STEEP Discussion Paper No 36
Underdevelopment of markets a barrier
to technological efficiency? A comparative study of ministeel plants in
India and the UK
Suma S Athreye
Centre for Business Research Department of Applied Economics University of Cambridge
Science Policy Research Unit
University of Sussex
East Sussex BN1 9RF
Tel: +44 (0)1273 686758
Fax: +44 (0)1273 685865
? Suma S Athreye 1996
1 Introduction 1
2 Relative underdevelopment and technological dynamism 2
3 Steel markets and differences in the UK and Indian steel markets 9
4 Firm behaviours: costs of specialisation or diversification to steel firms 11
5 Explaining the variation in input productivity among firms 18
6 Implications of the lack of industrial deepening 25
7 Summary and conclusions 26
1: Associated lines of business activity for the firms in the sample 13 2: Variable names, description and units of measurement 15 3: Performance data for the sample of firms, by nationality 16 4: Results of the one way ANOVA by variable IND 16
5: Principal components extractionwith real input consumption variables 20 6: Principal component extraction with cost share variables 23
How could relative underdevelopment pose a constraint to the harnessing of technological change and increased output productivity due to technological developments? In this paper one possible chain of causation in the understanding of this question is suggested. It is argued that fragmentation of markets and the absence of industrial deepening, both factors characteristic of developing country markets, could constitute important barriers to the adoption of firm strategies that promote technological efficiency in producer goods sectors. The fragmentation of markets (due to poor infrastructure) prevents specialisation and thus impedes technological learning through this route. The lack of industrial deepening implies that a large proportion of demand for intermediate producer goods emerges from households in the economy rather than firms. To the extent that firm-based demand is more likely to penalise poor performance and quality, which are essential for effective technological learning, an absence of industrial deepening reinforces the effects of non specialisation for technological learning by firms in the producer goods sector. Both these factors may cumulatively act to produce a lower level of technical efficiency for firms in underdeveloped countries, both within the particular sector and for sectors dependent on the producer good sector. The evidence used in the construction of this argument comes from a comparative survey of Indian and UK electric steel producers.
The successful catch-up of growth rates and levels of per capita incomes across countries has been a much studied phenomenon (Abramovitz, 1986, Barro and Sala-i-Martin, 1992). The large literature on this area acknowledges the role of traditional factors like rates of physical capital accumulation and rate of growth of population in explaining this process of convergence (Mankiw et al, 1992) of growth rates of per capita incomes across countries.
Also acknowledged as important in explaining the catch-up process are non-traditional factors, such as, the rate of technological change, the nature of technology between countries (Bernard and Jones, 1994, 1996) and the rate of accumulation of human capital and education and increasing returns thereof in explaining the closure of initial gap in per capita incomes between latecomers and early industrialisers (Lucas, 1988).
The catch-up literature also acknowledges two other facts about the growth across countries. Firstly, successful catch up appears to take place when the initial differences in per capita levels and the gap in the initial levels of development between countries are not very high. Convergence of per capita incomes between countries is thus "conditional" upon structural similarity (in preferences technologies, rates of population growth, government policies, etc). Thus, OECD countries show convergence, while an expanded sample of OECD and non-OECD countries does not show convergence (Baumol, 1986; Barro, 1991; Sala-i- Martin, 1996). Secondly that, divergence and falling behind is as common as convergence and catching up. This is the so-called "club convergence" hypothesis that the per capita incomes of countries that are identical in their structural characteristics converge to one another in the long run provided their initial conditions are similar as well (Durlauf and Johnson, 1995; Quah,
11996). These authors generally also predict both convergence and divergence as possible.
Somewhat different from the above theoretical literature, is the work of economic historians such as Abramovitz (1986, 1993), Maddison (1979, 1982, 1991) and much earlier work of
1A recent volume of the Economic Journal (July 1996) has an excellent review of all the issues involved in the
convergence and catch-up literature.
Gerschenkrohn (1962). This literature has stressed the scope for catching up as contained within the social capability of an economy to absorb technologies. All these findings together suggest that relative underdevelopment itself may constitute an important constraint in preventing a catch-up to other countries in terms of economic growth. To the extent that the rate and nature of technological change is important to such a process of catch-up and growth, relative underdevelopment could be a barrier to the harnessing of technological changes to promote growth as well.
How could relative underdevelopment pose a constraint to the harnessing of technological change and increased output productivity due to technological developments? In this paper one possible chain of causation in the understanding of this question is suggested. I will argue that fragmentation of markets and the absence of industrial deepening, both factors characteristic of developing country markets, could constitute important barriers to the adoption of firm strategies that promote technological efficiency. The evidence used in the construction of this argument comes from a comparative survey of Indian and UK electric steel producers.
The paper is organised in the following way: Section 1 surveys briefly the literature that addresses the problem of technological dynamism and underdevelopment, and the role of market size in explaining the relation between the two. Sections 2-5 describe the empirical evidence used in the paper and Section 6 draws out some conclusions.
2 Relative underdevelopment and technological dynamism
An early and profound perspective on how development and technological dynamism were inextricably linked was provided by Adam Smith in the Wealth of Nations. Smith argued that
the extension of exchangeable market scale, promoted the division of labour within a firm which in turn improved productivity. Thus, the increased division of labour was directly related to the extent of the 'market', or the opportunities that were available for exchange.
Further such an increased division of labour between different tasks in the pin factory (relative to the situation when one workman made all the pins), permitted an increase in output productivity, through:
(i) an improvement in the dexterity of workmen carrying out a single operation
(ii) a saving in the 'time commonly lost in passing from one sort of work to
(iii) the introduction of new machines that facilitate and shorten the work.
It can be seen that in one bracket three different sources associated with improved productivity are discussed, which are often a part of any discussion on technological dynamism. The first is what is now commonly understood as specialisation and the acquisition of job-specific skills. This can also come from training, but division of labour beyond a point can result in the splitting of tasks into relatively simple routines that require no more than simple skills. Smith's argument was however was not the same as allocating workers to the job that they did best. It clearly saw skills and comparative skills as arising from the division of labour under the factory system, which in turn was permitted by the possibility of (profitable) exchange. In other words, in the optimal sized market, each worker could be fully employed in the task that was allotted to him. At the other extreme was the small market where one man produced all the pins to keep himself fully employed, and this was necessarily less productive of output.
The second source relates to the co-ordinating and organisational dimensions of the factory system, as opposed to craft manufacture. The job of entrepreneurship is thus removed from the domain of physical production to one of co-ordinating and selling. In this way there is also a division of labour between the entrepreneur and the workers, but this division is different from the fragmentation of tasks that accompanies the division of labour for the worker. In a way, the combining of these various tasks becomes a task in itself. More interesting we find here the perception that increasing scale of production may require corresponding changes in the organisation of that technology of production. Indeed in recent times the move away from
mass-production in factories based on assembly line technologies to flexible production based on niche markets and micro-electronic technology has shown the relatedness and importance
2of organisational changes to technological changes.
The third source of increasing productivity due to the increasing specialisation of tasks, is the possibility of mechanisation of some tasks. In its extreme form it is an extension of the division of labour with some parts of the production tasks being substituted by machines. Output productivity is enhanced by labour displacement and its substitution of labour power by machine power through automation.
Thus, the three sources of productivity gain discussed in Smith touch upon the different dimensions of technological dynamism that have been discussed in economic and industrial economics literature. Economic history of the advanced industrial countries is witness to the importance of mechanisation and machination of production. Labour skills and organisational factors have been seen to be important in explaining the technological successes of firms in the East Asian countries
It is also worth noting however that in the Smithian perspective development proceeds through an expansion of exchange relations which in turn permit productivity increases which in turn further increases the possibility of an expansion in market scale. The source of the productivity increases are due to the different impacts of increasing division of labour and the consequent change in techniques and organisation of production. The profundity of the Smithian argument rests in the fact that this division of labour refers to both intra-firm and inter-firm division of labour. At the inter-firm level, division of labour could take the form of the capacity for specialisation in different branches of production, and the emergence of specific industries. This understanding underlies several notions about development. Thus, for example, the importance of industrialisation for economic growth and development is an
2This strand of the Smithian argument and its application to the understanding of current debates on flexibility is well reviewed and analysed in Morroni (1992).
extension of the inter-firm division of labour argument. Industry is subject to (limitless) increasing returns because of the possibilities of division of labour and the associated productivity increases. This is unlike agriculture where there are physical limits to increased productivity due to the scarcity of land.
Similarly, that division of labour and functional specialisation could lead to vertical disintegration of process technologies, and the differentiation of industry was pointed out by Stigler (1951). In the face of increasing returns, division of labour can take place at a inter-firm level and cause firms to become specialised in some stages of a long production process. Markets then could emerge in the intermediate stages of a complex product (eg, component manufacturers and cars, or hardware and software producers of computers). Though his analysis was directed to the life of a single industry they are equally applicable to economies. To quote:
"Young industries are often strangers to the established economic system. They
require new kinds or qualities of materials and hence make their own; they must
overcome technical problems in these of their products and cannot wait for potential
users to overcome them; they must persuade customers to abandon other commodities
and find no specialised merchants to undertake this task. These young industries must
design their specialised equipment and often manufacture it, and they must undertake
to recruit (historically often import) skilled labour. When the industry has attained a
certain size and prospects, many of these tasks are sufficiently important to be turned
3over to specialists."
The more often this happens and there is such vertical disintegration and specialisation, the greater is the size of the industrial sector, and the greater is the industrial deepening of the economy. Industrial deepening in the sense I use the term in this paper refers to the increasing roundaboutedness of production in the economy. This is a consequence of the increasing penetration of exchange relations and related to the growth of specialisation in the economy. One consequence of industrial deepening is that we expect other firms to be more important as customers for products in the producer goods sectors, than final demand in household.
3Stigler (1951), p190.
Another way of putting it is to say that derived demand is more important than final demand from households for firms in the producer goods sector.
Industrial deepening and a predominance of derived as opposed to final goods demand are characteristics that separate developed industrial markets from relatively underdeveloped ones. The two features are related in the following way. A large scale final demand is necessary for any level of derived demand to emerge. Indeed it may even be argued that two-sector models of growth sought to apply this understanding in a pro-active way to cause development by the setting up of a capital goods sector in the economy. However in being pro-active they saw the supply side deficiency and sought to correct it. The experience of countries like India who "planned" to grow in this way, points to the problems caused by the neglect of the demand
4creation side of the process. By the mid sixties growth based on such expansion had slowed
down and capital goods industries in particular were characterised by excess capacities and high unit costs, which could be traceable ultimately to a narrow base of domestic demand for industry (Nayyar, 1976; Khanna, 1989). Equally the overwhelming evidence about the important dynamic effects of export markets in explaining the growth of the Far East Asian countries is also testimony to the importance of the expansion in the scale of demand through international trade in countries that have a small scale of the domestic market.
Rosenberg (1963) explicitly linked the presence of such industrial differentiation and the existence of economies of specialisation to effective market size and to technological dynamism. In the context of the capital goods sector, he argued that the technological development of the machinery goods sector (which was reflected primarily in the range of goods that it could produce) in developing countries, is limited by the scale of demand for such capital goods. Since technical development has the strongest 'technological' linkages with the rest of the economy he saw this central problem as being of crucial importance in explaining why developing countries could not develop the labour -using technologies that
4See for an example of an early criticism of this kind, Bhagwati and Desai (1970).
they needed for their growth, in a manner similar to the American development of capital-using technologies to economise on scarce labour.
More important than this central argument, for this paper, are the more general observations that Rosenberg makes about the economies of specialisation, their importance for generating technological dynamism and the importance of market scale in this process. The first observation relates to the importance of economies of specialisation. Economies of specialisation arise due to the concentration of production on a relatively narrow range of products, which in turn require a relatively homogenous collection of firm-level resources in production. How is such specialisation important for technological dynamism in the economy? Again to quote:
".... there is an important learning process involved in machine production, and a high
degree of specialisation is conducive not only to an effective learning process but to an
effective application of what is learnt. This highly developed facility in the designing
and production of specialised machinery, is perhaps, the most important single
characteristic of a well-organised capital goods industry and constitutes an economy of
5enormous importance to other sectors of the economy."
The second observation is about the importance of geographical market integration, for the development of a specialised machinery goods sector. Indeed it is pointed out that the development of a specialised machinery producing sector in America coincided with the laying down of a railway network.
"The growth in the size of the market for individual producers of machines, resulting
from the reduction of freight costs, was peculiarly important to the process of
6specialisation in the production of capital goods."
5Rosenberg 1963, p220. Though Rosenberg in making these arguments was careful to distinguish between producer durable goods and intermediate goods like steel which may be subject to traditional economies of scale, we will argue that parts of the steel market such as the special steel sector have the same characteristics as the capital goods sector. 6Rosenberg 1963, p222.