April 2002


The ability of economies to increase the welfare of their citizens depends upon their ability to generate economic growth and employment.

Governments, independently of their philosophical predisposition, have an important role to play in the stimulation of their economies. The globalisation of the markets, the recognition of the importance of innovation for growth and the support of rival industries by their governments provide some of the challenges.

Monitoring of and reacting on programmes and policies of rival countries are important. However, within the global economy, the national location and character of particular economic activities matter, because different economic activities provide different growth and employment potentials. Governments amidst the rivalry of competing theories and practices have to keep pace with rapidly changing socio-economic realities and provide leadership.

This submission to the Portfolio Committee on Trade and Industry and the select Committee on Economic Affairs purports to lead towards a more effective and efficient industrial policy in South Africa.

More specifically the submission aims to provide answers to questions such as:

The submission first describes the legitimate reasons and rationale for government interference in the market place. Knowledge of the raison-d'etre for government intervention is the cornerstone for the identification and development of an optimum set of actions.

The next section provides a conceptual framework for the development of industrial policy. The classic approach of identifying policy targets and instruments is utilised. The width of the field of industrial policy and the cross responsibility of government departments and agencies point towards particular policy avenues.

The section A Time of Change focuses on more recent factors affecting the development of appropriate industrial policies. The issues of technological growth, globalisation and diffusion of services and manufacturing activities are in the centre of the section. The document ends with a number of suggestions for Reaching our Goals.

The Rationale for Industrial Policy

One of the amazing weaknesses of most contemporary industrial policy- making is the lack of a solid theoretical foundation and the lack of systematic learning from experience. In spite of the common tendency to justify actions in terms of theory and/or experience, the simple fact is that learning from theory and experience is accidental and sporadic.

Bert provides an incisive perspective of the character of industrial policy according to the wider theoretical framework underpinning it. He states:

>From the free-market perspective industrial policy implies subsidisation of inefficient enterprises, from the orthodox Keynesian perspective it is associated with market interference for stabilisation purposes, from the mixed economy perspective it suggests correcting for market failure with public planning, and from the socialist economic perspective it means substituting planning for markets as a superior mechanism for coordinating resource allocation. But since these perspectives share the plan-market dichotomy, none adequately captures the role of industrial policy in post-war Japan. Here the purpose of industrial policy is to promote internationally competitive business enterprises in markets that are continuously being reconstituted by strategically aware competitors.

The divergence of opinions on the role and emphasis of industrial policy (not withstanding the calls for less regulation (see Box 1) , is manifested in the establishment of a multiplicity of national programmes aiming to support industrial objectives. An OECD database monitoring the governmental support of industries contains 30 to 40 programmes for each member country. The Australian Ministry for Industry, Technology and Regional Development cites 43 programmes in their 1994 report and so on.

Despite the lack in theoretical underpinning and the ‘shotgun’ approaches used internationally, the knowledge of the justification for government intervention and its limitations are the cornerstones for the identification and development of optimum and consistent sets of interventions.

Traditionally, governments’ interference with the market has been justified as a remedy for the existence of public goods, of externalities, increasing returns to scale and informational asymmetries.

Public goods have the property that their consumption by one person does not diminish their availability to others. This implies that their price to consumers should be zero, since their consumption by one party is at no cost to others. In many cases, though, the production of public goods is costly, which creates a contradiction. Private firms are not supposed to charge for the public goods, in which case they will not produce them, or if they are able to charge for them, there will be too little of them consumed. A free market alone will not produce an optimal result so the state must intervene to do so.

The Carnegie Commission has proposed a list of societal goals which relate to public goods. The main topics are:

  1. quality of life, health, human development;
  2. a resilient, sustainable, and competitive economy;
  3. environmental quality and sustainable use of natural resources; and
  4. personal, national and international security.

An externality occurs when production or consumption by one firm or individual influences the well-being of another and the effect is not valued on any market. There is a tendency for overproduction of negative externalities, such as pollution, and a likelihood of under-production of positive externalities (such as knowledge).

Government intervention related to returns to scale refers to the investments that need to be made at the outset for the establishment of a new industry or industrial sector for which the costs may be prohibitive from the perspective of any one firm or involve high-risk long-term return potential. The extensive subsidies to Airbus consortiums by European governments and to RAM chips by the Japanese government are cases in point. Support for activities benefiting collectively small and medium enterprises may be justified on the basis of returns to scale.

Informational asymmetries may arise in a number of circumstances and government interference may be required for correction purposes. Borrowers usually know better than lenders how risky their projects are. Consequently, interest rates, determined by lenders, may be too high from a social point of view. Government subsidisation could correct the 'adverse selection' and investments will be undertaken on projects that would have been acceptable if interest rates had been at the optimal social level.

The advent of competitiveness in a global economy sets the background for a different and additional role for governments. Governments in this context have the responsibility to promote and protect the interests of their industries. Game theory provides a natural way to think about the interactions of nations. As governments from different nations have their own aims and strategies, and multilateral organisations (e.g. WTO) attempt to put some rationalisation in the international markets, governments of competing nations have to devise appropriate strategies and tactics for their own industries. The ‘outlawing’ of certain trade policy instruments (e.g. tariffs) with the success of trade liberalisation agreements among countries, provide further impetus for governments’ involvement in industrial policy.

Magaziner and Reich’s statement that "our standards of living can only rise if 1) capital and labour increasingly flow to industries with high value added per worker and 2) we maintain a position in those industries that are superior to that of our competitors" is probably the pinnacle of competitiveness related industrial policy. Of course, opinions to the contrary are not lacking (see Box 2).

This dual justification for the development of industrial policy – correcting market failures at home and promoting industrial competitiveness in the international market, may present governments with intractable dilemmas.

Returns to scale, for example, may point towards the establishment of an oligopolistic market behaviour with benefits for the oligopolists rather than a net local social gain. If rents, however, can be collected from foreign consumers and they exceed the losses to domestic consumers’ misallocation, an economic case could be made for subsidisation of domestic producers. A precondition for such policies is that other governments do not retaliate.

It is within this context that Bruce advocates international rationalisation through monitoring:

Industrial subsidy programmes in different countries have effects on international economic efficiency - that is, on the flow of goods and services across international borders in response to their highest commercial value. More fundamentally, industrial subsidy programmes can be as important as trade policies in the realm of international economic relations, and they may be becoming more so with the success of trade liberalisation agreements among countries that limit their abilities to impose tariffs. ... Industrial subsidies, like trade policy, can lead to a 'prisoner's dilemma' situation in which each country appears to serve its own best interest by implementing such programmes but, when all countries do so, every country is made worse off. Thus, industrial subsidies have already become the object of international negotiation toward mutually beneficial arrangements in which the subsidies are removed on a multi- or bi-lateral basis. It is this possibility that presents the most pressing need for the comprehensive and consistent measurement of industrial subsidies across countries.

The final factor that has to be elaborated in this context is the balance between the cost of intervention and the expected benefits. Government intervention does not occur without costs. Costs occur because the state has to collect and process all information necessary for the development of appropriate policies and because of the expenses of resources being directed into unproductive activities by private agents in order to capture rents generated by state intervention. While these costs have to be less than the expected benefits in order for the intervention to be considered successful, lack of information, monitoring and assessment make any programme ineffective and unjustifiable. As a rule of thumb, the OECD recommends that 2% of each programme should be spent for monitoring and evaluation.

Towards a Conceptual Framework for Industrial Policy

The development of an appropriate industrial policy requires the existence of a framework defining its domains and functions.

We suggest that industrial policy can be seen as the effort to maximise an ophelimity function of a particular group (a nation, region, organisation, etc.). The word 'ophelimity' comes from the Greek work 'ophelia' which can broadly be translated as 'usefulness'.

The suggested ophelimity function is a collective one consisting of the sum of the ordinary ophelimity functions of the individuals constituting the referred nation, region, organisation and so on. The ophelimity function of an individual is constituted by a set of attributes providing utility to the particular individual. It can be the pleasure for manufacturing an item or the utility accruing from a more productive manufacturing technique. To put it in a mathematical format, if O denotes the general interest, in whatever sense that may be taken, then O will be a function of a certain number of r variables which we may call 'target variables' [O=F(Y1,Y2,...yJ].

A certain numerical value for a variable will be called a 'target'. The targets (yi) will be chosen in such a way as to make O(y) a maximum. Activities aiming at attaining this maximum may be called optimum policy as other sub-optimal policies may be available.

It has to be realised that the fixation of the function O is a difficult matter. Difficulties for fixating O arise when it is considered to be a sub-function of the functions of economic policy, defence, social and cultural policies, etc.; from the difficulty of creating an ophelimity function which takes into account the 'interests' of all individuals in society, and so on.

In practice, the stage of trying to maximise O, will be passed over and the targets (yi) will be chosen directly. Thus, although the ultimate aim may be to increase the economic growth in the country, the fixation of 0 may be limited to deciding, say, an acceptable level for competition, the level of technological balance of payments, the rate of inflation, and so on.

Variation in the value of the ophelimity function can be achieved through two different kinds of activities. These can be characterised as qualitative and quantitative policies. By qualitative policy we mean the alteration of certain qualitative aspects of the structure of the system (the way the variables are intermingled). Examples of qualitative changes are:

A quantitative policy denotes the change, within the qualitative framework of the given structure, of variables such as the number of new products produced (innovation), the level of exports/imports of particular goods, the country's share in the international trade, and so on.

The target variables, although not directly amenable to manipulation by the policy makers, are affected by changes in variables under the command of the government, which may be called 'instruments'.

To use a mathematical format, the targets yi are functions of a number of instruments Ik, yi=f(I1,I2,

Examples of such instruments may be the level of taxation for the attraction of foreign companies, R&D grants for the production of new products, the provision of funds for training, and so on.

It should be emphasised that the definition of variables as targets and instruments is not standard or permanent. A variable may be characterised in one case as a target, while in another as an instrument. Further, there are policies which have both quantitative and qualitative characters.

It is not possible to define a priori which combination of qualitative and quantitative actions will produce the optimum policy. In principle, if only qualitative changes were considered, O could be estimated for each policy and the alternative showing the highest O value could be chosen. Knowing the structure required for an optimum policy, the instrument variables could be controlled in order to reach the values of the target variables which maximise O. In practice, however, the treatment of problems of qualitative policy meets with great difficulty. The main reason is that our original knowledge of human behaviour under different structural conditions is relatively restrictive.

Moreover, the debate concerning the importance of various qualitative and quantitative changes is still on. The lessaiz-faire school of thought argues that the maximum of general well-being will automatically be obtained by complete freedom. On the other extreme, the socialist argument is that only full state intervention can guarantee the extraction of maximum benefit from industry. The empirical evidence for qualitative policies is always debatable.

Quantitative policies by nature are more amenable to scientific inquiry and an increasing literature is accumulating on various subjects such as on the effects of tax subsidies for the R&D investments by the private sector.

The issues concerning quantitative policies are the fixation of the desirable level of the target variables and the availability of appropriate instrument variables to affect the target with minimum side-effects. In practice, however, not all variables affecting industrial performance are under the control of the particular authority concerned with industrial policy and defensive instruments may be employed in order to counterbalance the adverse effects of policies instituted for different objectives or interdepartmental bodies have to be instituted.

A number of advantages derive by approaching industrial policy according to the exposed way:

A Time of Change

We face change on many fronts and change characteristically engenders both opportunity and uncertainty. Rapid technological change and the ongoing information revolution both enable and demand new ways of doing and understanding of business. Globalisation and the emergence of competitive economies in Europe and Asia put new stresses on local employment and industries. Service activities like distribution, design and advertising break their distinction from traditional manufacturing and are becoming major providers of jobs and wealth.

Industrial policies are affected directly by these changes and have to be adjusted appropriately.

Technology is changing fundamentally our understanding of economic growth and trade. There has been an explosion of research papers in the innovation literature, but - like all new knowledge - it is taking time to absorb and to work out all the implications, some of which appear to be far-reaching. Based on findings from work in this area, a number of researchers are rewriting some of the fundamental assumptions of classical and neoclassical economics, which were originally based on trade in simple technologically unchanging commodities. It is increasingly clear that a quite new economic paradigm is beginning to emerge from this developing body of work on innovation.

Dosi, Pavitt and Soete in The Economics of Technical Change and International Trade develop a new theoretical approach grounded on study of a wide range of empirical data relating to innovation. These authors develop a new model of world trade patterns:

. . . international technology gaps, reflecting superior and inferior techniques and... cost based adjustment mechanisms [are determined by international differences in innovative capabilities], ... the sources and uses of innovations, [and] corporate strategies and institutional conditions. [These gaps are] of fundamental importance in explaining the participation of each country in international trade flows, and international differences in income levels.

They challenge neoclassical economic theory, arguing that:

. .. a good deal of analysis on technology, growth and international trade has a 'reductionist' flavour, attempting more often than not to squeeze genuine dynamic problems of innovation, learning, uncertainty and change into the more familiar [neoclassical] cloth of endowments, relative scarcities and optimisation under budget constraints. ... Many of the pioneering investigations on the complex dynamics linking technical change, growth and transformation... have often been neglected in [this] broad reductionist vein, ... particularly in Anglo- Saxon countries. Our central interest in innovative phenomena brings us ... to present... a framework which focuses on change, through time and differences between countries, instead of on timeless equilibria and uniformity between countries.

Furthermore, they take significant steps towards the development of a new macro-economic theory which emphasises the importance of a dynamic approach. They argue:

... technology plays a major role in two fundamental topics pertaining to economic analysis: i) the problem of coordination and interdependence between agents and, by implication, between countries; and (ii) the patterns of change and transformation of each economy ... [We present] ... a theoretical analysis of innovation, trade and growth, which differs in some important respects from most traditional analyses ... First, ... [we consider] ... that the widely accepted representation of 'technical progress' as a shift in the production function resulting from disembodied or embodied technical change inadequately represents the more complex and interesting reality that emerges from a variety of industry and firm-based studies. ... Second, we reject the assumption that the generation of technology is independent of investment and production. In most sectors, it is strongly dependent on them. ... Third, the assumption that firms' technical choices are exogenously determined and optimal is rejected in favour of the proposition that such choices are generally discretionary and non-optimal, given the impossibility of foreseeing the nature and likelihood of all possible future technological and market developments.

Technological dominance also appears to create a self-reinforced advantage in favour of technological advanced nations vis-a-vis laggards. Arthur, of Stanford University, argues that in a growing number of industries there is a natural tendency for the market leader to get further ahead causing a monopolistic and geographical concentration of business.

These businesses have three things in common. First, they have high fixed costs, such as R&D, but low variable costs. For example, the cost of writing a computer programme is the same regardless of the number of copies sold, so the higher the sales, the greater the profit margin. The same is true for many other industries that are heavy in know-how and light in material resources from pharmaceuticals to defence.

The second common characteristic is what is called 'network externalities'. In the software industry, for example, this means that the more widely an operating system is used, the more likely it is to become a standard for the industry and the more people will want to use it to ensure that their software is compatible with that of other network users. This makes it harder for rivals to compete.

The third factor which strengthens a leader's grip on a market is the customer lock-in effect. Many high-tech products are difficult to use, so once a customer has learnt how to use a computer programme, say, he loathes to switch. If all three factors are present, Arthur argues, increasing returns will magnify the market leader's advantage. By cutting its price, a leader can grab a bigger share of the market, earn bigger profits and spend more on R&D than its rivals, sharpening its edge even further.

The technological revolution calls for an economic revolution too, and the new thinking finds its way on industrial and trade policy. In a number of occasions prominent American academics and politicians link technology to industrial policy.

Branscomb, in an effort to resolve the utilisation of technology for the pursuit of social goals (e.g. pollution), without compromising the international competitiveness of the US industry, suggests that:

To the extent that the externalities are the primary motive for the programme, the right policy would be to encourage world-wide industry to use the new technologies; to the extent that the goal is the competitiveness of US manufacturers, tight control should be retained on the intellectual property.

Similarly, Laura Tyson states:

... The question of [automobile] emissions is not only an environmental problem but a competitive problem ... Companies that come up with clean cars will gain market share in the global industry. There is a real incentive for industry and government to work together to advance this as both a public and a private good.

The US government, under the banner of the Economic Espionage Act of 1996, tightened legislation related to the protection of trade secrets and the OECD is cautioning its community against techno-nationalism.

With respect to technology support policies, governments should try to avoid falling into the trap of techno-nationalism - of promoting only domestically-owned firms and excluding foreign-owned companies, even though the latter may make substantial intangible investments domestically and have useful worldwide research and networking links. National policies to support what are considered by governments to be 'strategic' technologies, for example, through subsidies and discriminatory public procurement, are becoming a source of increasing international friction … We recommend ... that the issue of equal access for domestic and foreign firms to publicly funded research on an equivalent basis be addressed, including the desirability of granting access to government- sponsored consortia and programmes.

The facts and effects of globalisation are equally impressive. During the 1980s, trade has been growing twice as fast as output and foreign direct investment three times as fast. More economies than ever before have opened up their markets to trade and capital and have embraced market- friendly economic reforms.

Manufactured goods now account for almost 60% of the developing and emerging economies, up from 5% in 1995. The Third World's share of world exports of manufactures jumped from 5% in 1970 to 22% in 1993 and for the first time in history developing countries have become an independent source of growth in the world economy.

Economists and politicians argue about the effects of these changes. Has the mobility of technology and capital, which allows firms in rich countries to locate wherever in the world is cheapest, undermined the theory of comparative advantage which underpins the case for free trade? What is the effect of global competition on regional jobs in general, and on low wage/low skilled jobs specifically?

Globalisation, coupled with the technological revolution, presents special challenges for policy. The mobility of capital and technology across borders may give low wage developing countries access to the best production techniques. At the same time, modern telecommunications allow firms in manufacturing and even in services, to shift production wherever labour is cheapest without losing touch with head office. The only constraint is education and infrastructure in low-wage countries which may lag many years behind those of high-wage countries. Different infrastructures, however, are decisive for productivity. Furthermore, productivity will vary from sector to sector across countries, for example, because of differences in R&D intensity. In some sectors, such as clothing or consumer electronics, productivity in a developing country may be almost as high as in the developed world. Consequently, its low-wage structure (determined by average productivity across the whole of manufacturing) will give firms in those industries a cost advantage relative to those in rich countries. However, productivity differences across sectors will be decisive in determining the country's comparative advantage.

Effective policies in the short term will have to differentiate and handle differently sectors lagging, vis-a-vis those with competitive advantage internationally, while in the long term human and physical infrastructures will determine competitiveness.

An additional factor to be taken into consideration is the international market that is dominated to an increasing extent by products with high knowledge intensity. Between 1986 and 1992 the value of high-tech exports of major trading nations increased by 46%, compared with 38% for other exports. These increased their share of the total exports from 41 % to 43%. High- and medium- tech products in the 1990s cover more than 80% of the international trade, and OECD estimates that more than half of total GDP in the rich economies is now knowledge-based.

Knowledge-intensive products, however, are difficult to tax (Internet transactions may slip away from the tax-man), regulate and measure.

Manufactured goods are particularly sensitive to the changing world. The boundaries between manufacturing and services have always been tenuous. Software written in-house by a car manufacturer counts as manufacturing, yet if software is made by a software house it logs as service.

Wyckoff of OECD argues that information technology is encouraging the convergence of manufacturing and services.

Manufacturing is becoming more like services: customer service is getting more important, and products are increasingly being tailored to the needs of individual customers. One example is Levy Strauss's service offering personally tailored jeans for women. Measurements are transmitted over the Internet to the factory, which allows the finished jeans to be delivered to the customer within days.

At the same time, services are becoming more like manufacturing. In the old days, production and consumption of services had to coincide in time and space. Most services were like a medical check-up: a patient has to go to the doctor's surgery and wait his turn. But now the codification of knowledge in some services makes direct contact between producer and consumer unnecessary. Such services can therefore be held as inventories and traded internationally. Examples include expert computer systems that can perform routine legal tasks such as drafting a will, and accountancy packages that plan the client's finances and file his tax return.

Apart from this convergence, a substantial part of the value of manufactured products is contributed to services. Over 70% of the value of a typical manufactured product is already contributed by service activities such as design, sales and advertising. Obviously, policies handling manufacturing separately from services would address at best part of the problems.

We have touched only briefly the effects of technology, globalisation and service - manufacturing diffusion on the development of trade and industrial policies. It becomes profound that traditional reductionist policies do not conform with contemporary understanding and needs. Our technological revolution calls for an economic revolution too.

Reaching our Goals

The Department of Trade and Industry has to provide direction to an uncharted as yet future. At the same time, governments' role in the field is sharply questioned internationally. Societies everywhere look to a new order of leadership.

These earthshaking changes require new approaches and principles - addressing conflicts in the values people hold, helping them face tough realities and mobilising them to learn new ways.

The challenges we face demand that we set and keep focused on our two goals:

To reach our objectives, we propose a coherent, integrated set of approaches involving government, industry, the research and educational community, and all our citizens. We are all stakeholders in the national vision and we must focus on a shared commitment.

These are:

1. Support for a Technologically-based Future

The Department should recognise that scientific and technological advances are the wellspring of innovations whose benefits are seen in all aspects of economic and social life.

The Department should, in collaboration with the Department of Arts, Culture, Science and Technology, undertake a sustained, continuous effort to strengthen the nation's economic competitiveness by fostering partnerships between and leveraging the strengths and unique capabilities of government, industry, labour and education institutions.

The Department should aim to accelerate technology development and commercialisation, through the promotion and the diffusion of productivity enhancing technologies, technical know-how, and modern business and human resource practices across the industrial sector.

Generic technologies/sectors of importance for future development and prosperity, such as biotechnology and information technology, should receive particular attention.

2. Programme Specificity

Policy instruments are by their nature weak and the targets aimed at are usually substantive. Policy instruments (i.e. programmes) should aim at and be measured by the success of a limited number of targets and preferably only by one target.

Short- and long-term priorities should be distinguished, and different instruments should be utilised for each group. For example, direct funding for innovation is a short-term intervention, while tax incentives address the long-term innovation capabilities and competitiveness of the country.

3. Leadership Provision

In a global knowledge-intensive economy "catch-up" is becoming increasingly difficult. Competitiveness will require the early identification of new scientific and technological frontiers and the creation of appropriate national infrastructure. Biotechnology and nanotechnology appear to be the new scientific/technological frontiers promising the new revolution. The relevant authorities should make explicit, announce, and fund appropriately the new competitiveness frontiers.

The future cannot be created without leadership.

4. International Monitoring

South Africa does not have a coherent approach to monitoring and reporting on industrial, trade and technology policy of other nations. Consequently, policy-makers and South African firms are 'informationally' disadvantaged while many other governments and foreign companies have been successful in commercialising ideas and technologies that originated outside domestic borders.

To assist in the development of appropriate policies and increase industry's awareness of promising and effective practices developed in other nations, the Department of Trade and Industry should, in collaboration with other related departments, develop appropriate mechanisms facilitating monitoring and access to best practice abroad (management, technical and scientific).

5. Programme Efficiency and Effectiveness

The Department should pay particular attention in promoting efficiency and effectiveness in all its activities and it should impose a real and sustained pressure on and within each programme for continuous improvement in the value of money obtained in the delivery of policy and services.

Programmes should be required to establish quantifiable objectives and sunset clauses well in advance, and monitoring and assessment should become integral parts of all efforts.