Biased Policy Professionals

Abstract

Although the decisions of policy professionals are often more consequential than those of individuals in their private capacity, there is a dearth of studies on the biases of policy professionals: those who prepare and implement policy on behalf of elected politicians. Experiments conducted on a novel subject pool of development policy professionals (public servants of the World Bank and the Department for International Development in the UK) show that policy professionals are indeed subject to decision-making traps, including the effects of framing outcomes as losses or gains, and, most strikingly, confirmation bias driven by ideological predisposition, despite having an explicit mission to promote evidence-informed and impartial decision making. These findings should worry policy professionals and their principals in governments and large organizations, as well as citizens themselves. A further experiment, in which policy professionals engage in discussion, shows that deliberation may be able to mitigate the effects of some of these biases.

How Research Affects Policy: Experimental Evidence from 2,150 Brazilian Municipalities

“…While most such research studies beliefs in lay populations to identify systematic biases and heuristics (see Benjamin (2018) for a review), we add to the smaller body of work studying the beliefs of experts such as central bankers (Malmendier et al, 2017), academics (DellaVigna & Pope, 2018, and judges (Chen et al, 2016). 10 In this sense, our study is most closely related to Banuri et al (2017), Nellis et al (2019), and Vivalt & Coville (2019), who study how the beliefs of policy professionalsprogram officers, aid-agency workers, and government officials-respond to research findings and new data. Complementing their work, we study the extent to which academic research changes elected heads of government's beliefs; the extent to which they themselves value access to research, and how policy adoption ultimately responds to research findings.…”

“…11 Banuri et al (2017), Nellis et al (2019), and Vivalt & Coville (2019) study the belief-formation of (mostly U.K. and U.S.-based) policy professionals. Like Nellis et al (2019), we find evidence of fairly sophisticated processing of information, with little evidence of obvious deviations from Bayesian learning.…”

Motivating Bureaucrats With Non-Monetary Incentives When State Capacity Is Weak: Evidence From Large-Scale Field Experiments in Peru

“…Some developing countries are starting to The use of behavioral insights is particularly relevant, given increasing evidence suggesting that civil servants are affected by the same set of behavioral biases that have been documented by behavioral economists in other populations. 7 Civil servants exhibit biases in information processing, value assessment, and decision-making (Banuri et al, 2017), despite special provisions that exist in legislation that regulate the behavior of civil servants regarding qualities such as objectivity and impartiality that they are expected to follow. 8 Despite this, little has been done to incorporate these behavioral dimensions in efforts to improve civil servants' performance.…”

“…International organizations like the World Bank and the United Nations also have similar principles. See Banuri et al (2017) for more details.…”

“…This paper is also related to a very new literature that explores the role of behavioral biases in the case of civil servants. Recent papers have systematically documented the existence of present bias (Andreoni et al, 2016), status quo bias (Celhay et al, 2015), confirmation bias, framing effects, inattention bias and optimism bias (Hallsworth et al, 2018 andBanuri et al, 2017) among bureaucrats. Rather than documenting the existence of such biases, this paper tests a set of strategies to deal with them.…”