Contents - Previous - Next

This is the old United Nations University website. Visit the new site at http://unu.edu

The maintenance of firms' capital equipment

Lest we give the impression that local governments and firms always choose the best allocation of scarce resources to advance science and technology, and foreign donors never, we shall single out one activity internal to productive organizations in which the latter appear to have made the better choice: this is the activity which leads to the maintenance of the productivity of capital, both physical and human. In singling out maintenance and training we shall first examine their roles in the activities of Sub-Saharan African firms, public and private; and then report on our and others' observations of the current contributions made by foreign donors.

Our evidence on the roles of maintenance and training is drawn chiefly from two countries, Ghana and Tanzania. The Ghanaian source is Lall et al. (1994); the Tanzanian is Mjema and Kundi (1993), on maintenance, and our own interviews with Tanzanian educators and foreign consultants (on training). So far as maintenance in Ghana is concerned, Lall and his colleagues investigated both the maintenance of product quality and of equipment. Of 30 firms in four industries (textiles and garments, food processing, wood working and metal working), 18 had no employee assigned to quality control and 14 no one to equipment maintenance (ibid.: Table 6.2). None of the 14 firms without equipment maintenance personnel fell into Lall's category of 'technologically competent' fines (ibid.: Table 4.3; 11 firms for which data were reported in his Table 6.2); the 11 'technologically competent' firms all assigned a higher percentage of their total employees to quality control and equipment maintenance than the others.

In Tanzania the situation regarding maintenance is quite similar. Mjema and Kundi's sample of firms was larger, comprising over 50 firms, although their inquiry was limited to the maintenance of equipment. Of the 50 firms answering the questionnaire, exactly half replied that they kept no records of maintenance; of the half who reported (two very large, 23 medium-and small-sized firms) 90 per cent repaired equipment when it broke down, 50 per cent carried out some preventative maintenance (regular oil changes, lubrication, etc.), but only 10 per cent of the respondents had a system of predictive maintenance. In those cases where a machine fell out of service because of a failed part, the tendency was to order a replacement only after the failure occurred. The larger firms employed qualified engineers within a maintenance department, but 40 per cent of the respondents relied upon technicians or artisans. Finally, asked reasons for the existence of problems in maintaining equipment the two most frequent answers were a lack of spares (75 per cent of respondents) and of skilled personnel (70 per cent).

So far as the maintenance or augmentation of human capital is concerned, Lall and his associates found that one factor leading to technological competence in the firms they surveyed was the deliberate investment in creating skills and information (ibid.: 85). More highly educated entrepreneurs and production managers were found in the more competent firms (ibid.: 86), as were higher proportions of scientists, engineers and technicians in their workforce (ibid.). They found it difficult to judge the effects of external training on firm efficiency, although the extent to which external training is practiced by Ghanian firms seemed very low by world standards. Internal training, chiefly through the traditional method of apprenticeships, but also, on the parts of affiliates of foreign firms and local firms managed by foreigners, through on-the-job training schemes, is common (ibid.: 111 and Table 5.6); although some firms in the wood and metal working industries appear to employ apprentices as ordinary workers at low wages. In summary, it is only the larger and more progressive firms that approach maintenance and training systematically and devote any efforts to their conduct.

As to the contributions that foreign agencies make to maintaining the productivity of physical and human capital in Sub-Saharan Africa, our impression is that they are proportionately more substantial and more effective than those of local firms and governments, but substantially less so than similar contributions in the rapidly growing countries of Asia and Latin America. The main vehicle for the contributions in this applied area of science and technology is the industrial consulting firm, hired by the foreign donor to help restore the facilities and improve the performance of large firms (public and private but more often public) engaged in manufacturing and the provision of services (but more often services). Examples of the recipient African institutions are railways, highways, ports and harbours, post and communications, hospitals, schools and universities and public administration. The foreign consulting firms arrive knowing that maintenance is one of the integral functions of any enterprise and aware that Sub-Saharan African firms are ill-equipped, both intellectually and professionally, to carry out the function. (In Swahili there are no words for 'maintenance', 'reliability', 'availability'.) They therefore expect to devote a good portion of their consulting to establishing systems of maintenance, training managers and operators, obtaining implementation, and persuading all involved, at all levels of authority, of the necessity for undertaking the task. Given their terms of reference the consultants are required to work on improving activities within the firm to which they are attached; but their employers, the foreign donors, may also finance ancillary, outside projects, such as the publicizing of recommendations and the provision of courses (e.g. the introductory courses on maintenance laid on by TEMDO in Tanzania).

We do not know how effective these foreign-inspired efforts to promote the adoption of programmes of maintenance of physical and human capital are, but we can conclude that, even if they are extraordinarily effective within their own terms of reference, they, and their local counterparts, are a tiny fraction of what is needed. One comparison will illustrate this conclusion: maintenance can only be performed by technical personnel, who are so very scarce in Sub-Saharan African countries. Tanzania has an enrolment of approximately four thousand students in its technical colleges; South Korea, albeit with half again as many inhabitants, has nearly one hundred times as many. The deficiency in technical training is of an order of magnitude higher even than in R&D, to which so much of our, and others', attention has been directed.

The effectiveness of R&D efforts

Yet we return to R&D, for it is to R&D that most of our data apply. The selection of products-with-potential, towards which R&D should be directed, may be a necessary condition for its contribution being as large as possible, but it is not alone sufficient. Whatever scarce resources are allocated to R&D should be applied efficiently. What does our research reveal about the effluency with which the countries in our sample conduct their R&D, in the course of Structural Adjustment?

Questions concerning efficiency are notoriously difficult to answer when R&D is carried out in developed countries; in developing countries answers are even more difficult, but also even more important, to obtain. It is the importance of the answers that leads us to attempt them, even if the confidence that we can place on them is meagre. We shall focus on three matters, for which we gathered some evidence: two of these three have to do with the sources and uses of funds for R&D; the third with the allocation, within a single organization, of effort among competing activities.

Let us consider first the sources of finance for advancing science and technology. In Ghana, Kenya, Tanzania and Uganda private firms and individuals have little money to spare for such problematic activities as R&D, with such uncertain returns. The author of one study, for Kenya during a relatively prosperous period (see Chapter 4), estimated that private agents carried out no more than 10 per cent of the country's total R&D, a paltry fraction.

Domestic finance for R&D sterns from the government, either directly through the budget or indirectly through levies on the revenues of parastatal firms. The former public source is generally the larger in total, the latter generally the more closely related to need. The greater volume of public funds devoted to R&D has been documented in the previous chapter: the Ghanaian government spends approximately as much public money on R&D institutes as do the para-statal bodies; the Kenyan government spends approximately twice as much; and the Tanzanian and Ugandan governments account for almost all domestic R&D expenditures.

The close tuning of para-statal contributions to needs is illustrated by our data on Ghana's R&D institutes' requests for and authorizations of funds. Table 9.2 provides the data; juxtaposed are requests and authorizations for one para-statal research organization and four public R&D institutes. The results are clear; the para-statal organization, financed indirectly by levies on sales of the para-statal's crop, had approved, on the average, 86 per cent of all its requests, whereas the public institutes, financed directly from the government's budget, had approved, again on the average, 71 per cent of their requests for recurrent funds and 56 per cent for development funds.

Moreover, the data in Table 8.2 reveal that there was less fluctuation year by year in the para-statal's percentages, particularly where Development items were concerned. This observation deserves some comment, since it seems to be general across Sub-Saharan African countries and since it has implications for the efficiency with which resources are applied in R&D. To conduct R&D effectively it is necessary to plan activities so that objectives are agreed upon, and matched with resources, and so that the resources are available in the types and at the times that they are required. To plan coherently one needs, at a minimum, stability, in the senses both of a relatively assured and steady source of finance and of a core of experienced personnel. If the income of an R&D institute is insecure, and if it fluctuates widely from year to year, coherent planning is not possible. Goals cannot be set with any likelihood of their being attained, resources cannot be procured in synchrony with the plan, and standards of performance cannot be established, let alone attained. In such an uncertain environment, a systematic allocation of resources, so vital for an activity with such a long gestation period as R&D, becomes impossible and what resources as are applied are used inefficiently.

Table 9.2 Ghana approvals as a percentage of requests for expenditures for para-statal and public R&D institutes 1981-1991

Sources:
Tables 3.9-3.12
Note:
^a: The Industrial Research Institute (IRI), Scientific Instrumentation Centre (SIC), Food Research Institute (FRI) and Technology Transfer Centre (TTC)
^b: Simple arithmetic averages of the four public institutes, and simple arithmetic averages over the 11 years

Let us consider another matter covered by the statistics above. Commonly, in Sub-Saharan African research institutes, the major (usually almost the sole) use on which recurrent items are expended is wages and salaries of the employees of the institutes. The figures for Ghana over the period 1974-1981 are typical: in the eight-year interval wages and salaries consumed from 88 to 98 per cent of the total expenditures (95 per cent on the average) of all the country's public R&D institutes (see Table 3.16). In the other three countries in our sample, much the same phenomenon was observed for recent years. To be sure, these figures exclude foreign loans and gifts, most of which are for development items, but the conclusion is that almost all local funds are devoted to paying the wages and salaries of the institutes' personnel. Negligible sums are left for buying books, subscribing to scholarly journals, ordering supplies of reagents and other working materials, constructing laboratories and field stations, and procuring laboratory equipment (corroborating evidence is provided by Gaillard, 1991: 68 75). Even those pieces of capital equipment that the institutes do procure, usually with funds from foreign donors, cannot be properly maintained and repaired; the institutes themselves may not employ capable technicians ('... Africa seems to be in the worst position since over half [51 per cent] of the institutions on this continent do not have the technical staff required to ensure that their scientific equipment will work well...' ibid.: 70) or may not have been allocated the foreign exchange necessary to secure spare parts and/or hire technicians from abroad ('more than two-thirds of the scientists [68 per cent] had to wait 5 months or more to have their equipment repaired by a foreign technician, and more than one-fourth had to wait 10 months or longer,' ibid.: 72). With almost all their funds consumed in paying wages and salaries, the R&D institutes can tick over; but, isolated, uninformed and impecunious, they cannot fulfil their potential.

There is even some evidence that the allowances for Recurrent items are not even sufficient to keep some institutes ticking over. In such cases, it is those employees not entrenched in the resident bureaucracy who tend to be released, agricultural extension workers in Ghana and Kenya having been one such vulnerable group (Duncan and Howell, 1992). One can hardly imagine an act more likely to delay the dissemination of results of R&D in agriculture, and hence more likely to reduce R&D's efficiency, than cutting extension staff. To the extents that meeting the conditions on Structural Adjustment loans requires a reduction in government expenditures, that a reduction in government expenditures leads to a reduction in funds for public R&D, and that a reduction in funds for public R&D bears most heavily on 'development' items and on extension services, the outcome will be a reduction on the effectiveness with which the remaining R&D is carried out.

The above argument rests on the assumption that scholarly books and journals, laboratory supplies, scientific equipment, extension staffs, etc. cannot be replaced by more scientists. To be sure, it does appear that there is an abundant supply of university-trained scientists in Sub-Saharan Africa, and that employment in a public R&D institute is a career to which most aspire; but the question does arise as to whether or not scientists, maintained in employment with domestic funds, can substitute for the other factors of production requiring scarcer resources. Putting the question in the marginal terms familiar to economists, would an extra scientist contribute more or less to the output of an R&D institute than (the last) 50 scholarly books or (the last) five research journals or the last three extension workers? We suspect that the answer, given the conditions prevalent today in Sub-Saharan Africa, is that the marginal scientist would contribute less to R&D output than the alternative uses of expenditure.

A counterfactual experiment

Having addressed the issues of the choice of product towards which R&D should be directed, the efficiency with which R&D is carried out, and the extent to which there is substitutability in R&D between more abundant and less abundant factors of production, we shall now try to combine these within a single frame of analysis. The frame we shall use is that of Fung and Ishikawa, described in Chapter 2 and applied already in Chapter 7.

We have argued in this chapter that the consequences of the adoption of Structural Adjustment Programmes are leading to R&D being directed increasingly towards products that have unfavourable prospects for the Sub-Saharan countries and, perhaps, to its being performed with lower efficiency (because of domestic constraints on its financing, the constraints being felt most acutely on non-wage and salary components of expenditures and, within the wage-and-salary component, on bureaucratically isolated staff). Let us now use Fung and Ishikawa's model to illustrate these consequences.

Within the model, the consequences of Structural Adjustment are represented by three specifications: first, R&D is directed towards good Y (for which the country's potential is unfavourable), rather than good X; secondly, the efficiency with which resources are employed in R&D is lower; and thirdly, the substitutability of inputs in production (of R&D, and consequently of R&D's output, intermediate goods) is reduced. In order to illustrate these alterations, we shall construct Case III, and compare its implications to those of Case I, which are the most attractive for a Sub-Saharan African country. The differences in assumptions are as follows:

1 in Case III, R&D is allocated to advancing science and technology in good Y (the good with unfavourable prospects for Sub-Saharan Africa) whereas in Case I R&D is allocated to good X (with favourable prospects);

2 more R&D is carried out in Case III than in Case I (foreign grants more than compensating, in Case III, for any reduction in local expenditure);

3 the efficiency with which R&D is carried out (measured by the parameter d ) is lower in Case III than Case I; and

4 the substitutability of factors of production in R&D, and subsequently in intermediate goods (measured by the parameter b) is lower in Case III than Case I.

For both cases we shall consider the long run, over which R&D comes to fruition and the terms of trade for good Y, vis-à-vis good X, steadily deteriorate.

Four figures are necessary to display fully the comparison of Cases I and III. The first compares the possible production patterns, the second and third, trading postures, and the fourth consumption patterns. Common to both Cases I and III are identical volumes of domestic consumption of good Y (the good with unfavourable prospects: we could think of it as coffee); and identical relative prices of goods X and Y (i.e. identical terms of trade), based upon assumptions that the country is 'smell' (i.e. does not, by its own behaviour, influence world market prices) and 'open' (i.e. that domestic prices are set by world market prices). Over time, the world market and domestic price of Y falls relative to the price of X.

Looking first at Figure 9.1, we represent the initial production pattern (at t (o) the 'o' standing for the present), common to both cases, by the heavy solid curve (X, Y)^c, this curve representing the locus of all possible pairs of outputs of X and Y. given current (c) technology and resources.

Future production patterns will differ, depending upon the direction of R&D, as well as the portion of the country's total resources allocated to it, the efficiency with which they are applied, and the substitutability of scarce inputs one for another. So far as the direction of R&D is concerned, we are assuming that it is towards good X (the good with favourable prospects) in Case I and towards good Y (the good with unfavourable prospects) in Case III; so far as the volume of resources devoted to R&D is concerned, we are assuming no difference between the two cases; and so far as both efficiency of R&D and substitutability of scarce inputs are concerned, we are assuming that they are higher in Case I than III.

Figure 9.1 Production possibility curves for the present and future under Cases I and III

Given these assumptions, the production possibility schedules (drawn in light solid lines in Figure 9.1) at some future date will be (X, Y)^f_I and (X, Y)^f_III the superscript 'f' standing for some distant future date, and the subscripts indicating the cases. In Case I, over time, the production possibility curve has swung out to the right, pivoting on its intersection with the vertical axis, indicating the growing productivity of the country's resources in producing good X, consequent upon its advancing technology. In Case III, in a similar manner, the production possibility curve moves upward, pivoting on its original intersection with the horizontal axis. Over time, as a comparison of the two curves (X, Y)^f_I and (X, Y)^f_III, show, the productive potentials of the country become more and more different: in Case I it is an ever-increasingly efficient producer of X, in Case III of Y.

Let us turn to Figures 9.2 and 9.3, in order to compare the different effects of increasing efficiencies of production on the country's international trade. Remember that the terms of trade are identical for both cases, and have deteriorated over time for good Y, vis-à-vis good X. The original terms of trade are 1:1 (as drawn in Figure 9.1), which was represented graphically by a (dashed) line of 45 degrees: let us represent the terms of trade in the distant future as a (dashed) line of 60 degrees. (At time zero, one unit of Y commands the same price in world markets as one unit of X: hence terms of trade 1:1. In the distant future, because of the deterioration in the terms of trade, the price of a unit of Y will be, say, only half that of a unit of X: hence terms of trade 1:2. Terms of trade of 1:2 are represented in Figures 9.2 and 9.3 by (dashed) lines of slope 2, or of angle 60 degrees.)

The slopes of the relative price (the dashed) lines are significant, for they represent the country's future possibilities for trade. Assuming that the country is to be in balance of payments equilibrium (i.e. value of total imports equal to value of total exports), the country can, in Case I (see Figure 9.2), trade anywhere along the (dashed) line passing through p(X, Y)^f_I, and, in Case III (see Figure 9.3), anywhere along the (dashed) line passing through p(X, Y)^f_III,. (The reason is that the relative price or terms-of-trade lines are the loci of points of equal total value of outputs of X and Y. Between any two points along the relative price lines the increment in value of the good whose quantity is increasing is equal, but opposite in sign, to the increment in value of the good whose quantity is decreasing.

Figure 9.2 Terms of trade, exports and imports at some future date under Case I

Figure 9.3 Terms of trade, exports and imports at some future date under Case III

For example, in Figure 9.2, the value of the additional quantity of good X between the points p(X, Y)^f_I and c(X, Y) ^f_I is equal, but opposite in sign, to the value of the reduction in quantity of good Y. The country can, by exchanging Y for X (or conversely X for Y) move from point p(X, Y) ^f_I to c(X, Y) ^f_I (or conversely from c(X, Y) ^f_I to p(X, Y) ^f_I and remain in balance of payments equilibrium. Let us imagine that the country, in Case I, produces at point p(X, Y) ^f_I. The fetter 'p' stands for production, and, assuming that production is efficient, p(X, Y) ^f_I will lie along the production possibility curve (X, Y) ^f_I. Let us also assume that the country wishes to export an amount of Y equal to Y ^e_I ('e' for export). At world market paces, the country can exchange this amount of Y for an amount of X equal to X ⁱ_I ('i' for import). Therefore by giving up, to foreigners, the amount of Y equal to Y ^e_I the country can acquire for its own citizens an extra amount of X equal to X ⁱ_I. Any point along the (dashed) relative price line is a point of potential consumption; the point c(X, Y) ^f_I ('c' for consumption, the use to which all goods are assumed to be put) is selected to represent the country's actual choice. The country produces goods at point p(X, Y) ^f_I, consumes at point c(X, Y) ^f_I, exports Y(X, Y) ^e_I, imports X ⁱ_I, and is in balance of payments equilibrium.

The outcome for the country in Case III is determined analogously, and portrayed in Figure 9.3. Total exports Y ^e_III, and imports X ⁱ_III are considerably larger in volume than they are in Case I, a result to be expected since, in Case III, the country is specializing in the production of the good that had unfavourable prospects (a good like coffee); but, because both cases operate under identical terms of trade, the ratios X ⁱ_I¸ Y ^e_I and X ⁱ_III¸ Y ^e_III are equal (to 1 ¸ 2).

We have yet to explain how the production points p(X, Y) ^f_I and p(X, Y) ^f_III and consumption points c(XY) ^f_I and c(X, Y) ^f_III were selected, and how the final outcomes compare. The production points were selected so as to maximize the value of total national output, given the country's technological potential, resources and relative prices (i.e. terms of trade). Graphically they are the points of tangency between the production possibility curves, (X, Y) ^f_I and (X, Y) ^f_III, and the (common) relative price line.

The consumption points are selected so as to maximize satisfaction from total consumption of X and Y. according to the community's relative preferences for the two goods. In both cases, the quantity of good Y preferred is equal, at c(Y) ^f_I and c(Y) ^f_III in Figure 9.4. (Thinking of it as a good like coffee, domestic demand can be assumed, as a first approximation, to be independent of the price it commands in world markets.) As a result of this assumption there is no ground to prefer the outcome in Case I or Case III, or vice versa, so far as consumption of good Y is concerned.

Figure 9.4 Total consumption at some future date under Cases I and III

It is in the consumption of good X (the good which had favourable prospects and towards which R&D was directed in Case I, but not in Case III) that the difference between the two cases arises: having directed its R&D towards the good with favourable prospects the country can consume more of it in the future (c(X) ^f_I exceeds c(X) ^f_III in Figure 9.4). The outcome is unambiguously superior in Case I: consumption of the same amount of good Y. and a greater amount of good X. Focusing R&D on the good with improving terms of trade has been the contributor to the happy outcome.

Summary

In this chapter on the effects of Structural Adjustment within organizations, we have focused on those organizations whose chief function is the pursuit of science and technology. The major issues confronting these organizations are, in our judgement: first, towards which products/processes should they direct their efforts? and secondly, how should they improve their efficiency? We argued that the direction of R&D should be towards those products/processes with the most favourable prospects for the future, where favour will be determined not only by movements in the terms of trade, but also by comparative advantage in production and distribution, its efficiency in conducting R&D, and ability to disseminate the results. The data we collected and analyzed suggested means by which organizational efficiency could be improved: by maintaining, year by year, relatively stable funding and by re-allocating internally some of the funds from wages and salaries to the non-wage and salary components (supplies, reference material, equipment, maintenance, etc.). Dissemination of results could be more effective if employment of core staffs, rather than extension workers, was curtailed.

Where the proper direction of R&D efforts was concerned, we argued that the choice should be made deliberately, with several factors in mind trends in terms of trade for individual commodities; potential for improvements of production methods through R&D, both in the country under consideration and elsewhere; support for, or opposition to, the expansion of production and export, etc. Such deliberations should be done systematically and with the long-run interest of the developing country in mind. We shall consider this matter further in Chapter 11, where we will see, unfortunately, that proper choices are extremely difficult to make, both bureaucratically and politically.

Although we did not attempt to make any such choices ourselves, we did illustrate, via a counterfactual experiment, the consequences of right and wrong choices ('right' and 'wrong' from the point of view of the developing country). Case I illustrated the right choice of direction of advance of science and technology, Case III the wrong. Comparison of the two cases revealed quite different patterns of production (specialization on the production of the good without favourable prospects in Case III as against production of the good with favourable prospects in Case I), quite different volumes of international trade (large exports of the good with unfavourable prospects in Case III as against small and diminishing exports of the same good in Case I), and a lower level of consumption (equivalent to a lower standard of living) in Case III than in Case I. To be sure, this was only an illustration; but we know of no better way of making explicit the effects, over the long-run, of choosing the right rather than the wrong realms in which to advance science and technology.

Contents - Previous - Next