Download Technology Policy and Innovation: Choices and Challenges and more Study notes Literature in PDF only on Docsity!
CHAPTER 3
BACKGROUND AND CONTEXT FOR THE STUDY
3.1 THE THEORETICAL CONTEXT
3.1.1 Introduction
This section, based on the international literature, presents the key findings and controversies of the literature and their relevance to less favoured regions. The analysis is based on four topics, where major ideas are presented i.e. a reference to the names of the most important researchers associated with the field, followed by a brief outline of the major findings in each topic.
The topics are:
- clarification of the basic concepts, which lead to the emphasis in growth theory on technology.
- macro-approaches, which deal with productivity, growth, employment and trade.
- micro-issues, which include mainly the key notions of appropriability and diffusion, as well as conclusions on sectors and size.
- technology and innovation management tools, such as finance and human resources.
Objective 1 regions 1 in the European jargon are regions whose development is laging behind, in that they have a GDP per head which is lower than 75% of the EU average. Objective 6 regions are defined by scarce population density. Both are well ahead of developing countries and regions as they already have some competitive productive activities and a generalised system of education. They merit a different theoretical approach than either developing or developed countries. Thus, the tools and observations from developed countries can only be applied after being tested under LFR conditions. Besides, all observations demonstrate that if technological input and output measures are taken into consideration, the difference is even more striking than GDP:
- GERD is lower than 0.5% of GDP
- BERD to GERD (with few exceptions) is less than 40%
- there is a very limited number of patents granted to national companies (which strongly inhibits empirical research)
- very few radical innovations emerge in these regions.
(^1) Sometimes referred to as ‘less favoured’ regions (LFRs).
The particular features of LFRs that distinguish them from the core populations of the Triad, where concepts and empirical measurements for the economics of technological change are developed, are the following:
The structure of the productive sector is a limiting factor
- manufacturing activities are limited so that agglomeration and spillovers are less obvious
- manufacturing activities are concentrated in traditional sectors, so science based and specialised suppliers industries are the exception, rather than the rule
- productive companies are SMEs, small rather than medium, with very few indigenous bigger ones: this affects the debate about the relevance of the size of companies for innovative behaviour
- in particular, the combination of sectoral specialisation and size eliminates part of the debate on the importance of small dynamic companies
- finally perfect markets are the exception rather than the rule and regulatory frameworks have their own dynamics
The regional culture in relation to technology is different
- uncertainty is traditionally higher than in regions/countries which have benefited from long periods of sustainable growth, thus short termism in company behaviour is very pronounced; this is a good explanation for low propensities to invest in technology
- trust relations among companies are also limited because of the episodic rather than continuous growth, which inhibits long lasting co-operation
- policy making had traditionally very limited resources and state- corporatism reduced funds for structural interventions; in this sense, the CSFs make the case of European LFRs a unique social experiment, where one can study the impact of unexpectedly rapid increases in technology funding
- finally, more often than not in Europe, national governments intervene very strongly and the degree of regional autonomy is limited.
Despite these characteristics, LFRs are now obliged to change their policies because:
- their wage costs have increased and they can no longer outcompete developing countries, so higher specialisation and value added is their only choice
- protectionism has vanished with the opening of the Single Market.
For these reasons technology policy becomes an imperative for their regional development but although the economics of technological change have in recent decades improved our knowledge of their endogenous role for economic development, they still seem to need further work of adaptation for the special case of LFRs.
qualitative advantages that result from adaptation to technological change. In the centre is the behavioural innovator-entrepreneur rather than the optimising innovator. His decisions are based on routines that are proper to him and not symmetric to all innovators with similar characteristics. Technological accumulation is not random, but evolves in a structured fashion (Sahal, Dosi, Pavitt). Similarly to firms, the evolutionary policy maker adapts rather than optimises and his central concern is the innovation system, the operation of the set of institutions within which technological capabilities are accumulated, diversion increases and selection becomes more efficient. In conclusion, the generation of new knowledge is a necessary but not a sufficient condition for economic progress. Technological regimes, determined by opportunity conditions, appropriability conditions, cumulativeness of technological knowledge and the nature of the relevant knowledge base, affect the dynamics of market structure, more than firm size or demand.
Systems of innovation is the most recent of all approaches (Nelson, Lundvall, Edquist, Teubal). The idea that lies behind this approach is that although the firm is the key element of the system of innovation, firms alone cannot promote new knowledge. There is a whole system of interconnected actors, such as firms, knowledge infrastructure and interfaces, as well as institutions that are active in the process of knowledge creation and transfer. At the national level we have national systems of innovation co-ordinated mostly by non-market mechanisms. An efficient national system of innovation may lie behind the economic success of a nation, while less efficient systems explain limited innovative behaviours and economic backwardness. Efficient here does not relate to any kind of optimality, but rather to the ability of the elements of the system to interconnect so that they become an engine of progress. Formal and informal networking are the mechanisms that allow systems to operate. Universities and educational establishments play a crucial role in this system and make complementary contributions to the knowledge produced by firms. Learning takes a central role on how innovation patterns emerge. In some countries, the learning system contributed to such diversities, that more systems than one, may operate in parallel in the same country (Malerba).
The national systems literature has been opposed by authors who consider the regional or sectoral level more appropriate. Although the national system constitutes a natural boundary for many technological systems, it sometimes makes sense to examine regional or local technological systems. In other cases, the technological systems are international, even global. Where the boundaries are drawn depends on the circumstances, the technological and market requirements, the capabilities of various agents, the degree of interdependence among agents etc. (Carlsson and Stankiewics). Thus, the level of analysis and intervention depends on the questions to be solved and the circumstances and is not replicable from one region to the other. Characteristics of the national systems apply more to the formal legal framework and many institutions (like the technical educational system in Germany), while the interaction of agents may be of high importance at subregional level (like in the provinces of Third Italy).
Issues arising for the LFRs
Under in equilibrium theories, policies for LFRs would require positioning on the basis of optimality calculation, under evolutionary and systems of innovations perspectives the exercise becomes more complicated: initial conditions that determine evolution are deficient and path dependencies (David) suggest that a quick change is the exception rather than the rule. Subsidies become less important in their direct influence on correcting market failures and take the form of incentives to trigger institutional change that will in its turn improve the participation in the process of technological change. The influence on propensities to network maybe the most important element of such a policy. Finally by definition these approaches challenge the idea that what is applicable in core regions is automatically applicable to LFRs as well. Regional innovation systems in LFRs have different absorption capabilities and some of them have more rapid rates of diffusion than others. Unless these differences are studied and policies are adapted accordingly, it is likely that the increase in RTDI funding will end up in limited additionality or with no influence at all.
3.1.3 Macro-approaches
3.1.3.1 R&D, productivity and growth
It may be conventional wisdom that RTDI has a major economic impact, but economic research is still far from being able to econometrically capture and quantify returns of RTDI investment under various circumstances, let alone determine their optimal level.
Many attempts have been made to correlate basic research, industrial R&D and other proxies to productivity and economic growth, both at the level of industries and of national economies. The initial impetus goes back to historical case studies (Kuhn, Freeman). More recently a wider block of economic literature has tried to analyse and assess private and social rates of return to R&D investments through measurements of scoring, patenting and using production function models (Grilliches, Lichtenberg, Terlesckyj). On the whole, these studies demonstrate high rates of return for both private and public investments, but coefficients vary substantially across countries and industries and criticism of the reliability and limits of the sources, be it national accounts or patents, inhibits policy implications.
The concepts behind the above average returns of RTDI investments in comparison to tangible investments are spillovers and positive externalities (Romer, Bernstein & Nadiri, Scherer). This means that the level of productivity of one firm or industry does not depend only in its own research efforts, but also on the knowledge accessible to it. Thus, externalities may arise within or outside a specific industry and access to information becomes a relevant feature for positive correlation between RTDI and social returns on investment. Similar models are used to estimate the contribution of universities to industrial productivity in general. R&D spillovers may be
jobs may not match the old ones either with respect to skills or to location. Severe mismatch leads to structural unemployment. Schumpeter (1939) gave a new twist to the whole debate with his conception of ìsuccessive industrial revolutionsî when new technologies were diffusing through the productive system.
The complexity of the relationship between growth and employment has led to a long lasting debate, in particular because of the long term persistent structural unemployment in all OECD countries in the last decade. Innovations may prove to be major sources of growth (Denison, Solow) but there are both generation and destruction effects related to the adaptation of technological change. The relationship between innovation and employment are seldom direct and usually mediated by a number of offsetting factors (Vivarelli) and there have been certain typologies facilitating estimates in earlier periods. The differentiation between product/employment creating and process/labour saving innovations lies in the centre of the traditional analysis, complemented with the role of R&D for employment, as well as multiplier effects (Sylos- Labini). Such effects include multipliers on demand in other industries, real income effects and adjustments in labour markets.
In a systems of innovation approach more detailed distinctions are suggested. A conceptual framework is suggested for the discussion of various types of process and product innovations and their relationship to employment, with the conclusion that process innovations, although labour saving, should not be hindered, as it would have devastating consequences for productivity. Organisational process innovations increasing capital productivity are most relevant from the employment policy point of view, but they are still under research. If product innovations dominate, there is a tendency to increase employment and a reallocation of resources from process to product innovations is expected to have positive effects. In particular, R&D intensive sectors are employment generating and this includes R&D intensive service sectors which are correlated with product innovation (Edquist, Hommen & McKelvey).
Issues arising for the LFRs
The structure of industry in LFRs is rather predisposed to process than product innovations and in that sense there is a danger that adaptation to technical change will increase unemployment in the short term and important policy adjustments are expected to offset these negative effects. But at the same time preventing industry from process innovation will only aggravate problems of competitiveness. Thus, an active human resources policy for adjustment purposes and emphasis on product rather than process innovations result as policy recommendations.
3.1.3.3 Technological Change, International Trade and Foreign
Direct Investments
Beside its effect on economic growth the concept of externalities has strongly influenced the theory of international trade. Empirical measurements relating export patterns to total factor productivity (reflecting technological capabilities) and the introduction of the product life cycle (Vernon) have driven out the traditional theory, which explained trade patterns through resource differences. Although the life cycle theory proved to have certain limitation, the introduction of the linkage of trade, foreign direct investments and uneven patterns of development has triggered severe criticism to the free trade benefits for lagging countries. New trade theories accept that technological differences are major engines of trade. It is now conventional wisdom that differences in technology rather than differences in resource endowments are the determinants of comparative (competitive) advantage (Krugman, Grossman, Helpman). Adding the concepts of externalities and limited access to knowledge these theories conclude that the real income of the leading country (models are constructed for two countries the rich, technologically advanced North and the poor South) is higher than in a world in which the lagging country had full access to the latest technology. Empirical work has confirmed that countries tend to export in industries in which they overproportionally invest in R&D and/or demonstrate a higher share of patents (Gruber, Hirsch, Dosi). The demonstration of the possibility of continuous steady-state growth thus allows for accumulating benefits and increasing disparities through international trade. The empirical work describing the long lasting benefits from clusters (Porter) falls under this category.
Nevertheless, the role of technology diffusion (and for that matter technology transfer) allows also for countries to catch up, as the observed case of European countries and Japan treated originally as latecomers compared to the US in the original Vernon model. Once one takes deliberate investment in imitation into account, the lagging countries tend to play a harder game to catch up. By expanding resources in reverse engineering and gradually creating their own relative RTDI intensities, these countries may create their own strong start up positions and benefit from externalities and the self- reinforcing character of technological advantage.
The problem with new trade theories is that they allow for two radically opposed interpretations: the failure of the assumption that all countries benefit
globalisation tendencies and the absence of any degrees of freedom in reintroducing protectionism. As we have to take the accumulative character of technology and benefits from international trade for granted, the only option of LFRs is to heavily invest in imitation to create their own headstart positions in sectors that are more competitive than their current specialisation, yet not global leaders. Restructuring cannot be avoided, but the models suggest that it is unlikely that such countries/regions will be able to invest in the really ìstrategicî sectors, that allow maximisation of benefits. Investment is not a sufficient condition given the characteristics of technological change. Increasing inputs alone will not ipso facto lead to improved output, thus a better insight into technology and industrial organisation is needed to better understand the process on how to improve the technological level in Objective 1 regions. What they can learn from the new trade theories is the need to pool resources and benefit from improved performance of a few industries, rather than heavily invest in high tech industries, where their relative position starts in disadvantage.
Regarding the new academic literature stressing the rise of different patterns of location of multinational companies where flexibility, networking and performance-based competition emerge as relevant concepts, is whether LFRs are attractive locations for that type of investment. The infrastructure to attract quality-seeking inward investment becomes critical (Cantwell) but it has been observed that certain plants in LFRs gradually acquire and develop research and technical skills, often associated with local problem solving and adaptation to local markets (Hakansson, Amin and Tomaney).
3.1.4 Technology and industrial organisation
Much discussion concentrates in the way markets operate and the relation between industrial organisation and technological change. Market failures (Arrow), i.e. indivisibilities of high initial R&D cost, uncertainties of outcome and the danger of spillovers to competitors lead either to underinvestments in R&D and/or monopoly tendencies. Market failures are not necessarily the same in all types of markets. One may assume that the small size of firms and higher uncertainties in LFRs may have particular effects in that respect.
3.1.4.1 The concept of appropriability
The ability of innovators to commercialise their innovations and make profits out of them is the incentive that leads them to ignore uncertainties and invest in R&D. Thus inventors have every interest to appropriate their knowledge and inhibit its diffusion to their current or potential competitors. If they cannot, they lose interest in investing in R&D. For that reason knowledge is not accessible to all firms, on the contrary, early movers have every interest to keep it for themselves. So, appropriability is an important feature for investing in technological change.
There is no doubt there are differences in appropriability among sectors and types of economic activity, depending on the nature of the technology, the relevance of tacit knowledge (that is the ease of transmitting knowledge) and
barriers to entry in a business. The most effective barriers to imitation are natural ones, that is high cost and long time to reinvent a technology. One should note here that diffusion and appropriability may be seen as a different angle of the same process. While rapid diffusion is necessary for achieving spillovers and thus increasing social returns on investment, companies themselves try to limit free transmission of their knowledge and extend appropriability periods and monopoly rights as long as they can. In this process they donít rely on the natural characteristics of easily appropriable knowledge, but act proactively.
Patent grants (Mansfield, Pakes) are the oldest and the only legally binding way to deal with appropriability. Through the creation of property rights, patents diminish the risk of costless imitation and help create a market for knowledge. The monopoly rights granted permit R&D investors to profit from their innovation long enough not only to amortise it, but ideally to keep their market leadership. Thus, the key questions for the economic literature in this area are related to the optimal duration and coverage of patents in respect to maximising public returns. But patents are not the only means and often not the strongest one to assure appropriability. Many studies have shown that patents are a pertinent means for only a few industries, because of limitations in their effectiveness to protect new knowledge. While not all studies converge to exactly the same conclusions, it is by now conventional wisdom that patents are most widely used for the protection of pharmaceuticals. In organic chemicals, plastics or chemical processes, patents may prove relevant but not always, while finally, in traditional industries, they play a minor role. Other forms of appropriability are more effective, like trade secrets, lead time, first mover advantages or the use of knowledge with generic or specific complementary assets (Levin, Teece). Vertical strategies or targeted alliances are ways used by companies to improve their complementary assets.
Despite firm strategies there are many ways knowledge diffuses partly through scientific personnel mobility, partly through informal networks of employees and partly through deliberate networking which allows companies to share pre-competitive research costs. Spillovers are likely to occur in technologically adjacent sectors and in firms that have their own R&D. The old perception that R&D and licensing or spillovers are alternative ways of accessing the same technology has been reversed. It is now widely accepted that since technologies are cumulative spillovers; they are likely to complement own R&D (Foray, Jaffe). To put in Geroski¥s words, only innovative firms benefit from spillovers.
In terms of economic policy one typical solution to the appropriability problems leading firms to sub-optimal investments are RTDI subsidies or tax exceptions. But judging the appropriate level of subsidies requires policy makers to calculate the gap between public and private rates of return, an exercise which is difficult to do, not least because the gap is likely to vary tremendously across industries (Geroski). More recently, public policy favours co-operative research in order to spread benefits to more industries or create national champions. In some advanced countries a system of sharing R&D among potential competitors has worked (industry induced research
In seeking to understand why sectors differ in the degree to which they engage in RTDI investment and innovative behaviour, explanatory variables such as market demand, technological opportunity and appropriability conditions have been suggested. Evidence from capital goods industries suggests that cycles in the output of capital goods and in capital expenditures by downstream industries, lead cycles in time series on relevant capital goods patents (Schmookler), a pattern which did not apply in chemical industries (Walsh). These studies have opened the technology push-demand pull debate, which is currently researched with sophisticated instruments like price elasticities of demand and distinctions in product and process innovation. Technological opportunity in its turn, suggests that trajectories determine RTDI investments and these trajectories, studied with hindsight by economic historians, may be modelled in different ways (such as elasticity of unit costs in respect to R&D spending) to capture quantitative information on such trends.
More research is undertaken in the sectoral aspects related to technological trajectories and paradigms (Dosi), linking elements of a larger whole which also evolves from an early phase through growth and maturity. The life cycle of such a techno-economic paradigm is composed of a series of interrelated technology systems and the change of techno-economic paradigms leads to transition periods of high turbulence (Freeman and Perez). A new field of research is the connection between technological regimes and sectoral systems of innovation, composed of those firms that are active in the innovative activities of a sector (Breschi and Malerba).
Issues arising for the LFRs
Although explanations about the intensity and direction of technological change among industries differ, measurements suggesting the few sectors dominating R&D activity are unanimous, and these sectors are not well represented in LFRs, or to the extent that they are, they constitute either isolated cases or they represent activities at the end of the production chain. Software may be the only exception.
But, as there are sectoral patterns of innovation and evolutionary changes during periods of paradigm transition, a temporary window of opportunity is available for latecomers (Perez and Soete).
3.1.4.3 The relevance of the size of firms
The relation between innovativeness and firm size goes back at least to Schumpeter, who relates it to market concentration arguments. The dominant idea has been for a long period of time that innovative activity increases more than proportionately with firm size (Galbraith, Scherer). Several explanations have been used: market concentration arguments, better access to finance, high economies of scale in the exploitation of technology, economies of scope in diversified firms and last not least, the relevance of complementary assets. Later, more sophisticated analyses differentiated between the relevance of firm size and business unit size and between the relevance of size in different
industries (Cohen, Mansfield, Soete). It was thus suggested that it is the size of the business unit, rather than the size of the firm that matters and that chemicals are typically the industry where R&D rises with firm size. But counterarguments have also emerged in the literature related to diminishing R&D efficiency with size, because management loses control and/or individual incentives are reduced. Some authors have demonstrated that smaller firms tend to account for a disproportionately high share of innovation in relation to their size and that R&D productivity tends to decline with size (Pavitt, Acs & Audretsch). But these models have been criticised as strongly biased because they seem to be able to get data only for successful SMEs. Measurement difficulties, i.e. criticism to the reliability of the data used expressing formal R&D and underestimating R&D incorporated in other parts of the firm, and the ability of bigger firms to operate more R&D and spread it over a variety of projects inhibits final conclusions about R&D productivity and firm size. But there are at least a few points on which the theory is unanimous: there is hardly any R&D below a certain threshold size and innovation in small firms is usually measured in biased samples.
In the Schumpeterian tradition the majority of studies which examine the relationship between market concentration and R&D, have found a positive relationship (Mansfield), few have found evidence of a negative impact of concentration (Williamson) and others suggested more complex, non monotonic relationships (Scherer). More recently, some authors suggested that causality runs from innovation to concentration and not vice versa (Phillips), since technology has a cumulative character and thus market power, accruing from successful innovations, is not transitory, as Schumpeter originally suggested. But again more detailed analysis shows technology classes and technology maturity to strongly affect the explanatory potential of the argumentation (Utterback).
Going beyond the debate of size and market structure, the recent literature emphasises the dynamics of companies: success and is not a matter of size but of management of technology by companies. Persistent innovators as opposed to opportunistic ones (Malerba, Lissoni, Breschi), entrepreneurs that shift technology to the core of their management process (Coombs, Dogdson), and adhere to the right networks (Best) are more likely to keep international competitive edges. Continuous innovations increase appropriability and allows firms to remain internationally competitive in global markets, either expanding or locked in an SME size limited to niche markets through specialisation.
Issues arising for the LFRs
Although literature in core countries has not resolved the debate about the importance of size for R&D productivity, it may be argued that reduced efficiency is likely to be endemic as firms in LFRs are either too small, if the positive correlation is taken into consideration, or too traditional, if the argument of the above average efficiency of successful SMEs is the most valid.
successful than in the US. Many reasons may be found for this but institutional explanations and the difference between market induced business angels in the US and regulated schemes in Europe, account for much of the difference in performance.
Issues arising for the LFRs
Finance is a key barrier for innovation, but ample finance does not necessarily improve innovative performance. Venture capital is the best suited instrument under certain circumstances, but is has not always worked well in advanced countries. Thus a key issue is to find the best systems of innovation suited to each region, trying to gear the banking sector and the public incentives towards a higher acceptance of innovation finance. But the relevant point is to chose the right financial mechanisms and not to offer more finance.
3.1.5.2 Human resources
Growth theories based on savings rates and capital/output ratios have been overcome by the emerging literature on the role of human capital (Schultz, Becker). All modern economic theories recognise this ìnew assetî for competitiveness and development and most econometric evidence suggests that it is precisely the countries with high skills/high wages that benefit from the highest growth rates (Lundvall). The quality of human capital depends on the whole range of educational services from the primary school to continuous post doc training. A good basic educational system lies behind any technological leapfrogging, but at the same time continuous training is necessary in order to permanently adapt to technological change.
Innovation and training in modern economies are inextricably linked. Their respective strengths are often as much a reflection of industrial performance as a cause of it. The connections are reciprocal and complex. Training not only requires micro-level investment in developing people, but also macro- level investment for creating a training infrastructure, so that external economies can be achieved. Innovation may, however, sometimes precede training as new skills are necessary to operate complex new machinery (Sorge et al .).
A nation which fails to develop skills, risks the inability either to promote innovations or to take advantage of them in the form of technology diffusion or transfer. Investments in research equipment are useless if there are no research skills at all levels. In particular, when the economy picks up in the trade cycle such bottlenecks hamper growth. In high tech sectors even in conditions of recession employers complain that they cannot recruit enough qualified staff. Such skills are particularly short supply at the intermediate skill level. More often than not given the failure of domestic producers to service the home market, imports flow in to further worsen the trade gap (Warner).
It is thus not only the public educational system but more so companies that need to constantly retrain their personnel (Campbell et al.). National systems where companies were willing to bear the cost of restructuring, like in the
Nordic countries (Lundvall) or where the intermediate training has a long and successful tradition as in Germany, have coped better with challenges of technological change and paradigm shifts. Porter explains Japanese, German and Swiss industrial performance by companies participating in educational activities. In Germany every significant company participates in apprenticeship programmes involving local technical schools, whereas in Japan advanced factor creation is mostly within firms. Recent evidence suggests that innovation and management training are most relevant and joint technical, business and social skills are best suited for success. The development of hybrid skills and the motivation of knowledge workers, require a new approach to training which is based on the concept of continuous learning. With rapid changes in technology, shorter product life cycles and more solution-oriented and user needs, many knowledge workers need to upgrade their work skills on a life-time basis (Rajan).
Issues arising for the LFRs
Human capital is so closely interrelated with technological change that it is difficult to separate them. A solid basic education is necessary for regional employment and continuous upgrading and life-long training are necessary to accompany structural adaptation and growth. But the two key issues resulting from recent literature are that:
trained researchers alone may prove insufficient for growth and competitiveness, although they will contribute to the improvement of the research system. Thus, a design based on skill improvement and tailored to productive systems and needs is a prerequisite for technological development.
- Intervention through a system of fine tuned and tailor made incentives can help to mobilise regional productive systems. Thus, the literature suggests certain general guidelines and principles to be adopted universally at the Obj. 1 level, and also specific incentives for the national (change the formal institutional framework) and regional levels (change the type and scope of intervention). Incentives need to be very strong if they are expected to change formal and informal rules, as already pointed out by Hirschmann in early development theories.
Bibliography Abramovitz, M. (1993) ëThe search for the sources of economic growthí Journal of Economic History , 53, 217-43. Acs, Z. J., D.B. Audretsch and M. Feldman (1992) ëReal effects of academic research: Commentí The American Economic Review 82(1), 363-67. Acs, Z., D. Audretsch and M. Feldman (1992) 'Real effects of academic research' American Economic Review 82, 363-7. Amin, A. and J. Tomaney (1995) `The regional development potential of inward investment in the less favoured regions of the European Communityí. In Amin, A. and J. Tomaney (eds.) Behind The Myth of European Union: Prospects for Cohesion, Routledge: London. Arora, A. And A. Gambardella (1991) ëComplementarity and external linkages: The strategies of the large firms in biotechnologyí The Journal of Industrial Economics 38, 361-80. Arrow, K.J. 'The economic implication of learning by doing' Review of Economic Studies 29(3), 155-73. Becker G., (1975) Human Capital, a theoretical and empirical analysis with special references to education , Columbia University Press Bernstein, J. and M. Nadiri (1988) 'Inter-industry R&D spillovers, rates of return and production in high-tech industries', American Economic Review , Papers and Proceedings 78, 429-34. Bernstein, J.I. (1989) 'The structure of Canadian interindustry R&D spillovers and the rates of return to R&D', Journal of Industrial Economics 37(3),315-28. Bernstein, J.I. and the M.I. Nadiri (1988) 'Interindustry R&D spillovers, rates of return and production in high tech industries', American Economic Review Paper and Proceedings 78(2), 429-34. Bernstein, J.I. and the M.I. Nadiri (1989) 'Research and development and intra-industry spillovers: an empirical application of dynamic duality', Review of Economic Studies 56, 249-69. Best, M. (1990) The New Competition , Polity Press, Cambridge. Breschi S., F. Malerba (1997) Sectoral Innovation Systems, in Edquist, Charles (1997), Systems of Innovation , Pinter. London and Washington Camagni, R. (1992) 'Development scenarios and policy guidelines for the lagging regions in the 1990s', Regional Studies , 26, 4, 361-374. Campbell A., Werner M (1992), New Technology, Skills and Management , Routledge, London Cannell W., B. Dankbaar ed. (1996), Technology Management and Public Policy in the European Union, Office of the Publications of the EC and Oxford University Press, New York Cantwell, J. (1989) Technological Innovation and Multinational Corporations , Oxford, Blackwell.
- (1992) 'The Internationalization of Technological Activity and its Implications for Competitiveness' in O. Granstrand, L. Hakanson and S. Sjolander (eds), Technology Management and International Business , Chichester, Wiley. Carlsson B. and Stankiewics R. (1995) On the nature, function and composition of technological systems, in B. Carlsson Technological systems and economic performance: the case of factory automation, Dordrecht: Cluver Cohen, W., R. Levinthal (1989) 'Innovation and learning: the two faces of R&D', Economic Journal 99, 569-96. David, P.A. (1985) ëClio and the economics of QWERTY’ American Economic Review Proceedings 75, 332-7. Dodgson M and R. Rothwell, (1994), The Handbook of Industrial Innovation , Edward Elgar, Cheltenham UK- Brookfield US Denison, E. F. (1967) Why growth rates differ , Washington D.G., The Brookings Institution.
- (1979) Accounting for Slower Economic Growth , Washington D.G., The Brookings Institution.
- (1985) Trends in Americal Economic Growth 1929-1982, Washington D.G., The Brookings Institution. Dosi, G. (1982) 'Technological paradigms and technological trajectories: a suggested interpretation of the determinants and directions of technical change', Research Policy 11.
- K. Pavitt and L. Soete (1990) The Economics of Technical Change and International Trade , Hemel- Hempstead, Harvester-Wheatsheaf.
- C. Freman, R. Nelson, G. Silverberg and L. Soete (eds) (1988 ) Technical Change and Economic Growth , London, Pinter. ++ Drucker, P.E. (1990) 'The emerging theory of manufacturing', Harvard Business Review , May-June, 94-102. Edquist, Charles (1997), Systems of Innovation , Pinter. London and Washington Edquist, C. and S. Jocobsson (1988) Flexible Automation : The Global Diffusion of New Technology in the Engineering Industry , Oxford, Basil Blackwell. and B.-A. Lundvall (1993) 'Comparing the Danish and Swedish systems in Innovation' in R. R. Nelson (ed.) National Innovation System : A comparative study, Oxford, Oxford University Press. Foray, D. and C. Antonelli (1991) 'The economics of technological clubs', Economics of Innovation of New Tecnology 1, 16-27. Freeman, C. (1984) The economics of industrial innovation , London, Pinter.
- (1987) Tecnology policy and economic Performance, London, Pinter.
- and C. Perez (1988) 'Structural crisis of adjustment business cycles and investment behavior' in Dosi et al. (eds), Technical Change and Economic Theory , London, Pinter.
- and L. Soete (1987) Technical Change and Full Employment , Oxford, Basil Blackwell. ++ (Geroski, P. (1991) 'Innovation and the sectoral sources of UK productivity growth', Economic Journal 101, 1483-
- (1992) 'Vertical relations between firms and industrial policy', Economic Journal 102, 138-47.
- (1993) 'Antitrust policy towards cooperative R&D ventures', Oxford Review of Economic Policy 9, 58-71.
- and S. Machin, J. Van Reenan (1993) 'The profitability of innovating firms' Rand Journal of Economics 24 (2), 198-211. Griliches, Z. and J. Mairesse (1983) 'Comparing productivity growth: an exploration of French and U.S. industrial and firm data', European Economic Review 21, 89-119.
