Journal Issue
Back Matter

Back Matter

Natalia Tamirisa, and Christoph Duenwald
Published Date:
January 2018
  • ShareShare
Show Summary Details
Annex 1. Data on Public Wage Bills and Related Institutional Arrangements for the Middle East and Central Asia

Data Sources

Data on government wage bills and key economic variables were sourced mainly from the IMF’s World Economic Outlook (WEO) database. The indicators included GDP (nominal, real, and purchasing power parity (PPP)), government expenditure (total, current, capital, and compensation of employees), government revenue, exports (total and fuel), gross government debt, demographic data (population, labor force, and unemployment rate), and commodity prices. Data on public wage bills were cross-checked against the database created for the IMF’s 2016 paper on Managing Government Compensation and Employment and the World Bank databases. The WEO aggregates for public wage bills and other fiscal indicators were for the general government.

Data on employment and compensation were collected from several other sources. Public employment and compensation data are both scarce and inconsistent in definitions and coverage. A new data set was constructed for the purposes of this paper to combine the existing information with data from national statistics websites, a survey of IMF desk economists (see below), the International Labour Organization (ILO), and the World Bank database (Annex Table 1.1). Labor market indicators for other comparator countries were obtained from the ILO as well as from the Fiscal Affairs Department (FAD) databases.

Annex Table 1.1.Public Sector Data Sources and Coverage
MENA Oil Importers
AfghanistanAfghanistan Central Statistics OrganizationNumber of Current Employees Male and Female by Ministries and Departments2002–16;
EgyptILOTotal Public Sector Employment2005–14
JordanJordan Department of StatisticsNumber of employees in the public sector2000–14
MauritaniaQuestionnairePublic sector employment2005–16
MoroccoNational Statistics OfficePublic sector employment2004–15 Staff estimates for 2004–06 obtained by extrapolating national statistics data for 2007 and ILO-based growth rates.
TunisiaCountry authorities1Public sector employment2008–16
West Bank and GazaCountry authorities1Public sector employment2007–15
GCC MENA Oil Exporters
BahrainBahrain labor market indicators published by Labor Market Regulatory AuthorityPublic sector employment2002–16
KuwaitKuwait Central Statistical Bureau, annual statistical abstracts, various issuesEmployees in government civil service (a sum of government offices and departments, attached budgets, independent budgets, and state-owned enterprises)2000–14
OmanOman National Center for Statistics and InformationEmployees in the public sector2002–13
QatarMinistry of Development Planning and Statistics, labor force surveys, various issuesEmployees in government departments and government companies/corporations2001–15
Saudi ArabiaCountry authorities12007–16
United Arab EmiratesILOPublic sector employment2007,2010
Non-GCC MENA Oil Exporters
AlgeriaCountry authorities1Public sector employment2004–13
IranCountry authorities1Public sector employment2005–15
IraqCountry authorities1Public sector employment2006–16
CCA Oil Importers
ArmeniaNational Statistics Service of the Republic of Armenia, Labor Market Indicators, various issuesPublic sector employment2001–15; data for 2001–03 provided by the authorities
GeorgiaNational Statistics Office of Georgia, Distribution of Employed by Institutional SectorEmployment in the public institutional sector2002–16; Data for 2002–05 is from thr World Bank Database2
Kyrgyz RepublicILO and World Bank database2Public sector employment2000–15Data for 2000–07 from the World Bank database, and from ILO afterwards
TajikistanNational Statistics AgencyPublic sector employment2000–15
CCA Oil Exporters
AzerbaijanState Statistics Committee of the Republic of AzerbaijanPublic sector employment2000–15
KazakhstanILOPublic sector employment2009–13

“Country authorities” refers to data provided by the authorities to IMF country teams or the IMF Fiscal Affairs Department. Data for missing years were extrapolated from the available data for adjacent years.

Note: CCA 5 Caucasus and Central Asia; GCC 5 Gulf Cooperation Council; MENA 5 Middle East and North Africa; ILO 5 International Labour Organization.

“Country authorities” refers to data provided by the authorities to IMF country teams or the IMF Fiscal Affairs Department. Data for missing years were extrapolated from the available data for adjacent years.

Note: CCA 5 Caucasus and Central Asia; GCC 5 Gulf Cooperation Council; MENA 5 Middle East and North Africa; ILO 5 International Labour Organization.

Data on structural indicators were collected from external databases. The World Bank World Development Indicators (WDI) and Global Competitiveness Index (GCI) databases were used to retrieve indicators of labor market development and business climate, as well as fuel exports. Data on security, social unrest, and political stability came from the World Bank World Governance Indicators (WGI) database.

Weaknesses in the availability and quality of data for the Middle East and Central Asia region hindered the analytical work for this paper. Some countries in the region (Libya, Pakistan, Syria, Uzbekistan) do not provide public wage bill data, which required excluding them from the sample used for this paper. The West Bank and Gaza is also excluded from aggregates because its PPP GDP data are not available. WEO data were collected for the period from 1990 to 2016. Yet for most countries in the region data start in the mid-1990s, with data for Caucasus and Central Asia oil exporters and some economies in conflict (Afghanistan, Iraq, West Bank and Gaza) start in the early to mid-2000s. Data on employment and population from national websites or submitted by country authorities covered a shorter period than the WEO data, creating additional challenges for analytical work.

Desk Survey

To enhance the data set used in this paper, a survey was developed and sent to IMF country desks to gather more detailed public wage bill data for countries in the region. The survey requested desks to provide data on civil service wages and employment; wage bills for education, health care, and security sectors; and basic salary, allowances, and social contributions. IMF desk economists filled in the survey using country authorities’ sources as well as external country-specific sources. The survey had limited success, and detailed public wage bill data for many countries was not obtained (Annex Table 1.2). The average response rate was roughly 50 percent. There is no apparent bias in the response rate based on geography; however, oil importers had an average response rate of 57 percent, while oil exporters averaged 32 percent.

Annex Table 1.2.Survey Responses Regarding Data on Public Wage Bills(Latest observations; data reflects 2000–16 averages)
Government Wage Bill (percent of GDP)Government Wage Bill (percent of total spending)Government employment (percent of working-age population)Wage BillType of CompensationEmploymentSocial Benefits
TotalEducationHealthCivil ServantsSalariesAllowances and OtherTotalEducationHealthCivil ServantsTotalEducationHealth
Gulf Cooperation Council (GCC) and Algeria
Saudi Arabia13.540.016.3
United Arab Emirates4.715.716.5
Other Middle East and North African (MENA) Oil Exporters
Middle East, North Africa, Afghanistan, and Pakistan (MENAP) Oil Importers
West Bank and Gaza15.245.00.0
Caucasus and Central Asia (CCA) Oil Exporters
Caucasus and Central Asia (CCA) Oil Importers
Kyrgyz Republic9.823.921.0
Source: National authorities; IMF staff calculations.
Source: National authorities; IMF staff calculations.

The survey was also used to collect information on the current institutional arrangements underlying government hiring and compensation. The following questions were presented to IMF desks:

  • How are wages determined?
  • How strong are public sector unions at the national level (for example, are they able to influence legislation and regulations related to labor issues)?
  • Is the percentage of public sector workers covered by the terms of public sector collective wage agreements high, moderate, or low or there are no collective wage agreements?
  • Are base wages indexed?
  • Does the government have a ceiling on the total wage bill?
  • Does the government have an overall ceiling on the total number of government employees?
  • Can government employees be made redundant/laid off?
  • Can staff be employed on a contractual basis?
  • Who approves the creation of new staff positions?

The response rate for these questions was higher than that for the detailed public wage bill data, averaging close to 70 percent across all questions (Annex Table 1.2). There is no apparent bias in the response rate based on geographic or other factors (for example, whether a country is an oil exporter or importer).

Despite high cross-country variation in institutional arrangements, there are some areas of commonality (Annex Table 1.3). These include (1) the lack of wage indexation, (2) highly centralized decision making on wage increases, and (3) hiring of contractual labor. A systematic difference is also apparent between countries in the Middle East, North Africa, Afghanistan, and Pakistan (MENAP) and in the Caucasus and Central Asia (CCA). CCA countries tend to have some ceilings on government employment and wage bills. They also permit laying off public employees, and decisions on creation of new positions are highly centralized, either through the president/prime minister or the legislature. By contrast, MENAP countries (particularly in the Gulf Cooperation Council [GCC]) have broader decision-making bodies for the creation of new positions, usually at the departmental level or through a commission. These countries have few or no established ceilings on employment and the overall public wage bill.

Annex Table 1.3.Survey Responses to Institutional Questions(Latest observations)
Are wage increases centralized?Are labor unions strong?Are public employees covered by collective wage agreements?Are base wages indexed?Is there a ceiling on total wage bill?Is there a ceiling on public employment?Can government employees be laid off?Can staff be employed on a contractual basis?Who approves the creation of new positions?
Gulf Cooperation Council (GCC) and Algeria
Saudi Arabia
United Arab Emirates
Other Middle East and North African (MENA) Oil Exporters
Middle East, North Africa, Afghanistan, and Pakistan (MENAP) Oil Importers
West Bank and Gaza
Caucasus and Central Asia (CCA) Oil Exporters
Caucasus and Central Asia (CCA) Oil Importers
Kyrgyz Republic
No responseCivil Service Commission
President/Prime Minister
Source: National authorities; IMF staff calculations.
Source: National authorities; IMF staff calculations.

Country Groupings and Comparators

The region’s country groupings are skewed. In countries with similar characteristics, one country tends to be much larger than the rest. This can be seen in Annex Table 1.4, in which Kazakhstan is by far the largest country among CCA oil exporters, Egypt among MENAP oil importers (the relevant data for Pakistan, another heavyweight in this grouping, are missing), Saudi Arabia among the GCC countries, and Iraq among other MENA oil exporters (Iran was not included in this group because of idiosyncrasies). These large-weight countries account for at least 45 percent of the respective subgroup’s weight. The most balanced category is CCA oil importers, where Armenia, Georgia, and the Kyrgyz Republic have broadly similar weights.

Annex Table 1.4.Weights for Middle East and Central Asian Countries, by Grouping, 2000–16(Percent, period average)
GCC and AlgeriaAlgeria16.4
Saudi Arabia45.2
United Arab Emirates17.6
Other MENA Oil ExportersIraq77.6
MENAP Oil ImportersAfghanistan2.4
West Bank and Gaza0.0
CCA Oil ExportersAzerbaijan22.7
CCA Oil ImportersArmenia24.9
Kyrgyz Republic20.7
Source: IMF staff calculations.Note: Based on purchasing-power-parity GDP. Zero weights indicate missing data. CCA 5 Caucasus and Central Asia; GCC 5 Gulf Cooperation Council; MENAP 5 Middle East and North Africa, Afghanistan, and Pakistan.
Source: IMF staff calculations.Note: Based on purchasing-power-parity GDP. Zero weights indicate missing data. CCA 5 Caucasus and Central Asia; GCC 5 Gulf Cooperation Council; MENAP 5 Middle East and North Africa, Afghanistan, and Pakistan.

The emerging market and developing economies country group corresponds to the WEO definition. The full list of countries can be located at the WEO portal on the IMF website:

The fuel exporters group represents countries with average net oil and gas exports exceeding 1 percent of GDP during 2007–15. The data on net fuel exports were sourced from the World Bank World Development Indicators database. The list of countries is provided in Annex Table 1.5.

Annex Table 1.5.Fuel-Exporting Economies Outside the Middle East and Central Asia
CountryNet Fuel Exports, 2007–15 (percent of GDP)
Brunei Darussalam56.0
Trinidad and Tobago28.1
Source: World Bank, World Development Indicators.
Source: World Bank, World Development Indicators.

Some key observations for countries outside the region are missing. Among advanced economies, these include Japan and the United Kingdom, which represent, respectively, 12 percent and 6 percent of the advanced economy grouping. Canada and the United States also have missing data for the latest year, 2016. Among emerging market economies, China and Mexico are the largest countries with missing data, representing 32 percent and 6 percent of the emerging market group, respectively.

Several large countries stand out in global groupings because they carry at least a 5 percent weight. For the advanced economies these standouts are the United States (40 percent weight), Germany (11), France (8), Italy (7), South Korea (5), and Spain (5). Among emerging market economies, India (23), Russia (15), Brazil (13), Indonesia (9), and Turkey (5) stand out. Low-income countries Bangladesh (26), Kenya (7), Ethiopia (6), Tanzania (6), and Ghana (6) are notable for a higher weight.

Annex 2. Experience with Public Wage Bill Reforms in the Middle East and Central Asia


Iraq’s public spending is high in international comparison, driven by the public wage bill. The public sector payroll lacks centralized control and, as a consequence, has been growing rapidly. The number of positions financed may not reflect the number of employees performing a meaningful job in the absence of systems that would prevent absenteeism or fraud. Relatively high public sector wages and job security make other alternatives appear undesirable, curtailing private sector job growth.

Institutional Background

The central government has limited control over the actual size of the public sector. There is no central authority exercising hiring, promotion, redundancy and retirement rights. While a legal framework establishing the condition of employment exists, the extent to which spending units observe the rules is difficult to control. There is no central registry of public sector employees, and there is a discrepancy between reported payrolls financed by the budget and the actual number of workers.

In terms of the size and cost of the payroll, significant uncertainty is attributable to sectoral and governance issues. The size of the defense and security sector is not fully known, partly because of the changing relationship between the government and non-governmental militias and partly because the reporting standards in this sector are even less developed than in the civilian areas, with potentially large numbers of fictitious employees and rank inflation. The incomplete flow of information between the central government and the Kurdistan Regional Government also makes it difficult to establish the size and structure of public sector employment in the country as a whole.

Allowances represent a high portion of the compensation package, weakening the performance-remuneration link. Allowances range between 35 percent and 150 percent of the base salary, with the relative weight of allowances growing with grades. Given that some allowances are linked to marital status, number of dependents, and other socioeconomic factors, total compensation does not reflect position or performance. This compensation profile ensures that the wage premium—including allowances and supplements—compared with the private sector is constant, at approximately 11 percent, in all grades.2

Trends and Issues

Over the past 10 years, Iraq’s public wage bill has grown very large, fed by abundant hiring and generous remuneration policies. In response to the recent plunge in oil prices, Iraqi governments have made large expenditure cuts, but these have focused on investments, goods, and services and have left personnel and social expenditures structurally unchanged. Any reductions in salaries, pensions, or social welfare transfers stemmed mostly from the government’s inability to pay. By 2016, Iraq’s spending on public sector compensation (wages and pensions) had grown above that of various comparators: geographic comparators (Middle East and North Africa, MENA), oil-exporting emerging economies, and transition economies (Annex Figures 2.1 and 2.2).

Annex Figure 2.1.General Government Spending by Expense, 2016

(Percent of total expenditure)

Sources: Country authorities; IMF, World Economic Outlook; and IMF staff estimates.

Note: EMs = emerging market economies; MENAP = Middle East and North Africa, Afghanistan, and Pakistan; OECD = Organization for Economic Cooperation and Development.

Annex Figure 2.2.Iraq: General Government Spending by Category, 2004–16

(Percent of total expenditure and net lending)

Sources: Country authorities; and IMF staff estimates.

While both the expansion of the payroll and the increasing real compensation levels contributed to the growing wage bill, the payroll played a more important role. The wage bill rose almost 2.5 times between 2006 and 2016, and the real value of employee compensation increased by 40 percent in the same period (Annex Figure 2.3). Iraq, on an international scale, has become an outlier in terms of the size of its wage bill (Annex Figure 2.4).

Annex Figure 2.3.Iraq: General Government Wage Bill, 2006–16

(Index 2006 = 100)

Sources: Country authorities; and IMF staff estimates.

Annex Figure 2.4.Wage Bills and Public Sector Employment, 2016

(Percent, latest available data)

Sources: Country authorities; IMF, World Economic Outlook; and IMF staff estimates.

Note: Data labels in this figure use International Organization for Standardization (ISO) country codes. EMDE = emerging market and developing economies; MENAP = Middle East and North Africa, Afghanistan, and Pakistan; OECD = Organization for Economic Cooperation and Development.

The largest employers are the education sector,3 defense, and interior ministries (Annex Figure 2.5). These comprise approximately 1.8 million employees, or 78 percent of the total payroll.4 Whereas security-related employment is large as a share of the total, payroll expansion in this sector took place mostly before 2011 and has continued only among non-security-related ministries since then.

Annex Figure 2.5.Iraq: Public Sector Employment by Ministry, 2016

(Thousands of people)

Sources: Country authorities; and IMF staff estimates.

In addition to the expansion of the payroll, the growth of the wage bill has been driven by an increase, in real terms, of average compensation per employee. The real value of the average total compensation received by public employees grew, between 2006 and 2016, by 60 percent, and the number of public employees also grew over 50 percent. The overall picture conceals differences between the federal government and the Kurdistan Regional Government: whereas the government payroll went up by 64 percent in the institutions under Baghdad’s control, the same indicator increased by 22 percent in those controlled by the Kurdistan Regional Government.

Reform Challenges

To reduce the public sector wage bill and to control the drivers of its potential future growth, the government needs to implement measures addressing both the size of the payroll and, to a lesser extent, the level of compensation. Given that wage bill growth observed over the past 10 years was driven primarily by payroll expansion, reform measures should focus mostly on this aspect of wage bill management.

Although the low-inflation environment makes it harder to reduce real compensation, restructuring and recalibrating public sector wages and supplements can make room for curtailing further wage growth. The public sector wage premium is relatively low at approximately 10 percent. The lack of private sector alternatives and the availability of nonwage benefits (such as health insurance and eligibility for generous old-age pensions) allow for a gradual decrease in allowances without risking a large-scale exodus from public employment. Efforts in this area should focus primarily on strengthening the relationship between performance and compensation. But such measures should not rule out such emergency measures as a temporary cap on allowances, freezes on promotions, and seniority wage increases.

Sector-specific approaches and a business environment conducive to private sector employment will be needed to address the oversized public sector payroll. Addressing the oversized public wage bill, while improving the quality of public goods provided by the state will require sector-specific approaches. For instance, health care is critically understaffed, whereas employment levels in secondary and higher education can be justified only if enrollment rates improve. Demobilization is expected to exert supply pressure on the labor market, and the government must resist the fiscally irresponsible urge to expand the public sector and employment by state-owned enterprises. Such a differentiated approach will require a functional and procedural review and a medium-term human resource strategy. It will also mean improving the business environment to generate jobs for Iraq’s growing labor force and those employed in the public sector whose jobs cease to be necessary or affordable.

In cases of both voluntary and involuntary separation, risks to service quality should be considered. Voluntary departures may lead to various risks, including adverse selection (the most productive and best-skilled employees leave), improperly priced severance packages (for example, overly generous or insufficiently generous), and the risk of rehiring (especially when such programs are inappropriately targeted or are implemented without sufficient analysis of future staffing needs).

In terms of short-term employment measures, the government needs to implement a staff reduction plan through natural attrition. This approach would not amount to a full hiring freeze but would prevent automatically filling positions that become vacant due to retirement or other reasons. The effectiveness of this measure depends on the age structure of public employees and the hiring flexibility retained at higher decision-making levels. In cases where clear staffing indicators can be established (such as population-health personnel ratios and minimum operating staff numbers), the government may link the expiration of the partial hiring freeze to reaching these indicators.

The government needs to urgently complete a payroll census. Under the current Stand-By Arrangement, the government will conduct a payroll audit. Given the number of public employees, it is crucial to establish a risk-based audit timetable whereby a sample would first inform the Board of Supreme Audit of the sectors where fraud is most frequent and then focus on these areas early on. It is equally important to have a standardized action plan in case of identified cases of fraud, with clearly established legal and financial consequences affecting not only the worker involved (in the form of termination of employment) but, more important, officials who were or who could reasonably be expected to be aware of the fraud.

The government may need to consider outright staff reductions, coupled with voluntary early retirement, as a one-off measure. This measure could be best implemented in sectors where service outcomes are unrelated to population growth, economic activity, or the security situation or where service has been reduced or contracted out to the private sector. At the same time, voluntary retirement programs can achieve their objectives only if there is a credible threat of layoff.

Hiring decisions, as a matter of commitment control, should be subject to fiscal authorization. For the sake of simplification, this may take the form of joint authorization by the hiring department and the Ministry of Finance. Over the medium term, a different institutional arrangement should be introduced. Uniform hiring, promotion, and termination rules would be handled by a central human resource management unit or a civil service and public employee commission. Regulations establishing legal and financial liability for unauthorized and unlawful hiring decisions—potentially retroactively with a short grace period—are needed. These would penalize decision makers and all officials who could reasonably be expected to be aware of such decisions.

To better reflect performance and to make the wage scale more transparent, total remuneration needs to rely less on allowances. A system of allowances is a common approach to blur the actual remuneration of public sector employees, to grant more discretion to supervisors for differentiating emoluments without necessarily reflecting performance, and to decrease the link between remuneration and responsibilities and performance of employees. As part of the human resource strategy, allowances should be merged with wages, restricting nonwage remuneration to performance-based, occasional bonuses.

Kyrgyz Republic5

Over the past several years, the public wage bill in the Kyrgyz Republic has been rising, reaching almost 10 percent of GDP by 2016—which is high compared with the country’s peers (Annex Figures 2.6 and 2.7). The 2011 attempt to reform the civil service was unsuccessful. There is now an urgent need for a comprehensive strategy to improve the efficiency of public wage bill spending and to contain its rising cost, given the tight fiscal constraints faced by the country. With a high debt level, the Kyrgyz Republic needs to create fiscal space to rebuild buffers for dealing with future unexpected shocks and to promote inclusive growth. Reforming public wage bill spending will require strong political commitment at the highest level.

Annex Figure 2.6.Kyrgyz Republic: Public Sector Employment by Sector, 2010–15

(Thousands of people)

Sources: Country authorities; and IMF staff estimates.

Annex Figure 2.7.Kyrgyz Republic: General Government Wage Bill, 2000–16

(Percent of GDP)

Sources: Country authorities; and IMF staff estimates.

Institutional Background

A reform strategy for the civil service was developed in 2011, but has been only partially implemented. The main objective of the strategy was to increase the wage level at the low end of the pay scale to align it to nationally identified subsistence levels. A second objective was to move toward a performance-based compensation framework. Overall, the objectives of the strategy appear to have been only partially achieved, since wage dynamics have mostly been driven by a series of large ad hoc increases in 2011, 2013, and 2015. The main outcomes of the 2011 reform were (1) an increase in the share of base salaries in total compensation and a decrease in allowances; (2) the separation of pay scales in health and education to better tailor pay progression and composition; (3) elimination of the 13th month of pay; and (4) some progress in introducing incentive-based pay by linking bonuses to performance reviews.

The government wage bill is set by the legislature as part of the overall budget process. It comprises public wage bill spending by the central government, autonomous and semiautonomous agencies, municipalities, and other public agencies that are fully or partially funded by the budget. However, it excludes wages paid by the social funds, particularly the Mandatory Health Insurance Fund (MHIF), as well as by smaller government agencies. The MHIF represents close to 20 percent of the public wage bill. The lack of an institutional framework to guide wage policies has enabled several large discretionary increases in public wages across several sectors in the past years.

The structure of public employee compensation (the shares of the base salary, allowances,6 and incentive bonuses) varies dramatically among sectors. Prior to 2011, base salary represented less than 20 percent of wages, but since the 2011 reform, the base salary share has increased and the allowance system has been streamlined. In 2015, base salary represented more than 80 percent of total compensation in some sectors, such as recreation, culture, and religion, but only about 55 percent in social security.

The education and health sectors absorb the largest share of the wage bill, yet the outcomes of education and health service delivery are uneven (Annex Figures 2.82.10). Despite the large amount of spending, outcomes in education are poor as indicated by the Programme for International Student Assessment (PISA) scores below. Outcomes in the health sector are comparable to those in the Kyrgyz Republic’s peers, but there is room for containing costs. The wage premium to work in the public sector is negative, and the level of wages across sectors varies considerably. The highest average wage is paid to government employees working in public administration, the lowest to employees in the education sector. Compared with the private sector, medical staff in the public sector appear to have the highest wages, while the opposite holds for teachers and other government workers.

Annex Figure 2.8.Kyrgyz Republic: Average Public Sector Wages, 2000–14

(Ratio to average private sector wages)

Sources: Country authorities; and IMF staff estimates.

Annex Figure 2.9.Kyrgyz Republic: General Government Wages by Sector, 2014

(Local currency)

Sources: 2014 Combined Household and Labor Force survey, Kyrgyz Republic; and IMF staff estimates.

Annex Figure 2.10. Figure 2a.10.Education Spending and Performance, latest available data

(X-axis: percent of GDP; Y-axis: units)

Sources: Country authorities; IMF, World Economic Outlook database; and IMF staff estimates.

Note: KGZ = Kyrgyz Republic; PISA = Programme for International Student Assessment.

Trends and Issues

Overall public spending on wages is high and has been rising since 2011 (Annex Figure 2.11). Increases in compensation have been the main driver behind the rising public wage bill rather than increases in employment. Average public wages have grown at a rate much faster than inflation. Specifically, wages in education grew the fastest, albeit catching up from a low base, followed by wages in the health sector; in other sectors they were close to inflation. In addition, a wage drift, such as seniority-related increases, estimated at 1.7 percent a year, has contributed significantly to public wage growth.

Annex Figure 2.11.Benchmarks for Wages and Employment, 2015

(Percent of GDP; unless indicated otherwise)

Source: IMF, FAD Expenditure Assessment Tool (EAT), excluding Uzbekistan.

Note: CCA = Caucasus and Central Asia.

1 Countries are Armenia, Azerbaijan, Georgia, Kazakhstan, Tajikistan, Turkmenistan and Uzbekistan.

Notwithstanding the 2011 reforms, a comprehensive approach to public wage bill management is lacking. The wage bill has surged due to discretionary increases in compensation, which makes it difficult to manage public wage bill spending and even more difficult to contain it. The country lacks a clear policy framework for setting public wages and employment and, more broadly, a medium-term strategy for its civil service. This reflects in part the absence of systematic information on government employment and compensation, particularly a comprehensive registry of public employees. Medium-term fiscal planning has been in place since 2009, but there is no link between medium-term wage policies and the budgeting process. The lack of a human resource management information system is also an impediment to proper public wage bill management. With each service delivery unit being responsible for its own payroll management, control over overall public payroll is weak.

Given its high debt level, the country needs to create fiscal space to avoid crowding out expenditure, promote inclusive growth, and build up buffers for unforeseen future shocks. Fiscal consolidation is necessary to keep the country’s debt sustainable. Containing the public wage bill is one of the most important elements of the consolidation program because the public wage bill is rising without producing the desired public service outcomes, especially in education. In that area, increasing access to postsecondary education and vocational training would be key to reducing skills mismatches and boosting the country’s medium-term growth prospects (Said, forthcoming).

Reform Challenges and Lessons

Given the rising trend in public wage bill spending, the authorities decided to take action in 2016. In the context of the latest Extended Credit Facility arrangement, which started in 2014, IMF staff members have been supportive of the authorities’ initiative to reduce the public wage bill as a share of GDP and have included this measure in the program’s conditionality. Although some basic measures, such as hiring or freezing salary increases, can deliver wage bill adjustment in the short term, over a longer horizon, only structural measures can put spending on a sustainable path. The main objectives of structural wage bill reforms are to enhance efficiency and control wage spending, get wage policies in sync with overall fiscal targets, and increase the predictability of wage dynamics. The IMF recommended several short- and medium-term measures to achieve the wage bill reduction without having to lay off employees or cut wages, including (1) only partially replacing departing workers, (2) limiting growth of allowances and bonuses as well as of base salaries to inflation, and (3) suspending seniority-based allowance and promotion increases.

The authorities recently designed an action plan to reform the staffing and remuneration policy, which is expected to reduce the public wage bill over 2016–19. The first step envisaged under the action plan is to create an inventory of all employees along with their wages and allowances. This should help identify and eliminate inconsistencies in the overall remuneration. Next, the authorities plan to conduct a functional analysis of the different sectors to streamline the number of employees and identify outsourcing options. The last stage of the reforms includes the development of an automated system, which is expected to be integrated with the financial management information system to control and manage the wage bill efficiently.

So far, progress in reforming the public wage bill has been slow, and tangible results have not yet materialized. Since 2017 was an election year, , reforms in this sensitive area have lacked traction. Experience from other countries shows that political will at the highest level is key for successful implementation of a challenging reform agenda in the area of public wage bill management. In addition, to be successful, reform should be comprehensive, well-sequenced, and coordinated, with a carefully thought-out communication strategy.


Since 2011, Tunisia has made great strides politically, yet economic progress has lagged—slow job creation, high unemployment, and security threats (terrorism and the conflict in Libya) have added fuel to social discontent. Macroeconomic vulnerabilities have been rising as low growth, exogenous shocks, and policy slippages have led to high fiscal and external deficits and accumulating debt. One of the main factors contributing to the deficits has been Tunisia’s rapidly expanding wage bill: today it stands at 14.1 percent of GDP, representing about two-thirds of tax revenue and about half of total expenditure. The government intends to comprehensively reform the civil service to reduce the wage bill to 12 percent of GDP by 2020, based on wage restraint, limits on recruitment, and voluntary head count reduction.

Institutional Background

The central government wage bill is the largest spending category of the budget, accounting for about half of total expenditure. The Parliament’s annual budget law contains the central government wage bill, covering wage spending by central administrative units, executive agencies, and other agencies that depend on budget financing. It excludes the wages of local governments and state-owned enterprises.

Collective bargaining determines wages in the public sector. The Union Générale Tunisienne du Travail (UGTT)—the umbrella organization of sectoral unions—negotiates on behalf of all public employees with the government, typically over a three-year period. In addition, sectoral unions negotiate in a decentralized manner sector-specific pay with line ministries and the government. Since 2011, the bargaining process has led to several exceptional pay increases to respond to social tensions.

Recruitment is based on competitive procedures. The Statute of the Public Service mandates a competitive and centralized recruitment process (concours) for entry into the public service (Assemblée Nationale, 1998). However, recruitment became more decentralized since 2011, as line ministries and agencies hired directly, effectively bypassing competitive recruitment processes. New recruits are hired on a permanent basis, after an initial internship of one year (new graduates and contractual workers) or two years for all other civil servants. Overall, hiring tends to be based on ad hoc requests instead of a systematic evaluation process that assesses personnel needs.

The current framework does not adequately support redeployment nor encourage mobility. A law authorizing redeployment was passed in 2014, but the implementing regulations have met with significant resistance from unions.

Tunisia’s public sector compensation system is complex. It combines a base salary with numerous allowances, which vary according to employees’ professional group, category or unit, and rank. Allowances make up about 60 percent of average pay. All public servants receive common allowances, such as for mileage and family members, and a performance bonus. Some allowances are specific to various professional groups and statutes and others apply on special occasions (such as religious holidays and the beginning of the school year).

Promotions are largely based on seniority, making them almost automatic. Public employees automatically advance a level within their grade based on a timetable defined for each professional group (usually 1–2 years). The annual performance rating lacks transparency: the assessment often reflects personal sentiment instead of a transparent comparison of achievement of previously agreed on work objectives.

Trends and Issues

Tunisia’s wage bill increased to 14.1 percent of GDP in 2016 (from 10 percent in 2010) and is now among the highest in the world (Annex Figure 2.12). Tunisia’s public wage bill has traditionally been high because governments used public sector employment as a means of distributing public resources and securing political support. After 2011, this trend became more pronounced when successive governments resorted to hiring and wage increases, partly in response to social pressure. Annex Figure 2.13 decomposes the increase in the wage bill since 2011 into contributions from employment and wages. The figure suggests that the authorities’ policy response since the revolution went through two phases:

  • Strong recruitment (2011–14). In the years following the 2011 revolution, the number of employees in the civil service rose from 430,000 to about 590,000 by the end of 2015, an increase of about 35 percent.8 Hiring reflected in large part social demands to improve the status of contractual workers and compensate families of political opponents. Hiring slowed in 2015, except for more security personnel as Tunisia stepped up its efforts against terrorism.
  • High real wage increases (2015–16). The government and the influential labor union UGTT agreed on several wage settlements for all civil servants, the most recent one covering 2016–18. In addition, specific professions benefited from additional improvements in pay and conditions. As a result of the multitude of wage settlements, average nominal wage growth in the civil service reached 12 percent in 2016, about 7 percentage points above the average inflation rate. Yet the average increase conceals significant differences across civil servants. For example, primary school teachers received pay increases of 11 percent, whereas medical doctors saw increases of only 5 percent.

Annex Figure 2.12.Tunisia: General Government Wage Bill, 2000–16

(Percent of GDP)

Sources: Country authorities; and IMF staff estimates.

Annex Figure 2.13.Tunisia: Contribution to Yearly Wage Bill Growth, 2011–16

(Thousands of dinars)

Sources: Country authorities; and IMF staff estimates.

Significant wage premiums arose in the public sector. The overall public sector wage premium compared with private sector wages was about 20 percent, based on detailed household salary information drawn from the 2012 labor force survey. At 34 percent, the wage premium was even higher for employees of state-owned enterprises. Split across levels of education, the wage premium is highest for university graduates and close to zero for workers without a high school diploma. In fact, the public sector (including state-owned enterprises) employs a disproportionate share of university graduates; attractive jobs in the private sector remain too rare. About 30 percent of university graduates are officially unemployed (twice as many as the national average), often waiting for an opportunity to enter the public sector and earning meager wages in the informal economy in the meantime. The positive wage premium therefore reflects Tunisia’s unemployment challenge.

Strong growth in government employment and compensation has been associated with a decline in productivity and negative spillovers to the private sector (Annex Figures 2.142.16). Labor productivity in the civil service (measured as value added per worker) declined by about 10 percent between 2010 and 2015. The productivity slump is related to strong hiring during 2011–13, especially for low-skill jobs. Tunisian graduates tend to continue to prefer a career in the civil service over a job in the private sector. The resulting high reservation wages tend to slow entrepreneurial initiative, dent overall competitiveness, and contribute to the slow pace of job creation in the private sector.

Annex Figure 2.14.Tunisia: Nominal Wages, 2010–16

(Thousands of dinars; unless indicated otherwise)

Sources: Country authorities; and IMF staff estimates.

Annex Figure 2.15.Tunisia: Public employment, 2010–16

(Thousands of people)

Sources: Country authorities; and IMF staff estimates.

Annex Figure 2.16.Tunisia: Productivity Growth in Public Sector, 2011–15


Sources: Country authorities; and IMF staff estimates.

Reform Challenges and Lessons

The authorities have expressed commitment to bring down wage bill spending to 12 percent of GDP by 2020, with the aim of channeling freed-up resources to priority public investments that stimulate inclusive growth. Wage restraint is expected to deliver about half of the total adjustment. The recently adopted comprehensive civil service reform strategy contains the following priorities:

  • Wage setting: When the current wage agreement expires in 2019, the government intends to consider civil service wage increases only if economic growth outperforms the current forecast. Equally important, the government will simplify the compensation system to reduce the number of profession-specific bonuses and allowances (currently more than 100 regimes exist).
  • Hiring control: The number of people hired fell below the number of retirees in 2017, leading to a permanent reduction in head count. In 2018, the government plans to implement a hiring freeze. The office of the Head of Government will have full control over all hiring decisions in line ministries, ensuring consistency between personnel plans at the departmental level and central government objectives.
  • Departure programs: The government has committed to identifying up to 25,000 employees during 2017–18 who are willing to leave the civil service, which will save the public wage bill about 0.4 percent of GDP. Two types of departure programs are planned: (1) A voluntary separation program for employees younger than 57 will offer a compensation package in return for permanently leaving the civil service (the authorities are currently seeking financing for the one-off costs). (2) An early retirement program for workers 58 and older will advance by one or two years the date of retirement. Unlike the voluntary departure program, the early retirement program will not lead to significant net savings over time as higher pension transfers will offset the savings in wages.
  • Human resource management: Functional reviews of the main ministries will systematically identify areas for efficiency gains. A dashboard, fed by a modern information technology system, will monitor the allocation of human resources across ministries and functions. Finally, the creation of la haute fonction publique will improve performance management and recruitment for strategic positions in the civil service.

Reforming the civil service will require overcoming economic and political obstacles. Tunisia’s economy currently suffers from high unemployment, slow job creation, and strong social pressure to improve living conditions quickly. In this environment, many stakeholders question the rationale for reducing the role of the civil service as the country’s most stable employer. Many Tunisians fear that less employment in the public sector will add to unemployment, exacerbating the already difficult social climate. Forthcoming comprehensive civil service reform must therefore coalesce into a broad coalition in favor of the reform. For that reason, the reform will emphasize the expected improvements in public service quality and the gains from higher public investment, including in the form of more jobs in the private sector.

United Arab Emirates9

A key goal of the UAE National Agenda is a more diversified and productive economy supported by a skilled and healthy population. To distill lessons from international experience, it is crucial to consider the Emirates’ special characteristics, including a relatively young population, a large share of migrants in the labor force, and the importance of the private sector in the provision of health and education services. Accounting for these characteristics, the analysis suggests that (1) better educational outcomes are possible while containing costs, (2) a gradual increase in public resources for health care may be necessary to reach the Agenda’s goals, and (3) equitable outcomes require coordination to equalize education and health care standards and outcomes across regions and systems.


Public expenditure on education is high relative to Organisation for Economic Co-operation and Development (OECD) peers. Education spending is estimated at 1.6 percent of GDP a year. Although at first glance this seems low compared with the OECD average of 4.5 percent of GDP, a proper comparison requires adjusting for large differences in the role of the private sector (in the United Arab Emirates, 80 percent of primary and secondary students are enrolled in private institutions compared with 31 percent in OECD countries) and demographics (in the Emirates, the school-age population is 17 percent of the working-age population compared with 29 percent in the OECD) (Annex Box 2.1). Accounting for these differences, the United Arab Emirates spend more on public education than the most generous OECD countries (Norway, Denmark, Finland). Expenditure per student is above $22,000, more than twice the average in OECD countries (Annex Table 2.1).

Annex Table 2.1.Public Education Spending
Adjustment for private sector size and demographics
Public education spending (percent of GDP)Percent of students in private educationSchool-age population to population 20–64 years oldAdjusted public education spending (percent of GDP)Expenditure per student ($)Students per teaching staffPISA scores (average of science, reading, and math)
OECD average4.516290.99,05214492
Sources: UAE Ministry of Education; Organisation for Economic Co-operation and Development; and IMF staff calculations.Note: OECD = Organisation for Economic Co-operation and Development; PISA = Programme for Interational Student Assessment; and UAE = United Arab Emirates.
Sources: UAE Ministry of Education; Organisation for Economic Co-operation and Development; and IMF staff calculations.Note: OECD = Organisation for Economic Co-operation and Development; PISA = Programme for Interational Student Assessment; and UAE = United Arab Emirates.

However, education outcomes are substantially lower than those in the OECD peers. The considerable resources devoted to education have not yet translated into strong outcomes. For example, the UAE’s PISA scores are at the bottom of those in the OECD economies. Importantly, in all subjects, more than 40 percent of students are at or below level 2—a proficiency level deemed by the OECD as necessary to participate fully in a globalized world.

Going forward, the challenge is to address the performance gap within the same resource envelope while ensuring equitable outcomes. In the United Arab Emirates’ institutional context, the Ministry of Education sets the broad national guidelines and regulations and administers public schools in Dubai and the Northern Emirates; the Abu Dhabi Education Council (ADEC) manages public schools and oversees private schools in Abu Dhabi; and the Dubai Knowledge and Human Development Authority is responsible for the quality of private education in Dubai.

  • Increasing the efficiency of education spending: The performance gap is not explained by expenditure levels. As a share of GDP and in terms of inputs (student-per-teacher ratios are comparable to the OECD levels), the United Arab Emirates are well positioned to achieve strong outcomes. Ongoing initiatives could help bridge the performance gap without raising costs by implementing new quality standards for teachers and schools, developing curriculums, introducing a common framework for school evaluation, and promoting science, technology, engineering, and math as well as innovation. As performance monitoring is enhanced, granting greater autonomy to schools could be considered.
  • Ensuring equitable outcomes: The education system is largely divided between the types of providers (public schools with 27 percent of students, private schools with the rest), curriculum (Ministry of Education or international), and jurisdiction (Abu Dhabi with 35 percent of the students, Dubai and the Northern Emirates with the rest). The fragmentation of the education system risks causing disparities in outcomes. For example, in Dubai, private schools that follow international curricula perform better on the PISA tests than private or public schools that follow the Ministry of Education curriculum. Furthermore, even within public schools, there are substantial differences: average spending per student is estimated to be 50 percent higher in the ADEC public schools than in other public schools. Furthermore, differences in gender outcomes are rising among nationals, with lower rates of secondary school graduation and postsecondary enrollment for males. It is crucial to continue monitoring the evolution of these outcomes, aiming to lift all boats. To this end, the establishment of a national exam (piloted in 2017 and planned to be adopted in 2018) should help monitor the evolution of performance across jurisdictions (Abu Dhabi, Dubai, and other Emirates) and systems (public/private).

Health Care

Public health care expenditures have been rising but remain lower than in OECD peers. Public health care expenditures in the general government increased by more than 0.5 percentage point of GDP during 2011–16, largely reflecting efforts by the government of Abu Dhabi to improve the health care infrastructure. Overall, the World Health Organization estimates public health expenditures at 2.6 percent of GDP, substantially below OECD peers, even after accounting for the younger population in the United Arab Emirates (Annex Box 2.1). At nearly $1,500 a year, total health expenditures per capita are also below the $3,800 OECD average (about $2,200 adjusted for demographics) (Annex Table 2.2).10

Annex Table 2.2.Public Health Spending
Adjustment for demographics
Public health spending (percent of GDP)Share of population age 0–4Share of population age 60 and olderAdjusted public health spending (percent of GDP)Total health expenditure per capita ($)Doctors (per 100,000 population)Nurses (per 100,000 population)Healthy life expectancy (years)Infant mortality (per 100,000 births)
Sources: World Health Organization; United Nations; and IMF staff calculation.Note: OECD = Organisation for Economic Co-operation and Development; and UAE = United Arab Emirates.
Sources: World Health Organization; United Nations; and IMF staff calculation.Note: OECD = Organisation for Economic Co-operation and Development; and UAE = United Arab Emirates.

Fiscal pressure associated with health care is likely to rise. Better health care outcomes call for higher inputs, including raising the number of doctors and nurses over time. To approach OECD levels, this would cost as much as 1.3 percentage point of GDP over the next few years. A gradual approach is warranted given fiscal and implementation constraints, and additional resources must demonstrably improve outcomes. Nevertheless, the gap in outcomes and expenditures compared with OECD countries highlights the importance of protecting health care expenditures during fiscal consolidation.11

Achieving greater integration across health care models could improve service delivery. Currently, the health care system is a mix of a mandatory health insurance model (in Abu Dhabi and Dubai, including special insurance plans for government employees and nationals) and a government-funded model (in the Northern Emirates). A more integrated system could prevent future health care disparity and avoid duplication of services. Integration could be achieved through better coordination by insurers and service providers across regions and eventually by a national insurance market. To this end, plans for separating the regulatory and service provision arms of the Ministry of Health would be a step in the right direction, similar to those in the education sector (Italy, Sweden, and the United Kingdom also separate such roles).

Enhancing monitoring and oversight is crucial to unlock the potential for a universal insurance model at an affordable cost. Robust regulation can help maintain proper standards across public and private service providers. To contain the growth of health care costs, efforts to increase price transparency and implement episode-based payment, which bundles payments to providers by medical conditions, can increase the efficiency of service delivery. Promoting the use of health information technology, which allows for data sharing across insurers and providers, and benchmarking costs would also be useful. In the medium term, reviewing the incentive and payment structure for public providers would contain fiscal costs and improve the quality of care. Today, public providers receive a budget allocation for current and capital expenditure. Over time, these allocations could be linked to indicators of quantity and quality of care to place public providers on the same footing as private providers. Care must be exercised when considering the use of public-private partnerships to finance health care initiatives. The Ministries of Finance and Health could develop an inventory of all such partnerships, clearly identifying existing commitments and the exposure of the government (for example, through demand guarantees).

Annex box 2.1.Education and Health Care Adjustments

Public education in percent of GDP can be expressed as the product of expenditure per student to GDP per working-age population, public enrollment to total enrollment, enrollment to school-age population, and school-age population to working-age population (A2.1).

Following this identity, it is possible to adjust the OECD expenditure to account for differences in public school enrollment (multiplying expenditure by the ratio of public school enrollment in the United Arab Emirates to that in the OECD) and demographics (multiplying this result by the ratio of the school-age to working-age population in the United Arab Emirates to that in the OECD). That is, 4.5 percent of GDP X 30.7/84.4 X 16.5/28.8 = 0.9 percent of GDP. This adjusted expenditure is a counter-factual, interpreted as the level of public education expenditure the average OECD country would have assuming similar private school coverage and a demographic profile similar to that in the UAE.

Public health spending in percent of GDP can be expressed as the product of health spending per capita for the population ages 40–44 to GDP per capita and the sum of the product of the share of the population in each age group and the ratio of health spending per capita of that group to the health spending per capita for the population ages 40–44 (A2.2).

Using this identity, it is possible to control for the different demographic profiles of the UAE and OECD countries. This can be done by changing the demographic profile (the share of the population at each age group, corresponding to the first term in the sum in (2)). Using the UAE’s demographic profile would reduce health spending from 6.8 to 3.9 percent of GDP. This adjusted expenditure is a counterfactual, interpreted as the level of public health care expenditure the average OECD country would have assuming demographic profile to that in the UAE.

Annex 3. International Experience with Public Wage Bill Reforms

This annex summarizes country experiences with reforming public wage bills based on case studies undertaken by the IMF and the World Bank in the context of technical assistance, country surveillance, or IMF-supported programs (see IMF 2016b). Examples presented in the table below cover both short-term employment and compensation measures, as well as structural and institutional reforms of public wage bills. The overall outcome of these reforms is difficult to quantify because it depends on country-specific factors—such as hiring practices, type of wage indexation, and the age structure of public employees. In addition, the overall economic conditions may substantially alter the impact of wage bill reforms as, for example, wage freezes generate larger savings in a highly inflationary environment. Finally, because the reforms of the wage bill are typically introduced as a package of measures—affecting both public sector wages and employment—the contribution of individual measures is difficult to quantify.

In almost all countries covered, the key motivation for reforms stemmed from high levels and/or fast growth in public wage bill spending, which was crowding out other spending and threatening to undermine fiscal sustainability. Controlling for skills, education, and other characteristics, a high and persistent public-private pay gap also had to be addressed in some countries (Honduras, Ireland, Jamaica).

Slowing growth in public wage bills required identifying and addressing the country-specific practices driving it. For example, in France, the public wage bill was growing rapidly, partly reflecting decentralization as well as hiring and promotion practices at the local level, while in Jamaica the main reason was excessive reliance on contractual and temporary workers, who comprised a significant share of the government workforce.

When designing reforms, countries across all income groups and regions relied heavily on wage freezes. Control of public sector wages through nominal or real freezes delivered fiscal savings while narrowing the gap between public and private wages (Côte d’Ivoire, Honduras, Ireland, Jamaica, Kenya, Moldova, Netherlands, Portugal, Romania). In Ireland, wage reductions resulted in the wage bill falling by about 4 percentage points of GDP by 2015,1 but frozen wage scales did not prevent public wages from a significant increase in France. The potential gains from freezing nominal wages and wage scales depend on the size of inflation. For example, wage-bill-to-GDP ratios diminished by less than 0.5 percentage point of GDP in the case of France and the Netherlands thanks to very low inflation rates after 2010. The potential gains from wage freezes are larger in countries with higher rates of inflation. Nevertheless, fully freezing public wages in a highly inflationary environment may dramatically erode the purchasing power of public employees after a few years. However, the extensive use of wage freezes may have made public sector jobs less attractive and undermined the morale of public workers. In addition, wage freezes often led to nontransparent increases in bonuses and allowances and further reinforced distortions in the public compensation system (Côte d’Ivoire, Ghana, Portugal). In several countries (Côte d’Ivoire, Jamaica, Moldova, Netherlands, Portugal, Romania, Senegal) the effect of wage freezes was partly or fully reversed by abandoning these measures and returning to previous practices, such as in Portugal and Romania, where the impact of wage freezes and reductions was reversed by constitutional courts, resulting in recurring wage pressure.

To strengthen control over the wage bill and to provide competitive, equitable, and transparent remuneration, some countries reformed government compensation structures. In Ghana, Portugal, Romania, and Zambia, reforms focused on consolidating allowances and bonuses into base pay and mapping various professional categories and sectors to a common base pay. The implementation of these reforms is typically gradual and requires adequate administrative capacity—and it may also involve up-front wage costs, as happened in Ghana and Portugal. In addition, in South Africa and Romania reforms of compensation structures did not deliver expected reductions in wage pressures.

Some countries consolidated employment levels by limiting new hiring and by relying on attrition-based employment reductions. Over the long horizon, the size of the public sector should develop in line with the size of the population or the working age-population. Temporarily limiting public employment growth to levels below population and/or working-age population growth may significantly reduce the public wage bill in young economies whose populations are growing rapidly—for example, Millennium Challenge Account countries. As part of fiscal consolidation efforts, Ireland relied on a public sector moratorium on recruitment and a number of incentivized retirement plans. In France, employment was cut through the introduction of employment caps beginning in 2006 and the implementation of attrition targets (replacing only 1 of 2 retiring civil servants) during 2007–12. Moreover, as many of those measures were phased out or reversed, the wage bill rose slightly again. To attain desired employment levels, governments in Portugal and Romania replaced only a fraction of all retirees, while in Ghana, Kenya, Moldova, and South Africa attrition rules were implemented in specific sectors.

A few countries embarked on public sector restructuring over the past decade. Effective restructuring takes time, requiring technical capacity and strong political will. In Honduras, to strengthen oversight and to identify irregular (ghost) workers, the government implemented a census of public employment, and in combination with wage and hiring freezes, the wage bill declined by about 2.5 percentage points of GDP during 2014–15. In Portugal, the government also made structural changes to improve oversight, address inefficiencies, and increase flexibility in the management of human resources. One important step was the development of a comprehensive government employment and remuneration database in 2011. This enhanced oversight and transparency and became an important tool in identifying inefficiencies and potential areas for reforms.

Strengthening human resource management can reduce the wage bill while improving control over employment and compensation. In Malaysia, the government decided to rebalance the highly centralized human resource management system and to give spending ministries more flexibility over staff management to achieve better outcomes. To examine the pay and working conditions of civil servants and make recommendations on the best practices, a remuneration commission was created in South Africa in 2013. In Honduras, the government implemented a census in 2014, followed by automation of payroll payments to validated workers. In many countries in sub-Saharan Africa, censuses also proved to be key for identifying ghost workers and paved the way for removing them in the 1990s. More recently, new technology, such as biometric monitoring of government employees, has proved effective in identifying ghost workers, as implemented in Kenya and Nigeria. Nevertheless, the process of developing effective payroll management systems can take a long time. As in Jamaica, a full roll-out of a centralized human resource management and payroll system was expected to take 4–5 years. In Portugal, a new wage grid that integrates all careers in a single wage scale and replaces the existing arrangement of over 115 base wage levels and exceptions was adopted in 2014. Yet the roll-out of the reform has been postponed, and it will not generate savings in the short and medium term.

The short-term impact of the reforms on the size and dynamics of the wage bill was in general positive and significant. Specifically, wage and hiring freezes proved effective in containing growth of the wage bill in the short run. However, the reforms did not provide expected and lasting savings for countries such as Côte d’Ivoire, France, Zambia, and Romania—often because reform was phased out or reversed, as in France, Romania, and Portugal. Country experience suggests that a medium-term approach to wage bill management supplemented by structural reforms of public employment and pay are key for sustaining short-term gains from the reforms.

Annex Table 3.1.International Experience with Public Wage Bill Reforms
IssueMeasures ImplementedImpactLessons
Cote d’lvoire- Steadily growing wage bill crowding out other expenditure.- Medium-term strategy for 2014–20 to lower the wage bill from 44 to 35 percent of tax revenue.- The measures may not provide sufficient savings in the short term.- Integrating the administrative and financial management of human resources is key for wage bill sustainability.
- Increasing public employment in health and education sectors.- Move from a short-term approach to a more medium-term approach to ensuring fiscal sustainability.- Medium-term approach to wage bill management can help put the wage bill on a sounder footing.
- Fragmented management of government compensation and employment.- Pay in full previous wage commitments, cancel wage arrears, shift some of the pension contribution to the employees, reinstate automatic progression in the wage grid, boost wages for high-skilled staff and priority sectors, reduce recruitment targets, and commit to introducing a new wage grid to address equity issues.
France- High wage bill (a quarter of total spending) due to high employment levels.- Reduce employment by attrition and employment caps.- Limited employment effect due to policy reversal.- Decisions on wage increases are made without linkages to the budget process. Greater coordination could facilitate wage bargaining and oversight.
- Soaring wage bill of local governments due to loose hiring practices and rapid promotions.- Across-the-board wage freeze.- Limited impact of wage freezes due to low inflation.- A more stringent legal control of local governments’ employment practices by the central government offices is required.
Ghana- Expenditure on public compensation increased dramatically in 2008–14 due to introduction of a new wage structure.- Introduction of a single spine salary structure to increase equity, ensure fiscal sustainability of the wage bill, simplify negotiations and better connect pay and productivity.- Most of general government employees covered under a common compensation framework.- The fragmentation of human resource management across services and agencies makes it difficult to advance reforms.
- Increase in the wage bill to GDP ratio by 2 percentage points.- The introduction of a new pay scale requires fiscal space.
Honduras- Wage bill to GDP between 11 to 15 percent of GDP over the past 15 years

- High public sector wages, especially in education.
- Wage freezes in nominal terms in 2014 and 2015.

- Elimination of vacancies.

- Enhancement of oversight following a census of public employment.
- The wage bill declined by 2.3 percentage points of GDP.

- In sectors other than health care and education more than 80 percent of unfilled vacancies were eliminated.

- Identification of irregular (ghost) workers.
- A flexible employment framework is critical to downsizing in the public workforce.

- The reform agenda needs to be focused on an exit strategy from the short-term measure such as across-the-board freezes.
Ireland- Growing public wage bill between 2000 and 2009.- Cutting incomes of public employees; moratorium on recruitment and promotion; incentivized retirement schemes; reduction in paid benefits such as sick leave and annual leave; and productivity-enhancing measures.- By 2015 public sector wage levels have only fallen back to 2005 levels and the wage bill remains high.- Introducing significant reductions in public sector wages is challenging.
- Significant positive wage premium in the public sector.- An agreement on the need for wage reductions was achieved without significant industrial action.
- Sharp increase in government employment between 2000 and 2009.- Since 2009, public sector employment levels have fallen and remain below the average of advanced economies.
Jamaica- Fluctuating wage bill due to accumulation of arrears that are cleared every few years.

- Contractual and temporary workers comprise a significant share of the government workforce.
- Wage freezes between 2010 and 2012. In 2013, an agreement with labor unions to further freeze wages in order to reduce wage spending to 9 percent of GDP.

- The filling of vacated posts being reviewed and approved by the Post Operations Committee (the Ministry of Finance and Planning, the Office of Services Commission, and public sector unions).
- The medium-term goal is to keep the wage bill at 9 percent of GDP.- Hiring restrictions appear to have reduced the relative number of younger full-time permanent employees.
Kosovo- Significant public-private pay gap.

- Fast-growing wage bill due to wage hikes and additional hiring.
- Since 2018, a budget rule to limit the increase in the wage bill to no more than nominal GDP growth.- Tighter link between the wage bill and overall economic developments.- Changes in nominal GDP translate into caps on the public sector wage bill with a considerable lag.
Mali- Steadily increasing wage bill due to increasing recruitment in priority areas and increase in the base pay and allowances.

- Compensation payments to civil servants not properly recorded.
- Reforms linking pay to employee performance.

- Improved monitoring of public workforce.
- Attempted, but failed, to reform financial incentives for staff from the tax and customs departments.

- As of 2016 the outcomes of the staff census not disclosed.
- Vigorous resistance has largely stalled policy reforms on human resource management.
Moldova- High wage bill at 9 percent of GDP due to high public employment.

- Wages low relative to the private sector while affected by a political cycle.
- Short-term measures to contain the wage bill, e.g., postponing wage increases, freezing employment, and cutting permanently vacant positions.

In 2012, merit-based promotions and basic salary increases became more formalized.
- Budgetary savings in 2010–11.

- The cost of the civil service reform was not trivial. While being lower than the 2009 peak, the total wage bill in 2012 was 0.3 percentage points of GDP higher than the 2011 level.
- Nonstructural reforms can contain the size of the public wage bill only in the short run.

- Strong political will is important for the reforms to succeed; most of the reform efforts have been weakened during preelection periods.
Philippines- Enhanced public service delivery and accountability.- Replacing across-the-board bonuses to all civil servants with performance-based bonuses.- The wage bill has gradually decreased since 2011 as a percent of GDP indicating that the introduction of performance-based bonuses was not fiscally costly.

- Improved performance at unit and individual levels.
- The system of performance bonuses has improved public service delivery at a low fiscal cost.

- The overall view of the performance- based bonus system was positive among government employees.

- Measuring performance in the public sector is challenging.
Portugal- High wage bill due to generous wages.

- Organization of civil service characterized by large inefficiencies.
- Across-the-board attrition and cuts in the number of temporary workers.

- Introduction of progressive wage cuts and suspension of the 13th and 14th month salaries.

- Promotions, performance bonuses and mobility-related salary changes have been frozen.

- Single salary scale introduced.

- Structural reforms to improved oversight, limit inefficiencies, and increase flexibility in the management of human resources.
- In 2010–2014, the wage bill was reduced by 2 percentage points.

- General government employment declined from 14 to 11 percent of the working-age population.

- Attrition was the main instrument for employment rationalization.
- These reforms have been largely effective in containing the wage bill.

- Wage cuts proved less successful in curtailing compensation spending due to successive adverse Constitutional Court rulings and recent reform reversals.
Romania- Fast-growing wage bill due to both wages and employment.

- Fragmented wage bill management.
- Nominal wage cuts.

- Attrition-based reduction in public employment.

- Rationalization of nonbase pay, removal of special salaries, discounting payments for overtime work, and elimination of 13th month salaries and holiday bonuses.

- Strenghtened wage-setting processes for state-owned enterprises.
- Many of measures were phased out or reversed and the wage bill rose slightly again.- Structural reforms as well as consistency between policy measures are needed to rein in the wage bill over the medium and long term.

- Some of the more drastic measures were ruled unconstitutional.

- Beyond legal challenges, the reforms failed to address important distortions in the pay and employment system.
South Africa- Growing wage bill due to a new compensation framework.- Employment freezes in 2015 and 2016.

- Introduction of a voluntary severance package, combined with attrition and redeployment.

- Pension benefits restructured.

- Allowances consolidated into the base salary.

- Salary scales simplified.

- Job categories consolidated.
- Close to a 10 percent reduction in public employment by 1999.- Political will and social dialogue are key for the success of the reforms.

- Across-the-board employment restrictions could yield some savings in the short term, but they are not efficient.

- Achieving a rapid and sustainable reduction in the wage bill requires a combination in the wage bill requires a combination of wage and employment measures.
United Kingdom- Fast-growing wage bill.- Pay and recruitment freezes.

- Reform of pension benefits: raising retirement age, increasing contribution rates, and moving from final salary to career average defined-benefit plans.

- Modernized 1 D46 terms and conditions of employee contracts.

- Reform automatic progression pay.
- Reforms have been largely successful in controlling the wage bill.- Having clear policy goals facilitates reforms.
Zimbabwe- High and fast-growing wage bill due to high employment and wages.

- Extensive allowances.
- Hiring freezes.

- Nominal wage freeze since 2014.

- A number of direct interventions based on salary and employment audits.
- The freeze in positions has not been successful in containing growth in the civil service. Employment has continued to grow as service delivery priorities have overridden the commitment not to increase staff numbers.

- Wage growth has been driven mainly by the need to restore public service real wages that were eroded by hyperinflation during the 2000s.
- It is challenging to contain the wage bill while ensuring service delivery.

- A clear and credible wage bill target should be anchored in fiscal objectives.
Annex 4. Model-Based Simulations of Public Wage Bill Policies

A small dynamic stochastic general equilibrium (DSGE) model is used to simulate the impact of alternative policies on the labor market, growth, and fiscal position. The model allows the public and private sectors to interact and is a suitable tool for examining the effects of public wage bill policies on the economy.1 The model has been used to analyze the impact of lower public employment and higher public investment (Scenario 1) and then an increase in private sector productivity (for example, owing to structural reforms) under different public wage-setting rules (Scenario 2).

  • Scenario 1—Limiting the number of public vacancies while increasing public investment boosts private sector output and employment. Annex Figure A.4.1 (red dashed line) shows the impact of a policy mix that consists of a permanent decrease in public vacancies by one-fifth and a temporary autoregressive increase in public investment by 0.5 percentage point of GDP. Since public employment decreases following a reduction in public vacancies, unemployment rises. Nonetheless, as both private sector production and employment increase (0.5 percent) in response to higher public investment, most of the unemployed are absorbed by the private sector. Such a policy mix proves to be effective in creating fiscal space in the longer run (0.8 percent) and providing much needed public investment. The outcome of this policy mix is contrasted to an alternative scenario (blue line) in which public vacancies decrease permanently and public investment remains unchanged.
  • Scenario 2—Higher private sector productivity increases private sector wages, reduces the public wage premium, and motivates the unemployed to search for jobs in the private sector, leading to lower employment in the public sector. The increase in private employment—in response to improved productivity—outweighs the reduction in public employment and leads to lower unemployment (blue line). In addition, lower public employment translates into a lower public wage bill and improved overall fiscal position. However, in a special case when wages in the public sector are set to protect the size of public wage premiums, wages in the public sector follow wage developments in the private sector but also influence their development (yellow line). While such a wage-setting rule helps smooth out responses of public wages and other economic variables over time, it might lead to a spiral of ever-increasing public and private wages.

Annex Figure 4.1.Decreasing Public Wage Bills by Lowering Public Vacancies and Increasing Investment through a Temporary Shock to Government Investment

Source: IMF staff estimates.

Note: All variables are reported at monthly frequency, as deviations from the steady state. Unemployment, public employment, and private employment are expressed in percent of the labor force. Public investment and private and public production are in percent of GDP.

Annex Figure 4.2.Different Public Wage Rules in Response to a Permanent Positive Shock to Private Private Sector Productivity

Source: IMF staff estimates.

Note: All variables are reported at monthly frequency, as deviations from the steady state. Unemployment, public employment, and private employment are expressed in percent of the labor force. Public investment and private and public production are in percent of GDP.

Annex box 4.1.Simulating Complementary Policies to Public Wage Bill Management Policies

This box investigates economic effects of public wage bill policies using the Global Integrated Monetary and Fiscal (GIMF)1 model of the IMF calibrated to Saudi Arabia.1 Scenario 1 presents the impact of a mix of two policies—decreasing the number of public vacancies and increasing the level of public investment—with a neutral impact on the overall fiscal balance. Scenario 2 presents the same mix of policies complemented by an increase in targeted social transfers to liquidity-constrained households.

Annex Figure 4.1.1.Real GDP

(Percent difference)

• Scenario 1—A decrease in public employment is accompanied by a budget-neutral increase in public investment of 1 percentage point of GDP. The two policy measures have the opposite impact on the economy. On the one hand, higher public investment translates to a larger stock of public capital, which enhances private sector productivity and increases GDP. On the other hand, reduction in public employment reduces real GDP. Specifically, in the short run, higher public investment (1 percentage point of GDP) leads to real GDP that is above the baseline level by 0.2 percent. In the medium term, real GDP falls below the baseline level mainly due to lower employment in the public sector. However, the long-run effect of these two policy measures turns positive as real GDP reaches levels above the baseline mainly due to faster accumulation of public capital.

• Scenario 2—In response to negative effects of lower public employment and higher unemployment, the government introduces social transfers targeted to liquidity-constrained households (1 percentage point of GDP). Compared with Scenario 1, the negative adjustment in real GDP—due to lower public employment—is less dramatic thanks to larger public spending in the form of targeted transfers. However, the higher real GDP is accompanied by a permanent deterioration in the overall fiscal balance by 1 percentage point of GDP.

Annex Figure 4.1.2.Real GDP

(Percent difference)
1 The GIMF model is a finite horizon DSGE model augmented by sticky prices and wages, real adjustment costs, the presence of liquidity-constrained households that all imply an important role for fiscal policy. For details, see Kumhof and others 2010 and Anderson and others 2013.

The World Bank’s government wage bill and employment database is available online:


Prepared by Csaba Feher.


The wage premium, in most countries, is positive for public sector employees in lower grades and positions but gradually turns negative at higher grades, which limits the public sector’s ability to retain senior civil servants. Although the 10–11 percent wage premium is close to that observed in comparator countries, on average, it remains constant along the public sector wage distribution.


The education sector includes institutions under the Ministry of Education and under the Ministry of Higher Education and Scientific Research.


These numbers exclude the Kurdistan Regional Government since no information is available regarding the sectoral distribution of public employees in that jurisdiction.


Prepared by Claire Gicquel with inputs from Maura Francese, Chad Abdallah, Christopher Ben, and Liv Bjornestad.


Allowances reward seniority, education, or the difficulty of working conditions.


Prepared by Robert Blotevogel. The case study draws on IMF 2016b.


A negative contribution in 2012 reflects a decline in average wages in the civil service owing to the recruitment of less-qualified workers at lower wages.


Prepared by Mauricio Soto, based on the Selected Issues Paper for the IMF 2017 Article IV Consultation.


The analysis does not distinguish between nationals and expatriates.


Health care expenditures must be protected in order to improve health outcomes and because of demand drivers in the local population, such as population growth and aging, and the emergence of so-called lifestyle diseases, which are likely to put more demands on the health care system over time.


In addition to significant cuts in the incomes of public employees, the Irish government introduced a public sector moratorium on recruitment, employment control frameworks, and incentivized retirement plans.


Model simulations presented in this appendix are an outcome from a small DSGE model drawing on Gomez (2015). The model is expanded by search and matching frictions and calibrated to a representative economy in the region. Steady-state levels of labor income tax, the debt-to-GDP ratio, unemployment, public employment and private employment have been set to correspond to those for the Egyptian economy. Parameters characterizing labor market frictions replicate the stylized facts of labor markets in the region.


    AfonsoAntonio and PedroMaia Gomes. 2014. “Interactions between Private and Public Sector Wages.” Journal of Macroeconomics39 (PA): 97112.

    • Search Google Scholar
    • Export Citation

    Albino-WarMariaSvetlanaCerovicFrancescoGrigoliJuanCarlos FloresJavierKapsoliHaonanQuYahiaSaidBahromShukurovMartinSommer and SeokHyunYoon. 2014. “Making the Most of Public Investment in MENA and CCA Oil-Exporting Countries.” IMF Staff Discussion Note SDN 14/10International Monetary FundWashington, DC.

    • Search Google Scholar
    • Export Citation

    AlesinaAlbertoStephanDanninger and MassimoRostagno. 1999. “Redistribution Through Public Employment: The Case of Italy.” IMF Working Paper 99/177International Monetary FundWashington, DC.

    • Search Google Scholar
    • Export Citation

    AlganYannPierreCahuc and AndréZylberberg. 2002. “Public Employment and Labour Market Performance. Economic Policy (April).

    AminMagdiRaguiAssaadNazaral-BaharnaKermalDervisRaj M.DesaiNavtej S.DhillonAhmedGalalHafezGhanemCarolGrahamDanielKaufmanHomiKharasJohnPageDjavadSalehi-IsfahaniKatherineSierra and Tarik M.Yousef. 2012. After the Spring. Economic Transitions in the Arab World. Oxford University Press.

    • Search Google Scholar
    • Export Citation

    AndersonDerekBenHuntMikaKortelainenMichaelKumhofDouglasLaxtonDirkMuirSusannaMursula and StephenSnudden. 2013. Getting to Know GIMF: The Simulation Properties of the Global Integrated Monetary and Fiscal Model. Washington, DC: International Monetary Fund.

    • Search Google Scholar
    • Export Citation

    BeharAlberto and JunghwanMok. 2013. “Does Public-Sector Employment Fully Crowd Out Private-Sector Employment?IMF Working Paper 13/146International Monetary FundWashington, DC.

    • Search Google Scholar
    • Export Citation

    BorjasGeorgeJ.1984. “Electoral Cycles and the Earnings of Federal Bureaucrats.” Economic Inquiry22 (4): 44759.

    BouhlilaDoniaSmaali. 2015. “The Quality of Educational Performance of Students in the Middle East and North Africa: Much Remains to be Done.” Economic Research Forum Policy Brief.

    • Search Google Scholar
    • Export Citation

    BrixiHanaEllenLust and MichaelWoolcock. 2015. “Trust, Voice, and Incentives: Learning from Local Success Stories in Service Delivery in the Middle East and North Africa.” World BankWashington, DC.

    • Search Google Scholar
    • Export Citation

    BrockmeyerAnneMahaKhatrouch and GaëlRaballand. 2015. “Public Sector Size and Performance Management: A Case-Study of Post-Revolution Tunisia.” World Bank Policy Research Working Paper 7159Washington, DC.

    • Search Google Scholar
    • Export Citation

    CahucPierre and StephaneCarcillo. 2012. “Can Public Sector Wage Bills be Reduced?National Bureau of Economic Research Working Paper 17881Bureau of Economic ResearchCambridge, MA.

    • Search Google Scholar
    • Export Citation

    CerisolaM.A.AbdallahV.Davies and M.Fischer. 2015. “Assessing the Impact of Fiscal Shocks on Output in MENAP Countries.” Technical Notes and ManualsInternational Monetary FundWashington, DC.

    • Search Google Scholar
    • Export Citation

    Clements BenedictJ.DavidCoadyStefaniaFabrizioSanjeevGuptaTrevorSerge Coleridge Alleyne and Carlo A.Sdralevich. 2013. Energy Subsidy Reform: Lessons and Implications. Washington, DC: International Monetary Fund.

    • Search Google Scholar
    • Export Citation

    ClementsBenedictSanjeevGuptaShamsuddinTareq and IzabelaKarpowicz. 2010. “Evaluating Government Employment and Compensation.” IMF Technical Notes and Manuals 10/15International Monetary FundWashington, DC.

    • Search Google Scholar
    • Export Citation

    CrivelliE.D.Furceri and J.Toujas-Bernaté. 2012. “Can Policies Affect Employment Intensity of Growth? A Cross-Country Analysis.” IMF Working Paper 12/218International Monetary FundWashington, DC.

    • Search Google Scholar
    • Export Citation

    DahlbergMatz and EvaMork. 2011. “Is There an Election Cycle in Public Employment? Separating Time Effects from Election Year Effects.” CESifo Economic Studies57 (3): 48098.

    • Search Google Scholar
    • Export Citation

    EckardtSebastian and ZacharyMills. 2014. “What Goes Up Must Come Down: Cyclicality in Public Wage Bill Spending.” Policy Research Working Paper 6760World BankWashington, DC.

    • Search Google Scholar
    • Export Citation

    El MassnaouiKhalid and MhamedBiygautane. 2011. “Downsizing Morocco’s Public Sector: Lessons from the Voluntary Retirement Program.” In Case Studies in Governance and Public Management in MENA No. 3. World Bank, Dubai School of Government.

    • Search Google Scholar
    • Export Citation

    FeldmannHorst. 2009. “The Unemployment Effects of Labor Regulation around the World.” Journal of Comparative Economics37 (1): 7690.

    • Search Google Scholar
    • Export Citation

    Fernández-deCórdobaJavierJ.Pérez and José L.Torres. 2009. “Public and Private Sector Wages Interactions in a General Equilibrium Model.” ECB Working Paper 1099European Central BankFrankfurt.

    • Search Google Scholar
    • Export Citation

    GattiRobertaMatteoMorgandiRebekkaGrunStefanieBrodmannDiegoAngel-UrdinolaJuanManuel MorenoDanielaMarottaMarcSchiff Bauer and ElizabethMata Lorenzo. 2013. “Jobs for Shared Prosperity: Time for Action in the Middle East and North Africa.” World BankWashington, DC.

    • Search Google Scholar
    • Export Citation

    GomesPedro. 2015. “Optimal Public Sector Wages.” Economic Journal125: 1425451.

    GundersonMorley. 1979. “Earnings Differentials between the Public and Private Sectors.” Canadian Journal of Economics12 (2): 22842.

    • Search Google Scholar
    • Export Citation

    HertogSteffen. 2016. “Rent Distribution, Labor Markets and Development in High Rent Countries.” LSE Kuwait Programme Paper Series 40.

    • Search Google Scholar
    • Export Citation

    HertogSteffen. 2017. The Political Economy of Distribution in the Middle East: Is There Scope for a New Social Contract?International Development Policy 7.

    • Search Google Scholar
    • Export Citation

    HeynemanStephenP.1997. “The Quality of Education in the Middle East and North Africa.” International Journal of Educational Development17 (4): 44966.

    • Search Google Scholar
    • Export Citation

    HolmlundBertil. 1993. “Wage Setting in Private and Public Sectors in a Model with Endogenous Government Behavior.” European Journal of Political Economy9 (2): 14962.

    • Search Google Scholar
    • Export Citation

    International Monetary Fund. 2014a. Republic of Kazakhstan: 2014 Article IV Consultation—Staff Report. IMF Country Report 14/242Washington, DC.

    • Search Google Scholar
    • Export Citation

    International Monetary Fund. 2014b. The Caucasus and Central Asia: Transitioning to Emerging Markets. Washington, DC.

    International Monetary Fund. 2015a. “Adjusting to Lower Commodity Prices.” World Economic OutlookWashington DC.

    International Monetary Fund. 2016a. “Managing Government Compensation and Employment—Institutions Policies and Reform Challenges.” Washington DC.

    • Search Google Scholar
    • Export Citation

    International Monetary Fund. 2016b. “Case Studies on Managing Government Compensation and Employment—Institutions Policies and Reform Challenges.” Washington DC.

    • Search Google Scholar
    • Export Citation

    International Monetary Fund. 2017. Regional Economic Outlook Middle East and Central Asia. Washington DC, October.

    JewellAndrew. 2014. Meeting Algeria’s Fiscal Challenges. IMF Country Report 14/342.

    JewellAndrewMarioMansourPrithaMitra and CarloSdralevich. 2015. “Fair Taxation in the Middle East and North Africa.” IMF Staff Discussion Note 15/16Washington, DC.

    • Search Google Scholar
    • Export Citation

    KraayAart and CarolineVan Rijckeghem. 1995. “Employment and Wages in the Public Sector—A Cross-Country Study.” IMF Working Paper 95/70International Monetary FundWashington, DC.

    • Search Google Scholar
    • Export Citation

    KumhofMichaelDirkMuirSusannaMursula and DouglasLaxton. 2010. “The Global Integrated Monetary and Fiscal Model (GIMF)—Theoretical Structure.” IMF Working Paper 10/34International Monetary FundWashington, DC.

    • Search Google Scholar
    • Export Citation

    LamoAnaJavier JPérez and LudgerSchuknecht. 2007. “The Cyclicality of Consumption, Wages, and Employment of the Public Sector in the Euro Area.” ECB Working Paper 757European Central BankFrankfurt.

    • Search Google Scholar
    • Export Citation

    LamoAnaJavierJ Pérez and LudgerSchuknecht. 2008. “Public and Private Sector Wages: Co-movement and Causality.”

    MatschkeXenia. 2003. “Are There Election Cycles in Wage Agreements? An Analysis of German Public Employees.” Public Choice114: 10335.

    • Search Google Scholar
    • Export Citation

    MitraPrithaAmrHosnyGoharMinasyanMarkFischer and GoharAbajyan. 2016. “Avoiding the New Mediocre: Raising Long-Term Growth in the Middle East and Central Asia.” IMF Departmental Paper Middle East and Central Asia DepartmentInternational Monetary FundWashington DC.

    • Search Google Scholar
    • Export Citation

    MuellerRichardE.1998. “Public–Private Sector Wage Differentials in Canada: Evidence from Quantile Regressions.” Economics Letters60 (2): 22935.

    • Search Google Scholar
    • Export Citation

    PérezJavier and AntonioJesús Sánchez Fuentes. 2010. “Is There a Signaling Role for Public Wages?European Central Bank Working Paper 1148Frankfurt.

    • Search Google Scholar
    • Export Citation

    RotherBjörnGaëllePierreDavideLombardoRistoHerralaPriscillaToffanoErikRoosGregAuclair and KarinaManasseh. 2016. “The Economic Impact of Conflicts and the Refugee Crisis in the Middle East and North Africa.” IMF Staff Discussion Note 16/08International Monetary FundWashington, DC.

    • Search Google Scholar
    • Export Citation

    SaidYahia. Forthcoming. “Economic Transformation: Human Centered Diversification StrategyIMF Selected Issues PaperInternational Monetary FundWashington, DC.

    • Search Google Scholar
    • Export Citation

    SdralevichCarloRandaSabYounesZouhar and GiorgiaAlbertin. 2014. “Subsidy Reform in the Middle East and North Africa: Recent Progress and Challenges Ahead.” International Monetary FundWashington DC.

    • Search Google Scholar
    • Export Citation

    SommerMartinGregAuclairArmandFouejieuInutuLukongaSaadQuayyumAmirSadeghiGaziShbaikatAndrewTiffinJuanTrevino and BrunoVersailles. 2016. “Learning to Live with Cheaper Oil.” IMF Departmental Paper Middle East and Central Asia DepartmentInternational Monetary FundWashington DC.

    • Search Google Scholar
    • Export Citation

    World Bank. 2008. “The Road Not Traveled: Education Reform in the Middle East and North Africa.” MENA Development ReportWashington, DC.

    • Search Google Scholar
    • Export Citation

    World Bank. 2009. “From Privilege to Competition: Unlocking Private-Led Growth in the Middle East and North Africa.” Washington, DC.

    • Search Google Scholar
    • Export Citation

    World Bank. 2013. “Opening Doors: Gender Equality and Development in the Middle East and North Africa.” MENA Development ReportWashington, DC.

    • Search Google Scholar
    • Export Citation

    World Bank. 2014. “More Jobs, Better Jobs: A Priority for Egypt.” World Bank Report 88447-EGWashington, DC.

    World Bank. 2015. “Towards a New Social Contract.” MENA Economic Monitor (April) Washington, DC.

    World Bank. 2016. “Public Employment and Governance in MENA.” World Bank Report ACS18501Washington, DC.

    YazbeckAbdoS.Tamer S.Rabie and AakaPande. 2017. “Health Sector Reform in the Middle East and North Afria: Prospects and Experiences.” Health Systems & Reform3 (1): 16.

    • Search Google Scholar
    • Export Citation

    Other Resources Citing This Publication