Pierre Veltz Livres Articles Tribune Contact International

Pierre Veltz, Ecole Nationale des Ponts et Chaussées, Institut d’Etudes Politiques Paris

Business services in the global service economy

Symposium international sur l’économie des services, TianJin (Chine) juin 2005


It is a commonplace to distinguish different stages in the pathway of a national economy: after the agricultural stage comes the industrial stage, and then the so-called post-industrial stage, characterized by the predominance of information (rather than material) processing sectors, knowledge (rather than cost) based competition and a dramatic shift in the occupational structure. So, the US and the European economies are supposed to be post-industrial: 85 % of jobs in the US belong to the service sectors (either personal or business services), 75 % in France.

This view, and the related narrative, seems “natural” in the case of old industrialized countries, given their history, and the rather slow pace of the changes in their economic structure along the past century. The deep and often painful restructuring processes that took place over the last decades in mature industrial sectors, facing increased competition in the new global arena, enforces in the mind of a majority of citizens, even in the decision-making spheres, the idea that manufacturing is definitely out-of-date, and that “de-industrialization” (in favor of emerging countries like China) is our natural fate. Some think there will be a (more or less stable) new division of labor between “post-industrial” societies and a set of countries specialized in manufacturing and/or in standardized low-skilled (“manufacturing-like”) services. My argument in this paper is that these views are basically wrong. I would argue on the contrary that : (a) even in countries where (direct) manufacturing employment is getting a very small part of the overall employment, it is much more relevant to characterize the present state of the economy and the society as “hyper-industrial”, rather than “post-industrial” ; in other words, the service activities and a deeply reshaped manufacturing sector (service-oriented rather than merely supplier of material goods) are merging in a complex structure where the traditional distinction between the two spheres, far from being a basic framework as in the post-industrial theory, becomes misleading and confusing, both for the understanding of what is happening and for the design of relevant policies; (b) for developing and emerging countries, the simultaneous (and not the sequential) development of manufacturing and services is certainly the winning policy, even if there are many different ways to combine and articulate the trends observed in more advanced economies ; the vertical division of labor (roughly speaking : intellectual labor and the upper parts of the value chains in the most advanced countries; manual or routine labor and the lower part of the value-chains in developing countries) will not disappear, but will move through different successive geographical patterns; the rapid growth of a very numerous high-skilled workforce in countries like China and India, combined with the possibilities offered by new generation IT, will on the contrary lead to the creation of many areas in the developing world displaying the same comparative advantages as the leading regions in the developed world, and entering therefore in the highly cooperative and competitive sphere of global innovation, in horizontal and distributed value chains; and, even for the regions which will not rank among this upper set of leading regions, there is no reason to reproduce the historical pathway of Europe or the US, and not to go directly to the “hyper-industrial” structure, bypassing the former stages of traditional manufacturing.

The post-industrial thesis has different roots and relies on a vast literature. D. Bell, an American sociologist, has coined the term in 1973, emphasizing not only the changes in the occupational system in advanced economies (and mainly in the US) but the fact that knowledge and science based technology replaces progressively the physical energy of human labor (and traditional machinery) as the real basis of growth and as the core process of social development. Sauvy, a French demographer, explained how the increase of productivity generated by the growing automation in the manufacturing sectors results, through a complex process he called “productivity overspill”, in the creation of new jobs in the sectors of services where automation is impossible. Many economists, like Baumol in the US, have emphasized this distinction between two sectors: a sector characterized by high productivity (broadly speaking: manufacturing) and a sector where the productivity is by nature stagnant. Finally, the rather universal division existing in the statistical taxonomy between manufacturing and services gives this division the status of a quasi-natural dichotomy. So, why should we prefer the idea of a “hyper-industrial” society to the idea of a “post-industrial” one? For several reasons:

De-industrialization in advanced societies is a largely a myth .

Of course, manufacturing employment has decreased dramatically but the share of the added value of industrial sectors in GNPs remains, roughly speaking, constant, if not growing. One should add that industrial employment has never been, in the past, as dominant as the political weight of industrial workers and of big manufacturing firms seems to indicate: only in England during the 19 th century has the ratio of industrial workers reached about 50% of the global workforce. Moreover, this stable ratio of manufacturing in the economy is obtained despite two major counter-effects: the decline in relative prices for manufacturing goods, compared to “immaterial” goods ; the massive outsourcing of service activities that were previously part of the internal organization of manufacturing firms, which has reduced artificially the boundaries of the industrial sector as measured by statistics. This last point must be stressed. Of course, there are several mechanisms explaining the rapid increase of business services since the 80s. The growing complexity of the overall economic process, largely driven by the fragmentation of production networks in the globalization (see M. Delapierre’s contribution) and the related needs for coordination, on the one hand, the specific and relatively autonomous expansion of the financial sphere (with the creation of new markets and a wholly new range of financial products, like the futures and the derivatives), on the other hand, are probably the main factors. But the outsourcing factor is also an important one, given the large (and ever enlarging) scope of activities concerned by this movement. And it is very difficult to separate analytically and statistically these trends, for different reasons: the lack of relevant data, but also the fact that the outsourcing of services like IT, facilities management, human resource management, accounting, entire “business processes” (in so-called BPO: business process outsourcing), creates in itself new needs for coordination, exactly as the fragmentation of the manufacturing process does. In the developed countries, the share of the business services in the intermediate consumption of the firms was, broadly speaking, 10% in 1970, 20% in 1990 and 30% in 2000 ; but this ratio is even higher in some tertiary sectors (banking and insurance, transportation) than in the manufacturing sectors.(See N. May’s paper).

The result is therefore a structure in which large and diversified arrays of focused units (either blue-collar or white-collar) – design centers, logistical infrastructure and info-structure, production units, commercial units, routine service suppliers, specialized service suppliers – constitute intertwined networks crossing the border of both industry and services. For a growing part, these networks cross also the national borders, and this raises the question of the geographical patterns of the relationships between buyers and suppliers of producer or business services. The opportunities for local service suppliers in the case of global networks are discussed in M. Delapierre’s paper. Empirical evidence is that the dramatic increase of FDI by multinationals (either in manufacturing or in service sectors), which is the core process of the globalization of production, is only one part of the picture: it goes along with a strong development of local subcontracting, which explains for instance why the part of intra-firm trade in international trade remains more or less constant, in spite of the dramatic rise of FDI (Milberg, 2004). The breaking up of the production processes into small parts rises also the sensitivity to labor costs differentials: even is the labor cost is a relatively small part of the overall process, the focus put on specific segments of the value chain (generally labor intensive in the case of service activities) brings to light new opportunities of cost cutting, and therefore reinforces the trend towards externalization, in a positive feed-back. (Krugman,1995).

Let me finally stress that this idea of a “hyper-industrial structure”, where the boundaries between services and manufacturing are greatly blurred does not mean that the development of always more sophisticated sets of services is only a consequence of the growing complexity of manufacturing. There is also a powerful trend of self-generated expansion in some sectors of services: the best example is the huge development of the finance industry after the invention of new financial instruments and the creation of new markets in the 80s, which boosted the growth of employment in places like London and New York.

The economy as a whole is shifting to a “service economy”

The descriptive and organizational approach showing the close and intertwined links between different sectors catches only one aspect of the economic reality. The most important point is to go beyond the dichotomy between goods and services, putting apart traditional distinctions that seem relevant only in a very superficial view, and to understand the deep convergence between manufacturing and services at the level of the process of wealth-creation itself, considered as a whole. This convergence is generally summarized under the terms of “service economy” (see for instance Giarini, 1987). And this thesis is of course compatible with the idea of “hyper-industry” (so we should characterize the present economy as a “hyper-industrial service economy”).

1) The first set of major changes concerns the firms (especially in manufacturing) under the pressure of global competition, during the 80s, in connection with the technological changes (the new possibilities of IT, but also the development of an always more complex and fragile machinery). In this truly new context, a silent (though generally painful in terms of employment) revolution occurred in the large firms of advanced countries. The core of this revolution is the shift from a price-based competition to a much more complex pattern, where quality, diversity and innovation, for goods and services, are becoming key-factors of economic survival; therefore the traditional difference between price-based competition and differentiation is getting less and less clear, except for small niches. Firms involved in global competition have generally no choice between price-based and differentiation-based competition; they must face both, at the same time. And this is linked with a major underlying change in the dominant efficiency paradigm (Veltz, 2000). Competitiveness lies no longer in the mere intensification of work and traditional productivity effects. Efficiency depends less and less on pure division of labor (which was the main driver of productivity for centuries). It depends more and more on the quality of the cooperation processes, of which only a small part can be standardized and mechanized, as in Taylorist industry. Criteria such as innovation or quality of goods and services, and the reliability of sophisticated machinery, which is the key factor of the productivity of capital, depend crucially on the quality of formal and informal communication among the actors of a value-chain, among the different components of a firm, among the firm, his suppliers and his clients, among engineering, production and marketing, etc., and eventually between the firm and its overall environment.

In this context, manufacturing firms, even in mature sectors like steel industry, move progressively to new models where “service” is a key strategic issue. Firstly, the traditional concept of “products” is challenged: corporations are engaged in the production of complex packages in which goods and services are mixed. Secondly, the focus of the strategy is no more merely to develop new products and then try to sell them to anonymous customers, but to create channels in which the transactions between suppliers and customers, at the various stages of a value chain, can be optimized. Efficiency gains and competitiveness are no longer confined to material processing, but extend to the management of relationships (inside the boundaries of the firm, but also outside, with the suppliers and with the intermediate or final customers), this latter element becoming the crucial one. In other terms, firms move from “product out” to “market in” strategies. Steel-makers sell no more steel, but solutions to problems for their clients, who are frequently involved in upstream cooperative R&D processes with them. Fine tuned knowledge about the habits and the supposed needs of the customers becomes strategic (as in Customer Response Management) as well as logistics and distribution (Supply Chain Management), both approaches being increasingly linked. There are plenty of examples where the actors of distribution become the real center of power in the value chain, because of their proximity to the final market (this is for instance the case in the textile industry in Europe or in the US). “Service economy”, in this context, refers no more to a sector-based definition. It means deepening the interactions between providers and users, suppliers and buyers, at every stage: interaction before the production step, in order to identify the packages which will generate the most additional value ; interaction during the production process, to adapt the product to the changing needs and “customize” it ; interaction after the delivery, to facilitate use and detect new opportunities for the next cycle of development.

2) At the same time manufacturing firms move towards a service model, service industries, for a large part of theme, move to efficiency patterns that are very close to the manufacturing practices. There is no fundamental difference between designing and operating large and complex IT infrastructure for the car industry, retail or banking. Standardization of procedures, quality control, etc. spread in services as well as in manufacturing. A major change, with far reaching consequences, is the following: the traditional “non-tradability” of most services – requiring providers and users to be in the same place at the same time - is deeply challenged by IT. Information can be stored, and instantly moved through long distance at very low price to be processed. The result is the quick rise of “off-shoring” in the field of information processing services (which means potentially a very large part of service activities), either intra-firm (captive) or externalized. Of course, every task is not likely to be off-shored : many complex activities require structurally face-to-face contacts and cannot be achieved through routine an impersonal interactions. But, on the other hand, the development of IT will continue to increase the tradability of services, and, at the same time, the convergence between traditional industry and service industries. The continuous shift of FDI towards services is remarkable: in the early 70s, the services accounted for one quarter of the world FDI stock; by 2002, it had risen to about 60% (UNCTAD, 2004). The fastest growth has taken place in the US and in Europe, reflecting the fact that most service FDI is market-seeking rather than efficiency-seeking.

The knowledge-based economy: an ambiguous mantra

In this context, what is the meaning of “knowledge-based economy”? Often featured as a radical novelty, “knowledge-based” economy is, in one sense, as old as industry, or economy. The Parisian economy of the 17 th or 18 th century, devoted to the production of luxury goods, was a knowledge-based economy, in the sense that the basis of his efficiency lied, at base, in the minds of the craftsmen. In the context of the present, and in a narrower sense, to talk about a “knowledge-based economy” refers generally to two changes: the increasing role of shortened R&D cycles in the competitiveness, given the rapid spread of newly born technologies all around the world, and the limited effectiveness of intellectual property protection; the closeness, in time and in nature, between scientific research and market-oriented technological development. In the mind of many, a strategic orientation towards “knowledge-based competition” means, for a city or a country, a shift from old-fashioned manufacturing to “high tech services”. (In this view, popular among the politicians, Bangalore is a knowledge-based city, Chicago, or Nagoya, with their manufacturing concentration, is not). But this view is naive. We should remain aware of the fact that manufacturing, even in mature but highly competitive sectors, and precisely because of this competitiveness requirement, which makes innovation a matter of life or death, is still the main driver for technological investment, far before the service sectors. (We must also, on the contrary, remind the bias introduced by the fact that R&D can be easily isolated and therefore measured in manufacturing, but is much more difficult to catch in services, where innovation is “soft” and distributed along the whole value chain (Callon, al, 1997). Once more, innovation and knowledge-based economy has to be replaced into the framework of the “hyper-industrial service economy” rather than on the background of obsolete dichotomies. A very interesting study of the high tech employment in American cities (Markusen, al. 2001) shows that the results are deeply different if the criterion used is the share of the high tech sectors in the local economy or the share of high tech employment throughout all sectors: on this second, occupational, base, the first high tech American city is Chicago, largely outdoing Silicon Valley! And the study ranks older industrial cities such as New York and even Detroit higher than places celebrated as high tech like Phoenix or Austin.

Diversified metropolitan areas are the breeding ground of hyper-industrial service economy

The development of business services plays a major role in the rise of employment in large cities (see N. May’s contribution). In some cases, this development has given birth to occupational structures dominated by transnational corporate headquarters, international finance, transnational institutions and a complex range of business services related, directly or indirectly, to the needs of co-ordination of the global or “regional” economy. These “world-cities” (Knox, Taylor, 1995) represent the upper layer of urban hierarchies; at their top, we find the so-called “global cities” described by Sassen (Sassen, 1991), New York, London, and Tokyo, with their huge concentration of international finance. On the one hand, these places are really major hubs of co-ordination of the world economy; but, on the other hand, the City in London or the Wall Street district in New York are simply “industrial districts” specialized in some parts of the finance industry, just like Detroit is specialized in the car industry, LA in movies, Rotterdam in maritime logistics, Boston in medical research. And it has to be stressed that, even in metropolitan areas like New York or London, where headquarters of multinationals are concentrated, many financial and business services cluster not only, as it is often supposed in the literature, as direct suppliers of these headquarters, but for the same reason specialized sector tend to concentrate, independently of HQ locations: economies of scale, cross-sector fertilization, shared labor market, etc. As J.V. Henderson puts it: “New York is a giant shopping center for firms looking for creative financial and service products. Much of this is sold on an export basis” (Henderson, 2004). The link between service economy and large cities is therefore, in my opinion, a much broader phenomenon than this trend to hyper concentration of advanced corporate services in some particular cities. My thesis is that the economic and organizational trends briefly sketched above are specially consistent with the environment of large diversified metropolitan areas, and therefore that there is a mutual reinforcement between the spatial polarization process around these big cities and the reshaping of the economy itself. To put it briefly (for more detail, see Veltz 2005): (a) there is a basic process, already experimented in former stages of economic history, which is that the decrease in communication costs does not by itself lead to a spreading and diffusion of activities; on the contrary, it entails their polarization. The explanation of this paradox is, at base, very simple. A world where communication costs are high is a world of closed and separate compartments, which strongly limits the level of competition between the firms, creates spatial monopoly rents, and hinders the expression of economies of scale and of the advantages of agglomeration, for consumers and producers. On the contrary, as the fluidity of goods and information grows, these positive effects in terms of increasing returns and of agglomeration externalities can unfold freely and appear on the front stage. What demonstrates the metropolitan growth process? Basically the overwhelming strength of agglomeration externalities, often underestimated by the economists, either pecuniary (i.e. linked with the price mechanism) or “technological”. (b) This basic mechanism is a generic one, and it is relevant for many periods of history. In the present period, however, this rationale is strongly reinforced by the specific features of what I have described as the shift to the hyper-industrial service economy. This is particularly visible if we focus on the dynamic effects, in contrast with a sometimes rather static vision of economies of agglomeration. Because cities offer not only complementary assets (like in the input/output structure of a local production system) and static coordination between actors of the economy, but the possibility of reshaping rapidly and efficiently the networks of actors and the value chains themselves. Metropolitan areas can be likened to powerful hubs or switchboards, which permit the constant creation and reshaping of the chains linking producers, consumers, and different kinds of indirect players of the economy (like universities or public agencies of all kind). They are especially efficient in accelerating the search processes that are the basis of growth in a Schumpeterian context. As for final consumption markets, large cities are the laboratory for new products, new services, new forms of fragmentation and coordination along the value chains, new ways of management of the relationships between providers and users of goods and services. Cities, in that respect, are the principal suppliers of the relational resources that fuel these open-ended coordination processes, which cannot be set up either by the decision making of a centralized structure or through sheer market forces. They provide many forms of relational externalities, based or not on informal communities, based or not on face-to-face. The shift occurring in “service-oriented” organizations from hierarchical and pyramidal structures to more open network-shaped structures acts like a positive feed-back in this context: large cities are the adequate ecosystem for the continuous restructuring of such networks. All that is of course strongly dependent on size, “deepness” and flexibility of the labor market, which are in fact the main competitive advantages of big cities, compared to smaller ones.

Ports and logistical services as crucial nodes of global service economy

Global supply chain management and logistics are of crucial importance in the global economy, as far as they ensure the physical coordination (based on an always growing and expanding network of sophisticated information processing activities) throughout the fragmented landscape of global production and distribution systems. The cross-border development of logistics, and especially maritime logistics, constitutes at the same time the result of this transnational fragmentation of production and, together with international finance and international legal regulations, one of the fundamental pillars sustaining the global economic system as a whole. Therefore, the efficiency, the smoothness, the reliability and the safety of the maritime logistical system are crucial components of the competitiveness for every region or country (in China, approximately 90% of the country’s international trade, in volume, is handled through maritime transport, and 20% of the world’s container traffic occurs from Asian ports). At this point, we must stress a paradoxical effect. Globalization relies basically on the reduction of trade costs, including tariff costs, freight cost and all kind of transaction costs, which cannot be measured directly but appear indirectly (such as “border-effects” in officially unified markets). The general lowering of tariff barriers implies that transportation costs, in spite of their own sharp decrease, represent a constant, if not increasing part of trade costs, and are more carefully scrutinized by the investors (Hummels, 1999 ). A second effect is that the global efficiency of maritime transportation chains depends mainly on port efficiency. A study of bilateral trade in the case of Latin America and the US in relation to port efficiency shows the crucial importance of this element: improving port efficiency (lessening bureaucracy, fighting organized crime, improving infrastructure and time-based reliability) has macro-economic effects ; bad ports (under the 25 th percentile) are equivalent to being 60% farther away from markets for the average country. (Clark, Dollar, Micco, 2004).

Finally, rather than simple technical infrastructures, or than the location of a set of specialized functions, seaports must be analyzed as elements in a value-driven chain system or constellation, delivering service to shippers and to third-party service providers. They can also be seen from a cluster perspective, as concentration of commercial activities, “information hubs”, and many different services which capture value for both the local city and the users of the port. (Robinson, 2002 ; de Langen, 2004);

Therefore far-looking development strategies in seaports must cope with the multiple and diversified opportunities of the hyper-industrial service economy as a whole.




Callon M., Laredo P., Rabeharisoa V., Que signifie innover dans les services ? Paris, La Recherche, février 1997

Clark X., Dollar D., Micco A., Port Efficiency, Maritime Transport Costs and Bilateral trade, NBER WP N° 10353, March 2004

De Langen P., Governance in Seaport Clusters, Maritime Economics and Logistics, volume 6, Issue 2, 2004, 141-156

Giarini O. (ed), The Emerging Service Economy, Oxford, Pergamon Press for SWF, Geneva, 1987

Hummels D. Toward a Geography of Trade Costs, University of Chicago, January 1999

Henderson J. V., What Makes Big Cities Tick ? A look at New York, paper presented at LSE conference on Resurgent Cities, April 19-21, 2004

Knox P.L., Taylor P.J., (eds) World Cities in a World-System, Cambridge University Press, 1995

Krugman P., Growth in World Trade: Causes and Consequences, Brookings Papers in Economic Activity, 1995:1, 327-377

Markusen A., Chapple K., Schrock G., Yamamoto D., Yu P., Gauging Metropolitan “High-Tech” and “I-Tech” Activity, WP, University of Minnesota, December 2001

Milberg W., The Changing Structure of International Trade Linked to Global Production Systems: What are The Policy Implications ? International Labour Office, Geneva, May 2004

Robinson R., Ports as Elements in Value-Driven Chain Systems: The New Paradigm, Maritime economics and Logistics, volume 29, Issue 3, 2002, 241-255

UNCTAD, World Investment Report. The Shift to Services, Geneva, 2004

Sassen S., The Global City, Princeton University Press, 1991

Veltz P., Le nouveau monde industriel, Paris, Gallimard, 2000

Veltz P., Mondialisation, villes et territoires. L’économie d’archipel, Paris, PUF, Quadrige Essais, 2005