Cumulative versus punctuated patterns of R&D learning, productivity and R&D persistence

This paper analyses how R&D persistence is affected both by direct learning determinants and indirect determinants through enhanced productivity. Advancing previous research, we allow for three distinct forms of R&D experience through which learning occurs: the firm’s current within-spell learning; learning derived from the firm’s cumulative R&D experience; and the potential detrimental effect of punctuated learning, where a firm’s cumulative R&D experience is divided into multiple R&D spells. Our findings suggest markedly different learning processes depending on firm size and technology. Small and large firms, and firms operating in low and med-tech sectors benefit from cumulative previous R&D but medium-sized and high-tech firms do not. However, a history of fragmented or intermittent R&D is negatively associated with R&D persistence for all firm sizes, but not for low-tech firms. While cumulative R&D experience improves productivity for all firms, there is evidence of an indirect (productivity) association of R&D experience on subsequent R&D persistence only for SMEs and low-tech firms.