Introduction

This chapter will discuss some of the potential disadvantages of utilizing voluntary compliance. For example, when relying on voluntary compliance, which is based on limited monitoring and enforcement by the state, an increase in inequality between “good” and “bad” people may rise, as many people will continue to obey while others might not. This could lead to a situation where there is a growing gap between the formal law and social norms. In addition, while in previous chapters the notions of voluntary compliance and trust-enhancing regulatory approaches were associated with substantial changes in trust in society, this chapter focuses on their potential harmful effects to society and the rule of law. For example, if governments need to persuade people that laws should be obeyed, then this practice may be counterproductive as it expects states to be more intrusive. Of course, in many studies on factors such as legitimacy and fairness, where states behave in a trustworthy way, changing people’s attitudes toward the law by explaining its importance is not always necessary.Footnote ¹

In addition, this chapter will examine to what extent the reduction in the usage of sanctions by states will be replaced by nonformal sanctions which could, in some contexts, be more problematic than formal sanctions (e.g., reliance on reputation might be more harmful to some people than to others).Footnote ² While most of the discussion on crowding out assumes some interplay between extrinsic sanctions and intrinsic motivation, we will argue that in many contexts this could be the opposite, where formal sanctions will be replaced by informal ones.Footnote ³ Those sanctions, while being somewhat more effective and less costly to enforce, could have far more problematic results in terms of procedure, evidentiary rules, and proportionality.Footnote ⁴ Many of the discussions about the power of communities and social norms are such that there is a very limited opportunity to understand the control mechanisms at play. For example, there is a great variation in the strictness of the community in which one lives. Community-based governance can lead to significant variations in how closely people’s behavior is monitored.Footnote ⁵

Heterogeneity: The Challenge to Voluntary Compliance

Given the expected heterogeneity in population response, the main challenge for policymakers will be how this gap affects the level of trust both among cooperators and among noncooperators as well as regarding the interaction between them. Furthermore, as discussed in the next paragraph, research on cooperation shows that many of the cooperators are, in fact, conditional cooperators,Footnote ⁶ and hence at a certain point, many may shift their behavioral patterns based on the cooperation of others. To that end, estimating what proportion of the population are conditional cooperators is difficult. In one of the most influential papers on this topic, Fischbacher and colleagues demonstrate that 50 percent of participants in public good experiments are conditional cooperators – individuals who increase their contributions when they observe higher average contributions from other group members.Footnote ⁷ While many studies in this field identify a substantial proportion of conditional cooperators in their populations, they often refrain from specifying exact percentages. This approach is understandable, as the precise proportion of conditional cooperators can vary significantly based on multiple factors, including experimental design, cultural context, and participant demographics.Footnote ⁸

Conditional cooperators make up a significant portion of participants in public goods experiments, with estimates ranging from 25 percent to 62 percent. These individuals exhibit sophisticated behavior, adjusting their contributions based on others’ actions to maximize group earnings while avoiding inefficient outcomes.Footnote ⁹ Their decisions are influenced by both financial incentives and social norms, demonstrating a nuanced understanding of social dilemmas. Conditional cooperators demonstrate increased contributions when informed about the presence of other cooperative individuals in their group. This behavioral pattern indicates that transparency regarding group composition could effectively promote cooperation. However, the strategic interactions between conditional cooperators can produce complex outcomes – potentially enabling free-riding behavior or transforming collective action situations into coordination problems. The research also highlights the importance of distinguishing genuine conditional cooperation from confusion-driven behavior or a general tendency to follow others.Footnote ¹⁰ These findings underscore the need for careful consideration of the heterogeneity in cooperative behaviors when designing policies or interventions aimed at promoting collective action in public goods scenarios.

To deal with the negative consequences of an approach focusing on intrinsic motivation or even just coercion-free voluntary compliance, states will, in certain contexts, still need to employ monitoring mechanisms to determine if they can continue with the “hands-off” approach. As such, states must maintain oversight to prevent exploitation of trust-based regulation – essentially adopting a “trust but verify” approach. However, this oversight may diminish some of the efficiency benefits that pure trust-based systems aim to achieve. Furthermore, voluntary compliance could lead to situations where there could be what has been termed a “chilling effect,” where some people might be unsure of what is expected of them and, ultimately, comply to a greater extent than is required.Footnote ¹¹ For example, people might cover their faces with masks even in contexts where the risk is minimal. Alternatively, others may enjoy the more lenient approach and would comply less, leading to a situation where “good” law-abiding people will suffer more, while “bad” people will suffer less.

Such a shift may suggest that the gap between good doers and bad doers will only arise when regulatory approaches, primarily relevant to the bad doers, are reduced. Within a simple schematic perspective of “good” and “bad” doers, the undermining of extrinsic measures and the increase of intrinsic measures, as part of the voluntary compliance approach, creates an unbalanced approach toward the population. Part of the regulatory dilemma is that current research on the proportion of good and bad doers is quite limited. Even in well-studied, easy-to-quantify contexts such as tax evasion, where states could reasonably be expected to know how many people evade taxes, it is very hard for policymakers to come up with concrete numbers as to how many tax evaders there are.Footnote ¹² The net benefit of transitioning to trust-based regulation depends on two key factors: the proportion of compliant versus noncompliant actors in society for the regulated activity and the relative social and behavioral benefits of voluntary compliance compared to the costs of reduced enforcement efforts associated with voluntary compliance. A responsible regulation with a sequential approach that uses increasingly severe enforcement actions is clearly not a perfect solution, as the information on who will cooperate in each situation is missing and can even be misleading.

An additional point of concern regarding voluntary compliance relates to the definition of compliance as based on intrinsic motivation. If the state needs to cause people to intrinsically believe in the law, this may require the state to engage in a far more intrusive process of educating people and leading them to see the value in obeying. Thus, the state will need to not only cause people to change their behavior but also their belief system, a far more problematic process from many points of view.

There is, of course, a difference between the state providing information on how tax revenue is used versus attempting to explain the reasoning behind a certain law. The latter could raise concerns about a slippery slope, where states feel compelled to shape public beliefs to align with government policies. However, if the state focuses solely on regulating behavior rather than attempting to influence attitudes, the fear of brainwashing is diminished. By not viewing itself as responsible for aligning public attitudes with state values, the government can avoid overreach.

Indeed, if the state aligned its laws to fit with the values of society, rather than the other way around,Footnote ¹³ this is clearly something that doesn’t raise the criticism suggested. However, the focus of the argument regarding the fear of intrusive intervention is the reverse, where states would have chosen to change public preferences. For example, during the COVID pandemic, many individuals’ preferences were to move openly, engage in social interactions, and fly abroad. States needed to demonstrate to their citizens how noncompliance would undermine public health efforts to contain the pandemic. While often overlooked in behavioral public policy literature, this process of building public understanding is central to enhanced approaches to nudge theory such as the “nudge plus.”Footnote ¹⁴ The need to focus on deliberation and reflection, rather than merely changing behavior, becomes evident when we consider the implications of people internalizing values differently.

When governments attempt to project trustworthiness, particularly when unwarranted, it raises significant ethical concerns. Such behavior may mask ulterior motives behind a facade of public trust. When a regime relies solely on public goodwill for cooperation, it might resort to manipulation – alternating between propaganda and appeasement to secure public support.

Voluntarily Cooperating: Variance in People’s Responses

As stated in the introduction, this chapter focuses on how people react differently to law in the absence of sanctioning systems.

Even with current enforcement approaches, there is significant variation in how people respond to legal requirements. Some individuals are more adept at evading enforcement and more likely to do so based on factors like their attitudes toward risk and their respect for the rule of law.Footnote ¹⁵ Therefore, concerns about distributive justice are not unique to voluntary compliance or reduced coercive measures.

Various theories of social preference compare people’s preference to cooperate, free ride, and respond to cooperation (conditional cooperators).Footnote ¹⁶ Given the prevalence of conditional cooperators discussed, governments that wish to treat the public using trust-based approaches face a significant challenge. However, the argument put forward here is that without extrinsic motivation such as sanctions, there is more room for intrinsic motivation to influence cooperation. As suggested with regard to the existence of different types of cooperators – conditional ones, those who tend to free ride, and those who always comply – intrinsic motivation is expected to have far greater variation between people based on their personalities, culture, and contextual factors.Footnote ¹⁷ In contrast, price as affecting compliance, being external to the person, is more likely to create equalization between people,Footnote ¹⁸ even if their economic status obviously differs.

This heterogeneity could create a few problems:

• First, the proportion of wrongdoers may be too high to be tolerated by the state, leading to increased costs to the majority of the public.

• Second, many cooperators are what’s called conditional cooperators,Footnote ¹⁹ where their cooperation depends on that of others.Footnote ²⁰ With an increase in the number of wrongdoers who might exploit the trust given to them by the state, conditional cooperators might follow and stop cooperating as well.Footnote ²¹

• Third, we might encounter a situation where people misinterpret the true content of social norms, either due to false signals or overestimation of wrongdoing. As a result, they might stop cooperating, not because others are actually uncooperative, but because they believe others would likely not cooperate under less scrutiny from authorities.Footnote ²²

Heterogeneity in Honesty

Beyond differences in cooperation levels, there’s also significant variation in honesty among people. This heterogeneity likely affects how much governments can trust the public without more specific data on the distribution of cooperative and honest individuals in each population.

In my previous book,Footnote ²³ I examine differing accounts in the literature. Some studies suggest a substantial level of dishonesty albeit with variation across the different experimental paradigms,Footnote ²⁴ while Abeler and colleagues seem to believe that honesty is less prevalent.Footnote ²⁵ The relationship between honesty and cooperation is an important consideration, as discussed in Chapters 1 and 2.Footnote ²⁶

In that context of individual difference, Jacquemet and colleagues found that individuals who engage in occasional or moderate dishonesty – rather than systematic lying – are influenced by a truth-telling oath as a nonprice commitment device.Footnote ²⁷ Through experiments examining income reporting and tax declarations, researchers used response time analysis and sequential decision patterns to demonstrate that moderate dishonesty reflects a weak preference for profitable honesty. The oath effectively transforms these occasional deceivers into committed truth-tellers. However, in my work with Eyal Pe’er (discussed further in Chapter 4), we discovered a broader effect: pledges reduced dishonesty more effectively than fines, impacting even habitual liars. This effect proved consistent across multiple time periods, varying degrees of dishonesty, and different individual characteristics.Footnote ²⁸

What Is the Prevalence of Dishonesty in Society?

Regarding the prevalence of dishonesty in society, research has examined statistics on unethical behavior in the US.Footnote ²⁹ For example, in a 2004 poll done by the popular magazine Readers’ Digest,Footnote ³⁰ 93 percent reported engaging in one or more kinds of dishonesty at work or school, such as calling in sick when not feeling ill (63 percent), taking office supplies from work (63 percent), and lying on their résumés (18 percent).

In contrast to the perspective which views unethicality as prevalent and contagious among normative people, Abeler and colleagues combined data from ninety experimental studies in economics, psychology, and sociology, and showed that, in fact, people lie surprisingly little.Footnote ³¹ The study’s empirical evidence suggests that a preference for being seen as honest is the main motivation for truth-telling.

Similarly, Ostrom presents a comprehensive overview of research, and some of the key insights include the fact that not all individuals are purely rational egoists and that social norms can evolve to support cooperation.Footnote ³² For example, in a prisoner’s dilemma experiment, 40 percent of the subjects ranked the cooperative outcome higher than the outcome of defecting while the other cooperates, and 27 percent were indifferent between these outcomes, even though their individual payoff was substantially higher in the latter. According to Ostrom, this finding confirms that not all players enter a collective action situation as pure forward-looking rational egoists who make decisions based solely on individual outcomes. Some bring with them a set of norms and values that can support cooperation. In addition, Ostrom shows that conditional cooperators are apparently also a substantial proportion of the population, with rates ranging from 40 to 60 percent.

In another important question regarding honesty from a more economist perspective, Gibson and colleagues sought to solve the question of whether the world is populated by exactly two fixed types: economic types and ethical types, and whether there is heterogeneity in the preferences of the people to lie or to tell the truth, depending on consequences.Footnote ³³ The type-based model and the model with heterogeneous preferences for truthfulness lead to very different implications.Footnote ³⁴ Therefore, it is important to determine which of these two models offers a more accurate description of reality.

To address this question, they conducted a decision-theoretic laboratory experiment in which each participant was placed in the situation of a CEO who had to announce their firm’s earnings to a passive market. The participants were informed of the true level of earnings. Participants were informed that misreporting higher earnings was illegal but would result in larger payouts compared to truthfully reporting their lower earnings. It was expected that economic types would always lie in our experiment because truthfulness was designed to be economically costly; also, it was expected that ethical types would always tell the truth.

They observed that, in a situation where the standard economic model predicts that everybody will lie, 32 percent of the participants chose not to do so, thus forgoing a larger variable compensation. Importantly, the aggregate percentage of truth-tellers decreased as the costs of truthfulness increased. That implies that the marginal effect of a cost increase on the probability of an individual telling the truth is significantly negative, even after controlling for various demographic and psychological factors. These results are at odds with the type-based model but are consistent with a model that posits heterogeneous preferences for truthfulness. The models’ primary contribution, therefore, is to provide evidence for the notion that people occupy a spectrum of preferences for truthfulness rather than only two opposite positions.

Why Do So Many People Cheat?

Competing literature varies between different meta-analysis studies, as well as in more field-oriented studies that show that in certain contexts a large proportion of the population lies.Footnote ³⁵ Jacobsen and colleagues present six main theories explaining why people cheat:Footnote ³⁶

1. 1. The Economic Theory: This represents a rational approach to cheating, where individuals weigh the costs and benefits of dishonest behavior.

2. 2. The Moral Balance Theory: This introduces a psychological perspective, suggesting that people engage in a form of ethical accounting, allowing themselves to cheat if they feel they have accumulated moral credit.

3. 3. The Self-Maintenance Theory: This focuses on preserving one’s self-image, indicating that individuals will cheat only to the extent that it doesn’t force them to revise their self-perception as honest.

4. 4. The Self-Justification Theory: This extends the idea of self-maintenance, proposing that people can rationalize their cheating behavior, even after the fact, to maintain cognitive consistency.

5. 5. The Moral Disconnection Theory: This highlights the human capacity for moral disengagement, where individuals exempt themselves from the ethical standards they expect others to follow.

6. 6. The Limited Morality and Moral Blindness Theory: This addresses the role of awareness and self-reflection in ethical behavior, suggesting that some individuals may cheat due to a lack of moral awareness or failure to critically examine their own actions.

Together, these theories illustrate that cheating is not simply a matter of economic calculation, but a complex interplay of psychological, moral, and cognitive factors that can vary significantly across individuals and situations.

Jacobson and colleagues concluded that people regularly engage in dishonest behavior, but that this behavior is highly malleable because it is sensitive to elements such as decision contexts, behavior of others, state of mind, and ego depletion.Footnote ³⁷ In other words, changes in regulatory culture or industry norms can significantly impact behavior. For instance, an industry where cutting corners becomes the norm might see a rapid increase in such practices.

While the concept of conditional cooperators has been extensively studied, the idea of conditional cheaters has received comparatively less attention.Footnote ³⁸ Despite this imbalance in research focus, there are notable similarities between these two bodies of literature. These similarities allow us to speculate that environmental factors likely influence the prevalence of dishonesty in ways similar to how they affect cooperation.

The key contrast lies in the depth of research: conditional cooperation is well documented, while conditional cheating, though studied,Footnote ³⁹ lacks the same level of comprehensive investigation. However, the parallels between these phenomena suggest that social and environmental cues probably shape dishonest behavior just as they do cooperative behavior.

To sum up, research on heterogeneity of honesty and cooperation is diverse and many models of human cooperation have accounted for the fact that different people come with different motivations. However, individual differences are not the only concern; equally important is the potential distributive effect when trust is extended to those who shouldn’t be trusted. As we discuss in Chapter 7, this creates a paradox: If privacy and autonomy considerations force us to trust everyone uniformly, we might ultimately end up trusting no one.

In line with the perception against dichotomy and for a spectrum, an important study claims that there are more diverse profiles of liars than has been accepted so far in the literature.Footnote ⁴⁰

In an experimental study examining dishonest behavior, participants were instructed to roll a die, with financial rewards proportional to the reported outcome. The findings revealed distinct patterns of honest and dishonest behavior. While one group maintained complete honesty, others exhibited various forms of dishonest behavior: Some participants rolled multiple times without misreporting (reporting the first, less profitable result); others misreported their single roll; and a third group both rolled multiple times and misreported their results.

The researchers identified a behavioral spectrum across these categories and explored participants’ motivations. Of particular interest was the distinction between those who repeatedly rolled until achieving their desired outcome versus those who directly misreported results without rolling. This behavior suggests different psychological mechanisms regarding self-image maintenance when dealing with actual versus hypothetical dishonesty.

The study demonstrated consistency in dishonest behavior profiles across different populations while highlighting that honesty is not a unidimensional trait. Individuals often simultaneously violate rules while maintaining a self-perception of honesty. This heterogeneity in honest behavior, previously considered problematic for voluntary compliance, may inform the development of optimal policy solutions, particularly in the context of Chapter 11’s compliance mechanisms.

Heterogeneity in Compliance

The previous sections have discussed heterogeneity in prosocial behavior and dishonesty. Now, we’ll examine these factors in the context of compliance. Compliance is not merely a combination of prosocial behavior and honesty; it also involves people’s perceptions of their relationship with authority and their duty to obey the law. In Chapter 2 we discussed the work of van Rooij and colleagues in the context of compliance motivation, which is also important for the heterogeneity.Footnote ⁴¹ In that context, we have examined variations in people’s sense of obligation to obey the law (OOL). The OOL plays a vital role in people’s responses to the legal system and their compliance decision-making. So far, most studies have linked the OOL to people’s experiences and views of the legal system.Footnote ⁴² This chapter seeks to expand our understanding of the OOL by analyzing how such variation relates to personal and moral differences, differences in socialization in higher education, and differences in political preferences. Throughout a series of 4 studies with a total of 5,085 participants, Fine and colleagues found that the OOL is not solely dictated by experiences with and perceptions of the legal system but is comparatively more related to personal and political differences.Footnote ⁴³ Indeed, as discussed in more detail in this paper, an important source of variation between people in the US is related to political orientation, where conservative moral values, such as loyalty, authority, and purity,Footnote ⁴⁴ may lead to a preference for obedience,Footnote ⁴⁵ stricter punishment, and more trust in law enforcement.Footnote ⁴⁶ Research consistently links political affiliation with perceptions of the law and law enforcement,Footnote ⁴⁷ with Republicans being more “tough on crime” compared to Democrats. One could only predict that such variation will have a huge impact on heterogeneity in compliance which might end up increasing the gap between the different political groups. Such an increase in the gap might be exacerbated when accounting for the rise of polarization in many countries around the world.Footnote ⁴⁸

Heterogeneity in Moral Reasoning

The propensity to morally disengage refers to an individual’s tendency to use cognitive mechanisms that allow them to distance themselves from their own unethical actions. It involves a set of cognitive processes that enable individuals to justify or rationalize their unethical behavior, thereby reducing feelings of guilt or responsibility.Footnote ⁵³

Moral disengagement is a common phenomenon that can be found in all segments of society. Drawing on Bandura’s moral disengagement theory,Footnote ⁵⁴ research has examined this phenomenon across schools, businesses, and prisons. For example, a study found that 85 percent of college students used moral disengagement strategies to justify cheating on exams.Footnote ⁵⁵

In the seminal work by Moore and colleagues,Footnote ⁵⁶ they examined the influence of individuals’ propensity to morally disengage on a broad range of unethical organizational behaviors. They concluded that considering an individual’s propensity for moral disengagement can help in understanding and predicting a wide range of undesirable behaviors. This variation in moral disengagement acts as another predictor of behavior. Consequently, without enforcement, we are likely to see a much greater gap between people’s actions.

Heterogeneity in Social Value Orientation

As discussed in Chapter 3, social value orientation plays a crucial role in predicting cooperative behavior and compliance. Within the Big Five personality model, social value orientation represents the most recent addition that is particularly relevant to understanding voluntary compliance behaviors.Footnote ⁵⁷

Another important personality dimension related to cooperation is honesty-humility, which appears as the first dimension in the HEXACO (honesty-humility [H], emotionality [E], extraversion [X], agreeableness [A], conscientiousness [C], and openness to experience [O]) personality model.Footnote ⁵⁸ This trait is defined as the tendency to be fair, genuine, and sincere in dealing with others.

For example, Hilbig and Zettler investigated the predictive ability of the honesty-humility personality trait in relation to economic and cooperative behavior.Footnote ⁵⁹ They hypothesized that this trait would explain how individuals allocate valuable resources between themselves and others in two different scenarios: the dictator game and the ultimatum game.Footnote ⁶⁰ Furthermore, they predicted that honesty-humility would account for the differences in behavior between these two games, where the recipient’s power to retaliate varies.

Their study involved 134 participants and yielded results that supported both hypotheses. Participants who scored low in honesty-humility made more self-centered decisions and only shifted toward a more equitable distribution when the other party had the power to punish uncooperative behavior. In contrast, those who scored high in honesty-humility consistently chose a fairer allocation, even when they could have acted selfishly without fear of repercussions.

Lastly, the study found that social value orientations partially mediated the observed effects, suggesting that an individual’s preference for certain social outcomes played a role in the relationship between honesty-humility and economic decision-making.

An additional study, by Lainidi and colleagues, found that people who were high in honesty-humility were more likely to cooperate with others in a prisoner’s dilemma game, even though they knew that they could potentially earn a larger reward by defecting. This suggests that people who are high in honesty-humility are more likely to cooperate with others, even when it is not in their best interests to do so.Footnote ⁶¹

In a meta-analysis of studies on the relationship between honesty-humility and prosocial behavior, Fang and colleagues found that honesty-humility is positively correlated with prosocial behavior such as helping others in need, cooperating with others, and donating to charity.Footnote ⁶²

Heterogeneity of Social Norms

In other studies, conducted for example with Barak-Corren and Gidron,Footnote ⁶³ we show how pluralistic and nonhomogeneous are the effect of norms on different parts of society. From the perspective of voluntary compliance, this suggests an important caveat to consider, that is, how much we can rely on social norms to replace formal laws.

Additionally, many argue that religious mandates have contributed to a decline in public support for religion and have triggered public resistance to perceived attempts at top-down control.Footnote ⁶⁴

This argument suggests that for practices like circumcision, Leil HaSeder observance, and refraining from driving on Yom Kippur, there is no need for legal enforcement. The social norms surrounding these practices are strong enough to guide behavior without legal intervention. However, in areas where there is likely to be a prohibition, such as the law against eating bread during Passover, noncompliance is prevalent.

The Yom Kippur example in Israel illustrates the complexities of relying solely on social norms for regulation. While there’s a strong norm among secular Jews not to drive on Yom Kippur, transforming roads into spaces for cycling,Footnote ⁶⁵ this practice has led to unforeseen consequences. With 20 percent of Israel’s population being Arab and not observing this norm, there have been tragic accidents in recent years.

This situation highlights a key challenge: societies often comprise diverse groups with different beliefs and practices. The effectiveness of norms in changing behavior is limited when society isn’t homogeneous. In the Yom Kippur case, the absence of cars led children to cycle on roads without considering potential dangers from the minority still driving.Footnote ⁶⁶

This example relates to the taxonomy discussed in chapter 11 of The Law of Good People, where we compared legal contexts requiring cooperation from everyone (like trade secrets), a few (whistleblowing), or the more the better (e.g., tax compliance). The Yom Kippur scenario falls into the category needing universal cooperation, making it difficult to rely solely on intrinsic motivation without formal compliance measures.

Furthermore, when there’s no consensus in the population and a law is passed to enforce a behavior that only some are motivated by social values to follow, much of its effectiveness may be lost due to social reactance theory.Footnote ⁶⁷

Regulating Heterogeneous Populations Effectively: Insights from Research on Honesty Pledges

Along with research on the factors that predict the likelihood of cooperation among different individuals, there is hope in the work I have done with Eyal Pe’er on the effect of pledges on brazen liars, whom one might predict would be less affected by trust-based regulation such as pledges. We have analyzed whether pledges work primarily for honest people or also for brazen liars. We hypothesized that pledges, which raise awareness of honesty’s morality, might only be effective for those with a strong commitment to ethics. Previous findings also suggested that pledges might only affect minor cheaters.Footnote ⁶⁸ However, we found that pledges significantly influenced behavior across individual differences in rule-following tendencies and cheating extents. This implies that trust-based regulation could be equally effective for people with low commitments to the law. The overall reduction in brazen liars was similar, with no interaction found between individual differences in commitment to obey the law. The percentage of “high-extent cheaters” (defined as 75 percent or higher) was largest in the self-report group, considerably smaller with a pledge or fine, and smallest with both. These differences remained significant when the “high-extent cheater” threshold was defined between 60 percent and 90 percent, suggesting that pledges and fines affect even those who exhibit high degrees of cheating. This contradicts the perspective that trust-based regulation might increase the gap between ethical and nonethical people.

Social Punishment as a Replacement for State Punishment

A possible solution to the problem we have raised about voluntary compliance without punishment and monitoring by the state in a heterogeneous society, where not all members are equally committed to the cooperate with the requests of the state, is social punishment, which could be substituted for state enforcement.

Social punishment is widely discussed in the field of behavioral game theory where the idea of altruistic punishment is being developed and it was argued that it could lead to stable cooperation.Footnote ⁶⁹ Some other studies have focused on the usage of shaming as an alternative to state enforcement in various domains.Footnote ⁷⁰ However, the question is whether such altruistic punishment can really be used to ensure voluntary compliance given the complexity of many forms of compliance, especially with regards to corporations,Footnote ⁷¹ as will be developed in more detail in Chapter 7. This approach was subject to lots of criticism for not always being accurate and proportional.Footnote ⁷²

Undesirable Outcomes of Heterogeneity in Voluntary Compliance

In voluntary compliance regimes, nonintrinsically motivated actors may increase their noncompliant behavior. This can trigger a cascade effect where conditional cooperators, observing free-riding or misconduct, reduce their own compliance. Such dynamics can escalate perceived distributive injustice as compliance gaps between different groups widen.

The impact of others’ behavior on compliance varies significantly across regulatory domains. Tax compliance demonstrates a particularly strong “sucker effect,” where individuals are highly sensitive to others’ free-riding behavior and may withdraw their own compliance in response. Environmental regulations, however, often elicit more independent compliance decisions, less influenced by others’ behavior.

In domains requiring collective cooperation, such as protecting trade secrets within professional communities, uneven compliance can undermine the entire system’s effectiveness. For instance, when some business association members breach confidentiality while others maintain it, the value of information protection diminishes for all participants.

Governments face a fundamental dilemma in addressing these varying cooperation levels. While effective policy implementation requires detailed information about individual backgrounds and behavioral patterns (as Chapter 8 will explore), collecting such comprehensive personal data raises significant privacy concerns.

The Fear of Brainwashing

In the research about nudges there is always the fear that there will be harm to autonomy, but in voluntary compliance the fear might be greater for various reasons related to overreliance on intrinsic motivation. When many people need to be intrinsically committed so are more likely to engage in social enforcement, cooperation, and compliance, the government might need to resort to various techniques to cause as many people as possible to be committed to the purposes of the state. When the government focuses on sanctioning and monitoring such a shift might be less needed.

Overrelying on Moral Compliance

Focus should be given to some of the models reviewed in this chapter which are less likely to interfere with other models and carry unintended effects, as was demonstrated regarding sanctions or incentives. For instance, the concept of procedural justice, widely studied by scholars such as Tyler and others,Footnote ⁷⁵ is likely to increase legitimacy and compliance with less likelihood of interfering with effective functioning of deterrence. Similarly, informing people of the harm associated with their behavior may be relevant for some people without inducing resentment toward the law. Nevertheless, even with these approaches, some scholars propose the possibility that emphasizing morality might give people the impression that the state is unable to enforce the law; that is, these approaches may potentially backfire.Footnote ⁷⁶ Thus, regulators should strive to identify the policy which will target as many motivations as possible while acknowledging the challenge of achieving complete success in this mission.

The Problem of Biased Morality

An additional challenge with intrinsic motivation and moral conviction arises from allowing individuals to use moral reasoning to guide their behavior. This approach assumes a universal agreement on moral standards, which is problematic. While there may be clear consensus on straightforward issues like the obligation to pay taxes, many situations present more complex dilemmas. In these cases, we must consider how people will ultimately apply their moral judgments and whether they will do so appropriately.

As already suggested in my previous book, The Law of Good People, since people’s interest is to not view themselves in a negative light, they are likely to engage in various biased cognitive process related to reasoning, memory, and vision. In that book, I outline various biases which could explain the distortion of people’s evaluation of what is moral.Footnote ⁷⁷ Thompson and Loewenstein have shown that people were more likely to remember information that was related to their own position, with the result that their opinion of what was fair aligned with their interest in the outcome of a settlement.Footnote ⁷⁸ The fact that these biases operate without awareness makes it difficult for people to notice the process. Moore and Loewenstein were among the first to show that self-interest and concern for others affect behavior through different cognitive systems and that self-interest, unlike concern for others, is automatic, viscerally compelling, and often unconscious.Footnote ⁷⁹ By comparing people’s private beliefs and public behavior, Moore demonstrated that people truly believed their own biased judgments, and did not recognize any problems in their responses.Footnote ⁸⁰

As this review reveals, there is a growing recognition that many ethical decisions are the result of implicit, not explicit, choices.Footnote ⁸¹ Given that people’s unethical behavior is frequently accompanied by a limited and distorted view of their own conduct, it is particularly important to focus on legal violations by otherwise good employers in the context of employment law. Indeed, this study will show that many of those mechanisms, which are particularly likely to increase the likelihood of good people behaving with limited awareness of the full legal and ethical meaning of their actions, are especially problematic in the context of employment. In the employer–employee relationship, issues such as ambiguity, repeated smaller violations, and the strong effect of workplace norms, which may not necessarily conform to those of the employees’, are likely to have significant impacts on employees’ conduct.Footnote ⁸²

Regulatory Mistakes and Public Safety

An essential aspect to consider when discussing voluntary compliance, as explored in Chapter 4, is the complex interaction between regulation, trust, and public safety. A significant concern in regulatory research is the potential harm to the public when regulators place excessive trust in businesses. This issue arises from the possibility that regulated entities might exploit the increased flexibility granted by regulators to act against public interests.Footnote ⁸³

This concern aligns with our earlier discussion on heterogeneity in compliance motivations. When the government extends trust to regulated entities that don’t genuinely merit public confidence, the consequences can be severe:

1. 1. It may jeopardize public safety, which regulators are primarily tasked with protecting.

2. 2. It could erode public trust in both the regulatory system and the regulated industries.

3. 3. It might discourage public engagement in economic activities due to perceived risks.

The cost of regulatory mistakes in such scenarios extends beyond immediate safety concerns. It can lead to long-term damage to the relationship between the public, businesses, and regulatory bodies, potentially undermining the entire regulatory framework’s effectiveness.

The Limits of Fairness and Altruism

Another possible criticism on the reliance on voluntary compliance is related to studies which show that it is unreliable to focus on fairness in costly decisions.Footnote ⁸⁴ How likely are people to base their behavior on intrinsic motivation when what’s being requested from them is very costly and requires a sacrifice? Is it true that most regulators currently believe voluntary compliance motivations are not truly reliable? Indeed, research on honesty usually shows that the amount of money is not crucial, but is it clear enough to show that costs don’t matter?Footnote ⁸⁵ Other lines of research, for example those developed in economics and behavioral economics on fairness as a constraint on profit seeking, do tend to understand the tension between profit and fairness.Footnote ⁸⁶ Furthermore, there is an obvious problem with the fact that experimental lab studies are limited in measuring high prices.Footnote ⁸⁷ Another important debate about fairness concerns the self-interested interpretation of what constitutes “fair,” as explored in The Law of Good People. According to this view, fairness is inherently limited because it is subjective and therefore may not effectively constrain opportunistic behavior.Footnote ⁸⁸ However, Tyler’s research on procedural justice suggests that perceptions of fairness can transcend self-interest – both those who benefit from a decision and those who do not (e.g., winners and losers in court) tend to view procedurally fair decisions as more legitimate.Footnote ⁸⁹

Managing Heterogeneity Risks

Finally, it is important, as will be developed in more detail in Chapter 11, that there are various mitigations to many of the concerns discussed thus far. Without the focus on people’s background, it will be very hard for governments to know whom they can trust. In work with Orly Lobel and Ori Aronson,Footnote ⁹⁰ discussed in more detail in Chapter 7, we found one approach which could help governments deal with this issue. We identified personality traits of people we can trust more. In that work we explore various ways to examine what kind of people are more likely to be cooperative and receptive to voluntary compliance initiatives. However, it is not clear whether, given privacy concerns, it is realistic for governments to treat people based on personality traits.

An alternative approach to determining trustworthiness was developed in collaboration with Yotam Kaplan.Footnote ⁹¹ We argued that it is more effective to adopt a differentiated approach that compares situations rather than one that evaluates people based on their past behavior. In Chapter 7, we explore technological approaches to voluntary compliance and discuss the potential for creating a more nuanced approach to voluntary cooperation. This approach recognizes that different situations pose varying levels of risk for voluntary compliance, potentially offering a mitigation strategy for the risks discussed earlier.

In higher risk situations, where the disparity between voluntary and nonvoluntary compliance might be more significant, we might reduce our emphasis on voluntary compliance and allocate more resources to alternative enforcement strategies. Thus, understanding the contexts in which people are less likely to voluntarily cooperate with the government could provide at least a partial solution to addressing the gaps discussed at the beginning of this chapter.

Conclusion

The heterogeneity in people’s responses to voluntary compliance presents significant challenges to regulatory approaches that rely on intrinsic motivation and trust. This chapter has highlighted various dimensions of this heterogeneity, including differences in honesty, moral reasoning, social value orientation, and compliance attitudes. These variations among individuals can lead to undesirable outcomes, such as increased inequality between “good” and “bad” actors, erosion of trust among conditional cooperators, and potential exploitation of trust-based systems by those less inclined to comply voluntarily.

While voluntary compliance approaches offer many benefits, including reduced enforcement costs and increased social capital, they also carry risks. These include the potential for brainwashing, overreliance on moral conviction, and the limits of fairness and altruism in costly decisions. The chapter has also explored how social punishment might replace state punishment in some contexts, but this too comes with its own set of challenges. Despite these concerns, the potential benefits of voluntary compliance suggest that rather than abandoning such approaches, regulators should seek ways to mitigate risks and tailor strategies to account for population heterogeneity.

To address the challenges posed by heterogeneity in voluntary compliance, future research should focus on several key areas. First, more comprehensive studies are needed to accurately measure the distribution of different types of cooperators (e.g., conditional cooperators, free riders, and altruists) across various regulatory contexts. Second, research should explore how to effectively combine voluntary compliance approaches with targeted enforcement strategies, perhaps using AI and big data to identify high-risk individuals or situations while respecting privacy concerns. Third, investigations into the long-term effects of voluntary compliance strategies on social norms and individual behavior are crucial to understanding their sustainability. Finally, cross-cultural studies examining how heterogeneity in compliance attitudes varies across different societies could provide valuable insights for designing more universally effective regulatory approaches. This research agenda would help policymakers develop more nuanced, adaptive regulatory strategies that can harness the benefits of voluntary compliance while mitigating its risks in heterogeneous populations.