Abstract
In this paper, we shed new light on the links between firm-level innovation and growth. We introduce data that capture a difficult-to-observe aspect of firms' innovative activity – new product/service launches – at scale. We show that our novel measures complement existing innovation metrics. We build a simple framework covering firm-level innovation, launches and revenue productivity. Then, we show positive linkages between past patenting and launches and between launches and performance for a large panel of small and medium-sized enterprises (SMEs) in the UK. We go on to explore the roles of age, size, industry and product/service quality in these relationships. A subset of SMEs with high-quality launches explains our results.
1. Introduction
There is decades’ worth of research exploring the role of innovation in explaining economic performance. Endogenous growth theory helps explain country-level innovation-productivity linkages (Lucas, 1988; Romer, 1990). Schumpeterian frameworks highlight the roles of entrepreneurial entry, competition and factor reallocation (Schumpeter, 1962; Aghion and Howitt, 1992). They also provide firm-level microfoundations linking R&D activity, innovation and new products and services (see Akcigit (2017) for a review). Evolutionary perspectives emphasise that firms’ capabilities vary greatly. Firms’ resources accumulate over time, and it is difficult to shift them (Nelson and Winter 1982, Dosi et al 2000). This variation in firms’ capabilities makes predictors of average firm performance difficult to identify (Nightingale and Coad, 2013).
7. Conclusions
A vast field of literature explores the links between innovation and economic performance (Romer, 1990; Aghion and Howitt, 1992; Coad, 2009; Akcigit, 2017). Four streams of empirical work unpack connections at the firm level. An established set of studies uses R&D, patents and innovation surveys; newer analyses use trademarks, text-based measures or product-level data. However, this body of work has two constraints: the informality of much innovative activity (Hall et al, 2014) and practical limits in processing richer text or product-level information. Our paper makes three practical and empirical advances on this literature. First, we develop novel product-level innovation metrics that extend existing studies. Second, we show that our new measures complement existing, formal IP metrics. Third, we find positive links between SME launch activity and revenue per worker. Importantly, we also find that industry, size, IP strategy and launch quality differences moderate our main results.
