## SCALING UP PUBLIC INVESTMENT, GROWTH AND DEBT SUSTAINABILITY (_cr16391)

## Source details

**Canonical URL:** [SCALING UP PUBLIC INVESTMENT, GROWTH AND DEBT SUSTAINABILITY (_cr16391)](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2016/_cr16391.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2016/_cr16391.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2016/_cr16391.pdf.json)

---

### A. Introduction
- Burkina Faso faces significant infrastructure gaps; estimated need of 11 percent of GDP in infrastructure investment over 10 years to close the gap with the continent’s leading performer.
- Natural resource wealth (gold and other commodity exports) can accelerate development but scaling up public investment financed in part by external borrowing can generate substantial debt risk (historical precedent: 1970s oil-financed investment followed by 1980s debt crises).
- Central policy question: how to use resource revenues to support the economy while maintaining debt sustainability given factors such as commodity prices, absorptive capacity, rate of return and efficiency of projects, private sector response (crowding in/out), Dutch disease, financing availability, debt composition, and revenue/expenditure rules.
- Analytical approach: use a model to examine empirical evidence and trade-offs of scaling up public investment, focusing on macroeconomic effects, financing options, and impact of resource-sector shocks.

### B. Country Context and Key Statistics
- Growth and volatility:
  - Rolling five-year average real GDP growth rose from 3.0 to 5.5 percent between 1990 and 2015.
  - Rolling five-year standard deviation declined from 3.8 to 1.4 percent.
  - Trend growth rate expected to resume and improve at 6-7 percent over the medium term as stability returns, investment increases and energy supplies improve.
- Resource sector:
  - Gold mining accounted for 58 percent of exports and 14 percent of fiscal revenues in 2014; earlier peak of 75 percent of exports and 19 percent of fiscal revenues in 2011-12 when international prices were nearly 30 percent higher.
- Authorities’ investment objectives:
  - National Economic and Social Development Plan (PNDES) aims to raise public investment to about 17 percent of GDP annually until 2019 versus a baseline of 12.5-13 percent of GDP, implying a substantial projected increase in the debt-GDP ratio.
- Empirical estimates:
  - Literature estimates growth impact of public investment in the range of 0.2-0.5 percent onto the growth rate for each additional percentage point of public investment.
- Investment execution and absorptive capacity:
  - Investment budget execution rates fell to 67.3 and 75.5 percent in 2014 and 2015 respectively.
  - Over 2013-15, investment spending declined from 14 percent of GDP to 7.7 percent of GDP.
  - Only about 44 percent of total expenditure in the investment budget actually results in fixed capital formation due to budget classification issues.
  - Domestic arrears identified at about 1 percent of GDP linked to weak project execution and procurement.
- Identified bottlenecks: weak project selection and preparation; long and complex procurement; inefficient execution monitoring/reporting; excessive use of exceptional procedures (e.g., “deblocage de fonds”, single sourcing).
- Policy tool: authorities interested in a Public Investment Management Assessment (PIMA) to provide comprehensive recommendations.

### C. Principles and Determinants of Scaling-Up Impact
- Investment efficiency and absorptive capacity:
  - Each additional unit of investment spending may not equal an equivalent increase in productive public capital due to corruption, scarce technical capacity, absorptive capacity constraints, insufficient maintenance; depreciation assumed to increase proportionally when effective investment fails to maintain capital.
- Crowding in vs. crowding out:
  - Higher public capital can raise marginal product of private capital (crowding in), but use of domestic resources for public investment can crowd out private investment and consumption.
  - Tax increases or reductions in transfers to finance scaling-up lower consumption and welfare; failure to cover recurrent infrastructure costs with user fees exacerbates debt risk.
  - Short-term transition likely to see crowding out dominate; long run can see crowding in if projects are high quality and address bottlenecks.
- Fiscal strategy alternatives:
  - Conservative approach: moderate public investment targets; public debt increases less as share of GDP and may decline in longer term; no future fiscal adjustment implied.
  - Gradual approach: public debt rises to a medium-term peak then declines as public capital builds and non-resource output expands; limited fiscal adjustment required.
  - Aggressive approach: immediate large increase in public investment leads to bigger medium-term debt build-up and would require a painful fiscal adjustment to service accumulated debt.
- Natural resource revenue uncertainty:
  - Resource revenue is volatile and uncertain; a gradual scaling-up that anticipates some but not all future potential revenue is recommended to translate mining revenues into non-resource growth without compromising fiscal stability.
  - Aggressive approach unlikely to deliver better growth outcomes due to absorptive constraints and poses threats to debt sustainability.

### D. Model Simulations (DIGNAR) — Scenarios and Core Results
- Model framework:
  - Uses DIGNAR (Debt, Investment, Growth and Natural Resources) model combining Buffie et al. (2012) debt-sustainability framework and Berg et al. (2012) natural resource model.
  - Includes natural resource, traded and non-traded sectors; calibrated for investment efficiency variability, limited absorptive capacity, Dutch disease, and detailed fiscal buffers.
- Scenario definitions:
  - Scaling-up (PNDES/aggressive) scenario: authorities’ 2017 budget proposal raising capital spending to 17 percent of GDP, declining gradually to 15 percent of GDP by 2026 and remaining at 15 percent until 2035.
  - Baseline (gradual) scenario: investment reaches 12 percent of GDP in 2017, rises to 14 percent of GDP in 2026, then declines gradually towards 10 percent of GDP by 2036.
  - Investment efficiency parameter: baseline assumed 50 percent; PNDES/aggressive scenario assumes efficiency drops to 35 percent due to absorptive capacity constraints and decline in quality of spending.
- Key simulation outcomes:
  - Lower investment efficiency in the aggressive scenario produces relatively lower increments to the public capital stock, reducing productivity in traded and non-traded sectors and leading to lower long-run GDP growth than baseline.
  - Debt outcomes:
    - Under the aggressive PNDES scenario, debt/GDP ratio increases and breaches the WAEMU 70 percent limit sometime around 2030.
    - Baseline projects debt/GDP ratio stabilizing at around 50 percent of GDP by the end of the simulation period.
  - Welfare and consumption:
    - Private consumption ultimately declines by 7 percent relative to baseline under the aggressive scaling-up scenario as higher taxation is required to maintain fiscal sustainability.

### E. Scenario Variants and Sensitivity Findings
- Two variants analyzed:
  - Gold price drop variant:
    - Gold price assumed to drop by about 30 percent relative to the baseline.
    - Resource revenues fall from 15 to 10 percent of total revenues.
    - Debt/GDP ratio increases by around 12 percentage points.
    - Slight decline in GDP and in welfare due to increased indirect taxation to offset revenue shortfalls.
  - No tax increase variant:
    - Assumes no increase in taxation to finance the higher investment spending in the baseline (tax rate fixed).
    - Principal effect is a larger increase in the debt-to-GDP ratio than in the scaling-up case; public debt would reach almost 90 percent of GDP by 2030, well beyond the WAEMU criterion.
    - The increase in the debt-to-GDP ratio is a full 35 percentage points above the baseline value and 3 points above that of the scaling-up scenario assuming partial fiscal adjustment.
    - The increase in debt is accounted for by increases in both external non-concessional and domestic debt, underscoring the need for vigilance in debt management.
- Overall sensitivity conclusion:
  - Aggressive "big-push" public investment scaling-up faces significant absorption capacity constraints that reduce investment efficiency and public capital accumulation.
  - Model simulations for the aggressive scaling-up scenario implied by the PNDES are likely inconsistent with long-term fiscal and debt sustainability.
  - Even the moderate gradual scaling-up baseline faces non-negligible risks (e.g., commodity price shocks).
  - Strengthening institutions that shape planning, allocation, and implementation of public investments could close up to two-thirds of the public investment efficiency gap (IMF, 2015).

### Policy Recommendations on Investment Management
- Good investment budget management is critical for successful implementation of the PNDES.
- Recent measures to improve investment budget execution include a revised procurement code.
- A Public Investment Management Assessment (PIMA) exercise is to be launched to strengthen key institutions involved in public investment.
- Fiscal strategy recommendation: prefer gradual scaling-up that anticipates but does not fully rely on uncertain resource revenues, combined with institutional reforms to raise investment efficiency and absorptive capacity.

### The Wage Bill — Evolution, Risks, and Reform Options
- Recent evolution and key figures:
  - Government wage bill as a share of total tax revenue rose from 35.8 percent in 2013 to 50.4 percent in 2015.
  - This ratio exceeds the WAEMU convergence criterion of 35 percent.
  - Public wage bill represented 7.3 percent of GDP in 2015, up from 5.9 percent of GDP in 2013.
  - As of 2014, there were 58,074 permanent civil servants and 80,075 contractual staff.
  - The implementation of the new civil service code (Law 81) adopted in November 2015 permanently increased the wage bill by an estimated 0.4 percent of GDP in 2016.
- Short-term and medium-term projections and targets:
  - The wage bill is expected to represent 46 percent of tax revenues and 7.4 percent of GDP in 2016.
  - Authorities aim to bring the wage bill down to 38.3 percent of tax revenues and 7.2 percent of GDP in 2017, predicated on containing nominal wage bill growth to 6 percent in 2017.
  - Public employment growth has averaged 7 percent per year historically; achieving the 6 percent nominal wage bill growth target in 2017 implies a decrease in average nominal compensation of about 1 percent in 2017.
- Long-term trends and risks:
  - Public wage bill grew from 4.7 percent of GDP in 2000 to 7.4 percent of GDP in 2016.
  - Public employment rose from 0.9 to 1.7 percent of the working-age population over the same period.
  - Decomposition of the wage bill increase between 2000 and 2016: employment contributed +3.6 percent of GDP; average public sector compensation contributed -0.9 percent of GDP (mitigating effect).
  - By 2025, public employment to working-age population is expected to reach 2.3 percent, which would raise the wage bill to 10.1 percent of GDP if current wage levels are maintained.
- Relative compensation levels (as presented):
  - Government wages are about eight times larger than GDP per capita.
  - Government salaries are about 4% to 20% higher than average private sector salary.
  - Government compensation is 668% higher than GDP per capita (Table 1 entries as presented).
  - Government compensation is 90% higher than average household consumption expenditure (urban).

### Reform Options to Manage the Wage Bill
- Short-term employment measures:
  - Attrition measures (e.g., replacement rules such as one hire per two exits).
  - Ad hoc wage freezes or cuts.
  - Departure incentives.
  - Freeze career progressions.
  - Reduce staff hours.
  - Improve HR management.
- Short-term compensation measures:
  - Consolidate benefits with base pay.
  - Eliminate redundancy.
  - Rationalize benefits and sectoral overlap.
  - Link salaries with performance.
- Trade-offs and country-specific considerations:
  - Attrition measures can mitigate employment growth but may harm service delivery and future capacity in sectors like education and healthcare.
  - Successful reforms often combine attrition with elimination of redundant positions, stronger worker mobility, and retraining.
  - In Burkina Faso, public employment growth was most pronounced in secondary and higher education, primary education, and security services; growth was strongest among mid-level and managerial categories and contractual workers, who accounted for virtually all change in public employment between 2010 and 2014.
- Allowances and pay structure:
  - In Burkina Faso, allowances represent almost half of total public compensation.
  - Figure 4 shows allowances constitute 40 percent of the average total compensation in each employment category.
  - Adopting a unified pay scale would bring limited benefits in Burkina Faso because allowances are well-regulated and behave as an informal, proportional extension of the base salary.
- Performance-based pay:
  - Recommendation: use performance-based allowances to align public sector compensation with productivity, especially in education, health, and tax administration.
  - Design considerations: performance criteria should be set objectively and in advance; careful design is needed to ensure outputs reflect quality and are not subject to gaming.
  - Evidence: Philippines example where a World Bank survey found most civil servants had a positive opinion of performance pay.

### Short-Term Implementation Priorities for Wage Management
- Conduct job audits as a first step toward a medium-term employment strategy; use audits as a basis for a future functional review of public administration and to identify sectors where performance pay could be implemented or expanded.
- Continue reforms on payroll management and control:
  - Automate payroll payments and move towards non-cash payments.
  - Conduct an administrative census and allowances reviews at regular intervals to keep the payroll database up to date and minimize undue payments.

*Source: IMF staff analysis from the chapter “SCALING UP PUBLIC INVESTMENT, GROWTH AND DEBT SUSTAINABILITY” (November 30, 2016).*

### References _____________________________________________________________________________ 11

### SCALING UP PUBLIC INVESTMENT, GROWTH AND DEBT SUSTAINABILITY

### A. Introduction
- Burkina Faso faces significant infrastructure gaps; estimated need of 11 percent of GDP in infrastructure investment over 10 years to close the gap with the continent’s leading performer.
- Natural resource wealth (gold and other commodity exports) can accelerate development but scaling up public investment financed in part by external borrowing can generate substantial debt risk (historical precedent: 1970s oil-financed investment followed by 1980s debt crises).
- Central policy question: how to use resource revenues to support the economy while maintaining debt sustainability given factors such as commodity prices, absorptive capacity, rate of return and efficiency of projects, private sector response (crowding in/out), Dutch disease, financing availability, debt composition, and revenue/expenditure rules.
- Analytical approach: use a model to examine empirical evidence and trade-offs of scaling up public investment, focusing on macroeconomic effects, financing options, and impact of resource-sector shocks.

### B. Country Context
- Pre-2014/15 sociopolitical events: sustained growth with improved macro management, stronger institutions, increased aid, higher human and physical capital investment.
- Key statistics and facts:
  - Rolling five-year average real GDP growth rose from 3.0 to 5.5 percent between 1990 and 2015; rolling five-year standard deviation declined from 3.8 to 1.4 percent.
  - Trend growth rate expected to resume and improve at 6-7 percent over the medium term as stability returns, investment increases and energy supplies improve.
  - Gold mining accounted for 58 percent of exports and 14 percent of fiscal revenues in 2014; earlier peak of 75 percent of exports and 19 percent of fiscal revenues in 2011-12 when international prices were nearly 30 percent higher.
  - Authorities’ National Economic and Social Development Plan (PNDES) aims to raise public investment to about 17 percent of GDP annually until 2019 versus a baseline of 12.5-13 percent of GDP, implying a substantial projected increase in the debt-GDP ratio.
  - Literature estimates growth impact of public investment in the range of 0.2-0.5 percent onto the growth rate for each additional percentage point of public investment.
- Investment execution and absorptive capacity:
  - Investment budget execution rates fell to 67.3 and 75.5 percent in 2014 and 2015 respectively.
  - Over 2013-15, investment spending declined from 14 percent of GDP to 7.7 percent of GDP.
  - Only about 44 percent of total expenditure in the investment budget actually results in fixed capital formation due to budget classification issues.
  - Domestic arrears identified at about 1 percent of GDP linked to weak project execution and procurement.
- Identified bottlenecks: weak project selection and preparation, long and complex procurement, inefficient execution monitoring/reporting, excessive use of exceptional procedures (e.g., “deblocage de fonds”, single sourcing).
- Policy tool: authorities interested in a Public Investment Management Assessment (PIMA) to provide comprehensive recommendations.

### C. Impact of Public Investment Scaling-Up: Principles and Underlying Determinants
- Investment efficiency and absorptive capacity:
  - Each additional unit of investment spending may not equal an equivalent increase in productive public capital due to corruption, scarce technical capacity, absorptive capacity constraints, insufficient maintenance; depreciation assumed to increase proportionally when effective investment fails to maintain capital.
- Crowding in vs. crowding out:
  - Higher public capital can raise marginal product of private capital (crowding in), but use of domestic resources for public investment can crowd out private investment and consumption.
  - Tax increases or reductions in transfers to finance scaling-up lower consumption and welfare; failure to cover recurrent infrastructure costs with user fees exacerbates debt risk.
  - Short-term transition likely to see crowding out dominate; long run can see crowding in if projects are high quality and address bottlenecks.
- Fiscal strategy alternatives:
  - Conservative approach: moderate public investment targets; public debt increases less as share of GDP and may decline in longer term; no future fiscal adjustment implied.
  - Gradual approach: public debt rises to a medium-term peak then declines as public capital builds and non-resource output expands; limited fiscal adjustment required.
  - Aggressive approach: immediate large increase in public investment leads to bigger medium-term debt build-up and would require a painful fiscal adjustment to service accumulated debt.
- Natural resource revenue uncertainty:
  - Resource revenue is volatile and uncertain; a gradual scaling-up that anticipates some but not all future potential revenue is recommended to translate mining revenues into non-resource growth without compromising fiscal stability.
  - Aggressive approach unlikely to deliver better growth outcomes due to absorptive constraints and poses threats to debt sustainability.

### D. Model Simulations (DIGNAR)
- Model description:
  - Analysis uses DIGNAR (Debt, Investment, Growth and Natural Resources) model combining Buffie et al. (2012) debt-sustainability framework and Berg et al. (2012) natural resource model.
  - Framework includes natural resource, traded and non-traded sectors; calibrated for investment efficiency variability, limited absorptive capacity, Dutch disease, and detailed fiscal buffers.
  - Simulates macro outcomes of scaling up public investment financed by borrowing and resource revenues; previously used for Chad, Republic of Congo, Mozambique, Kazakhstan, and four fragile West African countries.
- Scenario definitions and assumptions:
  - Scaling-up (PNDES/aggressive) scenario: authorities’ 2017 budget proposal raising capital spending to 17 percent of GDP, declining gradually to 15 percent of GDP by 2026 and remaining at 15 percent until 2035.
  - Baseline (gradual) scenario: investment reaches 12 percent of GDP in 2017, rises to 14 percent of GDP in 2026, then declines gradually towards 10 percent of GDP by 2036.
  - Investment efficiency parameter: baseline assumed 50 percent; PNDES/aggressive scenario assumes efficiency drops to 35 percent due to absorptive capacity constraints and decline in quality of spending.
- Key simulation results:
  - Lower investment efficiency in the aggressive scenario produces relatively lower increments to the public capital stock, reducing productivity in traded and non-traded sectors and leading to lower long-run GDP growth than baseline.
  - Debt outcomes:
    - Under the aggressive PNDES scenario, debt/GDP ratio increases and breaches the WAEMU 70 percent limit sometime around 2030.
    - Baseline projects debt/GDP ratio stabilizing at around 50 percent of GDP by the end of the simulation period.
  - Welfare and consumption:
    - Private consumption ultimately declines by 7 percent relative to baseline under the aggressive scaling-up scenario as higher taxation is required to maintain fiscal sustainability.

### E. Sensitivity Analysis: Risks to the Baseline from Resource Revenue Volatility or Fiscal Policy
- (Section header present; detailed sensitivity results and numbers for this subsection are contained later in the source and are not included verbatim in the extracted content provided here.)

*Source: IMF staff analysis from the chapter “SCALING UP PUBLIC INVESTMENT, GROWTH AND DEBT SUSTAINABILITY” (November 30, 2016).*

### 16.      The analysis also includes 2 variants on the

### _cr16391 - 16.      The analysis also includes 2 variants on the

### Scenario variants and fiscal impacts
- Two variants on the baseline scenario are analyzed:
  - Gold price drop variant: gold price assumed to drop by about 30 percent relative to the baseline.
    - Resource revenues fall from 15 to 10 percent of total revenues.
    - Debt/GDP ratio increases by around 12 percentage points.
    - Slight decline in GDP and in welfare due to increased indirect taxation to offset revenue shortfalls.
  - No tax increase variant: assumes no increase in taxation to finance the higher investment spending in the baseline (tax rate fixed).
    - Principal effect is a larger increase in the debt-to-GDP ratio than in the scaling-up case; public debt would reach almost 90 percent of GDP by 2030, well beyond the WAEMU criterion.
    - In the standard model approach, consumption taxes are set to increase to close the fiscal financing gap from scaled-up investment spending not covered by additional borrowing. When the tax rate is fixed at its initial level, additional (non-concessional) borrowing is required.
    - The increase in the debt-to-GDP ratio is a full 35 percentage points above the baseline value and 3 points above that of the scaling-up scenario assuming partial fiscal adjustment.
    - The increase in debt is accounted for by increases in both external non-concessional and domestic debt, underscoring the need for vigilance in debt management.

### Key empirical and modeling findings
- Aggressive "big-push" public investment scaling-up faces significant absorption capacity constraints that reduce investment efficiency and public capital accumulation.
- Model simulations for the aggressive scaling-up scenario implied by the PNDES are likely inconsistent with long-term fiscal and debt sustainability.
- Even the moderate gradual scaling-up baseline faces non-negligible risks (e.g., commodity price shocks).
- Strengthening institutions that shape planning, allocation, and implementation of public investments could close up to two-thirds of the public investment efficiency gap (IMF, 2015).

### Policy recommendations on investment management
- Good investment budget management is critical for successful implementation of the PNDES.
- Recent measures to improve investment budget execution include a revised procurement code.
- A Public Investment Management Assessment (PIMA) exercise is to be launched to strengthen key institutions involved in public investment.

### The wage bill — recent evolution and challenges
- Rapid increase in government wage bill:
  - Government wage bill as a share of total tax revenue rose from 35.8 percent in 2013 to 50.4 percent in 2015.
  - This ratio exceeds the WAEMU convergence criterion of 35 percent.
  - Public wage bill represented 7.3 percent of GDP in 2015, up from 5.9 percent of GDP in 2013.
- Contributors to the wage bill increase:
  - Pressures to increase public sector wages and incorporation of contractual workers as civil servants following the new civil service code (Law 81) adopted in November 2015.
  - The implementation of the new code and sectoral wage agreements permanently increased the wage bill by an estimated 0.4 percent of GDP in 2016.
  - As of 2014, there were 58,074 permanent civil servants and 80,075 contractual staff.
- Short-term and medium-term projections and targets:
  - The wage bill is expected to represent 46 percent of tax revenues and 7.4 percent of GDP in 2016.
  - Authorities aim to bring the wage bill down to 38.3 percent of tax revenues and 7.2 percent of GDP in 2017, predicated on containing nominal wage bill growth to 6 percent in 2017.
  - Public employment growth has averaged 7 percent per year historically; achieving the 6 percent nominal wage bill growth target in 2017 implies a decrease in average nominal compensation of about 1 percent in 2017.
- Long-term trends and risks:
  - Public wage bill grew from 4.7 percent of GDP in 2000 to 7.4 percent of GDP in 2016.
  - Public employment rose from 0.9 to 1.7 percent of the working-age population over the same period.
  - Decomposition of the wage bill increase between 2000 and 2016: employment contributed +3.6 percent of GDP; average public sector compensation contributed -0.9 percent of GDP (mitigating effect).
  - By 2025, public employment to working-age population is expected to reach 2.3 percent, which would raise the wage bill to 10.1 percent of GDP if current wage levels are maintained.
- Relative compensation levels:
  - Government wages are about eight times larger than GDP per capita.
  - Government salaries are about 4% to 20% higher than average private sector salary.
  - Government compensation is 668% higher than GDP per capita (Table 1 entries as presented).
  - Government compensation is 90% higher than average household consumption expenditure (urban).

### Reform options for managing the wage bill
- Short-term employment measures:
  - Attrition measures (e.g., replacement rules such as one hire per two exits).
  - Ad hoc wage freezes or cuts.
  - Departure incentives.
  - Freeze career progressions.
  - Reduce staff hours.
  - Improve HR management.
- Short-term compensation measures:
  - Consolidate benefits with base pay.
  - Eliminate redundancy.
  - Rationalize benefits and sectoral overlap.
  - Link salaries with performance.
- Trade-offs and country-specific considerations:
  - Attrition measures can mitigate employment growth but may harm service delivery and future capacity in sectors like education and healthcare.
  - Successful reforms often combine attrition with elimination of redundant positions, stronger worker mobility, and retraining (examples include Ireland).
  - In Burkina Faso, public employment growth was most pronounced in secondary and higher education, primary education, and security services; growth was strongest among mid-level and managerial categories and contractual workers, who accounted for virtually all change in public employment between 2010 and 2014.

_Italic: Source — Excerpt from IMF staff report content unit _cr16391._

### 12.      The authorities should design a medium-term strategy to contain the wage bill,

### 12.      The authorities should design a medium-term strategy to contain the wage bill,

### Medium-term strategy to contain the wage bill
- Priority: ensure increases in public employment reflect actual service needs.
- Core elements:
  - Develop a medium-term plan of employment needs by ministry and by function.
  - Rationalize hiring and eliminate sectoral overlap (downsizing or merging overlapping departments and positions).
  - Identify areas where the private sector could take the lead instead of government.
  - Remove legal and practical impediments to worker reassignment and retraining to ease redeployment across departments and geographic areas.
  - Move towards a position-based employment system rather than a career-based system to allow greater flexibility in adjusting employment levels and composition and to mitigate pressure for wage increases as the current workforce ages.
- Implementation note: these reforms imply a significant coordination effort across government departments and require time to implement (e.g., recent reforms in Ghana and Côte d’Ivoire saw new employees often fail to receive wages for several months because of poor coordination across hiring, budgeting, and payroll departments, as described in IMF (2016b)).

### Compensation reforms: short-term vs. medium-term considerations
- Short-term tools:
  - Nominal wage freezes, or limiting wage increases below real GDP growth or inflation, can be effective in restraining wage pressures in the short-term.
  - Examples of countries where such measures were used: Côte d’Ivoire, Kenya, Jamaica, Portugal and Romania.
- Drawbacks of across-the-board nominal wage freezes:
  - Can affect morale and the government’s ability to attract and retain qualified workers.
  - Disrupt existing wage adjustment mechanisms and distort compensation structure.
  - In Côte d’Ivoire and Senegal, prolonged base wage freezes were partly offset by rising pay supplements and bonuses.
  - They imply substantial long-term costs and can postpone necessary structural adjustments.
- Recommendation: consider more structural reforms focused on achieving wage bill sustainability in the medium-term, given the importance of increasing publicly-provided services in sectors like health and education.

### Allowances, pay scale, and Burkina Faso specifics
- Allowances:
  - In Burkina Faso, allowances represent almost half of total public compensation.
  - The number and extent of allowances by employment category is explicitly defined in the law.
  - Figure 4 shows allowances constitute 40 percent of the average total compensation in each employment category.
  - Allowances can attract staff in areas or sectors with skill shortages but can be used to circumvent wage caps and reduce transparency.
- Unified pay scale:
  - Adopting a unified pay scale would bring limited benefits in Burkina Faso because allowances are well-regulated and behave as an informal, proportional extension of the base salary.
  - Recent reforms in other countries (Ghana, Zambia, Portugal) consolidated allowances into base pay, but upfront costs (e.g., Portugal and Ghana) and limited gains in Burkina Faso suggest this is not a priority reform at this point.

### Performance-based allowances and pay-for-performance
- Recommendation:
  - Use performance-based allowances to align public sector compensation with productivity, especially in education, health, and tax administration.
  - Link some allowances to individual performance, particularly for jobs with quantitative outputs (teachers, health care providers, tax administrators, public infrastructure).
- Design considerations:
  - Performance criteria should be set objectively and in advance.
  - Careful design is needed to ensure outputs reflect quality of public service provision and are not subject to gaming (see IMF, 2016b).
- Evidence:
  - Example: Philippines reforms where a World Bank survey found most civil servants had a positive opinion of performance pay.

### Suggested short-term policies
- Conduct job audits as a first step toward a medium-term employment strategy; use audits as a basis for a future functional review of public administration and to identify sectors where performance pay could be implemented or expanded.
- Continue reforms on payroll management and control:
  - Automate payroll payments and move towards non-cash payments.
  - Conduct an administrative census and allowances reviews at regular intervals to keep the payroll database up to date and minimize undue payments.

### Lessons from country cases

- Box 1 — Lessons from Wage Bill Reform in Côte d’Ivoire
  - Wage freeze experience:
    - A wage freeze for over 20 years did not stop the wage bill from growing and had several drawbacks:
      - Reduced purchasing power among civil servants.
      - Created inequalities through targeted wage increases benefiting certain categories and incoherent bonuses and allowances.
      - Postponed larger adjustments, resulting in social unrest and later large wage increases and recognition of significant wage arrears.
  - Institutional recommendation:
    - Centralize wage bill management rather than allocate human resources and financial management to separate ministries; decentralization led to inconsistent worker headcounts and non-integrated human resource processes.
  - Medium-term strategy:
    - In 2014, Côte d’Ivoire adopted a medium-term strategy for 2014–2020 to lower the wage bill from 44 to 35 percent of tax revenue.
    - The strategy features: cancelling wage arrears, reducing recruitment targets, increasing employee pension contributions, reinstating automatic career progression, introducing a revised wage grid, and providing a formal framework for wage bargaining.
  - Caveats:
    - Effective medium-term strategies require strong political will.
    - The new strategy sharply increased the wage bill in 2014–2015 while postponing reductions to the end of the decade and relies on optimistic projections (real GDP growth of 10 percent), increasing risks of fiscal slippages.
  - Table 1. Key Historical Trends
    - Wage Bill (percent of GDP): 2005 6.3    2015 7.4
      - Driven by: Higher employment; Wage increases
    - Government Employment (percent of working-age population): 2005 1.5    2015 2.0
      - Driven by: Education and Health; Security services

- Box 2 — Lessons from Wage Bill Reform in the Philippines
  - Performance pay design:
    - The Philippines introduced a performance-based bonus system in 2012 replacing uniform bonuses.
    - Preconditions: departments had to meet 90 percent of their agreed performance targets for the year and comply with governance guidelines.
  - Performance and departmental/individual joint ranking:
    - Departments ranked: “Best” (10 percent), “Very Good” (25 percent), “Good” (65 percent).
    - Employees ranked into three categories; final bonus depends on joint department and individual ranking.
    - Table 1. Performance-Based Bonus Calculation (as a share of base bonus payment)
      - Department ranking headers: Best Department (10 percent), Very Good Department (25 percent), Good Department (65 percent)
      - Row: Best Performer (10 percent) — 7x (Best Dept), 4x (Very Good Dept), 2x (Good Dept)
      - Row: Very Good Performer (25 percent) — 5x (Best Dept), 2.7x (Very Good Dept), 1.4x (Good Dept)
      - Row: Good Performer (65 percent) — 3x (Best Dept), 2x (Very Good Dept), 1x (Good Dept)
  - Outcomes and limits:
    - Performance pay is well-accepted where jobs have measurable outputs.
    - Issues: subjectivity in ratings and potential for gaming (e.g., splitting advisories to inflate counts).
    - More effective and accepted in sectors with easily measured and independently verified outputs (teachers, health workers, infrastructure projects).

*Source: IMF staff analysis as presented in the original chapter.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2016/_cr16391.pdf_