Prometheus: Vol 30, No 1 (2012)

Click to expand.
notes an article is available as an Open Access pdf.
notes an article is free to download.

Editorial
Stuart Macdonald
Pages: 1-3

In March 2010, the Science Policy Research Unit (SPRU) at the University of Sussex held a conference in honour of Nick von Tunzelmann. The Nickfest, as it came to be called, was quite an occasion, as well it should have been: two full days of papers from those who know and admire Nick and his work, many of them once Nick’s colleagues or students. The Nickfest was an entirely fitting way to mark not just Nick’s retirement, but also his contribution of so much to so many. Prometheus is happy to add its mite to the praise of Nick von Tunzelmann in its selection of several papers from the conference that represent the spirit of the Nickfest.

Despite the Prussian name, Nick von Tunzelmann is actually a New Zealander. He began his academic life at the University of Canterbury. Not many people know this, but there is actually a von Tunzelmann Point somewhere in Antarctica, named after a roving relative. Nearly fifteen years at Cambridge followed before Nick fetched up at SPRU in 1984. He is now an emeritus professor there. Technology has always been his subject, how it changes, how it can be transferred, how it emerges and develops into complex systems. Nick has always been – as so many genuine scholars are – an historian: to begin to understand where we are and where we might be going, we must first understand how we got here. Nick’s work always seeks – and generally finds – context.

As economic historian, Nick has done sterling service for years in annoying orthodox economists; indeed, he harries anyone who shackles an argument by obliging it to observe the convenience of convention. He has resisted both the lure of management as an expanding subject in universities, and the temptation to destroy such an easy target with his merciless logic. Those of us who enjoy blood sports have always regretted the latter. In the SPRU of the last century, Nick found an intellectual home, a place where ideas about technological change ricoched from the very walls. What SPRU lacked in efficiency under the leadership of Chris Freeman, it more than made up for in scholarship. The best in the world in science and technology policy flourished in this atmosphere; Nick von Tunzelmann breathed deep and thrived.

The six papers in this issue of Prometheus are not representative of Nick’s work, nor are they intended to be. Rather they are representative of the inspiration and guidance that Nick gave to so many in their own work. SPRU advised the world on science and technology policy, and consequently found visitors and students from round the world arriving at its doorstep to drink from the font. Nick relished the interaction – as he still does – and delighted in relating his own perspectives on change to those of so many others. It was no surprise at all that the conference in his honour involved many dozens of speakers and some hundreds of delegates. It is from this well that we have drawn, aided by Maria Savona, the chief organiser of the event. Steven Henderson from Solent University has undertaken the editing of the collection with all the skill and determination the task requires.

The use of such intellectual property rights as patents and copyright by universities has increased steadily over the last two decades. Birgitte Andersen and Federica Rossi examine what seems to be deviant behaviour: why should public institutions, supported by public money for the public good seek to make private property out of what they produce? The study of intellectual property rights is deep in the SPRU tradition and their paper would have intrigued Keith Pavitt, Nick’s colleague at SPRU. Andersen and Rossi find huge market failure in university use of IPR and advocate the open distribution of academic knowledge rather than its privatisation. This is not the way to the heart of the modern vice chancellor.

Isabel Freitas, Aldo Geuna and Federica Rossi also look at relations between universities and firms. They focus on their collaboration, and find that contractual relations between institutions are much less effective than contractual relations involving individual academics. The observation is pertinent to the tension between the professional and the managerial in the university. Where the professional has given ground – as it has throughout the UK – there may be acceptance that it is the role of the institution to determine the role of the individual. This may satisfy the modern vice chancellor more than it satisfies the firm wishing to deal with an individual rather than an institution. Developing a similar theme, Valeria Arza and Claudia Vazquez, from Argentina, investigate the links between public research organisations and the firms that use the information they produce. They discover that information flows in a variety of ways, and that interaction is most productive when it flows in both directions. As a temple of knowledge the public research organisation is isolated from the outside world.

Two papers concern themselves with small firms of various sorts. Massimo Colombo, Annalisa Croce and Massimiliano Guerini, from the Politecnico di Milano, examine new technology-based firms (NTBFs). They find that NTBF policy serves large firms best in the north of Italy. The small NTBF is neglected. In the south of the country, all NTBFs are neglected. Young, small NTBFs benefit greatly from public subsidy when they get it, but this just does not happen often. Luca Grilli and Samuele Murtinu also ask whether public subsidies really do help the NTBF. The argument is clear: the market fails to appreciate the value, especially the public value, of so much novelty. So, there is apparently a role for public subsidy. However, their discovery is that public subsidy is effective only when it is selective rather than automatic, and when it is directed towards enhancing the R&D performance of the NTBF. In other words, policy makers must know what they are doing if they are to devise programmes capable of distinguishing R&D worthy of public support. Their record in this area is not good. Giovanni Cerulli and Bianca Potì, from the National Research Council in Rome, also explore the Italian situation. Their paper looks at the impact of a specific R&D policy instrument, the Fondo per le Agevolazioni della Ricerca, on industrial R&D and technological output at the firm level. They find, as have many others, that policy seems to benefit large firms rather than small. Now why should this be?

We hope Nick von Tunzelmann will find not only some interest in these papers, but also some of his own interests. We hope he will feel that he had some hand in inspiring not just the area of inquiry, but also the questioning that makes these papers worth reading. Nick will get some satisfaction from that, satisfaction that Nick, the kindest of men, thoroughly deserves.

General Editor

Inefficiencies in markets for intellectual property rights: experiences of academic and public research institutions
Birgitte Andersen & Federica Rossi
Pages: 5-27

ABSTRACT

The formal use of such intellectual property rights (IPR) as patents and registered copyright by universities has increased steadily in the last two decades. Mainstream arguments, embedded in economic theory and policy, advocating the use of IPR to protect academic research results are based on the view that IPR marketplaces work well and allow universities to reap significant benefits. However, there is a lack of evidence-based research to justify or critically evaluate these claims. Building upon an original survey of 46 universities and public research organizations in the United Kingdom, this study analyses the quality of the institutions underpinning the markets for patents and copyright, investigating potential inefficiencies that could lead to underperformance of the IPR system. These include ‘IPR market failures’ with respect to search processes and transparency; price negotiation processes; uncertainties in the perception of the economic value of IRP and the relationship with R&D cost. Further sources of underperformance may include ‘institutional failures’ with respect to enforcement and regulation. Particular attention is paid to the role of governance forms (e.g. alternative types of licensing agreements) through which IPR exchanges take place. We find that a high share of universities report market failures in IPR transactions and that the choice of IPR governance forms matter for the obstacles that are encountered. Given the importance of widely disseminating university research outcomes to foster innovation and economic development, the presence of inefficiencies in IPR markets suggests that such objectives could best be achieved by encouraging open distribution of knowledge, rather than privatization of academic knowledge.

Additional information
Notes

  1. The UKNOW database was developed as part of European Commission research project ‘Understanding the Relationship between Knowledge and Competitiveness in the Enlarged EU (UKNOW)’, Work Package 3.2: An IPR Regime in Support of a Knowledge Based Economy, a project of the EU 6th Framework Programme.
  2. In the following analysis, we use the term ‘proprietary IP’ [or, equally, ‘intellectual property rights’ (IPR)] to identify IP upon which restrictions on use, sharing, copying and modification are enforced by legal means, and ‘non-proprietary IP’ for IP on which some or all of these restrictions are relaxed.
  3. The problem of under-exploitation of IPR is common in commercial firms as well. Rivette and Kline (2000) identified ‘a staggering $1 trillion in [ignored] intellectual property asset wealth’ in the USA, while the PATVAL survey of European inventors found that while 11% of a random sample of European Patent Office patents had been licensed, an additional 7% could have been licensed, but were not, and a study by consulting firm BTG International found that 35% of patented technologies (valued at $115 billion) were ignored by the firms that developed them. A survey of US firms found that more than a third of total IPR inventory rated as available for licensing but unlikely to be licensed (Cockburn, 2007).
  4. Examples are non-software copyrighted materials (articles, reports, books, lecture notes, presentations); software (source level code as well as executable programmes developed by researchers in the course of their research work); materials (synthesized by researchers working in the fields of chemistry and materials); database rights; cell lines; new plant or animal varieties; registered and unregistered designs; photographs and videos; research questionnaires; and finally, tacit knowledge (know-how), which is hard to codify and transfer, but which is nonetheless valuable to third parties (Baghurst et al., 2009).
  5. Throughout this paper, we refer to this sample as ‘UK universities’, for the sake of simplicity. Higher education colleges and public research organizations comprise less than 25% of the sample and of respondents, as evidenced in Table 4.
  6. The number of academic staff and total staff (academic, non-academic, atypical) of the institution (relative to 2007/08), the share of academic staff employed in scientific fields (engineering and technology, medicine and natural sciences, in the same period), and the income of the institution were supplied by the Higher Education Statistics Agency. The year of foundation of the technology transfer office and the number of staff employed (relative to 2007) within were drawn from the HE-BCI survey.
  7. The categories are the following: old universities (founded before the mid-nineteenth century); red brick universities (founded between the mid-nineteenth century and the mid-twentieth century); plate glass universities (founded between the 1960s and the end of the 1980s); former polytechnics (institutions formerly designated polytechnics which changed their status to universities in 1992); and modern universities (founded after 1992, not formerly designated polytechnics).
  8. This is consistent with results discussed by Cockburn (2007) when studying patent licensing deals in the US and Canada: here, only about 10% of survey respondents cited uncertainty about the strength or scope of IP rights, and less than 5% cited other structural issues, such as there being too many parties involved in the negotiation. What really matters is the ability to reach agreement on financial and non-financial terms of the licensing contract; again, this is consistent with results found in our survey.
  9. Universities were not asked to agree with the statements ‘difficulty in finding the best IPR’ and ‘lack of clarity of the IPR document’ with reference to copyright as these obstacles were not considered relevant to the case of copyright.
The governance of formal university–industry interactions: understanding the rationales for alternative models
Isabel M. Bodas Freitas , Aldo Geuna & Federica Rossi
Pages: 29-45

ABSTRACT

This article develops a conceptual framework to explain the economic rationale underpinning the choice of different modes of governance of formal university–industry interactions: personal contractual interactions, where the contract regulating the collaboration involves a firm and an individual academic researcher, and institutional interactions, where the relationship between the firm and the academic is mediated by the university. Although institutional interactions, for numerous reasons, have become more important, both governance modes are currently being implemented. We would argue that they have some important specificities that need to be understood if university–industry knowledge transfer is to be managed effectively and efficiently.

Additional information
Acknowledgements
Comments from participants at the Technical Change: History, Economics and Policy conference held in honour of Nick von Tunzelmann in Brighton in March 2010 are gratefully acknowledged. The authors thank the Piedmont Chamber of Commerce for help with data collection and Barbara Barazza for her support and comments. The authors are indebted to Federico Caviggioli, Cornelia Meissner and Marco Riva for their database creation skills. The UIPIE and PIEMINV databases were created with support from the project IAMAT, coordinated by Fondazione Rosselli. Financial support from the European Commission (FP6) Project NEST-2006-PATH-Cul CID is also gratefully acknowledged.

Notes

  1. This article is an abridged and radically developed version of Bodas Freitas et al. (2011).
  2. Another trend, not explored in this paper, is the progressive increase in the number and importance of the interactions between firms and universities, compared with the relative decline in importance, since the 1990s, of the interactions between firms and public research centres. Such a trend has been noted in many European countries, including Ireland, Denmark, the UK, Iceland, Italy and Hungary, although notably not Germany (Senker et al., 1999).
  3. Academic consulting activities are also often mediated by the university institution, which channels consulting income through its accounts and may apply overheads (Beath et al., 2003; Perkmann et al., 2009). The personal contractual interactions examined in this paper are formal (contract-based) agreements between individual academics and firms, which are different from university-mediated consultancy activities and consultancy based on informal personal relationships.
  4. See the special issues of Industrial and Corporate Change, 2007, 16, 4; Oxford Review of Economic Policy, 2007, 23, 4; and Journal of Economic Behavior & Organization, 2007, 63, 4.
  5. The UIPIE questionnaire was administered in autumn 2008 to a representative sample of 1058 firms in the Piedmont region; we obtained 1052 valid responses (a response rate of 99%). The sample was developed and validated by the local chamber of commerce, which sent out our questionnaires with its quarterly regional economic foresight survey.
  6. The PIEMINV questionnaire was administered in autumn 2009 and spring 2010, to inventors with a Piedmont address who had applied for a patent to the European Patent Office in the period 1998–2005 (about 4000 patents and 3,000 inventors in Piedmont). We obtained 945 valid responses from 2583 questionnaires sent (a response rate of 36%).
Firms’ linkages with universities and public research institutes in Argentina: factors driving the selection of different channels
Valeria Arza & Claudia Vazquez
Pages: 47-72

ABSTRACT

Knowledge flows between public research organisations (PROs) and firms may occur through various channels. Channel selection may have different drivers and effects. Although much research has been carried out on the drivers of firms and researchers to connect with each other, less attention has been paid to the determinants of the selection of different channels of interaction. This research analysis factors driving firms’ selection of different channels of interactions with public research organisations (PROs), both public research institutes (PRIs) and universities (UNIs). The paper estimates bi-variate probit models with sample selection using micro data for 2007 from a representative survey of Argentinean firms. The classification of channels is based on previous research for Latin America and includes four types according to the main goals that firms and public research organisations seek when interacting: traditional, service, commercial and bi-directional channels. We find that factors driving the selection of the bi-directional channel are different from those driving selection of the others. In particular, firms choosing this channel employ a more skilled workforce and generally interact with PRIs and UNIs in order to benefit their own innovative activities. Thus, this commitment to knowledge capabilities and innovation when firms use the bi-directional channel may enhance the potential of PRO–firm interactions to upgrade the national innovation system (NIS).

Additional information
Acknowledgements
The authors acknowledge the Argentinean National Institute of Statistics (INDEC) and in particular Jorge Souto and Germán Herrera for all their efforts in building up the database. The National Council for Science and Technology Research (CONICET), Argentina, the National Agency for the Promotion of Science and Technology (ANPCyT), Argentina, and the International Development Research Centre (IDRC), Canada, helped fund this project.

Notes

  1. See Arza and Vazquez (2010) for the Argentinean study, Fernandes et al. (2010) for Brazil, Orozco and Ruiz (2010) for Costa Rica and Dutrénit et al. (2010b) for the Mexican case study.
  2. For example, three Argentinean scientists working in Argentinean PROs won Nobel Prizes in Science and a Mexican graduate of a Mexican public university won the Nobel Prize in Chemistry for research at MIT. Similarly, the share of publications by authors from these countries in total world publications indexed in the ISI Web of Science is much higher than the share of these countries in the patents database USPTO (United States Patent and Trademark Office). Moreover, the increase of scientific production observed in the last decade has not been accompanied by an increase in patent applications.
  3. Recent research indicates that these modes continue to be preferred by some Latin American countries (Dutrénit and Arza, 2010).
  4. Varsavsky (1973) proposed what López (2007) called a linear model but ‘the other way around’: he argued that society had to set the productive priorities, from which technological needs were to be derived. These needs should be satisfied by S&T. In turn, Sábato (1973) developed the triangle model to emphasise the need for public policies to integrate the three vertexes – the state, the productive sector and the scientific sector. Sábato’s ideas set a precedent for the triple helix notion of Etzkowitz and Leydesdorff (1997).
  5. Notably, private sector investments in R&D are low (Thorn, 2005). In 2007, for example, firms participated in less than 30% of total expenditures in innovative activities, a share lower than that in Brazil (45.5%), Chile (45.8%) and México (41.5%) (http://www.ricyt.org/).
  6. See Arza and Vazquez (2010) for the Argentinean study, Fernandes et al. (2010) for Brazil, Orozco and Ruiz (2010) for Costa Rica, and Dutrénit et al. (2010b) for the Mexican case study.
  7. Kruss (2012) uses the same taxonomy to relate channel with benefit and also with risk of interactions for the biotechnology sector in South Africa.
  8. UNIs and PRIs differ in their mission within the NIS. Consequently, we analyse results separately for these types of institutions.
  9. PRIs included INTA, INTI and ANPCyT.
  10. This group was built to resemble as closely as possible the linked group in size and sector affiliation.
  11. Despite the sampling methods for the unlinked group, non-responses created a significant difference in the average size of both subsamples.
  12. Firms had to assess research outputs on a four-point Likert scale (1–4). Table 1 re-scales original values by dividing them by 4, thereby creating a 0.25–1 scale.
  13. For the sake of precision, we consider choosing a channel of interaction to be when the firm assessed any of the modes which comprised the channel as at least moderately important (i.e. a value >0.5 in the 0.25–1 scale).
  14. Results correspond to a probit model on the probability of connecting to PROs. The estimates for the selection equations of all the models, whose regression equation is presented in Table A.1, are very similar and are available upon request.
  15. It is important to remind the reader that the group of unlinked firms that was included in The Survey 2006 was selected to resemble the size and sector characteristics of linked firms. Thus, size and sectoral affiliation are somehow already controlled for and we would not expect strong differences in size and sectoral variables between linked and unlinked firms just because of the way the sample was constructed.
  16. Not important, of little importance, moderately important and very important for each of the modes of the channel.
  17. R&D managers informed us, in personal interviews, that collaboration with PRIs is always easier and smoother than with UNIs, because of the complicated bureaucratic procedures involved, especially among the bigger UNIs (such as UBA).
  18. Although in this case the effect is larger for PRI interactions.
Is the Italian Government effective in relaxing the financial constraints of high technology firms?
Massimo G. Colombo , Annalisa Croce & Massimiliano Guerini
Pages: 73-96

ABSTRACT

The present work analyses the effect of public finance on firm investments in a longitudinal sample of 293 Italian unlisted owner-managed, new technology-based firms (NTBFs), observed over a 10-year period from 1994 to 2003. We find that large, old NTBFs and those located in the north of Italy are not financially constrained, while small, young NTBFs and those located in the south of Italy rely significantly on internal capital to finance their investments. Public finance may play a prominent role for these latter firms. Indeed, empirical evidence shows that receipt of public subsidies by financially constrained NTBFs results in a reduction of investment–cash flow sensitivity in the long run. We interpret these results as an indication of the relaxation of financial constraints. Moreover, we find that, after receiving public finance, young and small firms increase their investment rate while NTBFs located in the southern regions do not. Nonetheless, small and young NTBFs benefit greatly from public intervention, but are less likely to obtain public support than their larger and older peers. Italian policy measures have also paid particular attention to NTBFs located in the south of Italy, but we find that public finance has no effect on the investment rate for southern NTBFs. This evidence raises some doubts about the overall efficacy of Italian governmental intervention in this domain.

Additional information
Notes
Current address: DESE, Università di Pisa, Pisa, Italy.

  1. Some studies assessing the effects of financial constraints on R&D and innovation have relied on direct survey-based measures of the existence of financial constraints (e.g. Savignac, (2008) and Tiwari et al., (2007). These studies show that financial constraints do indeed hamper innovation.
  2. An automatic scheme gives financial assistance to all applicants fulfilling all the requirements specified in the law. In contrast, a selective scheme provides financial support to selected applicants; applicants compete for financial subsidies and their projects are judged by committees of experts appointed by the national authority.
  3. Nonetheless, the finding that SBIR grants foster growth was not replicated by Wallsten (2000). After controlling for endogeneity of public support in a multi-equations framework, it was found that SBIR grants did not positively influence firms’ employment growth, while crowding out firms’ private R&D expenses.
  4. In the north category, we include the following Italian regions: Valle d’Aosta; Piemonte; Lombardia; Liguria; Emilia Romagna; Friuli-Venezia Giulia; Trentino-Alto Adige; Veneto. In the South category we include: Lazio; Marche; Toscana; Umbria; Abruzzo; Basilicata; Calabria; Campania; Molise; Puglia; Sardegna; Sicilia.
  5. The descriptive statistics presented here are consistent with the official statistics for the Italian industrial system, provided by Istituto Nazionale di Statistica (ISTAT, 2009).
  6. Measures in support of the south date back to the early 1950s, when the Italian Government created the Cassa del Mezzogiorno (Fund for the South), a public agency devoted to financing industrial development and public infrastructure in the region. For a discussion of Italian policy measures targeted to NTBFs, see Appendix 1.
  7. We measure investments by the increase in the book value of tangible and intangible assets net of depreciation.
  8. Other authors have used ex-dividend cash flows (e.g. Manigart et al., 2003). We opted for cash flows before dividends because our sample is composed of unlisted firms. Managers of listed firms are more constrained than those of private firms because of the dividend paid to shareholders, as any reduction may be perceived as a negative signal by investors. Conversely, in private firms, dividends have no signalling role and all cash flows can be reinvested if profitable investment opportunity arises.
  9. These indicators represent the share of RITA NTBFs that obtained public funds out of the total number of RITA NTBFs located in the same geographical area and operating in the same industry as firm i.
  10. Note that we make the assumption that the clustering of public funds in specific industry or geographic markets is exogenous − that is, it is not driven by investment opportunities unobserved by third parties faced by the NTBFs that are in those markets and are potential candidates for receiving public funds. Therefore, it is uncorrelated with the error term of the Euler equation.
  11. The evidence presented here is limited to the effect of public finance on investments in tangible and intangible assets. However, a positive effect of public intervention can materialise in different forms for firms in the south of Italy (e.g. productivity or R&D investments).
  12. In Italy, several governmental institutions were responsible for the administration of public subsidies. They include the Ministry of Economics and Finance, the Ministry of Industry, the Ministry of University and Research, the Ministry of Labour and Welfare, the Ministry of Agricultural Food and Forest Policies, the Ministry of International Trade, and the Institute for Foreign Trade (ICE). In Italy, unlike in other European countries, there is no public agency in charge of innovation policy measures.
  13. Law 808/1985 accounted for another 25%, but mainly benefited large established firms. The remaining 25% was dispersed among a plethora of schemes.
  14. According to Italian fiscal law, firms may decide whether to treat R&D expenses as investments or to expense them when they are incurred. This latter option is more favourable for firms with positive net income.
Do public subsidies affect the performance of new technology-based firms? The importance of evaluation schemes and agency goals
Luca Grilli & Samuele Murtinu
Pages: 97-111

ABSTRACT

New technology-based firms (NTBFs) greatly contribute to the dynamic efficiency of the economic system. To perform this role, NTBFs need external financing. However, private financing of this type of firm is particularly subject to market inefficiencies. This seems to recommend policy intervention and NTBFs often find support through payment of direct public subsidies. When these are based on ex ante selective screening procedures of applicants and awarded competitively, direct public support may exert a positive effect on the performance of NTBFs beyond the amount of the subsidy. By picking promising projects, governments may signal the quality of a firm to third parties, thereby lowering information asymmetries. This paper contributes to the literature on the differing impact of various subsidies on firm performance by crossing the evaluation dimension (i.e. selective vs. automatic subsidies) with the dimension of the specific goal (R&D-enhancing vs. other measures) for which a subsidy may be implemented. Our results show that the evaluation mechanism and the goal of the subsidy are both important dimensions in the policy design domain and that selective R&D subsidies outperform other types of scheme in fostering NTBF performance.

Additional information
Acknowledgements
We gratefully acknowledge the support of PRIN 2006 and UNICREDIT research funds. We are indebted to Massimo G. Colombo and participants of the 2010 Conference ‘Technical Change: History, Economics and Policy’ in honour of Nick von Tunzelmann at SPRU for helpful comments on this and related works. The usual disclaimer applies.

Notes

  1. Definitions of static and dynamic efficiency have been provided by several streams of academic literature: organisation theory (Burns and Stalker, 1961), technology and operations management (Abernathy, 1978), strategic management (Heskett, 1987) and economics (Stigler, 1939; Hart, 1942; Marschack and Nelson, 1962; Jones and Ostroy, 1984; Klein, 1984; Carlsson, 1989). In this paper, we adhere to the ideas of Klein (1984). The author defines static efficiency as the optimal combination of disposable inputs subject to the constraints imposed by a given (fixed) production function. Conversely, dynamic efficiency is defined as ‘changing the production function in profitable directions’ (p.46). To some extent, NTBFs might engender technological shifts on the production function through product, process and organizational innovations.
  2. The Small Business Innovation Research program in the US is a typical example of a selective subsidy. The French Credit Impôt Recherche and Jeunes Entreprises Innovantes schemes are emblematic examples of automatic subsidies.
  3. The responsibility for administration of these policy measures was assigned to different governmental institutions. They include the Ministry of Economic Development, the Ministry of University and Research, the Internal Revenue Service, the Ministry of Labour and Welfare, the Ministry of Agricultural Food and Forest Policies, the Ministry of International Trade, the Institute for Foreign Trade, SIMEST (Italian Society for Foreign Firms) and Finance Company for Entrepreneurs located in the North-East. There was no single public agency in charge of innovation policy measures.
  4. In 1950, the Italian Government founded the Cassa del Mezzogiorno (Fund for the South), a public agency to construct public works and infrastructure for the development of the south of Italy; it ceased operations in 1992.
  5. In Italy, data provided by official national statistics do not include a reliable description of the universe of Italian NTBFs. Most individuals defined as ’self-employed’ by official statistics are actually salaried workers with atypical employment contracts. Thus, on the basis of official data, such individuals cannot be distinguished from the entrepreneurs who created new firms.
  6. Note that we aggregate the previously exposed sectors into three macro-industries to give a sufficient number of observations in each industry to estimate our performance variable.
  7. It is calculated as the average of the following indices at NUTS (Nomenclature of territorial units for statistics) 3 level: per capita value added; share of manufacturing of total value added; employment index; per capita bank deposits; automobile:population ratio; consumption of electric power per head. For a seminal discussion and a critical review of the empirical literature on the relationship between public infrastructure capital and firm productivity, see Holtz-Eakin (1994) and Fernald (1999).
  8. For a survey of the various estimation techniques for total factor productivity and a more detailed description of Olley and Pakes’ methodology, see Levinsohn and Petrin (2003) and van Biesebroeck (2007).
  9. Firms choose their current input levels knowing possible unobserved productivity shocks, which are known to the firm, but unobserved by the researcher. This leads to a correlation between production inputs and the composite error term of the production function and to a biased estimation of the coefficients of production inputs through ordinary least squares estimation. For a detailed discussion on this aspect, see Eberhardt and Helmers (2010).
  10. Note also that the use of survey information necessarily implies a potential survivorship bias in our data. We tested the possible presence of the problem by adapting to our specific framework a recent methodology proposed by Semykina and Wooldridge (2010) for testing the existence of selection bias in panel data in the presence of unobserved heterogeneity and endogenous regressors. We estimated the two models of Table 3 with the addition of an inverse Mills ratio type of firm exit (for details, see Colombo et al., 2009). Its coefficient turns out not to be statistically significant and excludes the presence of any remarkable survivorship bias (p > |Z| = 0.247 and p > |Z| = 0.164 in model I and model II respectively).
  11. In addition to France, many other countries give automatic subsidies for R&D. For instance, in Canada, the Scientific Research and Experimental Development provides tax credits to businesses conducting R&D (they must meet the Frascati definition of R&D). In Belgium, there is a partial exemption of advance tax payments in favour of companies employing researchers. In Norway, tax support for industrial R&D (the SkatteFUNN scheme) was established in 2002.
The differential impact of privately and publicly funded R&D on R&D investment and innovation: the Italian case
Giovanni Cerulli & Bianca Potì
Pages: 113-149

ABSTRACT

This paper explores the impact of a specific R&D policy instrument, the Italian Fondo per le Agevolazioni della Ricerca (FAR), on industrial R&D and technological output at the firm level. Our objective is threefold: first, to identify the presence or absence of private R&D investment additionality/crowding-out within a pooled sample and in various firm subsets (identified by region, size, level of technology, and other features), while also taking into account the effect of single policy instruments or mixes of them. Secondly, to analyse the output (innovation) additionality by comparing the differential impact of privately funded R&D and publicly funded R&D expenditure on applications for patents filed by firms. Thirdly, the paper will compare the structural characteristics of firms showing additionality with those of firms showing crowding-out, in order to determine the firm characteristics associated with successful policy interventions. Our results suggest that FAR is effective in the pooled sample, although no effect emerges in some firm subsets. In particular, while large firms seem to have been decisive for the success of this policy, small firms present a more marked crowding-out effect. Furthermore, the firms’ growth strategies and ability to transform R&D input into innovation output (patents) seem to have a positive effect in terms of additionality.

Additional information
Acknowledgements
This paper is part of the FIRB 2005–2008 strategic project (Models and Tools for Evaluating the Short and Medium Term Impact of Firm R&D Investments on the Italian Productive System), financed by the Italian Ministry of University and Research. We wish to thank all those who participated in the project and the anonymous referees for their useful suggestions, which substantially improved the paper. All remaining errors are our own.

Notes

  1. FIRB 2005–2008, code RBNE03ETJY.
  2. Martin and Scott (2000) suggest that policy interventions to promote R&D should be targeted and sector-specific rather than widespread and generic. They make use of Pavitt’s taxonomy (1984) to identify: (1) main sectoral innovation modes; (2) sources of sectoral innovation failure; and (3) suitable policy instruments.
  3. It is worth stressing that, although the main concern of the literature is with input additionality (the direct effect of an R&D support programme on firm R&D expenditure), another two kinds of additionality are relevant: output additionality (referring to the downstream effects of R&D incentives on firm innovativeness, productivity or profitability), and behavioural additionality (referring to structural/strategic changes in the way a firm operates after receiving a subsidy; for example, by becoming a patenting firm, by modifying technological specialisation, and so on). While both input and output additionality are generally measured through quantitative econometric techniques, behavioural additionality is usually detected by qualitative surveys (interviews and questionnaires) as well as case studies [for an in-depth analysis of these aspects, see IPTS (2002)].
  4. In particular, they distinguish between contracts and grants as they are different incentive tools on the part of the government. In this paper, we focus on grants, although many of our conclusions can be extended to contracts too.
  5. Actually, David et al. (2000) maintain that the MCC curve starts flat and rises only after a given threshold. This form of the MCC curve is attributable to the self-financing effect: firms start by using their retained earnings (the flat part) and only when these have run out do they turn to debt and/or equity markets (the upward part). In other words, they embrace the pecking order approach to firm investment financing (see Myers and Majluf, 1984).
  6. The distinction among these forms of subsidisation is significant. In particular, the analysis of contracts greatly differs from that of grants. According to Lichtenberg (1987) and David and Hall (2000), two main elements contribute to the occurrence of additionality/crowding-out effects in the case of contracts: the first is based on an increase in research input costs attributable to changes in the labour demand for scientists and engineers brought about by the contract (especially when the total supply of researchers is assumed to be fixed and the government is budget-constrained); the second concerns the spillover effects generated by contracts, especially when they are the basis for future (expected) contracts and/or when firms plan to sell products to the government at the end of an R&D programme. Both causes can bring about additionality as well as crowding-out, even though the former (labour market effect) seems likely to generate crowding-out, whereas the latter (spillover effect) is likely to cause additionality [for a formal model, see David and Hall (2000)].
  7. This percentage (13%) does not represent the share of collaborative projects within FAR, but only within our dataset.
  8. Here ‘oriented’ means that more than 75% of R&D activity is devoted to either research or development. In the other cases – the majority – the level is 61%.
  9. In what follows, we use the terms ‘treated’ and ‘untreated’ as synonyms for ‘supported’ and ‘non-supported’.
  10. We work with data on public subsidy commitment and not with subsidy outlays (i.e. subsidy allocation) since the latter are not fully available and are less reliable.
  11. Vector x 1 represents the agency’s selection criteria, usually including firm/project characteristics as well as welfare objectives. In our case, only the first type of variable is included.
  12. Vector x 2 represents variables referring to firm R&D choices/strategies and should include the DHT variables from the previous section.
  13. The average own R&D expenditure of the untreated units is about 570,000 euros: (801–570)/570≈0.40.
  14. Since benefits from tax credit are calculated on past R&D activities, we have checked for the presence of additionality/crowding-out for this fiscal measure by allowing for one and two-time lags of own R&D expenditure, obtaining the same negative result as in the case of Table 8.
  15. The results might be attributable to two factors: not all collaboration projects are included in our dataset (which was created by merging three different initial datasets), and collaboration projects include top-down programmes dominated by large firms.
  16. We use the words ‘instrument’ and ‘measure’ interchangeably.
  17. The introduction of covariates reduces the number of observations because of numerous missing values.
  18. Since the NNM already makes use of the covariates used in the regression analysis, in the Poisson regression we use as covariates region, sector, time, and size.
Erratum
Correction

Support the Manchester Manifesto: a case study of the free sharing of human genome data
Huanming Yang
Page: 151

This article refers to:Support the Manchester Manifesto: a case study of the free sharing of human genome data

Support the Manchester Manifesto: a case study of the free sharing of human genome data

In this Response, published in Prometheus, Volume 29, Number 3, pp. 337–341 (DOI:10.1080/08109028.2011.631275), Figure 2 is incorrect. The correct Figure 2 is displayed below. Taylor & Francis would like to apologise for this error.

Figure 2 The impact of free sharing of the rice genome sequence data on rice research in developing countries