## Wage Bill Conditionality in PRGF Countries, 2003–05

## Source details

**Canonical URL:** [Wage Bill Conditionality in PRGF Countries, 2003–05](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06106.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06106.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06106.pdf.json)

---

### Introduction: context and purpose
- Meeting the Millennium Development Goals (MDGs) requires both more government spending and better policies.
- Ethiopia example (Mattina’s (2006) estimate):
  - Total cost of achieving the MDGs amounts to almost six times the country’s current annual GDP.
  - Roughly three-quarters of this would have to come in the form of additional public spending, financed by foreign aid.
  - Per capita aid example: scale up from around US$11 in 2005 to over US$65 by 2015.
- Human resource shortages:
  - Meeting the MDGs in Sub-Saharan Africa may require a tripling of the health-sector labor force, or adding more than one million workers over time (Berg and Qureshi, 2005).
- Central question:
  - Have wage bill ceilings in IMF-supported (PRGF) programs been detrimental to hiring additional personnel in key social sectors when additional donor resources were available?
- Main empirical conclusion:
  - The paper did not find that wage bill ceilings in IMF-supported programs have prevented expanding public employment in priority sectors when donor funds were made available.
- Recommended complementary actions emphasized:
  - More transparent accommodation of additional donor funds for wage spending within wage bill ceilings.
  - Donors should provide clear indications about their plans to provide aid, including over the medium term.
  - Recipient governments should strengthen policy dialogue between sectoral ministries and macroeconomic policy agencies to address macroeconomic and sectoral absorption constraints and increase the flexibility and efficiency of wage spending.

### Scaling up aid and the IMF’s role
- Short-term constraints and mitigation:
  - Country capacities to absorb additional aid funds in key sectors are often limited (skill mix shortages, HIV/AIDS impacts).
  - Examples of mitigation strategies:
    - Education: certifying teachers without requiring a teaching degree.
    - Health: leveraging doctors and nurses with community health workers.
  - Evidence suggests these strategies can be cost effective, though capacity and quality constraints remain.
- Macroeconomic challenges:
  - If aid is not directly spent on imports, it can add to inflationary pressures and not raise net imports sufficiently.
  - Effective macroeconomic management may require real exchange rate appreciation to increase net imports and ease domestic inflation (see IMF, 2005b).
  - Entering longer-term spending commitments on the basis of historically volatile aid raises concerns about fiscal sustainability and aid dependency.
  - Hiring decisions are harder to reverse if aid flows decline (Foster, 2004; Heller, 2005).
- IMF’s role:
  - Support via surveillance, technical assistance, and access to financing to help governments address development needs consistent with macroeconomic stability.
  - IMF advises on creating fiscal space (resource mobilization, reduction in unproductive spending, and noninflationary financing) and collaborates with other institutions on sectoral considerations.
- Debate and perception:
  - IMF has been criticized for focusing on controlling fiscal deficits and limiting budgetary flexibility needed to meet the MDGs.
  - Center for Global Development survey of 353 health-sector professionals: “IMF caps” listed as a concern in only 1 percent of responses; leading concerns were lack of political will (29 percent), poor national coordination (28 percent), shortcomings of the health care delivery system (14 percent), national absorptive capacity constraints (8 percent), and policy confusion (7 percent).
- Donor aid trends:
  - Following Monterrey (2002), net ODA from OECD member states rose 15 percent in real terms during 2001–04, reaching US$80 billion in 2004.
  - In 2005, net ODA estimated to have increased by 23 percent in real terms, to US$98 billion.
  - Official promises project a further 31 percent real-term increase through 2010, to US$128 billion (OECD-DAC, 2006).

### Why wage bill ceilings?
- Policy rationale:
  - Composition of spending matters; wage bills can become sources of macroeconomic volatility and pressures:
    - High government wages and large employment can push up the wage bill and crowd out other spending.
    - Public wage increases can feed into general wage-price spirals undermining competitiveness.
    - Wage increases can result in fiscal slippages.
  - Macroeconomic impact depends on private sector wage responses and exchange rate regime (under fixed exchange rates impacts are harder to absorb).
- Use in PRGF-supported programs:
  - Some programs include wage bill ceilings to complement other fiscal targets.
  - Country examples:
    - Ghana: PC on the wage bill in 2005 to bring wages under control while a computerized payroll was developed.
    - Mozambique: quantitative benchmark on the wage bill in 2004 to focus attention on factors underlying wage increases and link civil service hiring to broader reforms and donor commitments.
    - Nicaragua: introduction of a PC on the wage bill against a background of public sector wages being well above comparable private sector wages.
- Concerns motivating ceilings:
  - Avoiding long-term expenditure commitments without long-term donor commitments.
  - Addressing risk of fiscal loss of control from ballooning wage bills.

### Case evidence — Zambia (Box 1 summary)
- Allegation:
  - Global Campaign for Education claimed Zambia was forced to stop hiring teachers and health workers to qualify for debt relief; alleged IMF told government to reduce public sector wage bill to no more than 8 percent of GDP and ban new hiring.
- IMF response and factual points presented:
  - Up until 2004, Zambia’s domestic debt and debt service costs were rising unsustainably; wage bill increased from 5.3 percent of GDP in 2000 to 8.4 of GDP in 2003, driven by large increases in wages and allowances.
  - April 2003: government wages raised and a new housing allowance introduced; measures were not in the budget and would have cost an additional 2½ percent of GDP.
  - July 2003: staff and authorities agreed a wage bill of such size would jeopardize economic recovery because it had to be financed by monetary expansion and additional borrowing, putting pressure on prices and interest rates and squeezing priority spending.
  - Large wage increase in 2003 constrained hiring as some ministries lacked budget room to recruit; attrition savings of some 4,000 staff during the year were used for hiring in central administration rather than priority sectors.
  - Structural factors impeded new hiring: in 2004 over 7,000 teachers were ready to take retirement, but the government could not afford termination benefits; donors stepped in and these teachers remained on the payroll while not working, delaying hiring of newly qualified teachers.
  - The 2004 PRGF-supported program did not include a limit on wage spending for teachers and health workers; a 2002 arrangement included a freeze on civil service hiring that specifically exempted hiring of teachers, doctors, and nurses.
  - The wage bill ceiling under the present arrangement explicitly accommodates hiring of new teachers and includes allocations for frontline staff by the Ministry of Health.
- Budgetary outcomes cited:
  - 2005 budget included extra resources for hiring additional teachers and health workers and a retention scheme for nurses and clinical staff.
  - By end-2005, government had recruited an additional 8,000 teachers and 1,336 medical personnel.
  - Proposed 2006 budget increases allocation to education and health to 45 percent of the discretionary budget from 36 percent in 2005, allowing recruitment of an additional 4,578 teachers and 800 medical personnel.

### Empirical finding highlighted in the text
- The paper’s empirical investigation did not find that wage bill ceilings in IMF-supported programs have prevented expanding public employment in priority sectors when donor funds were made available.
- Reconciling wage bill ceilings with expanded hiring requires careful policy balancing and greater transparency on donor funding plans and program modalities.

### Policy implications and options
- Transparency and coordination:
  - IMF actions:
    - Program documentation should explicitly link projections of donor support to wage bill ceilings.
    - Link between wage bill ceilings and the country’s poverty-reduction strategy should be discussed with authorities and reported in PRGF program documents.
  - Government actions:
    - Strengthen collaboration and exchange of information between hiring agencies and budget/macro policy agencies.
    - Promote greater coordination with aid donors.
  - Donor actions:
    - Make aid disbursements less volatile and more predictable.
    - Share disbursement schedules with recipient governments to enhance coordination and allow better budgetary planning and execution.
- Design and longer-term measures:
  - Increase transparency in how additional donor funds for wage spending are accommodated within wage bill ceilings.
  - Donors should provide clearer medium-term indications of aid plans to enable better fiscal planning.
  - Strengthen policy dialogue between sectoral ministries and macroeconomic agencies to address macroeconomic and sectoral absorption constraints and increase flexibility and efficiency of wage spending.
  - Consider phasing out ceilings when economic circumstances change and replace ceilings with more sophisticated conditionality (e.g., civil service reform) or drop conditionality where appropriate.
  - Address mismatch between donor commitment horizons and lasting implications of expanding workforce or pay:
    - Encourage donors to extend commitment periods.
    - Establish reserves to smooth aid volatility.
  - Enhance wage spending flexibility within governments:
    - Use temporary and flexible contracts, contracting out services, and efforts to raise efficiency of public service delivery.

### PRGF terminology overview (Box 2)
- Performance criteria (PCs):
  - Apply to clearly-specified variables or measures that can be objectively monitored and are so critical for achievement of program goals that disbursements should be interrupted in cases of non-observance.
  - Quantitative PCs: refer to variables reasonably under authorities’ control; typical quantitative PCs refer to foreign exchange reserves, monetary policy, fiscal policy, and external debt.
  - Structural PCs: refer to policy actions critical for success of the program or monitoring implementation; generally limited to macro-critical measures.
- Adjusters:
  - Protect a program from relatively small deviations from program assumptions (e.g., shortfalls or delays in external financing).
  - Designed to ensure balance between adjustment and financing in response to adverse shocks.
  - Apply to quantitative PCs and not to benchmarks.
- Waivers:
  - Waivers for nonobservance of PCs are supported by staff and granted by the Board when (i) noncompliance is minor and does not affect program objectives, and (ii) deviation is self-correcting or authorities have taken measures to correct it.
- Structural benchmarks:
  - Used for measures that cannot be objectively or adequately monitored as PCs, or for small steps in reform that would not individually warrant interruption of Fund financing; lighter form of conditionality than PCs.
  - Adjusters do not apply to them; can be set on numerical thresholds (quantitative benchmarks) or structural measures.

### Escalation of wage bill conditionality — Four country cases (Box 3)
- Ghana:
  - Prior: wage spending overruns; wage bill rose quickly during 2000–05.
  - 2003–06 PRGF: initially targeted an increase of wage spending (to 9 percent of GDP); overruns occurred due to unbudgeted salary increases for some 30,000 staff and unbudgeted base wage payments of subsidized agencies.
  - 2005 measures:
    - Strengthen payroll management, including computerizing employee records.
    - 2005 program established a quantitative PC on the wage bill including all allowances.
    - Ceiling allowed increase in wage spending to 9.1 percent of GDP from 8.5 percent of GDP in 2004 to encourage capacity building and higher productivity.
    - Government committed to comprehensive civil service reform expected to reduce the wage bill to around 8.5 percent of GDP over time.
- Nicaragua:
  - Prior: public sector wages grew rapidly, approaching about 8½ percent of GDP in 2005 and reaching just over 1/3 of total expenditure in 2005.
  - Policy response:
    - Authorities decided to lower compensation growth and eliminate vacancies in a number of government agencies.
    - December 2005: introduced a new quantitative PC on the wage bill under the PRGF-supported program; according to the 2006 budget, wages will grow somewhat above inflation.
    - Status: PC was introduced recently and it is too early to assess its impact.
- Chad:
  - Prior: wage bill not unduly large, but ceiling under the 2000–04 PRGF aimed to underpin payroll management improvements to support decentralization.
  - 2005–08 PRGF: elevated quantitative benchmark to a quantitative PC, reflecting a rise in the wage bill as a share of non-oil GDP and prospective changes in army payment system.
  - 2005 projection: wage bill projected to rise by some 8½ percent, reflecting recruitment in priority sectors (out of an estimated 2,931 new hires, 2,625 were to be in the priority sectors) and implementation of the military wage scale (with a reduction of supplementary allowances).
  - Structural conditionality: creation of a payroll and civil service roster.
- Malawi:
  - 2004 civil service wage reform:
    - Consolidated previously untaxed numerous allowances into basic pay; unified pay grades across core civil service; reduced grades to 18 from 75; realigned and decompressed pay line.
    - Resulted in a 25 percent increase in gross wages and a 50 percent increase in the minimum wage; pay to top-tier contractuals was frozen.
    - Fiscal outcome: reform proved significantly more costly than anticipated, resulting in a cost overrun of 1½ percent of GDP a year; reform also increased pension benefits significantly.
  - 2005–08 PRGF program:
    - Introduced a quantitative PC on the wage bill to keep domestically financed wage bill at or below 8 percent of GDP over the medium term.
    - Cap includes an adjuster for wage spending funded under the health SWAp.
    - Adjuster design: allows additional wage spending when wage-related disbursements under the multidonor SWAp arrangement for the health sector are larger than expected; adjuster is symmetric and without caps.
    - Status at program approval (August 2005): first review not yet completed; preliminary information suggests SWAp disbursements have lagged behind baseline projections, implying the wage bill ceiling may need downward adjustment.

### Design options for wage bill conditionality (from Box 3)
- Explicit adjuster (conditional automaticity):
  - Purpose: increase the wage bill ceiling—defined by a baseline on available donor funds—to accommodate additional hiring or wage increases in specific sectors if additional donor funds above baseline are available.
  - Mechanism: ceiling could be adjusted downward by the amount donor funds fall short of the baseline to avoid loosening conditionality; maximum allowed adjustment (upward or downward) could be capped if required.
  - Considerations: an upward adjuster for donor-financed hiring in health could be capped at levels that do not risk a wage-price spiral.
  - Practical experience: explicit adjusters require well-specified, clear, and monitorable links between donor funds and wage spending; frequent program reviews reduce the need for explicit adjusters when unexpected aid increases between reviews are small.
  - Usage in sample: of eight PRGF-supported programs with a quantitative PC on the wage bill, only Malawi includes an explicit adjuster (symmetric and without caps).
- Limiting coverage of the targeted wage bill ceiling:
  - Purpose: exclude specific sectors or levels of government from the ceiling so increases there are not subject to program ceilings.
  - Risks:
    - Reduces effectiveness of wage bill ceiling because partial ceilings are easier to circumvent than comprehensive ones.
    - Timely data on distribution of the wage bill across sectors often unavailable, complicating monitoring.
  - Practice: wage bill ceilings with limited coverage have been used extensively in the sample, often reflecting data limitations rather than deliberate exclusion of priority sectors.

### Empirical experience, program mechanics, and outcomes (from Box 3)
- Program reviews and adjustability:
  - Program reviews (typically three to six months ahead) are used to adjust conditionality when aid timing or levels change; PCs are set for the next review, targets for subsequent reviews are usually indicative and revisited.
  - Since aid-financed hiring often takes time, there is usually sufficient time to revise program conditionality to accommodate hiring once aid timing and modalities are clarified and funds made available.
- Waivers and breaches:
  - If a PC is breached, IMF’s Executive Board may grant a waiver when warranted; a waiver is only needed when a PC is breached.
  - In the sample: all PCs on the wage bill were met, and no waivers were needed. PCs were met in Dominica, Guyana, and Honduras. Wage bill PCs were introduced recently in Chad, Ghana, Malawi, Nicaragua and have not yet been subject to Executive Board review.
- Evidence and remaining concerns:
  - Recent experience suggests wage bill ceilings have not restricted the use of available donor funds, but evidence is not conclusive.
  - With donor funding for hiring set to increase (e.g., expanding vertical programs), tensions from aid scaling up under wage bill ceilings may become more acute.
  - Lack of predictability and transparency of future aid flows are main obstacles to effective wage bill conditionality; aid volatility and uncertainty hamper budgeting and make ministries of finance reluctant to precommit to funding additional priority spending.

### Appendix: countries without wage bill conditionality and structural measures
- Countries listed in Appendix Table 2 (as shown in columns):
  - AFR: Cameroon; Cape Verde; Congo, Republic of; Ethiopia; The Gambia; Lesotho; Madagascar; Rwanda; São Tomé & Príncipe; Tanzania; Uganda.
  - APD: Bangladesh; Lao People's Dem. Rep.; Mongolia; Sri Lanka; Cambodia.
  - EUR: Albania.
  - MCD: Armenia; Georgia; Kyrgyz Republic; Mauritania; Pakistan.
- Footnote on region definitions (verbatim):
  - 1/ Regions are shown in columns and are defined according to IMF classification: AFR: Sub-Saharan Africa; APD: Asia Pacific; EUR: Europe; MCD: Middle East and Central Asia; WHD: Western Hemisphere.
- Structural conditionality measures to improve payroll management (selected examples):
  - Develop a payroll and civil service roster based on a civil service census or update the registry of staff records (Chad).
  - Complete a census of the wage bill for all budgetary sector employees (Albania).
  - Transfer of the budgetary and administrative oversight on the pay, and transfer of paymasters to, the Treasury (Democratic Republic of Congo).
  - Cabinet to approve the final plan for the civil service reform covering human resource policy, reviewing the organization and structure of the civil service and addressing wage policy and payroll management (Ghana).
  - Establish public service remuneration board (Malawi).
  - Completion of a financial audit of the wage bill (Niger).
  - Introduce photo verification system for civil servants and teachers (Sierra Leone).
  - Submission to cabinet of a timetable for the design of a comprehensive medium-term civil service reform (Azerbaijan).
- Measures to streamline computation of wages (selected examples):
  - Determine new mechanism to determine salaries of public officials (Kenya).
  - Consolidate all allowance into the salary structure (Malawi).
  - Reorganize procedures for paying civil servants based on recommendations of external audit of the payroll system (Democratic Republic of Congo).
  - Develop detailed schedule of wage adjustments by grade (Malawi).
  - Issue regulations to integrate supplementary benefits into the overall salary (Honduras).
- Measures to rationalize employment levels (selected examples):
  - Eliminate identified ghost workers (Democratic Republic of Congo).
  - Implement plan to eliminate vacancies (Nepal).
  - Reduce the number of employees in the education sector by 5 percent (Tajikistan).
  - Outline the process for reducing the wage bill by 5 percent through retrenchment (Dominica).
- Overview of wage bill trends (Appendix IV — chart coverage):
  - PRGFs with Wage Bill Conditionality — countries charted include: AFR — Benin; Burkina Faso; Burundi; Chad; Congo, D.R.; Ghana; Kenya; Mali; Malawi; Mozambique; Niger; Senegal; Sierra Leone; Zambia. APD — Nepal. MCD — Azerbaijan; Tajikistan. WHD — Dominica; Guyana; Honduras; Nicaragua.
  - PRGFs without Wage Bill Conditionality — countries charted include: AFR — Cameroon; Cape Verde; Congo, R.; Ethiopia; The Gambia; Lesotho; Madagascar; Rwanda; São Tomé & Príncipe; Tanzania; Uganda. APD — Bangladesh; Lao, P.D.R.; Mongolia; Sri Lanka. EUR — Albania. MCD — Armenia; Georgia; Kyrgyz; Pakistan.
  - Sources for charts: "IMF’s World Economic Outlook database and IMF staff estimates." (verbatim)

*Source: Excerpts from “Wage Bill Conditionality in PRGF Countries, 2003–05.”*

### 1. Wage Bill Conditionality in PRGF Countries, 2003–05 .......................................................9

### Wage Bill Conditionality in PRGF Countries, 2003–05

### Introduction: context and purpose
- Meeting the Millennium Development Goals (MDGs) requires both more government spending and better policies.
- Example: Mattina’s (2006) estimate for Ethiopia: total cost of achieving the MDGs amounts to almost six times the country’s current annual GDP; roughly three-quarters of this would have to come in the form of additional public spending, financed by foreign aid.
- Per capita aid example: scale up from around US$11 in 2005 to over US$65 by 2015 (Ethiopia example).
- Human resource shortages in key sectors are a major impediment to MDGs:
  - Meeting the MDGs in Sub-Saharan Africa may require a tripling of the health-sector labor force, or adding more than one million workers over time (Berg and Qureshi, 2005).
- Central question of the paper: have wage bill ceilings in IMF-supported (PRGF) programs been detrimental to hiring additional personnel in key social sectors when additional donor resources were available?
- Main empirical conclusion summarized in the text: the paper did not find that wage bill ceilings in IMF-supported programs have prevented expanding public employment in priority sectors when donor funds were made available.
- Recommended complementary actions emphasized:
  - More transparent accommodation of additional donor funds for wage spending within wage bill ceilings.
  - Donors should provide clear indications about their plans to provide aid, including over the medium term.
  - Recipient governments should strengthen policy dialogue between sectoral ministries and macroeconomic policy agencies to address macroeconomic and sectoral absorption constraints and increase the flexibility and efficiency of wage spending.

### Scaling up aid and the IMF’s role
- Short-term constraints:
  - Country capacities to absorb additional aid funds in key sectors are often limited (skill mix shortages, HIV/AIDS impacts).
  - Examples of mitigation strategies:
    - Education: certifying teachers without requiring a teaching degree.
    - Health: leveraging doctors and nurses with community health workers.
  - Evidence suggests these strategies can be cost effective, though capacity and quality constraints remain.
- Macroeconomic challenges of scaled-up aid:
  - If aid is not directly spent on imports, it can add to inflationary pressures and not raise net imports sufficiently.
  - Effective macroeconomic management may require real exchange rate appreciation to increase net imports and ease domestic inflation (see IMF, 2005b).
  - Entering longer-term spending commitments on the basis of historically volatile aid raises concerns about fiscal sustainability and aid dependency.
  - Hiring decisions are harder to reverse if aid flows decline (Foster, 2004; Heller, 2005).
- IMF’s role:
  - Support via surveillance, technical assistance, and access to financing to help governments address development needs consistent with macroeconomic stability.
  - IMF advises on creating fiscal space (resource mobilization, reduction in unproductive spending, and noninflationary financing) and collaborates with other institutions on sectoral considerations.
- Debate and perception:
  - IMF has been criticized for focusing on controlling fiscal deficits and limiting budgetary flexibility needed to meet the MDGs.
  - Example empirical evidence cited: Center for Global Development survey of 353 health-sector professionals found “IMF caps” listed as a concern in only 1 percent of responses; leading concerns were lack of political will (29 percent), poor national coordination (28 percent), shortcomings of the health care delivery system (14 percent), national absorptive capacity constraints (8 percent), and policy confusion (7 percent).
- Donor aid trends context:
  - Following Monterrey (2002), net ODA from OECD member states rose 15 percent in real terms during 2001–04, reaching US$80 billion in 2004.
  - In 2005, net ODA estimated to have increased by 23 percent in real terms, to US$98 billion.
  - Official promises project a further 31 percent real-term increase through 2010, to US$128 billion (OECD-DAC, 2006).

### Why wage bill ceilings?
- Policy rationale:
  - Composition of spending matters; wage bills can become sources of macroeconomic volatility and pressures:
    - High government wages and large employment can push up the wage bill and crowd out other spending.
    - Public wage increases can feed into general wage-price spirals undermining competitiveness.
    - Wage increases can result in fiscal slippages.
  - Macroeconomic impact of public wage increases depends on factors such as private sector wage responses and exchange rate regime (under fixed exchange rates impacts are harder to absorb).
- Use in PRGF-supported programs:
  - Some programs include wage bill ceilings to complement other fiscal targets.
  - Country examples from the text:
    - Ghana: PRGF-supported program included a performance criterion (PC) on the wage bill in 2005 to bring wages under control while a computerized payroll was developed.
    - Mozambique: quantitative benchmark on the wage bill in 2004 to focus attention on factors underlying wage increases and link civil service hiring to broader reforms and donor commitments.
    - Nicaragua: introduction of a PC on the wage bill against a background of public sector wages being well above comparable private sector wages (contextual note in text).
- Concerns motivating ceilings:
  - Avoiding long-term expenditure commitments without long-term donor commitments to finance them.
  - Addressing risk of fiscal loss of control from ballooning wage bills.

### Case evidence and illustrative box (Zambia)
- Box 1 summary: “Did IMF Policies Prevent the Hiring of Teachers in Zambia?”
  - The Global Campaign for Education claimed Zambia was forced to stop hiring teachers and health workers to qualify for debt relief; alleged IMF told government to reduce public sector wage bill to no more than 8 percent of GDP and ban new hiring.
  - IMF response points (as presented in the text):
    - Up until 2004, Zambia’s domestic debt and debt service costs were rising unsustainably; leading contributor was the government wage bill, which increased from 5.3 percent of GDP in 2000 to 8.4 of GDP in 2003, driven by large increases in wages and allowances.
    - April 2003: government wages raised and a new housing allowance introduced; measures were not in the budget and would have cost an additional 2½ percent of GDP.
    - July 2003: IMF staff and authorities agreed that a wage bill of such size would jeopardize economic recovery because it had to be financed by monetary expansion and additional government borrowing, putting pressure on prices and interest rates and squeezing priority spending.
    - Large wage increase in 2003 constrained hiring as some ministries lacked budget room to recruit; attrition savings of some 4,000 staff during the year were used for hiring in central administration rather than priority sectors.
    - Structural factors impeded new hiring: in 2004 over 7,000 teachers were ready to take retirement, but the government could not afford termination benefits; donors stepped in and these teachers remained on the payroll while not working, delaying hiring of newly qualified teachers.
    - The 2004 PRGF-supported program did not include a limit on wage spending for teachers and health workers; a 2002 arrangement included a freeze on civil service hiring that specifically exempted hiring of teachers, doctors, and nurses.
    - The wage bill ceiling under the present arrangement explicitly accommodates hiring of new teachers and includes allocations for frontline staff by the Ministry of Health.
    - Budgetary outcomes cited:
      - 2005 budget included extra resources for hiring additional teachers and health workers and a retention scheme for nurses and clinical staff.
      - By end-2005, government had recruited an additional 8,000 teachers and 1,336 medical personnel.
      - Proposed 2006 budget increases allocation to education and health to 45 percent of the discretionary budget from 36 percent in 2005, allowing recruitment of an additional 4,578 teachers and 800 medical personnel.

### Empirical finding highlighted in the text
- The paper’s empirical investigation did not find that wage bill ceilings in IMF-supported programs have prevented expanding public employment in priority sectors when donor funds were made available.
- The text emphasizes that reconciling wage bill ceilings with expanded hiring requires careful policy balancing and greater transparency on donor funding plans and program modalities.

### Policy implications and options (as discussed)
- Increase transparency in how additional donor funds for wage spending are accommodated within wage bill ceilings in IMF-supported programs.
- Donors should provide clearer medium-term indications of aid plans to enable better fiscal planning.
- Strengthen policy dialogue between sectoral ministries and macroeconomic agencies to:
  - Address macroeconomic and sectoral absorption constraints.
  - Increase flexibility and efficiency of wage spending to support fiscal sustainability.
- Link civil service/hiring decisions to overall development objectives and comprehensive civil service reform to avoid hiring distortions and improve targeting to priority sectors.

*Source: Excerpts from “Wage Bill Conditionality in PRGF Countries, 2003–05.”*

### Box 2. Overview of Terminology in PRGF-Supported Adjustment Programs

### Box 2. Overview of Terminology in PRGF-Supported Adjustment Programs

### Performance criteria (PCs)
- PCs apply to clearly-specified variables or measures that can be objectively monitored and are so critical for the achievement of the program goals or monitoring implementation that disbursements under the arrangement should be interrupted in cases of non-observance.
- Quantitative PCs:
  - Refer to variables that are reasonably under the authorities’ direct or indirect control, and can be measured in a timely way.
  - Typical quantitative PCs refer to foreign exchange reserves, monetary policy, fiscal policy, and external debt.
- Structural PCs:
  - Refer to policy actions that are critical for the success of the program or for monitoring implementation, and should be defined unambiguously.
  - Are in general limited to measures that are macro-critical.

### Adjusters
- Protect a program from relatively small deviations from program assumptions, such as shortfalls or delays in external financing or privatization receipts that could be accommodated within the program without affecting the program’s main objectives.
- Designed to ensure an appropriate balance between adjustment and financing in response to adverse shocks.
- Apply to quantitative PCs and not to benchmarks.

### Waivers
- Waivers for the nonobservance of PCs are supported by the staff and granted by the Board on the basis that:
  - (i) the noncompliance is minor and does not affect significantly the program objectives; and
  - (ii) the deviation is self-correcting (temporary) or the authorities have taken the necessary measures to correct the deviation.

### Structural benchmarks
- Used for measures that cannot be objectively or adequately monitored to be PCs, or for small steps in the overall reform process that would not individually warrant an interruption of Fund financing in case of breach—they are a “lighter” form of conditionality than PCs.
- As they are less critical than PCs, adjusters do not apply to them.
- Can be set on numerical thresholds (quantitative benchmarks) or structural measures.

*Based on International Monetary Fund (2002).*

### Box 3. Escalation of Wage Bill Conditionality—Four Country Cases

### Box 3. Escalation of Wage Bill Conditionality—Four Country Cases

### Country cases: implementation and outcomes
- Ghana
  - Prior situation: wage spending overruns were a long-standing problem; wage bill rose quickly during 2000–05.
  - 2003–06 PRGF-supported program initially targeted an increase of wage spending (to 9 percent of GDP); overruns occurred on two occasions due to unbudgeted salary increases for some 30,000 staff in the education sector and unbudgeted base wage payments of subsidized agencies.
  - 2005 measures:
    - Government committed to strengthening payroll management, including computerizing employee records.
    - 2005 program established a quantitative PC on the wage bill including all allowances.
    - The ceiling allowed for an increase in wage spending to 9.1 percent of GDP from 8.5 percent of GDP in 2004, mainly to encourage capacity building and higher productivity by attracting and retaining qualified personnel.
    - Over the medium term, government committed to comprehensive civil service reform expected to reduce the wage bill to around 8.5 percent of GDP over time.

- Nicaragua
  - Prior situation: public sector wages grew rapidly, approaching about 8½ percent of GDP in 2005 and reaching just over 1/3 of total expenditure in 2005.
  - Concerns: public wages rose well above private sector wages, risking a wage-price spiral and undermining competitiveness.
  - Policy response:
    - Authorities decided to lower compensation growth and eliminate vacancies in a number of government agencies.
    - December 2005: introduced a new quantitative PC on the wage bill under the PRGF-supported program; according to the 2006 budget, wages will grow somewhat above inflation.
    - Status: PC was introduced recently and it is too early to assess its impact.

- Chad
  - Prior situation: wage bill not unduly large, but a ceiling under the 2000–04 PRGF-supported program aimed to underpin civil service payroll management improvements to support administrative decentralization.
  - 2005–08 PRGF-supported program: elevated the quantitative benchmark on the wage bill to a quantitative PC, reflecting a rise in the wage bill as a share of non-oil GDP and prospective changes in army payment system (from lump-sum to a wage scale based on seniority and grade).
  - 2005 projection: wage bill projected to rise by some 8½ percent, reflecting recruitment in priority sectors (out of an estimated 2,931 new hires, 2,625 were to be in the priority sectors) and implementation of the military wage scale (with a reduction of supplementary allowances).
  - Structural conditionality: creation of a payroll and civil service roster.

- Malawi
  - 2004 civil service wage reform:
    - Consolidated previously untaxed numerous allowances into basic pay; unified pay grades across core civil service; reduced grades to 18 from 75; realigned and decompressed pay line.
    - Resulted in a 25 percent increase in gross wages and a 50 percent increase in the minimum wage; pay to top-tier contractuals was frozen.
    - Fiscal outcome: reform proved significantly more costly than anticipated, resulting in a cost overrun of 1½ percent of GDP a year; reform also increased pension benefits significantly.
  - 2005–08 PRGF program:
    - Introduced a quantitative PC on the wage bill to keep domestically financed wage bill at or below 8 percent of GDP over the medium term.
    - Cap includes an adjuster for wage spending funded under the health SWAp.
    - Adjuster design: allows additional wage spending when wage-related disbursements under the multidonor SWAp arrangement for the health sector are larger than expected; adjuster is symmetric and without caps.
    - Status at program approval (August 2005): first review not yet completed; preliminary information suggests SWAp disbursements have lagged behind baseline projections, implying the wage bill ceiling may need downward adjustment.

### Design options for wage bill conditionality
- Explicit adjuster (conditional automaticity)
  - Purpose: increase the wage bill ceiling—defined by a baseline on available donor funds—to accommodate additional hiring or wage increases in specific sectors if additional donor funds above baseline are available.
  - Mechanism: ceiling could be adjusted downward by the amount donor funds fall short of the baseline to avoid loosening conditionality; maximum allowed adjustment (upward or downward) could be capped if required.
  - Considerations: an upward adjuster for donor-financed hiring in health could be capped at levels that do not risk a wage-price spiral.
  - Practical experience: explicit adjusters require well-specified, clear, and monitorable links between donor funds and wage spending; frequent program reviews reduce the need for explicit adjusters when unexpected aid increases between reviews are small.
  - Usage in sample: of eight PRGF-supported programs with a quantitative PC on the wage bill, only Malawi includes an explicit adjuster (symmetric and without caps).

- Limiting coverage of the targeted wage bill ceiling
  - Purpose: exclude specific sectors or levels of government from the ceiling so increases in wages or employment there are not subject to program ceilings.
  - Risks:
    - Reduces effectiveness of wage bill ceiling because partial ceilings are easier to circumvent than comprehensive ones.
    - Timely data on distribution of the wage bill across sectors often unavailable, complicating monitoring.
  - Practice: wage bill ceilings with limited coverage have been used extensively in the sample, often reflecting data limitations rather than deliberate exclusion of priority sectors.

### Empirical experience, program mechanics, and outcomes
- Program reviews and adjustability
  - Program reviews (typically three to six months ahead) are used to adjust conditionality when aid timing or levels change; PCs are set for the next review, targets for subsequent reviews are usually indicative and revisited.
  - Since aid-financed hiring often takes time, there is usually sufficient time to revise program conditionality to accommodate hiring once aid timing and modalities are clarified and funds made available.

- Waivers and breaches
  - If a PC is breached, IMF’s Executive Board may grant a waiver when warranted by circumstances; a waiver is only needed when a PC is breached, not when a benchmark is missed.
  - In the sample: all PCs on the wage bill were met, and no waivers were needed. PCs were met in Dominica, Guyana, and Honduras. Wage bill PCs were introduced recently in Chad, Ghana, Malawi, Nicaragua and have not yet been subject to Executive Board review.

- Evidence and remaining concerns
  - Recent experience suggests wage bill ceilings have not restricted the use of available donor funds, but evidence is not conclusive.
  - With donor funding for hiring set to increase (e.g., expanding vertical programs), tensions from aid scaling up under wage bill ceilings may become more acute.
  - Lack of predictability and transparency of future aid flows are main obstacles to effective wage bill conditionality; aid volatility and uncertainty hamper budgeting and make ministries of finance reluctant to precommit to funding additional priority spending.

### Policy recommendations and practical measures
- Coordination and transparency improvements
  - IMF actions:
    - Program documentation should explicitly link projections of donor support to wage bill ceilings.
    - Link between wage bill ceilings and the country’s poverty-reduction strategy should be discussed with authorities and reported in PRGF program documents.
  - Government actions:
    - Strengthen collaboration and exchange of information between hiring agencies (sectoral or line ministries) and budget/macro policy agencies (ministries of finance).
    - Promote greater coordination with aid donors.
  - Donor actions:
    - Make aid disbursements less volatile and more predictable.
    - Share disbursement schedules with recipient governments to enhance coordination and allow better budgetary planning and execution.

- Longer-term reforms and contingency measures
  - Aim to make wage bill conditionality redundant via lasting structural measures (noting difficulty in phasing out PCs; sample shows no examples where wage bill PCs were phased out but five countries introduced PCs).
  - Consider phasing out ceilings when economic circumstances change (e.g., countries exiting post-conflict environments) and replace ceilings with more sophisticated conditionality (e.g., civil service reform) or drop conditionality where appropriate.
  - Address mismatch between donor commitment horizons and lasting implications of expanding workforce or pay:
    - Encourage donors to extend commitment periods.
    - Establish reserves to smooth aid volatility.
  - Enhance wage spending flexibility within governments:
    - Use temporary and flexible contracts, contracting out services, and efforts to raise efficiency of public service delivery to help adjust spending in response to financing changes and safeguard fiscal sustainability.

*Italic source: Box 3. Escalation of Wage Bill Conditionality—Four Country Cases (excerpt) from the IMF PDF _wp06106*

### Appendix Table 2. Countries without Wage Bill Conditionality 1/

### _wp06106 - Appendix Table 2. Countries without Wage Bill Conditionality 1/

### Countries listed in the Appendix Table 2 (as shown in columns)
- AFR      APD                        EUR         MCD
- Cambodia  Bangladesh               Albania     Armenia
- Cameroon  Lao People's Dem. Rep.   Georgia
- Cape Verde  Mongolia               Kyrgyz Republic
- Congo, Republic of  Sri Lanka      Mauritania
- Ethiopia                                                                                     Pakistan
- The                                                                                         Gambia
- Lesotho
- Madagascar
- Rwanda
- São Tomé & Príncipe
- Tanzania
- Uganda

### Footnote on region definitions (verbatim)
- 1/ Regions are shown in columns and are defined according to IMF classification: AFR: Sub-Saharan Africa; APD: Asia Pacific; EUR: Europe; MCD: Middle East and Central Asia; WHD: Western Hemisphere.

### Structural conditionality on wage bill policy (Appendix III — substantive measures listed)
- Measures to improve payroll management
  - Develop a payroll and civil service roster based on a civil service census or update the registry of staff records (Chad).
  - Complete a census of the wage bill for all budgetary sector employees (Albania).
  - Transfer of the budgetary and administrative oversight on the pay, and transfer of paymasters to, the Treasury (Democratic Republic of Congo).
  - Cabinet to approve the final plan for the civil service reform covering human resource policy, reviewing the organization and structure of the civil service and addressing wage policy and payroll management (Ghana).
  - Establish public service remuneration board (Malawi).
  - Completion of a financial audit of the wage bill (Niger).
  - Introduce photo verification system for civil servants and teachers (Sierra Leone).
  - Submission to cabinet of a timetable for the design of a comprehensive medium-term civil service reform (Azerbaijan).
- Measures to streamline the computation of wages
  - Determine new mechanism to determine salaries of public officials (Kenya).
  - Consolidate all allowance into the salary structure (Malawi).
  - Reorganize procedures for paying civil servants based on the recommendations of the external audit of the payroll system (Democratic Republic of Congo).
  - Develop detailed schedule of wage adjustments by grade (Malawi).
  - Issue regulations to integrate supplementary benefits into the overall salary (Honduras).
- Measures to rationalize employment levels
  - Eliminate identified ghost workers (Democratic Republic of Congo).
  - Implement plan to eliminate vacancies (Nepal).
  - Reduce the number of employees in the education sector by 5 percent (Tajikistan).
  - Outline the process for reducing the wage bill by 5 percent through retrenchment (Dominica).

### Overview of wage bill trends (Appendix IV — charts and country coverage)
- Appendix IV presents wage bill trend charts in two groupings:
  - PRGFs with Wage Bill Conditionality — charts and axes include "In percent of total spending (lhs)" and "In percent of GDP (rhs)". Countries shown include (by region headings and chart labels): AFR — Benin; Burkina Faso; Burundi; Chad (note: "1/ Wage bill in percent of non-oil GDP is used."); Congo, D.R.; Ghana; Kenya; Mali; Malawi; Mozambique; Niger; Senegal; Sierra Leone; Zambia. APD — Nepal. MCD — Azerbaijan; Tajikistan. WHD — Dominica; Guyana; Honduras; Nicaragua.
  - PRGFs without Wage Bill Conditionality — charts with same axes. Countries shown include (by region headings and chart labels): AFR — Cameroon; Cape Verde; Congo, R.; Ethiopia; The Gambia; Lesotho; Madagascar; Rwanda; São Tomé & Príncipe; Tanzania; Uganda. APD — Bangladesh; Lao, P.D.R.; Mongolia; Sri Lanka. EUR — Albania. MCD — Armenia; Georgia; Kyrgyz; Pakistan.
- Sources for charts: "IMF’s World Economic Outlook database and IMF staff estimates." (verbatim)

*Source: _wp06106 - Appendix Table 2. Countries without Wage Bill Conditionality 1/ (IMF PDF content provided).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06106.pdf_