Designing and Enforcing Preliminary Agreements

Article - - Issue 3

Preliminary agreements—variously labeled as memoranda of understanding, letters of intent, term sheets, commitment letters, or agreements in principle—are common in complex business transactions. They document an incomplete set of terms that the parties have agreed upon, while anticipating further negotiation of the remaining provisions. They often create legal obligations, particularly a duty to negotiate in good faith. This duty has been the subject of a substantial number of judicial opinions over the past few decades and yet continues to be regarded as a confusing and unpredictable issue in contract law. Legal scholarship is hamstrung in its analysis of the case law because it has focused on only one purpose for this good faith duty: protecting the parties’ reliance investments in the bargaining process. This Article broadens the analysis by introducing multiple goals that parties may seek in imposing legal obligations on their negotiation process and by shifting the focus to what the courts have identified as a necessary feature of the duty to negotiate in good faith: the expectation of some fidelity to the agreed-upon terms specified in the preliminary agreement. The ease with which the parties may deviate from these terms in their negotiations is the essence of the good faith standard. Once parties have searched for and chosen their respective contracting partner, they need the incentives and flexibility to tailor and optimize the terms of their deal while also efficiently constraining value-claiming behavior and allocating exogenous risks. The recognition of such broader objectives (beyond protection of reliance investments) allows us also to justify how and why courts are willing to enforce the obligation with the more robust remedy of expectation damages instead of the reliance damages that are advocated by prior scholarship. We show that, by choosing whether to agree to a duty to negotiate (in good faith or otherwise) and by selecting the appropriate damages measure, the parties can achieve the desired level of “stickiness” while addressing concerns about the uncertainty of a flexible legal standard such as good faith.


Contract formation in commercial transactions can be an expensive and intricate process involving multiple stages and players, as well as significant investments in expertise and information. In complex asset purchases, intellectual-property licenses, leases, corporate acquisitions, or venture-financing transactions, to name just a few types, it is practically impossible for the parties to execute a fully stipulated and binding contract in a single meeting or over a very short period of time.[1] Negotiations of these transactions are typically sequenced, with a subset of issues being addressed at each stage by numerous agents with different expertise. Midstream in their negotiations, the parties frequently enter into preliminary agreements,[2] often labeled memoranda of understanding, agreements in principle, commitment letters, letters of intent, or term sheets. These agreements document the parties’ agreement-to-date on some or all of the core provisions for the underlying transaction, but they also contemplate, and to some extent regulate, the parties’ remaining negotiation. Preliminary agreements should therefore be thought of as setting ground rules for negotiations, which may include obligations of confidentiality, disclosure, and exclusivity. Significantly, these agreements can also create either express or implied legal duties to negotiate in good faith. The invocation of the good faith (or similar) standard invites the court to police the parties’ negotiations, especially if they break down.

Over the past few decades, a significant volume of litigation has arisen over the enforcement of duties to negotiate in good faith. Although contract law polices bargaining activity that leads to concluded contracts (such as through the doctrines of misrepresentation, coercion, duress, unconscionability, and unjust enrichment), U.S. law does not have a general duty to bargain in good faith that courts would enforce if negotiations break down before an enforceable agreement has been reached.[3] Nevertheless, parties have the ability to invoke such a duty in a preliminary agreement, and the duty to negotiate in good faith in U.S. law is rooted in the parties’ intent. Courts are called on to determine whether the parties have in fact agreed to legally binding good faith obligations, what is required by their good faith standard, and the remedy for breach. Despite many judicial opinions addressing these questions, the law in this area has not been satisfactorily explained or rationalized. In particular, the existing legal scholarship has been hamstrung in its ability to explain the law of preliminary agreements because of its focus on one purpose: encouraging relationship- or deal-specific investment in the contemplated transaction, including due diligence investigation into whether the transaction is profitable and expenditures in contract design that increase its profitability. The current scholarly explanation is that the enforcement of a good faith obligation protects such investment from opportunistic holdup by the noninvesting party in subsequent stages of negotiation.

Although this rationale appears in some judicial opinions, it seems to be only a small part of the explanation of the existence and enforcement of obligations to bargain in good faith. Parties opt into a standard such as good faith because it can address a variety of other actions and goals. Had the protection of reliance investment been the sole objective, more straightforward methods could protect reliance expenditures: for example, a simple promise to reimburse reasonable expenses if negotiations fall through.[4] We also observe that the courts’ reluctance to award expectation damages stems more prominently from concerns about proof than the nature of the protected interest. Many courts have indicated their willingness to compensate for lost expectancy if the plaintiff could establish the terms of the contract that would have been entered into if the defendant had acted in good faith. Moreover, the facts of many judicial cases reflect that parties agree to negotiate in good faith even when anticipated investments after the preliminary agreement are small relative to those that preceded it. The presence of other goals is reflected in several fact patterns and features of the case law and can be explained by analyzing the parties’ motivations to stage their negotiations.

By focusing on the protection of reliance, legal scholarship leaves unexplained a number of important features of preliminary agreements in practice and in the courts. This Article introduces and describes a more robust and complete framework of the motivation for preliminary agreements and the duty to negotiate in good faith. We suggest that these agreements are tools for regulating the negotiation process when parties, following search and diligence activity, are reasonably confident that they have found the partner with whom they wish to deal. In these circumstances, they seek to regulate their negotiations in one or more of the following respects beyond the protection of reliance investments.[5] In addition to encouraging value-creating reliance investment, the parties wish to discourage attempts or investments by either side to improve its bargaining position and thereby claim a larger share of the transaction surplus. It is noteworthy in this respect that express promises to negotiate in good faith or with best efforts are often seen in agreements complementing more explicit exclusivity promises not to shop an offer or negotiate with any other prospective party for a specified duration of time.[6]

While shifts in the distribution of the transaction surplus may result from the deliberate investments of the parties in improving their alternative options, they may also be caused by exogenous changes in markets and surrounding conditions. The preliminary agreement therefore also provides the means by which the parties can efficiently allocate selected risks of changed circumstances that would affect the distribution of surplus from the deal. Although time may be needed to complete the terms of the contract, it may be valuable for the parties to allocate some risks sooner. A relevant example is a preliminary agreement in a loan commitment that pegs the interest rate against changes in the market rate, while the parties negotiate other terms, such as representations, warranties, and covenants. Indeed, the enforcement of preliminary agreements in a series of leading cases in the Second Circuit was motivated by the allocation of interest-rate risk and not the protection of reliance investments.[7] As we explain in this Article, the parties may seek in their preliminary agreement to protect the risk allocation while allowing the parties to create value by continuing their contract negotiations.

The aforementioned goals of promoting efficient specific investment, discouraging value-claiming activity, and efficiently sharing exogenous risks must be balanced against the flexibility needed to reach an optimal agreement on the remaining terms or to respond to previously unforeseen contingencies. This is, of course, an instance of the existential tension in contracts between commitment and flexibility. In this regard, the preliminary agreement is a mechanism by which the parties can fine-tune this balance by giving weight to their agreed-upon terms. We argue that a negotiation duty based on a contextual contract standard—whether good faith or best efforts—can be well suited to address this multifaceted regulation of the negotiation process that parties often wish to invoke. The parties (or the court) would be correspondingly unlikely to invoke such a commitment in a preliminary agreement when reliance, risk allocation, and danger of rent extraction are minimal or absent.[8] Once we recognize the beneficial role that a contextual standard can play under the right circumstances, we can justify the court’s enforcement of such an open-ended promise with a stronger remedial measure than reliance damages, including the award of expectation damages. Drawing from prior work in which we explain how parties can exploit the benefits of standards and avoid the potential costs of litigation and judicial error, we respond to the concerns of lawyers and scholars about the litigation of vague standards.[9]

This Article is organized as follows. Part I briefly summarizes the relevant common law and judicial policy and highlights the neglected relationship of the good faith duty with the other provisions in the preliminary agreement. In this respect, we show that a central feature of the duty is a requirement of fidelity to the terms settled in the preliminary agreement. Part II provides guidelines for the parties’ design of multistage contracting and their use of preliminary agreements to regulate the remainder of their negotiation process by imposing some stickiness on their settled terms. In particular, we describe the common scenario in which, after investing in a search, the parties determine that they wish to transact with each other, invest in the design of optimal transaction terms while deterring value-claiming investments, and allocate certain risks that would affect the surplus division. We tailor our discussion closely to the case law introduced in Part I, particularly from courts in the Second Circuit. Subpart II(D) shows how granting expectation damages for breaching the duty to negotiate in good faith can be consistent with the existing contract law doctrines. Part III describes the benefit of legal standards, such as good faith or best efforts, to support the parties’ objectives in staged negotiations. The use of standards that are costly to verify and susceptible to judicial error is often dismissed in legal scholarship in favor of crisp and clear rules. We have rebutted this belief in other work and draw on that work here to show how in some circumstances, a legal standard may yield superior results to alternative approaches, whether full enforcement, low-level sanctions, or nonenforcement. We summarize our contribution in the conclusion and provide suggestions for future inquiry.

I. The Modern Law of Preliminary Agreements

The law of preliminary agreements can be summarized in three sets of questions that courts face in enforcing them: (1) did the parties have a duty to negotiate in good faith; (2) what behavior constitutes bad faith when negotiations break down; and (3) what is the appropriate judicial remedy? With respect to the first question, the modern law relating to preliminary agreements in most U.S. jurisdictions is reflected in a taxonomy of three types of incomplete agreements. The first type is a mere agreement to agree, which reflects an insufficient meeting of the minds and creates no legally enforceable obligation.[10] In this type of agreement, the parties memorialize their mutual understanding but have either not reached the point of wishing to invoke legal enforcement or not given the court sufficient basis on which to regulate their negotiations.[11] Even without legal consequence, however, these agreements play important roles and may rely on forces other than legal enforcement, such as moral, relational, or reputational sanctions.[12] Under the second type of incomplete agreements, the parties agree upon many, if not most, of the material terms, intend to be bound by them, and anticipate no significant future negotiation (or renegotiation) of terms.[13] This agreement (sometimes called a Type I preliminary agreement) is “fully” enforceable as a contract, unless the parties expressly intend otherwise.[14] In enforcing, the courts will fill any missing terms with legal default or commercially reasonable terms (especially if the transaction is common and the terms are relatively standard)[15] and award the usual contract law measure of expectation damages. Conversely, if a court finds that the parties did not intend to create any legally binding obligations, the court usually allows the parties to walk away from the negotiations without any liability.

Historically, judges, scholars, and commercial parties were dissatisfied with the binary choice between full enforcement of a contract and no enforcement of an agreement to agree, and they sought a middle ground that incorporates a negotiation standard such as good faith.[16] At the same time, American law has rejected a general duty to negotiate in good faith that exists in some civil law jurisdictions.[17] This was based on concerns that it would chill negotiations, create uncertainty, and add undue pressure on parties to conclude their negotiations.[18] Under the law that emerged, U.S. courts enforce obligations in preliminary agreements, including good faith in negotiations, but only if those obligations can be found in the objective intent of the parties to be bound, rather than being triggered simply by the start of negotiations.[19] In contrast to the fully enforceable contracts (Type I), these are sometimes referred to as Type II agreements.[20]

In fact, many commercial parties enter into preliminary agreements in which they assent to a course of conduct during negotiations, often by expressly invoking a standard such as good faith or best (or reasonable) efforts, while settling on some of the core terms of their exchange. Explicit promises to bargain up to express standards are common in many types of transactions, including asset purchases, intellectual-property licenses, leases, bank loans,[21] venture-capital financing,[22] and corporate mergers and acquisitions.[23] To be sure, not all preliminary agreements incorporate promises to negotiate in good faith.[24] In searching for the parties’ intent, most courts are inclined to enforce clear and express language agreeing to or disclaiming a duty of good faith.[25] But, sometimes the language of the parties’ preliminary agreement is ambiguous (or rendered ambiguous by other indicators), and courts may look more broadly at relevant evidence, including: (a) the parties’ intent as revealed in the language of the agreement; (b) the context of the negotiations; (c) the existence of open terms or lack of clarity, particularly essential, material, or major terms; (d) the conduct of the parties, including partial performance; and (e) the customary use of formalities in this type of transaction.[26] Even if the language of the preliminary agreement seems to exclude binding obligations, extrinsic statements (including internal communications within one party’s organization) or conduct may lead the fact finder to find that the parties in fact intended to create a good faith duty. Given the highly fact-dependent nature of the inquiry and the immaturity of the good faith doctrine in this area, commentators have noted that the case law is characterized by inconsistent results, and lawyers warn clients about the consequent traps for the unwary in letters of intent and similar documents.[27]

Although courts are no strangers to standards of good faith or best efforts, commentators have noted that the content of this standard remains difficult to specify and predict.[28] The failure to conclude an agreement itself is clearly not a sufficient indicator of bad faith, even if due to the defendant’s self-interested and strategic bargaining.[29] Good faith invokes an element of motive and a requirement of “honesty in fact”[30] that precludes deliberate, material factual misrepresentations but does not require disclosure of private information.[31] While some courts have interpreted good faith as precluding negotiations with competing bidders, other courts require exclusivity to be explicitly specified in the preliminary agreement.[32] In fact, agreements often combine an express obligation to negotiate in good faith with covenants of confidentiality and exclusivity.[33]

Although negotiations might break down because one party simply walks away, they more commonly fail as a result of one party’s unreasonable (from the perspective of the other party) insistence on or rejection of proposed terms. That party’s insistence or rejection might be motivated by the goal of extracting a much better deal or simply avoiding an unprofitable transaction. Given that the duty to negotiate in good faith is rooted in the parties’ mutual intent, whether that party’s behavior is justified is appropriately based on the parties’ expectations at the time of the preliminary agreement.[34] Notably, courts decline to enforce good faith obligations in the absence of a framework or settled terms in the preliminary agreement.[35] They use the settled terms as a benchmark for determining whether proposals or rejection of proposals are made in good faith. Insisting on terms that vary or do not conform with terms settled in the preliminary agreement may be bad faith.[36] This is especially likely in the face of evidence that such insistence is a pretext for scuttling the deal because of dissatisfaction with the settled terms. Yet, as explained below, it would not be desirable for the duty to be so invasive as to bar any deviation from the settled terms in all circumstances.[37]

The centrality of settled terms is prominent in many opinions. For example, in Teachers Insurance & Annuity Ass’n of America v. Tribune Co.,[38] Judge Leval stated that “parties can bind themselves to a concededly incomplete agreement in the sense that they accept a mutual commitment to negotiate together in good faith in an effort to reach final agreement within the scope that has been settled in the preliminary agreement.”[39] Under this agreement, a party may demand that its counterparty “negotiate the open terms in good faith toward a final contract incorporating the agreed terms.”[40] Judge Leval further stated that:

Each [party] was obligated to seek in good faith to conclude a final agreement within the terms specified in the commitment letter, supplemented by such representations, warranties and other conditions as are customary in such transactions. Teachers would not have been free to walk away from the loan by reason of a subsequent decision that the transaction was not in Teachers’ interest.[41]

Similarly, in Adjustrite Systems, Inc. v. GAB Business Services, Inc.,[42] the Second Circuit Court of Appeals held that a party did not negotiate in good faith when it “insisted on conditions that do not conform to the preliminary writing.”[43] More recently, in Brown v. Cara,[44] the Second Circuit described it as an “obligation to negotiate the open issues in good faith in an attempt to reach the [contractual] objective within the agreed framework.”[45] This is especially true if the condition being insisted upon was important and could have reasonably been included in the preliminary document. The Seventh Circuit Court of Appeals held in A/‌S Apothekernes Laboratorium for Specialpraeparater v. I.M.C. Chemical Group, Inc.[46] that the obligation to negotiate in good faith “can only be determined from the framework the parties have established for themselves in their letter of intent.”[47] In SIGA Technologies, Inc. v. PharmAthene, Inc.,[48] the Delaware Supreme Court held that the prospective licensor, “SIGA[,] failed to negotiate in good faith for a definitive license agreement in accordance with the terms of the [term sheet].”[49]

In sum, preliminary agreements serve to regulate negotiations by limiting the freedom of the parties to depart from terms settled therein, and the courts are willing to enforce this intent in their application of the good faith or best efforts standard. This central feature of the duty to negotiate in good faith has not been examined in legal scholarship and remains underspecified. Parties to a preliminary agreement are in the midst of negotiating their deal, and it is therefore unlikely that they intend to set in stone the terms on which they have agreed at that point. But, their invocation of legally enforceable negotiation duties signals their intent to give considerable weight to these terms. Good faith negotiation requires fidelity to terms specified in the preliminary agreement. It is not absolute, thereby making the preliminary terms sticky rather than fixed. In any given dispute, the court is called upon to assess (a) the degree of deviation and (b) the justification for such deviation. A coherent framework for making these assessments is lacking and we provide such a framework in Part II by examining the parties’ objectives in preliminary agreements.

A number of federal and state court opinions—as well as legal scholars—state that a goal of the duty to negotiate in good faith is to protect and encourage efficient deal- or relationship-specific investments and minimize the risk of opportunistic holdup behavior. An example is the statement in a recent First Circuit opinion that:

Modern transactions often involve significant up-front investments in deal structuring and due diligence, and parties may wish to protect those investments in some measure. Without any such protection, a rapacious counter-party may attempt to take advantage of the other party’s sunk investment by trying to retool the deal at the last minute.[50]

Indeed, some courts suggest that judicial enforcement of a contract to negotiate is inapt—even in the face of bad faith—in the absence of significant reliance in between the contracting stages; these are likely to be treated as mere agreements to agree.[51] As we explain in section II(C)(2), reliance is an incomplete rationale for the good faith duty: parties often make specific investments in negotiations and deal preparation without contractual protection[52] and the good faith duty also promotes other goals.

Judicial pronouncements as to the appropriate measure of damages for the breach of the good faith duty also reflect the presence of justifications other than reliance. Where the protection of specific investment motivates the arguments in favor of contracts to negotiate in good faith, reliance damages are the usual measure.[53] However, a number of opinions have indicated that expectation damages may be appropriate if the plaintiff provides evidence that good faith efforts would have led to an agreement and if the terms of that hypothetical agreement are clear enough that such damages can be calculated.[54] As we discuss in the next part, the openness to expectation damages reflects a recognition that parties use preliminary agreements to pursue objectives other than the protection of specific investment, including the allocation of some risks. Other scholars have focused on the objective of protecting reliance. We believe that, while the goals vary across contexts, reliance protection by itself is in fact not usually the driving goal. We present a more complete framework in Parts II and III.

II. The Structure of Multistage Contracting

Two challenges face commercial parties in designing the process of contracting in multiple stages. The first is whether a court will accurately find manifest intent to create or disclaim a contract to negotiate in good faith or with best efforts. This challenge is the focus of lawyers’ advice to clients but the easier of the two, especially given that parties can create or disclaim such a duty through an express agreement. The second, and more significant, challenge is that the parties should determine what effect they wish to assign to their preliminary agreement. In many cases, the parties give this second question insufficient consideration, and they are often confused or of two minds. They may simply want to get the best of both worlds by binding their counterparty while preserving their own freedom to walk. Or, they would like some mutual constraint but are unsure about how much and how to combine legal with nonlegal means to achieve it. In this Article, we focus on two questions facing the parties in any given transaction: (1) how sticky should the preliminary provisions be and (2) what mechanism (legal or extralegal) should make these provisions sticky? To begin with, we ask why parties often conduct negotiations in stages and why they may choose to use a preliminary document.

A. Why and How to Sequence Negotiations

The negotiation agenda, including the sequencing of issues, can affect the deal outcome and, indeed, may be the subject of bargaining over process.[55] In many cases, negotiations are sequenced, and a subset of issues is addressed at each stage. If nothing else, sequencing is a response to cognitive barriers to negotiating all issues at once. There is a trade-off between the benefit of being able to logroll across issues to exploit differences in preferences and endowments and the cognitive load of doing so.[56] When deals are broken up into manageable parts, how the issues are divided and sequenced is a distinct and important question. One approach is that of gradualism or incrementalism, under which easier issues are settled first in order to build trust and a positive atmosphere, to create “momentum” to face the more difficult issues in subsequent stages.[57] The terms that are more difficult to resolve are more likely to be the major terms that, in turn, courts are more likely to require in an enforceable contract to negotiate.[58]

An alternative strategy moves in the opposite direction, tackling first the potential roadblocks associated with essential terms before time is spent on the easier matters. This seems to be a common sequence in corporate transactions. Together with early diligence, agreement on the main deal terms—particularly price and deal structure—ensures and signals that the parties are confident they have found the right partner.[59] The benefit of this sequencing is reinforced by the need for costly lawyers, accountants, architects, consultants, and other experts. It makes sense to have the essential elements of the bargain reflect the prospect of a profitable transaction before these investments are incurred. Although the first stage addresses the major terms—such as what is being sold and at what price—the second stage is not trivial. Even if the second stage consists of lawyers hammering out representations and warranties, covenants, closing conditions, remedies, and termination rights, these can contribute significant value to the transaction.[60] Yet, because essential terms are settled first, a key question is the degree to which the parties can adjust them in negotiating the second stage.

The sequencing of the monetary price of a transaction is particularly interesting, especially to the degree that it is difficult to revise a settled price (because of the enforcement of the duty to bargain or other constraints).[61] Price is usually a critical deal term and unique in two important respects. First, it is likely to be the most divisible consideration in a contract: it can be adjusted by dollars while a quality measure cannot. Second, it is typically the distributional term in the contract with the least impact on the size of the aggregate contracting surplus. In some transactions, price is set after the nonprice (often the lawyers’) terms have been set: for example, master agreements for services or goods that are followed by agreements (such as purchase orders) that specify quantity and price. Many other business transactions set price before nonprice terms. For example, in commercial loans, private equity investments, and corporate acquisitions, many terms are agreed upon after the price is settled: representations and warranties, covenants, termination rights, choice of law and forum, etc.[62] In the first stage of these negotiations, the parties agree to price and key nonprice provisions, often without their lawyers, and then turn over the second stage to their lawyers to work out these contractual details. These terms, though valuable, are usually settled without adjustment to price. This arrangement leads to a peculiar process in the second bargaining stage between lawyers in which the parties are limited to bartering nonprice provisions.[63] The stickiness of price constrains the ability of the parties to maximize the efficiency of their contract design, particularly when the parties are not using standard or market terms.[64] They cannot adjust their terms to respond fully to either new information or changes in their environment. Yet, as explained below in subpart II(C), the parties must balance the risk of these ex post efficiency losses against a variety of ex ante efficiency advantages from constraining flexibility. Overall, we describe a framework under which parties agree to distributional parameters and allocating some risks in the preliminary agreement, while working toward maximizing the contracting surplus in their final contract.

B. Purpose of Preliminary Agreements

When the parties choose to divide their negotiations into multiple stages, there is a separate question as to whether and when to memorialize their progress, as well as whether to give such memorialization any legal impact. The most straightforward sense in which they may choose to invoke a legal obligation is in express provisions that regulate their negotiation process.[65] If the parties anticipate continuing their diligence investigation after the preliminary agreement, the party being investigated may agree to cooperate by releasing relevant, nonpublic information while the investigating party may promise to keep information confidential.[66] Promises of confidentiality and nondisclosure are common. The parties may also agree to a period of exclusivity (or “no shop”), during which they agree not to shop, solicit, or discuss a similar transaction with any third party.[67] There is little doubt that the parties can enter into these binding promises, and their advantage is relatively clear. The focus of this Article is the obligation to negotiate in good faith, which may be provided for in combination with such other express terms that the parties intend to be legally binding.[68] As we have emphasized in this Article, the central feature of the good faith duty lends some degree of stickiness to the deal terms that are included in the preliminary document.

The decision to sequence negotiations presumably requires that the first-stage terms would have to be somewhat settled before beginning the second stage. Otherwise, how would the parties maintain the psychological benefits of a gradual compromise or the fragmentation of complex issues referred to above? What would be the value of addressing roadblocks first if they could reappear later in the negotiations? Moreover, parties would negotiate more seriously before the preliminary agreement if they knew that there would be a cost to renegotiating or walking away from its terms. Presumably, memorializing the settled terms for the benefit of future negotiations supports this staging process. A related purpose of the memorialization of a term sheet is to communicate the settled terms with third parties who were not part of the negotiations,[69] and this requires some degree of understanding that the counterparty is not at liberty to insist on significant changes in these terms.

One view held by deal lawyers is that preliminary agreements are “signposts. They mark a moment in the deal’s lifecycle when enough uncertainty and complexity has been resolved that the deal is likely to go forward . . . . [They are] markers for the accumulation of deal momentum.”[70] According to this view, these documents are not legally binding; instead, they indicate that momentum has been achieved and that a completed deal is prospectively along the lines of the thereto settled terms.[71] The parties use this information as a sign that they can proceed to incur further negotiation costs.[72] Indeed, these preliminary documents serve as evidence and provide information to the deal team of lawyers, accountants, and other advisers, as well as lenders, investors, and other third parties who will contribute to the project.[73] In this respect, commentators sometimes refer to these preliminary documents as mechanisms for “organizing” these third parties.[74]

In our analysis, the content of the duty to negotiate in good faith depends on the content of the memorialized deal terms and the court’s determination of the degree of flexibility that the parties had to adjust them in subsequent negotiations. As noted in Part I, courts repeatedly refer to this central feature in holding that the parties must negotiate the subsequent stage “within the scope of,” “within the terms specified in,” “in accordance with,” or “incorporating the agreed terms in” the preliminary agreements.[75] Given that the parties need flexibility to complete their negotiations and design the optimal terms of exchange, the optimal stickiness is a matter of degree and of type. As we elaborate in the next section, when the parties conclude their search for the right partner, they enter into a preliminary agreement to crystallize, at least to some degree, the distribution of (expected) contracting surplus, while they incentivize investment in designing the optimal terms of trade that will maximize that surplus. The imposition of a duty to bargain in good faith may provide a useful, costly signal of each party’s confidence that they have searched and found the appropriate counterparty.[76] The stickiness can accomplish several other objectives: encouragement of reliance investments, deterrence of wasteful bargaining behavior, and allocation of risks. We refer to these collectively as ex ante efficiency considerations, while we refer to the optimization of deal terms and exchange as ex post efficiency.[77] We describe these goals and the balancing objective in the next section.

C. Stickiness of Settled Terms in Preliminary Agreements

We have identified the stickiness of deal provisions in preliminary agreements as a key element of the obligation to negotiate in good faith. In this section, we discuss how the parties might choose the optimal degree of stickiness by balancing ex post and ex ante efficiency considerations. Then, in Part III, we address the parties’ choice between legal and nonlegal mechanisms to achieve their desired stickiness of preliminary terms. Contracting parties would opt for legal enforcement of their preliminary terms, through a duty to bargain in good faith or otherwise, if (a) they wished their terms to be sticky and (b) they preferred legal to nonlegal measures to achieve this. The optimal answers to these questions vary across circumstances, which is why we observe good faith duties in some preliminary agreements but not in others.

The less sticky the preliminary terms, the more flexibility is available to the parties to optimize the terms of the contract at the end of negotiations so as to maximize the contractual surplus (ex post efficiency), given the conditions that exist at the conclusion of their negotiations. However, stickiness in major deal terms promotes three types of ex ante efficiency goals: (1) encouraging relationship- or deal-specific investment that helps them identify those (ex post) optimal terms; (2) discouraging inefficient surplus-grabbing or rent-seeking behavior; and (3) promoting efficient allocation of risk that may arise from changes in circumstances. We discuss the ex post and ex ante efficiencies below.

1. Flexibility and Ex Post Efficiency.—Contracting parties typically do not wish to make the terms of their preliminary agreements fully binding because they need flexibility to accommodate another significant stage in negotiation.[78] Given the requirements that contract law imposes, even on consensual modifications to binding contracts,[79] the parties have greater flexibility if they agree to negotiate in good faith from preliminary terms rather than if they enter into a binding contract with the view of subsequently modifying it.[80] Simply put, when the parties fail to modify an existing contract, they will be bound by the initial terms; whereas if they fail to modify a nonbinding preliminary agreement, they can walk away from the deal. This flexibility is particularly important if the parties anticipate that they will acquire new information, or be subject to changed circumstances, that would reveal a more efficient transactional structure or set of terms or that the transaction is inefficient under any terms. In a corporate acquisition, for instance, when the parties expect to uncover new information through the due diligence process or face an unforeseen hurdle in obtaining regulatory or shareholder approval, it becomes particularly important for them to remain flexible to accommodate the future challenges.[81] Any stickiness in the initial terms—especially price terms, as noted above—limits the opportunities to logroll and agree to all efficient terms.[82] This flexibility should not be surrendered without the existence of offsetting benefits from imposing legal weight on settled terms.

Contract theorists generally presume that commercially sophisticated parties are able to renegotiate when new information comes to light.[83] The parties agree to adjust previously settled terms and implement the deal structure that is ex post efficient, even if the settled terms are fully binding. However, this relies on assumptions that the new information is revealed at no cost, there are no obstacles to modification, and the parties symmetrically acquire information in the renegotiation stage. These assumptions obscure important differences between renegotiating a legally binding agreement and a newly enforceable contract. They are also stylized and not borne out in reality. Indeed, one can readily imagine scenarios in which costly renegotiation would not take place. Suppose, for example, that the cost of diligence and renegotiation to reveal and adjust to new information must be borne by one of the parties in the amount of $7. If the modified terms would increase the joint surplus by $10 and the parties enjoy even bargaining power (so as to enjoy a $5 incremental surplus each), this party would not have the incentive to invest.

Even if a contracting party has the incentive to invest in diligence and modification efforts, there may be another type of inefficiency arising from the private information it acquires about the ex post optimal deal terms.[84] On this basis, the party would propose a modification with a new or additional deal term. The uninformed counterparty, however, would not know whether the proposed term is surplus increasing in fact or the informed party is acting opportunistically by attempting to sell a term at a higher price than it is worth.[85] For example, a seller may attempt to use its superior information about the value of warranties to a buyer to dupe the buyer into accepting a broader (but unnecessary) warranty at a higher price. Anticipating this rent-seeking incentive, the uninformed party may rationally reject any offer of a term from the informed party, even when it may be in fact efficient. The prospect of such renegotiation failure raises an obstacle to (ex post) efficient modification, which the parties would want to address ex ante.

Staged contracting can affect the impact of asymmetric information in interesting ways. The hazard from asymmetric information is greater if the terms of the preliminary agreement are fully binding because a fully binding agreement to transact ensures the informed party its expected profit under that deal.[86] The informed party may thereby have little or nothing to lose by opportunistically offering new terms solely to extract a larger surplus from the counterparty.[87] If the uninformed party rejects the offer, the informed party can simply fall back on the initial terms and still realize its expected profit. Similarly, if the uninformed party can fall back on its expected profit from the initial agreement, it can afford to be more skeptical of the new proposal and much more inclined to simply reject the modification proposals. In short, given this skepticism in the presence of asymmetric information, it could be (often prohibitively) difficult for the parties to renegotiate the fully binding initial terms so as to adopt a more efficient set of terms.

In contrast, if the terms of the preliminary agreement are not binding, the incentive of the informed seller to dupe a buyer into a broad warranty term is at least partially attenuated because neither party has an enforceable contract to fall back on. In this case, if the uninformed party rejects the proposed terms, the parties no longer have a deal, and neither party will be able to capture any contractual surplus. While such a downside risk can go a long way in terms of deterring the informed party’s opportunism, it may not be sufficient, especially when the extra return an informed party can realize by acting opportunistically is substantial.[88] There is no easy way of guarding against such opportunistic behavior under these conditions (except perhaps for simply walking away completely from the deal, which could be inefficient).

The intermediate regime—a Type II preliminary agreement with the duty to negotiate in good faith or best efforts—can better address the problem of (re)negotiation failure in many circumstances. It does so by punishing the party who attempts to renegotiate the terms of the initial agreement solely for the purpose of extracting rent from the counterparty. Under the good faith standard, the parties invite the future court to examine the motives behind the informed party’s new proposal by asking whether the informed party acted dishonestly. They thereby create a new avenue for the uninformed party to police the informed party’s opportunism that is valuable, as we explain in Part III below, even when judicial enforcement is costly and error-prone.[89] When the parties are reassured that opportunistic behavior is deterred, it is easier for them to supplement or modify the terms when it is in fact efficient to do so. Furthermore, when the parties realize that the deal is no longer efficient, given that the initial agreement is not binding, they are not stuck with a fully binding initial agreement; they can avoid liability if they negotiate in good faith.

2. Relationship-Specific Investment and Ex Ante Efficiency.—As indicated earlier, the conventional justification given for enforcing negotiation duties in preliminary agreements is to protect and encourage relationship-specific or reliance investments that would increase the expected deal surplus.[90] Often, the investment contemplated by courts, scholars, and lawyers is the cost of negotiating and contracting over the terms of the deal, conducting due diligence, and retaining experts to advise on performance (lenders, consultants, architects, etc.).[91] The classic concern of contract theory is that if price is negotiated or renegotiated after one party has made such reliance investment, the other party may engage in a holdup to deprive the investing party of the expected return on its investment. Anticipating such holdup, the investing party would decline to make even a beneficial investment. If the terms of the bargain have not been settled, this holdup hazard may go unchecked, so the parties may wish that the courts police it by enforcing an agreement to negotiate in good faith.

In their article on preliminary agreements, Alan Schwartz and Robert Scott focus on a similar problem. They examine incentives in “exploratory” preliminary agreements, under which the parties agree to make simultaneous investment in information that would reveal whether and which among a set of contemplated projects would be profitable to pursue.[92] For example, a seller would investigate its cost of delivering a good and the buyer would investigate the value it would receive, and the parties would then share the acquired information to determine whether the contemplated trade is efficient. Without enforcement of the promise to invest, each party has the incentive to cheat: to let the other party invest first and decide later whether to invest.[93] The authors believe that the good faith duty can encourage investment by ordering the reneging party to pay the reasonable costs of the performing party’s preliminary investment.[94] Schwartz and Scott remark that the predominant legal approach, via the contract to negotiate, “is deficient, however, because it is unnecessary to require parties to bargain in good faith. As we show, efficiency would be enhanced if the law were simply to protect the promisee’s reliance interest.”[95] In their analysis, they assume that the reliance expenditure is (at least partially) verifiable, as is its reasonableness, so the good faith standard is an unnecessarily indirect means to create the efficient investment incentive.[96] To the extent that what the parties really care about is their investment, their contract, or the law, should obligate them to make the reasonable amount of reliance investment. By contrast, the existing law on preliminary agreements concerns the duty to negotiate in good faith, and this is the obligation often expressly adopted by the contracting parties. This suggests that the parties are likely to be more concerned about other features in the process of negotiation and contract design than simply protecting reliance investment.[97]

The use of preliminary agreements to protect reliance investments is more complicated than presented in much of the commentary. Indeed, investments that assess whether a contemplated transaction with a particular counterparty will be profitable or that determine the basic framework of the deal may occur before or after the preliminary agreement. Commercial parties regularly make specific investments before entering into any agreement and expect to recover them from the profits of completed deals.[98] Contract theory is over-stylized when it posits that specific investments are deterred by the threat of holdup.[99] In fact, from a survey of counsel to merger and acquisition deals, Cathy Hwang observes that “[s]cholars assume that preliminary agreements are first steps . . . before investigation, and before making relationship-specific investments. In reality, parties sign preliminary agreements slightly later in the deal process, after most initial investigation is done.”[100] Their ability to continue conducting diligence can be enhanced in preliminary agreements by the binding specific promises described earlier, providing one’s counterparty with access to records and related materials, the sharing of expenses, and adhering to nondisclosure requirements. In light of Hwang’s finding that most of the investment in search has been completed before the preliminary agreement, it suggests that the parties have confidence that they have selected the optimal partner and that the transaction is likely to be efficient.[101] The open question is, under what terms? In these cases, the significant reliance investment after the preliminary agreement is in the negotiation and design of the deal terms that increase the shared surplus.

There is another respect in which the scholarly analysis of specific investment is incomplete. Recognizing the holdup hazard described above, negotiating parties may wish to provide incentives for investment in discovering the parameters of efficient deal terms (such as the optimal representation or warranty, closing condition, or even a revised deal structure) that will increase the size of the transaction surplus. However, making the preliminary terms sticky to protect such investment may lead to another type of inefficiency. In cases in which one party’s investment improves the other party’s return from the transaction (for instance, the seller uncovers new warranty terms that will increase the buyer’s value),[102] the investing party will not be able to capture any benefit from the investment if the preliminary agreement causes the other party to refuse to renegotiate the initial terms. Moreover, as discussed in section II(C)(1) above, if the investing party privately uncovers relevant information, the uninformed counterparty may hesitate to renegotiate the terms due to the possibility of being duped by the informed party. In such scenarios, the parties will face a dual challenge of providing proper investment incentive and deterring rent-extracting behavior.[103] In either case, some flexibility to renegotiate the terms (in favor of the investing party) is necessary in these cases to encourage the beneficial investment.[104]

3. Discouraging Inefficient Investment in Bargaining Power.—Negotiation experts contrast strategies of value-creation and value-claiming. The foregoing discussion concerns value-creation: how the parties can design their deal terms to maximize their transaction surplus. Investment leading to efficient deal terms is desirable in this respect and the parties should seek to encourage such beneficial investment. Value-claiming strategies, in contrast, aim to capture a larger share of the surplus by changing the perceived bargaining range between the parties’ reservation prices. They usually entail costly actions or investments that yield a private return to the actor at the expense of the counterparty, yielding a net reduction in the transactional surplus. First, a party can improve its alternatives to the agreement (known as BATNA[105]), raise its reservation price, and thereby improve its share of the surplus.[106] It can do so by seeking and developing negotiations with alternative counterparties. For instance, even though a party knows that closing the transaction with the counterparty is efficient, it can attempt to secure a competing offer from a third party solely for the purpose of seeking concession from its counterparty. To be sure, investing in alternatives can reveal a more efficient deal elsewhere, but the motivation of concern here is to improve bargaining power. Second, one party’s selfish interest is served by becoming better informed about its counterparty’s reservation price and changing its counterparty’s perception of its own.[107] Of course, the counterparty has the corresponding incentive to conceal that value. Expenditures incurred to conceal one’s own or reveal one’s counterparty’s reservation price do not contribute to efficiency unless they produce information relevant to whether the deal and its terms are optimal. Third, a party may take actions to hurt the counterparty’s no-agreement alternative or BATNA and thereby change the bargaining range in the first party’s favor.

The parties’ goal of deterring value-claiming investments is distinct from encouraging value-creating investments. As noted earlier, incentives for surplus-enhancing investments can be created by a simple promise to share expenses or reimburse if the deal falls through because the party incurring the cost has the incentive to provide evidence that they were made and were reasonable.[108] It is much more difficult to deter value-claiming investments because they are difficult both to observe and verify. It is not surprising, therefore, that preliminary agreements often have no-shop, no-talk, or other exclusivity restrictions on developing alternatives during a time set for negotiations alongside the contract to negotiate in good faith.[109] The good faith standard is broader than the specific exclusivity restrictions, and the court’s investigation into the good faith or best efforts of each party can police value-claiming behavior not caught by the specific prohibitions.[110] For example, it is easier to verify the seller’s marketing of an asset than the buyer’s search for alternative acquisitions.

As an alternative to policing the value-claiming activities themselves, the parties could seek to neutralize any advantage gained after the signing of the preliminary agreement by locking in the distribution of the surplus while encouraging the parties to invest in maximizing the surplus. Where this is an important goal, the parties incur the loss of some logrolling flexibility by agreeing on price terms in the preliminary stage and limiting the ability of either party to demand a large revision in the price or other key terms. Each party has a correspondingly limited incentive to improve its BATNA or impair its counterparty’s BATNA. The parties might allow each other to walk away from the deal but not to propose modified terms that shift the surplus-sharing proportion significantly in favor of one party or the other. Or, they may simply close off the ability to take advantage of an outside offer in any respect by adding an exclusivity clause.

4. Risk Allocation.—Another reason that parties may prefer stickiness in their preliminary agreement terms—also distinct from protecting relationship-specific investment—is to manage risks that affect the size and distribution of the deal surplus. Risk allocation is a well-known, core function of contracts. Even if a deal is not completely negotiated, allocating some risk of changes in circumstances during the negotiation period may be efficient, and there may be gains to putting this allocation in place sooner rather than later. New circumstances or information can lead to (a) a change in the expected distribution of the contracting surplus; (b) a change in the expected size of the joint gains from the contemplated transaction; and (c) a change that could render the transaction inefficient. Optimally, the parties would seek to avoid the deal in the last contingency and to preserve it in the first and second contingencies, perhaps with modified or additional terms in the second. Indeed, this ex ante challenge is familiar generally in the design and enforcement of contracts in the face of changed circumstances.[111] The same principles apply to preliminary agreements and the enforcement of obligations to bargain in good faith.

If the parties settle terms that allocate an exogenous risk, either explicitly or implicitly, the court should enforce that allocation. In this light, good faith may be breached when a party attempts to escape the adverse materialization of a risk allocated to it under the preliminary agreement by breaking off negotiations or demanding an unreasonable modification of the agreed-upon terms.[112] A party breaches its good faith obligation when it unreasonably insists on changing the settled terms, particularly if it thereby seeks to either escape or modify the deal because of the materialization of a risk that was otherwise allocated to that party.[113] In Venture Associates Corp. v. Zenith Data Systems Corp.,[114] Judge Posner drew a significant distinction between preliminary terms that are “open” and “closed,” finding that the price in the preliminary agreement in that case was an open term that the parties could continue to negotiate.[115] Parties allocate risks by agreeing to close terms.[116] Such settled terms may allocate a range of risks: from the core terms of price to less essential legal conditions—such as the availability of regulatory provisions like tax or accounting treatments—that may affect the value or cost of performance to either side.

a. TIAA Cases.—The leading cases concerning agreements to negotiate in good faith in the Southern District of New York concern loan commitments entered into by the Teachers Insurance and Annuity Association of America (TIAA), and they provide good examples of the risk allocation function of preliminary agreements. The commitment letters in those cases memorialized agreement on the basic loan terms, including the rate of interest, but left other clauses to be negotiated in the next stage, such as closing conditions, covenants, events of default, remedies, and default prepayment fees.[117] The letters stated that these other provisions would be negotiated “within the scope that has been settled in the preliminary agreement.”[118] The court found that the letters were Type II agreements because the parties did not intend to enter into fully binding loan agreements but sought to police negotiations by agreeing to bargain in good faith.[119] During the negotiation of the open provisions, market interest rates fell and the courts found that the borrower in each case sought to escape or renegotiate the rate of interest.[120]

In one case, the court found that the borrower objected to the lender’s proposal to include a default prepayment fee as “a pretext for not going forward with the loan,” instead of counteroffering or otherwise negotiating over that term.[121] The court held that the “refusal to negotiate with respect to the Default Prepayment Fee Language was simply a last-ditch attempt to scuttle the loan agreement . . . .”[122] In a second case, the borrower insisted that the deal was conditional on its ability to report the loan in its financial statements as an off-balance-sheet item.[123] The court found that the borrower unreasonably insisted on a “condition that was outside the scope of the bargain.”[124] In the third case, the borrower argued that the lender had inappropriately insisted on a call-protection right outside the scope of the commitment letter, but the court held that the borrower’s argument was a pretext for escaping an unfavorable interest rate.[125] The court observed that the “borrower took a negotiating stance allegedly designed to alter or scuttle the transaction, and finally refused to continue negotiating with TIAA, claiming that TIAA had ‘walked’ from the deal.”[126]

Significantly, the New York court in two of these cases awarded TIAA with expectation damages equal to the difference between the expected interest income under the contract and what it would get by lending to a third party at the time of breach.[127] Although the court expressed concern about the lender’s specific investment,[128] which might have been compensated with reliance damages, its approach in the face of market change was to enforce the allocation of interest rate risk that the parties had agreed to in their commitment letters. Therefore, it held that expectation damages were appropriate despite the fact that many provisions were left to be negotiated.[129]

b. SIGA Techs., Inc. v. PharmAthene, Inc.—In other cases, whether and how the parties intended to allocate risk of changes in their environment is not as clear as in the loan commitment context of the cases described above. Without such a clear indication, the court can look for evidence of such allocation in the text or circumstances of the parties’ agreement, or it could make an educated inference as to the parties’ likely intent. Whether explicit or implicit, risk allocation in commercial transactions is typically meant to promote efficiency. The basic principle underlying analogous elements of contract doctrine is that the superior risk bearer is likely to be the party who can control the probability or magnitude of the contingency or who can better insure against it.[130] This principle provides useful guidance in some cases. As noted earlier, preliminary agreements are often executed in the middle of diligence activity. Presumably, the risk of error in prior diligence should normally fall on the party conducting the diligence (barring some act of concealment or fault of the other party), while the risk of surprises from future diligence would not. Indeed, preliminary agreements sometimes state that final agreement is contingent on the report of a pertinent expert, whether engineer, accountant, architect, or lawyer.[131]

The parties’ intent to allocate risks through settled terms in preliminary agreements is not always as clear as in the case of loan commitments. The Delaware case of SIGA Technologies, Inc. v. PharmAthene, Inc. provides a good example that highlights the importance of objectives other than the protection of specific investments, particularly the allocation of risk. In that case, there were no significant reliance expenditures after the term sheet and the court opined that expectation damages were appropriate if the trial court found that the parties would have concluded a contract but for the defendant’s bad faith.[132] Yet, the implicit allocation that would be thereby enforced is debatable under the facts provided in the court’s opinion.

SIGA acquired an antiviral smallpox drug in 2004, but it encountered difficulties developing the drug and was running out of money by the end of 2005.[133] Desperate for an infusion of cash, it entered into discussions with PharmAthene, who was interested in either acquiring SIGA or receiving a worldwide license of the smallpox drug.[134] While negotiating a merger, the parties signed a two-page license agreement term sheet (LATS) in early 2006 that contained all the deal terms.[135] They subsequently entered into a merger agreement and a contract for $3 million in bridge financing.[136] Under both of these agreements, the parties agreed that if the merger failed to close, they would negotiate a license agreement in good faith “in accordance with” the term sheet.[137] In fact, the merger did not close and the parties resumed negotiating the license agreement.[138] However, SIGA had meanwhile met with considerable and unexpected success with the drug: it was awarded approximately $20 million in new grants from the National Institutes of Health and announced that the drug had provided full protection against smallpox in a trial with primate subjects.[139] It had also sold 2 million shares of stock at a price more than three times its 2005 share price.[140] Armed with good fortune, SIGA then presented PharmAthene with an offer for a license agreement on terms that were radically more favorable to SIGA than the term sheet.[141] When they were unable to bridge the gap, PharmAthene terminated the negotiation and brought action in the Delaware Court of Chancery.[142]

At trial, the Chancery court held that SIGA acted in bad faith when it proposed these radically different terms and breached its duty to negotiate in good faith.[143] On appeal, the Delaware Supreme Court found that:

Evidence that “SIGA began experiencing ‘seller’s remorse’ during the merger negotiations for having given up control of what was looking more and more like a multi-billion dollar drug” bolsters the Vice Chancellor’s finding that SIGA failed to negotiate in good faith for a definitive license agreement in accordance with the terms of the LATS.[144]

By pointing to the “seller’s remorse” in having given up huge profits from the multibillion-dollar drug, the court implicitly interpreted the term sheet as having allocated most of the upside to PharmAthene. SIGA could not use the negotiation over the definitive license agreement to recover those profits by modifying the terms any more than the borrowers in the TIAA cases could use their post-commitment-letter negotiations to benefit from declining market interest rates. Accordingly, the Delaware Supreme Court in SIGA sought to protect PharmAthene’s inchoate expectation in the term sheet and not simply its reliance interest. This recognition of an expectation interest parallels that of the New York court in the TIAA cases. The Delaware Supreme Court went on to state that:

[W]here the parties have a Type II preliminary agreement to negotiate in good faith, and the trial judge makes a factual finding, supported by the record, that the parties would have reached an agreement but for the defendant’s bad faith negotiations, the plaintiff is entitled to recover contract expectation damages.[145]

The implication of the court’s opinion is that a party such as SIGA could not use supervening events that would improve its bargaining position to extract better terms from PharmAthene.

Cases such as this illuminate the purpose of good faith bargaining promises that go beyond the protection of reliance investments. The good faith promise protects the distribution of contracting surplus from both value-claiming behavior and the realization of allocated risks while encouraging the parties to focus on maximizing the size of the surplus by designing optimal contract terms. The court’s decision also protects this distribution from the materialization of certain risks. In SIGA, PharmAthene’s argument implicitly stated that the term sheet also conferred on it a significant share of the upside risk of the drug’s effectiveness, although it would have been under a duty to negotiate to an agreement if the drug’s prospects had dimmed. Whether this risk had in fact been allocated through the provisions in the term sheet could be disputed. The parties might well have contemplated a renegotiation of the surplus distribution if the prospects improved as dramatically as they in fact did. Given that the improvement was due to SIGA’s post-term-sheet efforts, it is plausible that the parties would have wanted to provide SIGA with a larger share of the incremental value.

D. Remedies

In legal scholarship and in past judicial decisions, reliance is the common measure of damages for breach of the duty to negotiate in good faith in a preliminary agreement.[146] This has been justified by two sets of reasons: (1) the general contract law limitations on expectation damages and (2) the sometimes narrow focus in the good faith duty to negotiate on protecting relationship-specific or reliance investments. With respect to the first, contract law requires generally that the plaintiff prove that losses were caused by the breach. The Restatement (Second) of Contracts, for instance, provides that “the injured party has a right to damages based on his expectation interest as measured by . . . the loss . . . caused by [the other party’s breach] plus . . . any other loss, including incidental or consequential loss, caused by the breach . . . .”[147] Without proof of causation, the court may award no damages or merely a nominal award.[148] Contract law also requires that the plaintiff provide evidence establishing the measure of damages with reasonable certainty. The Restatement (Second) of Contracts provides that “[d]amages are not recoverable for loss beyond an amount that the evidence permits to be established with reasonable certainty.”[149] Significantly, where lost profits are uncertain, the law provides that the reliance (or perhaps restitution) measure may be appropriate, unless the defendant can prove that the plaintiff would have incurred a loss rather than a profit from performance.[150] Just like any other contractual obligations, these contract doctrines apply to the choice of remedy for breach of the duty to negotiate in good faith.

Where the plaintiff seeks a remedy for breach of the good faith duty to negotiate, these requirements lead courts to award expectation damages only if the plaintiff proves that a contract would have resulted had the defendant acted in good faith (causation) and provides sufficient evidence of the likely terms of the hypothetical contract to provide a basis for calculating the plaintiff’s lost expectation (reasonable certainty).[151] The requirements of causation and reasonable certainty present added obstacles to the plaintiff suing for breach of a promise to negotiate in good faith compared to a concluded deal such as an asset sale. With respect to causation, for instance, if the defendant can show that the breakdown of negotiation was caused not by its bad faith actions but by other intervening events, such as changes in circumstances that made the transaction no longer in the parties’ interests, the plaintiff will not be able to recover expectation damages.[152]

In addition to the requirement of causation, if the plaintiff cannot show the damages with reasonable certainty, the court falls back on reliance, restitution, or even nominal damages, depending on whether the plaintiff can satisfy the reasonable certainty requirement. In Brown v. Cara, for example, very few substantive terms were settled in the Memorandum of Understanding, in which the parties expressly agreed to “work together” to execute a contract for the development of a real estate joint venture.[153] Although the plaintiff was able to show causation—that the defendant’s bad faith behavior prevented them from reaching an agreement—it was not able to establish what the deal terms would have been and its expectancy loss.[154] Of course, a court might be able to use legal defaults or standard terms in the industry. However, the court may apply instead the reasonable certainty requirement and award reliance, rather than expectation, damages.[155] By contrast, if the parties intended to bind themselves to the transaction (Type I), they are likely to have agreed to more of the terms (than under a preliminary agreement) and would have more clearly signaled their preference for legal or industry defaults. Under such a Type I agreement, the court is more likely to find that expectation damages have been established with sufficient certainty.

The second reason for the common use of the reliance measure is that scholars and many courts have limited their conception of the purpose of the good faith bargaining duty to the task of minimizing the underinvestment in specific investments. The reliance measure follows from this purpose.[156] In this Article, we have criticized this narrow view of the good faith duty. Indeed, we have noted that, if the court can measure the amount invested in reasonable reliance (i.e., the amount of reasonable reliance is verifiable by the court), the parties can protect specific investments simply by promising to reimburse each other’s reasonable reliance if the negotiations break down. We have described a broader range of purposes for the good faith duty to negotiate and explained how the expectation measure of damages (subject to the usual common law requirements) better vindicates these purposes. If the parties’ intent is to preserve the distribution of surplus or the allocation of risk or to maintain flexibility to adapt to new information, then the expectation measure may be more suitable than reliance. When risk allocation and flexibility are the primary objectives, granting the injured party reliance damages would likely be suboptimal.

At the stage of a preliminary agreement, the parties have typically bargained part of the way toward a deal that should yield a surplus for the parties. The parties expect that further negotiation will increase the size of the surplus but not dissipate it. As we have explained, a danger exists that one party may seek to increase its share of the surplus, even at the risk of destroying the deal. In scuttling the deal, the misbehaving party only loses its share of the expected surplus; the rest is a loss borne by its counterparty. When such behavior is a breach of the good faith duty, expectation damages force the breaching party (party acting in bad faith) to internalize the negative externality (loss of counterparty’s contractual surplus). Expectation damages thereby deter destructive value-claiming behavior. The idea is similar to the concept of negligence or gross negligence in tort law.[157] If a tortfeasor’s negligence is shown to have caused an accident that inflicts harm on a victim, then the victim is entitled to be compensated for the entire loss caused by the negligence. Similarly, in a contract setting, when one party acts in bad faith and the bad faith actions deny the counterparty the contractual surplus, the party acting in bad faith should be responsible for the injured party’s lost contractual surplus, provided that the injured party establishes causation and proves the loss with reasonable certainty.[158]

The foregoing discussion suggests that expectation damages are appropriate where preliminary agreements allocate risks and the causation and reasonable certainty of expectation losses can be verified. These conditions are likely to be satisfied when the parties wish to allocate risks of market fluctuations while they complete their negotiations. The case of Teachers Insurance & Annuity Ass’n of America v. Tribune Co. discussed earlier is exemplary. The parties in that case decided to allocate a market risk by entering into a loan commitment letter, which contained the interest rate, among other key terms, but left less economically significant terms to be determined through future negotiations.[159] Since the intent to allocate risk is relatively clear and the exogenous change in interest rates is easy to verify, the court would award expectation damages if it found that the borrower’s bad faith behavior caused negotiations to break down.[160] As we have observed, risk allocation was less clear in SIGA but the court found that the defendant acted in bad faith by insisting on radically revised terms and the bad faith actions led to the negotiation failure. Although the risk that materialized was not a market risk, the parties had agreed to a fairly detailed term sheet. It was correspondingly easier for the plaintiff to establish both that SIGA’s bad faith behavior prevented them from reaching an agreement and what the plaintiff’s profit would have been if they had concluded a contract.[161]

While expectation damages may be appropriate in many cases, reliance damages might be favored by the contracting parties simply because they are usually less than expectation damages and therefore moderate the degree to which the parties will feel stuck to the terms of their preliminary agreements. If so desired, the parties can make appropriate adjustments in their agreements. In this sense, the content of the duty itself and the remedy are two levers that the parties can pull in tailoring their contracting and negotiating strategy: they can adjust the stickiness of their preliminary terms by adjusting the duty, the remedy, or both. In this sense, the parties have considerable choice within the intermediate Type-II range existing between a nonbinding agreement to agree and an agreement to execute their transaction (with defaults filling any gaps). They can choose not only among substantive obligations such as the duty to negotiate in good faith but also among various damages measures.

The duty to negotiate in good faith is similar to other standard-based obligations.[162] Commercially sophisticated parties often enter into contracts with vague, open-ended obligations in place of more specific alternatives. For instance, many corporate-acquisition agreements include covenants obligating a party to exert reasonable efforts or best efforts in securing the regulatory approval necessary to close the transaction. That party’s promise is to exert reasonable efforts rather than to actually secure the approval. In the event of a breach of such efforts, the counterparty is entitled to recover damages, possibly expectation damages, subject to the usual requirements of causation (including foreseeability and mitigation) and reasonable certainty.[163] In some transactions, the parties calibrate damages by agreeing to pay break-up or reverse break-up fees if they fail to receive the specified regulatory approval.[164] By doing so, they eliminate the hurdles of having to prove causation and damages with reasonable certainty. We address such calibration of damages in the next Part.

III. The Use of Legal Standards to Vindicate the Goals of Staged Contracting

While lawyers warn their clients against agreeing inadvertently to legally binding terms in the preliminary agreements, it is clear that parties often do contemplate some form of commitment.[165] Commentators observe that even when preliminary agreements are not binding, parties tend not to depart materially from terms settled in those agreements in their subsequent negotiations.[166] How the stickiness is achieved is a matter of some speculation, but it is often phrased in nonlegal terms. Frequently, commentary suggests that parties use them to bind counterparties “morally,” “psychologically,” or “ethically.”[167] Of course, some business entities hope to use preliminary agreements to bind their counterparty while retaining freedom for themselves to withdraw with impunity.[168] It is also possible that the reputation of lawyers (and other agents, including financial advisers) provides the discipline discouraging the parties from walking or demanding unwarranted changes in settled terms.

In this Article, we have analyzed partially binding preliminary agreements—which impose a duty to negotiate in good faith or with best efforts—as formal instruments that govern the parties’ negotiation process and have identified objectives beyond the protection of relationship-specific investments. They create value by establishing parameters for negotiation to promote the objectives discussed in the previous Part: (a) encouraging efficient relationship-specific (reliance) investment; (b) preserving relative bargaining power by deterring value-claiming or rent-extracting behavior; (c) enforcing the risk allocation desired by the parties during the negotiations; and (d) allowing flexibility to use new information and expertise to build surplus from the deal.[169] In many circumstances, the parties are unlikely to wish for full enforcement of their preliminary agreements, under which the court would fill the open terms with the legal defaults. This could impede their incentive to invest and their flexibility to adopt improvements in their deal terms. Such enforcement would not balance well the multiple goals that often motivate the parties in the preliminary agreement; optimal enforcement is thus something short of full contract enforcement. While there is a broad (although vague) consensus in favor of a middle ground, there are a range of approaches suggested by legal scholars, many of which combine both legal and nonlegal enforcement. In this Part, we review suggestions that scholars and practitioners have advanced for the mechanism by which to reach this intermediate level of enforcement. We then explain the obligation to negotiate in good faith as a legal standard that promotes this objective.

At the other end of the spectrum from a fully binding preliminary agreement, as mentioned above, Cathy Hwang observes that the terms of preliminary agreements are sticky even though enforcement by courts is likely to be nonexistent. She suggests that lawyers or financial advisors may be the gatekeepers whose reputations are harmed when a party walks away from such an agreement contrary to the parties’ expectations.[170] As described in the previous Part, Alan Schwartz and Robert Scott advocate for protection of specific investments through reliance damages, particularly in the specific context they describe in which parties commit to making simultaneous investments.[171] Under their approach, the party who fails to make their investment must reimburse the nonbreaching party’s investment (reliance) costs. In a subsequent article, Ronald Gilson, Charles Sabel, and Robert Scott advocate for this scheme of reliance damages as “low-powered” legal enforcement, which can combine with extralegal forces to implement the desired degree of freedom with which the parties can deviate from the substantive terms of the contemplated deal.[172]

Jonathan Barnett describes an interesting alternative configuration of legal and extralegal discipline. He examines the context of Hollywood movie deals and suggests that parties are disciplined by a combination of industry norms and the possibility that a plaintiff would succeed in obtaining enforcement from a court.[173] He refers to these deals as “soft contracts” that are possibly but not certainly subject to legal liability.[174] Parties can achieve their preferred level of enforcement and, conversely, transactional flexibility by calibrating the level of formalization used to memorialize the terms, and thereby controlling the probability that they will be legally binding.[175] Barnett describes that parties face “a calculated tradeoff—with respect to each deal element, deal stage, and deal participant—that weighs the marginal transactional flexibility [from the ability to walk away] and cost savings from reduced formalization against the marginal increased risk of holdup and other forms of counterparty opportunism.”[176]

Practitioners, courts, and scholars are all aware of the existence of nonlegal sanctions that visit the party who unreasonably withdraws from negotiations or makes outlandish demands. Sometimes, these extralegal forces achieve the desirable combination of commitment and flexibility. For example, in his discussion of soft contracts between Hollywood movie studios and star actors, Jonathan Barnett reports that there are “norms” deterring the most egregious forms of holdup that are available once a studio has begun filming with a star.[177] He observes that “[i]n practice, nothing close to this extreme form of holdup behavior actually occurs: even in the absence of a signed deal, talent attorneys report that they renegotiate open terms following production but refrain from renegotiating the fixed compensation.”[178] This is consistent with the good faith negotiations that are sometimes required in preliminary commercial agreements, which impede the renegotiation of settled terms but provide flexibility in the negotiation of open terms.

In other contexts, however, lawyers and scholars acknowledge that nonlegal sanctions do not completely deter such behavior and might need to be paired with some degree of legal enforcement. Scott Baker and Albert Choi observe that many extralegal sanctions impose costs on both the party that makes the sanction and the one that receives it.[179] To impose a meaningful relational or reputational sanction by suspending or even terminating a relationship, the enforcer often must forego a profitable transaction or relationship with the transgressor. Furthermore, for reputational sanctions to be effective, the enforcer must obtain fairly reliable information as to whether the putative transgressor indeed behaved opportunistically. In contrast, the outcome of legal enforcement is a payment of expectation or reliance damages between the parties, with some deadweight loss for litigation. The parties, ex ante, may prefer the latter system, especially when reputational sanctions can be triggered by faulty information. In addition, in many contexts, the legal system is a reasonably effective producer of information about the conduct of the alleged bad actor, with access to different and sometimes more superior information than what could be obtained by the counterparty and third parties in the reputational community. The fact finding of the courts may inform the discretion of these parties to impose nonlegal sanctions.[180]

In short, notwithstanding the presence of various extralegal mechanisms, such as social or commercial norms or reputational sanctions, the parties may desire some form of legal enforcement (possibly as a supplement) when negotiating a complex transaction. Under one approach discussed earlier, the parties would agree to a binding set of deal terms (with appropriate default terms to fill gaps), but anticipate subsequent renegotiation and modification. If they wish to promote flexibility in renegotiation, they might limit damages to reliance and thereby make it less costly for either party to walk away from the preliminary terms. Each party would be bound by these preliminary terms but only to the extent of the counterparty’s reliance. The use of the reliance measure here, however, is somewhat arbitrarily chosen because it is less than expectation damages and is otherwise unconnected to the parties’ objectives.

The alternative approach invokes a legal standard—such as good faith or reasonable efforts—to police the negotiation process.[181] We have identified several goals that the parties may seek to promote in regulating their negotiation process beyond the protection of reasonable reliance investment. These require a highly contextual approach that befits a standard. Moreover, they call for a stronger enforcement remedy than reliance damages. While reliance damages could be sufficient to protect and encourage beneficial relationship- or deal-specific investment,[182] they will often be insufficient to achieve desired risk allocation, to deter value-claiming or rent-extracting behavior, or to implement the most efficient deal terms.[183] Perhaps in response to these concerns, many courts have recognized that expectation damages are appropriate if the plaintiff can show that a contract would have been concluded if the defendant acted in good faith and within the likely terms of that contract.[184] This approach follows from the contract law’s general requirements of causation and reasonable certainty. As noted at the end of Part II, expectation damages are often a superior mechanism for enforcing risk allocation as well as discouraging investment in bargaining power while encouraging efficient relationship-specific investment.[185]

Good faith and best efforts are familiar standards in contract and commercial law. However, lawyers and scholars are often critical of vague standards that call for fact-specific determinations because they introduce the costs of unpredictability, judicial error, and litigation cost.[186] This hostility is somewhat curious, especially given the frequency with which complex commercial contracts utilize vague, open-ended standards and obligations, such as the covenant to exert reasonable or best efforts (for instance, in securing regulatory and shareholder approvals), the material adverse change condition, the promise to operate the business in ordinary course of business, etc.[187] Noting that contracting parties regularly agree to standards of conduct even when more precise rules are available, we have addressed these concerns in other work.[188] The key lies in the design of corresponding contract provisions that calibrate the relative levels of litigation costs and liquidated damages for breach, in such a manner as to efficiently set both litigation and performance incentives.

To illustrate, suppose that a court will accurately determine whether a promisor has breached the duty to negotiate in good faith, as intended by the parties, two-thirds (2/3) of the time. That is, one-third (1/3) of truly breaching promisors escape liability while 1/3 of truly nonbreaching promisors are held liable.[189] Suppose also that the parties know whether there has been a breach, that damages are stipulated at an amount, d, and that the litigation cost for the plaintiff is l > 0.[190] By suing, a plaintiff invests l for a “lottery ticket” that gives it a 2/3 chance of winning d if the promisor has really acted in bad faith and a 1/3 chance of winning d if the promisor has acted in good faith. If d is set such that 3l > d > (3/2)l, then the plaintiff will sue only when the promisor has in fact breached.[191] This is a fairly broad range, and the parties can manipulate liquidated damages d and litigation costs l (for instance, by contracting over choice of forum, arbitration, or procedural rules) to some degree.[192] When this condition is satisfied, the sanction of d provides an accurate deterrent to breach of the good faith or best efforts standard, as it was intended by the parties, even in the face of a substantial risk of judicial error (1/3 of the time the court errs). Thus, where the parties adopt a negotiation standard of good faith or best efforts, the parties can realize the benefits of their contextual application while avoiding the downsides.[193]

The foregoing example suggests that even if the courts are error-prone, the litigation process as a whole may be designed to yield efficient enforcement and incentives. Furthermore, as we have argued in our earlier works,[194] in designing the remedy (d from the example or when the court is to determine the appropriate remedy) and the process of adjudication (which affects l), the objective of the remedy is not compensation of the victim per se but creating the screening mechanism to produce more effective deterrence.[195]


In this Article, we have presented a function for preliminary agreements, where contracting is done in multiple stages, that improves the efficiency of complex business transactions. Parties may use the preliminary agreement to regulate their negotiations by including terms such as the duty to negotiate in good faith or with best efforts. We envisage a scenario in which the parties have conducted enough of their search and diligence activities to be confident that they have found the right counterparties. At this point, they seek to deter value-claiming and encourage value-creating investments where value-creation includes the discovery of optimal deal terms. They also often seek to efficiently allocate risks of future changes in their circumstances. The terms provided in the preliminary agreements must have some moderate degree of commitment in order to satisfy this set of goals. We suggest—consistent with the substantial amount of case law—that contract standards such as good faith and best efforts can provide a desirable level of stickiness. We also show how such standards can better address the hazard of information asymmetry, where one party has (or is perceived to have) an informational advantage over the counterparty in recognizing the optimal deal terms. Good faith and best efforts obligations can effectively deter the party from abusing its informational advantage solely for the purpose of extracting a bigger rent from the other. In contrast to the prior scholarship and concerns expressed by practitioners, we have argued that enforcing such duty with simple reliance damages would often be inadequate and, in pursuing goals such as risk allocation, expectation damages would be proper, subject to the usual requirements of causation and reasonable certainty. Finally, we explain the refinement offered by liquidated damages in addressing judicial uncertainty stemming from a vague standard such as good faith and the possible overdeterrence of expectation damages.

  1. .In an asset sale or corporate acquisition, for instance, it is customary for the parties to conduct often extensive due diligence while they are negotiating over the sale agreement. As the due diligence investigation uncovers more information, the new information can affect the nature and character of the terms, such as representations and warranties, covenants, conditions, termination, and indemnification.
  2. .Although we follow convention by using the label “preliminary agreement,” this is sometimes a misnomer because the parties may reach such an agreement after a significant amount of negotiation. As we discuss later, business terms are often settled after the legal terms (e.g., master services agreements) and at other times before (e.g., loan term sheets). See infra note 63 and accompanying text.
  3. .To be sure, noncontract doctrines in tort (e.g., fraud or misrepresentation) and unjust enrichment may be available in some cases of bargaining breakdown. The merits of a general obligation to negotiate in good faith have been debated in legal scholarship. See, e.g., E. Allan Farnsworth, Precontractual Liability and Preliminary Agreements: Fair Dealing and Failed Negotiations, 87 Colum. L. Rev. 217, 220 (1987) (explaining adverse effects of a general duty to negotiate in good faith); Charles L. Knapp, Enforcing the Contract to Bargain, 44 N.Y.U. L. Rev. 673, 686–88, 690 (1969) (arguing in favor of a duty to bargain in order to protect reliance); Leon E. Trakman & Kunal Sharma, The Binding Force of Agreements to Negotiate in Good Faith, 73 Cambridge L.J. 598, 599–600 (2014) (criticizing English courts’ reluctance to enforce agreements to negotiate in good faith).
  4. .Good faith could contribute to a more refined protection than blanket reimbursement of each party’s reliance, similar to the combination of negligence and contributory negligence in tort law, because the defendant would be required to compensate only if negotiations broke down because of her lack of good faith.
  5. .As discussed in section II(C)(2), one category of surplus-increasing investment is in the design of the remaining deal terms, typically by lawyers, such as representations and warranties, termination rights, remedies, etc. In addition to the conventional holdup problem, this type of investment raises a different obstacle by giving the investing party beneficial private information. The investing party could use its acquired information to extract a larger share of the surplus—for example, by offering a warranty at a higher price. Anticipating this asymmetry, the uninformed party will be skeptical of all offers from the informed party (even surplus-increasing ones), and this in turn discourages the informed party from making the investment in the first place. In a companion paper, we analyze the problem of asymmetric information and surplus-producing investment in depth using game-theoretic analysis. Albert Choi & George Triantis, Relationship-Specific Investment, Asymmetric Information, and the Role of Good Faith Obligations (Va. Law & Econ., Research Paper No. 2019-02, 2019), https://‌‌/sol3‌/papers.cfm?abstract‌_id=3330973 [https://‌‌/N9RA-8YCH].
  6. .See infra note 109 and accompanying text.
  7. .See infra subsection II(C)(4)(a).
  8. .The parties are also unlikely to agree to negotiate in good faith if they perceive the court to be very unreliable or deem default remedies, such as expectation or even reliance damages, to be inefficiently large. See infra text accompanying notes 186–95.
  9. .E.g., Albert Choi & George Triantis, Completing Contracts in the Shadow of Costly Verification, 37 J. Legal Stud. 503 (2008) [hereinafter Choi & Triantis, Costly Verification]; Albert Choi & George Triantis, Strategic Vagueness in Contract Design: The Case of Corporate Acquisitions, 119 Yale L.J. 848 (2010) [hereinafter Choi & Triantis, Strategic Vagueness].
  10. .See, e.g., Copeland v. Baskin Robbins U.S.A., 117 Cal. Rptr. 2d 875, 877, 880–81 (Cal. Ct. App. 2002) (distinguishing between an unenforceable “agreement to agree” and a written agreement to negotiate in good faith); Keystone Land & Dev. Co. v. Xerox Corp., 94 P.3d 945, 948, 950 (Wash. 2004) (holding that an agreement to agree is not enforceable and that, in a contract to negotiate, “[t]he parties did not exchange promises to conform to a specific course of conduct during negotiations, such as negotiating in good faith, exclusively with each other, or for a specific period of time”). Ridgway v. Wharton is a classic statement that an agreement to agree is not enforceable. (1857) 10 Eng. Rep. 1287, 1297, 1314–15; VI H.L.C., 238, 263–64, 306–09.
  11. .See Keystone, 94 P.3d at 948 (quoting Sandeman v. Sayres, 314 P.2d 428, 429 (Wash. 1957)) (stating that an agreement to do something that requires a further meeting of the minds is not enforceable).
  12. .See Cathy Hwang, Deal Momentum, 65 UCLA L. Rev. 376, 398–99 (2018) (suggesting that formal enforcement may be unnecessary given various types of nonlegal sanctions).
  13. .See Restatement (Second) of Contracts § 27 (Am. Law Inst. 1981), which provides: “Manifestations of assent that are in themselves sufficient to conclude a contract will not be prevented from so operating by the fact that the parties also manifest an intention to prepare and adopt a written memorial thereof; but the circumstances may show that the agreements are preliminary negotiations.”
  14. .See, e.g., Vacold LLC v. Cerami, 545 F.3d 114, 129 (2d Cir. 2008) (finding that the parties had addressed relevant contingencies such as buyer financing and thus had a Type I agreement); Arnold Palmer Golf Co. v. Fuqua Indus., Inc., 541 F.2d 584, 587–88 (6th Cir. 1978) (finding evidence that the parties intended to enter into an enforceable agreement that would later be formalized); V’Soske v. Barwick, 404 F.2d 495, 499 (2d Cir. 1968) (drawing on the language of the parties’ correspondence to find that they intended to create a binding agreement); Learning Annex Holdings, LLC v. Whitney Educ. Grp., Inc., 765 F. Supp. 2d 403, 410 (S.D.N.Y. 2011) (explaining that, even if a more formal agreement is not produced, each party may demand performance under a Type I agreement); Cohen v. Lehman Bros. Bank, 273 F. Supp. 2d 524, 528 (S.D.N.Y. 2003) (quoting Adjustrite Sys., Inc. v. GAB Bus. Servs., Inc., 145 F.3d 543, 548–49 (2d Cir. 1998)) (stating that the parties’ intention is “key” in determining whether they have entered into a Type I agreement); Larwin-S. Cal., Inc. v. JGB Inv. Co., 162 Cal. Rptr. 52, 60 (Cal. Ct. App. 1979) (quoting Burrow v. Timmsen, 35 Cal. Rptr. 668, 671 (Cal. Dist. Ct. App. 1963)) (noting that not every term of an enforceable agreement must be present in a contract); Loppert v. Windsortech, Inc., 865 A.2d 1282, 1287, 1291 (Del. Ch. 2004) (finding that the communications between the parties established an enforceable agreement because, in part, there was no evidence that they agreed to only be bound by a formal document), aff’d sub nom. Windsortech, Inc. v. Loppert, 867 A.2d 903 (Del. 2005) (unpublished table decision); Berg Agency v. Sleepworld-Willingboro, 346 A.2d 419, 423–24 (N.J. Super. Ct. App. Div. 1975) (stating that if the “basic essentials” of an agreement are “sufficiently definite,” then the parties’ intention will not be frustrated by any remaining gaps in the contract). In New York, the dichotomy between binding versus nonbinding preliminary agreements was laid out by the Second Circuit in V’Soske v. Barwick, 404 F.2d 495 (2d Cir. 1968):Two rules on this subject are well established: first, if the parties intend not to be bound until they have executed a formal document embodying their agreement, they will not be bound until then; and second, the mere fact that the parties contemplate memorializing their agreement in a formal document does not prevent their informal agreement from taking effect prior to that event.

    Id. at 499.

  15. .If there are open terms and the parties intended to be bound, the court can fill the gaps, even significant ones like price. See, e.g., U.C.C. § 2-305 (Am. Law Inst. & Unif. Law Comm’n 2017) (stating that the price is a “reasonable price at the time for delivery if . . . the price is left to be agreed by the parties and they fail to agree”); id. §§ 2-308, 2-309 (filling gaps with respect to place and time for delivery of goods).
  16. .For decades, contract scholars have been calling for middle ground exemplified by a good faith standard. See, e.g., Farnsworth, supra note 3, at 286 (articulating that there is not an adequate reason for refusing to give the explicit intention of the parties to negotiate effect); Knapp, supra note 3, at 716, 728 (arguing for recognition of a middle ground that would direct the attention of courts to the good faith of the nonperforming party). The emergence of promissory estoppel offers some softening of the sharp edge between a full contract and no contract. See Restatement (Second) of Contracts § 90 (Am. Law Inst. 1981) (providing for enforcement of a promise that would be reasonably anticipated to induce action or forbearance). But it is relatively rare in use when bargaining has failed.
  17. .The duties of good faith and fair dealing in the Uniform Commercial Code and Restatement (Second) of Contracts do not extend to precontract negotiations. The civil law duty to negotiate in good faith is often referred to as culpa in contrahendo, and American courts have declined to adopt it, leaving it to the parties to opt in if they choose. Steven J. Burton & Eric G. Anderson, Contractual Good Faith: Formation, Performance, Breach, Enforcement 330–31 (1995); see also Teachers Ins. & Annuity Ass’n of Am. v. Tribune Co., 670 F. Supp. 491, 497 (S.D.N.Y. 1987) (“[P]rime significance attaches to the intentions of the parties and to their manifestations of intent.”). While the civil law jurisdictions (presumably) have an opt-out regime, common law countries have the opt-in rule with respect to the duty of good faith and fair dealing. In contrast, the common law duty of good faith in the performance and enforcement of contracts does apply when the parties attempt to modify an existing contract. See infra note 79.
  18. .Allan Farnsworth explained the policy behind the absence of wholesale regulation of negotiation in the U.S. as follows:The difficulty of determining a point in the negotiations at which the obligation of fair dealing arises would create uncertainty. An obligation of fair dealing might have an undesirable chilling effect, discouraging parties from entering into negotiations if chances of success were slight. The obligation might also have an undesirable accelerating effect, increasing the pressure on parties to bring negotiations to a final if hasty conclusion.

    Farnsworth, supra note 3, at 242–43.

  19. .Applying Illinois law, Itek Corp. v. Chicago Aerial Industries, Inc., 248 A.2d 625 (Del. 1968), is perhaps the opinion that initiated the trend toward judicial enforcement of contracts to negotiate. In that case, the letter of intent stated that the parties “shall make every reasonable effort to agree upon and have prepared as quickly as possible a contract providing for the foregoing purchase . . . embodying the above terms and such other terms and conditions as the parties shall agree upon.” Id. at 627. A third party bid higher than the buyer and the seller terminated negotiations even though the buyer agreed to all additional terms proposed by the seller, including price adjustments and other concessions. Id. at 628. Since then, a very significant number of opinions have enforced agreements to negotiate, many of which we refer to in the footnotes. E.g., Brown v. Cara, 420 F.3d 148 (2d Cir. 2005) (applying New York law); Venture Assocs. Corp. v. Zenith Data Sys. Corp., 96 F.3d 275 (7th Cir. 1996) (applying Illinois law); Newharbor Partners, Inc. v. F.D. Rich Co., 961 F.2d 294 (1st Cir. 1992) (applying Rhode Island law); Channel Home Ctrs., Div. of Grace Retail Corp. v. Grossman, 795 F.2d 291 (3d Cir. 1986) (applying Pennsylvania law); Tribune Co., 670 F. Supp. 491 (applying New York law); Copeland v. Baskin Robbins U.S.A., 117 Cal. Rptr. 2d 875 (Cal. Ct. App. 2002) (applying California law); SIGA Techs., Inc. v. PharmAthene, Inc., 67 A.3d 330 (Del. 2013) (applying Delaware law); Logan v. D.W. Sivers Co., 169 P.3d 1255 (Or. 2007) (applying Oregon law). Based on our survey of case law, it seems that fewer than a dozen states have yet to enforce agreements to negotiate.
  20. .The Type II label originated in the Second Circuit. Judge Leval distinguished between Type I and Type II preliminary agreements in Teachers Insurance & Annuity Ass’n of America v. Tribune Co. With respect to Type I agreements, Judge Leval stated:[A Type I agreement] occurs when the parties have reached complete agreement (including the agreement to be bound) on all the issues perceived to require negotiation. Such an agreement is preliminary only in form—only in the sense that the parties desire a more elaborate formalization of the agreement. The second stage is not necessary; it is merely considered desirable.

    Tribune Co., 670 F. Supp. at 498 (emphasis added). The Second Circuit reiterated this categorization in later cases, including Arcadian Phosphates, Inc. v. Arcadian Corp., 884 F.2d 69, 72 (2d Cir. 1989), and Adjustrite Systems, Inc. v. GAB Business Services, Inc., 145 F.3d 543, 548 (2d Cir. 1998). With respect to Type II agreements, Judge Leval stated:

    The second and different sort of preliminary binding agreement is one that expresses mutual commitment to a contract on agreed major terms, while recognizing the existence of open terms that remain to be negotiated. Although the existence of open terms generally suggests that binding agreement has not been reached, that is not necessarily so. For the parties can bind themselves to a concededly incomplete agreement in the sense that they accept a mutual commitment to negotiate together in good faith in an effort to reach final agreement within the scope that has been settled in the preliminary agreement.

    Tribune Co., 670 F. Supp. at 498 (emphasis added). More recently, the Second Circuit court described Type II agreements as follows:

    “Type II” preliminary agreements . . . are “binding only to a certain degree,” reflecting agreement “on certain major terms, but leav[ing] other terms open for further negotiation.” Type II agreements “do[] not commit the parties to their ultimate contractual objective but rather to the obligation to negotiate . . . in good faith in an attempt to reach the . . . objective within the agreed framework.”

    Brown, 420 F.3d at 153 (second, third, and fifth alterations in original) (citations omitted) (quoting Adjustrite, 145 F.3d at 548) (applying New York law); cf. Frazier Indus., L.L.C. v. Gen. Fasteners Co., 137 F. App’x 723, 734–35 (6th Cir. 2005) (finding that a preliminary agreement bound the parties to negotiate in good faith but declining to find a breach of that duty); IDT Corp. v. Tyco Grp., S.A.R.L., 918 N.E.2d 913, 915 n.2, 917 (N.Y. 2009) (criticizing the Type I versus Type II distinction and finding a binding obligation to negotiate in good faith but not a breach of that obligation).

  21. .See infra subpart II(C).
  22. .In venture capital financing, for instance, expressly opting into the duty to negotiate in good faith is quite common. In the model forms published by the National Venture Capital Association, the model term sheet expressly stipulates (as part of a no-shop clause) that “[t]he [financing recipient] agrees to work in good faith expeditiously towards a closing.” Nat’l Venture Capital Ass’n, Term Sheet 14 (2019), https://‌‌/files‌/Uploads‌/Images‌/CICHandout
    040412.pdf [https://‌‌/2UT3-C6MX].
  23. .Practitioners identify several advantages from entering into a letter of intent in merger and acquisition (M&A) transactions, including: (1) allowing the parties to “test the waters” before incurring the costs of negotiating a definitive agreement and performing due diligence; (2) morally or ethically obligating the parties to key terms; (3) dealing with complexity; (4) satisfying premerger notification requirement under the Hart–Scott–Rodino (HSR) Act; and (5) securing necessary financing. 2 Mergers & Acquisitions Comm., Am. Bar Ass’n, Model Stock Purchase Agreement with Commentary 91–92 (2d ed. 2010) [hereinafter 2 ABA, Model Stock Purchase Agreement]. First, as part of a “match right,” whenever a target corporation receives an unsolicited offer from a third party, the clause obligates the target corporation to negotiate in good faith with the purchaser so as to render the third party’s offer no longer attractive. Such match rights are common in corporate-acquisition transactions. See, e.g., Mergers & Acquisitions Comm., Am. Bar Ass’n, Model Merger Agreement for the Acquisition of a Public Company 169–71 (2011) [hereinafter ABA, Model Merger Agreement] (imposing the duty to negotiate in good faith with the purchaser before the target board can change its recommendation to its shareholders); Fernán Restrepo & Guhan Subramanian, The New Look of Deal Protection, 69 Stan. L. Rev. 1013, 1031–33 (2017) (documenting the rise and the prevalence of match rights in acquisitions involving public target corporations). Second, as part of a dispute resolution mechanism primarily over valuation, transacting parties will also agree to negotiate in good faith. For instance, in purchase-price-adjustment mechanisms, which allow the transacting parties to adjust the purchase price that the buyer has to pay the seller after closing (based on updated accounting statements), if the parties were to proceed to dispute resolution, they will agree to negotiate in good faith. See 1 Mergers & Acquisitions Comm., Am. Bar Ass’n, Model Stock Purchase Agreement with Commentary 67–68 (2d ed. 2010) (stating that the “parties shall negotiate in good faith in order to seek agreement on the procedures to be followed by the Independent Accountants, including procedures with regard to the presentation of evidence”). A similar duty is seen frequently in earn-out agreements. 2 ABA, Model Stock Purchase Agreement, supra, at 135. Contracting parties often expressly opt into the duty to negotiate in good faith in other preliminary agreements, such as the letter of intent and memorandum of understanding.
  24. .Letters of intent in M&A transactions commonly include express stipulations as to which provisions are binding (such as exclusivity, confidentiality, and expense reimbursement) and which are not (such as purchase price and other deal terms). 2 ABA, Model Stock Purchase Agreement, supra note 23, 97–106; see JamSports & Entm’t, LLC v. Paradama Prods., Inc., 336 F. Supp. 2d 824, 828–29 (N.D. Ill. 2004) (examining a letter of intent that expressly stipulates that only the confidentiality, exclusivity, and publicity provisions are binding). In many cases, parties will expressly disclaim any obligation to negotiate in good faith, to negotiate further, or to enter into a definitive proposal. ABA, Model Merger Agreement, supra note 23, at 342–43.
  25. .According to Vice Chancellor Laster of the Delaware Chancery Court:Letters of intent mean something . . . . [P]arties enter into letters of intent for a reason. They don’t enter into them because they are gossamer and can be disregarded whenever situations change. They enter into them because they create rights.

    . . . [I]f parties want to enter into nonbinding letters of intent, that’s fine. They can readily do that by expressly saying that the letter of intent is nonbinding, that by providing that, it will be subject in all respects to future documentation . . . .

    Glob. Asset Capital, LLC v. Rubicon US REIT, Inc., No. 5071-VCL, at *5–6 (Del. Ch. Nov. 16, 2009) (order granting temporary restraining order); see, e.g., Newharbor Partners, Inc. v. F.D. Rich Co., 961 F.2d 294, 299 (1st Cir. 1992) (explaining that the enforceability of an obligation to act in good faith “turns on the clarity with which such intent is expressed in the instrument”); Emma Robinson, Getting Out of a Bind: Making Sure Your Non-Binding Letter of Intent Is Actually Non-Binding, Weil, Gotshal & Manges LLP: Global Priv. Equity Watch (Mar. 24, 2016), https://‌‌/getting-bind-making-sure-non-binding-letter-intent-actually-non-binding [https://‌‌/DG8D-M2TM] (advising that a letter of intent should include a “clear, unambiguous statement” that the parties do not intend for it to be binding).

  26. .E.g., Brown v. Cara, 420 F.3d 148, 157 (2d Cir. 2005); Arcadian Phosphates, Inc. v. Arcadian Corp., 884 F.2d 69, 72 (2d Cir. 1989); Learning Annex Holdings, LLC v. Whitney Educ. Grp., Inc., 765 F. Supp. 2d 403, 411 (S.D.N.Y. 2011); Teachers Ins. & Annuity Ass’n of Am. v. Tribune Co., 670 F. Supp. 491, 499–503 (S.D.N.Y. 1987). This five-factor test in determining whether the parties have entered into a Type II preliminary agreement is a modification from the four-factor test previously used by Second Circuit courts, e.g., Winston v. Mediafare Entm’t Corp., 777 F.2d 78, 80 (2d Cir. 1985), to determine whether the parties have entered into a binding preliminary agreement. Tribune Co., 670 F. Supp. at 498–99. In applying the Type II test, courts have held that the most important factor in determining whether an intermediate agreement is binding is the language used by the parties in the document. See, e.g., Budget Mktg., Inc. v. Centronics Corp., 927 F.2d 421, 426 (8th Cir. 1991) (applying Iowa law to find that the parties did not bind themselves to a duty to negotiate in good faith because the intention expressed in their letter of intent was clear); R.G. Grp., Inc. v. Horn & Hardart Co., 751 F.2d 69, 75 (2d Cir. 1984) (placing “considerable weight” on a party’s explicit statements of intent).
  27. .See, e.g., Browning Jeffries, Preliminary Negotiations or Binding Obligations? A Framework for Determining the Intent of the Parties, 48 Gonz. L. Rev. 1, 15–16 (2012) (noting the inconsistent treatment by courts toward good faith claims); Gregory J. Marsden & George J. Siedel, The Duty to Negotiate in Good Faith: Are BATNA Strategies Legal?, 14 Berkeley Bus. L.J. 127, 154–55 (2017) (highlighting the concern of unknowingly agreeing to a duty to negotiate in good faith).
  28. .See, e.g., Farnsworth, supra note 3, at 259–60 (“It would be difficult to find a less predictable area of contract law.”); Ronald J. Gilson et al., Braiding: The Interaction of Formal and Informal Contracting in Theory, Practice, and Doctrine, 110 Colum. L. Rev. 1377, 1427 (2010) (“[T]he courts’ experience so far provides little normative guidance concerning the breadth of the enforceable obligation . . . .”); Alan Schwartz & Robert E. Scott, Precontractual Liability and Preliminary Agreements, 120 Harv. L. Rev. 661, 675 (2007) (“This modern approach provides too little normative guidance.”).
  29. .See, e.g., A‌/S Apothekernes Laboratorium for Specialpraeparater v. I.M.C. Chem. Grp., Inc., 873 F.2d 155, 159 (7th Cir. 1989) (stating that a “duty to negotiate in good faith does not encompass an automatic duty to approve the final deal”). Applying Illinois law, the Seventh Circuit in Feldman v. Allegheny International, Inc., 850 F.2d 1217 (7th Cir. 1988), stated: “So one cannot characterize self-interest as bad faith. No particular demand in negotiations could be termed dishonest, even if it seemed outrageous to the other party. The proper recourse is to walk away from the bargaining table, not to sue for ‘bad faith’ in negotiations.” Id. at 1223; see also Venture Assocs. Corp. v. Zenith Data Sys. Corp., 96 F.3d 275, 279 (7th Cir. 1996) (stating that “[s]elf-interest is not bad faith” (citing Feldman, 850 F.2d at 1223)); L-7 Designs, Inc. v. Old Navy, LLC, 964 F. Supp. 2d 299, 307 (S.D.N.Y. 2013) (quoting Venture Assocs., 96 F.3d at 279) (same); Jenkins v. Cty. of Schuylkill, 658 A.2d 380, 385 (Pa. Super. Ct. 1995) (quoting Apothekernes, 873 F.2d at 158–59) (same).
  30. .Honesty in fact is the core of the Uniform Commercial Code definition of good faith. U.C.C. § 1-201(b)(20) (Am. Law Inst. & Unif. Law Comm’n 2017).
  31. .Some courts have held that the good faith obligation does not of itself require the defendant to disclose activity in competition to the negotiated transaction. SuperValu Inc. v. Associated Grocers, Inc., 428 F. Supp. 2d 985, 991 (D. Minn. 2006); Schwanbeck v. Federal-Mogul Corp., 578 N.E.2d 789, 798–99 (Mass. App. Ct. 1991), rev’d on other grounds, 592 N.E.2d 1289 (Mass. 1992); see also Trovare Capital Grp., LLC v. Simkins Indus., Inc., 646 F.3d 994, 1000–01 (7th Cir. 2011) (holding that misrepresentation is a basis for finding breach of duty to negotiate in good faith under a letter of intent, especially when part of a strategy to avoid a deal).
  32. .See, e.g., Venture Assocs., 96 F.3d at 277 (observing that good faith might entail an obligation not to entertain other offers). But see Gas Nat., Inc. v. Iberdrola, S.A., 33 F. Supp. 3d 373, 384 (S.D.N.Y. 2014) (defendant’s failure to disclose the existence of bidders other than the plaintiff did not constitute bad faith, particularly because the parties contemplated and rejected including an exclusivity clause); SuperValu, 428 F. Supp. 2d at 992 (noting that the obligation to negotiate in good faith does not preclude competition between the parties where their letter of intent explicitly provides for it).
  33. .The National Venture Capital Association term sheet incorporates the duty to negotiate in good faith as part of a no-solicitation (or no-shop) obligation. Nat’l Venture Capital Ass’n, supra note 22, at 14–15.
  34. .In their treatise on contractual good faith, Steven Burton and Eric Andersen suggest that, in the context of the duty to negotiate, good faith means that a party cannot reject a proposal with respect to an open term pretextually because of dissatisfaction with or regret over a closed term (including the emergence of a better offer or an adverse market change) and that a party may not object to all objectively reasonable proposals for the open terms. Burton & Andersen, supra note 17, at 369–70, 379.
  35. .A‌/S Apothekernes Laboratorium for Specialpraeparater v. I.M.C. Chem. Grp., Inc., 873 F.2d 155, 158 (7th Cir. 1989) (“The full extent of a party’s duty to negotiate in good faith can only be determined . . . from the terms of the letter of intent itself.”); Clark Res., Inc. v. Verizon Bus. Network Servs., Inc., No. 1:10-cv-1119, 2011 WL 1627074, at *5 (M.D. Pa. Apr. 29, 2011) (acknowledging “the danger of applying a duty of good faith, without some limiting principles, to the process of complex commercial negotiations”); Sea Hawk Seafoods, Inc. v. City of Valdez, 282 P.3d 359, 368–69 (Alaska 2012) (granting summary judgment for the defendant after finding no breach of duty to negotiate in good faith when the letter of intent did not include a framework for negotiations against which any such obligation could be evaluated); 2004 McDonald Ave. Realty, LLC v. 2004 McDonald Ave. Corp., 858 N.Y.S.2d 203, 205 (N.Y. App. Div. 2008) (holding that “[n]o objective criteria or standards against which the defendant’s efforts can be measured were stated in the [letter of intent], and they may not be implied from the circumstances of this case”); Jenkins v. Cty. of Schuylkill, 658 A.2d 380, 381–82, 385 (Pa. Super. Ct. 1995) (quoting Apothekernes, 873 F.2d at 158–59) (explaining that in the absence of agreed-upon terms, or even a negotiating framework, the parties were free to insist on or reject any proposed terms to the contract); see also Burton & Andersen, supra note 17, at 360 (observing that “[t]he better view is that a general duty to negotiate, without more, is too indefinite to be enforced”).
  36. .Farnsworth divides Type II preliminary agreements into two subcategories: (1) agreements with open terms and (2) agreements to negotiate. Agreements with open terms bind the parties with respect to the settled terms (such as the interest rate in the Tribune Co. example) while obligating the parties to negotiate in good faith the unsettled terms. Agreements to negotiate, on the other hand, obligate the parties to negotiate in good faith all the terms of the deal. See Farnsworth, supra note 3, at 250–51.
  37. .See, e.g., Karns v. Jalapeno Tree Holdings, L.L.C., 459 S.W.3d 683, 694 (Tex. App. 2015) (noting that the duty of good faith did not prevent the parties from revising covenants in their letter of intent, only that it required they do so in good faith and actively engage in the negotiation process).
  38. .670 F. Supp. 491 (S.D.N.Y. 1987).
  39. .Id. at 498 (emphasis added). The court also stated that the obligation to negotiate in good faith prevents a party from “insisting on conditions that do not conform to the preliminary agreement.” Id. At the same time, the court recognized that the duty to negotiate in good faith does not necessarily guarantee a binding agreement. According to the court:[The good faith] obligation does not guarantee that the final contract will be concluded if both parties comport with their obligation, as good faith differences in the negotiation of the open issues may prevent a reaching of final contract. . . . The obligation does, however, bar a party from renouncing the deal, abandoning the negotiations, or insisting on conditions that do not conform to the preliminary agreement.


  40. .Id. (emphasis added).
  41. .Id. at 500–01 (emphasis added).
  42. .145 F.3d 543 (2d Cir. 1998).
  43. .Id. at 548 (emphasis added); see also EQT Infrastructure Ltd. v. Smith, 861 F. Supp. 2d 220, 232 (S.D.N.Y. 2012) (finding that the insistence on a condition that was not provided in the letter of intent—that the seller find a buyer for another asset—can indicate absence of good faith); Juanes v. Lyzwinski, 875 F. Supp. 2d 155, 162 (N.D.N.Y. 2012) (quoting Tribune Co., 670 F. Supp. at 498) (stating that a refusal to agree to terms in a preliminary agreement and attempts to add terms may indicate lack of good faith).
  44. .420 F.3d 148 (2d Cir. 2005).
  45. .Id. at 157 (emphasis added) (quoting Adjustrite, 145 F.3d at 548).
  46. .873 F.2d 155 (7th Cir. 1989).
  47. .Id. at 159.
  48. .67 A.3d 330 (Del. 2013).
  49. .Id. at 347 (emphasis added). Following the release of the SIGA opinion, lawyers at Fried Frank wrote that abandonment of negotiations without good faith or insistence on materially inconsistent terms with respect to those featured in the term sheet constitute a breach of the obligation to negotiate in good faith, especially if there is evidence of regret. Andrew J. Colosimo et al., Fried, Frank, Harris, Shriver & Jacobson LLP, Practice Points for Term Sheets, Letters of Intent, and Undertakings to Negotiate in Good Faith—Based on Delaware Supreme Court’s SIGA Decision 4 (2016), https://‌‌/siteFiles‌/Publications‌/FINALv8-2-8-2016-Practice%20Points%20on%20Use%20of%20Term%20Sheets
    %20and%20Letters%20of%20Intent.pdf [https://‌‌/HJ7R-EYHY]. However, Fried Frank noted that a good faith disagreement on a material term not included within the parties’ letter of intent is not likely to constitute a breach. Id. Lawyers at Lincoln Gustafson wrote that the good faith obligation would be breached by “attempt[s] to change fundamental points of the deal in order to take advantage of changing external conditions and extract more favorable terms than were initially agreed upon in the letter of intent.” Patrick Klingborg, When a “Non-Binding” Letter of Intent Is Binding After All, Lincoln, Gustafson & Cercos, LLP (June 1, 2016), http://‌‌/when-a-non-binding-letter-of-intent-is-binding-after-all‌/ [https://‌‌/D2WQ-PFZX]. While the decision in SIGA came as a surprise to many practitioners, it was not unprecedented. In RGC International Investors, LDC v. Greka Energy Corp., No. CIV.A.17674, 2001 WL 984689 (Del. Ch. Aug. 22, 2001), for example, the parties had expressly agreed to negotiate in good faith and the Delaware Chancery Court held that this was breached when the defendant broke off negotiations. Id. at *1; see also Glob. Asset Capital, LLC v. Rubicon US REIT, Inc., No. 5071-VCL, at *4, 14 (Del. Ch. Nov. 16, 2009) (order granting temporary restraining order) (stating that “the duty to negotiate in good faith . . . is one that this Court recognizes, is one that is of commercial importance, and is one that this Court will protect” and that “radio silence is not negotiating in good faith”).
  50. .Butler v. Balolia, 736 F.3d 609, 615 (1st Cir. 2013) (citation omitted); see, e.g., Brown v. Cara, 420 F.3d 148, 157 (2d Cir. 2005) (noting that enforcement of preliminary agreements encourages up-front investments by ensuring that counterparties will benefit from at least some protection during the negotiation of a fully enforceable agreement); Burbach Broad. Co. of Del. v. Elkins Radio Corp., 278 F.3d 401, 408–09 (4th Cir. 2002) (“[M]any state courts . . . have recognized the pragmatism and commercial necessity of recognizing such agreements . . . .”); Venture Assocs. Corp. v. Zenith Data Sys. Corp., 96 F.3d 275, 279–80 (7th Cir. 1996) (seller could demand a higher price “provided that it was not trying to scuttle the deal or take advantage of costs sunk by Venture in the negotiating process”) (citation omitted); Copeland v. Baskin Robbins U.S.A., 117 Cal. Rptr. 2d 875, 885 (Cal. Ct. App. 2002) (finding enforceable an agreement to negotiate in good faith and concluding that “damages for breach of a contract to negotiate an agreement are measured by the injury the plaintiff suffered in relying on the defendant to negotiate in good faith”).
  51. .See, e.g., Burbach Broad. Co., 278 F.3d at 407 (asserting that “courts will not find enforceable binding contracts” from an “unspecified agreement to agree at some time in the future”); Skycom Corp. v. Telstar Corp., 813 F.2d 810, 815–18 (7th Cir. 1987) (raising concern about making transactions riskier without need to protect reliance).
  52. .Judge Easterbrook, for example, writing the majority opinion for the Seventh Circuit in Empro Manufacturing Co. v. Ball-Co Manufacturing, Inc., 870 F.2d 423 (7th Cir. 1989), observed that:Empro [the buyer of business assets] claims that it is entitled at least to recover its “reliance expenditures”, but the only expenditures it has identified are those normally associated with pre-contractual efforts . . . “in negotiating with defendants, in investigating and reviewing defendants’ business, and in preparing to acquire defendants’ business.” Outlays of this sort cannot bind the other side any more than paying an expert to tell you whether the painting at the auction is a genuine Rembrandt compels the auctioneer to accept your bid.

    Id. at 426.

  53. .See 3 E. Allan Farnsworth, Farnsworth on Contracts § 12.17 (Zachary Wolfe ed., 4th ed. 2019) (stating that a party’s specific investment includes its opportunity losses that may be compensated for in the measure of reliance damages); infra subpart II(D).
  54. .These courts generally require that the plaintiff show that an agreement would have been reached if the defendant had bargained in good faith (causation requirement) and provide sufficient evidence as to the terms that would have been agreed to (reasonable certainty requirement). In these two requirements, the courts are simply applying the common law conditions of causation and reasonable certainty. See Venture Assocs., 96 F.3d at 278 (“[I]f the plaintiff can prove that had it not been for the defendant’s bad faith the parties would have made a final contract, then the loss of the benefit of the contract is a consequence of the defendant’s bad faith, and, provided that it is a foreseeable consequence, the defendant is liable for that loss—liable, that is, for the plaintiff’s consequential damages.”); see also Teachers Ins. & Annuity Ass’n of Am. v. Ormesa Geothermal, 791 F. Supp. 401, 415–16 (S.D.N.Y. 1991) (finding that defendant breached its duty to negotiate in good faith and that plaintiff’s damages were mathematically straightforward); Teachers Ins. & Annuity Ass’n of Am. v. Tribune Co., 670 F. Supp. 491, 508 (S.D.N.Y. 1987) (“[T]he borrower undertook a binding commitment to negotiate open terms in good faith and breached that commitment.”); Teachers Ins. & Annuity Ass’n of Am. v. Butler, 626 F. Supp. 1229, 1235–36 (S.D.N.Y. 1986) (finding that defendants breached their duty of good faith and that plaintiff was entitled to damages calculated from the difference in interest rates); United House of Prayer for All People v. Therrien Waddell, Inc., 112 A.3d 330, 345 (D.C. 2015) (explaining that expectation damages are appropriate if the trial court finds that a contract would have been concluded if the defendant had acted in good faith and that there is a basis for calculating lost profits); SIGA Techs., IncvPharmAthene, Inc., 67 A.3d 330, 350–51 (Del. 2013) (holding that if the parties have an agreement to negotiate in good faith and the judge makes a factual finding supported by the record that the parties would have reached an agreement but for the bad faith negotiation of one party, then the other party is entitled to expectation damages); RGC Int’l Inv’rs v. Greka Energy Corp., No. CIV.A.17674, 2001 WL 984689, at *13 (Del. Ch. Aug. 22, 2001) (holding that it was bad faith for the defendant to attempt to force the plaintiff to give up a specifically negotiated provision in the Term Sheet); see also Restatement (Second) of Contracts § 347 (Am. Law Inst. 1981) (damages are measured by “the loss . . . caused by [the other party’s] failure or deficiency, plus . . . any other loss . . . caused by the breach”); id. § 352 (“Damages are not recoverable for loss beyond an amount that the evidence permits to be established with reasonable certainty.”).
  55. .Game theorists have shown that the decision to negotiate issues simultaneously or sequentially is likely to affect the probability of success and the negotiated outcome. See, e.g., Thomas C. Schelling, The Strategy of Conflict 31 (1960) (stating that when two objects to negotiate are done so simultaneously, the outcome is affected); Mehmet Bac & Horst Raff, Issue-by-Issue Negotiations: The Role of Information and Time Preference, 13 Games & Econ. Behav. 125, 125 (1996) (noting that the American Automobile Association encourages buyers to first focus on negotiating the price of a car, deferring discussion of financing, factory rebates, and trade-in allowance); P.V. Balakrishnan et al., Toward a Theory of Agenda Setting in Negotiations, 19 J. Consumer Res. 637, 648–49 (1993) (implying a preference for simultaneous or sequential negotiations if an asymmetric power relationship is present); Chaim Fershtman, The Importance of the Agenda in Bargaining, 2 Games & Econ. Behav. 224, 237 (1990) (showing that parties’ differences in preferred outcomes will affect their behavior in sequential negotiations); Younghwan In & Roberto Serrano, Agenda Restrictions in Multi-Issue Bargaining, 53 J. Econ. Behav. & Org. 385 (2004) (arguing that bundling offers can prevent parties from exploiting trade-offs among issues in a negotiation).
  56. .See Howard Raiffa et al., Negotiation Analysis: The Science and Art of Collaborative Decision Making 91 (2002) (describing the cognitive difficulty of negotiating multiple issues at once).
  57. .See Omri Ben-Shahar, “Agreeing to Disagree”: Filling Gaps in Deliberately Incomplete Contracts, 2004 Wis. L. Rev. 389, 390 (2004) (“[Parties sometimes leave] issues that were difficult to resolve for future completion. In these situations, contractual incompleteness is neither a result of haste nor of unforeseeability, but rather a deliberate choice to temporarily disagree over some matters, to sidestep difficult issues over which consensus could not be reached.”).
  58. .See supra note 20 and accompanying text.
  59. .While the common sequence in corporate transactions is to agree on pricing and deal structure provisions first and then negotiate on other nonprice terms, many transactions are negotiated in the reverse order. In corporate public debt or equity transactions, for instance, parties agree on the nonprice terms before the pricing of the instrument. See infra note 63 and accompanying text.
  60. .The value that lawyers contribute may be difficult to quantify but, notwithstanding the aphorism to “let the lawyers take care of the details,” the fact that they can command significant hourly rates for work performed after preliminary documents suggests that the market values their contributions to contracts in the second stage. The various substantive rights and obligations added in the second stage can significantly affect the value and successful execution of the deal. For instance, in an M&A transaction, whether the target can entertain a third-party bid (as stipulated in the “no-shop” clause) or whether the buyer can walk away from the deal when the target suffers a materially adverse event (as defined in the “material adverse change” clause) will bear on the likelihood of successful closing of the deal. While some intermediate agreements, such as letters of intent and term sheets, contain exclusivity or no-shop clauses, other clauses, such as warranties, closing conditions and termination rights are negotiated in the later stage and appear in the definitive agreement. See Choi & Triantis, Strategic Vagueness, supra note 9, at 863–66, 870–72 (explaining material-change clauses, reverse termination fees, and other termination rights).
  61. .See Albert Choi & George Triantis, The Effect of Bargaining Power on Contract Design, 98 Va. L. Rev. 1665, 1694–95 (2012) (explaining how bargaining power might affect nonprice contract terms if price is settled first).
  62. .But see James C. Freund, Anatomy of a Merger: Strategies and Techniques for Negotiating Corporate Acquisitions 59 (1975) (suggesting that a seller’s counsel should delay agreement on purchase price until other material terms have been settled).
  63. .It is interesting in this respect to contrast two types of debt contracts that settle price early or late in the process. In a typical commercial bank loan, the parties first negotiate a term sheet that contains the maturity, interest rate, and other fees, in addition to a handful of major terms. The interest rate is rarely changed subsequently during the negotiation of the covenants, etc. In contrast, in the sale of a bond or debenture, the covenants are settled before the price is determined. Given the advantages of the price term and constraints on nonprice bartering identified in the text, one might speculate that, all else equal, the design of bond covenants would be more efficient. One might also anticipate more innovation in the provisions of nonbank debt, where the market can reward valuable contract terms with a lower yield.
  64. .Moreover, it is generally perceived that the second stage presents the opportunity for the party with the bargaining power at that stage to seize more rents because the price cannot be reopened. See Choi & Triantis, supra note 61, at 1694 (suggesting that “bargaining power is a determinant of the nonprice terms” in this context because of “the inability to trade off risk allocation against a price adjustment”).
  65. .An exclusivity agreement will, for instance, prohibit one or both parties from negotiating with a third party. Under another type of agreement, one party (e.g., the seller) promises to disclose various nonpublic information for the purpose of due diligence while the agreement also imposes a confidentiality obligation on the recipient not to disclose or misuse such information. ABA, Model Merger Agreement, supra note 23, at 345–52, 385–88. Parties may include either or both exclusivity or confidentiality obligations within a preliminary agreement, such as a letter of intent that may additionally obligate the parties to negotiate in good faith. See, e.g., JamSports & Entm’t, LLC v. Paradama Prods., Inc., 336 F. Supp. 2d 824, 828–29 (N.D. Ill. 2004) (documenting a letter of intent that bound the parties to negotiate exclusively with one another and in good faith). In some cases, the exclusivity and duty to bargain is phrased more categorically. See, e.g., Channel Home Ctrs., Div. of Grace Retail Corp. v. Grossman, 795 F.2d 291, 293 (3d Cir. 1986) (detailing a letter of intent that covered most of the significant shopping mall lease terms and provided that the prospective lessor would “withdraw the Store from the rental market, and only negotiate the . . . leasing transaction to completion”).
  66. .ABA, Model Merger Agreement, supra note 23, at 345–52.
  67. .Id. at 385–88.
  68. .In some preliminary agreements, the parties rely on specific provisions without the duty to bargain in good faith. See, e.g., Logan v. D.W. Sivers Co., 169 P.3d 1255, 1262 (Or. 2007) (stating that, without any obligation to bargain in good faith towards a completed sale, the seller agreed to provide due diligence documents within a specified time in consideration for the buyer’s promise to review the documents in good faith).
  69. .For instance, in corporate-acquisition transactions where financing is involved, the purchaser will often forward the preliminary agreement to the investors and underwriters so as to line up the necessary financing before closing the sale.
  70. .Hwang, supra note 12, at 382–83; see also Alan P.W. Konevsky, A US Perspective on the Risks of Term Sheets, 26 Butterworths J. Int’l Banking & Fin. L. 353, 353 (2011) (noting that practitioners frequently affirm that preliminary agreements foster a moral commitment between the parties and create deal momentum).
  71. .Of course, the signal of momentum is not a particularly reliable one in the economic sense because it is relatively cheap (or even costless) to send unless there is some sanction, legal or nonlegal, on the party that deviates from the terms of the agreement, thereby rendering them “sticky” to some degree.
  72. .See, e.g., 2 ABA, Model Stock Purchase Agreement, supra note 23, at 91 (explaining that letters of intent are used to test the waters” for the prospects of a definitive agreement “before incurring the costs of negotiating a definitive agreement”).
  73. .Included in these third parties are regulators whose approvals are condition precedent to the closing of the deals and who need the time to review (e.g., antitrust review). 2 Comm. on Negotiated Acquisitions, Am. Bar Ass’n, Model Asset Purchase Agreement with Commentary 118 (2001).
  74. .See, e.g., Hwang, supra note 12, at 408–09 (entering a preliminary agreement serves both an internal and external organizational purpose); Ralph B. Lake, Letters of Intent: A Comparative Examination Under English, U.S., French, and West German Law, 18 Geo. Wash. J. Int’l L. & Econ. 331, 332 (1984) (explaining that letters of intent assist in “bringing order to the complexity”); see also 1 Comm. on Negotiated Acquisitions, Am. Bar Ass’n, Model Asset Purchase Agreement with Commentary 165 (2001) (explaining that the parties promise to cooperate in obtaining necessary consents and regulatory approvals).
  75. .See supra text accompanying notes 38–49.
  76. .Richard Craswell, Precontractual Investigation as an Optimal Precaution Problem, 17 J. Legal Stud. 401, 403 (1988).
  77. .We are adopting the labels used in contract theory. Although the delineation between ex post versus ex ante efficiency is sometimes unclear in the literature, the purpose is to distinguish between the parties’ behavior before a given event, such as realization of a market shock or arrival of new information that changes the optimal terms, and behavior after the event. For instance, the parties could perform a due diligence investigation (ex ante behavior) that could lead to discovery of new information that they could attempt to use to modify the deal terms or deal structure (ex post behavior).
  78. .See Hwang, supra note 12, at 384 (“[M]ost preliminary agreements in M&A deals are signed but non-binding.”). Hwang also notes that deals are sticky even though enforcement for breach is weak. Id.
  79. .Under the Uniform Commercial Code, an enforceable contract modification requires not only mutual assent but also good faith which, between merchants, includes observance of reasonable commercial standards of fair dealing in the trade. U.C.C. § 2-209 cmt. 2 (Am. Law Inst. & Unif. Law Comm’n 2017). Although new consideration is no longer necessary to support an enforceable contract modification, “modifications . . . must meet the test of good faith imposed by [U.C.C. § 1-304].” Id. Further:The effective use of bad faith to escape performance on the original contract terms is barred, and the extortion of a “modification” without legitimate commercial reason is ineffective as a violation of the duty of good faith. Nor can a mere technical consideration support a modification made in bad faith.

    Id.; see Restatement (Second) of Contracts § 89(a) cmt c. (Am. Law Inst. 1981).

  80. .Although contract law requires the parties to negotiate in good faith when they are modifying an existing contract, modifying a contract will be more difficult than negotiating a newly binding one because at least one of the parties is “guaranteed” to realize a return from the existing contract. Choi & Triantis, supra note 5, at 10–14.
  81. .The recent experience of Walgreens and Rite Aid is illustrative. Initially, the parties entered into a merger agreement, subject to regulatory and shareholder approval conditions. But when they faced a significant antitrust challenge, they decided to change the structure of the transaction so as to allow Walgreens to purchase a fraction of Rite Aid stores as an asset purchase instead of a merger. See Walgreens Boots All., Inc., Current Report (Form 8-K) (June 29, 2017) (stating in Item 7.01 that Walgreens had made a new agreement with Rite Aid to purchase stores and other assets—an agreement that replaced and revoked the previous merger agreements between the two companies).
  82. .Modifying or renegotiating the price term is considered to be more difficult because the price term deals more with how the contractual surplus is being divided between the parties. Hence, modifying the price term tends to create a zero-sum bargaining structure. By contrast, nonprice terms tend to affect both the size of the contractual surplus and its division.
  83. .See, e.g., Schwartz & Scott, supra note 28, at 703 (stating that one party’s investment may yield information that makes a project profitable, leading to a potential renegotiation between the parties).
  84. .Investment in due diligence yields a combination of symmetric and private information. For instance, a buyer of goods can uncover not only information that the seller knows but also some information that the seller does not know. The discussion in this paragraph concerns the private information revealed to the investigating party.
  85. .More precisely, if the due diligence investigation would produce relevant private information with less-than-certain probability and only the investigating party is aware of the acquisition of information, the counterparty (the uninformed party) would not know whether it is dealing with a party that indeed has the relevant information to increase the contractual surplus or with a party that does not have the information but is merely attempting to grab more contractual surplus.
  86. .Choi & Triantis, supra note 5, at 11–16.
  87. .More precisely, even if the informed party knows that the proposed terms will not increase the contractual surplus for both parties, given that the terms will allow the informed party to grab a larger fraction of the surplus and that the uninformed party’s rejection will simply allow the informed party to realize the profit under the initial terms, the informed party has little or no incentive not to (opportunistically) offer the new terms. The uninformed party, rationally expecting this, will likely reject all modification offers, and this will, in the end, prevent the parties from engaging in efficient renegotiation.
  88. .For instance, if the uninformed party, trusting the deterrence incentive based on the downside risk, were to always accept the modification proposal, there is no reason for the informed party not to act opportunistically. Hence, the uninformed party still needs to reject the modification proposal with some frequency.
  89. .In fact, the cost of dispute resolution (borne by the party behaving opportunistically) can function as an additional deterrent. At the same time, the dispute resolution cost on the uninformed party will function as a screen against frivolous or meritless litigation.
  90. .In a short essay in the first issue of the Negotiation Journal of the Harvard Law School Program on Negotiation, Howard Raiffa suggested a process by which parties who had entered into a negotiated settlement would jointly go to an intervenor (from a “Contract Embellishment Service”) who would propose an alternative that would replace the original one only if both parties would prefer it over the original. For her efforts, the intervenor might be paid a slice of the extra surplus she created. The intervenor would meet separately with each party and keep the content of conversations in strict confidence, in order to create incentives for honest disclosure. Howard Raiffa, Post-Settlement Settlements, 1 Negot. J. 9, 9–10 (1985). This Article addresses the more common version of this idea: the parties reach an agreement on the basics and then seek to increase the surplus through joint effort rather than retaining an intervenor.
  91. .“Without such legal recognition, parties would be obliged to expend enormous sums negotiating every detail of final contract documentation before knowing whether they have an agreement, and if so, on what terms.” Teachers Ins. & Annuity Ass’n of Am. v. Tribune Co., 670 F. Supp. 491, 499 (S.D.N.Y. 1987); see also Hwang, supra note 12, at 389 (“Some enforcement of a preliminary agreement means that the parties can rely on their preliminary bargains as they engage in the costly process of solving deal complexity.”). In client letters, law firms advise clients to “determine whether they have a meeting of the minds on the material terms of a deal before proceeding with the more detailed, prolonged, and costly effort of definitive documentation.” Colosimo et al., supra note 49, at 2. And, to “make sure they are on the same page as to the significant points of a deal before they undertake the time and expense to prepare a detailed contract to express their complete agreement.” Klingborg, supra note 49.
  92. .Schwartz & Scott, supra note 28, at 666.
  93. .Id.
  94. .Id. at 667.
  95. .Id.
  96. .Id. at 667 n.13.
  97. .Schwartz and Scott also write that:Rational parties will pursue efficient projects and abandon inefficient projects. They will disagree, if at all, over whether a party should be compensated for a reliance expense. If they disagree, they may call upon a court to resolve the dispute, and it should determine whether a promise to invest simultaneously has been breached and, if so, what fraction of the injured party’s reliance should be reimbursed.

    Id. at 667. The analysis assumes that the contracting parties are symmetrically informed regarding the contractual surplus, which, in turn, allows them to always adopt an ex post efficient arrangement. Given their premise that the parties can always implement ex post efficient arrangement, it is not surprising that the authors perceive the role of the court as simply to protect and encourage reliance investment. The analysis places much confidence in the parties’ ability to achieve ex post efficiency with the optimal deal terms. If the parties are not symmetrically informed, however, this premise does not hold. Choi & Triantis, supra note 5, at 4.

  98. .See, e.g., Empro Mfg. Co. v. Ball-Co Mfg., Inc., 870 F.2d 423, 425–26 (7th Cir. 1989) (noting that what the plaintiff characterized as “reliance expenditures” were merely expenditures “normally associated with pre-contractual efforts”). See supra note 52. Especially with respect to investments that are made before a contracting partner has been identified, parties could be properly incentivized in finding the right match even without contractual protection. See Lucian Arye Bebchuk & Omri Ben-Shahar, Precontractual Reliance, 30 J. Legal Stud. 423, 453, 456 (2001) (proposing that even without precontractual liability, parties will enter negotiations where there will be a positive surplus).
  99. .In explaining hold-up, contract theory conventionally assumes that (i) the investing party’s return from within the relationship, at the margin, must be strictly higher than the return from the market (outside the relationship), and (ii) both the outcomes of the investment (e.g., the value to the buyer and the cost to the seller) and the cost of the investment are nonverifiable (and, hence, noncontractible). See Oliver Hart, Firms, Contracts, and Financial Structure 73, 75 (1995) (noting the inherent difficulty of specifying some contract terms); Sanford J. Grossman & Oliver D. Hart, The Costs and Benefits of Ownership: A Theory of Vertical and Lateral Integration, 94 J. Pol. Econ. 691 passim (1986) (discussing how investment costs and outcomes are often nonverifiable and noncontractible); Oliver Hart & John Moore, Incomplete Contracts and Renegotiation, 56 Econometrica 755, 755–56 (1988) (identifying how certain elements of a deal are nonverifiable due to the prohibitive cost of specifying them).
  100. .Hwang, supra note 12, at 386; see also Victor P. Goldberg, Protecting Reliance, 114 Colum. L. Rev. 1033, 1042–45 (2014) (discussing the balance in the enforcement of preliminary agreements between protection of reliance and flexibility to adapt contract terms to new information).
  101. .See Hwang, supra note 12, at 393 (suggesting that since parties “almost always” sign a definitive agreement after signing a preliminary agreement, new information rarely reveals that they are not suitable partners).
  102. .In the contract-theory literature, this type of investment is known as “cooperative” investment. Yeon-Koo Che & Donald B. Hausch, Cooperative Investments and the Value of Contracting, 89 Am. Econ. Rev. 125, 125 (1999).
  103. .This is an example where the presence of asymmetric information and reliance investment creates both ex ante and ex post efficiency challenges. Choi & Triantis, supra note 5, at 3.
  104. .In light of the threat of holdup in contract design, it is curious that parties in many transactions jockey to be the authors of the first draft of an agreement. There are, of course, advantages to being the initial drafter, such as the familiarity of its team with the template document and the tendency of negotiations to anchor on offered terms. In addition, the initial drafter of the agreement may also acquire superior information about the deal structure that the party could exploit to its advantage.
  105. .BATNA stands for Best Alternative To a Negotiated Agreement. Roger Fisher & William Ury, Getting to Yes 101 (1981).
  106. .See David A. Lax & James K. Sebenius, The Manager as Negotiator: Bargaining for Cooperation and Competitive Gain 51, 55, 119–21 (1986) (discussing the concept of “reservation price” and how parties can alter counterparties’ perceptions to project a more favorable position).
  107. .G. Richard Shell, Bargaining for Advantage: Negotiation Strategies for Reasonable People 104–05 (3d ed. 2018).
  108. .See supra text accompanying note 4.
  109. .See, e.g., ABA, Model Merger Agreement, supra note 23, at 148–50 (providing a model no-solicitation or no-shop covenant). Even with a no-shop clause, however, the target corporation preserves the right to respond to an unsolicited offer from a third party. Whether the target corporation can do so depends, again, on the target board’s “good faith” determination that the third party’s offer may be superior. Hence, even with a no-shop clause, the good faith duty plays an important role. See id. at 169–71 (regarding a target corporation’s good faith duty to negotiate with the purchaser in response to a third party’s offer); see also Nat’l Venture Capital Ass’n, supra note 22, at 14 (including the obligation to negotiate in good faith within a term sheet’s no-shop clause).
  110. .But see SuperValu Inc. v. Associated Grocers, Inc., 428 F. Supp. 2d 985, 992 (D. Minn. 2006) (applying Washington law) (“A generalized duty to negotiate in good faith, as found in the language in the Letter of Intent, does not require disclosure of competing negotiations.”); Schwanbeck v. Federal-Mogul Corp., 578 N.E.2d 789, 798 (Mass. App. Ct. 1991) (citing Feldman v. Allegheny Int’l, Inc., 850 F.2d 1217, 1223 (7th Cir. 1988)) (stating that requiring negotiating parties to disclose their positions would be counter to the “cause of ebullient business competition”), rev’d on other grounds, 592 N.E.2d 1289 (Mass. 1992).
  111. .For instance, under the Restatement (Second) of Contracts § 89(a) (Am. Law Inst. 1981), contract modifications are binding when done “in view of circumstances not anticipated by the parties when the contract was made” and no consideration is necessary.
  112. .In the contract-modification context, see Varouj A. Aivazian et al., The Law of Contract Modifications: The Uncertain Quest for a Bench Mark of Enforceability, 22 Osgoode Hall L.J. 173, 186–92 (1984).
  113. .See, e.g., A‌/S Apothekernes Laboratorium for Specialpraeparater v. I.M.C. Chem. Grp., Inc., 873 F.2d 155, 158 (7th Cir. 1989) (“For instance, a party might breach its obligation to bargain in good faith by unreasonably insisting on a condition outside the scope of the parties’ preliminary agreement, especially where such insistence is a thinly disguised pretext for scotching the deal because of an unfavorable change in market conditions.”). See supra notes 34–49 and accompanying text.
  114. .96 F.3d 275 (7th Cir. 1996).
  115. .Id. at 279–80.
  116. .In cases in which a party demands the inclusion of a term omitted in the preliminary agreement, a more complicated determination is whether the omission of such term is “closed” in that sense. See, e.g., Teachers Ins. & Annuity Ass’n of Am. v. Tribune Co., 670 F. Supp. 491, 498–99 (S.D.N.Y. 1987) (declaring that parties may only negotiate open terms whose existence as such is recognized).
  117. .See Teachers Ins. & Annuity Ass’n of Am. v. Ormesa Geothermal, 791 F. Supp. 401, 414 (S.D.N.Y. 1991) (observing that despite many open terms, all of the “crucial economic terms” were set forth in the parties’ commitment letter); Tribune Co., 670 F. Supp. at 496 (noting the plaintiff’s contention that the open terms of “minor economic significance” included in the parties’ commitment letter by general reference, such as representations and warranties, closing conditions, and other covenants, did not render it nonbinding); Teachers Ins. & Annuity Ass’n of Am. v. Butler, 626 F. Supp. 1229, 1234–36 (S.D.N.Y. 1986) (stating that pursuant to their duty to negotiate in good faith, the parties are expected to negotiate the terms of open provisions, such as acceleration clauses, default prepayment fees, and lender’s remedies).
  118. .Tribune Co., 670 F. Supp. at 498.
  119. .Id. at 498–99 (concluding that the commitment letter represented a binding preliminary agreement, obligating the parties to pursue a final loan agreement by negotiating in good faith to resolve open terms). In Farnsworth’s typology, these commitment letters would likely be preliminary agreements with open terms, which would obligate the parties with respect to the settled (or closed) terms while requiring them to negotiate the open terms in good faith. See supra note 36.
  120. .Ormesa Geothermal, 791 F. Supp. at 402–03; Tribune Co., 670 F. Supp. at 496; Butler, 626 F. Supp. at 1236.
  121. .Butler, 626 F. Supp. at 1234.
  122. .Id. at 1235.
  123. .Tribune Co., 670 F. Supp. at 506. In the case, Tribune, the borrower, sold one of its real estate holdings in Chicago on an installment basis and wanted to offset that income with the borrowing from TIAA. Id. at 492. This was apparently for the purposes of making its balance sheet look healthy for its upcoming initial public offering. Id. at 492–93.
  124. .Id.
  125. .Ormesa Geothermal, 791 F. Supp. at 407–08.
  126. .Id. at 402–03.
  127. .Id. at 415; Butler, 626 F. Supp. at 1236; see also Tribune Co., 670 F. Supp. at 498 (underscoring that, “of course,” contract law’s aim is to “gratify, not to defeat,” expectations arising from an intended agreement, regardless of the need for further negotiations).
  128. .Tribune Co., 607 F. Supp. at 499 (“Without such legal recognition, parties would be obliged to expend enormous sums negotiating every detail of final contract documentation before knowing whether they have an agreement, and if so, on what terms.”).
  129. .It is true, however, that expectation damages in these cases mimic a reliance measure that accounts for TIAA’s foregone opportunity to lend the money elsewhere at the time it made the commitment. See supra note 53.
  130. .A classic example is the doctrine of commercial impracticability as explained in Richard A. Posner & Andrew M. Rosenfield, Impossibility and Related Doctrines in Contract Law: An Economic Analysis, 6 J. Legal Stud. 83, 90–91 (1977).
  131. .See, e.g., Larwin-S. Cal., Inc. v. JGB Inv. Co., 162 Cal. Rptr. 52, 54 (Cal. Ct. App. 1979) (noting that the buyer’s approval of its engineering report was a closing condition).
  132. .SIGA Techs., Inc. v. PharmAthene, Inc., 67 A.3d 330, 350–51 (Del. 2013).
  133. .Id. at 334.
  134. .Id.
  135. .Id. at 335.
  136. .Id. at 337.
  137. .Id. at 337–38.
  138. .Id. at 339.
  139. .Id. at 338–39.
  140. .Id. at 339.
  141. .Id. at 339–40.
  142. .Id. at 340.
  143. .Id. at 340–41.
  144. .Id. at 347.
  145. .Id. at 350–51.
  146. .See supra notes 90–97 and accompanying text.
  147. .Restatement (Second) of Contracts § 347 (Am. Law Inst. 1981) (emphasis added).
  148. .The causation requirement in contract law is sometimes considered to follow that in tort law (including both but-for and proximate causation) and is also subject to mitigation and foreseeability limitations. See, e.g., Wright v. St. Mary’s Med. Ctr. of Evansville, Inc., 59 F. Supp. 2d 794, 799 (S.D. Ind. 1999) (“Causation-in-fact is generally a prerequisite to recovery in both contract . . . and tort . . . .”). But see 3 Farnsworth, supra note 53, at § 12.01 (“[C]ausation in contract does not necessarily correlate with causation in tort . . . .”). Proving causation in contract law is typically not a serious hurdle for the plaintiff (consider, for instance, a seller who breaches by not delivering a promised product or by delivering a defective one). It is a more significant issue when the promise is a standard, such as good faith or best efforts (suppose the seller promises to exercise its best efforts to deliver a conforming product).
  149. .Restatement (Second) of Contracts § 352 (Am. Law Inst. 1981).
  150. .Id. § 349 cmt. a.
  151. .According to the court in Venture Associates Corp. v. Zenith Data Systems Corp., 96 F.3d 275 (7th Cir. 1996), damages for breach of an agreement to negotiate in good faith[m]ay be, although they are unlikely to be, the same as the damages for breach of the final contract that the parties would have signed had it not been for the defendant’s bad faith. If, quite apart from any bad faith, the negotiations would have broken down, the party led on by the other’s bad faith to persist in futile negotiations can recover only his reliance damages—the expenses he incurred by being misled, in violation of the parties’ agreement to negotiate in good faith, into continuing to negotiate futilely. But if the plaintiff can prove that had it not been for the defendant’s bad faith the parties would have made a final contract, then the loss of the benefit of the contract is a consequence of the defendant’s bad faith, and provided that it is a foreseeable consequence, the defendant is liable for that loss—liable, that is, for the plaintiff’s consequential damages.

    Id. at 278.

  152. .The case of Williams Cos. v. Energy Transfer Equity, L.P., 159 A.3d 264 (Del. 2017), is exemplary. In that case, Energy Transfer Equity (ETE) had an obligation to put in “reasonable best efforts” in obtaining a tax opinion before the parties could close the acquisition transaction. Id. at 268. When ETE did not obtain the required opinion and the transaction fell apart, Williams sued for breach of contract. Id. at 269–70. Although the court ruled that ETE likely breached its “reasonable best efforts” obligation, the failure did not “materially contribute” (i.e., did not cause) to the failure of the closing condition. Id. at 267–68.
  153. .Brown v. Cara, 420 F.3d 148, 151 (2d Cir. 2005). Based on the “work together” language, the court stated that:We cannot imagine more clear evidence of an intention to be bound to the MOU [Memorandum of Understanding] as a general framework in which the parties will proceed in good faith toward the goal of developing [the real estate joint venture] while preserving for later negotiation the specific details of necessary business, design, construction, financing, and management terms.

    Id. at 158.

  154. .Id. (stating that although the plaintiff was entitled to demand that the defendant negotiate remaining terms in good faith, “considerable efforts” would be required to reach a definitive agreement).
  155. .Id. at 155 (stating that “the MOU reaches almost none of the terms . . . that require negotiation”). The parties settled after the case got remanded to the lower court following the circuit court’s failure to indicate the appropriate damages remedy. Victor P. Goldberg, Brown v. Cara, the Type II Preliminary Agreement, and the Option to Unbundle, in Rethinking Contract Law and Contract Design 207, 210 (2015).
  156. .See Schwartz & Scott, supra note 28, at 667, 702–04 (arguing that, by using the enforcement of good faith to protect verifiable investments, the parties encourage exploration of potentially profitable ventures). According to Farnsworth, the appropriate remedy should depend on whether the parties have a binding preliminary agreement with open terms (Type 1) or a preliminary agreement with a duty to negotiate in good faith (Type II). Farnsworth, supra note 3, at 286. While expectation damages would be appropriate in the first case, Farnsworth argues that reliance damages would be more appropriate since there is “no larger agreement” between the parties and there is no way of knowing what its terms would have been or whether the parties would ever have arrived at an ultimate agreement. Id. at 263. Knapp also states that the greater uncertainty with respect to damages that flow from the breach of a “contract to bargain” makes expectation damages unlikely. Knapp, supra note 3, at 723. He also states, however, that “[i]n some cases . . . the main terms of performance, including quantity, quality and price, may have been so agreed upon by the time of breach that an expectation remedy can be computed with as much certainty as is usually required.” Id.; see also Bebchuk & Ben-Shahar, supra note 98, at 451 (noting that by restricting recovery to sunk investments, courts can provide remedies even where terms are “severely incomplete”).
  157. .One could argue that acting in “bad faith” is more akin to being either “grossly negligent” or even “reckless,” because bad faith actions usually accompany bad or ill intentions. Under tort law, when a tortfeasor’s “reckless” action causes harm, the victim may be able to recover punitive damages. By contrast, under contract law, punitive damages are not allowed unless the breach also constitutes a tort. Restatement (Second) of Contracts § 355 (Am. Law Inst. 1981). This Article is not arguing that breach of duty to negotiate in good faith should constitute a tort. Hence, we focus on compensatory, and not punitive, damages.
  158. .On the flip side, when the defendant shows that it acted in good faith, even if the negotiations fell apart and the plaintiff did not get to realize any contractual surplus, the defendant will not be liable. Similarly, under the negligence-based liability in tort law, when the defendant shows that it was not negligent, the plaintiff will not recover for any harm it suffered.
  159. .See supra subsection II(C)(4)(a).
  160. .The other TIAA cases, cited supra note 117, are similar in this respect.
  161. .According to the court:[W]here the parties have a Type II preliminary agreement to negotiate in good faith, and the trial judge makes a factual finding, supported by the record, that the parties would have reached an agreement but for the defendant’s bad faith negotiations, the plaintiff is entitled to recover contract expectation damages.

    SIGA Techs., Inc. v. PharmAthene, Inc., 67 A.3d 330, 350–51 (Del. 2013) (emphasis added).

  162. .For instance, under either a requirements or an output contract, the Uniform Commercial Code imposes a good faith obligation on the party with discretion over the quantity. U.C.C. § 2-306 (Am. Law Inst. & Unif. Law Comm’n 2017). If the party, for instance, refuses to make or take any delivery, the court will have to deal with similar issues of: (1) whether the refusal was in bad faith (even though making or taking no delivery in good faith is allowed) and (2) what the appropriate remedy is. Similar issues will arise in an exclusive agency setting (without a minimum sale condition) where the agent has an obligation to put in best efforts in making sales. In another context, a typical merger agreement, the covenants section is replete with standard-based obligations. It may require the target corporation to exert reasonable best efforts (or other similar effort standard) in securing shareholder approval, disclosing material information to the acquirer, obtaining regulatory approval, filing the necessary forms with the SEC, and otherwise dealing with the buyer to consummate the transaction. Similarly, the section will typically obligate the buyer to exert reasonable best efforts in obtaining regulatory approval, obtaining necessary financing, and getting certain securities filings (such as registration statement S-4) approved by the SEC. See, e.g., Akorn, Inc. v. Fresenius Kabi AG, No. 2018-0300-JTL, 2018 WL 4719347, at *91 (Del. Ch. Oct. 1, 2018) (noting that the standard for reasonable best efforts requires a party to take all reasonable steps to consummate the transaction), aff’d, 198 A.3d 724 (Del. 2018) (unpublished table decision); ABA, Model Merger Agreement, supra note 23, at 165, 189–90 (demonstrating that merger agreement covenants typically require both the seller and the buyer to use commercially reasonable efforts in filing forms with the SEC, seeking regulatory approval, and dealing with each other).
  163. .See Williams Cos. v. Energy Transfer Equity, L.P., 159 A.3d 264, 273 (Del. 2017) (discussing the parties’ obligation to put in reasonable efforts to obtain a Section 721 tax opinion); see also Verified Complaint at 6–7, Anthem, Inc. v. Cigna Corp., No. 2017-0114-JTL (Del. Ch. filed Feb. 17, 2017) (accusing Cigna of breaching the covenant to exert “reasonable best efforts” in securing antitrust approval and “sabotaging” the merger); Verified Complaint at 44–45, Cigna Corp. v. Anthem, Inc., No. 2017-0109-JTL (Del. Ch. filed Feb. 17, 2017) (accusing Anthem of failing to take “reasonable best efforts” to satisfy closing conditions as required by the merger agreement).
  164. .If the break-up fees are not stipulated (or determined by the court) to be the “sole and exclusive” remedy, the plaintiff can seek expectation damages or even specific performance for the other’s breach of covenant. ABA, Model Merger Agreement, supra note 23, at 289.
  165. .See, e.g., Glob. Asset Capital LLC v. Rubicon US REIT, Inc., No. 5071-VCL, at *5 (Del. Ch. Nov. 16, 2009) (order granting temporary restraining order) (explaining that parties enter into preliminary agreements to create legal rights); ABA, Model Merger Agreementsupra note 23, at 22 (stating that the target may require the buyer to deposit shares of the buyer’s common stock and cash prior to the merger); Nat’l Venture Capital Ass’nsupra note 22, at 1 (providing in the term sheet a limited set of alternatives for distributing dividends).
  166. .See, e.g., Hwang, supra note 12, at 393 n.57 (“Dealmakers [that were interviewed] with a wide breadth of experience—at firms and in-house, working with repeat players and one-off deal parties, in private and public deals, in a variety of firms and cities, representing financial parties and strategic parties—report that preliminary agreements have exceptional binding power.”).
  167. .See 2 Comm. on Negotiated Acquisitions, supra note 73, at 107 (“The parties may also feel morally, if not legally, obligated to key terms [in a letter of intent] if those terms are set down in writing.”); Freund, supra note 62, at 60 (operating as “anti-renegotiation insurance,” a letter of intent represents an “explicit moral obligation” between the parties); Lena G. Goldberg & Mary Beth Findlay, Harvard Bus. Sch., Just an MOU or a Real Deal? 3 (2011) (“Unless there has been a material adverse change, it may be difficult, psychologically, to renegotiate terms that are expressly included in an MOU.”); Farnsworth, supra note 3, at 258 (noting that preliminary agreements make it harder for parties to let negotiations fail, especially when the prospect of success has been made public); Knapp, supra note 3, at 679 (arguing that business persons consider themselves bound, if not legally, at least morally or ethically).
  168. .Mark K. Johnson, Enforceability of Precontractual Agreements in Illinois: The Need for a Middle Ground, 68 Chi.-Kent L. Rev. 939, 939–40 (1993).
  169. .See Butler v. Balolia, 736 F.3d 609, 615 (1st Cir. 2013) (“[P]arties may wish to build in safeguards . . . . This can be accomplished by binding themselves sufficiently such that they feel comfortable investing resources into the deal, but without inextricably committing themselves to a transaction that is still inchoate. Contracts to negotiate can satisfy this need.”); Ben-Shahar, supra note 57, at 407 (“The precontractual commitment enables a party to commit to a specific partner and a specific negotiation protocol without committing to specific terms.”).
  170. .Hwang, supra note 12, at 378–79, 405.
  171. .See supra notes 92–97 and accompanying text.
  172. .Gilson et al., supra note 28, at 1385–86.
  173. .Jonathan M. Barnett, Hollywood Deals: Soft Contracts for Hard Markets, 64 Duke L.J. 605, 607 (2015).
  174. .Id. at 607–08.
  175. .Id. at 608–09.
  176. .Id. at 644.
  177. .Id. at 640–41.
  178. .Id. at 641.
  179. .Scott Baker & Albert Choi, Contract’s Role in Relational Contract, 101 Va. L. Rev. 559, 562–63 (2015). The analysis shows that while reputational sanctions impose the deterrence benefit–cost ratio of one, legal sanctions, by contrast, can allow the parties to achieve a better deterrence benefit–cost ratio, especially given that the plaintiff would be willing to bring suit only when the expected return from litigation is higher than the expected cost. Id. at 601–02; see also Scott Baker & Albert H. Choi, Reputation and Litigation: Why Costly Legal Sanctions Can Work Better than Reputational Sanctions, 47 J. Legal Stud. 45, 47 (2018) [hereinafter Baker & Choi, Reputation and Litigation] (showing how legal sanctions generate an “inframarginal” deterrence benefit that reputational sanctions lack as well as generalizing and expanding the analysis on the informational role played by legal sanctions).
  180. .Legal and reputational sanctions can, therefore, be complements and not substitutes. Baker & Choi, Reputation and Litigation, supra note 179, at 47.
  181. .Omri Ben-Shahar focuses on incomplete agreements where the parties deliberately leave terms to be agreed upon later. Ben-Shahar, supra note 57, at 390. Instead of having the court police the specific investments of the parties, he advances a novel proposal that the party seeking enforcement of a deliberately incomplete agreement would have an option to enforce the transaction under the agreed-upon terms supplemented by terms that are the most favorable (within reason) to the counterparty. Id. Ben-Shahar argues that this facilitates sequential negotiations and deters unilateral retractions or threats to precontractual investments. Id. at 392.
  182. .At the same time, protecting each party’s investment through reliance damages, when not appropriately limited, could lead to overreliance. See Bebchuk & Ben-Shahar, supra note 98, at 434 (explaining that a strict liability regime for reliance damages would likely result in “excessive” reliance investment). In contract law, this issue is partially dealt with by the requirement that reliance must have been reasonable.
  183. .Reliance damages deal largely with precontractual or ex ante behavior and are not suited to tackling ex post (renegotiation) inefficiency. Optimally calibrated reliance damages can achieve ex ante efficiency, but the parties will still have to deal with the hurdles of achieving ex post efficiency, which reliance damages can address only indirectly.
  184. .See supra notes 151–55 and accompanying text. Given that the optimal damages may sometimes be less than expectation damages, expectation damages can serve as a penalty default rule, providing incentive for the parties to contract around the default of expectation damages, for instance, through liquidated damages.
  185. .As discussed in subpart II(D) on remedies, this is consistent with how certain courts (such as in Tribune Co. and SIGA) have been willing to grant expectation damages, conditional on proving causation (subject to foreseeability and mitigation principles) and reasonable certainty. Of course, if the court believes that encouraging beneficial reliance is a more important goal, then the court can award reliance, rather than expectation, damages.
  186. .Good faith is inherently ambiguous and calls for investigation into the parties’ intent, thereby opening the door to costly consideration of evidence extrinsic to the contract document. See, e.g., JamSports & Entm’t, LLC v. Paradama Prods., Inc., 336 F. Supp. 2d 824, 848 (N.D. Ill. 2004) (“The concept of good faith appears also to require an inquiry into the breaching party’s intent.”). According to the Uniform Commercial Code, good faith means “honesty in fact and the observance of reasonable commercial standards of fair dealing.” U.C.C. § 1-201(b)(20) (Am. Law Inst. & Unif. Law Comm’n 2017). Notwithstanding the cost and uncertainty of the necessary factual determination, complex commercial contracts are replete with such standard-based obligations. See, e.g., ABA, Model Merger Agreement 124, 189–90 (obligating the target corporation to use “commercially reasonable efforts” in preserving intact the target’s business, submit regulatory filings “as promptly as practicable,” and take “all actions necessary” in consummating the merger).
  187. .See supra notes 162–63 and accompanying text for a discussion of such open-ended obligations in corporate-acquisition contracts.
  188. .Of course, the costs of litigation and unbiased judicial error are avoided when parties settle their dispute, as they often do. Even in cases where more precise and predictable arrangements would have been possible, we have suggested that commercially sophisticated parties would still opt for a vague standard. Choi & Triantis, Costly Verification, supra note 9, at 504, 519–20; Choi & Triantis, Strategic Vagueness, supra note 9, at 854–55.
  189. .The probability of 1‌/3 captures both false positives (type I error) and false negatives (type II error).
  190. .While the conventional treatment of verifiability is binary (a factor is either verifiable or it is not), our analysis assumes that the verification is costly, but not impossible, and that the cost of verification includes: (1) possibly erroneous determination and (2) litigation cost. Although liquidated-damages provisions are uncommon in intermediate agreements with the duty to negotiate in good faith, the parties sometimes stipulate expense reimbursement or other similar remedy provisions. Contracting parties often back standard-based obligations, such as an obligation to exert reasonable best efforts in securing regulatory approval, with liquidated damages in other types of agreements. See supra notes 162–64 and accompanying text.
  191. .In case the promisor has breached, promisee’s expected return from litigation is (2‌/3)d − l; whereas if the promisor has not breached, the expected litigation return is (1‌/3)d − l. When (2‌/3)dl > (1‌/3)d, the first expression is positive while the second expression is negative. That is, the promisee will bring suit only when the promisee knows that the promisor has breached. Note also that when desired screening is achieved, the cost of dispute resolution also plays an important deterrence role. This is because the promisor who breaches will incur the cost of for certain, while the promisor who does not will incur no dispute resolution cost.
  192. .In corporate acquisitions, for instance, when the parties agree to resolve their disputes over valuation using arbitration overseen by independent accountants, the parties can tailor their procedures and rules of evidence. See LJL 33rd Street Assocs., LLC v. Pitcairn Props. Inc., 725 F.3d 184, 194 (2d Cir. 2013) (acknowledging that when parties agree to arbitrate “it is indisputably correct that arbitrators are not bound by the rules of evidence”).
  193. .In this sense we are allied with Barnett, supra note 173, at 635, who similarly sees virtue in probabilistic enforcement; but we endorse the use of substantive standards in the terms (good faith, best efforts) rather than the uncertainty from incomplete formalization of the deal.
  194. .E.g., Choi & Triantis, Costly Verification, supra note 9; Choi & Triantis, Strategic Vagueness, supra note 9.
  195. .In an ideal world, where the court does not err and litigation is costless, full compensation for the injured party also achieves efficient deterrence. Compensatory damages may no longer be efficient when the court is likely to err and litigation is costly, and this is where the screening and deterrence objectives play a more prominent role.