A cross-country comparison of innovation efficiency

The purpose of this paper is to highlight the weakness of innovative activities and guide the improvement of innovation efficiency at country-level through carefully comparing innovation efficiency across countries. Following the conceptual framework which divides innovation processes into knowledge production process (KPP) and knowledge commercialization process (KCP) and applying dual network-DEA models, this paper tries to take economic benefit of R&D outputs into account. Moreover, we construct the production frontier of the innovation processes and two component processes under different assumptions (e.g., constant returns-to-scale, variable returns-to-scale and non-increasing returns-to-scale) for 35 countries over the period 2007-2011. Based on the production frontier, we do not only estimate technical efficiency and scale efficiency for each country but also investigate and verify whether returns-to-scale of each country are decreasing or increasing. Furthermore, we add together the radial movement and the slack movement to acquire input redundancy. We decompose the input redundancy into two parts: redundancy caused by technical inefficiency (R_TI) and redundancy caused by scale inefficiency (R_SI), and carry out a detail analysis of the input redundancy. We find specific circumstances of inefficiency and redundancy vary with the different countries' characteristics and development stages. Moreover, innovation efficiency statistically mainly depends on the KCP efficiency. In particular, the study reveals that China suffers scale inefficiency is attributed to insufficient macro-level coordination, malfunctioning funding system, and flawed evaluations and incentives. Finally, public policy implications are proposed for the inefficient countries.