We propose libertarians to compensate for external effects

Nudging as a Political Instrument - Good Intent or State Invasion?

Behavioral economics and its practical implications are increasingly becoming the focus of German politics. Individual decisions should be influenced “gently” in the sense of “libertarian paternalism”. The "nudges" consist of standard specifications, self-commitments and the provision of information. However, their application requires the clarification of some questions: Who can presume to make “wise” decisions for economic agents? In which decisions can interventions by another authority be justified? With what time horizon and on the basis of what welfare considerations is a decision defined as "correct"? Which concept of rationality is behind the concept?

Is it in the citizens' interest if their government implements nudges?

Lisa V. Bruttel, Florian Stolley

Not only since the book “Nudge - How to initiate smart decisions” by Richard Thaler and Cass Sunstein has been known 1 that people have difficulties in making optimal decisions for themselves. They exercise less and eat less healthily than they intended, they don't put enough money aside for old age, and they postpone unpleasant tasks and decisions for longer than they would like. In addition, experimental economic research has repeatedly shown that people make systematic errors even when judging alternative decisions in a supposedly rational manner.2 According to Thaler and Sunstein, this lack of self-control and rationality can be compensated for by so-called “nudges”. The idea is to use a cautious, quasi "minimally invasive" redesign of the decision-making situation, e.g. to facilitate complex decisions or reduce self-control problems, without de facto curtailing the freedom of choice. Many people, according to Thaler and Sunstein, are genuinely grateful for well-constructed nudges, such as placing healthy food at eye level in a cafeteria, which makes eating unhealthy food only slightly more difficult. If such nudges are used by the state, Thaler and Sunstein speak of a concept of “libertarian paternalism” as opposed to “real” paternalism, such as the prohibition of unhealthy food. Examples of situations where individuals are acting out of their own interest and where nudges can help are numerous. We would like to introduce a selection in more detail first:


The most well-known type of nudges are certainly standard specifications, the defaults. These are intended to "nudge" people in a recommendable direction by offering an optimal standard as a pre-selection in decision-making situations. In their book, Thaler and Sunstein describe a few examples that clearly show how strongly standard guidelines influence the decision-making of individuals. The best-known and most cited example is participation in the company pension scheme in the USA. If joining the company pension scheme is automatic and an employee who does not want to have to actively object to joining, significantly more employees opt for a company pension scheme than in the opposite case.3 In Germany, however, this question hardly arises because it is here The pension system ensures significantly better old-age provision compared to the US - incidentally, a form of paternalism that is rightly hardly questioned


But nudges are much more than the well-known defaults. In the examples of a lack of self-control mentioned above, the core of the problem often lies in time-inconsistent behavior.5 The decision to do sport and thus the weighing up between the "costs" (exertion, opportunity costs of time) and the "benefits" (fitness, health ) is optimally taken every year at the New Year. However, if the decision is to go to sport right now, the optimality calculus is a different one, since the "costs" are incurred directly and are felt to be relatively high immediately, while the long-term "benefit" only occurs with a delay and thus tends to be underweighted in the current optimality calculation becomes. Nudges that change the incentive structure - offered by the state or the private sector - can help to cope with these self-control problems. For example, betting with other people as to whether you will be able to act in a certain way in the future (e.g. quitting smoking) works comparatively well. Losing such a bet has direct consequences, both social (admitting failure to an acquaintance) and possibly financial (handing over the winnings to an acquaintance). The consequences of a bet change the incentives, which can change behavior.6 This type of self-restraint is chosen completely freely, so that freedom of choice is granted ex ante. In a sense, such self-control nudges are therefore more harmless than the more popular defaults.


Nudges can also improve people's decisions if they come in the form of well-prepared information. As documented in particular by Kahneman and Tversky since the 1970s, 7 various cognitive biases and incorrectly applied rules of thumb often lead people to make irrational decisions. Without describing these distortions and rules of thumb in detail, it should at least be mentioned that people have difficulties primarily with assessing and analyzing complex and rare decision-making situations. Nudges can help here by making it easier for people to filter out the important and central information, which makes it easier to assess and compare different alternatives. For example, in the insurance sector, precise information about probabilities and damage sums is necessary in order to make rational decisions. Information nudges, which provide this information in a clear form - possibly even individualized for the respective situation - can help here

Even if you can see nudges as very useful, there is some skepticism in the public discussion about state-used nudges. The criticism ranges from the accusation of paternalistic paternalism in private matters to the presumed exercise of coercion through nudges.9 This skepticism may appear justified in the light of a paternalistic tendency in German and European politics that can be observed. While Thaler and Sunstein are still reluctant to propose the introduction of intelligent lamps and cables that are intended to show consumers their electricity consumption vividly and thereby reduce consumption, the European Union bans the sale of lightbulbs.10 Smoking bans can be further examples of paternalistic tendencies in politics State-11 or at the state level12 in Germany or the obligation to have child preventive examinations in some federal states, which was introduced as a result of publicized cases of neglected or abused children.

To be clear: we are by no means opposed to these measures. It is only intended to make it clear that there is currently a tendency towards paternalistic politics to be observed in Germany and the EU and that this tendency may even increase the skepticism with which the use of nudges is viewed.13 The fear that nudges could be on the part of be used by the government to subconsciously influence citizens - even against their will - and the state could increasingly interfere in private affairs through the use of nudges, at any rate seems entirely justified.14 In certain decision-making situations that are particularly ethically controversial, for example, Many citizens rightly feel uncomfortable when their government tries to influence decisions. One example is the discussion that recently flared up again about the shortage of donated organs, and the question of whether a contradicting solution that would lead to a dramatically higher number of organ donors 15 would be morally justifiable instead of the currently valid consent solution.16

The example of organ donation shows that the state must use nudges carefully and transparently and that particular attention should be paid to the question of which area these decisions affect. Failure to observe these principles would generally increase criticism of the use of nudges. There are enough application examples in which the sensible and transparent use of nudges is supported by almost all sides. A blanket demonization or praise is therefore not effective. Nudges have to be assessed in a more differentiated manner according to their type and purpose.

Information nudges

A broad consensus for the use of government nudges is particularly to be expected if the goal of a nudge is to empower citizens to make more informed choices by providing them with more information. Nudges in the form of the provision of information can be roughly divided into two groups: information that serves to protect weaker market participants, and information that is intended to steer behavior in a desirable (individually and socially) direction.

Provision of information to protect weaker market participants makes sense when relevant information that enables a simple comparison of the alternatives is not easily available in decision-making situations. For complex decisions in particular, the state can stipulate clear rules as to which information the stronger or better informed market participant has to pass on to the weaker or less informed market participant and in what form. A so far unsuccessfully established, but very promising proposal for such an information nudge, which consumer advocates vigorously advocated17, is the food traffic light. This should make it easier for the consumer to obtain health-relevant information about the ingredients of food. However, the labeling was considered to be overly simplistic and in parts misleading, so that the European Union has initially spoken out against the mandatory introduction of such a food traffic light. 18

Information nudges with the aim of controlling behavior can be useful, among other things, in the area of ​​energy consumption. As already mentioned, the EU has decided against a nudge and in favor of a "classic" ban in order to reduce the use of light bulbs, which consume a lot of energy. We believe that Thaler and Sunstein's proposal to give citizens direct feedback on their electricity consumption can have a positive effect on behavior. Electricity consumption is something very abstract and an annual billing hardly enables anyone to correctly estimate the costs of everyday consumption. The installation of intelligent electricity meters in private households, which can not only control consumption in a sensible way, 19 but also provide direct, easily perceptible feedback on everyday energy consumption, would enable consumers to control their electricity consumption and realize savings potential without them to make any prescriptions. 20

Self-control nudges

We also consider cautious government support with regard to self-control problems to be harmless. An example of a nudge that has already been introduced by the state is the self-lock when gambling in casinos in Germany. In addition to government offers to help with gambling addiction, players who fear that they will lose control of themselves in the future can apply for a self-lock, which will deny them access to casinos throughout Germany.21 Self-lock is completely voluntary, so the The individual's freedom of choice is preserved.

Furthermore, we could imagine that the state - for example in the form of the Federal Center for Health Education (BZgA) - supports the establishment of an Internet platform that makes it easier for interested citizens to make resolutions - such as quitting smoking or doing more sport - to implement. The privately run American website StickK.com could serve as a model here, which enables its users to conclude binding contracts to solve self-control problems.22 Here, a project is specified and an acquaintance named who checks whether the project was successfully completed , and an amount of money blocked, which the participant will get back after successful implementation of the resolution. If the project does not succeed, the money goes to a previously determined institution or person. The desire not to lose the money and not to have to admit failure to friends successfully serves as a motivation to actually implement the resolution. Similar to the self-lock in gambling, participation in this offer would of course be completely voluntary - and a state-run website might even be more trustworthy than a private provider.

Protection against the disadvantageous commercial use of nudges

In addition to the possible active use of nudges, the state also has an important role to play in protecting its citizens from exploitation through the commercial use of nudges, especially in the area of ​​default nudges. Because the state not only has the option to set standard specifications, but also to prohibit certain ones, e.g. to protect weaker market participants. A number of sensible measures have already been implemented in Germany and Europe. In addition to a consumer-friendly legal situation for distance selling contracts, which is intended to prevent the exploitation of information advantages for retailers, retailers were forbidden, for example, from ticking an online customer's subscription to a newsletter.23 The situation is similar with flight bookings on the Internet where the Purchase of additional insurance may not be preset. 24


Nudges are more than just setting defaults. Good nudges can reduce self-control problems or, by providing information, improve and facilitate decisions. However, there is a certain skepticism about the state use of nudges, as certain variants can also be understood as manipulation and paternalism. We therefore advocate that their use must be cautious and transparent and that a discussion is necessary as to the areas for which the use of a nudge can be justified. The view that the state should protect its citizens from the disadvantageous use of commercial nudges by stipulating the provision of easily understandable information or regulating the use of defaults is certainly largely undisputed. There are also areas of application in which we consider it harmless, if not even desirable, for government agencies to actively construct nudges themselves. This includes in particular those nudges that provide citizens with clear information (e.g. about their electricity consumption) and thereby make them more responsible in their decisions, as well as nudges that offer citizens supportive offers to get their self-control under control if they themselves make this resolution grasp.

  • 1 R. H. Thaler, C. R. Sunstein: Nudge - How to initiate smart decisions, Berlin 2009.
  • 2 Cf., among others, the many contributions by Daniel Kahneman and Amos Tversky, summarized in: D. Kahneman: Fast thinking, slow thinking, Munich 2012.
  • 3 Cf. B. Madrian, D. Shea: The Power of Suggestion: Inertia in 401 (k) Participation and Savings Behavior, in: The Quarterly Journal of Economics, 116th vol. (2001), no. 4, p. 1149 -1187.
  • 4 As a further example of a default nudge, for which most citizens are very grateful, Thaler and Sunstein name the time change in summer, which makes it easier to implement the resolution to get up earlier in the warm season. For 67% of Germans who would prefer to abolish summer time, see O.V .: Hohles Promise, in: Stern, No. 14 of March 27, 2014, p. 21, however, that is clearly too much of a paternalism.
  • 5 Cf. inter alia D. Read, B. Leeuwen: Predicting Hunger: The Effects of Appetite and Delay on Choice, in: Organizational Behavior and Human Decision Processes, Volume 76 (1998), No. 2, p. 189 -205.
  • 6 Computer programs, for example, with which one can block certain websites for a pre-set time in order to work more efficiently, and which have probably already saved several students from submitting their thesis on time, have a similar effect.
  • 7 D. Kahneman, loc. Cit.
  • 8 In addition to a better informed decision by individuals, these nudges could also be used to make only certain information clearer and thus more present and thus to influence the individual decision in a direction desired by the state. In many respects, this corresponds to what companies have always tried to achieve with advertising, and it is more in the direction of pure paternalism.
  • 9 P. Plickert, H. Beck: Chancellor seeks behavioral researcher, http://ow.ly/DQsSj (August 26, 2014); P. Plickert: The great tutelage by behavioral economists, 2012, http://ow.ly/DQsY6 (March 16, 2012).
  • 10 Regulation (EC) No. 244/2009 of the Commission of March 18, 2009 implementing Directive 2005/32 / EC of the European Parliament and of the Council with regard to the specification of requirements for the environmentally compatible design of household lamps with unbundled light.
  • 11 Law on the introduction of a smoking ban in federal facilities and public transport (Federal Non-Smoking Protection Act - BnichtrSchG) of July 20, 2007.
  • 12 An overview of the individual smoking bans in the federal states can be found here: http://ow.ly/DQtoC (21.10.2014).
  • 13 K. Horn: Slave owners of the future, http://ow.ly/DQt7f (11.3.2013).
  • 14 Cf. J. Schnellenbach: Benevolent pushing: What can be achieved with liberal paternalism and what are its side effects ?, in: Perspektiven der Wirtschaftsppolitik, 12th year (2011), no. 4, pp. 445-459.
  • 15 Cf. E. Johnson, D. Goldstein: Do Defaults Save Lives ?, in: Science, 302. Jg. (2003), pp. 1338-1339.
  • 16 O.V .: Number of organ donors falls to a historic low, http://ow.ly/DQrBI (15.1.2014); M. Bartsch: Proposed law: ministers want to declare every German to be an organ donor, http://ow.ly/DQrIU (1.6.2011).
  • 17 B. Weitzel: Customers choose red-yellow-green, http://ow.ly/DQqMS (16.6.2009).
  • 18 D. Kuhr: Aigner rejects food traffic lights, http://ow.ly/DQr8g (3.7.2013).
  • 19 O.V .: Federal government says goodbye to intelligent electricity meters, http://ow.ly/DQrgo (October 1, 2014).
  • 20 Such information nudges are also often used in the health sector. In doing so, the state tries to make health risks clear and to change behavior with regard to these risks. The Federal Center for Health Education (BZgA), for example, provides information on correct behavior during pregnancy and on how to deal with small children that expectant parents are offered at various points. For example, there are campaigns against smoking during pregnancy or campaigns that point out the dangers of shaking babies (see www.kindergesundheit-info.de or www.familienplanung.de).
  • 21 §8 Player ban, State Treaty on Gambling in Germany (State Treaty on Gambling - GlüStV) dated December 15, 2011.
  • 22 This platform was founded by professors from Yale University.
  • 23 Federal Court of Justice, judgment of July 16, 2008, VIII ZR 348/06.
  • 24 District Court Frankfurt am Main, judgment of January 22, 2014, Az. 2-06 O 379/13.

Nudging: Authority thinking and administration in a new guise?

Werner Güth, Hartmut Kliemt

In “Asterix bei den Goten” 1 we find Ovid's sentence “video meliora proboque deteriora sequor” (I see the better and my name is good, I will pursue the worse). Walter Mischel used his marshmallow test - children who have to resist the sweet temptation in order to receive double the amount afterwards - not originally for diagnostic purposes, 2 but to find out how children already react to their “too great” weakness for the obvious arm yourself against the distant. These strategies for dealing with emotions are what are interesting, not the fact that we are tempted. Odysseus can be tied to the mast, the overweight person removes the chocolate from the household so that he cannot give in to immediate temptation. Some of us don't drink alcohol before sunset or, like Leonid Breshnew, have a cigarette box, the contents of which can only be accessed once an hour.

We are even willing to pay for self-commitment opportunities. Arms in particular seem to do that. In developing countries, for example, they take out loans at high interest rates in order to immediately reinvest the amount they will need in the future at a bank at much lower interest rates. They pay to become the potential victim of debt collectors just to be able to achieve the long-term goal regardless of their everyday weakness.3 In the developed world, of course, the problem also exists, and, as everywhere, not only for the poor. Specialized self-binding contracts are even offered under legally developed institutions. 4

If Odysseus asks for a mast to which he wants to be tied, from the standpoint of individual decision-making autonomy, nothing can be said against it, if he is made appropriate offers on free contract markets. If, however, there should be market failure and no suitable self-commitments are offered, temptations arise for the state to act as a provider. He should do this in the event of market failure, if it is. It is a completely different matter, however, if the state does not only act as a mast builder and mast provider, but wants to get Odysseus to commit or even oblige him to commit himself. This is exactly what so-called “nudging” is all about. It is assumed that there are insufficient opportunities to bond or that these are not used adequately due to human weakness. As a political countermeasure, Odysseus is not only offered a mast, but is tied with light shackles that allow him to be delivered without much effort. This happens in his favor and without his prior consent ("paternalism"), but with the option of breaking free from the bond ("libertarian").

A weighty example

Imagine someone who is overweight and someone who has cleared the refrigerator, but who could stock up again at any time. It is good for him not to be able to give in to temptations immediately. Because he's slow to make decisions, he doesn't go to refill the refrigerator with purchases, but he could do that at any time. That sounds harmless and at first it is. If, for example, “the buddies” clear out the icebox in close friendship, they are doing him a favor that he can evade at any time. He can insist that the well-filled refrigerator return to normal (default). Friends cannot always be relied on in such matters and the state may want to intervene to help overweight people in the struggle with their own pounds. The enactment of a “refrigerator filling ordinance” for people with a body mass index that is too high would be “typical”. Since it would require prior encroachments on privacy, one would hardly want to accept such a thing. The absurdity of the example is not accidental, however. She points out that acceptable nudging not only requires the option to exit at low cost, but also the ability to initiate (the nudge) without massive interference in the private and / or public spheres. However, this will only be possible in exceptional cases.

Building up low-calorie foods in the school cafeteria and then, later and more difficult to reach, high-calorie foods is one thing, getting all cafeterias to do this is quite another. If we were to require the state that all coffee houses have to do this, then they might simply not want to include low-calorie products in their programs or we would have to order all coffee houses to include low-calorie products in their programs. Many turns of the intervention and regulation screw threaten. And who then "nudged the nudgers"?

Dangerous nudging

On closer inspection, many of the phenomena associated with nudging are surprisingly complicated. This indicates that the implementation of the concept is problematic regardless of the initial plausibility of many examples. As soon as nudging is made a program of politics, it will no longer contain advice, but fixed regulations. In any case, the simple resignation or resignation option alone can no longer ensure the libertarian character. Any nudging carried out by private individuals in relation to private individuals within the framework of established rights remains libertarian for the time being. Creating more information about how we can manage ourselves better and have ourselves nudged at our own request is certainly not to be criticized. It becomes problematic if one wants to implement by the state pre-contractual or “in contrahendo”, nudges, which then cannot go back to pre-contractual approval. The argument that cannot be avoided, that an option must be defined as “default” in a decision-making situation and that it will occur if nothing is done, is also valid in the case of decisions influenced by the state.

If one considers the enforcement of a decision to be legitimate, then it is certainly sensible to design the “default option” in such a way that it protects the interest of the decision maker or that of others who have a legitimate interest in the matter. The previously exercised coercion is by no means libertarian, but if the state has to exert coercion in certain spheres anyway - and everyone who is not an anarchist must accept this - then its interest-oriented design is certainly desirable. The compulsion as such always requires justification. The exit option and the alignment of interests mitigate the inevitable obsessional evil, but we must not allow the popularity of nudging to desensitize us to the negative side of compulsion. Let us take a closer look at the matter.

Dangerous bonds

We expect modern constitutional states to provide opportunities for legal self-binding. In order to support the provision of this service, the legislature must be empowered to enact rules. The authorized legislature can, however, under the guise of wanting to expand our free self-determination and protect us from self-damaging abuse, become an instrument of “paternalistic” interest politics. We ourselves know that the greater the power of attachment, the greater the greater the risks for us. These problems, related to the downside of nudging, need to be kept in mind when properly determining the state's role.

But check who binds himself forever

In our law, there are special rescission clauses for door-to-door sales and others for entering into contracts on the Internet. The citizen is protected from rushing to bind himself. He can even pronounce a ban on himself that forbids him to enter casinos. The legal handling of the game is otherwise instructive. Because the legal system does not generally enforce gambling debts. It also does not allow it to be possible to gamble on credit (so that the individual who enters a casino can partially control his own spontaneous desires with a previously counted sum of money). All of these regulations have a (self-) paternalistic aspect. The so-called freedom of contract is a state-defined and limited authorization to modify legal claims. Not all contracts are enforced by the state, only certain ones. Such a restriction includes the waiver of government enforcement of classes of contracts, not the prohibition of corresponding voluntary transactions. In the end, the citizen should be able to decide for himself about his ties. In certain cases, however, it is protected from state production of permanent effects. Long-term contracts that do not contain a withdrawal or exit clause are problematic because of the risks they create. In itself, one could imagine that the state offers to enforce marriage contracts with as well as those without divorce. It would then be left to the responsible citizens to choose which “mast” they want to be tied to. But even Wilhelm von Humboldt thought that an earlier self should not be allowed to prevent the later manifestations of the same person from reorienting in this final way.5 This “paternalistic” element in the range of instruments of self-attachment seems inevitable.

However, the legal system appears to be less sensitive to the automatic renewal of contracts. On the one hand, like the citizens, it wants adults to be able to conclude savings contracts that are automatically renewed. In view of their inclination not to make a decision, the citizen can commit himself to the continuation of his saving decision in a mild way. He foresees that the amounts will be debited from the account and that he will probably not have the resolve to change the status quo of contractual obligations. On the other hand, the providers of mobile phone contracts, for example, understand very precisely how they can use the corresponding psychological mechanisms for themselves. Shouldn't one protect citizens from certain self-renewing contracts and suggest other contracts? It is clear that there must be no general constitutional rule against self-extending contracts if one wants to preserve the good sides of such soft bonds, for example in the case of self-extending savings contracts. You need very good reasons why the state can discriminate against similar types of contract. Finally, an essential function of the private legal system is to enable private subjects to self-bind and then to let the corresponding bindings become effective with the means of state enforcement. But the discrimination of certain ties and the promotion of other ties is comparatively harmless in view of the much stronger state interventions, which are now to be brought closer to us as nudging.

Commitments and real nudges

In Germany, the so-called “statutory insurance” is sometimes mentioned as an example of nudges. Compulsory health insurance and compulsory pension insurance, like automobile liability insurance, are not to be understood in the sense of "nudging". If they were mere nudges, then all human beings would have to be allowed to opt out of the systems completely. That's not the case. It is not a question of (self-) paternalistic regulations, but of compulsory devices to ward off externalities. It is true that statutory health insurance is often represented with the argument that those who cannot take out health insurance for themselves because they are too unreasonable or because they are too poor must be supported for their own long-term benefit. But compulsory insurance does not primarily protect those with compulsory insurance, but above all society. We would be subject to a Samaritan dilemma.6 For we would want to pay because, under the rule of law, we cannot stand to let citizens die for their mistakes. There are good reasons for earmarked health taxes. But wanting to see nudges in them leads to perversions such as the release of the wealthy from the tax and its redistributive component, which is borne solely by the less wealthy. In the case of real nudging, the state must not deprive us of our final power of disposal. However, compulsory insurance does just that.

Final consideration

It seems to be entirely in keeping with Franz Böhm's7 regulatory conception of a private law company if the state enforces any self-commitments, but leaves their choice and scope to the citizens themselves. But even Wilhelm von Humboldt in his “Ideas for an attempt to determine the limits of the effectiveness of the state” 8 rejected the effective enforcement of any treaty by the state because of the inherent risks. Even in his libertarian approach, citizens are paternalistically protected from the dangers of excessive legal self-commitment through resignation and restriction clauses. It should be clearly understood that one wants to go further than Humboldt in the so-called “libertarian paternalism” that is associated with nudging today. The decision-making situation in which individuals decide for themselves about self-commitments should be influenced by the state. True nudging is not about forcing something on the citizen, but in pre-contractual decision-making situations in which he can still decide against certain actions that are generally desirable, to promote the choice of certain actions. Especially when this restricted intervention is based on “choice editing” based on behavioral theory, it has positive sides. But it remains more problematic than the state's refusal to allow certain contractual obligations. The latter, more harmless means should be used first. Especially when it comes to protecting an increasingly aging population from sticking to their well-worn contractual relationships, the state may well refuse to make certain types of contracts legally enforceable. The main problem will be to do this according to general principles that do not encourage age-discriminatory politics of interests.

  • 1 R.Goscinny, A. Uderzo: The adventures of Asterix the Gauls - Asterix near the Goten, Vol. 3, Paris 1963 or Berlin 1970, p. 20.
  • 2 See W. Mischel: The Marshmallow Test: Mastering Self-Control, 2014.
  • 3 See A. Banerjee, E. Duflo: Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty, Reprint, New York 2012.
  • 4 See I. Ayres: Carrots and Sticks: Unlock the Power of Incentives to Get Things Done, New York 2010.
  • 5 W. von Humboldt: Ideas for an attempt to determine the limits of the effectiveness of the state, Stuttgart 1851/1967.
  • 6 J. M. Buchanan: The Samaritan's Dilemma, in: E. S. Phelps (Ed.): Altruism, Morality and Economic Theory, New York 1975, pp. 77 ff.
  • 7 F. Böhm: Privatrechtsgesellschaft und Marktwirtschaft, in: ORDO, Vol. 17, 1966, pp. 75-151.
  • 8 W. von Humboldt, loc. Cit.

Implications of nudging for consumer welfare

Steven Bosworth, Simon Bartke

Three recently advertised positions at the Federal Chancellery show that the Federal Government is interested in nudging. Staff in this unit and related decision-makers should be aware that creating successful nudges requires an understanding of how human decision-making works and how it affects well-being. This article therefore introduces anomalies in the economic decision-making that nudging is intended to target. In addition, the possible effects of nudges on the well-being of consumers are systematically analyzed and thus an orientation for the development of appropriate nudges is provided.

A nudge is an intervention that is intended to influence human decisions without limiting the number of possible options. The concept of nudging stems from behavioral economics that show that people do not always make the decisions that maximize well-being. This can occur because people are unable to include all relevant aspects in the decision, they have unreasonable expectations about the consequences of their decisions, or because they lack the willpower to actually implement the preferred alternative. Furthermore, preferences can depend on the environment in which decisions are made. These biases in decision-making, hereinafter referred to as “anomalies” (biases), have been discovered, tested, and modeled in decades of behavioral research.1 Different types of anomalies require different nudges and have different implications for well-being.

One source of anomalies in decision-making are psychological or cognitive costs that arise from comparing different decision alternatives with one another. These costs can increase with the complexity of the decision problem and can be viewed as a type of transaction cost. Other anomalies result from distorted attitudes. The behavioral economic evidence shows the tendency of people to be too optimistic about their own abilities compared to others. People also tend to see patterns in completely independent events in which no pattern can exist. The preferences of decision-makers can also change in a predictable way between the moment the decision is made and the moment the consequences of the decision become apparent (anyone who has tried dieting or wanted to quit smoking can certainly put themselves in the shoes of it ). In addition, people assess the attractiveness of different decision alternatives by comparing them with a reference state. However, this reference state also changes over time depending on one's own circumstances. One finding from the “Prospect Theory” 2 is that a given loss in relation to this reference state has a stronger influence on the attractiveness of a decision option than an equally high gain.

People's wants and needs can be influenced by the context in which they are at the time the decision is made. This can be due to how a decision-making environment is perceived or to the interaction between the environment and the affective state of the decision maker (e.g. a positive attitude due to good weather). Subtle changes in the presentation of the decision options can therefore have decisive effects on the final decision. The susceptibility of human decision-making behavior to such factors increases the difficulty of assessing the well-being of the decision maker after such an influenced decision and of assessing whether a nudge that led to this decision has increased welfare.

Overcome anomalies with the help of nudges

Nudges that maximize consumer welfare can mean costs or additional regulation for producers. These can degrade producers' welfare. This issue should be considered by relevant decision-makers, especially because of possible political-economic effects, but is beyond the scope of this article. Since the decision-making anomalies of the actors can lead to decisions that, after careful consideration, they would have liked to have made differently, measures can be used by political decision-makers to limit the harmful effects of these anomalies. This reasoning motivated traditional paternalistic measures such as taxes on cigarettes, the ban on narcotics, and regulation on usury. The nature of these decision anomalies, however, also opens the door to "libertarian paternalism". In contrast to pure paternalism, this does not influence decisions by forbidding or restricting options, the freedom of choice should remain completely intact. Traditionally, economic measures are judged on whether they increase welfare, which is a well-defined criterion in the absence of anomalies. Classically, economists are of the opinion that the decisions of agents show exactly what these agents want. Thus, from this perspective, policies can only improve well-being in situations where the decisions of one party cause damage to another party (externalities).

However, assessing the welfare implications of libertarian paternalism-style interventions is much more complicated. Thaler and Sunstein3 state the normative ideal case of a nudge as follows: A “pure” nudge enables people to make a decision as if they were free from decision anomalies. In our opinion this is a good theoretical benchmark, but we also find inherent difficulties in implementing this ideal case. The welfare effects of a decision influenced by nudging should also be classically assessed according to whether they reduce externalities such as environmental pollution. In addition, however, they are to be assessed according to whether the results of “satisfactory” decisions ultimately correspond more to the actual preferences than if the anomalous decision maker had decided without the nudge. This is a difficult undertaking, as the anomalies distort the correspondence between goals and preferences and so everything must first be diagnosed: decision anomalies with their negative effects on consumer wellbeing and the resulting externalities. In addition to this identification problem, however, normative questions for decision makers also arise. Which consumer preferences should be seen as the actual preferences? Which nudges should influence these preferences in which direction?

Preset decisions

First, we examine the nudges that target decision anomalies that arise in relation to the cost of decision making. As an example, consider the problem of deciding for or against the status of an organ donor. We call the costs involved in making a decision c. These costs can vary from person to person. Let's assume that in the absence of an active decision, the Non-Donor option is automatically selected for each person. In this case, those who would prefer donor status for themselves, but not so much that they would be willing to c accept, automatically do not get donor status. However, those who prefer not to be a donor are not required to wear a c. Individuals who strongly prefer to be donors - to a degree greater than c - will continue to choose donor status. If the pre-set decision (default) is now set to the status of organ donor, then the well-being of those who prefer the status of organ donor only slightly improves, i.e. to a degree less than c. This welfare gain will not exceed the costs of making the decision to donate organs independently, as otherwise you would have made this decision and thus have shown strong preferences for organ donation. On the other hand, changing the default decision can potentially worsen the well-being of those who have slight preferences for not being a donor. If their number and decision-making costs are comparable to those who slightly prefer the status of donor for themselves, changing the default decision to the “organ donor” status can destroy any welfare gain. Here it becomes clear that nudges that change the preset decision only increase well-being if the decision costs are high and preferences are approximately homogeneous. As part of this nudge, decision-makers should therefore pay attention to this criterion. A well-known example is the automatic participation in a retirement savings plan in the USA.4 In the case of pension decisions for old age, there seem to be fairly homogeneous anomalies in the decision-making process. In connection with the decision in favor of financial products, anomalies due to decision costs also become clear: Who has the time and leisure to understand and compare all of the fine print in detail?

Active choice

The same logic follows the analysis of other nudges that could address decision costs. A measure is considered that provides for an active choice for everyone, so that no preset decision has to be made. This measure ensures that everyone who does not want to be an organ donor does not become one, and that everyone who does not want to be an organ donor becomes a donor. This measure also appears to be attractive from a political and strategic point of view, as the state does not have to enforce its view of the preset decision. It should be noted, however, that only groups with slight preferences will end up with a preset decision that is unfavorable for them (see above). These groups didn't want to wear c and so didn't make a decision. A measure by which everyone has to make a choice is clearly welfare-reducing if one assumes that people correctly express their preferences for donor or non-donor status.

Measures that are intended to reduce the costs (in) decision-making are, for example, the clear and understandable presentation of relevant product properties on the product or the removal of bureaucratic hurdles. As a result, consumers with weak preferences for a preset option will be more likely to express their preferences through active choice and their wellbeing will improve. At the same time, nobody who prefers the preset option has to bear higher costs. This measure clearly increases well-being, and decision-makers do not need to know the true distribution of preferences here either.

On the other hand, distorted ideas about the consequences of a decision are a relatively simple problem for decision-makers who are considering the use of nudges. For example, suppose households are convinced that conventional incandescent lamps are less expensive than efficient LED lamps over their entire lifespan . This opinion can arise if the respective costs of energy consumption over the service life are not sufficiently taken into account and too much emphasis is placed on the initial purchase price. Assuming that wellbeing depends on actual, not expected, costs, clearing and understanding the life cycle cost of both options can increase consumer wellbeing. This would better align the consumer's decision with the costs ultimately experienced. Those consumers who actually prefer conventional incandescent lamps - perhaps because of the color of the light - will not be affected by this information as long as the preference for the light color compensates for the true cost of the lamp.

The environment and the presentation of a decision problem can have had a profound influence on the decision made. Since people perceive and evaluate profits and losses differently, nudges can be used to present decisions either in a profit or loss context (framing). In the incandescent lamp example, the different energy costs of classic incandescent lamps and LED lamps can either be shown as: a. not switching to LED bulbs will cost you 50 euros every year or b. By switching to LED lamps, you save 50 euros every year. Since people are more sensitive to losses, the first representation should be chosen. Analyzing the welfare effects of such a changed representation is not trivial, since preferences depend on the decision-making environment. Decision-makers have to determine whether they will respect the pre-nudge preferences of consumers for whom they were satisfied with the previous lamps. However, if only post-nudge preferences are viewed as relevant, then this nudge increases welfare as the consumer will not return to the inefficient lamps.

An important empirical problem in the context of nudging is initially to identify distorting anomalies that affect large parts of the population and influence consumption decisions. Only then can nudging to influence these anomalies be justified in the first place. In empirical studies, for example, Hausmann and Dubin and McFadden base their identification strategies for anomalies on comparing the relative market shares of different goods.5 A current summary can be found at Allcott.6

Conventions of a society

Using a nudge to influence the decision-making environment can be linked to another approach: the tendency to obey societal norms. An example of this are US studies of the content of electricity bills.7 The bills that were sent to households as part of the experiment provided information about how high their own electricity consumption is compared to that of their neighbors. This information was also provided with emoticons that showed a happy face if the respective household consumed less energy than the neighborhood average. This nudge proved to be effective in persuading households to move closer to the average with their consumption. It also had the effect of reducing overall energy consumption.

Assuming people have consistent preferences to follow societal norms, then the nudge of informing about the average energy consumption of the environment leads them to shift their own consumption in the direction of actual preferences. This nudge can also change people's attitudes towards their energy consumption. If you previously only had your personal needs and costs in mind when consuming energy, this could change in such a way that you now think that your consumption has normative significance. Without comparison with the environment, the household benefits from energy consumption; with the comparison, the household could benefit from “doing the right thing”. An analysis of the well-being under these two possibilities is not easy, since these two kinds of benefit have a different origin. Such nudges can have uncertain welfare effects. However, if this avoids serious negative externalities such as environmental pollution, this may need to be weighted more heavily.

Guide to the conception and use of nudges

The guide consists of the following questions to be clarified:

  1. What is the problem to be solved? What is the decision anomaly that makes it possible for a nudge to improve welfare?
  2. What are the real wants and needs of people who are to be genudged? How homogeneous or inhomogeneous are these between the individuals?
  3. Does the nudge lead the desires and decisions of individuals to converge? Do those affected want to be genudged?

The answers to these questions are by no means trivial, but they do provide a good guide for future nudgers.

  • 1 Cf. M. Allais: Le comportement de l’homme rationnel devant le risque: critique des postulats et axiomes de l’école Americaine, in: Econometrica, 21st vol.(1953), H. 4, pp. 503-546; H. A. Simon: Models of man; social and rational, New York 1957, chap. 14, p. 287 ff .; D. Kahnemann, A. Tversky: Prospect Theory: An analysis of decision under risk, in: Econometrica, 47th Jg. (1979), H. 2, pp. 263-291.
  • 2 Ibid.
  • 3 See R. H. Thaler, C. R. Sunstein: Nudge: Improving decisions about health, wealth, and happiness, New Haven, CT, 2008.
  • 4 Cf. B. C. Madrian, D. F. Shea: The Power of Suggestion: Inertia in 401 (k) Participation and Savings Behavior, in: Quarterly Journal of Economics, 116th vol. (2001), no. 4, pp. 1149-1187.
  • 5 Cf. J. Hausmann: Individual Discount Rates and the Purchase and Utilization of Energy-Using Durables, in: Bell Journal of Economics, Vol. 10 (1979), H. 1, pp. 33-54; See J. Dubin, D. McFadden: An Econometric Analysis of Residential Electric Appliance Holdings and Consumption, in: Econometrica, 52nd year (1984), H. 2, pp. 345-362.
  • 6 See H. Allcott: Paternalism and Energy Efficiency: An Overview, NBER Working Papers, No. 20363, National Bureau of Economic Research, Cambridge MA 2014.
  • 7 Cf. DL Costa, ME Kahn: Energy Conservation "Nudges" And Environmentalist Ideology: Evidence From A Randomized Residential Electricity Field Experiment, in: Journal of the European Economic Association, 11th year (2013), no. 3, p. 680-702; H. Allcott: Social Norms and Energy Conservation, in: Journal of Public Economics, 95th vol. (2011), H. 9-10, pp. 1082-1095.

Incomplete rationality is not a sufficient justification for paternalistic intervention

Jan Schnellenbach