SOCIAL PROTECTION & JOBS | P  OLICY & TECHNICAL NOTE NOVEMBER 2020 | No. 23 SUMMARY Where is the money The unprecedented and ongoing coming from? scale-up of social protection responses Ten stylized facts on financing to the COVID-19 pandemic dwarf social protection responses the response to the Great Recession. to COVID-19 But how are countries financing such scale-up efforts? This note lays out 10 stylized findings from a rapid review Mohamed Almenfi, of social protection financing sources Melvin Breton, Pamela Dale, in 31 countries, including in terms Ugo Gentilini, Alexander Pick, of composition between external and domestic resources, and specific and Dominic Richardson1 modalities within each. I. Introduction Since the onset of the COVID-19 health crisis, and associated stringency measures, governments have acted in support of businesses and individuals by activating fiscal stimulus measures and introducing or expanding social protection programs. Initiatives to collect, analyze and monitor social protection responses have generated a wealth of evidence on the most aspects 1 The note presents preliminary findings for discussion and only reflects the personal views of the authors. We are grateful to Penny Williams, Amjad Zafar Khan, Delphine Prady and David Coady for precious comments and support. Almenfi and Gentilini are with the World Bank; Breton, Dale and Richardson with UNICEF; and Pick is with the OECD. For more info: ugentilini@worldbank.org. 1 SOCIAL PROTECTION & JOBS | POLICY & TECHNICAL NOTE  NOVEMBER 2020 | No. 23 of the crisis (Gentilini et al 2020; ILO 2020; IMF 2020a; Second, depending on the scale and affordability of the Oxford University 2020). However, limited information is initial responses — and learning directly from responses available on how these responses are financed. to the Global Financial Crisis of 2007–08 in large parts of the world — how these were financed may influence the This rapid review maps and compares sources of likelihood of austerity following stimulus. Social protection financing for COVID-19 social protection measures in a measures have not escaped austerity measures in the selection of countries. Its objective is to understand how recent past, which can impose a grave cost on the most COVID-19-related social protection responses are financed, vulnerable in society 4 but it may not go into extensive detail on the implications for sustainability and affordability2. Third, understanding finance choices, particularly where a range of options was available — will help determine both Practically, this rapid review involved web-based searches the adequacy of the response, and the opportunity costs for official reports from governments or international for selection of one source of finance over another. Where organizations on the finance sources for the 41 countries global finance mechanisms have been used, assessments with the largest horizontal expansion since the onset of the of adequacy, opportunity cost, and strategy can also COVID-19 crisis3. Findings are reported for 31 countries be considered. where sufficient evidence was available. This is only our initial attempt to map and compare sources of finance, Fourth, understanding the sources of financing during the and the results reported here are preliminary. Preparation initial response phase — and potentially responses during for a more comprehensive initiative is underway. future waves — can inform the design of measures to strengthen the responsiveness and resilience of a social Understanding how COVID-19 responses are financed is protection system in the future. important to the future of social protection in at least five ways. First, the financial sustainability of social protection Finally, any recommendations that follow from the work responses is increasingly unclear in many countries, noted above, or other sources attempting to assess the especially given that there is no end in sight to the global efficacy of national or international response to COVID-19, pandemic. In examining this query, it is important to bear are more likely to be actionable with a fuller understanding in mind that most programs are short-term by design and of the constraints on financial resources. The remainder not intended to ‘compete’ with established schemes. In a of the note sets out emerging findings, followed by way, there is a question on whether, and to what extent, brief conclusions. programs that worked well during the crisis, and that addressed preexisting gaps in coverage, could become II. Emerging Stylized Findings permanent. Or whether other emergency programs would subside, become ‘dormant’ and be activated in crises only. We compare financing across domestic and external sources. The former includes spending restructuring such as reallocation or reprioritized committed budget as well as tax relief and exemptions; national debt and deficit 2 The authors are developing a comprehensive initiative generating in-depth, detailed information on the financing of social protection. 3 See Gentilini et al (2020), IMF (2020a) and Oxford University (2020). 4 See Tirivayi et al (2020) and OECD (2019). 2 SOCIAL PROTECTION & JOBS | POLICY & TECHNICAL NOTE NOVEMBER 2020 | No. 23 measures; and tapping state reserves5 and contingent  risis spending on social protection is 20% Fact 2. C funds. Externally, international financing institutions have higher now than a decade ago but uneven between countries. provided funds in the form of concessional loans and grant Based on estimates from administrative data by Gentilini assistance, while bilateral and multilateral development et al (2020), about 18% of stimulus spending is devoted actors provided most assistance in the form of grants. Ten to social protection. This amounts to about $790 billion, a stylized facts emerge from the analysis. level 21% higher than the $653 billion for 2008–09 (or between 24–27% of economic package spending for that Fact 1. Current pandemic spending is more than twice as large as the response to the Great period) estimated by Ortiz and Cummins (2013). Recession of 2008-09. Based on data on overall stimulus packages from Prior to COVID-19, low- and middle-income countries spent 41  countries, economic measures average 9% of GDP an average of $242 million and $4.8 billion, respectively, in 2020, against 4% of GDP in 2008–09 (figure 1). on social protection6. The difference is particularly striking if we consider that spending in 2008–09 refers to 18 months, while spending Average COVID-19 social protection spending is $243 per for 2020 covers just over a semester (March-October capita — ranging from $695 in high-income countries to 2020). The minimum national spending levels for the only $4 in low-income settings. The latter amounts to only two periods are similar (about 0.2–0.3% of GDP), while 0.51% of GDP per capita. maximum levels for 2020 include countries like Japan that spent an unprecedented 42% of GDP on fiscal stimuli.  pending helped ramp up an unprecedented Fact 3. S but short-term response. Figure 1. Economic stimulus packages, 2008–09 Programs like cash transfers, which represent 31% of and 2020 the 1,179 social protection measures implemented in 212  countries or territories, have been scaled up considerably. This includes a planned or actual 45% 42.28% horizontal expansion in coverage by 217% relative 40% to pre-COVID-19 levels and covering 17% world’s Fiscal spending (% of GDP) 35% population (1.3 billion people). Benefit levels also nearly 30% doubled, with an increase in the size of cash transfers 25% 19.92% of 93%7. Where data on implementation progress is 20% available, or for 33  programs, performance is rather 15% 10% 8.87% encouraging, with the difference between planned and 5% 4.07% actual coverage levels amounting to just 3.7 percentage 0.35% 0% 0.22% points. However, 68% of programs are new; the average duration of programs is short (3.3 months); and only 1) 1) =4 =4 (n (n 08 20 average 20 20 6 Calculated as average country spending using the 1.5% of GDP 5 Savings and reserves can often be used interchangeably. However, figure as per World Bank (2018) estimates. the latter term should not be confused with contingency reserve, 7 Pre-COVID data shows that safety net transfers accounted for which is a specific part of the annual budget allocation in certain 26% of the income or consumption of people living on less than countries to allow for unforeseen disasters in a given year. $1.9/day. 3 SOCIAL PROTECTION & JOBS | POLICY & TECHNICAL NOTE  NOVEMBER 2020 | No. 23 Domestic financing External financing Countries Spending Debt and deficit State reserves/ IFIs Bilateral and reallocation contingent funds/ multilateral fiscal savings development actors AFR 2 1 7 4 Congo, Rep. X X Kenya X X Liberia X X Mauritania X X X Nigeria X X Sierra Leone X South Africa X X EAP 3 6 2 1 Hong-Kong X Indonesia X X X Japan X Malaysia X Philippines X Singapore X South Korea X X X Thailand X ECA 3 2 2 2 1 Serbia X X X Ukraine X X X Uzbekistan X X X X LAC 4 4 1 8 7 Argentina X X X X Bolivia X X Colombia X X X Costa Rica X X X X Dominican Republic X X X X Ecuador X X X El Salvador X X Peru X X MNA 2 2 2 1 Egypt X X Morocco X X X X Turkey X North America 1 US X SAR 1 1 Pakistan X X Grand Total 15 14 7 21 13 4 SOCIAL PROTECTION & JOBS | POLICY & TECHNICAL NOTE NOVEMBER 2020 | No. 23 16  out of 340 programs were extended for a second Fact 5.The level of domestic financing wave of support (for an average of 3.4 months). varies remarkably across the country income spectrum. Those shares increase from nearly zero in LICs, to 37% in Fact 4. Financing is diversified in sources8. LMICs, and to 47% in UMICs (and 100% in HICs10) (figure Based on information available for 31 countries and 70 data 2). However, in some cases, the injection of external points, we divided sources into domestic and international, financing has been limited and most spending would each including a select number of sub-categories9 (table come from domestic sources. For example, Pakistan spent 1). The most prevalent domestic modality was restructuring approximately $872 million on its emergency response. or re-prioritizing budget lines in 15 countries; 14 countries Out of this, $109 million went on additional support incurred in domestic debt and deficit spending; while for the 4.5 million regular safety net beneficiaries and 7  tapped state reserves, contingent funds and fiscal $763 million to 10.5 million new beneficiaries via the savings. These strategies were not mutually exclusive Ehsaas program. Within such response, support from the and 48% of countries pursued mixed-source financing. World Bank represented only 3% of total government In general, 32% of countries tapped domestic sources as spending on the emergency response. Similarly, the the only source of financing and 19% relied on external Indian government’s first package of social protection resources only. Figure 2. Share of domestic and external financing by country income group (left) and region (right) Domestic nance External nance 100% Share of source of nancing 8% 90% 80% 43% 40% 50% 53% 70% 63% 57% 57% 63% 60% 77% 50% 100% 100% 100% 92% 40% 30% 57% 60% 50% 47% 20% 43% 43% 38% 37% 10% 23% 0% ) ) ) ) ) ) 2) ) 7) 8) 1) 8) ) =1 31 31 =5 13 =3 =3 =2 = = = = =1 = n= (n n= (n (n (n (n (n (n (n (n (n (n l( ic a C A P l( A C R C R EA HI IC IC LA SA AF LI EC N ta ta er LM M UM To To Am al ps on h ou rt gi No gr Re e m co In 8 While ‘diversified’ in the context of social protection financing can sometimes evoke the balance between contributory or specific part of the annual budget allocation in certain countries non-contributory, we here focus mostly diversification within to allow for unforeseen disasters in a given year). IFIs (WBG and/ social assistance. or IMF) includes receiving external funds from the WB and/or IMF, 9 Spending restructuring (domestic) includes measures such as whether in the form of loan or grant, to finance fiscal measures cutting spending, reallocation of committed budget, tax relief/ including social protection policies. Finally, UN, bilateral donors, exemptions, etc. National debt/deficit (domestic) includes other international actors includes receiving external funds from borrowing domestically and increase the country’s domestic debt other international actors, whether in the form of loan or grant to and budget deficit. State reserves, contingent funds and fiscal finance fiscal measures including social protection programs. savings (domestic) includes tapping on country’s saving and 10 In some cases, however, European high-income countries have reserves to spend on fiscal measures (contingency reserve is a... borrowed from supra-national institutions like the BCE. 5 SOCIAL PROTECTION & JOBS | POLICY & TECHNICAL NOTE  NOVEMBER 2020 | No. 23 assistance was about $23 billion, with nearly $1 billion Insurance Fund’s surplus. In Indonesia and the Philippines, supported by the World Bank. the response reallocated funds from nonurgent goods expenditures and those not yet programmed. Singapore, Fact 6.There are several innovations in lower as mentioned, is an example of utilizing reserves: it income settings. drew SP$900 million from past reserves to fund its In Mauritania, a special fund for social solidarity was $1.1 billion Solidarity Payment response (universal created with a state contribution of $170 million. This one-off cash transfer). Uzbekistan tapped on $324 million helped support 206,000 households with a cash transfer of  from their sovereign wealth fund (the Fund for the $60/person. Now there is a possible extension of the fund Reconstruction and Development of Uzbekistan) and for an additional $13.5 million. In Uzbekistan, $44.8 million restructured their budget, such as postponing the was generated through public-private partnerships, non-priority projects and expenditures to post-crisis period, including donations and charity from individuals and to finance COVID-19 response. businesses being directed to the “Kindness and Support” program of the Ministry of Family Support. Fact 10. It’s no just about social assistance. Social insurance programs can be effective and responsive Fact 7.In high-income contexts, there are different counter-cyclical measures as they typically replace a ways financing universal cash transfers. meaningful proportion of income, kick in automatically In Singapore, a one-off cash transfer of $1.1 billion was and there is little likelihood of them being removed financed from reserves and contingent budgets; Hong Kong post-crisis. As mentioned, for instance, worth noting that spent $9.16 billion for cash payouts, all via deficit spending; South Africa’s response was partly financed by the UIF’s in Serbia, the universal cash benefit cost $712  million, surplus. Similarly, in the US Moffitt and Ziliak (2020) show which was part of the 3.9 billion stimulus package. Half a sharp uptick in unemployment insurance alongside of that was financed by a Eurobond valued at €2 billion, the core social assistance instrument, the supplemental re-prioritized existing budget, and currency reserves. nutrition assistance program (SNAP).  ocial protection spending can come at the Fact 8. S expense of other social services. III. Some Concluding Reflections In Ukraine, part of the process included cuts in subsidies, regional budgets, social services, the planned census, At the time when new confirmed cases of COVID-19 are at among others. In order to meet higher expenditures, the an all-time high, many countries are engaging in an “all-in” project budget deficit was increased to 7.5% of GDP. Most effort to save the economic texture of societies. Concerns ministries experienced a reduction of respective budgets, around fiscal space, sustainability and financing sources for except a few like Ministry of Health, Finance, Social Policy, those measures, including for social protection, have not and Internal Affairs. yet emerged as first-order priorities in the response. The objective of this note is not necessarily to wave red flags Fact 9. Tapping reserves and re-prioritizing spending on macroprudent policies and fiscal sustainability; there is a frequently adopted strategy. is already an existing stream of analytical work pursuing South Africa announced an emergency spending package those issues (IMF 2020b). Instead, the note is motivated by of $30 billion, about 29% of which was financed through an interest to provide a broad-based view on main sources reprioritization and drawing from the Unemployment 6 SOCIAL PROTECTION & JOBS | POLICY & TECHNICAL NOTE NOVEMBER 2020 | No. 23 of financing for social protection responses to COVID-19, to access. In the case of the pandemic, these are typically and to reflect on emerging themes from those modalities. lumped together with broader stimulus financing and are not easily distinguishable. Future initiative could help All countries have scaled up social protection in response provide precious detailed and comparable information to COVID-19. However, the extent of this scale-up, the on financing sources and modalities, such as initiatives relative importance of different types of social protection in to (i) clarify the level of financing and standardize the emergency response, and the sources of financing for templates and definitions of financing sources; (ii) this scale-up differ between countries. There are questions enhance our understanding of the composition of social on whether temporary responses will become permanent: protection financed during both the pandemic and non- in part, this would depend on understanding how diverse crisis times; and (iii) document how successful countries sources of funding interact over time. For instance, running have pursued a sustainable mix of financing modalities. large deficits now might lower the prospects of future Combined, these initiatives would help in enhancing expansion; some financing comes from cutting other accountability and evidence-based policymaking. services and reprioritized spending lines will only become fully clear once the year is over. Looking ahead, the conversation on financing dove tails to how future crisis responses can be made more While financing is a key factor, information on funding effective and efficient. Efforts to strengthen crisis sources for social protection — whether domestic, preparedness and informing scale up efforts based external, or a combination thereof — is not always easy on objective and verifiable triggers would be another important and promising area for further exploration. 7 SOCIAL PROTECTION & JOBS | POLICY & TECHNICAL NOTE  NOVEMBER 2020 | No. 23 References Gentilini, U., Almenfi, M., Dale, P., Palacios, R., Natarajan, H., Galicia Rabadan, G., Okamura, Y., Blomquist, J., Abels, M., Demarco, G. and Santos, I. (2020) “Social Protection and Jobs Responses to COVID-19: A Real-Time Review of Country Measures”. Living paper version 13 (September 18, 2020). Washington DC. IMF (2020a) “Policy Tracker”. Online tool. Washington, DC. IMF (2020b) ”Fiscal Monitor October 2020 – Policies for the Recovery”. Washington DC. Moffitt R. and Ziliak, J. (2020) “Covid-19 and the US Safety Net” NBER, Working Paper 27911. Cambridge. OECD (2019) “Can Social Protection Be an Engine for Inclusive Growth?” Paris. Ortiz, I. and Cummins, M. (2013) “The Age of Austerity: A Review of Public Expenditures and Adjustment Measures in 181 Countries”. Columbia University, Initiative for Policy Dialogue and the South Centre, Working Paper. New York. Oxford University (2020) “Oxford Supertracker: The Global Directory for COVID Policy Trackers and Surveys”. Oxford. Tirivayi, N., Richardson, D., Gavrilovic, M., Groppo, V., Kajula, L., Valli, E. and Viola, F. (2020) “A rapid review of economic policy and social protection responses to health and economic crises and their effects on children: Lessons for the COVID-19 pandemic response”. UNICEF Innocenti Center Working Paper 2020-02. Florence. World Bank (2018) The State of Social Safety Nets 2018. Washington DC. Zhang, Y., Thelen, N. and Rao, A (2010) “Social Protection in Fiscal Stimulus Packages: Some Evidence.” UNDP/Office of Development Studies, Working Paper. New York. © 2020 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: +1 (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, its Board of Executive Directors, or the governments they represent. The World Bank does 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 judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. RIGHTS AND PERMISSIONS The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications: The World Bank Group 1818 H Street NW Washington, DC 20433, USA fax: +1 (202) 522 2625 e-mail: pubrights@worldbank.org. Project 75335 8