percent increase over the 2019 budget) through simplified and digitalized tax collection despite falling commodity prices. In the first 11 months of 2019, over 50 percent of total CIT collection came from mining companies. In addition, the approved 2020 budget expects to collect MNT 2.4 trillion in social security contributions, a substantial increase over MNT 1.7 trillion in 2019, largely due to a 1 percent increase in the contribution rate agreed under the IMF Extended Fund Facility program. However, recently the president vetoed the last phase of this gradual increase in the contribution rate to the social insurance program, which had been approved by Parliament in December 2019. An immediate impact of the veto would be that revenue collection for 2020 will decline by about 0.4 percentage points of GDP. The authorities recently indicated that this 17 The 2017 legal changes include (i) reducing the age for eligibility for pension insurance benefits by five years for herders; (ii) calculating each year of herders’ contributory service as 1.5 years instead of one year; (iii) introducing a second round of service buy-back options for herders and the self-employed following the first buy-back in 2012; (iv) awarding 1.5 service years for retiring mothers per child; and (v) guaranteeing social insurance premiums for mothers looking after a child under three years of age, financed by employers and the Social Insurance Fund. The service buy- back for herders and self-employed accounts for 90 percent of increase in Pension Insurance Fund expenditure. | 20 | Mongolia economic update increase in contribution rate will be implemented in 2021 instead of 2020. Moreover, there are proposals by members of Parliament to revise the VAT Law, which could be a threat to the fiscal gains of the last three years (see box I.2). Box I.2. Proposed amendments to the VAT are likely to be detrimental to revenue mobilization Two draft amendments to the VAT law aimed to decrease the value-added tax (VAT) statutory rate and increase refunds were recently introduced in the parliament. Specifically, the proposals intend to reduce the statutory rate from 10 percent to 5 percent and increase the refund on VAT paid to consumers for requesting vendors of a certified bill of purchase to 70 percent from the current 20 percent (of the VAT paid). Moreover, the proposal includes increasing the ceiling of total VAT refunds into total VAT revenue to 40 percent from 30 percent. Lawmakers argue that the proposed changes will alleviate the tax burden on the registered taxpayers but also ensure further contraction of the shadow economy and thus broaden the tax base. They also argue that VAT collection is at risk in the medium to long term since the number of VAT payers has risen at a slower rate since the amendment of the VAT law in 2015. But there are several weaknesses in the proposed changes. A reduction of the VAT statutory rate could Figure I.20. Mongolia’s VAT statutory rate has been low have a significant immediate adverse effect by international comparison on VAT revenue collection. In January 2007, the VAT statutory rate (introduced in 1998) was reduced from 15 percent to 10 percent This resulted in a drop in the VAT revenue estimated at 0.7 percentage point of GDP between 2006 and 2007. Mongolia’s VAT statutory tax rate is already low compared to its peer countries (figure I.20). The current 10 percent statutory rate is also below the average applied rate in neighboring countries, including China (a five-tier VAT structure of 3 percent, 5 percent, 6 percent, 10 percent, and 16 percent), Kazakhstan (12 percent), and Russia (a differentiated structure with 10 percent applied on basic Sources: MoF; IMF; World Bank database; World Bank staff foodstuffs, medicines, children’s stuffs, and estimates. 20 percent on any other goods). In addition, the low tax burden in Mongolia is further evidenced in the 2020 Doing Business report, as total tax and contribution rate (percent of commercial profits) stands at 25.7 percent, compared to 33.6 percent for East Asia and Pacific and 39.9 for OECD high income countries. A more generous VAT refund system could further erode Mongolia’s VAT revenue mobilization. The rollout of the VAT system supported by internet-based electronic receipt system (e-barimt), resulted in a substantial increase in the number of registered business entities. It also led to an expansion of the tax base as domestic VAT revenue increased by twofold to MNT 1.1 trillion in 2019 from about MNT 500 billion in 2015 (figure I.21). However, Mongolia’s VAT refunds have also increased substantially during the same period.a Specifically, the 20 percent VAT refund to individuals has accelerated, while refunds to corporates have been declining (figure I.22). | 21 | Macroeconomic Stability and Credit Growth Figure I.21. VAT return filers (number of businesses Figure I.22. Revenue collection and refunds have during 2016–19) substantially increased 87,119 Domestic VAT & Refund 78,963 1,200 400 20% refund & lottery to individuals (billion MNT): RHS 1,088 71,171 Corporates (billion MNT): RHS 1,000 Domestic VAT collection (billion MNT): LHS 898 300 45,015 Refund (billion MNT): RHS 40,206 800 734 218 30,902 202 24,869 26,809 551 553 189 21,255 600 200 17,622 483 14,231 10,981 6,494 129 83 400 138 164 38 95 108 0 39 100 0 1-Jan-2016 7-Feb-2016 15-Jun-2018 30-Jun-2015 16-Jul-2016 17-Jan-2016 24-Jan-2016 31-Jan-2016 14-Feb-2016 31-Dec-2015 31-Dec-2016 31-Dec-2017 31-Dec-2018 31-Dec-2019 200 95 108 91 106 64 54 0 0 2014 2015 2016 2017 2018 2019 Sources: MoF; General Department of Taxation; World Bank staff estimates. The VAT system needs reform, but not the reforms currently under discussion. First, the statutory tax rates are too low not too high. Second, the authorities should integrate the e-barimt system with customs declaration database and other tax information for ease of paying tax and filing tax return. Third, e-barimt system security and sustainability issues should be addressed by introducing ISO standards. Fourth, the e-barimt system should be made more accessible and convenient to end-users. Fifth, the tax authorities should focus on business entities that evade the e-barimt system altogether or fail to print out e-barimt to end-users. Moreover, the refund framework should be revised, following a detailed assessment of the costs and benefits of the current system. Particularly, the 20 percent VAT refund on final consumers is an income transfer. It would be useful to estimate the amount and determine if it makes sense both from a revenue and a policy standpoint. In short, the main VAT priority is to continue the rollout of the electronic receipt system and gradually reduce the refund. Source: Summarized and updated by the authors based on World Bank (2018). Note: a. There are three types of refunds that are operated under the current VAT system: (i) normal VAT refunds through the input tax credit; (ii) up to a 20 percent VAT refund to consumers who upload e-receipts of the VAT including purchases from registered VAT payers; and (iii) the lottery (applicable to all regardless of VAT payer) for all e-receipts. Box I.3. Summary of the 2020 budget The 2020 budget assumes (a) real economic growth of 8 percent in 2019 and 6 percent in 2020, (b) strong demand for coal and relatively lower production of copper concentrate, and (c) that tax reforms are fully implemented from 2020. It also assumes that inflation will be contained at 8 percent in 2020 as part of the monetary policy guidelines. Under the macroeconomic assumptions, the 2020 budget provides the revenue projections and corresponding expenditure plans shown in table I.1. | 22 | Mongolia economic update Table I.1. Summary of the 2020 approved budget   2019 Budget Plan 2020 Approved Budget   Billion MNT % of GDP Billion MNT % of GDP A. Total revenue & grants 11,067 31.3 12,900 31.7 B. Total structural revenue & grants 9,676 27.4 11,798 29.0 Tax revenue 8,586 24.3 10,656 26.2 Non-tax revenue 1,091 3.1 1,142 2.8 C. Total expenditure & net lending 11,589 32.8 13,873 34.1 Current expenditure 8,217 23.3 10,019 24.6 Wage bill 2,215 6.3 2,628 6.5 Goods and services 1,732 4.9 2,158 5.3 Interest payments 865.6 2.4 960 2.4 Subsidies 252 0.7 375 0.9 Transfers 3,152 8.9 3,899 9.6 Capital expenditure 3,244 9.2 3,806 9.4 Domestic investment 2,300 6.5 2,890 7.1 Externally funded investment 944 2.7 916 2.3 D. Structural balance: B–C (1,913) -5.4 (2,075) -5.1 E. Overall balance: A–C (522) -1.5 (973) -2.4 F. Stability and Heritage Funds: E–D 1,391 3.9 1,102 2.7 Key highlights of the 2020 budget include the following: • The budget is moderately expansionary. Primary spending (total expenditure net of interest payments) will increase by 1.4 percentage points of GDP in 2020. In the 2019 budget, primary spending rose by 5 percentage points relative to 2018. • Revenue projections, especially corporate income tax collection, are unrealistically high, given the slowdown in the economy and in commodity prices. The government has argued that the impact of tax reforms (simplification and digitalization) supports their revenue projections. • The substantial increase in recurrent spending is largely explained by an 8 percent increase in wages and pensions. This will feed through to social insurance, and social welfare benefits to create future spending commitments. • Capital spending is projected to reach 9.4 percent of GDP, a slight rise from 2019, mostly to complete existing projects. • The approved budget also includes the provision of MNT 125 billion (US$46 million) for fiscal stabilization funds and MNT 977 billion (US$361 million) for the Future Heritage Fund. Sources: MoF; 2020 budget document; World Bank staff calculations. A5. The balance-of-payments situation has improved, but external buffers remain weak Balance-of-payments pressures moderated in 2019 due to robust FDI inflows and import deceleration. The current account deficit decreased to US$1.7 billion in 2019 from US$2.2 billion in 2018. This mainly reflects trade balance improvements following the robust growth of exports (8.7 percent, y/y, in 2019) and a sharp deceleration of import growth (from a peak of 45.5 percent, y/y, in May 2018 to 4.3 percent in 2019). Relatively high commodity prices (especially for coal and iron ore) and stronger external demand contributed to the robust export performance. Meanwhile, the impact of the macroprudential measures to contain credit growth to households (effective | 23 | Macroeconomic Stability and Credit Growth January 2019) and slowing demand from the manufacturing sector explain the sharp deceleration of imports (figure I.23). The current account deficit has remained fully financed by FDI, with net inflows closing the year at US$2.11 billion, virtually unchanged from 2018. In addition, portfolio investment increased to US$368 million in the same period, primarily reflecting the international bond issue of the Mongolia Mortgage Corporation in January 2019.18 As a result, the balance-of- payments turned around from a deficit of US$141 million in 2018, to a surplus of US$453 million in 2019 (figure I.24). Figure I.23. Macroprudential measures have Figure I.24. The balance-of-payments situation has significantly contained growth of imports improved in 2019 Exports, imports, and bank credit growth (%, y/y), External Balance (% of GDP), 2006–19 September 2015–December 2019 60 Total Imports Manufacturing IP Credit Growth Overall BoP 50 30% Net FDI 55% 40 CA balance 25% KA & FA balance 30 40% 20% 20 25% 15% 10 10% 0 10% 5% -10 -5% -20 0% -20% -30 -5% -40 -35% -10% -50 Sep-15 Sep-16 Sep-17 Sep-18 Sep-19 Mar-16 Mar-17 Mar-18 Mar-19 Dec-15 Dec-16 Dec-17 Dec-18 Dec-19 Jun-16 Jun-17 Jun-18 Jun-19 2019e 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Sources: BoM; World Bank staff estimates. Note: CA = Current Account; KA = Capital Account; Note: IP = Industrial Production. FA = Financial Account. Gross international reserves further strengthened in 2019, mainly reflecting easing balance- of-payments pressures (figure I.25). In 2018, reserve accumulation was the result of a bond issue (not sovereign guaranteed) by the Development Bank of Mongolia (US$500 million) in October, and historically large amount of Bank of Mongolia gold purchases (about US$800 million) from individuals and companies. In 2019, gross international reserves further expanded to US$4.3 billion (about seven months of imports), supported by an improved trade balance, the international bond issue of the Mongolian Mortgage Corporation, and official sector support from development partners. 18 In January 2019, the Mongolian Mortgage Corporation issued a US$300 million non-sovereign guaranteed bond at a rate of 9.75 percent with a three-year maturity. | 24 | Mongolia economic update Increased foreign exchange interventions in Figure I.25. Gross international reserves increased 2019 led to a real appreciation of the tugrug Gross international reserves and exchange rate (US$ against the U.S. dollar and the Chinese billion), January 2012–December 2019 renminbi (RMB). Gross Reserves (bn US$): LHS Nominal exchange rate (MNT/US$): RHS In nominal terms, the tugrug depreciated by 4.5 3.4 percent against the U.S. dollar and by 2.1 2700 Joint Donor Program 4.0 3.5 percent against the Chinese RMB in 2019, 3.0 2500 compared to depreciation of 9 percent and 2.5 2.0 2300 3.4 percent, respectively, in 2018. The Bank of 1.5 2100 Mongolia is estimated to have sold US$2.86 1.0 0.5 1900 billion to defend the tugrug in 2019, up from 0.0 1700 US$1.5 billion for 2018 (box I.4).19 Pressure on Jun-14 Jun-15 Jun-16 Jun-17 Jun-18 Jun-19 Dec-14 Dec-15 Dec-16 Dec-17 Dec-18 Dec-19 the exchange rate was moderate during the first half of the year but accelerated thereafter, Sources: Balance-of-payments/external debt statistics partly reflecting general U.S. dollar strength (Bank of Mongolia); World Bank staff estimates. against the currencies of Mongolia’s peers and neighbors (figure 1.26) and partly explained by some market panic prompted by the gray listing of Mongolia by the Financial Action Task Force (FATF). Meanwhile, with Mongolia’s inflation higher than that of major trading partners (China and Russia), the real effective exchange rate appreciated by 5 percent in 2018 and 3.7 percent in the first 10 months of 2019, which may have affected the export competitiveness of the non-mining sector (figure I.27).20 Figure I.26. The U.S. dollar strengthened globally in Figure I.27. The REER has appreciated relative to 2019... major trade partners Nominal exchange rate per US$ (percent change) NEER, REER, and nominal exchange rate (December 12% 2017=100, index), December 2017–December 2019 Chile 113 REER NEER MNT/USD Kasakhstan 3% 111 Mongolia 109 3% Depreciation 107 Euro 2% 105 China 2% 103 101 South Africa 1% 99 Botswana 1% 97 95 Malaysia -1% 93 Japan -3% 91 89 Indonesia -3% Aug-18 Aug-19 Oct-18 Oct-19 Jun-18 Jun-19 Apr-18 Apr-19 Dec-17 Feb-18 Dec-18 Feb-19 Dec-19 Russia - 7% Sources: Balance-of-payments/external debt statistics Sources: BoM; World Bank staff estimates. (Bank of Mongolia); World Bank staff estimates. 1920 19 During this period (July–August), Chile’s currency depreciated by 5 percent, Russia’s by 4.1 percent, and China’s by 3 percent. 20 According to Gan-Ochir and Munkhbayar (2019), the real exchange rate was undervalued for the period 2000-03 and 2007-08, and there is a sign that the underestimation of the domestic real exchange rate started to occur in the end of 2018. | 25 | Macroeconomic Stability and Credit Growth Box I.4. Exchange rate development and FX intervention in Mongolia The Bank of Mongolia (BoM), the country’s central bank, has been intervening in the foreign exchange market to stabilize the tugrug in response to recent market reaction to the gray listing by the Financial Action Task Force (FATF). Since April 2018, the exchange rate has come back under depreciation pressure, facilitated by an easy credit policy. The tugrug depreciated by 10.5 percent between March 2018 and March 2019, sliding back to MNT 2,644 per U.S. dollar. Between April and September 2019, the exchange rate was relatively stable amid strong export performance, robust FDI, and slowing import growth despite a strong U.S. dollar. However, the tugrug depreciated sharply in early October with the announcement of Mongolia’s inclusion on the gray list by the FATF. As the exchange rate approached MNT 2,700 per U.S. dollar, the BoM increased its foreign exchange (FX) intervention. In Q4 2019, BoM intervened to the tune of US$1 billion, a historical high (figure I.28). Figure I.28. Recent foreign exchange market Figure I.29. FX loans have consistently declined interventions in historical perspective since September 2018 FX deposits/total deposits FX loans/total loans The BoM's net sales of foreign exchange: 40% 1,100 3,000 LHS 900 MNT/US$ (eop): RHS 35% 2,500 700 30% 2,000 500 USD millions 25% 300 1,500 100 20% 1,000 -100 15% 500 -300 10% -500 0 Dec-13 Dec-14 Dec-15 Dec-16 Dec-17 Dec-18 Dec-19 Jun-14 Jun-15 Jun-16 Jun-17 Jun-18 Jun-19 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Sources: Balance of payments/external debt statistics Sources: Banking sector balance sheet statistics from Bank of Mongolia; World Bank staff estimates. from Bank of Mongolia; World Bank staff estimates. Note: LHS = left-hand side; RHS = right-hand side. In addition, the BoM recently raised FX reserve requirements to ease depreciation pressure on the tugrug. On December 20, 2019, the Monetary Policy Committee maintained the central bank’s policy rate at 11 percent amid moderate inflation and a slowing global economy. The BoM argued that in the context of high FX holdings, the interest rate channel of monetary transmission is impaired. Hence, to discourage FX demand, the BoM further increased the reserve requirement ratio on FX deposits to 15 percent from 12 percent. High provisioning for FX loans, on the other hand, has led to a sharp decline in their share in total credit since September 2018 (figure I.29). Given limited reserves, the BoM will not be able to sustain recent levels of intervention in the FX market. If market participants anticipate this, the mere fact of intervention may prompt additional FX demand, as prospective buyers expect future devaluation, rendering intervention less effective. A policy to discourage dollarization would need to be based on gradually moving towards a more flexible exchange rate, so that depositors and lenders know they face a two-sided bet. | 26 | Mongolia economic update A6. Monetary policy: Monetary conditions have gradually tightened, but risks to financial sector sustainability remain high Monetary policy stance has improved since late 2018, lowering the risks to the balance of payments. Monetary policy was tightened in late 2018 and the BoM reaffirmed its commitment to a more flexible exchange rate. This followed a credit boom in 2018, when domestic credit grew at 26.5 percent in 2018, fueling rising imports and growing external imbalances. The sharp rise of credit reflected a series of factors including the lagged response to lower policy rates, improved market sentiment, and an apparent exuberance following the findings of the Asset Quality Review, with a lower-than-expected capital shortfall. Monetary authorities’ response to the surge in imports and credit growth included a series of macroprudential measures aimed at limiting the debt-to-income ratio of individual borrowers from as high as 100 percent to 60 percent, reducing the maturity on non-mortgage household loans, and raising the risk rating on unhedged foreign currency borrowing. In addition, the BoM raised the policy rate by 100 basis points to 11 percent in November 2018.21 As a result of these efforts, credit growth moderated to less than 5 percent (y/y) at the end of 2019 (figure I.30 and figure I.31). Figure I.30. Monetary policy was tightened in late Figure I.31. The macroprudential measures have 2018… contained credit and import growth Consumer Loan Growth (y/y) Consump �on imports, y/y CPI Inflation (y/y): LHS 28% Credit growth: LHS 17% 35% 25% Policy Rate: RHS 15% 22% 20% 19% 13% 16% 11% 13% 5% 9% 10% 7% 7% -10% 4% 5% 1% 3% -2% -25% Sep-16 Sep-17 Sep-18 Sep-19 Mar-16 Mar-17 Mar-18 Mar-19 Jun-16 Jun-17 Jun-18 Jun-19 Dec-15 Dec-16 Dec-17 Dec-18 Dec-19 1% Jun-16 Jun-17 Jun-18 Jun-19 Dec-15 Dec-16 Dec-17 Dec-18 Dec-19 Mar-16 Mar-17 Mar-18 Mar-19 Sep-16 Sep-17 Sep-18 Sep-19 -5% -8% -1% Sources: BoM; World Bank staff estimates. Note: LHS = left-hand side; RHS = right-hand side. Interest rates have been on a downward trend. Interbank rates have closely tracked changes in the BoM’s policy rate, declining substantially between mid-2016 and end 2018 and increasing moderately since then (figure 1.32). Weighted average lending rates of new domestic currency loans fell to 17 percent in 2019 from about 17.2 percent in 2018 and 19 percent in 2017. Domestic currency deposit rates—which were stable at around 13 percent on average during most of 2017— fell to 12 percent in 2018 and to 10.5 percent in 2019 (figure I.33).22 21 Although introduced in late 2018, the effectiveness date of the above macroprudential measures was January 2019. Banks massively front-loaded loan issuances in the last quarter of 2018 to take advantage of the delayed effectiveness of the measures. The debt service to income, set at 60 percent, remains higher than the international standards. 22 Nearly flat deposit rates reflect strong competition among banks for large deposits concentrated in the hands of a few depositors. | 27 | Macroeconomic Stability and Credit Growth Figure I.32. Interbank rates have increased Figure I.33. Average lending rates have gradually moderately in tandem with the policy rate… declined amid relatively robust funding conditions Interbank market rates and BoM policy rate (%) Weighted average lending and deposit rates (domestic 12 week T-bills Interbank Deposit Rate currency, %) Overnight lending rate BoM policy rate Interest spread (lending - deposit rate, bps): RHS Average bank lending rate 18 Average bank deposit rate BoM policy rate 16 21 800 19 700 14 17 600 12 15 500 13 400 10 11 300 8 9 200 7 100 6 5 0 Aug-15 Aug-16 Aug-17 Aug-18 Aug-19 Apr-16 Apr-17 Apr-18 Apr-19 Dec-15 Dec-16 Dec-17 Dec-18 Dec-19 Aug-16 Aug-17 Aug-18 Aug-19 Apr-16 Apr-17 Apr-18 Apr-19 Dec-16 Dec-17 Dec-18 Dec-19 Source: BoM. Money supply (M2) growth has decelerated in line with slowing credit. Broad money increased by 7 percent in 2019, down from 23 percent in 2018 (figure 1.34). On the asset side, credit to the public sector declined, while net foreign assets became a main driver of money supply growth. Meanwhile credit to the non-bank private sector decelerated. On the liability side, money supply was mainly driven by the growth of local currency deposits, which averaged nearly 25 percent in 2018 and 13 percent during 2019 (figure I.35). Figure I.34. The deceleration in broad money was Figure I.35. … and decelerating local currency driven by slowing credit … deposits M2 growth decomposition by assets (y/y, %), December M2 growth decomposition by liability (y/y, %), April 2016–December 2019 2016–December 2019 Non-bank private sector Govt & Public Sector 40% Currency outside banks MNT deposits Net foreign assets M2 growth (%, y/y) Foreign currency deposits M2 growth (y/y, %) 65% Net domestic credit (y/y,%) 35% 55% 30% 45% 25% 35% 25% 20% 15% 15% 5% 10% -5% 5% -15% -25% 0% Aug-16 Aug-17 Aug-18 Aug-19 Apr-16 Apr-17 Apr-18 Apr-19 Dec-16 Dec-17 Dec-18 Dec-19 Oct-17 Oct-18 Oct-19 Apr-17 Jun-17 Apr-18 Jun-18 Apr-19 Jun-19 Dec-16 Dec-17 Dec-18 Dec-19 Feb-17 Feb-18 Feb-19 Aug-17 Aug-18 Aug-19 -5% -10% Sources: BoM; World Bank staff estimates. | 28 | Mongolia economic update The quality of bank assets remains weak The quality of banking sector assets remains Figure I.36. FigureI.36. Bank Bank asset asset quality quality remains remains weak… weak… Figure I.37. weak, with a high level of nonperforming 13 NPLs (% of Total Loans) 16 Corpo loans. Between 2014 and 2018, NPLs in the Past Due Loans (% of Total Loans) 14 banking sector more than doubled and, by 12 official accounts, have remained above 10.1 9 10 percent in 2019 (figure I.36). Private sector 23 8 loans have accounted for the bulk of bad 5 6 loans despite the strong economic recovery 4 2 in the past two years. The persistently high 0 level of NPLs suggests that underlying capital 1 Apr-13 Aug-13 Apr-14 Dec-13 Dec-14 Dec-15 Dec-16 Dec-17 Dec-18 Dec-19 Jun-14 Jun-15 Jun-16 Jun-17 Jun-18 Jun-19 shortfalls identified during the Asset Quality Review (AQR) remain, raising the urgency of Sources: BoM; World Sources: BoM; World Bank staff estimates. Bank staff estimates. Note: NPLs = nonperforming loans. completing the bank recapitalization program. Note: NPLs = nonperforming loans. Ensuring that the banks are fully capitalized sector. will be crucial to restore confidence in the financialBox recently, there More Management I.5. Asset have been Company calls – An to effective solution t create an asset management company (AMC) to address Mongolia’s nonperforming loan issues, Mongoliaweaknesses which may not be appropriate given lingering structural is considering in the banking creating (box sectorManagement an Asset Comp I.5). problems in the banking sector, including the rise in nonperfor Mongolian NPLs is not easy; estimates vary from the official figure o twosolution Box I.5. Asset Management Company – An effective to three to banking thethe times citedproblems? sector number by several local experts. A draft Mongolia is considering creating an Asset Management in Parliament to establish Company (AMC) thetoAMC. addressSince AMCs are costly to est structural benefits need to be carefully problems in the banking sector, including the rise in nonperforming loans. Estimating the size assessed and the preconditions of for su Mongolian NPLs is not easy; estimates vary from the official need figure to be of 10.1 in place. percent can Mongolia in December 2019 to draw on international experience two to three times the number cited by several local experts. A draft resolution was recently introduced in Parliament to establish the AMC. Since AMCs are costly to establish A recent study on and aoperate, their toolkit for expected public AMCs,benefits a building on nine cas need to be carefully assessed and the preconditions for successful establishment of an AMC need to be found that a successful AMC requires certain preconditions. T in place. Mongolia can draw on international experience to design its AMC. comprehensive reform of the banking sector; (b) a systemic problem A recent study on a toolkit for public AMCs, building a diagnostic on nine and critical case mass studies around assets; of impaired from (d) a tradition of ins the world, accountability; and found that a successful AMC requires certain preconditions. These a robust(a) (e) include legal framework for a commitment tobank resolution, comprehensive reform of the banking sector; (b) a systemic problem and public funds at risk;b (c) a solid diagnostic and critical mass of impaired assets; (d) a tradition of Mongolia to be independence institutional seems far from the and public above-mentioned requir accountability; and (e) a robust legal framework for bank resolution, debt recovery, and creditors’ rights. independence and public accountability, recent experience of mana Mongolia seems to be far from the above-mentioned has not been successful (for requirements. example, In terms the Development Bank o of institutional Corporation). There is also a need independence and public accountability, recent experience of management involving the government to improve the regulatory capacit has not been successful (for example, the Development Bank the prudential of Mongolia, supervisor Mongolian Mortgagebanks’ in assessing business models, their Corporation). There is also a need to improve the regulatory capacity the and institutional performance independence of its regulatory and of supervisory the prudentialfunction. Moreove supervisor in assessing banks’ business models, their an instrumentand sustainability asset quality, to liquidate and the performance misidentified NPLs at preferential pricing fo of its regulatory and supervisory function. Moreover,and risk exists of abusing AMCs as an instrument to the investors. liquidate NPLs at preferential pricing for politically well-connected debtors and investors. Another key factor for a successful AMC is a robust legal fram recovery, and creditors’ rights. In Mongolia, although significant p IMF’s Extended Fund Facility program in building banking laws and r standards and improving the capacity of banking supervisors, th prudential supervisor remains weak, and the legal and regulato 23 of NPLs, ranging from the official figure of 10.1 percent to There is considerable debate surrounding the correct sizeimprovements. several multiples of it cited by local experts (box I.5). In addition, an AMC is a remedial solution that should always prevent a buildup of future NPLs. In that respect, | 29 | the prudentia substantial capacity limitations. Addressing more structural issues priority. These include a highly concentrated banking sector (the “too lending to few economic sectors (concentrated lending portfolios po Macroeconomic Stability and Credit Growth Another key factor for a successful AMC is a robust legal framework for bank resolution, debt recovery, and creditors’ rights. In Mongolia, although significant progress was achieved through the IMF’s Extended Fund Facility program in building banking laws and regulations in line with international standards and improving the capacity of banking supervisors, serious shortcomings remain, particularly in the implementation of the new framework and the political independence of the BoM. In addition, an AMC is a remedial solution that should always be accompanied by measures to prevent a buildup of future NPLs. In that respect, the prudential supervisor in Mongolia still faces substantial capacity limitations. Addressing more structural issues of the banking sector should be a priority. These include a highly concentrated banking sector (the “too big to fail” problem), concentrated lending to few economic sectors (concentrated lending portfolios pose risks to asset quality, profitability, and stability of banks), and weak internal capacity to assess emerging risks, sound lending practices, NPL valuation, and collateral valuations. Source: Summarized and adapted from “Public Asset Management Companies: A Toolkit” (World Bank 2016). Note: a. World Bank 2016; b. Establishing an AMC is not warranted unless weaknesses in the financial system are systemic and threaten to put public funds at risk. The public AMC’s role is to limit the ultimate cost to the public sector of resolving financial sector weaknesses, by recovering proceeds from assets that have lost value temporarily. Available liquidity indicators suggest that the banking system as a whole remains liquid (figure I.37). According to BOM, the system-wide capital adequacy ratio is currently above the minimum threshold of 12 percent. But a forensic audit of the recapitalization process suggests raised doubts over the quality of bank capital, suggesting several systemic banks remain severely undercapitalized. Bank reserves stood at about 27 percent of total deposits in December 2019, above the 10.5 percent reserve requirement ratio. Meanwhile, the ratio of liquid assets—including bank reserves, central bank bills, and government bonds—to total bank liabilities was 37.2 percent in December 2019, above the minimum threshold of 25 percent imposed by the central bank. Figure I.37. Banking system appears liquid Figure I.38. Rising currency mismatch in the banking system 40% FX deposits/total deposits FX loans/total loans Bank Reserves/Deposits (%) 40% Liquid Assets/Total Assets(%) 35% 35% 30% 30% 25% 25% 20% 20% 15% 15% 10% 10% Sep-15 Sep-16 Sep-17 Sep-18 Sep-19 Jun-15 Jun-16 Jun-17 Jun-18 Jun-19 Mar-15 Mar-16 Mar-17 Mar-18 Mar-19 Dec-14 Dec-15 Dec-16 Dec-17 Dec-18 Dec-19 Dec-13 Dec-14 Dec-15 Dec-16 Dec-17 Dec-18 Dec-19 Jun-14 Jun-15 Jun-16 Jun-17 Jun-18 Jun-19 Sources: BoM; World Bank staff estimates | 30 | Mongolia economic update The banking system remains exposed to a high risk of currency mismatch (figure I.38). The size of the foreign liabilities of the banking system is more than twice the size of foreign assets, exposing the banks to high currency mismatch risks. However, credit risks associated with exchange rate fluctuation have significantly declined, as foreign currency loans accounted for 12.6 percent of total loans at the end of 2019, but a significant portion of foreign currency loans were likely provided to unhedged borrowers. This substantial decline in FX loans is mainly attributed to macroprudential measures by the monetary authorities (high provisioning for FX loans), especially since September 2018 when the measure was introduced. B. Outlook and Risks With the global economy slowing and commodity prices falling, growth in Mongolia is likely to decelerate as well albeit remaining at par with its main regional peers (box I.7). Supported by a stable macroeconomic environment, robust domestic demand, large inflows of foreign direct investment, and relatively robust commodity exports, economic growth is projected to be around 5.8 percent in 2019 (revised downward from our initial projections of 6.9 percent). This revision is explained by the recent fall in copper prices, lower gold content of OT’s copper production, and falling coal production, especially during November–December 2019. Average growth over the medium term (2020- 22) is expected to be around 5.6 percent, mainly supported by the mining sector, especially with imminent developments of the Tavan Tolgoi mining project, and the remaining export potential of Oyu Tolgoi phase 1, despite a delay in phase 2. Buoyed by its links with mining, the services sector is also expected to enjoy robust growth (table I.2). Monetary policy is expected remain relatively tight, given persistent inflation and external sector pressures. Inflation is likely to be within the central bank’s target mainly due to moderating domestic demand, despite the immediate and lagged impact of oil and food prices (mainly meat prices) and exchange rate depreciation. Moreover, increasing public wages and relatively higher fiscal spending in 2019–20 compared to the previous two years could generate inflationary pressures. Our base case is built on the continued commitment of the monetary authorities to price stability keeping inflation within the central bank’s target, which would eventually help lower inflation expectations. Interventions in the foreign exchange market are expected to be limited to smoothing excessive volatility, allowing more flexibility in exchange rate movements, and rebuilding international reserves. The 2020 budget and the medium-term fiscal framework are broadly consistent with fiscal consolidation and the debt reduction objective. The 2020–22 Medium-Term Fiscal Framework projects an overall budget surplus of about 1 percent of GDP by 2022.24 We estimate the fiscal deficit (using the World Bank/IMF definition) more conservatively (given less optimistic revenue projections) to average around 0.3 percent of GDP during 2020–22. Financial support from donors is expected to continue to contribute to the fiscal consolidation process. Concessional budget support loans from multilateral and bilateral donors would allow expensive domestic debt to be replaced. Interest payments, reflecting projected concessional financing, are forecast to decline to about 1.5 percent of GDP during 2020–22 from 24 According to the Fiscal Stability Law, the structural budget balance is defined as the difference between structural rev- enue (that is, total government revenue minus transfers to the Fiscal Stabilization Fund and the Future Heritage Fund) and total government spending (excluding projects financed by Development Bank of Mongolia). Meanwhile, the IMF and World Bank monitor the overall fiscal balance, including on the revenue side, proceeds to the saving funds | 31 | Macroeconomic Stability and Credit Growth 2.3 percent of GDP in 2019, 3.2 percent in 2018, and 4.1 percent in 2017.25 Government debt is projected to fall further, supported by real GDP growth and a primary surplus (figure I.39). Assuming fiscal consolidation is sustained – and any revenue shortfalls are compensated by additional spending restraint – the government debt to GDP ratio should continue to decline to around 60 percent of GDP, falling below the target set in the Economic Recovery Program. Compared with its peers, Mongolia’s debt ratio remains elevated, however. Figure I.39. Government debt-to-GDP ratio is Figure I.40. The government is expected to pay a projected to fall further but remains higher than higher amount of debt due on external bonds during many selected peers 2021–24 Debt-to-GDP ratio in percent (US$ million) 90 80 Chile 600.0 2,900.0 Kazakhstan 70 800.0 60 Russia 50 1,000.0 Malaysia 40 30 Peru 20 500.0 Philippines 10 0 Mongolia Total 2024 2021 2022 2023 2019f 2020f 2021f 2022f 2010 2011 2012 2013 2014 2015 2016 2017 2018 China Sources: IMF (Fiscal Monitor, October 2019); World Bank Sources: MoF, BoM; World Bank staff estimates. staff estimates. Despite significant progress in lowering public sector debt, pressures on the balance of payments are likely to continue in the medium term. Although the current account deficit is expected to gradually narrow in 2020–22, an expected slowdown in commodity prices and a relatively large deficit in the services and income accounts will likely keep the current account in the red in the coming years. Unless further external financing is timely and successfully mobilized, a higher current account deficit could easily translate into a deficit in the balance of payments. Fortunately, no large external debt repayments are due on the public sector side until 2021, except the renewal of the People’s Bank of China (PBOC) swap line with the BoM.26 However, a total of US$2.9 billion in sovereign external debt is due in 2021-24 (figure I.40). In addition, delayed implementation of ongoing reforms could affect Mongolia’s sovereign ratings and reduce the odds of refinancing under favorable conditions. 25 One key feature of the government’s debt management strategy is to substitute expensive domestic debt with con- cessional borrowing and foreign debt obtained through refinancing on preferential terms, resulting in a considerable decline in interest payments in 2017–18. 26 Expected public debt repayments during 2020–24 include the following: PBoC swap line (US$1.7 billion, due in June 2020), Mazaalai Bond (US$500 million, due in 2021), Gerege Bond (US$800 million, due in November 2022), Chinggis Bond (US$1 billion, due in December 2022), Samurai Bond (US$268 million, due in May 2023), and Khuraldai Bond (US$600 million, due in March 2024). | 32 | Mongolia economic update Table I.2. Key macroeconomic indicators   2016 2017 2018 2019f 2020f 2021f 2022f    Annual percent change unless indicated otherwise  Real GDP growth, at constant market prices 1.4 5.4 6.8 5.8 5.3 5.6 5.8 Private Consumption -2.2 5.3 6.2 6.3 6.2 6.3 6.4 Government Consumption 10.6 -1.8 0.5 14.0 13.2 4.5 2.8 Gross Fixed Capital Formation 0.5 35.6 22.4 16.0 11.0 8.7 9.3 Exports, Goods and Services 13.8 14.8 13.4 4.6 0.6 4.4 6.5 Imports, Goods and Services 12.7 24.8 20.8 7.1 4.0 5.9 7.7 Real GDP growth, at constant factor prices 1.2 5.3 7.2 5.8 5.3 5.6 5.8 Agriculture 6.2 1.8 4.5 4.5 4.4 4.4 5.0 Industry (incl mining) -0.4 0.7 7.4 3.4 3.1 5.6 5.5 Services 1.1 7.7 5.0 6.8 6.2 6.0 6.7 Inflation (CPI, end-period) 0.9 6.4 8.1 5.2 6.2 7.0 7.3 Current account balance (% of GDP) -6.3 -10.2 -16.9 -12.5 -11.9 -10.9 -9.8 Financial and Capital account (% of GDP) 7.6 23.4 16.7 17.4 12.1 11.0 7.4 Net Foreign Direct Investment (% of GDP)* 1.1 12.7 16.4 15.4 13.1 12.1 8.8 Fiscal Balance (% of GDP)** -15.3 -3.8 2.6 1.4 -1.9 -0.1 1.2 Primary Balance (% of GDP) -10.1 -1.3 5.8 3.7 0.3 1.9 2.7 Debt (% of GDP)*** 87.6 84.7 73.3 68.2 65.2 60.0 57.9 * In 2016, Net FDI number excluded the transactions of OT-2 project financing in May-June, 2016. **DBM spending is excluded from fiscal balance and monitored separately. ***General government debt data excludes SOE's debt and central bank's liability from PBOC swap line. Risks The economic outlook, while positive, is subject to considerable political and macroeconomic risks. The major risks from domestic and external sources include (a) political uncertainty due to the 2020 election, (b) the potential impact of the U.S.A.-China trade deal on the demand for Mongolia’s key export commodities (particularly copper and coal), (c) climate shocks (drought/ flooding, harsh winter), (d) renewal of bottlenecks at the border with China, (e) limited progress on recapitalization of the banking sector, (f) impact of the recent inclusion of Mongolia in the “gray listing” of the Financial Action Task Force, and – most recently - (g) the impact of the coronavirus epidemic. In the short run, pre-election fiscal spending increases (due to an accelerated capital budget execution in Q4 2019 and a relatively less stringent 2020 budget), a decline in confidence following Mongolia’s recent “gray listing,” and concerns over the banking sector, could increase the demand for U.S. dollars and put pressure on limited international reserves. In addition, the government’s decision to write off pension-backed loans is a threat to recent fiscal gains (box I.6). Over the medium term, political uncertainty and its impact on economic management, the large external financing requirements during 2020–24, and potential negative external trade shocks present the highest risks. Moreover, given strong trade and investment links with China, additional downside risk to economic outlook would depend on the magnitude and duration of the impact of coronavirus outbreak in China. Some possible transmission channels to the Mongolian economy could be through a decline in commodity prices and demand, and a slowdown in activities such as tourism and meat processing following the closing of key border posts with China. Some signs of pressures recently observed on food prices may be fueled by panic related to the percieved risk of the coronavirus epidemic spreading to Mongolia On the positive side, progress on the U.S.A- Mongolia trade agreement following the July 2019 visit of the Mongolian president to the U.S.A could hlep support investor confidence. | 33 | Macroeconomic Stability and Credit Growth Box I.6. The macro-fiscal implications of writing off loans of pensioners The National Security Council recently announced the decision to write off pension-backed loans. The Council argued that its decision aims to improve the inclusiveness in sharing the proceeds of the country’s natural resources as indicated in one of key amendments to the constitution, approved in December 2019. In January 2020, the Parliament passed two laws describing the pension loans forgiveness and its implementation strategy. According to the approved laws, pension loans up to MNT 6 million (about US$2200) will be written off. In addition, pensioners with no pension-backed loans will be eligible to receive securities equivalent to MNT 1 million (US$365) which can be liquidated starting from May 2021. Regarding the funding scheme, Erdenes Mongol LLC (a state-owned enterprise) is expected to issue a MNT 900 billion bond (US$330 million) collateralizing future proceeds from one of its mines (Salkhit silver mine).a The bond would be swapped for the outstanding pension loans with the commercial banks. The banks can make repurchase agreements (usually called repos) with the central bank, and any potential loss incurred by the central bank in relation to this bond is expected to be compensated by the government.b While the goal of increasing inclusion in Mongolia is laudable, the proposed write-off has several drawbacks. Central bank financing of any public sector losses incurred in the above transaction can be viewed as a resumption of quasi-fiscal activities, against the principles of the central bank law. According to the newly approved law on the one-time paying off of pension-backed loans by the State (Clauses 5.4 and 5.7), the central bank can buy including through open market operations, the corporate bond issued by Erdenes Mongol to write-off the pension loans. This could contravene the principles of the central bank law on quasi fiscal activities and the spirit of the commitments made under the Economic Recovery Program. The quasi-fiscal costs of fully monetizing the bond issue could be up to 2.5 percent of GDP. The implementation of the above scheme would pose substantial risks to the fiscal gains of the past two years. Since the government is liable to compensate for any loss by the central bank incurred on the bond, it adds to the government’s contingent liability. Moreover, borrowing against future mining revenues is also ill-advised, given the volatile nature of commodity prices. In case of a loss in the value of the bond in 2020, it would most likely create a burden on the budget for 2021, and the fiscal gains of 2018 and 2019 could be eroded. Note: a. A five-year bond of MNT 900 billion with a 6 percent annual coupon rate would be guaranteed by the Development Bank of Mongolia. The weighted average interest rate or return on outstanding pension loans is 17 percent, and the central bank’s policy rate is 11 percent as of end-2019. In addition, the approved law indicates that to reduce the above interest rate differential, the government can deposit at subsidized rates its social insurance fund reserves in commercial banks with pension loan assets. b. Repurchase agreements (repos) are instruments used by the central bank to address commercial bank short-term liquidity issues. The maximum maturity in Mongolia is 12 weeks. Significant vulnerabilities remain. The most important vulnerabilities pertain to the path for GDP growth, as well as fiscal and monetary policy management and the impact on the exchange rate. Slower economic growth or a higher real exchange rate depreciation could push the debt- to-GDP-ratio back above 80 percent in the medium term. Mongolia’s net international investment position remains weak, primarily reflecting limited foreign assets at the central bank (gross foreign exchange reserves are only 25 percent of GDP) and very high external debt (which has risen to 240 percent of GDP from just 40 percent in the last 10 years). Moreover, Mongolia faces substantial external debt repayments amounting to US$14.3 billion during 2020–24, of which 34 percent is public sector debt including a US$1.8 billion swap line from China coming due in 2020. The conditions under which these debts can be rolled over are critical to the country’s balance-of- payments outlook and long-term debt sustainability. | 34 | Mongolia economic update Box I.7. Global and regional outlook and risksa Global growth in 2019 has been downgraded to 2.4 percent, 0.2 percentage points below previous forecasts, reflecting continued weakness in global trade and investment. Growth is projected to gradually rise to 2.5 percent by 2020 as investment and trade gradually recover. These estimates pre-date the coronavirus outbreak, the impact of which has not yet been fully assessed. Global economic conditions are expected to improve marginally over a longer forecast horizon, with the global growth rate projected to increase to 2.6 percent in 2021 and to 2.7 percent in 2022. After slowing to 1.4 percent in 2020, growth in advanced economies is projected to improve slightly to 1.5 percent in both 2021 and 2022. Growth in emerging market and developing economies (EMDEs) is expected to recover over the next two years. This projection is predicated on a rebound in several large EMDEs, some of which are emerging from deep recessions or sharp slowdowns but remain fragile. Figure I.41. Real GDP growth (percent) Figure I.42. World commodity price forecast (Index=nominal U.S. dollars, 2016=100) 7 Coal, Australia (Thermal) EMDEs World Advanced economies 170 Crude oil 6 160 Copper 5 150 Gold 140 4 130 3 120 2 110 1 100 0 90 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019f 2020f 2021f 2022f 2019f 2020f 2021f 2016 2017 2018 Sources: World Bank 2019a, 2019b. Global economic conditions are expected to remain challenging over the forecast period figure I.41). A further escalation of trade tensions involving major economies could lead to a sharp increase in trade barriers and weigh on confidence and investment. Global financing conditions are expected to remain volatile, even if generally more supportive. This reflects the fact that major central banks have provided accommodation in response to softening economic prospects. However, EMDEs with low credit ratings have not benefited from the global decline in borrowing costs. Portfolio outflows resumed in the second half of last year. A rising share of EMDE currencies are at their lowest level against the U.S. dollar in a decade. | 35 | Macroeconomic Stability and Credit Growth Financial market volatility will continue to have the strongest impact on countries with high vulnerabilities, weak growth prospects, and elevated policy uncertainty. Oil prices averaged US$60 per barrel in 2019, a 12 percent decline from 2018 and 9 percentage points below previous projections, with production cuts by the Organization of the Petroleum Exporting Countries (OPEC) and its partners. Oil prices are expected to decline slightly to an average of US$59 per barrel in 2020–21, with high uncertainty around the outlook (figure I.42). Prices for most base metals weakened in the second half of 2019, primarily reflecting weaker global growth and elevated trade tensions. Despite subdued industrial commodity demand, metals prices are expected to decline further in 2020. Agricultural prices declined in the second half of 2019 on improved weather conditions and elevated stock levels for grains. Agricultural prices are expected to stabilize in 2020, with risks to the forecast broadly balanced. Growth in the region is projected to slow Figure I.43. Regional growth from an estimated 5.8 percent in 2019 to 5.7 percent in 2020, and to moderate 8 2018 further to 5.6 percent in 2021–22 (figure 7 2019f I.43). Easier financing conditions and fiscal 6 2020f policy support will partly mitigate the 5 negative impact of ongoing trade tensions 4 and associated policy uncertainty. In China, 3 growth is expected to slow gradually, from 2 an estimated 6.1 percent in 2019, to 5.9 1 percent in 2020, and to 5.7 percent by 2022. 0 In the rest of the region, growth is expected Mongolia Vietnam China Myanmar Cambodia Lao PDR Malaysia Thailand to slightly recover to 4.9 percent in 2020 and firm further to 5 percent in 2021–22. Source: World Bank 2019a. Downside risks to regional growth have intensified. They include a contraction in global trade due to a further escalation of trade tensions; a sharper- than-expected slowdown in major economies; and a sudden reversal of capital flows due to an abrupt deterioration in financing conditions, investor sentiment, or geopolitical relations. An upside risk to the forecast is the possibility of a de-escalation of trade tensions between the United States and China, which would likely boost regional trade. Risks to the outlook. Global growth, which weakened to an estimated 2.4 percent in 2019, is projected to edge up to 2.5 percent this year, following an expected recovery of trade and investment. However, risks to the global outlook are firmly on the downside. Rising trade barriers and protracted policy uncertainty could further weigh on global activity. Amid financial sector vulnerabilities, major economies could slow more than expected. EMDEs remain at risk of financial stress, especially those with elevated debt, while some EMDE regions could be affected by geopolitical tensions, social unrest, large swings in commodity prices, or increasingly volatile weather patterns. On the upside, progress in the resolution of ongoing trade tensions could mitigate global policy uncertainty and bolster activity. Source: World Bank 2019a. Note: a. This box was prepared by Ekaterine Vashakmadze, Senior Country Economist, World Bank. | 36 | Mongolia economic update ii. UNDERSTANDING MONGOLIA’S BANKING SECTOR | 37 | Macroeconomic Stability and Credit Growth CREDIT DEVELOPMENT Private sector credit growth in Mongolia has recently moderated. But the rapid credit expansion in the past several years and the slow progress on bank recapitalization process call for heightened vigilance. While monetary policy tightening since late 2018 has helped to contain credit growth, further macroprudential measures can be effective to preempt emerging risks. Mongolia will benefit from ongoing efforts to strengthen financial supervision and its macroprudential framework, and to bolster its crisis management and resolution frameworks. A. Context The importance of the banking sector has increased over the past decade. Total assets in the banking sector reached 97 percent of GDP in 2019 compared to 56 percent in 2008. Mongolia’s financial sector is highly dominated Figure II.1. Mongolia ranks well in financial by the banking system (close to 90 percent as sector development of H1 2019),27 and consequently, loans from 0.7 Financial Development Index: Selected Countries commercial banks are the main financing source 0.6 Average of all countries Asia and Pacific for both individuals and businesses. The banking 0.5 sector consists of 14 banks (now 13 banks with 0.4 one bank in the resolution process since April 0.3 2019) with 1,413 branches and 1 development 0.2 bank. Total outstanding loans in the banking 0.1 sector are equivalent to 56.3 percent of GDP 0 Botswana Malaysia Moldova Cambodia Tunisia Egypt, Arab Rep. Bolivia Ukraine Honduras Peru Philippines Albania Indonesia Pakistan Suriname Myanmar Kyrgyz Republic Kazakhstan Mongolia Bangladesh Chile Nigeria in 2019. Meanwhile, total deposits and current accounts (individuals and companies) represent 51 percent of GDP. Sources: IMF; World Economic Forum Financial Development Index (2017); World Bank staff In the last decade, Mongolia experienced estimates. rapid growth of the banking sector. This was driven by growing incomes from mining resource rents, which generated a greater domestic source of funding and created opportunities for financial intermediation and deepening. Mongolia consequently ranks well in terms of financial sector development among comparable countries in the region (figure II.1). Mongolia faces structural characteristics that challenge its financial sector development and integrity. These challenges interact with money laundering risks. The Mongolian banking sector is highly concentrated in five large banks, which as of Q3 2019 together held around 87 percent of total banking assets. There are six systemically important banks and seven small banks.28 The deposit base is narrow and concentrated among a few large clients. 27 The total financial sector (banks – 88 percent, NBFI’s – 1.4 percent, stock market – 6.6 percent, and insurance market – about 1 percent) accounted for 122 percent of GDP in H1 2019. 28 A bank that holds over 5 percent of the banking system assets for the past 6 months is considered a systemically important bank. | 38 | Mongolia economic update The loan portfolio remains concentrated in five sectors that have strong links with the mining sector. Loan portfolios in these economic sectors account for about half of total loans, while household loans including mortgage, consumption, and salary loans constitute the remaining half. Reflecting a sharp economic downturn in 2015-16, banking sector NPLs rose from less than 3 percent before 2014, to about 9 percent in 2016 and to more than 10 percent in 2018. Despite the subsequent economic recovery, NPLs have remained persistently high, raising questions over the underlying health of the sector. In January 2018, as part of the IMF’s Extended Fund Facility program, third-party auditors completed an Asset Quality Review of the banking sector on behalf of the BoM, which revealed some serious weaknesses in the current supervision practice and raised some red flags. B. Recent developments in bank credit B1. Current status of banking sector credit The credit-to-GDP ratio is lower than the average for upper middle-income countries suggesting some financial deepening potential. The credit-to-GDP ratio averaged over 70 percent in 2016–17, which is comparable to that of the lower-and middle-income groups (64 percent). It is lower than the average in upper middle-income countries but high relative to some of its regional peers (Indonesia, Kazakhstan, and Myanmar) (figure II.2). The total credit-to-GDP ratio has increased significantly in the past 15 years, driven by consumer loans (figure II.3).29 The average credit-to-GDP ratio increased from around 30 percent in 2006 to about 40 percent in 2007–12. Then it jumped significantly in 2013 and hovered around 60 percent, reflecting the subsidized housing mortgage program of the BoM. By 2019, individual loans accounted for about 58 percent of total loans, while corporate loans accounted for 42 percent, their respective shares reversed relative to the period before 2013. As housing mortgage loan growth started to decelerate since early 2017, non-mortgage individual loans (especially salary loans) were a key driver of the high growth of individual loans in 2017–18. 29 Total credit is defined as the credit extended by banking institutions to both firms and households, and the outstanding amount of mortgage credit of the Mongolian Mortgage Corporation. | 39 | Macroeconomic Stability and Credit Growth Figure II.2. Mongolia’s credit-to-GDP ratio is Figure II.3. …relative to inflation, it has been higher comparable to that for lower middle-income than the credit-to-GDP ratio for some regional countries… comparators Credit-to-GDP ratio in selected countries, average of Credit-to-GDP ratio by institutional sector, 2004–18 2016–17 (in percent) 16 180 15 Egypt 14 Nigeria Suriname 160 Upper middle income - 153% 13 Ukraine Inflation (%) 140 12 High income - 136% 11 Inflation: 120 Upper Middle 10 100 9 Myanmar Income 8 Mongolia 80 Lower middle income - 64% 7 60 6 Moldava Tunisia Low income - 43% 40 5 Kazakhstan 4 4 Vietnam 20 3 2 Chile Malaysia 0 1 Cambodia Kyrgyz Mongolia Albania Vietnam El Salvador Moldova Kazakhstan Botswana Georgia Bhutan Kyrgyz Republic Malaysia Egypt, Arab Rep. Tunisia Cambodia Myanmar India Pakistan Chile Indonesia Peru Ukraine Bangladesh Nigeria Philippines Bolivia Honduras Suriname 0 0 20 40 60 80 100 120 140 160 Credit to GDP ratio (%) Sources: World Bank (World Development Indicators); Sources: World Bank (World Development Indicators); IMF. IMF. Mongolia also stands out with high credit growth (figure II.4). Credit growth averaged around 22 percent during 2009–18 compared to 12.1 percent in its peer countries. Moreover, high credit growth was accompanied by weak lending standards, which resulted in excessive leverage (individuals and corporates) asset price bubbles (due to rapid growth of mortgage loans in 2013– 15), and relatively high inflation. Figure II.4. Credit growth in Mongolia is high relative Figure II.5. Mongolia has experienced several to peers episodes of high credit growth Credit growth (in percent, average of 2009–18) (in percent, y/y) 40 35 80 Nominal Credit Growth 30 Real Credit Growth 25 22 60 20 15 40 10 5 20 0 0 Myanmar Indonesia Nigeria Bhutan Suriname Kyrgyz Republic Cambodia Tunisia Ukraine Mongolia Philippines Peru Botswana Albania Bolivia Pakistan Honduras Bangladesh Malaysia Moldova Egypt Chile Kazakhstan -20 Apr-05 Mar-06 Dec-08 Sep-11 Aug-12 May-15 Apr-16 Mar-17 Dec-19 Feb-07 Nov-09 Oct-10 Feb-18 Jul-13 Jan-08 Jun-14 Jan-19 Sources: World Bank (World Development Indicators); IMF. Underlying strong average credit growth have been several sharp credit cycles, most recently during the second half of 2018. Mongolia has experienced several episodes of rapid credit expansion in the past, including episodes in 2005–08, 2010–11, and 2013–14 (figure II.5). The recent rapid credit expansion was channeled mainly to individuals (figure II.6). Over the past decade, bank credit in Mongolia has usually been a key driver of domestically financed private investment (total domestic investment excluding government investment) (figure II.7). In 2018, bank credit to the economy grew annually by about 26.5 percent in nominal terms and 15.4 percent in real terms. | 40 | Mongolia economic update Figure II.6. Most of the credits were channeled to Figure II.7. Credit expansion was mostly beneficial individuals to domestic private investment Credit to Private Sector (December 2015=100) Credit Intensity and Domestic Private Investment 100 190 90 Corporate Loans 80 170 Individual Loans Domestically funded-private 70 investment growth 150 60 50 130 40 30 110 20 10 90 0 Feb-16 Feb-17 Feb-18 Feb-19 Aug-16 Aug-17 Aug-18 Aug-19 Jun-16 Oct-16 Jun-17 Oct-17 Jun-18 Oct-18 Jun-19 Oct-19 Dec-15 Dec-16 Dec-17 Dec-18 Dec-19 Apr-16 Apr-17 Apr-18 Apr-19 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 Credit Intesity (Nominal private sector credit growth/NGDP growth) Sources: BoM; World Bank staff estimates. Note: NGDP = Nominal Gross Domestic Product. A tightening of monetary and credit policies brought the most recent credit boom to a halt in 2019 (figures II.8 and II.9). The BoM’s response to the surge in imports and credit growth in 2018 included a series of macroprudential measures aimed at limiting the debt-to-income ratio, reducing the maturity of non-mortgage household loans, and raising the risk weights on unhedged foreign exchange borrowing. In addition, the BoM adopted a tightening bias in November 2018 as it raised its policy rate by 100 basis points to 11 percent. Partly in response to these policy actions, credit growth to individuals and the private sector fell from its recent peak of 35 percent and 17 percent, respectively, in December 2018, to 4 percent and 6 percent, respectively, in December 2019. Given persistent external pressures, monetary and credit policies will likely need to remain tight. Figure II.8. A sharp deceleration of credit growth Figure II.9. The quality of corporate loans has pushed the credit-to-GDP ratio below its long-term significantly deteriorated trend 60 18 Corporate Loans Individuals 16 55 14 50 12 10 45 8 40 6 Actual Credit-to-GDP Ratio (%) 4 35 HP Trend (1 sided, Lambda = 1600) 2 30 0 Aug-13 Aug-14 Aug-15 Aug-16 Aug-17 Aug-18 Aug-19 Apr-13 Dec-13 Apr-14 Dec-14 Apr-15 Dec-15 Apr-16 Dec-16 Apr-17 Dec-17 Apr-18 Dec-18 Apr-19 Dec-19 Dec-08 Dec-09 Dec-10 Dec-11 Dec-12 Dec-13 Dec-14 Dec-15 Dec-16 Dec-17 Dec-18 Dec-19 Jun-08 Jun-09 Jun-10 Jun-11 Jun-12 Jun-13 Jun-14 Jun-15 Jun-16 Jun-17 Jun-18 Jun-19 Sources: BoM, World Bank staff estimates. Sources: BoM, World Bank staff estimates. Note: The credit-to-GDP gap is defined as the difference between the credit-to-GDP ratio and its long-run trend. It captures the buildup of excessive credit in a reduced form (Definition of the Bank of International Settlement). HP = Hodrick – Prescott Filter. | 41 | Macroeconomic Stability and Credit Growth NPLs are concentrated in the corporate sector, reflecting weak lending standards during previous credit expansions and suggesting underlying financial sector vulnerabilities (figure II.10). The NPL ratio of the corporate sector reached 19 percent, its highest level, during the 2008– 09 global financial crisis. But in December 2019, the NPL ratio of the corporate sector approached 18 percent again. Corporate sector NPLs are mainly concentrated in a few risky sectors such as mining, manufacturing, construction, and trade, which are highly vulnerable to a volatile mining sector. By contrast, the NPL ratio of the household sector has been stable at around 3.5 percent over the past four years. Loan loss provisioning (LLP) remains low for many key sectors.30 The mining, construction, trade, and real estate sectors account for over 80 percent of total NPLs, but LLP of these sectors ranged between 71 and 76 percent as of December 2019. However, manufacturing sector LLP stood at 83 percent. Meanwhile, pension loans have exhibited LLP equaling or exceeding 100 percent (figure II.11).31 Figure II.10. Provisioning remains below average for Figure II.11. Provision of individual loans increased many key sectors Corporate loans Total Corporate loans 79% Total individual loans 82% Other 70% Salary loans 74% Trade 70% Consump�on loans 75% Real estate 71% SMEs 85% Mining 75% Construc�on 76% Mortgage 86% Manufacturing 83% Pension loans 131% Source: BoM. Source: BoM. Note: SMEs = Small and Medium-Sized Enterprises. B2. Main drivers of credit development Bank deposits, GDP growth, and commodity prices were the key drivers of credit growth. A recent empirical assessment by the BoM and the World Bank used a Vector Autoregression model based on quarterly data spanning 2000–18 to define factors that affect private sector credit growth.32 It found that commodity prices and economic growth were the strongest contributing factors to private credit growth, reflecting the cyclical characteristic of credit growth. However, the BoM also actively supporting faster credit growth by engaging in quasi-fiscal activities, many of which have since been withdrawn. In this regard, credit growth and macroeconomic shocks 30 The provision coverage ratio is used to determine how banks are protected from possible losses on nonperforming credits. 31 NPLs for pension loans have been relatively low, which causes high LLPs. 32 To define the main drivers of credit development in Mongolia, a Vector Autoregression model was employed using quarterly data between Q1 2000 and Q4 2018. The endogenous variables included real GDP; real deposits; real credit; the nonperforming loan ratio; the interbank market rate; the nominal effective exchange rate; and the exogenous variable, the export price index. The main results, including the impulse response functions and variance decomposition, are shown in the annex. | 42 | Mongolia economic update have been closely intertwined, often feeding on each other to amplify the boom and bust cycles in Mongolia. B3. Growing importance of non-bank Figure II.12. Growing importance of non-bank financial institutions financial institutions (NBFIs) NBFI's Credit (% of GDP): RHS NBFI's Credit Growth (y/y, %): LHS Risks from volatile credit growth in Mongolia 60 4.0 are further complicated by the growing 50 3.5 3.0 importance of non-bank financial institutions 40 2.5 (NBFIs) (figure II.12). While banks are the main 30 2.0 providers of private credit, loans by NBFIs 20 1.5 increased as a share of GDP from 1.2 percent 10 1.0 in 2010 to about 3.7 percent in September 0.5 0 - 2019 (loans increased by almost 13 times, from Jun-11 Jun-12 Jun-13 Jun-14 Jun-15 Jun-16 Jun-17 Jun-18 Jun-19 Dec-11 Dec-12 Dec-13 Dec-14 Dec-15 Dec-16 Dec-17 Dec-18 Mar-11 Mar-12 Mar-13 Mar-14 Mar-15 Mar-16 Mar-17 Mar-18 Mar-19 Sep-11 Sep-12 Sep-13 Sep-14 Sep-15 Sep-16 Sep-17 Sep-18 Sep-19 about MNT 100 billion to about MNT 1.3 trillion during this period). While this sector is too small Source: BoM. to constitute a systemic risk, vulnerabilities of NBFIs should be closely monitored and addressed, as NBFIs largely serve the riskier part of the population.33 Particularly, following the macroprudential measures by the BoM since mid-2018, credit (especially individual loans) may have gradually migrated to the less regulated non-banking sectors, including NBFIs. Thus, regulation and supervision of the NBFIs need to be strengthened and, if possible, coverage of macroprudential limits should be expanded to NBFIs. C. Sectoral allocation C 1. Household sector Individual credit is dominated by mortgage loans and overdrafts. With the housing mortgage program, mortgage loans have risen sharply since 2013 (figure II.13). The share of outstanding mortgage loans in total household credit reached 51.5 percent in 2016 before moderating to 38 percent in 2019. Salary loans have become a key driver of individual loans (26 percent of individual loans in 2019) (figure II.14). Pension and other consumption loans have also picked up since late 2016. Consequently, the average household debt-to-income ratio was estimated at over 90 percent at the end of 2018, up from 39 percent in 2012, reflecting a substantial increase in household loans. However, with macroprudential measures, the growth of individual loans has sharply decelerated since early 2019. Meanwhile, loans to household small businesses or small and microenterprises decreased substantially over 2010–17, as most banks opted to extend consumption loans (mainly salary), given a higher return and limited risk. However, with macroprudential measures on consumption loans, small and microenterprise loans have recently increased again. 33 NBFIs provide loans to the segment of the population that does not meet bank requirements or requirements for small and medium-sized enterprise loans. Hence, they usually issue vehicle loans, leases, and business loans. | 43 | Macroeconomic Stability and Credit Growth Figure II.13. A large part of outstanding individual Figure II.14. Salary loans have been a key driver credit is for mortgage and salary loans of individual credit growth since 2016 Composition of outstanding individual loans (in percent) Contribution to anuual growth of individual loans (in 100% percentage points) 90% Others 65 Mortgage incl RMBS 80% 55 Salary loans SME Pension credit 70% Consumption loans 45 60% SMEs Consumption Loans Other 50% 35 Total 40% Pension Loans 25 30% 15 Salary Loans 20% 5 10% Mortgage 0% -5 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2012 2013 2014 2015 2016 2017 2018 2019 Source: BoM. Source: BoM. Note: RMBS = residential mortgage-backed securities; SMEs = small and microenterprises. C2. Private sector Corporate credit has increased sharply since mid-2018 amid an economic recovery and the BoM’s efforts to contain individual loans. However, sectors with a higher share of GDP receive a smaller share of credit (figures II.15 and II.16). Corporate credit growth to the trade and mining sectors accelerated (16.2 and 20.2 percent, respectively) in 2019. Although accounting for a total of 44 percent of GDP in 2018, the mining, agriculture, and manufacturing sectors held about 35 percent of total corporate credit in 2019. The trade (wholesale/retail) and construction sectors held 43 percent of total credit while accounting for only 13 percent of GDP. Figure II.15. Corporate credit is concentrated in Figure II.16. Corporate credit growth has been the mining and mining-related sectors dominated by trade and mining activities since early 2018 Composition of outstanding corporate loans, by Contribution to annual growth of corporate loans, by sector (in percent) sectors (in percentage points) 25 100% Mining Manufacturing Trade Construction Agriculture Real Estate Others Corporate Loans 90% 20 80% 15 70% Mining 10 60% 5 50% Manufacturing 0 40% -5 30% Construction -10 20% -15 10% -20 0% Trade 2015Q1 2015Q2 2015Q3 2015Q4 2016Q1 2016Q2 2016Q3 2016Q4 2017Q1 2017Q2 2017Q3 2017Q4 2018Q1 2018Q2 2018Q3 2018Q4 2019Q1 2019Q2 2019Q3 2019Q4 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Others Source: BoM. Source: BoM. Persistent tugrug depreciation and implementation of macroprudential measures have led to a sharp reduction in foreign currency lending. In fact, with the economic difficulties during 2013–16, the tugrug depreciated significantly, resulting in substantial losses for unhedged borrowers. Accordingly, the credit dollarization has constantly declined since 2009 except for a brief period in 2017-18 (figure II.17). The share of total foreign currency loans dropped to 9.5 percent at the end of 2019, from 33 percent in 2013. | 44 | Mongolia economic update Figure II.17. Credit dollarization has been declining Figure II.18. …for both corporate and individual since mid-2018… borrowers Credit dollarization by corporate and individual loans ratio FX credit to total credit (LHS) mnt/ us$ (percent) 50 3,000 45 2.8 Exchange rate (MNT/US$) 2.6 40 2,500 40 2.4 35 2,000 2.2 30 30 2 1,500 25 1.8 20 1,000 1.6 20 Corporate Loans: LHS 1.4 10 500 15 Individual Loans: RHS 1.2 - 0 10 1 2004 2006 2008 2010 2012 2014 2016 2018 2019 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 2017 2018 2019 Source: BoM. Source: BoM. Note: LHS = left-hand side; RHS = right-hand side. Credit dollarization has been declining for both corporate and individual borrowers (figure II.18). Almost all individual loans (99 percent as of the end of 2019) were issued in the local currency, reflecting the fact that mortgage and salary loans dominated individual loans. Meanwhile, about 21 percent of corporate loans were denominated in foreign currency as of December 2019, substantially down from about 40 percent in Q3 2016. This reflects the persistent depreciating trend of the tugrug and macroprudential measures introduced since September 2018 (for example, greater risk weights for foreign exchange loans). C3. Credit concentration Another characteristic of credit in Mongolia is its concentration in a limited number of firms. The majority of total credit in the banking system went to only a few large companies. Only 13 percent of total corporate borrowers acquired more than 80 percent of the total outstanding corporate loans (figure II.19). Basically, corporate loans are heavily concentrated in a few large companies, which indicates serious risk of exposure to a limited number of borrowers. Moreover, a significant majority of individual borrowers have taken more than half of the total individual loans. Over 90 percent of the total household borrowers have acquired credit up to MNT 20 million (about US$7,500), representing about 60 percent of the total outstanding individual loans. In contrast, only 10 percent of the individual borrowers own about 40 percent of the loans (figure II.19). Also, only a minority of individuals own more than half of bank individual deposits, pointing to banks’ dependence on a few large depositors. In fact, on average, only 1 percent of household depositors holds deposits exceeding MNT 20 million (US$7,500). This exposes banks to the risk of sudden shifts in their deposit base (figure II.19).34 34 An additional risk for high concentration in the banking sector is related party lending whereby bank owners are also the biggest borrowers and the biggest depositors. In fact, amendment of the banking law in January 2018 indicates that a related party’s loan should not exceed 5 percent of the bank’s own capital and total loans to a related party and to his/her related parties in total should not exceed 20 percent of the bank’s own capital. | 45 | Loan and Figure II.19. Stability Macroeconomic deposit and concentration Credit Growth by allocation and client Figure II.19. Loan and deposit concentration by allocation and client Over 80 percent of total credit in the banking system went to only a few large companies… Over Figure 80 .percent II.19 Loan andof total creditconcentration deposit in the banking system went to only by allocation andaclient few large companies… Corporate loans by loan Over 80 percent of total credit in the banking system went to only a fewof size Structure corporate large borrowers, companies… Corporate loans by loan size Structure categorized of by corporate the size of borrowers, their loans Structure of corporate borrowers, Corporate loans by loan size Up to MNT 1 bn categorized by the size of their categorized by the size of their loans loans 16% Above MNT 1 bn Up to MNT 1 bn 13% Above MNT 1 bn 16% 13% Above MNT 1 bn Up to 84% Above MNT 1 bn Up1 MNT to 84% bn MNT 1 87%bn 87% Over 90 percent of individual borrowers were only eligible for small loans (up to MNT 20 million or about Over percent 90 500) US$7, Over 90 fromof percent individual the of banks… borrowers individual wereonly were borrowers onlyeligible for small loans eligible for (upto loans (up toMNT MNT 20 20 million million oror about about US$7,500) from the banks… US$7,500) from the banks… Individual loans by loan size Structure of individual borrowers, Individual loans Individual by loans loan by size loan size categorized Structure by of Structure the their loans size of borrowers, individual of individual borrowers, Above MNT 100 mn categorized categorized by by MNT the the sizeof oftheir size 50-100 their loans loans 16% MNT 20-50 mn Above MNT 100 mn Above MNT 100 mn 5% MNT 50-100 0% 16% MNT 20-50 mn Above MNT 100 mn MNT 50-100 mn 5% 0% 7% MNT 50-100 mn 7% Up to MNT 20 mn MNT 20- 58% Up to MNT 20 mn Up to MNT 20 mn 50 mn 94% MNT 58% Up to MNT 20 mn 19% 20- 50 mn 94% 19% Only 1 percent of depositors each hold deposits over MNT 20 million (US$7,500), reflecting an Only 1 percent unequal of depositors distribution each of bank hold deposits over MNT 20 million (US$7,500), reflecting an unequal deposits… distribution of bank deposits … Only 1 percent of depositors each hold deposits over MNT 20 million (US$7,500), reflecting an unequal distribution of bank deposits Household sector: … Household Sizeof sector:Size ofdeposits deposits Household Household sector: sector: Numberof Number ofdepositors depositors Household sector: Size of deposits Above MNT Household sector: 20 mn of depositors Number 1% Above MNT 20 mn 1% Up to MNT 20 mn 38% Up to MNT 20 mn 38% Above MNT 20 mn Up to MNT 20 mn 62% Above MNT 20 mn 99% Up to MNT 20 mn 62% 99% BoM; World Bank Sources:BoM; Sources: staffestimates. Bank staff estimates. Note: bn = billion; mn = million. Sources: BoM; World Bank staff estimates. 46 46 | 46 | Mongolia economic update D. Credit cycle and business cycle Adequate credit management is crucial to economic growth as credit and GDP growth rates are generally closely correlated in Mongolia (figure II.20 and box II.1). Credit development in Mongolia was largely driven by economic recoveries at the end of 2005, the beginning of 2010, and 2017. In 2011, however, a greater inflow of FDI was a key factor. Since 2013, the government subsidized mortgage program fueled credit growth. More recently, non-mortgage consumer loans were the key driver of credit growth (box II.1). Overall, financial cycles seem to be more persistent, with a larger amplitude compared to business cycles. This is consistent with the literature regarding other emerging and developing countries, which shows a strong co-movement between business and credit cycles (figure II.21). Figure II.20. Credit growth and economic Figure II.21. Credit cycle and business cycle performance Hihg commodity price, Global High FDI Serious IMF-led Macroprud .20 Favorable weather Crisis Quasi Fiscal economic program ential tools Credit cycle slowdown Measures .15 Business cycle Credit Growth 20 110 Commodity Prices (metal) .10 Economic Growth: RHS 90 15 .05 70 10 .00 50 5 -.05 30 10 0 -.10 -10 -5 -.15 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 (I-III) 00 02 04 06 08 10 12 14 16 18 Sources: BoM; NSO; and World Development Indicators. Note: Shaded areas illustrate the contraction phases of business cycles. Box II.1. Credit booms in Mongolia Fluctuations in commodity prices and capital flows have driven three credit booms since early 2000. The government decided to grant mining licenses for several large coal and copper deposits including Oyu Tolgoi (OT) and Tavan Tolgoi (TT) after the 2009 global financial crisis. These factors resulted in a surge in FDI, particularly to the mining sector. High domestic demand triggered by high copper prices and public sector salaries drove the credit boom in 2008. Favorable global copper prices during 2004–08 not only increased export revenue but also supported raising public sector wages through a higher tax income. Consequently, the private sector raised its wages. With rising wages, households expanded their consumption and acquired greater salary loans, which resulted in a significant rise in domestic purchases that later boosted domestic investment and output. Thus, the private sector’s demand for credit also increased. Substantial quasi-fiscal activities of the government resulted in a credit boom in 2014. Unlike the previous credit boom in 2008, Mongolia faced negative external shocks in 2014, on both the trade and financial sides of the balance of payments. Since the second half of 2013, global commodity prices started to decline due to the slowdown in Chinese demand, coinciding with a significant drop of FDI in the mining sector. As a result of these negative shocks, net foreign assets shrank by 12 percent in 2013. | 47 | Macroeconomic Stability and Credit Growth To mitigate the negative impacts of those shocks and support the economy, the authorities decided to loosen the monetary and fiscal stance. For instance, the monetary policy rate was decreased by 75 basis points in January and by 100 basis points in both April and June of 2013. In addition, under an unconventional monetary policy framework, the BoM implemented “The Pricing Stabilization Program,” with two subprograms: the “Affordable Housing Finance Program,” and the “Price Stabilization Program by removing the supply chain constraints.” Meanwhile, the government issued Chinggis Bonds worth US$1.5 billion at the end of 2012. As a result, M2 increased by 91 percent in 2014. Thus, macroeconomic policies played a main role in the credit cycle of 2014-16. During 2017-18, excessive liquidity in the banking system and repayment of government domestic bonds created room for renewed household credit expansion. Commodity prices started to increase in the second half of 2016 after almost four years of continuous decline. Higher commodity prices and increasing export volumes supported tax revenues. Moreover, the IMF’s Extended Fund Facility program improved national credibility among foreign investors and gradually attracted FDI. With its commitments to the IMF, the government reduced its expenditures and repaid the domestic bonds with the financial resources from the IMF and other organizations. The excessive liquidity from the bond repayment caused the domestic banks to increase their credit, particularly to the household sector since the end of 2017. E. Policy recommendations The following findings emerge from the special topic of this Mongolia Economic Update: (a) credit growth has been excessive and volatile, (b) credit policy was procyclical, (c) household and corporate debt are high, (d) funding and loans are highly concentrated, (e) level of provisioning and savings are low, (f) unconventional credit policies are common, (g) fiscal dominance is observed,35 and (h) credit allocation and the diversification agenda are not matched. The significant swings in real economic conditions complicate monetary and financial policies. Large FDI inflows and high commodity prices combined with an accommodative monetary policy and lax supervision have contributed to rapid credit expansion. A significant share of bank lending is channeled to the mining and mining-related sectors (including construction, trade, and real estate) and to individual loans (mainly pension and salary loans). Credit development is highly affected by external shocks, including commodity prices and foreign direct investments through macro-financial links. The rapid credit growth has fueled a sustained rise in real estate prices and imports, leading to the buildup of substantial credit and liquidity risk in the banking system. Ensuring price stability and smoothing out the credit cycle under these conditions is challenging. Coupling traditional monetary policy with appropriate macroprudential policy is therefore warranted. To address these challenges, the authorities will need a combination of structural and macroeconomic policy tools. To curb excessive and volatile credit growth, monetary policy will need to be carefully calibrated and supported with adequate macroprudential measures. The recent stance of the BoM has helped limit financial sector risks and vulnerabilities and in light 35 Fiscal dominance is an economic condition that occurs when a country has a large government debt and deficit such that monetary policy targets keep the government from bankruptcy, as opposed to economic targets such as inflation, growth, and employment. Over the period of study, we noticed that risks of fiscal dominance or pressure on a central bank to take actions that are not in line with the inflation objective appear high. For instance, government domestic borrowing in the period of 2013-16 affected credit dynamics which crowded out the private sector to some extent. | 48 | Mongolia economic update of continuing external vulnerabilities should be continued, whilst focusing on keeping inflation close to the BoM’s target. Fiscal policy also will have an important role to play in complementing monetary and macroprudential measures to smooth out excessive economic volatility. Over the medium term, Mongolia may benefit from designing a more holistic macroprudential framework. On the structural side, upgrading prudential standards in line Basel III requirements is a priority, to strengthen the BoM’s surveillance framework, together with steps to strengthen the regulator’s capacity and political independence. Moreover, in light of persistent NPLs and the identified capital shortfalls, completing the bank recapitalization program as foreseen under the ERP will be critical to strengthen financial sector resilience. Table II.1. Mongolia: Managing credit growth volatility - some policy options MACROECONOMIC POLICIES Policy Areas POLICIES MACROECONOMICPolicy Options Policy Areas Set policy rate is at the appropriate level MACROECONOMIC to stem POLICIES Policy credit growth volatility while maintaining Options Monetary Policy price stability Policy Areas Set policy rate is at the appropriate level to stem Policy credit growth volatility while maintaining Options Monetary Fiscal Policy Policy Implement countercyclical fiscal policies to mitigate boom and bust cycle price stability Set policy rate is at the appropriate level to stem credit growth volatility while maintaining Monetary Fiscal Policy Policy Implement countercyclical fiscal policies PRUDENTIAL to mitigate boom and bust cycle POLICIES price stability Fiscal Policy Implement countercyclical fiscal policies PRUDENTIAL to mitigate boom and bust cycle POLICIES Complete implementation of adequate requirements of Basel II and III to meet the PRUDENTIAL POLICIES appropriate capital requirements Complete implementation of adequate requirements of Basel II and III to meet the appropriate capital requirements Complete Enhance implementation the of adequate relevant regulations requirements and supervision of Basel of NBFIs II and in line III with to meet the commercial banks Banking and NBFIs appropriate capital requirements Regulation, Improve Enhance thea framework to monitor and relevant regulations measure links and supervision among of NBFIs infinancial line with institutions commercial (banks banksand Banking and NBFIs Supervision, and NBFIs) Regulation, Improve Enhance thea framework to monitor and relevant regulations measure links and supervision among of NBFIs infinancial line with institutions commercial (banks banksand Banking and NBFIs Resolution Strengthen the monitoring of corporate loans given the high level of NPLs in the private Supervision, and NBFIs) Regulation, Improve a framework to monitor and measure links among financial institutions (banks and Resolution sector, and develop appropriate macroprudential and supervision tools Strengthen NBFIs) the monitoring of corporate loans given the high level of NPLs in the private Supervision, and Diversify sector, andcredit portfolio develop by reviewing appropriate sector concentration macroprudential and supervision tools Resolution Strengthen the monitoring of corporate loans given the high level of NPLs in the private Utilize Diversifymicro level credit data including portfolio the by reviewing credit information bureau in macroprudential analysis sector concentration sector, and develop appropriate macroprudential and supervision tools and streghten top-down stress testing Utilize Diversifymicro level credit data including portfolio the credit by reviewing sectorinformation bureau in macroprudential analysis concentration Crisis Preparation and streghten top-down stress testing Utilize micro Strenghten level the data including financial the credit system oversight information framework bureau Financial including in macroprudential analysis Stability Committee and Management and streghten top-down stress testing Crisis Preparation by addressing priority risks and improving existing crisis prevention system and procedures Framework Strenghten the financial system oversight framework including Financial Stability Committee and Management Crisis Preparation by addressing priority risks and improving existing crisis prevention system and procedures Framework Strenghten the financial system oversight framework including Financial Stability Committee and Management by addressing priority risks and improving existing crisis prevention system and procedures Framework Note: NBFIs = non-bank financial institutions. | 49 | Macroeconomic Stability and Credit Growth Annex: Estimating the drivers of bank credit in Mongolia To define the main drivers of credit in Mongolia in the short and medium term, a Vector Autoregressive (VAR) model with 2 lags is estimated on quarterly data between Q1 2000 and Q4 2018.36 The VAR consists of endogenous variables that include real gross domestic product (GDP), real deposits, real credit, the nonperforming loan ratio, the interbank interest rate, the nominal effective exchange rate, and an exogenous variable of the export price index. The estimation results reveal that total credit increases due to higher export prices, GDP, total deposits, and the nominal effective exchange rate (appreciation of the domestic exchange rate), while it decreases following a hike in the policy rate and increases in nonperforming loans (NPLs) (figure A.1). Figure A.1. Impulse response of total credit to selected shocks Response to export price index Response to GDP Response to total deposit 1.2 .3 .8 .2 1.0 .6 .1 .4 0.8 .0 .2 0.6 -.1 .0 0.4 -.2 -.2 0.2 -.3 -.4 0.0 -.4 2 4 6 8 10 12 14 16 18 20 2 4 6 8 10 12 14 16 18 20 2 4 6 8 10 12 14 16 18 20 Response to NEER Response to policy rate Response to NPLs 1.0 .004 2 0.8 .000 1 0.6 -.004 0 0.4 0.2 -.008 -1 0.0 -.012 -2 -0.2 -0.4 -.016 -3 2 4 6 8 10 12 14 16 18 20 2 4 6 8 10 12 14 16 18 20 2 4 6 8 10 12 14 16 18 20 Sources: BoM and World Bank staff calculations. Note: Dotted lines are asymptotic 1 standard deviation confidence bands; NEER = nominal effective exchange rate; NPLs = nonperforming loans. Variance decomposition shows that credit in Mongolia is mainly explained by fluctuations in deposits (figure A.2). For instance, total deposits explain 70.9 percent of the fluctuations in total credit. This is consistent with the fact that about 90 percent of the financial market consists of the banking sector, and deposit rates in Mongolian banks are relatively high (around 12 to 18 percent) and inflexible due to a high concentration of deposits in the hands of only a few depositors. In the short term, bank credit seems highly sensitive to the volatility of commodity prices, which explains the 6.6 percentage point variations in the past. As the economy is highly dependent on the mineral sector, bank credit is strongly caused by commodity prices. 36 Lag length criteria show different results; some criteria select 1 lag, some criteria select 2 lags, and some criteria select 3 lags. A lag of 2 was chosen following the empirical studies conducted on Mongolian data. | 50 | Mongolia economic update For instance, it was estimated that a 1 percentage point positive shock in the export price index could boost bank credit by up to 2.4 percent within three quarters, explaining 12 percent, on average, of the variations of total credit. The variance decomposition analysis Figure A.2. Main drivers of credit growth (Variance suggested that real GDP shock Decomposition of VAR model) explains about 10 percent of the credit 30% 3% 4% fluctuations. For instance, a 1 percentage 25% 2% 3% 5% 5% 5% 5% point positive shock in real GDP could 3% 4% 3% 3% 3% 2% expand bank credit by around 0.4 percent 20% 6% 6% 4% 4% 3% after one quarter shock observed, sustaining 7% 8% 9% 4% the effect in the medium term. 15% 7% 9% 9% 7% Monetary policy also played a critical role 10% 17% 6% 10% 6% for bank credit development. The policy 5% 1% 13% 10% 5% rate shock explained about 3.4 percent of 7% 7% 6% 5% 4% 4% the variations in total credit. 0% The NPLs and exchange rate fluctuations Periods are also important factors for bank Export price index GDP NEER Policy rate NPL ratio credit. It is found that a 1 percentage point Source: BoM and World Bank staff estimations (2019). increase in the NPL ratio decreases bank Note: NEER = nominal effective exchange rate; NPLs = credit by up to 1.5 percent in the short term, nonperforming loans. while the nominal exchange rate shock explained 3.4 percent of the historical variations of bank credit. | 51 | Macroeconomic Stability and Credit Growth References Bank of Mongolia. 2019a. “Banking Credit: Trends, Issues, and Policies.” Background paper prepared jointly with World Bank staff. Ulaanbaatar, Mongolia. ―――. 2019b. Monetary Policy Guideline for 2020. Government Document. Ulaanbaatar, Mongolia. ―――. 2019c. “Inflation report (2019 Q3).” Government Document. Bank of Mongolia, Ulaanbaatar, Mongolia. Drehmann, M., C. Borio, L. Gambacorta, G. Jiménez, and C. Trucharte. 2010. “Countercyclical Capital Buffers: Exploring Options.” BIS Working Papers, No. 298, Bank for International Settlements, Basel. Gan-Ochir and Munkhbayar. 2019. “Measuring the natural rate of interest in a commodity exporting economy: Evidence from Mongolia.” International Economics. IMF (International Monetary Fund). 2014. “Staff Guidance Note on Macroprudential Policy— Detailed Guidance on Instruments.” IMF Policy Paper, International Monetary Fund, Washington, DC (November). IMF (International Monetary Fund). 2019. “Fiscal Monitor: How to Mitigate Climate Change.” IMF, International Monetary Fund, Washington, DC (October). ―――. 2018a. “Mozambique: Macro-financial Linkages and Bank Vulnerabilities.” IMF Selected Issues. International Monetary Fund, Washington, DC. ―――. 2018b. “Sri Lanka’s Credit Expansion: Engineering a Soft Landing While Financing Growth.” IMF Selected Issues. International Monetary Fund, Washington, DC. ―――. 2019. “Mongolia: Article IV Consultation.” Staff report. International Monetary Fund, Washington, DC (September). MoF (Ministry of Finance). 2019. “Medium Term Fiscal Framework for 2020–22.” Government Document. Ministry of Finance, Ulaanbaatar, Mongolia (December). MoF (Ministry of Finance). September 2019. “2020 budget document.” Government Document. Ministry of Finance, Ulaanbaatar, Mongolia. NSO (National Statistics Office). December 2019. “Monthly Bulletin of Statistics.” Government Document. National Statistics Office, Ulaanbaatar, Mongolia. Schmillen, Achim, and Nina Weimann-Sandig. 2018. Perceptions of precariousness: A Qualitative Study of Constraints Underlying Gender Disparities in Mongolia’s Labor Market. Washington DC: World Bank Group. | 52 | Mongolia economic update Vashakmadze, E. 2019. “Global and Regional Outlook and Risks – Key Highlights.” Box prepared for the Global Economic Prospects report (January 2020). Washington, DC: Washington, DC. World Bank. 2016. “Public Asset Management Companies: A Toolkit.” World Bank, Washington, DC. ―――. 2018. “Mongolia: Public Expenditure Review.” World Bank, Washington, DC. ―――. 2019a. Global Economic Prospects (January 2020). Washington, DC: World Bank. ―――. 2019b. World Bank Commodity Outlook (October 2019). Washington, DC: World Bank. World Economic Forum. 2017. The Global Competitiveness Report 2016–17. Insight Report. Edited by Klaus Schwab. Geneva: World Economic Forum. | 53 | Macroeconomic Stability and Credit Growth | 54 | 5th Floor, MCS Plaza Building, 4 Seoul Street, 14250 Ulaanbaatar, Mongolia Tel: +(976)70078200 Web: www.worldbank.org/mongolia Twitter: WorldBankMgl