International Conference ‘Small Firms Strategy for Innovation and Regional Problems’ Algarve, 4 and 5 December 2003 Measuring the portuguese ICT sector at a local level Flávio Nunes Department of Geography, University of Minho (Guimarães, Portugal) [email protected]

1. Introduction There is a widespread belief concerning the impact of ICT investments on productivity and general living standards. However several studies based on empirical analysis failed to find and prove that link, in some cases it was even detected a slow down on productivity growth during the period in which large investments in ICT occurred (Haltiwanger and Jarmin, 2000; Moulton, 2000). Some explanations can be advanced for this “productivity paradox”1, like the inability from official statistics to capture all the improvements in cost savings associated with improvements in information flows and reduced transaction costs between organizations, or the possible and well known lag between investments in such innovations and related productivity increases. Nevertheless the lack of ability to evaluate if firms have finally see returns on investments in ICT, it is unquestionably that technological change is a essential engine of economic growth, and since knowledge “pushes back the frontiers of technology” (Quah, 2000, p. 31), a knowledge-based economy is a concept and a reality progressively recognised. One broad area of research related to the knowledge-based economy is concerned with the need to understand how these new equipments and services have not only enabled businesses to do their work in new ways (promoting organizational readjustments), but like Moulton (2000, p. 35) says “has led to the creation of new firms and even entire industries”, what it is usually defined as the ICT sector. The purpose of this study is mainly concerned with the detailed characterization of the portuguese ICT sector. To achieve this goal this research has been conducted in order to understand the organization and location patterns of these new forms of business and production, and how they influence changes in portuguese economic structure as well as their impact in local labour markets. Here are some of the questions that we will try to answer: What kind of activities are included in portuguese ICT sector? What is the size of it? How fast is it growing? How does it contribute to economic activity? How do portuguese municipalities distinguish in terms of specialization in this kind of production and services provided? To answer these questions and in order to identify carefully all firms of the portuguese ICT sector, in the first section of this work we will try to clarify what kind of economic activities are included in it. This is not an easy and consensual task, on one hand we have a recognized lack of synchronization about the contents of this sector, and on the other hand to measure the knowledge-based economy with comparable statistics it is important to point out that the currently used classifications of economic activities do not have yet a clearly recognized ICT sector. Just before studying with geographical and economic detail the portuguese ICT sector, we will do a short macro level approach, in order to make a comparative analysis 1

about the gradually significance of ICT enterprises on economic key figures, not only with US and EU averages, but also with each one of the others EU-15 countries. To conclude this work and to measure and evaluate the portuguese ICT sector at a local level, it will be presented a detailed cartography analysis with a very geographical desegregation, in order to better understand the spatial diffusion of ICT enterprises and establishments, as well as the relevance of the sector to the local employment and sales volume. In our opinion a strict measuring of the portuguese knowledge-based economy implicate a local level analysis, not only to identify some clustering tendencies affecting these activities, but also because in terms of local development there is an evident challenge related to the connections between enterprises included on ICT sector (specially ICT-services) and the local business environment. It is more and more consensual that the effect of ICT on increasing local and regional productivity is not confined to dotcoms enterprises performance, but their most important impact will be on existing companies, revolutionizing the ways they do business, the ways they interact with one another and with their staff and customers. Like Cairncross (2001, p. 128) says “these new relationships (…) will clearly be an important –perhaps the most important– competitive advantage for businesses in the new century”. But to promote these relationships upgrading, the traditional business district must be recipient of services intended to enable the function of information processing and communication by electronic means. As they integrate the services provided by these new economic activities, existing companies will change and will be gradually included in the new economy. So, it is essential to determine if this kind of new activities are locally available throughout Portugal. 2. How can the ICT sector be defined? Specially in consequence of the recent excitement and subsequent crash in the market value of dot.com firms, “reliable and comprehensive indicators are needed to track developments in new information technologies and understand their impact on our economies and societies” (OECD, 2002, p.3). As ICT has only be recognised as a major source of economic and social change in recent years, statistics on the Information society are still under development and this is one of the most demanding and urgent challenges for the official statistical community. It is important to remind, in order to measure the knowledge-based economy, that the currently used classifications of economic activities (ISIC Rev.3)2 do not have yet a clearly recognized ICT sector. Recognising this problem, since 1997 the OECD Working Party for Indicators on Information Society (WPIIS) in collaboration with Eurostat Task Force on Information Society Statistics, has been addressing the need for international standards for ICT statistics, with the purpose of establishing a set of definitions and methodologies to facilitate the compilation of internationally comparable data for measuring various aspects of the information economy, specifically goods and services produced, consumed or traded and the entities involved in those activities. If it seems to be easier to come to an agreement on the definition of the ICTservices3, for the ICT industry it seems to be more complex to come to a common stance, since difficulties arise in the borderline discussions relating to specific ICT goods (Gardin, 2000). However, since it was crucial to come to a definition that could be applied to actual compilation of internationally data as soon as possible, the WPIIS adopted some guiding principles for the delineation of the manufacturing parts of the ICT sector:

2

“For manufacturing industries, the products of a candidate industry must: - be intended to fulfil the function of information processing and communication by electronic means, including transmission and display; or - use electronic processing to detect, measure and/or record physical phenomena, or to control a physical process. Components primarily intended for use in such products are also to be included.” (Gardin, 2000, p.3) In Table 1 is given the existing list of ICT activities according to these guidelines4. Since users generally require data to be comparable over time and to be able to link domestic production data with international data, this table shows the concordance between ISIC Rev.3 (International Standard Industrial Classification) and the classification used in all EU member countries, NACE Rev.1 (Statistical Classification of Economic Activities in the European Community) and CAE Rev.2 (Portuguese Statistical Classification of Economic Activities) TABLE 1. Activity-based definition of the ICT Sector ISIC Rev.3

NACE Rev.1 / CAE Rev.2

3001 - Manufacture of office machinery 3000 - Office, accounting and computing 3002 - Manufacture of computers and other machinery information processing equipment 3130 - Insulate wire and cable 313 - Manufacture of insulate wire and cable 3210 - Electronic valves and tubes and 321 - Manufacture of electronic valves and other electronic components tubes and other electronic components 3220 - Television and radio transmitters 322 - Manufacture of television and radio and apparatus for line telephony and line transmitters and apparatus for line telephony telegraphy and line telegraphy 3230 - Television and radio receivers, 323 - Manufacture of television and radio sound or video recording or reproducing receivers, sound or video recording or apparatus, and associated goods reproducing apparatus, and associated goods 3312 - Instruments and appliances for 33203 - Manufacture of instruments and measuring, checking, testing, navigating appliances for measuring, checking, testing, and other purposes, except industrial navigating and other purposes, except process equipment industrial process equipment 3313 - Industrial process control 333 - Manufacture of industrial process equipment services control equipment services 5150 - Wholesaling of machinery, 5164 - Wholesaling of machinery, equipment equipment and supplies and supplies 6420 - Telecommunications

642 - Telecommunications

7133 - Renting office machinery and equipment (including computers) 721 - Hardware consultancy 722 - Software consulty and suplly 723 - Data processing 72 - Computer and related activities 724 - Database activities 725 - Maintenance and repair of office, accounting and computing machinery 726 - Other computer related activities Source: based on European Commission (2002) and OCDE (2002)

ICT - manufacturing

ICT - wholesales ICT telecommunications

7123 - Renting office machinery and equipment (including computers

ICT - business activities

One of the most important aspects of the OECD ICT sector definition is that it fractures the long-established ISIC dichotomy between manufacturing and services activities5, since “activities producing or distributing ICT products or services can be found everywhere in the economy” (OECD, 2002, p. 81). However, this kind of readjustments that will support measures and analysis to determinate to what extent a society has been informatized is not consensual. Some sceptical views of the 3

information age, like Christopher May (2002), disagree with general claims that everything is organized on the basis of information and knowledge and argues that “underlying substance of our socio-economic system remains largely the same” (2002, p. 1). This author criticizes the rearrangements of various economic and employment statistics in order to firmly establish that the information society is emerging. In his view the emerging ICT sector is rarely concerned with new types of enterprises or products, but is more often the result of the progressing division of tasks into their constituent elements, that were spun off or outsourced (May, 2002). Despite this kind of argument being much connected with the most sceptical views of the information age, we also consider that more important than studying an economic sector in its global dimension, and corroborate international trends and patterns, it seams more significant, in order to assist a more comprehensive and realistic analysis, to promote a disintegration of ICT activities in four main categories: ICTmanufacturing; ICT-wholesales; ICT-telecommunications and ICT-business activities (see the right column of Table 1). In our opinion, this analytical approach will allow to obtain valuable information regarding the portuguese ICT sector development, in order to interpret it with reliability, since it will allow comparisons not only across time and space, but also between the diverse sort of activities grouped in it. Just before starting with our empirical analysis it seams important to clarify that if the progressive diffusion of ICT has led to structural changes in our labour force and in our economy, the ICT sector is just one of the categories used in measuring knowledgebased economies. For the last forty years several studies6 emphasized the substantial increase of workers engaged in tasks that require skill in the use of information, and in the production and distribution of knowledge. Nevertheless measuring the impact of these knowledge-producing activities it is not an easy and consensual task, mostly for the reason that there is a recognized lack of synchronization about its contents (Table 2). TABLE 2 – The categories included in the knowledge-based economy7 Machlup (1962) -Education -Research & development -Media and communication -Information machines -Information services

Porat & Rubin (1977) -Knowledge production and inventive industries -Information distribution & communication industries -Risk management -Search & coordination industries -Information processing & transmission services -Information products industries -Selected government activities -Support facilities in information industries

OECD (1981) -knowledge producing -Search, coordinating & risks management industries -Information distribution & communication industry -Consumption & intermediate goods -Investment goods

Castells (1996) -Research & development -Information and communications technology -Finances sector -Biotechnology

Quah (2000) -Intelectual property (patents, copyrights, advertising, financial and consulting services…) -Information and communications technology -Electronic libraries and databases -Biotechnology and pharmaceuticals

Despite this lack of agreement around the research focused on attempts to define better ways of measuring the size of workforce engaged in knowledge -or informationintensive occupations, the ICT sector has been always a common reference, since among the more common orientations to the information society are those that focus on diffusion of computer and telecommunications technologies as the defining 4

characteristics. In the opinion of Foray and Lundvall (1996, p. 13) basic features of the knowledge-based economy, like “the new dynamics in the formation of tacit and codified knowledge; the growing importance of networked knowledge; and the acceleration of processes of interactive learning”, are based on the increasing use of ICT. Beyond this general assumption that both phenomena (ICT and the advent of the knowledge-based economy) are strongly interrelated, in this research the option to study the ICT sector is mainly related to the recognised relevance of the impact of ICT on local economic development, which extend far beyond the growth of high technology companies. Nowadays, for all over the territory, all kind of firms are integrating computers and computer-controlled tolls into their operations, as well as the increase use of web-based purchasing technologies, with comparable competitive advantages. Like Fuller and Souther (1999, p. 290) said, the major reason for the interest which policymakers show on the take-up of ICT by small firms “is associated with contemporary notions of regional and national competitiveness”. This means that companies shifting needs should be recognized for the whole portuguese territory, and ensure that ICT services (wholesales, telecommunication and business activities) are locally offered. On the other hand, the ICT enables high technology companies (most of them big multinationals firms, or great national companies traditionally located on the biggest urban regions) to split off key functions throughout the world, in order to cut costs and obtain a higher level of efficiency, profitability and competitiveness8. Even small and peripheral countries, like Portugal, specially if they can offer some location advantages such as qualified workforce and technological infrastructures or tax incentives, can now focus on and compete for key form functions, whether they involve manufacturing, research and development, logistic or sales. Has Portugal been an attractive destination to this kind of investments? Where are they located? How large is the size of the workforce involved in this kind of activities? 3. An international comparison on the emerging ICT sector. The international success history of the Irish ICT sector. If we use the share of people employed in the ICT sector and the value added obtained by each one of these employees, the European Union average is still far away in comparison with the US reality. TABLE 3. Main figures of US and EU ICT sector

Share of ICT employment in total employment, 1999 (%) Value added at factor cost in the ICT sector per person employed, 1999 (thousand EUR) Source: data collected from European Commission (2002)

EU-15

US

2,8 81

3,5 131

Soete (2000) summarizes some of the most significant reasons that explain this lag of productivity rates, and shares of labour force engaged in the ICT production, distribution and related services. In Soete opinion, the transition to the knowledgebased economy demands substantial efforts concerning public and private investments, and if between Europe and US there aren’t such big differences related with the public support, during the 90’s the European private sector investments were not comparable, mostly in consequence of a fewer availability to assume risk investments and to deliver capital to small and medium enterprises engaged with new business opportunities. Other kind of reasons are related with the less demand of ICT, but not only by the business environments since the ICT penetration involves the ICT usage in households, enterprises, government, health, education and so on. In what concerns european 5

enterprises, they have been much more slower in comparison with the americans and the japanese in the progressive learning process about how can they readapt their organizations into the ICT challenges. As a consequence of this less demand, the ICT activities (manufacturing and services) were not moving fast enough to reach the US position, specially in what concerns the more advanced ones, those who enable the higher value added. Despite the US’s lead over the EU slowness in investment and in using ICT, this overall performance does, however, mask the significant differences which exist between the different EU countries. According to the European Commission (2002) in 1999, Sweden, Finland and Ireland where the EU member states where the ICT sector accounted for the largest share of employment (respectively 4,5%, 4,4% and 4%), even higher than US figures. In order to better distinguish between different economic performances of the ICT sector, we present the position of each one of the EU-15 member states, using 2001 data related to share of ICT exports in total amount of national exports, as well as the ICT trade balance relative to GDP at market prices (Figure 1 and Chart 1).

6

Chart 1. The EU-15 countries according to share of ICT exports and ICT trade balance (2001) ICT trade balance relative to GDP at market prices (%) 2001

12

IRL 10

8

6

4

L

FIN

2

S 0

0

10 F D

5

EL

I B E

DK A P

15

NL

20

UK

25

30

35

40

-2

-4

ICT exports. Share in total exports (%) 2001 Note: The axes crosses at EU-15 average for the share of ICT exports in total exports (12,3%), and on 0 for the ICT trade balance Source: Data collected from European Commission (2002)

In 2001 the portuguese ICT exports rounded 2,47 billion EUR, about 9,3% of total national exports. This figure is lower than EU average (12,3%), and results from a slower growth between 1996 and 2000, since EU ICT exports rose by 18,4% on average per year, while in Portugal it grew only 14,7% per year. The most competitive european economies on the ICT sector are Ireland, Luxembourg, Finland and Sweden (Figure 1). These four countries are the only EU member states where the ICT trade balance is positive, and the share of ICT exports is higher than the EU average. Ireland distinguishes itself because of the higher share of ICT in total exports (round about a third of the total Irish exports are related with ICT goods, what represent 32,6 billion EUR), and with the most positive ICT trade balance (the Irish exports of ICT goods exceeded the imports of the same goods by around 12 billion EUR, which represent 10,4% of the GDP at market prices). An intermediate group of countries, only composed by the United Kingdom and the Netherlands, present a negative ICT trade balance but a share of ICT exports higher than EU average. Although some important diversities, that can be detected by the careful analysis of Chart 1, all of the others nine countries recorded a negative ICT trade balance and a share of ICT exports lower than EU average. There are five EU countries with a lower share of ICT exports (Denmark, Belgium, Italy, Spain and Greece) than Portugal, associated with the presence of some units of multinationals companies (as we will see bellow), but only Greece has a more negative ICT trade balance. The portuguese imports of ICT goods exceeded its exports by 1,69 billion EUR in 2001 (1,4% of GDP at market prices), this trade deficit in 2001 was smaller than in 2000, and may signify a reversal of the trend to larger trade deficits in ICT goods observed in 1999 e 2000. The ICT sector in Ireland is an undoubted international success story. In order to look at the best practices in other international markets we will give some attention to how ICT has become the engine of growth in Ireland, how it has been subject to the current fragility of the whole global economy, and what are the strategies presented by the Irish government in order to preserve the competitiveness advantages in the ICT sector, and encourage Ireland growing reputation as a place where ICT companies can operate with very high levels of effectiveness. 7

In recent years the Irish government has placed an important emphasis on attracting foreign owned ICT companies, specially those subsidiaries of US-based multinationals9, which have progressively promoted the appearance of many small indigenous companies. This policy, that transformed Ireland in the major European base for USbased ICT firms (Dobbins, 2001), is considered pivotal to the recent high rates of economic growth and impressive job creation. According to Green and Galway (2002), in the 90’s, Ireland attracted over a fifth of all electronics FDI (Foreign Direct Investment) in Europe and half of all software FDI. As a result Ireland became the fifth largest exporter of PC’s in the world, and the biggest exporter of software products with a third of the global market10. In 2001, however, in response to a serious slowdown in the US technology sector, a number of American ICT multinationals started the implementation of cost-cutting measures. This downturn in the US technology sector, followed by the current fragility of the whole global economy, has had great implications in terms of job losses in Ireland. What seams to explain this Irish vulnerability it is an already expected adjustment process, in Dobbins opinion (2001, p. 3) “as history shows, the economic system is subject to periodic crisis of profitability caused by increased competition, which leads to over-production, and, then, an eventual ‘shake-out’ characterized by cost-cutting, redundancies, and even plant closures”. Nowadays, the ICT sector in Ireland is facing its most challenging period in its history, a trend related to an already losing lower-end technologies to low cost locations overseas. In order to survive and surpass the global economic downturn, and ensure that Ireland remains a worldwide leader in ICT, the Irish government has recognised the need to immediately launch a major policy orientation (A strategic vision for the ICT sector in Ireland 2002-2005) with a range of actions to promote an adaptation to the changing circumstances. In this political agenda several priorities have been identified, among the key recommendations are the following ones: - an expansion of ICT related research within third level institutions and the establishment of a national research institute for ICT; - increase second level student numbers (male and female) taking science and mathematics by 25% over the next five years; - broadband access should be provided to all business parks and towns of 3000 people and over by end 2004; - the VAT rate for e-commerce should be brought into line with competitor countries; - increase the number of high potential ICT start-ups by 15-20% on an annual basis over the period 2002 to 2005; - develop a statistical profile highlighting the critical importance of the sector to the Irish economy. (IBEC, 2002) Concluding, for Ireland, and for the same reason for Portugal because of its scale and local market limited size, indigenous ICT companies that intended to compete in the global market can be succeed only by addressing specific market niches in which they have specialised knowledge of the application area. In this kind of countries, in order to benefit from the beginning upturn in the ICT market, the more competitive advantage must be no longer based on low labour costs but on the people skills and creativity. 4. Measuring the portuguese ICT sector at a local level According to data provided by Portuguese Ministry for Social Security and Work11, in 2000 Portugal had 3221 ICT establishments (occupying 63090 employees), corresponding to 1% of the portuguese total establishments. 8

TABLE 4. Share of ICT establishments and employees in the portuguese economy Establishments 2000 (nº) % Employees 2000 (nº) Portugal - total 311,317 100 2,688,477 Portugal – ICT sector 3,221 1,03 63,090 Source: data collected in Labour, Employment and Professional Training Statistics Department, 2003

% 100 2,35

Figure 2 represents the portuguese municipalities by crossing two main indicators: the relative importance under the ICT total establishments and the establishments dimension in terms of employees. Analysing the map we may conclude that this economic activity sector has a pattern of diffuse localization, with ICT establishments all over the country. Despite this general trend there are 27% of the total municipalities without ICT establishments which are specially located in the less developed areas, particularly inside the following NUT’s III: Douro, Pinhal Interior Norte, Pinhal Interior Sul and Alto Alentejo. This territorial diffusion is illusionary, since almost all ICT establishments (including the biggest ones in terms of employees) are located in the metropolitan areas and some middle-sized cities. To demonstrate better this obvious concentration throughout the municipalities where there are the best levels of road accessibility, we should mention that around 60% of all municipalities with ICT establishments have less than 0,1% of portuguese total ICT establishments12. To exemplify this general trend we can say that around 42% of those establishments are located only in four municipalities, three of them belong to Lisbon Metropolitan Area (Lisbon, Oeiras and Sintra) and the other to Oporto Metropolitan Area. Table 5 also shows some important but less relevant municipalities outside the two portuguese metropolitan areas, it is the case of Braga, Leiria, Coimbra and Aveiro. TABLE 5. The TOP 15 portuguese municipalities in terms of ICT establishments ICT establishments (2000) No % Portugal - ICT sector 3,221 100 Lisboa 757 23,5 Porto 260 8,1 Oeiras 217 6,7 Sintra 116 3,6 Amadora 92 2,9 Braga 76 2,4 Cascais 76 2,4 Matosinhos 73 2,3 Leiria 73 2,3 Almada 72 2,2 Coimbra 64 2,0 Loures 59 1,8 Vila Nova de Gaia 58 1,8 Aveiro 57 1,8 Maia 53 1,6 TOP 15 municipalities 2103 65,3 Source: data collected in Labour, Employment and Professional Training Statistics Department, 2003

9

10

Portuguese ICT sector is mainly composed by micro-firms, over 70% of ICT establishments have less than 9 employees, and only 16 companies have over than 500 employees (Chart 2). This small dimension has been reinforced over the recent years. Between 1995 and 2000 only the smallest firms have increased in number. It is an economic sector with a low intensive use of workers, as the majority of the portuguese firms. CHART 2. Portuguese ICT establishments according to the number of employees (1995-2000) 80,0

60,0

1995 40,0

2000

20,0

0,0 1 to 9

11 to 49

50 to 99

100 to 499

o ver 500

N o o f e m plo ye e s by e s t a blis hm e nt

S ource: data collected in Labour, Employment and Professional

Training Statistics Department, 2003

ICT sector is nowadays the one of the most dynamic of the portuguese economic branches (Table 6). If between 1995 to 2000 the establishments growth rate for the all tertiary activities was 39,9%, in ICT sector that rate was 87,3%13. Such expressive growth was only comparable with the one recorded on the Real estate, renting and business activities (86,4%) and construction (81,4%). TABLE 6. The recent evolution of establishments and employees in the different portuguese economic sectors14 1995 Portugal - Total Portugal – ICT sector Primary sector Secondary sector Tertiary sector Agriculture, hunting and forestry Fishing Mining and quarrying Manufacturing (*) Electricity, gas and water supply Construction Wholesale and retail trade; repair of motor vehicles, motorcycles and personal and household goods (*) Hotels and restaurants Transport, storage and communication (*) Financial intermediation Real estate, renting and business activities (*) Public administration and defence; compulsory

Establishments 2000 95/00 (%)

1995

Employees 2000

95/00 (%)

223,356 1,720 8,304 64,536 150,516 8,018 286 992 40,945 658 21,941 80,411

311,317 3,221 10,733 89,977 210,607 10,430 303 1,154 48,404 619 39,800 102,490

39,4 87,3 29,3 39,4 39,9 30,1 5,9 16,3 18,2 -5,9 81,4 27,5

2,230,451 56,834 44,326 1,090,591 1,095,534 38,595 5,731 13,952 844,980 19,596 212,063 450,890

2,688,477 63,090 50,768 1,186,234 1,451,445 46,100 4,668 15,081 848,090 16,926 306,137 556,935

20,5 11,0 14,5 8,8 32,5 19,4 -18,5 8,1 0,4 -13,6 44,4 23,5

24,138 7,424 5,807 14,896 181

33,282 11,993 7,761 27,766 261

37,9 61,5 33,6 86,4 44,2

143,987 136,667 86,812 125,638 2,496

186,436 157,136 84,779 242,391 3,780

29,5 15 -2,3 92,9 51,4

11

social security Education 2,674 3,494 30,7 39,586 51,039 Health and social work 6,776 11,264 66,2 61,512 102,484 Other community, social and personal service 8,205 12,292 49,8 47,921 66,448 activities Private households with employed persons Extra-territorial organizations and bodies 4 4 0,0 25 17 Source: data collected in Labour, Employment and Professional Training Statistics Department, 2003

28,9 66,6 38,7 -32,0

The ICT employment had a much slower growth comparing to the establishments dynamic. However in 2000 it represented almost 2,5% of portuguese employees, which is more people employed than in whole portuguese primary sector. Nevertheless this positive dynamic employment should be looked with more detail, once it hides some dramatic and significant joblessness in some municipalities. Figure 3 illustrate that during the recent years it was recorded a decrease of this kind of employment in 53 municipalities. Valongo, Sintra and Castelo Branco are the three municipalities with the biggest lost, each one of them loosing over a thousand employees just in a period of five years. The ICT employment recent evolution is characterized by a gradually and clear super-concentration in the two metropolitan areas (Table 7). All the municipalities with huge increase of ICT employees are located in Lisbon Metropolitan Area (the most dynamic is Oeiras with almost 5 thousand new employees just in five years) or Oporto Metropolitan Area or nearby (Ovar, an adjacent municipality, is the most dynamic with around 1300 new employees). This territorial agglomeration can also be confirmed by the large employment decrease on all non metropolitan middle-sized cities, which had in 1995 a significant number of ICT employees (Braga, Guimarães, Guarda, Castelo Branco, Aveiro, Coimbra and Faro), the only exception is Évora where the ICT activities created almost 400 new jobs. TABLE 7. Municipalities with a growth or a decrease higher than 200 ICT employees, between 1995 and 2000 Municipalities with a growth of ICT Number of new Municipalities with a decrease of Number of ICT employees ICT employees ICT employees joblessness Oeiras + 4,863 Valongo - 1473 Seixal + 1,541 Sintra - 1271 Ovar + 1,314 Castelo Branco - 1196 Lisboa + 1,150 Guarda - 898 Porto + 1,009 Santo Tirso - 853 Vila do Conde + 682 Braga - 514 Palmela + 563 Gondomar - 373 Évora + 389 Setúbal - 359 Vila Nova de Gaia + 205 Amadora - 310 Maia - 305 Loures - 294 Guimarães - 273 Source: data collected in Labour, Employment and Professional Training Statistics Department, 2003

As a conclusion we can say that agglomeration of ICT employment is higher than the territorial agglomeration of ICT establishments: 15 municipalities collect a little more than 65% of the total ICT establishments (table 5), and the same number of municipalities grouped over than 77% of the total ICT employment (table 8).

12

13

TABLE 8. The TOP 15 portuguese municipalities in terms of ICT employees ICT employees (2000) No % Portugal – ICT sector 63,090 100 Lisboa 17,033 27,0 Oeiras 6,092 9,7 Braga 4,541 7,2 Porto 4,336 6,9 Seixal 2,352 3,7 Palmela 2,228 3,5 Évora 1,998 3,2 Cascais 1,772 2,8 Ovar 1,525 2,4 Amadora 1,468 2,3 Sintra 1,418 2,2 Vila Nova de Gaia 1,134 1,8 Maia 1,006 1,6 Matosinhos 1,005 1,6 Almada 940 1,5 TOP 15 municipalities 48,848 77,4 Source: data collected in Labour, Employment and Professional Training Statistics Department, 2003

Table 8 shows also the national supremacy of Lisbon, this city concentrate over than a quarter of all portuguese ICT employment, while Oporto, the second portuguese city in population and also in ICT establishments, is only fourth in terms of ICT employees, supersede by Oeiras and Braga. Lisbon through its status as the capital city is also the leading financial and administrative centre and concentrates a large and mixed supply of advanced services, with the head offices of the largest domestic and multination firms being found there. Oporto productive structure is not so rich or diverse in terms of services, and its industrial profile is more traditionally based, but strongly export orientated. Évora is the only municipality present in this Top 15 that is located outside the two metropolitan areas15. Its national importance results not only on the positive dynamic of new ICT jobs, but also from its relative position in terms of portuguese total ICT employment. The key factors to explain the prominence of Évora and Oeiras municipality in the portuguese ICT sector In Évora there are located two important companies producing electronic components. Both of them result directly from the Siemens strong investment, that begun in 1969 with the construction of a manufacturing facility for communications products and the admission of around one thousand employees. In 1999 occurred the purchase of the relay production unit in Évora by Tyco International, with the continuing manufacture of electro-mechanical relays for the automobile industry. This firm became the top plant in Europe for manufacturing automotive relays, after high-tech manufacturing equipment and a third fully-automatic production line were added. At another site in Évora, in 1997 Siemens build a new plant (EPCOS) for the production of tantalum chip capacitors with joint-venture partner Matsushita. The success of this firm is consequence of the use of this chip capacitors in automotive industry and computer industries, but also by the increased demand of this electronic components to produce cellular telephones. Initially Eastern Europe was also a consideration for the site of the new tantalum plant, however, the distribution infrastructure already established by Siemens electromechanical plant was a key point in the choice of Évora. Sources: THUERMER, K. E. (1997). Siemens builds on long history in Portugal, to the benefit of both [http://www.supplychainbrain.com/archives/10.97.siemens.htm?adcode=5] accessed 19-10-2003 [http://www.ec-central.org/magazine/PDF/art_2_jan_feb_00.pdf] accessed 11-11-2003 Oeiras attractive capability for the new ICT companies is strongly related to the Taguspark, established in the municipality in 1992. This is the oldest and largest science park in Portugal, which had in 2002 about 5,000 people employed in 190 companies The main factors that can be identified as being critical to its successful operation are: the location near Lisbon and close to the motorway network around the capital, the infrastructures developed by the

14

local authority covering several aspects of the development of Taguspark and the surrounding area, the combination of public and private sector partners and the developing of Taguspark image as a prestigious location. The first priority was to attract large ICT company ‘anchor’ tenants, as Portugal Telecom or the call centre and IT operations of a major bank (BCP). After that phase the emphasis switched to attracting smaller technology-based companies. Since 1999 the business incubator started operations, and nowadays there are some 120 companies being assisted, being ICT the main activity, but there are also electronics, energy-related, biotech and other new enterprise projects, some of them with strong links with academic institutions. Source: CENTRE FOR STRATEGY & EVALUATION SERVICES (2002). Benchmarking of Business Incubators [http://europe.eu.int/comm./enterprise/entrepreneurship/support_measures/incubators/Portugal_case_study_2002.pdf] accessed 10-11-2003

ICT is a very diversified sector containing 17 different economic activities, that can be grouped into four main typologies: ICT-industry; ICT-wholesales; ICTtelecommunications and ICT-business activities. It is the ICT-business activities (in particular software consultancy and supply) that collect the majority of those companies (Chart 3). Nevertheless we have a different reality in terms of employment. We can affirm that in Portugal ICT-telecommunication and ICT-industry (especially manufacture of electronic valves and tubes and other electronic components, one of the most important activities of the portuguese automotive components industry) employed the majority of ICT-employees. It is the manufacturing activities that collect the highest number of employees by company. The enormous growth of ICT establishments presented above is not justified by the appearance of new ICT manufacturing activities (Chart 4). Indeed the whole ICT industry loosed 33 establishments between 1995 to 2000. All other ICT activities, excepting Data Processing, have increased their number, especially wholesaling of machinery, equipment and supplies (from 625 to 1013 establishments), as well as telecommunications activities (from 201 to 662 new establishments, with the highest jump between 1996 and 1997).

(%)

Chart 3. Activities included on portuguese ICT sector, 2000 45,0 40,0 35,0 30,0 25,0

share of ICT total establishments, % (2000)

20,0

share of ICT employees, % (2000)

15,0 10,0 5,0

ICT - manufacturing

725

Cae 726

724

723

722

721

642

7133

333

5164

323

33203

322

321

313

3002

3001

ICT - business activities

ICT - wholesales

ICT - telecommunications

ICT - manufacturing

0,0

ICT - business activities ICT - telecommunications ICT - wholesales

Source: Data collected in Labour, Employment and Professional Training Statistics Department, 2003

15

Chart 4. Recent evolution of ICT establishments (1995-2000)

1000

Number of establishments

800

1995 1996

600

1997 1998 1999 2000

400

200

0 3001

3002

313

321

322

323

33203

333

5164

642

7133

721

722

723

724

725

726

Activities included in the ICT sector

Source: Data collected in Labour, Employment and Professional Training Statistics Department, 2003

The effect of automobile industry in the progress of portuguese ICT activities Since the early 60’s with the compulsoriness of fitting up the automobiles for the internal market, and after that with the attraction of Renault companies, it started the development of portuguese automobile components industry. However the consolidation of this economic sector was only achieved with AutoEuropa company, created in Palmela municipality in 1991, from a joint venture between Ford and Volkswagen16. This investment in Palmela is responsible for a volume of production around 130 thousand vehicles/year and have changed the way firms relate themselves in a network, with a larger impact on productive structure in Portugal17. These new methods are based on a high quality guarantee and on a very efficient logistic system that allows frequent deliveries (from components suppliers) to a supplier-park in close proximity to the assembly plant, which allows a sequential just-in-time production. The AutoEuropa has been responsible for a multiplicative effect on the emerging of several important firms of automobile components suppliers. Some of the more important ICT-manufacturing activities, like manufacture of electronic components, manufacture of cables or manufacture of radio receivers, belong to the network of institutions and firms throughout Portugal that have developed new forms of co-operation with the automobile industry. The manufacture of electronic components located in Portugal produce almost exclusively for the automobile industry. Some examples of multinational companies present in Portugal are: Delphi (belongs to General Motors), Yazaki Saltano, Grunding, Siemens, Preh, Roederstein, United Technologies Automotive or Ford Electronics. The manufacture of cables is a very dynamic activity, and allows Portugal to be known as the biggest European producer and exporter of cables. Cablinal, Lear and Leonische are the main cable producers present in Portugal. The national producers of radio receivers are not exclusively dedicated to the automobile industry and export the majority of their goods. The Portuguese production of auto-radios is concentrated in only four firms, three of them are placed in Braga District and belong to three biggest european multinationals: Bosh, Grunding and Ford Electronics. Sources: CHORINCAS, J. (2002). O Cluster Automóvel em Portugal, Lisboa: Ministério das Finanças – Departamento de Prospectiva e Planeamento, 63 p. MONIZ, A. B.; KRINGS, B.-J.; HOOTEGEM, G. V. (2002). Technological practices in the European Auto Industry: exploring cases from Belgium, Germany and Portugal [http://www.univ-evry.fr/labos/gerpisa/rencontre/S05Moniz.pdf] accessed 10-11-2003

Analysing the territorial distribution of ICT employment in the four main ICTtypologies we can now discover some significant differences (see table 9 and figure 4). First of all if in the portuguese middle-sized cities are located the majority of ICTmanufacturing employment, for the other three ICT typologies it is clear the agglomeration effect from the biggest cities prevail. 16

17

TABLE 9. Spatial distribution of ICT employment by the four main typologies ICT employees (2000)

ICT sector (total)

ICT manufacturing

ICT wholesales

ICT telecommunications Nº (%) 19,906 100

ICT business activities Nº (%) 14,385 100

Nº (%) Nº (%) Nº (%) Portugal (total) 63,090 100 19,212 100 9,527 100 6 municipalities containing cities 28,618 45,4 4,854 25,2 4,403 11,484 7,877 46,2 57,7 with more than 100.000 inhabitants 33 municipalities containing middlesize cities, more 17,456 27,7 8,703 2,443 25,6 3,680 18,5 2,630 45,2 than 25.000 and fewer than 100.000 inhabitants 269 other 17,016 27,0 5,715 29,7 2,681 28,1 4,742 23,8 3,878 municipalities Source: data collected in Labour, Employment and Professional Training Statistics Department, 2003

54,8

18,3

27,0

As we already shown the ICT is an economic activity sector with a pattern of diffuse localization, now we can state that this reality is overall a result of the spatial distribution of telecommunication services. But this spatial distribution respond to a double and inverse tendency, on one hand we have more than 57% of all telecommunication services in the 6 portuguese biggest cities, and on the other hand is the ICT activity present in the highest number of municipalities (much related to the diffusion of establishments to sell and assist mobile phones). Moreover in the highest number of municipalities, telecommunication join up the majority of local ICT workforce (see figure 5). In opposition there is a strong concentration of ICT manufacturing activities. As figure 4 illustrate 243 of 308 total municipalities do not have this kind of activities (for instance none of these establishments are placed in Azores Islands). Despite a not so evident territorial cleavage the ICT wholesales and the ICT-business activities are also badly represented on the more rural and less developed areas. For instance, in 176 municipalities there are no ICT business activities, particularly outside the most dynamic area that extends from litoral boarder between Viana do Castelo and Setúbal Peninsula. TABLE 10. Share of ICT sales volume in the portuguese economy Enterprises 2000 (nº) % Total sales 2000 (million escudos) Portugal - total 268,701 100 45,546,049 (around 227,2 billion EUR) Portugal – ICT sector 2,431 0,9 2,705,916 (around 13,5 billion EUR) Source: data collected in Labour, Employment and Professional Training Statistics Department, 2003

% 100 5,94

The ICT sector with 0,9% of all portuguese enterprises18 is responsible for a sales volume around 13,5 billion EUR, which is almost 6% of the all national sales volume. Comparing with other sectors we can conclude that ICT have a very high productivity rate, to exemplify we can say that despite the recognised tourism importance in Portugal the ICT sales volume is twice the amount recorded by tourism activities, particularly hotels and restaurants.

18

19

TABLE 11. The recent evolution of enterprises and total sales in the different portuguese economic sectors 1995/2000 (%)

Enterprises % under the portuguese total enterprises (2000)

Total sales 1995/2000 % under the (%) portuguese total sales (2000)

Portugal - Total 39,8 100 54,3 Portugal – ICT sector 82,8 0,90 170,2 Primary sector 30,7 3,74 44,8 Secondary sector 42,1 31,56 39,7 Tertiary sector 39,2 64,70 64,1 Agriculture, hunting and forestry 31,5 3,64 48,3 Fishing 6,6 0,10 10,3 Mining and quarrying 18,9 0,36 8,4 Manufacturing 19,5 17,01 28,6 Electricity, gas and water supply 43,8 0,04 28,7 Construction 85,3 14,15 101,7 Wholesale and retail trade; repair of motor 25,7 31,57 63,4 vehicles, motorcycles and personal and household goods Hotels and restaurants 36,2 11,50 76,6 Transport, storage and communication 70,9 3,53 73,1 Financial intermediation 43,0 0,58 39,6 Real estate, renting and business activities 83,2 8,80 97,5 25,0 0,08 159,2 Public administration and defence; compulsory social security Education 38,4 1,07 158,2 Health and social work 65,8 3,46 138,1 Other community, social and personal service 50,1 4,11 115,5 activities Private households with employed persons Extra-territorial organizations and bodies 0,0 0,00 -100,0 Source: data collected in Labour, Employment and Professional Training Statistics Department, 2003

100 5,94 1,08 35,37 63,55 1,00 0,08 0,38 24,09 2,94 7,96 36,65 2,33 6,02 11,00 4,98 0,01 0,55 0,92 1,10 0,00

In order to identify some clustering tendencies shaping this economic sector, figure 6 illustrates the portuguese municipalities where are located the high number of different ICT activities (from an universe of 17 branches), in addition we also signalised the municipalities with a location quotient higher than 1 (in terms of ICT employment) as well as more than 2% of the total portuguese ICT sales volume. TABLE 12. The TOP 15 portuguese municipalities in terms of ICT sales volume

Portugal – ICT sector Lisboa Vila do Conde Oeiras Évora Seixal Braga Palmela Amadora Matosinhos Porto Cascais Maia Loures Vila Nova de Gaia Sintra

Total sales of ICT enterprises (2000) (millions escudos) 2,705,916 (around 13,5 billion EUR) 1,108,108 339,685 308,568 274,237 137,270 82,982 77,724 58,532 36,257 35,576 28,774 25,991 24,994 22,485 18,672

% 100 41,0 12,6 11,4 10,1 5,1 3,1 2,9 2,2 1,3 1,3 1,1 1,0 0,9 0,8 0,7

TOP 15 municipalities 2,579,855 95,3 Source: data collected in Labour, Employment and Professional Training Statistics Department, 2003

20

The municipalities with the highest levels of ICT diversity and the highest rates of employees in this kind of activities, can be aggregated in two groups, each one with four municipalities. The first one is composed by those which have more than 10% of total portuguese ICT sales volume, they are: Lisbon, Vila do Conde, Oeiras and Évora. The other one is composed by those which have more than 2% but less than 10% of total portuguese ICT sales volume, they are: Seixal, Braga, Palmela and Amadora. Concluding, the composition of the ICT sector varies considerably across portuguese territory. Patterns of specialization and comparative advantage seems to suggest the presence of economies of agglomeration in some metropolitan municipalities. 5. Conclusion The portuguese ICT sector is very heterogeneous in its contents. If it is essentially composed by micro-firms and this small dimension has been reinforced during recent years, there are also several huge companies presented on TOP portuguese companies. As a result of the limited size of their local market, some of these portuguese ICT firms with growing up ambitions, have already started a successful strategy of internationalisation, with economic profits as well as establishing new cooperation links. The main foreign destination markets are public administrations and local telecommunication enterprises, located in countries with some cultural and geographic proximity: portuguese speaking african countries (Angola and Mozambique), EU countries (some of them in eastern Europe), and Brazil (Valente, 2002). As well as portuguese ICT enterprises are presented in other markets, there are also some multinational ICT manufacturing companies that split off key functions throughout Portugal, specially firms of components suppliers to the automotive industry, which is a major use of ICT in Portugal. This explains that the relative specialisation of some municipalities in ICT production depends largely of foreign affiliates, for instance Évora with Tycos Electronics and Vila do Conde with Infineon Technologies. The importance of the ICT sector within portuguese economy grew expressively over the last years. It has been a major source of establishments growth, much more meaningful than its contribution to employment growth which have been leading by ICT services, while employment in ICT-manufacturing follow a decline trend. In part as a result of being composed by emergent activities, ICT is the portuguese economic activity sector more dynamic in terms of appearance of new establishments. Nevertheless our geographic detailed analysis allows us to say that this global positive dynamic hides some dramatic and significant joblessness in some municipalities, with strong economic and social impacts in local communities. Throughout Portugal this economic activity sector has a pattern of diffuse localization, essentially as a result of the dispersion of smallest telecommunication establishments, but without relevance on the main national figures of the sector. A strict analysis points out that it is an urban activity sector, and gradually becoming a metropolitan activity sector. In Portugal the dimension of two metropolitan areas enable the ICT establishments to benefit from a variety of agglomeration factors, like population, businesses, employment, R&D institutions, high qualified workforce, infrastructures and transport services. It is progressively recognised that growth of economies is in some way fuelled by ICT adoption, since it has altered the way businesses and consumers interact. Thereby, more important than analyse if ICT sector already accounts for a relative share of portuguese total volume sales, is to understand that it can make a relative large contribution to growth and productivity performance to the rest of the economy. To achieve this goal there is the need to generalise the access to modern means of 21

information and knowledge and enable firms to efficiently and profitably use these new technologies. However the local unavailability of ICT businesses activities in portuguese less developed and peripheral areas, difficult in some way the adoption of outsourcing tasks into services capable to process information and communication by electronic means, which may have some negative impacts once ICT are not just innovative tools but are also vehicles of further innovation. If we know that the development of ICT activities follows a demand-pull model, we can expect that portuguese non-urban firms are much more slower in the progressive learning process about how to readapt their organizations into the ICT challenges. In these municipalities, various policies and efforts are needed to help in the rapid diffusion of ICT and to get companies to adopt them.

Acknowledgments This essay is integrated on my developing doctorate project, and it draws on joint work with Profª Iva Pires who had decisively contributed to shaping my views of this subject, besides all the improved comments in the revision process. I have also had the benefit of excellent conditions, provided by Mrs. Ermelinda Lopes and Dr. Artur Simões, during the large period of databases fulfilling spend on the Labour, Employment and Professional Training Statistics Department (Ministry for Social Security and Work). References BELL, D. (1973). The Coming of Post Industrial Society, New York: Basic books CAIRNCROSS, F. (2001). The Death of Distance 2.0. How the Communications Revolution Will Change Our Lives, London: Texere CENTRE FOR STRATEGY & EVALUATION SERVICES (2002). Benchmarking of Business Incubators [http://europe.eu.int/comm./enterprise/entrepreneurship/support_measures/incubators/Portugal_case_stud y_2002.pdf] accessed 10-11-2003 CHORINCAS, J. (2002). O Cluster Automóvel em Portugal, Lisboa: Ministério das Finanças – Departamento de Prospectiva e Planeamento, 63 p. DOBBINS, T. (2001). Job losses hit information technology sector [http://www.eiro.eurofound.ie/2001/08/Feature/IE0108101F.html] accessed 31-10-2003 DORDICK, H. S.; WANG, Georgette (1993). The Information Society. A Retrospective View, Newbury Park: SAGE Publications EUROPEAN COMMISSION (2002). Information society statistics. Data 1996-2001, Luxembourg: Office for Official Publications of the European Communities, 107 p. FORAY, D.; LUNDVALL, B.-A. (1996). “The Knowledge-Based Economy: From the Economics of Knowledge to the Learning Economy”, in OECD Documents. Employment and Growth in the Knowledge-based Economy, Paris: OECD Publications, pp. 11-32 FULLER, E. C.; SOUTHERN, A. (1999). “Small firms and information and communication technologies: policy issues and some words of caution”, in Environment and Planning C: Government and Policy, vol.17, pp. 287-302 GARDIN, O. (2000). A definition of ICT manufacturing products. A proposal for discussion (DSTI/ICCP/IIS(2000)2), Paris: OCDE, 33 p. GREEN, R.; GALWAY, N. (2002). Where does knowledge come from in the knowledge-based economy? 22

[http://66.218.71.225/search/cache?p=ireland+ict+sector&ei=UTF8&url=TR4JkKPg_LMJ:www.irs…/2.pp] accessed 31-10-2003 HALTIWANGER, J; JARMIN, R. S. (2000). “Measuring the Digital Economy”, in BRYNJOLFSSON, E. and KAHIN, B. (eds.), Understanding the Digital Economy, Massachusetts: The MIT Press, pp. 13-33 IBEC (2002). Strategic Vision. The ICT sector in Ireland 2002 to 2005, Dublin: Irish Business and Employers Confederation MACHLUP, F. (1962). The Production and Distribution of Knowledge in the United States, Princeton: Princeton University Press MAY, C. (2002). The Information Society. A sceptical view, Cambridge, Polity Press MONIZ, A. B.; KRINGS, B.-J.; HOOTEGEM, G. V. (2002). Technological practices in the European Auto Industry: exploring cases from Belgium, Germany and Portugal [http://www.univ-evry.fr/labos/gerpisa/rencontre/S05Moniz.pdf] accessed 10-11-2003 MOULTON, B. R. (2000) “GDP and the Digital Economy: Keeping up with the changes”, in BRYNJOLFSSON, E. and KAHIN, B. (eds.), Understanding the Digital Economy, Massachusetts: The MIT Press, pp. 34-48 OECD (1981). Information Activities, Electronics and Telecommunication Technologies, Paris: OECD Publications OECD (2002). Measuring the Information Economy, Paris: OECD Publications, 93 p. PORAT, M. U.; RUBIN, M. (1977). The Information Economy, Washington DC: Government Printing Office QUAH, Danny (2000). “Characteristics of the new economy: Stability, Growth, and Polarisation”, in EUROPEAN COMMISSION, Report of the Workshop on 6-7 April 2000: The New Economy of the Global Information Society. Implications for Growth, Work and Employment, Brussels: European Commission/ Information Society Directorate-General, pp. 31-34 SOETE, L. (2000). “A Economia Baseada no Conhecimento num Mundo Globalizado”, in RODRIGUES, M. J. (ed.), Para uma Europa da Inovação e do Conhecimento. Emprego, Reformas Económicas e Coesão Social, Oeiras: Celta, pp. 3-31 THUERMER, K. E. (1997). Siemens builds on long history in Portugal, to the benefit of both [http://www.supplychainbrain.com/archives/10.97.siemens.htm?adcode=5] accessed 19-10-2003 VALE, M. (1999). Geografia da Indústria Automóvel num Contexto de Globalização – Imbricação Espacial do sistema AutoEuropa, Dissertação de Doutoramento VALENTE, T. (2002). Internacionalização do sector das TIE, Lisboa: ANETIE-Associação Nacional das Empresas das Tecnologias de Informação e Electrónica, 34 p. WORKING GROUP of STATISTICAL EXPERTS (2001). Measuring the knowledge-based economy: Global and regional developments [http://www.unescap.org/stat/cos12/wgse12/wgse12-05.htm] accessed 22-01-2003

Notes 1

The productivity paradox asks how productivity growth could have slowed during the 1970s and 1980s despite of technological improvements in computers and related ICT equipment. Moulton (2000) emphasizes the fact that the productivity slowdown was not limited to the United States, but was seen broadly across industrialized countries. 2 A major review of the currently used classifications of economic activities ISIC (International Standard Industrial Classification) is only foreseen for 2007 (Gardin, 2000). There is a general agreement that ISIC needs urgent improvement in respect of information content industries, ever since it does not have a class 23

for publishing leisure software and data base activities class is poorly defined (Working Group for Statistical Experts, 2001). 3 Services who “enable the function of information processing and communication by electronic means” (Working Group for Statistical Experts, 2001, p. 6). 4 The OECD´s activity-based definition of ICT sector was established in 1998. In April 2002 it was decided that although this definition gives only a first approximation of the ICT sector, it will be only subject to reconsideration at a latter date and in the context of the major revision of ISIC (OECD, 2002). 5 It can be seen as another critic to the three sector model. Those critics have started with the reclaiming proposal presented initially by Machlup (1962). This earlier approach argues that the three sector model should be update, with the subdivision of the tertiary activities through the addition of a fourth occupational sector to group all people holding knowledge-producing occupations. 6 For a good synthesis, with a retrospective view that compares these studies and its contribution to understand better how this informatization is shaping our society, see Dordick and Wang (1993). 7 Based on the comparison of the most expressive knowledge or information industries done by Dordick and Wang (1993) and here updated with some more recent reflections on this subject. 8 Cairncross (2001) highlights the two changes that are driving through the world economy: technological advance in computing and communications, and the fall in barriers to trade and investments. 9 The major companies in the Irish ICT sector, according to the number of workers as of January 2001 are: Dell Computers (5700 workers); Intel (4600 workers); Nortel (3500 workers); Compaq (2300 workers); Hewlett-Packard (2200 workers) and Apple (1000 workers), (Dobbins, 2001). 10 The limited size of the home market has forced Irish software companies to adopt an export-oriented, product focussed approach, with significant success particularly in the US. 11 In Portugal the best source of information for keeping track of the changing structures of markets, in terms of the enterprises location, size and workers characteristics, is the Labour, Employment and Professional Training Statistics Department (Ministry for Social Security and Work). This database is annual updating (with answers to compulsory questionnaires) and follows the universe of all portuguese firms, with location references by municipality. 12 This percentage increases to 85% if we consider the municipalities with less than 0,5% of the portuguese total ICT establishments. 13 To understand better this figure we should not forget that it is a sector composed by emerging activities, and it must be expected that the matured sectors do not reveal so high growth rates. 14 The ICT sector is composed by several activities included in the four economic sectors signalised with *. 15 Excluding Ovar, an adjacent municipality of the Oporto Metropolitan Area. 16 Although since 1999 VW assumed the control all over the company. 17 From the total 365 suppliers enterprises of AutoEuropa 44 of them are located in Portugal (Vale, M., 1999, cited in Chorincas, J., 2002) 18 In this database the sales volume is collected by enterprises. One single enterprise may group several establishments.

24