The Definition of Nudge and Libertarian Paternalism: Does ... · seller titled Nudge: Improving Decisions About Health,Wealth,andHappiness.Beforehandmostfa-mousfromMontyPython’ssketch“nudge,nudge,

RoskildeUniversity

The Definition of Nudge and Libertarian PaternalismDoes the Hand Fit the Glove?

Hansen, Pelle Guldborg

Published in:European Journal of Risk Regulation

Publication date:2016

Document VersionPublisher's PDF, also known as Version of record

Citation for published version (APA):Hansen, P. G. (2016). The Definition of Nudge and Libertarian Paternalism: Does the Hand Fit the Glove?European Journal of Risk Regulation, 7(1), 155-174. http://www.lexxion.de/pdf/ejrr/Pelle_2016_01.pdf

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain. • You may freely distribute the URL identifying the publication in the public portal.

Take down policyIf you believe that this document breaches copyright please contact [email protected] providing details, and we will remove access to thework immediately and investigate your claim.

Download date: 27. Feb. 2021

http://www.lexxion.de/pdf/ejrr/Pelle_2016_01.pdf

EJRR 1|2016 155The Definition of Nudge and Libertarian Paternalism

The Definition of Nudge and LibertarianPaternalism: Does the Hand Fit the Glove?

Pelle Guldborg Hansen*

In recent years the concepts of ‘nudge’ and ‘libertarian paternalism’ have become populartheoretical as well as practical concepts inside as well as outside academia. But in spite ofthe widespread interest, confusion reigns as to what exactly is to be regarded as a nudgeand how the underlying approach to behaviour change relates to libertarian paternalism.This article sets out to improve the clarity and value of the definition of nudge by reconcil-ing it with its theoretical foundations in behavioural economics. In doing so it not only ex-plicates the relationship between nudges and libertarian paternalism, but also clarifies hownudges relate to incentives and information, and may even be consistent with the removalof certain types of choices. In the end we are left with a revised definition of the concept ofnudge that allows for consistently categorising behaviour change interventions as such andthat places them relative to libertarian paternalism.

I. Introduction

“… ‘Soft Paternalism’ would refer to actions of gov-ernment that attempt to improve people’s welfareby influencing their choices without imposing ma-terial costs on those choices… We can understandsoft paternalism, thus defined, as including nudges,and Iwill use the terms interchangeablyhere.” (CassSunstein Why Nudge? The Politics of LibertarianPaternalism 2014, p. 58).“It is important to recognise that behavioural eco-nomics and so-called ‘nudges’ are distinct. The for-mer is a scientific subdiscipline; the latter is a par-ticular way to apply its findings to policy, which

holds that policy makers should avoid regulationsthat limit choice (bans, caps, etc.) but can use be-havioural science to direct people towards betterchoices.” (Regulatory Policy and Behavioural Eco-nomics, (Lunn 2014), OECD report)

Since the publication of Thaler and Sunstein’sNudge: Improving Decisions about Health, Wealthand Happiness1 the concepts of nudge and their par-ticular version of soft paternalism called ‘libertarianpaternalism’ have become concepts of increasing in-terest and debate among public policy makers andacademics alike.2,3,4,5,6 However, ensuing discus-sions7 concernedwithparticular applications, the im-

* Behavioral Scientist, Ph.D., Department of Communication,Business & Information Technologies, Roskilde University.

1 Richard H. Thaler and Cass R. Sunstein, Nudge: ImprovingDecisions about Health, Wealth, and Happiness, Revised andExpanded Edition (New York: Penguin Books, 2009).

2 Sarah Conly, Against Autonomy: Justifying Coercive Paternalism(New York: Cambridge University Press, 2012), at p. 29.

3 Daniel Kahneman, “Foreword”, in Eldar Shafir (ed.), The Behav-ioral Foundations of Public Policy, (Princeton NJ: Princeton Uni-versity Press, 2012), pp. VII et sqq., at p. VIII.

4 Pete Lunn, Regulatory Policy and Behavioural Economics (OECDPublishing, 2014).

5 Barry Schwartz, “Why not nudge? A Review of Cass Sunstein’sWhy Nudge”, 17 April 2014, available on the internet at: <http://thepsychreport.com/essays-discussion/nudge-review-cass-sunsteins-why-nudge/> (last accessed on 17 April 2014).

6 Cass R. Sunstein,Why Nudge?: The Politics of Libertarian Pater-nalism (New Haven: Yale University Press, 2014).

7 See e.g. Ryan Calo, “Code, Nudge, or Notice”, 99(2) Iowa LawReview (2014), pp. 773 et sqq; Henry Farrell and Cosma Shalizi,“Nudge No More”, New Scientist, 26 November 2011: <http://www.slate.com/articles/health_and_science/new_scientist/2011/11/does_nudge_policy_work_a_critique_of_sunstein_and_thaler_.html> (last accessed on 26 November 2014); Lunn, RegulatoryPolicy and Behavioural Economics, supra note 4; Thomas Ploug,Søren Holm and John Brodersen, “To nudge or not to nudge:cancer screening programmes and the limits”, 66(12) Journal ofEpidemiology and Community Health (2012), pp. 1193 et sqq;Anthony Randazzo, “The Case Against Libertarian Paternalism”,23 April 2013, available on the internet: <http://reason.com/archives/2013/04/23/the-case-against-libertarian-paternalism>(last accessed 26 November 2014); Mark D. White, The Manipu-lation of Choice: Ethics and Libertarian Paternalism (New York:Palgrave Macmillian, 2013); Mark D. White, “The richness ofpersonal interests: A neglected aspect of the nudge debate”, 23

EJRR 1|2016156 The Definition of Nudge and Libertarian Paternalism

plications and the acceptability of nudges in publicpolicy and their relationship to libertarian paternal-ism have paid very little attention to a series of dis-agreements and ambiguities with regard to the waythat the concept of nudge has been defined and howit relates to that of libertarian paternalism.

The confusion exists even amongst researchers.For instance, nudges and libertarian paternalism areoften discussed as synonymous, as indicated by thetwo quotes above.8 At the same time it is said thatsupermarkets, restaurants, and cafeterias are nudg-ing consumers all the time whether they recognisethis or not.9 Yet it seems that the choice architects ofsupermarkets, restaurants and cafeterias can hardlybe said to be libertarian paternalists (whether theyrecognise this or not, or whether the references tothem as choice architects who nudges are intendedas such). So are nudges always acts of libertarian pa-ternalism, and are libertarian paternalists alwaysnudging by definition? Researchers seem to be con-fused.

An extension of this confusion is that the applica-tionandacceptabilityofnudgesareusuallydiscussedwithoutdeterminingorbeing sufficiently clear aboutwhether certain requirements apply to the motivesof the choice architect for an intervention to countas a nudge.10 That is, does a nudge have to be rootedin the mind-set of libertarian paternalism? Is it onlytruly a nudge if applied with this mind-set? That is,is it true, as it is often held by critics, that any propo-nent of nudge interventions is also necessarily sug-gesting the doctrine that policy makers should avoidregulations that limit choice?

In addition, a series of conceptual specifics are un-derdetermined in the current literature. For instance,it is often unclear what the exact relation betweennudges and incentives is.11 According to Sunsteinand Thaler it depends on whether the material or“cognitive” costs are low enough to make a nudgeeasy or cheap to avoid.12 Yet, as illustrated by discus-sions such as in Mongin and Cozic 201413, that easi-ly becomes a little too vague to be useful in any prac-tical discussion. For instance, while imposing a taxis said not to be a nudge14, and the same goes forplacing candy in an obscure place in the supermar-ket15, choosing a charm price or asking costumers topay 5 cents for plastic bags both count as nudges.16

But, as someone with philosophical inclinationsmight ask, where is the objective point of differenceto be found between the nudge provided by a 5-cent

tax on plastic bags or placing candy at eye height,and a non-nudge of a 5-dollar tax on plastic bags orplacing candy behind the counter? Is there a strictline between nudges and other interventions, is it acontinuum, or can an intervention be both a nudgeand not a nudge at the same time (especially sincecognitive costs seem relative to the individual cogni-tive capabilities of those nudged)? All of this is cur-rently not clear from the literature.

Besides the confusion on incentives, it also re-mains in the shadows whether and under what con-ditions the additionor removal of choice optionsmaycount as a nudge and if so whether it should also beregarded as a policy based on libertarian paternal-ism.17 On the one hand, the definition of nudge pro-vided by Thaler and Sunstein does not excludeadding choices. On the other, the removal of choiceoptions barred by the definitionmay, as we shall see,in some cases seem to qualify as nudges. This vague-ness and its consequences became clearly illustrated

October 2013, available on the internet at: <http://blogs.lse.ac.uk/politicsandpolicy/the-richness-of-personal-interests-a-neglected-aspect-of-the-nudge-debate/> (last accessed 26 November 2014);Paula Zoido-Oses, “The problem with nudge policies is thatthreaten our freedom to choose to act well”, 9 July 2014, avail-able on the internet at: <http://blogs.lse.ac.uk/politicsandpolicy/the-problem-with-nudge-policies-freedom-to-choose/#Author>(last accessed on 26 November 2014).

8 See also Calo, “Code, Nudge, or Notice”, supra note 7, at p. 773,775, 783, 785, 786, and 795; Conly, Against Autonomy, supranote 2, at p. 29-31; Lunn, Regulatory Policy and BehaviouralEconomics, supra note 4; Farrell and Shalizi, “Nudge No More”,supra note 7; Ploug, Holm and Brodersen, “To nudge or not tonudge”, supra note 7; Sunstein,Why Nudge?, supra note 6, at p.58.

9 See e.g. Thaler and Sunstein, Nudge, supra note 1, at p. 3; andAlberto Salazar, “Libertarian Paternalism and the Dangers ofNudging Consumers”, 23(1) King's Law Journal (2012), pp. 51 etsqq.

10 See e.g. Salazar, “Libertarian Paternalism and the Dangers ofNudging Consumers”, supra note 9; and Pierre Schlag, “Nudge,Choice Architecture, and Libertarian Paternalism”, 108(6) Michi-gan Law Review (2010), pp. 913 et sqq.

11 See e.g. the debate concerning user financial incentives asnudges around Adam Oliver, “A nudge too far? A nudge at all?On paying people to be healthy”, 12(4) Healthcare Papers (2012),pp. 8 et sqq.

12 Sunstein and Thaler, Nudge, supra note 1, at p. 8 footnote.

13 Philippe Mongin and Cozic Mikaël, “Rethinking Nudges”, (HECParis Research Paper No. ECO/SCD-2014-1067, 2014).

14 Thaler and Sunstein, Nudge, supra note 1, at p. 8.

15 Thaler and Sunstein, Nudge, (Ibid).

16 Sunstein,Why Nudge, supra note 6, at p. 64-65.

17 Mongin and Cozic, “Rethinking Nudges”, supra note 13, at p. 6and Schlag, “Nudge, Choice Architecture, and Libertarian Pater-nalism”, supra note 10, at p. 917.

ww

w.le

xxio

n.eu

Pow

ered

by

TC

PD

F (

ww

w.tc

pdf.o

rg)


by the debate caused by former New York City May-or Michael Bloomberg’s Big Gulp Ban.18While May-orBloomberg referred to it as anudge,RichardThalerquickly tweeted “To state the obvious: a BAN is nota NUDGE. The opposite in fact. So don't blameBloomberg’s ban on large soda cups on us.”19 How-ever, commentators were not convinced on theoret-ical grounds by Thaler’s disclaimer.20

Finally, as we shall see, confusion is widespreadeven as to how nudges may be properly separated, ifat all, fromtrivial standardmeasures suchas themereprovision of information and rational persuasion.21

Such trivial interventions are obviously true mea-sures of libertarian paternalism and qualify asnudgeson theoriginal definitionpresentedbyThalerand Sunstein. But if they are nudges, then the ques-tion arises: what is new about nudging, if anythingat all?

Although these disagreements and ambiguitiesmay seem ‘just theoretical’ they actually pose seriousproblems for the on-going efforts to apply behaviour-al science to public policy and other pro-social do-mains. Without clear and consistent foundationalconcepts the new policy-paradigm of applied behav-ioural science may easily come to seem ill founded,leaving the concept of nudge as well as the ideology

of libertarian paternalism vulnerable to accusationsof slippery-slopes, claims of conceptual inconsisten-cy, and warnings that nudges may quickly turn intoshoves and so forth. It also renders current effortsvulnerable to a misuse or the dilution of the under-lying ideas.Anykindof interventionmayeasily seemto qualify as a nudge, making any practitioner claimthat there is little new to the nudge approach andthat he or she has always been nudging; and any ob-jection pertaining to libertarian paternalism mayseem to concern the use of nudges as well, and viceversa.

However, the prevailing confusion about thenudge concept and its relation to libertarian pater-nalism is quite understandable. As pointed out byHausman and Welch,22 the explicit definition ofnudge provided by Thaler and Sunstein inNudge on-ly provides two negative conditions (that is condi-tions sayingwhat nudges are not), a couple of heuris-tics for determining what counts as such, and a vastseries of examples distributed throughout thebook.23

Yet, to the person well versed in behavioural sci-ence, one of these heuristics provides a fundamentaltheoretical principle for defining nudges based up-on the discipline of behavioural economics. This isthe heuristic saying that “… a nudge is any factor thatsignificantly alters the behavior of Humans, eventhough it would be ignored by Econs”.24 In their pa-per “Debate: To Nudge or Not to Nudge” HausmanandWelch at one point revise Thaler and Sunstein’sdefinition of nudge for reasons of consistency withthis principle so as to exclude incentives broadly con-ceived. In their paper “Nudge and the Manipulationof Choice” Hansen and Jespersen revise the defini-tion even further so as to separate the concept ofnudges from mere accidental influences in order toaccommodate ethically relevant considerations, andin this process they note the principle as a founda-tional one as well.25 In the paper “RethinkingNudges”Mongin andCozic semantically distinguish-es this heuristic as a concept of nudge separate fromthat provided in the original definition.26

In this paper I argue, based on this principle, forrevising Thaler and Sunstein’s original definition of‘a nudge’ on a series of important points. That is, dis-tinct fromMongin and Cozic I do not treat it as a sep-arate concept of nudge, but as the primary sense ofit. The aim is to arrive at a more viable definition toguide the discussion of applications of nudges aswell

18 Michael M. Grynbaum, “Health panel approves the restriction onsale of large sugary drinks”, New York Times, 13 September2012.

19 Richard Thaler, Tweet on Twitter, 31 May 2012, available on theinternet: <https://twitter.com/R_Thaler/status/208273339507150849> (last accessed on 27 December 2014).

20 See e.g. Oliver Burkeman, “‘How Bloomberg's soda ban is aclassic example of ‘choice architecture’’, Blog on The Guardian,10 July 2012, available on the internet at: <http://www.theguardian.com/commentisfree/oliver-burkemans-blog/2012/jul/10/bloomberg-soda-ban-new-york-freedom> (last accessed 26December 2014) and Pelle Guldborg Hansen, “The ‘Big GulpBan’ – a nudge or not?”, 8 October 2012, available on the inter-net: <http://inudgeyou.com/the-big-gulp-ban-a-nudge-or-not/>(last accessed 26 December, 2014).

21 Mongin and Cozic, “Rethinking Nudges”, supra note 13; andSunstein,Why Nudge, supra note 6, at p. 58.

22 Daniel Hausman and Brynn Welch, “Debate: To Nudge or Not toNudge”, 18 Journal of Political Philosophy (2010), pp. 123 etsqq.

23 See Thaler and Sunstein, Nudge, supra note 1, at p. 6.


25 Pelle Guldborg Hansen and Andreas Maaløe Jespersen, “Nudgeand the Manipulation of Choice: A Framework for the Responsi-ble Use of the Nudge Approach to Behaviour Change in PublicPolicy”, 1 European Journal of Risk Regulation (2013), pp. 3 etsqq, at p. 6.

26 Mongin and Cozic, “Rethinking Nudges”, supra note 13.


as to explicate the relationship between the conceptof nudge and that of libertarian paternalism as un-derstoodbyThaler andSunstein. In theprocess I clar-ify how nudges relate to and differ from other inter-ventions such as the provision of factual informationand rational persuasion as well as how nudges mayincorporate incentives while even being consistentwith the removal of certain types of choice. Ultimate-ly I put forward the following definition of what anudge is:

A nudge is a function of (I) any attempt at influ-encing people’s judgment, choice or behaviour ina predictableway (1) that ismade possible becauseof cognitive boundaries, biases, routines andhabits in individual and social decision-makingposing barriers for people to perform rationally intheir own declared self-interests and which (2)works by making use of those boundaries, biases,routines, and habits as integral parts of such at-tempts.

I also conclude that in so far as a nudge serves thedeclared self-interests of those being nudged, it mayfurther be referred to as libertarian paternalism sincethe revised definition of nudge implies that people’sbehaviour is influenced in ways that work indepen-dently of (i) forbidding or adding any rationally rel-evant choice options, or (ii) changing incentives,whether regarded in terms of time, trouble, socialsanctions, economic incentives and so forth. In addi-tion, this definition also implies that libertarian pa-ternalism goes beyond nudging since it follows fromit that nudges (iii) work independently of the provi-sion of factual information and rational argumenta-tion, that fall squarely within libertarian paternal-ism.

II. The Two Concepts

1. Libertarian Paternalism

Richard Thaler and Cass Sunstein originally intro-duced the concept of ‘libertarianpaternalism’ in their2003 essay of the same name published in The Amer-ican Economic Review.27 Here they defined a policyas ‘paternalistic’ “if it is selected with the goal of in-fluencing the choices of affected parties in away thatwill make those parties better off”,28 where they in-tend by “better off” that this be “measured as objec-

tively as possible”29 (and not always equating re-vealed preference with welfare). According to Thalerand Sunstein, while many economists believe theterm paternalistic to be derogatory because theythink paternalism always involves some kind of co-ercion, this is not necessarily the case.30 Policies maybe selected with the goal of influencing the choicesof affected parties in a way that will make those par-ties better off, but where there is no coercion in-volved.31 They refer to this kind of paternalism aslibertarian paternalism and ultimately define it as

“… an approach that preserves freedom of choicebut authorizes both private andpublic institutionsto steerpeople indirections thatwill promote theirwelfare.”32

According to Thaler and Sunstein an approach likethat of libertarian paternalism “should be acceptableto even the most ardent libertarian”.33 Of course,many critics have pointed out that Thaler and Sun-stein’s notion of libertarian paternalism is neithertruly ‘libertarian’, nor truly ‘paternalistic’,34 and thatit is a contradiction in terms35. However, those dis-cussions will not concern us here, as it is the conceptof nudge and its relation to libertarian paternalismthat are in focus.

In their best-selling book Nudge the notion of lib-ertarianpaternalism is further refined. It is describedas a “movement” or “strategy” recapturing commonsense from dogmatists.36 The libertarian aspect ofthe strategy is said to lie in “the straightforward in-sistence that, in general, people should be free to dowhat they like – and to opt out of undesirablearrangements if they want to do so”.37 Borrowing a

27 Richard H. Thaler and Cass R. Sunstein, “Libertarian Paternalism”,93(2) American Economic Review (2003), pp. 175 et sqq.

28 Thaler and Sunstein, “Libertarian Paternalism”, supra note 27, atp. 175.

29 Ibid.

30 Ibid.

31 Ibid.

32 Ibid, at p. 179.

33 Ibid, at p. 175.

34 See e.g. Hausman and Welch, “Debate: To Nudge or Not toNudge”, supra note 22.

35 Gregory Mitchell, “Libertarian paternalism is an oxymoron”, 99Northwestern University Law Review (2005), pp. 1245, et sqq.


37 Ibid.


term from Friedman, Thaler and Sunstein say that“libertarian paternalists urge that people should be‘free to choose’” and strive to “design policies thatmaintain or increase freedom of choice.” 38 In par-ticular, Thaler and Sunstein say that by modifyingthe term paternalism with libertarian they simplymean liberty preserving, adding that “Libertarian pa-ternalists want to make it easy for people to go theirownway; theydonotwant toburden thosewhowantto exercise their freedom.”39According to Thaler andSunstein thepaternalistic aspect lies in the claim that“it is legitimate for choice architects to try to influ-ence people’s behaviour in order to make their liveslonger, healthier, and better.”40Hence they argue fora self-conscious effort by institutions “to steer peo-ple’s choices in directions that will improve theirlives.”41However, they alsomodify their understand-ing of paternalism compared with their 2003 paper,now - referring to Van De Veer42 – holding that, “apolicy is ‘paternalistic’ if it tries to influence choicesin a way that will make choosers better off, as judgedby themselves”43, rather than interpreting “better off”as earlier to be “measured as objectively as possi-ble”44.

According to Thaler and Sunstein “libertarian pa-ternalism is a relatively weak, soft, and nonintrusivetype of paternalism because choices are not blocked,fenced off, or significantly burdened.”45 Yet, it doescount as paternalism, or ‘soft paternalism’ because“private and public choice architects are not merelytrying to track or to implement people’s anticipatedchoices. Rather, they are self-consciously attemptingtomovepeople indirections thatwillmake their lives

better.”46 In particular, Thaler and Sunstein say thatby doing this “They nudge.”47

2. Nudge

But what do Thaler and Sunstein mean when sayingthat libertarian paternalist ‘nudge’? The concept wasoriginally suggested by the first editor approachedby Thaler and Sunstein, an editor who ultimately de-clined to publish the book that later became a best-seller titled Nudge: Improving Decisions AboutHealth, Wealth, and Happiness. Beforehand most fa-mous from Monty Python’s sketch “nudge, nudge,wink, wink” the concept of ‘nudge’ has now becomea term closely tied – if not almost synonymous with– Thaler and Sunstein’s concept of ‘libertarian pater-nalism’.48 But what does ‘nudge’ mean and how doesthe concept fit together with that of libertarian pa-ternalism?

In Nudge Thaler and Sunstein define a nudge asfollows:

“A nudge, as we will use the term, is any aspect ofthe choice architecture that alters people’s behav-ior in a predictable way without forbidding anyoptions or significantly changing their economicincentives. To count as amere nudge, the interven-tion must be easy and cheap to avoid. Nudges arenot mandates. Putting fruit at eye level counts asa nudge. Banning junk food does not.”49

The concept of nudge thus seems to fit like hand inglove with that of libertarian paternalism. Libertari-an paternalism is an approach that authorises bothprivate and public institutions to steer people in di-rections that will promote their welfare; a nudge al-ters people’s behaviour in a predictable way. Liber-tarianpaternalism is an approach that preserves free-dom of choice; a nudge works without forbiddingany options or significantly changing economic in-centives. Finally, libertarian paternalists nudge.

But does that mean that nudges are always liber-tarian paternalistic by definition? As noted in the in-troduction this has been a tacit assumption by manycommentators. It is thus often assumed that inter-ventions based on nudges and the strategy of liber-tarian paternalism are the same. That is, that nudgesare libertarian and paternalistic, and that libertarianpaternalistic measures are nudges. But is this thecase? In order to answer that question and resolve

38 Ibid.

39 Ibid.

40 Ibid.

41 Ibid.

42 Donald Van De Veer, Paternalistic Intervention: The MoralBounds on Benevolence, (Princeton: Princeton University Press,1986).


44 Thaler and Sunstein, “Libertarian Paternalism”, supra note 27, atp. 175.

45 Ibid.


47 Ibid.

48 See supra note 8.



the resulting confusion and ambiguities, we need tolook more closely into the concept and the theoreti-cal foundations underpinning it.

III. Nudge by Definition

It is often said that the concept of nudge is ill de-fined50 and may lead to slippery slopes51. Such ob-servations easily lead to critiques headed “Whennudge comes to shove” or something like it.52 How-ever, given Thaler and Sunstein’s definition quotedabove, such critiquesmay seem tomisfire. In fact, byrearranging the original definition provided byThaler and Sunstein a bit, it seems that they providea quite clear-cut definition saying that:

A nudge is any aspect of the choice architecturethat alters people’s behaviour in a predictable waywithout1) forbidding any options or2) significantly changing their economic incen-tives.

Still, definitions are not all of one kind. They serve avariety of functions. Thus e.g. some definitions aredescriptive,while others are stipulative. That is, somedefinitions try to describe or capture the actual us-age of the term, while others stipulate or impart ameaning to the defined term and thus involve nocommitment that the assigned meaning agrees withprior uses (if any) of the term.53 For this latter type,“ill defined” can only mean that a definition fails toprovide the wanted conceptual clarity and consisten-cy needed for it to work within its intended area ofapplication.

Obviously, Thaler and Sunstein’s original defini-tion is a stipulative one, and one that has since thenbeen broadly adopted. However, as this paper willshow, it is not always true for Thaler and Sunstein’sdefinition of nudge that it provides the intended clar-ity and consistency. In particular it fails to serve itsfunction relative to its theoretical foundations in be-havioural economics and its relation to the conceptof libertarian paternalism. Of course, that does notimply that the concept is fundamentally flawed ornecessarily will lead to slippery slopes. Rather it im-plies that it can be improved upon. Throughout thissection I thus offer an explication of the definitionof nudge resulting in what I see as an absolute im-provement of an existing, imperfect concept.

1. Econs and Humans

To explicate the definition of ‘nudge’ we first need tounderstand the purpose intended to be served by itsintroduction and to which it should be aligned. InNudge Thaler and Sunstein explicitly motivates itsintroduction with the findings of four decades of be-havioural economics.54 This is not surprising sinceThaler is one of theworld’s leading behavioural econ-omists and Cass Sunstein is, amongst many otherthings, a pioneer and leading scholar in BehaviouralLaw, a movement explicitly rooted in behaviouraleconomics and the Biases andHeuristics programmeofKahnemanandTversky.Thisbackgroundbecomesparticularly obvious in Thaler and Sunstein’s sectionin Nudge titled ‘Humans and Econs: Why NudgesCan Help”55 that follows directly after their section‘LibertarianPaternalism’56 that endsbyproviding theoriginal and explicit definition of nudge quotedabove.

The distinction between Econs and Humans intro-duced by Thaler and Sunstein is a distinction con-trasting traditional conceptions of human behav-iours as derivatives of axioms of reason and rational-ity with the insights into human behaviours emerg-ing from cognitive psychology and behavioural eco-nomics in the last four decades. Reason and rational-ity have been analysed most intensely in philosophy

50 See e.g. Mongin and Cozic, “Rethinking Nudges”, supra note 13;and House of Commons, Public Health: Twelfth Report of Session2010-12, Vol. 1: Report. Together with Formal Minutes (GreatBritain: Parliament: House of Commons: Health Committee2011), at p. 84.

51 See e.g. Mario J. Rizzo and Douglas Glen Whitman, “LittleBrother Is Watching You: New Paternalism on the SlipperySlopes”, 51 Arizona Law Review 2009, pp. 685 et sqq; and AdamC. Smith and Todd J. Zywicki, “Behavior, Paternalism, and Policy:Evaluating Consumer Financial Protection”, George Mason Law &Economics Research Paper No. 14-05 (2014), at p. 12.

52 See e.g. Tim Adams, “Nudge economics: has push come to shovefor a fashionable theory?”, The Guardian, 1 June 2014, availableon the internet at <http://www.theguardian.com/science/2014/jun/01/nudge-economics-freakonomics-daniel-kahneman-debunked> (last accessed on 27 December 2014); John Tierney,“A Nudge (or Is it a Shove?) To the Unwise”, New York Times, 24March 2008, available on the internet at <http://tierneylab.blogs.nytimes.com/2008/03/24/a-nudge-or-is-it-a-shove-to-the-unwise/> (last accessed on 27 December 2014).

53 See Anil Gupta, “Definitions”, in Edward N. Zalta (ed.), TheStanford Encyclopedia of Philosophy, (Fall 2014 Edition), avail-able on the internet <http://plato.stanford.edu/archives/fall2014/entries/definitions/> last accessed (28 December 2014).

54 Thaler and Sunstein, Nudge, supra note 1, at p. 6-8.

55 Ibid.

56 Thaler and Sunstein, Nudge, supra note 1, p. 4-6.


and economics in a movement that ultimately led toneo-classical axiomatisations of rational or econom-ic behaviour. According to these axiomatisations, be-haviour is epitomised by rational choice or decision-makingbasedupon individual agents reasoning theirway by deduction to appropriate actions based ontheir desires orpreferences togetherwith theirbeliefsand information about the world. Keeping it simple,options, preferences, and beliefs are thus all that in-fluence the behaviours and actions of a rationalagent.57

In passing it may be noted that in the neo-classi-cal view of decisions and behaviour there is no roomfor nudges andno reason for libertarianpaternalism.This is perhaps most clear from the adoption of Re-vealed Preference Theory (RPT) in the neo-classicaleconomic analysis of real world behaviour. RPT as-sumes that the preferences of consumers are re-vealed by their purchasing habits. If a consumerchooses a particular item it is inferred that this out-come is observed as a result of the consumer’s pref-erence for that item above others given her availableinformation. This means that the intention-actiongap that motivates the libertarian paternalist strate-gy and, as we shall see, within which the nudge ap-proach to behaviour change works, is ruled out a pri-ori.

However, as pointed out byThaler and Sunstein,58

during the last four or five decades the disciplines ofbehavioural economics, cognitive, and social psy-chology have provided ample evidence or behaviour-al insights that much of our behaviours and choicescannot be explained convincingly as being consis-tent with the neo-classical axioms of rational behav-iour. For instance, behavioural economics has shown

that human behaviour and choice exhibit boundedrationality, bounded self-interest, and boundedwillpower.59 This helps explain not only why peopleoften fail to reason theirway to the right conclusions,but also why we often fail to act upon these conclu-sions when reached, causing the gap between ourgood intentions and our actual behaviour - the gapthat motivates the doctrine of libertarian paternal-ism.

Now, to some, “bounded” may seem to carry theconnotation that we are merely limited in our cogni-tive skills at exerting rationality, self-interest, andwillpower. That is not the case. Behavioural econom-ics runs on experimental evidence pioneered in cog-nitive and social psychology, showing that much ofour individual and social behaviour is due to ourbrains processing information in ways that are notonly bounded but also cognitively biased, where acognitive bias is a systematic pattern of deviation injudgment or decision-making. Implicit in this con-cept is a “pattern of deviation” from a baseline orstandard of comparison.60 In this case the baselineis made up by the bundle of neo-classical assump-tions and their derivatives giving rise to the rationalagent Homo Economicus - or as Thaler and Sunsteindub them in plural: Econs.

Of course, the empirical evidence and theoreticalconstructs of biases invoked to explain our behav-iours only make up a consistent alternative to neo-classical economics and its derivatives such as RPT,if an alternative theory exists that not only predicts,but better explains the facts. Here, like others, Thalerand Sunstein rely on Dual Process Cognitive Theories(DPTs), especially as portrayed by Kahneman61 as aresult of Kahneman and Tversky’s work on the Bias-es and Heuristics programme.

DPTs assert that the human brain functions inways that invite a distinction between two kinds ofthinking: one, which is intuitive and automatic, andanother, which is reflective and rational. Kahnemandubs these ways of thinking System 1 and System 2,respectively; we choose, however, to follow the leadof Thaler and Sunstein when referring to thesemodes of thinking as automatic thinking and reflec-tive thinking. Automatic thinking is characterised bybeing fast, intuitive, and usually not associated withexperiences that one would describe as thinking. Re-flective thinking is associatedwith the deliberate andconscious processing of information. It is slow, ef-fortful, and dependent on concentration. It is associ-

57 See also Thomas Gilovich and Dale Griffin, “Introduction -Heuristics and Biases: Then and Now”, in Thomas Gilovich, DaleGriffin and Daniel Kahneman (eds.), Heuristics and Biases: ThePsychology of Intuitive Judgment (Cambridge: Cambridge Univer-sity Press, 2002), pp. 1 et sqq; Martin Peterson, An Introductionto Decision Theory, (Cambridge: Cambridge University Press,2009).

58 Thaler and Sunstein, Nudge, supra note 1, p. 7.

59 See also Christine Jolls, Cass R. Sunstein and Richard Thaler, “ABehavioral Approach to Law and Economics”, 50(5) Stanford LawReview (1998), pp. 1471 et sqq.

60 Gilovich and Griffin, “Introduction - Heuristics and Biases”, supranote 57.

61 See e.g. Daniel Kahneman, “Maps of Bounded Rationality:Psychology for Behavioral Economics”, 93(5) The AmericanEconomic Review (2003), pp. 1449; and Daniel Kahneman,Thinking, Fast and Slow, (London: Allen Lane 2011).


ated with self-awareness, the experience of agency,autonomy, and volition. The key features of each sys-tem are shown in Table 1.1.

DPTs explain why we – beingHumans rather thanEcons – not only fall short of rational decision-mak-ing, but actually may systematically deviate from itsnormative prescriptions (even in our reflected deci-sions at times) due to our behaviour and decisionsbeing biased by seemingly irrelevant factors and as-pects of decision-making and behavioural contexts.At its most general, the concept of nudge is devisedto capture the fact that human decision-making andbehaviour are influenced by cognitive boundariesand biases in ways that may be utilised for promot-ing particular behaviours - and Thaler and Sunstein’soriginal definition is their attempt to capture this in-sight.

2. The Principle from BehaviouralEconomics

However, the discussion of Human and Econs ulti-mately leads Thaler and Sunstein to state a corollaryof their original definition that I will refer to as theprinciple from behavioural economics. This says:

“In accordance with our definition, a nudge is anyfactor that significantly alters the behavior of Hu-mans, even though it would be ignored byEcons.”62

This principle of behavioural economics, I believe,although misleadingly stated as a derivative of theexplicit definition63, is what provides us with theclear-cut and foundational criterion of what shouldcount as a nudge. In accordance with the general gistof Thaler and Sunstein’s book, its adoption takes theBiases and Heuristics programme as the core moti-

vation for devising a concept of nudge for referringto any contextual feature causing Humans to deviatefrom the prescriptions of normative rationality. As acriterion it states that if we know what behaviour orchoice would count as rational in a given situationprovided agents’ preferences and information, andwe observe judgment, choice, or behaviour deviatingfrom this prediction, then the contextual feature re-sponsible for such deviation should be categorisedas a nudge. If accepted as a foundational criterion, itthus follows that any viable definition of a nudgeshould be aligned with it. See Figure 1.

Granted the criterion as a fundamental one, theconcept of nudge is thus devised to refer to featuresthat influence behaviour in ways not in accordancewith that of economic rationality. Since economic ra-tionality follows from the dictates of principled rea-son, one could add that nudges work in ways thatshould not work in principle, or in ways that ‘ought’not influence us. That is, in principle a nudgeshouldn’t influence behaviour, but in practice it does.In particular, it is obvious that condition (1) and (2)of Thaler and Sunstein’s original definition are in-tended to exclude those features that influence thebehaviour, not only of humans, but also of rationalagents. Hence, it seems that it is conditions (1) and(2) of the original definition that are derivatives ofthe principle, rather than vice versa. Of course, theacceptability of this proposed hypothesis about thefoundations of the concept of nudge, rests on the abil-ity of the principle from Behavioural Economics to



Table 1.1

Figure 1


provide a consistent and clear framework for a moreviable definition of the concept than that providedby Thaler and Sunstein’s explicit definition. I believeit does, and the remainder of this paper provides theargument for this. However, instead of constructinga definition anew from the foundational principlefrom Behavioural Economics, I will start in mediasres and revise the original definition so as to recon-cile this with the principle.

IV. Reconciling the Definition of Nudgewith the Principle from BehaviouralEconomics

1. Adding some Positive Conditions

As a first step to reconcile Thaler and Sunstein’s de-finition of nudge with the principle from Behaviour-al Economics, we may notice that Thaler and Sun-stein’s original definition of a nudge says nothingabout cognitive boundaries and biases (Humans) orthe baseline from neo-classical economics (Econs). Itonly says what a nudge is not. But given what hasjust been explained about DPTs and the Biases andHeuristics programme, we can now see why DanielHausman and Brynn Welch in a later article add the“positive” condition to the definition of a nudge say-ing that:

“They [nudges] are called for because of flaws inindividual decision-making, and work by makinguse of those flaws”.64

In fact, being a conjunctive sentence, this additioncontains two positive conditions. One of function ormotivation – that “nudges are called for because of…”and a second condition, which explicates the under-lying principle by specifying that nudges “work bymaking use of those flaws”. That is:

(i) They [nudges] are called for because of flaws inindividual decision-making, and (ii) work bymak-ing use of those flaws.

Thus, as a first step to reconcile the definition ofnudge with the principle from Behavioural Econom-ics, we may integrate Hausman andWelch’s positiveconditions into the former. But rather than calling bi-ases “flaws”, I prefer calling them by their propername: cognitive bias. Also, it should be added that bi-ases do not necessarily only pertain to individual de-cision making, but also may be argued to pertain tosocial decisionmaking, i.e. when aggregate group be-haviour deviates from the interests of each individ-ual group member.65 In addition, it is not only bias-es, but also cognitive boundaries, habits and routinesthatmay systematically leadHumans todeviate fromthe behaviour of Econs.66 Finally, we should be a bitmore specific than Hausman and Welch are, so in-stead of saying “making use of those flaws”, we spec-ify that biases should be an integral part of the choicearchitecture.

Given these adjustments the positive conditionsmay be added to the original definition so that wenow have a definition with four conditions:

A nudge is any aspect of the choice architecturethat alters people’s behavior in a predictable waywithout(1) forbidding any options or(2) significantly changing their economic incen-tives.(3) Nudges are called for because of cognitiveboundaries, biases, routines, and habits in individ-ual and social decision-making, and (4) work bymaking use of those boundaries, biases, routines,and habits as integral parts of the choice architec-ture.

2. Incentives (Condition (2))

Still, more work needs to be done to reconcile the de-finition of nudge with its theoretical underpinnings.As noted above, the fundamental principle for deter-mining if something is a nudge is by determiningwhether it alters the behaviour ofHumans butwouldbe ignored by Econs.67 Fundamental to Econs is thattheir preferences may be captured by a suitable util-ity-function. As a fundamental construct in econom-ic theory it is not surprising that a utility function is

64 Hausman and Welch, “Debate: To Nudge or Not to Nudge”,supra note 22, at p. 126.

65 See e.g. Vincent F. Hendricks and Pelle Guldborg Hansen,Infostorms: How to Take Information Punches and Save Democ-racy (New York: Springer, Copernicus, 2014); and Cass R. Sun-stein Infotopia: How Many Minds Produce Knowledge (NewYork: Oxford University Press, 2006).

66 See e.g. Judith A. Oullette and Wendy Wood, “Habit and Inten-tion in Everyday Life: The Multiple Processes by Which PastBehavior Predicts Future Behavior”, 124(1) Psychological Bulletin(1998), pp. 54 et sqq.



sensitive to economic incentives – hence, Thaler andSunstein’s condition (2) saying that a nudge influ-ences behaviour without “significantly changing[people’s] economic incentives”.

Yet, as pointed out by Hausman and Welch,68 ra-tional agents are not only responsive to economicincentives. For instance, the utility-function of a ra-tional agent is determined by the prospect of painas well as penalties. This is not captured by the def-inition as it stands since, if taken at face value, thedefinition would render a 10.000 voltage elec-troshock, or a public beating to count as a ‘nudge.’Because this would be a rather uncharitable inter-pretation of what Thaler and Sunstein mean by anudge, as well as not in accordance with the princi-ple from Behavioural Economics, it is only reason-able to follow Hausman and Welch’s suggestion ofbroadening the definition so as to encompass all oth-er types of incentives affecting a rational utility func-tion as well.

Thus, we may revise the definition of a nudge asfollows:

A nudge is any aspect of the choice architecturethat alters people’s behaviour in a predictable waywithout(1) forbidding any options or(2) significantly changing incentives, whether re-garded in termsof time, trouble, social sanctions,economic and so forth.(3) Nudges are called for because of cognitiveboundaries, biases, routines, and habits in individ-ual and social decision-making, and (4) work bymaking use of those boundaries, biases, routines,and habits as integral parts of the choice architec-ture.

a. “Significantly” Changing Incentives

Still, onemay ask why the original condition (2) says“significantly” (as well as why Hausman and Welchrestates this as “appreciably”). To the best ofmy judg-ment this is done to accommodate insights from Be-havioural Economics about how changing incentivesmay be used to influence behaviour of Humans inways that do not affect Econs.

Let me give three examples of how this may hap-pen.1. Lotteries. First, disproportionally large behaviour-

al effects on Humans may be obtained by makingsmall, in principle insignificant changes in incen-

tives thatwouldn’t have an effect onEcons (at leastnot to the same extent). Say for instance you wantto conduct a survey amongst 500 people. To getpeople to respond you consider giving them someproper incentive. But your budget is only $500.One possibility is to offer each potential respon-dent $1 to complete the survey. Another possibili-ty is to offer each potential respondent a ticket fora lottery draw where the prize is an iPad. Now, toa rational agent this should make no differencesince the expected utility in both scenarios wouldbe $1. However, for Humans it makes a world ofdifference. Being offered a lottery ticket increasesour tendency to participate since we tend to over-estimate small probabilities because of, amongstother factors, the availability heuristic, anchoringon arbitrary priors, and a tendency to bemore sen-sitive toprobability changesclose to0 than toprob-ability changesaway from0.69 Inadditiononemayalso argue that Humans often perceive direct pay-ment differently than they perceive lottery tickets(usually regarding the latter as more like a gift,thereby activating norms of reciprocity, contraryto the former which we regard as impersonal pay-ment, possibly with some suspicion and compar-ison to alternative activities and the profit gainedby the payer).

2. A second example is that of rearranging incentivesin ways that shouldn’t have any effect on us if wewere Econs, but does affect us, as we are Humans.For instance, to an Econ it shouldn’t affect con-sumption patterns whether a tax payback is givenas a lump sum or distributed in a series over time.However, due to what is calledmental accounting,there is a tendency for humans to treat a lump sumpayment very differently.70 For instance, we mayregard it as house-money, or decide to use it forlarger investments, such as travelling, rather thanintegrate it into our everyday consumption. In ad-dition, incentives may be distributed over time

68 Hausman and Welch, “Debate: To Nudge or Not to Nudge”,supra note 22, at p. 123-136.

69 Zach Burns, Andrew Chiu and George Wu, “Overweighting ofsmall probabilities”, in James J. Cochran (ed),Wiley Encyclopediaof Operations Research and Management Science (New York:Wiley, 2010).

70 Shefrin, H. and Richard Thaler, “The behavioral life-cycle hypoth-esis”, 26 Economic Inquiry (1988), pp. 609 et sqq; Richard H.Thaler, “Mental Accounting Matters”, 12 Journal of BehavioralDecision Making (1999), pp. 183 et sqq.


causing so-called present-bias, whereHumans dis-count valuations hyperbolically rather than expo-nentially, that is, where valuations fall very rapid-ly for small delay periods, but then fall slowly forlonger delay periods, rather than being discount-ed at a constant rate.71

3. A third example is that of “charm prices”, e.g. ad-vertising a price of $0.99 rather than $1 therebyutilising the ‘left-digit-right-digit’ effect. Thisprice change shouldn’t lead to the significantchanges in consumer behaviour that we actuallyobserve.72

Effects like these three are not reconcilable with thestandard economic model of Econs. Yet, all do seemto involve changing incentives. However, looks maybe deceiving. Being more precise we may point outthat example 1 is not really about changing incen-tives, but rather about restructuring incentives interms of a lottery inways thatwouldnot affect Econs.Likewise example 2 is about rearranging incentives,rather than changing them. But what about example3 – is this not obviously about changing incentives?Well, in a sense changing incentives does create thebehavioural effect. Yet, the behavioural effect is notcredibly derivable from an agent’s utility functionalone. The incentives are not changed significantlyin creating a significant effect. The nudge is identi-fied by the deviation from the predictions of the stan-dard model from observed behaviour – a deviationcaused in this case by the salience effect - and not thefull behaviour change as such.

It is most likely in order to allow such insightsfrom behavioural economics of how insignificantchange in incentives may create effects beyond whatmay be explained as rational that Thaler and Sun-steinadopt thequalification “significantly” in the2ndcondition. However, if we substitute “work without”with “thatworks independently of”we can retain thispoint without the need of this qualification that hassometimes confused commentators. Also, to empha-sise more clearly that it is not the full intervention orchoice architecture as such that is evaluated as anudge - e.g. it is not the incentive as such, but the

structuring of an incentive as a lottery - we may sub-stitute the somewhat vague term ‘aspect’ with thatof ‘function’ which is also a more standard term inthe behavioural sciences. Hence, we may say:

A nudge is a function of the choice architecturethat alters people’s behaviour in a predictable waythat works independently of(1) forbidding any options or(2) changing their incentives, whether regarded interms of time, trouble, social sanctions, economicand so forth.(3) Nudges are called for because of cognitiveboundaries, biases, routines, and habits in individ-ual and social decision-making, and (4) work bymaking use of those boundaries, biases, routines,and habits as integral parts of the choice architec-ture.

3. Choices (Condition (1))

So far the only part of the original definition that wehave not revised is that concerned with the rulingout or forbidding of options, i.e. condition (1). Obvi-ously, this condition has a strong affinity with theidea of libertarian paternalism. That is, its formula-tion seems primarily intended to target this complexof ideas, and only secondarily at capturing the un-derlying principle of a nudge as expressed by the re-vised definition. In particular, by emphasising thata nudge works without forbidding any options it fol-lows that nudging neither involves bans or man-dates, which is a basic characteristic of libertarian pa-ternalism. In practice such a ban or mandate may of-tenbe reduced to a positive or negative incentive cou-pled with a symbolic dimension of social disap-proval.

From a conceptual point of view, however, condi-tion (1) turns out to be incomplete along several di-mensions. For instance, it does not say anythingabout whether onemay add options to the choice set.This raises the question of whether adding choicesmay qualify as a nudge?

a. Adding Choices: Predictively Rational

To investigate this, assume that Marge wants to in-fluence Homer in a way so as to stop him from eat-ing all that cake. Also assume that in the status quoHomer has the options or choice-set A comprising a:

71 Edward Cartwright, Behavioral Economics (New York: RoutledgeAdvanced Texts in Economics and Finance, 2011), p. 143-147.

72 Manoj Thomas and Vicki Morwitz, “Penny Wise and PoundFoolish: The Left-Digit Effect in Price Cognition”, 32(1) Journal ofConsumer Research (2005), pp. 55 et sqq.


“not eating cake” and b: “eating cake”. Finally we al-so assume that Homer prefers b to a. That is,

Ah = (a, b), a <h b.Next, assume that Marge in order to influence

Homer decides to serve a bunch of sandwiches - i.e.,she adds c: “eat sandwiches” to the choice-set - know-ing that Homer prefers sandwiches to cake. Thus wehave that,

Ah = (a, b, c), a <h b <h c.The question is whetherMarge’s addition of sand-

wiches to the choice-set should be regarded as anudge.

Obviously, Marge attempts to influence Homer’sbehaviour in a predictable way. Also, she is not for-bidding any options. Hence (1) is not violated. Nor isshe significantly changing the incentives so (2) is re-spected as well. Thus, based on the original part ofthe definition Marge’s addition would seem to qual-ify as a nudge.

However, it is obvious that Homer’s preferencesare rational in the sense that he is consistent in pre-ferring sandwiches to cake. Hence, the interventionwould not only influence Humans - or Homer in thiscase - but also Econs. This is because the influence isnot made possible because of cognitive boundariesor bias in individual and social decision-making, orroutines or habits, posing a barrier for Homer to per-form rationally. Instead, the influence is made possi-ble by Homer’s rational capacities. Also, it does notwork by making use of those boundaries and biasesas an integral part of Marge’s attempt at creating be-haviour change. In conclusion, Marge’s attempt isnot a nudge since it does not satisfy conditions (3)and (4) that have been explicated on the basis of theprinciple from behavioural economics.

b. Adding Choices: Predictively Irrational

But does that mean that we always have to rule outtheadditionof choiceoptionsas anudge?No,becauseone may also add choices to the existing choice-set soas to influence people in a “predictively irrational”way.Variousdecoyeffectsviolating the independenceof irrelevant alternatives axiom of decision theory re-veal this. Decoy effects are one family of biases inwhich the asymmetric dominance effect and the com-promise effect resulting from extremeness aversionhave been shown to be among the most stable.73

To illustrate the asymmetric dominance effect,imagine that you are faced with the choice between

two laptops that vary in price and storage:

Choice-set 1

A B

Price $400 $300

Storage 300 GB 200GB

Given this choice set, you may either prefer A for itsgreater storage capacity or B for its lower price. Now,suppose that a new laptop, C, is added to the choiceset, where C is more expensive than both A and Band has more storage capacity than B, but less thanA:

Choice-set 2

A B C

Price $400 $300 $450

Storage 300 GB 200 GB 250 GB

If you are like most other people, the addition of Cto the choice-set will make you become more attract-ed to A. This is because, while you are likely to avoidC since you can get laptop A with more storage for alower price, C affects your preferences by acting asa basis of comparison for A and B. Because A is bet-ter than C in both respects, while B is only partiallybetter thanC,more consumerswill preferAnowthandid before. C is therefore a decoy whose sole purposeis to increase sales of A. This is what is called theasymmetric dominance effect.

Theasymmetric dominance effect canalsobeusedto influence your preferences so that you come toprefer laptop B. Imagine that instead of introducinglaptop C, a laptop D is introduced into the choice-setwith the attributes shown in Choice-set 3:

73 See Joel Huber, John W. Payne and Christopher Puto, “AddingAsymmetrically Dominated Alternatives: Violations of Regularityand the Similarity Hypothesis”, 9(1) Journal of Consumer Re-search (1982), pp. 90 et sqq; Itamar Simonson and Amos Tversky,“Choice in context: Tradeoff contrast and extremeness aversion”,29 Journal of Marketing Research (1992), pp. 281 et sqq.


Choice-set 3

A B D

Price $400 $300 $350

Storage 300 GB 200 GB 150 GB

Faced with this set of choices the asymmetric domi-nance effect is likely to influence your preferencessuch that while you will disregard D, you will nowfind B more attractive than A.

So is the asymmetric dominance effect to be con-sidered a nudge? The effect is not one that may bereconciled with the rational preferences of Econs.74

That is, the effect of adding an, in principle, irrele-vant choice option does not affect Econs. Still, it doesaffect experimental subjects and hence Humans. Ac-cording to the principle from behavioural econom-ics, then, it counts as a nudge.

But does our revised definition of nudge alignwiththis conclusion? Obviously the addition of an irrele-vant choice option does not rule out or forbid any op-tions from the original choice-set. Hence (1) is not vi-olated. Nor does it significantly change the incentivesand hence (2) is respected as well. Instead the influ-ence is made possible because of cognitive bias in in-dividual decision-making. This means that condition(3) is satisfied. Finally, the addition of the irrelevantchoice option is an intervention that works by mak-ing use of those biases. That is, it is not a case ofmere-ly de-biasing choice and thus condition (4) is satisfied.

Accordingly, relative to our revised definition,adding choices in this way is a case of a nudge – notonly because it is stipulated so, but because it is con-sistent with the underlying principle saying that anudge is any factor that significantly alters or influ-ence the behaviour ofHumans, even though it wouldbe ignored by Econs. However, before revising the

definition so as to allow for the addition of rational-ly irrelevant choice options and bar the addition ofrelevant ones, we will address an interesting issue.

c. Nudging by Removal of Irrelevant Alternatives

If onemaynudgebyaddingan irrelevant choice, thenit seems reasonable that the samemay be the case byremoving such an irrelevant choice. For instance, go-ingback to the asymmetric dominance effect: if threelaptops are available in the status quo as in choice-set 2 and 3, and we then remove the choice-option Cand D, respectively, are these interventions then in-stances of nudges?

To answer this question we should, according tothe definition, first ask whether the removal of a ra-tionally irrelevant choice option like this significant-ly alters thebehaviourofHumans, even thoughEconswould ignore it. To determine this wemay look at theexamples above in reverse. Of course, since some ofthe experiments on the asymmetric dominance effectare between-subjects design, the effect of removingthe rational irrelevant options is the reverse of thatobservedwhenadding it. But even forwithin-subjectsdesigns theeffectoccurs.75Wethenaskwhether thesebehaviour-changeswouldbe consistentwith the stan-dard rationality axioms fromwhich the behaviour ofEcons is derived – the answer, of course, is no.

To be precise, according to the standard axiom ofdecision theory referred to as the independence ax-iom,76 rational choices should be independent of ir-relevant alternatives. This axiom is sometimes alsoreferred to as Sen’s property stating that:77

if an alternative x chosen from a set T, and x is al-so an element of a subset S of T, then x must bechosen from S.

That is, eliminating some of the unchosen alterna-tives should not affect the selection of x as the bestoption. In our case of the asymmetric dominance ef-fect the independence axiom implies that Econsshould be unaffected by the removal of C and D. Yet,as the evidence for the asymmetric dominance effectand other decoy effects shows, Humans are influ-enced by this removal. Hence, in such cases, removalof irrelevant choice options would qualify as nudgesaccording to the principle frombehavioural econom-ics.

It is in light of this conclusion that the definitionof a nudge above needs to be further refined so that

74 Huber, Payne and Puto, “Adding Asymmetrically DominatedAlternatives”, supra note 73; Simonson and Tversky, “Choice incontext”, supra note 73.

75 See e.g. Huber, Payne and Puto, “Adding Asymmetrically Domi-nated Alternatives”, supra note 73.

76 Peterson, An Introduction to Decision Theory, supra note 57, at p.99.

77 Amartya Sen, Collective Choice and Social Welfare (San Francis-co: Holden Day, Inc., 1970), at p. 17.


besides barring only the addition of relevant options,it also allows for the removal of rationally irrelevantones. While the removal of irrelevant choices as anudge has no bearing on the other conditions, it callsfor revising condition (1) by adding the qualificationof relevancy, such that:

A nudge is a function of the choice architecturethat alters people’s behaviour in a predictable waythat works independently of(1) forbidding or adding any rationally relevantchoice options or(2) changing their incentives, whether regarded interms of time, trouble, social sanctions, economicand so forth.(3) Nudges are called for because of cognitiveboundaries, biases, routines, and habits in individ-ual and social decision-making, and (4) work bymaking use of those boundaries, biases, routines,and habits as integral parts of the choice architec-ture.

4. Information

Earlier it was mentioned that only three basic ele-ments influence the behaviour of anEcon. Choice op-tions, preferences over the outcomes associated withthese, and information or beliefs. So far we have re-vised the definition relative to choice options andpreferences. But what about information? Just like itwas the case for the addition of choice options, theoriginal definition of a nudge provided byThaler andSunstein says nothing about this. So do informationand various forms of communication count asnudges?

a. The Provision of Factual Information

Beginning with the mere provision of factual infor-mation we may observe that in his recent book WhyNudge? Sunstein is quite straightforward on thispoint:

“Provision of information is certainly a nudge, butit may or may not qualify as paternalistic…”78

But is Sunstein right in this? Obviously in the sensetargeted here, the provision of factual information isusually intended to influence behaviour without for-bidding or adding any relevant choice options or sig-nificantly changing incentives. Hence, the provision

of factual information would seem to qualify as anudge given Thaler and Sunstein’s original defini-tion taken at face value. In addition, if offered as wellintended advice it should definitely also count asboth libertarian and paternalistic.

However, if we look to the principle from behav-ioural economics saying, “a nudge is any factor thatsignificantly alters the behaviour of Humans, eventhough it would be ignored by Econs” it becomesclear that, despite Sunstein’s assertion to the con-trary, the provision of factual information shouldusually not qualify as a nudge. This becomes evenclearer if we look at the two final conditions of thedefinition that captures this principle:

(3) Nudges are called for because of cognitiveboundaries, biases, routines, and habits in individualand social decision-making, and (4) work by makinguse of those boundaries, biases, routines, and habitsas integral parts of the choice architecture.

For one, the provision of factual informationwould not necessarily be called for because of cogni-tive boundaries and biases. It would merely be moti-vated by a person’s limited information. Second, un-der normal circumstances it would be intended towork in a way that would seek to influence Econs aswell as Humans. So while it is possible that the pro-vision of factual informationmay qualify as a case oflibertarianpaternalism, it doesnotqualify as anudge.

Perhaps a reason why nudges and the provision ofinformation may sometimes be confused is becauseinformationmay be provided in ways that are not in-tended as the provision of pure factual informationthat one did not have access to beforehand. Take re-minding someone of something, say, that I have anappointment with my dentist as in studies like thatof Altman and Traxler79. Here, conditions (1) – (3)would be satisfied. It would be an attempt to influ-ence my behaviour in a predictable way without for-bidding or adding any rationally relevant choice op-tions or changing my incentives, whether regardedin terms of time, trouble, social sanctions, economicand so forth. The reminder would be called for be-cause of my limited memory, which in turn may beconceived of as the result of cognitive boundaries giv-en Econs are assumed to have perfect memory –

78 Sunstein,Why Nudge?, supra note 6, at p. 55

79 Steffen Altmann and Christian Traxler, “Nudges at the Dentist”,(IZA Discussion Papers 6699, Institute for the Study of Labor,2012).


hence an Econ would never need a reminder. Butwould condition (4) be satisfied? That is, would thereminder be an attempt at influencingmy behaviourin a predictable way intended to work bymaking useof cognitive boundaries, biases, routines or habits asan integral part of that attempt? That is, am I not justbeing reminded in a way that could be described as‘de-biasing’? Following Sunstein reminding does sat-isfy condition (4) since itmakesuseof salience.80Thatis, it is not the information as such that causes the be-havioural effect or function, but that of making theinformation salient as a means of memory retrieval.Thus, as with the structuring of incentives, the inter-vention consists of two components - an information-al component and a behavioural function in terms ofsalience. In fact, this means that the whole so-callednotice paradigm as described by Ryan Calo qualifiesas nudging.81 This includes the infamous “fly in theurinal”which also satisfies condition4bymakinguseof salience; although it is called for due to anothercognitive boundary viz. limited attention, rather thanlimited memory. In general, such notice-tactics arepart of the nudge approach, although theymay oftenseem like mere provision of information. Still, thegeneral conclusion holds: the provision of factual in-formation in and by itself does not count as a nudge.

b. Arguments and Fallacies

Another tactic that is likely to give rise to similar con-fusion is that of rational persuasion or argument.Does this qualify as a nudge? Well, the answer is ev-ident given the term “rational”. For rational persua-sion and logical argumentation (1) and (2) will be sat-isfied as well as perhaps (3); though (3) will not besatisfied because of ‘limited memory’, but perhapsrather because of ‘limitedprocessing capacity’ in con-necting the facts. However, condition (4) will not besatisfied: rational persuasion or argument is not in-tended to work by making use of cognitive bias aspart of the attempt at influencing people’s behaviourin a predictable way. In so far as rational persuasionworks – not by providing new premises by means ofthe provision of factual information, but by connect-ing these premises into a conclusion – it should atmost be considered as a de-biasing strategy.

However, the reason to pick up the subject of ar-guments here is that not all persuasion is rational.One could make use of logical fallacies in order topersuade. So would that qualify as a nudge? Obvi-ously, the use of such fallacies would not convince aperfectly rational agent, that is an Econ. Yet, as iswelldocumented, fallacies are often used to convince andmislead Humans. Hence the use of fallacies to per-suade ultimately qualifies as nudging when consid-ering the principle from behavioural economics.

The same conclusion is also reached by applyingthe revised definition. The use of logical fallacies sat-isfies conditions (1) and (2) as the interventionwouldbe intended to influence behaviour, broadly con-ceived of, without forbidding or adding any rational-ly relevant choice options or changing incentives.Likewise conditions (3) and (4) aremet sincepersuad-ing someone by means of a logical fallacy would on-ly be possible due to cognitive boundaries and bias-es andwouldwork bymaking use of those biases andboundaries as an integral part of the attempt to per-suade. Hence, the use of logical fallacies qualifies asa nudge on this definition as well.

Given the conclusions reached in this discussionof information and fallacies, we may now revise thedefinition of nudge as follows:

A nudge is a function of the choice architecturethat alters people’s behavior in a predictable waythat works independently of(1) forbidding or adding any rationally relevantchoice options,(2) changing their incentives, whether regarded interms of time, trouble, social sanctions, economicand so forth, or(5) the provision of factual information or ratio-nal argumentation.(3) Nudges are called for because of cognitiveboundaries, biases, routines, and habits in individ-ual and social decision-making, and (4) work bymaking use of those boundaries, biases, routines,and habits as integral parts of the choice architec-ture.

5. Conclusion: Minimal Definition ofNudge

To reconcile the definition of nudge provided byThaler and Sunstein with the principle from behav-ioural economics we have revised most elements of

80 Sunstein,Why Nudge?, supra note 6, at p. 39-44.

81 Calo, “Code, Nudge, or Notice”, supra note 7.


the definition. However, at this point one may havenoticed that the negative conditions concerning op-tions, preferences, and information are derivativesfrom this principle, rather than vice versa. To empha-sise this as well as pre-empt the possibility that evenfurther negative conditions may be identified, I sug-gest the following re-arrangement of what I put for-ward as the minimal definition of a nudge:

Minimal DefinitionA nudge is a function of the choice architecturethat alters people’s behaviour in a predictable way(1) that is called for because of cognitive bound-aries, biases, routines, andhabits in individual andsocial decision-making and which (2) works bymaking use of those boundaries, biases, routines,and habits as integral parts of the choice architec-ture.Thus a nudge amongst other things works inde-pendently of:(i) forbidding or adding any rational relevantchoice options,(ii) changing incentives, whether regarded interms of time, trouble,social sanctions, economic and so forth, or(iii) the provision of factual information or ratio-nal argumentation.

V. Nudge and Libertarian Paternalism

Despite having reached a minimal definition ofnudge, it is not yet possible to determine the exactrelationship betweennudge and libertarian paternal-ism. Obviously a clear overlap exists between thesetwo concepts as it follows from the definition of anudge that nudges work independently of (i) - (iii).However, the question remains whether any particu-lar motivation, if any motivation at all, is to be tiedto nudges and how to deal with this conceptually inorder to achieve the wanted clarity and consistency.The purpose of this section is to examine that ques-tion.

1. Intentionality (A Nudge, or “to Nudge”)

A first answer to the question is found in the discus-sion about whether the concept of nudge should betreated as a noun or a verb.82 Initially, this may seema ridiculous point to make, but it turns out to have

some important bearings when one turns to theoret-ical issues as well as discussions of the ethics ofnudge.

In Thaler and Sunstein’s original definition, theconcept of ‘nudge’ is treated as a noun referring to“any aspect of the choice architecture” (my empha-sis). Thaler and Sunstein’s formulation of the princi-ple from behavioural economics takes the samestance since, there, a nudge is referred to as “any fac-tor that significantly alters the behavior of Humans,even though it would be ignored by Econs” (my em-phasis).

However, as in theMontyPythonsketch, a “nudge”usually serves as a verb designating an action.Hence,“to nudge” is an action intended by a subject. This al-so makes it possible to say, “nudging the ball” just asyou would say, “hitting the ball”. Likewise, as Thalerand Sunstein say, libertarian paternalists nudge,meaning that a libertarian paternalist may be nudg-ing. Of course, using the term nudge in this way doesnot exclude derivative uses implying a lack of inten-tionality, such as “he hit the wall” in the more unfor-tunate sense. But notice, using the term this waywould be derivative and not primary (like when wesay that “she unintentionally poisoned him”, we donot mean that she actually poisoned him). Also, in-tentionality does not imply that when someone“nudges” he necessarily does so with this concept inmind whether in sensu composito or in sensu diviso.You don’t need the concept of ‘bragging’ to actuallydo this, nor do you need to intend to ‘brag’ in indi-vidual instances in order to do it.

From the title of their paper “Debate: To Nudge orNot to Nudge”83 Hausman and Welch actually seemto take the “verb” approach to the question. Howev-er, on a closer read, it turns out that they oscillate be-tween treating “nudge” as a verb and as a noun. Ingeneral they treat nudges as “factors”.84 Yet, in theirsummary they refer to nudges as “ways of influenc-ing choice”. This formulation may in turn be under-stoodasanounreferring to systematic relations (bias,boundaries, etc.) in an objective world (as when say-ing “ways in which gravity affects the earth”), or asa verb describing the use of systematic competences

82 See Hansen and Jespersen, “Nudge and the Manipulation ofChoice”, supra note 25.

83 Hausman and Welch, “Debate: To Nudge or Not to Nudge”,supra note 22.

84 See e.g. Ibid, at p. 126.


involving some kind of intentionality. Either way, astheypoint out by the endof their paper there remainsan important difference between choices that are in-tentionally influenced by a third party, and choicesthat are not.85

It is for reasons resulting from this last point thatHansen and Jespersen argue for consistently treatingthe concept of nudge as a verb:

… there seems to be a clear and important distinc-tion to be made between a given context that ac-cidentally influences behaviour in a predictableway, and someone – a choice architect – intention-ally trying to alter behaviour by fiddlingwith suchcontexts.86

The reason why Hansen and Jespersen make thispoint is that inmatters of normative justification onesimply cannot dispense with the issue of intention-ality and, by extension, agency. Intentionality is aconceptual precondition of normative evaluation. Ig-noring it would render the notion of responsibilitysuperfluous.Unfortunately, asHansen and Jespersenalso point out, this is exactly what is often indicatedwhen treating the concept of “nudge” as a noun,rather than as a verb. Such a conceptual move blursa crucial distinction at the heart of normative justifi-cation as to the notion of responsibility. Hence theseemingly ridiculous question ofwhether to treat theconcept of “nudge” as a noun or a verb makes a cru-cial difference because it introduces intentionalityand thus the normative dimension of responsibilityinto the debate about the nudge approach to behav-iour change.

For this reason Hansen and Jespersen argue forstipulating the definition of nudge as being one of averb in order to ensure that this important distinc-tion is not lost.

Thus, we suggest that a nudge henceforth is bestunderstood as the intentional attempt at influenc-ing choice, while it is accepted that the settings ofany given decision-making context may acciden-tally influence choice andbehaviour inpredictableways aswell. …The notion of “nudge” then, shouldonly applywhen someone intentionally tries to in-

fluence our behaviour without the use of regula-tion or fiddling around with incentives.87

So returning to the revised definition we may nowfollow Hansen and Jespersen 2013 by integrating amore precise and viable condition substituting thepassive aspect of “choice architecture”with condition(I) stipulating that a nudge is a function of any (in-tentional) attempt at influencing people’s behaviour.While modifying this part of the definition we mayalso note that it may be beneficial to make explicitwhat is implicitly contained in the concept of behav-iour - namely judgment, choice as well as overt be-haviour - in order to avoid confusions on that point.

A nudge is a function of (I) any attempt at influ-encing people’s judgment, choice or behaviour ina predictable way (1) that is called for because ofcognitive boundaries, biases, routines, and habitsin individual and social decision-making, andwhich (2) works by making use of those bound-aries, biases, routines, and habits as integral partsof such attempts.Thus a nudge amongst other things works inde-pendently of:(i) forbidding or adding any rationally relevantchoice options,(ii) changing incentives, whether regarded interms of time, trouble,social sanctions, economic and so forth, or(iii) the provision of factual information or ratio-nal argumentation.

As we will see below, this at the same time clarifieswhy the terms nudge and bias are not logically equiv-alent and this hence precludes a logical substitutionthat otherwise would make for a conceptual oddity.

2. Who’s Calling? Or, Why Nudge?

So far we have determined that a nudge is an inten-tional attempt at influencing people’s behaviours ina predictable way under certain constraints. Howev-er, this raises the next question of whether a specialmotivation for forming this intention is required aswell in order for such an act to qualify as a nudge.For instance, is a libertarian paternalistic motive re-quired, or could other motives do as well? The addi-tion by means of condition (1) in the revised defini-tion adopted might indicate so by stating that

85 Ibid, at p. 133.

86 Hansen and Jespersen, “Nudge and the Manipulation of Choice”,supra note 25, at p. 10.

87 Ibid.


“nudges are called for” – but what position on thisprovides consistency and clarity?

Hausman and Welch’s formulation “nudges arecalled for” definitely seems to invite an interpretationin terms of themotivation behind the nudge. That is,it seems to refer to someone’s reason for intervening– the motive behind the (I) attempt at nudging. Forthis reasonwemay revise condition (1) so that it says:

(1) that is motivated because of cognitive bound-aries, biases, routines, andhabits in individual andsocial decision-making

But what kind ofmotive, if any, is required by the de-finition? Here, Hausman & Welch’s attempt at a de-finition says that nudges are called for “because offlaws in individual decision-making”.88 This meansthat condition (1) may be read as identifying a pre-condition for when a nudge is called for: given cog-nitive boundaries, biases (recall that we dropped theconcept of ‘flaws’), routines, and habits in individualand social decision-making, nudges are called for. Butwhy would that be? One possible and very likely rea-son would be that since boundaries, bias, routines,and habits may prevent one from acting accordingto the standards of rationality thennudges “are calledfor” because these biases pose barriers for people toperform (act, judge, evaluate) rationally. If we followthis interpretation we might thus consider revisingcondition (1) such that we get:

(1) that is motivated because of cognitive bound-aries, biases, routines, andhabits in individual andsocial decision-makingposingbarriers for peopleto perform rationally in their own declared self-interest.

Now this addition explicates the notion of cognitiveboundaries and biases, as well as certain routines,and habits. But why does the occurrence of cognitiveboundaries, biases, routines, and habits result in areason to nudge? That is, who or what is calling forthe need to nudge? As it turns out, a lot hangs on theinterpretation of this subtlety only implicit even inHausman andWelch’s attempt at explicating the de-finition provided by Thaler and Sunstein. I find thattwo possible interpretations are readily available.

a. Nudging as Libertarian Paternalism

First, we may interpret the definition as requiringthat the motive be paternalistic. That is, because of

cognitive boundaries and biases in individual and so-cial decision-making, the nudged agent is preventedfrom acting rationally according to his own interestas judged by the one nudging or as judged by him-self – and this may be taken to call for a paternalisticintervention, which as a matter of definition is donein the interest of the people nudged (as judged by theone nudging or as judged by himself). This is the po-sition taken by Sunstein in his recentWhy Nudge?89

If this interpretation is chosen, we have that:A nudge is a function of (I) any attempt at influ-encing people’s judgment, choice or behaviour ina predictable way according to their own self-de-clared interests (1) that is motivated because ofcognitive boundaries, biases, routines, and habitsin individual and social decision-making posingbarriers for people to perform rationally in theirown self-declared interests, and which (2) worksby making use of those boundaries, biases, rou-tines, and habits as integral parts of such attempts.Thus a nudge amongst other things works inde-pendently of:(i) forbidding or adding any rationally relevantchoice options,(ii) changing incentives, whether regarded interms of time, trouble,social sanctions, economic and so forth, or(iii) the provision of factual information or ratio-nal argumentation.

If choosing this interpretation of themotive in (1) weessentially marry the concept of nudge to that of lib-ertarian paternalism. Hence we may refer to this de-finition as the LP-definition. Under this definitionnudges are a subset of libertarian paternalism. SeeFigure 2.

b. Nudging in the Technical Sense

Alternatively we may interpret the definition as notmaking any specific requirements as to the motive,other than that the motive is dependent on or madepossible by the fact that cognitive boundaries, bias-es, routines, and habits in individual and social deci-sion-making pose barriers for people to perform ra-

88 Hausman and Welch, “Debate: To Nudge or Not to Nudge”,supra note 22, at p. 126.

89 Sunstein,Why Nudge?, supra note 6.


tionally. If we choose this strictly technical interpre-tation we get a definition saying that:

A nudge is a function of (I) any attempt at influ-encing people’s judgment, choice, or behaviour ina predictableway that is (1)made possible becauseof cognitive boundaries, biases, routines, andhabits in individual and social decision-makingposing barriers for people to perform rationally intheir own self-declared interests, and which (2)works bymaking use of those biases, routines, andhabits as integral parts of such attempts.Thus nudges amongst other things work indepen-dently of:

(i) forbidding or adding any rationally relevantchoice options,(ii) changing incentives, whether regarded interms of time, trouble,social sanctions, economic and so forth, or(iii) the provision of factual information or ratio-nal argumentation.

Obviously, “madepossible” isweaker than “called for”and thus, different from the LP-definition, the tech-nical definition is not married to the notion of liber-tarian paternalism. Hence we refer to it as the tech-nical, or the neutral definition. Under this definitionthe set of nudges overlaps with that of libertarian pa-ternalism, but is not contained in the latter. See Fig-ure 3.

Now, to decide which definition to prefer, onecould turn exegetic. However, inmy opinion, re-read-ing what has been written by Thaler and Sunsteinabout nudges to search for clues in one or the otherdirection would not lead to any decisive conclusions.Whether they favour the LP-definition or the techni-cal definition does not decide the question of whichstipulative definition we should opt for. However, asI have argued elsewhere there is one set of exegeticreadings that lend evidence to the technical defini-tion as the best stipulative definition of the conceptof nudge given the way that this concept has cometo be adopted in discussions surrounding the nudgeapproach to behaviour change.90

c. “Nudge for Good”

First, the technical definition is consistent with cer-tain central commentsmade by Thaler and Sunstein.

In general Thaler signs copies of Nudge with theslogan “nudge forgood”.Thisprecautionary call clear-ly indicates that the notion of ‘nudge’ is not neces-sarilymarried to that of ‘libertarian paternalism’, butmay instead be cast in terms of the technical defini-tion – because if thisweren’t the case, Thaler’s phrasewould be nothing more than a tautology. Somethingsimilar goes for the title of Sunstein’s recent bookWhy Nudge? After all, if the LP-motive were part ofthe definition of a nudge, then one would not haveto answer the question of ‘why nudge?’ – one wouldonly have to define the term. In addition one mayobserve that when Sunstein says that the “Provisionof information is certainly a nudge, but itmay ormaynot qualify as paternalistic…”,91 this seems to reveal

90 Pelle G. Hansen, “Nudge for good”, Policy Options, 3 Jun 2013,pp. 22-23 et sqq.

91 Sunstein,Why Nudge?, supra note 6, at p. 55.

Figure 2

Figure 3


that nudges that fall outside the scope of libertarianpaternalism exist and hence the set of nudges can-not be a subset of libertarian paternalistic measures.These considerations support the adoption of thetechnical definition.

Second, the technical definition is consistent withpoints about nudges and marketing consistentlymade by commentaries. If we opt for the technicaldefinition it makes sense to refer to the many mar-keting tricks used to fool us into buying things wedon’t need as nudges as well. That is, the technicaldefinition recognises that it is possible to nudge “forbad” or “for profit” as well as “for good”. This obser-vation is both in line with many comments made inthe general literature and mentioned here in the in-troduction as well as many formulations in the workof Thaler and Sunstein themselves.

Thirdly, when the nudge approach to behaviourchange is regarded as a sub-branch of libertarian pa-ternalism or as synonymous with this, there is a dan-ger that anyone in favour of the former must also bein favour of the latter. That is, if one sees nudges asvaluable measures for creating behaviour change,one is often taken as favouring the political ideologyof libertarian paternalism - an ideology that by someis said to hold that policy makers should avoid regu-lations that limit choice (bans, caps, etc.) but can usebehavioural science to direct people towards betterchoices. However, this is not necessarily so on thetechnical definition. Here one may adopt nudges asvaluable measures to behaviour change withoutadopting a particular political ideology.

There are probably other reasons for stickingwiththe technical definition of nudge. However, I will ac-cept the three mentioned above as sufficient as theylend evidence to the technical definition as the beststipulative definition of the concept of nudge, refin-ing what has come to be standard usage in the recep-tion of Nudge. That is, while one may choose eitherone as one’s favourite definition, it seems that there

is a conceptual need for adopting the technical onesince it is better at providing clarity when discussingapplications of behavioural economics and other be-havioural sciences with the aim of creating behav-iour change. For this reason I conclude that the mostsuitable definition of nudge is as follows:

A nudge is a function of (I) any attempt at influ-encing people’s judgment, choice or behaviour ina predictableway, that is (1)madepossible becauseof cognitive boundaries, biases, routines, andhabits in individual and social decision-makingposing barriers for people to perform rationally intheir own self-declared interests, and which (2)works by making use of those boundaries, biases,routines, and habits as integral parts of such at-tempts.Thus a nudge amongst other things works inde-pendently of:(i) forbidding or adding any rationally relevantchoice options,(ii) changing incentives, whether regarded interms of time, trouble,social sanctions, economic and so forth, or(iii) the provision of factual information and ratio-nal argumentation.

In so far as a nudge serves the self-declared interestof those being nudged, it may further be referred toas libertarianpaternalism since the general definitionof nudge implies that people’s judgment, choice, orbehaviour is influenced in ways that work indepen-dently of (i) forbidding or adding any rationally rele-vant choice options, (ii) changing incentives,whetherregarded in terms of time, trouble, social sanctions,economic incentives and so forth. In addition, thisdefinition also implies that libertarian paternalismgoes beyond nudging since it follows from it thatnudges work independently of (iii) the provision offactual information and rational argumentation,which fall squarely within libertarian paternalism.

Documents

The Definition of Nudge and Libertarian Paternalism: Does ... · seller titled Nudge: Improving Decisions About Health,Wealth,andHappiness.Beforehandmostfa-mousfromMontyPython’ssketch“nudge,nudge,