METHODOLOGICAL DEBATE ON THE ‘REGIONAL INNOVATION SCOREBOARD’.
METHODOLOGICAL DEBATE ON THE ‘REGIONAL INNOVATION SCOREBOARD’.
a lower impact on private R&D and business innovation in environments where companies' capacity to absorb knowledge is lower (Castel- lacci and Natera, 2013), while a positive inter - action between public expenditure and private R&D expenditure is observed in environments where companies have high capacity to absorb knowledge. Despite recognising their impor - tance, the RIS does not go further in assigning different relative weights to each indicator, but rather gives all variables the same relative weight, which implies, in operational terms, that the same relative importance is assigned to all factors of the system, regardless of its context (i.e. uniform weights for disparate territories). This trade-off between ensuring consistency in the relative weights in order to compare the territories being analysed and capture the differences between them (i.e. not all variables have the same relative importance in each territory) is particularly complex in the European sphere, with its 27 countries and 240 regions. The decision to assign identical weights to all variables and all territories (as the European Commission does in the SII/RIS calculation) is undoubtedly one way to remain neutral in this debate. However, this arbitrary position does not clarify when the differences in the positions of the two regions in the final inno - vation performance ranking can be attributed to this random decision or to significant differ - ences in the performance of their associated systems. In this first analysis, we will try to answer whether it can be concluded that the differences in the innovation performance of the European regions measured using the RIS synthetic index are due to the relative weight of its variables, or whether some regions always perform better than others, regardless of the weights assigned to these variables. To shed light on this first debate, we will study the sta - bility of the RIS in the event of changes in the weights of the variables, and propose a robust ranking for these weights.
Finally, a third debate envisages the possibility of interactions (i.e. complementarity effects (positive), or substitution effects between factors (negative)) between the indicators that measure an innovation system’s different activities and functions (Tarabusi and Palazzi, 2004); Lafuente et al. , 2021). The synthetic indices calculation assumes perfect substitut- ability between the components of an innova- tion system. However, as seen in the previous paragraph, considering that inputs can be sub- stituted by outputs could be problematic from an economic and public policy perspective. Furthermore, the fact that there are comple- menting factors in the innovation system (cap- tured in the RIS variables) could lead to “bottle- necks”, with development being slowed down by certain aspects of the innovation system (Ács et al. , 2014). These bottlenecks represent the weakest functions of the system (i.e. the function that acts as a constraint for the whole system), and therefore penalise system perfor- mance. Identifying these bottlenecks ensures poilcy-makers in each territory count on clear guidelines in terms of the interventions re- quired in order for the overall system to benefit from public action (i.e. additionality).
A second debate, not completely unrelated to the first one, refers to the problem of adding variables that are not designed to measure exactly the same phenomena (Edquist and Za - bala-Iturriagagoitia, 2015; Edquist et al. , 2018). Although the RIS indicators are undoubtedly related to innovation activity, there is a consen- sus that they measure different aspects of the innovation system. One of the most commonly used classifications (although not the only one) of the indicators in the RIS divides them into three groups: facilitators, business activities, and results (European Union, 2016). From the conventional perspective of the production function, the first two groups can be seen as “inputs” (internal and external to the company, respectively) of the regional innovation sys- tem's innovation activity, while the third one is for the “outputs” produced by the associ- ated system. Despite these clear differences between types of indicators, the synthetic index offered by the RIS does not differentiate between input and output variables, attributing the same relative weight to all of them. This methodological decision could lead to unrea- sonable situations from the point of view of economic analysis. For example, let us imagine a hypothetical context where two regions find themselves in opposing situations: the first has very high values for input indicators and very low values for output indicators, while the second shows the opposite, i.e. very few inputs and many outputs. Although the RIS synthetic index would attribute the same performance level to both regions (since they would have the same arithmetic mean), we will all agree that both cases represent two opposing examples of a system working efficiently in generating innovation results. In a second analysis, we will therefore develop a methodological approach to analyse the efficiency of regional innovation systems in Europe, allowing us to see, from this perspective, which territories make better use of the investment in their respective sys- tems, indicating that their respective innovation systems work better.
22
23
Made with FlippingBook - Online Brochure Maker