Middleton Motors, Inc., a struggling Ford dealership near Madison, Wisconsin, sought managerial and financial assistance from Lindquist Ford, Inc., a successful Ford dealership located in Bettendorf, Iowa. The ensuing negotiations centered on Middleton’s need for management services and a cash infusion from Lindquist. The two dealerships generally agreed that Craig Miller, Lindquist’s general manager, would take over as manager of Middleton and that Lindquist would be compensated for these services based on Middleton’s profits after Miller turned the dealership around. A more specific agreement, however, was not reached.
In the meantime, Miller assumed management responsibilities over both dealerships, and the parties continued to discuss *472 the details of the compensation for Miller’s services, the contemplated cash investment by Lindquist, and other terms of a possible joint venture. The negotiations ultimately fell apart because Lindquist did not come forward with any cash. Middleton, still sustaining losses, fired Miller without having paid for his services.
Lindquist and Miller 1 sued Middleton for breach of contract, promissory estop-pel, quantum meruit, and unjust enrichment, seeking recovery for the eleven months of management services Miller provided Middleton. The district court granted summary judgment for Middleton on the first two claims, and the latter two claims proceeded to trial. The court excluded a large amount of background evidence, believing that the only issues for trial were whether Middleton could overcome a “presumption” that compensation was owed and the amount of damages. After a bench trial, the court entered judgment for Lindquist under both quantum meruit and unjust enrichment; damages were awarded based on the court’s determination of the market rate of compensation for auto-dealership general managers or consultants in the field. Middleton appeals.
We reverse. Quantum meruit and unjust enrichment are both quasi-contractual theories, but the two claims have different elements and damages measures under Wisconsin law. The district court got off on the wrong foot by misconstruing these causes of action. This was understandable given some confusing phraseology in Wisconsin caselaw, but as a result of its misstep, the court failed to try the key issues, erroneously excluded relevant evidence, and failed to weigh the particular equities at stake in the commercial circumstances of this ease. Accordingly, we remand for retrial.
I. Background
A. A Failed Business Relationship
Middleton Motors, Inc., is a Ford dealership located in Middleton, Wisconsin, and owned and operated by brothers Robert, Dave, and Dan Hudson. In the months leading up to the events at issue in this suit, Middleton was experiencing heavy financial losses, and the brothers disagreed about how to best manage the dealership. For assistance Middleton looked to Lindquist Ford, Inc., a Ford dealership located in Bettendorf, Iowa. In 2002 and early 2003, Dave Hudson and Craig Miller, Lindquist’s general manager, spoke generally about the possibility of Lindquist becoming involved in Middleton’s operations and ownership. Meanwhile, Middleton’s situation worsened. In March 2003 Dave Hudson told Miller that he and Robert had placed Dan on a leave of absence, that Middleton continued to sustain losses, and that Middleton needed Lindquist’s help. Before entering into more serious negotiations, the parties signed a confidentiality agreement drafted by Lindquist. It included the following relevant provisions:
In connection with the interest of [Lindquist], in exploring the possible acquisition (the “Transaction”) of all or a portion of the business (the “Business”) owned by you, We are requesting that you or your representatives furnish certain information relating to the Business ....
6.) We acknowledge and agree that unless and until a written definitive agreement concerning the Transaction *473 has been executed neither you, any of your Representatives, us nor any of our Representatives, will have any liability to the other with respect to the Transaction, whether by virtue of this agreement of [sic] any other written or oral expression with respect to the Transaction otherwise.
On April 17, 2003, the two dealerships met to hammer out a deal. Middleton sought Miller’s services as a general manager and a cash infusion from Lindquist in exchange for a profit-sharing agreement. A general understanding was reached that Miller would take over as general manager of Middleton and Lindquist would be paid for these services on a percentage-of-net-profit basis, but the specifics of an agreement were not resolved at this meeting. Nevertheless, the parties agreed that Miller would begin working at Middleton on April 21, 2003, and they would negotiate further terms and commit the agreement to writing sometime later. On April 21 Miller started working as general manager of Middleton while maintaining the same position with Lindquist.
The next attempt at a written agreement came in a June 2 fax from Miller to Middleton. This proposal specified that “[t]he only compensation to [Lindquist] will be the Fee, the use of one vehicle, and the reimbursement of travel, meals and lodging costs.” The “Fee” was defined as 45% of Middleton’s profits; payment was to begin the first month that Middleton showed a net profit. On July 1, 2003, Middleton’s accountant sent an email to Lindquist’s accountant explaining that he (Middleton’s accountant) had met with Miller and rejected the June 2 proposal because it did not require Lindquist to make an up-front cash investment in Middleton. The email also asserted that Lindquist understood from the April 17 meeting that its compensation for Miller’s services would come only from Middleton’s profits once the dealership was in the black.
On August 28 Middleton’s accountant circulated a letter of understanding “for the relationship among the parties to be legally formalized at a later point.” The letter stated that the parties “have agreed to enter into an agreement whereby [Lind-quist] would provide a cash infusion into [Middleton] and take over management of the operations for the fees discussed below.” Those fees included, first, 15% of profits “for recovery of expenses and time associated with the assistance provided by Lindquist” and, second, 22.667% of the remaining real income to be paid for “management of the operations.” The letter of understanding reiterated that payment would begin the first month that Middleton reported a net profit. The letter also called for Lindquist to invest $500,000 in exchange for a 25% equity stake in Middleton.
Over the next several months, the parties continued to negotiate but never came to terms on the specifics of an agreement; Lindquist never made a cash investment in Middleton. On March 24, 2004, fed up and still sustaining losses, Middleton fired Miller. On May 11, 2004, Miller sent a letter to Middleton demanding payment for his services. Despite Middleton’s persistent losses, Miller asked for $32,627.84 as “final payment for the calendar year 2003,” “50% of adjusted profits per the ‘Letter of Understanding’ ” for 2004-2005, and an additional 50% of adjusted profits for 2006. Middleton rejected Miller’s demand, saying it owed nothing because Miller never turned the dealership profitable.
B. Lindquist Files Suit; Middleton Moves for Summary Judgment
Lindquist commenced this action in Iowa state court seeking recovery for Miller’s services. Middleton removed the *474 case to federal court, and the parties agreed to transfer venue to the Western District of Wisconsin. Lindquist filed an amended four-count complaint asserting claims for breach of contract, promissory-estoppel, quantum meruit, and unjust enrichment. Middleton moved for summary judgment on all counts. The district court granted summary judgment for Middleton on the breach-of-contract and promissory-estoppel claims — decisions not challenged on this appeal.
Proceeding to the quantum-meruit claim, the judge framed “[t]he ultimate inquiry” as “whether the parties came to a mutual agreement by their words, conduct or course of dealing, as shown by [the] parties’ external expressions of intention.” Citing
Theuerkauf v. Sutton,
Middleton had argued that because the parties expected compensation to be based on profits and there were no profits, Lind-quist could not recover. The judge rejected this argument, concluding that the parties’ expectations became irrelevant once the breach-of-contract claim was dismissed. The court also denied summary judgment on the unjust-enrichment claim, holding that there were material facts for trial on the value of Miller’s services and the equities of permitting Middleton to retain the benefit of his services without paying for them.
C. Trial
Lindquist’s quantum-meruit and unjust-enrichment claims then proceeded to a bench trial. Before opening statements, Lindquist made two motions in limine relevant to this appeal. First, it asked the district court to exclude as irrelevant any evidence of the parties’ understandings and expectations that Lindquist would be paid for Miller’s services based on profits alone. Picking up on an aspect of the district court’s summary-judgment ruling, Lindquist argued that this evidence was immaterial because the breach-of-contract claim had been dismissed. Lindquist’s counsel maintained that “the elements of [the quantummeruit claim] are that the plaintiffs requested the services and that the services were rendered,” and because those elements were essentially undisputed, “all we need to prove is the reasonable value of the services.” Second, Lindquist asked the court to exclude large portions of the proposed expert testimony of Middleton’s accountant, including his conclusion that Lindquist’s decision not to come through with the contemplated $500,000 investment in Middleton had increased Middleton’s risk.
The district court granted the motions in limine. The judge emphasized that the scope of the trial was limited:
[W]e’re going to try the question whether there was any benefit received by defendant for Mr. Miller’s services while he was there. There is — that is the single question that is at issue, was Middleton Motors enriched by Mr. Miller’s services, or as it feared, if the agreement didn’t go through, did Mr. Miller run it into the ground....
So what we’re going to be deciding today is was Middleton Motors enriched; did it receive a benefit from Mr. Miller’s *475 services. If it didn’t, it doesn’t owe Lindquist for any money for those services. If it was enriched, it does owe for whatever benefit it received.
The district court reiterated this understanding of the issues for trial later in the pretrial hearing: “We’re not looking at the terms of the contract that was never entered into to determine what the value of the services were and how they would be calculated.”
The three-day bench trial that followed thus focused narrowly on whether Miller conferred a benefit on Middleton and if so, what it was worth. In an oral ruling, the district court concluded that Miller’s services benefited Middleton and entered judgment in Lindquist’s favor. The court awarded damages of $160,000 based on a determination that this represented the going rate for general managers of auto dealerships in the relevant market and alternatively, that Miller would have charged a similar amount had he been hired as a consultant. The judge did not mention the unjust-enrichment claim in her oral ruling.
The court later amended the judgment by written order. Recapping the pretrial and trial proceedings, the judge said that when summary judgment on the quantum-meruit claim was denied, the court had determined that Lindquist had established all three elements set forth in
Theuerkauf,
In the order amending the judgment, the court also held that Middleton would be unjustly enriched if it did not pay for Miller’s services. The court determined that the damages on this claim were identical to the damages on the quantum-meru-it claim “because that is what a general manager’s services were worth in the marketplace and defendant failed to prove that it received less value from Miller than what a manager is worth in the marketplace.” This appeal followed.
II. Discussion
Middleton challenges the district court’s handling of almost every aspect of this case. It argues that: (1) it was entitled to summary judgment on the quantum-meruit and unjust-enrichment claims; (2) the district court erroneously excluded large amounts of relevant evidence at trial; (3) the district court erred in granting judgment for Lindquist on both claims; and (4) the district court erred in calculating damages. We review the denial of summary judgment de novo.
Guzman v. Sheahan,
A. Confidentiality Agreement
We first address one of Middleton’s central, but weak, liability arguments. Middleton claims that the confidentiality
*476
agreement precludes liability altogether and that the district court erred by disregarding it. According to Middleton, the parties drafted the confidentiality agreement to be as inclusive as possible, and they explicitly agreed that neither party would be liable to the other in the absence of a “written definitive agreement.” Because there was no written agreement, Middleton contends it is not liable for Miller’s services. We review de novo the interpretation of an unambiguous contract.
Estate of Sustache v. Am. Family Mut. Ins. Co.,
Like the district court, we conclude that the language of the confidentiality agreement does not support Middleton’s interpretation. The agreement is not as broad as Middleton suggests. The no-liability provision only bars liability “with respect to the Transaction.” (Emphasis added.) The contract, in turn, defines “Transaction” as the “explor[ation of] the possible acquisition ... of all or a portion of [Middleton].” Hiring Miller as a general manager lies outside the “Transaction” as defined in the agreement. Furthermore, the confidentiality agreement only bars liability based on “this agreement o[r] any other mitten or oral expression with respect to the Transaction otherwise.” (Emphasis added.) It does not preclude liability based on the provision of services, especially those that stem from a later agreement.
B. Quantum Meruit and Unjust Enrichment Under Wisconsin Law
With that preliminary question resolved, we come to the heart of this appeal, which requires a close analysis of quantum meru-it, unjust enrichment, and the difference between implied-in-fact and implied-in-law contracts under Wisconsin law. We note for starters that Wisconsin caselaw in this area can be confusing; the nomenclature and elements of proof are sometimes mixed up, leading to misconceptions about the nature and requirements of these discrete causes of action. 2 This has produced considerable confusion in this case. We think it helpful, therefore, to restate some first principles.
Generally speaking, if the parties have made an enforceable contract and there is no ground for rescission, then breach-of-contract principles will govern the dispute. In the absence of an enforceable contract, however, a plaintiff may turn to quasi-contractual theories of relief.
Watts v. Watts,
1. Unjust Enrichment
In Wisconsin unjust enrichment is a legal cause of action governed by
*477
equitable principles. The action is “grounded on the moral principle that one who has received a benefit has a duty to make restitution where retaining such a benefit would be unjust.”
Watts,
The measure of damages under unjust enrichment is limited to the value of the benefit conferred on the defendant; any costs the plaintiff may have incurred are generally irrelevant.
Mgmt. Computer Servs., Inc. v. Hawkins,
2. Quantum Meruit
Like unjust enrichment, quantum meruit is a legal cause of action grounded in equitable principles.
Tri State Home Improvement Co. v. Mansavage,
To take advantage of the more liberal recovery rule of quantum meruit, a
*478
plaintiff must prove two elements, both relating to the parties’ course of conduct. As explained by the Wisconsin Supreme Court in
Ramsey,'
to recover under quantum meruit, the plaintiff must prove that “the defendant requested the [plaintiffs] services” and “the plaintiff expected reasonable compensation” for the services.
Ramsey’s second element is potentially problematic as phrased. We have not found any Wisconsin case denying recovery under quantum meruit because the plaintiff expected unreasonable compensation. This makes sense. Suppose a defendant asks a plaintiff to paint his house and the plaintiff complies, expecting compensation. Suppose further that the plaintiff expected an unreasonable rate of compensation — say, $100,000, when the house is small and the painting services are worth far less. That the plaintiff subjectively expected “unreasonable compensation” rather than “reasonable compensation” should not necessarily defeat recovery under quantum meruit, properly understood. Rather, the outcome in this hypothetical case should be an award of damages for the plaintiff, albeit it at a lower rate based on the community standard. We suspect what the Ramsey court meant was that the plaintiff must reasonably expect compensation, not that he must expect reasonable compensation. See 26 RiohaRD A. LORD, Williston ON CoNtracts § 68:1, at 24 (4th ed.2007) [hereinafter Williston] (“The courts have generally allowed quasi-eon-tractual recovery for services rendered when a party confers a benefit with a reasonable expectation of payment.”).
Furthermore, we must not lose sight of the fact that quantum meruit is rooted in equity. If equity does not lie with the plaintiff, he will not recover under quantum meruit. As the leading contracts treatise puts it:
Although the remedy of quantum meruit was developed as part of the common law of contracts to avoid unjust enrichment under a contract implied by law, equitable considerations influence the determination of whether recovery is warranted in a given case. The duty to pay arises not from the intent of the parties but from the law of natural justice and equity.
26 Williston § 68:1, at 25 (footnote omitted);
see also Seegers,
Several of Wisconsin’s quantum-meruit cases involve women who provided household services to unrelated decedents and then sued their estates after the decedents’ deaths.
E.g., Brooks v. Steffes (In re Estate of Steffes),
Based on this line of cases, Lindquist suggests that there is recovery in quantum meruit under Wisconsin law whenever a plaintiff does not render the services gra-tuitiously. We think this argument goes too far. In the cited cases, the courts considered the underlying domestic factual setting. In that context, under ordinary circumstances, it is reasonable to believe that the plaintiffs either expected to be paid or did not expect to be paid; the services are either gratuitous or not, and there is little room for a middle ground. This generalization does not necessarily apply across the board or so easily translate to the commercial sphere where the negotiations of sophisticated parties focus on contingencies and other complex considerations.
An example highlights our concern about extending the “gratuitous” factor outside the domestic context in which it usually appears. Imagine that a client asks a lawyer to represent her in litigation. The lawyer agrees to accept the case on a contingent-fee basis but demands 70% of the verdict. The client accepts the contingent-fee arrangement but balks at the 70% figure. They battle back and forth without agreeing on percentage. The two decide nevertheless that the lawyer will represent the client. The case goes to trial, and the client loses.
Under Lindquist’s theory of quantum meruit, because the legal services were not performed gratuitously, the lawyer prevails in a quantum-meruit action against the client. The inequity of this result is readily apparent. Contingent-fee payments are well established in legal practice, and under any conceivable understanding, the lawyer would have recovered nothing when his client lost. In the language of quantum meruit, the lawyer reasonably expected compensation,
but only if he won the case;
he did not expect compensation — or his expectation of compensation was unreasonable — if he lost. Now suppose the lawyer had won the case and he seeks to collect under quantum meruit. The court equitably supplies a price term, looking to the parties’ negotiations, the percentage other lawyers collect in the community for similar work, the prevailing ethical standards in the profession, and the like.
Cf. Tonn v. Reuter,
In this example, if courts were to pay a lawyer under quantum meruit when he wins
and
when he loses, the lawyer will be grossly overcompensated. He is better off, or as well off, in either state of the world
because
he did not enter into the contract. This result in effect requires the client, perhaps too poor to have paid the lawyer by the hour, to supply insurance against a risk the two parties appreciated when they formed their relationship. Furthermore, such a result twists incentives.
Ex ante
the lawyer now prefers
not
to contract and is more indifferent to his client’s success, undermining a key rationale for contingency arrangements— whether for a lawyer, as in this hypothetical, or a general manager, as in our case. The only fair and administrable rule is to let the lawyer take the bad with the good. If the contingency does not materialize, the lawyer should lose on the quantum-meruit action.
See Liss v. Studeny,
This is not to suggest that the lawyer necessarily cannot recover under quantum meruit when the contingency does not materialize. Suppose the client frustrates the lawyer’s ability to win the case or fires the lawyer on the eve of a winnable trial. The result will depend on the facts and circumstances of each case. The trial court is generally in the best position to consider the facts and weigh the equities, and appellate courts should generally affirm when convinced that the correct equitable considerations have been regarded. Nevertheless, it should be clear from this discussion that the parties’ failed negotiations are relevant under quantum meruit — and, of course, under unjust enrichment, whose elements expressly include equity — because they can show, perhaps decisively, what the plaintiff expected when he rendered services. They may also show whether any expectation of compensation was reasonable.
3. Contracts Implied in Fact and Implied in Law
Before we apply these principles to this case, we pause to clarify some terminology that often creates confusion in this area of the law. As we have noted, quantum meruit and unjust enrichment are quasi-contract actions (obligations imposed by law in the absence of a contract).
Ramsey,
Wisconsin also recognizes contracts “implied in fact,” but a “quasi-contract or contract implied-in-law differs markedly from a contract implied-in-fact.”
Stromsted v. St. Michael Hosp. of Francis
*481
can Sisters (In re Estate of Stromsted),
... requires, the same as an express contract, the element of mutual meeting of minds and of intention to contract. The two species differ only in methods of proof. One is established by proof of expression of intention, the other by proof of circumstances from which the intention is implied as matter of fact.
Id.
(quoting
Wojahn v. Nat’l Union Bank,
Thus, Wisconsin recognizes two types of “implied contracts,” but only implied-in-fact contracts rely on contract-formation and breach principles. Compounding the confusion that arises from this linguistic similarity, some Wisconsin cases on quantum meruit invoke implied-in-fact contract cases to describe when a presumption of expected compensation arises.
See, e.g., Steffes,
C. The District Court’s Treatment of the Quantum-Meruit Claim
Much of the confusion in this case arose from the parties’ and the district court’s reliance on
Theuerkauf v. Sutton.
As Lindquist acknowledged in its brief in this court,
Theuerkauf
is an implied-in-fact contract case. It mentions quantum meru-it only once and does so at the end of the opinion when it quotes
Mead v. Ringling,
The district court’s misplaced reliance on Theuerkauf permeated this case. In denying summary judgment, the court cited Theuerkauf to describe the elements of quantum meruit. Accordingly, the court thought the value of the services to defendant established a presumption bearing on “[t]he ultimate inquiry[:] whether the parties came to a mutual agreement by the words, conduct or course of dealing, as shown by [the] parties’ external expressions of intention.” This substitution of the Theuerkauf implied-in-fact contract elements in place of the proper Ramsey quantum-meruit elements was a mistake of law that affected the court’s summary-judgment ruling.
The conceptual confusion continued throughout the trial. The district court excluded all evidence relating to the parties’ negotiations, reasoning that such evidence was irrelevant because this was not a breach-of-contract case. This was a mistake. It is true that this is not a breach-of-contract case (there was insufficient proof of either an express or implied-in- *482 fact contract), but the background evidence remains highly relevant. The parties’ course of conduct, their actions, and their failed negotiations all bear on whether Lindquist reasonably expected compensation at the time Middleton requested, and Miller rendered, his services.
The district court also erred in its order amending its previous judgment. Recounting the earlier proceedings, the court noted that Lindquist had established a re-buttable presumption of intended fair payment, which “left the possibility that at trial, defendant could avoid liability ... by coming forward ‘with evidence sufficient to rebut and overcome the presumption of the existence of an implied contract in fact.’ ” (Citing Theuerkauf and emphasis added.) We have already explained why this importation of implied-in-fact contract principles was improper here.
Accordingly, we conclude that the district court misconstrued the liability principles of quantum meruit under Wisconsin law and consequently mistried the claim. This was understandable given the inconsistencies in some of the caselaw. Nevertheless, the claim must be retried; the district court’s legal error led to the exclusion of relevant evidence on the central question of whether Lindquist reasonably expected compensation for Miller’s services.
See Dandridge v. Williams,
D. The District Court’s Treatment of the Unjust-Enrichment Claim
In contrast to the quantum-meruit claim, the district court correctly identified the three elements for liability under unjust enrichment. The question for us is whether the court correctly applied these elements. It was not clear error to find that Miller conferred a benefit to Middleton. Both sides presented evidence of what transpired at the dealership during Miller’s tenure. The judge found Lindquist’s evidence and argument on this point to be more credible and persuasive. On the second element of the claim, there is no question that Middleton accepted whatever benefit Miller conferred.
Our concern lies with the third element of unjust enrichment — whether “it would be inequitable [for Middleton] to retain the benefit without payment.”
See Seegers,
It also appears that the court excluded evidence that Middleton rolled back Miller’s operational changes as soon as he left, perhaps to undercut Miller’s claim that his efforts would eventually turn Middleton around. This evidence would tend to favor Lindquist’s position. These evidentiary decisions flowed from a legal error: the *483 court’s too-narrow view of the equitable element of unjust enrichment. Accordingly, this claim too must be retried.
If the court determines on remand that Lindquist expected to be paid only if Miller turned Middleton profitable and that Miller did not turn Middleton profitable after a fair attempt, then the court should enter judgment for Middleton under both quantum meruit and unjust enrichment. If the facts are as Middleton describes them, then Lindquist gambled and lost on its bet. Equity requires that it internalize the consequences.
E. Damages
We conclude with a brief word about damages, which we need not fully address in light of our decision on liability. The district court calculated damages in part by comparing Miller to a consultant who charges auto dealerships a couple thousand dollars a day for his services. We see nothing in the record to suggest that any general manager is compensated in this way. Also, while the court expressed concern about compensating Miller based on Middleton’s profits when none existed, the parties should be able to provide evidence of how general managers are paid in dealerships that lose money. Finally, Middleton argues that because Miller worked at several dealerships at one time, he should be paid as if he worked only part-time. Here again, additional facts could help the analysis. (Does a dealership pay its general manager less if he works for several companies at once? Are general managers instead given a certain amount per dealership, which is multiplied for each dealership a manager services?) There should be no need to resort to conjecture when the answers are obtainable.
III. Conclusion
For the foregoing reasons, we Reverse the district court’s judgment on both the quantum-meruit and unjust-enrichment claims, and Remand for further proceedings consistent with this opinion.
Notes
. Steven Lindquist, one of Lindquist's owners, is also a plaintiff. We will refer to the plaintiffs collectively as "Lindquist” unless the context requires otherwise.
. Compare Watts v. Watts,
. We note an apparent anomaly in Wisconsin law. As phrased, the third element of unjust enrichment appears to require an equitable determination by the court. In the famous case of
Hoffman v. Red Owl Stores, Inc.,
. Because it is sometimes difficult to establish whether a plaintiff expected payment, Wisconsin cases have allowed for a rebuttable presumption of payment in some instances. In the cited cases, for example, the court would presume that a plaintiff expected to be paid when she performed household services to an unrelated man.
E.g., Gename,
The use of presumptions in these cases is unsurprising because presumptions are employed where it is difficult to prove the existence of a fact. In the case at hand, the parties and the district court focused too heavily on presumptions and unnecessarily complicated the analysis. There is no need for a presumption where, there is direct or circumstantial evidence about expectation of payment; applying a presumption in such a case is not only unnecessary but inappropiate. Here, there is both direct and circumstantial evidence about Lindquist's expectation of payment, and therefore no presumption applies.
