Employment in the Service Sector in Sub-Saharan Africa

Myron Frankman and Edwin Charlé

Note: Footnotes are missing 

One characteristic which all countries presently hold in common - whatever their level of income - is the tendency for a growing proportion of the population to be employed in the 'service sector' of their economies rather than in 'secondary' or industrial activities. The purpose of this short article is to suggest that a consideration of certain features of the expanding employment in services in low-income countries may help to evaluate two commonly identified situations. First, in regard to manufacturing activities there has been a persistent tendency towards increasing capital intensity, despite the frequent assertion that resort should be made to more labour intensive technology in order to ëspreadí jobs more widely. Secondly, the pace of urbanisation has continued to accelerate, despite widespread concern over high rates of urban unemployment and under-employment that seem a consequence of excessive migration to the cities.

The term 'service sector' is used here to refer to all economic activities other than those in agriculture, mining, and manufacturing. Thus, it includes retail and wholesale trade, domestic service, general government employment, the professions, personal services, banking and insurance, construction, transport and communications. As most of these activities do not result in the direct production of tangible goods they have frequently been regarded as 'unproductive' or even ëparasiticí, and have been seen as contributing less to a country's economic growth potential than agriculture, mining, or manufacturing.

We wish to suggest that employment in services - rather than being an appropriate object for social condemnation or reluctant acceptance - might be seen both as economically rational from the standpoint of the individual participant, and as economically beneficial to the society. This line of argument seems relevant to understanding two apparent paradoxes: growing capital intensity in manufacturing in the face of an abundant labour supply, and growing urbanisation despite large-scale urban under-employment.

Although a number of countries in sub-Saharan Africa have arable land which is not at present overcrowded, rapid population growth has been accompanied almost everywhere by an even faster urban increase. The growth rate of cities with 20,000 or more inhabitants has been estimated to exceed 5 per cent per year in 21 of the 27 countries of sub-Saharan Africa for which data are available, and to be greater than 10 per cent per year in 13 of these countries.

Only a minority of the population (typically 4-9 per cent) derives its livelihood from mining and manufacturing. Moreover, this proportion seems likely to decline in the near future, for rarely is more than 2 per cent of the annual population growth being matched by new employment in manufacturing. Jobs in the extractive industries are expanding even more slowly, and have actually declined during the last decade in Ghana, Kenya, Malawi, South Africa, Rhodesia, and Tanzania. In contrast, available data suggest that from two to four times as many individuals in the countries of sub-Saharan Africa are economically involved in the service sector as in mining and manufacturing.

It is generally recognized also that a shortage of capital equipment coexists with a relative abundance of job-seekers. Under such circumstances it is ordinarily assumed that rational decision-makers - in both the public and private sectors - will attempt to economize on the use of capital by employing techniques which use labour quite liberally.

In respect to industrial operations in sub-Saharan Africa, however, both case-studies and aggregate statistics show a persistent tendency for jobs to increase less rapidly than output, and there are numerous instances of declining employment with expanding production. This trend is probably attributable in part to the well-known fact that, in established industrial enterprises, opportunities to reduce capital by substituting labour (while maintaining output levels) are technological rarities. Technical considerations are reinforced by the fact that the cost of the labour input as seen by employers is higher than would be suggested merely by the large numbers of persons seeking work. This is partly due to 'artificially' high wage rates associated with minimum wage laws and collective bargaining agreements. It also results from the relatively low productive potential of those who have recently left a 'traditional' environment. Poor health and illiteracy may be serious handicaps. In addition, the development of punctuality, time-regulated responsibility, and the acceptance of various mechanistic requirements of machine handling, require a period of cultural conditioning during which the productivity of employed workers may rise dramatically without any technological innovation.

Increased capital intensity may also be encouraged by the fact that employers who want more control over their labour force, especially to resist strikes, may take steps to minimise the number of workers they employ, even though this involves less than optimal factor combinations. This tendency may be reinforced by artificially low interest rates for large borrowers, and by tax laws which favour capital intensive installations.

Such factors suggest that the prevailing labour utilisation ratios are likely to be consistent with maximum profit goals for individual firms. It further implies that, at least for some time to come, most of the workers leaving a rural-traditional environment are likely to find productive opportunities outside of mining or manufacturing, i.e. in services.


Trade constitutes by far the most important non-agricultural occupation in sub-Saharan Africa. It may also be an important activity for those who are statistically recorded as being employed in agriculture, since it is not uncommon for many producers to market their own crops. Commercial trading takes place in urban and rural locations, encompasses the vending of almost every kind of product, and varies from the control of extensive volumes to those which are almost unbelievably small. Descriptive studies of these activities identify several salient economic characteristics:

  1. Most traders are self-employed.
  2. A very large number of individuals, including women, children, and the aged, can find employment in trade. Lack of formal or even literacy is 'ordinarily no obstacle to entry, and there are opportunities for part-time work to supplement other income, or to 'fill in' while seeking another job. Traders often begin operations with extremely small amounts of invested capital. Frequently there are kinsmen or friends able to supply goods for sale and useful advice. Barriers to entry may include government licensing requirements, the need for formal permission to operate in a designated market, or the informal exclusiveness of ethnic 'in-groups'. For most individuals, however, such restrictions would not bar entry to all areas of trade, especially in the cities.
  3. The earnings of traders are often small. This has frequently been attested to, and complements the evidence of open entry and effective competition.
  4. Trade offers perhaps a maximal degree of variability in the amount of capital utilised per employed worker and per unit of output. It can be carried on with or without shops, furnishings, or transport, and with a large or small inventory. The general lack of storage and preservation facilities is not a problem, because most traders are willing to deal in small quantities, and low-income families are unable to hold large stocks of non-durable consumer goods.
  5. Trading provides economically useful experience in commercial calculation, and facility in personal relationships.
  6. Such trading profits as do occur are often reinvested. Undoubtedly this usually takes the form of inventories of goods to be sold, but other investments include vehicles, apparatus for new types of production, and houses.
Aside from trade - and, of course, apart from agriculture, mining, and manufacturing - most other economic activities are a heterogeneous assortment labelled 'services' in the statistics of the United Nations. Included here are personal services (notably in restaurants and hotels), domestic work, teaching, many government occupations, and the 'learned professions'. Employment in banks and insurance companies, which is included officially with trade, might logically be considered here also.

Certain economic generalisations seem appropriate for this somewhat disparate group of human undertakings. In the first place, a wide variety of entry-level positions can be identified. For the uneducated, these include entry into domestic service, apprenticeship to barbers or photographers, unskilled tasks in medical or religious institutions, employment as messengers and labourers in commercial houses and government offices. For school-leavers there are clerical and administrative posts in both the public and private sectors. For those with more education, there are openings, for example, in law, medicine, civil engineering, and university teaching.

Secondly, it is clear that such callings provide a variety of routes for advancement. Few require important physical qualifications for professional progress, or systematically bar the aged from participation. Apprenticeship may be a formal necessity for some trades. Clerical positions frequently include opportunities for the acquisition of skills, and many other jobs need not lead to a professional 'dead end', especially where facilities for continuing education are available.

As a third generalisation it is apparent that many of these activities involve relatively small physical capital requirements (per unit of return) for new entrants. This would seem to be the case, for instance, with butchers, hairdressers, or tailors, as well as in the 'skilled professions'. There is, moreover, a high degree of variability in the amount of capital with which these activities may be carried on.

In most sub-Saharan countries, construction is estimated to provide from 40 to 100 per cent as much employment as does manufacturing, and transport and communications from 30 to 60 per cent as much. Entry is likely to be easy into many facets of both construction (including carpenters, plumbers, brick-layers) and transport (hand-carriages, carts, taxis, lorries), and relatively few restraints on advancement seem inherent in the nature of these jobs. Here there are opportunities for self-employment and a greater range of capital requirements than in the manufacturing sector.


The concentration of capital-intensive techniques in a narrow range of productive industries, concurrent with extensive employment in services (broadly defined), would seem to entail several predictable economic consequences. In those lines characterised by high capital concentration, ordinary competition cannot be relied on to limit profits or ensure technical efficiency. Instead, there is a clear possibility of pricing and output policies which are detrimental to consumers, and buying policies which are disadvantageous to the sellers of inputs. (In this respect, government-imposed minimum wages, or rates reached through collective bargaining with strong unions, in so far as they affect these industries with particular force, provide an offset to monopsonistic power.) The fact that heavily capitalised industries would be few in number and highly 'visible' (and frequently sponsored, at least in part, by the government) suggests that political controls could be invoked to protect the public.

Elsewhere in the economy the level of capital concentration would be lower, the incidence of self-ownership higher, and the ease of entry greater. Here would exist the likelihood of effective competition, and the protection of the public inherent therein. Here, also, the distribution of income would tend to be governed principally by labour inputs, rather than on the basis of capital ownership. This tendency towards equity might be reinforced by frequent private transfer payments from urban migrants to traditional villages. For the capital intensive sector, the likelihood of government participation or control suggests that the distribution of rewards may be guided by political pressures rather than by economic power. Such circumstances would seem compatible with the popularly expressed goals of many African states which accept government participation in the productive processes and which, even when espousing capitalism, advocate a relatively equitable distribution of income.

Finally, the coexistence within an economy of a few capital-intensive firms, and a large number of business units using little capital per man, may constitute a favourable environment for growth. The large firms will be able to take advantage of the technical and managerial experience of similarly organised establishments in high-income countries. The small units can provide both economic flexibility and a training ground for commercial skills and entrepreneurship. If the basic short-run need of the low-income countries is to economise their decision-making skills, it may be advantageous if most economic activity is carried on by small firms, guided and constrained by the competition of those who are self-employed and self-motivated.

The saving - and investment - potential of such a system constitutes a key indicator of its aggregate growth potential. If individual earnings are so low in the competitive sector that they are just sufficient to cover some minimal level of necessary consumption expenditure, savings will be at a low rate. However, other organisational patterns might lead to even less savings. If the alternative to employment in low-compensation services is unemployment, and if it is assumed that individuals will not be allowed to starve, unemployed workers will have to be supported by public or private transfers. Employment in low-paying service activities may therefore result in a net output gain for the society without affecting aggregate savings. Alternatively, where the purveyors of services earn more than minimal amounts, increased savings for investment may occur, and aggregate growth may be augmented. One should not underestimate the ability of the poor to make discretionary expenditures, occasionall y at the expense of calory intake, for such capital equipment as bicycles and sewing machines.

In some areas of the service sector, however, individual earnings might be considerably above subsistence levels under circumstances which would not encourage savings. The obvious case is employment in the public service. Here the government may force saving through taxation. Similarly, for the capital intensive sector, the rate of saving may be politically determined.

When urban immigration exceeds the rate of growth of manufacturing employment, the resulting expansion of work in the service sector is likely to be accompanied by a reduction in output per capita. This decline in productivity might be viewed as reflecting the fact that, given the variable capital-output ratio in services, the immigrants combine their efforts with less and less capital per head. Even if the rate of reinvestment in services is high, the rate of rural migration in most countries might easily be expected to surpass this. Since the capital-output ratio in the manufacturing sector remains relatively inflexible, the movement to the towns will not lead to a similar decline in output per worker there.

Individuals employed in manufacturing will typically add more to production than the value of their minimum subsistence needs. Whether new workers in the service sector actually produce such a surplus will depend largely on their rate of entry to the towns. Beyond some point, as immigration forces more employment into services, extra workers may not produce enough to cover their subsistence costs, and an income transfer to them may be needed from those with higher incomes.


Various circumstances suggest that employment in services may be less destructive of traditional ways of life than work in factories or mines. Employment in trade, especially, may be compatible with autochthonous customs and life-styles. Men and women, young and old, all may take part, and opportunities would seem unusually great for members of families to work together. Part-time employment may make it possible to divide time, for example, between the office or shop and the care of children. Easier control of oneís daily routine may make it feasible for urban immigrants to travel more freely to their former homes, and thus preserve traditional bonds. Customary systems of stratification may survive more easily, and the indigenous prestige of women may be maintained.

The incidence and intensity of some major social and psychological ills are influenced by the nature of the work undertaken by the urban labour force. With manufacturing (as in mining) it has frequently been noted that the arbitrary division of life into work and leisure is accentuated, and that many employers are unable to see a direct relationship between the final outcome and their own labour. Frequently, moreover, they are too poor to enjoy the consumption of those goods they have helped to create. Such circumstances give rise to frustrations and manifestations of social anomie, such as a rising incidence of alcoholism and family instability. But these problems are influenced by the nature of the economic opportunities available: employment in the services may provide links between labour inputs and economic outputs which are easier to understand, and the beneficial reinforcement of personal relationships may be more consistently maintained.

Today, in countries with a high income per capita, the service sector shows an increasing proportion of productive activities, and includes those jobs which appear most satisfying to human aspirations. If service employment in low-income countries could prepare workers for the potential attainment of desirable jobs, and if the growth of output need not necessarily be impaired in the process, this would seem an appropriate course for national economic growth. The fact that service employment emphasises relationships between people, rather than between men and machines, may suggest its relative superiority over factory work as a training ground for increasingly satisfying occupations. Moreover, the emphasis on self-control and self-motivation, rather than on outside pressures for routine performance, might reinforce this appropriateness in equipping individuals for more responsible positions.


In countries where few people spend any substantial portion of their lives in the formal educational system, improvements in the quality of labour force occur principally through work experience. As the majority of those leaving traditional agricultural pursuits in sub-Saharan Africa are likely to find employment in services, it is necessary to look to this sector for widely effective changes in the human factors of production. And it is here, moreover, where there is less blind destruction of traditional ways of life than occurs with ëmassí employment in mines and factories.

This suggests that an appropriate social goal for many African countries might be to facilitate economic involvement in service activities for those otherwise unemployed, and to provide for the continued growth of those so engaged. Various programmes might be designed to this end, for example: (i) monopoly exclusions should be proscribed which limit employment on the basis of race, ethnic background, or use of economic power; (ii) artificial barriers to professional achievement, such as examinations and licenses, should be re-examined to ensure their appropriateness to ëprotectí the public; (iii) school courses should be devised to assist more effectively in the attainment of useful skills; and (iv) more adult schemes of training should be inaugurated for the acquisition of relevant expertise and competence.

New opportunities for employment could be created by legislation, or by administrative action, to enable more people to perform useful services in certain professional fields, even although they might not be fully ëqualifiedí. Thus para-medical and veterinary posts could be established to enable trained teams to inoculate both villagers and their animals, or to diagnose specific diseases, or even to treat certain illnesses and injuries. Comparable legal skills could be developed to enable individuals to prepare deeds or wills, or to give advice concerning specific administrative procedures. Teaching assistants could be trained to help in a great variety of fields, including reading and mathematics.

The availability of such specially designed courses to impart these skills, and the acceptance of the qualifications of those so trained, might provide extensive new opportunities to increase human productivity. Such programmes could, of course, be arranged to provide work while learning. Even though an individual so trained did not have the competence of an academically qualified professional, he could satisfy needs not currently being met, and could provide for meaningful development of his skills in the process.

The achievement of such employment objectives does not require any African nation to turn its back on modern technology. A multifaceted approach is required which would include as a key component the promotion of significant and remunerative human involvement in service activities.