Behavioral Public Administration ad fontes: A Synthesis of Research on Bounded Rationality, Cognitive Biases, and Nudging in Public Organizations

• DOI: http://doi.org/10.1111/puar.12994
• Author: Paola Cantarelli,Nicola Bellé,R. Paul Battaglio,Paolo Belardinelli
• Published date: 01 May 2019
• Date: 01 May 2019

Public Administration Review,

Vol. 79, Iss. 3, pp. 304–320. © 2018 by

The American Society for Public Administration.

DOI: 10.1111/puar.12994.

304 Public Administration Review • May | J une 2 019

R. Paul Battaglio, Jr.

University of Texas at Dallas

Paolo Belardinelli

Bocconi University

Nicola Bellé

Sant’Anna School of Advanced Studies

Paola Cantarelli

Sant’Anna School of Advanced Studies

Behavioral Public Administration ad fontes:

A Synthesis of Research on Bounded Rationality,

Cognitive Biases, and Nudging in Public Organizations

Abstract: This article provides a comprehensive overview of how policy makers, practitioners, and scholars can

fruitfully use behavioral science to tackle public administration, management, and policy issues. The article

systematically reviews 109 articles in the public administration discipline that are inspired by the behavioral sciences

to identify emerging research trajectories, significant gaps, and promising applied research directions. In an attempt

to systematize and take stock of the nascent behavioral public administration scholarship, the authors trace it back to

the seminal works of three Nobel Laureates—Herbert Simon, Daniel Kahneman, and Richard Thaler—and their

work on bounded rationality, cognitive biases, and nudging, respectively. The cognitive biases investigated by the

studies reviewed fall into the categories of accessibility, loss aversion, and overconfidence/optimism. Nudging and choice

architecture are discussed as viable strategies for leveraging these cognitive traps in an attempt to alter behavior for the

better, among both citizens and public servants.

Evidence for Practice

• Understanding how public decisions may predictably go wrong is imperative to improve the architecture of

public organizations and services.

• Cognitive biases systematically affect public policy and management decisions.

• Behavioral science illuminates the gap between how people should behave and how they actually behave, thus

moving beyond traditional models of full rationality in decision making.

• Nudging and choice architecture represent viable tools for policy makers and public managers interested in

altering the behaviors of citizens and public employees, respectively, for individual and societal better.

“A choice architect has the responsibility for

organizing the context in which people make

decisions ... there is no such thing as a ‘neutral’

design.”

—Richard H. aler and Cass R. Sunstein (2008, 1).

Decisions in the public sector are often

shaped by a complex array of forces. This

is especially true in the fluid environment

of the information age, when public managers are

inundated with countless challenges (e.g., Kelman,

Sanders, and Pandit 2016). Ready access to data,

adaptable technology, and an ever-combative political

environment contribute to the complexity of decision

making in the public sector. Taking action on public

policy means public managers must overcome not

only these complexities in their environment but also

their own cognitive limitations and moral impasses.

Understanding how supposedly irrelevant factors of

choice architecture may alter public decision making

in predictable ways (Thaler 2017; Thaler and Sunstein

2008) is an increasingly germane topic for further

research (Gordon, Kornberger, and Klegg 2009;

Kelman, Sanders, and Pandit 2016; Moynihan, Herd,

and Harvey 2014; Vlaev et al. 2016). Influencing

public managers’ decision processes from a more

informed assessment of cognitive biases and libertarian

paternalism has the potential to improve effectiveness

through strategic choices that shape goal attainment.

A more robust analysis of the micro-level foundations

(e.g., Jones 2003) of public decision making is timely,

given the numerous and often insurmountable

complex influences facing public managers (Kelman,

Sanders, and Pandit 2016). Behavior modification

techniques offer a better understanding of how we

might influence decision making through heuristics

that nudge public managers in ways that result

in outcomes that are more favorable. Structuring

decision making in ways that positively influence

cognitive biases has the potential to moderate

complexity in the public sector environment,

subsequently reducing learning, psychological, and

Nicola Bellé is assistant professor in the

Institute of Management of the Sant’Anna

School of Advanced Studies, Pisa, Italy.

His research focuses on behavioral public

administration and management.

E-mail: nicola.belle@santannapisa.it

Paolo Belardinelli is a PhD candidate

in public policy and administration at

Bocconi University, Milan, Italy. His research

focuses on behavioral public policy and

administration.

E-mail: paolo.belardinelli@unibocconi.it

R. Paul Battaglio, Jr. is professor of

public and nonprofit management at the

University of Texas at Dallas and co–editor

in chief of

Public Administration Review

.

His research interests include public human

resource management, organization

theory, behavioral public administration,

comparative public policy, and research

methods.

E-mail: battaglio@utdallas.edu

Paola Cantarelli is a postdoc at the

Sant’Anna School of Advanced Studies, Pisa,

Italy. Her research focuses on behavioral

management and work motivation in

mission-driven organizations.

E-mail: paola.cantarelli@santannapisa.it

Research Article

Behavioral Public Administration ad fontes 305

compliance costs (Cantarelli, Bellé, and Belardinelli 2018). From

a practical standpoint, such cognitive strategies have the potential

to nudge public managers in a direction that improves individual

performance, overall productivity, and informs evidence-based

policy (Clement 1987; Vlaev et al. 2016).

A robust discussion has begun on decision-making biases in public

management, administration, and policy. Most scholars have

investigated how citizens make informed assessments of government

policies (see, e.g., Andersen and Hjortskov 2015; Geys and Sørensen

2017; Grosso, Charbonneau, and Van Ryzin 2017; Jilke, Van Ryzin,

and Van de Walle 2016; Marvel 2015b; Olsen 2017b). Fewer

studies, however, have reviewed the decision processes of public

managers (e.g., Bellé, Cantarelli, and Belardinelli 2017) and policy

makers (e.g., Moynihan and Lavertu 2012). This research supports

the need for continued systematic literature reviews of cognitive

biases as a means for providing useful information for how we might

affect individuals’ estimates, judgments, preferences, and behaviors.

Indeed, recent work conducted by the Organisation for Economic

Co-operation and Development (OECD) has highlighted the

usefulness of behavioral and cognitive sciences toward sustainable

public administration. Analyzing 159 case studies from 60

public bodies in 23 states and 2 international institutions, the

OECD (2017) reported that attempts to use behavioral insights

to inform policies are underway across numerous policy areas,

including, consumer behavior, education, energy, environment,

finance, health and safety, labor market, service delivery, taxes, and

telecommunications.

However, seeing through a glass darkly may be a more apt

assessment of the broader contribution of behavioral sciences to

public administration. A systematic review at this time in the field

provides an opportunity for moving beyond how systematic errors

work to pinpointing which have proven useful in our research and

the means for moving toward more fruitful heuristic interpretations.

Thus, our study aims to provide greater clarity in an effort to

avoid wrong assumptions about the use of choice architecture to

curtail biases in decision making. Specifically, this article provides

a research synthesis of the public administration, management,

and policy studies linked to the work of the three Nobel Laureates:

Herbert Simon, Daniel Kahneman, and Richard Thaler. By looking

at their seminal works on bounded rationality, cognitive biases,

nudging and choice architecture, our comprehensive synthesis traces

behavioral public administration scholarship back to its sources (ad

fontes) and highlights promising research directions and practical

implications. Simon’s theory of bounded rationality provides a

more circumspect assessment of factors that limit rational choices

by individuals. Kahneman’s work, which is empirically driven and

descriptive in nature rather than normative, unveils a number of

heuristics that people adopt to make difficult decisions and a series

of cognitive biases that systematically lead us astray. Thaler’s nudge

theory systematizes the use of behavioral science to influence high-

stake choices through low-powered incentives, thus paving the way

toward libertarian paternalism.

Bounded Rationality, Cognitive Biases, and Nudging

Several decades of behavioral research have buttressed Simon’s

(1947, 1956) claim that we are endowed with bounded rationality

and, in the face of information that is either intractable or

incomplete, tend to find solutions that are adequate rather

than optimal (e.g., Olsen 2015b). Simon’s conceptualization

of this “satisficing” strategy paved the way for later work aimed

at providing a more realistic representation of human decision

processes compared with postulates by rationalistic models, such

as Bernoulli’s ([1738] 1954) expected utility theory. In particular,

Kahneman and Tversky’s prospect theory (1979) has extended

Bernoulli’s utility theory along several dimensions. First, they

demonstrate that utility does not depend exclusively on the amount

of wealth one has at any given time, but rather on whether that

wealth is the result of a gain or a loss from a particular reference

point—an irrelevant supposition according to rational choice

models. Another supposedly irrelevant factor—whether the same

piece of information is framed in terms of prospective losses or

in terms of prospective gains—makes individuals risk seeking,

thus violating the tenet of risk aversion that underpins expected

utility theory (Tversky and Kahneman 1981). The fact that these

deviations from rational decision making tend to be systematic—

hence predictable—under specific conditions (e.g., Kahneman

2011) brings with it the possibility of strategically exploiting

cognitive biases, for better or worse (Thaler and Sunstein 2008).

Rational choice models and behavioral theories portray two different

types of agents, which Thaler and Sunstein (2008) identify as “Econs”

and “Humans,” respectively. Econs do not follow fashion and make

unbiased estimates. More precisely, their estimates are not necessarily

perfect, because this would convey omniscience; rather, when their

judgments are wrong, they are not systematically so in a predictable

direction. Conversely, Humans are social animals (i.e., they are

influenced by the behaviors of others) and make predictable errors.

Kahneman (2011) traces systematic patterns of deviation from

rational decision making to system 1 thinking, which, along with

the perceptual system, presides over intuition. The perceptual

system processes percepts, deals with stimulations in the moment

during which they are administered, and is stimulus-bound.

System 1, instead, deals with conceptual representations; can refer to

past, present, and future; and is evoked by language. Both are fast,

automatic, effortless, slow learning, and associative. They produce

impressions of the attributes of objects that are to be evaluated

automatically, involuntary, and even without the need of being

verbally overt. Reasoning, instead, happens in system 2 thinking,

which is slow, controlled, effortful, flexible, and rule governed. Like

system 1, system 2 uses cues; can refer to past, present, and future;

and is evoked by language. System 2 thinking produces judgments

based on either intuitions or deliberate reasoning. Judgments

are an intentional and explicit process, regardless of whether

they are verbally expressed or not (e.g., Kahneman 2002, 2011).

Overall, individuals make decisions through one of the following

mechanisms, in order of likelihood: (1) an intuitive judgment is

elicited and endorsed by system 2; (2) an intuitive judgment is

evoked and serves as an anchor to be adjusted by system 2 in light

of other situational features; (3) a deliberate judgment is created

by system 2 because no intuitive judgment is accessible; or (4) a

deliberate judgment is generated by system 2 because the intuitive

judgment that came to mind is identified as incorrect. System 1 and

2 are also known as automatic and reflective systems, respectively

(Thaler and Sunstein 2008).