During the 1970s and 1980s, the wage bill of many developing country governments ballooned as new ministries, departments, and state-owned enterprises were created. These countries found there was little left over for public investment programs after paying the wage bill. Resources were spread so thinly that service delivery suffered.

As countries approached multilateral lending institutions, they were pressured to reduce the public wage bill, while simultaneously improving services and developing capacity to formulate policy. The issue of civil service pay and, more generally, the incentives for performance, have remained high on the civil service reform agenda ever since.

This Web page outlines some key issues for consideration: How are civil servants' pay levels set? What constitutes total pay and non-pay rewards of civil servants? How do civil servants' pay and rewards compare with other options before them? Do some of the incentives stemming from civil service pay change as officials progress through their career?

Determination of civil service pay                                      [ TOP ]

The pay assigned to a particular position is determined by government -- often with the advice of an independent commission that considers the content, experience and skills necessary for each job, as well as market comparators. Frequently, the process also involves extensive bargaining by employees' unions. Government must attempt to strike a balance between fair remuneration for employees and public accountability for expenditures on civil service pay. Often "spikes" are superimposed on the underlying remuneration scheme as the result of ad hoc grants conferred by government to respond to the demands of an employee category, or following the judgment of an administrative tribunal.

Pay policy also reflects the changing priorities of government. After independence, many developing country governments set out to be "model" employers during the 1960s and 1970s. However, few developing country governments actively pursue this objective today (Chew 1993). A number of governments recently began experimenting with the introduction of civil service pay flexibility as a managerial tool (Kessler 1998). This issue is discussed further below.

Civil service salaries are revised annually or biennially in most industrial market economies. In contrast, they are adjusted infrequently in many developing countries (ILO 1989).

Incentives to be a civil servant                                           [ TOP ]

Government pay determination is widely supposed to be transparent (as public money is being used) and fair (by paying equally for jobs of equal value -- such paying men and women equally for the same work). However, there are several components to civil servants' compensation and rewards. As the chart below shows, the total compensation package is a combination of current rewards and future expectations, contractual and intangible rewards:

This website uses the following conventions:

• personal emoluments = cells 1 and 4 (current monetary rewards and allowances)
• personal disposable income = personal emoluments minus employer deductions for income tax withholding, provident fund/pension contributions, etc.
• total compensation = cells 1, 2, 4 and 5 (contractually provided current rewards and allowances)
• total rewards = cells 1 to 9 inclusive (contractual and non-contractual, current and expected rewards and allowances)

In addition, the following terms require a definition:

Base pay. Every civil servant receives this regularly (usually fortnightly or monthly) from the government by virtue of being on the payroll. Base pay (cell 1) is usually linked to the employee's position and is uniform across similar positions. For example, the base pay could be 1000 currency units for the post of Agricultural Extension Officer, and 1500 currency units for the post of Deputy Permanent Secretary in the Agriculture Ministry. It would also be 1500 currency units for the post of Deputy Permanent Secretary in other ministries.

The base wage is often cited to compare wages in the public and private sectors. It is, however, only one component of civil servants' total rewards.

Allowances. The purpose of allowances (e.g., transport, housing, telephone, travel) can be to encourage employees to live in remote areas (housing), encourage public transportation (transport), and ensure accessibility in an emergency (telephone). When these allowances are paid in cash (see cell 4), they are often part of civil servants' taxable income. When they are provided in kind (e.g., a free telephone at the residence, or rent-free accommodation; see cell 5), they are untaxed.

In-kind allowances can be provided across-the-board to all civil servants (e.g., subsidized government employee health or life insurance; cell 2). However, allowances also can differ sharply for employees at different grades, and for those in the same grade but in different ministries.

The proportion of allowances to base pay varies widely between countries; but it is not uncommon for them to equal or exceed base pay. In the OECD, the basic wage component averages nearly 90% of the total pay and rewards of civil servants, ranging from being 100% of total in Switzerland and Iceland, 65% to 90% in Finland, to 20% in Greece (OECD, 1997). In Jordan, however, where civil service wages have essentially been frozen since 1988, a variety of allowances (such as hardship, responsibility) amount to 70% or more of the basic salary (World Bank 1999). In Indonesia, allowances (many of them discretionary) can represent up to 92% of pay (Manning 2000). Meanwhile, total allowances as percentage of total salaries ranged from 12 and 107 percent in eight African countries during the 1980s.

Retirement benefits. In addition to current rewards, civil servants usually can count on receiving a pension (cell 7) and/or a lump-sum gratuity among retirement. Usually, pension benefits are paid from general revenues of the state, thus increasing the level of government consumption. The rules for calculating pensions differ markedly from one country to the next, but benefits are generally tied to the number of years worked, and the base pay drawn by the civil servant just prior to retirement. In some countries, retirees may receive land or a house either free or at highly subsidized prices (cell 8).

Intangible benefits. Civil servants enjoy additional benefits to which it is difficult to attach a price tag (cells 3, 6 and 9). Among these are job security, social privileges (such as preferential access to government services), and hard currency savings from trips abroad. Where private sector jobs are hard to obtain, job security in government is a very attractive feature. And when governments provide health care or higher education at subsidized prices, civil servants' access to these services for themselves and their families can generate substantial financial gains.

Intangible benefits may continue even after retirement. Officials in the civil service may be able to take high-level positions in public corporations or private industry upon retirement, as a direct result of their former position in government. The Japanese amakudari system ("descent from heaven") is one prominent example (Inoki 1993).

Comparing civil servants' incentives to private sector options    [ TOP ]

A general perception held by most civil servants and government employee unions is that state employees are paid less than their private sector counterparts. Accurate comparisons, however, are not easily made. Information on job requirements in each sector is often unavailable; and fair comparisons must take into account all aspects of public employees' rewards.

It is certainly true that in many countries civil servants' remuneration has declined in real terms. In Tanzania, for example, the real average civil service pay in 1998 was 70% of what it was in 1969, rising from a mere 27% in 1985 (Valentine 2001). However, in spite of declining real average government salaries, evidence from seven CFA franc countries shows that average civil service remuneration at the beginning of 1990s -- inclusive of both cash and in-kind benefits -- was higher than the average private sector remuneration. In Malawi, Zambia and Zimbabwe, surveys have shown that salaries at the lower end of the civil service salary scale were higher than those paid for comparable jobs in the private sector (Lienert 1998). In Korea, average pay in the civil service in the early 1990s was two-thirds that of the private sector, with the pay differential rising in higher civil service grades. However, civil service salaries in Hong Kong and Singapore ran ahead of the private sector during the same time (Chew 1993).

Incentives change during civil servants' career progression. Many developing countries pay higher-than-private sector wages at the lower end of the grade spectrum, and therefore attract talent which they may subsequently fail to retain. Whether or not an employee continues in the civil service will be influenced by whether civil service pay is too "compressed" to provide adequate incentives for advancement.

Pay scales typically are organized in a pyramid of hierarchical grades that differentiate people and positions according to the skill level and knowledge required for particular jobs. Additionally, the whole civil service can be disaggregated vertically into narrower pyramids representing specific occupational cadres. For example, the medical cadre could comprise doctors and paramedics employed by government. Often different cadres have their own range of grades organized according to skill requirements that are difficult to compare across cadres. The combinations of grades and cadres provide a series of horizontal and vertical organizational structures.

The table below shows how the Ethiopian civil service was organized into cadres and grades in the 1990s. Each grade is assigned a salary scale, with incremental steps leading from a minimum salary to the maximum of that grade. (Salary scales for different grades often overlap.) Each roman numeral in the example above represents a whole scale. Salary scales reward length of service and the accumulation of experience. Incremental (within scale) salary increases are often standard, while promotion to the next grade is contingent upon satisfying certain requirements and receiving a positive job evaluation.

Linking rewards and performance                                        [ TOP ]

An approach being followed in some OECD countries is to link a civil servant's pay to individual performance, as well as to the position. Over and above the base pay and any extras, a civil servant can be offered a bonus depending on the degree to which performance goals are achieved. It is important to note, however, that in OECD countries performance pay constitutes only a small share (up to 10%) of the civil servant's total pay. A survey conducted by the International Personnel Management Association (IPMA) indicates that the U.S. government is shifting somewhat from paying workers based on seniority to paying them for performance (IPMA 2001).

Rewards to encourage good performance are uncommon in developing country civil services. But examples do exist. In Singapore, employees receive an annual bonus -- depending on national economic growth. Chinese civil servants receive an annual bonus; but due to the difficulty in assessing individual performance accurately, virtually the same amount has been paid to each employee. In Thailand, each year 15% of officials in each grade can be awarded one extra increment for exceptional performance; but there is some skepticism about the reliability of how staff performance is evaluated (Chew 1993).

In a recent World Bank survey, public officials in sixteen countries were asked whether they knew of any employee in their organizations who had been rewarded for good performance in the preceding twelve months. In all countries surveyed, the majority of civil servants reported not knowing of any rewards for good performance (Mukherjee and Gokcekus 2001).

Conclusion                                                                            [ TOP ]

In order to determine whether state employees are being paid too much or too little compared to their private sector counterparts, it is first necessary to have an accurate measure of their total rewards. This includes the monetary value of all in-kind allowances and the net present value of future pensions. It is not uncommon for government administrators to be paid more than their private sector counterparts, while doctors and accountants are paid considerably less.

The incentives created by different forms of pay and reward -- and their impact on performance -- can be complex. As an illustration, job security is generally an important element of a civil servants' total rewards. Thus, governments wishing to adopt new-public-management style contractual arrangements (to reward performance) may have to offer a higher base pay to compensate for lost job security.

Recommended readings                                                       [ TOP ]

• Chew, David C. E. 1993. "Civil Service Pay in the Asian-Pacific Region." Asian Pacific Economic Literature 7(1), May.
• Inoki, Takenori. 1993. "Japanese Bureaucrats at Retirement: The Mobility of Human Resources from Central Government to Public Corporations." EDI Working Paper No. 93-32, The World Bank, Washington, D.C.
• International Labour Organization. 1989. World Labour Report 4. Geneva, International Labour Office.
• International Personnel Management Association. 2001. IPMA/NASPE Benchmarking Survey Results, 2000-2001.
• Kessler, Ian. 1993. "Pay Determination in the British Civil Service since 1979." Public Administration 71(Autumn): 323-340.
• Lienert, Ian. 1998. "Civil Service Reform in Africa: Mixed results after 10 years." Finance and Development 35(2).
• Mukherjee, Ranjana, and Omer Gokcekus. 2001. "Living with Political Patronage: Civil Service Reform Options When Patronage Cannot be Eliminated." unpublished paper.
• OECD. 1997. "Trends in public sector pay in OECD countries." Paris: OECD.
• Robison, Derek. 1990. Civil Service pay in Africa. Geneva: International Labour Office.
• Valentine, Theodore R. 2001. "On the Consequences of Successful pay Reform: Evidence from Tanzania."
• World Bank. 1999. Report No. 19664-JO of the World Bank.

This page was prepared by Ranjana Mukherjee of the World Bank with Nick Manning. It was submitted on 22 May 2000.