REINHARDT, Circuit Judge.
This case presents the question whether a publisher retains the right to reject an author’s manuscript written pursuant to a standard industry agreement, even though the manuscript is-of the quality contemplated by both parties. In this case, attorney Rafael Chodos entered into a standard Author Agreement with the Bancrofts Whitney Publishing Company under which he agreed to write a treatise on the intriguing subject of the law of fiduciary duty. The agreement is widely used in the publishing industry for traditional literary works as well as for specialized volumes. Bancrofh-Whitney thought that the treatise would be successful commercially and that it would result in substantial profits for both the author and the publisher. After Chodos had spent a number of years fulfilling his part of the bargain and had submitted a completed manuscript, Ban-crofb-Whitney’s successor, the West Publishing Company, came to a contrary conclusion. It declined to publish the treatise, citing solely sales and marketing reasons. Like a good lawyer, Chodos responded by suing for damages, first for breach of contract, and then, after amending his complaint to drop that claim, in quantum me-ruit. The district court held that under the terms of the contract West’s decision not to publish was within its discretion, and granted summary judgment in West’s favor. Chodos appeals, and we reverse.
I. BACKGROUND
Rafael Chodos is a California attorney whose specialty is the law of fiduciary duty.
His practice consists primarily of matters involving fiduciary issues such as partnership disputes, corporate dissolu-tions, and joint ventures. Prior to being admitted to the bar in 1977, Chodos worked as a software engineer. Beginning in approximately 1989, Chodos began developing the idea of writing a treatise on the law of fiduciary duty that included a traditional print component as well as an electronic component that incorporated search engines, linking capabilities, and electronic indexing. Chodos sought to draw on both his legal and technological expertise, and was motivated in part by the fact that there was, and continues to be, no systematic scholarly treatment of the law of fiduciary duty.
In early 1995, Chodos sent a detailed proposal, which included a tentative table of contents, to the Bancrofb-Whitney Corporation. Bancroft was at the time a leading publisher of legal texts. William Farber, an Associate Publisher, promptly
responded to Chodos’s proposal, and informed him that the Bancroft editorial staff was enthusiastic about both the subject matter and the technological features of the proposed project. In July, 1995, Bancroft and Chodos entered into an Author Agreement, which both parties agree is a standard form contract used to govern the composition of a literary work for hire.
The Author Agreement provided for no payments to Chodos prior to publication, and a 15% share of the gross revenues from sales of the work. Farber informed Chodos that a typical successful title published by Bancroft grossed $1 million over a five-year period, although Chodos’s work, of course, might be more or less successful than the average. Chodos sought publication of the work not only for the direct financial rewards, but also for the enhanced professional reputation he might receive from the publication of a treatise, which in turn might result in additional referrals to his practice and increased fees for him.
From July, 1995 through June, 1998, Chodos’s principal professional activity was the writing of the treatise. He significantly limited the time spent on his law practice, and devoted several hours each morning as well as most weekends to the book project. Chodos estimates that he spent at least 3600 hours over the course of three years on writing the treatise and developing' the accompanying electronic matérials. He did so with the guidance of Bancroft staff. For example, in late, 1995 or early 1996, Farber instructed Chodos that because Bancroft viewed the book as a practice aid and not as an academic work, he should delete an introductory chapter that was primarily historical and disperse the historical material throughout the text, in footnote form. As Chodos completed each of the chapters, he submitted them to Bancroft on a CD-ROM; the seventh and final chapter was sent to the publisher in February, 1998. When finished, the book consisted of 1247 pages.
In mid-1996, Bancroft-Whitney was purchased by the West Publishing Group, and the two entities merged at the end of the year. The Bancroft editors, now employed by West, continued to work with Chodos in preparing the work for publication, although West did establish a management position that ultimately had a direct bearing on Chodos’s career as a treatise-writer, that of Director of Product Development and Management for the Western Market Center. Between February and June, 1998, after the entire treatise had been submitted, Chodos reviewed the manuscript to ensure that the formatting was consistent and that no substantive gaps existed. In the summer of 1998 the West editors provided him with detailed notes and suggestions, to which he diligéntly responded. In November, 1998, West again sent Chodos a lengthy letter including substantive editorial suggestions related to the organization of the book. In early December, 1998, West sent Chodos yet another letter, this time apologizing for delays in publication, and assuring him that publication would take place in the first quarter of 1999. Burt Levy, who replaced Farber as Chodos’s editor, informed Cho-dos that copy editors were preparing the manuscript for release in the early part of that year.
After receiving no communication from Levy in January, 1999, Chodos contacted West to check on the status of his treatise. On February 4, 1999, Chodos received a response from Nell Petri, a member of the marketing department. Petri informed Chodos that West had decided not to publish the book because it did not “fit within [West’s] current product mix” and because of concerns about its “market potential.”
West admits, however, that the manuscript was of “high quality” and that its decision was not due to any literary shortcomings in Chodos’s work.
The decision not to publish the treatise on fiduciary duty was made by Carole Gamble, who joined West as Director of Product Development and Management for the Western Market Center at about the same time that Chodos completed the manuscript. In late 1998, West developed new internal criteria to guide publication decisions. Applying these criteria, Gamble decided not to go forward with the publication of the treatise. She did not in fact read what Chodos had written, but instead reviewed a detailed outline of the treatise and the original proposal for it. Gamble did not prepare a business analysis prior to making her decision. After Chodos informed West that in his view the publisher had breached its contract, West did prepare an economic projection that concluded that the publication of Chodos’s work would be an unprofitable venture. Thus, this legal action was born.
Proceedings Below
Chodos filed an action against West for breach of contract in Los Angeles Superi- or Court in March, 1999, shortly after the publisher’s decision not to publish his work, and West removed the case to federal court ■ on the basis of diversity jurisdiction.- . Chodos immediately moved for summary judgment, which was denied. Shortly thereafter, he amended his complaint to seek restitution on a quantum meruit basis and dropped the breach of contract claim. West moved to dismiss the amended complaint for failure to state a claim,.and the motion was denied. At the conclusion of discovery, West moved for summary judgment, and Chodos sought to amend the complaint again, in order to add a claim for fraud.. The district-court granted West’s motion and entered judgment in its favor; it simultaneously denied Chodos leave to amend his complaint.
II. DISCUSSION
Chodos makes two alternative arguments: first, that the standard Author Agreement is an illusory contract, and second, that if a valid contract does exist, West breached it. Under either theory of liability, Chodos contends that he is entitled to recover in quantum meruit.
A. The Author Agreement Is Not Illusory.
In support of his first argument, Chodos correctly notes that in order for a contract to be enforceable under California law, it must impose binding obligations on each party.
Bleecher v. Conte,
29 Cal.3d 345, 350, 213 Cal.Rptr. 852, 698 P.2d 1154 (1981). The California Supreme Court has held that “if one of the promises leaves a party free to perform or to withdraw from the agreement at his own unrestricted pleasure, the promise is deemed illusory and it provides no consideration.”
Mattel v. Hopper,
51 Cal.2d 119, 122, 330 P.2d 625 (1958). Chodos contends that because the contract required him to produce a work of publishable quality, but allowed West, in its discretion, to decide unilaterally whether or no.t to publish his work, the contract violates the doctrine, of mutuality of obligation and.is therefore illusory.
California law, like the law in most states, provides that a covenant of good faith and fair dealing is an implied term in every contract.
Carma Developers (Cal.) v. Marathon Dev. Cal., 2
Cal.4th
342, 372-73, 6 Cal.Rptr.2d 467, 826 P.2d 710 (1992);
see Russell v. Princeton Laboratories, Inc.,
50 N.J. 30, 231 A.2d 800, 805 (1967) (noting that a “majority, of the courts” read a good-faith obligation into contracts providing one party with discretion);
Boston Road Shopping Center v. Teachers Ins. and Annuity Ass’n of America,
13 A.D.2d 106, 213 N.Y.S.2d 522, aff'd. 11 N.Y.2d 831, 227 N.Y.S.2d 444, 182 N.E.2d 116 (1962). Thus, a court will not find a contract to be illusory if the implied covenant of good faith and fair dealing can be read to impose an obligation on each party.
See, e.g., Third Story Music v. Waits,
41 Cal.App.4th 798, 805-06, 48 Cal. Rptr.2d 747 (1995) (“[T]he implied covenant of good faith is also applied to contradict an express contractual grant of discretion when necessary to protect an agreement which otherwise would be rendered illusory and unenforceable.”). The covenant of good faith “finds particular application in situations where one party is invested with a discretionary power affecting the rights of another.”
Carma,
2 Cal.4th at 372, 6 Cal.Rptr.2d 467, 826 P.2d 710..
It is correct that the agreement at issue imposes numerous obligations on the author but gives the publisher “the right in its discretion to terminate” the publishing relationship after receiving the manuscript and determining that it is unacceptable. However, we conclude that the contract is not illusory because West’s duty to exercise its discretion is limited by its duty of good faith and fair dealing.
See, e.g.,
Asmas
v. Pacific Bell,
23 Cal.4th 1, 15-16, 96 Cal.Rptr.2d 179, 999 P.2d 71 (2000);
Third Story Music Inc.,
41 Cal.App.4th at 803-04, 48 Cal.Rptr.2d 747;
see also Wood v. Lucy, Lady Duff-Gordon,
222 N.Y. 88, 91, 118 N.E. 214 (1917) (“We are not to suppose that one party was to be placed at the mercy of the other”). More specifically, because the standard Author Agreement obligates the publisher to make a judgment as to the quality or literary merit of the author’s work — to determine whether the work is “acceptable” or “unacceptable” — it must make that judgment in good faith, arid cannot reject a manuscript for other, unrelated reasons.
See Third Story Music,
41 Cal.App.4th at 804, 48 Cal. Rptr.2d 747. Thus, Chodos’s first argument fails.
B. West Breached the Agreement.
Chodos’s alternative argument— that a contract exists and it was breached — is more persuasive. West contends that the Author Agreement allowed it to decline to publish the manuscript after Chodos completed writing it for
any
good-faith reason, regardless of whether the reason was related to the quality or literary merit of Chodos’s manuscript. However, West’s right to terminate the agreement is a limited one defined in two related provisions of the agreement. The first, the “acceptance clause,” establishes that West may decline to publish Chodos’s manuscript if it finds the work to be “unacceptable” in form and content. The acceptance clause, paragraph eight of the agreement, provides that:
After timely receipt of the Work or any portion of the Work prepared- by Author, Publisher shall review it as to both form and content, and notify Author whether it is acceptable or unacceptable in form and content under the terms of this Agreement. In the event that Publisher determines that the Work or any portion of the Work is unacceptable, Publisher shall notify Author of Publisher’s determination and Publisher may exercise its rights under paragraph 4.
The second relevant provision (referred to in the acceptance clause as West’s “rights under paragraph 4”) allows West 'to terminate the publishing agreement if the author does not cure a failure in performance
after being given an opportunity to do so. This provision, numbered paragraph four of the contract and entitled “Author’s Failure to Perform,” states:
[I]f Publisher determines that the Work or any portion of it is not acceptable to publisher as provided in paragraph 8 [the acceptance clause] ... [ajfter thirty (30) days following written notice to author if Author has not cured such failure in performance Publisher has the right in its discretion to terminate this Agreement.
The district court agreed with West that in determining whether a manuscript is satisfactory in form and content under the acceptance clause of the standard Author Agreement, the publisher may in good faith consider solely the likelihood of a book’s commercial success and other similar economic factors. We unequivocally reject the view that the relevant provisions of the Author Agreement may be so construed in the absence of additional language or conditions.
The expansive reading of the acceptance clause suggested by West, is inconsistent with the language of the two contract clauses. Under the agreement, the publisher may deem a manuscript unacceptable only if it is deficient in “form and content.” Thus, had Chodos submitted a badly written, poorly researched, disorganized or substantially incomplete work to West, the publisher would have been well within its rights to find that submission unacceptable under the acceptance clause — as it would were it to reject any work that it believed in good faith lacked literary merit. A publisher bargains for a product of a certain quality and is entitled to reject a work that in its good faith judgment falls short of the bargained-for standard. Nothing in the contract, however, suggests that the ordinary meaning of the words “form and content” was not intended, and nowhere in the contract does it state that the publisher may terminate the agreement if it changes its management structure or its marketing strategy, or if it revises its business or economic forecasts, all matters unrelated to “form and content”.
To the contrary, the fact that the contract required West to afford Chodos an opportunity to cure any deficient performance supports our straightforward reading of the acceptance clause as a provision that relates solely to the quality or literary merit of a submitted work.
As noted above, if West determined that Chodos’s submission was unacceptable, he was to be given a period of time to cure his failure in performance. The inclusion of this provision indicates that a deficiency in “form and content” is one that the author has some power to cure. Chodos has no power to “cure” West’s view that the marketplace for books on fiduciary duty had changed; nor could he “cure” a change in West’s overall marketing strategy and product mix; nor, indeed, could he be expected to do much about a general downturn in economic conditions. The text and structure of the contract thus demonstrate that West’s stated reasons for terminating the agreement were not among those contemplated by the parties.
The uncontroverted evidence in this case is that Chodos worked diligently in cooperation with West — indeed, with West’s encouragement — to produce a work that met the highest professional standard, and that he was successful in that venture. His performance was induced by an agreement
that permitted rejection of the completed manuscript only for deficiencies in “form and content.” Chodos thus labored to complete a work of high quality with the expectation that, if he did so, it would be published. He devoted thousands of hours of labor to the venture, and passed up substantial professional opportunities, only for West to decide that due to the vagaries of its internal reorganizations and changes in its business strategies or in the national economy or the market for legal treatises, his work, albeit admittedly of high quality, was for naught. It would be inequitable, if not unconscionable, for an author to be forced to bear this considerable burden solely because of his publisher’s change in management, its poor planning, or its inadequate financial analyses at the time it entered into- the contract, or even because of an unexpected change in the marketplace. Moreover, to allow a publisher to escape its contractual obligations for these reasons would be directly contrary to both the language and the spirit of the standard Author Agreement.
West urges us to affirm the district court’s ruling because, in its view,' it is well-accepted thafi' regardless of the contract’s failure to mention economic circumstances or market demands, publishers have broad discretion under the acceptance clause of the standard Author Agreement to reject manuscripts for any good faith commercial reason. For this proposition, the district court cited two cases from the Second Circuit involving that same- clause. Although at least one of the cases contains dicta that would support the.district court’s decision, both are distinguishable factually and legally. Moreover, to the extent that either case suggests that a publisher bound by the standard Author Agreement may terminate the contract for
any
reason so long as it acts in good faith, we respectfully reject that view.
In
Doubleday & Co. v. Curtis,
763 F.2d 495, 496 (2d Cir.1985), a publisher rejected a manuscript by the well-known actor but neophyte author, Tony Curtis, on the basis of its poor literary quality. There, as here, the publishing agreement allowed the publisher to reject a submission if it was not satisfactory as to “form and content”..
Id.
However, in
Doubleday,
in direct contrast to the circumstances here, it was agreed that the manuscript was
unsatisfactory
in form and content.
Id.
at 5Ó0. In
Doubleday,
Curtis’s claim was that the publisher had a good-faith obligation under the contract to re-write his admittedly unsatisfactory manuscript and to transform it into one of publishable quality.
Id.
The Second Circuit held that a publisher’s good faith obligation does not stretch that far; thus, the Second Circuit’s essential holding in
Doubleday
has no bearing on the present case.
It is true that the Second Circuit appears to have stated its holding in
Doubleday
more, broadly than the case before the court warranted. The court said:
[W]e hold that a publisher may, in its discretion, terminate a standard publishing contract, provided that the termination is made in good faith, and that the failure of an author to submit a satisfactory manuscript was not caused by the publisher’s bad faith.
Id.
at 501. Still, read in context, the holding does not make it clear whether the court meant that a publisher may reject a manuscript for reasons wholly unrelated to its literary worth or that it may do so only if it determines in good faith that the submitted work is unsatisfactory on its
literary merits. If the former is 'the Second Circuit’s view of the law, we respectfully disagree.
The district court also relied on
Random House, Inc. v. Gold,
464 F.Supp. 1306 (S.D.N.Y.1979). That case is more apposite than
Doubleday
in that the district court there held that a publisher may consider economic circumstances when evaluating a manuscript’s “form and content” under the standard publishing agreement.
Id.
at 1308-09. Although we disagree with that holding for the reasons set forth above, and are certainly not bound by it, we note that even in
Random House
thé court did not go so far as to state that economic considerations may be the
sole
reason for a publisher to décline to publish a manuscript that is in every other respect acceptable. In
Random House,
as in
Doubleday,
the submitted manuscript was not of publishable quality. In contrast to Cho-dos’s work, the editor at Random House considered the manuscript at issue to be “shallow and badly designed.”
Id.
at 1308.
In sum, we reject the district court’s determination that West acted within the discretion afforded it by the Author Agreement when it decided not to publish Cho-dos’s manuscript. Because West concedes that the manuscript was of high quality and that it declined to publish it solely for commercial reasons rather than because of any defect in its form and content, we hold as a matter of law that West breached its agreement with Chodos.
C. Chodos May Pursue A Quantum Meruit Claim.
The district court ruled that if West breached the contract, Chodos could proceed in quantum meruit, but only if the damages were not determinable under the contract. It also stated that a question of material fact existed as to whether contract damages'were determinable. It then granted West summary judgment on the quantum meruit claim because it held that there was no breach of contract. As we have already determined above, the district court erred in finding that no breach occurred. Accordingly, we must consider
the remaining issues relevant to Chodos’s quantum meruit claim.
Under California law, a party who has been injured by a breach of contract may generally elect what remedy to seek. In a leading case on election of remedies, the California Supreme Court stated:
It is well settled in this state that one who has been injured by a breach of contract has an election to pursue any of three remedies, to wit: He may treat the contract as rescinded and may recover upon a quantum meruit so far as he has performed; or he may keep the contract alive, for the benefit of both parties, being at all times ready and able to perform; or, third, he may treat the repudiation as putting an end' to the contract for all purposes of performance, and sue for the profits he would have realized if he had not been prevented from performing.
Alder v. Drudis,
30 Cal.2d 372, 381-82, 182 P.2d 195 (1947) (internal quotation marks omitted).
In employment contracts and contracts for personal .services, like the one before us, the first option, an action in quantum meruit, is generally limited to cases in which the breach occurs after partial performance and the party seeking a recovery does not thereafter complete performance. “Where [a party’s] performance is not prevented, the injured party may elect instead to affirm the contract and complete performance. If such is his election, his exclusive remedy is an action for damages.”
B.C. Richter Contracting Co. v. Continental Casualty Co.,
230 Cal. App.2d 491, 500, 41 Cal.Rptr. 98. (1964) (citing
House v. Piercy,
181 Cal. 247, 251, 183 P. 807 (1919)). Thus, if a plaintiff has fully performed a contract, damages for breach is often the only available remedy.
Oliver v. Campbell,
43 Cal.2d 298, 306, 273 P.2d 15 (1954).
The California Supreme Court has, however, recognized an exception to the general rule. In
Oliver,
the court stated:
The remedy of restitution in money is not available to one who has fully performed his part of a contract, if the only part of the agreed exchange for such performance that has not been rendered by the defendant is a sum of money constituting a liquidated debt;
but full performance does not make restitution unavailable if any part of the consideration due from the defendant in return is something other than a liquidated debt.
Id.
at 306, 273 P.2d 15 (adopting Restatement of Contracts § 350) (emphasis added).
Assuming that Chodos fully performed his end of the bargain by delivering a completed manuscript to West,
then whether Chodos can recover on a quantum meruit claim turns on whether the 15% of the gross revenues provided for in the
agreement constitutes a “liquidated debt”. According to Black’s Law Dictionary, “[a] debt is liquidated when it is certain what is due and how much is due. That which has been made certain as to amount due by agreement of parties or by operation of law.”
Black’s Law Dictionary
931 (6th ed. 1990). The term “liquidated debt” is similar to the term “liquidated damages,” which the California courts have defined as “an amount of compensation to be paid in the event of a breach of contract, the sum of which is fixed and certain by agreement. ...”
Kelly v. McDonald,
98 Cal.App. 121, 125, 276 P. 404 (1929) (citation omitted),
overruled in part on other grounds, McCarthy v. Tally,
46 Cal.2d 577, 297 P.2d 981 (1956).
Chodos’s entitlement to 15% of the revenues from his book on fiduciary duty is not a liquidated debt under California law, as it was not a certain or readily ascertainable figure. The mere existence of a fixed percentage royalty in a contract does not render that royalty a “liquidated debt,” if the revenues to which that percentage figure is to be applied cannot be calculated with reasonable certainty. Here, it is impossible to determine even now what those revenues would have been had West not frustrated the completion of the contract. Had. West honored its contractual obligations and published the treatise, the revenues would have depended on any number of circumstances, including how West chose to market the book, and how it was received by readers and critics.
Accordingly, under
Oliver,
Chodos is entitled to sue for restitution for the time and effort he reasonably invested in writing the manuscript.
See also O’Hare v. Peacock Dairies,
26 Cal.App.2d 345, 79 P.2d 433, 442-43 (1938) (holding future profits to be unascertainable where plaintiffs are owed a future revenue stream
from a dairy that had ceased operation). We express no opinion as to how restitution should be calculated in this case,-nor do we intimate any suggestion as to the appropriate amount of such recovery.
D. The District Court Did Not Err in Denying Chodos’s Second Motion to Amend His Complaint.
The denial of a motion to amend a complaint is reviewed for an abuse of discretion.
Simon v. Value Behavioral Health, Inc.,
208 F.3d 1073, 1084 (9th Cir.2000). It is generally our policy to permit amendment with “extreme liberality,”
Morongo Band of Mission Indians v. Rose,
893 F.2d 1074, 1079 (9th Cir.1990), although when a district court has already granted a plaintiff leave to amend, its discretion in deciding subsequent motions to amend is “particularly broad.”
Griggs v. Pace Am. Group, Inc.,
170 F.3d 877, 879 (9th Cir.1999). When considering a motion for leave to amend, a district court must consider whether the proposed amendment results from undue delay, is made in bad faith, will cause prejudice to the opposing party, or is a dilatory tactic.
Foman v. Davis,
371 U.S. 178, 182, 83 S.Ct. 227, 9 L.Ed.2d 222 (1962).
Here, Chodos sought leave to amend his previously-amended complaint to add a claim of fraud against • West, contending that he had learned new facts that supported that claim shortly before the close of discovery. The district court denied the motion, finding that those “new” facts had been available to Chodos even before the first amendment to his complaint. Given this finding, the district court’s conclusion that the motion to amend was made after undue delay did not constitute an abuse of its discretion.
See Lockheed Martin Corp. v. Network Solutions,
194 F.3d 980, 986 (9th Cir.1999). Moreover, the district court also found that the addition of a fraud claim would result in prejudice to West, and that the motion was a dilatory tactic. Those findings also do hot constitute an abuse of discretion. We therefore affirm the district court’s denial of Cho-dos’s motion.
III. CONCLUSION
Because West breached its contract with Chodos by rejecting his manuscript for a reason not permitted by the contract between the parties and because Chodos is entitled to recover for the breach in quantum meruit, we REVERSE the district court’s grant of summary judgment in West’s favor, and REMAND the case to the district court with instructions to enter summary judgment as to liability in Cho-dos’s favor, and for further proceedings consistent with this opinion.
See Bachelder v. America West Airlines,
259 F.3d 1112, 1132 (9th Cir.2001) (reversing a district court’s grant of summary judgment for the defendant and directing that judgment be entered for the plaintiff where both sides had moved for summary judgment below.).
We AFFIRM the district
court’s denial of Chodos’s motion to amend his complaint.
AFFIRMED in part, REVERSED in part, and REMANDED.