Printer Friendly

Fixing unfair contracts.

INTRODUCTION
I. THE PROBLEM OF UNFAIR TERMS
   A. Unfair Terms
   B. Three Solutions
      1. The most reasonable term
      2. The most unfavorable term
      3. The minimally tolerable term
II. MINIMALLY TOLERABLE TERMS: A CONCEPTUAL INTRODUCTION
   A. Minimally Tolerable Terms and the Freedom to Bargain
   B. Grounds for Intervention
   C. Cross-Doctrinal Foundations
   D. Identifying the Tolerable Threshold
III. DOCTRINAL APPLICATIONS
   A. Severability Clauses
   B. The Doctrine of Partial Enforcement
   C. Unconscionability
      1. Price and interest rates
      2. Arbitration clauses
   D. Liquidated Damages
   E. Conditions in Insurance Contract
IV. WHY ARE MINIMALLY TOLERABLE TERMS USED?
   A. Market Power and Bargained-For Advantage
   B. Unclear Boundaries
   C. Hidden Advantage and Bad Faith
   D. Duress and Contractual Hold-Up
CONCLUSION


INTRODUCTION

Imagine the following situation: As a student in a law school course, you are regularly assigned a daily reading load of 20 pages in preparation for the next day's class. Today, however, you received an extraordinary assignment of 200 pages, clearly more than you can feasibly prepare in one day. What should you do? What would be a principled mode of action, given that you cannot renegotiate or otherwise modify the command? Should you resort to the normal, "reasonable" practice of preparing 20 pages out of the assigned 200? Or should you, perhaps, disregard the unreasonable command of 200 altogether and in the absence of any other affirmative instruction read 0 pages? Or, perhaps yet, should you disregard only the unreasonable excess of the command and prepare the maximum level that is still tolerable, of say 50 pages?

This dilemma, I argue, is similar to one that is at the core of several basic doctrines of contract law. While this metaphorical dilemma is not set in a bargained-for exchange but instead in a hierarchical relationship, it captures the challenge created by one-sided, unfair commands. When a contract contains a term that is excessive or unfair, the law might consider it unenforceable, but it also has to set a substitute provision. Should the unfair term be replaced by the most reasonable majoritarian term (analogous to 20 pages in the above hypothetical)? Should the dictating party be "punished"--incentivized not to go too far--by replacing the unfair term with something least favorable to her (analogous to 0 pages)? Or, should the unfair term be reduced merely to the highest level that the law considers tolerable (analogous to 50 pages)?

This Article explores the problem of how to fix unfair terms and illustrates its solutions in existing law. It is an everyday problem that comes up in numerous settings, but, surprisingly, it has not received a systematic theoretical treatment. There is no single compelling approach to this problem and, indeed, as Part I shows, all three solutions--and only these three--can be traced across different contexts and legal traditions. Still, the analysis in this Article focuses primarily on the third regime--the one that intervenes to the least extent and fixes the unfair term only so much (or so little) as to make it minimally tolerable. This regime has received less notice in commentary than it deserves.

It might plausibly be conjectured, before reading this Article, that the minimally tolerable term criterion is esoteric, an academic curiosity at best. Surely, so goes the conjecture, if a court takes the trouble to correct an unfair term in the contract, it would "finish the job": it would naturally replace the offensive term with the most reasonable alternative, not with a barely tolerable one. Why let a party who overreached get away with the maximum allowable advantage? This Article demonstrates, perhaps surprisingly, that the minimally tolerable term criterion is quite prevalent. It shows how courts use it in a variety of legal contexts as a mainstream solution to the problem of unfair and unconscionable terms. One such doctrine is partial enforcement (or its archaic predecessor, the "Blue Pencil Rule"), and its application in the context of covenants not to compete. I show that when noncompete clauses are excessive, they are generally brought down to the maximum allowable level. Another doctrine studied here is unconscionability: what do courts do when a distributive term is struck as unconscionable? Quite often, it turns out, the vacated term is replaced with the minimally tolerable one, most favorable to the strong party. Yet another example involves the judicial supervision of liquidated damages. In some legal traditions, excessive liquidated damages are reduced to the maximum allowable level--the measure closest to the agreed sum, such that if it were the one agreed upon in the first place, the court would have enforced it.

Before it surveys the doctrinal prevalence of the minimally tolerable term criterion, this Article sets up the context in which the criterion operates, and provides a theoretical foundation for this regime. Part I identifies the minimally tolerable regime as one of three discrete and conceptually coherent solutions to the problem of how to fix unfair terms. Part II then analyzes the conceptual grounding of the minimally tolerable term criterion--the legal principles with which it is consistent and how it ties with other practices in various areas of the law. It shows that what underlies the minimally tolerable principle is a specific conception of severability (or divisibility) of contractual provisions, and that the principle of waiver can lead to a similar solution. Further, this Part demonstrates the prevalence of similar solutions in other areas of the law outside contracts.

Part III offers a brief cross-doctrinal survey and invites the reader to recognize that the minimally tolerable principle has broader application than one might expect. Recognizing the instances in which this regime applies and when it is rejected sets the stage for the normative inquiry in Part IV, which suggests several justifications for this practice. The normative defense is anything but straightforward. Admittedly, there is something objectionable about a legal rule that accords the strong party, who went too far in exploiting its bargaining power, the maximum allowable advantage. Why not reform the contract more aggressively? If courts already step in, why not take the opportunity to undo the effects of unfettered bargaining power? Legal solutions that favor the weak party, that level the playing field, are usually more appealing. Moreover, how do courts account for the bad incentives that minimally tolerable terms generate? Namely, the incentive to draft excessive terms, knowing that at most courts will only strike down the excrescence?

The key observation made here, in justifying the use of minimally tolerable terms, is that the problem of fixing unfair contracts is merely a species of gap filling. The court that vacated an unfair term has to decide how fill the newly created gap. If the court merely provides a gap filler, it cannot be too ambitious and it cannot undo the effect of uneven bargaining power (the source of which is unpacked in Part IV), or else its policy might backfire.

Indeed, the theme of this Article is part of a more general thesis concerning a principle of gap filling in contracts between parties that have unequal bargaining, or market, power. This thesis, which I develop elsewhere, (1) suggests that standard gap-filling approaches do not provide a workable prescription when the gap in the contract involves a purely distributive aspect that parties bargain over, such as price. The standard gap-filling criterion prescribes terms that are "majoritarian"--those that are used and sought by most parties. In law and economics, this criterion equates with surplus-maximizing gap fillers, because parties would naturally prefer provisions that accord them maximum joint gains, which they can then divide in any agreed-upon way by setting the price. (2) The problem with this surplus-maximization criterion is that it offers no recipe in the case in which the missing term is the price itself, or some other purely distributive clause. Missing prices cannot be filled with surplus-maximizing terms because more than one term maximizes the surplus. In fact, the choice of price is surplus-neutral.

To resolve this indeterminacy, a new principle of gap filling is needed, one of mimicking the bargain: the division of the surplus that would have been struck between these parties, given the allocation of market power. Here, courts cannot rely on average market prices because those do not mimic the bargain that would have been explicitly struck between these parties. This principle instead requires courts to fill gaps with terms that are sensitive to the division of bargaining power, and are more favorable to the party with the greater bargaining power. Any other method of filling gaps in distributive terms would be undone by the parties, at an increased transaction cost. If gap fillers do anything other than mimic the term this party could have dictated, they will have the ex ante effect of inducing this party to dictate the term in order to preempt any adverse allocation that would otherwise result from the gap filler. Perhaps even more than in other contexts, when the distribution is at stake it is likely that the stronger party will insist on contracting around a nonmimicking gap filler.

Thus, continues the argument, if courts have to fill a gap that arises from the elimination of an unfair term, the bargain-mimicking conception would dictate the term closest to the hypothetical bargain. More aggressive intervention would fail to achieve any redistributive results that might superficially or naively underlie it, and would be circumvented by the parties while imposing some deadweight loss.

In the end, though, this Article recognizes that any normative justification for the minimally tolerable term must account for the incentive problem and for the concern that this rule would induce strong parties to deliberately overreach and draft overly excessive terms, counting on courts to at worst remove only the excess. The analysis concludes by showing that this concern limits the application of the regime, both in theory and in practice.

This Article intends to fill a vacuum in the study of unconscionability and related doctrine. Much ink has been spilled on the question of what constitutes an unfair contract. In the last decade this question returned to the fore when various types of mandatory arbitration terms and other terms in consumer contracts were held to be unconscionable. But no systematic discussions emerged regarding the remedial aspect--how to repair contracts that contain excessively one-sided terms--and courts reach solutions that are ad hoc and inconsistent.

This Article provides a conceptual framework to consider this problem (by identifying the three competing solutions) and takes a first shot at justifying one possible fix.

I. THE PROBLEM OF UNFAIR TERMS

A. Unfair Terms

Contracts contain terms that design the transaction (for example, the quality and quantity of the goods) and terms that divide the surplus (for example, price). Many terms affect both the size of the surplus and its division, but the presence of pure distributive terms makes it possible to focus on the distributive effect of the contract in isolation. Specifically, my interest is in contracts that are distributively unfair--that allocate too much of the surplus to one party and too little (perhaps even negative payoff) to the other. While it is a basic premise of contract law that courts ought not evaluate or even inquire into the "adequacy of consideration" (3)--namely, what each party received from the deal--many doctrines do in fact police the distribution of surplus.

There are situations in which the one-sidedness of the deal is considered intolerable and courts are willing to step in. This might be true even for simple, easy-to-understand terms such as price, although such instances are extremely rare. (4) Intervention is more likely to occur when the excessive terms are less conspicuous than the price and are less well understood by the weak party, suggesting that flaws existed in the manner in which assent was reached. This may be the case for late payment terms and late fees, disclaimers, and exclusionary clauses. Intervention is even more justified when the boilerplate terms frustrate the induced legitimate expectations of the weak party--a phenomenon that is of particular concern in insurance contracts. (5) More recently, an increasing number of courts find arbitration terms in consumer and employment contracts, which are excessively favorable to the drafters (sellers or employers), to contain unconscionable elements. (6)

The unconscionability doctrine is not the only way to limit unfair terms in contracts. In some specific areas, terms may be struck without reference to the unconscionability standard. For example, liquidated damages that are clearly overcompensatory are considered punitive and unenforceable. (7) Although the standard used to determine what constitutes unreasonably large liquidated damages can at times be identical to the "shock the conscience" standard of unconscionability, (8) it is generally less strict. In another area, covenants not to compete with a business or an employer may be found intolerable if the duration or the geographic scope of the noncompete obligation is too long or too broad. (9) Finally, in some areas of contracting there are statutory caps or standards that determine the maximum allowable advantage a party can bargain for. Usury laws and price gouging acts are typical examples of a maximum constraint; lemon laws and minimum wage laws are examples of minimum constraints. If the contract contains bargained-for terms that are outside of these regulatory limits, the terms may be struck.

Much ink has been spilled on the question of what is and ought to be the limit to private contracting and what constitutes unfair terms. It is not my intent to contribute to this inquiry. Rather, assuming that a court applies any given fairness-based constraint on private contracting and invalidates terms that violate these constraints, an additional inquiry is needed. Once the unfair term is invalidated, there is a gap in the contract that calls for gap filling. How should this gap be filled?

It should be noted, though, that courts have not always been ready to fill gaps that arise from invalidated illegal terms and historically elected not to do so. Instead, when a contract contained an unconscionable element, or was unfair due to coercion, the entire contract was rendered unenforceable. (10) It was not considered the role of the courts to write the contract over in a more reasonable fashion. Even today, when the unconscionability of some terms is linked to flawed assent, as in the case of duress or fraud, courts may refuse to reform the term and instead vacate the entire agreement. (11) Beyond such nonenforcement consequences, overreaching can give rise to civil and criminal penalties such as disgorgement, fines, and damage multipliers. (12)

Modern courts have less trouble reforming the contract and enforcing it. Unconscionability statutes provide clear authority to do so. For example, section 2-302 of the Uniform Commercial Code allows courts to "limit the application of any unconscionable term as to avoid any unconscionable result." (13) Even more explicitly, under the Principles of European Contract Law, a court may "adapt the contract in order to bring it into accordance with what might have been agreed had the requirements of good faith and fair dealing been followed." (14) There remains some debate as to whether courts can affirmatively replace the offending term with a different one not drafted by the parties, or whether they are restricted to crossing out existing terms and letting the legal gap fillers set in. Part III reviews some of this debate. It demonstrates that many courts believe that they have sufficient authority to reduce an excessive term to any level they deem appropriate.

When a court decides to enforce the contract and reform the unfair term, it has to apply some principled policy in choosing the new term. True, this is not the "pure" gap-filling scenario of a contract that contains a lacuna and needs supplementation. Rather, this is an artificial gap that arises out of a legal policy that eliminates an existing express term. Nevertheless, this is a situation in which technically there is no longer an express provision in an otherwise enforceable contract and a new provision needs to be supplied. In the same way that courts need to turn to a gap-filling methodology when they knockout nonmatching express terms in a battle of the forms, (15) courts need to follow a systematic gap-filling pattern when repairing unconscionable terms.

B. Three Solutions

When the excessive term is struck down and a gap is created in the contract, how should this gap be filled? Courts generally follow one of three possible approaches.

1. The most reasonable term

The standard criterion for filling gaps in contracts is to supply the most reasonable, majoritarian term. While this criterion is more often identified with gap filling in indefinite contracts or as a solution to the battle of the forms, it is also a sensible solution to filling the gap arising from the invalidated excessive term. Thus, if the price is too high, replace it with a reasonable, intermediate-level market price. (16) If a liquidated damages remedy is excessive and punitive, replace it with standard expectation damages, measured by the contract-cover differential or by lost profits, excluding any uncommonly high consequential damages. If an interest term is usurious, replace it with the market rate.

Often, this solution of supplying the most reasonable midrange term arises implicitly. A court might strike the offensive term and make no affirmative substitution. The default rule would then apply, effectively supplying a reasonable provision. For example, if an unconscionable arbitration term is struck, the court need not select an affirmative gap filler; in the absence of an arbitration provision, parties resolve their dispute in court. If a disclaimer of warranty in a sales contract is invalid, it is struck and in its absence the standard implied warranty of merchantability applies. Or, if an exclusionary clause in an insurance policy is unconscionable, it falls and the insurer's obligation is read without it. Here, even if the court does not pay explicit attention to the principle underlying the gap-filling process, it often achieves an outcome that is consistent with a most reasonable term approach to default rules.

There is much to be said, of course, in support of this regime. The most reasonable terms are, by definition, the most compelling ex post solution. Contractors are allowed to deviate from them, within limits, and a court would normally let such deviations stand. But if the court deemed the situation fit for intervention, why not take the opportunity and write the most balanced contractual term for the parties? Moreover, there is usually good evidence to identify the most reasonable, majoritarian terms, rendering this approach easy to administrate.

2. The most unfavorable term

If the drafting party overreached by trying to secure an unfair gain, the court can "punish" this behavior by depriving this party of the entire advantage. The contract is enforced, but the excessive term is supplemented by the term least favorable (within reason) to the drafting party. Thus, for example, if a creditor bargained for an excessive, usurious interest rate, a court can replace it with zero percent interest. If the duration of a noncompete clause in an employment contract is excessive, the court can strike the clause altogether, effectively replacing it with zero duration. In fact, in some scenarios the example mentioned above of a vacated arbitration term can be regarded as an illustration of the most unfavorable term approach. The most reasonable gap filler, it can be argued, is a fair and balanced arbitration arrangement that would replace the unfair arbitration term that was struck. Eliminating arbitration altogether and sending the dispute to court is a way to punish the overreaching party and tilt the result in favor of the other party.

This gap-filling approach is punitive in the same way as the doctrine of contra proferentem, which construes an ambiguous contract term against the party who drafted or imposed it. It is not intended to identify the most balanced outcome, ex post, nor to follow the most common market arrangement. Instead, it is intended to induce the drafter to make drafting choices that would not overreach and would not necessitate court intervention in the first place. It is a species of penalty default rule, and as such it is consistent with the same policy concerns that supposedly justify penalty rules: forcing a party who enjoys a bargaining advantage (here, superior bargaining power) to forgo some of the gains that he can extract. In fact, the punitive principle can be taken one step further, imposing punitive damages or other sanctions on the drafting party. Some consumer protection acts follow this route. (17)

3. The minimally tolerable term

Finally, if a term is considered unfair, it can be broken down to two distinct components: the allowable portion, and everything beyond it. Once the second component--the excessive increment--is eliminated, the remainder is no longer unfair or unconscionable (even if still relatively one-sided), and does not necessitate further intervention. This remainder--the minimally tolerable term--would be enforceable.

To compare the three solutions, consider a situation in which the reasonable price of a service is $500. If purchased under conditions of a thick market-where there are many competing sellers and prices converge to a competitive equilibrium level--the price would always be $500. A buyer that was charged more was probably the victim of fraud. But situations arise in which one of the parties may experience urgency or vulnerability, or, alternatively, enjoy bargaining leverage, such that the price for the same service may reflect those circumstances. Assume that it is not unreasonable to charge as much as $750 to a buyer in need, or to pay as little as $250 to a desperate seller. Thus, any price within the range of $250 and $750 would be enforceable. But what if the service provider exploits her bargaining power to charge an unreasonable price of $1000, and this price gets struck by court? Under the most reasonable term approach, the gap filler would be $500---the term that most comports with the community standards of fairness. (18) Under the unfavorable term approach, the gap filler would be $250 (or perhaps even lower, if a harsher penalty is deemed necessary). And under the minimally tolerable term approach, the gap filler would be $750--the maximum price within the tolerable range.

II. MINIMALLY TOLERABLE TERMS: A CONCEPTUAL INTRODUCTION

Two of the three solutions discussed above are familiar to most readers. Both the most-reasonable-term and the unfavorable-term approaches have an intuitive appeal and are based on premises that are shared by more prominent theories of gap filling and contract interpretation. (19) That is, they can be viewed as analogous to the two familiar criteria for default rules: majoritarian and penalty defaults. Thus, when the law supplies a majoritarian term, it merely replicates what most parties do most of the time. And, when the law supplies a least favorable term, it gives incentives to the overreaching drafter to write more reasonable contracts. I have nothing more to say about these approaches and I will therefore focus in the remainder of this Article on the third regime--the minimally tolerable term. This regime is based on a principle that many might find, upon first reflection, objectionable. It allows a party who behaved unfairly by securing an unconscionable advantage to escape with minimum sanction. This leniency might seem both unjust and a weak deterrent. Since my goal here is to dispel some of this intuitive suspicion of the minimally tolerable principle, I begin in this Part with analysis of its conceptual grounding. I hope to show that that the minimally tolerable principle is consistent with some fundamental premises regarding contract enforcement. This will help explain, as illustrated later, why this rule is so surprisingly prominent in practice.

A. Minimally Tolerable Terms and the Freedom to Bargain

Parties are entitled to engage in tough negotiations, maneuver for advantages, and insist on self-serving contractual terms. Drafting a contract that contains terms other than the most reasonable ones is not illegal nor is it uncommon. It is only when these advantages are excessive--when they reach beyond a level that is regarded as tolerable--that the law steps in to invalidate them. Thus, if a court is to reform the unfair contract, it is only the illegitimate element of the one-sided term that needs to be struck. Effectively, then, the court would fill the gap with a term that is still one-sided, still favorable to the same party who dictated the original excessive term, but moderated sufficiently so that it would be tolerable--so as to fit it within the range that is considered legitimate.

This solution preserves to the maximum extent possible the bargaining advantage secured in the contract. It is therefore the one most consistent with the idea that the bargain ought to be respected, not undone. In a companion article, I examine the merits of a new criterion for filling gaps in purely distributive terms (such as price). (20) Under that criterion, gaps should be filled in a way that reflects the relative bargaining power of the parties. Specifically, the court-supplied term ought to resemble as much as possible the term that the parties would have negotiated expressly, even if such a term clearly favors the stronger party. This idea is based on normative grounding: if there is a range in which parties are allowed to bargain, the best that default rules can do is mimic the point within this range that the parties would hypothetically choose. (21) The minimally tolerable term that is discussed here is consistent with this more general bargain-mimicking idea because it reflects the relative bargaining power of the parties. To be sure, it is not a pure bargain-mimicking term. The perfect mimicking term was in fact written in the contract, and yet it was found unenforceable under a policy aimed at limiting the reach of bargaining power. Obviously, the court should not reinstate the same term it has just struck down. What the court should mimic here is the hypothetical bargain that parties negotiating over a truncated domain would reach. This is the term that the parties would choose (or, that the strong party would dictate), had they anticipated that courts would enjoin them from enforcing the invalid one. This solution accords with a "general 'duty of the court to preserve so much of a contract as may properly survive its invalid and ineffective provisions'" (22)--namely, to enforce the entire contract minus the excessive increment.

B. Grounds for lntervention

Under doctrines like unconscionability, which are intended to provide relief from unfair terms, there is no clear authority or justification for courts to provide more than the minimum necessary relief. The court's authority to intervene in the contract and to police its terms arises from the fact that an express term lies beyond society's tolerable range. The further out this term is relative to what is tolerable, the greater the justification for intervention. Once the offensive term has been pushed into the tolerable range, even if only barely so, there is no remaining justification for intervention.

One way to reinforce this idea is by considering the following synthetic illustration: Imagine again the case in which the contract contains an excessive price, $1000, and suppose that the process of adjusting the price involves a gradual examination of successively lower prices. That is, after deciding that $1000 is too high, the court considers one price lower, say $990. If that price is also unconscionable, the court considers a further incremental reduction, to $980. It continues similarly, step by incremental step, until it reaches a price that is no longer considered intolerable. Once that price is reached, the process of adjusting the price downwards would then stop. There will be no remaining justification--at least not under direct policy grounds that give rise to the unconscionability redress--for further adjustments of the price beyond this threshold.

Put differently, if we analogize the process of judicial intervention in the contract to a force that pulls the price from its current intolerable level towards the permissible region, the force gradually weakens as the price gets closer to the tolerable level, and vanishes completely as soon as this level is hit. The point where this adjustment process runs out of justification is not the midrange, majoritarian, most-balanced term. Rather, it is the minimally tolerable price: it is still a one-sided term, albeit one not as bad as the original term.

Here is an illustration: Suppose an insurance policy requires the insured to notify the insurer within twenty-four hours if he has been involved in an accident. What if the insured was a victim of a hit and run accident and was in critical care for a week? The twenty-four-hour term is intolerably short and needs to be expanded, but should it be expanded beyond the time after which the disability was removed? Should the insured be allowed, say, a full, reasonable one month to notify? As one court held, the disability excuses the delay, but once the grounds for the excuse are no longer present, further delay is not justified. (23) The twenty-four-hour unfair term is replaced with a minimally tolerable duration of one week. The insurer is entitled to insist on a short notification term, but only one within the tolerable range.

Conceptually, this argument is consistent with the language that authorizes court intervention in unconscionable contracts. If, as the Uniform Commercial Code instructs, courts are authorized to "limit the application of any unconscionable term as to avoid any unconscionable result," (24) this authority to further tinker with any clause expires once the term is no longer unconscionable.

This argument goes some distance towards justifying the utilization of minimally tolerable terms. It is based on the logic that if legal intervention in the contract is justified by a particular distributive concern, it is also limited by this very concern. There are several equivalent ways to articulate this claim: One is to compare a contract that contains the minimally tolerable price term with a contract that is all else equal, but contains an even worse price term. If there is no good reason to intervene in the former, is there a reason to intervene in the latter beyond fixing it to look like the former? As we shall see later, there may be incentive-based reasons for a more aggressive intervention in the latter contract. But those are different than the distributive concerns justifying the intervention in the first place. Another equivalent way of saying this is to focus on the complaint of the weak party. This party has no reasonable grounds to demand more than the minimally tolerable redress. Once accorded this adjustment, what basis does he have for demanding additional relief? (25)

C. Cross-Doctrinal Foundations

This conceptual defense of the idea of minimum necessary relief accords with other deep-rooted legal practices in general, and contract law in particular, which entitle a party to concede a greater contested claim in order to secure a smaller uncontested claim. For example, the doctrine of remittitur deals with excessive jury verdicts in civil trials. The judge determines the portion of the verdict that is excessive and gives the plaintiff the option to remit--to concede--this increment, or face a new trial. The verdict is not entirely voided, nor is it replaced with the most reasonable amount. Rather, only the excrescence-the sum that exceeds "the highest amount which the jury properly could have awarded"--is lopped off. (26) Many courts have specifically rejected the more intrusive approach, which reduces the excessive judgment to the level that is most fair. Instead, they prefer the minimal intervention approach, reducing the judgment "to the maximum that would be upheld by the trial court as not excessive." (27)

There is more than geometric resemblance between remittitur and the minimally tolerable principle. Both are based on a premise that an "outcome" can lie within a range, and as long as it is within the range there is no ground for intervention. In the remittitur context, the outcome is the jury's judgment regarding damages. It does not have to be a balanced midrange compromise between the litigants' opposing positions. It can lean towards one party, but not in an unreasonable, intolerable fashion. If it goes outside this region, it is pushed back to the minimally tolerable level.

Another procedural practice that applies the minimally tolerable criterion can be found in conflict of laws. If a foreign court that had jurisdiction over a civil matter issued an unreasonable judgment that violated the forum state's public policy, one solution might be straight dismissal. But the typical solution is to enforce the judgment only within the acceptable level. As the German high court explained, the fact that a foreign judgment exceeds the damages that German courts would normally award is not a bar to its recognition. What matters, as the court announced in Bundesgerichtshof, is whether the judgment conflicts so strongly with fundamental German concepts as to "make it seem intolerable." (28) In that case, the court struck the punitive damage element of a U.S tort judgment but enforced a high compensatory judgment that included elements not normally awarded in Germany.

The practice emerging from the minimally tolerable principle in contract law is also strikingly similar to that of severability, and the use of "fallback provisions" in constitutional law. (29) When a court invalidates a law, it has to replace it with a substitute provision, and does so by either severing only the invalid part, or replacing it with a fallback arrangement that was included in the original enactment. In either case, the court is not searching for the most reasonable substitute, nor is it using penalty defaults to deter legislative overreaching beyond the constitutional limits. It seeks to moderate the intervention to the minimum necessary extent.

Within contract law, too, one can find the roots of the idea that excessive provisions can be cured by incremental, rather than total, invalidation. A party may concede a gap in the contract in favor of the other party in order to cure indefiniteness and enforce the conceded contract. (30) Indeed, as we shall see in the next Part, it is quite common for parties who lose their cases in defense of an unconscionable term to concede the offensive element and ask the court to enforce the remainder. Many courts are receptive to such requests. (31)

More generally, the idea of minimum necessary relief can be embedded in the doctrine of waiver. The drafting party is treated as accepting a reduction of the self-serving term, waiving her right to insist on full, unlimited enforcement of that term. With the waiver in place, there is no remaining ground for intervention. As the Supreme Court of Texas recognized in this context: "[E]ven though the contract may be illegal and unenforceable as written, one of the parties may make it legal and enforceable by offering to take out of it the offending provision that makes it illegal." (32)

D. Identifying the Tolerable Threshold

While the criterion underlying the minimally tolerable terms--the upper bound of the legitimate range of contracting--may be conceptually coherent in the abstract, is it implementable by courts? Is it possible to adjudicate this criterion and identify the threshold? There are aspects to this inquiry regarding the practicality of the criterion that implicate the normative discussion, and will be postponed. It might be possible to identify the minimally tolerable threshold but only at an increased adjudication cost, in which case the normative argument in favor of this criterion would weaken.

I believe there are many scenarios in which it is possible to identify a tolerable threshold. Part III below demonstrates how courts manage fairly easily to follow this criterion. In some situations the threshold is easy to identify because it is expressly established by regulations. For example, in some states the maximum scope of noncompete clauses in employment contracts is set by statute. Likewise, the maximum interest rate to be charged for credit or for late payment is regulated in every jurisdiction. Contracts that exceed these thresholds can easily be reformed to the statutorily permissible levels. We will see, however, that in situations in which the threshold is crystal clear, there is also a stronger suspicion that the overreaching occurred in bad faith. In these cases, there is a stronger argument that the exceeding party ought to be punished and deprived of more than just the excess. But, interestingly, the law can separate the deterrence and the contract-reformation measures. To deter overreaching, the culpable party would be punished with harsh administrative sanctions (e.g., revocation of a license). But, to achieve the desired reformation of the unfair contract, the law instructs the minimal necessary relief. (33)

In other situations, where the maximum threshold is not set forth in a statute, courts can follow the principle of minimally tolerable terms without pinpointing the precise threshold. The premise underlying doctrines like unconscionability is that courts can distinguish ad hoc between what is and is not tolerable. Specifically, courts can look at the high end of what other merchants do in the market, (34) or at a majoritarian term boosted by a reasonable premium. As long as they do not err above the maximum, they can cautiously set a term that, while still favorable to the drafting party, is nevertheless not unreasonable. In the example above, the court does not need to decide whether $750 is the uppermost tolerable level that can replace the struck down price of $1000. The defendant might point to another amount, say, $650, as tolerable. (Perhaps the defendant already collected $650 and is willing to stop there; or proves that other merchants charge this price.) As long as the defendant's proposed term is within the tolerable range, it is consistent with the minimally tolerable criterion.

In some situations, a party drafting a mass-market contract may "experiment" with a one-sided term (e.g., arbitration term, or exclusionary clause) that is eventually held by courts to be unfair. In time, the drafting party will modify the boilerplate contract and offer a less extreme version of the one-sided term to new customers. But what about the old customers that accepted the original, excessive term? In these settings, the existence of a new, modified version that survives the scrutiny of courts makes a good candidate for the minimally tolerable term. This modified term would likely still be one-sided relative to the most reasonable or balanced term, but it would be tolerable.

Moreover, there is ample evidence that the parties themselves believe that the principle of minimally tolerable terms is implementable. As I will mention in the next Part, many contracts include severability or savings clauses that instruct courts to enforce any provision that is found to be excessive to the maximum extent permitted by law. If an interest rate is excessive, the maximum permissible rate should apply; if a warranty disclaimer is too harsh, the maximum permissible disclaimer applies; and so on. The parties themselves, it appears, believe in this criterion.

In the end, it might still be difficult in some situations to identify a discrete threshold for the minimally tolerable term. For one, it is possible that the choices cannot be ordered along some linear or other simple yardstick. All the examples used above involve price or other quantitative single-dimensional terms, whereby the preference ordering over the set of possible terms is simple: the closer to the original term, the better for the drafting party. But in other contexts this may not be the case. If, say, the contract stipulates mandatory arbitration and designates a biased arbitrator, what is the next best term? Who is the "minimally tolerable arbitrator"? Or, if the unconscionable term is a compound or multidimensional term, there are many ways to reform it and make it minimally tolerable. In these situations, and perhaps generally, it might be easier to resort to market data and apply the most reasonable, majoritarian terms. But the thesis of this Article does not fail even under this shortcoming. The thesis is proposed in a more tentative fashion: minimally tolerable terms ought to be considered as a way to repair unconscionable contracts to the extent that the court has information about the tolerable threshold. Such information, in other words, ought not to be ignored.

III. DOCTRINAL APPLICATIONS

This Part examines some existing practices that are consistent with the minimally tolerable terms approach. It also identifies some instances in which this criterion was expressly rejected. Overall, I hope to show that in many subtle ways contract law applies an approach that is close to a minimally tolerable terms regime.

A. Severability Clauses

Before turning to legal rules that determine how to repair excessive terms, it is important to note that such principles can be enacted in the contracts themselves. Minimally tolerable gap fillers can emerge in practice as a result of contractual drafting that instructs courts to apply such a criterion. Specifically, a party who drafts a self-serving standard form contract often adds a boilerplate severability clause, stating:
   If any provision of this Agreement shall be held illegal or
   unenforceable, that provision shall be limited or eliminated to the
   minimum extent necessary so that this Agreement shall otherwise
   remain in full force and effect and enforceable. (35)


Many contracts contain similar language, rendering a term enforceable "to the maximum extent permitted by law." For example, a lending term may stipulate that the creditor is entitled to the "maximum rate" and that if the charged interest exceeds the maximum permissible rate, the creditor should refund only the amount of such excess. (36) Or, a warranty/liability term can disclaim all warranties or damages to the maximum extent permitted by law. (37) These are situations in which the parties anticipate the possibility that a one-sided term might be struck, and instruct the courts specifically how to fill gaps in the agreement. The instruction is for the court to follow a one-sided methodology, picking terms that are equivalent to the minimally tolerable provisions.

Of course, courts can disregard such severability clauses, perhaps on the basis that such terms oust the inherent jurisdiction of the court to choose a remedy for wrongful behavior. Or, courts might consider the effect of such clauses to be undesirable and contrary to public policy. One way in which such clauses are undesirable is their vagueness. They refer to a fallback that equals the maximum extent permitted by law, but they do not define this substitute clearly, and render it difficult for contracting parties to determine the provision that will ultimately govern.

To illustrate this vagueness problem of the severability clause, imagine a hypothetical arbitration term that reads: "Any dispute arising out of this agreement will be arbitrated under the arbitration rules of the AAA and take place in the state that is least convenient for the plaintiff, so long as this is not too inconvenient to constitute an unconscionable burden."

Can this clause, with its built-in, unconscionability-proof language, still be struck as unconscionable or unfair? Upon first blush, it would be difficult to view this clause as unconscionable because by its very language it refers to a place that is not unconscionable. To resolve the paradox, it can be argued the flaw in such a clause is its vagueness: it does not give potential disputants sufficient notice where they may be required to arbitrate. (38) Alternatively, the problem with such drafting is bad faith: its sole purpose is to place the maximum feasible burden, to artificially reduce access to remedy.

B. The Doctrine of Partial Enforcement

The doctrine of partial enforcement is a method that enables courts to enforce a term that is otherwise unreasonable or extreme in a partial, tolerable manner. The archaic "Blue Pencil Rule" was the origin of this method. Under this rule, when a contract contained an invalid term, the invalid portion would be literally crossed out (by the metaphoric blue pencil). If the language that remained was grammatically meaningful, it would be enforced. Otherwise, if the remainder was not coherent without some affirmative rewriting, it was entirely invalid. (39) Here, if the contract was selectively enforced, the remainder did not follow a principled criterion. The outcome had an arbitrary, inconsistent aspect.

Yet there was an appealing feature to the Blue Pencil Rule that the more modem approach to partial enforcement sought to preserve. It was a technique that allowed courts to depart from the older and even more rigid all-or-nothing regime, which simply voided any unreasonable provision in its entirety. If the provision were divisible, why not sever only the offensive part--the minimally necessary part--and enforce the remainder? While such divisibility was the policy underlying the Blue Pencil Rule, the mechanical procedure of the rule significantly limited its effectiveness. A better rule for partial enforcement could implement the divisibility policy without constraining courts by a grammatical criterion.

Under the modem partial enforcement doctrine, a court is authorized to reform an unreasonable term in a contract and enforce it to the extent necessary to avoid the unreasonableness. The court does not have to use the judicial blue pencil. It can do more than just cross out some language and enforce the remainder. It can, in addition, substitute the offensive language with a different provision. The underlying goal is to give maximum effect to the parties' agreement, subject to the constraint of avoiding unreasonableness. (40) This goal is often implemented by amending the excessive term with the minimally tolerable term.

Perhaps the most striking (and most common) application of this partial enforcement technique involves covenants not to compete. When an employee enters an employment contract, he often signs a noncompete clause with the employer, applicable in the event that the employment ends. Similarly, when a business is sold with its goodwill, the seller often promises not to compete with the buyer. These restraints are at times too harsh, either by setting too long a duration of the noncompete period, or by defining the geographical boundaries of the noncompete region too broadly. Old decisions in the all-or-nothing tradition used to void these as unreasonable restraints altogether, and leave the parties free of any restraint. (41) Even today, in some European legal traditions such excessive restraints are considered void and may not be adjusted by courts. (42) Blue pencil decisions--somewhat less strict--granted relief by partially enforcing the restraints if they were grammatically meaningful without the offensive words. But most courts nowadays substitute an offensive term with a minimally tolerable one. (43)

At times, the minimally tolerable level is defined explicitly by statute. Some states have enacted bright-line rules stating the maximum duration of noncompete clauses in employment contracts. In these states, when the contract contains a noncompete term that is longer than the statutory cap, it is normally truncated to be equal to that cap. (44) That is, only the increment of the restraint that is socially intolerable is eliminated; the rest stands. In other states there is no bright-line statute. There, too, courts reduce the noncompete term, bringing it down to a level that is tolerable. The restraint is "not enforceable beyond a time or area considered reasonable by the [c]ourt." (45) The permissible duration varies across circumstances and jurisdictions. Case law provides numerous examples for the cap being set between six months and ten years. (46)

This approach can, and in fact does, apply to other contexts in which excessive terms place unreasonable restraint on a party. Thus, for example, in a German case dealing with an exclusive supply arrangement, in which the buyer was obligated to purchase its entire requirements over twenty-four years from the seller, the court reduced the exclusivity term to the minimally tolerable duration of sixteen years. (47)

The stated aim of this jurisprudence is to protect one party's interest, as displayed in the contract, without inflicting unreasonable hardship upon the other party. While this is a cost-benefit test, it would be wrong to conclude that the courts are enforcing the most efficient or most reasonable gap filler. (48) Rather, courts view their role as securing the bargain that the parties struck, recognizing the superior bargaining power of one of the parties. The cost-benefit test is used to identify not the surplus-maximizing term but the limits on the fairness analysis. A limitation that burdens an employee but also accords a substantial benefit to the employer is deemed reasonable. But when there is no efficiency rationale for the limitation, there is room for the fairness calculus. Thus, in eliminating the unreasonable portion of the restraint, the court is setting not the most reasonable or common term, nor a majoritarian or average provision, but rather the minimally tolerable one.

It is important to recognize that the application of minimally tolerable terms in this context is limited by a safety valve, which will be discussed in more detail in the normative analysis below. The concern is that the doctrine of partial enforcement and the application of minimally tolerable terms might give drafters incentives to dictate overly oppressive restraints, expecting to lose at worst only the excessive increment and to keep it anytime it is not challenged. In light of this concern, courts are ready to invalidate the entire noncompete clause (namely, replace the duration term with zero) if there is evidence of deliberate overreaching. (49) Minimally tolerable gap fillers apply only if the crossing of the boundary was done without bad faith. It might be difficult at times to ascertain whether there was bad faith. A presumption of bad faith may exist when the noncompete restraint is unlimited in duration, and often courts indeed vacate such restraints completely. (50)

C. Unconscionability

1. Price and interest rates

When a term in the contract is struck under the doctrine of unconscionability, courts have broad discretion how to repair the contract. A common solution would be to strike the offensive term and replace it with nothing, enforcing the remainder of the contract. This, of course, is an implicit choice to replace the unfair term with the default rule--the gap filler that applies in the absence of an express term. Thus, when the unfair cross-collateral term was struck in Williams v. Walker-Thomas Furniture Co., (51) it was replaced with the default rule that granted the seller no special security in the buyer's prior purchases.

This solution--replacing the offensive term with the legally supplied default provision--would rarely end up with the minimally tolerable term. The legally supplied gap fillers are normally the majoritarian, most reasonable provisions and do not favor the drafting party. But courts have the choice not to strike the entire offensive term, but to trim only the bit of it that is unfair. The remaining component could still be one-sided, but no longer unconscionable. Corbin seems to have recognized this criterion when he explained, in the context of a loan of money, that "a contract that requires a payment of a very high interest will be enforced, up to the point at which unconscionability becomes an operative factor." (52)

A striking illustration of this approach is Vockner v. Erickson. (53) Here, a land purchase contract was unconscionable because the payment terms were too favorable to the buyer, representing an interest rate that was too low. The court reformed the terms minimally, explaining that "the aim of reformation in these circumstances is to bring the contract in conformity with minimal standards of conscionability." (54) In another context, courts use a version of minimally tolerable terms when protecting uninsured patients against excessive medical and hospital bills. In one case, an unconscionable fee was substituted by reference to what other medical providers charge, choosing a point in the high end of the distribution of prices. (55)

Anecdotally, this was also the result in the classic door-to-door sale case, Toker v. Westerman, in which the buyer agreed to pay over $1200 for a refrigerator that normally sold for $400. (56) After the buyer paid more than $650, he sought relief from the oppressive price. The court indeed struck down the price as unconscionable, but allowed the seller to keep the money already paid. True, the court did not directly hold that the seller was entitled to the maximum price that is still tolerable. The $650 figure happened to be the amount already paid when the case was initiated. The buyer asked only for the remainder to be un enforceable; he did not seek restitution of some of the money previously paid. But in another case, in which the price was not yet paid, the court specifically reversed a lower court's stipulation of a low net-cost price, and allowed the seller to collect a price that included all indirect costs plus a reasonable profit. (57) For a product that has a thick market, where the same model is offered by many vendors and prices tend to converge, this comes close to a maximum price.

Another implicit application of this approach comes from the decision of the lower court in the famous Batsakis v. Demotsis case. (58) Recall that this case involved a loan in Greek currency made in Greece, which, in nominal terms, amounted to the equivalent of $25. In return, the debtor promised to pay $2000 plus interest after the war. When the time to pay the debt came, the debtor reneged. The lower court found that the promise to pay $2000 for a loan of $25 was not enforceable, for lack of consideration. Sympathizing with the debtor, but recognizing that the contract could not simply be voided and undone, the lower court ordered the debtor to pay $750 to satisfy the debt. (59) This result, we know, was overturned by the court of appeals, which reinstated the obligation to pay $2000. (60) For our purpose, however, it is the lower court's decision that is of interest, because it is only this court that found grounds for intervention in the first place. This court effectively invalidated the $2000 price and thus needed to fill a gap. It did not turn to the most reasonable and balanced term, nor to a term that comports with community standards of fairness. Rather, it used a one-sided term, just within what it perceived to be the tolerable range.

In credit transactions, the interest rate cannot exceed the statutory threshold or else it is considered unenforceable usury. But what happens when it does? What is it replaced with? Under the Alabama Small Loan Act, for example, interest cannot exceed three percent per month. If the contract violates this cap, it is unenforceable "to the extent, but only to the extent, of the violation." (61) In a California unconscionability case, the court struck the agreed upon interest rate of 200% (that was intended to apply over a short term but ended up applying over a longer duration), and replaced it with a 24% rate. It is not clear whether this was the maximum allowable rate, but it was significantly higher than the prevailing market rates. (62) In Austrian law a similar maximum rate is supplied: a usurious interest rate is adjusted, brought down to double the basic interest rate. In French Law, the excessive interest rate is reduced only to the extent that it exceeds the interest rate allowed by law. (63) In all these examples, a minimally tolerable criterion underlies the practice.

2. Arbitration clauses

The same result of implementing the minimally tolerable term is obtained when the court decides to sever the offensive component of a term and enforce the remaining part. This practice has been widely followed in the context of unconscionable arbitration clauses. Consider one of the leading cases--Brower v. Gateway 2000, Inc. (64) There, the arbitration term in a consumer contract was unconscionable because it placed unreasonable filing costs and location burdens on the consumer. The court vacated these elements but stopped short of eliminating the entire arbitration term (as the consumer would have liked). Instead, it remanded the case to the lower court to figure out a more reasonable arbitration forum. Specifically, it acknowledged Gateway's proposal to use a somewhat less onerous arbitration proceeding (one that Gateway offered its new customers in a modified version of the company's "shrinkwrap" agreement) and instructed the trial court to evaluate whether this proposed venue came within the tolerable range.

Many courts apply a similar approach to repair other elements of unconscionable arbitration clauses. (65) In one case, the court severed only the element of the arbitration clause that required the buyer to reimburse the seller for its arbitration and attorney fees, but left everything else intact, including the one-sided authority of the seller over the choice of arbitration. (66) In another case, when the cost of arbitration was unaffordable to one party, the court allowed the other party to fix the problem by making an ad hoc concession to pay "what [it] needled] to pay to make [arbitration] fair"--even if the amount was only so much as to make the arbitration affordable to this particular plaintiff. (67) Generally, when the arbitration agreement is found to be patently unconscionable because it violates individuals' federal statutory rights (such as the fight to seek punitive damages), courts have reached results that are consistent with the minimally tolerable criterion, either by giving full force to the severability clauses and vacating only the offensive exclusionary elements, (68) or by allowing the party seeking arbitration to waive the elements that are unreasonable. (69)

But courts do not always resort to this principle of striking down only the minimum amount necessary to bring the unconscionable term within the tolerable range. In the leading California case, Armendariz v. Foundation Health Psychcare Services, Inc., the court found some elements of an employment arbitration clause to be unconscionable and upheld the trial court's decision to strike down the entire clause, effectively filling the gap with a no-arbitration term. (70) The employer argued in vain for the court to sever and eliminate only the unlawful elements in the arbitration clause. The court rejected this solution, explaining that it has no vested power to "reform" a contractual term. (71) In this case, the offensive element was an asymmetry--only the employee's claims were directed to arbitration. The court explained that it would have to add a new, non-bargained-for term (that both sides must arbitrate their claims) instead of the existing, unreasonable term (that only the employee must arbitrate its claims). The court found this affirmative augmentation to be beyond its authority and consequently chose to void the entire arbitration clause. (72) It thus rejected the application of the minimally tolerable regime.

This is a puzzling justification. Surely the contract is "augmented" by a new term even under the court's approach. Once the term is struck, the court must supply a non-bargained-for gap filler--here, a no-arbitration clause. The Armendariz court voiced the concern that it is not for the court to write the contract over for the parties. But "[p]artial enforcement [of a contract term] involves much less of a variation from the effects intended by the parties than total non-enforcement would." (73) The question, then, is not whether the court has the power to reform the contract; it clearly does. (74) The question is how much of the bargain needs to be eliminated: only the minimum, rendering the remainder tolerable, or more than the minimum, rendering the remainder even fairer and more balanced?

It is likely that what drove the court's decision in Armendariz was not formalistic minimalism a la the Blue Pencil Rule. Rather, it was the drive to reform the contract in a way that accomplishes more than the minimum necessary, to attain a result that is more balanced and fair, rather than one that comports with bargaining power. It was, in other words, a masqueraded preference for the most reasonable term solution to the problem of repairing unconscionable contracts, possibly justified by the perceived deliberate, bad-faith drafting on the part of the employer. But despite the precedent that this case set for defining what constitutes unconscionability in arbitration clauses, its approach to severability was not generally followed. Indeed, in other cases, the same California Supreme Court invoked the severability principle and vacated only the offensive elements of the arbitration clause. (75)

D. Liquidated Damages

Another application of the minimally tolerable term idea concerns liquidated damages. It is well known that courts do not enforce liquidated damage terms that are clearly excessive and punitive. But what is the damage term that courts supply instead? While the textbook answer is "compensatory" damages, it is often the case that compensatory damages can be assessed with more or less accuracy. Court-imposed "compensatory" damages thus lie within a fairly broad range of reasonableness, from low estimates (that rule out consequential damages and types of avoidable harm) to high estimates (that include generous measures of potential lost profit). A minimally tolerable term would replace the unenforceable liquidated measure, not with the average or most reasonable compensatory measure, but with the high-end estimate of expectation damages.

American law largely rejects this notion. When an excessive liquidated damages clause is held unenforceable, it is wholly invalidated. In these situations, the most that courts are willing to award is the ex post proven expectation damages. Courts refuse to apply a method of reducing the liquidated damages to bring them within the reasonable range. (76)

But other legal traditions fix penalty clauses differently. (77) Under Israeli law, for example, courts are instructed merely to reduce excessive damages to the level reflecting the loss reasonably expected at the time of contracting. (78) In one case, a damages clause required the seller to pay $700 per day of delay in fulfilling his obligations. The seller was late by one hundred days. The court found the liquidated amount to be excessive, and held that actual damages were probably zero or close to it. Still, the court decided to reduce the damages, not to the actual harm of $0, but instead to $200 per day, explaining that "$200 per day is the maximal amount that the parties could have anticipated as possible harm from delay." (79) A leading commentary states that excessive liquidated damages should be reduced "to the highest level that the court regards as reasonably related the harm anticipated at the time of contracting ...; that is, reduced to the measure closest to the agreed sum, such that if that measure were the one agreed upon in the first place, the court would not have been justified in reducing it." (80) This, in other words, is the minimally tolerable level.

E. Conditions in Insurance Contract

An insured's right to collect insurance benefits is always conditional on the insured's satisfaction of strict conditions stipulated in the policy. These conditions are quite often technical and demanding, and the insured's failure to perfectly comply with them leads to forfeiture of the insurance coverage. Recognizing this harsh and intolerable consequence, the law is using a variety of tools to soften the effect of these conditions. One solution is to interpret them as obligations, not as conditions, such that a failure to perfectly conform to them would result in less than complete forfeiture. (81) Another solution, which is relevant to the present analysis, allows the court to "excuse" a condition by vacating it. But once the condition is excused, what is it replaced with? Does the court eliminate it altogether, or merely replace it with the minimally tolerable condition?

Case law in the insurance arena provides illustrations of both approaches. For example, in scrutinizing a harsh notification condition, one court in a leading case reformed the condition only minimally. (82) There, the condition stipulated that the insured must notify the insurer of the accident within twenty-four hours. The insured was in intensive care for seven days and was released from the hospital twenty-three days after the accident. She notified the insurer only four months after the accident, more than three months after her release from the hospital. The lower court found that the twenty-four hour notification condition was too harsh and struck it. In the absence of an affirmative condition, the lower court found no reason to deny the insured of coverage. The court did not specify a precise gap filler for the notification period, but merely held that four months was reasonable. (83) The logic that guided the lower court in devising this solution was the approach I labeled "the most unfavorable term"---construing ambiguity strictly against the insurer.

The Massachusetts Supreme Judicial Court reversed, opting instead for a solution resembling the minimally tolerable term. If the insured's disability was the reason for tolling the running of the notification period, the time extension need not stretch beyond this constraint. The contract intended to put time pressure on the insured, and this time pressure should be maintained once the disability is removed. (84) The practical result is that the insured was obligated to file the notification immediately after being released from the hospital, and the three-month further delay violated that obligation.

IV. WHY ARE MINIMALLY TOLERABLE TERMS USED?

The discussion so far has said little about the normative justifications for fixing unfair terms only minimally. It is time to address this issue, if only because it is now clear that this regime is not merely an intellectual curiosity but rather a prevalent solution in a variety of contexts. Why is this solution so surprisingly ubiquitous? What is the policy rationalizing this approach? In order to evaluate its normative justifications, it is important to distinguish the different problems that unfair terms might present and the different reasons for their emergence.

Saying that terms are "unfair" may be a label for two distinct problems. First, terms are unfair when they violate a substantive distributive criterion. For example, a price of $1200 for an appliance that ordinarily sells for less than $500 gives one party too much (and the other too little) of the surplus. The substantive distributive criterion here may be: a party shall not be allocated a negative expected surplus in a transaction where the other party profits. Parties are free to choose any distribution of the surplus within the constraint of the substantive distributive criterion.

But saying that a term is "unfair" may pack a different meaning--an objection to an unnecessary burden on one party that is not matched by a corresponding sufficient advantage to the other party. Here, the problem is the inefficiency of the arrangement: its cost, which is borne by one party, is unjustified only because it is too weighty relative to the benefit enjoyed by the other party. The term is not merely distributive; it also wastes some of the total surplus. For example, a noncompete clause may be wasteful because it restrains the employee's ability to practice his occupation without creating much of a benefit to the employer.

The case for minimally tolerable terms can be made in a coherent fashion when the term is unfair in the purely distributive sense; it is harder to make when a term is unfair in the second sense, measured by the metric of overall surplus. In the purely distributive case, there is indeed a range of permissible or tolerable outcomes, and the substantive distributive criterion defines the threshold. There, the idea of limiting the unfair term not to exceed the threshold is well defined. To justify it, I will use the idea of "bargained-for advantage" and argue that any fix to such distributively unfair terms that goes beyond the minimally necessary intervention is not desirable and could only hurt the protected party.

The case for minimally tolerable terms is less coherent when the underlying problem in the contract is the inefficiency of the particular burden. An inefficient term needs to be replaced by an efficient one. There is no such thing as a range of tolerably inefficient terms. The notion of a minimally tolerable outcome has neither conceptual nor normative grounding when the yardstick is total surplus.

In reality, many of the situations in which courts strike unfair terms are hybrid: the unfair term violates a distributive criterion but might also reduce the total surplus. Often, both elements are found. There is a way to fix the distributive violation that is consistent with the principle of minimally necessary intervention, but it is not always true that such intervention also resolves the totalsurplus concern. Accordingly, in the remainder of this Part, I will examine various reasons for the emergence of unfair terms and discuss whether the notion of minimally tolerable intervention makes sense in each of these scenarios.

A. Market Power and Bargained-For Advantage

One-sided terms are often purely distributive. A party may be in a position to insist on favorable terms because it has what can loosely be described as greater bargaining power (e.g., concentrated market power, or better outside options), or because it traded off other concessions to get the terms it wanted. In general, absent a defect in assent, the terms, even if strongly one-sided, represent a bargain.

When the bargained-for term is unfair, it may be thought that the law should repair it not just minimally, to the tolerable level, but more ambitiously, to the most fair and balanced level. Rather than let the strong party get away with most of the surplus and trim off only the stretched excess, let's put the parties in equal positions, splitting the surplus among them in the most reasonable, not just the narrowly tolerable way. This view is misguided because it does not promote the interest of the weak party whom it intends to protect. In cases in which one party has significant market power, the more aggressive ways to redress her overreaching are unhelpful because they do little to resolve the underlying imbalance of bargaining power. If the asymmetry between the parties remains, it will continue to manifest itself elsewhere in the contract. In fact, the strong party may now be induced to extract value in ways that not only are self-serving but also waste some of the total surplus, or which require a more costly drafting and transaction procedure.

For example, if a creditor cannot charge too high an interest rate, it might use other techniques to increase its profit, such as delayed fees, upfront charges, or requiring borrowers to purchase additional unwanted products and services like credit insurance and ancillary privileges. If an insurer cannot exclude a particular coverage, it might charge a higher premium. Or, if an employer cannot exploit its power to set low wages, it can increase the density of the tasks; if its termination powers are constrained, it might also change its hiring policy; and if it cannot impose arbitration terms, it can reduce wages by the cost of litigation insurance. The concern that these examples illustrate is that what might appear as a pro-consumer or pro-employee legal intervention merely shifts the use of rent-seeking terms to domains in which they are permissible and tolerable. At best, the distributive intervention is irrelevant. At worst, it imposes excess transaction costs or forces an inefficient redesign of the transaction. As long as one party's market power is maintained, it is used to dictate the distribution of the surplus.

It is possible that this argument proves too much. In arguing that distributive intervention should not go beyond the minimal necessary relief, I may appear to be saying that it is futile to engage in any redistributive intervention over some terms of the contract when at the same time parties are allowed to rewrite other terms. In fact, it is also possible that any such intervention would have an unintended distributive effect within the protected group. For example, if a pawnbroker cannot charge high interest rates, it can instead increase the upfront fees. Short-term borrowers will now be subsidizing in part the credit of long-term borrowers. Likewise, eliminating arbitration terms in consumer contracts helps some consumers--those that will invoke the legal right to sue--but perhaps at the expense of another group, the nonlitigious consumers, who will pay higher prices. The more aggressive the legal intervention, the greater this cross-subsidy.

There may be good reasons why some terms in contracts should not be permitted, even if their removal leads to other unbalanced terms. For example, excessive interest rates may represent a different concern than high loan-underwriting fees. Marketing unreasonably unsafe products (and using the contract to limit the manufacturer's liability for physical injury) represents a different type of social cost than high prices. Since parties have the ability to undo some of the distributive effects intended by legal intervention, it is naive to think that courts' power to intervene would simply repair the underlying distortion. But it could shift the distortion to a more acceptable and transparent domain. However, to the extent that similar distributive problems would re-emerge elsewhere in the contract, and perhaps with the additional problem of cross-subsidies, the intervention power of courts should be used minimally.

B. Unclear Boundaries

Not all bargains are permitted. Legal standards determine the boundaries of permissible terms and what constitutes unfair contracting. But, these boundaries are not always obvious. For example, arbitration clauses in employment contracts that were traditionally enforceable, and still are in many jurisdictions, are occasionally held to be unconscionable. A merchant operating on a national scale might find that the class action waivers in its consumer contracts are held to be unconscionable in some states but not in others. (85) There is a degree of uncertainty as to which elements of arbitration are unconscionable and would be struck down when challenged in court. Even when the legal standard is clear, parties who draft contracts are not always sure how it will apply to their particular case. For example, if it is permissible to write a noncompete clause in an employment contract, is a twenty-five-mile radius marking the noncompete territory excessive? (86) Further, parties may be uncertain about the effect of a particular provision in the contract because the effects could depend on subsequent circumstances. (87)

In these settings of uncertainty, there is only so much that legal intervention can do to change the essence of the bargain. If the drafting party is punished for overreaching by, say, replacing the excessive term with one significantly below the maximum permissible level, it faces a familiar dilemma: any additional sliver of the surplus it tries to appropriate through an incrementally more self-favorable term has a small upside if the term is enforced, but a downside of potentially significant magnitude if the term is held to be excessive and replaced with a midrange term.

This asymmetry between the benefit and the risk of one-sided drafting could have two effects on the drafting party. First, it could induce more cautious drafting, to avoid bumping against the maximum permissible boundary. (88) When bargaining for an advantage, the party will maintain a safety cushion by forgoing part of the surplus that it would otherwise extract. This effect alone is not a social cost, but it may be inefficient if it also affects other terms of the deal. Second, the drafting party would have an added incentive to invest in information that would enable it to assess the exact location of the boundary. Such information has the private value of avoiding the safety cushion, but it has low (or zero) social value. (89)

It is the first effect that requires attention. The incentive to moderate a term to avoid bumping against the permissible limit may induce the party to shift its market power to other terms of the contract. In extreme cases, the moderating party might not enter the contract in the first place unless it can secure another favorable term. In less extreme cases, it might insist on other bad terms that are permissible, but no less costly to the other party. For example, if the law were to apply the minimally tolerable terms, an employer who is interested in a mandatory arbitration clause or a noncompete restraint would experiment with drafting such terms, knowing that at worst, if a court views the term as excessive, it would be corrected only incrementally. But if the law replaces an unfair arbitration term with one that is unfavorable to the employer (say, no arbitration at all, or elimination of the noncompete restraint), the employer might be reluctant to draft such terms in the first place, and would use any market power it has to insist on lower wages, bonds, or some other costly burden on the employee. Or, in consumer credit sales, a vendor can replace the option to finance the sale at usurious rates with a technique of quoting separate "cash price" and "credit price." Under the common law's "time price doctrine," the quoting of a separate credit price evades usury scrutiny even if the credit price exceeds the cash price plus lawful interest. (90) The result is equally costly and probably less convenient for the consumer, but it passes legal scrutiny. Generally, an aggressive legal intervention in one area of the contract can shift bargain dominance to other areas, affording no true relief to the weak party ex ante. (91)

This concern is particularly acute in insurance contracts. Here, it is not even clear that the drafter of a particular nasty term has greater market power. It is likely that the terms of the policy are reflected in the price, such that more restrictive terms correspond to more affordable premiums. If courts tinker with a single term that is deemed unreasonable, the bargain will change not only in the dimension of this term, but elsewhere too, at least with respect to price.

C. Hidden Advantage and Bad Faith

Another form of superior bargaining power, which accounts for the presence of unfair terms, is greater sophistication. The sophisticated party understands the terms of the contract; the naive party does not. Some interventions in contracts are justified when the sophisticated party takes advantage of this asymmetry and drafts a term that its counterpart fails to appreciate or conceals some of the implications of the contract. Door-to-door sales cases like Toker v. Westerman (92) and some of the consumer unconscionability cases illustrate this scenario.

Here, in the absence of legal intervention, there is no guarantee that the term would be chosen efficiently. Or, put differently, a legal intervention that strips the strong party of the entire benefit from the hidden term could well be efficient. Of course, the mere fact that the effect of the term was unknown to the less sophisticated party does not render it unconscionable (or else the bulk of consumer contracts would be unconscionable). The law requires an independent substantive criterion to establish that the particular hidden term is unconscionable. Some terms are merely distributive (e.g., the high price in Toker), while other terms affect the overall surplus (e.g., noncompete clauses), so the substantive criterion could be based on either distributive or efficiency grounds. When the unfair term is also inefficient, there is no reason to use minimum intervention. The new term should be the optimal term, recognizing though that the effect of more aggressive intervention could be reflected in other distributive terms.

Of course, without articulating the precise substantive criterion for intervention, it is impossible to say which level of intervention in the contract is "correct." All that can be said is that the intervention should not go beyond addressing the particular substantive criterion justifying it. Many of the cases applying minimally tolerable terms invoke the notion that there is no justification for intervention more aggressive than that which is prescribed by the substantive criterion.

There is an important scenario, however, in which more aggressive intervention would be desirable, even if the unfair term is purely distributive. A party who drafts a hidden advantage may be doing so knowing that the term violates the substantive criterion of unconscionability and that it exceeds the boundary of tolerable terms. For example, an employer hiring hourly workers and paying them less than the minimum wage is deliberately crossing the permissible threshold. Or, a creditor charging interest exceeding the usury limits is often cognizant of the boundary being crossed. Here we encounter an incentive problem: if the drafting party expects that the court would only strike the excessive increment, what incentive does it have to avoid overreaching? (93) At best, the express term will stand; at worst, it will be replaced with the most favorable term permitted by law. Why draft a term that reflects this permissible standard if you can get away with--in fact, benefit from--drafting an excessive term? In contrast, were the law to intervene more aggressively, perhaps even with a contra proferentem provision, the drafter would have more to lose from overreaching. If the social goal is to reduce the drafting of hidden, nasty, bad faith terms, a more aggressive intervention is necessary.

This is a powerful insight, recognized by many commentators and courts. (94) Ironically, it is almost too powerful to succeed under its own terms, namely, when it is invoked to support the most-reasonable-term regime. A wrongdoer who deliberately drafts unreasonable terms can withstand even the more aggressive reformation of the contract in those few cases in which he is challenged, continuing his illicit business otherwise. When the contract terms are challenged, he will lose more than the minimal increment when the contract is reformed to the most reasonable term or to the contra proferentem term. But if there is a pattern of unconscionable behavior, this occasional loss will likely be dwarfed by the upside of the illicit gain. To deter such calculated violations, something more is needed, such as punitive damages or antifraud measures. (95) Indeed, some usury laws provide a striking example: a contract with an excessive interest rate is repaired only minimally, to the maximum allowable rate, but the creditor may also be subject to the stiff sanction of license revocation. (96) The law addresses the incentive problem while using minimally tolerable gap fillers.

The law ought to be sensitive to this incentive problem. Indeed, in many areas, the law distinguishes between deliberate and inadvertent overreaching. Not all hidden unconscionable terms are drafted in bad faith. The line between what is permissible and what would be considered intolerable is not clearly marked and not always recognized by the drafting party. At times, a party will cross the line knowingly and in bad faith, counting on the majority of parties to capitulate and exploiting the high likelihood that such overreaching will go unchallenged. Often, though, the boundaries are not deliberately and maliciously crossed. Rather, the strong party chooses one way to exploit its greater sophistication (or even hide it)--but it may have refrained from doing so had it recognized the boundary.

When the boundary of permissible terms is known and nevertheless crossed, the term should be replaced in a way that provides deterrence. A good example is the minimum wage law, which sets a clear and bright line between tolerable and intolerable wages. An employer that pays less than minimum wage cannot plead ignorance. Accordingly, the statute awards the aggrieved employee more than the minimally tolerable wages--it doubles the unpaid wages. (97) But if the boundary is fuzzy and was violated without bad faith, the law should only reduce the excessive term back to the boundary--to the minimally tolerable level. Here, a good example is reformation of covenants not to compete. (98) Most drafters don't know the exact permissible threshold and cross it inadvertently.

Sometimes it is clear that an unreasonable term was inserted into the contract deliberately, in bad faith. But other times it may be difficult for courts to determine whether the boundary was crossed deliberately, and thus whether there was bad faith that would require a more aggressive intervention. Can the legal rule that dictates how to repair the contract turn on such a distinction regarding the "mens rea" of the drafting party? Several heuristics can aid with this determination. First, as just mentioned, it matters whether the substantive boundary of what is minimally tolerable is set by a bright-line rule or not. Bright lines are more likely to be known or cheaply assessed than vague standards, and crossing them is easy to prove. (99) Presumably, then, a crossing of a bright-line threshold is done deliberately and deserves harsher treatment. (100)

Second, it matters how egregious the excessive term is. Terms that cross the boundary by a wide margin deserve to be reformed more aggressively by courts for at least two reasons: they are more likely to represent deliberate overreaching, even of a fuzzy standard, and a greater sanction is needed to offset the larger gain to the wrongdoer. In other words, how to fix unfair terms depends on how bad they are.

Third, it matters whether the drafter is a repeat transactor. There are at least three reasons why the repeat-player characteristic justifies a more aggressive reformation of the contract. One reason is that an experienced transactor is generally more informed and more likely to know where the boundary lies. Another reason is that even if this party doesn't know the boundary, a legal rule that induces her to be more cautious and to spend more on acquiring information about the boundary is less of a waste, as this knowledge would be used more often. Finally, a repeat transactor benefits repeatedly from the excessive term and thus it takes a greater sanction to deter the wrongful behavior.

Fourth, it matters what terms other parties who are in similar market positions draft. If the excessive term is commonly used, it is more plausible that the drafting party intended to follow a prevailing practice--perhaps even copied someone else's contract--rather than intentionally experimented with an unconscionable term. Normally, when competitors use less onerous terms we count on competition to eliminate the nasty term. But since we are assuming here that the term is hidden, it is possible that it would survive even in the presence of competition. (101)

Even when there is strong indication of bad faith, it still matters whether the drafting party is likely to shift the added cost of the legal intervention to other aspects of the contract, and whether it can do so retrospectively by using its one-sided power to modify existing contracts. Deterrence of one kind of overreaching can divert rent-seeking behavior to other areas of the contract without producing the desired redress for consumers.

D. Duress and Contractual Hold-Up

Another reason a party may be able to dictate an unfair term is that it is able to exploit the vulnerability of its counterpart, recognizing that the counterpart is desperate for the contract and would agree to any terms. Batsakis v. Demotsis and other usurious loan cases illustrate this scenario. (102)

It is tempting to rationalize such intervention on fairness grounds, but it is important to examine what degree of intervention would be in the interest of the parties, ex ante. For it may well be that the anticipation of an overly aggressive intervention by the law would backfire and induce the strong party to refrain from any kind of dealings with its counterpart. If the contract is reformed with a term that, while ex post reasonable, does not leave enough of an incentive for the stronger party to enter it, such intervention is not in the ex ante interest of both parties. In Batsakis, as in many high-interest loans, the creditor is taking advantage of the borrower's urgent need for funds. But it would be misleading to examine the scenario ex post, when it is already known that the borrower was able to secure the loan, and conclude that the price was excessive. Ex ante, the high price was necessary in light of the uncertainty over repayment. The court should reform the contract to a term not lower than the minimum term the creditor would have agreed to. (103)

If the court has perfect information about this minimum acceptable term to the creditor, the only normative criterion that can be derived is that the intervention should not go farther than this threshold. More lenient intervention is fine; more aggressive intervention is harmful. This criterion does not directly correspond to the minimally tolerable principle, nor to the other intervention principles outlined above. It is based on an ex ante analysis of incentives, not on ex post conception of distributive fairness. But if the court has imperfect information about the minimum acceptable term, intervention in the contract should be done with greater care. If the coerced party represents an unusual risk, one that requires a price premium, it would be better to play it safe and use a minimally tolerable term, thereby reducing the chance that anticipation of the more aggressive relief would deter the transaction ex ante.

CONCLUSION

The purpose of this Article is not to advocate for the general use of minimally tolerable gap-filling methodology, but to identify it as a conceptual and practical possibility and discuss some arguments in support of such a regime. Upon first encounter, I imagine, readers are likely to be skeptical. If a court already goes to the trouble of reforming an unfair contract, why not provide the most reasonable repair? And indeed, the argument in favor of the most reasonable term is compelling, intuitive, and quite easy to make. Yet despite this inclination, I set out in this Article to explore an alternative. I found surprising pervasiveness in the use of minimally tolerable terms, in scattered areas of American contract law, as well as in comparative law. That this principle managed to permeate the law so broadly suggests to me that in some more subtle ways it can be justified, or at least grounded, in broader existing practices, and it can trump even the hardwired predisposition in favor of the most-reasonable-terms criterion.

In the end, I find the incentive problem to be the most troubling one. Would this regime of minimally tolerable terms encourage parties to draft unfair and unconscionable terms? If this concern is the crucial one, and I think it is, it marks the limits of this approach. If the boundary of permissible contracting is easily known and nevertheless crossed, the term should be replaced in a way that provides deterrence--something significantly less then the minimally tolerable term. In fact, true deterrence may require some kind of punitive response. But if the threshold is fuzzy and was violated without bad faith, the law would only reduce the excessive term back to the threshold--to the minimally tolerable level.

This principle of minimally tolerable terms is part of my more ambitious thesis, that bargaining power matters for contract doctrine. In a companion article, I attempted to justify the thesis that bargaining power ought to affect the way courts supplement true gaps in contracts. (104) The gap fillers should mimic the bargain that the parties would have struck, even if that bargain favors one of the parties. Minimally tolerable terms, in my view, are a species of bargain-mimicking terms. They are favorable to the party with the bargaining power, mimicking to the maximum permissible level the bargain of the parties.

(1.) See Omri Ben-Shahar, A Bargaining Power Theory of Default Rules, 109 COLUM. L. REV. 396 (2009).

(2.) See RICHARD A. POSNER, ECONOMIC ANALYSIS OF LAW 98 (6th ed. 2003) ("[C]ontract law cannot readily be used to achieve goals other than efficiency. A ruling that fails to interpolate the efficient term will not affect future conduct; it will be reversed by the parties in their subsequent dealings. It will only impose additional--and avoidable-transaction costs."); see also FRANK H. EASTERBROOK & DANIEL R. FISCHEL, THE ECONOMIC STRUCTURE OF CORPORATE LAW 17-22 (1991); E. ALLAN FARNSWORTH, CONTRACTS [section] 7.16, at 486 (4th ed. 2004) (noting that courts are to provide terms "that an economist would describe as maximizing the expected value of the transaction"); Mark P. Gergen, The Use of Open Terms in Contract, 92 COLUM. L. REV. 997, 1064-72 (1992) (suggesting that the default rule should be a joint maximization rule).

(3.) See RESTATEMENT (SECOND) OF CONTRACTS [section][section] 79 cmt. c, 208 cmt. d (1981).

(4.) See 1 JAMES J. WHITE & ROBERT S. SUMMERS, UNIFORM COMMERCIAL CODE [section] 4-5, at 223 (4th ed. 1995) ("[T]he reported litigation based on excessive price has dwindled to a trickle.").

(5.) See, e.g., C&J Fertilizer, Inc. v. Allied Mut. Ins. Co., 227 N.W.2d 169 (Iowa 1975); RESTATEMENT (SECOND) OF CONTRACTS [section] 211(3) cmt. f(1981).

(6.) See, e.g., Armendariz v. Found. Health Psychcare Servs., 6 P.3d 669 (Cal. 2000); Brower v. Gateway 2000, Inc., 676 N.Y.S.2d 569 (App. Div. 1998); Susan Randall, Judicial Attitudes Toward Arbitration and the Resurgence of Unconscionability, 52 BUFF. L. REV. 185 (2004).

(7.) See U.C.C. [section] 2-718(1) (2003) ("A term fixing unreasonably large liquidated damages is void as a penalty.").

(8.) See, e.g., Leeber v. Deltona Corp., 546 A.2d 452, 454 (Me. 1988).

(9.) See, e.g., FLA. STAT. ANN. [section] 542.335(1)(d)(1) (West 2010) ("[A] court ... shall presume unreasonable in time any restraint more than 2 years in duration.").

(10.) See, e.g., Earl of Chesterfield v. Janssen, (1750) 28 Eng. Rep. 82 (Ch.) (Eng.).

(11.) See Richard Craswell, Property Rules and Liability Rules in Unconscionability and Related Doctrines, 60 U. CHI. L. REV. 1 (1993) (comparing the two approaches).

(12.) See Fair Labor Standards Act of 1938, 29 U.S.C. [section] 216 (2006); Federal Price Gouging Prevention Act, H.R. 1252, 110th Cong. [section][section] 3-4 (2007).

(13.) U.C.C. [section] 2-302(1) (2003); see also UNIF. CONSUMER CREDIT CODE [section] 5.108(3), 7 U.L.A. pt. 3, at 231 (2002).

(14.) COMM'N ON EUR. CONTRACT LAW, PRINCIPLES OF EUROPEAN CONTRACT LAW, art. 4:109(2) (Ole Lando & Hugh Beale eds. 1999).

(15.) It is common to apply gap fillers in situations in which the gap arises not from indefiniteness in drafting, but from the knockout of an express term. Compare U.C.C. [section] 2-204(3) (2003) (applying gap fillers to "open terms"), with U.C.C. [section] 2-207(3) (applying gap fillers as a result of the knockout of express terms).

(16.) See, e.g., Aristides N. Hatzis & Eleni Zervogianni, Judge-Made Contracts: Reconstructing Unconscionable Contracts (Dec. 26, 2006) (unpublished manuscript), available at http://papers.ssm.com/sol3/papers.cfm?abstract_id=953669.

(17.) See JONATHAN SHELDON & CAROLYN L. CARTER, UNFAIR AND DECEPTIVE ACTS AND PRACTICES 112 (5th ed. 2001). For example, under some usury laws creditors who charge excessive interest can have their licenses revoked. See, e.g., ALA. CODE [section] 5-18-9(a) (2010).

(18.) See RESTATEMENT (SECOND) OF CONTRACTS [section] 204 cmt. d (1981).

(19.) See, e.g., Ian Ayres, Default Rules for Incomplete Contracts, in 1 THE NEW PALGRAVE DICTIONARY OF ECONOMICS AND THE LAW 585, 586-87 (Peter Newman ed., 1998); Richard Craswell, Contract Law: General Theories, in 3 ENCYCLOPEDIA OF LAW AND ECONOMICS 1, 1 (Boudewijn Bouckaert & Gerrit De Geest eds., 2000).

(20.) Ben-Shahar, supra note 1.

(21.) For a model conceptualizing reasonableness as a range, see Avery Katz, Your Terms or Mine? The Duty to Read the Fine Print in Contracts, 21 RAND J. ECON. 518 (1990).

(22.) Anders v. Hometown Mortg. Servs. Inc., 346 F.3d 1024, 1032 (11th Cir. 2003) (citing Exparte Celtic Life Ins. Co., 834 So. 2d 766, 769 (Ala. 2002)).

(23.) See Royal-Globe Ins. Co. v. Craven, 585 N.E.2d 315, 317-18 (Mass. 1992).

(24.) U.C.C. [section] 2-302(1) (2003).

(25.) Corbin recognizes this rationale: "[T]he line [representing the enforceable term] must be drawn somewhere, and it is drawn at the point where the protection to which the buyer is justly entitled ends." Arthur L. Corbin, A Comment on Beit v. Beit, 23 CONN. B.J. 43, 46 (1949).

(26.) Dick v. Watonwan Cnty., 562 F. Supp. 1083, 1108 (D. Minn. 1983); see also 11 CHARLES ALAN WRIGHT, ARTHUR R. MILLER & MARY KAY KANE, FEDERAL PRACTICE AND PROCEDURE 167 (2d ed. 1995).

(27.) Earl v. Bouchard Transp. Co., 917 F.2d 1320, 1328 (2d Cir. 1990).

(28.) Bundesgerichtshof [BGH] [Federal Court of Justice] June 4, 1992, ENTSCHEIDUNGEN DES BUNDESGERICHTSHOFES IN Z1VILSACHEN [BGHZ] 118, 1993, translated in 32 LL.M. 1327, 1333 (1993).

(29.) See Michael C. Dorf, Fallback Law, 107 COLUM. L. REV. 303, 303 (2007).

(30.) See 1 E. ALLEN FARNSWORTH, FARNSWORTH ON CONTRACTS [section] 3.29 (2d ed. 1998); Omri Ben-Shahar, "'Agreeing to Disagree": Filling Gaps in Deliberately Incomplete Contracts, 2004 WIS. L. REV. 389, 421.

(31.) See infra text accompanying notes 64-69.

(32.) Lewis v. Krueger, Hutchinson & Overton Clinic, 269 S.W.2d 798, 800 (Tex. 1954) (Calvert, J., concurring).

(33.) See, for example, Alabama's Small Loan Act, a typical usury law that provides for a reduction of excessive interest only to the extent of the violation, but also allows a revocation of the offending creditor's license. ALA. CODE [section][section] 5-18-9, 5-18-21 (2010).

(34.) One court held that medical charges below the seventy-fifth percentile of what other providers charge are "within the range of the overall market." Colomar v. Mercy Hosp., 461 F. Supp. 2d 1265, 1270 (S.D. Fla. 2006) (internal quotation marks omitted); see also Mark A. Hall & Carl E. Schneider, Patients as Consumers: Courts, Contracts, and the New Medical Marketplace, 106 MICH. L. REV. 643, 686-89 (2008).

(35.) PayPal Payflow Gateway Services for Resold Merchants [section] 14.3, PAYPAL, https://cms.paypal.com/us/cgi-bin/?cmd=_render- content&content_ID=ua/PayflowResold_full (last updated Oct. 20, 2010).

(36.) See, e.g., 2 J. ROBERT BROWN, JR. & HERBERT B. MAX, RAISING CAPITAL: PRIVATE PLACEMENT FORMS & TECHNIQUES 13.01P (3d ed. 2008) ("Payee shall never be deemed to have contracted for or be entitled to receive ... as interest on this Note ... any amount in excess of the Maximum Rate, and, in the event that Payee ever receives ... as interest any such excess.... then any remaining excess shall forthwith be paid to Maker.").

(37.) See, e.g., Terms and Conditions, NAPSTER, http://www.napster.com/terms.hlrnl (last visited Jan. 6, 2011) ("To the extent that in a particular circumstance any disclaimer or limitation on damages or liability set forth herein is prohibited by applicable law, then, instead of the provisions hereof ... Napster ... shall be entitled to the maximum disclaimers and/or limitations on damage and liability available at law or in equity.... " (emphasis added) (capitalization omitted)); see also End User License Agreement, REALNETWORKS, http://www.realnetworks.com/uploadedFiles/Support/helix-support/eula-Helix- MServerContProv.pdf (last visited Jan. 6, 2011) ("To the maximum extent permitted by applicable law, RealNetworks further disclaims all warranties...." (emphasis added) (capitalization omitted)); Altova Online Training Terms of Use, ALTOVA, http://www.altova.com/orderaotterms.html (last visited Jan. 11, 2011) ("Because some states or jurisdictions do not allow the exclusion or the limitation of liability for consequential or incidental damages, in such states or jurisdictions, Altova and our suppliers' liability in such state or jurisdiction shall be limited to the extent permitted by law." (emphasis added) (capitalization omitted)).

(38.) For a similar problem in constitutional law, see Doff, supra note 29, at 311.

(39.) See FARNSWORTH, supra note 2, [section] 5.8, at 344.

(40.) See Corbin, supra note 25, at 49-50.

(41.) The earliest case is Mitchel v. Reynolds, (1711) 24 Eng. Rep. 347 (Eng.). This rule remained in force in England and in the United States until the end of the nineteenth century. See C.T. Drechsler, Enforceability of Restrictive Covenant, Ancillary to Employment Contract, as Affected by Territorial Extent of Restriction, 43 A.L.R.2d 94, [section] 6(a) (1955).

(42.) See Hatzis & Zervogianni, supra note 16, at 5 (citing German law).

(43.) See, e.g., Solari Indus., Inc. v. Malady, 264 A.2d 53, 61 (N.J. 1970); see also Harlan M. Blake, Employee Agreements Not to Compete, 73 HARV. L. REV. 625, 646-51 (1960).

(44.) See, e.g., FLA. STAT. ANN. [section] 542.335(1)(d)I (West 2010) ("[A] court ... shall presume unreasonable in time any restraint more than 2 years in duration."), enforced, Flickenger v. R.J. Fitzgerald & Co., 732 So. 2d 33, 34 (Fla. Dist. Ct. App. 1999); LA. REV. STAT. ANN. [section] 23:921C (2010); S.D. CODIFIED LAWS [section] 53-9-11 (2010).

(45.) Justin Belt Co. v. Yost, 502 S.W.2d 681,685 (Tex. 1973). In other countries the unlawfulness of the restraint is a matter for the courts to determine--but again with the consequence that if found unlawful, the restraint is not struck down in toto, but reduced to its minimally tolerable level. For a French decision, see Cour de cassation [supreme court for judicial matters] sot., Oct. 21, 1960, JCP 1960, II, 11886, discussed in HUGH BEALE ET hE., CONTRACT LAW 312-13 (2002).

(46.) See, e.g., Kramer v. Robee, Inc., 824 F. Supp. 508, 512-13 (E.D. Pa. 1992) (reducing a three-year restraint to two years and enforcing a geographical limit of the entire United States); John Roane, Inc. v. Tweed, 89 A.2d 548, 557 (Del. 1952) (reducing, under Maryland law, a five-year restraint to four years); Foltz v. Struxness, 215 P.2d 133 (Kan. 1950) (enforcing a ten-year restraint); Fullerton Lumber Co. v. Torborg, 70 N.W.2d 585, 592 (Wis. 1955) (reducing a ten-year restraint to three years).

(47.) BGH Sept. 16/17, 1974, NEUE JURISTISCHE WOCHENSCHR1FT [NJW] 2089, 1974; see also BEALE ET. AL., supra note 45, at 313-14.

(48.) In many contexts, the most common duration gap filler is at-will termination--a zero restraint on the right of an employee to walk away. See, e.g., Payne v. W. & Atl. R.R., 81 Tenn. 507, 519 (1884) (termination of employment contract); U.C.C. [section] 2-309(2) (2003) (at-will termination in sales contracts).

(49.) See, e.g., Cent. Adjustment Bureau v. Ingram, 678 S.W.2d 28, 37 (Tenn. 1984).

(50.) E.g., Harvest Ins. Agency v. Inter-Ocean Ins. Co., 492 N.E.2d 686, 690 (Ind. 1986).

(51.) 350 F.2d 445, 445-46 (D.C. Cir. 1965).

(52.) 1 ARTHUR LINTON CORBIN, CORBIN ON CONTRACTS [section] 129, at 556 (1963) (emphasis added) (internal quotation marks omitted).

(53.) 712 P.2d 379 (Alaska 1986).

(54.) Id. at 383-84.

(55.) See Hall & Schneider, supra note 34.

(56.) 274 A.2d 78 (N.J. 1970); see also Bank of Ind. Nat'l Ass'n v. Holyfield, 476 F. Supp. 104 (S.D. Miss. 1979); Jones v. Star Credit Corp., 298 N.Y.S.2d 264 (Sup. Ct. 1969) (allowing the buyers to stop payments after $620 was paid on a $300 freezer, even though it was purchased for over $1400).

(57.) Frostifresh Co. v. Reynoso, 281 N.Y.S.2d 964 (App. Term 1967), rev'g 274 N.Y.S.2d 757 (Sup. Ct. 1966).

(58.) 226 S.W.2d 673 (Tex. Civ. App. 1949).

(59.) Id. at 674.

(60.) Id. at 675.

(61.) ALA. CODE [section] 5-18-21 (2010).

(62.) See Carboni v. Arrospide, 2 Cal. Rptr. 2d 845 (Ct. App. 1991). The prime rate around that time was 9.5%, Mortgage (ARM) lndexes: Prime Rate: Historical Data, MORTGAGE-X, http://www.mortgage-x.com/general/indexes/prime.asp (last visited Jan. 5, 2011), and the California tax penalty interest was 10%, What Are Previous and Current Interest and Estimate Penalty Rates?, ST. CAL. FRANCHISE TAX BOARD, http://www.ftb.ea.gov/individuals/faq/ivr/617.shtml (last visited Jan. 5, 2011).

(63.) See Hatzis & Zervogianni, supra note 16, at 7-8.

(64.) Brower v. Gateway 2000, Inc., 676 N.Y.S.2d 569 (App. Div. 1998).

(65.) See, e.g., Swain v. Auto Servs., Inc., 128 S.W.3d 103 (Mo. Ct. App. 2003).

(66.) See Missouri ex rel. Vincent v. Schneider, 194 S.W.3d 853 (Mo. 2006).

(67.) Anders v. Hometown Mortg. Servs., Inc., 346 F.3d 1024, 1029 (11th Cir. 2003).

(68.) See id. at 1031-32; Ex parte Thicklin, 824 So. 2d 723, 734 (Ala. 2002). But see Sosa v. Paulos, 924 P.2d 357 (Utah 1996) (holding that a substantively unconscionable clause may render the entire agreement unconscionable).

(69.) See Exparte Celtic Life Ins. Co., 834 So. 2d 766, 768 (Ala. 2002) ("[A] party to a contract can waive a contractual provision beneficial to that party.").

(70.) 6 P.3d 669 (Cal. 2000).

(71.) Id. at 697.

(72.) Id. ("Because a court is unable to cure this unconscionability through severance or restriction, and is not permitted to cure it through reformation and augmentation, it must void the entire agreement.").

(73.) Corbin, supra note 25, at 50.

(74.) See U.C.C. [section] 2-302 cmt. 2 (2003) ("Under this section, the court, in its discretion, may ... strike any single term or group of terms which are unconscionable .... ").

(75.) See, e.g., Little v. Auto Stiegler, Inc., 63 P.3d 979 (Cal. 2003).

(76.) For an explicit rejection of the reduce-and-enforce methodology in penalty clauses, see Cad Cam, Inc. v. Underwood, 521 N.E.2d 498, 503 (Ohio Ct. App. 1987).

(77.) Under French law, judges can moderate punitive damages clauses by reducing-though not eliminating--the penalty, to avoid only the excessive element of the penalty. See CODE CIVIL [C. CIV.] arts. 1152, 1226; see also Charles Calleros, Punitive Damages, Liquidated Damages, and Clauses Penales in Contract Actions: A Comparative Analysis of the American Common Law and the French Civil Code, 32 BROOK. J. INT'L L. 67, 106-07 (2006).

(78.) Contract Law (Remedies for Breach of Contract), Sec. 15(a), 5730-1970, SH No. 609 p. 13.

(79.) CA 539/92 Zaken v. Ziva 1, 4 [1994] (unpublished decision) (on file with author) (emphasis added).

(80.) URI YADIN, HOK HAHOZIM: TERUFOT BESHEL HAFARAT HOZEH 1970 [CONTRACT LAW: REMEDIES FOR BREACH OF CONTRACT 1970] 132 (2d ed. 1979) (emphasis added); see also EYAL ZAM1R ET AL., HAPERUSH HAKATSAR LEHUKIM BAMISHPAT HAPRATI [BRIEF COMMENTARY ON LAWS RELATING TO PRIVATE LAW] 302 (2d ed. 1996) ("[T]he measure of reduction of liquidated damages ought to be to the level for which the element of excessiveness no longer applies ... [such that] if that level was set in the first place, it would not have been reduced by the court.").

(81.) See RESTATEMENT (SECOND) OF CONTRACTS [section] 227 (1981).

(82.) See Royal-Globe Ins. Co. v. Craven, 585 N.E.2d 315, 317-18 (Mass. 1992).

(83.) See id. at 317.

(84.) See id. at 317-18.

(85.) Compare Szetela v. Discover Bank, 118 Cal. Rptr. 2d 862 (Ct. App. 2002) (holding a credit card arbitration clause unconscionable because it forbade class actions), with Hutcherson v. Sears Roebuck & Co., 793 N.E.2d 886 (Ill. App. Ct. 2003) (holding the same clause not unconscionable).

(86.) See, e.g., Brecher v. Brown, 17 N.W.2d 377 (Iowa 1945).

(87.) For example, short-term credit can have a very high interest rate that reflects not only the time value of money, but also the "closing" fees. If the debt is to be paid off as scheduled, the high interest rate would be tolerable. But if it is not paid off in the short term, the interest rate can become unconscionable. See, e.g., Carboni v. Arrospide, 2 Cal. Rptr. 2d 845 (Ct. App. 1991).

(88.) See Richard Craswell & John E. Calfee, Deterrence and Uncertain Legal Standards, 2 J.L. ECON. & ORG. 279, 280 (1986).

(89.) The argument that uncertainty over legal standards can lead parties to invest excessively in acquitting information has been developed in the literature on the negligence rule. See, e.g., Louis Kaplow & Steven Shavell, Accuracy in the Assessment of Damages, 39 J.L. & ECON. 191 (1996); Louis Kaplow & Steven Shavell, Private Versus Socially Optimal Provision ofExAnte LegalAdvice, 8 J.L. ECON. & ORG. 306 (1992).

(90.) See, e.g., Carper v. Kanawha Banking & Trust Co., 207 S.E.2d 897, 910 (W. Va. 1974).

(91.) Ex post, some weak parties will be winners. For example, some employees benefit from access to courts rather than arbitration; some employees benefit from elimination of noncompete restraints. But other weak parties will be net losers. For many employees, the above-mentioned benefits are less relevant. From their perspective, the more aggressive the ex post legal intervention, the worse off they are.

(92.) 274 A.2d 78 (N.J. 1970).

(93.) See WRITE & SUMMERS, supra note 4, at 234-35; Craswell, supra note 11, at 1617.

(94.) For example, in Central Adjustment Bureau, Inc. v. Ingram the court notes:
   We recognize the force of the objection that judicial modification
   could permit an employer to insert oppressive and unnecessary
   restrictions into a contract knowing that the courts can modify and
   enforce the covenant on reasonable terms .... [T]he employer may
   have nothing to lose by going to court, thereby provoking needless
   litigation. If there is credible evidence to sustain a finding that
   a contract is deliberately unreasonable and oppressive, then the
   covenant is invalid.


678 S.W.2d 28, 37 (Tenn. 1984) (citation omitted); see also Jenkins v. Jenkins Irrigation, Inc., 259 S.E.2d 47, 51 (Ga. 1979); Walker v. Sheldon, 179 N.E.2d 497 (N.Y. 1961). The same practice is followed in Canada and England. See Machtinger v. HOJ Indus. Ltd., [1992] 1 S.C.R 986 (Can.) ("If the only sanction which employers potentially face for failure to respect the minimum notice periods is an order that they comply with the Act, they will have little incentive to make contracts with their employees that meet the statutory standards."); Mason v. Provident Clothing & Supply Co., [1913] A.C. 724 (H.L.) (Eng.) (severing the entire noncompete clause when drafted in deliberately unreasonable fashion); cf. Goldsoll v. Goldman, [1915] 1 Ch. 292 (H.L.) (Eng.) (severing only the unreasonable increment, with no evidence of deliberate overreaching).

(95.) The Federal Price Gouging Prevention Act proposed using punitive damages in cases of deliberate price gouging. See H.R. 1252, ll0th Cong. [section] 3 (2007) (permitting enforcement by the Federal Trade Commission with fines of up to $3 million); id. [section] 4 (imposing stiff criminal penalties such as frees up to $150 million for corporations and fines up to $2 million and ten years of imprisonment for individuals).

(96.) See, e.g., Small Loans Act, ALA. CODE [section] 5-18-9 (2010).

(97.) See Fair Labor Standards Act of 1938, 29 U.S.C. [section] 216(b) (2006).

(98.) See FARNSWORTH, supra note 2, at 347 ("[A]bsent a showing that [the excessive clause] was drafted in good faith ... the court may fix a lesser restraint than it would have allowed the parties themselves to fix in their agreement.").

(99.) See generally Louis Kaplow, General Characteristics of Rules, in 5 ENCYCLOPEDIA OF LAW AND ECONOMICS 502 (Boudewijn Bouckaert & Gerrit De Geest eds., 2000).

(100.) In the German Civil Code a usurious interest rate is presumptively bad faith and replaced with zero percent. See BURGERLICHES GESETZBUCH [BGB] [CIVIL CODE], Aug. 18, 1896, REICHSGESETZBLATT [RGBL] 195, as amended, [section] 138 (providing that contracts against "good morals" are void); see also WALTER ERMAN, HANDKOMMENTAR ZUM BURGERLICHES GESETZBUCU [COMMENTARY ON THE GERMAN CIVIL CODE] [section] 138, at 333 (11th ed. 2004) (citing case law that establishes zero percent interest when a loan contract stipulated unconscionable interest rate).

(101.) Terms can survive in competition even if not hidden. See, e.g., Xavier Gabaix & David Laibson, Shrouded Attributes, Consumer Myopia, and Information Suppression in Competitive Markets, 121 Q.J. ECON. 505 (2006); Ezra Friedman, Competition and Unconscionability (Oct. 2009) (unpublished paper) (on file with author).

(102.) 226 S.W.2d 673 (Tex. Civ. App. 1949); see also supra text accompanying note 58.

(103.) See Oren Bar-Gill & Omri Ben-Shahar, The Law of Duress and the Economics of Credible Threats, 33 J. LEGAL STUD. 391 (2004); Steven Shavell, Contractual Holdup and Legal Intervention, 36 J. LEGAL STUD. 325 (2007).

(104.) See Ben-Shahar, supra note 1.

Omri Ben-Shahar, Frank and Bernice Greenberg Professor of Law, University of Chicago Law School.

I am grateful to Richard Brooks, Richard Craswell, Robert Hillman, Avery Katz, Doug Laycock, Barak Medina, Ralf Michaels, Mitch Polinsky, Ariel Porat, Scott Shapiro, Steven Shavell, Peter Siegelman, Eyal Zamir, Kathy Zeiler, and workshop participants at Columbia, Duke, Harvard, the Hebrew University, Iowa, Michigan, and Stanford for helpful suggestions.
COPYRIGHT 2011 Stanford Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Ben-Shahar, Omri
Publication:Stanford Law Review
Date:Apr 1, 2011
Words:18601
Previous Article:Guns, fruits, drugs, and documents: a criminal defense lawyer's responsibility for real evidence.
Next Article:Should the Supreme Court stop inviting amici curiae to defend abandoned lower court decisions?
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters