Framing Effects under Different Uses of Performance Information: An Experimental Study on Public Managers

Abstract

Combining insights from public administration, accounting, and psychology, this article explores the microprocesses by which public managers use performance information, investigating whether the type of performance information use and the request to justify decisions affect the way in which information is processed. The study draws on data from a series of artifactual survey experiments with Italian municipal executives. Findings show that managers process information differently under ex post rather than ex ante performance information uses. More specifically, managers are more likely to be subject to framing bias under ex post than under ex ante uses of performance information. This interaction seems to be robust when subjects are asked to provide justification for their decisions. Evidence for Practice• Managers will process performance information differently depending on its use. • Managers are less prone to framing bias under ex ante uses of performance information (e.g., allocation of resources) than under ex post uses (e.g., performance evaluation). • Asking managers to provide justification for their decisions may not be sufficient to reduce framing effects. • Increased attention should be given to avoid framing effects, especially in ex post uses of performance information, as this could affect the fairness of evaluation processes, performance-related sanctions, and rewards.

The role(s) of accounting and performance measurement systems in contemporary public administration

“…More recently, performance measurement, and especially the use of performance information, has also attracted significant attention in behavioural public administration research, where scholars have embraced insights from behavioural science (and psychology in particular) in exploring the effects of performance information on individuals' perceptions, use and decision-making behaviours. Mainly applying experimental designs, these studies provide new insights into how individuals process performance data (see Nielsen and Baekgaard 2013;James and Mosley 2014;Olsen 2015Olsen , 2017Barrows et al 2016;Belardinelli et al 2018;Moynihan 2018). While there has been a strong focus on exploring a range of cognitive biases (e.g., negativity bias, anchoring effects) related to the presentation/ processing of performance information (e.g., Belardinelli et al 2018;Moynihan 2018;Battaglio et al 2019), scholars have also started to look at the micro-foundations of the role of broader organizational and institutional factors, such as institutional pressures (Nielsen and Moynihan 2017;Bellé et al 2019), in influencing the use of performance information in decision-making.…”

Ministerial Leadership and Endorsement of Bureaucrats: Experimental Evidence from Presidential Governments

“…The importance of agency heads’ leadership skills lies in the fact that they are not only political appointees from the president's perspective but also managers from the perspective of civil servants. Research in public administration shows that the role of agency heads has proven consequential in improving agency performance (Andrews and Boyne ; Balla and Gormley ; Belardinelli et al ; Boyne ; Boyne et al ; Wolf ) as well as in changing agency policies and influencing policy outcomes (Connolly ; Lewis ; Scholz and Wood ; Wood ; Wood and Waterman , ). Agency heads may interpret the vague and sometimes conflicting laws passed by the legislature and translate them into policy; their decisions about budget requests to the legislature, rulemaking, personnel, and the allocation of resources inside the agency can significantly influence policy; they monitor bureaucratic activity; and they can also improve bureaucratic performance by taking actions that advance their agencies’ long‐term interests and foster a sense of cooperation among agency managers and employees (Lewis , 7; Balla and Gormley , 282; Wolf ).…”

When Are Nudges Desirable? Benefit Validity When Preferences Are Not Consistently Revealed

“…As is now well known, subtle biases can shape perceptions of job candidates and the assessment of their performance. In particular, whether framings are positive or negative can have substantial effects on the ex post interpretation of performance data (Belardinelli et al ). Being aware of the sources of possible biases, if not explicitly countering them with nudges, is thus potentially important for fair and effective personnel systems.…”