Innovation, productivity and growth in US business services: A firm-level analysis

Abstract

This paper examines the impact of innovation on the performance of US business service firms. We distinguish between different levels of innovation (new-to-market and new-to-firm) in our analysis, and allow explicitly for sample selection issues. Reflecting the literature, which highlights the importance of external interaction in service innovation, we pay particular attention to the role of external innovation linkages and their effect on business performance. We find that the presence of service innovation and its extent has a consistently positive effect on growth, but no effect on productivity. There is evidence that the growth effect of innovation can be attributed, at least in part, to the external linkages maintained by innovators in the process of innovation. External linkages have an overwhelmingly positive effect on (innovator) firm performance, regardless of whether innovation is measured as a discrete or continuous variable, and regardless of the level of innovation considered.

Introduction

In the last decade an increasing body of research has begun to examine the nature, types, and causes of innovation in services. However, there is much less research on the impact of service innovation on business performance, especially at the firm level. As Cainelli et al. (2006) point out, this is partly because of the difficulties involved in obtaining micro-level data, which is well developed in manufacturing but less so in services,¹ and partly because of methodological problems relating to the availability of appropriate indicators to measure innovation activities in services. Metrics which are standard in manufacturing, such as R&D and patents, may be less meaningful in the case of services (Evangelista and Sirilli, 1995; Djellal and Gallouj, 1999; Love and Mansury, 2007).

It is increasingly recognised that models of innovation developed principally for manufacturing may not apply easily to services. For example, the traditional distinction between product and process innovation is less useful in services, which are often processes that cannot be easily disentangled from the outcomes they produce. And the way in which service firms innovate is often different from manufacturing firms. Tether (2005) analyses data from the European Innobarometer, a telephone survey of managers in over 3000 firm, and found substantial differences in the way manufacturing and service firms performed innovation. Service firms were much more likely to regard organisational change as important and to develop innovations in collaboration with customers and suppliers, while manufacturers tended to stress the importance of their in-house R&D and research links with universities. In addition, manufacturers tended to emphasise ‘hard’ strengths such as R&D competence and flexibility of production methods while service providers more frequently stressed ‘soft’ skill such as workforce skills and collaborative interactions.²

An important issue is therefore whether and how the different ways in which service firms perform the process of innovation affect the economic performance outcomes which result from innovation. This is the focus of the present paper, which examines the impact of innovation on the economic performance of a sample of service sector firms. The paper adds to our knowledge of service sector innovation in three ways. First, we use data from the United States: most of the previous studies of the effects of innovation in services have been from Europe (e.g. Cainelli et al., 2004, Cainelli et al., 2006). Secondly, we distinguish between different levels of innovation (new-to-market and new-to-firm) in our analysis, and allow explicitly for sample selection issues. Finally, reflecting the literature which highlights the importance of external interaction in service innovation (Howells and Tether, 2004; Tether, 2005; Kanerva et al., 2006), we pay particular attention to the role of external innovation linkages and their effect on business performance.