اندازه گیری مشکلات سیستمیک در سیستم نوآوری ملی.برنامه کاربردی برای تایلند

The paper contributes to research on innovation systems in general and, in particular, to the current debate on rationales for innovation policy by providing a framework to identify systemic problems in a given system of innovation and test the framework empirically. The data were drawn from the Thai Community Innovation Survey in the period after which a major change in the country's innovation system policy had been initiated. By hierarchical factor analysis, systemic problems are identified and grouped into four components: institution, network, Science and Technology infrastructure and other support services. The analysis allows researchers to investigate the mismatch between policies and problems and identify policy gaps.

Innovation system (IS) research is increasingly important to innovation policy making. Since the approach was flagged by the OECD in the mid nineties, an increasing number of governments have adopted IS explicitly in their innovation policies (Mytelka and Smith, 2002). However, applying the concept in practice is a daunting task (Chaminade and Edquist, 2006 and Chaminade and Edquist, 2010). Policies based on the IS approach often collide with old paradigms, rationales and instruments (Intarakumnerd and Chaminade, 2007) and, more often than not, end up being one-size-fits-all-policies rather than policies that take the specificities of the system into account.1 One of the reasons for this is that we know too little about how to identify and measure specific problems in the system (if at all possible), despite several fruitful attempts to define them. The literature on national systems of innovation (Lundvall, 1992, Edquist, 1997, Nelson, 1993 and Freeman, 1987) and more specifically the strand of literature dealing with rationales for innovation policy (Lipsey and Carlaw, 1998, Smith, 2000 and Chaminade and Edquist, 2006), has defined systemic problems as systemic imperfections that might slow down or even block interactive learning and other activities that are crucial parts of innovation process in a certain system of innovation (Woolthuis et al., 2005, 610). Despite the prior efforts to define what systemic problems are ( Carlsson and Jacobsson, 1997, Norgren and Hauknes, 1999, Smith, 2000 and Woolthuis et al., 2005), to our knowledge, no attempt has been made thus far to empirically identify or measure problems in a specific system of innovation. This paper aims at contributing to filling this gap by analysing problems in the Thai innovation system. Thailand is an interesting case study, since the country, unlike the East-Asian Tigers, is a less-successful country in terms of technological catching up with the forerunners. It has also been a latecomer in trying to adopt and implement the IS approach, despite suffering from very clear systemic problems ( Bell, 2002 and Intarakumnerd et al., 2002). The paper investigates whether there is a mismatch between the systemic problems of the Thai innovation system and the innovation policies implemented in the country since 2001. In doing so, we use data from the Thai innovation survey in 2003 which seems to allow a sufficient time lag for our analysis to identify systemic problems after a major political transition starting in early 2001, i.e., changing from a traditional research-based policy (pre-Thaksin administration) to a more explicit innovation system policy (Thaksin era). The Thai innovation survey has a particular advantage as it contains several detailed questions2 that seem to allow identification of some of the systemic problems in Thailand. We employed hierarchical factor analysis in identifying institutional, S&T infrastructure, support services and network components/problems. These system components were then linked to a qualitative description of the real situation in Thailand in the discussion of whether there is a mismatch between Thai innovation policy instruments and the systemic problems captured. The rest of the paper is organised as follows. In the next section, we give a brief summary of the IS approach and discuss its implications for innovation policy, and we introduce some major systemic problems as the prior studies pointed out. Section 3 provides an overview of the Thai innovation system and policy. Section 4 gives a general account of the Thai innovation survey and describes the dataset used and the questions selected to capture system subcomponents and system components. In Section 5, we provide descriptive evidence, present our hierarchical (two-stage) factor analysis, identify and measure problems of the innovation system and discuss them in the light of the recent transformation of the Thai innovation system and innovation policy. Section 6 matches the systemic problems found with some of the main current policies in Thailand. The paper is rounded up in Section 7 with conclusions and some final remarks.

6.1. Results from first-stage factor analysis In this section, we present the factor analysis for each of the 4 questions included in Table 2, which we want to reduce. The following Table 4, Table 5, Table 6 and Table 7 show, respectively, the results of the factor analysis for each of the 4 questions, that is, Question 1. Business environment for innovation in Thailand, Question 2. Government support for innovation in Thailand, Question 3. Obstacles to innovation in Thailand and Question 4. Sources of information for innovation in Thailand.We labelled the subcomponent 1 as ‘Knowledge Resource’, as it refers to system elements that supply knowledge for innovation activities in firms. This first sub-component loads highly on the technological sophistication of suppliers and the availability of suitable manpower. In particular, the results concerning the latter are consistent with the recent studies highlighting that firms in Thailand consider the lack of qualified human resource, especially within the area of Science and Technology, as a very serious problem (see for example, TDRI, 2004, Chalamwong et al., 2007 and NESDB, 2007). We labelled the subcomponent 2 as Technical support as it refers mainly to technical support and collaboration from other organisations. For innovative firms, consultancy support, which has a high factor loading for the subcomponent 1 (knowledge resource), seems to correlate also (though to a lesser extent) with other support shown with high factor loadings in the subcomponent 2, ‘Technical Support’. Besides consultancy support, this dimension includes support from and collaboration with universities and other organisations (for both innovative and non-innovative firms). This is reassuring because it is common that universities provide consultancy service to firms. Schiller (2006) found that this is particularly true in the Thai case as consultancy service is the most popular mode of university–industry linkage in Thailand. Generally, it can be said that this subcomponent points to the fact that interactions with the key knowledge producers in the system are important for the firm's innovation process (i.e., the need for knowledge transfer). The subcomponent 3 was labelled ‘Openness to Innovation’ as it refers to the attitude of the different Thai system actors’ (customers, suppliers and general public) towards innovation. This is an important subcomponent that captures a soft institutional aspect, which is not often addressed in studies based on innovation surveys. The fourth sub-component is labelled ‘Regulation and Other Institutional Conditions’ as it embraces indicators of failure acceptance, regulatory environment, intellectual property protection and finance for innovation. In the case of innovative firms, an overlap was found in stock exchange listing requirements as it has a factor loading shared about halfway between this and the last subcomponent, ‘Financial and IT Infrastructure’. Finally, the fifth subcomponent is labelled as ‘Financial and IT infrastructure’, as it includes government incentives for innovation, communication services for innovation and, as mentioned above, stock exchange listing requirements in case of innovative firms. The result differs somewhat for non-innovative firms as the subcomponent includes acceptance of failure, communication services and finance for innovation. These different results may reflect the share of larger companies among innovative firms versus that of smaller ones among non-innovative firms.19 Small firms are usually not listed on the stock exchange (e.g., as for the purpose of fund raising). Alternatively, they rely much more on other sources of funding (like business angels), usually located in close proximity to the firm (Crevosier, 1997).20Cooke (2002) uses the term ‘proximity capital’ for this sort of financial infrastructure and, further, suggests its correlation with physical infrastructure like telecommunication services, which supports the results obtained for the case of non-innovative firms. For innovative firms, the results are in fact coherent with our knowledge about the Thai economy. In Thailand, the requirements for stock exchange listing may be, on the one hand, regarded as a regulation and institutional condition (they provide the firms with access to external funding sources for their innovation activities) and as a type of government incentives, on the other. It should be noted, however, that this does not appear to work well for smaller firms. The Thai ‘Market for Alternative Investment’ (MAI), for example, has been set up to especially foster innovative SMEs since 1999, but the MAI gets little interest from SMEs in practice. One reason is that in this case the founding shareholders are reluctant to enact common stock rights issues that would effectively dilute their stakes in the listed companies (possibly referring to acceptance of failure). Moreover, many SMEs see that the MAI requirements tend to disqualify most small- and medium-sized enterprises for being below the minimum capitalisation level. This instigates a problem as it results in too few outstanding shares to trade adequately on the market (Freeman, 2000). The results of the first-stage estimate referring to government support for innovation (as discussed above, another set of indicators unique to the Thai case) are presented in Table 5. The factor patterns, which are similar for innovative and non-innovative firms, are summarised as follows. The label ‘Government Technical Support’ is given to the first subcomponent retained in this estimate. This subcomponent integrates different services provided by NSTDA and the Ministry of Industry, including information services, testing as well as analytical services and support for quality systems and human resource development. For non-innovative firms, this dimension also includes industrial consultancy services and technology transfer arrangements. 1. The second column presents the group of ‘Government Industrial Support’, which consists of loans and grants, technology transfer arrangements and industrial consultancy services. However, in the case of non-innovative firms, the last indicator (having a high factor loading in the first column) does not appear to correlate strongly with the other two indicators constituting this dimension. 2. We label the last subcomponent in this estimate ‘Tax Incentive’, as it combines two tax deduction programs for training and R&D activities which are reported with high factor loadings in both cases. We continue with a summary of the results in Table 6 reporting the estimations for the Question 3. Obstacles to innovation as perceived by innovative and non-innovative firms. Three subcomponents were retained for the two groups of firms (also with similar factor patterns). We label the first subcomponent ‘Financial Constraint and Uncertainty’ as it comprises the firm's perceived high cost and risk as well as monetary limitation. This dimension appears in a similar way for innovative and non-innovative firms. The second subcomponent is labelled ‘Lack of Information and Other Support’ as it includes the problems about qualified personnel, government and other support. In this case, the results for innovative firms show that the lack of information on market and technology tends to be relevant to this dimension. We label the last subcomponent retained as ‘Hampering Market Condition’, loads highly on the lack of domestic competition, the lack of customer's interest in innovation and moderately on the lack of information on markets. In the case of non-innovative firms, this subcomponent also seems to refer to the lack of information on technology, to some extent. Now we shift our attention to the last question on information sources for innovation in Thailand. The results of our last factoring estimate in the first stage, which are provided in Table 7 (available for innovative firms only, see a discussion above), present the following five subcomponents. We label the first sub-component as ‘Universities and Non-Profit Research’ as it combines information from universities and public as well as private non-profit research institutes. Subcomponent 2 is labelled ‘Supplier’ as it embraces information from both local and foreign suppliers. The third subcomponent is labelled ‘Professional Knowledge Sources and Internet’ as it brings together information from the literature, Internet, conferences and other events. The fourth subcomponent loads primarily on competitors and business and technical service providers. We label it ‘Industry’ as it includes many information sources within the industry. Note that this subcomponent also loads, though only modestly, on patent disclosures and private research institutes. The last subcomponent that we labelled ‘Intra-firm, Client and Competitor’ correlates mostly with information from clients and from within the company or group of companies, and to some degree with competitors. 6.2. Results from second-stage factor analysis As indicated in the analytical framework, factor scores for all system subcomponents detected in each first stage estimate were computed and used in the second stage factor analysis (innovative firms only, as noted above). In addition, four separate variables were included: (i) a dummy for venture capital/business angle investment received for the firm's innovation activities; (ii) innovation intensity in terms of innovation expenditure over total employees; (iii) R&D intensity in terms of R&D expenditure over total employees; and (iv) a proportion of knowledge workers including scientists and engineers in the firm. The results suggesting four distinct but related components in the Thai innovation system are provided in Table 8.