## htnea2025004 - Introduction and Main Messages

## Source details

**Canonical URL:** [htnea2025004 - Introduction and Main Messages](https://www.imf.org/-/media/files/publications/howtonotes/2025/english/htnea2025004.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/howtonotes/2025/english/htnea2025004.pdf.md)
- [Structured JSON version](/-/media/files/publications/howtonotes/2025/english/htnea2025004.pdf.json)

---

### Introduction and Main Messages
- Scope: Management of government spending on employee compensation in small developing states (SDS).
- Key premise: Employee compensation constitutes a substantial share of government spending in SDS and is central to fiscal sustainability and service delivery.
- Main messages and guiding principles:
  - Small states face specific challenges driven by diseconomies of scale and greater openness and volatility.
  - Flexibility is required to overcome diseconomies of scale and address inefficiencies; benchmarking should prioritize the domestic private sector but may require cross-country comparisons in cases of acute “brain drain.”
  - Consolidation of functions in central government can mitigate diseconomies of scale but implies trade-offs.
  - Small states have used varied approaches: prioritization of essential sectors, improved training, migration management, GovTech, remote-location adaptations, and cooperation with regional peers or larger countries.

### Definitions and Country Groupings
- SDS definition: low-income and emerging market economies with a population of less than 1.5 million.
- Microstates: population below 200,000.
- Small advanced economies: countries in the highest income group with population below 3 million.
- Government compensation: total wages, salaries, allowances, and any other benefits of government employees.

### How Is Government Compensation and Employment Different in Small States? — Key Findings
- Core drivers of difference:
  - Diseconomies of scale in public service delivery.
  - Greater openness and volatility.
  - Additional factors: higher risk of natural disasters and rising sea levels, low population density, high population dispersion, remoteness, narrow economic bases (tourism, fishing, mining, agriculture).
- Compensation spending (averages, latest values available):
  - Small states: 12 percent of GDP.
  - Microstates: 16.5 percent of GDP.
  - Advanced economies: 9.4 percent of GDP.
  - Emerging market economies: 8.2 percent of GDP.
  - Low-income developing countries: 6.2 percent of GDP.
- Government employment and compensation (Table 1, selected indicators, latest value available):
  - Small states (Obs.): 34; Government Spending on Compensation of Employees As a Share of GDP: 12.1; Government Employment As a Share of Public Spending: 28.3; Compensation Premium As a Share of Working-Age Population: 13.5; In % of Private Sector Compensation (Obs. 2): 41.0.
  - Microstates (Of which): Obs. 15; As a Share of GDP: 16.5; As a Share of Public Spending: 30.4; As a Share of Working-Age Population: 18.7; Palau (2014) 21.0.
  - Other small states: Obs. 19; As a Share of GDP: 8.6; As a Share of Public Spending: 26.6; As a Share of Working-Age Population: 10.7; Eswatini (2016) 61.8.
  - Small advanced economies: Obs. 10; As a Share of GDP: 10.9; As a Share of Public Spending: 26.1; As a Share of Working-Age Population: 13.9; In % of Private Sector Compensation (Obs. 7): 12.0.
  - Advanced economies: Obs. 27; As a Share of GDP: 9.4; As a Share of Public Spending: 22.6; As a Share of Working-Age Population: 12.5; In % of Private Sector Compensation (Obs. 24): 6.5.
  - Emerging market economies: Obs. 66; As a Share of GDP: 8.2; As a Share of Public Spending: 26.7; As a Share of Working-Age Population: 8.9; In % of Private Sector Compensation (Obs. 36): 14.4.
  - Low-income developing countries: Obs. 52; As a Share of GDP: 6.2; As a Share of Public Spending: 27.0; As a Share of Working-Age Population: 4.4; In % of Private Sector Compensation (Obs. 20): 15.5.
  - All (global sample): Obs. 189; As a Share of GDP: 8.7; As a Share of Public Spending: 26.4; As a Share of Working-Age Population: 9.5; In % of Private Sector Compensation (Obs. 80): 12.3.
  - Sources: FAD Government compensation and employment database and Government-Private Sector Compensation Premium Data set.
  - Note: Data as a share of GDP and Government spending refer to 2022 and 2021, respectively, in most cases. Data on government employment refer to the latest data available after 2015.
- “Too small to specialize”: small absolute government size limits specialization, making it difficult to justify full-time niche experts; multitasking can leave critical tasks underserved.
- Brain drain and limited labor pool:
  - High outward migration on average, particularly in Pacific and Caribbean.
  - Example: St. Kitts and Nevis — 2.4 nationals living abroad for each person living in the country.
  - Education abroad and non-return (Palau, Samoa) and pay differentials with larger countries exacerbate recruitment challenges.
- Minimum necessary positions and geographic necessities increase per-capita staffing costs (remote islands, full-time local posts).
- Funding and risk:
  - Many small states sustain high compensation-to-GDP ratios with substantial revenues from abroad, which are volatile.
  - Examples: Federated States of Micronesia spends 21 percent of GDP on compensation and receives substantial grants; Marshall Islands 22 percent; Palau 18 percent. Nauru spent about 21 percent of its GDP on compensation of employees in fiscal year 2022, supported by grants, fishing license fees, and revenues related to its Regional Processing Centre.

### How to Take into Account the Circumstances of Small States in Government Compensation and Employment Management — Policy Guidance
- Overarching principle: tailor standard public-sector compensation management recommendations to small-state circumstances; assess trade-offs in country-specific context.
- Managing government compensation with macroeconomic and fiscal volatility:
  - Heightened scrutiny is required given higher average compensation spending and historical large increases in small states.
  - Recommended tools (also used in other states): hiring freezes, limits on pay growth, reviews of spending; right-sizing through sectoral spending reviews (example: Eswatini).
  - Caution: major short-term reductions in employment risk losing skilled staff to emigration or other sectors; temporary pay restraint that is well communicated may have fewer long-term consequences.
  - Risk identified: using government employment as short-term social safety net can create hard-to-reverse increases in compensation spending and private-sector shortages during recovery.
  - Need for flexible yet robust frameworks that allow adaptation to shocks and natural disasters; governance, transparency, and data improvements are part of the reform package.
  - Institutional constraints: constitutional, legislative, and institutional rules in some small states reduce adaptability (example: Eastern Caribbean Currency Union Constitutions limit reducing remuneration, allowances, increments).
  - Good practice example: Vanuatu — school grant sizes proportional to enrollment and published on a dedicated website.
- Taking account of diseconomies of scale:
  - Centralization can mitigate diseconomies of scale; decentralization benefits are limited in small states while costs are greater.
  - Examples:
    - Cyprus centralized all police staff into one institution.
    - Palau: IMF technical assistance recommended considering merging some functions from federated states into national government (park policing, road maintenance, trash collection, vehicle registration).
  - Trade-offs: decentralization can limit power concentration and reduce clientelism and patronage; historical/political reasons for decentralization may make reversal difficult.
  - Regional universities as economy-of-scale solution:
    - University of the West Indies serves 17 English-speaking countries and territories with campuses in four countries and one online campus; financed by member governments based on student numbers.
    - University of the South Pacific is jointly owned by 12 members, with a campus in each and a single university structure.
  - Partnerships with larger countries can discharge state functions (postal, consular, defense) and leverage economies of scale to avoid additional compensation spending.

### Practical Approaches and Tools Used by Small States (summary of approaches mentioned)
- Prioritization of essential sectors and personnel.
- Improved training and capacity-building adapted to small-scale settings.
- Migration management policies to mitigate brain drain.
- GovTech and remote-service delivery innovations.
- Cooperation with regional peers or larger countries to support public service delivery without increasing compensation spending.

### Box 2. Free Association and Other Agreements between Countries
- Free association arrangements and partner-provided public services:
  - Free and formal agreements between a small sovereign or self-governing state and a larger country can yield partner-provided public services.
  - Examples:
    - The Federated States of Micronesia, the Marshall Islands, and Palau: Compact of Free Association with the United States. Under the Compact, the United States provides defense; services from the US Postal Service, Weather Service, and Federal Aviation Administration, among other agencies; and access to US grants and social services.
    - The Cook Islands and Niue: free association agreements with New Zealand; New Zealand has responsibility for defense and foreign affairs for both countries and acts on this responsibility only at the request of their governments.
    - Nauru: defense provided by Australia under an informal agreement.
    - Liechtenstein: citizens can rely on Switzerland’s consular services abroad; Switzerland can represent Liechtenstein’s diplomatic interests upon request.
    - San Marino: defense is provided by Italy.
  - Note: The Cook Islands and Niue are not IMF members.
- Prioritization and innovative solutions with scarce resources (health and education):
  - Prioritization examples and outcomes:
    - Maldives: chose to offer higher pay for health workers and employ more health workers after a large improvement in GDP and available revenue; this led to improvements in access to health services and health outcomes, but has created challenges around affordability and efficiency of spending (World Bank 2023).
    - Seychelles: prioritized health and education in public spending leading up to 2008, producing improvements in health outcomes and educational enrollment but raising continued concerns about efficiency and sustainability of spending (World Bank 2017).
  - Education in Pacific SDS:
    - Blackman and others (2023): education expenditure in the Pacific SDS tends to be high or very high relative to comparator countries, whereas educational outcomes tend to be below expectations across many measures.
    - Identified challenges: quality of teaching, reading opportunities at home, and basic resources in some Pacific states.
    - Policy responses and evidence:
      - Micronesia and Tonga developed tailored and affordable materials and training to support schools in remote locations; these approaches produced significant improvements:
        - After one year, the share of proficient students increased by about 20 percent both in reading and literacy in Micronesia.
        - Average reading scores improved by 0.19 standard deviations in Tonga.
- Tailored approaches to overcome brain drain issues (compensation, training, recruitment):
  - Benchmarking compensation:
    - Government compensation should generally be competitive with the domestic private sector; typically government pay can be somewhat below private-sector pay once individual characteristics are accounted for (Thévenot and Walker 2024).
    - In cases of acute competition with other countries, it may become necessary to benchmark government pay against other countries to retain staff and protect critical public services; this could imply paying government employees more than comparable domestic private-sector workers for critical skills (with trade-offs including crowding out private-sector employment and higher cost).
  - Training and expanding supply:
    - Many small states may need to train additional workers to offset emigration of essential workers.
    - Nurses training in the Caribbean has been important given high emigration (World Bank 2009); in the 2000s Guyana, St. Lucia, and St. Vincent and the Grenadines expanded training places and improved training quality.
    - Approaches to get destination countries to pay for the training of emigrating nurses have been used in SVG to offset domestic training costs.
  - Broader retention strategies:
    - Successful recruitment and retention look beyond remuneration to include job security, career prospects, working environment, job satisfaction, family ties, and quality of life (World Bank 2009).
    - Migration agreements were considered by Caribbean small states in the 2000s to reduce nurse outflows (World Bank 2012).
    - UK Department of Health for England introduced a 2001 code of practice banning “active” recruitment of health workers from low-income and emerging market countries unless covered by a memorandum of understanding (Khonje 2015).
  - Recruiting foreign skilled workers:
    - Palau (limited domestic higher education capacity) is considering fellowship programs to recruit foreign workers into government service (Walker and others 2024). IMF technical assistance recommended Palau harmonize employment and compensation policies across domestic and foreign workers to facilitate recruitment.
    - Guyana has actively recruited Cuban nurses and doctors while gradually training more national health workers.
- Using GovTech approaches to overcome delivery challenges:
  - General potential:
    - GovTech approaches can serve populations difficult to reach through standard service models (Barron Rodriguez and others 2022; Amaglobeli and others 2023).
    - SDS are on average well connected, with substantial variability across countries (Cantu-Bazaldua 2021).
  - Health:
    - Telemedicine and electronic medical records can enable access where otherwise challenging.
    - Bloch and Wang (2023): telemedicine, laptops/tablets, data, and drones have supported access to rural and remote areas in Brazil and sub-Saharan Africa (Blackman and others [2023] for telemedicine in the Pacific).
    - Sub-Saharan African policymakers are investing in biometric identification systems to enable data-heavy interventions (Parekh 2020).
  - Education:
    - Digital tech supports remote coaching and meetings; Kotze, Fleisch, and Taylor (2019) found virtual teacher training had similar positive effects to on-site training but was slightly cheaper and less logistically challenging (Rodriguez-Segura 2021).
    - Digital libraries can provide publicly available resources across member states (Organization for Eastern Caribbean States’ digital library; Myers and others 2022).
  - Public administration and payroll integrity:
    - Blockchain in Guinea-Bissau produced a single, more reliable set of compensation data, curbing discrepancies and reducing corruption opportunities through consistency and interoperability.
      - Result: the ratio of compensation spending to tax revenues declined from 84 percent to 50 percent between 2020 and 2024 (after implementation).
  - Underlying digital infrastructure:
    - Investments in connectivity examples include the Caribbean Regional Communications Infrastructure Program that expanded connectivity to Grenada, St. Lucia, and SVG.

### Understanding Country Context and Policy Implications
- No single cutoff identifies excessive public compensation spending; assessments depend on:
  1. The overall level of fiscal sustainability and revenue mobilization.
  2. The distribution and approach to public service delivery (for example, between central and local government).
  3. Policy design and objectives.
- Other considerations:
  - Concerns about a large public sector crowding out the private sector vary across small countries.
  - Inefficiency in public service delivery may stem from geographic or historical challenges or misallocation of resources—each requiring distinct policy responses.
  - Institutional capacity will influence reform feasibility and appropriateness.
- IMF support:
  - The IMF’s Fiscal Affairs Department offers capacity development to help manage government employment and compensation and identify measures consistent with efficient service delivery and fiscal sustainability.

### Annex: Determinants of government compensation spending (regression evidence)
- Empirical strategy:
  - Two regression analyses were conducted:
    1. Panel regression with random effects (generalized least squares) on a panel of 189 countries between 1980 and 2022 (5,646 observations). Explanatory variables included population (log) and population (log)^2.
    2. Cross-sectional linear regression focusing on the quartile of countries with the smallest populations (sample covers 41 countries with latest data since 2019).
- Key findings from regressions (Annex Table 2.1):
  - Regression details and estimates:
    - Observations: 5,646; 5,646; 41; 38 (for Regression 1, Regression 1 bis, Regression 2, Regression 2 bis respectively).
    - R2: 0.11; 0.12; 0.67; 0.72.
    - Population (log): −11.7**; −11.3**; −2.0; −3.2**.
    - Population (log)2: 0.4**; 0.4** (present in the panel regressions).
    - GDP per capita (log): −0.2**; −0.2**; −0.6.
    - Commodity exporter (yes): 7.0* (in one specification).
    - Latin America and the Caribbean: −6.1* (in one specification).
    - Emerging and developing Asia and Remote interaction: 8.4** (significant interaction term in the small-country cross-section).
  - Interpretation:
    - Population size has a statistically significant relationship with government compensation spending in the panel regressions, suggesting a negative relationship between population size and government compensation spending.
    - The positive and significant quadratic term (population (log)2) indicates nonlinearity: at higher levels of population the negative effect of population size on government compensation is reversed (consistent with Jugl 2022).
    - In the small-country cross-section, GDP per capita is not statistically significant (possibly due to small sample size). The interaction between remoteness and classification as emerging Asian economies is significant, highlighting specificity for small, geographically isolated countries.
  - Statistical notation: dependent variable is government spending as a share of GDP; **p < .01; *p < .1. In Regression 2 bis, variables with statistically insignificant estimates were excluded.

*Source: IMF | How to Note NOTE/2025/004 — "How to Manage Government-Sector Remuneration and Staffing in Small States", May 2025.*

### Introduction and Main Messages .........................................................................................

### htnea2025004 - Introduction and Main Messages

### Introduction and Main Messages
- Scope: Management of government spending on employee compensation in small developing states (SDS).
- Key premise: Employee compensation constitutes a substantial share of government spending in SDS and is central to fiscal sustainability and service delivery.
- Main messages and guiding principles:
  - Small states face specific challenges driven by diseconomies of scale and greater openness and volatility.
  - Flexibility is required to overcome diseconomies of scale and address inefficiencies; benchmarking should prioritize the domestic private sector but may require cross-country comparisons in cases of acute “brain drain.”
  - Consolidation of functions in central government can mitigate diseconomies of scale but implies trade-offs.
  - Small states have used varied approaches: prioritization of essential sectors, improved training, migration management, GovTech, remote-location adaptations, and cooperation with regional peers or larger countries.

### Definitions and Country Groupings
- SDS definition: low-income and emerging market economies with a population of less than 1.5 million.
- Microstates: population below 200,000.
- Small advanced economies: countries in the highest income group with population below 3 million.
- Government compensation: total wages, salaries, allowances, and any other benefits of government employees.

### How Is Government Compensation and Employment Different in Small States? — Key Findings
- Core drivers of difference:
  - Diseconomies of scale in public service delivery.
  - Greater openness and volatility.
  - Additional factors: higher risk of natural disasters and rising sea levels, low population density, high population dispersion, remoteness, narrow economic bases (tourism, fishing, mining, agriculture).
- Compensation spending (averages, latest values available):
  - Small states: 12 percent of GDP.
  - Microstates: 16.5 percent of GDP.
  - Advanced economies: 9.4 percent of GDP.
  - Emerging market economies: 8.2 percent of GDP.
  - Low-income developing countries: 6.2 percent of GDP.
- Government employment and compensation (Table 1, selected indicators, latest value available):
  - Small states (Obs.): 34; Government Spending on Compensation of Employees As a Share of GDP: 12.1; Government Employment As a Share of Public Spending: 28.3; Compensation Premium As a Share of Working-Age Population: 13.5; In % of Private Sector Compensation (Obs. 2): 41.0.
  - Microstates (Of which): Obs. 15; As a Share of GDP: 16.5; As a Share of Public Spending: 30.4; As a Share of Working-Age Population: 18.7; Palau (2014) 21.0.
  - Other small states: Obs. 19; As a Share of GDP: 8.6; As a Share of Public Spending: 26.6; As a Share of Working-Age Population: 10.7; Eswatini (2016) 61.8.
  - Small advanced economies: Obs. 10; As a Share of GDP: 10.9; As a Share of Public Spending: 26.1; As a Share of Working-Age Population: 13.9; In % of Private Sector Compensation (Obs. 7): 12.0.
  - Advanced economies: Obs. 27; As a Share of GDP: 9.4; As a Share of Public Spending: 22.6; As a Share of Working-Age Population: 12.5; In % of Private Sector Compensation (Obs. 24): 6.5.
  - Emerging market economies: Obs. 66; As a Share of GDP: 8.2; As a Share of Public Spending: 26.7; As a Share of Working-Age Population: 8.9; In % of Private Sector Compensation (Obs. 36): 14.4.
  - Low-income developing countries: Obs. 52; As a Share of GDP: 6.2; As a Share of Public Spending: 27.0; As a Share of Working-Age Population: 4.4; In % of Private Sector Compensation (Obs. 20): 15.5.
  - All (global sample): Obs. 189; As a Share of GDP: 8.7; As a Share of Public Spending: 26.4; As a Share of Working-Age Population: 9.5; In % of Private Sector Compensation (Obs. 80): 12.3.
  - Sources: FAD Government compensation and employment database and Government-Private Sector Compensation Premium Data set.
  - Note: Data as a share of GDP and Government spending refer to 2022 and 2021, respectively, in most cases. Data on government employment refer to the latest data available after 2015.
- “Too small to specialize”: small absolute government size limits specialization, making it difficult to justify full-time niche experts; multitasking can leave critical tasks underserved.
- Brain drain and limited labor pool:
  - High outward migration on average, particularly in Pacific and Caribbean.
  - Example: St. Kitts and Nevis — 2.4 nationals living abroad for each person living in the country.
  - Education abroad and non-return (Palau, Samoa) and pay differentials with larger countries exacerbate recruitment challenges.
- Minimum necessary positions and geographic necessities increase per-capita staffing costs (remote islands, full-time local posts).
- Funding and risk:
  - Many small states sustain high compensation-to-GDP ratios with substantial revenues from abroad, which are volatile.
  - Examples: Federated States of Micronesia spends 21 percent of GDP on compensation and receives substantial grants; Marshall Islands 22 percent; Palau 18 percent. Nauru spent about 21 percent of its GDP on compensation of employees in fiscal year 2022, supported by grants, fishing license fees, and revenues related to its Regional Processing Centre.

### How to Take into Account the Circumstances of Small States in Government Compensation and Employment Management — Policy Guidance
- Overarching principle: tailor standard public-sector compensation management recommendations to small-state circumstances; assess trade-offs in country-specific context.
- Managing government compensation with macroeconomic and fiscal volatility:
  - Heightened scrutiny is required given higher average compensation spending and historical large increases in small states.
  - Recommended tools (also used in other states): hiring freezes, limits on pay growth, reviews of spending; right-sizing through sectoral spending reviews (example: Eswatini).
  - Caution: major short-term reductions in employment risk losing skilled staff to emigration or other sectors; temporary pay restraint that is well communicated may have fewer long-term consequences.
  - Risk identified: using government employment as short-term social safety net can create hard-to-reverse increases in compensation spending and private-sector shortages during recovery.
  - Need for flexible yet robust frameworks that allow adaptation to shocks and natural disasters; governance, transparency, and data improvements are part of the reform package.
  - Institutional constraints: constitutional, legislative, and institutional rules in some small states reduce adaptability (example: Eastern Caribbean Currency Union Constitutions limit reducing remuneration, allowances, increments).
  - Good practice example: Vanuatu — school grant sizes proportional to enrollment and published on a dedicated website.
- Taking account of diseconomies of scale:
  - Centralization can mitigate diseconomies of scale; decentralization benefits are limited in small states while costs are greater.
  - Examples:
    - Cyprus centralized all police staff into one institution.
    - Palau: IMF technical assistance recommended considering merging some functions from federated states into national government (park policing, road maintenance, trash collection, vehicle registration).
  - Trade-offs: decentralization can limit power concentration and reduce clientelism and patronage; historical/political reasons for decentralization may make reversal difficult.
  - Regional universities as economy-of-scale solution:
    - University of the West Indies serves 17 English-speaking countries and territories with campuses in four countries and one online campus; financed by member governments based on student numbers.
    - University of the South Pacific is jointly owned by 12 members, with a campus in each and a single university structure.
  - Partnerships with larger countries can discharge state functions (postal, consular, defense) and leverage economies of scale to avoid additional compensation spending.

### Practical Approaches and Tools Used by Small States (summary of approaches mentioned)
- Prioritization of essential sectors and personnel.
- Improved training and capacity-building adapted to small-scale settings.
- Migration management policies to mitigate brain drain.
- GovTech and remote-service delivery innovations.
- Cooperation with regional peers or larger countries to support public service delivery without increasing compensation spending.

*Source: IMF | How to Note NOTE/2025/004 — "How to Manage Government-Sector Remuneration and Staffing in Small States", May 2025.*

### Box 2. Free Association and Other Agreements between Countries

### Box 2. Free Association and Other Agreements between Countries

### Free association arrangements and partner-provided public services
- Free and formal agreements between a small sovereign or self-governing state and a larger country can yield partner-provided public services.
- Examples:
  - The Federated States of Micronesia, the Marshall Islands, and Palau: Compact of Free Association with the United States. Under the Compact, the United States provides defense; services from the US Postal Service, Weather Service, and Federal Aviation Administration, among other agencies; and access to US grants and social services.
  - The Cook Islands and Niue: free association agreements with New Zealand; New Zealand has responsibility for defense and foreign affairs for both countries and acts on this responsibility only at the request of their governments.
  - Nauru: defense provided by Australia under an informal agreement.
  - Liechtenstein: citizens can rely on Switzerland’s consular services abroad; Switzerland can represent Liechtenstein’s diplomatic interests upon request.
  - San Marino: defense is provided by Italy.
- Note: The Cook Islands and Niue are not IMF members.

### Prioritization and innovative solutions with scarce resources (health and education)
- Prioritization examples and outcomes:
  - Maldives: chose to offer higher pay for health workers and employ more health workers after a large improvement in GDP and available revenue; this led to improvements in access to health services and health outcomes, but has created challenges around affordability and efficiency of spending (World Bank 2023).
  - Seychelles: prioritized health and education in public spending leading up to 2008, producing improvements in health outcomes and educational enrollment but raising continued concerns about efficiency and sustainability of spending (World Bank 2017).
- Education in Pacific SDS:
  - Blackman and others (2023): education expenditure in the Pacific SDS tends to be high or very high relative to comparator countries, whereas educational outcomes tend to be below expectations across many measures.
  - Identified challenges: quality of teaching, reading opportunities at home, and basic resources in some Pacific states.
  - Policy responses and evidence:
    - Micronesia and Tonga developed tailored and affordable materials and training to support schools in remote locations; these approaches produced significant improvements:
      - After one year, the share of proficient students increased by about 20 percent both in reading and literacy in Micronesia.
      - Average reading scores improved by 0.19 standard deviations in Tonga.

### Tailored approaches to overcome brain drain issues (compensation, training, recruitment)
- Benchmarking compensation:
  - Government compensation should generally be competitive with the domestic private sector; typically government pay can be somewhat below private-sector pay once individual characteristics are accounted for (Thévenot and Walker 2024).
  - In cases of acute competition with other countries, it may become necessary to benchmark government pay against other countries to retain staff and protect critical public services; this could imply paying government employees more than comparable domestic private-sector workers for critical skills (with trade-offs including crowding out private-sector employment and higher cost).
- Training and expanding supply:
  - Many small states may need to train additional workers to offset emigration of essential workers.
  - Nurses training in the Caribbean has been important given high emigration (World Bank 2009); in the 2000s Guyana, St. Lucia, and St. Vincent and the Grenadines expanded training places and improved training quality.
  - Approaches to get destination countries to pay for the training of emigrating nurses have been used in SVG to offset domestic training costs.
- Broader retention strategies:
  - Successful recruitment and retention look beyond remuneration to include job security, career prospects, working environment, job satisfaction, family ties, and quality of life (World Bank 2009).
  - Migration agreements were considered by Caribbean small states in the 2000s to reduce nurse outflows (World Bank 2012).
  - UK Department of Health for England introduced a 2001 code of practice banning “active” recruitment of health workers from low-income and emerging market countries unless covered by a memorandum of understanding (Khonje 2015).
- Recruiting foreign skilled workers:
  - Palau (limited domestic higher education capacity) is considering fellowship programs to recruit foreign workers into government service (Walker and others 2024). IMF technical assistance recommended Palau harmonize employment and compensation policies across domestic and foreign workers to facilitate recruitment.
  - Guyana has actively recruited Cuban nurses and doctors while gradually training more national health workers.

### Using GovTech approaches to overcome delivery challenges
- General potential:
  - GovTech approaches can serve populations difficult to reach through standard service models (Barron Rodriguez and others 2022; Amaglobeli and others 2023).
  - SDS are on average well connected, with substantial variability across countries (Cantu-Bazaldua 2021).
- Health:
  - Telemedicine and electronic medical records can enable access where otherwise challenging.
  - Bloch and Wang (2023): telemedicine, laptops/tablets, data, and drones have supported access to rural and remote areas in Brazil and sub-Saharan Africa (Blackman and others [2023] for telemedicine in the Pacific).
  - Sub-Saharan African policymakers are investing in biometric identification systems to enable data-heavy interventions (Parekh 2020).
- Education:
  - Digital tech supports remote coaching and meetings; Kotze, Fleisch, and Taylor (2019) found virtual teacher training had similar positive effects to on-site training but was slightly cheaper and less logistically challenging (Rodriguez-Segura 2021).
  - Digital libraries can provide publicly available resources across member states (Organization for Eastern Caribbean States’ digital library; Myers and others 2022).
- Public administration and payroll integrity:
  - Blockchain in Guinea-Bissau produced a single, more reliable set of compensation data, curbing discrepancies and reducing corruption opportunities through consistency and interoperability.
    - Result: the ratio of compensation spending to tax revenues declined from 84 percent to 50 percent between 2020 and 2024 (after implementation).
- Underlying digital infrastructure:
  - Investments in connectivity examples include the Caribbean Regional Communications Infrastructure Program that expanded connectivity to Grenada, St. Lucia, and SVG.

### Understanding country context and policy implications
- No single cutoff identifies excessive public compensation spending; assessments depend on:
  1. The overall level of fiscal sustainability and revenue mobilization.
  2. The distribution and approach to public service delivery (for example, between central and local government).
  3. Policy design and objectives.
- Other considerations:
  - Concerns about a large public sector crowding out the private sector vary across small countries.
  - Inefficiency in public service delivery may stem from geographic or historical challenges or misallocation of resources—each requiring distinct policy responses.
  - Institutional capacity will influence reform feasibility and appropriateness.
- IMF support:
  - The IMF’s Fiscal Affairs Department offers capacity development to help manage government employment and compensation and identify measures consistent with efficient service delivery and fiscal sustainability.

### Annex: Determinants of government compensation spending (regression evidence)
- Empirical strategy:
  - Two regression analyses were conducted:
    1. Panel regression with random effects (generalized least squares) on a panel of 189 countries between 1980 and 2022 (5,646 observations). Explanatory variables included population (log) and population (log)^2.
    2. Cross-sectional linear regression focusing on the quartile of countries with the smallest populations (sample covers 41 countries with latest data since 2019).
- Key findings from regressions (Annex Table 2.1):
  - Regression details and estimates:
    - Observations: 5,646; 5,646; 41; 38 (for Regression 1, Regression 1 bis, Regression 2, Regression 2 bis respectively).
    - R2: 0.11; 0.12; 0.67; 0.72.
    - Population (log): −11.7**; −11.3**; −2.0; −3.2**.
    - Population (log)2: 0.4**; 0.4** (present in the panel regressions).
    - GDP per capita (log): −0.2**; −0.2**; −0.6.
    - Commodity exporter (yes): 7.0* (in one specification).
    - Latin America and the Caribbean: −6.1* (in one specification).
    - Emerging and developing Asia and Remote interaction: 8.4** (significant interaction term in the small-country cross-section).
  - Interpretation:
    - Population size has a statistically significant relationship with government compensation spending in the panel regressions, suggesting a negative relationship between population size and government compensation spending.
    - The positive and significant quadratic term (population (log)2) indicates nonlinearity: at higher levels of population the negative effect of population size on government compensation is reversed (consistent with Jugl 2022).
    - In the small-country cross-section, GDP per capita is not statistically significant (possibly due to small sample size). The interaction between remoteness and classification as emerging Asian economies is significant, highlighting specificity for small, geographically isolated countries.
  - Statistical notation: dependent variable is government spending as a share of GDP; **p < .01; *p < .1. In Regression 2 bis, variables with statistically insignificant estimates were excluded.

*Source: IMF — Box 2. Free Association and Other Agreements between Countries (htnea2025004).*

---


_Source: https://www.imf.org/-/media/files/publications/howtonotes/2025/english/htnea2025004.pdf_