A taxonomy of innovation: How do public sector agencies innovate? Results of the 2010 European Innobarometer survey of public agencies
Arundel, A and Hollanders, H, A taxonomy of innovation: How do public sector agencies innovate? Results of the 2010 European Innobarometer survey of public agencies, European Commission, Maastricht, The Netherlands, pp. 1-38. (2011) [Report Other]
We use factor analysis followed by cluster analysis to identify different innovation strategies or methods that European public sector agencies use to innovate. The methodology identifies three methods. The first, used by 30.4% of the responding agencies, consists of a policy driven approach to innovation, with innovation implemented in response to mandated changes in the agency budget, new laws or regulations, new policy priorities, or the mandated introduction of new services. The second method, used by 34.3% of the agencies, is characterised by a ‘bottom up’ approach to innovation and is the opposite of the policy driven approach. Bottom up innovators have active management support
for innovation and have implemented several policies to encourage the development of innovative ideas and tests of their efficacy. The third method, external knowledge innovators, is used by 35.3% of the agencies. These agencies report high levels of barriers to innovation and seek to overcome them by drawing on sources outside of the agency for assistance with developing an innovation.
The effect of each innovation method on outcomes was analysed using descriptive analyses and regression techniques. The results show significantly poorer outcomes for policy driven innovators than for the other two innovation methods. A significantly lower percentage of policy driven innovators report each of five benefits from service innovations and four benefits from process/organisational innovations. For almost all benefits from innovation, the poorer performance of policy driven innovators remains after controlling for the effect of agency size, geographic area of responsibility (local, regional, national), the use of innovation groups and training to support innovation, and national per capita income. The exception is improved user satisfaction from innovative services, for which there is no significant difference between policy driven innovators and the other two innovation