VI. Public Sector Employment
- Ke-young Chu, and Richard Hemming
- Published Date:
- September 1991
What determines the number of public sector employees?
What is the relationship between the public and private sector labor markets?
Is public sector employment excessive? What implications does this have?
What measures can be taken to reduce employment in the short run and the long run?
The wage bill of government employees represents a sizable share of government expenditure in both industrial and developing countries (Table 1). Wage and employment policy in the public sector therefore has a significant impact on total expenditure. A proper grasp of this important area of expenditure policy has to start from the premise that employment policy is intimately linked with pay policy. For convenience, employment issues are dealt with in this note and issues relating to the level and structure of pay in the note on Public Sector Pay. The strong interconnections between these two subjects is reflected in the country illustrations included in the latter note.
|General government||Central government|
In relation to total population, employment at the general government level is much higher in most industrial countries than it is in most developing countries. In several Northern European countries, there were more than 10 general government employees per 100 inhabitants in 1978-80, and the OECD average was about 7 employees per 100 inhabitants. Regional and local governments account for more than half of the total in most industrial countries, and much of this employment is attributable to the health and education sectors. The government is often a direct provider of educational and health services, and the coverage of health insurance schemes and school enrollment levels are in both instances high. These statistics are in stark contrast to those of developing countries, particularly Africa: in most of the African countries for which data are available, there were fewer than 2 general government employees per 100 inhabitants. Nonetheless, general government employment in developing countries can represent a very large share of employment in the formal sector, particularly in sub-Saharan Africa. Thus, in a sample of five countries in this area, employment at the general government level accounted for between 15 and 45 percent of total formal sector employment; the share rises to 75 percent in Zambia if parastatal employees are included.
Employment Demand in the Public Sector
Labor market analysis starts from the premise that the private sector employs labor solely for the purpose of the value of its productive services, and the demand for labor services will depend on the cost of labor relative to the value of the output it produces and the cost of the other inputs into the productive process. With the exception of health and education sector personnel, however, labor employed in the public sector produces goods and services that generally cannot be profitably produced by the private sector. A government seeking to minimize the cost of a given volume of production of public goods would do so by making a choice between available techniques of delivering a public service based upon their technical characteristics and relative input prices.
It is very difficult to measure the volume of output and to estimate a production function for the public sector. Nonetheless, if the relative cost of labor and other inputs does not change, and the demand for publicly provided goods tends to increase with income, then public sector employment would tend to grow with GDP. Broadly speaking, this is what is observed; international cross-section data reveal that there is a strong relationship between employment per capita and real income per capita at the general government level.
There is a wealth of casual evidence that public sector employment decisions are influenced by far more than cost minimization, and that public employment has a dynamic of its own. If public sector managers are not subject to pressures to minimize costs, and they gain status and enhanced responsibilities from a larger staff, pressures to increase employment will emerge that are quite independent of any increase in demand derived from an increase in the demand for public services (see the notes on Public Expenditure and Resource Allocation and Public Expenditure Productivity for additional discussion of this point). Moreover, increasing employment in the public sector is also often seen as an integral part of national employment policy.
Impact on the Private Sector
The economic impact of additional employment in the public sector depends partially on how private sector labor markets function and the extent and nature of unemployment. In a fully employed economy with competitive labor markets, an increase in public sector employment will reduce both output and employment in the private sector. It does so by bidding up the price of labor, thus reducing the quantity of labor the private sector will demand. In effect, the increase in public sector employment crowds out private sector employment through its impact on private sector wage rates.
If the extra output of the public sector is deemed to be at least as valuable as the private sector production it displaces, there is nothing inherently undesirable about an expansion of public sector employment, although it does affect the functional distribution of income and may also affect the personal distribution of income and the rate of return to private sector investment. It should also be noted that the displacement of some private sector labor by additional public sector labor does not imply that public sector salaries are too high. The story is different in the presence of labor market rigidities or unemployment, because an expansion in public sector employment need not entail a reduction in private sector employment; instead, it results in an increase in total employment. As an employment policy, however, increases in public sector employment are clearly inferior to measures that improve the functioning of labor markets.
It has been argued that there may be some critical level of public sector employment in relation to private sector employment, above which public sector wage settlements begin to affect private sector negotiations. The importance of this phenomenon will depend on the mechanism of wage determination in the private sector. Private sector markets will be affected to the extent that higher wage settlements in the public sector lead either to an increase in the reservation wage of private sector workers or to an increase in search unemployment, as private sector employees remain or become unemployed to better their chances of employment in the public sector. The strength of the link between public and private sector pay awards may also vary with the degree of unionization in the private sector.
Excess Public Sector Employment
It was noted at the outset that it was inherently difficult to quantify the extent of excess employment in the public sector, because it is hard to measure output and hence the productivity of labor. Yet in many developing countries it is generally conceded that the level of employment exceeds any reasonable estimate of requirements by a substantial margin, and often public offices are filled with personnel with little to do. Such overemployment is often associated with inadequate levels of expenditure on other current goods and services. The available evidence for public employment in Africa also shows very substantial rates of growth: for example, in Ghana the average annual rate of growth of the civil service, excluding the educational service, police, and military, was 15 percent over the period 1975-82 (Table 2).
|Ghana||- Civil service||15.0|
|Malawi||- Civil service established posts||7.9|
|Mali||- Civil service||5.6|
|Nigeria||- Federal civil service||8.6|
|- Total public sector||15.8|
|Senegal||- Civil service and public enterprises||5.4|
|Zambia||- Central and local government||0.7|
Data refer to various sub-periods.
Data refer to various sub-periods.
The rapid growth of employment is usually associated with an increase in the wage bill as a percentage of government expenditure, in spite of the fact that average real public sector wages have generally declined. Employment growth has been most rapid for less-skilled groups of public labor; for some highly skilled groups such as doctors, employment levels have actually declined. This development has been attributed to the compression of the wage and salary structure that has come about as a result of attempts to favor the less well paid in periods of relative austerity.
Reducing Public Sector Employment
The realization that resources have been wasted on excess public employment has prompted experiments in many countries, particularly in Africa, with hiring freezes as well as wholesale reductions in the size of the civil service. Public sector employment has been restrained by the application of the following measures, more or less in increasing order of implementation difficulty:
- (i) Eliminating ghost workers: this measure has been implemented in several African countries, where it has been found that the payroll included the names of nonexistent or departed public employees. This particular problem is typically found only in countries whose administrative capacity is especially weak.
- (ii) Firing temporary workers: this has been an element in programs of public sector reform pursued in The Gambia and Jamaica, presumably because it is easier to dismiss temporary staff than full-time staff.
- (iii) Hiring freezes: these have been used in many countries, although usually only as a temporary measure. A variant of this approach is the policy of only partially replacing departing employees; in order to prevent upward pressure on the wage bill, the policy may take the form of restraining the remuneration of new employees to some fraction of the remuneration of the old. Its intent has been thwarted by replacing former regular employees with day laborers, as is the case in at least one African country.
- (iv) Abandonment or suspension of the policy of employer-of-last-resort for university graduates: this policy is easiest to apply when steps have already been taken to reduce excessive intake by universities and technical or professional schools, although the latter policy has a long lag before its effects are felt. Forward-looking educational planning can prevent the emergence of the problem in the first place (see the note on Education for discussion of imbalances in educational programs at the tertiary level).
- (v) Voluntary retirement or resignation: several African countries have tried to implement a policy of voluntary retirement or resignation, under which generous severance arrangements are used to induce public sector employees to quit their jobs; this type of policy requires considerable administrative expertise, may be quite costly, and is open to abuse. The more “voluntary” the retirement the more costly it has to be. This approach was tried in Guinea in 1987-88, but it met with little success.
- (vi) Early retirement: this policy has been promoted in some countries, including Nigeria, where it proved to be costly. If the main aim of the policy is a permanent reduction in the public sector’s labor costs it is necessary to ensure that the terms of retirement are not excessively generous. If there is an important lump-sum component to the early retirement or voluntary retirement bonus, the immediate effect of the policy will be to increase the size of the budget deficit, although if most of the bonus is saved the impact on aggregate demand would be comparable to an annuity. These sorts of policies would appear to have a greater chance of success the greater the opportunity for employment in the private sector.
- (vii) Dismissal: few countries have adopted a policy of outright dismissal. Those that have done so have tried to soften the blow by offering generous severance pay. Ghana has had some success in reducing excess employment at its cocoa marketing agency in this way.
- (viii) Privatization: public sector employment will fall as a result of the sale of public enterprises and the devolution of activities to the private sector; however, overmanning is more likely to be reduced if the market in which the privatized enterprise operates or the privatized activity takes place is relatively competitive (see the note on Privatization).
The experience with policies to reduce public sector employment points to a number of conclusions. Massive employment reductions are generally not feasible, although measures affecting selected groups of the public sector work force—for example, temporary employees or employees in a particular sector—have been successfully employed in some countries. It is easier to avoid hiring new employees than to dismiss existing employees, and the lower status of part-time or temporary workers makes their dismissal less unpalatable. Freezes on hiring have also worked, at least for a time. The reduction in public sector employment, although often a more durable means of lowering the public sector wage bill than wage restraint, has not been without short-run or medium-term costs. Many governments have had to design compensatory schemes for redundant public employees.
By way of a more general conclusion, employment policy is an area where the dictum that an ounce of prevention is worth a pound of cure is of particular validity. To avoid the creation of excessive employment or the aggravation of an already existing problem, it is necessary that there be centralized control of new appointments, by which requests for new positions can be subjected to some sort of check. Only budgeted positions should be authorized. To avoid the diversion of wages and salaries to unauthorized persons, a strict correspondence between public service records and payroll lists needs to be maintained.
Heller Peter S. and Alan A. Tait Government Employment and Pay: Some International ComparisonsIMF Occasional Paper No. 24 (Washington: International Monetary Fund1984).
Lindauer DavidL. “Government Wage and Employment Policy: an Analytical Treatment with Special Reference to Africa,” Discussion Paper No. 1987-3 (Washington: World BankMarch 1987).
Lindauer David L. Meesook Oey Astra and Parita Suebsaeng “Government Wage Policy in Africa—Some Findings and Policy Issues,” World Bank Research Observer (WashingtonJanuary 1988).