## 1sleea2022002

## Source details

**Canonical URL:** [1sleea2022002](https://www.imf.org/-/media/files/publications/cr/2022/english/1sleea2022002.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2022/english/1sleea2022002.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2022/english/1sleea2022002.pdf.json)

---

### Climate Change and Mining Activities — Nexus, Commitments, Financing, and Mining Risks
- Climate vulnerability and development
  - Sierra Leone is "highly vulnerable to the adverse impacts of climate change despite low level of exposure to climate-related hazards."
  - Physical exposure lower than world average; low human development limits coping capacity.
  - Notable events and impacts:
    - August 14, 2017 flooding and landslide in Freetown "took over a thousand lives."
    - World Bank damage and loss assessment: "a total economic loss of over US$31 million and recovery needs of over US$82 million (or 0.8 percent of GDP in 2016)."
    - Loss of "2,850 ha of protected forest in the watershed area" associated with a "40 percent drop in the water level of the Guma dam" between 2015 and 2021 (WFP, 2022).
  - Socioeconomic amplifiers:
    - Agriculture employs "54 percent of total population" and, with fishing and forestry, "constitutes 59 percent of GDP in 2020."
    - "About 56.8 percent of population in 2018 were living below US$1.90 per day (World Bank, 2020)."
    - "4.6 percent of population in severely food insecurity"; "Almost half of the population are not consuming sufficiently nutritious diet."
  - Environmental degradation:
    - Sierra Leone "lost 1.82Mha of relative tree cover (equivalent to a 32 percent decrease since 2000)" (Global Forest Watch).
  - Fiscal constraints:
    - "Overall fiscal deficit excluding grants is estimated at 9.3 percent of GDP in 2021 and is projected to remain above 6 percent of GDP until 2024."
  - Policy emphasis:
    - Mainstream adaptation into national development; prioritize projects with development and adaptation co-benefits; strengthen PFM to plan and govern climate spending.

- Climate commitments and institutions
  - NDCs and adaptation planning:
    - First NDC approved November 2016; updated version in 2021.
    - Mitigation targets: "reducing CO2 emissions by 5 percent before 2025, 10 percent before 2030, and 25 percent before 2050."
    - Adaptation target: "reducing vulnerability by half by 2030."
    - Initial National Adaptation Plan (iNAP) published 2021; follows UNFCCC-aligned blocks and identifies five categories of adaptation programs.
    - iNAP and NAPA prioritized "24 adaptation projects" but "Selection criteria to prioritize these programs and projects were missing in the iNAP and NAPA."
  - Regulatory institutions and implementation challenges:
    - Sierra Leone Environmental Protection Agency (EPA) established 2008.
    - National Disaster Management Agency (NDMA) set under Office of the President in 2020.
    - National Protected Area Authority (NPAA) formed in 2012.
    - Weak oversight capacity limits enforcement and conservation.

- Box 1 — Climate Change and Mining Activities (selected findings)
  - Climate risk to mining and commodity demand shifts:
    - Climate change expected to cause more frequent floods and droughts disrupting water supply and mining operations.
    - McKinsey estimate: 50 percent of global iron ore production exposed to high risks of flood occurrence.
    - Heavy rainfall in southeastern Brazil put 100 million tons of annualized iron ore supply at risk (~7 percent of seaborne supply; ~30 percent of Brazilian supply).
    - Green recovery and recycling pressure iron-ore and bauxite markets; low-grade ore (e.g., Tonkolili) in Sierra Leone vulnerable; high-grade ore (e.g., Marampa) favored.
  - Regional and national financing needs:
    - Sub-Saharan Africa total US$1,318 billion (79 percent of GDP in 2020) for NDCs.
      - Mitigation: US$571.6 billion (34 percent of regional GDP in 2020).
      - Adaptation: US$407.2 billion (25 percent of regional GDP in 2020).
      - Unspecified: US$339 billion.
    - Sierra Leone’s updated NDC: US$2.8 billion (66 percent of GDP in 2020) for 2021–2030 (up from US$900 million).
      - Mitigation US$1.7 billion; Adaptation US$1.1 billion.
      - Implied annual fiscal cost: about US$276 million at 6.6 percent of GDP or 43 percent of fiscal revenue in 2021.
    - Government plan: allocate 10 percent of annual budget to combat climate change and secure 40 percent of donor funding for adaptation priorities; no explicit financing plan provided.
  - Financing scenarios (IMF staff calculations, Table 1 — US$ billion)
    - Scenario assumptions:
      - Scenario 1: government allocates 10 percent of revenue and 40 percent of grants; remaining financed by budget deficit.
      - Scenario 2: government allocates 2.5 percent of revenue and 20 percent of grants; remaining financed by budget deficit.
      - Scenario 3: government allocates 2.5 percent of revenue and 20 percent of grants for adaptation only; remaining financed by budget deficit.
    - Table 1 estimates:
      - Total financing needs: Scenario 1 = 2.76; Scenario 2 = 2.76; Scenario 3 = 1.08.
      - Resource mobilized: Scenario 1 = 1.43; Scenario 2 = 0.53; Scenario 3 = 0.53.
        - -revenue: Scenario 1 = 0.74; Scenario 2 = 0.19; Scenario 3 = 0.19.
        - -grants: Scenario 1 = 0.69; Scenario 2 = 0.34; Scenario 3 = 0.34.
      - Annual deficit: Scenario 1 = 1.33; Scenario 2 = 2.23; Scenario 3 = 0.55.
      - percent of GDP annually: Scenario 1 = 3 percent; Scenario 2 = 4.9 percent; Scenario 3 = 1.2 percent.
    - Scenario interpretations:
      - Scenario 1: raise US$1.4 billion (US$0.7 and US$0.7 billion from revenue and grants) between 2021 and 2030; budget deficit average 3 percent of GDP; public debt becomes unsustainable in medium term.
      - Scenario 2: raise US$530 million (US$190 and US$340 million from revenue and grants) between 2021 and 2030; budget deficit average 4.9 percent of GDP; jeopardizes debt sustainability.
      - Scenario 3: mobilization same as Scenario 2 but limited to adaptation (US$1.076 billion); total financing needs decrease by 60 percent and annual deficit narrows to 1.2 percent of GDP; public debt remains unsustainable.
  - Fiscal space and mobilization required:
    - SIP simulations: additional 0.6 percent of GDP can be spent on climate adaptation without derailing fiscal consolidation and still hit medium-term debt anchor after 12 years.
    - Required additional mobilization:
      - 5.4 percent of GDP from revenue and grants to meet mitigation and adaptation commitments.
      - 1.8 percent of GDP to meet adaptation commitments.
    - Calibrated debt dynamics (non-iron ore GDP basis):
      - Calibrated overall balance ranges from -6.9 to -0.7 percent of non-iron ore GDP in 2025 before long-term surplus +0.8 percent.
      - Debt to GDP increases from about 76.2 percent of non-iron ore GDP in 2021 to about 78.5 percent in 2023 before converging to about 51 percent by 2033.
      - To meet adaptation (Scenario 3), additional revenue mobilization of 0.6 percent of non-iron ore GDP (or additional debt) required.
  - Policy recommendations and priorities:
    - Calibrate climate costing and prioritize projects; publish climate-related investment projects with clear appraisal criteria.
    - Develop realistic financing plan based on medium-term revenue and expenditure strategies; emphasize external climate grants.
    - Prioritize adaptation measures (prevention, alleviation, resilient finance) and use prioritization criteria: positive externalities, remove private adaptation barriers, address equity.
    - Enhance capacity to access external grants; strengthen national systems and governance.
    - Explore carbon credits and contingent finance; strengthen enforcement and project pipelines.
    - Adopt Green PFM reforms across budget preparation, execution, accounting, control and audit; preconditions: basic functional PFM, strong political support, some green expertise.

*Prepared by Saanya Jain, Rosalind Mowatt, Peter Wankuru, and Yue Zhou.*

### Reaping the Demographic Dividend — The Role of Education (Findings, Gaps, and Policy)
- Major demographic context and projections
  - Sierra Leone population is young: in 2020, "more than 40 percent were younger than 15, and more than half were younger than 20."
  - Working-age population projected from 4.5 million in 2020 to 8.5 million by 2050 (UN medium-variant).
  - Medium-variant assumption: fertility converges to "2.1 children per woman."
  - Africa projected to contribute "80 percent" of global population increase by 2100.
  - Demographic dividend not automatic; requires education, health, training, and jobs.

- Baseline education outcomes and gaps (2018 and trends)
  - 2020: Free Quality School Education (FQSE) part of NDP 2019-2023; government assumes costs for tuition, textbooks, furniture, infrastructure, teacher training and recruitment.
  - Public spending on education: "21.4 percent" in 2022 (meets Dakar target).
  - Benefit incidence analysis (BIA) 2018:
    - Primary: spending share to poorest quintile increased between 2003 and 2018; share to richest quintile decreased.
    - Secondary: distribution remains regressive.
  - Completion and enrolment:
    - From 2016 to 2019, primary completion rates increased from 62.8 to 69.4 percent.
    - Lower completion increased from 40.9 to 47.0 percent; upper completion from 14 to 19.6 percent.
    - In 2020 these gains reversed due to COVID-19.
  - Dropout reasons (2018 SLIHS):
    - Lack of finance: 41 percent of bottom income quintile and 22 percent of top quintile cited expensive education/lack of finance as primary reason for dropping out.
    - Other barriers: lack of interest, marriage/pregnancy, distance significant for poorest.
  - Resource gaps and partner support:
    - World Bank estimates significant financing gap to fully deliver FQSE in next four years.
    - Development partner contributions preserved exactly:
      - US$50 million (World Bank) and US$15.9 million (Irish Aid, European Union and FCDO) in October 2020.
      - US$17.2 million (GPE) grant in 2019.
      - US$38 million (GPE eligibility announced in 2021): US$22.47 (system transformation grant), US$15 million (multiplier grant), US$2 million (capacity grant).

- Policy priorities and recommendations
  - Increase and optimize public spending; higher weight for infrastructure and non-wage items.
  - Improve teacher quantity and quality; recruit more teachers and strengthen teacher management.
  - Strengthen sector management and governance; integrated, well-used EMIS.
  - Target interventions for poorest: address financing barriers, distance, reintegration of pregnant girls/teenage mothers.
  - Expand health and primary health care investments to support human capital.
  - Link school subsidies to performance; finalize new Education Sector Plan; consider constitutional affirmation of free, quality, compulsory primary and secondary education.
  - Use digital tools and satellite imagery for planning, monitoring, and resilience.
  - Selected measures (Box 1):
    - Accelerated learning for out-of-school girls; sanitary hygiene kits; safety policy and comprehensive sexuality education; digital learning hubs; Youth Advisory Group (YAG) for Education.

### COVID-19 Shock, Model Framework, and Policy Scenarios (Macroeconomic and Distributional Analysis)
- Model framework and calibration
  - Dynamic multi-sector heterogeneous agent general equilibrium model calibrated to Sierra Leone; baseline year 2019.
  - Households: urban (private/public workers, entrepreneurs) and rural (small/large farmers); private workers high- or low-skilled.
  - Sectors: agriculture (domestic consumption and exports), manufacturing, services, mining.
  - Data sources: 2018 Integrated Household Survey; 2014/2015 Labor Force Survey; 2013/2019 DHS; national accounts; fiscal statements.
  - Key calibration targets (selected, preserved exactly):
    - Agriculture share of GDP: Data 0.53, Model 0.54
    - Agriculture sector share in GDP (domestic): Data 0.40, Model 0.46
    - Agriculture export sector share in GDP: Data 0.14, Model 0.08
    - Manufacturing sector share in GDP: Data 0.02, Model 0.06
    - Service sector share in GDP: Data 0.41, Model 0.34
    - Mining sector share in GDP: Data 0.05, Model 0.06
    - Tax revenues to GDP 2/: Data 0.12, Model 0.28
    - Gini national: Data 0.37, Model 0.39
    - External grants to government to GDP: Data 0.03, Model 0.03
    - Trade balance to GDP: Data -0.26, Model -0.33

- COVID-19 shock channels and calibrated shocks
  - Human capital shock:
    - World Bank estimates: 1.32 percentage point increase in dropouts and 20 percentage points in students repeating grades.
    - Food insecurity: 6-percentage point increase between end-2019 and end-2020.
    - Productivity drops: low skilled -1.8 percent; high skilled -0.06 percent.
    - Urban private unskilled households increase by 1.32 percent to reflect dropouts.
    - Result: long-term GDP change -1.1 percent; services weight declines; manufacturing +0.5 percentage points; agriculture declines.
  - TFP shock:
    - TFP assumptions: small farmers -1 percent; large farmers -1 percent; manufacturing -2 percent; services -3 percent.
    - Result: long-term GDP change -2 percent; manufacturing share declines by -0.5 percentage points; income losses larger for higher-earning groups; modest decline in inequality.
  - Combined shocks:
    - Aggregate long-term impact: absent policy, long-term GDP permanently lower by 3 percent; manufacturing less developed.

- Policy scenarios and results
  - Context: 2020 government response enrolled 71,000 households into social safety net; one-time transfer to 29,000 households; social safety nets nascent and externally financed.
  - Policy scenario 1 — Boosting social safety nets:
    - Expand transfers to 30 percent of vulnerable population; package size 1 percent of GDP; boost bottom tercile income by ½ percent of GDP.
    - Results: urban worker incomes +4 percent; slight decrease in inequality; long-term GDP increases but structure not substantially changed.
  - Policy scenario 2 — Increasing investment in human capital:
    - Increase share of skilled population by 7 percentage points.
    - Results: GDP +6 percent; manufacturing weight +4 percentage points; incomes become more equal; rural Gini declines significantly.
  - Policy scenario 3 — Boosting agricultural productivity:
    - Expand program to about 0.5 percent of GDP targeted to poorest tercile of farmers.
    - Results: does not substantially increase long-term GDP; food prices lower by 0.5 percent relative to Covid shock; redistributes income toward small farmers, reducing inequality.
  - Combined policy package (social transfers + education + agricultural productivity):
    - Results: combined policies boost GDP by 11 percent; structural shift toward manufacturing; incomes rise across the board; inequality decreases significantly.

- Conclusions and policy implications
  - Main simulated shock outcomes absent policy: long-term GDP permanently lower by 3 percent; manufacturing less developed.
  - Permanent increases in human capital yield large gains and structural transformation toward manufacturing.
  - Expanded social safety nets and agricultural measures mitigate adverse impacts; combined policies can over-compensate the COVID shock.
  - Recommend whole-of-government systems approach, link education/training to labor market needs, and strengthen data systems.

### Digitalization, Financial Inclusion, and Public Financial Management (Findings and Recommendations)
- Digital adoption and metrics (preserved exactly)
  - Internet penetration: about 13 percent in 2014 to about 27 percent in 2019; estimated 30 percent in 2021.
  - Internet users in 2021: 2.4 million persons.
  - Social media users in 2021: 850,000 people.
  - Mobile connections in 2021: 100 percent of the population.
  - DFS adoption: 30 percent of adult population in 2019 (UNCDF, 2021); growth from about 9 percent in 2017 to about 30 percent in 2019.
  - Cost of 1 GB in 2021: US$3.26 (rank 152/230); network quality ranking: 168 out of 173.
  - Installed power capacity as of 2018: about 105 MW for population over 7 million.
- Government strategies and digital initiatives
  - "Mobile First" digital strategy launched 2019; ICT development strategy 2019-2029 with pillars: digital identity, digital economy, digital governance.
  - Giga project to connect approximately 10,387 schools; currently 205 schools connected; project estimates benefit of about 194 students and teachers and 273 local community members within 1 kilometer per connected school.
  - IFMIS coverage expanded to approximately 61 MDAs (from about 30 in 2020); upgrade to web-based IFMIS underway; linkage to Bank of Sierra Leone T24 system planned.
  - Biometric vetting used in wage and salary management; e-voucher subsidy pilots in agriculture; drone initiatives for medical deliveries.
- DFS usage and payments
  - Three MNOs with about 35,000 agents.
  - Value of mobile money transactions grew by 73.8 percent in September 2021 to Le 9.6 trillion from Le 5.5 trillion in 2019.
  - Registered customer accounts fell to 4.7 million in 2020 (from 6.2 million in 2019) due to clean-up; active users increased to 2.2 million (vs. 1.4 million).
  - In 2020 only 0.2 percent of total DFS transaction volumes were for digital loans.
- Public procurement, tax administration, and city-level innovation
  - NPPA launched electronic procurement system March 2021.
  - Ministry of Finance rolling out electronic contract profiling system (ECOPS) to track outstanding payments and upcoming procurement expenditure.
  - FCC property taxation IT system increased revenue ten-fold from about Le 7 billion in 2019.
  - NRA implementing web-based Integrated Tax Administration System (ITAS) for tax registration, returns, payments, and compliance.
- Policy recommendations for digitalization and DFS
  - Invest in electricity, complete fibre optic rollout, invest in roads/water/housing to attract IT skills.
  - Fast-track connectivity for education institutions; only about 1.5 percent of ~10,387 schools are connected.
  - Strengthen legal/regulatory frameworks (data protection, cybercrime, IP rights); finalize National Payment Systems Act (NPSA) and agent banking guidelines.
  - Improve regulator staffing/capacity (OFSID) and supervision of payments systems.
  - Digitalize credit-related services (Credit Reference Bureau) and improve credit infrastructure.
  - Use digital PFM tools (IFMIS, biometric vetting, e-vouchers) to improve spending efficiency, targeting, and cash management.

### Debt Dynamics, Fiscal Framework, and Prudential Approaches
- Debt facts and risks
  - Debt-to-GDP ratio increased by 28.7 percentage points to 76.3 percent of non-iron ore GDP in 2020 (from about 47.6 percent in 2015).
  - Debt service absorbed about 26.2 percent of revenue in 2020.
  - Foreign currency debt: approximately 50.3 percent of non-iron ore GDP (or 65.1 percent of total public debt stock).
  - Domestic debt stock end-2020: Le.10,634bn (or 26 percent of GDP).
    - Commercial banks hold 57 percent of domestic debt; BSL and non-banks 19 percent.
    - Approximately 40 percent of banking sector assets are government securities.
  - MTNDP implementation cost: about 12 percent of non-iron ore GDP per year; development spending averaged 7.3 percent of non-iron ore GDP over 2015-20.
- Medium-term fiscal anchor and calibration
  - Objective: reduce debt-to-GDP toward anchor of 51 percent of non-iron ore GDP in 12 years while keeping safety buffer (~19 percent) to avoid exceeding maximum debt ceiling of 70 percent of GDP.
  - Prudential approaches and consolidated assessment:
    - Three methods yield a credible maximum debt limit of about 70 percent of GDP.
    - Growth-maximizing debt with α = 0.25 implies optimal debt ~55 percent of GDP.
    - LICs-DSF PV total public debt benchmark (medium capacity) = 55 percent of GDP; nominal ceiling implied ~69 percent of GDP.
  - Calibrated debt anchor sensitivity (Text Table 4 results — percent of GDP)
    - Scenario outcomes for initial debt anchor: between 35.2 to 51.1 percent of GDP across scenarios.
    - Example alternative: safety buffer 18.9 percent implies debt anchor 51.1 percent (70.0 - 18.9).
  - SIP simulations for climate adaptation: additional 0.6 percent of GDP can be spent on adaptation without derailing fiscal consolidation and still hit medium-term debt anchor after 12 years.
  - Required additional mobilization to meet climate commitments:
    - 5.4 percent of GDP from revenue and grants to meet mitigation and adaptation.
    - 1.8 percent of GDP to meet adaptation commitments.
- Fiscal adjustment scenarios calibrated to converge to 51.1 percent of GDP in 12 years
  - Key parameters and inputs (preserved exactly)
    - Current debt ratio (% GDP): 76.2%
    - Target debt ratio (% GDP): 51.1%
    - Long-term real interest rate (%): 3.3%
    - Long-term real growth rate (2022-26) (%): 4.8%
    - Fiscal balance (% GDP) in 2021: -4.1%
    - Primary balance (pb): -6.9%
    - Convergence period for debt (years): 12
  - Scenario summaries
    - Scenario 1 (fast adjustment): debt decreases to hit MT anchor by 2033.
    - Scenario 2 (slow adjustment, five-year transition): primary balance ranges from -2.9 to +0.9 percent of non-iron ore GDP in 2022-25; set at +2.1 percent in 2026-33; implied long-term overall balance about 0.1 percent of non-iron ore GDP (2026-33). Scenario 2 described as most realistic.
    - Scenario 3 (slower adjustment, eight-year transition): primary balance ranges -4.1 to 2.2 percent (2022-28); set at 3.1 percent over 2029-33; implied long-term overall balance about 1.2 percent of non-iron ore GDP (2029-33).
  - Policy actions and preconditions:
    - Revenue mobilization to at least 20 percent of GDP by 2023 (MTNDP target) from about 13.8 percent in 2020.
    - Implement medium-term revenue strategy (MTRS); strengthen tax administration and exemptions governance.
    - Strengthen expenditure and commitment controls; no new domestic payment arrears; prioritize concessional external loans.
    - Deepen domestic financial markets, extend maturities, diversify investor base; PFM reforms and IFMIS expansion to support credibility.

### Inflation Dynamics, Monetary Policy, and Short-Run Drivers
- Historical inflation episodes and drivers
  - Two waves:
    - Great inflation (1984-93): inflation nearly quadrupled to 56.2 percent (from 15.7 percent in 1971-83).
    - Great disinflation (1996-2020): inflation receded to average 11.7 percent.
  - Long-run drivers (panel regression results):
    - M3 accounts for 63 percent of Sierra Leone’s long-term average inflation.
    - Net credit to government (NCG) accounts for 85 percent of M3 growth during 2002-2021.
    - NEER accounts for about a third of long-run inflation.
    - Deterministic drift explains about one quarter of historical inflation.
    - IT dummy coefficient -0.026 (adoption of inflation targeting lowers inflation by 2.6 percentage points).
  - Decomposition of episodes:
    - Great inflation contributors: NEER depreciation +47 percent; M3 expansion +34 percent.
    - Great disinflation contributors: NEER 37 percent; M3 30 percent; XR_regime 16 percent; residual 15 percent.
- Short-run dynamics (ARDL, SVAR, and impulse responses)
  - ARDL short-run findings (2009Q1–2021Q2):
    - Persistence dominates: lagged CPI coefficient = 0.95 (s.e. 0.03) in levels.
    - Reserve money coefficients small and not robustly different from zero.
    - Nominal exchange rate shows positive association in levels (0.08, s.e. 0.02) but weaker in first-difference specifications.
    - Interpretation: associations not causal; one percent increase in previous quarter’s CPI associated with higher current inflation (persistence effect).
  - Key SVAR impulse responses (one standard deviation shocks; variables in logs interpreted as percent deviations)
    - Price shock: immediate CPI increase ~13.6 percent in first quarter; 4.9 percent in second quarter; returns to baseline thereafter.
    - Monetary policy shock (tightening): CPI decreases ~2.3 percent in quarter two; ~3.0 percent in quarter three; returns by quarter four.
    - Terms of trade deterioration: no first-quarter impact; ~1.7 percent CPI increase in second quarter; returns by quarter three.
    - Exchange rate depreciation: no first-quarter impact; +0.9 percent in quarter two; peaks +3.3 percent in quarter three; returns by fifth quarter.
  - Variance decomposition of CPI (selected)
    - 1-Quarter: Price shock explains 100.0 percent of CPI variation.
    - 4th-Quarter: Price shock 0.91; Monetary policy shock 0.54; TOT shock 4.2; Exchange rate shock 4.4.
    - 12th-Quarter: Price shock 4.6; Monetary policy shock 11.4; TOT shock 6.0; Exchange rate shock 6.2.
  - Short-run summary:
    - A "0.54 percent increase in present quarter CPI" is consistent with inflationary expectations measured by past inflation.
    - Monetary base shows no contemporaneous impact; exchange rate depreciation associated with CPI increases but not always statistically significant.
    - Exogenous domestic price shocks (e.g., seasonal food supply) are important drivers of short-run inflation dynamics.
- Policy implications and recommendations
  - Maintain control of money growth and avoid deficit monetization; bank financing of the deficit crowds out private lending.
  - Monetary policy should be forward-looking; deepen BSL’s analytical capacity and inflation forecasting.
  - Establish well-articulated medium-term inflation objective; clarify BSL’s hierarchy with price stability as overriding objective.
  - Improve transparency and communication; improve monetary operations and liquidity management; coordinate with Ministry of Finance.
  - Enhance exchange rate flexibility, build reserve buffers, diversify foreign exchange sources.
  - Reduce reliance on food imports and enhance agricultural productivity to reduce domestic price shocks (sectoral policy, not central bank lead).
  - Fiscal prudence and PFM improvements necessary to support disinflation.

*Source: Extracted content from 1sleea2022002 (IMF staff chapter).*

### 1. Climate Change and Mining Activities _________________________________________________ 9

### 1. Climate Change and Mining Activities

### A. Nexus of Climate Vulnerability, Development, and Debt Sustainability
- Sierra Leone is "highly vulnerable to the adverse impacts of climate change despite low level of exposure to climate-related hazards."
- Physical exposure to climate-related hazards is lower than the world average, but low human development limits coping capacity.
- Notable climate-related events and impacts:
  - On August 14, 2017, flooding and a subsequent landslide in Freetown "took over a thousand lives."
  - World Bank damage and loss assessment estimated "a total economic loss of over US$31 million and recovery needs of over US$82 million (or 0.8 percent of GDP in 2016)."
  - Loss of "2,850 ha of protected forest in the watershed area" associated with a "40 percent drop in the water level of the Guma dam" between 2015 and 2021 (reported source: WFP, 2022).
- Socioeconomic vulnerabilities that amplify climate risk:
  - Agriculture employs "54 percent of total population" and, together with fishing and forestry, "constitutes 59 percent of GDP in 2020."
  - High dependence on rain-fed farming increases vulnerability to temperature rises and extreme weather.
  - "About 56.8 percent of population in 2018 were living below US$1.90 per day (World Bank, 2020)."
  - "Almost half of the population are not consuming sufficiently nutritious diet," including "4.6 percent of population in severely food insecurity."
- Environmental degradation:
  - Sierra Leone "lost 1.82Mha of relative tree cover (equivalent to a 32 percent decrease since 2000)" (Global Forest Watch), driven by urbanization, international timber demand, and weak regulatory oversight.
- Fiscal and institutional constraints:
  - "The overall fiscal deficit excluding grants is estimated at 9.3 percent of GDP in 2021 and is projected to remain above 6 percent of GDP until 2024."
  - Tight fiscal space and weak budget execution limit capacity for risk transfer, post-disaster relief, and reconstruction.
  - Recommendation emphasis: mainstream adaptation into national development agenda; prioritize projects that meet both development and adaptation objectives; strengthen Public Financial Management (PFM) to plan and govern climate spending.

### B. Climate Commitments, Plans, and Institutional Framework
- National commitments and targets:
  - Sierra Leone approved its first NDC in November 2016 and presented an updated version in 2021.
  - Mitigation targets: "reducing CO2 emissions by 5 percent before 2025, 10 percent before 2030, and 25 percent before 2050."
  - Adaptation target: "reducing vulnerability by half by 2030."
- Planning instruments:
  - In 2021 Sierra Leone published the initial National Adaptation Plan (iNAP) to operationalize adaptation identified in the updated NDC.
  - The iNAP follows UNFCCC-aligned blocks: i) identifying climate impacts, vulnerabilities, and risks; ii) adaptation-related policies, plans and programs; iii) mainstreaming of adaptation; iv) monitoring, evaluation and reporting.
  - The iNAP identified five categories of adaptation programs per vulnerable sector: physical investments, human capacity development, institutional strengthening, regulatory modifications, research.
  - Together with the National Adaptation Plan of Action (NAPA) developed in 2007, authorities have prioritized "24 adaptation projects with detailed rationales, project objectives, inputs, outputs, institutional arrangements, risk and barriers, monitoring, and cost estimates."
  - Gap noted: "Selection criteria to prioritize these programs and projects were missing in the iNAP and NAPA."
- Regulatory and implementation institutions:
  - Sierra Leone Environmental Protection Agency (EPA) established in 2008 to address illegal logging, farming and mining in protected forest areas.
  - National Disaster Management Agency (NDMA) set up under the Office of the President in 2020.
  - National Protected Area Authority (NPAA) formed in 2012 to oversee national parks and protected areas and promote sustainable land use and environmental management.
  - Implementation challenges: weak oversight capacity of regulatory agencies limits enforcement and effective conservation.
- Sector-specific note:
  - The mining sector "contributes an important portion to fiscal and export revenues" and "needs to adapt proactively to climate change" (referenced as Box 1).

### C. Financing Implications and Policy Priorities
- Climate commitments and adaptation actions "generate significant financing needs" while constrained fiscal capacity risks "large financing gaps that threaten debt sustainability."
- Policy priorities highlighted:
  - Integrate climate costing into fiscal space analysis.
  - Enhance capacity to access external grants to fulfill climate commitments while maintaining debt sustainability.
  - Gradually adapt PFM practices to integrate climate targets within and beyond the budget cycle to mainstream adaptation and prioritize climate projects.
  - Prioritize adaptation investments that deliver both development and climate resilience benefits (for example, WASH infrastructure after flooding).
- Institutional strengthening:
  - Strengthen regulatory oversight and enforcement capacity of EPA, NPAA, and other agencies to address deforestation and environmental degradation.
  - Develop transparent selection criteria to prioritize adaptation projects within iNAP and NAPA frameworks.

_Prepared by Saanya Jain, Rosalind Mowatt, Peter Wankuru, and Yue Zhou._

### Box 1. Climate Change and Mining Activities

### Box 1. Climate Change and Mining Activities

### Climate risk to mining and shifts in commodity demand
- Climate change is expected to cause more frequent floods and droughts, altering the supply of water to mining sites and disrupting operations of mining companies.
- According to McKinsey’s estimates, 50 percent of global iron ore production is exposed to high risks of flood occurrence.
- Example: heavy rainfall in southeastern Brazil has put 100 million tons of annualized iron ore supply at risk, roughly 7 percent of seaborne supply and about 30 percent of Brazilian supply.
- A green recovery after the pandemic will likely shift commodity demands towards those needed for low-carbon technologies.
- More energy-efficient processing and widespread recycling would put pressure on the iron-ore and bauxite markets, particularly lower-grade products.
- Under the Paris Agreement to limit global warming by 2 degrees, iron ore and bauxite will see growth from new decarbonization technologies offset by increased recycling rates and focus on metal production from recycling versus virgin ore.
- Country-specific implication: in Sierra Leone, low-grade ore (e.g. Tonkolili) will take a hit as it requires more carbon-intensive processing to meet market standards, while demand will shift towards high-grade ore (e.g. Marampa) as international communities decarbonize the industry.

### Regional and national fiscal costs of mitigation and adaptation
- Sub-Saharan Africa quantified financing needs: total US$1,318 billion (79 percent of GDP in 2020) for climate commitments in the NDCs.
  - Mitigation needs: US$571.6 billion (34 percent of regional GDP in 2020).
  - Adaptation needs: US$407.2 billion (25 percent of regional GDP in 2020).
  - Unspecified financing needs: US$339 billion.
- Sierra Leone’s updated NDC estimates climate-related spending needs at US$2.8 billion (66 percent of GDP in 2020) between 2021 and 2030, revised upward from US$900 million in the previous version (NDC, 2021).
  - Estimated costs: Mitigation US$1.7 billion; Adaptation US$1.1 billion.
  - Implied ambitious annual fiscal cost: about US$276 million at 6.6 percent of GDP or 43 percent of fiscal revenue in 2021.
- Government plans: NDC indicates Sierra Leone will allocate 10 percent of its annual budget to combat climate change and secure 40 percent of donor funding for adaptation priorities, but no explicit financing plan or realistic strategy is provided.

### Financing scenarios and fiscal implications (IMF staff calculations, Table 1)
- Scenario assumptions:
  - Scenario 1: government allocates 10 percent of revenue and 40 percent of grants for mitigation and adaptation actions. Remaining financed by budget deficit.
  - Scenario 2: government allocates 2.5 percent of revenue and 20 percent of grants for mitigation and adaptation actions. Remaining financed by budget deficit.
  - Scenario 3: government allocates 2.5 percent of revenue and 20 percent of grants for adaptation actions only. Remaining financed by budget deficit.
- Table 1 estimates (US$ billion)
  - Total financing needs: Scenario 1 = 2.76; Scenario 2 = 2.76; Scenario 3 = 1.08.
  - Resource mobilized: Scenario 1 = 1.43; Scenario 2 = 0.53; Scenario 3 = 0.53.
    - -revenue: Scenario 1 = 0.74; Scenario 2 = 0.19; Scenario 3 = 0.19.
    - -grants: Scenario 1 = 0.69; Scenario 2 = 0.34; Scenario 3 = 0.34.
  - Annual deficit: Scenario 1 = 1.33; Scenario 2 = 2.23; Scenario 3 = 0.55.
  - percent of GDP annually: Scenario 1 = 3 percent; Scenario 2 = 4.9 percent; Scenario 3 = 1.2 percent.
- Scenario results (interpretation in source):
  - Scenario 1: raise US$1.4 billion (US$0.7 and US$0.7 billion from revenue and grants respectively) between 2021 and 2030; budget deficit on average at 3 percent of GDP; public debt becomes unsustainable in the medium term.
  - Scenario 2: raise US$530 million (US$190 and US$340 million from revenue and grants respectively) between 2021 and 2030; budget deficit at 4.9 percent of GDP on average; jeopardizes debt sustainability.
  - Scenario 3: with same resource mobilization as Scenario 2 but limited to adaptation (US$1.076 billion), total financing needs decrease by 60 percent and annual deficit narrows to 1.2 percent of GDP between 2021 and 2030; public debt remains unsustainable.

### Fiscal space, anchors, and debt dynamics
- SIP simulations: an additional 0.6 percent of GDP can be spent on climate adaptation without derailing fiscal consolidation substantially and still hitting the medium-term debt anchor after 12 years.
- Required additional mobilization:
  - 5.4 percent of GDP from revenue and grants to meet mitigation and adaptation commitments.
  - 1.8 percent of GDP to meet the adaptation commitments.
- Calibrated overall balance and debt trajectory (non-iron ore GDP basis):
  - Calibrated overall balance ranges from -6.9 to -0.7 percent of non-iron ore GDP in 2025 before being set at a surplus level of +0.8 percent of non-iron ore GDP in the long-term.
  - Implied debt to GDP increases from about 76.2 percent of non-iron ore GDP in 2021 to about 78.5 percent in 2023 (allowing for scale-up of climate adaptation projects) before converging to about 51 percent of non-iron ore GDP by 2033.
  - To meet adaptation commitments in Scenario 3, authorities would require additional revenue mobilization of 0.6 percent of non-iron ore GDP (or additional debt).
- Analysis caveats:
  - NDC estimated costs are used for fiscal deficit analysis but are not the actual financing needs.
  - Mitigation and adaptation measures are assumed additional to development spending, requiring extra resources.

### Policy recommendations and priority actions
- Calibrate climate costing and prioritize climate projects:
  - Incorporate macro-fiscal implications of climate adaptation, including costs, residual risks, and benefits such as potential loss recovery.
  - Publish climate-related investment projects with clear and transparent project appraisal criteria based on cost-benefit analysis and distributional analysis.
  - Develop a realistic financing plan based on medium-term revenue and expenditure strategies, emphasizing securing external climate grants.
- Prioritize adaptation measures given limited resources; extend the three pillars approach to natural disasters to guide adaptation identification:
  - Prevention of climate risk (e.g., infrastructure).
  - Alleviation of residual climate risks (e.g., social safety nets).
  - Ensuring resilient finance (e.g., flooding insurance).
- Prioritization criteria for adaptation projects:
  - Invest or subsidize adaptations that have positive externalities.
  - Remove barriers to efficient private adaptation.
  - Address equity issues within and between countries.
- Enhance capacity to access external grants for adaptation:
  - Strengthen national systems, improve governance systems and other financial and administrative controls, and build staff expertise to manage funding.
  - Note: only US$65 billion committed to SSA for adaptation between 2015 and 2019.
- Explore a package of measures to boost social resilience and carbon finance:
  - Accelerate post-disaster reconstruction, improve capacities for post-disaster supports, develop contingent finance and reserve funds.
  - Explore carbon credits to finance actions; strengthening enforcement and project pipelines is needed.
  - Evidence: Hallegatte et al. (2016) shows that by building the resilience of the poor, Sierra Leone could reduce the climate-related GDP losses by 23 percent.
- Adopt and sequence Green Public Financial Management (Green PFM) reforms to integrate climate objectives into the budget cycle:
  - Budget preparation: integrate climate considerations into Public Investment Management appraisal templates and economic forecasts.
  - Budget execution, accounting, and reporting: enhance budget performance monitoring and SOE performance monitoring to include climate considerations.
  - Control and audit: use control and audit mechanisms to measure and monitor efficiency and effectiveness of climate policies.
  - Preconditions for Green PFM: basic functional PFM system, strong political support, some degree of “green” expertise, and appropriate sequencing of reforms.
- Financing implications: tapping external grants and concessional loans is critical to fulfil climate commitments while maintaining debt sustainability; international community support is required to avoid exacerbating debt situations.

*Source: Box 1. Climate Change and Mining Activities, 1sleea2022002 (IMF staff summary of content).*

### References

### Reaping the Demographic Dividend in Sierra Leone - The Role of Education

### Major findings and motivation
- Policies to support the demographic transition are critical to reaping the demographic dividend and achieving macroeconomic and development objectives (World Bank, 2016; Bloom, 2020).
- A growing workforce presents opportunities for development when accompanied by human capital accumulation (education, health, training).
- Some studies estimate that the demographic transition has driven 40-50 percent of India's per capita income growth since the 1970s (Aiyar and Mody 2011).
- Africa will contribute 80 percent of the projected increase in global population by 2100, presenting opportunities for industrialization and development.
- Sierra Leone's population is young: in 2020, more than 40 percent were younger than 15, and more than half were younger than 20.
- Under the United Nations medium-variant assumption on fertility and mortality, the working age population is projected to increase from 4.5 million in 2020 to 8.5 million people by 2050.
- Declining dependency ratios are expected as fertility and mortality rates continue to drop in line with past trends.
- Reaping the demographic dividend is not automatic; without sufficient education, training, health investments, and job creation, the youth bulge could become a source of social unrest.

### Key demographic projections and statistics (preserved exactly)
- 2020: more than 40 percent of Sierra Leoneans were younger than 15, and more than half were younger than 20.
- Working-age population projection: from 4.5 million in 2020 to 8.5 million by 2050 (UN medium-variant).
- The medium-variant assumption: fertility in Sierra Leone would converge towards the replacement level (2.1 children per woman).
- Variants: low variant assumes 0.5 births per woman below the medium level; high variant assumes 0.5 above; constant-fertility implies level observed on average in 2015-2020.

### Baseline education outcomes and gaps (2018 and recent trends)
- Adult literacy rate had been on the rise since 2013 but lagged behind the regional average.
- Enrolment rates for Sierra Leone were well below the median for the sub-Saharan African region and other developing countries.
- Student-per-teacher ratio in Sierra Leone was higher than regional peers, indicating need for more teachers relative to students.
- Almost all public expenditure in the sector goes towards recurrent costs; under-investment in the capital budget negatively affects education performance.
- Weak sector management and governance: (i) weak policy and regulatory environment; (ii) inadequate quality assurance systems across sub-sectors; (iii) an EMIS which is fragmented and under-used (World Bank, 2021).
- Benefit incidence analysis (BIA) for 2018:
  - Primary education: share of spending to poorest quintile increased between 2003 and 2018; share to richest quintile decreased.
  - Secondary education: distribution remains regressive despite changes.
- Gross vs. net enrolment gaps:
  - Primary: gap between gross and net enrolment decreased for poorer quintiles (suggesting slight improvement).
  - Secondary: enlarging gap driven by high inflow of individuals outside school age and probably high repetition rates, implying deteriorating secondary education quality.
- Reasons for dropping out (2018 SLIHS):
  - Lack of finance is the main barrier: 41 percent of the bottom income quintile and 22 percent of the top quintile reported expensive education and lack of finance as the primary reason for dropping out (primary and secondary).
  - Other barriers: lack of interest, marriage/pregnancy, having completed the desired level; distance to schools significant for the poorest.

### Recent policies, resource allocation, and support
- Free Quality School Education (FQSE) aims to increase access and quality across pre-primary, primary, secondary, and technical and vocational education; enshrined in National Development Plan (NDP) 2019-2023 “Education for Development”.
- Government assumes resource burden for tuition, core textbooks, furniture, learning materials, infrastructure, teacher training and recruitment.
- Resource allocation increases:
  - Sierra Leone meets the Dakar target of allocating at least 20 percent of government resources to education; public spending on education reached 21.4 percent in 2022.
  - World Bank (2021) estimates a significant financing gap to fully deliver the FQSE Program in the next four years, mainly reflecting infrastructure gap and salaries for additional teachers.
- Development partner support (examples preserved exactly):
  - Free Quality School Education Project multi-donor trust fund received US$50 million from the World Bank and US$15.9 million from Irish Aid, European Union and FCDO in October 2020 (World Bank 2020).
  - Global Partnership for Education (GPE) provided a US$17.2 million grant in 2019; GPE announced in 2021 Sierra Leone’s eligibility for a US$38 million funding, of which US$22.47 is a system transformation grant, a US$15 million multiplier grant and an additional US$2 million capacity grant.

### Selected government measures (Box 1)
- Expand the provision of accelerated learning to out-of-school girls
- Scale-up efforts for their reintegration into school as well as teenage mothers
- Provision of sanitary hygiene kits for girls to reduce period poverty
- Rollout of a comprehensive safety policy and comprehensive sexuality education
- Establishment of digital learning hubs
- Use of data to guide site selection for construction of schools or classrooms and planning for other interventions like distributions of learning materials
- Use of satellite imagery to monitor the impact of weather conditions on schools
- Establishment of the Youth Advisory Group (YAG) for Education to ensure that decision making at all levels of education happens with children and young people at the center

### Policy priorities and recommendations distilled from the paper
- Increase and optimize public spending on education, including a higher weight for school infrastructure and non-wage items in the public education budget.
- Improve teacher quantity and quality: recruit more teachers and strengthen teacher management to reduce absenteeism.
- Strengthen sector management and governance: improve policy and regulatory environment, quality assurance systems, and an integrated, well-used EMIS.
- Target interventions to improve access for the poorest: address financing barriers, reduce distance constraints, and support reintegration of pregnant girls and teenage mothers.
- Expand investments in health and primary health care (including vaccine coverage) to support human capital accumulation.
- Link school subsidies to school performance and outcomes; support education for disadvantaged groups.
- Develop and finalize a new Education Sector Plan to guide planning, management, and delivery of quality education services.
- Constitutional affirmation of free, quality, and compulsory primary and secondary education (subject to approval) to support progress toward SDG 4 and the NDP.

### Key statistics and figures cited (preserved exactly)
- 40-50 percent (estimate of India's per capita income growth attributable to demographic transition since the 1970s)
- 80 percent (Africa's contribution to projected increase in global population by 2100)
- 4.5 million (working-age population in 2020)
- 8.5 million (projected working-age population by 2050 under UN medium-variant)
- 2.1 children per woman (replacement-level fertility in medium-variant assumption)
- 21.4 percent (public spending on education reached this share in 2022)
- 41 percent (share of bottom income quintile citing expensive education/lack of finance as primary dropout reason)
- 22 percent (share of top quintile citing expensive education/lack of finance as primary dropout reason)
- US$50 million (World Bank contribution to FQSE Project multi-donor trust fund)
- US$15.9 million (contributions from Irish Aid, European Union and FCDO in October 2020)
- US$17.2 million (GPE grant in 2019)
- US$38 million (GPE eligibility announced in 2021), of which:
  - US$22.47 (system transformation grant)
  - US$15 million (multiplier grant)
  - US$2 million (capacity grant)

### References (as listed in the source)
- African Development Bank (AfDB). 2019. “Analysis of Adaptation Components of Africa's Nationally Determined Contribution.” Abidjan, Côte d’Ivoire.
- Climate Finance Advisors. 2021. “Increasing Sub-Saharan African Access to Climate Finance Facilities.” Washington, DC.
- Delevingne, Lindsay, et al. 2020. "Climate risk and decarbonization: What every mining CEO needs to know." Report McKinsey & Company.
- Food and Agriculture Organization. 2017. “Restoring Food Security to Flood-Affected Families in Sierra Leone.”
- Gonguet, Fabien, et al. 2021. "Climate-Sensitive Management of Public Finances—“Green PFM.” IMF Staff Climate Note. Washington, DC.
- Hallegatte, S., Vogt-Schilb, A., Bangalore, M. and Rozenberg, J., 2016. Unbreakable: building the resilience of the poor in the face of natural disasters. World Bank Publications.
- International Monetary Fund (IMF). 2016. “Small States’ Resilience to Natural Disasters and Climate Change— Role of the IMF.” IMF Policy Paper, Washington, DC.
- International Monetary Fund (IMF). 2019a. “Fiscal Monitor: How to Mitigate Climate Change.” Washington, DC, October.
- International Monetary Fund (IMF). 2020. " Regional Economic Outlook: Sub-Saharan Africa." Washington, DC.
- Islam, N. and Winkel, J., 2017. "Climate change and social inequality.”
- Irish Aid. 2015. “Sierra Leone Climate Report.”
- Novta, Natalija, Gemma Preston, and Sureni Weerathunga. Forthcoming. “Access to Climate Finance: The Pacific Island Countries.” IMF Departmental Paper, International Monetary Fund, Washington, DC.
- Reuters, March 2019. “With reservoirs at risk, Sierra Leone capital confronts water crisis.” https://www.reuters.com/article/us-leone-water-feature/with-reservoirs-at-risk-sierra-leone-capital-confronts-water-crisis-idUSKCN1V9007
- The Republic of Sierra Leone. 2021. “Updated Nationally Determined Contribution.” https://www4.unfccc.int/sites/ndcstaging/PublishedDocuments/Sierra percent20Leone percent20First/210804 percent202125 percent20SL percent20NDC percent20(1).pdf
- World Food Program. 2020. “Findings of Sierra Leone January 2020 Food Security Monitoring.”
- World Food Program, 2022. “Sierra Leone: Western Area Forest Reserve. 2022 Preliminary Assessment.”
- World Bank. 2020. “Poverty & Equity Brief: Sierra Leone.” Washington, DC. https://databank.worldbank.org/data/download/poverty/987B9C90-CB9F-4D93-AE8C-750588BF00QA/SM2020/Global_POVEQ_SLE.pdf

*International Monetary Fund — SIERRA LEONE: REAPING THE DEMOGRAPHIC DIVIDEND IN SIERRA LEONE - THE ROLE OF EDUCATION (selected content and references from the source PDF).*

### 12.      Indeed, first evidence is emerging that enrolment rates and learning outcomes have

### 1sleea2022002 - 12.      Indeed, first evidence is emerging that enrolment rates and learning outcomes have

### Recent outcomes in the education sector
- From 2016 to 2019, primary completion rates increased from 62.8 to 69.4 percent.
- Lower and upper completion rates increased from 40.9 to 47.0 percent and 14 to 19.6 percent, respectively.
- In 2020 these completion-rate gains reversed as COVID-19 posed additional challenges to learning.
- The number of children taking and completing the National Primary School Exam (NPSE) and the Basic Education Certificate Examination (BECE) increased overall, though 2018-19 numbers saw a dip as the government’s focus on reducing exam malpractice increased.
- Sierra Leone’s National Early Grade Reading and Mathematics Assessment Baseline Study (2021) noted significant improvement in early grade reading and math in 2021 compared to a similar study conducted in 2014.
- Students accessed digital learning tools millions of times during school closures (figures in text: "millions of times").
- The Ministry launched the Learning Passport; it was downloaded 4,000 times on the first day.

### COVID-19 interruptions and EdTech response
- Schools were closed for 14 weeks as part of containment measures.
- The Ministry of Basic and Senior Secondary Education (MBSSE) and the Teaching Service Commission (TSC) convened an Education Emergency Taskforce and introduced a Hybrid EdTech solution delivering learning via radio, television, mobile phone, the web, and print media.
- Radio teaching program design:
  - 30 minutes of lectures plus 30 minutes of interaction with teachers by toll free phone line.
- Mobile-based learning technologies developed and deployed included USSD/SMS Results Checker, Dictionary, and Placement Checker (at zero cost for students).
- The Government, through Giga and partners, has an ongoing project to expand school connectivity to approximately 11,000 schools.
  - Currently, 205 schools have been connected to the internet.
  - For every school connected to the internet, it is projected that about 194 students and teachers and 273 local community members within 1 kilometer will benefit.

### Key short- and medium-term macro and social impacts of COVID-19
- Annual per capita losses for emerging market and developing economies (excluding China) in 2020-2022 amounted to 6.3 percent.
- In Sierra Leone:
  - Per capita incomes are expected to return to 2019 levels only by 2022.
  - Tax revenues decreased from 12.3 percent in 2019 to 11.2 percent in 2020 (drivers: temporary import tax deferrals, deferrals on corporate and income tax, suspension of in-person tax enforcement).
  - The service sector was hit particularly hard in 2020.
  - In a Ministry of Finance survey, eight out of ten businesses in the tourism sector were running a loss compared to the period before COVID-19.
  - The shock to household incomes implies that poverty rates have likely risen from already high pre-crisis levels (56.8 percent in 2018).
  - Schools were closed for 14 weeks in Sierra Leone (compared to 31 weeks for the LIC average).
  - The share of people who are food insecure had increased significantly to about 60 percent of the population in June 2020.
  - Female-headed households experienced a larger economic shock and bought less food during the initial months of the pandemic.

### Model framework and analytical approach for longer-term impacts
- The paper calibrates a dynamic multi-sector heterogeneous agent general equilibrium model to Sierra Leone to analyze longer-term economic and distributional effects of COVID-19.
- Main model elements:
  - Households: urban (private sector workers, public sector workers, entrepreneurs) and rural (small and large farmers); private sector workers are high- or low-skilled; households face idiosyncratic and transitory income shocks; receive international remittances and government transfers; save using a simple one-period bond with borrowing limits; pay taxes.
  - Production sectors: agriculture (domestic consumption and exports), manufacturing, services, and mining; sectors require different inputs (labor, land, capital, energy, intermediate goods); households can produce informal services requiring only labor.
  - Sectoral productivity vulnerability: sectors differ in degree of contact exposure and vulnerability to shocks.
  - Government: collects consumption tax, labor and corporate income tax, energy sector taxes, taxes on agricultural exports, and tariffs; expenditures cover wages, recurrent expenditures, social transfers, and public investment.
- High-level model findings (summary from text):
  - The crisis likely dampened households’ income and long-term real GDP.
  - The shock to human capital alone would exacerbate income inequality somewhat, but the more broad-based decline in incomes across population groups suggests that long-term income inequality could modestly decrease.
  - Policies can mitigate losses: investments in human capital (e.g., Free Quality School Education program) can more than compensate for the loss of income; social transfers can reduce poverty and income inequality but need careful calibration and sizing.

### Conclusions and policy recommendations
- Strategic priority: Given demographic trends, investments in human capital will have high returns; policies should prioritize improving access to quality education and health.
- Education investment focus:
  - Invest at primary and secondary levels and develop tertiary education to meet demand for high-skill labor.
  - Expand Technical-Vocational training for youths who dropped out (e.g., early pregnancies) to increase employability or ability to create jobs.
  - Support complementary policies, such as the school feeding policy, to increase the return to education spending and serve as a safeguard against shocks.
- Selected policy recommendations from the World Bank's 2021 public expenditure review:
  - Optimizing spending: prioritize between salary and non-salary expenditures to improve availability of learning materials and a favorable learning environment.
  - Supporting disadvantaged groups: address weak infrastructure, teacher quality, shortage of learning materials, crowded classrooms, opportunity cost of schooling, distance to school, teenage pregnancy and (forced) marriage, and gender-based violence and sexual harassment; continue radical inclusion efforts and develop targeted policies.
  - Link school subsidies from the FQSE to school performance and outcomes (pupil and teacher attendance, pass rate in exams).
  - Improve teacher quality: Teaching Service Commission should only add teachers to the payroll if they meet minimum standards; strengthen teacher registration and licensing; expand supporting infrastructure (e.g., teachers' colleges).
  - Strengthen quality assurance: monitor quality beyond government schools and strengthen capacity of local agencies at the district level.
- Macroeconomic and fiscal recommendations:
  - A strong fiscal framework with a debt anchor could help allocate resources to promote education without jeopardizing macro stability.
  - Implement a medium-term revenue mobilization strategy and efficient revenue administration to raise revenue for education and structural reforms.
  - Follow up on PIMA and PEFA recommendations to improve spending efficiency.
- Cross-sectoral and institutional recommendations:
  - Continue a whole-of-government 'systems' approach to human capital development; strengthen collaboration across the two education ministries and the Ministry of Youth Affairs.
  - Complement education with investments in health, food security, and social protection; encourage development partners and the private sector to invest.
  - Design policies to create high-quality jobs (competitive labor and capital markets, human capital investments, infrastructure, sound macroeconomic management, careful trade policies, good governance).
  - Align education and training with labor market needs: let job-market skills guide curriculum design and training across all levels; ensure inclusivity for women, girls and persons living with disabilities.
  - Leverage digital offerings and nurture entrepreneurial spirit and innovation in response to the fourth industrial revolution.
  - Develop internationally recognized frameworks for certification of technical and vocational skills in preparation for AfCFTA implementation.
  - Strengthen data systems to capture changes across education and labor at higher frequencies to monitor outcomes, perform impact analysis, and guide policy and program design.

*Source: Excerpt from IMF staff paper and Sierra Leone country analysis as provided.*

### 7.       The baseline calibration aims to capture the main structure of Sierra Leone’s economy

### 7.       The baseline calibration aims to capture the main structure of Sierra Leone’s economy

### Baseline calibration and data sources
- Baseline year: 2019 (chosen to avoid distortions from past shocks such as the Ebola health crisis and commodity price shock).
- Datasets guiding calibration:
  - National account statistics: calibrate shares of the five sectors in overall production; CPI and CPI weights determine household consumption profile; balance of payments statistics inform share of imports in consumption.
  - 2018 Sierra Leone Integrated Household Survey: distributional target moments, including overall, rural and urban Gini coefficients.
  - Fiscal statements and World Bank public expenditure review on social spending: inform detailed expenditure and revenue categories and nature/distribution of social transfers.
  - 2014/2015 Labor force survey: shares of employment, rural vs. urban population shares, wage and productivity differentials; comparator country estimates as robustness checks.
  - 2013/2019 Demographic and Health Surveys: educational attainment to target shares of skilled and unskilled population.
- Key calibration comparison (Data targets 1/ based on 2019 data) — selected figures preserved exactly:
  - Agriculture share of GDP: Data 0.53, Model 0.54
  - Agriculture sector share in GDP (domestic): Data 0.40, Model 0.46
  - Agriculture export sector share in GDP: Data 0.14, Model 0.08
  - Manufacturing sector share in GDP: Data 0.02, Model 0.06
  - Service sector share in GDP: Data 0.41, Model 0.34
  - Mining sector share in GDP: Data 0.05, Model 0.06
  - Tax revenues to GDP 2/: Data 0.12, Model 0.28
  - Labor income tax: Data 0.04, Model 0.04
  - GST: Data 0.03, Model 0.03
  - Non-tax revenues to GDP: Data 0.02
  - Government spending to GDP: Data 0.15, Model 0.15
  - Gini national: Data 0.37, Model 0.39
  - Gini urban: Data 0.33, Model 0.31
  - Gini rural: Data 0.27, Model 0.27
  - External grants to government to GDP: Data 0.03, Model 0.03
  - Trade balance to GDP: Data -0.26, Model -0.33
- Note on fiscal/external balancing: 2/ "The deviation between actual data and the model is due to the external and fiscal balance condition needing to hold (no debt creating flows)."

### Caveats on model assumptions and data quality
- Model scope and simplifying assumptions:
  - Focus on medium- to long-term distributional outcomes; model does not allow statements about short-term impacts (e.g., lockdowns) or short-term policies.
  - Government budget and current account are balanced in the model; model abstracts from implications on debt accumulation and external stability.
  - Labor market clears in the model; shocks and policies do not impact the level of unemployment.
- Data quality issues in Sierra Leone:
  - National accounts and BOP statistics have methodological deficiencies.
  - Some apparent inconsistencies across datasets (e.g., weight of food in consumption basket vs. share of agriculture in GDP).
  - Fiscal and external sectors are constrained to balance in the model, requiring calibrated baseline deviations from actual data (example: remittances are only 2 percent of GDP in data, but the model requires additional private inflows to balance external sector).
  - These balancing assumptions are stated as not affecting the economic predictions of the model.

### C. Impact of the COVID-19 shock — Overview
- Two main channels analyzed: human capital (school closures, increased food insecurity, health impacts) and total factor productivity (TFP) in individual sectors (agriculture, manufacturing, services).
- Shock magnitudes and assumptions are drawn from cited evidence within the document (World Bank estimates and Barrett and others).

### Shock to human capital — assumptions and impacts
- Assumed education and dropout impacts captured:
  - World Bank estimates: 1.32 percentage point increase in dropouts and 20 percentage points in students repeating grades (World Bank 2021a).
  - Food insecurity: 6-percentage point increase in proportion of food insecure population between end-2019 and end-2020.
  - Modeling assumptions: drop in productivity of 1.8 percent for low skilled workers and 0.06 percent for high skilled workers.
  - Increase in percentage of private unskilled households in urban areas by 1.32 percent to reflect increased dropouts.
- Results from human capital shock:
  - Long-term GDP change: -1.1 percent.
  - Sectoral shifts: relative weight of services declines; manufacturing-to-GDP increases marginally by +0.5 percentage points; agriculture-to-GDP declines despite no change in farmer productivity (due to reduced food demand).
  - Distributional effects: rural income inequality declines (large farmers’ income falls); urban income inequality increases (urban low-skilled workers hit harder); overall income inequality increases slightly.

### Shock to total factor productivity (TFP) — assumptions and impacts
- TFP shock assumptions (reflecting sectoral vulnerabilities):
  - Small farmers TFP: -1 percent
  - Large farmers TFP: -1 percent
  - Manufacturing TFP: -2 percent
  - Services TFP: -3 percent
- Rationale: services sector most affected by COVID in Sierra Leone; rural/agricultural output less impacted.
- Results from TFP shock:
  - Long-term GDP change: -2 percent.
  - Sectoral shifts: manufacturing sector share declines by -0.5 percentage points.
  - Distributional effects: income losses larger for higher-earning groups (large farmers, entrepreneurs), leading to modest declines in income inequality in both rural and urban areas, and in aggregate.

### Combined shocks (human capital + TFP)
- Effect: amplifies loss to human capital; however, income inequality decreases modestly because lower TFP creates broader-based income declines.
- Aggregate long-term impact (reported in Conclusion): absent policy measures, long-term GDP is lower by 3 percent permanently and manufacturing sector is less developed.

### D. The impact of mitigating policies — policy scenarios and outcomes

- Context: Government response in 2020 enrolled 71,000 households into social safety net; one-time transfer to 29,000 households with informal urban workers; cash and in-kind support for people with disability and farmers. Social safety nets characterized as nascent, fragmented, and largely externally financed.

- Policy scenario 1 — Boosting social safety nets
  - Design:
    - Expand transfers to 30 percent of the vulnerable population by increasing transfers to the first tercile of both skilled and unskilled populations.
    - Overall package size: 1 percent of GDP.
    - Boost income of the bottom tercile of the income distribution in both groups by ½ percent of GDP, with external aid increased accordingly.
    - Second-step assumption: a proportionate increase in household spending on education, food and health for transfer recipients, and a proportionate increase in skills by one third of low and high-skilled workers, respectively.
  - Results:
    - Targets the poor effectively and has spillover effects for farmers via increased demand for agricultural goods.
    - Incomes for urban workers are boosted by 4 percent.
    - Slight decrease in income inequality.
    - Overall long-term GDP increases, but not enough to substantially change the structure of the economy given modeling assumptions.
    - Authors note assumptions likely underestimate human capital impacts because they do not model switching from low- to high-skilled households.

- Policy scenario 2 — Increasing investment in human capital (education)
  - Design:
    - Increase share of skilled population by 7 percentage points (consistent with observed increase in lower-secondary completion rates between 2015 and 2019).
  - Results:
    - Large gains: GDP increases by +6 percent.
    - Private sector investment share increases disproportionately.
    - Manufacturing sector weight increases by +4 percentage points as it absorbs a large part of skilled workers.
    - Incomes become more equal in both rural and urban areas; rural Gini declines significantly.

- Policy scenario 3 — Boosting agricultural productivity
  - Design:
    - Capture existing measures (rural roadwork, transfers and inputs) and expand program size to about 0.5 percent of GDP targeted to poorest tercile of farmers.
  - Results:
    - Does not substantially increase long-term GDP or change overall economy structure given package size.
    - Lowers food prices by 0.5 percent relative to the Covid shock.
    - Redistributes agricultural income toward small farmers, reducing aggregate inequality.

- Combined policy package (social transfers + education investment + agricultural productivity)
  - Results:
    - Combined policies boost GDP by 11 percent.
    - Accelerates transition toward a manufacturing base away from agriculture via increased formalization.
    - Incomes rise across the board; inequality decreases significantly.

### E. Conclusion — main findings and policy implications
- Model and approach:
  - General equilibrium model with heterogeneous agents calibrated to Sierra Leone-specific data to simulate medium- to long-term impacts.
- Main simulated shock outcomes (absent policy):
  - Long-term GDP permanently lower by 3 percent.
  - Manufacturing sector less developed.
  - Human capital shock alone increases income inequality; adding a TFP shock leads to a modest decrease in inequality because incomes fall more broadly.
- Policy options and effectiveness:
  - Permanent increases in human capital (education) yield particularly large gains — substantial increases in GDP and a structural shift toward manufacturing.
  - Expanded social safety nets and measures to boost agricultural productivity can mitigate adverse impacts; combined policies can over-compensate the COVID shock.
- Model capabilities and extensions:
  - Model allows simulation of further individual and combined shocks and policy packages (e.g., structural changes in remittances, changes in food/energy/commodity prices, distributional impact of sector-specific reforms, digitization, wage premium, health measures, and alternative social transfer designs).

*Source: Authors’ estimates.*

### References

### 1sleea2022002 - References

### Literature and data sources
- Cited institutional sources include AfDB, AUC/OECD, IMF, World Bank, UNCTAD, UNDP, WFP, FAO/IFAD/UNICEF/WHO, ILO, ITU, GSMA, Cable.co.uk, Speedtest, DataReportal, WDI, UNCDF, and peer-reviewed/preprint studies (medRxiv, Lancet Infect Dis, PNAS, BMJ Global Health, Science).
- Working papers and reports referenced: NBER Working Papers, IMF World Economic Outlook (October 2020 and October 2021), IMF Regional Economic Outlook: Sub-Saharan Africa (October 2021), IMF staff notes on labor market impacts and COVID-19 special series on gender equality.
- Country-specific and sectoral studies referenced: Sierra Leone COVID-19 High Frequency Phone Survey (World Bank), Sierra Leone 2021 Economic Update, Public Expenditure Reviews (Education; Social Assistance Programs), studies on tourism and hospital utilization in Sierra Leone, and multiple analyses of pandemic impacts on growth, poverty, remittances, and food insecurity.

### Pace of digitalization and adoption metrics in Sierra Leone
- Internet penetration increased from about 13 percent in 2014 to about 27 percent of total population in 2019 (GSMA, 2020) and is estimated to have reached 30 percent in 2021.
- Number of internet users in 2021: 2.4 million persons.
- Number of persons using social media platforms in 2021: 850,000 people.
- Mobile connections in 2021: 100 percent of the population.
- Adoption of digital financial services (DFS) estimated at 30 percent of adult population in 2019 (UNCDF, 2021).
- DFS adoption growth: from about 9 percent of adults using a DFS account in 2017 to about 30 percent in 2019.

### Digital infrastructure, affordability, and quality indicators
- 15 percent of the population has no mobile phone signal.
- 60 percent of the country has at least a 3G network coverage.
- Active mobile broadband subscriptions: 20 percent (ITU, 2020).
- Cost of 1 GB of internet data in 2021: US$3.26, ranking Sierra Leone 152nd out of 230 countries for cost of mobile data.
- Network quality ranking: 168 out of 173 countries for quality and speed (Speedtest, 2021).
- Sierra Leone’s installed power capacity as of 2018: about 105 MW for a population of over 7 million.
- ICT goods imports (2015-20): about 2 percent of total goods imports in Sierra Leone compared to the regional average of about 5 percent.

### Comparative digital adoption and gaps
- Sierra Leone lags the SSA average on the Digital Adoption Index (DAI) on a 0-1 scale (DAI panels referenced).
- Mobile connectivity broadly in line with regional average (Panel 3c) but below the regional average in terms of individuals’ use of the internet (Panel 3d).
- Sierra Leone ranks 126 out of 137 on firm-level technology absorption (WEF Global Competitive Report 2017-18).
- Average literacy rate context: SSA average about 65 percent over 2015-20; Sierra Leone lags in both average literacy rate and internet penetration compared to the regional average.

### Barriers constraining digitalization
- Affordability and quality of service remain relatively poor.
- Legal and regulatory challenges, including intellectual property rights and need for cybercrime reduction.
- Low private sector participation and limited access to finance.
- Lack of adequate IT skills and insufficient e-business skills.
- Insufficient electricity connectivity and unreliable electricity supply.
- Low access to IT equipment and related accessories.
- Measurement cautions: potential measurement errors and definitional scope for IT products may under- or overstate ICT goods import comparisons.

### Government strategies, initiatives, and recent progress
- Sierra Leone’s "Mobile First" digital strategy launched in 2019; ICT development strategy covering 2019-2029 with three pillars: digital identity, digital economy, and digital governance (DSTI).
- Priority projects include: ease of doing business, education data hub, integrated geographic information system, financial data mapping, GoSL appointment app, and environment.
- Legal reform: Cyber Security and Crime Act enacted in 2021.
- Infrastructure: rollout of fibre optic cable expected to increase access, quality, and speed of internet connectivity.
- International initiatives: Sierra Leone joined Giga (ITU and UNICEF) in 2021 to connect approximately 10,387 schools, eventually reaching over 3 million people around the schools in addition to the students (Giga, 2021).
- Recent unbundling of SALCAB to a private sector entity (Zoodlabs) expected to increase operational efficiency, improve wireless access, and reduce internet connectivity costs.

### Digitalization in public financial management (PFM) and service delivery
- IFMIS coverage expanded to approximately 61 ministries, departments, and agencies, from about 30 in 2020. IFMIS is being upgraded to a web-based version and will be connected to the T24 system at the Bank of Sierra Leone to allow electronic files to be transmitted from IFMIS to the Bank of Sierra Leone.
- Expected benefits of IFMIS upgrade: replace manual issuance of checks, shorten processing time, strengthen cash management and planning.
- Biometric vetting used in wage and salary management to help reduce ghost worker problems.
- Education: automated digital system for teacher assignments and student progress evaluation; scope to scale up to monitor teacher and health worker absenteeism.
- Agriculture: e-voucher based subsidy for small holder farmers to acquire seeds, fertilizers, tractor services, and chemicals.
- Health logistics and service delivery: historical use of mobile phones for salary payments to healthcare workers and targeting of cash assistance since Ebola; mobile apps used for tracing, monitoring, and managing COVID-19 outbreak.
- Drone initiatives: 2019 Drone Corridor (government and UNICEF) for medical supplies; 2021 launch of a medical drone delivery service (MDDS) by the Directorate of Science and Technology (DST) in collaboration with UAVaid and support from the Bill and Melinda Gates Foundation for delivery of medicines to remote parts of Sierra Leone.

### Policy implications and priorities identified
- Continue investment in human capital development and digital skills to ensure inclusive benefits from digitalization.
- Address affordability and quality of internet service to expand access and active mobile broadband subscriptions.
- Strengthen legal and regulatory frameworks to improve private sector participation, reduce cybercrime, and protect intellectual property rights.
- Improve electricity connectivity and increase power supply to support digital adoption.
- Expand access to IT equipment and reduce import-related constraints to hardware availability.
- Scale up DFS to increase financial inclusion, noting progress from about 9 percent DFS use in 2017 to about 30 percent in 2019, and the adult account ownership increase from about 20 percent in 2017 to about 24 percent in 2019 (account at a financial institution or mobile money provider).
- Leverage digital PFM tools (IFMIS, biometric vetting, e-vouchers) to improve spending efficiency, targeting, and cash management.

*Prepared by IMF staff; references and data cited within the chapter.*

### 15.      Publishing contracts online has improved transparency of spending. Sierra Leone has

### 15.      Publishing contracts online has improved transparency of spending. Sierra Leone has

### Transparency and public procurement
- Sierra Leone has continued publishing unaudited financial statements for its emergency financing response and key details of large procurement contracts related to crisis mitigation, and is reporting on their broader social and economic response.
- The National Public Procurement Authority (NPPA) launched the electronic procurement system in March 2021 to improve efficiency, mitigate corruption, transparency and accountability in procurement processing at public institutions.
- Publication of procurement and tender award information on the NPPA website is expected to bolster transparency and accountability in public spending.
- The Ministry of Finance is rolling out an electronic contract profiling system (ECOPS) that will track all outstanding payments for contracts and capture upcoming procurement expenditure from planning to payment stage. This is expected to improve the efficiency and transparency of public procurement transactions.

### Tax administration and IT modernization
- The National Revenue Authority (NRA) is implementing a new web-based Integrated Tax Administration System (ITAS) that digitalizes taxpayer registration, returns filing, payment of taxes, compliance management and other domestic tax management functions.
- ITAS is expected to reduce compliance costs for taxpayers and lower administration costs for the NRA, and to improve governance and reduce corruption by reducing face-to-face engagement.
- Integration of ITAS with third-party data is expected to enhance monitoring of tax compliance.
- ITAS interface with the Automated System for Customs Data (ASYCUDA) World will enable Customs and the domestic tax department (DTD) to share information on importer Tax Identifier Number (TINs), import GST declarations and withholding income tax paid at customs, facilitating taxpayer profiling, assessment and reconciliation.

### City-level innovation: Freetown City Council (FCC) property taxation
- FCC developed a new IT system leveraging satellite imaging to manage property taxation from data collection, valuation, billing, payments, appeals, to enforcement, supported by International Growth Centre and International Centre for Tax and Development.
- Results: revenue increased 10-fold from about Le 7 billion in 2019, enabling the city budget to be self-sufficient and boosting service delivery.
- This model is presented as a best practice for property and real-estate taxation in SSA.

### Digitalization and financial inclusion: overview
- Traditional banking services are challenging to access; financial inclusion remains low—less than the Sub-Saharan Africa average.
- Percentage of the adult population with an account at a financial institution or mobile money provider increased to 24 percent in 2019 from 20 percent in 2017.
- Percentage of adults with an account at a formal financial institution is much lower (12 percent in 2017, the latest year available).
- Women and rural residents lag behind in financial access.
- DFS can provide convenient payment and withdrawal channels as well as savings and borrowing channels, with potential to spur growth, entrepreneurship, and reduce poverty.
- DFS users grew at an average of 82 percent every year.

### Mobile money and DFS statistics and usage patterns
- Sierra Leone currently has three MNOs with about 35,000 agents in operation.
- Value of mobile money transactions grew by 73.8 percent in September 2021 to Le 9.6 trillion from Le 5.5 trillion in 2019.
- Number of registered customer accounts fell in 2020 to 4.7 million (down from 6.2 million in 2019) due to a clean-up of accounts; number of active users increased to 2.2 million (vs. 1.4 million), indicating improved DFS usage during Covid-19.
- The most popular MNO services are cash-in, cash-out and domestic peer-to-peer transfers.
- In 2020 only 0.2 percent of total DFS transaction volumes were for digital loans in Sierra Leone (UNCDF, 2021).

### Banks, microfinance and digital credit
- Commercial banks are expanding ATM access points, internet banking, mobile bank applications and USSD to grow customer base and complement brick-and-mortar services.
- Microfinance institutions are beginning to roll out some digital services.
- Microfinance institutions have partnered with MNOs to introduce digital credit products (examples: Orange Money Lajor Loan and AfriMoney Africredit).
- Regulatory reforms and more data collection on microfinance activities are needed to increase supply and uptake of digital credit services.

### Payments systems, interoperability and infrastructure
- By 2013 Sierra Leone had introduced RTGS for large-value transactions and ACH for retail transactions; ACH reduced settlement times from T+9 to T+1.
- Most commercial banks and the Bank of Sierra Leone (BSL) still do not provide complete straight-through-processing (STP) of transactions.
- The National Payments Switch, allowing interoperability of payment platforms (including MNOs, but excluding other non-bank financial institutions), is due to come into operation before the end of 2022.
- Interoperability is expected to promote competition, reduce fixed costs and enable economies of scale for payment services.

### Regulatory environment, supervision and risks
- BSL introduced a regulatory sandbox for fintech startups in 2017, supported by UNCDF, to encourage innovation and competition.
- In May 2018 four businesses were selected to enter the regulatory sandbox; two were granted licenses to operate (Mikashboks and InvestEd) after assessment.
- As of February 2022, fifteen sets of fintech products are in the sandbox, with five having been granted licenses to test their product in the financial system (Afro Mobile Money, Soft Pay, Pay Online, B ‘N’ B Cash and Stochastic).
- New regulatory work includes the National Payment Systems Act (NPSA, currently in Parliament), guidelines for mobile money, remittances regulations and agent banking guidelines; the NPSA will give BSL the power to regulate and oversee the National Payment System as a whole, including MNOs.
- Regulatory gaps remain in data management and data protection, consumer protection, and ensuring sufficient regulatory capacity and understanding of sector risks.
- Cybercrime is a growing problem; safeguarding of data and the safekeeping of customer funds require attention even if e-money deposits are relatively small compared with commercial banks.

### Credit infrastructure and digitalization gaps
- BSL launched a web-based Collateral Registry in June 2017 to improve access to finance for MSMEs; scope broadened to include immovable assets and non-incorporated entities.
- Registry enables collateral registration for inventories, farm products, timber, consumer goods, receivables, movable assets (including vehicles), machineries and immovable assets (land and buildings).
- Registry reforms need to be complemented by reforms to commercial court processes.
- Delays in obtaining credit reports from the Credit Reference Bureau (housed at BSL) persist because the bureau relies on an Excel database and manual report production.
- There remains scope for the BSL to improve credit infrastructure services through enhanced digitalization.

### Conclusion: opportunities and constraints
- Factors limiting digitalization progress: prohibitive cost of internet, low quality, legal and regulatory challenges, low private sector participation, lack of adequate IT skills, and supportive infrastructure gaps (fibre optic cable and electricity connectivity).
- Adoption of digitalization has picked up in government operations, providing a platform for efficient, transparent and effective PFM, improved service delivery in health, education and agriculture, and enhanced revenue mobilization.
- Health sector: mobile and drone technology helps reach previously excluded populations.
- Agriculture: e-vouchers ensure that input subsidy is well targeted.
- Digitalization presents numerous opportunities to improve allocation of scarce public resources and public service delivery.

### Policy recommendations (specific)
- Investment in key infrastructure and complementary social amenities:
  - Pursue universal access to electricity at affordable prices to enable broader connectivity and digitalization.
  - Accelerate completion of laying of the fibre optic cable to cut internet cost and raise connectivity speed.
  - Invest in roads, water, and housing to attract and retain technology and IT skills.
- Investment in human capital:
  - Fast-track connectivity for education institutions; only about 1.5 percent of approximately 10,387 schools are connected to the internet.
  - Prioritize public secondary schools and technical/vocational institutes given the low base.
  - Integrate information technology into the education curriculum from elementary to tertiary levels over the medium to long-term.
- Recalibrate the policy, legal and regulatory environment:
  - Review legislation to strengthen measures to limit cyber-attacks and cybercrime and to establish data protection.
  - Review regulatory regime to ensure competitive pricing, cyber security, data protection, and business continuity.
  - Address lack of supportive intellectual property rights cited by the private sector.
- Build e-business skills and support behavior change:
  - Enable public sector workers to acquire digital skills and garner support for change and digital adoption.
  - Complement automation of public service delivery and strengthened PFM controls with staff ethics and change management to avoid exemptions and reversion to manual processes.
  - Provide capacity building to instill e-business skills across all public servants to uphold systems' integrity.

*Italic: IMF staff summary extracted from the Sierra Leone country report chapter on digitalization and financial inclusion.*

### 33.      A wider adoption of DFS would increase the ease and speed of payments as well as

### 1sleea2022002 - 33. A wider adoption of DFS would increase the ease and speed of payments as well as

### Digital Financial Services (DFS): findings and policy recommendations
- A wider adoption of DFS would increase the ease and speed of payments as well as enhancing financial inclusion. However, due consideration also needs to be given to the regulation of these financial services.
- Improve data collection on DFS and make data publicly available.
  - One of the challenges faced under the previous strategy was a lack of data and measurement tools.
  - Improved data collection will be key to monitoring financial inclusion going forward, including data on DFS offered both by banks and non-banks.
  - Greater transparency would benefit investors and customers, and foster a more competitive DFS market.
- Leading through example.
  - Increasing the volume of government payments made using DFS could encourage broader acceptance of this medium, as opposed to cash.
- Address regulatory gaps.
  - Address regulations on consumer protection.
  - Address regulatory gaps on microfinance institutions to facilitate more partnerships with DFS and enhance the role of DFS in financial inclusion.
- Complete the National Switch project and upgrade supervision of payments systems.
  - The new NPSA, guidelines for mobile money, remittances regulations and agent banking guidelines are areas where authorities are working to improve the legal and regulatory environment for digital financial services.
  - The NPSA will give the BSL the power to regulate and oversee the National Payment System as a whole.
  - The BSL will need to upgrade its supervision of payments systems, including developing an effective oversight framework, an important priority supported currently by Fund CD.
- Improve staffing and capacity at the regulator.
  - The recent FSSR found that the Other Financial Institutions Supervision Department (OFSID) at the BSL, which is responsible for supervising MNOs among others, is severely under-resourced and that staff need capacity building in DFS.
  - The UNCDF (2021) noted that results from a survey of financial service providers timelines for approval of DFS may be lengthy and non-transparent.
  - Beyond OFSID, staff in other departments, such as the Financial Stability, Monetary Policy, and Research Departments, will need to be aware of the potential impact of DFS on their areas of operation.
- Digitalize important credit-related services such as the Credit Reference Bureau.
  - This should remove one of the obstacles to obtaining access to credit in Sierra Leone.

### Debt dynamics: key facts and numeric indicators
- Total public debt stock has accelerated in the last five years, driven primarily by wider fiscal deficits amidst large macroeconomic shocks.
  - Debt-to-GDP ratio increased by 28.7 percentage points to 76.3 percent of non-iron ore GDP in 2020 (from about 47.6 percent in 2015).
- In the aftermath of the Ebola epidemic:
  - Real GDP growth contracted by some 21 percent in 2015, resulting in a steep jump in the debt to GDP ratio by 9 percentage points.
  - A terms of trade shock in 2016 due to collapse in the price of iron ore led to a further increase in the debt to GDP ratio by some 7 percentage points of GDP in 2016.
- Debt sustainability assessment:
  - Sierra Leone remains at high risk of debt distress with sustainability predicated on ambitious fiscal adjustment and continued reliance on concessional financing, largely grants.
  - Debt service obligations absorbed about 26.2 percent of revenue in 2020.
- External vs domestic debt composition (end 2020):
  - Foreign currency debt accounts for approximately 50.3 percent of non-iron ore GDP (or 65.1 percent of total public debt stock).
  - Among multilateral creditors, the largest share of debt is owed to the Fund and the World Bank.
- Domestic debt (end 2020):
  - Total domestic debt stock stood at Le.10,634bn (or 26 percent of GDP).
  - Marketable instruments account for the lion’s share; non-marketable instruments (mostly zero-coupon bonds) and pre-April 2018 domestic payment arrears account for the balance.
  - From holders’ perspective: commercial banks account for 57 percent of domestic debt; the Bank of Sierra Leone (BSL) and non-banks make up 19 percent.
  - Approximately 40 percent of banking sector assets are held in the form of government securities.
- Development financing needs vs budget:
  - MTNDP implementation cost is estimated at about 12 percent of non-iron ore GDP per year.
  - Development spending allocations averaged around 7.3 percent of non-iron ore GDP over 2015-20.
- Financing mix and interest burden:
  - Shift toward increased reliance on relatively expensive domestic debt.
  - 99 percent of government securities are 364 days T-bills.
  - Demand for medium to long-term bonds is very weak.
  - Interest payments to revenue and effective interest rates are trending up.
- Macroeconomic volatility:
  - Sierra Leone has historically faced large macroeconomic volatility and higher fiscal shocks compared to peers and LICs average (selected comparators include Burkina Faso, Central African Republic, Chad, Guinea, and Rwanda).
  - Volatility is large for changes in terms of trade (ToT), changes in real effective exchange rate (REER), and real GDP growth.
- Near-term outlook risks:
  - Medium-term growth prospects are subject to increased uncertainty due to vulnerability to commodity price fluctuations and the COVID-19 pandemic.
  - Commodities exporters, including Sierra Leone, are exposed to ToT shocks and loss in mining revenue receipts.
  - Domestic financial markets remain underdeveloped; scope for raising more funds from commercial banks is narrowing.

### Medium-term fiscal framework and debt anchor: approach and recommendations
- Objective:
  - Develop a medium-term fiscal anchor that can reduce the debt-to-GDP ratio towards an anchor of 51 percent of non-iron ore GDP in 12 years while maintaining a safety buffer of about 19 percent so that Sierra Leone’s maximum debt ceiling of 70 percent of GDP is not exceeded in all circumstances.
- Rationale:
  - The credibility of this framework depends on ongoing reforms to raise potential output, enhance revenue mobilization, and improved expenditure and commitment controls.
- Context:
  - MTNDP 2019-2023 commits to investing in human and physical capital and maintaining public debt at or below 70 percent of GDP in nominal terms.
  - Debt rose to 76.3 percent of non-iron ore GDP in 2020, breaching the 70 percent ceiling.
- Determining a maximum debt limit and a debt anchor:
  - The difference between the maximum debt limit and the safety buffer yields a debt anchor that provides a medium-term fiscal path.
  - The debt anchor should balance economic costs of excessive debt and development financing needs, allowing policymakers time to take corrective actions.
  - The paper uses the Fund’s “How to Calibrate Fiscal Rules note” framework and runs stochastic simulations to calibrate a debt anchor that provides a safety buffer below a known maximum debt limit.
- Approaches used to determine the maximum debt limit:
  - (i) Prudential approach
  - (ii) Growth maximizing approach
  - (iii) Bank-IMF’s debt sustainability framework for LICs
  - The government’s maximum debt limit of 70 percent of GDP was found appropriate and within ranges proposed by these methods.
- Prudential approach note:
  - Estimating a debt anchor using a prudential approach places weight on economic costs and risks of bearing excessive debt.
  - Several prudential approaches can be used; Box 1 (in source) considers two applicable to Sierra Leone, a country with relatively nascent international market access.

*Source: IMF staff report (content from the provided PDF chapter).*

### Box 1. Prudential Based Approaches

### Box 1. Prudential Based Approaches

### Stabilizing debt with a feasible fiscal effort
- Maximum debt limit formula: d* = (maximum primary balance) / (interest-growth differential) as stated in the source.
- Policy rationale: Policymakers cannot realistically assume they can “do whatever it takes” to generate arbitrarily large primary surpluses during severe distress because tax increases may be ineffective beyond certain levels and steep expenditure cuts may be politically/socially unacceptable.

### Maximum Primary Balance Surplus, 2005
- Highest observed primary surplus in Sierra Leone over 1990-2020: 1.44 percent of GDP in 2005.
- Debt to GDP in 2005: 130.8 percent.
- Interest-growth differentials under stress (literature-based reference values): David et al. 2022 finds 95th percentile for WAEMU around 3.3 percent; a broad range for emerging and developing countries under stress could be 2-3.3 percent.
- Resulting implied debt limits from this approach: between 44 to 72 percent of GDP.

### Debt maximum limit based on debt carrying capacity (interest expenses to tax revenues)
- Tax revenue to GDP for Sierra Leone (2022-28): 13.8 percent.
- Effective interest rate used: 3.2 percent.
- Implied debt limit from this tax/revenue approach: between 69 and 82 percent of GDP.
- Empirical note: Hadzi-Vaskov and Ricci (2019) find that a 10 percentage points increase of public debt to GDP is associated with 100-120 bps higher spreads across all emerging markets; lack of sovereign spreads for SLE limits direct use.

### (ii) Debt level that maximizes growth (bottom-up approach)
- Theoretical basis: Checherita-Westphal (2014) model and empirical regularity of non-linear relation between per capita income growth and annual changes in public debt to GDP.
- Estimated output elasticity of public capital stock examples:
  - Euro area: 0.23
  - OECD: 0.28
  - WAEMU (David et al. 2022): about 0.30
  - Panel estimates for LICs/MICs (incl. Sierra Leone) over 1990-2015: alpha estimated at 0.25 (used for policy calculation)
  - Range of alpha found in the paper: 0.16 for LICs to 0.30 for high-income countries
- With alpha = 0.25:
  - Growth-maximizing (optimal) debt for LICs/MICs: about 55 percent of GDP.
  - Implied optimal debt anchor for HICs: about 72 percent of GDP.
- Text Table 1 (Country FE model / Optimal debt anchor / Number of countries):
  - α: 0.16 / 0.25 / 0.30
  - Optimal debt anchor: 28 / 55 / 72
  - Number of countries: 22 / 65 / 1

### (iii) Thresholds inferred from the latest Sierra Leone Debt Sustainability Analysis (DSA)
- LICs-DSF PV total public debt benchmark for Sierra Leone (medium debt carrying capacity): 55 percent of GDP.
- Ratio of nominal debt to PV debt in 2020 for Sierra Leone: about 1.25 (sustained over the 10-year horizon).
- Implied maximum nominal debt ceiling from DSA benchmark: about 69 percent of GDP.
- DSA note: Sierra Leone’s debt carrying capacity has been borderline between weak and moderate in recent reviews; a downgrade to weak is possible with multiple reviews and would affect maximum debt limits.

### Consolidated assessment of maximum debt limit
- Combining methods: a maximum debt limit of 70 percent of GDP appears credible for Sierra Leone.
- Average across three methods suggests Sierra Leone’s stated maximum debt limit equal or less than 70 percent of GDP in nominal terms is appropriate.
- Comparative note: David et al. 2022 finds a maximum debt limit of about 80 percent of GDP for the WAEMU countries.

### Calibrating the Debt Anchor for Sierra Leone
- Method: estimate distribution of macroeconomic and fiscal shocks from a multivariate normal distribution using annual data on terms of trade gap, maximum primary balance, ratio of foreign debt to total debt, ratio of concessional debt to total foreign debt, real effective interest rates, contingent liabilities, and stock-flow adjustments (SFA). Use shocks to simulate debt trajectories via standard debt dynamic equation and fiscal reaction function (Baum et al. 2018).
- Contingent liabilities assumption: estimated at 3 percent of GDP every 6 years (Bova et al, 2019; David et al. 2022).
- Safety buffer concept: determined by stochastic debt distribution and policymaker’s willingness to accept probability of breaching the maximum debt limit.
- Scenario example (from text):
  - Under a 10 percent risk tolerance and accounting for contingent liability of 3 percent of non-iron ore GDP, the safety buffer required to avoid breaching a maximum nominal debt limit of 70 percent of GDP would be about 33 percent of non-iron ore GDP, suggesting a debt anchor of 37 percent of non-iron ore GDP.
- Alternative presented result:
  - The paper also reports that a safety buffer of 18.9 percent of GDP (accounting for historical shocks and contingent liabilities) implies a debt anchor of 51.1 percent of GDP (difference between maximum debt limit 70 percent and safety buffer 18.9 percent).
  - Maintaining this safety buffer of about 18.9 percent of GDP would allow Sierra Leone to stay below the maximum debt limit of 70 percent of GDP with a 90 percent probability given typical historical shocks.

### Sensitivity analysis (Text Table 4: Scenarios 1–5)
- Parameters (Maximum debt limit / Contingent liabilities / Likelihood exceeding debt limit):
  - Scenario 1: 70.0 / 5.0 / 10.0
  - Scenario 2: 70.0 / 3.0 / 10.0
  - Scenario 3: 70.0 / 3.0 / 20.0
  - Scenario 4: 70.0 / 3.0 / 15.0
  - Scenario 5: 70.0 / 3.0 / 30.0
- Calibrated results (Debt anchor (initial) / Median debt anchor at year 6 / Range of the fan chart):
  - Scenario 1: Debt anchor (initial) 35.2; Median at year 6 47.9; P75-P25 13.6; P95-P5 20.0; P80-P20 16.7
  - Scenario 2: Debt anchor (initial) 37.0; Median at year 6 47.9; P75-P25 14.0; P95-P5 23.0; P80-P20 17.4
  - Scenario 3: Debt anchor (initial) 45.4; Median at year 6 55.8; P75-P25 14.3; P95-P5 23.2; P80-P20 17.8
  - Scenario 4: Debt anchor (initial) 40.6; Median at year 6 51.8; P75-P25 14.5; P95-P5 23.8; P80-P20 17.9
  - Scenario 5: Debt anchor (initial) 51.1; Median at year 6 60.4; P75-P25 15.5; P95-P5 35.8; P80-P20 18.8
- Overall range across scenarios for initial debt anchor during the first year: between 35.2 to 51.1 percent of GDP.
- Median debt anchor at year 6 across scenarios ranges between 47.9 to 60.4 percent of GDP.

### Calibrating the Primary Balance
- Determinants of the primary balance needed to converge to the debt anchor of 51.1 percent of non-iron ore GDP:
  - Macroeconomic fundamentals: higher real growth allows smaller primary balance (larger overall fiscal deficit) to still guide debt toward the anchor; large exchange rate depreciation requires larger primary balance.
  - Financing mix: higher reliance on domestic borrowing requires larger primary balance because domestic debt is relatively expensive; external debt is more concessional with substantial grant element (at least 35 percent or more).
  - Gap between current debt level and medium-term debt anchor: larger current debt requires larger fiscal consolidation.
  - Desired pace of adjustment: shorter convergence period requires larger fiscal effort. For Sierra Leone, convergence period set at 12 years.
- Convergence parameters and baseline:
  - Baseline macro forecast aligned to April 2022 WEO; current debt ratio, primary balance, and overall balance values in percent of non-iron ore GDP taken as at 2021; long-term real interest rate and real GDP growth are averages over medium-term range (2022-26); convergence period set at 12 years.
- Three paths consistent with attaining medium-term debt anchor of 51.1 percent of GDP by 2033:
  - Scenario 1 (fast adjustment):
    - Primary balance improves from about -4.1 percent of GDP in 2021 to 1.3 percent in 2023.
    - Overall deficit improves from about 6.9 percent of non-iron ore GDP in 2021 to about 0.6 percent in 2023.
    - Debt is expected to decrease to hit the medium-term debt anchor by 2033.

*Source: IMF staff estimates and text from Box 1. Prudential Based Approaches in the Sierra Leone chapter.*

### 0.6 percent in 2023 (Figure 22) and

### 1sleea2022002 - 0.6 percent in 2023 (Figure 22) and

### Calibrations and Scenario Summaries
- Three scenarios are calibrated to converge public debt to the MT debt anchor (around 51.1 percent of GDP) in 12 years.
- Scenario 1 (fast adjustment): debt is expected to decrease to hit the MT debt anchor by 2033 (Figure 23).
- Scenario 2 (slow adjustment within five-year transition period):
  - Calibrated primary balance ranges from -2.9 to +0.9 percent of non-iron ore GDP in 2022-25.
  - Primary balance then set at +2.1 percent of non-iron ore GDP in the long-term (2026-33).
  - Implied long term overall balance is about 0.1 percent of non-iron ore GDP (2026-33) (Figure 22).
  - Debt should converge to its MT debt anchor in 12 years (Figure 23).
  - Scenario 2 is described as the most realistic and is consistent with the WEO projection and the latest DSA.
- Scenario 3 (slower adjustment within eight-year transition period):
  - Primary balance ranges from -4.1 to 2.2 percent of non-iron ore GDP (2022-28).
  - Primary balance set at 3.1 percent of non-iron ore GDP over the medium term (2029-33).
  - Implied long term overall balance is about 1.2 percent of non-iron ore GDP (2029-33) (Figure 22).
  - Debt should converge to its MT debt anchor in 12 years (Figure 23).

### Key Parameters and Calibration Inputs (Text Table 5)
- Current debt ratio (% GDP): 76.2%
- Target debt ratio (% GDP): 51.1%
- Long-term real interest rate (%): 3.3%
- Long-term real growth rate (2022-26) (%): 4.8%
- Fiscal balance (% GDP) in 2021: -4.1%
- Primary balance (pb): -6.9%
- Overall balance (ob): 12
- Convergence period for debt (years): 2
- Fast fiscal adjustment period (years): Slow fiscal adjustment period (years) 5 8
(Note: parameter table formatting in source lists these values in the sequence shown above.)

### Fiscal Adjustment Paths and Levels (Selected Figures and Text)
- The calibrated primary balance under scenario summaries:
  - Scenario 1: (implied) ranges from -4.1 to +0.9 percent of non-iron ore GDP in 2025 before being set at +2.1 percent of non-iron ore GDP in the long-term (2026-2033).
  - Scenario 2: ranges from -2.9 to +0.9 percent of non-iron ore GDP in 2022-25; set at +2.1 percent in the long-term.
  - Scenario 3: ranges from -4.1 to +2.2 percent of non-iron ore GDP (2022-28); set at 3.1 percent over 2029-33.
- Overall balance implications:
  - Scenario 2 implied long term overall balance: about 0.1 percent of non-iron ore GDP (2026-33).
  - Scenario 3 implied long term overall balance: about 1.2 percent of non-iron ore GDP (2029-33).

### Credibility, Risks, and Preconditions for Adjustment
- Credibility of the fiscal adjustment path depends on ongoing structural reforms to expand potential growth and revenue mobilization.
- The three-year average fiscal adjustment over 2022-25 reflects a structural break recovery from a sharp deterioration in the primary balance in 2020 and 2021 due to the COVID-19 shock, higher energy subsidies, and wage pressures in education, security and health.
- Projected fiscal adjustment from 2026 onwards is considered realistic and feasible because of:
  - Expected recovery in domestic revenues as full efficiency gains from ongoing automation and modernization reforms materialize.
  - Ongoing investment in agriculture and mining, as well as services, expected to raise potential output.
  - Gradual efficiency improvements including improved appraisal, selection, and execution of priority development projects.
- High risks to realization include still evolving COVID-19 that could stall progress in reforms and impact output in the mining sector.

### Supporting Public Financial Management (PFM) Reforms
- Effective PFM processes are central to implementation of the MT fiscal anchor.
- Multi-pronged action points include:
  - Revenue mobilization and securing more budget grants.
  - Strengthening expenditure and commitment controls.
  - A policy on no new domestic payment arrears.
  - Keeping new external loans explicitly concessional for debt sustainability.
  - Advancing structural reforms to raise growth and promote economic transformation and efficiency.
  - Reforms to bolster the domestic financial market, reduce the sovereign bank nexus, and de-risk government securities.
- Revenue mobilization specifics:
  - MTNDP pegs achievement of development goals on increased domestic revenue mobilization to at least 20 percent of GDP by 2023 (from about 13.8 percent in 2020).
  - A 2019 World Bank tax gap study finds Sierra Leone collects less than 70 percent of its tax potential, with a tax gap of about 4 percent of GDP.
  - Recommended actions: broaden the tax base through new taxpayer registration and strengthen governance of tax expenditure/exemptions; develop and implement an effective framework for administering exemptions; create a complete and updated register of taxpayers benefiting from concessions, exemptions and other investment incentives.
  - Development of a medium-term revenue strategy (MTRS) by fall of 2022 is expected to improve tax structure and laws.
  - Strengthening compliance and tax administration will benefit from automation: Integrated tax administration system (ITAS) and installation of electronic cash registers for GST liable establishments.
- PFM execution and systems:
  - Sierra Leone is in its 5th year of implementing the PFM reform strategy 2018-2021, including budget execution, reporting, monitoring and evaluation pillars.
  - IFMIS coverage extended to some 61 MDAs; testing protocols for a web based IFMIS underway.
  - Authorities exploring top-down strategic budgeting to align expenditure commitments with available resources to enhance budget credibility.
- Domestic financial market deepening:
  - Financing mix shifting towards increased reliance on relatively expensive domestic debt, increasing interest bill and annual rollover risks because domestic debt is contracted at much higher interest rates and short tenors.
  - Reforms to extend tenure and diversify investor base are ongoing with Fund MCM TA but are medium to long term.
  - Possible domestic debt treatments include fiscally sustainable securitization of legacy arrears, 364 days T-bill sequester operations to restructure, and issuing other market instruments to extend the yield curve (as suggested in the 2019 MCM Technical report).

### Conclusions and Policy Recommendations (Section G)
- The debt anchor set at around 51.1 percent of GDP is deemed appropriate and achievable within 12 years, predicated on:
  - Expected recovery on growth.
  - Revenue mobilization.
  - Improved expenditure and commitment controls.
  - Other PFM improvements.
- Scenario 2 is identified as the most realistic medium-term fiscal policy guide post the ECF:
  - Calibrated primary balance ranges from -2.9 to +0.9 percent of GDP in 2022-2025; set at +2.1 percent of GDP in the long-term.
  - Consistent with WEO projection and the latest DSA.
- Main policy recommendations:
  - Fast track ongoing structural reforms to raise potential growth through effective implementation of projects in agriculture, human and physical capital; leverage digitalization of government operations for efficiency and transparency in budget execution.
  - Revenue mobilization: exploit identified tax revenue potential; a 2019 World Bank tax gap analysis indicates Sierra Leone collects less than 70 percent of its tax potential (4 percent of GDP in potential revenue); MTRS under development should sequence and guide tax policy and administrative measures.
  - Improve expenditure controls and spending efficiency to create fiscal space for priority spending; maintain spending in health and social support and continued investments in human and physical capital; reform energy sector and SOEs to contain subsidy and transfer spending.
  - Advance reforms to develop and deepen domestic financial markets to de-risk government securities and reduce banking system exposure; extend maturity tenure of government bonds and diversify investor base to reduce rollover risks.

*Source: IMF staff estimates and computations as presented in the provided content.*

### 17.7 percent annually since 1973) was 2½ times higher than both EMDE (7.2 percent) and SSA

### 1sleea2022002 - 17.7 percent annually since 1973) was 2½ times higher than both EMDE (7.2 percent) and SSA

### Historical inflation patterns
- Sierra Leone’s historical inflation is characterized by two waves:
  - Great inflation (1984-93):
    - Inflation nearly quadrupled to 56.2 percent (from an average of 15.7 percent in 1971-83).
    - Sierra Leone’s 1984-93 inflation placed it at the 92nd percentile of the SSA group—next only to DRC (134.7 percent), Zimbabwe (60.2 percent), and Uganda (58.8 percent).
  - Great disinflation (1996-2020):
    - Inflation receded to average 11.7 percent over this period.
    - SSA’s average inflation fell from 18.3 percent to 10.7 percent over the disinflation period.
- Sierra Leone’s NEER depreciation episodes tended to be more abrupt and disorderly compared with EMDE and SSA peers, where exchange rate adjustments have been more continuous.
- Periods of persistent depreciations in Sierra Leone closely coincide with steeper monetary expansions than in EMDE and SSA.

### Monetary policy framework and institutional features
- The Bank of Sierra Leone (BSL):
  - Price stability is the primary objective, but de jure and de facto objectives include maintaining financial stability, meeting fiscal financing needs, mitigating exchange rate volatility, achieving reserve targets, and financial inclusion.
  - Operational independence exists under legal framework but is constrained by expectations that it will help government objectives.
- Evolution of monetary policy:
  - BSL moved away from money targeting; introduced a monetary policy rate (MPR) in 2011.
  - Since 2016 BSL maintained a corridor around the MPR with Standing Deposit Facility and Standing Lending Facility; MPR set quarterly by the Monetary Policy Committee.
  - Monetary aggregates remain relevant under the IMF ECF-supported program, which includes a floor on net domestic assets and a ceiling on net foreign assets.
- Exchange rate regime:
  - According to Ilzetzki et al (2017) classification, Sierra Leone’s de facto exchange rate regime shifted from “flexible” to “fixed” in 2003 (crawling band narrower than +/-2 percent), helping stabilize inflation expectations.

### Cross-country panel regression (long-run drivers)
- Empirical setup:
  - Panel covers all 130 EMDEs from WEO for 1970–2020; fixed-effect panel regression specified in equation (1).
  - Regressors include CPI, M3, NEER, DebtGdp, Energynet, Foodnet, TradeOpen, IT (inflation targeting dummy), XRregime (de facto exchange rate regime dummy), CPagriculture; Dum_break (structural break dummy, 1 if t>93).
- Main long-run results for Sierra Leone:
  - M3 (broad money) accounts for 63 percent of Sierra Leone’s long-term average inflation.
  - Net credit to government (NCG) accounts for 85 percent of Sierra Leone’s M3 growth during 2002-2021.
  - Components of M3 growth during 2002-2021: NCG 85 percent; private sector credit (PSC) 21 percent; Net Foreign Assets (NFA)+Other Items Net (OIN) -5 percent.
  - NFA+OIN contribution: 60 percent of M3 growth until 2009; turned to -13 percent since 2010.
  - NEER accounts for about a third of long run inflation.
  - Deterministic drift (country fixed effect) explains about one quarter of historical inflation.
  - Estimated coefficient of IT dummy is -0.026 (interpretation: adoption of IT permanently lowers inflation by 2.6 percentage points).
  - Model caveats: possible endogeneity bias between CPI and NEER/M3; model chosen without structural break despite rolling-window evidence because structural-break specifications fit Sierra Leone poorly.

### Decomposition of the two secular episodes
- Great inflation (1984-93):
  - Annual average inflation rose by 40.6 percentage points from the 1971-83 average.
  - Contributions to the rise:
    - NEER depreciation: +47 percent.
    - M3 expansion shocks: +34 percent.
  - Frequent devaluations and monetary accommodation produced destabilizing inflation-exchange rate feedback loops.
- Great disinflation (1994-2020):
  - Inflation declined from 56.2 percent (1984-93, period average) to 11.7 percent (period average).
  - Largest contributors to disinflation:
    - NEER: 37 percent.
    - M3: 30 percent.
    - Exchange rate regime dummy (XR_regime): 16 percent.
    - Unexplained residual: 15 percent.
  - Sierra Leone decelerated M3 growth more forcefully than SSALIC peers and introduced more exchange rate stability from 2003 onwards.

### Short-run inflation dynamics and ARDL results
- Motivation:
  - Short-run dynamics examined because recent uptick in inflation is a central macroeconomic concern.
  - Data limitations prevent use of Phillips curve and mark-up models (lack of unemployment, output gap series).
- ARDL specification:
  - Dependent: CPI (quarterly, seasonally adjusted, logs).
  - Explanatory variables: reserve money (FASMB), nominal exchange rate (ENDA), government spending (GOS), terms of trade (TOT). Optimal lag length selected as one.
  - Bounds test: do not reject null of no cointegration; ARDL method used to model short-run dynamics.
- ARDL / short-run regression findings (Table 1; dataset 2009Q1–2021Q2, 50 observations):
  - Persistence dominates: previous period’s CPI is the main short-term correlate of current CPI.
    - Levels model: coefficient on lagged CPI = 0.95 (standard error 0.03) — statistically significant.
    - First-difference models and short-run OLS similarly show strong persistence (coefficients 0.54 and 0.57 with standard errors 0.11 and 0.12 respectively).
  - Reserve money (FASMB) coefficients small and not robustly different from zero across specifications (e.g., Levels: 0.02, s.e. 0.02).
  - Nominal exchange rate (ENDA) shows positive association in Levels (0.08, s.e. 0.02) but weaker in first-difference specifications.
  - Terms of trade (TOT) has small positive contemporaneous association in Levels (0.04, s.e. 0.02) and first-difference (0.06, s.e. 0.02).
  - Model fit and diagnostics:
    - Levels Adj. R-squared = 1.00; First-Difference Adj. R-squared = 0.50; Short-run OLS Adj. R-squared = 0.70.
    - Durbin-Watson statistics: 1.6 (Levels), 2.0 (First-Diff), 2.3 (Short-run OLS).
    - Akaike info criterion: -6.4 (Levels), -6.3 (First-Diff), -6.6 (Short-run OLS).
  - Interpretation: associations, not causal estimates; one percent increase in previous quarter’s CPI is associated with a (model-implied) increase in current inflation (persistence effect).

### Additional empirical and interpretive notes
- Rolling-window analysis suggests the relationship between CPI, M3 and NEER may have changed since mid-1990s for the EMDE group, but structural-break specifications fit Sierra Leone poorly, implying Sierra Leone is an outlier from the trend of structural changes.
- Short-run regressions including crude oil prices, government financing needs, and government spending were not statistically significant at level or lag values.
- The study emphasizes data limitations and potential identification issues at high frequency; results from ARDL and other short-run specifications are suggestive rather than definitive.

*Source: IMF staff estimates based on regression (1), Sierra Leone chapter (IMF).*

### 0.54 percent increase in present quarter CPI, holding all other factors constant, consistent with

### 1sleea2022002 - 0.54 percent increase in present quarter CPI, holding all other factors constant, consistent with

### Key short-run findings (ARDL and SVAR analyses)
- A 0.54 percent increase in present quarter CPI is consistent with inflationary expectations (measured by previous periods inflation).
- No contemporaneous impact of the monetary base on short-run inflation; monetary base is significant at lagged value with a small point estimate.
- Depreciation in the exchange rate is associated with an increase in the CPI, but results are not statistically significant.
- The sign on terms of trade is not correct in reduced-form regressions, suggesting potential omitted variable bias.
- Exogenous (domestic) price shocks are an important driver of short-run inflation dynamics (e.g., reduced seasonal supply of domestic food items).

### Structural VAR (SVAR) specification and identification
- Model estimated over 2009-2021 using complete quarterly data.
- Variables: all enter in natural logs, seasonally adjusted and in first difference (most are I (1)); government spending is an exception (trend stationary).
- Lag length selected: three (Akaike information criteria).
- Structural shocks identified (four):
  - Own price shock (captures domestic supply disruptions such as delayed harvests).
  - Monetary policy shock (tightening).
  - Commodity terms of trade (ToT) shock.
  - Nominal exchange rate shock.
- Short-run identification restrictions:
  - Domestic prices respond contemporaneously to terms of trade shock but not to reserve money or exchange rate shocks.
  - Terms of trade respond contemporaneously only to its own shock (exogenous for a small open economy).
  - Reserve money can respond contemporaneously to terms of trade and domestic price shocks, but not to exchange rate shocks.
  - Exchange rate can respond contemporaneously to all other variables.
- Model diagnostics:
  - Stable with all roots of the characteristic AR polynomials inside the unit circle.
  - Residuals well behaved (no serial correlation and are normally distributed).

### Impulse Response Function (IRF) results (inflation responses to one standard deviation shocks)
- Interpretation: variables are in logs, so deviations from baseline can be interpreted as percent deviations.
- Price shock:
  - Immediate increase in CPI by about 13.6 percent in the first quarter.
  - 4.9 percent increase in the second quarter.
  - Returns to baseline thereafter.
- Monetary policy shock (tightening):
  - Decrease in inflation by about 2.3 percent in quarter two.
  - Decrease by about 3.0 percent in quarter three.
  - Returns to baseline by quarter four.
- Terms of trade shock (deterioration):
  - No impact on CPI in the first quarter.
  - Associated with a 1.7 percent increase in CPI in the second quarter.
  - Returns to baseline by quarter three.
- Exchange rate shock (depreciation):
  - No impact on CPI in the first quarter.
  - Associated with increase in CPI in quarter two by 0.9 percent.
  - Peaks at 3.3 percent in quarter three.
  - Returns to baseline by the fifth quarter.
- Blue dashed lines in IRF plots indicate 95 percent confidence interval.

### Variance decomposition of CPI inflation (percentage points)
- Own price shocks:
  - Account fully for the variation of CPI in the first quarter and explain dynamics in subsequent quarters.
- Monetary policy shocks:
  - No contemporaneous impact on CPI variation in the first quarter due to slow transmission.
  - Impact picks up in quarter four, explaining approximately 10.5 percentage points of the variation in CPI.
- Terms of trade and exchange rate shocks:
  - Negligible immediate impact on CPI variation but persistent lagged effects.
  - Explain about 4.2-4.2 percentage points of the variation in CPI in the fourth quarter.
  - Explain about 6.0 -6.2 percentage points in the 12th quarter.
- Table excerpted variance decomposition (Forecast S.E. and contributions):
  - 1-Quarter: Forecast S.E. 0.011; Price shock 100.0; Monetary policy shock 0.0; TOT shock 0.0; Exchange rate shock 0.0
  - 4th-Quarter: Forecast S.E. 0.018; Price shock 0.91; Monetary policy shock 0.54; TOT shock 4.2; Exchange rate shock 4.4
  - 8th-Quarter: Forecast S.E. 0.0277; Price shock 7.4; Monetary policy shock 11.2; TOT shock 5.6; Exchange rate shock 5.8
  - 12th-Quarter: Forecast S.E. 0.0276; Price shock 4.6; Monetary policy shock 11.4; TOT shock 6.0; Exchange rate shock 6.2

### Unit root test results (ADF, sample 2009Q1-2021Q2; critical value at 5%: tau = -3.50)
- CPI: Tau -2.91 — Non stationary
- Reserve money: Tau -3.27 — Non stationary
- Government spending: Tau -6.99 — Trend stationary
- Exchange rate: Tau -1.96 — Non stationary
- Terms of trade: Tau 0.82 — Non stationary
- Spot crude oil: Tau -2.79 — Non stationary

### Conclusions (historical context and main drivers)
- Two secular waves in Sierra Leone inflation history:
  - Great inflation from mid-1980s to mid-1990s.
  - Great disinflation from mid-1990s to present.
- Sierra Leone’s inflation and disinflation episodes have been more accentuated and long-term inflation higher and more volatile than peers.
- Structurally high inflation largely associated with persistently high monetary growth.
- Reactive and disruptive exchange rate adjustments frequently set inflation-depreciation spirals in motion; monetary accommodation has aggravated these spirals and de-anchored inflation expectations.
- Disinflation gains can be explained by an improved policy framework that minimized inflation-depreciation spirals.
- Since 2009, exogenous shocks (exchange rate depreciation, terms of trade deterioration, domestic price shocks) are important short-run drivers of inflation; impacts are typically lagged.
- Fiscal variables did not show a clear short-run relationship with inflation in the regressions, but credit to government from the banking system clearly drives money growth and crowds out private sector lending.

### Policy implications and recommendations
- Maintain control of money growth and avoid deficit monetization; the relationship between money and inflation holds in both short and long run.
- Recognize that a tightening of reserve money growth may not generate an immediate deflationary response; impacts tend to be felt one quarter later.
- Monetary policy should be forward-looking; deepen BSL’s analytical capacity and inflation forecasting (work ongoing with Fund TA).
- Improve monetary policy framework:
  - Establish a well-articulated medium-term inflation objective.
  - Clarify hierarchy of the BSL’s objectives and that price stability is the overriding objective.
  - Improve transparency and communication (timely publication of statistics, economic reviews, and monetary policy statements).
  - Improve monetary operations and liquidity management, including closer coordination with the Ministry of Finance.
- Exchange rate policy:
  - Consistency in exchange rate policy is important; market-determined regime has improved predictability and helped anchor inflation.
  - BSL may have a role to intervene to smooth volatility as it improves foreign exchange market functioning (including moving to a daily Leone/US$ exchange rate calculation), but weak external position constrains intervention capacity.
  - Increased reserve buffers and a better external position would provide more policy flexibility.
  - Enhance exchange rate flexibility and diversify sources of foreign exchange to avoid exchange rate pressure episodes that feed back to inflation.
- Reduce reliance on food imports and enhance agricultural productivity to reduce frequency and severity of domestic price shocks (but central bank should not lead these sectoral policies).
- Fiscal prudence and containing bank financing of the deficit are necessary to bring inflation down; government borrowing crowds out bank lending to private sector and undermines monetary policy transmission.

### Data summary (selected series, descriptive statistics)
- DLOG(CPI_SPL): Mean 0.114341; Max 6.484239; Min -1.299368; Std. Dev 363
- DLOG(M3) Broad money: Mean 0.175057; Max 4.837324; Min -6.597936; Std. Dev 474
- DLOG(NEERINS_IMP) Nominal effective exchange rate: Mean -0.06155; Max 2.245908; Min -8.224448; Std. Dev 338
- D(DEBTGDP_WEO_IMP) Public debt-to-GDP ratio (Δ): Mean 0.341676; Max 604.3264; Min -657.9196375
- D(ENERGYNET_SPL2010) net energy export (percent of GDP): Mean -0.00373; Max 328.8514; Min -373.4388015
- FOODNET_SPL2010 net food export (percent of GDP): Mean 0.385497; Max 156.5032; Min -92.98798371
- D(TRADE_OPEN): Mean 0.360795; Max 341.3552; Min -303.8718019
- IT inflation targeting dummy: Mean 0.072157108925
- DLOG(NEERINS_IMP)*XR_REGIME1 cross effect: Mean -0.01117; Max 1.727221; Min -3.701037651
- DLOG(CP_AGRICULTURE): Mean 0.024853; Max 0.492264; Min -0.214119555

### Box: New Consumer Price Index (summary)
- New CPI to be released at end-February 2022, referenced to December 2021=100.
- Weights based on 2018 Sierra Leone Integrated Household Survey (updating previous weights based on 2003 data).
- New series: 83 new items and 357 items retained from previous basket.
- Expanded data collection centers to include the Northwest region; coverage will include all four provinces plus Freetown/Western Area (data collection remains largely urban; rural coverage planned later).
- Revised weights show higher relative expenditure on communication and increased weight for restaurants and hotels to reflect consumption at informal eateries.

*Source: IMF staff computation (1sleea2022002).*

### References

### 1sleea2022002 - References

### Journal articles and book chapters
- Bangura, Morlai, Eugene Caulker, Sandy Pessima, 2012, “Exchange Rate Pass-Through to Inflation in Sierra Leone: A Structural Vector Autoregressive Approach”, Journal of Monetary and Economic Integration, 12(1), 93-123.
- Bangura, Morlai, Augustine Ngombu, Sandy Pessima and Isatu Kargbo, 2021, “Bank Lending Channel of Monetary Policy: Dynamic Panel Data Evidence from Sierra Leone”, Modern Economy, 12, 1035-1058.
- Blanchard, Olivier and Danny Quah, 1989, “The Dynamic Effects of Aggregate Demand and Supply Disturbances”, The American Economic Review, 79(4), 655-673.
- Bernanke B.S., and F.S. Mishkin, 1997, “Inflation Targeting: A New Framework for Monetary Policy?”, NBER Working Paper 5893.
- Ghosh A.R., Gulde A., Ostry J.D., and H.C. Worf, 1997, “Does Nominal Exchange Rate Regime Matter?”, NBER Working Paper 5874.
- Ha, Jongrim, Anna Ivanova, Peter Montiel and Peter Pedroni, 2019, “Inflation in Low-Income Countries”, World Bank Policy Research Working Paper 8934 (Washington: World Bank).
- Ilzetzki, Ethan, Carmen M. Reinhart, and Kenneth S. Rogoff, 2021, “Rethinking Exchange Rate Regimes,” in Gopinath, Helpman and Rogoff (eds.), Handbook of International Economics vol. 5.
- Jackson, Emerson, Edmund Tamuke and Mohamed Jabbie, 2019. “Disaggregated Short-Term Inflation Forecast (STIF) for Monetary Policy Decision in Sierra Leone”, in Financial Markets, Institutions and Risks, 3(4), 32-48.
- Kallon, Kelfala M., 1994, “An Econometric Analysis of Inflation in Sierra Leone”, Journal of African Economies, 3(2), 199-230.
- Kovanen, Arto, 2006, “Why Do Prices in Sierra Leone Change So Often? A Case Study Using Micro-Level Price Data”, IMF Working Paper 06/53 (Washington: International Monetary Fund).
- Pesaran, M.H. and Shin, Y. 1999, “An Autoregressive Distributed Lag Modelling Approach to Cointegration Analysis.” Econometrics and Economic Theory in the 20th Century: The Ragnar Frisch Centennial Symposium, Strom, S. (ed.) Cambridge University Press.
- Svensson L.E.O.,1997, “Inflation Forecast Targeting: Implementing and Monitoring Inflation Targets”, NBER Working Paper 5797 (https://www.nber.org/papers/w5797).

### IMF, World Bank, AERC, and policy briefs / working papers
- Danladi, Jonathan D., 2020, “International Commodity Prices and Inflation Dynamics in Sierra Leone”, AERC Policy Brief (Nairobi: AERC).
- Debelle G., Masson P.R., Savastano M., and S. Sharma, 1998, “Inflation Targeting as a Framework for Monetary Policy”, IMF Economic Issues.
- Gottschalk, Jan, Kadima Kalonji and Ken Miyajima, 2008, “Analyzing Determinants of Inflation When There are Data Limitations: The Case of Sierra Leone,” IMF Working Paper WP/08/271 (Washington: International Monetary Fund).
- Hooley, John, Lam Nguyen, Mika Saito, and Shrin Nikaein Towfighian, 2021, “Fiscal Dominance in Sub-Saharan Africa Revisited”, IMF Working Paper WP/21/17 (Washington: International Monetary Fund).
- International Monetary Fund, 2015. Evolving Monetary Policy Frameworks in Low-Income and Other Developing Countries (Washington: International Monetary Fund).
- ________________________, 2020, “Improving Monetary Policy Effectiveness in Sierra Leone”, Sierra Leone—Selected Issues (Washington: International Monetary Fund)
- _______________________, 2022, “Nigeria’s Inflation Persistence—Causes and Policy Response”, Nigeria—Selected Issues (Washington: International Monetary Fund).
- ____________________, 2022, Regional Economic Outlook. Sub-Saharan Africa; a new shock and little room to maneuver (Washington: International Monetary Fund).
- Korsu, Robert Dauda, 2014, “The Inflationary Effects of Fiscal Deficit in Sierra Leone: A Simulation Approach”, AERC Research Paper 290 (Nairobi: AERC).
- Ree J., 2020, “Nigeria—Strategy for a Monetary Policy Reset”, IMF, Nigeria Selected Issues, Country Report No. 2021/034 (https://www.imf.org/en/Publications/CR/Issues/2021/02/05/Nigeria-Selected-Issues-50065).
- Ree J., 2021, “Nigeria’s Inflation Persistence—Causes and Policy Response”, IMF, Nigeria Selected Issues, Country Report No. 22/24 (https://www.imf.org/-
/media/Files/Publications/CR/2022/English/1NGAEA2022002.ashx)
- World Bank, 2019, “Inflation in Emerging and Developing Economies, edited by Ha J., Kose, M.A., and Ohnsorge F. (https://www.worldbank.org/en/research/publication/inflation-in-emerging-and-developing-economies).

### Regional and country-focused studies on Sierra Leone and Nigeria
- Bangura, Morlai, Eugene Caulker, Sandy Pessima, 2012, “Exchange Rate Pass-Through to Inflation in Sierra Leone: A Structural Vector Autoregressive Approach”, Journal of Monetary and Economic Integration, 12(1), 93-123.
- Bangura, Morlai, Augustine Ngombu, Sandy Pessima and Isatu Kargbo, 2021, “Bank Lending Channel of Monetary Policy: Dynamic Panel Data Evidence from Sierra Leone”, Modern Economy, 12, 1035-1058.
- Danladi, Jonathan D., 2020, “International Commodity Prices and Inflation Dynamics in Sierra Leone”, AERC Policy Brief (Nairobi: AERC).
- Gottschalk, Jan, Kadima Kalonji and Ken Miyajima, 2008, “Analyzing Determinants of Inflation When There are Data Limitations: The Case of Sierra Leone,” IMF Working Paper WP/08/271 (Washington: International Monetary Fund).
- Jackson, Emerson, Edmund Tamuke and Mohamed Jabbie, 2019. “Disaggregated Short-Term Inflation Forecast (STIF) for Monetary Policy Decision in Sierra Leone”, in Financial Markets, Institutions and Risks, 3(4), 32-48.
- Kallon, Kelfala M., 1994, “An Econometric Analysis of Inflation in Sierra Leone”, Journal of African Economies, 3(2), 199-230.
- Kovanen, Arto, 2006, “Why Do Prices in Sierra Leone Change So Often? A Case Study Using Micro-Level Price Data”, IMF Working Paper 06/53 (Washington: International Monetary Fund).
- Korsu, Robert Dauda, 2014, “The Inflationary Effects of Fiscal Deficit in Sierra Leone: A Simulation Approach”, AERC Research Paper 290 (Nairobi: AERC).

*Source: 1sleea2022002 - References*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2022/english/1sleea2022002.pdf_