Competitiveness has been a keyword of the economic discourse in the last decade, equally applied to the situation of firms and of countries. It can be defined as the ability of a country to sell its goods in world markets while maintaining or increasing the real incomes of its people. It is a relative concept, comparing a country's productivity and export performances against those of others.

In an integrating world, with more globalized economic activities and more open markets, the pressure of international competition has become a major constraint. As traditional export markets have been lost and domestic markets have been penetrated by foreign goods, more effort has been put in reducing the costs of domestic production, in upgrading products and processes, in searching for new markets, in developing new activities and the framework conditions and infrastructures needed for supporting them.

As world trade and demand have been growing in the past decades more rapidly than the economy of most countries, the ability to be competitive in the world market has often been considered as a key for assuring growth and employment in the national economy. While some industrializing countries have entirely relied on their competitive strength in some industries as the engine of national growth and job creation, for the more advanced countries competitiveness has become more of a constraint, and a spur to upgrade their activities.

However, much confusion on the meaning and definition of competitiveness has remained. A number of different aspects of competitiveness, and policy views related to it, can be pointed out, with different impacts on domestic employment. They are considered in turn below.

1. Introduction

Competitiveness has been a keyword of the economic discourse in the last decade, equally applied to the situation of firms and of countries. It can be defined as the ability of a country to sell its goods in world markets while maintaining or increasing the real incomes of its people. It is a relative concept, comparing a country's productivity and export performances against those of others.

In an integrating world, with more globalized economic activities and more open markets, the pressure of international competition has become a major constraint. As traditional export markets have been lost and domestic markets have been penetrated by foreign goods, more effort has been put in reducing the costs of domestic production, in upgrading products and processes, in searching for new markets, in developing new activities and the framework conditions and infrastructures needed for supporting them.

As world trade and demand have been growing in the past decades more rapidly than the economy of most countries, the ability to be competitive in the world market has often been considered as a key for assuring growth and employment in the national economy. While some industrializing countries have entirely relied on their competitive strength in some industries as the engine of national growth and job creation, for the more advanced countries competitiveness has become more of a constraint, and a spur to upgrade their activities.

However, much confusion on the meaning and definition of competitiveness has remained. A number of different aspects of competitiveness, and policy views related to it, can be pointed out, with different impacts on domestic employment. They are considered in turn below.

Most concepts of competitiveness in the current public debate are narrowly confined to cost and price competitiveness; this approach however has severe limitations. Firstly, it is looking only at short term developments. Secondly, it focuses on firms performances and cost structure, while a country's economy is affected by broader economic mechanisms. For instance, cutting domestic wages and employment may well reduce costs for firms and increase their exports, but at the same time it may reduce domestic demand more than the increase offered by the foreign component, leading to a recession. Cutting wages is therefore not an appropriate way of increasing competitiveness, as it leads to reduced real incomes, lower demand and employment. Rather than focusing on costs and wages, the emphasis should be put on a country's productivity performances.

The technological level and economic structure of a country tells much about its competitive prospects. International competition varies across sectors and markets, with differences in the growth rates, in the number of rivals, in the elasticity of demand for a country's export, in the market power of national firms, in the degree of segmentation of markets. Countries whose exports concentrate in fields wher productivity levels, technological intensity and oligopolistic and market power are greater are much less exposed to competitive pressures. The sectoral specialization of countries - in export, but also in production and technological activities - is therefore an important factor in their competitive performances, and will be discussed at length below.

A currency devaluation automatically improves a nation's competitiveness (in such a narrow definition), although its overall consequences on the trade balance and the growth and employment prospects depend on several other factors. In particular, key issues include the price elasticities of exports and imports; the ability of exporting firms to expand volumes following the advantage offered by the devaluation, rather than increasing their profit margins on the same sales; the distributional impact of the increased import prices.

Competitive devaluations have been a frequently used policy adopted by countries in their initial stage of industrialization or confronted with balance of payments constraints, or more severe financial crises. They may be succesful in the short term in contributing to address existing unbalances, but in the longer term they tend to reinforce existing economic structures and patterns of specialization, with a possible worsening of a country's structural and technological competitiveness.

Competitiveness in the short term

As already pointed out above, costs reductions and currency devaluations are the main actions which can be taken in a strategy to increase competitiveness in the short term. The limitations of a short run perspective on competitiveness are evident; still, most of the policy debate is confined to such a view. Structural factors and sectoral specializations change much more slowly, and they represent a major constraint on the possibility to improve quickly a country's competitive position.

Competitiveness in the long term

In the long term, competitiveness depends on a country's ability to develop the factors that sustain productivity growth; in a dynamic view of competitiveness the key factors include innovation, R&D and the development of human resources needed to build in the long term competitive advantages in established and new industries (Porter, 1995).

In a long term perspective, the investment and innovation decisions of firms are crucial, combined with the ability of public policy to provide a right set of incentives, framework conditions and infrastructures. Institutional and organisational change is also important in improving the efficiency of the economic system.

The search for long-term competitiveness therefore involves important changes in a country's economy. First, structural change leads to exit from sectors with low export capacity and to the development of new competitive sectors. Second, technological change makes it possible the emergenge of new products and markets, as well as increased efficiency in the processes for producing existing ones. Third, demand changes are important as they need to accompany the evolution of the supply structure and support the emergence of new products, markets and industries.

Competitiveness as a zero-sum game

When production and trade in particular industries are stagnant, the effort at increasing competitiveness by one country may simply displace foreign producers, with no net gain - except in the form of some welfare gains for consumers of countries with access to lower cost goods, or a theoretical improvement in the efficiency of resource allocation (making the irrealistic assumption of perfect mobility of factors of production across sectors, regions, etc.). As the country losing out may respond with similar efforts at lowering costs, a downward spiral may result, leaving both rivals worse off; as labour cost and conditions are the easier target for action in the short term, the risk is a race to the bottom which offsets the economic gains promised by increased efficiency and competitiveness. The employment consequences of such strategies are particularly negative.

How serious is this risk? Official views are quick to dismiss such a possibility. The Report of the European Competitiveness Advisory Group stresses that "competitiveness is not a zero sum game". An increase in competitiveness in one country is not seen as coming at the expense of another, as the gains in productivity and efficiency in different countries can be mutually reinforcing (Competitiveness Advisory Group, 1996).

In reality the risk of a zero-sum game depends on the overall dynamism of particular industries. Competitiveness is generally referred to the tradable part of the economy, in particular manufacturing industry, which is a shrinking part of the economy in terms of value added and employment; moreover, within manufacturing there are a number of slow-growing sectors where efforts to squeeze large productivity gains and increase competitiveness may result in a zero-sum game. In fact, many manufacturing industries which in several European countries have registered net employment losses appear to be clear cases of competitive pressures resulting in a zero-sum game, with employment losses in European countries and job gains in other industrial areas.

Dynamic competitiveness as a positive-sum game

The gains in economic efficiency, growth and employment promised by increased competitiveness may be reaped when the focus is not on a static process of redistribution of market shares, but when is on a dynamic process of creation of new resources, technological innovation, development of new markets, new skills and competences. These are the processes which make competiveness a positive force in economic change, capable to sustain employment growth.

In aggregate terms, within a national economy, competitiveness is a positive-sum game when GDP growth is higher than productivity growth, allowing space for increased production capacity and employment. If this is not the case, there may simply be a redistribution of market shares among firms and labour-saving investment and restructuring.

The characteristics of long term competitiveness pointed out above also contribute to make it a dynamic process and a positive sum game.

Competitiveness as a policy objective

Countless policy reports have been prepared in several countries recommending ways for improving national competitiveness, starting with the US in the 1980s when the trade deficit had become ever larger (Council on competitiveness, 1988).

Later the European Union attached a key role to competitiveness in the strategy of European integration in the White Paper 'Growth, competitiveness and employment' (European Commission, 1993), and later established a Competitiveness Advisory Group (1996) to identify appropriate policies.

However, some critics have considered the debate on competitiveness as a "dangerous obsession", suggesting that the key issue to be addressed is in fact productivity (Krugman, 1996).

The policy issues for competitiveness and employment will be addressed again in a specific section below.

2. Competitiveness and economic theory

While standard economic theory has always considered export success to be a function of prices, the relevance of "non-price" factors of competitiveness was first pointed out by Kaldor (1978). The "Kaldor paradox" showed that in several countries export market shares increased together with relative unit costs or prices. In other words, in the long term, higher wages, costs and prices did not weaken the competitive position of industrial countries, but, on the contrary, sustained their growing export shares. Fagerberg (1996) showed that the same relationships identified by Kaldor can still be found today for a group of Western and Asian countries, and that growth of market shares is also positively related to the growth of productivity (measured by GDP per capita) and of R&D expenditures (as a share of GDP, an indicator of technological intensity).

The relevance of innovative activities and productivity growth for assuring a country's competitiveness in the long term has therefore been clearly established, while the emphasis on costs, wages and prices has been shown to be misleading.

The contribution of increasing exports to domestic growth has been emphasised by Keynesian views of exports as an exogenous component of demand, and by studies of growth in open economies, such as the model of balance of payments constrained growth developed by Thirlwall (1979), where the key determinants were identified in the expansion of international trade, changes in relative prices (price competitiveness) and the ratio of the income elasticity of exports to that of imports. The empirical tests of this model confirmed the key role of non-price factors in the competitiveness of Western countries.

A step further has been taken by Fagerberg (1988) who included indicators of technological activities as determinants of the domestic and foreign market shares of a country. The growth of innovation efforts and the potential for imitation were found to be crucial for the competitive performance of fifteen OECD countries in the period from the early 1960s to the early 1980s, while cost factors played a marginal role.

As the identification and quantification of the non-price factors of national competitiveness offered a reasonable explanation of the Kaldor paradox, more attention developed on the influence technology had on international trade, following the debate on the "Leontief paradox" (Leontief, 1953), with the emergence of neo-technological trade theories postulating differences in technological capabilities between advanced and backward countries (Posner 1961; Vernon 1966; Krugman 1979; Dosi, Pavitt and Soete 1990). The rate of innovation of rich countries was seen in these models as the key for sustaining high wages and incomes, constantly threathened by the low-wage cost competition of poorer countries relying on an imitation-led catching up process.

In recent years, neoclassical economists have also addressed these problem, developing the "new growth theory" which tries to explain growth patterns as a result of the endogenous creation of technological and learning activities, motivated by private incentives and with some degree of spillovers, while maintaining a rigid framework of general equilibrium analysis. The importance of learning by doing is emphasized by Romer (1986) and Lucas (1988), while R&D efforts are examined by Romer (1990) and Grossman and Helpman (1991).

A wide range of economic views therefore recognizes the importance of technological factors in countries' competitiveness, moving beyond a narrow focus on cost-based competition. Innovation is increasingly seen as a base for building national competitive advantages, which are the source of specialization patterns and export strangths, examined in the next section.

3. Competitiveness and trade specialisation

The traditional argument is that competitiveness is sustained by countries trading among one another following their established comparative advantages. Specialisation in the fields of greater national relative strength allows countries to achieve economies of scale and greater learning by doing, so that productivity is higher (and grows faster) than in industries with a lower (relative) concentration of activities.

This approach is able to explain patterns of inter-industry specialisation of a more traditional type, while more recent patterns have focused on the growing role of intra-industry specialisation. This is driven by the increasing variety of activities within each industry, including product differentiation and market segmentation, as well as by the growth of intra-industry trade associated to greater international integration and international production by large multinational firms.

A large body of research has addressed the question whether specialisation is good for growth. Positive links are found by Amable (1996), analysing a group of 39 countries, between trade specialisation (considering imports as well as exports) and productivity growth over the 1965-1990 period.

The opposite results are found by Busson and Villa (1994), from a larger sample of 98 countries. They have found a negative impact of inter-industry specialisation on growth and their policy conclusions suggest that either countries should specialise in products with fast growing world demand, or they should increase their intra-industry specialisation, which can lead to an increase of overall productivity.

Focusing on OECD countries, the structure and evolution from 1965 to 1992 of inter-industry export specialisation patterns is investigated by Dalum, Laursen and Villumsen (1996), addressing both the distribution of countries’ exports across industries, and the concentration of OECD exports of individual industries across countries. The findings point out the stability and slow change of specialisation patterns; the distribution of exports across sectors and countries tends to slowly become more similar. This can be considered as a result of the relative economic convergence among OECD countries, but is somewhat in contrast to the patterns of specialisation in technological activities, as measured by patents, which show signs of increasing specialisation across sectors (Pianta and Meliciani, 1996).

The patterns of export specialisation therefore appear to be the result of the specificity of national systems - and of the persistence of their characteristics - combined with the pressure towards a more similar trade pattern across industries in OECD countries.

The patterns of specialisation, however, are conditional to the sectoral breakdown used. The available industry classifications are increasingly unable to differentiate the nature of international flows of goods and services; in recent decades, as OECD economies have become more similar, trade within sectors has increased more than traditional inter-industry trade, raising new questions on the nature and patterns of specialisation.

The relevance of intra-industry trade specialisation has been addressed by Dalum, Laursen and Verspagen (1996) with data for 75 product groups assigned to 11 industries. The impact of specialisation on growth rates of sectors has been investigated, showing that within-sector specialisation in trade has some relevance for growth rates of production, together with variables such as the growth rate of employment, the investment-output ratio, the number of patents per employee, the gap in productivity (indicating the potential for catching up). However, the ability to explain growth patterns decreases over time and is very limited in the 1980s. The interpretation of the authors is that in the 1960s and 1970s catching up was mainly due to non-activity specific learning, while in the 1980s this has become specific to the activities carried out. In other words, while initially there was scope for growth through specialisation (based on comparative advantages) regardless of the field, in the last decade it does indeed matter what are the industries in which a country is specialised.

Furthermore, the evolution of high technology industries suggests that while there may be a continuing advantage for the established leaders (who have an absolute as well as comparative advantage), in some cases the "windows of opportunity" for catching up by (initially) de-specialised countries may be more important than apparent relative advantages by smaller and less active countries.

4. The impact on growth and employment

There are limits, however, on the possibility to achieve growth by relying on the mechanisms of comparative advatage. In a study of four EU countries, the US and Japan in the 1980-92 period, Pianta (1997) examined whether growth in technological activities, production and exports across manufacturing sectors has been higher in the industries where countries had a stronger relative specialisation. Quite surprisingly, the growth rates for patents in the US, value added and exports were found to be frequently negatively related to existing specialisation patterns. The countries considered, especially the European ones, have grown more in the sectors of relative de-specialisation. A sort of "decreasing returns" to specialisation appear to emerge which can hardly be reconciled to the promise of increasing benefits from comparative advantage. While some "negative returns to scale" may operate in the fields where countries have already developed extensive economic and technological activities, the presence of relative strengths does not result into stronger performances neither in technology, production or trade.

The evidence points out the limits of comparative advantage as a principle for understanding current economic change and as a policy prescription. While globalisation is expected to offer a greater role to countries’ comparative advantages, the evolution of technology, production and exports appears to be mainly driven by the pattern of structural change, rather than by comparative advantage. It should be pointed out, however, that in the 1980s the four large European countries have experienced relatively slow growth and a persisting specialisation in industries with lower growth rates; this can be expected to emphasise the "divorce" between specialisation structures and performance indicators. An analysis of more dynamic Asian countries, specialising in the "right" industries, where growth of patenting, production and export is fastest, may lead to more positive results on the importance of relative advantages.

The conclusions we can draw is that, in order to achieve higher economic and employment growth, it is important to be active in the "right", faster growing industries, rather than in the fields of greater relative specialization. Large mature economies can hardly rely on the operation of comparative advantage for this process to take place. Developing economies may have some room for catching up through competitive advantage in the more traditional and slow growing sectors of international trade. What appears to matter most however are the structural shifts brought about by economic change, including those associated to the emergence of new Information and Communication Technologies (ICTs).

In order to achieve long term growth and competitiveness, and reduce unemployment, more attention has to be paid to the ability to innovate in technologies, production, organisations and institutions. Such a view of the foundations of competitiveness suggests that long-term efforts to adapt the economic structure and to sustain the factors contributing to productivity growth are more relevant than short term actions aiming at reducing costs or increasing the efficiency of the operation of specific markets.

This is particularly important in the current debate on policies to counter the unemployment problem. Innovation and structural policies are more important than actions narrowly focused on labour costs and markets, aiming at increasing flexibility and downward mobility of wages, which have recently been undertaken in most countries.

A useful summary of policies for supporting national competitiveness is provided by Porter (1990, p.619); it includes the following recommendations:

• "Firms are the players in international competition, not nations", and government policy should be targeted to national firms.

• "A nation's competitive advantage in industry is relative", depending on the efforts of other countries in the same field.

• "Dynamisms leads to competitive advantage, not short-term cost advatages".

• "National economic prosperity demands that industries upgrade", otherwise they will succumb to foreign competitors with cost advantages and an innovative edge.

• "Competitive advantage in industries is often geographically concentrated".

• "Competitive advantage in a nation's industries is created over a decade or more", emphasizing the need for long-term strategies.

• "Nations gain advantage because of differences, not similarities", as domestic specificities in learning, capacities, economic activities and social and institutional settings are at the root of the emergence of competitive industries.

• "Many categorizations used to distinguish or prioritize industries have little relevance", as differences within industries may be equally important than cross-industry distinctions.

• "The process of sustaining advantage may be intensely uncomfortable for firms and those who work in them", as it implies a constant pressure for investment, improvement and productivity gains.

The above list of policy issues emphasizes the importance of a long term, dynamic view of competitiveness based on technological factors and on the specific strengths of national economies (be them of a geographical, social or institutional nature). Such an approach has generally positive implications for the reduction of unemployment, job creation and the maintenance and growth of wage levels and workers' rights. However, the downside of such views of competitive efforts is that a continuing effort to increase productivity is required, and some degree of mobility from declining to emerging sectors is needed.

Two sets of data are used in this section. The first include UN data for most countries of the world, taken from the Second European Report on S&T indicators, 1997, on employment and export of the total economy in 1985 and 1995 (or the closest available year).

The second set is taken from the OECD STAN database (1996) and refers to the total of manufacturing industry for all OECD countries from 1970 to 1995.

First, Table 1 shows for all the countries with available data the situation of employment and export in 1985 and 1995 and the rates of change over the period (care should be taken in considering data for Germany, as 1995 data includes the former GDR, and for the Netherlands, whose high employment growth hides the multiplication of part time jobs). Regional data are also included for the most important aggregations.

The most important regularity is that export growth is generally substantially higher than job growth. The fastest export growers are Asian countries, which have supported important increases in employment. European countries have expanded trade by 50-60% while jobs have increased modestly. A lower trade growth but higher job increases are found in North America. The worst performances are those of South American and Central and East-European countries, with frequent dramatic job losses and slow-growing or stagnant exports.

Figure 1 shows the position of each country in terms of the growth of export and employment. A general positive relation is evident, with extreme cases of some Asian countries relying on export-led growth, but generating lower than expected jobs, and of some oil-producing countries whose exports have been badly hurt by the fall of oil prices.