Something Else: Specific Relief for Breach of Human Rights Terms in Supply Chain Agreements

Something Else: Specific Relief for Breach of Human Rights Terms in Supply Chain Agreements

68 Am. U. L. Rev. 1751 (2019).

* Harold E. Kohn Professor of Law, Temple University-Beasley School of Law. Thanks to Susan Maslow, David Snyder, and William Woodward for comments on an earlier draft.  Erin McKevitt and Katherine Cordry provided valuable research assistance.  In the interest of full disclosure, I have participated in the work of the American Bar Association discussed in this Article; nevertheless, the views expressed here are mine, alone, as are any errors or omissions.  © 2019, Jonathan C. Lipson, all rights reserved.

Contracts may include clauses designed to address human rights related conduct, such as the treatment of a party’s employees, which are difficult to enforce.  Model Contract Clauses recently promulgated by the ABA Business Law Section Working Group to Draft Human Rights Protections in International Supply Contractsseek to address these challenges through innovative specific performance terms.  ThisArticle assesses these clauses and observes that, while they are unlikely to be enforceable in any ordinary sense, they nevertheless have value because they can induce more constructive settlements ex post and more thoughtful bargaining, ex ante.

“The jurisdiction of a court of equity, in decreeing a specific performance of an agreement, is a peculiar jurisdiction, in the exercise of which that forum becomes, of its own inherent strength, a court of conscience.”1Hudson v. King, 49 Tenn. (2 Heisk.) 560, 568 (1870).

“The duty to keep a contract at common law means a prediction that you must pay damages if you do not keep it,—and nothing else.”2Oliver Wendell Holmes,The Path of the Law, 10 Harv. L. Rev. 457, 462 (1897).

Introduction

Supply chain agreements (SCAs) are central to global trade; they are connective tissue in complex webs of relationships that have produced profound economic growth.  While SCAs present many interesting and important legal issues, I wish to focus on a new and potentially significant aspect of their use:  implementing human rights reforms.  In particular, I will consider model contract terms (“Model Contract Clauses,” or “MCCs”) to protect the human rights of those employed by firms that are parties to SCAs, developed by the ABA Business Law Section Working Group to Draft Human Rights Protections in International Supply Contracts (the “Working Group”).3See generally David V. Snyder & Susan A. Maslow, Human Rights Protections in International Supply Chains—Protecting Workers and Managing Company Risk, 73 Bus. Law. 1093, 1096 (2018) [hereinafter Model Contract Clauses].

From an academic perspective, the MCCs reflect a turning point in the use of contract.  We typically think of contract as a mechanism to promote joint economic gains through shared promissory commitments.  As Alan Schwartz and Robert Scott put it, “[t]he typical contract is dyadic:  it has two parties.”4Alan Schwartz & Robert E. Scott, Third-Party Beneficiaries and Contractual Networks, 7 J. Legal Analysis 325, 325 (2015).  Its law“should facilitate the efforts of contracting parties to maximize the joint gains (the ‘contractual surplus’) from transactions . . . [and] nothing else.”5Alan Schwartz & Robert E. Scott, Contract Theory and the Limits of Contract Law, 113 Yale L.J. 541, 544 (2003) (emphasis added).  Using contract to address problems of human rights abuses created by economic globalization is a deliberate effort to have contracts do “something else.”

Whether, and to what extent, mechanisms like the MCCsare able to do “something else”—e.g., to achieve their human-rights goals—depends in part on the efficacy of the remedies available.  But traditional contract remedies—money damages or specific relief—fit poorly in this context.

Take a simple example. Assume that U.S. buyer “B” is in a supply chain agreement with supplier/seller “S” in a foreign nation. The SCA makes the enforceable choice of the law of the U.S. state of B’s main operations.  Assume further that the SCA incorporates a human rights (“HR”) term6Like the Working Group, I focus mostly on labor-related human rights (HR) terms.  I note, however, that the MCCs could apply to substantive terms for a range of social, economic, and environmental problems. under which S agrees not to use “child, slave, prisoner or any other form of forced or involuntary labor.”7This language derives from a term developed by General Motors.  See Anti-Slavery and Human Trafficking Statement, Gen. Motors Co., https://www.gm.com/ content/dam/company/archive/docs/legal/General_Motors_Company_Anti_Slavery_And_Human_Trafficking_Statement.pdf (last visited June 1, 2019) (emphasis omitted) (“Seller further represents that neither it nor any of its subcontractors, vendors, agents or other associated third parties will utilize child, slave, prisoner or any other form of forced or involuntary labor, or engage in abusive employment or corrupt business practices, in the supply of goods or provision of services under this Contract.” (emphasis omitted)).  B learns from a story in the Wall Street Journal that despite supplying conforming goods at the agreed price, S has allegedly breached this HR term by using forced prison labor. B’s stock price plummets.

How, if it got to that, should a court craft a remedy?  Presumptively, we start with expectation damages—the difference in value between what S promised and what B got.  Because the goods themselves conformed to the specifications of the contract, the only material defect involved the manner of their production.  From a purely commercial perspective, and with respect to those goods, it might be hard to show expectation damages.

We might think that the breach caused consequential damage to B’s reputation.  But quantifying reputational harm flowing from breach of contract is difficult for courts.8See Redgrave v. Bos. Symphony Orchestra, Inc., 855 F.2d 888, 892 (1st Cir. 1988) (en banc) (stating that calculating reputational damages may be speculative and unascertainable).  If B’s shares were publicly traded, we might think that a drop in price on the news is a plausible proxy for reputational damage.  But many factors contribute to the rise and fall of share prices and, in any case, B is not likely to own most of the shares in question.  Its shareholders will, but they are not parties to the SCA, even though they may be the ones ultimately harmed by B’s now-tainted reputation.

These and similar problems may have led the drafters of the MCCs to consider specific performance and injunctive relief (collectively “specific relief”) to be attractive alternatives (or supplements) to money damages.9Model Contract Clauses, supra note 3, ¶ 5.3.  Thus, the MCCs provide that remedies for breach of HR terms may include:  (1) an injunction enforcing the HR term; (2) B’s right to require S to remove problematic employees or “Representatives” of S; and (3) B’s right to require S to terminate contracts with sub-suppliers.10Id. §§ 1.1, 5.3, at 1097, 1102–03.  See infra Part I.B (defining “Representatives”).

But specific relief is also problematic, especially under U.S. law, which tends to view it as “an extraordinary remedy never awarded as of right.”11Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7, 24 (2008) (referring to preliminary injunctions).  Courts make similar statements about specific performance. See Paloukos v. Intermountain Chevrolet Co., 588 P.2d 939, 944 (Idaho 1978) (stating that specific performance is a remedy only used when other remedies are not adequate).

I focus only on U.S. law.  I note that civil law systems may be more receptive to specific relief than common law systems.  Henrik Lando & Caspar Rose, On the Enforcement of Specific Performance in Civil Law Countries, 24 Int’l Rev. L. & Econ. 473, 474 nn.6–7 (2004). They do, however, add a layer of complexity beyond the scope of this Article.  Courts seem no more inclined to grant it for claims of reputational harm than to award money damages.12Bennington Foods LLC v. St. Croix Renaissance, Grp., 528 F.3d 176, 178–79 (3d Cir. 2008) (“[A] plaintiff in a breach of contract case cannot convert monetary harm into irreparable harm simply by claiming that the breach of contract has prevented it from performing contracts with others and that this subsequent failure to perform will harm the plaintiff’s reputation.”).  While a court can back up such an order with contempt powers,13Gene R. Shreve, Federal Injunctions and the Public Interest, 51 Geo. Wash. L. Rev. 382, 389 (1983) (discussing the “specter of civil and criminal contempt” for disobeying a federal injunction). it is unclear how effective those powers are, especially against defendants who may be in foreign nations.  In any case, if a U.S. court orders S to take action specifically contemplated by the contract—e.g., to fire an employee or change its employment practices—how will it know whether S has complied?  If the harm to B is reputational, isn’t B’s better solution to end the contract now and buy from someone else, meaning partial rescission?  Uncertainty about these remedial paths may lead multinational corporations (MNCs) to wonder why they should bother with the MCCs on specific relief at all.

This Article tries to address these questions by assessing the efficacy of the Model Contract Clauses’ specific relief terms in two steps.  First, I note that the MCCs are unlikely to be enforceable in any ordinary sense because they may induce two anxieties that have long troubled judges asked to impose specific relief:  the anxiety of indenture and the anxiety of entanglement.  The former, indenture, is most problematic when courts are asked to order a person to do (or to refrain from doing) something, as and to the extent the “something” starts to look like involuntary servitude.14See infra Section II.B.  The latter, entanglement, reflects judicial concerns about interfering with matters that are personal to the parties or that are political, such as value choice or the “public interest.”15Political entanglement may be especially worrisome.  See Owen M. Fiss, Foreword:  The Forms of Justice, 93 Harv. L. Rev. 1, 54 (1979) (“[T]he issue is not shrewdness, not the capacity of judges to devise strategies for dealing with these limiting forces, but rather the very need to devise these strategies and what the perception of this need does to their sense of independence. Judges realize that practical success vitally depends on the preferences, the will, of the body politic.”).  I discuss the anxiety of entanglement as it pertains to claims of religious liberty in Jonathan C. Lipson, On Balance:  Religious Liberty and Third-Party Harms, 84 Minn. L. Rev. 589, 593 (2000).  These anxieties of indenture and entanglement will constrain and channel the judicial imagination in crafting specific relief for the breach of HR terms.

These limitations might lead parties to eschew HR terms in SCAs. My second point is that they should not. Rather, parties and courts should see that granting specific relief for breach of HR terms shares important aspirations with the injunctive relief courts grant in institutional reform litigation, lawsuits in which beneficiaries of public programs or agencies have sued to reform and restructure those agencies.16See infra Section III.A.  Although not a perfect analogy, courts in those cases have developed experimental techniques such as monitoring and the use of neutral experts as forms of specific relief that could be adapted to cases involving breach of HR terms.  Thus, parties seeking to use HR terms in supply chain agreements (and perhaps future iterations of the Model Contract Clauses) might include some of those techniques, including the appointment of monitors, trainers, and other quality assurance mechanisms.

At the same time, specific relief terms often improve performance under relational contracts such as SCAs,17See infra Section III.B. and this should be true of HR terms within them.  Negotiating such terms may impose transaction costs, but the terms should have the transaction benefits of compelling parties to think more deliberately about whether they can make human rights related promises, and, if they do so, how they can be performed.  In the event of breach, the threat of specific relief tends to have a clarifying effect that promotes more efficient negotiated resolution, especially among parties who wish to preserve their commercial relationship.18See infra Section III.B.

The bottom line, then, is that the Model Contract Clauses’ specific relief provisions are innovative ways to solve difficult social problems created or exacerbated by globalization.  While they may be problematic to enforce as currently written, they can have important, practical value that parties who are serious about addressing human rights problems should consider.

This Article proceeds in three parts.  Part I provides brief background on the Model Contract Clauses, and why globalization has created demand for them.  Part II addresses some doctrinal challenges that the MCCs may present under U.S. law, and how those doctrinal challenges mask anxieties of indenture and entanglement. Part III acknowledges these challenges but suggests that such terms nevertheless have value as mechanisms to structure consent decrees in the event of litigation and, in any case, may productively channel the parties’ relationship regarding human rights and similar issues.

Background—Governance Gaps in Global Trade & the Model Contract Clauses

The benefits of globalization are well understood—chiefly in the form of increased wealth for more people. But, globalization has costs as well, articulated as problems of social, economic, and environmental responsibility (“SEER”).  These costs are often attributed to “governance gaps,” in John Ruggie’s words:

The root cause of the business and human rights predicament today lies in the governance gaps created by globalization—between the scope and impact of economic forces and actors, and the capacity of societies to manage their adverse consequences.  These governance gaps provide the permissive environment for wrongful acts by companies of all kinds without adequate sanctioning or reparation. How to narrow and ultimately bridge the gaps in relation to human rights is our fundamental challenge.19John Ruggie, Special Rep. of the Sec’y Gen., Promotion and Protection of All Human Rights, Civil, Political, Economic, Social and Cultural Rights, Including the Right to Development ¶ 3, U.N. Doc. A/HRC/8/5 (Apr. 7, 2008).

SEER problems have often been viewed as problems that government must solve.  But of course, there is no global government, and conventional, self-seeking strategies may create long-term externalities.  A great deal of global economic activity is undertaken by MNCs, and such corporations may for many purposes be “stateless.”20Hany H. Makhlouf, Multinational Corporations and Nation-States:  Managing Shared- and Conflicts of Interest, 4 J. Soc. & Admin. Sci. 139, 141 (2017) (noting that a multinational corporation resembles “a federation of different companies or semi-autonomous subsidiaries that are, at least, partially owned and controlled by a central unit”).  Among other things, this means that they will have an enhanced capacity to evade or arbitrage regulation intended to solve SEER problems.  Unless MNCs choose to cooperate, it will be difficult to achieve many SEER goals.

Some MNCs have chosen to cooperate in addressing at least some SEER problems through statements of corporate social responsibility (CSR).21Corporate social responsibility reflects “the economic, legal, ethical, and discretionary/philanthropic expectations society places on organizations at a given point in time.”  Ann K. Buchholtz & Archie B. Carroll, Business and Society:  Ethics, Sustainability, and Stakeholder Management 34 (9th ed. 2015).  Corporate law scholars have invested considerable energy debating the merits of CSR. Optimists such as Merrick Dodd,22E. Merrick Dodd, Jr., For Whom are Corporate Managers Trustees?, 45 Harv. L. Rev. 1145, 1160–61 (1932) (arguing in support of corporate social responsibility). David Million,23David Millon, Two Models of Corporate Social Responsibility, 46 Wake Forest L. Rev. 523, 524 (2011) (developing models of corporate social responsibility). and Lyman Johnson,24Lyman Johnson, Law and Legal Theory in the History of Corporate Responsibility:  Corporate Personhood, 35 Seattle U. L. Rev. 1135, 1140 (2012) (“[C]orporate law today has little to say about a subject of great societal significance—corporate responsibility.”). have argued that corporations can and should act for the benefit of all who are affected by the corporation’s activities, not merely their shareholders.  Corporations have a duty, in other words, to be socially responsible.  Others, such as Milton Friedman,25Milton Friedman, The Social Responsibility of Business is to Increase its Profits, N.Y. Times Mag., Sept. 13, 1970, at 2–3 (describing corporate social responsibility as “taxation without representation” and equating the use of social values, other than profit maximization, in business decision making as “socialist”). Jonathan Macey,26Jonathan R. Macey, Corporate Social Responsibility:  A Law & Economics Perspective, 17 Chap. L. Rev. 331, 332 (2014) (arguing that shareholder interests are presumptively primary and exclusive of interests of other stakeholders). and Adolph Berle see CSR as naïve—little more than a “pious wish that something nice will come out of it all,” as Berle famously put it.27A. A. Berle, Jr., For Whom Corporate Managers Are Trustees:  A Note, 45 Harv. L. Rev. 1365, 1368 (1932).  In the law and economics literature, the base cite is typically Michael C. Jensen & William H. Meckling, Theory of the Firm:  Managerial Behavior, Agency Costs and Ownership Structure, 3 J. Fin. Econ. 305 (1976), which developed the theory of shareholder primacy as a function of agency costs.

Efforts to use corporate law to induce MNCs to internalize the costs of globalization have had little legal success.28See, e.g., Order Denying Def.’s Joint Mot. to Dismiss at 1–2, Nat’l Consumers League v. Wal-Mart Stores, Inc., No. 2015 CA 007731 B, 2016 WL 4080541 (D.C. Super. Ct. July 22, 2016) (finding that statements of corporate policy are merely “aspirational in nature”).  The Working Group that drafted the Model Contract Clauses appears to understand this, and views contract terms as a way to help fill the governance gaps identified by Ruggie.29See Ruggie, supra note 19 and accompanying text; see also Model Contract Clauses, supra note 3, at 1094 (“The foundational idea behind the present work is to move the commitments that companies require, whatever they may be, from corporate policy statements to the actual contract documents where those policies may have greater impact.”).

A.  The Model Contract Clauses

The Model Contract Clauses seek to put teeth into CSR by providing contract processes that would implement such human rights standards as the ABA Model Business and Supplier Principles on Labor Trafficking and Child Labor (the “Principles”).30For a discussion of the Principles, see, for example, E. Christopher Johnson, Jr., Business Lawyers Are in a Unique Position to Help Their Clients Identify Supply-Chain Risks Involving Labor Trafficking and Child Labor, 70 Bus. Law. 1083 (2015); see also ABA Model Business and Supplier Policies on Labor Trafficking and Child Labor, ABA (Jan. 9, 2019), http://www.americanbar.org/groups/business_law/initiatives_awards/child_labor.  “The hope,” the Working Group explained, “is that following the steps outlined in the ABA Model Principles will help eradicate labor trafficking and child labor from supply chains, making a difference to real people—their health, safety, and freedom, and in some cases, saving lives.”31Model Contract Clauses, supra note 3, at 1094.

Although motivated by concerns about human rights abuses in the global supply chain, the MCCs do not specify the substance of those rights. Rather, they leave those to the parties in specific cases, to be incorporated in the contract in what the Working Group refers to as a “Schedule P.”32“Whatever moral and legal commitments companies want to require can be accommodated in what this Working Group entitles Schedule P, which the model clauses incorporate, but the actual content of Schedule P is beyond the scope of this Working Group.”  Id. at 1096.  Schedule P was chosen as a reference to the principles that might supply or animate the substantive standards in question.  Id. at 1096 n.13.  In principle, the Model Contract Clauses could provide redress for violation of any substantive terms that seek to solve SEER problems.

B.  Model Contract Clauses as a Contract for Procedure

While the Model Contract Clauses do not specify substantive SEER goals, they do offer a series of mechanisms for increasing the likelihood that parties achieve these goals by making them central to the SCA itself.  Sellers under the MCCs would represent and warrant that they—and their “Representatives” (defined below)—have complied with Schedule P33Id. at 1097 (“Supplier and its subcontractors and [to Supplier’s [best] knowledge] the [shareholders/partners, officers, directors, employees, and] agents of Supplier and all intermediaries, subcontractors, consultants and any other person providing staffing for Goods or services required by this Agreement [on behalf of Supplier] (collectively, the ‘Representatives’) are in compliance with Schedule P.”) (alterations in original). and that compliance with Schedule P is a “material” term of the agreement.34Id. at 1099 (“It is a material term of this Agreement that Supplier and Representatives shall strictly comply with Schedule P.”).  Noncompliance with Schedule P would, under the MCCs, “substantially impair[] the value of the Goods and this Agreement to Buyer” and would constitute “a fundamental breach of the entire Agreement.”35Id.

There are two gating issues with these provisions.  First, it is not clear whether or how a supplier (e.g., S, in the example from the Introduction) can assure that itssuppliers and Representatives have complied with Schedule P.  The Model Contract Clauses define a “Representative” as potentially including “shareholders/partners, officers, directors, employees, and agents of Supplier and all intermediaries, subcontractors, consultants and any other person providing staffing for Goods or services required by this Agreement on behalf of Supplier.”36Id. at 1097 (alterations omitted).  The goal appears to be to assure that the norms and values reflected in Schedule P are driven through S to all in some material relationship with S, including its suppliers.

This, in principle, is a laudable goal. Supply chains can be quite long. But one has to ask whether it is realistic to expect suppliers to know what their suppliers’ suppliers are up to? In a footnote, the Model Contract Clauses observe that an unqualified version of this definition “supports [the] Buyer’s goals to allocate the risk of undiscovered issues to Supplier and contractually encourage Supplier to gather accurate information about its subcontractors.”37Id. at 1097 n.15.  Whether, or to what extent, it is realistic to think that a tier one supplier can do so as to its suppliers, its sub-suppliers, and so on, is another matter.  Moreover, given the breadth of the definition, one might wonder whether it includes those in legally protected relationships with S, such as its attorneys. Is S’s counsel a “Representative” under the Model Contract Clauses?  The broad definition suggests they might be, but they may be surprised to learn that.

Second, the Model Contract Clauses treat the breach of the HR term (Schedule P) as a breach of terms that would otherwise involve the goods, themselves.  This is probably a necessary, but perhaps problematic, strategy because, as discussed below, specific remedies may not involve the goods at all.  Rather, the breach of Schedule P probably involves the process of their production.

Nevertheless, the Model Contract Clauses appear to assume that they would be governed by the Uniform Commercial Code (UCC) or the United Nations Convention on Contracts for the International Sale of Goods (CISG).38Id. at 1096 (“The text proposed assumes that buyers are located in the United States and that the applicable law is the Uniform Commercial Code (the ‘U.C.C.’) or the United Nations Convention on Contracts for the International Sale of Goods (the ‘CISG,’ a treaty to which the United States is a party).”); see also United Nations Convention on Contracts for the International Sale of Goods, art. 45–52, 61–65, Apr. 11, 1980, 1489 U.N.T.S. 59, 67–69, 70–71 [hereinafter CISG].  But those laws were drafted not with a view toward affecting the behavior and norms of parties to contracts involving the sales of goods, but instead to provide default rules regarding the sale of the goods, themselves, e.g., as to their quality, quantity, price, and legal title.39See, e.g., U.C.C. § 2-101 (Am. Law Inst. & Unif. Law Comm’n 2017) (“The arrangement of the present Article is in terms of contract for sale and the various steps of its performance.”); Id. § 2-102 (“Unless the context otherwise requires, this Article applies to transactions in goods.”).  Because the substance of Schedule P may have nothing to do with the characteristics of the goods, and because specific relief for the breach of Schedule P may not, either, it is not difficult to imagine disagreement over whether either of those bodies of law apply to these aspects of the Model Contract Clauses.

C.   Specific Relief under the Model Contract Clauses

I point these issues out not to quibble with the Model Contract Clauses, but instead to frame the challenges underlying what are perhaps their most innovative features—their forms of specific relief.  As one might expect, most of the remedial terms in the Model Contract Clauses involve money.  The MCCs provide for, among other things, liquidated damages,40See Model Contract Clauses, supra note 3, at 1103 (“[T]he parties have therefore agreed to liquidated damages in an amount calculated as follows: _______________.”). damage due to lost sales (profit) and/or to reputational harm, and indemnification for B’s losses due to S’s breach of Schedule P.41Id. at 1103–04.  The MCCs candidly recognize, however, that the “[b]uyer may face challenges with respect to proving damages.  This is common in claims for breach of contract, but Buyer may have special challenges with respect to the impact on its brand that results from violations of human rights policies.”42Id. at 1103 n.42.  Moreover, collecting money judgments against a foreign promisor may be difficult.

The Model Contract Clauses contain three provisions articulating specific relief:

(1)  First, they would have the Supplier agree that, in the event of a breach of Schedule P, Buyer may “obtain an injunction with respect to Supplier’s noncompliance with Schedule P, and the parties agree that noncompliance with Schedule P causes Buyer great and irreparable harm for which Buyer has no adequate remedy at law and that the public interest would be served by injunctive and other equitable relief.”43Id. at 1102.  (“Model Contract Clauses Injunction”).

(2)  Second, the Supplier would agree that the Buyer may “require Supplier to remove an employee or employees and/or other Representatives.”44Id. at 1103.  (“Model Contract Clauses Employee/Representative Termination”).

(3)  Third, the Supplier would agree that the Buyer may “require Supplier to terminate a subcontract.” (“Model Contract Clauses Subcontract Termination” and, with the Model Contract Clauses Employee/Representative Termination,” the “Termination Powers”).45Id.

While it is difficult to know, empirically, whether these sorts of terms are truly novel in supply chain agreements, it is not difficult to imagine that they will be challenging for courts to enforce by way of injunction or specific performance, which I explain in the next Part.

II.   Specific Performance and Injunctive Relief

The Model Contract Clauses appear to contemplate both specific performance and injunctive relief, although they articulate only the latter directly.  We tend to think that specific performance and injunctive relief are the same, legal substitutes for one another.46See Engemoen v. Rea, 26 F.2d 576, 578 (8th Cir. 1928) (“An injunction decree enjoining the breach of a contract is in effect a decree for its specific performance, and the principles which govern the granting of both remedies are generally the same.”); Raymond T.Nimmer & Jeff C. Dodd, Modern Licensing Law § 11:36 (2018) (“One can readily see that an injunction may well be the reverse side of the coin of specific performance:  forbidding a party from taking an action may well be all the performance that a party expects.”).  And, while they overlap significantly in their necessary elements and their functions, they are not coterminous.  Instead, they appear to be doctrinally distinct in two respects that might matter to courts asked to enforce them in this context and to parties that consider using them.

A.   Doctrine and its Limits

Contract doctrine conventionally presumes that specific performance is an “exceptional remedy,”47Melvin A. Eisenberg, Actual and Virtual Specific Performance, the Theory of Efficient Breach, and the Indifference Principle in Contract Law, 93 Calif. L. Rev. 975, 1016 (2005). to be granted only if damages would not be adequate.48See, e.g., Restatement (Second) of Contracts §§ 357, 359–60, 366–67 (Am. Law Inst. 1979) (stating it is within the court’s discretion to decide if damages would not be adequate).  U.S. courts will grant specific performance in cases where the aggrieved party can show that goods are “unique” or in “other proper circumstances.”49Andrea G. Nadel, Annotation, Specific Performance of Sale of Goods Under UCC § 2-716, 26 A.L.R. 4th § 2[a] (1983); see also Jason S. Kirwan, Appraising a Presumption:  A Modern Look at the Doctrine of Specific Performance in Real Estate Contracts, 47 Wm. & Mary L. Rev. 697, 701 (2005) (“It is also important to note that the general trend in U.S. contract law favors increased latitude for trial courts to grant specific performance as a redress for breach.”).  The underlying logic of specific performance derives from the view that substitutionary remedies—money damages—will usually be adequate.  As Farnsworth put it, “[o]ur system of contract remedies is not directed at the compulsion of promisors to prevent breach; it is aimed, instead, at relief to promisees to redress breach.”503 E. Allan Farnsworth, Contracts § 12.1 (2d ed. 1994) (emphasis omitted).

The principal point of doctrinal overlap between specific performance, as a contract remedy, and injunctive relief is “adequacy”:  Both remedies are said to be limited to circumstances where money damages would be “inadequate.”  In the case of injunctions, the element is referred to as “irreparable injury,”51To obtain an injunction, a plaintiff must in theory show:

[1] Success (or, in the case of a preliminary injunction, likely success) on the merits;

[2] Irreparable injury in the absence of injunctive relief;

[3] The balance of equities tips in plaintiff’s favor; and

[4] An injunction is “in the public interest.”

Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7, 20 (2008).but it has substantially the same import: a court must do something, the claimant says, because money itself will not be a sufficient substitute.

Yet, “adequacy” is a problematic standard. The classic case for specific performance of a contract term—conveying title to real property—is not necessarily one in which money damages are inadequate.  Deriving a dollar value for the property in question may be difficult, but it is not obviously more speculative than deriving damages in cases where the aggrieved party is a new business that lacks a track record of profitability.52See Chung v. Kaonohi Ctr. Co., 618 P.2d 283, 291 (Haw. 1980) (finding that recovery should not be denied simply because the business is new so long as the plaintiff can show future profits with reasonable certainty).  Conversely, the case in which money damages would be highly inadequate—breach of a personal services contract—is one in which a court would almost certainly not order specific performance.53Eisenberg, supra note 47, at 1017, 1036 (“[A] decree ordering an employee to specifically perform an employment contract would seem too much like involuntary servitude or peonage.”); see also Douglas Laycock, The Death of the Irreparable Injury Rule 169 (1991) (“The reason for [the rule that employment contracts will not be specifically enforced against employees] is a substantive law commitment to free labor.  Despite the vast social distance between chattel slavery and specific performance of contracts with professional athletes and entertainers, similar policies apply to both . . . .  An order to work on pain of contempt produces servitude that is involuntary when the services are performed.”).

Not surprisingly, analysts view the adequacy standard skeptically.  Melvin Eisenberg considered “inadequacy” as a predicate to obtaining specific performance to be a “virtually dead”rule.54Compare Eisenberg, supra note 47, at 1017 (“The adequacy rule, as a rule that simply bars the gate, is virtually dead and probably should be.  The commentators also generally agree that specific performance is granted more freely today than traditional doctrine suggests.”), with Douglas Laycock, The Death of the Irreparable Injury Rule, 103 Harv. L. Rev. 687, 703 (1990) (“Damages are inadequate if plaintiff cannot use them to replace the specific thing he has lost.  This is by far the most important rule in determining the doctrinal relationship among remedies.”).  Those who study equitable remedies say much the same.  Douglas Laycock, for example, has argued that the “irreparable injury” standard:

does not describe what the cases do, and it cannot account for the results.  Injunctionsare routine, and damages are never adequate unless the court wants them to be.  Courts can freely turn to the precedents granting injunctionsor the precedents denying injunctions, depending on whether they want to hold the legal remedy adequate or inadequate. Whether they want to hold the legal remedy adequate depends on whether they have some other reason to deny the equitable remedy, and it is these other reasons that drive the decisions.55Laycock, supra note 54, at 692.

B.   “Other Reasons” for and Against Specific Relief

This, then, invites the question:  What are those other reasons?  One way to answer the question is to map four inchoate, but important, factors underlying the substantive or policy disputes at issue.  These are whether the specific relief sought is (1) affirmative or (2) negative; and whether it involves (3) persons or (4) property.

Subject Matter/Relief Negative Affirmative
Property56This functional analysis would comport with the trend toward recognizing that equitable power—the power to enjoin or order performance—reaches both persons and property. Historically, it was limited to property.  Shreve, supra note 13, at 386 (“Until recently, equity’s protections were confined to property as opposed to personal rights.”). Attachment/lien;57I note that it is not clear whether an attachment or similar judicial lien would be a legal or equitable remedy.  Great-W. Life & Annuity Ins. Co. v. Knudson, 534 U.S. 204, 214 (2002) (characterizing a claim for restitution under ERISA as “restitution that . . . is not equitable—the imposition of a constructive trust or equitable lien on particular property—but legal—the imposition of personal liability for the benefits that they conferred upon respondents”). enjoin infringement58Pappan Enters., Inc. v. Hardee’s Food Sys., Inc., 143 F.3d 800, 806 (3d Cir. 1998) (discussing “irreparable harm [that] Hardee’s and MRO would continue to suffer as a result of Pappan’s non-consensual use of the ROY ROGERS [trade]marks”). Transfer title
Persons Enjoin harmful conduct (e.g., impermissible competition) Order personal services59See, e.g., Sampson v. Murray, 415 U.S. 61, 83 (1974) (relying on “the traditional unwillingness of courts of equity to enforce contracts for personal service either at the behest of the employer or of the employee”);Arthur v. Oakes, 63 F. 310, 318 (7th Cir. 1894) (holding that a decree of specific performance for breach of an employment contract would be involuntary servitude); The Case of Mary Clark, 1 Blackf. 122, 123, 126 (Ind. 1821) (denying specific enforcement of a promise to serve employer for twenty years); Laycock, supra note 54, at 745 (stating that the court will not grant specific performance for personal services contracts).

Plotting specific relief in this way helps to reveal, and perhaps helps to explain, two anxieties that seem to constrain judges who are asked to grant specific relief:  the anxiety of indenture and the anxiety of entanglement.  The anxiety of indenture reflects the deeply-held belief that specific performance is inappropriate in personal services contracts on free-labor grounds. Having abolished slavery, any effort to order a person to perform a promised service may look like peonage—a shameful condition abetted by long-discredited judges.60The Peonage Cases addressed the problem that arose after Reconstruction of southern courts convicting African Americans of a “crime” derived from breach of a debt or other contract and indenturing them to the creditor as the remedy.  See, e.g.Pollock v. Williams, 322 U.S. 4, 24 (1944); Taylor v. Georgia, 315 U.S. 25, 29 (1942); Bailey v. Alabama, 219 U.S. 219, 245 (1911).  Although Congress had forbidden peonage in the wake of the Civil War (Act of March 2, 1867, 14 Stat. 546 (1867) (codified as amended at 42 U.S.C. § 1994 (2012), 18 U.S.C. § 1581 (2012))), southern courts continued to defy it.  SeeUnited States v. Gaskin, 320 U.S. 527, 529–30 (1944) (reversing a lower court holding that one could not be convicted under the Anti-Peonage Act for merely arresting someone with the intent of returning them to peonage).  See generally William Wirt Howe, The Peonage Cases, 4 Colum. L. Rev. 279 (1904) (discussing lower court decisions under federal Anti-Peonage Act).  Some judges have suggested that involuntary servitude had to be akin to slavery in the specific sense of subjecting workers to physical or legal coercion.  See James Gray Pope, Contract, Race, and Freedom of Labor in the Constitutional Law of “Involuntary Servitude,” 119 Yale L.J. 1474, 1514 (2010).  As Laycock has observed, “the difficulty of coercing close personal relationships is a powerful reason for denying specific relief, even if damage remedies are inadequate.”61Laycock, supra note 54, at 748.

The anxiety of entanglement stems from fears that judges cannot plausibly intervene in the day-to-day work and lives of parties.  As Schwartz has noted, “specific performance [is] an unattractive remedy in cases in which the promisor’s performance is complex, because the promisor is more likely to render a defective performance when that performance is coerced, and the defectiveness of complex performances is sometimes difficult to establish in court.”62Alan Schwartz, The Case for Specific Performance, 89 Yale L.J. 271, 277 (1979).  Perhaps the most articulate expression of this anxiety arose in connection with federal efforts to enjoin steelworker strikes in the 1950s:

No doubt a District Court is normally charged with the duty of independently shaping the details of a decree when sitting in equity in controversies that involve simple and relatively few factors—factors, that is, far less in number, less complicated and less interrelated than in the case before us.  But a court is not qualified to devise schemes for the conduct of an industry so as to assure the securing of necessary defense materials.  It is not competent to sit in judgment on the existing distribution of factors in the conduct of an integrated industry to ascertain whether it can be segmented with a view to its reorganization for the supply exclusively, or even primarily, of government-needed materials.

United Steelworkers of Am. v. United States, 361 U.S. 39, 50–51 (1959) (Frankfurter & Harlan, JJ., concurring) (per curiam). Even the most Herculean judge cannot get off the bench and run the company.63Cf. Ronald Dworkin, Law’s Empire 239 (1986) (discussing the approach of a hypothetical “Herculean” judge to judicial action).  Interestingly, although there is ample literature on Dworkin’s famous enthusiasm for the wisdom and power of judges, see, for example, W. Bradley Wendel, Sally Yates, Ronald Dworkin, and the Best View of the Law, 115 Mich. L. Rev. Online 78, 79–80 (2017), none of it seems to focus on the mechanism by which judges would most directly exercise that power, specific relief.  Perhaps this reflects Dworkin’s interest in “serious” matters of constitutional interpretation, as distinct from seemingly more trivial matters of contract enforcement. See generally Ronald Dworkin, Taking Rights Seriously (1977) (analyzing different rights from a philosophy of law approach).  I leave that project to another day.

These anxieties are most severe in the lower right-hand quadrant of the figure above; and mildest in the upper left.  In some cases, the specific relief envisioned by the MCCs may take judges closer to the lower right-hand zone than they would like.  In that case, they would be unlikely to enforce such terms. The balance of this Part explains why; the following Part offers some alternative ways of thinking about the work that the MCCs, or other specific relief provisions, can perform in this context.

C.   Doctrinal Distinctions

The anxieties of indenture and entanglement have no strong doctrinal locus, even as they may matter to judges asked to enforce (or to refrain from enforcing) provisions like the MCCs.  Doctrinally, courts at least talk as if “adequacy” is the central question for both specific performance and injunctions, though the adequacy in question is that of the remedy, and not of the court’s capacity to effectuate it. Still, there are doctrinal distinctions between injunction and specific performance that should matter in the context of HR terms, in part because judges may use them to attempt to manage these anxieties.  This Part looks at some of the salient doctrinal distinctions in this context.

1.      “Other proper circumstances”—specific relief and supply chain agreements

Specific performance would appear to be easier to obtain than an injunction, as and to the extent it does not induce concerns about judicial compulsion of services or entanglement in personal relations.64See Nemer Jeep-Eagle, Inc. v. Jeep-Eagle Sales Corp., 992 F.2d 430, 433 (2d Cir. 1993) (citing Travellers Int’l AG v. Trans World Airlines, Inc., 722 F. Supp. 1087, 1104 (S.D.N.Y. 1989)) (observing that test for specific performance is “more flexible” than the test for a preliminary injunction).  See generally Shreve, supra note 13, at 387 (“The injunction does occupy an inferior hierarchical position in the law of remedies.  In an action for damages, plaintiff need not establish that the harm he suffers is substantial.  To obtain an injunction, however, plaintiff must show that the equitable remedy is necessary to avert substantial harm.”).  To obtain specific performance under the UCC, for example, the plaintiff need only show under section 2-716 that goods are “unique” or that there are “other proper circumstances.”65U.C.C. § 2-716(1) (Am. Law Inst. & Unif. Law Comm’n 2018).  Because we can assume that the goods in most cases involving the MCCs will not be unique, the questions are whether breach of an HR term constitutes “other proper circumstances” and how high a hurdle that sets for a plaintiff-buyer.

It is safe to say that the drafters of section 2-716 probably did not have HR terms in mind when they used the words “other proper circumstances.”  Section 2-716 “seeks to further a more liberal attitude than some courts have shown in connection with the specific performance of contracts of sale.”66Id. § 2-716 cmt. 1.  But, the main example given is the inability to cover—meaning that the focus is on the goods and not the conditions under which they were produced.67Id. § 2-716 cmt. 2 (“[I]nability to cover is strong evidence of ‘other proper circumstances.’”).  Where no market alternative is “adequate,” specific performance would be “other proper circumstances.”

Thus, while courts may order sellers to perform under long-term supply contracts, they may also worry about their real power to do so.  Courts are not necessarily unwilling to enforce such terms merely because they involve repeat performance.68See Laclede Gas Co. v. Amoco Oil Co., 522 F.2d 33, 39–40 (8th Cir. 1975) (finding that a long-term contract makes it difficult to estimate damages and specific performance offers an efficient way to attain the ends of justice); Copylease Corp. of Am. v. Memorex Corp., 408 F. Supp. 758, 759 (S.D.N.Y. 1976) (recognizing that unique goods are an exception to the general rule that limits specific performance as a remedy).  But, as Nathan Oman has observed, “specific performance represents a greater intrusion into personal freedom than do money damages,”69Nathan B. Oman, The Failure of Economic Interpretations of theLaw of Contract Damages, 64 Wash. & Lee L. Rev. 829, 869 (2007).  and this certainly seems to be true in long-term supply contracts. Although there appear to be no published opinions involving injunctions for breach of HR terms, case law suggests that courts are uncomfortable granting injunctions when the court faces the prospect of long-term engagement with the defendant.70Soinco v. NKAP provides an illustrative example outside the UCC.  See Soinco v. NKAP, Zürich Chamber of Commerce Arbitration Award ZHK 273/95 ¶ 173 (May 31, 1996), http://cisgw3.law.pace.edu/cases/960531s1.html (providing an example of a long-term supply contract).  Here, an international arbitration set in Zurich determined that there was no basis for claims of specific performance in Russian law, and so the buyer’s claim for specific performance was denied.  The buyers requested specific performance of several contracts of sale; if ordered, the seller, a Russian aluminum producer, would have had to produce and deliver aluminum to the buyers for between eight and ten years.  Id. ¶ 349.  The tribunal found that it “fail[ed] to see how specific performance could be an appropriate remedy for buyers in this case” because “[the buyers] can hardly expect to be able, under the New York Convention or otherwise, to have an award enforced in Russia . . . for the next eight or ten years . . . .”  Id. The tribunal denied the buyer’s claim for specific performance because ordering performance would require constant supervision, and thus confirmed that difficulty of enforcement would influence a court’s decision to order specific performance under the CISG.  See id. ¶¶ 348–49; see also Nayiri Boghossian, A Comparative Study of Specific Performance Provisions in the United Nations Convention on Contracts for the International Sale of Goods (CISG) 65 (Nov. 1999) (unpublished Master of Laws thesis, McGill University), https://www.collectionscanad a.gc.ca/obj/s4/f2/dsk2/ftp03/MQ64262.pdf.

Conversely, if the problem could easily be cashed out, any intrusion by the court into a party’s operations may be problematic. In Healthcare Corp. of America v. Data Rx Management, Inc.,71No. 2:12-2910 (KM), 2013 WL 1314736 (D.N.J. Mar. 28, 2013). for example, a health-care claims management company allegedly held funds that should have been remitted to a pharmacy benefits provider pursuant to a supply chain agreement.72See id. at *1.  Although the management company did not dispute that it held the funds that it owed to the benefits provider,73Id. at *4 (“HCA does not meaningfully dispute that it is holding a significant amount of money due to pharmacies who are not parties to this action.  HCA concedes that Data Rx has sent invoices for the Pharmacy Charges and Processing Fee, and it has conceded at oral argument that it does not challenge the amount of these invoices.  HCA admits that it has been paid by Middlesex County for past prescriptions filled by the pharmacies.  HCA has not turned these funds over to Data Rx.”). the court declined to grant the injunction:

Data Rx has alleged that the trust on which its supply chain is based will erode if a preliminary injunctionis not granted.  It has not, however, (1) presented any sworn statements from pharmacies stating that such a chilling effect is taking place, or (2) differentiated this supply chain from any other where suppliers take on financial risk by dispensing products in advance of receiving payment.  In short, Data Rx has not submitted proof of loss of reputation or demonstrated that its pharmacy network is different from “other types of commerce in such a way that normal breach of contract remedies could not provide a remedy.”74Id. at *5 (quoting Bennington Foods LLC v. St. Croix Renaissance, Grp., LLP, 528 F.3d 176, 179 (3d Cir. 2008)).

The court in Data Rx recognized that the benefits provider, the party seeking the injunction, may well have had a claim, but denied the injunction because liability could easily be established at a later point in the litigation, and it apparently faced no collection risk in the event it won.75E.g., id. at *1 (“That is not to say that Data Rx’s claims could not prevail on summary judgment or at trial, but at present, they do not furnish a basis forinjunctive relief.”); id. at *5 (noting that counsel represented that following an anticipated sale of the counterclaim-defendant, “the combined company is projected to have significantly greater revenues than HCA alone”).  The Data Rx court’s conclusion was also motivated in part by the Supreme Court’s holding in Grupo Mexicano that a preliminary injunction should not ordinarily issue for a breach of contract claim, even where the defendant may be preferentially transferring assets in anticipation of the litigation.  Id. at *4 (citing Grupo Mexicano de Desarrollo, S.A. v. All. Bond Fund, Inc., 527 U.S. 308, 332–33 (1999)).  Moreover, the request for the preliminary injunction was made as part of a counterclaim to a complaint alleging breach of one or more of “numerous” contracts that governed the parties’ relationships.76Id. at *2 (observing that the parties’ “relationships are governed by numerous contracts”).  It is not difficult to imagine that the lower court here did not wish to become entangled in the complex relationships of the parties, especially when it did not need to do so.

This is not to suggest that buyers could never specifically enforce HR terms on a theory that they present “other proper circumstances.” I mean only that courts may worry about their ability to do so without becoming impermissibly entangled in the affairs of the parties, or required to indenture persons over whom they have little, if any, direct control.

Consider a variation on the example from the Introduction.  Assume that S employs a manager “M” who knows that S has agreed not to use prison labor under its SCA with B.  Yet, M also knows that he cannot provide the promised goods at the promised price without using below-market labor, and so he leases inmates from a nearby prison, in breach of the SCA.  Assume further that B learns of this and demands that S terminate M pursuant to a provision akin to the Termination Powers in the Model Contract Clauses.  S refuses to terminate M.  Then what? Would a U.S. court order a promisor (S) to terminate an employee who, in S’s judgment, has not committed an offense that creates grounds for termination?  How could S do so without avoiding the risk of collateral litigation for wrongful termination or interference from M?77If M were a “key” employee, the problem would be even more complicated.  For a discussion on the enforceability of “key man” clauses generally, see In re Orion Pictures Corp., 4 F.3d 1095 (2d Cir. 1993).

It is hard to know what to do with any of the standard doctrinal tests on these facts.  It is easy to see that money damages might be inadequate, but that hardly tells a court how to balance the equities or otherwise to achieve some rough justice. Indeed, it is not hard to imagine a judge being very reluctant to provide affirmative relief because enforcing these terms might enmesh a judge in complex personnel matters.  While B may believe that the normative force of the underlying dispute constitutes “other proper circumstances,” it is easy to see why a court may disagree.  Anxieties of indenture and entanglement are likely to be high, here.

2.      The “public interest” and specific relief

A second, and perhaps more interesting, doctrinal difference between specific performance and injunctive relief appears to involve the role of the “public interest.” There is no formal requirement that courts entertaining a request for specific performance of a contract term consider the “public interest.”  By contrast, when courts consider whether to grant (or deny) injunctive relief (especially preliminary injunctions), they may do so.78See Restatement (Second) Contracts § 357 cmt. c (Am. Law Inst.1981) (“In granting [injunctive] relief, as well as in denying it, a court may take into consideration the public interest.”).

At one level, the difference is not surprising, since we typically assume that breach of contract problems are largely “private” in nature.79See, e.g., Orion Pictures Corp., 4 F.3d at 1097 (demonstrating the breach of contract dispute between Orion and Showtime, private parties); see also Daniel Markovitz, Theories of the Common Law of Contracts, Stan. Encyclopedia of Philosophy (Sept. 11, 2015), https://plato.stanford.edu/entries/contracts-theories.  While a few courts have considered the public’s interest in granting specific performance, they appear to be the exception.80See Pennsylvania R.R. Co. v. City of Louisville, 126 S.W.2d 840, 843 (Ky. 1939) (ordering the railroad to specifically perform the contract to make grade crossing safe in support of the public interest).  If the “public interest” is an element of a party’s request for specific performance, it will be because it is a predicate to the enforceability of the substantive term, such as a covenant not to compete,81DJR Assocs., LLC v. Hammonds, 241 F. Supp. 3d 1208, 1232 (N.D. Ala. 2017) (noting that when considering whether to grant injunction, the court observed that “the enforceability of non-compete covenants touches on the conflicting fundamental public policies of many states”). and not necessarily of the remedy.

Because specific relief is typically viewed as emanating from a court’s equity power, public interest and specific relief have long been linked and debated. The alliance has roots in the Supreme Court’s statement in 1937 that “[c]ourts of equity may, and frequently do, go much farther both to give and withhold relief in furtherance of the public interest than they are accustomed to go when only private interests are involved.”82See Virginian Ry. Co. v. Sys. Fed’n No. 40, 300 U.S. 515, 552 (1937).  A leading treatise observes that the “public interest” element of injunctive relief “is another way of inquiring whether there are policy considerations that bear on whether the order should issue.”8311A Charles Alan Wright et al., Federal Practice and Procedure:  Civil § 2948.4 (3d ed. 2013).  The treatise continues:

Thus, when granting preliminary relief, courts frequently emphasize that the public interest will be furthered by the injunction.  Conversely, preliminary relief will be denied if the court finds that the public interest would be injured were an injunction to be issued . . . .  If the court finds there is no public interest supporting preliminary relief, that conclusion also supports denial of any injunction, even if the public interest would not be harmed by one. Consequently, an evaluation of the public interest should be given considerable weight in determining whether a motion for a preliminary injunction should be granted.

Id.  Even within the sphere of private ordering, observers have argued that courts should account for the moral and policy implications of using specific performance.84See Eisenberg, supra note 47, at 978 (“Actual specific performance should be awarded unless a special moral, policy, or experiential reason suggests otherwise in a given class of cases . . . .”).

This both frames the central issue in the enforcement of HR terms, and invites difficult questions.  One of the key problems with such terms will be their “public” character.  Because they involve or affect persons who are not parties to the contract, they may require an assessment of interests that differ from those in ordinary contract disputes, and that may differ from those of the parties to the contract.  Whose public interest are we talking about—the buyer’s, the seller’s or the seller’s employees?  How is a court supposed to select and, once it does so, how is a court supposed to know what the interest is?  If, as seems likely in a global supply chain agreement, the employees at issue are in a foreign nation, how is a court in the United States (whether a state court or a U.S. district court in diversity) supposed to know what the public interest is there?  Surely, it cannot be the same as the public interest in the United States.

Consider, again, the example from the Introduction. Assume S’s manager, M, has leased prisoners from a local prison run by a relative, and it is common in S’s nation both to use leased prison labor and to tolerate intra-family dealing.  Both are, however, violations of Schedule P in the SCA between S and B.  B is embarrassed when the facts are revealed and seeks an injunction to enforce Schedule P and stop S from this practice.

B may start by pointing to language adapted from the MCCs, that the parties “agree that . . . the public interest would be served by injunctive and other equitable relief.”85See Model Contract Clauses, supra note 3, at 1102.  S may respond by producing an expert witness who explains that local law and custom tolerate S’s conduct, and it is therefore consistent with local public policy.86S may also point out that leasing prison labor remains a feature of the U.S. economy.  See Ifeoma Ajunwa & Angela Onwuachi-Willig, Combating Discrimination Against the Formerly Incarcerated in the Labor Market, 112 Nw. U. L. Rev. 1385, 1385 (2018) (“[P]rivate corporations rely on and profit from low-wage prison labor to argue that the state penal institutions that lease prisoners to such corporations should push for contractual agreements that stipulate that corporations relying on prison labor must revoke policies that bar employing the formerly incarcerated upon their release.”).  B may then produce an expert witness from a watchdog group who testifies to the opposite effect.  S may retort by arguing that an injunction halting the use of prisoners would force S to shut down, leaving all of its workers—some of whom are not leased inmates—out of work, a point to which B and its experts may have no good response.

At this point, it should be apparent that from a U.S. perspective, there are competing public interest goals at stake, and that some of them will be virtually impenetrable to a U.S. judge.  On one hand, there is a strong public interest in assuring the decent treatment of workers and (perhaps) in eschewing the use of prison labor.  On the other hand, courts are wary of taking action that may have the effect of eliminating jobs.87See Boomer v. Atl. Cement Co., 257 N.E.2d 870, 873 (N.Y. 1970) (refusing to enjoin operation of a cement plant that employed over three hundred people); see also Harrisonville v. W.S. Dickey Clay Mfg. Co., 289 U.S. 334, 338 (1933) (refusing to enjoin pollution from municipal sewer plant and stating that “[w]here an important public interest would be prejudiced, the reasons for denying the injunction may be compelling”) (footnotes omitted) (citations omitted).

And, this assumes that a U.S. sense of public interest is appropriate.  But why should we assume that?  Even if, as the hypothetical stipulates, the parties have chosen U.S. law, that does not necessarily mean that they have chosen U.S. public policy.  Indeed, B may have bargained for the HR terms in Schedule P precisely because it understood that public policy in S’s location tolerated conduct considered normatively offensive in the United States.  To ask a court to make these choices risks entangling the court not only in the business and affairs of the parties, but also the public policies of their respective nations.

In the face of these difficulties, a court may be tempted to do nothing.  But this would permit S to shirk its responsibilities, to embarrass B, and to harm those the HR term sought to protect. While B may have other remedies, including a liquidation of damages and indemnification, those will not necessarily have any greater bite than specific relief.  And, while B could walk away, rescission may have little remedial effect. Switching to a new supplier is likely to raise B’s costs and may leave S free to continue to use forced labor.

The bottom line, then, is that specific relief to enforce the remedial provisions of the MCCs presents a serious problem:  courts will find these terms difficult to use to remedy breach of HRterms, regardless of the doctrinal path taken.  Moreover, these paths provide little guidance for judges who may understandably fear forcing individuals to work (or to fire employees on uncertain grounds) or becoming entangled in problems that they feel are beyond their power or expertise.  But this then invites a more basic question:  If courts may not enforce specific relief for breach of HR terms, why use the MCCs at all?

III.    Restructuring and Relationalism

As is often the case with contract terms, there is usually life beyond the black letter law.  In the case of specific relief for breach of HR terms, that life may emanate from an analogy to the work courts do in institutional reform litigation and may reflect the relational effects that specific performance terms can have, independent of their doctrinal value.

A.   Institutional Reform Litigation

Although courts addressing contract disputes may not immediately see it this way, problems presented by the breach of HR terms in supply chain agreements will share certain important characteristics with litigants’ demands in institutional reform litigation (IRL).88The discussion in this subsection draws on Kathleen G. Noonan, Jonathan C. Lipson & William H. Simon, Reforming Institutions:  The Judicial Function in Bankruptcy and Public Law Litigation, 94 Ind. L.J. 491, 493–94 (2019); see also Fiss, supra note 15.  Institutional reform, or “public law” litigation, uses courts to correct the behavior of errant public agencies such as police departments, child-welfare systems, and public schools.89Abram Chayes, The Role of the Judge in Public Law Litigation, 89 Harv. L. Rev. 1281, 1303–04, 1303 n.93 (1976) (noting that public law litigation molds public perception of the legal system).  Structural injunctions in this context address a broad range of the operations of government agency defendants. These decrees are most strongly identified with civil rights claims, but they can be found in other areas.90They have occurred in many complex environmental controversies.  E.g., Charles M. Haar, Mastering Boston Harbor:  Courts, Dolphins, and Imperiled Waters 3 (2005) [hereinafter Haar, Mastering Boston Harbor] (discussing the court-induced clean-up of Boston Harbor).  They also have a long lineage in antitrust law.  See generally Richard A. Epstein, Antitrust Consent Decrees in Theory and Practice:  Why Less is More (2007) (focusing on the history of consent decree ligation in antitrust law).  And they have some resemblance to recent practice in which corporations agree to submit to monitoring and to adopt compliance procedures in return for deferral of prosecution for violation of, for example, the securities laws or the Foreign Corrupt Practices Act.  See, e.g., Brandon L. Garrett, Too Big to Jail:  How Prosecutors Compromise with Corporations 240, 267 (2014).  These litigations are most closely identified with the federal courts, but a substantial number of structural decrees have emerged from state courts, including some of the most ambitious.91E.g., Perez v. Bos. Hous. Auth., 400 N.E.2d 1231, 1234 (Mass. 1980) (approving decree reforming the Boston Housing Authority); Haar, Mastering Boston Harbor, supra note 90, at 3 (noting that the Boston Harbor clean-up was judicially supervised); Charles M.Haar, Suburbs Under Siege:  Race, Space, and Audacious Judges 8–9 (1996) (discussing the decades-long judicial efforts to induce reform of exclusionary zoning practices in New Jersey).

Like all forms of specific relief, IRL asks a court to order a party to do something, or to refrain from doing something, or some of both.92See Chayes, supra note 89, at 1292 (discussing the importance of equitable relief in public law litigation).  Like the Termination Powers under the Model Contract Clauses, the “something” in IRL is likely to involve governance and management of the defendant.  In both IRL and HR terms, the underlying breach probably involves a failure of assurances (whether through positive law or contract) about normatively sensitive matters, such as the treatment of workers or the environment.  In both cases, ordinary remedies, in particular money damages, are unlikely to be satisfactory. While IRL does not involve formal contract, courts in both contexts grapple with failed promises that reflect contested public policy choices and try to craft a set of workable remedies acceptable to the parties and those affected by their undertakings.

In most cases, IRL results not in a simple injunction enforcing a governmental obligation, or enjoining a governmental practice, but instead a consent decree negotiated by the parties against the backdrop of the threat of litigation or a deliberate judicial decision to ignore a material grievance.93See generally Noonan, Lipson & Simon, supra note 88, at 527.  In its earlier days, consent decrees were often a collection of many specific rules. A decree with respect to prison conditions might, for example, specify the minimum space for cells or the temperature of water in the showers.94See Malcolm M. Feeley & Edward L. Rubin, Judicial Policy Making and the Modern State:  How the Courts Reformed America’s Prisons 40–41 (1998) (discussing decrees that “specify many requirements in . . . painstaking or excruciating detail[, including] the wattage of the light bulbs in the cells, the frequency of showers, and the caloric content of meals”).  Modern decrees may still contain some such rules, but they tend to focus on general management functions of self-monitoring, assessment, transparency, and accountability, often derived through agreement of the parties, expert guidance, or both.95See, e.g., Kathleen G. Noonan et al., Legal Accountability in the Service-Based Welfare State:  Lessons from Child Welfare Reform, 34 Law & Soc. Inquiry 523, 524 (2009) (explaining the evolution of decrees in the context of child welfare cases).  An important goal of the decree is a higher-functioning organization that is sufficiently stable to self-correct based on a commitment to, and investment in, ongoing internal quality improvement practices and policies.96See id. at 530–31 (discussing past problems with poor functioning decrees).

The core requirements of framework decrees concern management practices of policymaking, monitoring, and reassessment.  While each of these may have different characteristics in IRL than they would in disputes over the breach of HR terms, it is not hard to imagine courts using these techniques to address the breach of human rights or other terms involving social, economic, or environmental welfare.  In the event of a dispute over breach of HR terms, the Model Contract Clauses on specific relief will facilitate each.

The use of a specific enforcement mechanism in the SCA will, for example, signal a policy commitment by the seller to do (or to refrain from doing) certain activities.  Obviously, breach calls its commitment to that policy into question, but to have articulated it in the first instance provides a basis from which a court and the parties may craft statements of policy that the parties consider workable.

Monitoring may be equally important.  IRL consent decrees often provide that the defendant agrees that a third-party expert will have access to the defendant’s operations and personnel, such as front-line staff, in order to determine how underlying problems arise, whether the proposed remedies in the consent decree are effective and, if not, insights into how to modify the decree to increase its likelihood of success.97See Noonan, Lipson & Simon, supra note 88, at 530.  Because the monitor is appointed by the court, she is likely less susceptible to bias than a party-appointed expert.  Because her appointment would be in a consent decree, a party’s refusal to cooperate with the monitor may be grounds for a finding of contempt.

The Model Contract Clauses do not provide for third-party monitoring, but could easily do so.  There is a growing field of experts who monitor companies for human rights compliance.98See, e.g., ISO 26000—Social Responsibility, Int’l Org. for Standardization, https://www.c2ccertified.org/get-certified/product-certification (last visited June 1, 2019) (providing businesses guidance on how to “operate in a socially responsible way”).  Of course, certification programs do not always work.  See Jonathan Webb, Supply Chain Audits Work for Corporations, but Not the Planet, Says New Report, Forbes (Jan. 16, 2017, 11:40 AM), https://www.forbes.com/sites/jwebb/2017/01/16/supply-chain-audits-work-for-corporations-but-not-the-planet-says -new-report (arguing that compliance certification of corporations is ultimately ineffective because facilities that pass compliance audits may still have substantial human rights violations).  If (or to the extent) parties do not contract for a human-rights monitor, a court that seeks to craft a consent decree may look to certification regimes that have already begun to develop in this context to guide the monitor’s work.  For example, the International Organization for Standardization (ISO) provides its own certification specifically dedicated to social responsibility (ISO 26000).99Id.  Many organizations indicate that they can certify companies and factories in these standards.100SA 8000—Social Accountability Certification, SGS, https://www.sgs.com/en/sustainability/social-sustainability/audit-certification-and-verification/sa-8000-certification-social-accountability (last visited June 1, 2019).  These include but are not limited to:  SGS International Certification services,101See id. (noting that an organization can certify to SA 8000 with an SGS audit). DNV Det Norske Veritas,102Corporate Sustainability in DNV GL, DNV GL, https://www.dnvgl.com/about/sustainability (last visited June 1, 2019) (noting that DNVL GL’s businesses are certified according to ISO 9001 standard and work toward sustainable development goals). BVQI Bureau Veritas Quality International,103See Social Responsibility:  Strengthen Your Company’s Reputation, Bureau Veritas Certification, https://www.bureauveritas.com/home/about-us/our-business/certification/sustainability/social-responsibility (last visited June 1, 2019) (noting that Bureau Veritas is an independent organization that performs social responsibility audits). Intertek Testing Services,104See Certification, Interek, http://www.intertek.com/certification (last visited June 1, 2019) (advertising that Interek offers certification programs for environmental and social accountability compliance). and TUV Rheinland Ltd.105See Certification and Auditing Services for Social Responsibility, TÜV Rheinland, https://www.tuv.com/usa/en/social-responsibility.html (last visited June 1, 2019) (offering services to assess and certify companies’ compliance with social responsibility and fair working conditions).

This monitoring and certification function then makes it possible for the parties to assess and reassess compliance with the goals of the original HR term.  In IRL, for example, the Seattle police settlement prescribes creation of a Community Police Commission with broad representation to review performance data and recommend policy changes.106Settlement Agreement and Stipulated [Proposed] Order of Resolution at ¶¶ 3–12, 119–25, United States v. City of Seattle, No. 12-CV-1282 (W.D. Wash. July 27, 2012).  The DOJ’s Principles for Promoting Police Integrity—which is a starting point for remedial design in many cases—demand continuous review of various data to determine “whether any revisions to training or practices are necessary.”  U.S. Dep’t of Justice, Principles for Promoting Police Integrity 5 (2001). Ecosystem decrees sometimes mandate increased use of “adaptive management.”107See San Joaquin River Restoration Program, Fisheries Management Plan: A Framework for Adaptive Management in the San Joaquin River Restoration Program:  Executive Summary 2 (2010), https://www.usbr.gov/mp/nepa/includes/documentShow.php?Doc_ID=7569.  “Adaptive management is an approach allowing decision makers to take advantage of a variety of strategies and techniques that are adjusted, refined, and/or modified based on an improved understanding of system dynamics.”  Id.  For example, the San Joaquin River Restoration decree altered the defendants’ water management practices to require a more rapid and nuanced response to indications of danger to the fish population.108See id.

The emphasis on provisionality and reassessment leads some courts to mandate explicit experimentation.  The New York police decree required the defendant to undertake a one-year “pilot project” with patrol officers wearing body-worn cameras in one precinct in each of the city’s five boroughs.109Floyd v. City of New York, 959 F. Supp. 2d 668, 685 (S.D.N.Y. 2013).  At the end of the year, the monitor was directed to report on results and deliberate with the parties over whether the practice should be adopted generally.110Id.

In none of these examples do courts in IRL manage or direct the operations of a defendant agency.  Instead, the consent decree provides a framework of procedures by which the parties can identify shared goals, mechanisms for determining progress toward those goals, and options in the event the defendant fails to achieve the agreed goals.  It is not hard to imagine a court asked to specifically enforce HR terms in supply chain agreements taking analogous steps.  It may not actually enforce the terms as written, but instead use them as a basis for crafting performative remedies that are realistic.  The judicial experience in IRL may provide a template for judges in cases involving specific performance of HR terms.

Indeed, future iterations of the Model Contract Clauses might include terms which designate a particular monitor in advance of breach, or who will inspect upon a claim of breach.  It may specify steps the parties agree to take in order to remediate the breach, such as improved training or certification by front-line or other relevant personnel.  It may provide feedback mechanisms under which parties can then assess the efficacy (or not) of the steps thus taken.111Such terms may be similar to those identified by Bernstein in her recent study of Midwestern Original Equipment Manufacturers. See Lisa Bernstein, Beyond Relational Contracts:  Social Capital and Network Governance in Procurement Contracts, 7 J. Legal Analysis 561, 563 (2015).

B.   Relationalism

A central feature of any consent decree must be “consent”—that is, both parties must agree to the resolution embodied in the decree.  It is easy to assume that if a seller breached HR terms in a supply chain agreement, it may have little interest in becoming party to a consent decree.  This is certainly possible, and in that event, the buyer is left to whatever litigation options it has.  If the seller believes that the buyer can make no progress in obtaining any remedy for breach, it is unlikely to come to the table, at which point rescission probably will be the buyer’s best (and perhaps most realistic) remedy.

But this also assumes that the seller is willing to sacrifice its relationship with the buyer, and perhaps other similarly situated buyers. This assumption may apply only to the most strategic and cynical sellers, in which case there are no options other than litigating and walking away.  But supply chain agreements are usually relational contracts, meaning agreements that are deliberately open-ended and flexible, intended not to be one-off affairs but the basis for sharing risks and rewards over a sustained period.112The literature on relational contracting is too large to cite here usefully.  Works relevant to this paper include:  id. at 566 (quoting Master Supply Agreement, Harley Davidson (Nov. 2004), https://www.h-dsn.com/genbus/PublicDocServlet?docID=22&docExt (“Th[is] [supply agreement] . . . describes in general terms how we work together with our suppliers . . . . [It] is not a long-term commitment; rather it is a commitment about how we will operate in the long-term.”); Amy H. Kastely, The Right to Require Performance in International Sales: Towards an International Interpretation of the Vienna Convention, 63 Wash. L. Rev. 607, 615 n.45 (1988) (citing Richard M. Brown, Specific Performance in a Planned Economy, in Papers and Comments Delivered at the Eight Annual Workshop on Commercial and Consumer Law 35, 37 (Jacob S. Ziegel ed., 1980)(observing that multinational supply agreements may be relational contracts)); see also Josh Whitford, New Old Economy:  Networks, Institutions, and the Organizational Transformation of American Manufacturing (2005).

Specific relief plays a special role in thinking about relational contracts.  Its directness and simplicity are generally thought to promote the formation and management of more fruitful relationships than money damages.  While negotiation over specific relief terms may be more costly than no negotiation at all—meaning, the default remedy of damages—the burden of negotiating these terms may be offset by the benefits of clarity.113See, e.g., Charles J. Goetz & Robert E. Scott, The Mitigation Principle:  Toward a General Theory of Contractual Obligation, 69 Va. L. Rev. 967, 1003–04 (1983) (“[A] carefully conditioned right of specific performance not only restrains evasion but also selectively filters the potentially opportunistic cases where the obligor’s cost of performance is substantially greater than the market value of performance.”).  Considering the possibility that HR terms will be specifically enforced is likely to lead parties to take such terms more seriously than if they thought the only remedy available would be rescission or money damages.  Given the uncertainty of the dollar amounts at issue, that remedy may be none at all, in which event the HR term would be wasted.  Specific relief terms may thus increase efficiency as they force parties to negotiate more thoroughly and carefully with one another.114See Schwartz, supra note 62, at 291 (arguing that specific relief terms “would minimize the inefficiencies of under compensation, reduce the need for liquidated damage clauses, minimize strategic behavior, and save the costs of litigating complex damage issues”).

Moreover, specific relief terms may compel parties to resolve their differences more quickly than would recourse to default or liquidated damages.  The breaching promisor threatened with a judicial decree compelling performance is in many cases likely to take that sanction more seriously than an award of money damages, because damages can be difficult to determine or to collect.115Id. at 291–92.  While enforcing specific relief will also not be easy, there is a view that in a broad range of cases, specific relief terms will induce more efficient and effective post-breach adjustment than damages, even without the adaptations suggested by IRL-type consent decrees.116See Thomas S. Ulen, The Efficiency of Specific Performance:  Toward A Unified Theory of Contract Remedies, 83 Mich. L. Rev. 341, 343–44 (1984) (“[U]nder specific performance post-breach adjustments to all contracts will be resolved in a manner most likely to lead to the promise being concluded in favor of the party who puts the highest value on the completed performance and at a lower cost than under any alternative.”); see also Laclede Gas Co. v. Amoco Oil Co., 522 F.2d 33, 35 (8th Cir. 1975) (determining the merits of Laclede’s argument that it is entitled to injunctive relief rather than damages); E. Air Lines v. Gulf Oil Corp., 415 F. Supp. 429, 442–43 (S.D. Fla. 1975) (concluding that specific performance is not an extraordinary remedy).

All of this seems especially true in cases involving HR terms in supply chain agreements among multinational corporations.  While some non-U.S. companies may want or need to shirk contractual promises to honor HR terms, one can imagine that including specific relief for breach of such terms might empower other, more willing promisors to seek to change conditions at home such that they could perform or develop a second-best remedy to preserve a value-creating relationship.  Having agreed to specific relief would not only alter the promisor’s relationship with the promisee, but also, potentially, others whose norms may conflict with the HR term in the SCA.

Crafting more intentionally experimental and incremental terms, along the lines found in institutional reform litigations, may reinforce the relational power of specific relief terms.  A promisor who has agreed to the presence of a pre-breach monitor will know that the promisee has a window into the promisor’s operations that makes defection difficult and probably enhances the integrity of the relationship.  The promisor that has agreed, ex ante, to a monitor (pre- or post-breach) should find it difficult to object to a request to specifically enforce such a provision; a court may view such a mechanism as an incremental step designed to save the parties’ relationship, and on that ground grant the request.

Conclusion

Constructing contract terms to prevent or ameliorate serious social problems, such as the use of forced labor, is an innovative and important development in the use of contract.  It will present challenges for contract doctrine and theory.  This brief Article has only scratched the surface of one facet of the problems presented by such terms, specific relief for their breach.

If nothing else, contract terms to remedy HR abuses are, themselves, unique, and their breach may well be irreparable in money damages. I have argued that parties that take such terms seriously should recognize that they may be doctrinally constrained.  Such constraints should not, however, lead the parties to abandon the terms or specific relief for their breach.  Instead, parties and courts should recognize the more limited, but still tangible, benefits that such terms can provide in this context.