A responsiveness-based (composite) indicator with an application
to countries’ innovative performance

Cerulli Giovanni
^{National Research Council of Italy}
Ceris -Institute for Economic Research
on Firms and Growth
Via dei Taurini, 19
00185, Roma
g.cerulli@ceris.cnr.it

 

Abstract: The aim of this paper is twofold: on one hand, from a methodological-statistical perspective, it develops a responsiveness-based index for a series of input factors on a specific target variable (assumed to capture the phenomenon the analyst wishes to look at), by means of an extended version of a random coefficient regression approach; on the other hand, it applies this methodology to the case of countries’ innovation performance, where the target variable is the country number of patents (as proxy of “innovativeness”), and where inputs are chosen according to the literature dealing with the measurement of country technological capabilities.

The novelty of the approach presented in the paper regards the possibility of extracting from data a country-specific “reactivity effect” or “responsiveness” (that is, mathematically, a derivative) to each single input feeding into the regression. Thus, the paper provides a promising approach for ranking countries according to their responsiveness to specific inputs, an approach that can be complementary to the analysis on “level” performed, for instance, in the canonical composite indicators’ literature.

As for results on countries’ innovation function, besides a (new) ranking of countries, this approach allows also for testing - in an original and straightforward way - the (possible) presence of increasing (decreasing) returns. Two years are considered and compared, 1995 and 2007, on 42 countries. Our tests conclude that in both years innovative increasing returns are at work, although in 2007 their strength drops considerably compared to 1995. According to a huge literature on the subject (both neoclassical and evolutionary), we conclude that a self-reinforcing mechanism in new knowledge production, absorption and diffusion is at the basis of these results. As for the structural change found between 1995 and 2007, we deem it to depend on the growing globalization of production and innovation processes and on the brilliant growth of some developing countries worldwide, with a remarkable role played – according to our results – by post-communist economies.

 

Keywords: Responsiveness, Country indicators, Random coefficient regression, Innovation function

 JEL-codes: C21, O31, Q01

  

Acknowledgments. A first draft of the present paper was presented at the Joint Research Centre - Econometric and Applied Statistics Unit of the European Commission in Ispra (Italy). I am grateful to Andrea Saltelli, Stefano Tarantola, Paola Annoni and Michaela Saisana for their useful comments and suggestions.