An often noted but relatively ignored feature of the global economy, is the rapid change in structural composition away from commodity production and in favour of services. This tendency has continued through the 1990s, despite the fact that the share of services in gross value added in some industrial countries was already extremely high at the end of the 1980s (Table 1).
Table 1 >>
 
An examination of the relative ranks in terms of services share in national income of a selected group of OECD countries shows that it is indeed influenced by relative levels of income, with the United States at the top of the table. Yet there are exceptions, as suggested by the position of a relatively less developed economy like Mexico, where the composition of economic activity has changed dramatically, even though income still lags. In fact, the experience across the developing world does suggest that liberalisation, which allows market signals to direct investment and reduces restrictions on the entry and nature of activity of foreign investors, encourages 'premature' diversification into services.
 
The relation between the level of development and the degree of diversification of economic activity in favour of services is much clearer when we examine figures on the share of services employment in total employment (Table 2). Here, not only do countries like the US and UK head the league table, with similar shares of income and employment being generated by services, but Mexico with a 66 per cent share of services in income drops way down in the league table with just 54 percent of employment being in the services sector.
Table 2 >>
 
A more long term analysis (Table 3) of changes in the structural composition of output in the US, which has recorded the highest degree of diversification into services, reflects a continuing and persistent setback to commodity production and a burgeoning of services. This process has, however, quickened in recent years. Between 1980 and 2000, for example, the share of private services producing industries in the National Income of the US rose by close to 14 percentage points, from 50.8 to 64.3 per cent, while the increase over the more than three decades between 1948 and 1980 was just 7.4 percentage points.
Table 3 >>
 
This growth in services has been primarily at the expense of the share of manufacturing in national income and not the result of the shrinkage of the private agricultural sector, whose share in private sector income stood at less than 5 percent even by 1960 (Table 4). However, the share of manufacturing income from private industries, that fell by little over 5 percentage points between 1948 and 1980, collapsed by a further 10.2 percentage points between 1980 and 2000.
Table 4 >>
 
Till some time back, this trend would have been considered an infirmity, since barring a few productive services, the strength of an economy was seen to be reflected by the expansion of and productivity increase in the commodity producing sectors. This was partly because the services sector was conventionally considered to be one in which there are substantial barriers to productivity increases. There are, however, two arguments that seek to counter the "productivity pessimism" that the evidence on the importance of services generates. First, there is reason to believe that the available evidence exaggerates the expansion of services. Precisely during the years when the new economy has allegedly emerged, a part of the rise in the share of services appears more statistical rather than real. It is not just that during these years the growth of manufacturing industries' GDP lagged behind that of services, resulting in a fall in manufacturing's share of GDP from 24 percent to 18 percent. Within manufacturing, while the share of intermediate transactions or costs paid out by individual activities remained constant at 43 per cent of industry gross output, the share of manufacturing intermediates declined from 22 per cent to 17 per cent and that of services rose from 21 per cent to 27 per cent (Jiemin and Planting 2000).
 
This growing services-intensity of manufacturing can be interpreted in two ways. On the one hand, it can be taken to reflect the growing productive role of services activity in the US economy. On the other, it can be seen to be a reflection of a growing trend towards outsourcing of services by US corporations. "Surveys conducted in the United States by the Outsourcing Institute show that companies with over USD 80 million in annual revenues increased outsourcing by 26% in 1997 to USD 85 billion. IT was the fastest growing activity being outsourced, accounting for 30% of total outsourcing expenditures. Human resources was the second largest (16%), followed by marketing/sales (14%) and finance (11%). Manufacturers accounted for nearly two-thirds of the outsourcing, with information and professional services each accounting for 13% of the total." (OECD 2000). If the latter were true, it would imply that the rise in the share of services and decline in that of manufacturing is in part a statistical rather than a real phenomenon, reflecting the splicing out of services that were earlier part of the value of manufacturing output. In France, for example, the combined contribution of manufacturing and industry-related services has increased only marginally from 27 to 29 per cent between the 1980s and 1990s, indicating that restructuring rather than expansion accounts for the growing share of services (OECD 2000).
 
This possibility notwithstanding, there are sound grounds to argue that the evidence does reflect an actual tendency on the ground, even if it exaggerates the growth in the share of services. But this is where the second challenge to "productivity scepticism" comes in. Advocates of that challenge suggest that, in the new technological and economic environment, service sector growth is as good as growth in the commodity producing sectors. A report prepared on the basis of a Business and Industry Policy Forum organized by the OECD summarized the position as follows: "The relative importance of manufacturing and services to economies, and the inter-relationship between the two have been the subject of much discussion through the years. Some have argued that the decline in manufacturing and the corresponding shift to services is unsupportable in the long run, since services depend critically on manufacturing for their existence. In the absence of manufacturing, service sectors are seen as collapsing. On the other hand, a forceful case was made at the Forum that services have become a major driving force in economic growth. Rather than services following and supporting manufacturing, manufacturing is seen as flowing to those countries and areas where the services infrastructure is efficient and well developed." (OECD 2000).
 
This position that service sector expansion drives economic growth stems from a reinterpretation of the nature and role of services in modern economies. It requires making a distinction between services as they prevailed in the earlier decades of post-war capitalist expansion and their nature and role in more recent times. Services, it is being argued are increasingly resembling commodities. The conventional idea that services differ from commodities because they cannot be stored, that they have to be consumed at the point of production and that their consumption requires the direct interaction of service providers and consumers is no more true of a range of services, the argument goes.