Over the past decades, governments have made major investments in systems and mechanisms aimed to hold public sector organizations accountable (Peters 2014). The reported effects of these systems and mechanisms, however, are mixed (Pollitt & Bouckaert 2011). Why accountability mechanisms produce positive effects on quality of decision making and performance in some instances, and negative effects in others, is only partially understood in the field of public administration. In the behavioral sciences, however, there is a long experimental tradition of accountability research investigating causal effects of accountability on behavior and decision making (Lerner & Tetlock 1999, Hall et al. 2015). The insights from this research can be of great value for public administration research and practice, providing advice on how to model accountability mechanisms in the public sphere, as well as directions for further research in the field.
This paper will present a systematic review of the experimental literature on accountability in the behavioral sciences. It will summarize the findings of 265 experiments exploring the effects of accountability mechanisms, presented in 210 studies published since 1970. These findings are deductively organized in four large themes: effects on decision-making, effects on behaviour, effects on outcomes, and effects of accountability mechanism characteristics. The findings from this exercise will then be discussed in relation to the public administration literature on accountability. The results of this systematic review will show that the criticism of accountability in the public administration literature might have been overstated, since the behavioral disciplines demonstrate primarily desirable effects of accountability on decision-making and behavior. We will conclude with a discussion of the implications of these findings for public administration, and an agenda for future research.