How European public sector agencies innovate: The use of bottom-up, policy-dependent and knowledge-scanning innovation methods

A. Arundel*, L. Casali, H. Hollanders

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Factor and cluster analysis are used to identify different methods that public sector agencies in Europe use to innovate, based on data from a 2010 survey of 3273 agencies. The analyses identify three types of innovative agencies: bottom-up, knowledge-scanning, and policy-dependent. The distribution of bottom-up agencies across European countries is positively correlated with average per capita incomes while the distribution of knowledge-scanning agencies is negatively correlated with income. In contrast, there is no consistent pattern by country in the distribution of policy-dependent agencies. Regression results that control for agency characteristics find that innovation methods are significantly correlated with the beneficial outcomes of innovation, with bottom-up and knowledge-scanning agencies out-performing policy-dependent agencies. © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license.
Original languageEnglish
Pages (from-to)1271-1282
Number of pages12
JournalResearch Policy
Volume44
Issue number7
DOIs
Publication statusPublished - 2015

JEL classifications

  • o32 - Management of Technological Innovation and R&D
  • o31 - Innovation and Invention: Processes and Incentives

Keywords

  • Innovation outcomes
  • Innovation survey
  • Public sector innovation
  • Taxonomy of innovation
  • Cluster analysis
  • Surveys
  • Bottom up
  • European Countries
  • Factor and cluster analysis
  • Innovation methods
  • Per capita income
  • Public sector
  • Public sector agencies
  • Scanning

Fingerprint

Dive into the research topics of 'How European public sector agencies innovate: The use of bottom-up, policy-dependent and knowledge-scanning innovation methods'. Together they form a unique fingerprint.

Cite this