همکاری دانشگاه و صنعت و نوآوری در صنایع نوظهور و بالغ در کشورهای صنعتی
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی|
|2380||2013||11 صفحه PDF||38 صفحه WORD|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Research Policy, Volume 42, Issue 2, March 2013, Pages 443–453
2. تعامل دانشگاه و صنعت و فن آوری چالش در صنایع رشد یافته و در حال رشد
3. زمینه همکاری دانشگاه و صنعت در برزیل
3. 1. علم و فن آوری و عملکرد رقابتی
3. 2. فعالیت های خلاقانه از شرکت های برزیلی و همکاری آنها با دانشگاه ها
3. 3. تکامل سیاست اقدامات حمایت همکاری دانشگاه و صنعت
5. همکاری دانشگاه و صنعت در برزیل
5. 1. درک محققان از انگیزه ها و موانع همکاری با صنعت
5. 2. مشوق های سازمانی برای کارآفرینی دانشگاهی و تعامل با صنعت
6. همکاری دانشگاه و صنعت :تفاوت پروژه ها در شرکت صنعتی در حال رشد چیست؟
7. بحث و نتیجه گیری
As the economies and indigenous technological capabilities of the new industrialized countries improve, national universities and public research organizations are expected to become increasingly important for supporting indigenous firms to move into more dynamic and high-opportunity industries. However, the characteristics of collaboration with universities may be very specific depending on whether the industry partner is engaged in mature or emergent activities. In this study, we explore and discuss the role of university–industry collaboration for the development of innovation in mature and emergent industries in new industrialized countries. Evidence from 24 research groups in science and engineering departments in universities and public research organizations in Brazil provides preliminary empirical corroboration for the proposal that the contexts and role of university–industry collaboration in mature and emergent industries are diverse. Knowledge networks are underdeveloped in emerging industries, and public support for research projects is dispersed. This means that university research and development projects with firms in emergent industries are less likely than projects with firms in mature industries to be the result of academic initiatives and public calls for research projects, or to be wholly financed by major public research sponsors. In emergent industries, the role of students and firm employees is crucial for mediating between public research organizations and companies. The policy implications of these preliminary findings are discussed.
Both managers and policy makers with responsibility for innovation at the firm and the country levels are interested in knowing the impact of pursuing different innovation strategies. A specific classification of innovation strategies that has received recent attention is the one that distinguishes between internal and external strategies. While it is well known that external sourcing and internal production are often used by firms in many areas of activity, the tendency for firms to use external sources of knowledge in their search for innovation is relatively recent (see Chesbrough, 2003) and a small but growing literature has started investigating the impact of these strategies upon innovation outcomes and performance (Cassiman and Veugelers, 2006 and Lokshin et al., 2008). This research has found that external R&D is productive in the sense that firms using external sources for their R&D strategies have better innovation outcomes, in particular if firms also undertake R&D in-house. However, not all research outcomes translate into profits. Studies that have examined more than one dimension of the research outcomes have found that the determinants of the creation and appropriation of value are not the same as those of the number of innovations or of sales of innovative products. For example, Belderbos et al. (2004) found the determinants of labor productivity growth and growth in sales of new and innovative products to be different, while Okamuro (2007) found that technological and commercial success have different determinants. Most of the research into the impact of alternative knowledge acquisition strategies has focused on how the changes in one variable of interest affect the mean performance of firms. However, the distribution of profits from innovation has been shown to be highly skewed, a small minority of innovations accounting for a disproportionate share of profits (Scherer and Harhoff, 2000). Given this typical shape of the distributions of gains from innovation, it is possible that different innovation strategies generate different distributions of performance. Knowing that a strategy may yield enormous returns in the few cases in which it works well is not the same as knowing that a strategy works well in most of the cases and provides positive albeit limited returns. In this paper we move beyond asking if different innovation strategies display different results on average, and we also ask questions such as: Do the different innovation strategies present different degrees of risk? Is one strategy more or less likely to create breakthroughs evinced by a more skewed distribution of performance? Is one strategy more likely to generate distributions of performance with many outliers? In other words, we ask whether these strategies affect the variability, the skew, and the heaviness of the tails of the resulting distributions. A simple way of attempting to answer such questions would be to compare distributions of returns for firms following different innovation strategies. This, however, would not take into account that firms are different in many dimensions other than research strategy. In order to control for these differences our empirical strategy is based on quantile regressions, which we use to compare the outcomes of internal and external innovation strategies against those of firms that do not pursue any formal innovation efforts. Quantile regressions provide a methodology for estimating the impact of a given variable upon different points of the distribution of interest, while controlling for other variables of interest (see Koenker, 2005 for a survey). We estimate quantile regressions for a wide range of quantiles of firm performance and, based on these estimates, we compute the impact of the innovation strategies upon measures of dispersion, skewness, and kurtosis of the distributions of performance. There are important implications from this knowledge. Even if a handful of firms benefit and the gains of those that benefit are large enough, from the society's point of view it should be desirable to pursue such strategies, as the losses of the many would be more than compensated by the gains of the few. However, if this is so, risk averse managers may not wish to engage in this type of activity, especially if their firms are small and lack the means to enter into a myriad of projects simultaneously. This may be particularly true if the strategies that lead to a breakthrough with high probability can also cause high losses with high probability. Managers may refrain from pursuing this strategy if they run the risk of being evaluated by the outcome of a few projects only. In such a case, policies should be designed to lead firms into activities that will lead to failure with very high probability. If most firms benefit, these policies are less needed. Even if distribution of gains is relatively symmetric, firms may be deterred from pursuing innovations if the distributions of gains have a very high kurtosis. In this case, the problem is not that only a handful of firms benefit but rather that, even if one average innovation pays off, the rate of convergence to the mean may be too slow and a firm may be required to engage in too many projects in order to have a reasonable degree of assurance of reaching positive outcomes. Concentration of research, or other mechanisms that offer some form of risk protection, seem to be needed if this is the case. Our findings indicate that innovation strategies affect the performance of firms in more ways than commonly recognized. In particular, external innovation strategies are significantly associated with increases in median profits relative to firms that do not conduct R&D. They are also significantly associated with increases in dispersion of profits and with kurtosis, reflecting the fact that external innovation strategies increase the likelihood of very extreme outcomes. No significant effect upon the skew of the profit distribution is detected, however. The same pattern holds for internal strategies, but the effects are estimated to be smaller and not statistically significant. The paper is organized as follows. In Section 2 we discuss the rationale for innovation strategies having an impact upon performance and the previous evidence on the topic. Section 3 presents the methodology. In Section 3.1 we discuss the quantile regression framework that is employed in the analysis and highlight how it can be used to help shed light on the impact of strategies upon the entire distribution of profit rather than on a single point of this distribution. Section 3.2 presents the data and Section 4 the results. Finally, Section 5 concludes the paper.
نتیجه گیری انگلیسی
We investigated the impact of internal and external innovation strategies on the profitability of firms. We found that external strategies (contracting external R&D or cooperating with external partners in R&D projects) exert a positive impact upon performance. The estimated magnitude of this effect is greater than the estimated impact of conducting in-house R&D activities only and the former is clearly significant while the second is not. We were, however, able to generate deeper insights of these effects by combining the results from regressions at different quantiles of the distribution of profits and were able to describe a more complex pattern of effects of internal and external innovation strategies than what is commonly reported. Our results also reveal that external innovation strategies are riskier. The distribution of profits is significantly more spread for firms with external innovation strategies than for firms with no formal innovation activities, whereas no such contrast could be found for firms with internal innovation only. A more skewed distribution might also be expected, given that some recent studies on the distributions of returns to innovations have found these distributions to be rather skewed. However, we found only a negligible and insignificant effect of both internal and external innovation strategies upon the skewness of the distribution of profits. Instead, we found that external strategies lead to distributions of profits with heavy tails (more leptokurtic), suggesting that these strategies can lead to a high incidence of highly successful projects, but also to a high number of unsuccessful projects. We suspect that the emphasis on skewness in earlier studies on the distribution of returns to innovation arises from the fact that those studies focus on the gross returns to innovation and largely neglect the cost side of these projects. Our finding that external innovation strategies are risky has several implications. On the one hand, it calls for more detailed investigations on the particulars of managing these strategies. While Chesbrough (2003) describes open innovation strategies as a source of superior performance in a number of firms, he also alerts us to a number of mistakes that firms following these strategies may easily commit and that may be responsible for poor performance. Furthermore, Laursen and Salter (2006) find that openness spurs innovation activity but the benefits to openness are subject to decreasing returns. This study also indicates that costs of pursuing external strategies do not decrease with the number of external R&D projects. “Learning” in terms of lowering costs given decreasing returns seems to be limited. Hence, the decreasing returns are not likely to be compensated through decreasing costs or ‘economies of scale’ in managing external ties. On the other hand, the finding that external innovation strategies significantly increase the kurtosis of the distribution of profits means that the observed results from innovation projects will converge to the true mean at a slower pace than would be the case if the distribution were less leptokurtic. Returns are thus less predictable and, unless one is able to assemble very large samples, it is difficult to assure convergence to the true mean. Sample size, in this context, refers to the number of projects in which a firm is involved, the implication being that firms entering cooperative arrangements in the pursuit of risk reduction should be advised to do so mostly if they are very large and can afford to be involved in a very large number of projects at the same time. Because external cooperation means increased productivity and profits, it may be in the interest of society to promote such collaborations. But if they increase private risk, as suggested by our findings, then rational managers may enter a lower number of such ventures than what would be optimal from the point of view of society as a whole. There may be a role for research policy here.