OPINION

We address the bankruptcy code’s provision on claims for pre-petition attorneys’ fees, 11 U.S.C. § 502 (b)(4).

FACTS

Charles Siller has been litigating with his brothers over his interest in the family business since 1982. The family business, Siller Brothers, Inc., held among its assets some 500 pieces of real estate, and Siller owned 40% of the stock. By 2001, Siller Brothers, Inc. had obtained a $10 million judgment against Siller, and it was threatening to execute on-his shares. To fight this litigation, Siller retained several law firms at various times.

In 2001, Charles Siller hired the law firm Cotchett, Pitre & Simon to represent him both in his lawsuit with his brothers and a legal malpractice dispute with some of his former lawyers. The Cotchett firm agreed to a contingent fee of 28% of the net settlement or trial award after subtraction of a $10 million judgment against Siller in favor of his brothers. The engagement agreement said that Siller was “without funds to pay hourly fees.” The arbitration provision, initialed by Sil-ler when he signed it, stated that “any dispute” relating to the fee agreement or the Cotchett firm’s performance of services would be submitted to arbitration.

Two and a half years later, Siller retained the Spiller • McProud law firm. Siller hired the Spiller firm “not as additional trial attorneys, but to assist, advise and discuss these legal matters personally” with Siller, and to act as “an interface” for Siller “with the attorneys at the Cotchett law firm.” The Spiller firm was to work to ensure that Siller could “fully understand and [be] in agreement with the Cotchett law firm’s trial strategy, trial preparation (including selection of experts), and conduct of the trial itself.” The Spiller firm was to serve as Siller’s “general counsel” and “to communicate to the Cotchett law firm [Siller’s] ideas, suggestions, and requests.”

Siller wanted Spiller • McProud to convince the Cotchett firm to pursue a theory that Siller’s brother’s death entitled Siller to purchase his brother’s shares for a small fraction of what they were worth, under a separate contract Siller had with his deceased brother. The Spiller firm agreed, but Siller and the Spiller firm expressly agreed that the contingent fee would not depend on the success of this theory. The Spiller firm was to get 8% of the same net amount from which the Cotchett firm’s 28% contingent fee was to be calculated. Siller and the Spiller firm also incorporated the other terms of Sil-ler’s agreement with the Cotchett firm, including the arbitration provision.

The Cotchett firm, consulting with the Spiller firm, won Siller’s case against Siller Brothers, Inc. After a failed mediation and a stay of execution on Siller Brothers, Inc.’s $10 million judgment against Siller, the case went to trial in California Superi- or Court. The judge issued a proposed judgment valuing Siller’s shares at over $56 million. Pending appeal and cross appeal, the parties settled for $10 million cash to Siller and $20.5 million worth of real estate in exchange for Siller’s shares. Consistent with Siller’s fee agreements with the firms, the contingent fees were to be based upon the $30.5 million value of the settlement. Documentation was delayed while Siller’s counsel, working with tax experts, created for Siller a new corporate spinoff, CWS Enterprises, Inc., so that Siller could mitigate the tax impact of his lawyers’ victory.

The firms also tried other cases that Siller insisted on. While the dissolution case was ongoing, the Cotchett and Spiller firms filed a separate action against Siller Brothers, Inc. and pursued Siller’s preferred theory (that Siller had acquired a right to buy his deceased brothers’ shares cheaply under a separate contract). They lost that case. The two firms also lost a malpractice case Siller brought against one of his previous lawyers, and they negotiated a $41,000 settlement in another case where Siller had refused to pay a different set of previous lawyers. (Siller then refused to pay even the $41,000 settlement, so those disputes remained pending after he moved on from the Cotchett and Spiller firms.)

Nor would Siller pay the 28% and 8% fees he had agreed to pay the Cotchett and Spiller firms, respectively. Siller’s failure to pay the Spiller firm its 8% contingent fee is the subject of the appeal before us. Siller and the two firms arbitrated-the fee dispute before a retired state judge acting as arbitrator, presenting two days of testimony and extensive argument.

Every aspect of the arbitration, including whether it should take place, was hotly contested. Despite his agreement to arbitrate “any dispute,” Siller refused to do so until the Superior Court denied his ex parte application to prevent such arbitration. Siller then continued his attempt to evade arbitration by way of unsuccessful motions in limine before the arbitrator.

During the arbitration, Siller’s counsel led off his cross examination of the attorneys’ lead witness, Mr. Pitre of Cotchett, Pitre, by asking what his and his associates’ hourly rates were. Siller’s lawyer then asked why they had contracted for a contingent fee. Pitre replied, “Because Mr. Siller didn’t have any money,” so the attorneys would have to be paid “[o]ut of a recovery, hopefully.” The lawyers advanced about $400,000 in expenses, as well as their time and effort. Pitre was initially reluctant to pursue the case against Sil-ler’s deceased brother under the buy-sell agreement (the theory under which Siller would buy his brother’s shares for a valuation price that was a small fraction of what the shares were worth) because he doubted that agreement’s enforceability. But, as Spiller had agreed to do in his retainer agreement, he consulted with Pitre and came up with a theory that the two firms could advance with a straight face. (They lost that portion of the case, as they did the malpractice case Siller had started against one of his previous lawyers.)

As Siller’s attorney presented his case during the arbitration, (1) the real estate accounting for two thirds of the settlement had declined in value since the settlement, so if a contingent fee applied at all, it should be against much less than the $30.5 million; (2) the value of the legal services was far less than the contingent fee would yield; (3) much of the work, including the failed lawsuits against the deceased brother’s estate and the failed lawsuit against one of Siller’s previous attorneys, produced no value, so the attorneys should not be compensated for it; (4) Spiller had participated actively in the successful trial, but had only been retained as “general counsel” to consult, so he ought not to be compensated for any of that time; and (5) the money went to the spinoff created for Siller to avoid taxes, not to Siller, and the spinoff had not signed the fee agreement, so no fees were due.

Siller’s case focused largely on the reasonable value of the Cotchett and Spiller firms’ respective services, both as an alternative to the contingent fee agreement and as justification for not holding him to his contingent fee agreement. Siller’s attorney urged the arbitrator to limit compensation to “the reasonable value of the services rendered” because the Cotchett and Spiller firms had “failed to fulfill” their contracts. He argued that “[i]f you have a breach of contract then the answer’s in quantum meruit.” “And even if it’s a contract, the law says you must prove reasonable value of the services, and that goes to the performance.” Siller’s attorney argued that the hours were also relevant to “the issue of eonscionability.” “[T]he consciona-bility of this fee will be determined also based upon ... [the] hours they devoted to matters that they failed to bring to conclusion through their own errors,” referring to the unsuccessful malpractice case against one of Siller’s prior lawyers.

The Cotchett and Spiller firms objected to all this evidence, which was the bulk of Siller’s case, on the theory that evidence as to quantum meruit was not necessary in a breach of contract action. But they conceded that Siller could “inquire about the amount of time put into the matters ... the time for their quantum meruit,” because it was an issue the arbitrator could reach if he deemed the contract void.

Siller’s attorney pointed out that the arbitrator had not yet ruled on whether the contingent fee agreement was binding, so “quantum meruit is still an issue.” He argued that if the contract was not deemed to be binding, then quantum meruit would determine the proper amount of the fee, and even if the contract were binding, counsel would still have to show a reasonable effort to justify the amount of the fee under state bar rules. So “[ultimately this comes down to hours,” Siller’s attorney argued, which apparently is why Siller’s presentation was largely a challenge to how many hours ought not to be compensated because they were either not eon-tracted-for or led to no success.

The arbitrator overruled the Cotchett and Spiller firms’ objections and let in all of Siller’s evidence going to the value of the attorneys’ services, including evidence concerning the reasonableness of the attorneys’ contingent fees, the hours the attorneys worked, and the reasonableness of the time they spent relative to the results (except for some details regarding valuation of the parcels of real estate they won for Siller), evidently because that evidence might bear on quantum meruit or uncon-scionability if the arbitrator decided the case on either basis.

Spiller testified that he put in 1,760 hours and that his usual rate was $250 per hour. He and Pitre had told Siller that the actions to enforce the buy-sell agreement and for legal malpractice were likely to be unsuccessful (and they were), but Siller insisted on pursuing them anyway.

After hearing all of this testimony and argument, the arbitrator chose to view the case as a claim on a contract, the written fee agreements. He concluded that the fee agreements were not unconscionable, that they were “reasonable,” and that the two firms were entitled to every penny of the fees they and Siller had agreed to, as well as the expenses the firms had advanced on Siller’s behalf. The Cotchett firm and Siller have since settled, so we need discuss only the Spiller firm’s claim.

The arbitrator found that Siller provided extensive advice almost daily for the three years of the litigation, working “full bore” for “over 1,760 hours on Siller’s behalf.” (The dispute had been litigated through multiple trials and proceedings in multiple courts on multiple theories, including Sil-ler’s disputes with several of his former attorneys.) The arbitrator also concluded that, as Siller himself admitted in his testimony, he had hired Spiller to assist trial counsel, not just to advise.

The arbitrator wrote a detailed, 26-page, single-spaced opinion explaining his decision. After rejecting Siller’s theories for why the arbitration should not proceed, and why it should not bind CWS (the corporate spinoff created so that Siller could avoid taxes on his $30.5 million settlement),' the arbitrator made findings of fact regarding the fee agreements. He found that Spiller “worked ... alongside [the Cotchett lawyers] on all litigation.” As for Siller’s contention that Spiller was not supposed to do that, just act as his “general counsel” to advise and inform him and the Cotchett firm, the arbitrator found otherwise. “Siller testified that he hired Spiller to ‘help Frank [Pitre] and not just to advise ... although Spiller did provide Siller with extensive advice about the progress of the litigation on an almost daily basis throughout three years of representation.” The arbitrator further found that Spiller spent less than 5% of his time on the unsuccessful malpractice litigation against one of Siller’s previous lawyers and sought no fees for that work.

As for not completing the contract, the arbitrator credited Spiller’s contention that Siller fired his lawyers “hoping to avoid paying his now former counsel” by “actions clearly designed to avoid payment of his legal obligations attendant to the extraordinary result obtained,” a $30.5 million settlement on a $45.7 judgment obtained after years of representation in complex litigation.

The critical question for this appeal, as argued on Siller’s behalf, is whether the arbitrator resolved only the issue of whether the contingent fee was unconscionable as applied to the $30.5 million result, or whether the arbitrator also determined the reasonable value of the legal services performed. The transcripts show, and the arbitrator found, that Siller’s attorney “devoted most of his cross examination to a detailed attack on how Mr. Pitre and Mr. Spiller spent their time on the case, on the theory that the Arbitrator might find the contract to be unconscionable (it is not) and that quantum meruit would be relevant.” The arbitrator concluded, though, that quantum meruit was not relevant because the contingent fee agreement was a valid contract. He expressly found that the contracted-for percentages were “reasonable” based on the work put into the case, the risks, and the need for counsel to finance the litigation.

Siller did not pay what the arbitrator concluded he owed, so the Cotchett and Spiller firms sought and obtained confirmation of the award in California Superior Court. The Superior Court entered a money judgment in their favor for the amount of the award. But Siller did not pay the judgment, either; instead, he prevented its execution by filing for chapter 11 bankruptcy.

The bankruptcy court took a fresh look at the “reasonable value” of Spiller’s services under section 502(b)(4) and applied a lodestar approach, multiplying Spider’s hourly rate by those hours that the bankruptcy court adjudged to be productive on the portions of the litigation on which Spiller was successful. Using this approach, the bankruptcy court concluded that Spiller’s fee was unreasonably high and should have been $440,250 (rather than the arbitrator’s figure of just under $2.5 million).

The bankruptcy court’s view was that there were “two tiers of reasonableness scrutiny,” first under state law, which if not satisfied required disallowance of the claim as “unenforceable,” and then under bankruptcy law, independently of state standards. The bankruptcy court did not articulate what the federal standard of reasonableness was, just that it was not necessarily satisfied by reasonableness under state standards. As the bankruptcy court saw it, the arbitrator had not determined reasonableness because he merely conducted a contract analysis under the contingent fee agreement. Responding to the attorneys argument that full faith and credit should be given to the Superior Court judgment, the bankruptcy court held that “as a matter of the Supremacy Clause, and regardless of the state preclusion law, the state-court judgment based on state law cannot trump the specific provision in Bankruptcy Code 502(b)(4).” As for claim and issue preclusion, the bankruptcy court’s view was that since section 502(b)(4) of the bankruptcy code' could not have been raised in the arbitration (because Siller had not yet filed for bankruptcy), the claim arbitrated did not involve the same “primary right.” The bankruptcy court thus could determine reasonableness on a clean slate without taking account of the arbitration award and the California Superior Court judgment.

The district court, on the Spiller firm’s appeal, reversed. The controlling determination in the district court decision was that “the similarity between the standard for determining the unconscionability of a contingent fee agreement, which was before the arbitrator, and the bankruptcy court’s standard for determining the reasonable value of the fees suggests that the issue of an appropriate fee for appellant’s work was necessarily determined and actually litigated in the formal arbitration that took place here.” The district court noted that the bankruptcy court “did not consider the transcript of the arbitration proceedings in determining what was before the arbitrator.” The district court did consider the transcript to see whether reasonableness was at issue. It noted that Siller’s position in the arbitration, fully litigated, was that, despite a contingent fee agreement, the attorneys still had to prove “reasonable effort ... to justify the fee.” The arbitrator heard evidence about the time spent and the complexities and risks of the litigation, not just the contract itself, and evaluated all of these factors for reasonableness. The district court concluded that “[b]y considering both the reasonable nature of the contingent fee contracts and rejecting the claim that the contracts were unconscionable, and applying California’s tests for both determinations, the arbitrator necessarily decided that the fees were reasonable within the contemplation of § 502(b)(4).”

After a trial in the bankruptcy court, Siller and the trustee argued in a second appeal to the district court that the issues adjudicated by the arbitrator were not “identical,” as required by California res judicata law, to the reasonableness determination under the bankruptcy code, so the judgment confirming the arbitration award should have no preclusive effect. But the district court, based on the arbitration transcript and arbitrator’s decision, found that “the arbitrator determined that Spiller’s fees were not unconscionable applying a test equivalent to the federal reasonableness test,” so the California Superior Court judgment was entitled to preclusive effect. Because the case required no further proceedings in bankruptcy court once the amount of Spiller’s claim was liquidated, and because that condition had occurred, the district court held that Spiller was entitled to the full amount of his claim as adjudicated.

Siller and his new spinoff corporation appeal.

ANALYSIS

Two statutes are at issue in this case, the bankruptcy code’s provision on claims for pre-petition attorneys’ fees, 11 U.S.C. § 502 (b)(4), and the Full Faith and Credit Act, 28 U.S.C. § 1738 . The first limits claims after the objection of an interested party. Here is its text:

Here is the text of the Full Faith and Credit Act:

There is little circuit court authority on section 502(b)(4) of the bankruptcy code. It limits claims for services rendered by the debtor’s attorney to the extent that such claims exceed “the reasonable value of such services.” The text leaves room for argument about how to apply it, especially in conjunction with any prior determination regarding attorneys’ fees and with the Full Faith and Credit Act. The district court concluded that a Tenth Circuit decision, Landsing Diversified Properties v. First National Bank and Trust Co. (In re Western Real Estate Fund, Inc.), provided a sound analysis, and so do we. There are a number of bankruptcy court, bankruptcy appellate panel, and district court decisions taking both consistent and different views. We adopt the Tenth Circuit’s position. Federal law benefits from consistency between circuits, and we agree with our sister circuit’s reasoning.

In Western Real Estate, the bankrupt entered into a pre-petition fee agreement with counsel for a reduced hourly rate plus a 25% contingent fee. Agreeing with our decision in In re Yermakov, the Tenth Circuit reasoned that, subject to section 502(b)(4), “a pre-petition contingency fee agreement between the debtor and an attorney is ... ‘like any other contract claim against the estate.’ ” Citing our decision in In re Pacific Far East Line, the Tenth Circuit concluded that the source of allowable contract damages for a breached attorney’s fee agreement is state law. Western Real Estate rejects the notion that under section 502(b)(4), federal law provides for a bankruptcy court to establish the reasonableness of an attorney’s fee in the first instance, independently of state law. Instead, 502(b)(4) works as a federal cap on a fee already determined pursuant to state law. The proper mode of analysis, Western Real Estate holds, is:

Western Real Estate holds that it is error for a bankruptcy court to bypass this analysis, as the bankruptcy court did in this case, and determine for itself in the first instance a reasonable contingent fee using the lodestar method.

Western Real Estate also holds that contingent fee agreements “provide reasonable alternatives to the hourly retainer, despite the fact that, as a result of their contingent and therefore risky nature, such agreements typically generate fees ... substantially in excess of’ lodestar calculations when the lawyer succeeds. Citing our decision in Venegas v. Skaggs, Western Real Estate holds that the contingent and therefore risky nature of a contingent fee is itself an element of the reasonableness analysis.

In the case before us, the Spiller firm’s fee had been reduced to judgment pursuant to state law in a California state court. It had also been “deemed allowed” under section 502(a) subject to Siller’s section 502(b)(4) objection. Siller had argued that because he fired the Cotchett and Spiller firms before the settlement was collected, he did not owe them their contingent fees, But the arbitrator found that this attempt by Siller to evade the fees should fail. The Superior Court judgment established that under California law, the Spiller firm was entitled to $2,497,325.07 for his fees and $800 for costs (plus 10% interest beginning November 25, 2008). So the first and second steps of the Western Real Estate analysis are complete.

Applying the section 502(b)(4) reasonableness test and reducing the claim to the extent of any excess are the third and fourth steps of the analysis, not the first. The bankruptcy court performed its reasonableness analysis from scratch, using the lodestar method, rather than treating it as a cap on the amount allowed under state law. We therefore conclude that the bankruptcy court’s analysis was mistaken in this case for the same reason as the bankruptcy court’s analysis in Western Real Estate.

So far, we have glossed over another question: whether section 502(b)(4) of the bankruptcy code leaves any room to reduce an attorney’s fee that a state court has deemed reasonable as a matter of state law, that is, whether the section 502(b)(4) reasonableness cap can ever require a reduction in such a fee. And we also have not yet spoken to the res judica-ta effect of a judgment entered prior to the filing of a bankruptcy petition.

The Third Circuit, in Anthony v. Inter-form, answered these questions for another of the section 502(b) exceptions. As in our case, the bankruptcy claim in Anthony was based on a pre-petition state court judgment confirming an arbitration award. Anthony construes section •502(b)(7) of the bankruptcy code, which limits a claimant’s damages from the termination of an employment contract to one year’s pay after termination or after the employer filed for bankruptcy (even if the employer breached a multi-year employment contract and left more than a year unfulfilled). The employer in Anthony thwarted the employee’s attempt to execute on his state court judgment by filing for chapter 11 bankruptcy relief, so the employee filed a bankruptcy claim for the amount of the judgment

Anthony rejects the proposition that the 502(b)(7) cap applies only to executory contracts and not to judgments. Even though the employee obtained a state court judgment amounting to several years’ pay after his employer’s breach, in bankruptcy he was entitled only to one year’s pay. The Anthony court looked through the employee’s state court judgment to the claim upon which it was based, applying the section 502(b)(7) cap to that underlying claim.

Reasonable arguments may be made against extending the Third Circuit’s position to the section 502(b)(4) reasonableness cap at issue in our case. The section 602(b)(7) cap is specific and calculable. The section 502(b)(4) cap, which limits attorneys’ fees to a “reasonable” amount, is indefinite in application. The section 502(b)(7) cap applies only to a future expectancy and only where the bankrupt has before filing paid the employee for services rendered. The section 502(b)(4) cap limits fees for services already performed.

Perhaps most strikingly, our sister circuit’s approach is inconsistent with the common law doctrine of merger. In California (as in most, if not all, common law jurisdictions), a claim that has been reduced to a judgment merges into the judgment. That is to say, an employee who was terminated before his employment contract ran out, having filed suit and obtained a judgment on that contract, no longer has a claim for unpaid wages. He now has a claim for what used to be called “debt on a judgment.” The employee can no longer sue for breach of contract, as he otherwise might prefer to do if that theory entitled him to more, because he no longer has a claim for breach of contract. All that remains is his entitlement to the debt owed him on the earlier judgment. The Anthony approach, which looks through the judgment to the underlying claim and then caps the amount of the judgment according to statutory criteria, is contrary to this common law doctrine.

On the other hand, merger is a common law principle, not a constitutional one. Subject to the Full Faith and Credit Act, discussed below, Congress has the power to promulgate bankruptcy law that supersedes what would otherwise be binding state law. The bankruptcy code’s one-year cap on claims for post-breach wage and salary, for example, balances the employee’s interest in recovering his damages against other creditors’ interests in their claims. There is also much to be said for aligning Ninth Circuit law with our sister circuits, removing the reward from forum shopping and providing law that is useful and predictably applied nationwide.

The distinction between the clear, mathematically calculable cap under section 502(b)(7), and the indefinite “reasonableness” cap under section 502(b)(4), might support distinguishing them, so that looking through a state court judgment might be appropriate under section 502(b)(7) but not under section 502(b)(4). We conclude, though, that this distinction should not make a difference. Reading section 502(b)(4) in its entirety, rather than limiting our analysis to the attorneys’ fees phrase, shows why.

The text of section 502(b)(4) limits to a “reasonable value” not only the services of attorneys but also the “services of an insider.” It imposes this limitation “if such claim is for services of an insider or attorney of the debtor.” Among others, “insiders” include family members, partners, and corporations the debtor controls. Collier suggests that Congress sought to protect creditors from “over-generosity,” an obvious hazard when the debtor is taking money from creditors to pay his own family members, partners, or corporation. And that concern arises even if the claim has been reduced to judgment, since the judgment might be a consent judgment or otherwise collusive. Such “over-generosity,” that is, intentionally paying more than a service is reasonably worth due to the close relationship between the debtor and creditor, may be likely with a relative, but it is not likely with the debtor’s attorney.

But, not least because they are in the same subsection, the reasonableness cap for payment of services to insiders should not be distinguished from the reasonableness cap' for payment of services to attorneys. There are other contexts where courts decide the reasonableness of attorneys’ fees due to, for example, the lack of an adversarial relationship or the risk of collusion, and such concerns may apply to an attorneys’ fee even if that fee has been reduced to a judgment. Among these contexts are cases where attorneys’ fees are shifted from the client to the losing party, as in English rule awards and class action settlements. We therefore adopt the Third and Tenth Circuit’s approaches to attorneys’ fees under section 502(b)(4). The bankruptcy code’s reasonableness cap limits a pre-petition obligation for a debt- or’s attorneys’ fees, even if such fees were allowable under state law, and even if such fees had been reduced to a state court judgment.

Finally, we must consider how the state court judgment in this case may have had preclusive impact on the “reasonableness” analysis under section 502(b)(4). The Full Faith and Credit Act applies in bankruptcy courts. The bankruptcy court in this case was thus required to give full faith and credit to the California Superior Court’s judgment entitling Spiller to his fees to the same extent that California res judicata law would give that judgment pre-clusive effect. The relevant question is not whether the California judgment for the amount of Spiller’s fees establishes that Siller was contractually obligated to pay him (it does), but whether Siller was precluded from arguing in the bankruptcy court that the amount of the California judgment exceeded the “reasonable value” of Spiller’s services. In other words, the question is not whether a state court judgment always has preclusive effect on the “reasonableness” analysis under section 502(b)(4) (we have already explained why it does not), the question is whether the state court judgment had preclusive effect on the reasonableness analysis in the particular circumstances of this case.

Issue preclusion, under California law, requires that (1) “the issue sought to be precluded from relitigation must be identical to that decided in a former proceeding,” (2) the “issue must have been actually litigated in the former proceeding,” (3) the issue “must have been necessarily decided in the former proceeding,” (4) “the decision in the former proceeding must be final and on the merits,” and (5) “the party against whom preclusion is sought must be the same as, or in privity with, the party to the former proceeding.” The party asserting preclusion bears the burden of establishing these elements. California courts have recognized exceptions to California’s preclusion law, but none applies here.

Using those standards, the district judge concluded that issue preclusion applied here, and so do we. The arbitration proceeding establishing the reasonableness of Spiller’s fee was fully contested and later confirmed by the judgment of a California court.

Siller argues that the arbitrator could not have decided section 502(b)(4) reasonableness because that is a bankruptcy standard and he had not yet filed for bankruptcy. We reject the notion that the word “reasonable” in section 502(b)(4) is a bankruptcy terra of art with a meaning different from its ordinary usage. There is no special definition of “reasonable” in the bankruptcy code. “Reasonable” under section 502(b)(4) means what it means in ordinary English. Determining a “reasonable” contingent fee cannot be reduced to a mechanical formulation. It calls upon the judgment of the tribunal in the particular circumstances of the matter before it. Some (but not all) of the factors bearing on reasonableness of a fee are whether the client had a fair opportunity to understand what he was agreeing to, the time the lawyer spent, the risk of not collecting, the need to advance time or expenses, the effect of representation on the lawyer’s ability to take on other cases, the ease or difficulty of working with a particular client, the legal market in the locality regarding hourly rates and contingent fee percentages, and—especially important here—whether the fee is contingent upon success.

Reasonableness, so understood, can be (and quite often is) decided in an arbitration proceeding like the one conducted here. Arbitration is a common method of resolving fee disputes under the bar rules of many states. Overall reasonableness is the usual criterion. And in the typical fee arbitration, the ex ante agreement between the lawyer and client is strong evidence of reasonableness, particularly where the fee was a contingent one.

Siller argues that even if reasonableness could have been determined in an arbitration proceeding, it was not determined in this one. The arbitrator decided that the parties’ fee agreement, not quantum meru-it, established the amount. But the arbitration consisted in large part of Siller presenting not just his arguments that the contract was unconscionable and Spiller failed to perform, but also his arguments and evidence to show that the Cotchett and Spiller firms’ fees were unreasonable. Siller devoted his presentation almost entirely to making this point. He cross-examined extensively and put on his own evidence, including an expert witness, to prove that the fees were so unreasonable as to be unconscionable, and to establish a lower number for a quantum meruit award if the arbitrator accepted his challenge to the fee agreements.

After hearing all of Siller’s arguments and evidence, the arbitrator concluded that Siller’s positions were wrong and the Cotchett and Spiller firms’ right. The quantum meruit analysis was irrelevant and Spiller’s fee was not unconscionable (a standard different from unreasonable). But that was not all the arbitrator concluded. He also concluded that Siller’s contract with the Spiller firm for an 8% contingent fee was “reasonable.” And he based his reasonableness determination on the work the parties anticipated, the work Spiller actually put into the case, the risks assumed by the parties, and Siller’s need for Spiller to finance his litigation.

Siller urges us to accept the bankruptcy court’s use of the lodestar method to determine the reasonableness of Spiller’s fee. But we explained in Venegas v. Skaggs that a continent fee may be reasonable where “it reflects] the risk of nonrecovery .,, assumed in accepting [a] case.” In this case, a lodestar fee would be unreasonable and could not, for that reason, serve as a cap under section 502(b)(4). No sensible attorney would undertake to represent a client at his usual hourly rate in these circumstances—where the client had been litigating for decades, had no money to pay, and had a history of declining to pay his lawyers and suing them for malpractice, the case was likely to take all or most of the lawyer’s time for the next several years, and the lawyer could get paid—if at all—only if he won. A “reasonable” fee must be reasonable for the lawyer as well as the client.

Had the arbitrator concluded that the amount of Spiller’s fee was unreasonable, but not so unreasonable as to make the contract’s formation or enforcement unconscionable, then the section 502(b)(4) reasonableness cap might have room to operate, because in that case the issue of 502(b)(4) reasonableness would not be identical to any of the issues arbitrated. But that is not how this case was arbitrated and decided. Under California law, one element of unconscionability is reasonableness, and a contingent fee contract can be rejected as unconscionable if it is sufficiently unreasonable in the circumstances. The arbitrator rejected that conclusion in this case. After hearing Siller’s arguments and extensive evidence to show unreasonableness, the arbitrator not only concluded that Spiller’s 8% contingent fee ought to be enforced, he specifically decided that it was “reasonable.”

So, although there might in some cases be room for a reduction, under section 502(b)(4)’s reasonableness cap, of a state court judgment confirming an arbitration award for a contingent fee, there is no room here. The performance in this case was difficult and demanding. And the relationship between the contracted-for amount and the service Spiller provided was not such as to make enforcement of the contract or payment of the fee unreasonable. The Full Faith and Credit Act requires, in the circumstances of this case, that the judgment of the state court confirming Spiller’s arbitration award for his fee be given full faith and credit in Siller’s bankruptcy proceeding.

The judgment of the district court is AFFIRMED.

. The firm would later be renamed Cotchett, Pitre & McCarthy. . 11 U.S.C. § 502 . . 28 U.S.C. § 1738 . . 11 U.S.C. § 502 (b)(4). . 28 U.S.C. § 1738 . . 922 F.2d 592 (10th Cir. 1990), modified sub nom. Abel v. West, 932 F.2d 898 (10th Cir. 1991). . See, e.g., In re Placide, 459 B.R. 64 (B.A.P. 9th Cir. 2011); In re Solar Trust of America, LLC, No. 12-11136, 2015 WL 1011548 (Bankr. D. Del. Jan. 12, 2015); In re Boulder Crossroads, LLC, No. 09-10381, 2010 WL 4924745 (Bankr. W.D. Tex. Dec. 1, 2010); In re Heritage Organization, L.L.C., No. 04-35574, 2006 WL 6508182 (Bankr. N.D. Tex. Jan. 6, 2006); In re Russell Cave Co., 253 B.R. 815 (Bankr. E.D. Ky. 2000); In re Nelson, 206 B.R. 869 (Bankr. N.D. Ohio 1997). . Western Real Estate, 922 F.2d at 594. . Id. (citing Yermakov, 718 F.2d 1465 , 1470 (9th Cir. 1983)). . Id. (citing Pacific Far East Line, Inc., 654 F.2d 664, 668-70 (9th Cir. 1981)). . See id. at 595-97. . See id. at 597. . See id. at 597-98. . Id. at 597. . 867 F.2d 527 (9th Cir. 1989), aff’d sub nom. Venegas v. Mitchell, 495 U.S. 82 , 110 S.Ct. 1679 , 109 L.Ed.2d 74 (1990). . Id. at 532 (citing Hamner v. Rios, 769 F.2d 1404, 1409 (9th Cir. 1985)). . 11 U.S.C. § 502 (a). . 96 F.3d 692 (3d Cir. 1996). . Id. at 693 . . 11 U.S.C. § 502 (b)(7). . Anthony, 96 F.3d at 693 . . Id. . Bankruptcy courts have gone both ways on this issue. Compare, e.g., In re Vic Snyder, Inc., 23 B.R. 185, 186-87 (Bankr. E.D. Pa. 1982) with In re Networks Elec. Corp., 195 B.R. 92, 99-100 (B.A.P. 9th Cir. 1996). . Anthony, 96 F.3d at 697 . . See id. at 695-97 . Anthony cited with approval our circuit’s Bankruptcy Appellate Panel decision in Networks Elec. Corp., 195 B.R. at 92 . . 28 U.S.C. § 1738 . . See Networks Elec. Corp., 195 B.R. at 100 . . 11 U.S.C. § 502 (b)(4). . Defined at 11 U.S.C. § 101 (31), the term “insider” includes— (A) if the debtor is an individual— (i) relative of the debtor or of a general partner of the debtor; (ii) partnership in which the debtor is a general partner; (iii) genéral partner of the debtor; or (iv) corporation of which the debtor is a director, officer, or person in control; (B) if the debtor is a corporation— (i) director of the debtor; (ii) officer of the debtor; ' (iii) person in control of the debtor; (iv) partnership in which the debtor is a general partner; (v) general partner of the debtor; or (vi) relative of a general partner, director, officer, or person in control of the debtor; (C) if the debtor is a partnership— (i) general partner in the debtor; (ii) relative of a general partner in, general partner of, or person in control of the debtor; (iii) partnership in which the debtor is a general partner; (iv) general partner of the debtor; or (v) person in control of the debtor; (D) if the debtor is a municipality, elected official of the debtor or relative of an elected official of the debtor; (E) affiliate, or insider of an affiliate as if such affiliate were the debtor; and (F) managing agent of the debtor. . 4 Collier on Bankruptcy ¶ 502.03[5][c] (Alan N. Resnick & Henry J. Sommer eds., 16th ed. 2012). . See, e.g., 42 U.S.C. § 1988 (b) (permitting the award of fees to the prevailing party under the English rule in many actions brought under federal law); Alaska R. Civ. P. 82 (requiring generally that the losing party pay the prevailing party's fees); Ariz. Rev. Stat. § 12-341.01 (permitting the award of fees to the prevailing party under the English rule in contract cases); APL Co. Pte. v. UK Aerosols Ltd., 582 F.3d 947, 957 (9th Cir. 2009) (applying the law of Singapore and noting that it follows the English rule for attorneys' fees). . 28 U.S.C. § 1738 ; In re Nourbakhsh, 67 F.3d 798 , 800 (9th Cir. 1995). . Nourbakhsh, 67 F.3d at 800. . In the past, California courts have referred to issue preclusion using the older terms "collateral estoppel” or "estoppel by judgment” or the broader term "res judicata." They now use the modem term “issue preclusion.” For an explanation of the use of these terms in California courts, see Lucido v. Superior Court, 51 Cal.3d 335 , 272 Cal.Rptr. 767 , 795 P.2d 1223 , 1225 n.3 (Cal. 1990) and Olson v. Cory, 134 Cal.App.3d 85 , 103 n.9, 184 Cal.Rptr. 325 (1982). . Lucido, 51 Cal.3d 335 , 272 Cal.Rptr. 767 , 795 P.2d at 1225 ; see also In re Harmon, 250 F.3d 1240, 1245 (9th Cir. 2001). . Lucido, 272 Cal.Rptr. 767 , 795 P.2d at 1225 ; Harmon, 250 F.3d at 1245 . . Such preclusion exceptions include public policy, unforeseeability, and the inability or lack of incentive to litigate the prior adjudication. See Olson, 134 Cal.App.3d at 103 n.9, 184 Cal.Rptr. 325 . . 867 F.2d at 534. . See Ketchum v. Moses, 24 Cal.4th 1122 , 104 Cal.Rptr.2d 377 , 17 P.3d 735, 742 (2001). . See Carlson v. Home Team Pest Def., Inc., 239 Cal.App.4th 619 , 191 Cal.Rptr.3d 29, 40 (2015).