Mapping and contesting peer selection in digitalized public sector benchmarking

Wai Fong Chua, Johan Graaf, Kalle Kraus

Research output: Contribution to journalArticleResearchpeer-review


This paper investigates the influence of digitalization on different modes of peer selection in public sector benchmarking. We do so in the context of a field study of the impact of “Kolada”—a digital database and benchmarking device comparing the performance of Swedish municipalities. We find that the municipal quality controllers often used algorithmically selected peer groups to identify “pure” performance gaps for a range of performance indicators. Politicians, departmental managers, and the citizenry, however, continued to prefer benchmarking against neighboring municipalities. Drawing on Gieryn's concept of cultural cartography, differences in peer selection are characterized as a form of credibility contest between digitally generated and local maps. Our paper contributes to the literature in three main ways. First, we demonstrate how peer selection involves a mutual interplay between new digitally generated, abstract maps of performance and local cartographic legacies sustained by complex social attachments. Second, our paper illustrates the importance of often overlooked social ties informing processes of peer selection, highlighting the importance of professional ties, neighborly familiarity, and affective relations. Third, our paper characterizes the power of “native truths.” More generally, our paper indicates the epistemic authority of digital “truths” is contestable and may be resisted. Ultimately, the coexistence of “old” and new epistemic maps confers choice, which contributes to the legitimacy of new technologies enabling digitalized benchmarking to persist in shifting and locally meaningful ways.

Original languageEnglish
Number of pages29
JournalFinancial Accountability and Management
Publication statusAccepted/In press - 2021


  • benchmarking
  • cultural cartography
  • digitalization
  • local government performance
  • peer selection

Cite this