It's Just Like a Contracts Outline, Except that I'm Doing It

I need to do an outline for contracts, but I'm compelled to post. So: Why not do my outline for contracts in a blog post? For those with an aversion to legal topics, this post will probably be pretty damn boring. But it might possibly contain some nugget of information you'd be interested in. I'd recommend the section on what makes a contract enforceable. It's an interesting little doctrine that could affect your life, but which few non-lawyers seem to know the details of. (Note to people who pick on grammar: I have made a conscious decision to capitalize important terms, like Consideration, regardless of whether they are proper nouns. This is to help these terms to stand out. I don't care if it's wrong; it's helpful.)

Will a contract be enforced?

In order for a contract to be enforceable, it must have three things: Offer, Acceptance, and Consideration. Offer and Acceptance are dealt with later, in the context of discussions of Bargaining. Consideration is the most important factor for this discussion. Lack of Consideration is the area where it is most likely that a judge will declare that no enforceable promise exists even though all the parties involve think, or at one point thought, there was a contract. Nonetheless, even if there isn't Consideration, a promise may nonetheless be enforced under Promissory Estoppel or the Material Benefit Rule.

Consideration

Consideration is difficult to define, precisely. The modern doctrine is that Consideration requires that a promise be bargained for. This is imprecise, however, because when determining whether there is consideration the courts do not ask "Did these parties sit down and haggle over this deal before it was made?" If they did, most contracts for sale would be unenforceable; when was the last time you bargained over the price of a Big Mac? What the courts actually ask is "Was this promise made in exchange for a reasonable return promise?" The difficulty in the consideration doctrine comes from the fact that courts tend to apply very close scrutiny to those return promises. It's tough to define what consideration is, precisely, because a lot of times on the margins it's difficult to determine whether the return promise is really a valid exchange or whether it's just a show designed to make a unilateral gift look like an enforceable contract. Thus, consideration is hard because a lot of times whether a return promise was valid Consideration or just a show is rooted in the peculiar circumstances of that particular contract.

The question is significant because courts do not enforce promises to give gifts. There's a pretty valid public policy reason for this. Laws affect how people behave. The laws of contract are byzantine. They subject the parties to heavy liability if they try to back out. Therefore, parties tend to be very cautious about entering contracts. We consider this to be a good thing; if we had no contract law, if people could just back out of contracts whenever they wanted, we'd have a lot of deadweight contracts, inefficient behavior, etc. and this would outweigh whatever gains we got from having more total promises. Because the cost of reneged upon contracts is pretty high, our policy goal is to minimize the number of such promises.

Promises to give gifts are a different matter. The most common gifts are between family members, and we can usually rely upon extra-legal means of enforcing these promises (the strains of commitment, as it were). We also tend to think that the cost of reneged upon promises to give gifts is a lot lower. When one side backs out of a contract, the other side is left having performed their part of the deal for no gain. When someone who promises to give a gift backs out, the other party is miffed, but hasn't really lost anything (Exception: See Promissory Estoppel, below). Finally, there's the concern that there would be a lot fewer gifts given if the law interceded to enforce such promises. When law involves itself in an area of our lives, we are much more cautious about what we say and do. If every promise to give a gift brought with it the spectre of a court case to enforce that promise, we'd make a lot fewer promises to give gifts. Thus, it is believed, the loss caused by fewer gifts that would result from enforcing such promises is greater than the costs of more reneged upon gift promises.

So the law doesn't enforce promises to give gifts. How, then, do you distinguish a promise to give a gift from an enforceable contract? This has become a tricky question because many know of the consideration doctrine and dress up promises to give gifts in the language of contracts, in an attempt to fool the courts. Some general rules:

Williston's Tramp

Performing an action that is necessary to receive the promise is not consideration. For instance: If I tell a homeless person that I will give them $20 if they walk with me to the ATM, their walking with me to the ATM does not constitute a return promise. I'm not asking the person to walk with me for the pleasure of their company; I'm asking them to walk with me so that they can receive the money. If I decide, upon getting there, that I don't want to give them the money after all, they haven't performed any return promise that the courts will consider to be enough to enforce my promise.

Matters of Form: Seals, Peppercorns, and Other Valuable Consideration

Broadly speaking, writing out a contract to give a gift and declaring that one party will give the gift, and the other party will, in return, give Consideration does not make the contract enforceable. Contracts like "I'll give you a house, you give me a peppercorn," or "I'll give you $1000 a month for the rest of your life, you give me $1 and other valuable consideration," are promises to give gifts with some pseudo-Consideration embellishments, and are unenforceable. "Other valuable Consideration" shows up a lot in contracts, often as an attempt to pull the wool over the judge's eyes.

There's also the doctrine of Seals, which the professor went on about a lot in class. There once was a time when Americans could attach a seal to a contract and BAM! it was enforceable, no matter what. You can still do this in England. The sealing process once involved an elaborate ritual, with signet rings and melted wax and such. Gradually, as Americans moved to the frontier and could less well afford such niceties, the ceremony for making a seal grew less formal, until eventually it reached a form where a party could put the letters L.S. (an abbreviation for a Latin phrase meaning something like "I Attach My Seal Here.") at the end of a pre-printed contract and that constituted a seal. The problem with this is that the elaborate ceremony is the point of the seal; it took a long time, during which the sealing party could think about it and decide to back out. The parties had to really think about what they were getting into before they could eternally bind themselves to a contract. Once you could put L.S. in the text of a contract and make all parties officially eternally bound, you had defeated the entire point of the seal. Therefore, the courts have now said "You know what, America? You abused your seal privileges, so now they're revoked." L.S. means nothing. Attaching your seal means nothing. The seal doctrine is null and void in America.

Seemingly Inadequate Consideration

The interesting thing about Consideration is that the courts are trying to get into the heads of the parties. Were these parties thinking of their exchange as a gift, or as a deal? What they're NOT looking at is the material value of the consideration. Hence the famous legal phrase that even a peppercorn can be valuable consideration. If you're honestly looking for a single peppercorn, and desperate for it, and offer your house in exchange, the courts will accept it. The question is whether you're really interested in the return promise, not whether the court thinks you're rational for giving what you give in exchange for that return promise. Thus you have cases like Batsakis v. Demotsis, where one party lent the other party $25 during the Greek resistance to the Nazis during World War II, and in exchange the other party promised to repay them $2000 + 8% interest per annum after the war ended. The court held the promise to be enforceable, because $25 might have been desperately needed during the war, and therefore, though the monetary values were disproportionate, it seemed like the borrowing party really wanted that $25.

Promissory Estoppel

Having said all that about not enforcing promises to give gifts, sometimes courts do. To return briefly to the justification for not enforcing promises to give gifts, it is generally assumed that the downside of reneging is fairly small. The promisee hasn't done anything for the promise, so she is presumed not to have lost anything. But what if she has lost something? This is where Reliance comes in. Reliance is an action or actions taken that assume that a promise will be performed. These actions will be of benefit if the promise is performed and of detriment if it is not. So a promisee may be put at a loss because of an unperformed promise to give a gift because she relied upon that promise. So one can go to a court and argue that a promise should be enforced because you relied upon it. Chances are, the courts won't buy it, but sometimes they do. There's good reason for courts to treat Reliance-based theories of enforceability skeptically. If they did, they would create an incentive to over-rely. Every promise made would be instantly followed by parties rushing to rely so as to make the promise enforceable, even if it's not rational for them to do so.

So how do courts handle Promissory Estoppel (the doctrine under which they enforce a contract on the basis of reasonable reliance)? It's pretty fuzzy, but in general the reliance has to be reasonable, it has to be foreseen by the promisor, and there has to be some sort of material change of condition (quitting a job because you expect to get money, etc.). But none of these factors is decisive. Indeed, a lot of promises enforced on Promissory Estoppel grounds really have a basis in Consideration. That is, a lot of the times there's an implied consideration, an inducement that can't be proved from the facts, but which is evidenced by the promisee's Reliance. Many times courts are primarily looking to Reliance as proof that there was Consideration, rather than using Reliance to substitute for Consideration.

Promissory Estoppel mostly comes up, these days, in incomplete bargains between businesses. One party relies upon the performance of the contract before the contract has officially been finished, either by acting on the expected contract or forbearing from seeking alternate contracts. This reliance must be reasonable, foreseeable, and substantial. Enforcement mostly comes when it is clear that one party, during the bargaining, clearly induced the other party to take an action in reliance upon the contract being fulfilled, e.g. a supermarket franchise inducing a small-business man to sell the family store and move to Topeka to show that he's committed to the job before making a franchise contract with him to run a store in Topeka.

The Material Benefit Rule

Here we're getting into very iffy territory. As a general rule, past consideration is no consideration. If I say to you, "Because you helped me through tough times, I'm going to give you $1000," the courts still consider that a promise to give a gift, not an enforceable contract. Even though there was a return value given, it isn't properly a return promise because it already occurred. It's not a bargain, it's not a deal to exchange promises. It's a gift that happens to have its origins in some previous action. This makes sense; arguably most promises to give gifts are based in reasons. If we allowed any past performance to be a valid consideration, we'd allow almost all gifts to be enforceable contracts through the back door.

Sometimes courts do enforce contracts based on past consideration, but it's very rare. There has to be a moral obligation to perform the contract. "Because you're a great grandson, I'm going to give you $1000" isn't enough. "Because you saved my life and, in the process, crippled yourself, I'm going to maintain you for the rest of your life" would be a moral obligation. But a moral obligation is not enough! It's necessary, but not sufficient. Essentially a huge number of vague factors come in, and most of the time courts won't enforce these contracts. But in rare circumstances they will. If the promisor carefully reflects on the promise AND received the benefit himself AND never repudiated the promise AND followed through on the promise for a long time AND the past consideration conferred a detriment on the promisee THEN the courts MIGHT think it's enough to enforce the contract. But don't count on it.

The Bargaining Process

As mentioned, an enforceable contract requires an offer and acceptance. This is another area where disputes can arise: Was there a proper offer? Was the offer properly accepted? To understand the trouble over offers, you need to understand that there are two kinds of contracts: Unilateral and Bilateral. A Bilateral Contract is what we often think of when we think of contracts: two parties learn that each can give the other something they want, so they sit down, hash out terms, and reach an agreement. Unilateral Contracts, however, are much more common. When McDonalds offers a Big Mac for $3, they're making an offer for a unilateral contract: Anyone who comes in and gives them $3 can get a Big Mac. They don't particularly care who does it, and they're not interested in haggling over terms. Moreover, there's not really a bargain or a deal; they have an open offer where they will perform some action (giving you a Big Mac) if you perform some action (giving them $3).

When it comes to Bilateral Contracs, however, it's somewhat rare to have a party issue an ultimatum offer and have the other side return with an immediate acceptance. Generally what you have instead is an opening invitation to negotiate. One party states their initial terms, the other party makes a counter-offer, and so on back and forth until we have a "meeting of the minds" and both parties agree to all the terms.

Disputes over offer and acceptance tend to revolve around when a proper offer was made and whether it was properly accepted. The major test for whether an offer is proper is completeness. If, for the particularly promises being exchanged, the offer is sufficiently complete as to encompass all material terms, it's an offer. If it includes only some of the terms, it's an invitation to negotiate. If you ask me for a quote on a pipe order, and I tell you the price, you can't simply say "Deal!" and enforce a contract on the simple terms "I give you some pipe and you pay me the quoted price." There are still questions about when the contract will be performed, quantities, warranty, etc. At the same time, if a department store puts an ad in the newspaper offering to sell a fur coat at a reduced price to the first customer who comes in on a given day, and quotes the price, describes the merchandise, and gives the terms of acceptance, they can't turn around and tell the customer who shows up that it was only an invitation to negotiate and therefore not a legally binding offer to be accepted. In that case, the terms given are sufficiently specific to constitute an offer.

Acceptance generally requires a specific communication by the accepting party to the offering party. Generally, though, offers contain within them the method of their acceptance; this can include signature, signature and delivery, verbal acceptance, commencing performance of the promise, etc. Generally, though, for a Bilateral Contract acceptance must be commicated to the offering party, while commencing performance is enough for a Unilateral Contract.

Contracts can generally be revoked through an official communication of an intention to revoke. Some offers are trickier to revoke, though, like Unilateral Contracts (nowadays, you have to revoke before the other party has commenced performance) or construction bids (which are, in some cases).

Offer and Counter-Offer

Once the Common Law governed this area. A Counter-Offer was taken to be a complete rejection of an offer and a return offer of new terms. This lead to such rules as the Mirror Image Rule (To be enforceable, both sides must agree to contracts that are mirror images of one another. That is, there was no contract until everyone explicitly agreed on everything) and the Last Shot Rule (Whoever sent the last counter-offer, the Last Shot, got their terms when the contract was enforced. That is, when a contract is considered to have gone into effect because both parties behave as though it is in effect, the enforced terms will be whatever the last counter-offer was).

The UCC changed this with Section 2-207. 2-207 is a complex little ball with three parts leading to three different outcomes. Under 2-207(1), a counter-offer is treated as an acceptance of the contracts on all the terms agreed to, coupled with a proposal for modifications to the contract on the altered terms unless the counter-offer is explicitly conditioned on acceptance of the alternate terms. Under 2-207(2), these terms become part of the contract IF the parties to the contract are merchants UNLESS the original offer was conditioned on complete acceptance of the umodified terms of the offer OR the terms materially alter the original offer OR the offering party notifies or has notified the counter-offering party of their objections to the altered or additional terms. Finally, if the parties decide not to work out their agreement explicitly and instead just proceed as though they have a contract, and if they end up with a dispute before a court, 2-207(3) dictates that only the undisputed terms of the contract will be considered part of the contract. Any gaps remaining will be filled by default provisions provided by the UCC.

So what does all this mean? It means in general that counter-offers are considered partial acceptances, leaving alterations open to further hashing out. The exception is if the parties are both merchants, in which case the alterations automatically become part of the contract unless they materially alter the deal. In either of these two cases, a party can prevent automatic acceptance by making it explicit that they won't accept alteration of their offer or counter-offer. Finally, if they let a dispute hang and just perform the contract, the court will decide things by throwing out the disputed terms and just using UCC defaults where needed.

This leads to so-called Battle of the Forms situations. Companies develop standard forms of sale and purchase and such that they throw at other companies when making agreements. These forms almost always say things like "Only the terms of this offer apply and no modifications or additions shall be accepted!" Further, since every company develops these forms, you will often see agreements between companies where each throws its forms at the other. The battling forms are unlikely to contain the same terms, and both are likely to contain clauses indicating a refusal to accept modifications. Further, nobody ever reads these forms, so it is unlikely anyone even knows there's a problem until a dispute arises and each party insists that his form wins out over the other's. Judges generally handle this by saying that the offer and counter-offer have both enacted the explicit rejection of modification clauses of 2-207(2) and (1), respectively, and therefore 2-207(3) applies. The disputed terms will be thrown out and the UCC applied.

There's some dispute over how Battle of the Forms cases apply in disputes over those additional terms you click through when you install software. Is buying the software an offer, and the additional terms a counter-offer? If so, the company's terms would probably be thrown out under 2-207(3). The other view is that the company is making the only offer, the terms in the license, and the user is accepting them. In this case, there is no Battle of the Forms problem and the company's terms govern. The latter interpretation seems to be more prevalent, though right now consumer advocates are fighting for a change to the UCC to get the former result, while Software Company lawyers are fighting to keep things as they are.

Relational Contracts: Long-Term Output Requirements, Exclusive Dealings, Conflicts of Interest, and Contractual Modifications.

Long-Term supply contracts allow companies to avoid risk and ensure stability of supply/demand. Generally the companies arrive at some determined price or price formula and arrange other standing terms (how buyers will request the product, when supplier will deliver them, what sort of quality, etc) and then the contract creates a relationship that exists until either the contract expires or the companies break it (generally leading to breach of contract damages).

Output Requirements

These contracts usually allow the buyer to demand a variable amount of product be supplied. Unless limits on quantity are set in the contract, UCC 2-306 has been interpreted, through its "Good Faith" requirement, to provide protection against either excessive demand (Eastern v. Gulf) or non-existent demand (Empire Gas v. American Bakeries) by the buyer . This is somewhat scant protection, however; it relies upon a judge's assessment of the buyer's motives in demanding too much or too little. If the judge deems the motives pure, or at least acceptable, the supplier may be forced to continue supplying more than they can profitably handle (or less than they had anticipated).

Exclusive Contracts

Exclusive Dealings contracts require that one party will supply the other party with a good, and that other party will be the sole dealer of that good. They bind the supplying party; they can't contract with another party to sell the good, and they can't sell the good themselves. The problem arises, though, that the retailing party could just screw their supplier. They could sell nothing and, through the contract, have taken a competitor off the market. To combat this, the UCC provides a default "Best Efforts" clause in such relational contracts (2-306(2)). This, however, is a vague standard that you'd have to go to a judge to get a specific ruling on. Generally, Best Efforts requires that the party take every reasonable effort to market and sell the product without going to extraordinary lengths. This is tough to rely on, though, and what consitutes Best Efforts for a given industry, product, and corporate situation will be difficult and expensive to prove in court. Parties to contracts therefore often set up alternative means and incentives to induce efficient behavior by the retailer and reduce agency costs.

Termination Clauses

One such means of controlling the other party is the Termination Clause. These generally allow a company to terminate the contract. This presents some problems, though. It reduces some of the incentive to contract in the first place, because there is the concern that ones supply or demand can be pulled out from under them at any moment. It therefore reduces some of the guaranteed business aspect of relational contracts that makes them attractive. To help combat this, there is an implied-in-law Good Faith and Fair Dealings clause in all contracts. Nonetheless, generally if a termination clause contains a provision that the contract may be terminated for no cause, the Good Faith and Fair Dealings clause does not override this and require a good cause for termination. Extraordinary situations may require otherwise, such as highly unequal bargaining power between the parties, but otherwise they courts tend to let No Cause provisions stand. They have, however, interpreted in the Good Faith and Fair Dealings clause a prohibition on termination for Bad Cause. This would tend to put the burden of proof on the terminated against party; the terminator has every incentive to just provide no reason for termination, and the terminated upon must then prove that their reasons were bad.

Non-Compete Covenants

Non-Compete Covenants are also used to control retailers and employees behavior. They provide that while the retailer or employee is the exclusive dealer or worker for the supplier, that retailer or employee will not work for or otherwise compete with the supplier. Further, these covenants tend to restrict their ability to compete after ending the contract; they can't pack up and sell all their information to a competitor. The law imposes heavy restrictions on these Non-Compete Covenants, however. They have to relate to contracts for employment or sale, there must be consideration for the covenant (That is, some portion of the contract must be explicitly in payment for the covenant. You can't, for instance, take an existing worker and demand that he sign onto a new non-compete covenant, you have to give him something additional in return), the covenant must be reasonably limited in time and territory, and it must be necessary to protect the employer. The burden is thus place on the party imposing the Non-Compete Covenant to prove that it is valid, necessary, and properly limited. This is supported by general rights of freedom of movement and contract. Forcing employees never to work in the industry again, anywhere, ever, take on a punitive character and practically treat the employee as chattel.

Modifications of Contracts

Under the Common Law, modifications to a contract not only had to be agreed to by both parties, they had to have consideration. That is, they were treated almost as seperate mini-contracts. If one party increased payment, for instance, with no alteration in the obligations of the other party, this would constitute a modification without consideration and would be thrown out. This was to prevent parties from getting into an advantageous position and then re-negotiating a contract while putting the other party under duress.

Under the UCC, this changed slightly. Modifications are allowed, provided all parties consent, and provided the modifications are made in good faith. This replaces a more strict rule (modifications require consideration) with a loose standard to be decided by a judge (contracts can be modified provided there are no shenanigans going on). This new standard provides much more freedom in contracting, but in turn creates more back-end cost if it goes to litigation.

Regulating the Bargaining Process: Duress, Fraud, Disclosure and Concealment, and Statute of Frauds Limitations on Oral Contracts

The law regulates the bargaining process because sometimes outside circumstances lead to the creation of "bad" contracts. Contracts where the parties are forced to sign on against their will (Duress), where parties are deceived into an agreement against their interests (Fraud and Concealment), or contracts where there simply isn't enough of a paper trail generated to make a reasonable decision at trial (Statute of Frauds).

Duress

As with the other items in this section, Duress is not a cut-and-dried matter. It is not enough merely that one side exerts pressure on the other to get them to contract, or that one side is experiencing outside pressure forcing them into a contract. That's considered a normal part of the hard bargaining process. The pressure exerted must be Wrongful, it must have definitely induced the agreement, and that inducement must have been reasonable. What is Wrongful is up to dispute; clearly this includes illegal or tortious actions, but it can also include immoral actions or actions that are situationally wrong. For instance, this could include actions which have no real purpose to the taker except to threaten the promisor. Second, the party claiming Duress must prove that the inducement taken actually caused them to alter their behavior. If they were planning on agreeing anyway, and the other party casually threatened them, that isn't enough to meet the standards for Duress. Finally, the party claiming Duress must have been left no other reasonable option but to agree to the inducing party's terms. If the party could just go to a competitor and hash out a deal with them, or take some other action to remedy the situation, Duress does not apply. Standards are much higher for Duress alledged to have occurred before hand than it is for Duress after the contract is made, because parties are bound to one another. There's no duty to negotiate in good faith, but there is a duty to re-negotiate in good faith.

Interestingly, Duress has also been applied to cases of situational monopolists, where one party takes advantage of another party's distress. For example, a party is stranded in the desert and a tow truck rescues them, charging a disproportionately exhorbitant fee for the rescue. In this case the party claiming Duress has, by outside circumstances, been placed in a position where he had no choice but to contract, and as such courts will consider the contract void insofar as it imposes unreasonable terms. There is some dispute over the wisdom of this; while it certainly seems immoral to take advantage of people in this situation, the removal of potential economic benefits to these otherwise unprofitable salvage-and-rescue operations may cause them to be performed less frequently, or not at all. Refusing to allow rescuers here to take excess profits from this job may mean there are no more rescuers to save future stranded individuals.

Fraud

Fraud is trickier than most people realize. To start with, Fraud has a higher standard of proof; it requires Clear and Convincing evidence, not just a Preponderance of evidence. Further, the rule is NOT "If you lie, you lose." The party alleging Fraud must prove 1. that there was a misrepresentation of fact, 2. that the statements made by the other party were knowingly and clearly false, 3. that they justifiably relied upon the fraudulent statements, and 4. that those statements actually caused the party to agree to the contract. Remember: Both of the cases for Fraud in the book seem to have been decided wrongly. Look to the dissents in Spiess v. Brandt and Danann Realty Corp. v. Harris. In the former, there should have been no fraud because the representations made were not necessarily false, and because the boys would probably have agreed to the deal on the same terms regardless of the alledged misrepresentations. In the later case, the company attempted to distribute the risk of fraud to the Plaintif by inserting a clause in the contract indicating that "the purchaser agrees that he relied upon none of our statements in deciding to purchase this building." The court bought this argument when they really shouldn't have; you can't create a contract shifting the risk from your malfeasance onto the party you wrong. On the other hand, you could make an argument that this was a sophisticated party who agreed to this term, and therefore was affirming that there was no causation, and therefore part 4 of the test was not satisfied. I don't buy it, but you could make the argument.

We don't automatically slam parties that misrepresent or exaggerate because we believe there is value in "cheap talk" and frank assessments. If every statement were being held up to judicial scrutiny, parties wouldn't want to make any statement that couldn't be independently proven, and contracting would be a much harder process.

Disclosure and Concealment

This is mostly covered by Restatement 161(b). You don't always have to disclose everything you know about everything. However, you could be hit on Concealment grounds if you fail to disclose a fact that 1. you know to be true, 2. if that disclosure would correct a mistake which is 3. a basic assumption the Plaintiff has made, and if 4. that nondisclosure was done in bad faith.

We don't require full disclosure, only disclosure of matters materially relevant to the contract, or matters which one party is unaware of and would alter their tendency to make the deal. We do this for several reasons. We believe that people have strong individual property rights in their information. Only in extraordinary circumstances, then, should we force them to disclose it. Further, we feel that not forcing disclosure forces people to take economically efficient actions to maximize their information and make the right decisions. If it's entirely reasonable that the party alledging concealment could have found the information on their own, or if, indeed, they ought to have known it from the start, it's not the Defendants's job to make them aware of every possible fact beforehand.

Statute of Frauds

Not what it sounds like! Statute of Frauds requirements are limitations on what oral contracts are legally enforceable (the idea being to cut down on the fraud caused by important contracts produced without a paper trail). 4 types of contracts must have a written memorial: 1. Property transfers. 2. Contracts answering the debt of another. 3. Contracts that can't be performed within 1 year. 4. Contracts of Sale. All are important and complex contracts, and all will run into evidentiary difficulties, so we want to maximize the paper trail. Covered by UCC 2-201.

In practice, what counts as a written memorial has been stretched a lot; See Monetti v. Anchor Hocking, where Posner combines a pre-contract memo with a post-contract memo on letterhead (which he counts as a signature) to create a synthesized written memorial, satisfying 2-201.

Contract Interpretation

This whole business is quite complex. The trouble is that it's not enough that we have a contract, we have to get into theories on how the terms of contracts should be interpreted. We can broadly classify the problem into two categories: What terms are included and what are not, and how to interpret the terms we include.

Inclusion and Exclusion of Terms

We start with the Parole Evidence Rule. This says that oral evidence about the terms of a contract cannot be included in figuring out what the terms of a contract are. Further, a contract can be in three states: Unintegrated, Partially Integrated, or Fully Integrated. If there's no final written contract, it's unintegrated and the Parole Evidence Rule doesn't apply. A partially integrated contract contains some of the final terms of the agreement, but not necessarily all of them. The terms that are integrated are fully integrated; you can't introduce new terms that contradict them, ever. If, however, the contract is partially integrated, it is theoretically possible to introduce oral evidence indicating that additional terms should be added, provided they are not contradictory. Finally, a contract may be fully integrated, in which case the terms of the contract are full and complete. No new terms may be introduced at all.

How do you determine if a contract is fully integrated, partially integrated, or unintegrated? First, at Common Law the Four Corners test determines partial integration. Once a contract can be shown to set forth the basic material terms needed for a contract, it is, at the least, partially integrated. If it excludes material terms it is clearly unintegrated, because there must be additional terms floating out there that cause the contract to make sense, so the Parole Evidence Rule doesn't apply. So the Four Corners rule determines partial integration. Many courts, however, reject the strict Four Corners rule, which only allows them to look at the document itself, and allow other outside evidence to determine if the contract is complete as to material terms. Next: Is a contract fully integrated or partially integrated. This is really tough to figure out, and must be divined from reading and interpreting the written document. Some contracts contain Merger Clauses, which indicate that this agreement is the full agreement and no other terms are floating out there. If accepted, this makes the contract fully integrated. However, courts tend to inspect these clauses very closely because they tend to make their way into contracts as boilerplate, and under the circumstances it's important to make sure that all parties fully understood and consciously agreed to the Merger Clause. Without a Merger Clause, things are more tricky, and it's really up to the judge.

Next, the question of including new terms based on prior agreements or oral agreements in a partially integrated contract. Here it matters significantly whether the contract is for a sale of goods under the UCC or whether it's covered by Common Law, and if the latter whether the jurisdiction the question is being decided in uses a hard parole evidence rule or a soft one. Under the Common Law, the rule is that additional terms must be agreed to by both parties, they must be consistent with the rest of the contract, and they must be the sort of terms that would be Naturally Omitted from a written contract. Under the UCC, the same standards of agreement and consistency apply, but it serves to only exclude those terms which would Certainly have been Included had they actually been agreed to.

Interpreting the Terms of the Contract

First, it should be pointed out that you first need to engage in the task of interpreting the existing terms before you can determine if a contract is integrated, at which point you can procede to the above step of determining what outside terms to include or exclude.

There are a number of interpretive techniques under the Common Law. The First Restatement suggests a pure plain meaning rule and, in the alternative, a rule interpreting the law according to how one party would believe the other would interpret the contract. The first standard applies to integrated contracts, the second to unintegrated ones. This can raise some trouble because it's difficult to determine if a contract is integrated without first adopting an interpretive standard.

The Second Restatement allows in context evidence in determining the meaning of terms. Terms are to be interpreted in light of the circumstances of the contract. The UCC is even more liberal: 2-202 sets forth interpretive standards that allow in outside evidence to explain the terms even of fully integrated contracts. Evidence as to Usage of Trade (industry-wide practices), Course of Dealings (prior dealings between the parties) and Course of Performance (how this contract was executed) are all permissible. The UCC is not unlimited in its acceptance of outside material, though. Contradictory evidence is impermissible, and additional terms can't be added to fully integrated agreements.

The UCC runs into problems where the Usage of Trade et al are considered to be part of the contract, but they can't contradict the contract's explicit terms. How can this be, if Usage of Trade et al tell us what the terms mean to begin with? They must have some universal meaning, but this is rejected by the UCC.

Mistake and Excuse

Mistake is the doctrine that if one or both of the parties are unaware or otherwise mis-interpret an existing state of affairs, the contract as written doesn't necessarily apply. Excuse is the doctrine that allows similar outs when an outside event occurs when renders performance impossible or pointless. In both cases, there is no out for the parties; rather the courts seek to alter the contract to accomodate the new state of affairs as the parties now understand them. They generally look to see what the majority of contracting parties in the same situation would set as their terms to deal with the facts as they are now understood, or to aportion the risk that these parties did not foresee. Excuse and Mistake are thus more doctrines of contract interpretation than they are regulations of the bargaining process, as discussed above.

Mutual Mistake is discussed in Section 152 of the Restatement. Unilateral Mistake is in Section 153, and rules of aportioning risk are in Section 154. Generally, if the mistake is mutual and materially affects the contract, the harmed party can void the contract, though judges are cautioned to try and find a way of reforming the contract to continue the relationship under the altered conditions. If the mistake is unilateral, the harmed party can still reform, but only if enforcement would be unconscionable (a high bar) or the other party knew of the mistake and didn't correct it. Finally, neither unilateral nor mutual mitake can be remedied if the party seeking the remedy had the risk apportioned to them under Section 154, by explicit contractual allocation, by awareness at the time of contracting that he was ignorant and failed to correct it, or by judicial allocation of risk.

Courts then have a tricky job in re-forming the contract. They have to put themselves in the shoes of the contracting party and try to construct the sort of agreement that the parties probably would have reached had they anticipated the problem that arose and hashed things out beforehand. Needless to say, a judge is not the two parties, if he's already ruled for one party he's likely to take their interests more into account than the others, and the judicially reformed contract is unlikely to be as satisfactory as a truly hashed out contract. Still, it is arguably better than the old contract. At least for one party.

Remedies

There are three basic kinds of cash remedies: Expectancy Damages, Reliance Damages, and Restitution Damages. Expectancy Damages require the breaching party to pay the breached upon party enough money to put them in the same situation they would have been in had the contract been performed. This does NOT include damages for poor mitigation on the part of the breached upon. If someone breaks the windshield of your car, and you don't put a garbage bag up, you can't sue them from the water damage that occurs when it rains that night. When you get breached upon, you're expected to take reasonable steps to mitigate losses. Also, mitigation is defined as ex post; you don't mitigate before a breach, you mitigate after it's occurred. This tends to put the breached upon at a disadvantage. Reliance damages compensate you just for the amount of money you put into reliance on the contract being fulfilled. Restitution Damages are just what you put into performing your part of the contract; Restitution and Reliance damages put you in the situation that you would have been in had the contract never happened in the first place. Expectancy Damages put you in the situation you would have been in had the contract been performed fully.

Courts prefer Expectancy Damages, even though they tend to encourage inefficient reliance (You know you'll get a full return whether the other party breaches or not, so why not rely to the hilt?). Expectancy encourages efficient decisions on the part of the breaching party, and reduces the trouble of finding out what the damages should be. Seller has the option to go ahead and establish a market value by going ahead and performing the action, then showing the result to the court. They then pay those damages.

There's the further question of whether damages for failure to perform up to specifications should be diminished value or cost of completion, that is, whether, if I don't perform exactly as specified, I should pay the difference in market value between what it should have been and what it is, or if I should pay the cost of hiring someone else to put it right (Plus the cost of finding that person, etc.). Generally courts prefer Diminished Value, because the assumption is that it's larger than Cost of Completion. When it's not, though, it seems to make sense to keep Diminished Value standard, since it would impose heavy burdens on the breaching party to little gain on the breached upon. There's also the concern of demanding Cost of Completion, in those circumstances, so as to pocket the extra money.

Finally, there's Specific Performance. Here, the court issues an injunction requiring the breaching party to perform the contract. Courts are reluctant to do this, because sometimes breaching and paying really is the most efficient outcome. They don't want to interfere with the economics of the situation. Specific Performance is generally only allowed in extremely thin markets, where the breacher is the only or one of the only sources for the required good or service, and that good or service is really required by the breached upon. The general assumption is that the buyer has the advantage in mitigating. But in cases of very thin markets, the Seller has the ability to shave the damages by showing the breached upon didn't mitigate enough.

Finally, some last points: Damages must be certain. No sentimental or aesthetic value. New business present a problem because profits are too speculative. Loss of Good Will and Reputation are also not considered. Damages must be foreseeable. You can't mail your acceptance of an offer to sell a valuable painting with a 34 cent stamp, then sue the Postal Service for the value of the painting when it gets lost on the mail. Why? Because the Postal Service had no idea how valuable it was. If they did, they'd have charged more in the form of insurance. Finally, there's the mitigation principle. Breached upon parties are expected to reasonably mitigate their damages after the breach. You can't claim damages that you could have prevented with reasonable mitigation.

February 2012
Sun Mon Tue Wed Thu Fri Sat
      1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29      

Contact Zach

Friends

Webcomics of Which I am a Fan

Sites I Read Daily: Politics

Sites I Read Daily: Video Gaming

Sites I Read Daily: General Miscellany

About this Entry

This page contains a single entry by Zach published on December 17, 2005 2:23 PM.

My Blog is Back and There's Gonna Be Trouble was the previous entry in this blog.

Pick-up Line is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.

Powered by Movable Type 5.04