آیا چرخه های تجاری بر ثبت اختراع تاثیر می گذارد؟ شواهد از دفتر ثبت اختراع اروپا  Do business cycles affect patenting? Evidence from European Patent Office filings

1. Introduction Two key issues motivate this paper: do business cycles affect patenting behavior and if so, is patenting pro-cyclical or countercyclical? Earlier, Griliches (1990) had observed the effects of oil shocks during the 1970s on patent applications. Likewise, the Great Recession of 2008 and 2009 has spurred interest in the impact of cyclical shocks on patenting.1 The relationship matters to institutional organizations that operate the global patenting system, such as national and regional patent offices, and to industries that provide complementary services to the patenting community; for example, legal, translation, and consulting services. Cyclical shocks affect the ability of these organizations to forecast accurately for purposes of planning and budgeting. Such shocks can therefore affect the supply of services and resources for patent procurement. Cyclical shocks can * Corresponding author. E-mail addresses: phingley@epo.org (P. Hingley), wgp@american.edu (W. Park). 1 See, for example, Bertenrath et al. (2011), European Commission (2011), Guellec and Wunsch-Vincent (2009), OECD (2009), and World Intellectual Property Organization (2010). also affect the demand side by influencing the investment and marketing decisions of firms and other potential patentees. The resulting imbalances in supply and demand could thus have repercussions for the nature and direction of innovation and commercialization.2 Much of the existing literature studying the determinants of patents and patenting propensity has not taken into account the role of cyclical shocks.3 Patents and innovation are often studied under the branch of economic growth theory, as drivers of long run productivity and technological change, where the emphasis is more on structural determinants and trend factors than on cyclical influences. Furthermore, innovation is viewed by many as being driven by longer term considerations, given that the duration of innovation projects is often longer than that of market cycles (see Heger, 2004). On the other hand, short run resource constraints may be binding for some innovators as patenting is costly, and the costs of procurement are often incurred upfront before a patented invention is exploited commercially. The literature on the relationship between innovation and business cycles is very limited, as will be surveyed in the next section. Few academic studies exist, focusing mostly on the input side of innovation – namely, research and development (R&D) expenditures. Research on the effects of business cycles on patenting is rarer. Furthermore, previous work has not formally derived and used a measure of business cycles, and instead has employed proxy variables, such as credit constraints or sales declines (although business cycles also include sales booms), or has compared conditions before and/or after an event, such as the Great Recession of 2008–2009. In this paper, we derive business shocks using standard filtering methods, and allow them to vary by country. A key novelty of our paper is that we apply our business cycle model to forecasting patent filings. This would especially be useful to the supply side we mentioned earlier. Failure to anticipate fluctuations leads these institutions and service providers to allocate resources poorly. Improved forecasting can lead to more accurate budgeting and greater cost effectiveness in services, and these increased efficiencies should potentially raise social welfare.