Research and development (R&D) beyond manufacturing: the strange case of services R&D

The share of business research and development (R&D) expenditure stemming from the services sector of the economy has been growing rapidly in many (though not all) OECD countries, according to official statistics. The same data sources also indicate, however, that services contribute less to R&D than would be expected given their large shares of employment and output in national economies. In part, incomplete sampling of services in R&D surveys may lead to some underestimation of their R&D activity, but this is unlikely to account for their apparently poor performance. This paper draws on an analysis of existing statistical sources, and on interviews and workshops with service firms' managers. It finds difficulties associated with the ways in which R&D has been operationalised in survey questions. Examination of survey questionnaires suggests that the formulation of these questions, focusing on technological R&D and ruling out much social scientific R&D, disproportionately reduces the reporting of R&D by service firms. But beyond this, the R&D concept itself has some problematic features where it comes to documenting innovation in service firms. This was investigated through a programme of interviews and workshops with service firms, where a lack of familiarity with the R&D concept and R&D management practices was found to be commonplace. R&D performance and innovation activities vary across services of different sorts, even though most service subsectors appear to be low R&D investors. It is thus important to examine services' innovation patterns and processes, to establish what types of R&D-like activity are underway in these subsectors. While some modification in R&D measurement would be desirable to capture services' activities, effort to understand the non-R&D elements of services innovation is also important - for management and for policy (given that governments are seeking to create R&D incentives and targets for services and other sectors). There are grounds for expecting (a) services' share of business R&D to continue to grow; (b) this share to continue to be well below what would be expected from the prevalence of services in economic activity, and (c) for many services' innovation to continue to rely heavily on sources that are not directly associated with R&D.