Aggregate Fiscal Impact of COVID-19 1 @2021 The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org This work is a product of the staff of the World Bank with external contributions. The findings, interpretations and conclusions expressed in this work do not necessarily reflect the views of the World Bank and its Board of Executive Directors. The World Bank do not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of the World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of the World Bank, all of which are specifically reserved. All queries on rights and licenses should be addressed to the Publishing and Knowledge Division, the World Bank, 1818 H Street NW, Washington DC, 20433, USA, Fax: 202-522-2625; email: pubrights@worldbank.org. Cover picture: hoaihungtv/Pixabay Contents Abbreviations 7 Acknowledgements 9 1. Summary 11 2. The impact of COVID-19 shock on revenues 12 3. The impact of the COVID-19 shock on public expenditure 20 4. Fiscal trends and the impact of the COVID-19 shock on deficit 27 and debt in 2020 and beyond 5. Economic outlook, fiscal out-turns, and policy implications 29 6. Recommendations 32 Annex I.1: Performance of main taxes 35 Annex I.2: Expenditure performance – general trends 38 Annex I.3: Additional figures and tables 42 Annex I.4: Mobility restrictions, investment disbursement, and regional growth 44 – indicative correlations Tables Table I.1. Tax-to-GDP ratio in selected EAP countries, 2019 and 2020 13 Table I.2. Economic support package – revenue measures 2020 14 Table I.3. Government revenue projections, 2021 18 Table BI.1.1. Sub-national ratio of contribution to the central government budget in 2020 19 Table BI.1.2. Planned transfer from the central government to subnational governments in 2020 20 Table I.4. Fiscal support package – 2020 – expenditure measures 22 Table I.5. COVID-19 intergovernmental transfer scheme in 2020 24 Table I.6. Public investment plan and execution in 2020 as of January 31, 2021 24 Table I.7. Central government public investment in 2019–2020 25 Table I.8. Public expenditure in 2019–2020 by region 27 Table I.9. Government bond portfolio, 2016–2020 28 Table I.10. Vietnam Key Economic Indicators – baseline scenario 30 Table I.11. PIT productivity among the ASEAN-4 36 Table I.12. All fiscal measures 42 Table I.13. Public investment execution in 2019–2020 43 Figures Figure I.1. Share of total revenue (% of total revenue) 12 Figure I.2. Revenues and expenditures as share of GDP (%) 12 Figure I.3. Tax collection in Vietnam and select EAP economies (% GDP) 13 Figure I.4. Tax collection in regional comparison 13 Figure I.5. Revenue composition in 2016–2020 (% GDP) 13 Figure I.6. Usefulness of government support policies 15 Figure I.7. Tax revenue composition in 2016–2020, (% GDP) 16 Figure I.8. Non-tax revenue composition in 2016–2020, (% GDP) 16 Figure I.9. Tax revenues in 2019–2020 by firm ownership (% and pps of GDP) 17 Figure I.10. Revenue composition by firm ownership in 2019 and 2020 (% of total revenue) 17 Figure I.11. Planned expenditure (% GDP) 21 Figure I.12. Implemented expenditure (% GDP) 21 Figure I.13. Estimated earning losses and planned support to households in Vietnam and its EAP peers in 2020 21 Figure I.14. Central government public investment composition in 2020 by administrative classification 25 Figure I.15. Public investment by region in 2019–2020 26 Figure I.16.CIT collection in regional comparison 35 Figure I.17. CIT productivity 35 Figure I.18. PIT collection in regional comparison 36 Figure I.19. VAT collection in regional comparison 36 Figure I.20. VAT C-efficiency comparison 36 Figure I.21. SCT collection in regional comparison (with other excise regimes) 37 Figure I.22. Planned expenditure (% GDP) 38 Figure I.23. Implemented expenditure (% GDP) 38 Figure I.24. Composition of planned expenditure 39 Figure I.25. Composition of actual expenditure (% of total actual expenditure) 39 Figure I.26. Recurrent expenditures by function, 2016–20 (% GDP) 40 Figure I.27. Recurrent expenditures by function, 2016–20 (% total expenditure) 40 Figure I.28. Public investment disbursement rate in 2018–2020 (percent of planned) 41 Figure I.29. Disbursement rate of capital investment in the first nine months, 2016–20 42 Figure I.30. Uneven public investment disbursement across provinces 43 Figure I.31. Mobility and disbursement 45 Figure I.32. Fiscal multiplier effects 45 Boxes Box I.1. Distributional impact of COVID-19 on revenues at the central and the subnational levels 19 Box I.2. Tax reform to increase and stabilize tax collection and increase fiscal space (adapted from the World bank recommendations to MOF on the Vietnam Tax Reform Strategy 2021-2030) 33 Aggregate Fiscal Impact of COVID-19 7 Abbreviations CIT corporate income tax COVID-19 coronavirus disease of 2019 EAP East Asia and Pacific EPT environmental protection tax FIEs foreign direct investment enterprises GDP gross domestic product GSO Government Statistics Office MOF Ministry of Finance MTIPS medium-term investment plans PIT personal income tax RGDP regional gross domestic product SCT special consumption tax TABMIS Treasury and Budget Management Information System VAT value-added tax VND Vietnamese dong WB World Bank y/y year-over-year 8 Note I Aggregate Fiscal Impact of COVID-19 9 Acknowledgements This note on Aggregate Fiscal Impact of COVID-19 has been prepared as part of the Vietnam Fiscal Update 2021, a second report in a programmatic series. This reports analyzes the impact of Covid-19 on Vietnam’s public finance during the first year of the crisis. It consists of four notes that aim to provide timely and relevant information and policy recommendations to the development of the Ministry of Finance’s ten-year financial strategy (2021-30) and technical assistance to the Government to adopt good practices in fiscal policy and public finance management reforms. Key contributors to this note were Dorsati Madani, Nguyễn Việt Anh and Nguyễn Thế Hoàng with inputs from Nguyễn Phương Anh. Quality review and assurance on earlier drafts have been provided by colleagues across the World Bank Group, including Arvind Nair (Senior Economist), Oleksii Balabushko (Senior Public Finance Specialist), Naira Melkumyan (Senior Country Officer), and Nguyễn Hồng Ngân (Senior External Affairs Officer). The report was also enriched by consultations with the government, development partners, and policy institutions. Comments were received from various departments of the Ministry of Finance (MOF). We would like to acknowledge the strong support and contribution from Mr. Nguyễn Minh Tân (Deputy Director, State Budget Department, MOF). The team is grateful for the overall guidance of Carolyn Turk (Country Director for Vietnam); Hassan Zaman (Regional Director, Equitable Growth, Finance and Institutions Practice, East Asia and the Pacific); Stefanie Stallmeister (Operations and Portfolio Manager for Vietnam); Deepak Mishra (former Practice Manager, Macroeconomic Trade and Investment Global Practice); Sebastian Eckhardt (Practice Manager, Macroeconomic Trade and Investment Global Practice); Fily Sissoko (Practice Manager, Governance Global Practice); Jacques Morisset (Program Leader); and Matthew Wai-Poi (Lead Economist). The report benefited from the administrative assistance of Le Thi Khanh Linh, and communications support from Nguyen Hong Ngan and Le Thi Quynh Anh of the World Bank Vietnam Country Office. 10 Note I Summary This note analyzes the short-term effects of the COVID-19 shock during the first year of the crisis and presents alternative medium-term recovery scenarios. It asks a set of questions, and in answering those questions, provides policy suggestions for the Vietnamese authorities’ consideration. These questions are: (i) How were tax revenues affected by the shock, and how, in turn, did they affect the fiscal space available to the authorities? What tax measures did the authorities implement to mitigate the shock and how successful were they? What are the implications for any future tax policy and tax administration reforms? (ii) How did the COVID-19 shock affect expenditures? What fiscal measures did the authorities implement to mitigate the shocks, and how successful were they? (iii) How did the shock affect fiscal balances and public debt? (iv) What are the growth and fiscal prospects in the medium term? What are the implications for any future tax policy and tax administration reforms? We find: (i) Taxes were strongly affected, not only by lower economic activity, but also by tax waivers associated with the economic support package. State-owned enterprise (SOE) revenue losses affected the government’s tax and non-tax revenues, as they are major contributors to the public coffers. The authorities resorted to raising more non-tax revenues to compensate for the loss of tax revenues, and by the end of 2020 had succeeded in ensuring a total revenue collection rate of about 98 percent compared to the 2020 planned budget. However, given uncertainties associated with the length of the crisis, the impact on taxes and non-tax revenues is expected to continue into 2021. (ii) The execution of the expenditure component of the fiscal support package was uneven. The program to support household income was relatively small (VND 36 trillion or 0.6 percent of GDP in 2020), short-lived, and was expected to reach over 13 percent of the population (13.3 million people). The social protection system remains small. In contrast, the authorities succeeded in supporting aggregate demand through an acceleration of the public investment disbursement. Here, the emphasis was mostly on projects in transport and agriculture and rural development, while the distribution and implementation of the projects remained heterogeneous across provinces. (iii) The economy is on a path to recovery, and given the discipline shown by the authorities on fiscal consolidation and debt reprofiling in the past few years, there is fiscal space and public debt is sustainable. However, given the resurgence of COVID-19 variants and the slow rate of vaccination, the global economic recovery is uncertain, and risks are on the Aggregate Fiscal Impact of COVID-19 11 downside. While fiscal discipline is warranted, moving to a neutral fiscal stance too early may deprive vulnerable households and small and medium-sized enterprises (SMEs) of needed support and affect the economy prematurely. Policy suggestions include: (i) On the revenue front, in the short-term tax administration reforms can help to restore and stabilize tax collection through stepping up compliance enforcement and implementing measures to increase the tax base. The recent decision to establish a Large Taxpayer Office to manage the most important group of taxpayers is a step toward more effective tax collection. In the medium term, the government is developing a new tax policy framework. A forward-looking tax policy would be needed to support Vietnam’s 2021-2030 SEDS’s vision of an inclusive, digitally transformative, and sustainable growth model. This would include strengthening international taxation, including digital taxation, and adopting environmental taxation, such as carbon taxation. Also, reform of the current tax structure, such as streamlining generous CIT incentives and introducing a property tax, will help expand the tax base and contribute to enhancing the efficiency and equity of the tax system. This could be an opportune time to also improving revenue projection practices and laying out a transparent practice of managing large cash balances. (ii) On the expenditure front, Vietnam has the fiscal space to spend more but needs to also spend better. Three areas deserve attention. First, develop an inclusive and responsive social protection system. Given the aging population and the coming to light of the weaknesses in the social protection system during the 2020 COVID-19 crisis, consideration should be given to developing a more integrated social safety net system, with digitalized intake and service delivery. Second, further improvements in expenditure efficiency, including further review of public investment management practices, appears warranted given the chronic under disbursement of the investment budget during the 2016–19 stability period. Third, in the medium term, given the critical infrastructure investment needs of the country and its objective to move to a more sustainable economic model, productive investment could be considered, especially in green and sustainable technology to create a base for such an economic transformation. (iii) On fiscal balances and fiscal stance: Vietnam has fiscal space and its debt is sustainable. In the short run, it may be premature to adopt a neutral fiscal stance given the ongoing effects of the COVID-19 shock. A more accommodative fiscal stance that supports vulnerable households and small businesses and continues to support aggregate demand may be more appropriate. In the long run the authorities may wish to consider developing automatic stabilizers to improve the ability of the system to deliver timely counter-cyclical response (both on the tax and expenditure sides). Development of these stabilizers would entail broad based taxation, and broad based and better targeted social protection programs. 12 Note I The impact of COVID-19 shock on revenues Pre Covid-19 Trends and Revenue Structure Vietnam’s revenue collection was mostly stable during 2016–19 but did not address the per- sistent structural fiscal deficit. Figure I.1 shows that the share of tax revenue as a component of total revenue stabilized at around 74 percent during 2016–19, while the share of non-tax revenues (capital revenues and fees and charges) averaged about 26 percent. The rise in non-tax revenues was partly attributed to the disposal of state assets (capital revenues), which is not a sustainable increase over time.1 The persistent fiscal deficits during this period (figure I.2) would suggest that the government should consider both enhancing the efficiency of public expenditures and raising and stabilizing tax revenue collection. Figure total revenue) I.2. Figure Revenues I.1. and Share of total expenditures revenue asrevenue) (% of total share of GDP (%) I.2. Revenues and expenditures as share of GDP (%) Figure 100% Grants 30100% 30 Tổng chi Expenditure Capital revenues 90% 90% 25 25 Fees, 80% charges & Tổng thu 80% Revenue non-tax Thu thuế 20 20 Taxes 70% 70% Taxes 60% 1560% 15 2020e 2016 2016 20172017 2018 2018 2019 2019 2020e 2020e 2016 2017 2018 2019 2020e Sources: WB staff estimates based on MOF and GSO data. Tax collection in Vietnam outperforms its East Asia and Pacific (EAP) peers and the group of emerging economies. Tax collection in Vietnam has been better than in most of its major neighbors (figure I.3), even during the pandemic (2020). However, driven by the policy choice to lessen the collection of capital income taxation to incentivize investment inflows, tax collection in Vietnam, which remained relatively flat during 2015-2018, compared to a rising trend in the EAP emerging and high-income economies (figure I.4). While overall tax collection is comparable to many EAP economies, individual tax performances diverge. The corporate income and value-added taxes are set relatively low at 20 percent and 10 percent, respectively, but perform adequately in terms of both collection (about 4.5 percent of GDP and 6 percent of GDP, respectively, in 2019) and productivity compared to other East Asian economies. In the case of the corporate income tax (CIT), this performance is affected by a substantial tax expenditure burden associated to incentives (about 30 to 40 percent of total CIT intake or about a 1 If viewed strictly from the Government Finance Statistics Manual (GFSM) 2014, the asset disposal would be excluded from revenue accounting because such transactions would not change the net worth position in government budget operations. Aggregate Fiscal Impact of COVID-19 13 quarter of the potential CIT without incentives). However, collection of the personal income tax (PIT) and special consumption tax (SCT) remains relatively low compared to other East Asian economies, reaching 1.85 percent of GDP and about 1.7 percent of GDP, respectively, in 2019 (Annex I.1). Figure I.3. Tax collection in Vietnam and select Figure I.4. Tax collection in regional comparison nd select EAP economies (% Figure GDP) I.4. Tax collection in regional comparison EAP economies (% GDP) 25 25 25 25 Việt Nam Vietnam 20 20 20 20 15 15 15 15 10 10 10 10 2015 2016 2017 2018 2015 2016 2017 2018 5 5 2020e 2016 phân vị Khoảng tứ2018 2017 2019 2020e Interquartile range Việt Nam Vietnam In-đô-nê-xia Vietnam Cambodia Thị trường China mới nổi ĐÁ-TBD Indonesia Emerging Markets in EAP Malaysia nhập cao ĐÁ-TBD Thailand Thu Philippines High Income in EAP Sources: WB staff estimates based on MOF data for Sources: WB staff estimates based on MOF and GSO data. Vietnam and Haver Analytics for other countries. Note: Revenue data for some of the other countries were unavailable beyond 2018. Thus, the comparison used the data range 2015–18 for regional groupings. Impact of the COVID-19 shock on revenues Tax collection in Vietnam was negatively impacted by the COVID-19 pandemic. The April nationwide lockdown, strict mobility measures in the early days, continued stringent international border closings, and the global recession contributed to an economic slowdown and the fall in tax collection, affecting overall revenues (figure I.5). However, while tax revenues fell from 18.6 percent of gross domestic product (GDP) in 2019 to 16.9 percent in 2020, they remained on par with neighboring countries (table I.1) Figure I.5. Revenue composition in 2016–2020 Table I.1. Tax-to-GDP ratio in selected EAP P) (% GDP) countries, 2019 and 2020 30 30 Tax/GDP ratio 2019 2020 2020/2019 25 25 Vietnam 18.6 16.9 91% 20 20 Cambodia 19.4 18.2 94% 15 15 China 15.9 15.2 96% 10 10 Indonesia 9.8 8.3 85% 5 5 Malaysia 12.0 10.9 91% 0 0 Mongolia 19.8 17.5 88% 020e 2016 2017 2018 2019 2020e Philippines 16.1 15.5 96% Thailand 16.2 15.5 96% huế Taxes Non-taxes Grants Total revenue Sources: Vietnam MoF and WB staff calculations. 14 Note I Tax revenue deteriorated, largely reflecting the economic slowdown. Taxes on oil producers dropped by 38.5 percent following the fall of oil prices in April 2020, reducing tax revenue by 0.4 percent of GDP (figure I.7). Weak domestic private demand, particularly in tourism-related sectors, caused the VAT on domestic and imported goods and services to drop by 0.6 percent of GDP. Taxes on international transactions fell by 0.4 percent of GDP, reflecting the deceleration of merchandise imports in 2020, due to lower domestic consumption and disrupted global value chains, as well as the impact of the ongoing tariff reforms associated with signed trade agreements. Non-oil CIT dropped by 0.1 percent of GDP due to lower revenue from SOEs. Despite the CIT reduction granted to SMEs, CIT collection from domestic private sectors in 2020 reached 1.4 percent of GDP, the same as in 2019. The PIT also remained virtually the same in 2020 as in 2019 at 1.8 percent of GDP, despite the tax relief offered by increasing PIT exemptions. The economic support package containing tax relief measures to businesses and households also contributed to the reduction in tax revenue collection. These measures included reductions and/or exemptions on a variety of taxes and fees and deferral of some tax payments to end-2020 (table I.2). The government reduced (i) the registration tax on domestically assembled cars, and (ii) 30 percent of the payable CIT in 2020 for SMEs under the VND 200 billion (US$8.7 million) turnover threshold, which accounted for 0.2 percent of GDP. These tax cuts reduced total tax revenue collected by about VND 30,000 billion (about US$1.3 billion), or about 3 percent. The government increased the monthly PIT exemption threshold from VND 9 million to VND 11 million, which reduced PIT collection. Table I.2. Economic support package – revenue measures 2020 Measure Planned Executed Execution rate % GDP % GDP % Revenue measures 4.0 1.8 44.8 Tax (CIT, PIT, VAT) and land rent deferral 2.9 1.1 37.4 Excise tax (SCT) deferral for automobiles produced 0.3 0.3 111.2 or assembled domestically Exemption and reduction of taxes, fees and charges, and land rents 0.8 0.4 47.6 CIT reduction 0.4 N/A N/A Raised family allowance in PIT 0.2 N/A N/A Exemption and reduction of other taxes, registration fees, and land rents 0.3 N/A N/A Reduction of fees and charges 0.0 N/A N/A Source: MOF (as of Jun 1, 2021). The tax deferral measures were highly welcomed by the business community. A business survey conducted by the Vietnam Chamber of Commerce and Industry (VCCI) completed in late 2020 confirmed the usefulness of the tax measures introduced by the government (figure I.6). The tax deferrals were applied to a wide range of sectors impacted by the pandemic, from the travel industry to agriculture and manufacturing. In addition, all small businesses in all sectors were entitled to tax deferrals. As a result, taxpayers took advantage of a total of VND 87,232 billion (US$3.8 billion) in tax deferrals. By the end of 2020, 96 percent of these deferred taxes had been collected by the tax authorities.2 2 Collection data from General Tax Department of the Ministry of Finance. Gia hạn nộp thuế GTGT 24 42 19 11 4 Gia hạn nộp thuế GTGT 19 36 24 17 5 Gia hạn nộp thuế TNDN 24 43 19 11 4 Gia hạn nộp thuế TNDN 21 36 24 16 4 Aggregate Fiscal Impact of COVID-19 15 Rất hữu ích Hữu ích Khá hữu ích Không mấy hữu ích Hoàn toàn không hữu ích Figure I.6. Usefulness of government support policies Domestic private sector Foreign-invested sector 0% Interest to Pay Salary 22 36 18 15 8 0% Interest to Pay Salary 18 29 21 24 8 Loan Extension 21 39 19 15 6 Loan Extension 15 34 22 23 6 Trade Union Fee Extension 17 41 22 15 5 Trade Union Fee Extension 17 31 23 23 5 Interest Rate Reduction 23 39 18 15 6 Interest Rate Reduction 16 35 23 21 5 Social Insurance Social Insurance 20 39 21 14 6 15 34 25 21 5 Postponement Postponement Land Fee Payment Extension 22 42 20 12 4 Land Fee Payment Extension 19 36 23 17 5 VAT Payment Extension 24 42 19 11 4 VAT Payment Extension 19 36 24 17 5 CIT Payment Extension 24 43 19 11 4 CIT Payment Extension 21 36 24 16 4 Extremely Useful Very Useful Somewhat Useful Not So Useful Not Useful at All Source: Vietnam Chamber of Commerce and Industry (VCCI) business survey – December 2020. While tax reduction and exemptions are also welcomed by the businesses, as indicated in the VCCI survey, they might not be a good policy choice for Vietnam. Since income tax exemptions or reductions will have impacts only on profit-making businesses, this policy is not a good choice if the objective is to support affected businesses, who could question the fairness of the tax stimulus. Therefore, the April 2021 tax relief package only includes tax deferrals and as of 1 June 2021, the government had not extended the tax reduction and exemptions measures in 2021. The 2021 tax relief package only includes tax deferrals. They are still the ideal choice as they strike a balance between business expectation and budget affordability. To offset the fall in tax collection, the government increased the collection of non-tax revenues, so that total revenues collected by end-December were about 98 percent of the revised plan for the year.3 Non-tax revenue increased by 5.2 percent in 2020 to reach over VND 442 trillion (7.0 percent of GDP) in 2020. On the one hand, fees and charges dropped by 14.4 percent due to a combination of fee reductions granted by the government and slower economic activity. On the other hand, capital revenues increased by about 12.8 percent4 in 2020, through assignment of land use rights.5 Higher revenues from the land use right assignment can be attributed in part to the implementation of Decree 79/2019/NĐ-CP issued on 26 October 2019 that came into force on 10 December 2019. This document substantially reduced the coverage of individuals eligible for five-year deferral of land use levy payment. Land rents also increased by 10.0 percent in 2020 (figure I.8). Proceeds from SOE equitization fell by 44 percent due to a slowdown in the privatization program in recent years.6 3 The revision was made on 12 November 2020, adding VND 26.7 trillion to the initial plan, which was prepared in 2019. 4 Excluding SOE equitization proceeds. 5 The institutional framework regarding assigning land use rights is as follows: the central government defines the rate structure, based on which the People’s Committee in each province (or city) set the land prices. Currently, this rate structure is considered to be underpricing land value. The provincial People’s Committees fully control the speed of granting land use rights and size of land rented out. The check and balance on this process is provided by the provincial (or City) People’s Committees. 6 Calculated using data from IMF (2016–2020) and MOF (2020). In GFS categorization, SOE equitization amounts are considered sales of assets and therefore registered financial assets (below the line). The MOF includes SOE equitization amounts as part of non-tax revenues. This note follows MOF methodology. 16 Note I Figure I.7. Tax revenue composition in 2016–2020, Figure I.8. Non-tax revenue composition in 2016– GDP) (% GDP) Figure I.8. Non-tax revenue composition in 2016–2020,2020, (% GDP) (% GDP) 20 20 8 8 15 15 6 6 10 10 4 4 5 5 2 2 0 0 0 0 2020e 2016 2016 2017 2017 2018 2019 2018 2019 2020e 2020e 2016 2017 2018 2019 2020e TNCN Phí & lệ phí Oil Tiền thuê Non-oil CIT PIT đất Fees & charges Land rents XNK Tiền VATsử dụng đất Excise Thu hồi Tra deDNNN vốn LUR assignment SOE privatization Thu ngoài Other thuế khác taxes Others Sources: MOF and WB staff estimates; SOE equitization proceeds in 2016–2019 from IMF Article IV. By firm ownership, revenues from domestic private and foreign direct investment enterprises (FIEs) (non-oil) remained relatively resilient in 2020, in contrast to a decline in revenues from SOEs. FIEs and domestic private firms increased their contribution to tax revenues by 1.1 percent year-over-year (y/y) and 3.7 percent y/y, respectively, which helped maintain their total tax contribution to GDP at 2019 levels (figure I.9). By comparison, tax revenues collected from SOEs dropped by 10.4 percent y/y (or 0.4 percent of GDP) in 2020. The decline came from lower the CIT and VAT on domestic goods and services (down 5.9 percent y/y and 10.5 percent y/y, respectively). However, total non-tax revenues from SOEs, including SOE equitization proceeds, increased by 4.9 percent, from VND 136.5 trillion in 2019 to 143.1 trillion VND in 2020 (about 2.3 percent of GDP), partly compensating for their lower tax contributions. Overall, total revenue collected from SOEs fell by 3.4 percent, and accounted for over 19 percent of total revenue in 2020, which is comparable to its share in 2019 (figure I.10). Figure Figure I.9. I.10. Tax revenues Revenue in 2019–2020 composition by firm in 2019 by firm ownership Figure andI.10. 2020Revenue composition (% of total by firm ownership revenue) y firm ownership (% and pps ownership of GDP) (% and pps of GDP) in 2019 and 2020 (% of total revenue) 8 2019 2020 8 2019 2020 2020 Change 2019-2020 6 DNNN DNNN 6 SOEs SOEs 19,3% 19,2% 19,3% 19,2% Khác Khác Others Others 4 FIESs FIESs 4 FIESs FIESs 48,1% 48,5% 48,1% 48,5% 17,3% 16,0% 17,3% 16,0% Ngoài quốc doanh Ngoài quốc doanh Private Private 2 16,4% 2 16,4% 15,3% 15,3% 0 0 Sources: MOF and WB staff estimates; SOE equitization proceeds in 2016–2018 from IMF Article IV. -2 -2 Note: Taxes whose data are disaggregated by SOEs, hác SOEs FIEs: Oil FIEs: Private Others FIEs, and private domestic enterprises include CIT, VAT Non-oil on domestic goods and services, excise tax, and natural resource tax. PPS: percentage point. Aggregate Fiscal Impact of COVID-19 17 Looking ahead, the government is expecting a further reduction in total revenue collection in 2021. Compared to the non-tax revenue collected in 2020, the government projects about 27 percent less collection from fees, charges, and non-tax revenues (table I.3). Given the impact of the shock on SOEs and continued expectation of lower profits and dividends in 2021, projecting lower non-tax revenues would be prudent. In addition, the government expects about one-third less capital revenues (revenues from the sale of state-owned houses, land use right assignment). On taxes, the government projects 3 percent less overall tax collections in 2021 compared to 2020. Most major taxes are expected to fall in 2021, except for the environmental protection tax (EPT). Fuel consumption, which is the majority of the EPT base, is expected to be stable, so there is an expected increase in EPT revenue. Lower 2021 overall revenue projections are partly in response to the uncertainty surrounding the COVID-19 crisis and partly standard practice in budget planning in Vietnam. The government is taking a cautious outlook that the COVID-19 pandemic will continue to negatively impact the economy, resulting in lower business profits, lower consumption, and lower personal income. Lower projected tax collection is also partly due to the authorities’ standing practice of projecting revenues conservatively. As experience shows, all provinces have an incentive to underestimate collection targets and then exceed those targets, which allows them to keep some of the “additional revenues” for investment purposes. Revenue collection in the first quarter of 2021 seems to confirm this is the case. However, the persistent under-projection of revenue targets raises the risk to budget credibility. This credibility is further affected by the retention of large cash reserve balances in the past few years. Large cash reserve accumulations appear to be associated with delayed implementation of the public investment programs in 2016-2019. 18 Note I Table I.3. Government revenue projections, 2021 Items 2020 2021 2021-Q1 Comparison between Collection in Q1- (2nd Est) (planned) VND trillions 2020 and plan 2021 2021 relative to VND trillions VND trillions (%) plan (%) Total revenue 1,507,845 1,343,330 403,700 89.1% 30.1% Current revenues 1,328,859 1,222,833 364,148 92.0% 29.8% Taxes 1,060,798 1,028,239 330,611 96.9% 32.2% Corporate income tax 259,329 247,304 89,403 95.4% 36.2% Personal income tax 115,213 107,796 38,825 93.6% 36.0% Land and housing tax 2,068 1,770 199 85.6% 11.2% Registration tax 34,822 33,871 8,677 97.3% 25.6% Value-added tax 342,029 330,881 106,353 96.7% 32.1% Excise tax 101,090 96,480 30,125 95.4% 31.2% Natural resources tax 30,148 27,042 6,842 89.7% 25.3% Agricultural land use tax 6 4 2 66.7% 50.0% Import-export tax, excise tax, and environmental protection 78,753 85,000 22,359 107.9% 26.3% tax on imports Environmental protection tax 60,631 64,391 14,208 106.2% 22.1% Revenue from lottery 36,709 33,700 13,618 91.8% 40.4% Fees, charges, and non-tax 268,061 194,594 33,537 72.6% 17.2% Fees and charges 35,099 36,562 9,448 104.2% 25.8% Land rents 39,370 24,415 5,595 62.0% 22.9% Miscellaneous revenues 193,592 133,617 18,494 69.0% 13.8% Of which: SOE equitization 29,846 proceeds Capital revenues (revenues from sale of state-owned houses, land 174,161 112,367 39,002 64.5% 34.7% use right assignment) Grants 4,825 8,130 550 168.5% 6.8% Sources: Vietnam MOF and WB staff calculations. Note: SOE = state-owned enterprise. Aggregate Fiscal Impact of COVID-19 19 Box I.1. Distributional impact of COVID-19 on revenues at the central and the subnational levels Central government revenues were likely disproportionally impacted by the COVID-19 shock as it collected less but distributed transfers as planned in the 2020 budget to subnational governments. The central government collects most taxes either entirely (oil revenues, VAT on imported goods and services, and taxes on international transactions) or partly (VAT on domestic goods and services, CIT, non-oil PIT, excise, and environment protection taxes). As shown in table BI.1.1, provinces and major cities contribute differentially to the national budget based on their level of development, with northern midlands and mountains contributing zero percent of collected taxes while HCMC only keeping 18 percent of collected taxes. Revenue collection was affected by the crisis (figure I.5 and figure I.7) . Despite lower revenues, the central government provided transfers to subnational governments according to the budget plan in 2020 (table BI.1.2). These amounted to VND 367.7 trillion (about 5.8 percent of GDP), or 36 percent of total planned revenues accrued to the subnational governments. The subnational governments were also able to collect virtually all land rents, and revenues from land use right assignments and the lottery, which could compensate to a large extent for their revenue losses due to their smaller tax collection caused by COVID-19. Poorer provinces appeared to be less affected by the pandemic as they were subsidized directly by the central government and indirectly by richer provinces. On the one hand, poorer provinces typically do not collect a lot of tax revenue, so declining tax collection would not affect them much. Instead, they rely more on substantial transfers from the central government, which accounted for more than half of total expected revenues for the North Midland and Mountains, and the Central Highlands. These transfers should not be affected by the central government’s lower revenue caused by the pandemic. On the other hand, richer provinces, mostly in Red River Delta and the Southeast, not only self-finance their operating budgets, but also contribute to the central government budget. As in the case of the central government, they rely more on taxes, and receive much less in intergovernmental transfers. Furthermore, the transfers they receive are targeted transfers. Consequently, it is likely that they were more affected by the fall in tax revenues caused by the pandemic. Table BI.1.1. Sub-national ratio of contribution to the central government budget in 2020 Monthly average income Average revenue per capita in 2019 contribution ratio in 2020 Current prices, thousand VND Northern Midlands & Mountains 2,640 0.0 Red River Delta 5,191 17.7 l Hanoi 6,403 65.0 North Central & Central Coast 3,331 5.9 l Da Nang 6,057 32.0 Central Highlands 3,095 0.0 Southeast 6,280 39.2 l Ho Chi Minh City 6,758 82.0 Mekong Delta 3,886 0.7 l Can Tho 4,713 9.0 Source: GSO, MOF and WB staff estimates Note: Revenue contribution ratio indicates the percentage of selective revenues stipulated by the Budget Law that sub- national governments must submit to the central government. 20 Note I Table BI.1.2. Planned transfer from the central government to subnational governments in 2020 Northern Red River North Central Southeast Mekong Midlands & Delta Central Highlands Delta Mountains & Central Coast Total revenue plan (trillion VND) 163.4 266.5 218.6 57.0 173.1 149.6 Composition (% total plan) Own collection 29.8 87.6 56.7 37.1 86.5 55.9 Transfer from central 70.2 12.4 43.3 62.9 13.5 44.1 Budget transfer 43.3 5.5 22.9 42.8 2.4 25.5 Salary reform transfer 4.2 0.8 1.9 1.9 0.0 1.9 Targeted transfer 22.7 6.1 18.4 18.2 11.1 16.7 Sources: MOF and WB staff estimates. The impact of the COVID-19 shock on public expenditure This section looks at expenditure performance in 2020. It will ask: How did the shock affect expenditures? What were the fiscal measures implemented by the authorities to mitigate the shocks, and how successful were they? We start with a brief background to this section as we summarize the policy and outturns for the 2016–20 stability period. 7 The 2016–20 stability period was marked by an effort to control overall expenditures and shift their composition from recurrent to investment purposes. Figures I.11 and I.12 show that on average total actual expenditures remained at about 27 percent of GDP, and slightly below the planned expenditures, except for the slight deviation in 2020. The financial plan required total expenditure to stay at about VND 8,025 trillion for the stability period, with public investment reaching 25 to 26 percent of total expenditure and recurrent expenditure (excluding interest payment) accounting for less than 64 percent of total expenditure by 2020. Consequently, current expenditures were planned to fall steadily from 65.7 percent in 2016 to 63.8 percent in 2019, and then 64 percent in 2020. Actual spending was maintained consistently close or below as planned, even though the actual spending typically includes carryover from previous years, which is not part of the plan. Despite the COVID-19 pandemic, 2020 recurrent expenditure was estimated at 63.1 percent of total expenditure, while public investment rose to 30.8 percent, thereby meeting the set targets. Total expenditure for 2016–20 was estimated at VND 7,600 trillion, five percent below the planned amount for the stability period. 7 Annex 2 provides a more in-depth review of expenditures in this period. Aggregate Fiscal Impact of COVID-19 21 DP) Figure I.12. Implemented expenditure (% GDP) Figure I.11. Planned expenditure (% GDP) Figure I.12. Implemented expenditure (% GDP) 30 3030 30 25 2525 25 20 2020 20 15 1515 15 10 1010 10 5 5 5 5 0 0 0 0 2020e 2016 2016 2017 2017 2018 2018 2019 2019 2020e 2020e 2016 2017 2018 2019 2020e t triển Rec urr Chi ent xuyên thường Ca pit Chi al tư phát triển đầu Recurrent Capital Interest payment Inte Chi trảp rest lãi nt ayme nợ Contingency Impact of the COVID-19 shock on expenditures In the wake of the COVID-19 shock, the authorities used expenditure measures in two waves and with different objectives. Initially, they announced a package of fiscal measures to attenuate the impact of the April national lockdown and mobility restrictions, continued strict border controls, and the global recession. Table I.4 summarizes the expenditure component of the planned fiscal package costed at about 0.7 percent of GDP to support poorer and most affected households and staff associated with quarantine and medical staff. Vietnam’s total fiscal support to households (including PIT reduction and exemption) was lower than its EAP peers and fell short of estimated earning losses (figure I.13). By the early summer 2020, and considering the flagging national demand affected by the lockdown and global crisis, the authorities accelerated the execution of public investment programs to boost aggregate demand. anned support to households in Vietnam and its EAP peers in 2020 Figure I.13. Estimated earning losses and planned support to households in Vietnam and its EAP peers in 2020 10 8 6 % GDP 4 2 0 Source: from World Bank’s East Asia and Vietnam Thailand Philippines Malaysia Indonesia Mongolia Phi-líp-pin Pacific Economic Update April 2021 Note: Earnings losses are calculated as loss in employment multiplied by changes in wages. Support to household includes by revenue (e.g. PIT reduction and exemption) and expenditure measures (e.g. cash transfer). Earnings loss Support to households 22 Note I The execution of the expenditure component of the fiscal support package was uneven. The programs to support household income were relatively small (VND 36 trillion or 0.6 percent of GDP in 2020), short-lived, and were expected to reach over 13 percent of the population (13.3 million people). It achieved an execution rate of 35.6 percent, partly associated with an underdeveloped social protection system and partly due to the difficulty of identifying and reaching new beneficiaries who work in a large informal economy and are not covered by the existing social safety net. This program is discussed in more detail in an accompanying note in this fiscal update volume. Overall, recurrent expenditure on social protection increased by 10 percent (or about 0.2 percent of GDP) in 2020, reversing the declining trend that started in 2017. Table I.4. Fiscal support package – 2020 – expenditure measures Measure Planned Executed Execution % GDP % GDP % Expenditure measures 0.7 0.3 39.4 Income support for households as per Resolution 42 and 154/NQ-CP 0.6 0.2 35.6 Allowances to medical and other staff and amenities for quarantined 0.1 0.1 59.7 individuals as per Resolution 37/NQ-CP Source: Ministry of Finance (as of 1 June 2021). Estimated recurrent health spending remained flat in 2020 compared to 2019, thanks to the government’s swift and decisive response. Succeeding in the management of the crisis meant lower additional health expenditures. Preliminary estimates from TABMIS8 show that recurrent health expenditures increased by 10 percent (y/y) in the first half of 2020, reflecting higher spending to address the health crisis. As the national measures succeeded in controlling the spread of the virus, however, health expenditure decreased by 12.7 percent (y/y/) in the second half. By the end of 2020, the government had spent about VND 5.3 trillion (or 0.1 percent of GDP) providing allowances to medical workers and other staff, and amenities for quarantined individuals, about 80 percent of its approved budget for these activities (VND 6.7 trillion). Overall, health spending was 1.4 percent lower in 2020 than in 2019.9 This decline may reflect the government’s successful reallocation of budgeted expenditures. It could also reflect the incomplete information on budget execution reported in TABMIS. The ongoing rationalization of current expenditures helped provide resources to cover additional costs of COVID-19 containment. Due to the COVID-19 pandemic, recurrent expenditures increased by 13.3 percent in 2020 compared to 2019, which is twice as much as the rates observed pre-pandemic. The government’s efforts focused on reallocating planned budgets to cope with COVID-19 rather than mobilizing new resources. Since May 2020, the Vietnamese authorities have required central and subnational governments to save up to 70 percent of meeting and travel costs and 10 percent of other remaining recurrent expenditures in 2020. They also postponed salary reforms for public sector workers to 2021. Overall, the central government managed to save VND 49.3 trillion (or 0.8 percent of GDP), which is significant compared to health and social protection expenses related to COVID-19. The central and subnational governments co-financed the expenses incurred to address the COVID-19 shock, proportionate to subnational revenue availability. In addition to the existing 8 TABMIS is the Treasury and Budget Management Information System of Vietnam. 9 Data from MoF, calculations by World Bank staff. Aggregate Fiscal Impact of COVID-19 23 fiscal transfer arrangements, the central government provided targeted support for expenses related to the outbreak containment measures, medical equipment procurement, household income support, and COVID-19 vaccine procurement.10 In general, the size of the transfers varied from zero to 70 percent of spending on COVID-19 by subnational governments, with poorer provinces receiving proportionately more financial support (table I.5). To meet the remaining financing needs, subnational governments could tap into their contingency funds, financial reserve funds, and unspent funds earmarked for the salary reform that was postponed. On the one hand, this arrangement helps provinces with limited revenue collection access necessary health supplies, execute prevention measures, and assist their residents through the crisis. On the other hand, the provinces most affected by COVID-19 might not be able to access needed financial assistance.11 For example, Da Nang city, the epicenter of the second outbreak of coronavirus community transmission, received only 30 percent of its actual COVID-19-related expenses from the central government, even though its total revenue fell by an estimated 16 percent in 2020, and it registered a deficit of about 0.4 percent of regional gross domestic product for the year. Starting in early summer 2020, the Vietnamese authorities significantly accelerated the execution of public investment programs to boost aggregate demand. A well-established literature suggests that public spending—including public investment—can affect aggregate demand and therefore have a “multiplier effect” on output (or GDP) especially during economic downturns. The multiplier effect is usually country specific and depends on structural factors that could amplify or reduce the impact of public expenditures (Annex I.4). Total approved public investment reached a record high VND 660 trillion (about 10.5 percent of GDP) in 2020, an increase of 40 percent compared to initial plans approved by the National Assembly (table I.6). This increase includes carryover from previous years (VND 91 trillion or 1.4 percent of GDP) and higher investment from subnational governments (VND 88 trillion or 1.4 percent of GDP). The higher investment from subnational governments was financed mainly by increased revenues from land use right assignments and a lottery in 2019 (about 1.1 percent of GDP), which are earmarked for public investment. By end-January 2021, VND 525 trillion (8.3 percent of GDP) was disbursed. This disbursement included 80 percent of the budgeted investment plans for 2020,12 which is a significant improvement compared to 2019 (74 percent of the budgeted investment plans) and 2018 (76 percent of the budgeting investment plan). This was achieved thanks to the streamlining of the Public Investment Law (new provisions became effective in early 2020) and the close monitoring of expenses by agencies and subnational governments, which are responsible for almost 75 percent of public investment projects in Vietnam. The government’s success in managing the health crisis, which enabled economic activities to resume quickly and prevented local outbreaks from having spillover effects to non-epicenter provinces, also played an important role in the acceleration of public investment. While it is empirically difficult to capture the direct effects of increasing public investment on growth, indicative estimates suggest it had a positive correlation with growth in Vietnam (Annex I.4). 10 The existing fiscal arrangement includes two types of transfers from the central government to subnational governments: (i) the central government provides subnational governments with very limited revenue collection (that is, those with a sharing ratio equal to 100) to cover their subnational budget deficits; and (ii) targeted transfers, which are earmarked to predetermined programs and given to virtually all provinces. 11 After tapping into all approved resources, if provincial budgets are still unable to meet additional spending requirements, then the central government would cover the gaps. 12 And about 77 percent of the carryover from previous years. Based on World Bank staff calculation and data from MOF’s public investment report number 3149/BTC-DT dated March 29, 2021, also see table I.6 below. 24 Note I Table I.5. COVID-19 intergovernmental transfer scheme in 2020 Province Revenue sharing Transfers from the ratio central government % (% of expenditure) Binh Duong, Dong Nai, Hanoi, and Ho Chi Minh City ≤ 50 0 Ba Ria-Vung Tau, Bac Ninh, Can Tho, Da Nang, Hai Duong, Hai Phong, Hung Yen, Khanh Hoa, Quang Nam, Quang Ninh, Quang Ngai, Vinh Phuc < 100 and > 50 30 Mountainous areas and Central Highlands 100 70 Other provinces 100 50 Source: WB staff compilation based on official data. Note: Revenue sharing ratio indicates the percentage of selective revenues stipulated by the Budget Law that subnational governments can keep for their own budgets. The remaining parts must be submitted to the central government. This scheme applies for allowances to medical and other staff and amenities for quarantined individuals under Resolution 37/NQ-CP, procurement of equipment; Decision 437/QD-TTg, household income support under Resolution 42/NQ-CP; and COVID-19 vaccine procurement under Resolution 21/NQ-CP. Table I.6. Public investment plan and execution in 2020 as of January 31, 2021 Planned Disbursed trillion % GDP trillion % GDP Disburse- VND VND ment rate Carried over from previous year 90.9 1.4 70.0 1.1 77.1 Total 2020 budget 568.5 9.0 455 7.2 80.0 Budget plan approved by NA (in 2019) 470.6 7.5 Cancelled by PM (in Oct. and Dec. 2020) 14.6 0.2 Increased by PM (in 2020) 10.1 0.2 Increased by local governments (in 2020) 87.8 1.4 Total 659.4 10.5 524.8 8.3 79.6 Source: WB staff estimates from MOF’s public investment report number 3149/BTC-DT dated March 29, 2021. Note: NA = National Assembly; PM = Prime Minister. At the central level, public investment concentrated on transport, national defense, and agriculture and rural development, which accounted for nearly three-fourths of total planned investment. Two development banks (Vietnam Development Bank and Vietnam Bank for Social Policies) accounted for over 10 percent, while health and national security made up another 10 percent. The Ministry of Transport improved its disbursement rates substantially to 91.6 percent in 2020, up from only 61.5 percent in 2019. This was followed by the Ministry of Defense and the Ministry of Security, and Ministry of Agriculture and Rural Development (MARD). Disbursement rates by the Ministry of Health remained starkly low compared to other ministries, and even slightly declined in 2020, while education and training and finance also saw declines (figure I.13, panels A and B; and table I.7). In the health sector, the authorities may have prioritized the containment of COVID-19 pandemic over the execution of investment projects. Also, there were some delays in the execution of several hospital construction projects. In the education sector projects are mostly ODA financed, which have experienced slow disbursement. Giao thông Giao thông, 32,6% 37,2% Các NHPT, Y tế, 5,6% 10,6% Các NHPT, 8,5% Quốc phòng, Aggregate Fiscal Impact of COVID-19 NNPTNN, 25 18,0% 13,1% Quốc phòng, 18,5% NNPTNN, 12,6% Figure I.14. Central government public investment composition in 2020 by administrative classification A. Planned (% total planned investment) B. Disbursed (% total disbursement) Others, 12.8% Finance, 1.3% Finance, 2.5% Edu. & Training, 1.9% Others, Edu. & Training, 2.6% 10.2% Security, 5.1% Security, 4.8% Health, 2.1% Transport, Transport, 32.6% 37.2% Health, 5.6% Development banks,10.6% MARD, 13.1% Defense, 18.0% Defense, 18.5% Development banks, 8.5% MARD, 12.6% Source: MOF and WB staff estimates. Note: Excluding VND 4.3 trillion cancelled by the Prime Minister. EVN = State-owned Vietnam Electricity Group; MARD = Ministry of Agriculture and Rural Development;. Table I.7. Central government public investment in 2019–2020 2019 2020 Planned Disbursed Dis. Rate Planned Disbursed Dis. Rate % GDP % GDP % % GDP % GDP % Central Government Total 1.91 1.22 64.2 1.76 1.42 80.5 Transport 0.63 0.39 61.5 0.57 0.53 91.6 Defense 0.38 0.22 57.4 0.32 0.26 82.9 MARD 0.25 0.20 82.9 0.22 0.19 83.6 Development banks 0.15 0.14 95.4 0.15 0.15 100.0 Health 0.12 0.04 31.8 0.10 0.03 29.8 Security 0.07 0.05 74.6 0.08 0.07 84.8 Edu. & Training 0.03 0.02 76.5 0.05 0.03 60.6 Finance 0.03 0.02 61.9 0.04 0.02 41.5 Others 0.26 0.14 55.5 0.22 0.14 64.5 Sources: MOF and WB staff estimates. Note: Excluding VND 4.3 trillion cancelled by the Prime Minister in 2020. At the subnational level, capital expenditures focused on more industrialized regions. About half of total subnational public investment in 2020 was planned in the two largest cities (Hanoi and Ho Chi Minh City) and their surrounding provinces (in Red River Delta and Southeast), which are the country’s industrial hubs. Red River delta (excluding Hanoi) doubled its planned investment due to a higher initial plan approved by the National Assembly and substantial increases during the year funded by subnational budgets (about 45 percent of its total plan). 26 Note I Figure I.15. Public investment by region in 2019–2020 30 Hà Nội; 10,9 While disbursement of national public investment improved, there was heterogeneity across 25 provinces. The improvement at the national level appeared to be driven by only half of the Thay đổi tỷ lệ giải ngân 2019-20 (pps) 20 (figure I.30). In fact, there seemed to be a negative correlation between growth in provinces planned15public investment and change in disbursement rate (figure I.15 and table I.8), signaling that implementation capacity may have Đông Nam bộ; 10,3disbursement. Hanoi, with lower planned hampered 10 investment, performed the TP Hồ Chí best Minh; in terms of accelerating execution as its disbursement rate 9,3 increased5 from 70 percent in 2019 to 93 percent in 2020. In contrast, Trung implementation du & mi slowed ền núi phía Bắc, 13,4 in Red River 0 Delta, the Central Highlands, and North Central & Central Coast, which experienced the highest -5 growth rates in planned ĐB sông C investment. ửu long; For DaMiNang, ền Trung; 17,9has been hard hit by the which pandemic, the planned investment was 13,6 91 percent higher in 2020 than in 2019. As of January 31, -10 Đồ bằng ngits sông Hồng, 20,6 2021, however, the city had managed to disburse 60 percent only4,0 Tây Nguyên; of budgeted investment for -15 2020, which is lower than in 2019 for the city and the lowest among all provinces in 2020. Although the city -20 was the epicenter of the second COVID-19 outbreak during August–October 2020 and was significantly -40 affected -20 by0a series20 40 storms of tropical in October 60 100 shocks 80 2020, these 120 do not seem to be the primary contributors Tăng trto ưởn g vốninvestment slow implementation. kế hoạch 2019-20 (%) Instead, difficulties in land clearance were the major bottleneck that caused the delays in public investment. Figure I.15. Public investment by region in 2019–2020 30 Hanoi; 10.9 Change in disbursement rate 2019-20 (pps) 25 20 15 Southeast; 10.3 10 Ho Chi Minh City; 9.3 5 Mid North Mtns., 13.4 0 North Central & -5 Mekong; 13.6 Central Coast, 17.9 -10 Red River, 20.6 Central Highlands; -15 4.0 -20 -40 -20 0 20 40 60 80 100 120 Planned investment growth 2019-20 (%) Sources: MOF and WB staff estimates. Note: Bubble size is proportionate to regional share of total subnational disbursement in 2020, which is shown in the bubble label. Excluding VND 10.3 trillion cancelled by the Prime Minister. Aggregate Fiscal Impact of COVID-19 27 Table I.8. Public expenditure in 2019–2020 by region 2019 2020 Region Planned Disbursed Disb. rate Planned Disbursed Disb. rate Tri. VND Tri. VND % Tri. VND Tri. VND % North Central & Central Coast 58.8 46.5 79.2 86.5 65.5 75.6 Red River Delta 43.0 39.2 91.2 85.7 75.3 87.9 Mekong Delta 49.5 41.0 82.9 58.2 49.9 85.8 Northern Mid. & Mountains 35.8 28.8 80.6 57.8 48.9 84.6 Southeast 36.4 25.7 70.5 49.0 37.7 76.9 Ho Chi Minh City 37.6 25.4 67.6 44.8 34.1 76.0 Hanoi 48.5 33.7 69.5 43.2 40.0 92.6 Central Highlands 13.0 10.9 83.6 18.6 14.5 77.7 Source: MOF and WB staff estimates. Note: Excluding VND 10.3 trillion cancelled by the Prime Minister. Fiscal trends and the impact of the COVID-19 shock on deficit and debt in 2020 and beyond During 2016–19, the government changed the trend, profile, and terms of publicly held debt, ensuring debt sustainability and an increasing fiscal space. During that period, public and publicly guaranteed debt was reduced from 59.7 percent of GDP to 55 percent13, placing it on a downward trajectory (table I.9) and well within the statutory level of 65 percent of GDP. In addition, the authorities succeeded in lengthening the maturity of the domestic government bonds being issued from 8.7 years in 2016 to 13.4 in 2019, leading to an increase in average time to maturity from 6 years to 7.4 years, reducing rollover risks (table I.10). The authorities also opted for domestic market instruments, increasing the share of domestic debt to over 60 percent by 2020, thus reducing foreign exchange risks, as well. The fiscal deficit increased significantly from 0.5 percent of GDP in 2019 to an estimated at 4.9 percent in 2020. While the fiscal stimulus contained appropriate policy measures to support businesses and households, its overall uptake appears to have been modest at about 2.1 percent of GDP. On the revenue side, the government was able to collect around 98 percent of the revised total revenues projected for 2020.13 The tax relief provided in the April package contained deferrals that ended by December, and the revenue losses associated with other tax exemptions and reductions in the program were largely replaced by the sizable non-tax revenues from the sale of public assets, land rentals, and assignment of land use rights in 2020 (14.2 percent of total revenues collected).15 13 This follows GFS methodology and is sourced from IMF. According to official data published by MOF, public debt was de- creased from 63.7 percent of GDP in 2016 to 55 percent in 2019. This calculation uses non-adjusted GDP numbers. 14 The plan revision was made on 12 Nov 2020, adding VND 26.7 trillion to the initial projections which were made in 2019 (as explained in the revenue section). 15 14.2 percent if excluding SOE equitization proceeds (as explained in revenue section), and 16.1 percent if including it. 28 Note I Table I.9. Government bond portfolio, 2016–2020 Item 2016 2017 2018 2019 Est. 2020 Average issuing coupon rate (%/year) 6.49% 5.98% 4.71% 4.51% 2.86% Average issuing maturity (year) 8.71 12.74 12.69 13.44 13.94 Bonds issued in tenors of 5 years or longer (% total issuance) 92% 100% 100% 100% 100% Average portfolio coupon rate (%) 7.13 6.70 6.28 5.93 5.13 Average time to maturity (years) 5.98 6.71 6.83 7.42 8.41 Source: MOF. On the expenditure side, the authorities reallocated funds from existing budget lines to the COVID-19 response to reduce the impact of COVID-19-related expenditures on the fiscal deficit. For instance, the authorities postponed planned civil service salary increases that were part of the medium-term civil service professionalization reform agenda. They also redirected unspent travel and conference expenditures to cover COVID-19 expenditures. The large increase in capital spending took place within the limits of the approved budget—the execution rate of the public investment program rose from about 74 percent to 80 percent from 2019 to 2020 and was funded by unspent cash reserves.16 As a result, the level of debt is estimated to have increased by only 0.3 percent of GDP, reaching around 55.3 percent at end-2020 (table I.10). The debt was financed mostly by new domestic borrowing (VND 324 trillion or 5.1 percent of GDP in 2020) with relatively low yields and long maturities, reflecting the public creditworthiness and abundance of liquidity in the domestic market.17 The government plans to rely on the domestic market as the major financing source, with maturities extended. Over the next five years (2021–25), the central government needs to raise VND 3,068 trillion to cover its budget deficit, refinance debt, and lend to subnational governments, public institutions, and firms. About 62 percent of this financing requirement will be sourced from issuing bonds in the domestic market, whose borrowing costs have recently been steadily lowered. The focus will be on long maturities to further lengthen the maturity of its borrowing portfolio. The average issuing maturity of its plan will be 9 to 11 years for the period of 2021- 2025 in general, and 14.1 years for 2021 in particular, which is slightly higher than in 2020. As the country graduated from the low-income country group, borrowing from official development assistance/concessionary loan share is expected to decrease from 20 percent during 2016–20 to about 17 percent during 2021–25. The remaining gap will be financed by existing cash reserves (that is, the State Treasury cash balances) and other sources, including the potential issuance of bonds in the international market. Limits on public debt and external debt are widened in the new five-year debt plan for 2021– 25. The new five-year debt plan for 2021–25 stipulates that total public debt should not exceed 60 16 Eighty percent includes VND 14.6 trillion cancelled by the Prime Minister in October and December 2020 in the investment plan. If we exclude this cancelled amount from the plan, the execution ratio would stand at 82.1 percent for 2020. 17 Government bond yields continued to decline amidst strong domestic demand. For example, the yield for five-year government bonds declined from 1.91 percent in 2019 to 1.17 percent on December 31, 2020, while the 10-year yield was down from 3.39 percent to 2.42 percent, and the 20-year yield was down from 4.02 percent to 3 percent during the same period. Aggregate Fiscal Impact of COVID-19 29 percent of revised GDP (or estimated 76 percent of non-revised GDP18), and central government debt will be limited to 50 percent of revised GDP (or an estimated 63 percent of non-revised GDP19). These new prudent ratios are slightly higher than those adopted in the last five-year period, 2016–20 (65 percent and 54 percent of non-revised GDP, respectively), and can give the government more space to cope with unanticipated shocks, a rapidly aging population, and other risks. On the external debt front, the plan imposes a 50 percent of revised GDP (or estimated 63 percent of non-revised GDP) limit on the country’s total external debt, which is also higher than the one set in 2016-2020 (50 percent of non-revised GDP). Economic outlook, fiscal out-turns, and policy implications20 The Vietnamese economy is expected to rebound in 2021 and has positive medium-term prospects. Assuming a continued rollout of vaccines in Vietnam and in the United States, China, and the European Union, the main markets for Vietnamese exports, and firming up of domestic demand, GDP growth is expected to rebound to an estimated 4.8 percent during 2021 (table I.10). Government first quarter GDP growth is estimated at 4.5 percent (y/y), suggesting a healthy growth dynamic and a gradual convergence toward the pre-pandemic trajectory. Growth is projected to be sustained at around 6.5 percent during 2022–24, reflecting continued improvements in external conditions for manufacturing exports and a firming of service sector demand, anchored in a growing middle class, resumption of domestic tourism, and steady contribution by the agriculture sector. The authorities have already adopted a neutral fiscal stance in 2021 and appear intent on resuming fiscal consolidation thereafter (table I.10). The authorities project lower revenue collection (in percentage of GDP) in 2021 and then gradual improvements in revenues afterward. The projected decline in 2021 is partly the result of falling non-tax revenues, which were exceptionally high due to a one-time rise in transactions in 2020. Domestic tax revenue was affected by the economic slowdown and the COVID-19-related tax relief provided, and the authorities did not expect a full recovery of tax revenues in 2021. In fact, given the Socialist Party and national elections during November 2020–May 2021 and the new government taking power, there is no expectation of tax policy reform during 2021–22. 18 World Bank estimation of non-revised GDP. 19 World Bank estimation of non-revised GDP. The official debt ceilings in 2021-2025 are denominated in the revised GDP. The government has started to publish aggregate revised GDP numbers that are approximately 26 percent higher than the before- revision GDP numbers used in this note as well as in the previous five-year debt plan. To make the thresholds in 2021-2025 comparable to those in the previous five-year period, thus, we convert them to ratios denominated in the before-revision GDPs, assuming that the revised GDP is 26 percent higher. 20 Assumptions about economic developments as of end as of Q2 2021. 30 Note I Table I.10. Vietnam Key Economic Indicators – baseline scenario21 2016 2017 2018 2019 2020e 2021f 2022f 2023f National Accounts (% change) Real GDP 6.2 6.8 7.1 7.0 2.9 4.8 6.5 6.5 Private consumption 7.3 7.4 7.3 7.4 0.5 4.0 7.4 7.3 Government consumption 7.5 7.3 6.3 5.8 6.2 6.9 5.0 4.9 Gross Fixed Capital Formation 9.9 10.2 8.2 8.7 4.1 7.6 9.6 9.5 Exports of goods and services 13.9 16.7 14.3 6.7 5.0 14.3 11.6 11.5 Imports of goods and services 15.3 17.5 12.8 9.5 3.4 14.1 12.3 12.1 GDP per capita 5.1 7.9 8.5 5.8 2.6 4.6 8.0 8.8 Sectoral contributions to growth (pps) Agriculture 0.2 0.4 0.6 0.3 0.4 0.5 0.3 0.3 Industry and construction 2.6 2.8 3.1 3.2 1.4 2.8 3.2 3.2 Services 2.7 2.9 2.7 2.8 0.9 1.1 2.3 2.3 Public Sector Finances (% GDP) Total revenue and grants 23.9 24.6 24.6 24.8 23.5 23.0 22.3 22.0 Tax revenue 17.9 18.3 18.4 18.6 16.9 16.1 16.4 16.6 Oil revenue 0.9 1.0 1.2 0.9 0.5 0.4 0.4 0.3 Non-oil tax revenue 17.0 17.3 17.2 17.7 16.3 15.7 16.0 16.2 Taxes on Goods and Services 9.2 9.1 9.2 9.2 8.3 8.1 8.2 8.2 Direct Taxes 5.0 5.1 5.3 5.6 5.5 5.5 5.6 5.7 Taxes on International Trade 2.1 1.9 1.6 1.6 1.3 1.4 1.3 1.3 Other Taxes 0.6 1.1 1.2 1.3 1.2 1.1 1.2 1.3 Non-tax and capital revenues 5.8 6.2 6.0 6.1 6.5 6.3 5.5 5.0 Grants 0.2 0.2 0.1 0.1 0.1 0.1 0.1 0.1 Total expenditure 27.8 27.1 25.9 25.3 28.4 28.9 28.3 27.4 Expenses 20.4 19.7 19.0 18.5 19.8 20.3 19.8 19.2 Education 4.0 4.1 4.0 3.9 4.1 4.0 4.0 4.0 Interest payment 1.9 2.0 1.9 1.8 1.7 1.6 1.7 1.7 Net acquisition of non-financial assets 7.4 7.3 6.9 6.8 8.6 8.7 8.4 8.1 Overall fiscal balance -4.0 -2.5 -1.3 -0.5 -4.9 -6.0 -5.9 -5.4 Primary fiscal balance -2.0 -0.5 0.6 1.3 -3.2 -4.3 -4.2 -3.7 Public and publicly guaranteed debt 59.7 58.2 55.0 55.0 55.3 58.3 59.0 58.8 Memorandum items GDP (nominal, trillion dong) 4,503 5,006 5,542 6,037 6,293 6,672 7,339 8,122 GDP (nominal, US$ billion) 205.3 223.8 245.2 261.9 271.2 286.0 311.5 341.4 Sources: WB staff estimates based on official data (Government Finance Statistics [GFS] methodology). Note: Projections for this table are based on assumptions about economic developments as of end Q2 2021. These calculations use non-adjusted GDP numbers. pps = percentage points. 21 Projections for this table are based on assumptions about economic developments as of end as of Q2 2021 Aggregate Fiscal Impact of COVID-19 31 To improve their revenue collection in the medium term, the authorities have just announced creation of the Large Taxpayer Office. Going forward, the government is also finalizing a new tax strategy for fiscal years 2021–30, which includes consideration of measures to broaden the tax base; streamline tax incentives and exemptions in the CIT; rebalance the tax load for efficiency and equity; and expand the environmental tax base and target pollution in production, use, and post- use. This could be an opportune time to also improve revenue projection practices and lay out a transparent practice of managing large cash balances. Expenditures are expected to rise slightly in 2021 and to fall thereafter. Continued civil service reforms including application of the “2 out, 1 in” rule to new hiring and contracting out within-government services are expected to help. Public investment is expected to revert to pre-pandemic levels. Given these revenue and expenditure trends, the fiscal deficit is estimated to increase from 4.9 percent of GDP in 2020 to 6 percent in 2021, before starting to decline in 2022. This temporary increase will require additional borrowing from the domestic and/or international markets and will lead to a rise in the public debt-to-GDP ratio to 58.3 percent at end-2021, and about 59 percent in 2022, before starting to decline again. Economic growth and the fiscal consolidation trajectory could be affected by delays in the rollout of vaccines22 or recurring or widening outbreaks. The slow rollout of national vaccinations could put Vietnam at a disadvantage compared to competitor countries if it delays the opening of international borders and the movement of people, affecting the recovery of the services sector. As of early May 2021, 500,000 vaccine doses had been administered. Vietnam expects to receive about 60 million doses of vaccine by end-2021. Recurring or widening outbreaks could lead to widespread economic disruptions, affecting the recovery dynamics in the industrial and services sectors. For instance, the fourth outbreak that started in late April 2021 has affected a number of provinces and the three main cities of Hanoi, HCMC and Da Nang and led to stringent mobility and social distancing measures, affecting the services sectors. It has also led to shutdown of some industrial zones and firms in the north of Vietnam and may affect the electronics global value chain. These health measures, while fully justified, affect businesses and the population through lower economic activities, loss of employment and lower household income. By extension, if the outbreak is not brought under control quickly, tax revenues may be affected while the government may decide to roll our further assistance to the affected businesses and households, potentially leading to an increase in the fiscal deficit in 2021. A slower recovery of the global economy can also affect economic growth and the fiscal consolidation trajectory. While the global economy is projected to rebound in 2021, there are still major uncertainties that could affect Vietnam’s major export markets, the United States, China, and the European Union. In the low-case scenario, Vietnam would recover more gradually, with GDP growth of around 3-3.5 percent in 2021, recovering to 6.5 percent from 2023 onward. While agriculture would continue its steady growth, the industrial sector and especially services could take longer to recover, limiting their contributions to growth. On the demand side, private consumption and investment would remain subdued in 2021, affected by global uncertainties, 22 As of early April 2021, Vietnam had started its phase 2 clinical trials for domestic vaccines. The government expects to obtain 60 million doses of vaccine in 2021, including 4 million doses through the COVAX (COVID-19 Vaccines Global Access) facility and the rest from AstraZeneca. 32 Note I the April 2021 outbreak and a slow vaccination rollout. A slower pace of global recovery would reduce the demand for Vietnamese exports, while imports would be affected by lower domestic final demand and input demands for exports. The authorities would be expected to continue their accommodative fiscal and monetary policy to support aggregate demand in 2021 before returning to a more prudent stance in the following years. Public debt would rise temporarily but is expected to fall in the medium term and would remain sustainable throughout the period. Recommendations Looking forward, proactive revenue and expenditure policies and management can help enhance revenues while supporting implementation of a sustainable and green development agenda without compromising fiscal sustainability. This note highlights the opportunities to improve the tax policy framework in the medium term and the quality of fiscal management, and discusses options for use of the existing fiscal space. n On the revenue front, while tax policy reforms are not expected to be adopted in the immediate future, as Vietnam is finalizing transition to a new government and a new parliament, tax administration reforms should be a high priority. Also, this could be an opportunity to increase budget credibility by improving revenue projection practices and retaining large cash reserves balances. n On tax administration, stepping up compliance enforcement and implementing measures to increase the tax base and restore and stabilize tax collection will be key to supporting a more inclusive and green recovery. The recent decision to establish a Large Taxpayer Office to manage the most important group of taxpayers would be a step toward more effective tax collection. The authorities will need to adopt a modern operating model for the new LTO to work effectively. Furthermore, they will also need to consolidate and build specialized skills and expertise for the LTO to address the largest taxpayers’ risks comprehensively. This process would take time before efficiency gains could be achieved. n On tax policy, the government is developing a new tax policy framework for the medium to long term. Vietnam’s 2021–2030 Socio-Economic Development Strategy (SEDS), adopted in February 2021, emphasizes digital transformation as a key driver to renew its growth model and enhance its competitiveness, and sustainability through promoting climate resilience and green growth. A forward-looking tax policy that would support such an inclusive, digitally transformative, and sustainable growth model would need to strengthen international taxation, including digital taxation, and adopt environmental taxation—such as carbon taxation—which would support greening of the economy in the medium to longer term. Also, policy reforms should be carried out to sustainably maintain the tax revenue performance from current taxes and improve the efficiency and equity of the system. Such reforms include streamlining generous CIT incentives and introducing a property tax to expand the tax base while enhancing the efficiency of land use. Additional suggestions are provided in box I.2. Aggregate Fiscal Impact of COVID-19 33 On the expenditure front, the relatively rigid expenditure framework and nested intergovernmental fiscal framework still provides the authorities opportunities for improvement. Spend more and spend better. In the face of a sharp economic shock and at a time when debt is on a sustainable path, Vietnam has the fiscal space to spend more. This space can be used to develop the country’s productive capacity and social protection network. However, to ensure effective use of the resources, improving quality of expenditure will be critical. n Develop an inclusive and responsive social protection system. Given the aging population and the coming to light of the weaknesses in the social protection system during the 2020 COVID-19 crisis, consideration should be given to developing a more systematic social safety net system, digitalizing it on both the intake and service delivery ends. n Improve expenditure quality and efficiency. This would include further review of public investment management practices, given the chronic under disbursement of the investment budget during the 2016–19 stability period. n Given the critical infrastructure investment needs of the country and its objective to move to a more sustainable economic model, productive investment could be considered, especially in green and sustainable technology to create a base for such an economic transformation. On fiscal balances and fiscal stance: Vietnam has fiscal space and its debt is sustainable in the medium term. n In the short run, it may be premature to adopt a neutral fiscal stance given the ongoing effects of the shock. A more accommodative fiscal stance that supports vulnerable households and small businesses and continues to support aggregate demand may be more appropriate. This may delay the fiscal consolidation in the short term but need not affect the consolidation trajectory in the medium term. n In the long run consider development and use of automatic stabilizers to improve the ability of the system to deliver timely counter-cyclical response. This would be the revenue and expenditure reforms suggested above, and entail broad based taxation, and broad based and better targeted social protection programs, including the use of active labor market policies. Box I.2. Tax reform to increase and stabilize tax collection and increase fiscal space (adapted from the World bank recommendations to MOF on the Vietnam Tax Reform Strategy 2021-2030) l Corporate income tax (CIT) reform could include measures to improve performance through rationalization of CIT incentives. The current rate of 20 percent is lower than the regional and global averages and acts as an incentive. Rationalizing tax incentives will help protect the base and minimize unintended non-neutrality in capital treatment. The widespread allowance of generous and ever-growing incentives, aggravated by fragmented legislation, usually leads to heightened fiscal costs, duplications and inconsistency across legal frameworks, and implementation complexity, which weaken incentives governance and make it harder to introduce sunset clauses into the existing tax incentive structure. 34 Note I l Personal income tax (PIT) reform could improve the equity of the tax system while enhancing tax administration efficiency. The PIT regime amounts of deductions, size of brackets, and marginal rates have been fixed since 2014. This allows inflation to deplete the real deductions and push incomes into higher tax brackets (the bracket creep problem), and thereby unfairly raises tax revenues at the cost of equity, disproportionately on low-wage workers. The PIT regime should be adaptive with the allowance for annual inflation adjustments for standard deductions and income brackets. Furthermore, the number of PIT brackets could be reduced to just three or a maximum four. Currently, the employment income tax table consists of seven brackets with rates ranging from 5 to 35 percent (with no exemption bracket, which is assumed to be implicit in the level of personal deduction). l Value-added tax (VAT) reform should focus on enhancing the efficiency of the VAT regime. The standard rate of 10 percent is already low relative to other regional countries and globally. The reduced rate of 5 percent should be abolished. Under the rate unification schemes, the MOF could review the current list of items in the 5 percent VAT category and determine which should be moved to the standard regime and which to the exemption list. With the same objective, Vietnam could consider narrowing the list of exemptions to only three specific groups: basic education, primary health care, and selected financial services. Exemptions are typically given to attain two purported objectives: (1) enhancing equity, and/or (2) excluding hard-to-tax sectors. Extensive exemptions would break the VAT chain and leads to risks—a narrower base or cascading taxationa (both would be efficiency reducing). Furthermore, zero-rating could be used instead of VAT exemption to help domestic production avoid cascading effect of VAT exemption. l Property tax. The property tax regime should move away from reliance on a one-time tax collection linked to property sales and the levy of transfer /registration charges to the more sustainable recurrent property taxation. Vietnam could consider reviewing and evaluating the phased implementation of non- agricultural land use tax policies to increase revenue mobilization from land and introduce housing tax. Such measures will encourage efficient use of land and housing, help reduce land and housing speculation, improve reasonable and stable revenue mobilization, and ensure an appropriate balance between Vietnam’s socio-economic conditions and good international practice. A priority for excise tax reform is a higher excise on tobacco products. In 2018, the tax in the retail price of Vietnam’s most popular brand of cigarettes was just less than 37 percent, significantly lower than the average of more than 58 percent for all middle-income countries. The World Health Organization has recommended that the total share of the tax burden on all tobacco products should be more than about 75 percent of the retail sales price. Similarly, Vietnam should consider applying a mixed structure with both specific and ad-valorem rates to improve the effectiveness of the excise tax on alcohol consumption. The rates on alcohol are comparable to the ones in advanced tax systems. A straightforward option would be to retain the existing ad valorem tax and add onto it a specific component. Such a combination would elevate the pros and mitigate the cons of these two types of rates, with expectedly better outcomes in both raising revenues and discouraging excessive consumption. Note: a. A cascade tax or cascading tax is a turnover tax that is applied at every stage in the supply chain, without any deduction for the tax paid at earlier stages. Aggregate Fiscal Impact of COVID-19 35 Annex I.1: Performance of main taxes A brief review of the performance of the main taxes provides insight into the overall tax policy and administration issues in Vietnam and suggestions for further reforms to increase and stabilize tax collection. Corporate income tax (CIT) performance The CIT-to-GDP collection ratio declined by 0.3 percentage points between 2015 and 2019. The decline is mainly due to the reductions in the CIT tax rate from 22 percent in 2015 to 20 percent in 2016, and to the substantial foregone tax revenue from CIT incentives. According to Ministry of Finance (MOF) estimates, foregone revenues from these incentives account for 30 percent to almost 40 percent of the total CIT intake (or about a quarter of the potential CIT without incentives). Furthermore, the incentives are granted to only about 3 percent of registered CIT taxpayers. To a lesser extent, reductions in oil revenue caused by decreases in crude oil production and crude oil prices also contributed to the fall in CIT over time.23 However, in a cross-country comparison of the level of CIT intake and productivity, Vietnam’s performance suggests a solid CIT-to-GDP collection ratio. Figure I.16 shows the relatively sound level of the Vietnam CIT-to-GDP collection compared to the regional averages, while figure I.17 highlights that CIT productivity in Vietnam is relatively high in international comparisons. With the CIT productivity ratio at 0.21, Vietnam ranks third among 14 countries in EAP and higher than most of the selected lower middle-income regional peers. Figure Figure I.17. I.16.CIT I.16 collection in regional comparison CIT productivity Figure Figure I.17. CIT productivity 6.0 6,0 0.35 6 0,35 6 0,30 5.0 5 0.30 5,0 5 0,25 4 0.25 0,20 4.0 4,0 4 3 0.20 0,15 3 0,10 3.0 2 0.15 3,0 0,05 1 2 0.10 0,00 2.0 0 2,0 1 0.05 Nhật Bản Thái Lan In-đô-nê-xi-a Phi-líp-pin Ga-na Dăm-bi-a Sing-ga-po Ma-lai-xi-a Hàn Quốc Cam-pu-chia U-dơ -bê-ki-xtan Môn-đô-va Niu Di-lân Lào Việt Nam Ốt-xtrây-li-a Mông Cổ Trung Quốc Hôn-đu-rát 1.0 0.00 1,0 0 Malaysia Philippines Vietnam Australia Indonesia Lao PDR Zambia Thailand Japan China Honduras Korea, Rep. New Zealand Cambodia Uzbekistan Moldova Ghana Singapore Mongolia 0.0 0,0 2015 2016 2017 2018 2015 2016 2017 2018 ĐÁ-TBD Thu nhập trung bình East Asia & Pacific Lower middle income Interquartile range Vietnam thấp Khoảng tử phân vị Việt Nam Efficiency ratio (LHS) CIT as % of GDP (RHS) Hệ số hiệu suất (LHS) TNDN Thuế in Emerging Market EAP(% GDP) (RHS) High Income in EAP Thị trường m ới nổi ĐÁ-TBĐ Thu nhập cao ĐÁ-TBĐ Sources: Vietnam MOF, IMF, and WB staff calculations. 23 Oil CIT/GDP declined by one-half a percentage point, from 1.2 percent of GDP in 2015 to 0.7 percent of GDP in 2019. 36 Note I Personal income tax (PIT) performance PIT collection in Vietnam has been relatively low compared to the three group EAP averages: the middle of the interquartile, average high income, and emerging EAP (figure I.18). However, the PIT-to-GDP ratio has increased by 0.5 percentage points since 2015, reaching 1.8 percent of GDP in 2019 despite deductions, tax brackets, and marginal tax rates remaining constant. The estimated PIT productivity of Vietnam ranks on the low side of the regional comparators (table I.11). omparison Figure I.18. PIT collection in regional comparison Table I.11. PIT productivity among the ASEAN-4 6.0 (1) (2) (3) = (2)/(1) 5.0 ASEAN 4 Standard PIT Efficiency PIT rate Revenue Ratio 4.0 (% of GDP) 3.0 Philippines 32 2.47 0.077 2.0 Malaysia 28 2.11 0.075 1.0 Thailand 35 1.79 0.051 0.0 Vietnam 35 1.57 0.045 2018 2015 2016 2017 2018 Indonesia 30 0.92 0.031 Interquartile range Vietnam ao ĐÁ -TBD Emerging Market in EAP High Income in EAP Average 32 1.77 0.056 Sources: Vietnam MOF, IMF, and WB staff calculations. Sources: Vietnam MOF, IMF, and WB staff calculations. Note: ASEAN = Association of Southeast Asian Nations. Value-added tax (VAT) performance The VAT performs relatively well. Vietnam’s VAT collection is above the average of the comparable EAP countries (figure I.19). Both the VAT productivity and C-efficiency ratios are robust.24 Figure I.20 shows that the C-efficiency in Vietnam’s VAT ranks on the side of high performers compared with other EAP and selected global peers. Figure comparison I.20. Figure VAT I.19. C-efficiency VAT comparison collection in regional comparison Figure I.20. VAT C-efficiency comparison 1,20 7.0 1.20 1,00 1.00 0,80 6.0 0,60 0.80 0,40 5.0 0.60 0,20 4.0 0.40 0,00 0.20 Phi-líp-pin Dăm-bi-a U-dơ -bê-ki-xtan Thái Lan Ma-lai-xi-a Sing-ga-po Nhật Bản In-đô-nê-xi-a Cam-pu-chia Ốt-xtrây-li-a Việt Nam Ni-ca-ra-goa Mông Cổ Trung Quốc Niu Di-lân Lào Hôn-đu-rát Môn-đô-va Hàn Quốc 3.0 0.00 2.0 Malaysia Thailand New Zealand Vietnam Japan Indonesia Lao PDR Philippines Zambia Mongolia Singapore Korea, Rep. Australia China Cambodia Honduras Nicaragua Moldova Uzbekistan Ghana 1.0 0.0 ĐÁ-TBD Thu nhập trung bình 2015 2016 2017 thấp 2018 East Asia & Pacific Lower middle income 2018 Hệ số hiệu suất C (2017) Châu Á & TBD (2010) C-efficency Ratio (2017) Asia & Pacific (2010) Interquartile range Vietnam cao ĐÁ -TBD Emerging Market in EAP High Income in EAP Sources: Vietnam MOF, IMF, and WB staff calculations. 24 C-efficiency refers to the collection efficiency ratio of the VAT. It is the ratio of the VAT revenue to aggregate consumption, divided by the standard VAT rate. Aggregate Fiscal Impact of COVID-19 37 Special consumption tax (SCT) performance The revenue intake from the SCT Figure I.21. SCT collection in regional comparison Figure I.21. SCT collection in regional comparison (with other excise regimes) fluctuated within a rather narrow band, (with other excise regimes) around 1.7 percent of GDP during 4,0 4,0 2016–19. This tax effort is notably lower 3,5 3,5 than the average in both the interquartile 3,0 3,0 range and the emerging EAP (figure I.21). 2,5 2,5 2,0 2,0 1,5 1,5 1,0 1,0 0,5 0,5 0,0 0,0 2015 2016 2017 2018 2015 2016 2017 2018 Khoảng tứ phân vị Việt Nam Interquartile range Vietnam Thị trường mới nổi ĐÁ -TBD Thu nhập cao ĐÁ -TBD Emerging Market in EAP High Income in EAP Sources: Vietnam MoF, IMF, and WB staff calculations. 38 Note I Annex I.2: Expenditure performance – general trends This section looks at expenditure performance.25 It will ask questions such as what where the global trends of current and capital expenditure in the years preceding the COVID-19 pandemic and how did the shock affect expenditures? What fiscal measures were implemented by the authorities to mitigate the shocks, and how successful were they? The 2016–20 stability period is marked by an effort to control overall expenditures and shift its composition from recurrent to investment purposes. Figures I.22 and I.23 show that, on average, total actual expenditures remained at about 27 percent of GDP, and slightly below the planned expenditures, except for the slight deviation in 2020. The financial plan required total expenditure to stay at about VND 8,025 trillion for the stability period, with public investment reaching 25 to 26 percent of total expenditure and recurrent expenditure (excluding interest payment) accounting for less than 64 percent of total expenditure by 2020 (figures I.23 and I.24). Consequently, current expenditures were planned to fall steadily from 65.7 percent in 2016 to 63.8 percent in 2019, and then 64 percent in 2020. Furthermore, the actual spending was maintained consistently close to or below as planned, even though the actual spending typically includes carryover from previous years, which is not part of the plan. Despite the COVID-19 pandemic, 2020 recurrent expenditure was estimated at 63.1 percent of total expenditure, while public investment rose to 30.8 percent, thereby meeting the set targets. Total expenditure for 2016–20 was estimated at VND 7,600 trillion, five percent below the planned amount for the stability period. ) I.23. FigureI.22. Figure Implemented Planned expenditure expenditure (% GDP) (% GDP) Figure I.23. Implemented expenditure (% GDP) 27,8 30 27.0 26,8 27.8 27,1 27.5 27.1 28,4 27.8 30 30 26.8 27.1 25.9 28.4 30 25,9 25,3 25.3 25 25 25 25 20 20 20 20 15 15 15 15 10 10 5 10 10 5 0 5 5 0 2016 2017 2018 2019 2020e 020e 0 0 2016 2017 2018 2019 2020e 2016 2017 2018 2019 2020e Recurrent Capital át triển Interest Chi trả nợ lãipayment Con Chi tingency đầu tư phát triển Recurrent Capital Total Chi thường xuyên Tổng chi Interest payment Total 25 This subsection summarizes global trends of current and capital expenditure in the years preceding the COVID-19 shock and the impact of the pandemic on expenditures in Vietnam. The main points of interest are: n On current expenditures: data permitting, a special focus on subsidies and economic transfers, social transfers, and intergovernmental transfers would help understand whether and how the COVID-19 shock has impacted these flows and, by extension, service delivery. The focus on these three areas in this note will help shed light on the implementation issues associated with the support package (current expenditures) and set the broader perspective for the other notes in this report. n In the case of capital expenditure, a special point of interest is whether the recent push to increase public investment to boost the economy is having the intended effects of boosting demand and economic activities at the subnational levels. Data permitting, we will look at the execution rate and potential multiplier effects associated with this increase in public investment. To make this part of the analysis manageable, we will focus on Da Nang city, as the epicenter of the July outbreak. Aggregate Fiscal Impact of COVID-19 39 Figure I.24. Composition of planned expenditure Figure I.25. Composition of actual Figure (% total I.25. Composition planned of actual expenditure (% of total expenditure) actual expenditure) expenditure (% of total actual expenditure) 100 100 100 100 80 80 80 80 60 60 60 60 40 40 40 40 20 20 20 20 0 0 2020e 2016 2017 2018 2019 2020e 0 0 hát triển 2016 2017 2018 2019 2020e 2016 2017 2018 2019 2020e Chi thường xuyên Chi đầu tư phát triển Current Capital Interest payment Contingency Current Capital Interest payment Chi trả nợ lãi Sources: WB staff estimates based on MOF data. Note: Data for 2020 are MOF’s second budget estimates, except: (i) the plan in 2020 excludes additional expenditures of VND 26.7 trillion approved by the National Assembly on November 12, 2020; and (ii) executed public investment in 2020 is estimated by total disbursement as of January 31, 2021, according to the MOF’s public investment report numbered 3149/BTC-DT dated March 29, 2021.. Lower recurrent expenditures were in part driven by cuts in administration and social protection spending. On the one hand, as figures I.25 and I.26 show, administration expenditure declined steadily, partly because the government embarked on reforms to make its administrative system leaner and more effective and to professionalize its civil service. Spending on social protection also did not keep up with total expenditure and economic growth, so that its shares have declined since 2017. Health expenditure doubled between 2016 and 2018, decreasing slightly thereafter. This trend continued in 2020, despite the COVID-19 shock, as the government’s early success in controlling community spread of the virus limited additional health expenditures. On the other hand, during the last five years, current expenditure on education and training was relatively stable at about 15 percent of total expenditure while current expenditure on science and technology made up about 0.8 percent of total spending each year. Economic affairs and environmental functions received a higher share of the budget. 40 Note I Figure I.26. Recurrent expenditures by function, Figure I.27. Recurrent expenditures by func- Figure unction, 2016–20 I.27. (% GDP) 2016–20 Recurrent expenditures by function, 2016–20 (% GDP) (% tion, total expenditure) 2016–20 (% total expenditure) 20 20 80 80 15 15 60 60 10 10 40 40 5 520 20 0 00 0 020e 2016 2016 2017 2017 2018 2018 2019 2019 2020e 2020e 2016 2017 2018 2019 2020e ng Quản lýtinhà Admistra on nước Kinh tế & & Economic environment môi trường Admistration Economic & environment Educa tion Giáo dục & Đào tạo Health Y tế Education Health Science & technology Khoa học & công nghệ Culture Văn hóa&& recrea tion giải trí Science & technology Culture & recreation Đảmprotec Social tion bảo xã hội Khácrecurrent Other Social protection Other recurrent Sources: WB staff calculations based on MOF and GSO data. Note: Dates in 2020 are from MOF’s second budget estimates, except: (i) expenditures in administration, culture and recreation, economic affairs and environment, health, and social protection in 2017 are from the second budget estimates, and during 2018–20 are estimated from expenditure growth extracted from TABMIS (Treasury and Budget Management Information System); and (ii) Other recurrent expenditures include those on national defense and security. The disbursement of public investment slowed during 2016–19 due to difficulties implementing the new Public Investment Law 2014, which mandated strict borrowing limits, fund allocation, and transfer procedures. As part of the consolidated fiscal policy initiated in 2016, the authorities set strict borrowing ceilings with the aim of reducing the debt-to-GDP ratio to more sustainable levels. Also, the rules and procedures introduced in the new public investment law turned out to be much more cumbersome than expected, leading to unexpected delays in the approval of investment projects, as their inclusion in the medium-term investment plan became mandatory for receiving funds. Allocation and transfer of funds were also subject to multiple layers of checks and balances that led to delays in the implementation of projects. One effect was that the execution rate has been much lower for projects financed by external borrowing than projects financed by domestic revenue (figure I.27).26 A comparison of the implementation of public capital investment in the first nine months of 2016–19 shows that less than 60 percent of approved public investment was executed in the first three quarters, significantly lower than the over 65 percent achieved in 2015. This finding applies even to 2020, when the authorities increased disbursement of public investment in early summer to support aggregate demand in response to the COVID-19 crisis.27 26 For example, in 2019, only 38.2 percent of externally funded projects were executed, while the corresponding figure for domes- tically funded projects was over 79 percent (World Bank, Taking Stock, December 2019.) 27 Budget estimates for the first nine months typically capture the implementation of public expenditure budgeted for the re- porting year, making the disbursement rate comparable across years (I.16). The practice of including the disbursement of public investment carried over from previous years in the final quarter of the current budget year makes it impossible to assess the implementation of capital expenditure just by comparing actual outlays with planned investment from budget reports for the entire current budget year. While focusing on the first nine months of the year overlooks the efforts expended by the government in the last quarter, it provides a useful comparison. Aggregate Fiscal Impact of COVID-19 41 The amendments made to the Public Investment Law in 2019 proved to be effective in accelerating the execution of capital investment programs. The revisions, which came into force in January 2020, introduced an annual update mechanism and a provision to ensure continuity of project preparation between consecutive medium-term investment plans (MTIPS).28 Also, provisions were made to delegate the review of fund allocation from the Ministry of Planning and Investment (MPI) and MOF to provinces for projects using provincial funds, and to allow spending agencies to allocate funds among projects to facilitate their implementation. As a result, disbursement rates improved significantly in 2020 (figure I.28); execution of ODA-funded projects (that is, external), particularly, improved significantly at both the central and subnational government level. The targeted investment transfer from the central government to subnational governments also improved. nt disbursement rate in 2018–2020 (percent of planned) Figure I.28. Public investment disbursement rate in 2018–2020 (percent of planned) 100 100 80 80 60 60 40 40 20 20 0 0 Nước ngoài Central Sub-national Domestic External vố n Tổng Implementing gov. Sources Total 2020 2018 2019 2020 28 This is done by allowing each MTIP to commit up to 20 percent above the total estimated investment cost for projects implemented in the following MTIP. 42 Note I Annex I.3: Additional figures and tables te of capital investment in the first nine months Figure I.29. Disbursement rate of capital investment in the first nine months, 2016–2020 80 80 60 60 40 40 20 20 0 0 8 2019 2020 2015 2016 2017 2018 2019 2020 Bình quân 2016-2020 Disbursement Av. 2016-20 Source: MOF. Table I.12. All fiscal measures Measure Planned Executed Execution rate % GDP % GDP % Revenue Measures 4.0 1.8 44.8 Tax (CIT, PIT, VAT) and land rent deferral 2.9 1.1 37.4 of which: VAT and PIT deferral for household businesses 0.1 0.0 18.9 Excise tax (SCT) deferral for automobiles produced or assembled domestically 0.3 0.3 111.2 Exemption and reduction of taxes, fees and charges, and land rents 0.8 0.4 47.6 CIT reduction 0.4 N/A Raised family allowance in PIT 0.2 N/A Exemption and reduction of other taxes, registration fees, and land rents 0.3 N/A Reduction of fees and charges 0.0 N/A Expenditure Measures 0.7 0.3 39.4 TIncome support for households as per Resolution 42 and 154/NQ-CP 0.6 0.2 35.6 Allowances to medical and other staff and amenities for quarantined 0.1 0.1 59.7 individuals as per Resolution 37/NQ-CP Total 4.7 2.1 44.0 Source: MOF (as of 1 June 2021). Aggregate Fiscal Impact of COVID-19 43 Table I.13. Public investment execution in 2019–2020 2019 2020 Planned Disbursed Disb. rate Planned Disbursed Disb. rate % GDP % GDP % % GDP % GDP % Total 7.3 5.4 73.7 9.0 7.2 80.0 Source of fund Domestic 6.3 5.0 79.3 8.1 6.8 83.5 External 1.0 0.4 38.2 1.0 0.5 50.0 Implementation government Central gov. 1.9 1.2 64.2 1.8 1.4 77.4 Domestic 1.5 1.0 68.6 1.5 1.2 82.6 External 0.4 0.2 46.8 0.3 0.2 54.8 Subnational gov. 5.3 4.2 77.9 7.2 5.8 80.4 Funded by subnational budgets 3.8 3.3 84.9 5.4 4.5 84.1 Targeted transfers from the central gov. 1.5 0.9 59.9 1.8 1.3 69.8 Domestic sources 0.9 0.7 73.7 1.2 1.0 81.7 External sources 0.5 0.2 36.2 0.6 0.3 46.4 Source: WB staff estimates from MOF data. Note: Disbursement in 2019 and 2020 is by 31 January 2020 and 2021, respectively (13 months). Figure I.30. Uneven public investment disbursement across provinces Disclaimer: The boundaries, colors, denominations, and other information shown on any map in this work do Source: WB staff estimates from MOF data. not imply any judgement on the part Note: Green indicates disbursement rate in 2020 was higher than or equal to the one in 2019, and otherwise is red. of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. 44 Note I Annex I.4: Mobility restrictions, investment disbursement, and regional growth – indicative correlations A well-established literature suggests that public spending—including public investment—can have a “multiplier effect” on output (or GDP); that is, changes in public spending lead to changes in output, though capturing the direct effect of such policies on growth is difficult. These fiscal multipliers tend to be larger during economic downturns than during expansion, because in downturns, public spending supports aggregate demand, while during expansion, public spending may crowd out private demand, leaving output unchanged. The effects of public spending multipliers are usually country specific and depend on a number of structural factors that could amplify or reduce the impact of public expenditures. The positive or negative effect of these factors on the multiplier are identified by a +/- as follows: size of the economy (+), labor market rigidities (+), flexibility exchange rate regime (-), openness of the economy (-), existence of automatic stabilizers (-), effectiveness of monetary policy (-), debt level (-), efficiency of public expenditure management, and revenue administration (+). There are alternative methods to assessing the effects of public spending on output. These include full macroeconomic models and demand-side models. A simple back-of-the-envelope method to estimate the short-term effects of public spending on growth is to use the “bucket method” (IMF, 2014).” Based on this method, given Vietnam’s level of public expenditure management and revenue administration, and that it is experiencing a negative output gap and has a crawling peg, it could be assessed as having a medium-range spending multiplier ranging from 0.3 to 0.6. This means that a VND 1,000 increase in public spending could have a VND300 to VND600 increase in GDP level. Preliminary regression analysis suggests that increased public investment had a positive correlation with regional growth (RGDP), although mobility restrictions appear to have affected its disbursement. A simple regression analysis shows that, on average, for every percentage-point increase in public investment disbursement rate, regional GDP growth are estimated to have increased by 0.06 in 2018 and 0.13 percentage points in 2020 (figure I.32). One potential explanation is that in 2018 the economy was performing at full potential so the impact of public investment should not be large. On the other hand, in 2020, given a slack economy, public investment was expected to have a larger impact growth more. In comparison, the same regression was run for 2018 when the economy was at full capacity, and the results show a positive but smaller impact. On the other hand, figure I.31 shows that as mobility decreased (from zero to -50 percent), so did public investment rates. Data available on mobility restrictions in places of retail and recreation were used as proxy for restriction measures at provincial levels. The finding suggests that restrictions on mobility also affected official business and contractors and workers implementing investment projects. Aggregate Fiscal Impact of COVID-19 45 Figure I.31. Mobility and disbursement I.33.Fiscal FigureI.32. Figure Fiscal multiplier multiplier effects effects = 0,37x + 90,05 y = 0,37x + 90,05 110 Hanoi 20 y = 0,06x + 3,55 20 R² = 0,07 R² = 0,07 Disbursement rate in 2020 (%) R² = 0,06 100 15 15 Tăng trưởng RGDP (%) 90 RGDP growth (%) 10 10 80 5 5 4 70 0 0 60 -5 -5 50 HCMC -10 y = 0,13x - 7,87 -10 Minh Da Nang R² = 0,11 40 -15 -15 -50 -40 -30 -20 -10 0 40 50 60 70 80 90 100 110 40 5 -10 0 Disbursement rate (%) Average change in mobility in places of retail & iải trí (%) recreation (%) 2020 2018 Source: WB staff estimates from MOF and Google Community Mobility data. Note: The estimated coefficient for public disbursement rate in 2020 is statistically significant at the 5 percent level. 46 Note I