CREATING FISCAL SPACE TO BUILD HUMAN CAPITAL Public Finance Review Benin Structure of the report Part I Macroeconomic dynamics: Drivers of fiscal space • Macroeconomic trends 01 • Revenue mobilization 02 • Expenditure 03 Part II Spending better to fulfil the potential of human capital • Education 04 • Health 05 • Social Protection 06 2 01 Rebuilding fiscal space is necessary to sustain growth and poverty reduction in a context of tightening financing conditions Benin’s development challenges are sizable and require sustained high growth and poverty reduction; creating fiscal space to achieve this goal is a precondition. In 2020, Benin officially transitioned to a lower middle-income country. Sustaining this path going forward requires strengthening the foundations to support the structural transformation of its economy and a more inclusive growth process. Poverty is still high : in 2019 38.5% of the population was living below the national poverty line; human development outcomes are low: in 2020 the Human Capital Index (HCI) was among the lowest globally. In the current global context (post COVID-19, war in Ukraine), sustaining human capital-oriented public spending (in health, education and social protection) without jeopardizing access to markets or the sustainability of its debt, will require rebuilding it fiscal space by addressing structural fragilities mostly through (1) greater revenue mobilization; (2) more efficient spending in social sectors that represent one third of total public expenditure (through both allocative and technical efficiency gains, and greater equity considerations). The PFR discusses these challenges and opportunities. The PFR compares Benin to structural and other peers chosen through a data-driven process. Structural peers are Togo, Rwanda and Senegal; while other peers include Morocco, Ghana, Sri Lanka and Tunisia. Benin will also be compared to WAEMU and ECOWAS regional peers and to the broader LMIC and SSA country groupings. 3 Part I Macroeconomic dynamics: Drivers of fiscal space Vulnerabilities have risen in the short-term due to the fiscal impact of pro- cyclical fiscal policy and tightening financing conditions The authorities continue to adhere to a sound macroeconomic policy framework, but vulnerabilities have risen in the short term Prudent fiscal policy before the COVID-19 crisis gave authorities significant space to increase spending: the fiscal deficit (incl. grants) was 0.5% of GDP in 2019, with the primary balance reaching -0.1% of GDP. An expansionary fiscal policy in 2020-21 prevented the pervasive effects of stronger economic deceleration. But short-term vulnerabilities have consequently increased: since 2019 the fiscal deficit has increased by 5.2 ppts and PPG debt by 8.6 ppts of GDP to reach 52.3 percent of GDP in 2022. According to the November 2022 DSA, Benin is currently at moderate risk of external and overall debt distress. However, the space to absorb shocks is limited, with external debt burden indicators breaching high-risk thresholds in selected stress scenarios related to exports and commodity prices. This highlights the importance of creating fiscal space to ensure debt sustainability in a context of increasing borrowing costs since the onset of the war in Ukraine and the tightening of monetary policy in the EU and US. Fiscal and primary balances (% of GDP) Debt service to revenue ratio PPG and external debt (% of GDP) 60.0 3 25 historical trend 2 50.0 1 20 0 40.0 -1 15 30.0 -2 -3 most extreme shock 20.0 10 -4 -5 10.0 -6 5 Most extreme shock: One-time depreciation 0.0 2019 2010 2011 2012 2013 2014 2015 2016 2017 2018 2020 2021 2022e 2017 2019 2021 2022e 2010 2011 2012 2013 2014 2015 2016 2018 2020 0 2022 2024 2026 2028 2030 2032 Fiscal balance (incl. grants) Primary balance (incl. grants) Total PPG debt External debt 4 Part I Macroeconomic dynamics: Drivers of fiscal space Fiscal policy will need to play a greater role in promoting inclusive growth Spending has been counter-cyclical but the public sector contribution to growth has been low Output gap and fiscal deficit (% of GDP) Demand-side growth decomposition Despite acting as a stabilizing force over the past 10 years, the scope of automatic stabilizers is very limited 7 4 8.0 because of the small government size and underdeveloped social programs, untargeted and 6 3 % of GDP, output gap sometimes regressive; also limiting the scope for an 5 2 inclusive fiscal policy. 4 1 3 Hence, the bulk of adjustments has come from 0 3.0 discretionary reductions in more elastic expenditures. 2 -1 For instance, the volatility of public investment 1 spending (as measured by its standard deviation) is 0 -2 about 4 times greater than that of social transfers. But 2012 2013 2010 2011 2014 2015 2016 2017 2018 2019 2020 2021 2023p 2022e -1 -3 2000-2010 2011-2019 2016-2019 discretionary fiscal policy can be poorly targeted and -2 -4 -2.0 timed. As a result, social and infrastructure spending have historically contributed little to GDP growth. Net exports Inventory Public investment contributions to growth – averaging Public investment Private investment 0.2 percentage points (ppts) of GDP over the last 10 Output Gap Fiscal deficit Government consumption Private consumption years and marked by large fluctuations – have been Real GDP growth below peers. Source: MFMod and authors’ calculations 5 Part I Macroeconomic dynamics: Drivers of fiscal space 02 Revenue mobilization remains the inevitable instrument for larger fiscal space Low revenue mobilization is a key challenge for greater spending. To maintain the level of public spending needed to sustain growth and poverty reduction, Benin needs to first and foremost focus on generating additional domestic revenue. Revenue levels remain extremely low : 4 percentage points (ppts) below the average in SSA. As such, they not only constrain spending, but they also limit the borrowing capacity. Tax revenue contributes the bulk of the government’s revenue and should increase in importance as income per capita levels continue to rise, but tax revenue to GDP is much lower than peer countries and well below the level expected for its GDP per capita, averaging 10.2 % over 2016-21 (and 10.9 % over the entire decade). Tax revenue to GDP (%) Revenues composition Benin’s tax revenue potential is estimated to be well above its actual level of tax mobilization, with scope to Tax revenue to GDP (%) increase revenue from taxes on goods 15 and services and on income 10 Gap in revenue collected as a % WAEMU SSA SSA LIC LIC of GDP with: 5 2019 1.4 1.3 2.7 2.1 0 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2015-19 1.9 1.8 3.1 2.5 Other Taxes Direct Revenues Table 2.1. Estimated tax gap between Benin and Taxes on Goods and Services Taxes on International Trade benchmark groups (% of GDP) Tax Revenues 6 Part I Macroeconomic dynamics: Drivers of fiscal space Tax policy has been simplified in recent years, but key tax bases and rates are still not structurally geared towards resilient revenue generation Benin is falling behind in terms of the absolute size of indirect tax revenue, collecting less than all its structural and aspirational peers, and well below SSA and WAEMU averages. Compared to WAEMU, it is significantly behind on VAT revenue to GDP. The same is true for excises. Aspirational peers such as Morocco collect more than five times the amount on excises (2.6 % of GDP), while Ghana collects three times that amount (1.5 % of GDP). Even compared to structural peers, Benin collects the lowest. It also compares unfavorably to its structural peers in terms of direct tax revenue collection. Further, and despite improvements, the productivity of Benin’s tax system is relatively low across all core tax instruments PIT and CIT levels remain well The efficiency of VAT is lower than peers CIT productivity is below peers despite similar rates below aspirational peers 40 3 PIT and CIT (% of GDP) vs aspirational peers (2019) 0.6 25 CIT rate (right axis) 35 8 0.5 20 30 7 2 6 0.4 25 C-efficiency 15 % 5 20 0.3 4 10 15 0.2 1 3 10 2 5 0.1 5 1 0 0 0 0 0 Benin Togo Rwanda Burkina FasoGhana Mali Senegal Côte d'Ivoire Togo Rwanda Benin Senegal Tunisia Morocco Ghana Benin Tunisia Tunisia Morocco Benin Morocco Ghana Ghana PIT CIT C-efficiency Standard rate 7 Part I Macroeconomic dynamics: Drivers of fiscal space Tax expenditures in key revenue-generating sectors undermine efforts to increase the tax base Revenue forgone due to tax expenditures is high in Benin, with several tax exemption schemes that appear inefficient. Total tax expenditures averaged 2% of GDP and 15.7% of tax revenue over 2016-20. Benin’s tax expenditure provisions are embedded in the General Tax Code, the Investment Code, the Customs Code, Mining and Petroleum Codes, Special Economic Zones, Finance Laws, Decrees and Conventions. Tax expenditures are embedded in major taxes, primarily customs, CIT and VAT. Domestic VAT still accounts for the bulk of tax expenditures (75.1 percent), with about 100 measures accounted for in 2020. CIT tax exemptions are fewer than VAT exemptions in number, but their revenue forgone has increased significantly in recent years. Most exemptions are granted through the General Tax and Investment codes, with reductions in statutory tax rates for industrial enterprises and reductions in rates for new enterprises contributing the largest shares of CIT tax expenditures. Close to 40% tax expenditures are on basic foodstuff Tax expenditures on goods and services are relatively high (2019) 6.0 5.0 22.1 17.3 Others 4.0 Basic foodstuff 3.0 3.8 Motorbikes and similar equipment 2.0 1.0 Agricultural inputs 18.4 38.3 Medicines and medical material Togo Benin Mali Senegal Burkina Côte Faso d'Ivoire Tax on goofd & services International trade tax Income tax FDI (rhs) Source: DGI, Annual tax expenditure report (2020); IMF 2022. 8 Part I Macroeconomic dynamics: Drivers of fiscal space High levels of labor informality significantly constrain the tax base… The tax base remains narrow due to the prevalence of informal employment; thus, only a small number of large taxpayers contribute the bulk of revenue. 73% of individuals in the highest income bracket are considered informal and thus do not contribute to personal income taxation. For indirect taxation, the situation is more complex due to the linkages between formal and informal Taxes paid by individuals represent 35 percent of total consumer markets, as well as informal trade with Nigeria. Further, statutory rates do not reflect actual tax payments, due government revenue, with most of the tax collection to the VAT design, which includes exemptions; evasion caused by informal transactions; and in general, to any other still relying heavily on large firms. Yet, Benin has the compliance matters in the VAT chain. Broadly, however, the CEQ estimates that consumption informality is high across the smallest number of CIT and VAT taxpayers of all board, averaging 77% of all purchases. At these rates, the non-inclusion of VAT net among the first eight deciles of the WAEMU countries expenditure distribution hovers around 77%, while the last two deciles present marginally lower rates. Still, 72 % of the consumption of the top decile is still considered informal. Direct and indirect taxes are mostly paid by the Share of informal consumption by income decile: Benin has one of the lowest number of taxpayers in richest households, but they still burden the poor 72 % of the richest decile is not paying VAT WAEMU, which limits the tax base 250,000 30 100% 79 78 Direct Tax Indirect Tax 86% 78 77 78 78 78 200,000 25 Share of informal consumtpion 78 77 77 80% 77 20 76 150,000 76 15 60% 75 100,000 74 10 34% 73 72 50,000 40% 5 72 0 0 16% 71 20% 12% Benin Burkina Togo Senegal Côte Mali 4% 5% 6% 8% 10% 7% 70 0%2% 0%3% 0% 1% 1% 1% 2% 3% (2021) Faso (2018) (2018) d'Ivoire (2018) 69 (2016) (2020) 0% 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Corporate Income Tax Personal Income Tax Expenditure deciles Value Added Tax Population (million) Source: EHCVM 2018/19 and authors calculations Source: EHCVM 2018/19 and authors calculations Source: IMF 2022b 9 Part I Macroeconomic dynamics: Drivers of fiscal space … but because of these structural fragilities, fiscal interventions remain broadly redistributive Because of these characteristics, fiscal interventions remain broadly redistributive. The findings of the CEQ analysis show that overall, the fiscal system reduces inequality by about 2 Gini points under the baseline scenario. Most taxes are progressive, which reduces the tax burden on the poor, with income taxation the most progressive element of the tax system. VAT and customs duties are also found to be progressive, given the high levels of informality (including the informal trade with Nigeria); as are excise taxes, as they are levied on luxury items or items not consumed by poorest households. They still, however, burden the poor as most uniform indirect taxation does, through the level of consumption that is formal and because of the pass through between formal and informal markets. Overall, fiscal systems that rely heavily on indirect taxation and where social spending is not targeted tend to increase poverty. Such correlations are also confirmed in Benin. Indirect taxes are concentrated similarly to income – not surprising given Direct personal income taxes collect little from the bottom 50 percent that they are levied on consumption and untargeted 0.8 0 0.7 Excise: genl. -2 Percent of pre-fiscal income 0.6 Concentration shares -4 Customs 0.5 0.4 In-kind user fees -6 Fuel taxes 0.3 Indirect taxes -8 Direct taxes 0.2 -10 VAT Prefiscal 0.1 -12 0.0 Payroll 1 2 3 4 5 6 7 8 9 10 -14 1 2 3 4 5 6 7 8 9 10 Decile of Pre-fiscal income PIT Decile of Pre-fiscal income Source: CEQ 2022, forthcoming Source: CEQ 2022, forthcoming 10 Part I Macroeconomic dynamics: Drivers of fiscal space Tax and customs administration reforms have supported the recent increases in tax revenue collection and should continue going forward. The reforms implemented since 2016 have allowed Benin to significantly modernize tax 1. processes and the tax administration. Key reforms included the institutional reinforcement of the tax authority through the creation of large taxpayers and tax policy units in 2018-19; and more recently the overhaul of the DGI’s institutional structure, including the creation of a new tax policy directorate including new units in charge of risk management, internal audit and the management of standardized invoices to combat fraud The digitalization of key processes was fundamental for increasing tax revenue. 2. Key measures have included the roll out of electronic VAT invoices, the e-declaration and e-payment of corporate taxes for most MSMEs and large firms, the mobile payment of the motor vehicle tax and property taxation, the linkage of the tax and customs IT-systems, and the creation of a unique taxpayer identifier (IFU). A new strategic plan (2022-25) for the Customs administration has been developed. It is 3. organized around nine axes defining the strategic priorities for reforms. The most important relates to the progressive implementation of the strict application of transactional values at customs to reduce the discretionary use of the adjusted valuation system, that creates distortions and revenue loss in the context of the informal trade with Nigeria. Improving the customs information system and its use to further improve the control of transactions; and strengthening and optimizing key customs functions should reduce evasion. 11 Part I Macroeconomic dynamics: Drivers of fiscal space 03 Constrained government expenditure emphasizes the importance of more efficient spending, including through greater public financial management Public spending in Benin is lower compared to its peers, constrained by the low performance of domestic resource mobilization. Over the last decade, with overall public spending at 15.8% of GDP, Benin significantly underperforms the WAEMU and SSA averages (excluding Benin) at 20.5% and 24.8% of GDP, respectively. The level is also significantly below all its structural peers, which are notably between 11 to 5 ppt above (Rwanda (26.2 % of GDP), Senegal (23.2 % of GDP) and Togo (19.5 % of GDP)) . Capex has been the most volatile, but has recently Government spending has historically been very low surged in response to COVID-19… 30 16 Capital Expenditures (% of GDP) Rwanda 14 %age of GDP 25 12 20 10 Senegal 8 15 6 Benin 10 4 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2 0 Benin SSA (Excl. Benin) WAEMU (Excl. Benin) 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Peer economies Aspirational economies Source: Authors’ calculations based on MFMod and WEO data Source: Authors’ calculations based on MFMod and WEO data 12 Part I Macroeconomic dynamics: Drivers of fiscal space Recurrent spending has historically represented the lion’s share of spending, even if it has declined since 2015 and its composition has shifted Education receives the largest allocation, but infrastructure sectors have seen the highest growth. Current expenditure represents 2/3 of total public spending, despite declining in recent years. Recurrent expenditure accounted for 72% of total government spending between 2010 and 2016. It has since started a gradual decline in line with downsizing plans, its overall share averaging 66% of expenditures in 2017-20. The wage bill has declined but interest payments have increased significantly until 2020, driven by the surge in public debt. Between 2011-15, social sectors – education, health, social protection, sports – represented on average 29 percent of total public spending. They have declined since. Energy and urban infrastructure has seen the largest increases in 2016-19 compared to the previous five years. Still, education remains the sector with the highest allocation, on average 20.9% of total budget in 2016-20. Interest payments averaged 1.4 percent of GDP between 2016- COVID-19 has accelerated the increase in capex. Education has the highest allocation, but 19, compared to 0.3 percent of GDP in the previous 5 years. 16.0 Composition of recurrent spending (% of GDP) % of total expenditure Share of total expenditure, main functions 14.0 12.0 100% 2221 20 10.0 8.0 6.0 50% 9 9 6 6 5 7 8 4.0 5 4 5 3 5 3 2.0 2 1 0.0 0% 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022p 2016-19 2020-22 Wages and Compensation Goods and Services Wages and Compensation Goods and Services Interest Payments Current Transfers 2010-15 2016-19 2020 Interest Payments Current Transfers Pensions Capital Expenditures Pensions Source: Authors’ calculations based on MFMod. 13 Part I Macroeconomic dynamics: Drivers of fiscal space Moderate budget rigidities underline the importance of greater efficiency. Budget components vary within a spectrum of flexibility, since not all can be easily modified by the authorities in the short term. Civil servant salaries and public benefit transfers (which includes pensions) fall under the high rigidity category, since these cannot be easily reduced. All capital expenditures and spending on goods & services (excluding professional services) are categorized as low rigidity. Public expenditure in Benin is currently moderately rigid, as it has been for the past decade, limiting the scope for fiscal adjustment across sectors. About 35-40% of its total expenditure can be classified as highly rigid, for a total of about 80% of spending with some level of medium-high rigidity. Over the last five years (2016-20), the budget share of non-discretionary items has also risen. Budget rigidity is comparable to other peer countries, however. About 35-40 percent of spending is highly rigid Average rigidities are similar to structural peers Overall Share of Expenditure by Rigidity Average expenditure, by rigidity level 78.7 82.2 76.4 31% 31% 34% 34% 34% 35% 36% 36% 36% 37% 40% 40% 37% 45% 41% 41% 39% 42% 41% 43% 48% 41% 21.3 23.6 17.8 29% 29% 26% 25% 25% 25% 23% 23% 21% 18% 16% 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Benin (2016-2020) Togo (2016-2019) Senegal (2016-2018) Low Rigidity Medium Rigidity High Rigidity Rigid Non-rigid 14 Part I Macroeconomic dynamics: Drivers of fiscal space Recent reforms to improve PIM processes and PFM are needed to improve efficiency of public spending The management of public investments has made progress in recent years. The Public Investment Management Assessment, PIMA 2020 underlined specific weaknesses in the operationalization despite a high quality and comprehensive legal and regulatory framework in place. Since 2019, the government has reinforced its capacity in various dimensions, including increasing transparency in the public procurement framework, the reinforcement of planning and monitoring capacities at the national and communal levels, and the reinforcement of statistical capacity. Similarly, procurement reforms have supported capex execution and need to be deepened. An E-Procurement initiative is being rolled out in all ministries by the end of 2024. PFM processes have also improved. For example, budgetary discipline has become better enforced due to the creation of a comprehensive framework for medium-term expenditure, in close articulation with the priority areas defined in the PAG. Program budgeting was introduced in 2022 and it should increase flexibility in the use of funds by line ministries as it is progressively rolled out. Reforms needs to be sustained. PIM was weaker than average WAEMU and SSA countries Capital expenditures are significantly under-executed in 2018, and it has improved slowly Deviation (Actual expenditure vs. Planned), by 0% category -5% -4% -7% Deviation (in % ) -10% -12% -12% -10% -10% -15% -15% -29% -38% -36% -46% -46% -44% -43% -50% -56% -61% -74% Current Capital Total 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Source: PIMA 2020 15 Part I Macroeconomic dynamics: Drivers of fiscal space 04 Achieving Benin’s human capital objectives requires reducing losses due to inefficient spending in social sectors Improving the quality of education, health and the strength of the social protection systems are a national objective to build a well-equipped labor force and achieve the human capital potential. The National Development Plan (NDP) 2018-25 and Government Action Plan (PAG II) 2021-2026 focus on the quality of education as a national priority for building a well-equipped labor force and on achieving access to quality and resilient healthcare at all phases of life, accessible to everyone by 2030. Similarly, as part of the PAG, a new social protection program is being developed – the Insurance for the Reinforcement of Human Capital, ARCH (Assurance pour le Renforcement du Capital Humain), with its flagship health insurance window (Assurancce Maladie-ARCH) to ensure effective health coverage to the poorest segments of the population. The challenge of overcoming a legacy of limited investment in human capital and social resilience systems is nonetheless large (World Bank 2022). Total social spending (% of GDP) With limited room to increase overall social 20 spending and given the imperative of improving 18 human capital outcomes, reducing inefficiencies 16 Social protection Health Education is a first order necessity in the short to medium 14 term. In absolute terms social spending is low, 12 constrained by the limited resources: it 10 averaged 5% of GDP compared to 8.2 percent in 8 LICs and it was lower on average across all 6 social spending categories. It has also declined. 4 Still, social spending is significant in relative 2 terms amounting to a third of total public 0 expenditure, largely driven by education (about 2011-15 2001-05 2006-10 2011-15 2016-19 2001-05 2006-10 2016-19 2001-05 2006-10 2011-15 2016-19 25% of total on average over the period), Additional fiscal space to spend in social sectors will first have to come through reducing EMDs LICs Benin technical and allocative inefficiencies and improving the distributional dimension of public Benin’s HCI is among the lowest in the world… Low spending in social sector highlights the importance of efficiency gains to spending in these key sectors. Source: World Bank 2022 move the needle. Source: IMF 2022 16 Part II. Spending better to achieve the human capital potential 16 Learning outcomes could increase by 41% in mathematics and 33% in language by increasing reducing inefficient spending. Estimates suggests that a more efficient use of resources in education could increase learning outcomes by at least a third. Estimates suggests that Benin could improve its learning outcomes by 41 percent in mathematics and 33 percent in language by increasing the efficiency of its education spending. In turn, this could significantly improve outcomes later in life given the existence of high returns to education in Benin. At the national level, an additional year of schooling yields a 10 percent return. Average estimated earnings increase threefold for the working age with post-secondary education compared to those with no formal education and the likelihood of underemployment is cut by half. Share of grade 6 students proficient in mathematics (left-hand) and language (right-hand) vs education expenditure 80.0% 100.0% Share of pipuls proficient in Maths 70.0% GABON 90.0% Share of pipuls proficient in GABON SENEGAL 60.0% BURUNDI 80.0% BURKINA FASO BENIN SENEGAL 70.0% BURKINA FASO 50.0% BENIN 60.0% CONGO language 40.0% TOGO 50.0% GUINEA GUINEA CAMEROON CAMEROON CONGO 40.0% TOGO 30.0% COTE D'IVOIRE 30.0% NIGER MADAGASCAR NIGER RDC BURUNDI 20.0% 20.0% CHAD RDC MADAGASCAR 10.0% CHAD COTE D'IVOIRE 10.0% 0.0% 0.0% 0.0% 1.0% 2.0% 3.0% 4.0% 5.0% 6.0% 7.0% 0.0% 1.0% 2.0% 3.0% 4.0% 5.0% 6.0% 7.0% Education expenditure as % GDP Education expenditure as % GDP 17 Part II. Spending better to achieve the human capital potential Greater efficiency in public health expenditure could increase average life expectancy by five years. Using a stochastic frontier model to measure the distance between Benin’s outcomes and the estimated frontier (best) outcome, suggests that life expectancy could be 3-5 years higher with more efficient spending. In terms of essential health service coverage, 8 percent more of the population could be covered with more efficient public health expenditure. Population coverage by essential health services is strongly correlated with higher level outcomes such as life expectancy, or child and maternal mortality. Realizing these gains would be equivalent to bringing Benin’s essential health service coverage to the level of Togo or Ghana. Overall inefficiencies are lower than in Senegal and Ghana (but these countries have much higher coverage at baseline), but much higher than in Togo, Rwanda, Morocco and Sri Lanka. Life expectancy could be 3-5 years higher if spending was more efficient…(world, left; SSA, right) 18 Part II. Spending better to achieve the human capital potential 05 Increasing technical and allocative expenditure efficiency in education, health and social protection can boost outcomes in human capital The quality of service delivery in education is poor with high geographical disparities, mostly due to inefficient human resource management Student-teacher ratios (STR) in primary public schools remain relatively high and there is variation across regions, reflecting technical inefficiency in the provision and deployment of teachers despite a recent reform to prioritize underserved regions - mostly resulting from the inadequate incentives to stay in hard-to-reach areas. Poor human resource management contrast with the large weight of total sectoral spending on wages and salaries for both education Spending on personnel constitutes the largest share of current spending in education, with most funds used to finance teachers’ salaries. Over 2017–19, the share of spending dedicated to wages and salaries was high across all levels of education accounting on average for 85 percent of current expenditures at the secondary level, 78 percent for both pre-primary and primary levels and 53 percent for higher education But HR management is poor, as evidenced by high The bulk of spending in education is on wages and salaries (2017-19) … Student-Teacher ratios in primary education are high 100% 60 8% 5% 8% 90% 13% 9% 50 16% 5% 19% 80% 5% 40 70% 38% 4% 60% 30 50% 3% 20 40% 81% 10 71% 69% 30% 20% 46% - Benin WAEMU - SSA LMIC Senegal Togo Rwanda Sri Lanka Ghana Tunisia 10% Excluding 0% BEN MEMP (pre-primary & MESTFP (secondary) MESRS (tertiary) All levels primary) Structural peers Aspirational peers Wages and salaries Good and services Transfers and subsidies Capital Source: Estimations based on BOOST data Source: UIS consulted in 2022 19 Part II. Spending better to achieve the human capital potential The inefficient allocation of human resources leads to significant economic losses in the health sector, estimated at 1% of 2019 GDP The number of medical personnel is insufficient, below that of structural peers and their geographical distribution is inefficient. For example, while Benin had 0.65 doctors per 10,000 inhabitants in 2018, Senegal, Rwanda and Togo had 0.88, 1.18, 0.83 respectively the same year. Geographical disparities are also stark. Inefficiency in the management of human resources generates significant losses. Estimates suggest that the unequal geographic distribution and low productivity of health professionals' amount to a loss of CFAF 82.6 billion to the health sector annually (i.e., 7.3 percent of the national budget 2019 or 1 percent of 2019 GDP). Although to a lesser extent, public expenditure on the health sector is also skewed towards human resources: they averaged 40 percent of the sector’s budget over 2017-20; suggesting efficiency gains can go a long way in improving health outcomes. On average wages and salaries represented the largest share of But human resources are unequally distributed (2021) health expenditure over 2016-21. 100% 25 20.0 90% Human Health Resources per 10,000 80% 20 70% 60% 15 50% 10 Inhabitants 40% 4.3 4.9 6.1 4.2 3.9 3.2 3.2 4.0 30% 2.9 3.9 5 2.3 20% 44% 41% 39% 37% 36% 31% 31% 10% 17% 0 0% Ouémé Atlantique Mono Plateau Atacora Collines Couffo Donga Zou Alibori Borgou Littoral 2016 2017 2018 2019 2020 2021 2016-2019 and 2016-2021 Average Average 2021 Publicly funded Funded with facilities' own resources WHO norm (23) Salaries and wages Goods and Services Credits, transfers, and grants Capital Source: Estimations based on BOOST data Source: Authors, using data from the 2020 Health Statistical Yearbook 20 Part II. Spending better to achieve the human capital potential Reducing technical inefficiencies in some sub-sectors can free up fiscal space for other areas There could be efficiency gains of up to 10 % of total ressources by bringing the regions towards the levels of the most efficient. A data envelopment analysis (DEA) was used to assess resource use based on efficiency scores. The overall efficiency level of Benin’s public primary education system is 90.6 percent; implying that 9.4 percent of resources are being wasted. Geographically, about two- thirds of Benin’s school districts are relatively inefficient in their use of resources. Map of efficiency scores by school districts Based on the latest household survey (2018), the losses due to repetition and dropouts are estimated at about CFAF 25 million at the primary level, or 25% of public spending at this level of education. The losses are higher in lower and upper secondary, at 29 and 36 percent of public spending respectively Year repetition and level-specific dropouts are costing the public sector 0.68% of GDP annually. 21 Part II. Spending better to achieve the human capital potential 06 Allocative inefficiencies in terms of sub-sector prioritization exist across social sectors Both health and education sectors display allocative inefficiencies in terms of sub-sector prioritization The share of the education budget allocated to secondary TVET (3.1 percent) and pre-primary (3.6 percent) is very low: below the averages for SSA and LICs (5 percent) and comparators like Rwanda (14 percent) for the former; and below the UNICEF-recommended levels of 10 percent for the latter. Allocating such a small share of the education budget to TVET is inefficient given the high unmet demand for skills training, the large numbers of out-of-school youth, and low levels of youth literacy Evidence has also shown that spending on pre-primary education is cost- effective as it increases primary school intake, strengthens efficiency (reduces repetition), and improves learning and equity Tertiary care represents more than half of health expenditure despite the need to reinforce primary care level. In 2016 and 2017, only 19 percent and 20 percent of the MoH’s budget was allocated to primary care However, low spending on primary care is an issue in a country with significant disparities in access to health and a high burden of disease. Expanding primary care tends to bring marked improvements. Most spending is directed to primary education Tertiary care receives the largest share, and primary care is underfunded… 100% 22% 22% 23% 21% 3.1% 3.0% 3.3% 80% 3.0% 27% 26% 26% 29% 60% 40% 44% 46% 44% 44% 54% 20% 44% 39% 32% 35% 19% 20% 3.6% 3.7% 3.6% 3.6% 0% 2017-19 (Avg) 2017 2018 2019 2016 2017 2018 2019 2020 2021 Average (2016-2021) Pre-primary Primary Secondary Gen. Secondary TVET Tertiary Primary Secondary Tertiary Source: Estimations based on BOOST and HR data from the Ministry of education 22 Part II. Spending better to achieve the human capital potential Most social protection spending is geared towards the public pension fund so reducing its operational deficit could free up resources Spending on social assistance is extremely low, in contrast, despite the development of adequate tools On average, 84% of total social protection spending is dedicated to financing the public pension fund deficit (Fonds national de retraites du Benin, FNRB). The fund only covers civil servants, and thus formal sector employees (although a program is in place to start covering informal employees). Reducing costs to formal social insurance can free up fiscal space that can be used towards more vulnerable sectors. Informal workers have an alternative pension system, but it is still incipient. Spending on social assistance is extremely low, in contrast, despite the development of a social protection framework since 2016. Spending on social assistance is extremely low and much lower than all other comparator countries, structural and aspirational peers combined. For example, in 2020 it was 72 times lower than Rwanda’s, at 0.03% of GDP. Low spending contrasts with the recent investments to improve the system. Social assistance spending is low compared to other Social insurance represents the largest share of total spending, countries with the same level of GDP per capita… although it has declined markedly. 35000 8% 7% Percentage of total expenses 30000 6% 25000 5% GDP per capita 20000 4% Sri Lanka 3% 15000 Moroc… 2% 10000 Ghana 1% 5000 Senegal 0% Togo 2018 2010 2011 2012 2013 2014 2015 2016 2017 2019 2020 0 0 1 2 3 4 5 6 7 Benin Social assistance expenditure as a percentage of GDP Social assistance Social Insurance Labour Market Programs 23 Part II. Spending better to achieve the human capital potential Contrasting with the size of personnel spending, capital and operational expenditures have been too low to keep up with demand Low capital spending in education has resulted in a shortage of classrooms and schools, especially at the secondary level. Capital expenditures in education amounted to only 12 percent of government expenditures over the period 2010-16 and decreased to 8 percent over 2017-19. They are particularly pervasive at the secondary level – where outcomes are significantly below peers – as they have left little space to absorb the increasing number of students who complete primary education. Non-salary recurrent spending is not directed towards learning material. Capital spending in health is insufficient but has increased due to the COVID-19 pandemic. The share of capital spending in total health spending decreased from about 45 percent in the early 2010s to 28 percent in 2017-19 Temporarily, due to the COVID-19 response, they increased in 2020, reaching 64 percent, as authorities focused on expanding capacity (building 12 laboratories, increasing the number of hospitals). . The population covered by essential health services has increased substantially between 2010-20 but remains low both in absolute and relative terms. Further, health facilities are not always able to provide essential services. The average operational capacity of Benin’s health facilities is estimated at 64 percent (Republic of Benin 2018). Limited infrastructure constrains the transition from primary to secondary Access to health facilities is well below the WHO standard (2020) 175% 171% 2.5 Health Facilities per 10,000 2 140% 120% 112% 1.5 102% Inhabitants 105% 1 77% 81% 74% 105% 70% 0.5 48% 87% 68% 0 35% 60% 55% 53% Alibori Atacora Atlantique Borgou Collines Couffo Donga Littoral Mono Ouémé Plateau Zou National 39% 35% 0% Total Boys Girls Poorest Richest Littoral Collines Zou Grade 6 Grade 7 National Hospitals Other Health Facilities WHO Norm Source: Estimations based on EHCVM (2018) 24 Part II. Spending better to achieve the human capital potential 07 A substantial share of social spending is regressive and inequitable, reducing the cost-effectiveness of public expenditure to reduce poverty and inequalities Out of pocket expenditures is high both in health and education reinforces disparities in access to health and education. According to the latest household survey (2018), household out-of-pocket (OOP) expenditure on education averaged 40 percent of total spending in the sector, and they are higher at pre-primary and primary levels. For health, OOP payments represent close to 50 percent of health expenditures, while public resources contribute less than one third. Benin’s share of health OOP is close to aspirational peers (Morocco, Sri Lanka) – countries with much higher income per capita. The high level of OOP reinforces disparities in access to health and education which are already high by income and other demographic characteristics. Out of pocket payments are high in the health sector Out of pocket payments on education are significant 100% 21% 20% 23% 15% 80% 31% 25% 100% 47% 40% 40% 40% 60% 80% 49.1% 49.4% 70.1% 65.5% 40% 60% 66% 43% 44% 45% 47% 51% 47% 46% 40% 20% 36% 50.9% 50.6% 12% 20% 34.5% 0% 29.9% 2016 2017 2018 2019 Senegal Rwanda Togo Ghana Morocco Sri Lanka 0% Benin Structural peers, 2019 Aspirational peers, 2019 Preschool Primary Secondary Post secondary Out-of-pocket payments External health expenditure Domestic general government health expenditure Volontary insurance and others Public Private Sources: Authors, using data from WHO's Global Health Expenditure Database Source: Estimations based on EHCVM 2018 25 Part II. Spending better to achieve the human capital potential The level of inequity in accessing health and education services is high, making significant shares of public spending in these sectors inequitable Public ressources in education are biased towards richer households except at primary level. The poorest quintile households allocate 8 percent of their total consumption spending to primary education. This share increases with each level of education, reaching 24 percent at upper secondary and 78 percent at tertiary level. Poorer households face greater difficulties in sustaining their investment in education, which undermines efforts to break the poverty cycle. Access to health is marred with significant disparities based on income and location, as a result subsidies to health facilities are regressive. Nationwide, 23 percent (32 percent in Littoral) of medical personnel are financed by the health facilities’ own resources, which increases disparities between richer and poorer areas. The lack of universal health coverage also reduces affordability for the lower deciles and existing free packages appear to be regressive Public spending on education tends to be poverty neutral at the primary The poorest (Q1) account for 17 percent of the need (hospitalization) level but regressive at both the secondary and tertiary levels but enjoy only 5 percent of benefits. 70% 100% 58% 25% 60% 80% 41% 47% Share of benefits/needs 50% 21% 60% 40% 29% 20% 27% 27% 26% 20% 25% 30% 23% 23% 40% 22% 21% 21% 20% 20% 20% 19% 18% 18% 17% 16% 16% 16% 15% 15% 14% 20% 18% 12% 12% 10% 20% 7% 10% 17% 4% 17% 2% 0% 5% 0% Pre-primary Primary Second Gen Second TVET Tertiary All levels Share of benefits Share of needs Q1 Q2 Q3 Q4 Q5 Poorest Poorer Middle Richer Richest Source: Estimation based on EHCVM 2018, EMIS 2017-19 and BOOST 2017-19 26 Part II. Spending better to achieve the human capital potential Subsidies and universal programs are regressive Several transfers and social assistance programs managed by other sectors (water, electricity, education) are regressive. A cost-effective use of social transfers would suppose the removal of more regressive subsidies… Evidence suggests that subsidies to electricity, water and tertiary education are 47.7 44.6 44.4 33.9 29.6 27.1 regressive (CEQ Analysis 2022). 22.0 19.4 19.0 8.1 5.0 In electricity for example, subsidies amount to 0.3 percent of GDP on average over - 2.0 - 4.6 - 8.6 - 19.0 - 21.4 2018-19 (although they have increased in - 24.0 - 35.2 2020-21 as part of the response to COVID- Payroll Water PIT Customs Health user fees School canteen Fuel taxes VAT Electricity Fuel subsidy Secondary educ. SPT Education user fees Primary healthc. Excise: genl. Tertiary educ. Primary educ. Hospital healthc. 19). Yet, they are regressive given the electricity consumption patterns in Benin. Source: authors’ estimates based on the EHCVM 2018/19, CEQ Analysis 2022. Notes: [1] The figure shows the Kakwani index for each main transfer and tax category. The Index is a summary statistic of progressivity. It is calculated for taxes by subtracting the Gini coefficient from the concentration coefficient, and for transfers by subtracting the concentration coefficient from the Gini coefficient. A value greater than 0 represents progressivity, and it’ s minimum and maximum values depend on the size of the Gini coefficient. [5] The Kakwani Index for a category of taxes or transfers cannot be calculated by aggregating the Kakwani Indices of its sub-components. 27 Part II. Spending better to achieve the human capital potential Targeted social transfers can increase equity in social spending while at the same improving the cost-effectiveness of public resources. The Unique Social Registry offers an opportunity to improve the targeting of social assistance, reduce inefficiencies and improve redistribution. To finance targeted cost-effective social transfers… (Gbessoke) a. Gbessoke Cash Transfer Target social transfers are an important instrument to increase equity while at the same improving 0 Incidence of pensions (%) 8 cost-effectiveness of scarce public resources. Incidence of pre-fiscal income Incidence of VAT (%) Given the vulnerability of the Beninese population, it is 6 -2 crucial to implement a social safety net with significant coverage, which can also provide the flexibility and 4 framework to prepare for and cope with shocks. -4 Simulations using the CEQ methodology suggest the 2 implementation of the Gbessoke program would improve -6 the redistributive nature of the Beninese fiscal system. 0 Currently, the limited degree of social transfer targeting is 1 2 3 4 5 6 7 8 9 10 one of the main limitations of the overall system. Gbessoke School canteen Indirect subsidies Incidence of pre-fiscal income Source: authors’ estimates based on the EHCVM 2018/19, CEQ Analysis 2022. c. PIT reform 28 Part II. Spending better to achieve the human capital potential 0 ) ) 08 Selected policy options Table I.1. Top 5 policy options on revenue mobilization and public spending systems (Part I) Reform area Time Area 1. Indirect taxation: Enhance policy coverage, remove exemptions and review tax rate mix 1. Significantly improve the productivity of VAT revenue generation (target 0.5% to GDP). Options include: (i) Review and rationalize the more than 100 VAT exemptions carefully considering equity and efficiency; (ii) eventually discontinue preferential rates and exemptions in the financial sector; (iii) better capture e- MT commerce in the VAT chain, also with a view to enhance the size of the formalized economy; (vi) reinforce compliance dialogue with VAT taxpayers following the positive impact of the risk-based VAT credit refund procedure. Area 2. Direct taxation: Expand tax bases, review the tax rate mix, launch a review of benefit-costs of certain tax expenditures 2. Seek higher inclusion of informal workers in the income tax bases. Reform options include create linkages between business permits and issuance of TIN to MT informal business; consider adjusting the rate mix of the personal business income (IBA); develop a compliance and audit strategy targeting top wage earners. Area 3. Deepening tax and customs administrative improvements 3. Explore and potentially implement the strict application of transactional values at customs MT Area 4. Enhanced quality, efficiency, and transparency of public investments. 4. Reinforce investment project preparation and appraisal. Publish all PIM related documentation, including the criteria for the appraisal and selection of major ST investment projects, along with feasibility studies reports, in the in the draft budget law documentation. Area 5. Improved procurement to enhance effectiveness and bidding competition 5. Finalize the rollout of the e-procurement framework. ST 29 08 Selected policy options (continued) Table II.1. Top 15 policy options on human capital expenditure Potential fiscal Reform area Time gains (+) / uses (-) Area 6. Address technical inefficiencies in education, health and social protection 1. Education: Reduce repetition and dropout rates for efficiency gains of up to 0.68 percent of GDP by establishing an automatic policy in specific grades of primary, ST (+) providing additional remedial programs, and make the school environment safe. 2. Education: Improve teacher management by working on three fronts: training, deployment, and attracting STEM teachers MT (+/-) 3. Education: Strengthen the education information management system to make it more functional and improve the Human Resource Information System to allow for a better management of education HR. This requires to develop a reliable EMIS and assessment system, and create a clear and comprehensive database for ST (-) education HR 4. Health: Introduce workforce plans and incentives (including specific incentives for women) to send health personnel to priority, remote areas; or recruit them MT (+) locally and Provide greater on-the-job training, mentorship and reward mechanisms and tighten supervision of clinical staff. 5. Social insurance: Carry out a parametric reform to the FRNB (e.g., reviewing the calculation basis of rights, minimum duration of contribution, generosity of rights) MT (+) Area 7. Improve allocative efficiency in education, health and SP 6. Health: Increase the share of public health expenditure allocated to primary and secondary care services, by introducing a funding mechanism linked to the spatial MT (+) demand for packages of essential services and community health workers; train personnel at the local level; and leverage digital technologies for telemedicine. 7. Education: Improve the allocative efficiency of the education budget to ensure that pre-primary and secondary TVET are adequately funded. MT Neutral 8. Allocate a budget line to social safety nets to support the most vulnerable households in a sustainable manner, financing it from more regressive subsidies. MT (-) Area 8. Improve equity, redistribution and cost-effectiveness of social programs 9. Adopt the USR as a common targeting mechanism to achieve greater efficiency and transparency in spending, minimizing targeting costs and allowing for MT (+) integrated management of different programs. The USR will need to be strengthened through updates, expansion, and further development of its procedures. 10. Implement an adaptive social safety net with substantial coverage to support the most vulnerable populations by limiting the impact during shocks. ST (-) 11. Education: Review the current scholarship and social assistance program offered at the tertiary level to make it more equitable and targeted. Benin should MT (+/-) consider introducing means-tested scholarships, while building better linkages between safety-net programs and school attendance. 12. Education: Prioritize vulnerable youth in the expansion of secondary TVET. Enhance the flexibility of TVET to ensure out-of-school youth can access TVET. MT (+/-) 13. Health: Increase the pro-poor dimension of health programs in terms of health services coverage and consider targeting to reduce their regressivity ST (+) 14. Health: Strengthen the operational aspects of AM-ARCH, including targeting, population outreach, and training of health professionals in health services. ST (+/-) 15. Improve the targeting of the electricity subsidy with a voucher mechanism MT (+) 30 HEADQUARTERS THE WORLD BANK 1818 H Street, NW Washington, DC 20433 USA Tel : (202) 473-1000