AN EVALUATION OF INDUSTRIAL POLICY PERSPECTIVES IN THE SOUTH AFRICAN CONTEXT

BACKGROUND

The Trade and Industry Chamber at NEDLAC has, as an item on its agenda, "Industrial Strategy in South Africa". The chamber has also established an industrial strategy task team, which is comprised of representatives of government, business and labour. All parties have submitted their particular input on industrial strategy to the task team, which has the responsibility to formulate a report for the chamber.

In order to engender a meaningful debate between the parties, NEDLAC has commissioned the Sociology of Work Unit at the University of the Witwatersrand to, in association with NEDLAC's research department, conduct an independent evaluation of different approaches to industrial policy.

We do not attempt to provide an exhaustive analysis of the shortcomings of South Africa's industrial structure, but refer to selected data to illustrate our arguments.

INDUSTRIAL STRUCTURE, STRATEGY, AND POLICY

Countries are becoming more economically interconnected as trade barriers between them are dismantled. For example, India reduced its average import tariffs from 82% in 1990 to 30% in 1997, and China from 43% in 1992 to 18% in 1997. South Africa's average import tariffs on manufactured goods were reduced from 14% in 1994 to 5,6% in 1998. In the context of trade liberalisation, the way in which goods and services are produced has changed significantly - commodity chains now span across national boundaries, and many products are assembled in several countries. Exported goods now average a value of $7 trillion. The result of this, as well as changes in the operation of financial markets globally, is a process of economic concentration and the growth of multinational corporations through mergers and acquisitions. In 1997 alone, $236 billion was spent on mergers and acquisitions. Many multi-national corporations now have annual sales totalling more than the gross domestic product of many countries.

In this context, the role of the state, particularly in developing countries, has come under renewed scrutiny. With the traditional model of Import Substitution Industrialisation (ISI) no longer available in its classical form, new approaches to industrial policy have emerged. These approaches focus specifically on how developing countries are affected by the global process of industrial restructuring - or how countries are 'inserted' into the global production process along commodity chains.Processes of policy analysis and research can no longer focus on the national level - the local and the national must be grounded in global processes of restructuring. Various states are 'inserted' into global commodity chains in different ways. Hence, one can no longer talk about 'best practise' in policy making in an unproblematic way. Commodity chains are not static, and are influenced by processes of national policy. The aim of industrial policy should therefore be to 'insert' the national economy into global commodity chains in a way that reduces vulnerability and risk - i.e. building on existing strengths and upgrading so that production can be grounded in less vulnerable segments of commodity chains.

Hence, industrial policy is not a matter of whether the state has a role to play, but what the appropriate role is. Historically, however, debates on the ability of industrial policy to shape development within a capitalist framework revolved around different perspectives on the nature of relationships between states and markets. On the one end of the continuum, institutions such as the World Bank, as well as some business-linked think tanks, traditionally support a neoliberal approach. In this model, industrial development should be market-driven, with a minimalist role for the state to play in influencing industrial structure. On the other end of the continuum, the model of the developmental state sees a central role for the state in shaping industrial structure through industrial policy measures such as the targeting of certain industries for investment or for other supportive measures (Evans, 1995; Chang, 1996).

For analytical purposes, it is important to distinguish between the concepts industrial structure, industrial strategy, and industrial policy. The industrial structure of a country relates to several variables:

• First, it relates to dominant sectors in the economy. Broad sectors include mining and agriculture, manufacturing, as well as financial and other services;
• Second it relates to the relationship between the state and markets in the economy. States are involved in the economy through, inter alia, its ownership of productive resources, its macroeconomic policies, public procurement, and the way it regulates the productive and reproductive functions of society in general;
• Third, industrial structure also relates to the linkages between sectors of the economy, and how these linkages are socially regulated.For example, many African economies are rich in minerals, but, because of colonialism and neocolonialism, have been unable to create strong manufacturing industries to add value to the commodities extracted in their countries for domestic and export markets. Instead, commodities are exported, and consumer goods have to be imported from other countries. Hence, this industrial structure leads to sustained dependency on other economies.

Industrial strategy implies a broad vision on how the state can alter the industrial structure of a country in order to facilitate industrial development. The aim of industrial strategy is to bring about, or to shape, a process of industrial restructuring with the view of contributing to national socio-economic goals, such as the creation of jobs and the alleviation of poverty. Different approaches to industrial strategy imply a range of different perspectives on what level of state "intervention" in the economy is required and feasible.

Whereas industrial strategy is more about a broad vision for industrial restructuring, industrial policy refers to "the role of governments in promoting industrial development beyond what would be possible without intervention" (Lall 1996:1). But what exactly this role entails, and what is considered as intervening, are subject to debate.

Chang (1994; 1998:53-54) distinguishes between "broad" and "narrow" definitions of industrial policy. Broad definitions include "every government policy that affects industrial performance". This implies that macro-economic, education and infra-structural policies, for instance, are all included under industrial policy. A narrow definition equates industrial policy with the targeting of certain industries. But there is also a definition that incorporates both these approaches. The "core" of industrial policy is seen as targeting, whereas general policies, such as human resource development policies, support for research and development (R&D), as well as macro-economic policies are included as general or functional industrial policy.

Chang (1994:61) prefers a more narrow definition, since a broad definition "overloads" the concept analytically. For this reason, he defines industrial policy as "a policy intended to affect particular industries to achieve outcomes that are perceived by the state to be efficient for the economy as a whole."

Hence, industrial structure refers to the actual structure of different sectors of the economy in relation to each other, as well as the way in which the interaction between the state and markets is structured. Industrial strategy is a broad vision of how the state can bring about, or facilitate, changes in the industrial structure, i.e. industrial restructuring. Industrial policy implies the actual policy instruments that can be used to implement an industrial strategy. There is some disagreement as to what constitutes these instruments, and how broadly the concept should be defined. The following section attempts to shed light on this by discussing the social context in which industrial policy became salient.

The rise of the developmental state

Debates and a conscious focus on industrial policy became prominent globally since the late 1970s. In the mid-1980s, the debate subsided to some extent, but in the 1990s, renewed attention was paid to the issue (Chang 1998:53). These debates took place in a certain social context. Processes at play in this context are certainly multi faceted and complex, but the discussion here is limited to three issues.

First, since the 1930s, governments tended to play a central role in their economies through demand-side macro-economic management. These demand-side measures included various ways of boosting domestic demand, such as minimum wages, coupled with the protection of local industries by import tariffs. The intention was to facilitate the development of local industries. This approach to industrialisation is known is Import Substitution Industrialisation (ISI).

However, this "golden age of capitalism" came under pressure because of rising inflation and the oil crisis in the early 1970s. Governments started to abandon these demand-side policies for neoliberal policies, implying a withdrawal from the economy through privatisation, trade liberalisation, and through the flexibilization of labour markets.

However, this economic restructuring process did not take place without pain. Chang (1998: 53) argues that market mechanisms seemed unable to channel resources released by declining industries into new industries. Even governments publicly opposed to state involvement in the economy, such as the Reagan administration and the Thatcher government, used industrial policy as a tool to shape this process of industrial restructuring.

Second, trade liberalisation exposed many developed countries to competition from certain Asian economies, notably those of Japan and South Korea. This drew attention to the role played by the governments of those countries in facilitating industrial development (Chang 1994; Lall 1996; Itoh, Kiyono, Okuno-Fujiwara, & Suzumura 1991).

Many of these governments used a combination of supply-side and demand-side measures to stimulate and to strengthen their manufacturing industries. Certain industries were targeted for export promotion. They would initially be protected to develop capacity to serve domestic markets, while receiving considerable support from governments through various supply-side measures, including state-funded research and development, financing, and tax incentives. Hence, the model of the developmental state challenged the idea that states are opposed to the functioning of markets. Instead, many argued, states should step in where markets fail.

Third, more recent debates on the impact of globalization on societies drew renewed attention to the need for governments to cushion some of the negative social effects of globalization. A re-evaluation of the relationship between state and markets took place, giving rise to new perspectives, some of which drew on the notion of industrial policy to shape the outcome of neoliberal globalization (see Evans, 1997).

In this regard, Castells (1996) argues that the integration of dominant sectors of different societies into the global economy has led to a repoliticization of the economic system: "By integrating countries into in a global economy, the specific political interests of the state in each nation become directly linked with the fate of economic competition for firms that are either national or located in the country's territory." Countries of which governments used the position of their firms in the global economy to further national interest succeeded in attaining developmental goals. These governments linked competitiveness, productivity and technology in explicit industrial policies, and established a dynamic interaction between co-ordinated supply-side and demand-side measures.

However, when using the concept globalization to frame a discussion for policy, one should be careful not to present the process of global integration as being neutral. For Castells, for example, the aim of an industrial policy would be to link a society to global informational capitalism - i.e., a strategy designed to be included. The alternative to this would be to be part of what Castells calls the "fourth world", or to be connected in ways that do not lead to sustainable development. Yet, totalising theories of globalisation often ignore the local (or, indeed, the regional) in the global. Notions of "best practice" are applied to diverse conditions, often at the expense of grounding policy practice in local realities.

Globalisation is an uneven process, and takes place in the context of highly unequal power relations - with regard to the position of firms, as well as governments. Often, selective protectionism in the North prevents developing countries from accessing major consumer markets, and government bureaucracies of developed nations develop sophisticated ways of playing the "trade game" in their favour. Also, different states have different labour regimes - in some basic labour rights are denied. From there the controversial debate on the social clause has arisen. There is always the danger that policy options are presented as a dichotomy instead of a spectrum of possibilities. Analysis is grounded in broad assumptions that do not ground analysis in local realities. Some economies are subordinate in the global economy, and hence, one cannot assume that all "best practice" industrial strategies apply to all circumstances - strategies for subordinate countries cannot be the same as for the developed world. Many developing countries are still locked into the low value-added segments of global commodity chains.

In Section B, we attempt to show how thinking around industrial strategy in South Africa has been shaped by various perspectives "imported" from elsewhere. We do not argue that industrial strategy formulation should take place in isolation. On the contrary, valuable lessons can be learned from the experiences of other countries and regions. However, a policy analysis should always be grounded firmly in local realities. We discuss the World Bank, post-Fordist, Porterist and political economy approaches. All these approaches have been influential to some extent, but have substantial weaknesses. We then discuss a more recent approach, the Global Commodity Chain (GCC) approach, and argue that, when grounded in an appropriate analysis of South (and southern) African realities, this approach can address to some extent the weaknesses of previous perspectives.

We also use country case studies of Malaysia, Mauritius and Ireland to illustrate how different countries have attempted to shape the way in which their economies are 'inserted' into the global economy to various degrees of succes.