Research SummaryWe develop an ex ante measure of commercial potential of science, an otherwise unobservable variable driving the performance of innovation-intensive firms. To do so, we rely on large language models and neural networks to predict whether scientific articles will influence firms' use of science. Incorporating time-varying models and the quantification of uncertainty, the measure is validated through both traditional methods and out-of-sample exercises, leveraging a major university's technology transfer data. To illustrate the methodological contributions of our measure, we apply it to examining the impact of university reputation and university privatization of science, finding that firms' reliance on reputation may lead to foregone opportunities, and privatization (i.e., patenting) appears to increase firms' use of the science of one university. We make our measure and method available to researchers.Managerial SummaryUsing machine learning, we develop a measure that estimates the probability that a scientific discovery will contribute to a commercially valuable innovation. This work addresses a key challenge: the inability to observe what scientific discoveries are worth pushing forward into commercial application. We illustrate the usefulness of this measure with two examples: 1.) firms' use of research from prestigious universities over equally promising work from less prominent ones; and 2.) how patenting affects the diffusion of commercially relevant science across firms. For practitioners, this measure can inform R&D, licensing, and other innovation related decisions by guiding firms' search for commercially relevant scientific research. The measure and the associated code are publicly available.

We study how the inventive capability of a firm conditions its participation in a division of innovative labor. Capable firms are, by definition, able to invent; for them, external inventions substitute for their own R&D. However, external knowledge is an input into internal invention, and thus, more valuable to firms with inventive capability. Using a simple model of innovation and imitation, we explore how inventive capability affects a firm’s R&D investments, and thus whether and how it innovates, imitates, or does neither. Further, we study how these outcomes are conditioned by the supply of external knowledge as well as the supply of external inventions. In an advance over the literature, we treat firm inventive capability as unobserved, and use a latent class multinomial model to infer its value. Using a recent survey of product innovation and the division of innovative labor among US manufacturing firms, we find that high capability firms tend to use internal, rather than externally generated inventions, to innovate, and they use external knowledge to enhance their internal inventive activity. By contrast, lower capability firms are more likely to introduce “me-too” or imitative products, and when they innovate, are more likely to rely on external sources of inventions. Our findings suggest the successful pursuit of R&D-led growth depends both on firm inventive capability and the external knowledge environment.

Download This Paper Open PDF in Browser Add Paper to My Library Share: Permalink Using these links will ensure access to this page indefinitely Copy URL Copy DOI

Do large firms produce more valuable inventions, and if so, why? After confirming that large firms indeed produce more valuable inventions, we consider two possible sources: a superior ability to invent, or a superior ability to extract value from their inventions. We develop a simple model that discriminates between the two explanations. Using a sample of 2,786 public corporations, and measures of both patent quality and patent value, we find that, while average invention value rises with size, average invention quality declines, suggesting, per our model, that the large firm advantage is not due to superior inventive capability, but due to the superior ability to extract value. We provide evidence suggesting that this superior ability to extract value is due to the greater commercialization capabilities of larger firms.

We suggest that, with proper design, innovation surveys can provide valuable data on innovation rates that inform judgments about whether the reported innovations are important, and in what sense, thus making such data more interpretable than claims that such innovations are simply “new.” First, we recommend asking respondents questions about a specific innovation in an identified line of business. Second, we recommend asking respondents to characterize their innovations in terms of different features that potentially link to the social welfare impact of the innovation. We propose five such features: technological significance, utility, uniqueness, imitability, and how different the innovation is from what the innovating firm has previously commercialized (what we call “distance” or the “implementation gap”). The chapter then illustrates the utility of our approach by, first, using newly collected data to construct measures corresponding to the proposed dimensions of innovation. We then use those measures to inform judgments about the importance of innovations in different industries. By recognizing the distinct features of innovations, we also showed how these features, when combined in novel and distinct ways, can provide a more nuanced view of innovation and its complexity.

Scholarly work seeking to understand academics’ commercial activities often draws on abstract notions of the institution of science and of the representative scientist. Few scholars have examined whether and how scientists’ motives to engage in commercial activities differ across fields. Similarly, efforts to understand academics’ choices have focused on three self-interested motives – recognition, challenge, and money – ignoring the potential role of the desire to have an impact on others. Using panel data for a national sample of over 2,000 academics employed at U.S. institutions, we examine how the four motives are related to patenting activities. We find that all four motives predict patenting, but their role differs systematically between the life sciences, physical sciences, and engineering. These field differences are consistent with differences in the payoffs from commercial activities, as well as with differences in the opportunity costs of time spent away from “traditional” research, reflecting the degree of overlap between traditional and commercializable research. We discuss implications for future research on the scientific enterprise as well as for policy makers, administrators, and managers.

Scholarly work seeking to understand academics’ commercial activities often draws on abstract notions of the academic reward system and the representative scientist. Few scholars have examined whether and how scientists’ motives to engage in commercial activities differ across fields. Similarly, efforts to understand academics’ choices have focused on three self-interested motives—recognition, challenge, and money—ignoring the potential role of the desire to have an impact on others. Using panel data for a national sample of over 2,000 academics employed at U.S. institutions, we examine how the four motives are related to commercial activity measured by patenting. We find that all four motives are correlated with patenting, but these relationships differ systematically between the life sciences, physical sciences, and engineering. These field differences are consistent with differences across fields in the rewards from commercial activities as well as in the degree of overlap between traditional and commercializable research, which affects the opportunity costs of time spent away from “traditional” academic work. We discuss potential implications for policy makers, administrators, and managers as well as for future research on the scientific enterprise. This paper was accepted by Toby Stuart, entrepreneurship and innovation.

Corporate research in the life sciences endures, despite diminishing in other fields of science. Corporate research in the life sciences endures, despite diminishing in other fields of science.

While organizational, economic, and innovation scholars have long understood the importance of coherence between a firm's overall strategy and structure (e.g., Chandler, 1962), much less is known about the link between a firm's innovation strategy and how innovation projects are structured within the firm. Given that a firm's innovation strategy can include multiple types of innovations (e.g., incremental improvements to existing products or substantially new innovations), we focus on different ways that management structures a firm's innovation projects (i.e., selecting organizational unit, project leader, investment timing, decision rights, organizational linkages, resource authority, and performance incentives). We introduce coherence between the type of innovation the firm pursues and how the innovation projects are structured. Coherence is proposed as a mechanism for explaining why some projects succeed while other projects fail. We argue that coherence increases when greater uncertainty in the innovation strategy (e.g., new technology) is accompanied by greater flexibility in the project structure (e.g., assigning project to a new organizational unit). To construct our arguments, we draw from prior literature on innovation management, together with exploratory evidence from 22 case studies across 9 large firms. We discuss the implications of our strategy implementation framework for organizational R&D, inertia, and learning.

Previous studies have investigated the correlation between firms' trademarking activity and innovation. However, there is a lack of formal investigation on which firms file for trademarks and the strategic reasoning behind their decision making process. In this paper we model firms' trademarking activity as a discrete choice in an innovation setting to understand why firms file for trademarks. We use a new sample based on firms in the United States and the USPTO trademark data to empirically test the various factors that condition firm's trademark decision. We show that trademarks help high quality firms hold on to their existing customers. During the diffusion process of a product innovation, trademarks facilitate the maintenance of the innovator's first mover advantage. As a result, innovators who face potential competition are more likely to file for trademarks than imitators. In markets that have larger proportion of early adopters, the first mover advantage is stronger, and the difference in incentive to file for trademarks is bigger.

Using firm data disaggregated by industry, we establish a set of regularities in the distribution of firm R&D intensities within manufacturing industries. We show how a simple probabilistic process, in which chance influences a key unobserved determinant of R&D and firm size conditions the returns to R&D, can account for these regularities and other features of the distributions. The model provides a unified, noncausal explanation of a series of long-observed relationships across mean R&D intensity, market concentration, and the coefficient of variation. It also offers a novel explanation for the inverse relationship between R&D productivity and firm size. (JEL 031)

In our response to Nelson’s important argument regarding the fit of research methods with the subject matter of various natural and social sciences, we highlight the complementarities offered by combining qualitative analysis with modeling and statistical analysis, focusing on economics. The argument is illustrated using a discussion of two studies on the economics of innovation.

Recent accounts suggest the development and commercialization of invention has become more "open." Greater division of labor between inventors and innovators can enhance social welfare through gains from trade and economies of specialization. Moreover, this extensive reliance upon outside sources for invention also suggests that understanding the factors that condition the extramural supply of inventions to innovators is crucial to understanding the determinants of the rate and direction of innovative activity.This paper reports on a recent survey of over 5000 American manufacturing sector firms on the extent to which innovators rely upon external sources of invention. Our results indicate that, between 2007 and 2009, 16% of manufacturing firms had innovated-meaning had introduced a product that was new to the industry. Of these, 49% report that their most important new product had originated from an outside source, notably customers, suppliers and technology specialists (i.e., universities, independent inventors and R&D contractors). We also compare the contribution of each source to innovation in the US economy. Although customers are the most common outside source, inventions acquired from technology specialists tend to be the more economically significant in term of their gross commercial value. As a group, external sources of invention make a significant contribution to the overall rate of innovation in the economy. Innovation policies, both public and private, should pay careful attention to the external supply of invention, and the efficiency of the mechanisms affecting the relationships between inventors and innovators. (C) 2016 Elsevier B.V. All rights reserved.

This article develops a model that shows how firm size—that most important firm-level correlate of R&D—moderates the impact of demand- and supply-side government policies that support R&D. The most robust result is that government support to product R&D will elicit less of a response the greater is average firm size within the industry. This result suggests, for example, that government support to upstream research in the life sciences will stimulate less downstream industrial investment in the development of new drugs to the extent that the firms in the downstream industry are larger. The model also predicts that where entry is less encumbered, we can expect demand-side subsidies to elicit greater product R&D within an industry. Thus, to the degree that, for example, barriers to entering the solar panel manufacturing industry are lower, we should expect R&D on solar panel development to rise more in response to tax credits for their purchase.

In addition to generating innovations, research and development (R&D) increases a firm's capacity to utilize existing information found within its environment. Although much research focuses on new innovation R&D, little research exists regarding the role that R&D plays in a firm's learning aptitude. A theoretical model is developed to examine implications of this dual role of research for R&D investment using the basic sources of technological knowledge a firm uses: such as the firm's own R&D, knowledge originating with its competitors' R&D spillovers, and knowledge created outside the industry. The model is tested using survey data on technological opportunity and appropriability conditions from Levin, et al. (1983, 1987) and data on business unit R&D expenditures from the Federal Trade Commission's Line of Business Programme. Results demonstrate that the influence of both appropriability and technical opportunity conditions are affected by determinants of the ease of learning, particularly the targeted quality of knowledge inputs. This suggests that the characteristics of knowledge that affect the ease of firm learning may represent an important class of determinants of R&D investments. (SFL)

This introductory essay of the Special Issue honoring Steven Klepper provides a synthesis of his work, and places the articles of the Special Issue within the context of his scholarship. Steven Klepper's pioneering work linked life cycle patterns in industry evolution to underlying micro-level firm innovation dynamics and employee entrepreneurship, and to the macro-level implications regarding geographical agglomeration and the growth of economies. Taken together, the lifetime of Klepper's research provides an integrated account of the key industrial mechanisms at work, and also sets the stage for future research, as is elaborated in individual contributions to the Special Issue.