Improving Compliance with Labor Laws: The Role of Courts

ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 2/1/2023 12:00 PM IMPROVING COMPLIANCE WITH LABOR LAWS: THE ROLE OF COURTS Guy Davidov† and Edo Eshet†† I. INTRODUCTION Enforcement of labor laws is inherently challenging.1 The same reasons that led to the enactment of labor laws—the inequality of bargaining power and the vulnerability of employees vis-à-vis their employers—also make the workers, in many cases, unlikely to complain or sue during the relationship. Additional barriers put significant limitations on the ability of workers to sue after the end of the relationship as well. The state can use inspectors and prosecute offenders, but this can affect only a small fraction of violations. For various reasons, the problem has exacerbated in recent years.2 As a result, new ideas have been proposed by scholars and some new tools introduced by legislatures, in an attempt to better prevent payments of wages below the minimum wage, as well as many other violations of laws that are crucial to the well-being of employees.3 † Guy Davidov is the Elias Lieberman Professor of Labor Law at the Hebrew University of Jerusalem. †† Edo Eshet is a Senior Lecturer at Sapir Academic College, Israel. We are thankful to Tammy Katsabian and participants at the conference on “Compliance With and Enforcement of Labor Laws” (Sep. 2020), organized by the Hebrew University, for excellent comments. 1. We use the term “labor law” in this article in the broad (non-American) sense, i.e. including employment law. 2. GUY DAVIDOV, A PURPOSIVE APPROACH TO LABOUR LAW Ch. 9 (2016). 3. For a recent extensive review of the theoretical literature see Tess Hardy & Sayomi Ariyawansa, Literature Review on the Governance of Work (ILO, 2019). For an overview and analysis of solutions see Guy Davidov, Compliance with and Enforcement of Labour Laws: An Overview and Some Timely Challenges, 3 SOZIALES RECHT 111 (2021). Notable contributions include David Weil, Public Enforcement/Private Monitoring: Evaluating A New Approach to Regulating the Minimum Wage, 58 ILR REV. 238 (2005); David Weil, A Strategic Approach to Labour Inspection, 147 INT’L LAB. REV. 349 (2008); Michael J. Piore & Andrew Schrank, Toward Managed Flexibility: The Revivial of Labour Inspetion in the Latin World, 147 INT’L LAB. REV. 1 (2008); Janice Fine & Jennifer Gordon, Strengthening Labor Standards Enforcement Through Partnerships with Workers’ Organizations, 38 POL. & SOC’Y 552 (2010); Tess Hardy & John Howe, Too Soft or Too Severe? Enforceable Undertakings and the Regulatory Dilemma Facing the Fair Work Ombudsman, 41 FED. L. REV. 1 (2013); DAVID WEIL, THE FISSURED WORKPLACE: WHY WORK BECAME SO BAD FOR SO MANY AND WHAT CAN BE DONE TO IMPROVE IT Ch. 9 (2014); Tess Hardy & John Howe, Chain Reaction: A Strategic Approach to Addressing Employment Noncompliance in Complex Supply Chains, 57 J. OF INDUS. REL. 563 (2015); MICHAEL PIORE & ANDREW SCHRANK, ROOT-CAUSE REGULATION: PROTECTING WORK AND WORKERS IN THE 21ST CENTURY (2018); Janice Fine & Tim Bartley, Raising the Floor: New Directions in Public and Private Enforcement of 71 ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 72 COMP. LABOR LAW & POL’Y JOURNAL 2/1/2023 12:00 PM [Vol. 43:71 It is possible to divide the efforts into two: methods to improve compliance and methods designed to improve enforcement. Although there are obviously strong connections between the two—the prospect of effective enforcement can be expected to push more employers to comply in the first place—there are also important differences. Efforts to improve the level of compliance focus on prevention by inducing a change of behavior among employers. The attempt is to structure the law in a way that will lead to more compliance, without investing resources in enforcement. For example, a law that requires the employer to keep records about working hours can include a provision shifting the burden of proof to the employer in case of violation, making it easier for employees to sue for overtime payments. Such an incentive structure is one possible way of increasing compliance. In contrast, instruments at the level of enforcement focus on which sanctions to use in case of violations, how to allocate inspection resources, monitoring mechanisms, etc. Obviously, this is also intended to change the behavior of employers, through deterrence, so the line is admittedly blurred.4 Without claiming that there is a clear line separating the two levels, our aim in the current contribution is to focus on the level of compliance, and specifically, on the role of courts. The literature in this area is mostly concerned with legislation and with enforcement agencies. A highly interesting and useful branch of the literature focuses on public-private initiatives, which add a role for private actors to support enforcement efforts. But the role of the judiciary has been relatively neglected. What can courts do (and to some extent are already doing) to push employers towards complying with labor laws? One aspect of the court’s role is deciding who should be considered an “employee” and an “employer.” These topics received enormous attention in the literature, and they are certainly important for tackling evasion; but they are not directly about compliance or enforcement. In any case, we do not discuss them in the current article. The question we wish to address is what judges can do to increase compliance in those cases in which the status of employer and employee are not in dispute. Our discussion assumes that judges have significant discretion to develop the law, which is especially true in common law systems. We draw our examples from Israel, which has a “mixed” legal system with common law origins. We believe that much of the discussion is relevant for civil law systems as well, but claim no expertise in this regard. Labor Standards in the United States, 61 J. OF INDUS. REL. 252 (2019); LEAH F. VOSKO ET AL., CLOSING THE GAP: IMPROVING EMPLOYMENT STANDARDS PROTECTIONS FOR PEOPLE IN PRECARIOUS JOBS (2020). 4. The term compliance is sometimes associated with preference for soft-law and self-regulation, or an assumption that violations are not intentional (see Leah F. Vosko et al., The Compliance Model of Employment Standards Enforcement: An Evidence-Based Assessment of Its Efficacy in Instances of Wage Theft, 48 INDUS. REL. J. 256 (2017)). We use the term here in a broader meaning, without any intention to suggest preference for self-regulation or an assumption that most employers are “good.” ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 2022] 2/1/2023 12:00 PM IMPROVING COMPLIANCE WITH LABOR LAWS 73 There are several conditions that have to be met to ensure compliance (or: several stages that we have to consider when working to improve compliance). First, an employer has to be aware of the law and understand what type of actions are required or prohibited by it. Second, the employer has to acknowledge the violation—to admit to himself that a certain action or choice violates the law. For that to happen, there must be no room (or at least as minimal room as possible) for self-serving rationalizations by the employer about its actions. Finally, the employer should be fearful of violating the law, either because economic analysis suggests that the violation is not beneficial for him and will result in a loss (given the expected sanction and the probability of its infliction), or because of other, noneconomic sanctions attached to the violation. It is possible to view these conditions as relevant for different types of people. In a recent book about compliance, Yuval Feldman divides wrongdoers into three groups: “‘erroneous wrongdoers,’ those whose wrongdoing can be attributed mainly to limited awareness due to errors and blind spots but they do not actively look for justification for their wrong doing; “situational wrongdoers,” those whose unethicality is primarily justified by their rationalizations for doing bad in a given situation; and ‘“bad people,’ or ‘calculative wrong doers, who deliberately engage in unethical behavior.”5 Feldman rightly adds that “the dividing line between these three groups is sometimes blurry,” but it is still useful to treat them differently in terms of the methods employed to ensure compliance. For the first group, the goal is to make sure that they know about the law; for the second group, the goal is to confront the rationalizations; and for the third group, of people who respond only to cost-benefit calculations, we need to use deterrence and other incentives. The challenge for the law is to build mechanisms to maximize compliance by all three groups. There are three bodies of literature that can help illuminate the potential of the law to induce compliance, which correspond roughly to the three conditions (or stages) mentioned above.6 First, the “expressive function of the law” is based on the view that proclamations by legislatures and by courts can have an impact—and contribute to compliance—quite aside from the fact that the law threatens with sanctions or that it represents an order from a legitimate authority. Second, behavioral ethics explains the psychological mechanisms that lead “good people” to break the law—especially by failing to acknowledge (through self-deception) violations that are serving their interests. Understanding these mechanisms can help us develop cures and improve compliance. Finally, economic analysis of law explains how 5. YUVAL FELDMAN, THE LAW OF GOOD PEOPLE: CHALLENGING STATES’ ABILITY TO REGULATE HUMAN BEHAVIOR 61 (2018). 6. This general framework relies heavily on Feldman, id., albeit with some variations. ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 74 COMP. LABOR LAW & POL’Y JOURNAL 2/1/2023 12:00 PM [Vol. 43:71 incentives can deter people from violations, and behavioral law and economics adds nuance that relaxes the assumption of rationality and better explains how incentives towards compliance can work as a matter of practice.7 The article starts with a brief review of these three bodies of literature, explaining the theories and related empirical findings concerning compliance with legal requirements, and their relevance for labor law (section 2). We then move to consider some specific contexts within labor law in which judges, in particular, can influence compliance. Relying on examples from Israeli case-law, we show that courts already take such considerations into account, and we argue that they should do so explicitly and more consistently. We discuss judgments that can be understood as an attempt to raise awareness to the law (section 3); the need to balance between ambiguity and specificity when judicially developing the law (section 4); the possibility of using a third party to pressure others into compliance (section 5); and ways to induce compliance through remedies (section 6).8 We should clarify that we do not argue that courts are more important for ensuring compliance than the other branches. Our claim is that they play a role in this regard, and they should play a role, alongside the other branches. In section 7 we briefly reply to a possible critique that judges lack the legitimacy to take compliance considerations into account in their judgments. II. HOW THE LAW CAN BE SHAPED TO INDUCE COMPLIANCE Labor laws are important, for the workers and for society at large, so obviously we want to make sure that they are obeyed. The three branches of government can all play a role in efforts to secure maximum compliance. While the current article focuses on the role of courts, we start with a more general examination of possible ways to induce compliance. The current section introduces three bodies of literature that seem relevant to this goal. We start, in sub-section 2.1, with the most basic and intuitive idea of deterrence through the threat of punishment, adding some behavioral insights that explain why it does not always work. We then move to consider two additional reasons that people have for complying with laws. Many people are “good” and will obey the law even without the credible threat of sanctions, simply because it is the moral thing to do. The field of behavioral ethics, briefly introduced in sub-section 2.2, explores why even good people 7. There are additional bodies of research related to compliance—for example, the view that procedural justice (perceiving the legal process as fair) is the key reason for people to obey the law (TOM R. TYLER, WHY PEOPLE OBEY THE LAW (2006))—but these seem less relevant for our current purposes. 8. We focus in this article on compliance by employers only. Improving compliance by employees is relevant for small parts of labor law, such as protection of trade secrets and non-compete agreements. It requires a different analysis which is out of the scope of this article. ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 2022] 2/1/2023 12:00 PM IMPROVING COMPLIANCE WITH LABOR LAWS 75 sometimes break the law, and what can be done to prevent this. Finally, people may also obey laws because of laws’ expressive function: the information they provide and their ability to secure coordination. This idea is briefly summarized is sub-section 2.3. A. From Economic Analysis of Law to Behavioral Law and Economics Standard economic analysis is based on the assumption that people act rationally, and strive to maximize their utility. The utility is expressed by the fulfilment of preferences, which in the case of employers can be assumed to be maximizing profits. This leads directly to the theory of deterrence. An employer presumably calculates the expected profit from violating an employment law, against the expected costs, which are calculated taking into account the chance of being caught, the possible punishment and other costs (such as harm to public image)9, and how swiftly the punishment will be applied. Deterrence is achieved when the expected costs outweigh the gains from violating the law.10 Such an analysis assumes that all actors have full information, make rational decisions, and only care about utility which can be calculated. In practice, however, deterrence does not always work, for several reasons. First, many people are not aware of the law and/or the expected punishment, and therefore will not be deterred. This is especially true for small employers without a human resource department and regular legal advice. In any case, studies suggest that the perceived likelihood of detection and the perceived punishment are more important than the actual ones. 11 Second, people sometimes fail to acknowledge that they have violated the law, again making deterrence ineffective. 12 Third, the utility function is not simple and unidimensional, but rather extremely complex.13 For example, to what extent do employers care about being seen in public as violating the law? Perhaps the reputational costs can be calculated if they have a direct impact on profits, but the indirect social cost might be different from one employer to another, depending on changing social and personal factors. We cannot assume that 9. Kevin Purse & Jillian Dorrian, Deterrence and Enforcement of Occupational Health and Safety Law, 27 INT. J. OF COMP. LAB. L. & INDUS. REL. 23, 36 (2011). 10. This is a very simplified description. For more nuance see Orley Ashenfelter & Robert S. Smith, Compliance with the Minimum Wage Law, 87 J. OF POL. ECONOMY 333 (1979); Yang-Ming Chang & Isaac Ehrlich, On the Economics of Compliance with the Minimum Wage Law, 93 J. OF POL. ECONOMY 84 (1985). And see Gary S. Becker, Crime and Punishment: An Economic Approach, 78 J. OF POL. ECONOMY 189 (1968) on “optimal” (efficient) deterrence. 11. For a recent review of the evidence on deterrence, regarding this point and others, see Tess Hardy, Compliance Defiance: Reviewing the Role of Deterrence in Employment Standards Enforcement, 37 INT. J. OF COMP. LAB. L. & INDUS. REL. 133 (2021). 12. See Feldman, supra note 5, at 69, 153. 13. See Walter Firey, Limits to Economy in Crime and Punishment, 50 SOC. SCI. Q. 72 (1969); Purse & Dorrian, supra note 9, at 38. ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 76 COMP. LABOR LAW & POL’Y JOURNAL 2/1/2023 12:00 PM [Vol. 43:71 every employer only cares about profits. Finally, and perhaps most importantly, the assumption of rationality is often unrealistic. It is here that behavioral law and economics offers important “qualifiers and modifiers” to the standard economic analysis.14 Behavioral studies reveal that people have biases which influence their perception of risk, and as a result lead to miscalculations. 15 For example, people are over-optimists, leading to “systematically perceive the probability of detection to be lower that it truly is.”16 Also, people tend to overestimate the probability of occurrences that are more available to them. For example, as Tversky and Kahneman explained, a person who was witness to a burning house would overestimate the probability of his own house catching fire, although nothing has really changed in the objective risk calculation. This is called the “availability bias.”17 Because it works in both directions, visible occurrences of noncompliance can contribute to employers believing that they are not likely to get caught, thus creating a norm of noncompliance.18 Moreover, it has been shown that repeat offenders in situations of lowprobability detection tend to rely on their experience (of not getting caught so far) and underestimate the probability of getting caught in the future.19 As a result, deterrence does not work well for repeat offenders, at least if the rates of detecting a violation are relatively low—which is the case in the labor law context. These limitations do not mean that deterrence never works. People do make calculations—even if, most often, not explicitly—and in many cases learning about a possible punishment or other costs will make them reconsider their actions. 20 However the impact of deterrence depends on raising awareness to the law (and the expected punishment); on ensuring that the punishment and indirect costs are significant enough for different types of employers; on raising the chances of detection (and perceptions about it); and on limiting as much as possible cognitive biases that lead to miscalculations. Some of these measures have to be targeted to specific types of employers to be more effective.21 14. See EYAL ZAMIR & DORON TEICHMAN, BEHAVIORAL LAW AND ECONOMICS 8 (2018). 15. Id., at 448. 16. Id., at 446. 17. Amos Tversky & Daniel Kahneman, Availability: A Heuristic for Judging Frequency and Probability, 5 COGNITIVE PSYCHOL. 207 (1973). 18. See Zamir & Teichman, supra note 14, at 448. 19. Id., at 449–450. 20. A recent study of State-level laws in the U.S. shows empirically that stronger penalties for “wage theft” succeed in creating effective deterrence. See Daniel J. Galvin, Deterring Wage Theft: AltLabor, State Politics, and the Policy Determinants of Minimum Wage Compliance, 14 PERSP. ON POL. 324 (2016). 21. For a more detailed and nuanced discussion see Hardy, supra note 11. ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 2022] 2/1/2023 12:00 PM IMPROVING COMPLIANCE WITH LABOR LAWS 77 A similar analysis can be made with regard to positive incentives. As predicted by economic analysis, people respond to incentives, and this can be an important factor of any strategy to improve compliance. At the same time, we cannot assume that employers have full information (including knowledge about the law), that they only care about profits, and that they always act rationally. Studies in behavioral law and economics help us understand these limitations. Alongside over-optimism and the availability bias, mentioned above, there are many other limitations of human rationality that can limit the effectiveness of incentives.22 One notable example that will prove relevant to our analysis is loss aversion. 23 People are much more worried about losing something they believe they have, then about not gaining something new, even if in fact the economic impact is the same.24 B. Behavioral Ethics Economic analysis assumes that employers only respond to deterrence and incentives to maximize profits. This may be true for some of them, but surely there are also people—including employers—that consider obeying the law a moral duty. Many of us are “good people” or at least potentially good people, in this sense. What is the thought process of those who are not “bad” but still decide to break the law? There are several options. First, it could be an “automatic” decision rather than a deliberative one, i.e. the employer can just do it without giving any thought to it and without making an explicit decision about it. Second, and probably most commonly, a decision to ignore the law could come with some rationalization—convincing oneself that the violation is not morally wrong—either in the specific circumstances (for the concrete case) or more generally (for a group of cases). That includes excuses such as (a) similar violations are common among peers/competitors (“everybody does it”); (b) the law is unfair—specifically because it places an unreasonable burden on employers; (c) the law is unjustified—specifically based on the view that employees will lose their jobs if the burden is too high on the employer, and therefore will be better off with the violation; and (d) the violation is temporarily justified because of some difficulties faced by the employer (losses etc.).25 22. See Zamir & Teichman, supra note 14 for an extensive overview and discussion. 23. There is some debate about whether this trait is irrational or not, but that is not important for current purposes. 24. Daniel Kahneman & Amos Tversky, Prospect Theory: An Analysis of Decision Under Risk, 47 Econometrica 263 (1979); Eyal Zamir, Law, Psychology, and Morality: The Role of Loss Aversion (2014). 25 .Although in theory, people do not believe that financial deprivation justifies cheating, when manipulated into thinking that they are financially deprived they tend to cheat more (see Eesha Sharma et al, Financial Deprivation Selectively Shifts Moral Standards and Compromises Moral Decisions, 123 Organizational Behav. & Hum. Decision Processes 90 (2014)). Another study has shown that people are ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 78 COMP. LABOR LAW & POL’Y JOURNAL 2/1/2023 12:00 PM [Vol. 43:71 A rich literature on behavioral ethics studies these processes.26 There is plenty of evidence showing that people break the rules more easily when they can maintain their self-image as honest, moral people. Research has shown that people conveniently forget the rules—or try to avoid inconvenient information—to help them feel better about themselves. Similarly, people tend to downplay the seriousness of their actions, or put the blame on others. Various psychological mechanisms such as “motivated reasoning,” “selfdeception,” “moral disengagement,” “bounded awareness,” “ethical blind spots” and “elastic justification” lead people to rationalize their unethical behavior.27 Some of this research relies on Daniel Kahneman’s distinction between two systems of reasoning: one fast, automatic, intuitive and mostly unconscious (“system 1”); the other slower and deliberative (“system 2”).28 The first system has been shown to be especially associated with self-interest, which is often people’s automatic instinct. However, the mechanisms noted above lead to unethical self-serving behavior in deliberative, conscious decisions as well.29 And while most of the research refers to decision-making by individuals, the same problems exist in organizations, because they are comprised of people who make the decisions.30 To increase the chances of compliance, we should therefore try to prevent “automatic” decisions, and push employers towards an internal deliberative process as much as possible, for example by requiring a hearing or a “cooling-off” period before a decision.31 We should also attempt to block the different routes towards selfserving rationalizations. There are various individual factors that determine the degree of “moral awareness” which varies from one person to another. But there are also two more likely to behave dishonestly when they face loss (Mary C. Kern & Dolly Chugh, Bounded Ethicality: The Perils of Loss Framing, 20 PSYCHOLOGICAL SCI. 378 (2009)). 26. According to David M. Messick & Max H. Bazerman, Ethical Leadership and the Psychology of Decision Making, 37 Sloan Mgm’t Rev. 9 (1996), there are three types of theories that people utilize when making decisions: theories about the world, about others, and about themselves. The authors focused on biases, but the same distinction is helpful in the current context as well. Another study has shown that eight different mechanisms are used for “moral disengagement”—i.e. for people to convince themselves that their actions are not immoral (see Albert Bandura, Moral Disengagement in the Perpetration of Inhumanities, 3 Personality & Soc. Psychol. Rev. 193 (1999)). The options listed above are those relevant in the current context and have some parallels in the list developed by Bandura. For further discussion see Feldman, supra note 5, at 52. 27. For an overview of the research see Feldman, supra note 5, at Ch 1-2. And see Max H. Bazerman & Francesca Gino, Behavioral Ethics: Toward a Deeper Understanding of Moral Judgment and Dishonesty, 8 Ann. Rev. of L. & Soc. Sci. 85 (2012). 28. Daniel Kahneman, Thinking, Fast and Slow (2011). 29. See Feldman, supra note 5, at 37-41 for a review of the evidence. 30. There are additional mechanisms related to decision-making by groups which will not be discussed here. While they add another layer of complexity which is important to understand, they do not solve the basic problems discussed here. 31. In many legal systems a hearing before dismissals is required by legislation. In Israel it was required by the courts; see section 4 below. On the idea of a “cooling-off” period before a decision to dismiss based on the behavior of an employee outside of work, see Tammy Katsabian, Employee’s Privacy in the Internet Age: Towards a New Procedural Approach, 40 BERKELEY J. EMP. & LAB. L. 203 (2019). ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 2022] 2/1/2023 12:00 PM IMPROVING COMPLIANCE WITH LABOR LAWS 79 important situational factors that have been revealed in the literature to be important for our awareness that an ethical problem exists. 32 The first is “issue intensity,” which is affected in particular by the magnitude of the consequences and the probability of harm. Making employers aware of the severity of their violation and the likelihood of harm to employees could thus be helpful in minimizing violations. The second situational factor is ethical infrastructure, which can be formal (e.g., a code of conduct) or informal (the ethical “climate”). For example, a firm with previous incidents of unethical decisions, and with a culture of using euphemistic language, 33 has an unethical climate. If we can impact the language that is used and the corporate culture, this can also raise moral awareness. Research has shown that the problem of “moral hypocrisy” (acting unethically but feeling ethical) can be minimized when the moral standard is made clear, and the level of self-awareness is high. 34 In an interesting experiment, people behaved more ethically when they were told explicitly what is the right thing to do and also a mirror was (literally) put in front of their face to ensure self-awareness. When people feel that they are being watched they tend to behave more ethically; in a striking study, people were asked to put money in jar when taking coffee at an office kitchen, and the payments were tripled when a picture of eyes was placed over the jar. 35 Arguably, a duty on employers to inform employees of their specific rights— and sometimes even to admit to violations—can serve to raise moral awareness.36 Thus far we explained how people often fail to acknowledge that they violated the law or that they acted immorally, in order to maintain their positive self-perception. A similar phenomenon is sometimes present already at the earlier stage of knowing what the law says. Research has shown that people sometimes engage in “information avoidance” in order to create for themselves some “moral wiggle room.”37 In other words, they turn a blind eye; they prefer not to know some aspect of the law. This allows them to 32. Ann E. Tenbrunsel & Kristin Smith‐Crowe, Ethical Decision Making: Where We’ve Been and Where We’re Going, 2(1) THE ACAD. OF MGM’T ANNALS 545 (2008). 33. For example, “shifting resources” instead of “stealing” (see FELDMAN, supra note 5, at 52). 34. C. Daniel Batson et al, Moral Hypocrisy: Appearing Moral to Oneself Without Being So, 77 J. OF PERSONALITY & SOC. PSYCHOL. 525 (1999). 35. Melissa Bateson et al, Cues of Being Watched Enhance Cooperation in a Real-World Setting, 2 BIOLOGY LETTERS 412 (2006). 36. A notable example is the remedy developed by the National Labor Relations Board in the U.S., requiring employers to post—or even read out loud before employees—information about freedom of association and about the employer’s violations. See Thomas C. Barnes, Making the Bird Sing: Remedial Notice Reading Requirements and the Efficiency of NLRB Remedies, 36 BERKELEY J. OF EMP. & LAB. L. 351 (2015); Benjamin I. Sachs, Law, Organizing, and Status Quo Vulnerability, 96 TEX. L. REV. 351, 366 (2017). 37. Jason Dana et al., Exploiting Moral Wiggle Room: Experiments Demonstrating an Illusory Preference for Fairness, 33(1) ECON. THEORY 67 (2007). See also Feldman, supra note 5, at 50. ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 80 COMP. LABOR LAW & POL’Y JOURNAL 2/1/2023 12:00 PM [Vol. 43:71 create—at least for themselves internally—some grey area to operate without acknowledging that they are breaking the law. Dan Arieli and his colleagues summarize the practical conclusions of this literature by proposing three interventions: reminding, visibility, and self-engagement.38 Reminding is intended to “provide cues that increase the salience of ethical criteria and decrease ability to justify dishonesty,” for example by reminding people why certain actions are immoral and how others can be harmed by them. Visibility is intended to “increase people’s feeling they are being seen and identified,” for example by making reporting processes personalized. Self-engagement is intended to “break down morality into concrete behaviors” and “obtain self-commitment to act morally prior to behavior,” for example by asking people to sign an honor code. C. The Expressive Function of the Law When the law creates mandatory norms—whether through legislation or judicial precedents—it sends a message to people. The most obvious message is: “you should do X (or avoid doing Y), otherwise you will face punishment or other sanctions.” For many people, this is enough, whether because they respect the rule of law and will follow the instructions of a legitimate authority, or because they are deterred by the threat of punishment. However, there is another message that the law sends, which is also important for understanding compliance. The law tells people what to do, and it has been argued by several scholars that this is important for two other reasons (unrelated to sanctions or legitimacy): promoting coordination and conveying information.39 The idea that what the law says has an impact on behavior in itself—it can lead people to comply—has been called an expressive theory of law’s effects.40 The coordination function is explained by Richard McAdams by relying on game theory.41 He starts with a simple example of a law telling people to drive on the right side of the road. In order to avoid collusion, we all need to coordinate on this matter, and the law provides this function. Whether there are sanctions or not, most people will look for this guidance and comply with the law in order to secure the gains of coordination.42 Even in situations of 38. Shahar Ayal et al., Three Principles to Revise People’s Unethical Behavior, 10 PERSPECTIVES ON PSYCHOL. SCI. 738 (2015). 39. RICHARD H. MCADAMS, THE EXPRESSIVE POWERS OF LAW: THEORIES AND LIMITS (2015). 40. Id., at 13. 41. Id., at Ch 2. 42. This situation is quite rare in the sense that people do not mind if the law chooses right or left. But the same idea also applies in “mixed motive” games, the much more common situation in which people have “a mutual desire to coordinate behavior but disagreement over how to coordinate” (id., at 23). The traffic light is a simple example: although we might all prefer to get priority to proceed at the intersection, most people will yield as instructed by a red light, even without sanctions—because we ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 2022] 2/1/2023 12:00 PM IMPROVING COMPLIANCE WITH LABOR LAWS 81 competition, the competitors desire coordination on the terms of the “game,”43 and in particular might follow the suggestion of a third party (such as a court) which constructs a “focal point” for them. 44 As explained by Robert Cooter, “in the case of social norms… law can influence where the system settles by coordinating expectations.”45 McAdams argues that a law’s expression can serve as a new focal point even in situations of an existing equilibrium: even if some custom was already created, it is still possible for the law to replace it (by the sheer expression, unrelated to sanctions and deference to legitimate authority).46 But he adds that situations of disequilibrium—in which an expression from the law is especially needed—are very common: “Technological and cultural change constantly upset previously settled expectations about what others will do. These states of disorder are the kind of situations in which legal change occurs, so a new law need not always compete with an otherwise unchallenged set of settled expectations.”47 The coordination function seems highly relevant for labor law, on two different levels. First, consider coordination between employers. Although some employers may wish to avoid all coordination with competitors regarding employment rights, others will find it helpful that the law creates a floor of rights which are taken out of the competition. Certainly the “good” employers who want to pay their employees well and respect their rights, would like to know that others are bound by similar norms and do not undercut them by competing on who will pay less. While direct coordination on such matters between competitors is prohibited by anti-trust laws, with regard to basic employment rights the law serves the same function. 48 Second, coordination is also needed between employers and their employees. With regard to terms explicitly agreed in advance and included in the understand that coordination in such cases is better than any other alternative. See also Edo Eshet, Coercion and Freedom in Labour Law: American, Canadian, and Israeli Perspectives, 33 INT’L J. OF COMP. LAB. L. & INDUS. REL. 489, 511 (2017). 43. MCADAMS, supra note 39, at 28. 44. Id., at 44 (relying on the theory first developed by THOMAS SCHELLING, THE STRATEGY OF CONFLICT (1960), and reviewing (in chapters 2-3) various empirical studies supporting this theory). See also Richard H. MCADAMS, The Expressive Power of Adjudication, 2005 U. ILL. L. REV. 1043, 1089 (2005). 45. Robert Cooter, Expressive Law and Economics, 27 J. OF LEG. STUD. 585, 594 (1998). 46. McAdams, supra note 39, at 97–99. 47. Id., at 99. McAdams further argues that law’s expressive powers are especially strong when social movements seek to destabilize an existing convention (as, for example, with the shift towards smoking limitations—where the law facilitated coordination between non-smokers who wanted to change the norm), and when the law renders an existing custom which is ambiguous more precise (id., at 100109). For an example of a judgment with expressive importance following a social movement (regarding gay marriage), see Kyle C. Velte, Obergefell’s Expressive Promise, 6 HOUS. L. REV.: OFF THE RECORD 157 (2015). 48. A recent Israeli case mentions coordination between employers as one of the justifications for a law allowing extension orders and requiring employers to pay fees to employers’ organizations. See (National Labor Court) R.S.L Electronics v. El v. Manufacturer’s Association of Israel, (2020) (Isr.). ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 82 COMP. LABOR LAW & POL’Y JOURNAL 2/1/2023 12:00 PM [Vol. 43:71 employment contract, cooperation is secured by the parties themselves. But often, especially in long-term “relational” contracts such as employment, questions will arise about issues that have not been settled in advance. If the law announces a solution, it solves the dilemma and creates a focal point around a solution that adjusts the expectations of both parties.49 The second expressive power of law is conveying information. The law tells us something about what the political actors (or other State authorities) believe, and this can change our own beliefs and affect behavior—often contributing to compliance. 50 This is relevant only when people actually know that the law exists and what it requires; that is, for this function to be effective, there has to be publicity to the new law, so a significant number of people become aware of it.51 The information can be about public attitudes towards the regulated behavior, signaling that the public disapproves of certain actions; or it can be about facts, for example letting people know that a certain behavior is especially risky or harmful to others.52 A recent study shows that giving reasons for the law—explaining its rationale—also advances compliance.53 The ability of the law to influence social norms by conveying information is especially relevant when we are trying to change existing problematic norms,54 and can be most effective when affecting the certainty of beliefs.55 Thus, for example, the American with Disabilities Act of 1990 arguably presented employers “with a different vision of disability identity than that previously held,” leading employers to update their knowledge and beliefs.56 Other common examples are non-smoking law57 and seatbelts laws.58 Obviously a change of attitudes is not always easy to achieve. As far as courts are concerned, their ability and willingness to 49. Compare to laws declaring a right to breastfeed in public, discussed by McAdams, supra note 39, at 86. And see, on the role of labor law in supporting trust between the parties, Simon Deakin & Frank Wilkinson, Labour Law and Economic Theory: A Reappraisal, in LEGAL REGULATION OF THE EMPLOYMENT RELATION 29, 56-59 (Hugh Collins, Paul Davies and Roger Rideout eds., 2000). 50. McAdams, supra note 39, at Ch 5. 51. Id., at 137, 179. 52. It has been argued that individuals are not impacted directly, but rather through their groups: the law can shape group values and norms, which in turn influence the values of individuals in this group. See Janice Nadler, Expressive Law, Social Norms, and Social Groups, 42 L. & SOC. INQUIRY 60 (2017). 53. Daphna Lewinsohn-Zamir et al., Giving Reasons as a Means to Enhance Compliance with Legal Norms, U. TORONTO L.J. (forthcoming). 54. Cass R. Sunstein, On the Expressive Function of Law, 144 U. PENN. L. REV. 2021 (1996). 55. Alex Geisinger, A Belief Change Theory of Expressive Law, 88 IOWA L. REV. 35 (2002). 56. Michael Ashley Stein, Under the Empirical Radar: An Initial Expressive Law Analysis of the ADA, 90 VA. L. REV. 1151, 1186 (2004). 57. For an analysis, see Lawrence Lessig, The Regulation of Social Meaning, 62 U. CHI. L. REV. 943, 1025ff (1995). 58. For empirical evidence that such laws have expressive impact, see Maggie Wittlin, Buckling under Pressure: An Empirical Test of the Expressive Effects of Law, 28 YALE J. ON REG. 419 (2011). ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 2022] 2/1/2023 12:00 PM IMPROVING COMPLIANCE WITH LABOR LAWS 83 engage in trying to change social norms depends on the institutional pressures that they face and the “social capital” they enjoy at a given time.59 Relatedly, Lawrence Lessig discussed the regulation of “social meaning”—which he defined as “the semiotic content attached to various actions, or inactions, or statuses, within a particular context.”60 He mentioned several techniques that can be used to change social meanings, among them “tying,”61 (i.e., transforming the social meaning of one act by associating it with the social meaning of another.) This can be relevant, for example, for attempts to change the social meaning of non-payment of wages or benefits, by associating such practices with theft.62 III. RAISING AWARENESS ABOUT THE LAW After reviewing different theories related to compliance, with some examples on how they can be relevant for labor law, we now turn more specifically to the role of courts. The next four sections discuss different possibilities for courts to influence compliance, which rely on the theoretical discussion above. We start, in the current section, with raising awareness. It is quite obvious that knowing about the law is a necessary precondition for compliance. But the previous section showed how crucial this awareness is, and helped direct our attention to several specific mechanisms. To achieve deterrence, people need to know what the law requires from them and also be aware of the expected punishment and other costs in case of violation. To minimize moral hypocrisy and other behavioral ethics failures, people have to acknowledge that they are violating the law—and we should make it difficult for them to use self-deception by turning a blind eye to such violations. To achieve the expressive effects of the law, people need to understand what the law says, in order to learn new information from it (for example, about changing public attitudes), and realize how it contributes to coordination. Legislation sometimes creates mechanisms for raising awareness: for example, the law can require employers to post information about workers’ 59. Jason Mazzone, When Courts Speak: Social Capital and Law’s Expressive Function, 49 SYRACUSE L. REV. 1039 (1999). 60. Lessig, supra note 57, at 951. 61. Id., at 1009. 62. The term “wage theft” is now widely used in the U.S. as a label for non-payments. See, e.g., Brishen Rogers, Toward Third-Party Liability for Wage Theft, 31 BERKELEY J. EMP. & LAB. L. 1 (2010); Matthew W. Finkin, From Weight Checking to Wage Checking: Arming Workers to Combat Wage Theft, 90 IND. L.J. 851 (2015); Llezlie L. Green, Wage Theft in Lawless Courts, 107 CAL. L. REV. 1303 (2019). And see Government Accountability Office, Wage and Hour Division’s Complaint Intake and Investigative Processes Leave Low Wage Workers Vulnerable to Wage Theft (2009), https://www.gao.gov/highlights/d09458thigh.pdf. For a discussion of possible ways to influence social norms against wage theft, see Nicole Hallett, The Problem of Wage Theft, 37 YALE L. & POL’Y REV. 93, 141ff (2018). ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 84 COMP. LABOR LAW & POL’Y JOURNAL 2/1/2023 12:00 PM [Vol. 43:71 rights on message boards. 63 This is intended to raise awareness first and foremost among employees, but is likely to achieve the same effect among employers as well. The Government can raise awareness to specific legal obligations through media campaigns, for example. What can courts do? We believe judges have a role to play in this regard as well. There are instances in which courts clearly try to “send a message” to the public (in our context, to employers). To be sure, employers are not likely to get the message directly from judgments; but they can get it through their legal advisors and through the media. We bring several examples below from Israeli case-law. While none of them addresses the need to improve compliance overtly, we believe it is justified for courts to take into account the need to raise awareness about the law, and this consideration should become more explicit. The judicial tools that can contribute to raising awareness are both procedural and substantive. Procedurally, the court can signal that a case is important by assigning an enlarged forum of judges to the case (a “full bench” or “expanded bench” where this is possible); 64 by inviting the Attorney General to present its opinion in a dispute between two private parties; by allowing other public interest parties to join the proceedings as “friends of the court” (amicus curiae); and by deciding to hear several cases which raise the same legal question together. To be sure, all of these procedural tools are needed first and foremost in order to improve the decision-making process— to contribute to a better judgment. At the same time, however, they serve an additional role, of signaling to the legal community and to the media that the specific case is important. In turn, the judgment is likely to receive more attention, helping to raise awareness. Consider for example the case of Omri Kiss, decided by the Israeli National Labor Court in 2018, which dealt with the legal status of tips (whether they count as wages for purposes of minimum wages and for purposes of workers’ compensation benefits). 65 The legal question is not settled in Israeli legislation. It was partially resolved several years earlier by the same court, in a way that allowed the counting of tips as part of the wage only if the tips are formally reported in the employer’s tax records.66 In the years following the previous judgment, compliance by employers was lacking, and it became clear that the judgment, which intended to protect employees (at least to some extent) in the labor law context, is harming them in the workers’ compensation context. Moreover, even compliant employers 63. In Israel, for example, such a duty exists in the Prevention of Sexual Harassment Regulations (Employers’ Duties) of 1998. 64. In Israel, the National Labor Court usually sits in panels of five: three professional judges and two lay judges, but the Article 20 of Labor Court Act of 1969 (Isr.), allows the Court’s President to decide on a larger panel, if the legal questions are especially important, difficult or innovative. 65. (National Labor Court) Omri Kiss v. The National Insurance Institute, (2018) (Isr.). 66. (National Labor Court) DGMB Eilat Restaurants v. Inbal Malka, (2005) (Isr.). ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 2022] 2/1/2023 12:00 PM IMPROVING COMPLIANCE WITH LABOR LAWS 85 issued wage slips on the minimum wage only, on the understanding that the rest of the tip amount is not part of the wage. Pension deductions, as well as vacation and sick payments, were made based on the minimum wage only. In the Kiss case the Court wanted to correct these failures, and eventually created a revised and more complete regulation of tips. Without getting into the details of the new judge-made law (which we do not necessarily support), we raise this case as an example of the Court using procedural tools to “send a message.” All the tools mentioned above were used simultaneously. The President of the Court, who also wrote the decision, decided to enlarge the panel to seven judges (instead of the regular five). Two unrelated cases representing two aspects of the problem (minimum wage and workers’ compensation) were joined together by a decision of the Court. Furthermore, on the Court’s initiative, the Attorney General, the Restaurants’ Association and the Waiters’ Union were all invited to submit opinions, even though they were not parties to the case. Moreover, the President of the Court chose this judgment to be delivered on his last day in office, together with his retirement celebrations, which is another common method used by Israeli judges to draw public attention to a particular case. And the decision delayed the entry into force of the new rules for a period of eight months, to allow employers to reorganize—which is perhaps another way of attracting attention to the importance of the new law. The judgment received significant media coverage as well as close attention (and critique) from restaurant owners. We believe that these procedural tools were important in raising the profile of the case, which in turn was helpful for the Court to raise public awareness to the problem and the new judge-made regulation. Substantively, we believe that courts sometimes “send a message” to employers by crafting legal rules that appear extreme and revolutionary, in order to generate a shift in managerial culture and attitudes. The rules are then adjusted into a more nuanced and balanced solution in later judgments, but the original decision, which can be described as a “door in the face,”67 is drafted in strong and unequivocal terms, signaling to employers and the public the court’s shift in attitude towards existing practices. Such an extreme stance attracts the attention of the legal community, employers and the media, thus helping to raise awareness about the law. Two cases of the Israeli National Labor Court can serve as examples for this argument. The first, Frumer v. Red Guard,68 dealt with restraint of trade covenants, also known as non-compete agreements. As in some other countries, the Israeli courts allow such covenants only to the extent they are “reasonable,” otherwise they are considered against public policy and 67. Guy Davidov & Maayan Davidov, How Judges Use Weapons of Influence: The Social Psychology of Courts, 46 ISRAEL L. R. 7, 17–19 (2013). 68. (National Labor Court) Dan Frumer v. Red Guard Ltd., (1999) (Isr.). ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 86 COMP. LABOR LAW & POL’Y JOURNAL 2/1/2023 12:00 PM [Vol. 43:71 therefore void. Over the years, the attitude of the National Labor Court gradually changed, as a result of increased importance given to freedom of occupation as well as a competitive market, both of which are harmed as a result of restrictive covenants. 69 Throughout the 1990s, the Court showed less tolerance for such covenants, decreasing the “regular” accepted period of restriction from three years to two years and later to one year. However, the message did not go through to employers. In the late 1990s, restrictive covenants were widespread in the Israeli labor market, even for low-wage unskilled employees. Then came the judgment of Frumer, in which the Court decided to change the default rule: restrictive covenants are now presumed to be unreasonable and void, unless the employer can show that a covenant is justified in the specific circumstances; and the Court gave a list of possible justifications (such as protecting trade secrets), making clear that they will each be construed narrowly. This was a dramatic decision which attracted a lot of public attention. We do not argue that this new legal rule was established only for reasons of improving compliance. The Court wanted to change the rule for substantive reasons. We do, however, believe that the choice to articulate the new rule in extreme, uncompromising fashion, can be seen as an intention to send a clear message to employers, the legal community and the public at large, that existing practices will not be tolerated any longer. Anecdotal evidence suggests that this was successful: the use of restrictive covenants decreased significantly. After a couple of years, the Court gradually started to soften its stance, showing more willingness to accept justifications for the covenants. 70 Currently the law is more balanced and nuanced, but the extreme Frumer decision was instrumental in minimizing abuse by employers. The second example of a judgment arguably used to raise awareness is the case of Pelephone,71 which dealt with freedom of association. Over the last three decades, union density in Israel has dropped sharply. New attempts of organizing have met with fierce objections from employers. The basic principle of freedom of association was established well before the Pelephone case, and included a prohibition on employers to interfere by way of dismissing or otherwise harming workers because of organizing.72 But this was hard to enforce, and the organizing drive by Pelephone employees on 69. With regard to freedom of occupation, one of the reasons for the increased importance was the enactment of Basic Law: Freedom of Occupation in 1992. 70. (National Labor Court) Erroca International Ltd. v. Anthony Cori, (2001) (Isr.); (National Labor Court) Girit Ltd. v. Mordechai Aviv, (2003) (Isr.). 71. (National Labor Court) The Histadrut v. Pelephone Communications Ltd., (2013) (Isr.). 72. See, e.g., (National Labor Court) Mifa’ley Tahanot Ltd. v. Israel Yaniv, (2011) (Isr.). These protections were codified in legislation in 2001 as part of the Collective Agreements Act of 1957 (new articles 33(8) ff). ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 2022] 2/1/2023 12:00 PM IMPROVING COMPLIANCE WITH LABOR LAWS 87 2012 provided ample proof that existing protections were insufficient. The employer used every trick from the union-busting playbook, including threatening employees who joined the union, preventing union officials from access to the workplace, sending false information about the union to all employees, creating “focus groups” with employees after which they were expected to revoke their union membership, and so on. The union went to the regional labor court several times, and each time Pelephone said it was a local initiative by one of the junior managers, and promised to stop it. Injunctions were given, but could not stop the next anti-union method that the employer turned to. The union could also sue for damages, but this requires a full legal process which takes years, and will not help the immediate goal of organizing. In any case, the damages awarded are never high enough to change the mindset of anti-union employers.73 When the issue came before the National Labor Court, the judges knew that the case is not exceptional; employer resistance to unions, with problematic and often illegal methods, became widespread in Israel. It became clear that new and stronger ways are needed to protect freedom of association. The issue is not concretely settled in legislation, leaving broad discretion to the courts to develop the law. In an important judgment that attracted broad attention, the Court decided that employers are not allowed to voice any objection to unions. That is, instead of prohibiting specific practices (such as threating employees who join a union), the new judgemade law prohibits all actions by employers against the union, even attempts to convince employees that the union will be bad for them, which were previously allowed. This clear bright line means that the employer’s freedom of speech is infringed, but the protection of freedom of association is much easier to enforce. The Court signaled from the beginning its intention to raise the “profile” of the case, by using the procedural tools mentioned above: inviting the Attorney General as well as an employers’ organization to submit opinions, and allowing labor unions who were not parties to the case to file amicus curiae opinions. Substantively, the Court chose an extreme approach, which does not leave any room for employers to engage in legitimate speech concerning the unionization drive. The one very minor exception mentioned in the judgment only served to show how extreme it was: even when the union is giving the employees false information about the employer, the latter is not allowed to respond without first getting permission from a labor court. When employers’ organizations tried to challenge the judgment before the Supreme Court, the highest Court refused to intervene, but noted that the 73. Although there is recently willingness to award more significant punitive damages than in the past; see section 6 below. ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 88 COMP. LABOR LAW & POL’Y JOURNAL 2/1/2023 12:00 PM [Vol. 43:71 National Labor Court can be expected to develop the new rule from case to case and gradually achieve a more nuanced solution.74 We do believe that the judgment was justified, at least for its time and place, despite being extreme.75 It was justified especially to improve enforcement of the rules protecting freedom of association. And for this purpose, awareness about the rules is crucial. Arguably, the choice of drafting the new rule as extreme and unequivocal, without exceptions, was made also to attract attention and raise awareness.76 There are also, of course, cases in which courts could do more to raise awareness to the law. Consider, for example, situations in which courts pressure the parties to agree to a settlement. This is common in the Israeli labor courts system (as in many other legal systems), sometimes even in suits concerning the violation of non-waivable employment laws. Obviously there are arguments in favor of such settlements; this is not the place to discuss the pros and cons of this practice. But it is pertinent to point out that one of the arguments against settlements achieved with the assistance (or pressure) of courts is the negative impact on awareness about the law. Although it is possible that an agreement by an employer to pay substantive sums as a result of labor law violations would reach the media and other employers, in most cases settlements will receive little publicity. Moreover, the lack of a clear statement from the courts that the law has been violated, reminding people about what the law requires and raising awareness to the costs of violations for the employees as well as the expected punishment, make any settlement a missed opportunity. It may be that in any given case the advantages for the plaintiff (and for the legal system) outweigh the disadvantages of a settlement; we cannot reject this possibility. Our point is simply that courts must take the consideration of improving compliance into account before proposing settlements—specifically, they must consider the cost in terms of public awareness about the law. 74. (National Labor Court) Coordinating Chamber of Economic Organizations v. National Labor Court, (2014) (Isr.). 75. For an analysis see Pnina Alon-Shenker & Guy Davidov, Organizing: Should the Employer Have a Say?, 17 THEORETICAL INQUIRIES IN L. 63 (2016). 76. Alongside the rule itself, the judgment explains in detail the importance of freedom of association. In other judgments as well, we can see courts explaining the justifications for the law in detail and sometimes passion, which can also be seen as an attempt to encourage compliance. Although employers are not likely to read the full reasons for judgments (as opposed to hearing about the bottom line), lawyers might read the full reasons and can be influenced by them to encourage their clients to comply. The detailed explanation about the importance of the law can thus be seen as a further tool to raise awareness. ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 2022] 2/1/2023 12:00 PM IMPROVING COMPLIANCE WITH LABOR LAWS 89 IV. A BALANCE BETWEEN AMBIGUITY AND SPECIFICITY Laws come in different degrees of specificity. Every legal system includes some combination of rules and standards. A rule is specific, for example “driving at a speed of over fifty km/h within city limits is prohibited.” A standard is open-ended, for example “driving at a dangerous speed is prohibited.” In the labor law context as well, we have rules (e.g., paying wages below 29.12 NIS per hour is prohibited), and standards (e.g., an employer must act in good faith towards the employee). Each technique has its own advantages.77 To a large extent the choice of legal technique is made by the legislature, but courts also play a major role in deciding the level of specificity. We argue that the choice of technique has an important impact on the level of compliance, which judges should explicitly take into account. Rules are generally seen as ensuring a higher degree of clarity, predictability and certainty; and as a result, they are easier to enforce, because it is easier for the public to know what the law requires, and easier for the enforcement agency to recognize violations. Otherwise put, if we want to deter a specific behavior (such as wages below a certain level), rules are better. Moreover, an open-ended standard opens more room for moral ambiguity, i.e. for “good” employers to deceive themselves that they are not violating the law.78 A further advantage of rules comes from the expressive theory: if we want to achieve coordination between employers, preventing a “race to the bottom” by taking some aspect of workers’ rights out of the competition, then employers need to be clear about what is required from them and what they can expect other employers to do. This also suggests a preference for clear-cut rules.79 At the same time, standards are much better at covering new and unforeseen problems, and responding to sophisticated evasion attempts, because they cast a much wider net. This is especially important when regulating long-term, ever-changing, and complex relations such as 77. The literature on this topic is vast. Among the notable contributions are Isaac Ehrlich & Richard A. Posner, An Economic Analysis of Legal Rulemaking, 3 J. OF LEG. STUD. 257 (1974); Cass R. Sunstein, Problems with Rules, 83 CAL. L. REV. 953 (1995); John Braithwaite, Rules and Principles: A Theory of Legal Certainty, 27 AUSTRALIAN J. OF LEG. PHIL. 47 (2002). For reviews of this literature as applied to the labor law context, see Davidov, supra note 2, at 159 ff; David Cabrelli, The Role of Standards of Review in Labour Law, 39 OXFORD J. OF LEG. STUD. 374 (2019). 78. See Feldman, supra note 5, at 102, 187. 79. McAdams, supra note 39, at 235. Note that McAdams qualifies this recommendation as relevant only for laws directed at a large and heteronomous audience. It might not be relevant for laws directed at specific sectors, where actors are more homogeneous and have a shared understanding about industry custom. It is also not relevant for laws that regulate relations between two specific parties, who might have a shared understanding of what the standard means (see id., at 236). So, from the perspective of the need to place obligations on an employer vis-à-vis its employees, the argument coming from the expressive theory of law in favor of rules is not valid. Still, to the extent that labor laws also seek to create a level playing field for employers, and to do so without regulating specific sectors separately, the argument in favor of rules has some force. ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 90 COMP. LABOR LAW & POL’Y JOURNAL 2/1/2023 12:00 PM [Vol. 43:71 employment. In a relationship characterized by inequality of bargaining power, which can be abused by the powerful party (the employer) in infinite ways, it is impossible to foresee in advance all the possible occurrences that need to be prohibited and list them as detailed rules. It has also been noted that specific rules could undermine intrinsic motivation, which is important for compliance by “good” employers, while standards are better at reminding people that they should act morally, rather than focus on concrete legal duties only. Relatedly, standards are considered better for fostering trust and confidence between contracting parties.80 The distinction between rules and standards is not a clear-cut dichotomy; there are different variations in-between, and various combinations of the two techniques. While in some cases, one of the legal techniques will clearly be better than the other, in other cases a balance is needed to maximize the advantages of both techniques. This can be done, for example, by deriving specific rules from a legislated standard, creating some clarity and determinacy about specific obligations, without undermining the existence of the standard as an additional open-ended obligation in the background.81 The Pelephone case mentioned in the previous section is an example of a judicial choice to derive a clear-cut rule from a standard. The principle of freedom of association, like other fundamental rights, can be infringed under Israeli law only for a legitimate cause and subject to the principle of proportionality. The limitation of proportionality creates an open-ended standard: there is a general obligation on employers to respect the right of employees to organize, and a rather ambiguous standard controls what counts as a prohibited violation. In Pelephone, the court decided to derive a specific rule from this standard: no employer speech with regard to unionizing is allowed. We have argued in the previous section that the extreme nature of this rule can be understood as “sending a message” to employers and increasing awareness to the law protecting freedom of association. We add here that the choice of a clear-cut rule rather than a more ambiguous standard can be explained by the need to ensure coordination between employers (prevent a race to the bottom and create a level playing field), and by the need to limit any moral wiggle room. It is interesting to add that people behave more unethically when facing the prospect of a loss, compared to facing the possibility of a gain. 82 In a non-unionized workplace, presumably the 80. For a comprehensive review of the behavioral studies related to the choice of rules vs. standards, see Feldman, supra note 5, at 184–189. 81. For a similar idea, supporting a middle ground by way of “catalogs” in which a standard comes with several examples alongside a residual category, see Gideon Parchomovsky & Alex Stein, Catalogs, 115 COLUM. L. REV. 165 (2105). 82. This is related to the phenomenon of loss aversion, mentioned above. See Zamir & Teichman, supra note 14, at 458–9. ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 2022] 2/1/2023 12:00 PM IMPROVING COMPLIANCE WITH LABOR LAWS 91 employer frames the entrance of a union as a loss. In contrast, where a union already exists, an employer that prefers to have unfettered control without bothering to negotiate with a union will frame the ousting of the union as a gain. Seen in this light, when fighting against a new organizing drive (the loss scenario) an employer is more likely to behave unethically. Therefore, more certainty (a rule) is needed, to limit the moral wiggle room, when protecting new organizing. At the same time, the new rule does not come instead of the general standard, which is still in the background and can be used to prevent other violations of freedom of association, unrelated to speech. Arguably, this allows the Court to secure the advantages of both rules and standards. Admittedly, there is a risk that in practice employers will see the rule as replacing the standard; some “good” employers might follow only the new specific rule and ignore the general standard. It is important in such cases to include in the judgments clear statements about the general standard and its continuing importance, and explain that the new rule only deals with a specific subset of cases. Another example of deriving a rule from a standard, again coming from the Israeli National Labor Court, is related to holding a hearing before dismissals. In Israel, although there are plenty of legislated protections against dismissals for various “bad” reasons, the default rule is employment at will. There is no legislated duty to hold a hearing before dismissals. However, there is a general legislated duty to perform contracts in good faith, and the labor courts have ruled over the years that this open-ended standard is especially important in relational, long-term contracts such as employment. In the 2002 of Herman, 83 the Court decided to derive from this general standard a specific rule: employers must hold a hearing for employees before making a decision about dismissals. This can be seen as judicial activism in the face of inaction by the legislature, after union density significantly declined, making such a rule increasingly necessary. But it is also an example of making the open-ended standard somewhat more specific. The Court could have used the good faith standard to rule in favor of the employee in the specific circumstances of the case. Instead, the Herman judgment maintained that as a general rule, not holding a hearing amounts to a violation of the good faith duty. Of course, the open-ended duty of good faith remains in force as well. Deriving a specific, relatively easy-to-apply rule from a standard can be contrasted with judicial attempts to create a set of detailed and complex rules, which appear much more problematic. In the case of Isakov, the Israeli National Labor Court had to decide whether an employer is allowed to look 83. (National Labor Court) Yosef Herman v. Sonol Israel Ltd., (2002) (Isr.). ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 2/1/2023 12:00 PM COMP. LABOR LAW & POL’Y JOURNAL 92 [Vol. 43:71 inside the e-mail correspondence of employees.84 Once again, the issue was not settled in legislation, except for a general protection of the right of privacy, with the open-ended principle of proportionality controlling infringements. The Court adopted a detailed set of rules, differentiating between a private e-mail box, an employer-owned box, and a “mixed” box; creating different and rather complex rules for each box; demanding that employers secure informed consent before looking into e-mail correspondence, both in general (consent for a general policy) and in some cases for the specific incidence as well. It is doubtful if such detailed regulation is appropriate for the judiciary, both in terms of legitimacy and in terms of competence. More importantly for current purposes, it is not clear that such a detailed set of complex rules is the best balance between rules and standards, given the goal of securing maximum compliance with the general duty to respect employees’ privacy. It is quite likely that employers will be able to work around these rules; the probable result will be that employers will see themselves allowed to read all e-mail correspondence, as long as they follow a set of specific instructions on how to do so (such as, for example, having every employee sign a contract drafted in line with the specific rules). V. THIRD PARTY LIABILITY Labor law is based on the assumption of bargaining power inequality between an employer and its employees. Internally, this is usually the case. At the same time, it does not mean that the employer is necessarily powerful externally. Some employers are subject to power dynamics that make them dependent on specific clients, often in situations of unequal bargaining power vis-à-vis those clients. In such cases, it is possible to increase compliance by exerting indirect pressure on the employer through its client. The law can achieve this by placing legal liability on clients for labor law violations by their contractors, at least to some extent. Such a legal regime creates incentives for the clients to serve as agents of enforcement, and use their power to ensure that contractors obey the laws. There are some similarities to this analysis in the relations between franchisors and franchisees, but also some differences which require a separate discussion. Our focus here is limited to the possibility of placing liability on lead companies that engage with contractors in client-contractor relations. As a general rule, placing liability on a client is hard to justify. If the client bears no blame whatsoever for the legal violations, why should it be responsible for them? However, sometimes the client is causing the violations, even if indirectly, by not paying the contractor enough. When the 84. (National Labor Court) Tali Isakov Inbar v. The State of Israel, (2011) (Isr.). ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 2022] 2/1/2023 12:00 PM IMPROVING COMPLIANCE WITH LABOR LAWS 93 client uses its power to eliminate the contractor’s ability to generate enough profits, the latter is likely to utilize its own power superiority, vis-à-vis the workers, and secure profits through labor law violations. At other times, although the client is not indirectly causing the violations, it is in a position to prevent them quite easily and at a minimal cost.85 This is the case when the client has some control over the contractor, for business reasons (supervising the provision of services), and can use the same control to ensure compliance with labor laws as well. The ability to cheaply prevent a violation may not be enough by itself to justify placing legal liability to do so, but with the addition of other factors creating a connection to the workers (for example, they perform their work regularly on the client’s premises), we believe this can be justified.86 Client liability can be imposed in legislation, and to some extent has been imposed in Israel, as will be mentioned shortly. But our focus here is on the role of courts. We bring several examples for judgments in which courts showed creativity and imposed liability on third parties for labor rights. These cases should be understood as creating a mechanism to increase compliance ex-ante by contractors. In a series of cases, the National Labor Court hinted that clients could be held responsible for workers employed indirectly for the benefit of the client.87 This approach culminated in the case of Shmuelov, decided by the Tel-Aviv Regional Labor Court in 2006.88 A major Israeli bank contracted a small contractor to provide cleaning services for one of its branches. The contractor employed Ms. Shmuelov who worked at the branch. She was paid less than the minimum wage, and many of her other employment rights have been violated. When she sued the contractor, she also filed a suit against the bank, claiming it was a joint employer. Based on the existing case-law concerning “who is the employer,” the Court did not find sufficient indicators that the bank had characteristics of an employer. At the same time, the Court decided that the bank had responsibility for the violations because it paid the contractor less than the minimum necessary to be able to pay Ms. Shmuelov 85. The idea of placing liability on the cheapest cost avoider is common in tort law theory. See Guido Calabresi & Jon T. Hirschoff, Toward a Test for Strict Liability in Torts, 81 YALE L.J. 1055, 1070 (1972). 86. For further discussion of this point see Davidov, supra note 2, at 234–238; Guy Davidov, Indirect Employment: Should Lead Companies Be Liable?, 37 COMP. LAB. L. & POL’Y J. 5 (2015); Brishen Rogers, Towards Third-Party Liability for Wage Theft, 31 BERKELEY J. OF EMP. & LAB. L. 1 (2010); Yossi Dahan et al., Global Justice, Labor Standards and Responsibility, 12 THEORETICAL INQUIRIES IN L. 117 (2011); Tess Hardy, Who Should Be Held Liable for Workplace Contraventions and on What Basis?, 29 AUSTRALIAN J. OF LAB. L. 78 (2016). 87. (National Labor Court) The National Kibbutz Department of Construction v. Halil Abed Al Rahman, (1995) (Isr.); (National Labor Court) Xue Bin v. A. Dori Construction Company, (2003) (Isr.); (National Labor Court) Dovrat Schwab v. The State of Israel, (2006) (Isr.). 88. (TA Regional Labor Court) Ayelet Shmuelov v. Moshe Poones Cleaning and Maintenance Services Ltd., (2006) (Isr.). ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 94 COMP. LABOR LAW & POL’Y JOURNAL 2/1/2023 12:00 PM [Vol. 43:71 the legal minimum wage. As explained in the judgment, the bank turned a blind eye to the fact that the amount it paid could not support legal employment of a cleaner. The bank’s actions contributed quite directly to the violations by the contractor. For this reason it was decided that the bank shall bear responsibility as a joint employer for the payment of minimum wages and other basic employment standards. The main idea of Shmuelov was to prevent “losing contracts” (or “deficit tenders”),89 at least in the context of a strong client and a sector known for labor law violations. If the client engages a contractor and pays an amount that is so low that the contractor is bound to lose money unless it opts for violating labor laws, then the client becomes legally liable directly towards the employees. This mechanism works in two ways. First, it creates deterrence for clients, and incentives to take steps to ensure compliance by their contractors. The new legal rule aims to change the incentive structure of calculating clients, from an interest in finding the cheapest contractor (to make maximum profit), to an interest in ensuring that contractors respect all their legal obligations towards employees. Second, this mechanism can confront a legal structure that cultivates moral hypocrisy. Previously, the client distanced itself from the workers, and felt that because they are not legally its employees, it had no moral duty towards them. The contractor, in turn, felt morally justified in paying sub-minimum wages, because it was forced to do so by the client. By placing liability on clients for losing contracts, we force them to “look in the mirror” and recognize their moral responsibility for the exploitation of workers. And if as a result, they pay the contractor more than the minimum necessary, the latter cannot deceive itself anymore that it has no moral responsibility for labor law violations. The lesson, we believe, is that courts can take the compliance consideration into account when deciding who should bear employer responsibilities. This should not be done lightly; some justification for placing liability on a third party must be present. But often such justifications indeed exist. In several judgments, the National Labor Court noted that a client could be held responsible directly towards the contractor’s employees, even without a “losing contract.”90 That is, even when paying the contractor sufficiently, the client (or “lead company”) can be held liable as a guarantor of sorts, or “residual employer,” when the contractor violates labor laws. These judgments referred to situations in which there was close proximity between the client and the employees. Indeed, they can be criticized for not recognizing the client as the “true employer” for all intents and purposes. 89. On this phenomenon see also Weil, supra note 3, at 137. 90. (National Labor Court) Dovrat Schwab v. The State of Israel, (2006) (Isr.); (National Labor Court) Israel Electricity Company v. Liah Naidorf, (2018) (Isr.); (National Labor Court) Minrav Construction and Engineering Ltd. v. David Moshe Buskila, (2020) (Isr.). ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 2022] 2/1/2023 12:00 PM IMPROVING COMPLIANCE WITH LABOR LAWS 95 However, they do open the door for placing liability on third parties who are not otherwise the employer, for example in sectors that are especially prone to violations and abuse and where the work is performed on the client’s premises. In the years following the Shmuelov judgment, the Israeli government and legislature took up the same cause of fighting “losing contracts.” Their response was limited to the cleaning and security sectors, notorious for systematic, ongoing violations of labor laws. In 2007 the Accountant General issued guidelines for public sector employers, which included a calculation of the cost of employment in the cleaning and security sectors (minimum wages plus other benefits, based on labor legislation and relevant extension orders), with a prohibition on public sector employers from buying services for less than those amounts. The same idea was later included in the Act to Improve the Enforcement of Labor Laws of 2011, this time broadened to cover private-sector employers as well, and crafted as civil liability of the client towards the contractor’s employees in case of violation (i.e. in case of a “losing contract”). 91 Some other legal systems have similar laws, sometimes broader in application (not only for cleaning and security workers92, and not limiting the third party liability for losing contracts93). Alongside direct client liability towards the workers, “losing contracts” can be prevented through procurement law. Under Israeli regulations enacted in 2009, in tenders by State agencies and some other public entities, proposals that will lead to violations of labor laws should be rejected. The European Union Directive on Public Procurement includes various provisions to the same effect. 94 The role of the courts during the procurement process is exemplified in the case of Ashkelon Municipality, decided by the Israeli Supreme Court in 2010.95 The municipality, which was not bound by the above-mentioned regulations, published a tender for cleaning services. Following the Accountant General guidelines “to the letter,” the tender 91. §28, Act to Improve the Enforcement of Labor Laws of 2011 (Isr.). Note that the Minister is yet to issue regulations for this provision (detailing the cost of employment), putting its validity into question. 92. See §2810, California Labor Code (2004). 93. See §558A ff, Australian Fair Work Act 2009, as added by the Fair Work Amendment (Protecting Vulnerable Workers) Act 2017. For an analysis see Tess Hardy, Big Brands, Big Responsibilities: An Examination of Franchisor Accountability for Employment Contraventions in the United States, Canada, and Australia, 40 COMP. LAB. L. & POL’Y J. 285 (2019). 94. See Directive 2014/24/EU On Public Procurement (2014), art. 18(2), 57, 71. For discussions as well as analyses of the CJEU decisions on this subject, see Catherine Barnard, To Boldly Go: Social Clauses in Public Procurement, 46 INDUS. L.J. 208 (2017); Lisa Rodgers, The Operation of Labour Law as the Exception: The Case of Public Procutement, in SMART PUBLIC PROCUREMENT AND LABOUR STANDARDS: PUSHING THE DISCUSSION AFTER REGIOPOST 141 (Albert Sanchez-Graells ed., 2018); Anja Weisbrock, Socially Responsible Public Procurement, in SUSTAINABLE PUBLIC PROCUREMENT UNDER EU LAW: NEW PERSPECTIVES ON THE STATE AS STAKEHOLDER 75 (Beate Sjåfjell & Anja Wiesbrock eds., 2015). 95. AAM 9241/09 Sheleg Lavan v. Ashkelon Municipality, (2010) (Isr.). ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 96 COMP. LABOR LAW & POL’Y JOURNAL 2/1/2023 12:00 PM [Vol. 43:71 included a minimum price of the exact amount needed to cover the minimum wages and benefits of the workers. The contract was then awarded to the cheapest proposal, from a contractor who offered the minimum price allowed by the tender. In a judgment explicitly acknowledging the role of procurement law in securing workers’ rights, and noting the forward-looking importance of the tender in ensuring compliance ex-ante, the Court annulled the decision of the municipality. It pointed out that the contract does not leave any room for profit for the contractor, so it is bound to lead to labor law violations. In effect, the judgment closed a gap in the Accountant General guidelines, using general administrative law doctrines that subject public entities to standards of reasonableness.96 VI. REMEDIES So far we discussed ways in which courts can contribute to compliance through substantive legal rules. We now turn to briefly consider how the same goal can be advanced through decisions concerning remedies. We bring two examples: the calculation of back payments owed in cases of misclassification, and the use of punitive damages. Misclassification of employees as independent contractors is a prevalent problem in Israel, as in many other countries. Sometimes the misclassified worker is paid as a contractor the same as comparable employees, or even less. But there are also cases in which workers are paid more—some “extra” payment compared to the wage of comparable employees—on the assumption that they are contractors. When this assumption is later corrected by a labor court, and employers have to make back payments to cover various employee rights and benefits, they usually argue that the “extra” contractor payment should be deducted from what they need to pay. In other words, they argue that the “extra” payment received by the worker should be offset against the labor law payments they owe. On recent years this question has led to a fierce debate among the judges of the National Labor Court. There is no relevant legislation that addresses this issue, so the Court had to come up with a solution on its own. Some of the judges have argued that an employer should not be required to pay twice; therefore, any payments already made, even under erroneous assumptions about the employee status, should be taken into account, and the 96. The contractor argued that this was not a “losing contract,” because it relied on the reasonable assumption that a significant number of employees will leave before the end of the first year, thereby saving some costs which accrue under Israeli law only after completing a year in employment. The Court refused to allow this assumption. For another recent decision in which a tender was annulled for not sufficiently taking employees’ rights into account, see (DC TA) Nursing Companies Association v. Ministry of Defense, (2019) (Isr.). ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 2022] 2/1/2023 12:00 PM IMPROVING COMPLIANCE WITH LABOR LAWS 97 offset should be allowed.97 In contrast, some other judges pointed out that such an offset significantly reduces the desired disincentives for misclassification. Rather than focusing on a just solution for the specific parties, this approach puts more emphasis on creating deterrence to prevent misclassification in the future. If employers bear the risks of paying the “extra” contractor payment on top of the labor law back payments, they will be much more careful before choosing to misclassify employees. 98 In judgments favoring this approach, judges explicitly noted the importance of securing compliance with non-waivable labor rights, including nonpecuniary rights (such as maximum hours, or vacation rights), during the time of the relationship. A decision to prevent any offset creates strong deterrence against de facto waivers of these rights. Moreover, because of the perception that employers who misclassify could face double payments, this approach has a punitive aspect, and as such, could help raise awareness among employers that such practices are considered by society illegitimate. Recently, an enlarged forum of nine judges has reached a compromise of sorts, allowing an offset to a limited extent but adding compensation for nonpecuniary damages—designed also to ensure deterrence against misclassifications.99 Another tool in the arsenal of remedies, which seems highly relevant for increasing compliance, is punitive damages. Israeli labor legislation includes several provisions which allow courts to award punitive damages in specific contexts. Such a provision is included in the Wage Protection Act of 1958, instructing the court to award punitive damages when wages are not fully paid in time, unless there are specific reasons not to do so.100 Moreover, new labor laws enacted over the last two decades—adding some specific protections concerning notice of employment terms, whistleblower protections, freedom of association and the right to seat during work—all include punitive damages provisions.101 In recent years the labor courts are using these provisions to award increasingly substantial sums.102 Alongside 97. (National Labor Court) Anat Amir v. The Israeli News Corporation, (2015) (Isr.). This approach does recognize an exception: the offset should not be allowed when employee status should have been obvious and the employer acted in bad faith. 98. (National Labor Court) Rafi Roffe v. Mirkan Insurance Agency Ltd., (2011) (Isr.). This approach as well allows an exception: when the employee acted in bad faith, the offset should be allowed. 99. (National Labor Court) Gavriel Kota v. Municipality of Ra’anana (2021) (Isr.). 100. §17-18, Wage Protection Act of 1958 (Isr.). 101. §5, Notice for Employees and Prospective Employees (Terms of Employment and Screening of Job Candidates) Act 2002 (as amended 2011) (Isr.); §3, Protection of Employees (Exposing Violations and Corruption) Act 1997, (as amended 2008) (Isr.); §4, The Right to Seating and Suitable Conditions at Work Act 2007 (Isr.); §33, Collective Agreements Act 1957 (11) (as amended 2009) (Isr.). 102. This is most notable in cases of freedom of association; see, e.g., (National Labor Court) Café Noir Ltd v. The Histadrut, (2019) (Isr.); (National Labor Court) Bait Balev Ltd. v. The Histadrut, (2019) (Isr.). In the context of the right to seat at work see (National Labor Court) Dizingof Club Ltd. v. Yaakov Zoeeli, (2011) (Isr.). ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 98 COMP. LABOR LAW & POL’Y JOURNAL 2/1/2023 12:00 PM [Vol. 43:71 the specific legislative provisions, the Israeli Supreme Court maintained that under Israel’s common law system, courts are authorized to award punitive damages as part of their general “innate” power. The Court added, however, that the use of punitive damages in civil litigation will be saved for exceptional circumstances, usually when dealing with intentional damage, severe wrongdoing, and profound breaches of constitutional rights. 103 Relying on this general power, the National Labor Court awarded punitive damages in sex discrimination cases. 104 Such awards require judges to creatively develop the law in an effort to induce higher levels of compliance. Punitive damages are considered an unusual remedy in civil litigation.105 Compensation in tort and contract law is generally designed to restitute damages, while punitive damages aim to secure deterrence and to some extent inflict punishment (i.e., retribution). There is also the obvious difficulty with giving the specific plaintiff a “windfall” of an award much beyond his real damages. Notwithstanding these challenges, economic analysis of law justifies punitive damages when (and only when) regular compensation schemes lead to under-deterrence. 106 It can also be used to “send a message” to employers, employees and the legal community about the importance of some rights, thereby contributing to a new societal perception about the severity of violating them.107 How would we know that there is under-deterrence, which can justify punitive damages? The idea of punitive damages as societal damages, developed by Catherine Sharkey,108 seems highly relevant to the labor law context. As she explained, punitive damages contribute to a wrongdoer’s better internalization of the societal cost of his wrong conduct. The legal lens should be expanded to catch not only the specific wrongdoer and the specific victim who are part of the litigation (individual harm), but also other parts of society (societal harm). Societal damages, therefore, serve first to acknowledge the hidden costs of those ignored victims, and second, to ensure a full internalization of the wrongdoer’s costs by also taking into account indirect victims that are otherwise ignored.109 103. CA 140/00 Michael Etinger Estate v. Jerusalem Old City Restoration Company Ltd, (2004) (Isr.); Berta Marziano Estate v. Yehoram Zinger, (2005) (Isr.); Amit Mentin Estate v. The Palestinian Authority, (2017) (Isr.). 104. (National Labor Court) Sharon Plotkin v. Eisenberg Brothers Ltd., (1997) (Isr.). 105. On the roots of punitive damages, see Joseph A. Seiner, The Failure of Punitive Damages in Employment Discrimination Cases: A Call for Change, 50 WM & MARY L. REV. 735, 743 (2008). 106. See A. Mitchell Polinsky & Steven Shavell, Punitive Damages: An Economic Analysis, 111 HARV. L. REV. 869 (1998). 107. Cass R. Sunstein et al., Assessing Punitive Damages (with Notes on Cognition and Valuation in Law, 107 YALE L.J. 2071, 2086 (1998). 108. Catherine M. Sharkey, Punitive Damages as Societal Damages, 113 YALE L. J. 347 (2003). 109. Id., at 365. See also Catherine M. Sharkey, Economic Analysis of Punitive Damages—Theory, Empirics and Doctrine, in RESEARCH HANDBOOK ON THE ECONOMICS OF TORTS 486, 488 (Jennifer Arlen ed., 2013). ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 2022] 2/1/2023 12:00 PM IMPROVING COMPLIANCE WITH LABOR LAWS 99 Sharkey proposed a system in which societal damages will reach other victims rather than the individual plaintiff. Exploring this option is beyond the scope of this article. We do, however, rely on the part of her analysis which identifies situations that lead to under-deterrence. Two of those situations seem especially relevant for our purposes: barriers to bring a case to court, and patterns of repeated wrongdoing. 110 Employees face wellknown barriers for self-enforcing their rights, notably lack of information, lack of resources to sue, and fear of reprisals.111 This can justify awarding punitive damages in order to ensure the full internalization of the social costs caused by the infringing employer. It is also quite common for employers who violate the labor rights of one employee to act similarly towards other employees. In such cases, knowing that many employees will not sue, the employer fails to fully internalize the societal costs of its conduct, again leading to under-deterrence.112 Punitive damages will not only raise the cost of the violation to employers, but will also raise the incentive for employees and their lawyers to detect the violation and sue. To be sure, there are additional methods to address the problems of barriers for self-enforcement and repeat offenders, notably unionization, criminal sanctions, and class action suits. But in many cases none of them is available or realistic. Punitive damages add an additional tool to this arsenal, which is useful especially when other techniques fail. VII. JUDICIAL LEGITIMACY AND DISCRETION In the previous four sections we have shown that courts play a role in pushing employers to comply with labor laws. Sometimes this is perhaps a by-product, but often, we believe, judges make choices designed to improve compliance, whether explicitly (as in the case of third-party liability and the case of remedies) or implicitly (as in the case of raising awareness and the choice between rules and standards). Given the crisis of enforcement, we believe that such efforts are justified: courts should join the other branches of government in the fight against non-compliance. One might question whether courts have the legitimacy to take considerations of increasing compliance into account in their judgments. We believe they do, for several reasons. First, we do not argue that courts should be granted with new discretion or new powers. We only suggest that when using discretion already granted to them, judges should consider the issue of increasing future compliance. Perhaps there is more room to do this in 110. Sharkey, supra note 108, at 366, 389. 111. See Davidov, supra note 1, at 226. 112. See C. Michael Mitchell & John C. Murray, The Changing Workplaces Review: An Agenda for Workplace Rights, Report for the Ontario Government 129 (May 2017), at https://files.ontario.ca/books/mol_changing_workplace_report_eng_2_0.pdf. ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 100 COMP. LABOR LAW & POL’Y JOURNAL 2/1/2023 12:00 PM [Vol. 43:71 common law systems, although judges obviously have broad discretionary powers in civil law systems as well. In any case, we do not maintain that considerations of future compliance must influence every decision. We simply argue that this is a consideration that should be taken into account where appropriate (and the examples discussed in the previous sections can help to illuminate this potential). Second, laws—whether made by legislation or by case-law—are designed to achieve social goals, and it is the courts’ role to give them effect. It seems only logical that courts will be able to use their powers in a way that advances compliance with these laws. Relatedly, it is important to remember that a decision not to act is also a decision. If a judgment does not include any component designed to ensure compliance, this is also a statement to employers, and a choice with an impact in the world. Consider, for example, the idea of using pressure on third parties (up to a point) to improve compliance by contractors. If a court refuses to place any responsibility on clients, this is not “neutral”; it is a decision with negative implications on future compliance. Third, for those who feel uncomfortable with a specific employer paying the price of an effort to improve future compliance—for example by way of punitive damages—note that we do not propose anything new here. Punitive damages exist in many legal systems and are designed to change the behavior of third parties as well (unrelated to the specific case). We hasten to add that we do not dispute the difficulty with placing a disproportional cost on a specific employer, but we believe this is something that judges can take into account as part of their discretion. There are also methods that can minimize the problem, such as deciding on a new law but delaying its entry into force, or announcing intentions (for example to impose high punitive damages) but applying them only gradually. Fourth, if the legislature is unhappy with a judge-made law, for any reason—including, unhappy with the way that judges chose to rely on considerations of improving compliance—the legislature can always change the law using legislation. Therefore, even if there are judicial mistakes creating unjustified costs, once they are identified by the legislature such costs can be stopped. VIII. CONCLUSION Labor standards, as Harry Arthurs summarized most eloquently, “ultimately succeed or fail on the issue of compliance. Widespread noncompliance destroys the rights of workers, destabilizes the labor market, creates disincentives for law-abiding employers who are undercut by law- ARTICLE 6_DAVIDOV&ESHET.DOCX (DO NOT DELETE) 2022] 2/1/2023 12:00 PM IMPROVING COMPLIANCE WITH LABOR LAWS 101 breaking competitors, and weakens public respect for the law.”113 In recent years compliance with labor laws attracted significant scholarly attention. The current article has added to this literature in two ways. First, by bringing to the fore three bodies of literature which explain, from different perspectives, what makes people comply (or not comply) with the law. We started with the intuitive idea of deterrence, and showed its importance but also limitations; we then added the studies in behavioral ethics, which explain why “good” people sometimes fail to obey the law; and concluded with the expressive theory which suggests that people can obey the law for reasons of cooperation and information. We made some brief comments about the relevance of these bodies of literature to the specific context of compliance with labor laws. The second and main contribution of this article was to focus on the role of courts in improving compliance, a topic hitherto neglected in the literature. Relying on examples from Israeli case-law, we discussed several ways in which courts can contribute (and have contributed) to compliance. First, they can raise awareness to the law by using procedural techniques to make it a “high profile” case and by making substantive decisions that “send a message” to employers. Second, courts should strive to find an optimal balance between rules and standards, and we argued that deriving a clear-cut rule from a standard can sometimes be useful for improving compliance. Third, we discussed the possibility of placing liability on third parties (such as clients) as indirect pressure to increase compliance. Finally, we gave examples for remedies that can create deterrence and also have an expressive function. This list is not exhaustive; surely there are additional ways in which judicial decisions can have an impact on future compliance. We have shown in this article that as a matter of practice, courts can be seen as taking steps to improve compliance with labor laws. We believe that such steps are warranted and justified, given the importance of these laws and the frequency of non-compliance, and have tried to dispel concerns about legitimacy. We hope to raise awareness to these possibilities, and bring to the attention of judges the relevant theories of compliance and how they play out in the specific context. We believe that these considerations should be taken into account in a more explicit and systematic way. This would eventually help the cause of making labor laws effective, rather than the illusory promise which too often they are. 113. Harry Arthurs, Fairness at Work: Federal Labour Standards for the 21st Century, Report for the Federal Government of Canada 53 (2006), at https://digitalcommons.osgoode.yorku.ca/cgi/viewcontent.cgi?article=1166&context=reports.