The role of science is the subject of controversial debate in the academic literature and public discourse. Views vary from seeing science as an essential input for corporate innovation to seeing it as an ivory tower type of activity with little practical value. Some well-known examples document that science can play an essential role in the development of technological breakthroughs. Ferdinand Braun and Guglielmo Marconi could not have developed the wireless telegraph before Heinrich Hertz showed the existence of electromagnetic waves. Likewise, the development of the transistor at the Bell Laboratories would hardly have been conceivable without the scientific understanding of the physics of semiconductors. However, sceptics have argued that these cases are the exception rather
Watzinger, Schnitzer considers the following as important:
This could be interesting, too:
Peri, Sasahara writes The effects of global warming on rural–urban migrations
Kromann, Sørensen writes Adoption of automation technologies
Madsen, Robertson, Ye writes Taking Malthus seriously
Acciari, Polo, Violante writes Intergenerational mobility in Italy
The role of science is the subject of controversial debate in the academic literature and public discourse. Views vary from seeing science as an essential input for corporate innovation to seeing it as an ivory tower type of activity with little practical value. Some well-known examples document that science can play an essential role in the development of technological breakthroughs. Ferdinand Braun and Guglielmo Marconi could not have developed the wireless telegraph before Heinrich Hertz showed the existence of electromagnetic waves. Likewise, the development of the transistor at the Bell Laboratories would hardly have been conceivable without the scientific understanding of the physics of semiconductors. However, sceptics have argued that these cases are the exception rather than the rule and that ideas for inventions usually come from sources other than science (Kline and Rosenberg 1986, von Hippel 1988). Sceptics also doubt the reliability of knowledge produced at universities and the effectiveness of knowledge transfer from universities to industry (Goozner 2005, Butler 2008, Harris 2011, Osherovich 2011, Freedman et al. 2015, Bikard 2018).
Measuring the value of science
Settling this controversy is difficult because measuring the contribution of science poses two problems. First, it is not obvious how to assign a monetary value to an invention. Second, science plays a larger role in some technologies than in others. The challenge is hence to distinguish how much of the value of an invention is due to science and how much of it is technology-specific.
In recent work, we measure the contribution of science to the value of patented private-sector inventions (Watzinger and Schnitzer 2019). To quantify the monetary value of patented inventions we rely on the work of Kogan et al (2017). They use abnormal stock-market returns around the date of publication of a patent to determine the value of an invention as seen by the financial market. To isolate the science component of the patent value, we use a metric for how much a patent is based on science. To characterise patents with respect to how close they are to science, we look at citation links between patents and scientific articles, building on Ahmadpoor and Jones (2017). A patent that directly cites a scientific article has a distance of one (D=1), a patent that cites a (D=1)-patent has a distance of two and so on. We then compare within a technology class the values of patents that are based on science with patents that are not to determine the value premium of science.
Science-based patents are more valuable, but also more risky
The resulting estimates for the value of science are shown in Figure 1.
A patent that directly builds on science is on average $2.9 million more valuable than a patent in the same technology class but unrelated to science. This value premium of science decreases monotonically with distance to science and is zero for patents with a distance to science larger than four. However, science-based patents are also more risky than patents not based on science. Figure 2 displays the shares of science and non-science-based patents over the percentiles of the value distribution. If the value distribution were the same for both types of patents, the share at each percentile should be 1%. However, as Figure 2 reveals, science-based patents are more likely to end up in the upper and the lower tails of the value distribution.
The novelty of patents as the missing link
What makes science-based patents particularly valuable? In the history of technology and innovation, inventions are often conceptualised as the outcome of successfully combining ideas, either by combining new ideas or resources or by combining existing ones in a novel way. For example, in A History of Mechanical Inventions, Abbott Payson Usher writes: “Invention finds its distinctive feature in the constructive assimilation of preexisting elements into new syntheses, new patterns, or new configurations of behavior” (Weitzmann 1998). This why we study, first, whether the value of a patent is related to the novelty of its content and, second, whether science contributes to the novelty of the content of a patent.
Following the concept of inventions as a novel combination of ideas or resources, we develop a new measure for patent novelty that is based on the content of the patent and that we make available in the Harvard Dataverse.1 More specifically, we measure how novel the combinations of words used in a patent are. For example, the word “mouse” combined with the word “trap” was used in patents since at least 1870. In contrast, the word “mouse” was combined with the word “display” for the first time in 1981 in the pioneering patents of Xerox. We call a patent “novel” if it contains word combinations that were seldom used in a patent before. Using this novelty measure, we find that patent novelty predicts the value of patents in a very similar way as the science-intensity of patents does. This is shown in Figure 3. The horizontal axis shows the average likelihood that the word combinations in a given patent were used before, adjusted for technology and year. The vertical axis shows the value of the patent. Patents that are more novel (i.e., have less likely word combinations) have a higher value.
How is novelty linked to science? Figure 4 contrasts the novelty of patents that are directly based on science (D=1) with the novelty of patents that are unrelated to science (D=4). As the graphs show, patents based on science are more novel than patents that are not. Taken together, these findings suggest that science-based patents are more valuable because science generates new ideas and new concepts that are useful for innovations in the private sector. Vannevar Bush apparently had a point when he claimed that “Basic science (...) creates the fund from which the practical applications of knowledge must be drawn. (…) [B]asic research is the pacemaker of technological progress” (Bush, 1945).
Understanding how much value science creates for society is fundamental for the case of public science funding. By illuminating the commercial value of science, our study provides a lower bound for its total value for society. Extrapolated to the US economy, we find this lower bound for the additional value created by science for marketplace inventions to be $720 per capita and year. This is about 25% of the total value of patented inventions in the US. Concurrent research also shows that not only the quantity, but also the quality of the science used in patents matters for their value (Poege et al. 2019). Overall, while scientists since Isaac Newton have been known to see further “by standing on the shoulders of giants”, our study suggests that many inventors in the private sector see further by standing on the shoulders of science.
Ahmadpoor, M and B FJones (2017), “The dual frontier: Patented inventions and prior scientific advance”, Science 357(6351): 583–587.
Bikard, M (2018), “Made in academia: The effect of institutional origin on inventors’ attention to science”, Organization Science 29(5): 818–836.
Bush, V (1945), Science, the endless frontier: A report to the President, US Government Publishing Office.
Butler, D (2008), “Translational research: crossing the valley of death”, Nature News 453(7197): 840–842.
Freedman, L P, I M Cockburn and T S Simcoe (2015), “The economics of reproducibility in preclinical research”, PLoS biology 13(6): e1002165.
Goozner, M (2005), The $800 million pill: The truth behind the cost of new drugs, University of California Press.
Hall, B H, A B Jaffe and M Trajtenberg (2001), “The NBER patent citation data file: Lessons, insights and methodological tools”, NBER Working Paper No. 8498.
Harris, G (2011), “Federal research center will help develop medicines”, New York Times, 22 January.
Kline, S J and N Rosenberg (1986), An overview of innovation. the positive sum strategy: Harnessing technology for economic growth, The National Academy of Science.
Kogan, L, D Papanikolaou, A Seru and N Stoffman (2017), “Technological innovation, resource allocation, and growth”, The Quarterly Journal of Economics 132(2): 665–712.
Osherovich, L (2011), “Hedging against academic risk”, Science-Business eXchange 4(15): 416–416.
Poege, F, D Harhoff, F Gaessler and S Baruffaldi (2019), “Science quality and the value of inventions”, arXiv preprint arXiv:1903.05020.
von Hippel, E (1988), The Sources of Innovation, Oxford University Press.
Watzinger, M and M Schnitzer (2019), “Standing on the shoulders of science”, CEPR-Discussion Paper 13766
Weitzman, M L (1998), “Recombinant growth”, Quarterly Journal of Economics 113(2): 331-360.
 The new measure of patent novelty is available for all US patents from 1980 to 2012 in the Harvard Dataverse (https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/L2BB9F)