Can Democratic States Trust the “The Behavioral Code”

Praise of the Book

First, full disclosure is called for before discussing this book. I am a big fan of the work of Benjamin van Rooij and Adam Fine, and partnered with them in establishing the ComplianceNet for disseminating compliance research.¹The Behaviorial Code: The Hidden Way through which the Law Makes Us Better or Worse (Hereinafter: The Book) written by Benjamin and Adam is distinguished by their intellectual prowess and boundless curiosity, which spills forth from its pages. Despite being an unlikely pair—one an anthropologist who studied China (Van Rooij) and the other an experimental developmental psychologist (Fine)—their collaboration results in a work that combines a comprehensive review with a thought-provoking analysis. The net cast by their collective expertise captures both quantitative and qualitative literature, resulting in a wide-ranging and nuanced exploration of the impact of law on behavior offering a much broader perspective than any other available today.

It is always wise to start a book review with some praise of the book, and this proves quite easy in this case. This impressive book offers a fresh perspective on a wide range of fields and disciplines that are essential for understanding compliance. The authors’ truly remarkable ability to seamlessly integrate various disciplines, from economics to psychology, sets this book apart from others. One of its unique features is the compelling storytelling accompanying the exceptional and memorable examples given. This descriptive prowess in no way undermines the care they take in presenting the scientific basis for the theories demonstrated in each example and in each field.

Indeed, while compliance is a challenging field that is often overlooked or misunderstood, this book takes the subject of compliance together the effect of law on behavior to the next level. With a fresh perspective and a holistic approach, the authors demonstrate the importance of compliance in both legal and ethical terms. They show how understanding the antecedent of compliance is critical for individuals, organizations, and society as a whole. Furthermore, The Book, in contrast to other recent works, is not satisfied with simply praising one concept, such as nudges or incentives. Instead, the authors’ holistic approach takes into account the myriad of factors that influence compliance. Rather than advocating for one overarching theory, the authors present a range of concepts, mostly developed by others, and examine how they are interconnected across different contexts.

Some reflections on the book

In reflecting on the main theme of the book, the challenge arises of how to take account of both the individual and the institutional context in compliance research. For example, the measure of efficacy, which is related to the belief that engaging in certain behavior is likely to improve reality,² is highly dependent upon the institutional contexts in which people operate. As a result, some of the book’s separation between the behavioral and the institutional contexts may appear artificial, given that any need that people may have to feel that what they do is relevant and beneficial is usually dependent on the presence of functioning institutions.

Moreover, when addressing this challenge, it is crucial to focus not only on what works but also on what does not work. Indeed, the problem of null effects also arises in compliance field studies, highlighting the extent of our knowledge gaps. We have to account not just for what we know, but also on what we don’t know. To fill these gaps, a holistic approach is needed to engender a meaningful discussion of trade-offs, which is missing in most current studies of the area.

The Behavioral Code draws from, among other things, the behavioral public policy literature, which shows that compliance cannot be achieved solely through price-based incentives. This approach has expanded the range of regulatory options available to policymakers, including nudges, framing, and pledges, all aimed at influencing behavior without coercion. While this increased array of tools complicates choosing an instrument, it also facilitates concrete, quantitative evaluations and comparisons of their effectiveness.³

For example, the nudge approach, based on the influential work of Thaler and Sunstein,⁴ represents an important addition to the regulatory choice dilemma, and has led numerous scholars to examine when nudges can be relied upon to replace coercive mandates.⁵ While nudges are perceived as a tool that maintains people’s freedom of choice, the fact that they are employed with only limited public awareness of their application could potentially lead to nudges being viewed as a more sophisticated form of coercion.⁶ The salience of the behavioral public policy literature, in the context of the behavioral code is that it unravels many of the dichotomies usually found in the regulation literature, whereby government either attempts to focus on external factors, such as prices, or on internal factors, such as morality.

Another challenge faced by those seeking to change public behavior using a behaviorally based regulatory approach is understanding not just how regulatory tools may have unforeseen or inadvertent effects, but also when and how the different tools may undermine the effect of other legal tools, thereby possibly creating a crowding out effect.⁷ The crowding-out effect, which is widely discussed in The Behavioral Code, suggests how complicated it is to predict when certain regulatory tools will work together positively or will impair each other’s effectiveness.

Challenges to The Behavioral Code approach

There are two levels of challenges posed by The Behavioral Code:

First, descriptively, there is the challenge of understanding how to predict the effects of different regulatory interventions. This challenge makes it difficult to determine their effectiveness, especially when trying to identify what is not working and why. Clearly, then, we need to understand how different regulatory tools interact with each other; only then can we begin to ascertain which regulatory tools work better.

The second challenge lies on the normative level. Even if the effect of a regulatory intervention on a given behavior is known, it is not clear how to determine the right balance of different approaches. Thus, it is crucial to try to understand how different regulatory approaches can be combined in order to achieve optimal outcomes. Later in this article, I discuss the trade-off approach as a way of addressing the uncertainty associated with the availability of so many regulatory approaches and the fact that many of them have the potential to cause damage in a given context.⁸

Can we predict theoretically what will work and when? Is this a problem?

An important theme of The Behavioral Code relates to the diversity of regulatory interventions which became associated with the behavioral public policy⁹. The challenge, discussed in Should States Trust the Public,¹⁰ is researchers and policy makers’ inability to predict the effectiveness of any particular regulatory instrument. One notable confirmation of the need for a trial-and-error approach is Milkman’s work comparing a number of interventions targeting various behaviors. Recently, Milkman conducted a mega-study on vaccinations involving 689,693 Walmart pharmacy customers, which revealed that text-based reminders could effectively encourage vaccination.¹¹ It also identified the most effective types of messages. Specifically, Milkman tested 22 different text reminders that used various behavioral science principles to nudge people to receive flu vaccinations, and found that reminder texts boosted vaccination rates by 6.8 per cent compared to a business-as-usual control condition. The best-performing messages were those that reminded patients that a flu shot was waiting for them and that were sent on multiple days.

Interestingly, Milkman reports that none of the experts (which are basically the people from the field who were invited to become co-authors of Milkman) were able to predict which intervention would be the most effective. In fact, experts were worse than laypeople in predicting the outcomes of the different interventions, suggesting how limited is the theory behind the regulatory tool box. This result inevitably leads to a number of questions. Is regulatory design in the behavioral age futile if experts are unable to predict what will work? Is it then necessary—and feasible—to constantly collect data? Clearly, any proposed program for a behavioral code must be prudent and based on empirical data. However, can data always be collected at the pace needed to make optimal choices? Is there any way to know in advance whether or not a regulation will work? Are there proxies that could help us in predicting which interventions are more likely to work? What are some of the limitations of attempting to predict what type of laws would succeed in making the intended behavioral changes? How do we deal with 1novel situations, when there is no past experience that can guide the design of new tools.

Should States Trust the Public? The promise and perils of voluntary compliance

At this point, it is useful to examine how the main argument from the trusting the public book, can offer a strategy for continuing from where the important work of The Behavioral Code ends. In short, in this book, I examine and compare competing relevant literatures on compliance, behavioral public policy, regulatory theory, trust, behavioral ethics, and cooperation concepts and their antecedents. A fundamental premise is that it is in the best interest of a democratic state to convince its citizens that the state should be obeyed voluntarily; hence, any legal instrument that aims to maximize voluntary behavior will be preferred. From a normative perspective, we might need to evaluate the various costs of encouraging the public to cooperate voluntarily.

The analysis in my new book seeks to set a higher standard than in the past, and examines not just what regulatory tool would work, but also the broader impacts of the law on behavior, resilience, and spillovers onto other domains. I propose integrating a new paradigm into the behavioral code approach whereby the answers to a series of questions helps us ascertain which regulatory tools to employ. Not just what works but also, to what end? What is it that we want to achieve? What variation are appropriate to address doctrines and cultures of different groups in society? Do we need the compliance to work in the long run? Is the quality of compliance more or less important than the quantity of compliance? Is it reasonable to expect that people will have the intrinsic motivation with regards to the behavior we are trying to change that might increase the likelihood of crowding out effect happening? What are the short-term and long-term negative effects of extrinsic interventions?

The emphasis on heterogeneity and the distributive consequences of law, for instance, indicates that groups in the population with lower levels of education and social capital may be more likely to change their views in response to an influencing campaign.¹² The long-term effects of such interventions on individuals’ feelings of trust in their government are associated with another moral effect arising out of the distributive effect of interventions, an issue discussed below in relation to crowding out effects. Policymakers are faced with the challenge of striking the correct balance between fostering individuals’ intrinsic motivation to obey the law and preventing increased legal evasion by lack of monitoring, particularly in light of the increasing evidence that in the long run enhancing intrinsic motivation is likely to lead to improved compliance.¹³

Both The Behavioral Code and Should States Trust the Public seek to meet the ongoing need to provide policy makers and regulators with a broad array of regulatory tools as well as a better understanding of how intrinsic and extrinsic measures interact. To better understand the impact of any proposed extrinsic measure on intrinsic motivation, it is important to determine the net overall gain between the regulatory effect and the direction and quality of the behavioral change. Such analysis could rely on existing research on the interplay between extrinsic and intrinsic motivation.¹⁴

One area that The Book focuses less attention on is how to make people not only engage in desirable behavior without coercion, as discussed in their seat belt compliance success example, but also on how to make people internalize and become intrinsically motivated to obey these specific laws,¹⁵ whether referring to the specific seat belt laws or possibly to laws in general. Furthermore, according to The Book, intrinsic motivation is the sense of morality inherent within the individual, whereas extrinsic motivation relies on exogenous incentives and rewards.¹⁶ For example, it could very well be that people vary in their internal level of commitment to ethical behavior,¹⁷ but there is debate as to what extent this variation is being taken into account in existing research.

Another question that needs to be considered is to what extent the discussion of intrinsic motivation in the compliance literature is comparable to that being conducted in the field of psychology. In psychology, the definition of intrinsic motivation is related to self-determination,¹⁸ and behaviors that increase peoples’ autonomy, and competence, benefiting them from engagement in certain behaviors, when the behavior in itself is somewhat rewarding for the individual.¹⁹ Extrinsic motivation is defined as behavior engaged in for external to the behavior such as incentives or reputation rather than internal reasons. Traditionally, the focus of motivation research in psychology has been on what is the internal source of motivation, and how it relates to the self, rather than on an entity like the law, which is, by definition, external to the person. Indeed, in the literature of regulation of law and economics, including work by scholars such as Robert Cooter on intrinsic motivation,²⁰ there is a broader definition of intrinsic motivation, which considers the possibility of inducing desired behavior without any need for external monitoring.

The difference between these two views of the interaction between intrinsic and extrinsic motivation, which reflect a continuation of the issues raised in The Behavioral Code, might be related to the question of how to treat factors such as fairness, legitimacy, and morality, which are associated with intrinsic motivation in the legal literature. However, according to purely psychological definitions, such as that of scholars like Ryan and Deci,²¹ these factors may not be considered part of intrinsic motivation, as they are not related to the activity itself but to questions such as whether the social institutions seeking change are viewed as just and are accepted by others. It is also important to note that when discussing compliance, the sought-after behavior is often driven not by individual choice but is imposed upon individuals, while in other cases, the behavior could result from motivations internalized by the individual.

How crowding out complicates the regulatory tool box

The crowding out of intrinsic motivation is a good demonstration of the descriptive challenges of a holistic approach such as that of The Book.

In general, the crowding out literature suggests that when individuals base their behavior on external rewards, they devalue any moral consideration behind their actions, thereby reducing any perceptible impact of intrinsic motivation. This theory, as applied to regulatory incentives, predicts that using monetary rewards or punishments as external incentives may have a negative impact on intrinsic motivations to comply.²² For instance, paying people in return for donating their blood might lead donors to view their contribution as a transaction rather than as a charitable act, thereby reducing the quality of blood donations (e.g., when one is sick and his blood might do no good).²³

Among the notable studies on the crowding out effect on incentives and enforcement,²⁴ Deci found that tangible rewards undermine intrinsic motivation for a range of activities.²⁵ In light of the importance of the sum of money being paid on the magnitude and direction of the crowding out effect, based on the size of incentives it is reasonable to assume that given the varying levels of intrinsic motivations among individuals, different sums of money will have correspondingly different effects on specific subgroups. In a previous work with Lobel on how to incentivize whistleblowers,²⁶ we demonstrated that those who were intrinsically motivated were not significantly affected by the framing of monetary incentives, whereas those who were low on intrinsic motivation were affected by it. Similarly, in my study with Perez, we found a direct relationship between the perception of incentives and intrinsic motivation levels: those who were low on intrinsic motivation were more likely to prefer environmental deposits to fines, whereas the opposite was true for those who had high levels of intrinsic motivation.²⁷

These results indicate that a better approach to the difficulty of predicting how a regulation will affect behavior is creating a detailed analysis of the trade-offs that will make policy makers aware of the pros and cons of each regulatory option. This method might be a valid compromise even in cases where collecting data prior to each regulatory intervention might be prohibitive. Making policy makers aware to the broad behavioral impact of each regulatory choice, is likely to improve the likelihood of success.

When making regulatory trade-offs, it’s crucial to consider a number of factors, including whether the decision maker should take account of long-term consequences. Another factor to consider is the quality of the compliance behavior (e.g., using face masks during the COVID-19 pandemic in a way which covers ones nose) and the potential impact it may have. Also important is trying to determine whether people will have intrinsic motivation to comply with the regulation and whether it’s necessary to create this motivation. Additionally, the costs of implementing extrinsic interventions must be weighed against the potential benefits. It is also worth considering how much information is necessary and available to inform the decision. Finally, the level of certainty on the efficacy of the interventions chosen must be assessed to ensure it is the most effective and sustainable option. Regulatory trade-offs can help the regulatory toolbox become more effective by taking these factors into account.

Taxonomies

Based on the behavioral trade-offs and enforcement dilemmas just reviewed, this section offers some preliminaries taxonomies that will facilitate the use of psychological and cognitive factors in the design, implementation, and enforcement of legal policy. Use of this taxonomy will make it easier to match legal policy to what we know about the behaviors, cognitions, and motivations of individuals to which it is targeted.²⁸

A taxonomy for regulatory initiatives should include the following relevant factors: type of regulated behavior, desired durability of the behavioral change, proportion of the target population whose cooperation is required for successful implementation of the mechanism, cost of enforcement, and cost of noncompliance. Let us examine each factor in turn.

Nature of the regulated behavior

The first question to be addressed is what type of behavior are we looking to change. Is it the kind of behavior for which we need to rely on intrinsic motivation for people to perform well or is sustainability of key importance, given the high cost of enforcement? Do we want to increase the quantitative aspects of the regulated behavior—the extent to which one engages in it—such as in the case of recycling? Or is the qualitative nature of the behavior more important, as in whistle-blowing or even blood donations (where we need people to be truthful about their health histories), where the level of commitment to the action is critical? In the latter types of behaviors, we might give more weight to the decision-making process rather than to optimize the outcome. Furthermore, in legal contexts where “extra-role” activity—that which is beyond the level required by law—is called for, the likelihood of reducing intrinsic motivation increases. Therefore, regulators should be more cautious in introducing measures or nudges that rely on extrinsic motives and that may crowd out internal motivation.

Durability of the behavioral change

Whether the sought-after behavior requires a one-time decision or choices repeated over time will help determine the appropriate type of regulatory mechanism. In contexts such as enrolling in or choosing a pension plan, people are less likely to reverse their decision once they have made it. With sustainability less important in areas like health or nutrition, where choices need to be reaffirmed on a daily basis, it is therefore more important to persuade people to make the right choice (outcome being the dominant focus).

What proportion of the target population needs to cooperate

In designing legal policy, one of the most important factors to consider is whether the population is likely to have high levels of intrinsic motivation to comply. This is naturally more challenging when little is known about the target population or when their level of intrinsic motivation varies. When a large proportion of the population has high levels of intrinsic motivation, policy makers can rely on noncoercive explicit measures or on softer types of implicit interventions, such as nudges. Van Rooij and Fine try to determine which policies may be effective. In contrast, in my work I try to examine which regulatory intervention is more likely to lead the desired behavior by higher proportion of honest citizens, cooperators, or compliers. The proportion question helps us ascertain the proposed regulations’ effectiveness. From a normative perspective, it is important to determine what proportion of cooperators are needed with each doctrine and what regulatory tools are needed to obtain that level of cooperation. For example, in the context of wearing face masks or getting a vaccine, obtaining a majority might be enough. However, regarding quarantines, a few non-cooperators might cause the pandemic to spread. This conditional cooperation could be analyzed differently from the perspective of public health that focuses only on how much compliance we need, or from a normative perspective—what will happen to society.

Implementation in Four Contexts

Finally, such analysis could be demonstrated with regard to four legal contexts which vary across the domains created thus far.

Trade secrets

In the area of trade secrets, everyone needs to be motivated to keep information confidential and not disclose the secret. If only a few people are subject to legal consequences for disclosure, keeping company knowledge proprietary may be a futile endeavor.²⁹ Regulations should be designed to foster cooperation among the entire target population, including individuals with varying levels of intrinsic motivation. Therefore, trade secret approaches should prioritize the needs of those with the lowest level of intrinsic motivation, specifically those who may be less inclined to stay loyal to the company. As individuals with lower motivation are more prone to making “self-serving mistakes,” which can be costly, it becomes crucial to emphasize achieving at least minimal compliance. Consequently, it may be necessary to prioritize ensuring loyalty even among those without intrinsic motivation, even if it means potentially undermining the intrinsic motivation of dedicated employees.

Whistleblowing

The context of whistleblowing is the exact opposite, where we need the cooperation of only some of the employees to go forward when some illegal activity occurs within the organization. Therefore, policymakers can focus on those with high-intrinsic motivation.³⁰ It might not be advisable to provide incentives to individuals lacking intrinsic motivation due to concerns about potential false reports from bounty hunters who are solely driven by maximizing rewards.

Recycling

In the context of recycling, the primary focus is on achieving long-term aggregate outcomes where a large number of people engage in extensive recycling. In this scenario, there is no preference for individuals with high or low intrinsic motivation. The policymaker must strike a balance and decide whether to promote extrinsic motivation and determine the appropriate types of incentives to encourage recycling. Additionally, while the ultimate objective may be to encourage environmentally responsible behavior in as many people as possible, the costs associated with the non-compliance of a single individual are relatively low. In other words, the failure of a particular regulation to increase recycling rates across the entire population may not be necessary since the goal is to raise the average level of recycling among the population.

Public Health

Regarding the issue of face mask wearing during the COVID-19 pandemic, quantitatively, the goal was to have everyone wear masks, and qualitatively, the goal involved how well people covered their faces with their masks. The considerations are different with the COVID vaccinations, where we just need more people to get it and the quality of shot is not dependent on the individual. In addition, not everyone needs to get the vaccine. For such context intrinsic motivation is not really necessary to improve adherence and it is possible to rely on extrinsic measures, such as incentives or mandates. On the other hand, a 100 per cent vaccination rate is not necessary with regard to most diseases which might suggest that the measures taken need not be too intrusive.

Conclusion

The Behavioral Code represents a significant advancement in our understanding of the intricate relationship between law and behavior. This comprehensive and analytical book draws upon various forms of literature to provide a comprehensive account of compliance dilemmas. However, this complexity presents additional challenges when formulating concrete legal policies. This brief note outlines several factors that can aid in mitigating this complexity. It reviews the factors that regulators should consider when selecting the appropriate regulatory tool. When the costs of errors are high and cooperation is crucial, relying on fewer trust-based regulations may become necessary. Conversely, when the public displays high intrinsic motivation and support for a specific law, regulators may find it easier to depend on trust-based regulations. Hopefully, Van Rooij and Fine’s outstanding book will inspire further empirical and normative research on compliance, facilitating the identification of the optimal regulatory tools for specific behavioral outcomes.

Footnotes