concurring and dissenting:
I join Parts I, 11(A), 11(B)(1), and 11(D) of the majority’s opinion, concur in the opening section of Part 11(B), concur in part and dissent in part from Part 11(B)(2), and dissent from Part 11(C).
*1173 Enforceability of releases
In the opening section of Part 11(B) of its opinion, the majority concludes that “the language in the releases” referring to “ ‘any and all claims [Ritchie] might have arising under federal, state or local law,’ ” is broad enough to preclude her from filing this qui tam action. Maj. Op. at 1167. More specifically, the majority concludes, citing both the language of the FCA and Vermont Agency of Natural Resources v. United States ex rel. Stevens, 529 U.S. 765, 773, 120 S.Ct. 1858, 146 L.Ed.2d 836 (2000), that “[t]he portion of the claim assigned to the relator, namely the amount the relator is entitled to recover in a successful action, belongs to the relator to a sufficient degree that it should be considered under the releases signed in this case as a claim Ritchie ‘might have under federal ... law.’ ” Maj. Op. at 1167.
I respectfully disagree with this conclusion. To be sure, the Supreme Court concluded in Vermont Agency, in the context of analyzing a qui tam relator’s standing to maintain a suit under the FCA, that “[t]he FCA can reasonably be regarded as effecting a partial assignment” from the government to the qui tam relator. 529 U.S. at 773, 120 S.Ct. 1858. Importantly, however, the Court emphasized that this partial assignment was only “of the Government’s damages claim.” Id. (emphasis added). Moreover, the Court noted that “[a] qui tam relator has suffered no ... invasion” of a “legally protected right,” id. at 772-773, 120 S.Ct. 1858, and emphasized that “the ‘right’ he seeks to vindicate does not even fully materialize until the litigation is completed and the relator prevails,” id. at 773, 120 S.Ct. 1858. Further, the Court characterized “the relator’s suit” as one simply for “bounty,” id., and distinguished between “a bounty and an express cause of action,” id. at 777, 120 S.Ct. 1858.
Interpreting the Court’s language in Vermont Agency, the Second Circuit recently concluded, and I agree, that “while the False Claims Act permits relators to control the False Claims Act litigation, the claim itself belongs to the United States.” United States ex rel. Mergent Serv. v. Flaherty, 540 F.3d 89, 93 (2d Cir.2008) (citing Vermont Agency of Natural Res. v. United States ex rel. Stevens, 529 U.S. 765, 774-75, 120 S.Ct. 1858, 146 L.Ed.2d 836 (2000)). Applying that principle to the facts presented here, the FCA claims ultimately asserted by Ritchie belonged to the United States, not Ritchie. Thus, it is far from certain in my view that the release language quoted by the majority encompassed those FCA claims.
Having said this, I note that Paragraph 11 of each release provided:
Employee understands that if this Agreement were not signed, Employee would have the right to voluntarily assist other individuals or entities in bringing claims against Employer. Employee hereby waives that right and will not provide any such assistance other than assistance in an investigation or proceeding conducted by the United States Equal Employment Opportunity Commission. Employee and Employer further agree that Employee may provide information pursuant to any valid subpoena, provided that Employee first notifies Employer in writing of the proposed disclosure in order to provide Employer an opportunity to seek a Protective Order.
In my view, this language was sufficiently broad to have encompassed Ritchie’s right to serve as a relator in a qui tam action.
Does federal policy preclude the enforcement of these releases?
I agree with the majority’s conclusions, in Part 11(B)(2), that “there is a federal interest in protecting the federal government’s right to recover under the FCA,” *1174Maj. Op. at 1169, that “a uniform federal rule is appropriate to protect the federal interest at stake,” id. at 1169, and that, accordingly, we must “apply federal common law to determine the enforceability of the releases,” id. As to the federal common law standard to be applied in this case, I also agree with the majority that the Ninth Circuit’s analysis in United States ex rel. Green v. Northrop Corp., 59 F.3d 953 (9th Cir.1995), is thorough and well-reasoned and that we should adopt it as our own.
I disagree with the majority’s decision, however, to adopt the Ninth Circuit’s analysis in United States ex rel. Hall v. Teledyne Wah Chang Albany, 104 F.3d 230 (9th Cir.1997). Several related public interests are undoubtedly at stake under the FCA’s enforcement scheme. These include (1) incentivizing insiders to come forward with information about fraud against the Government, (2) supplementing Government efforts to recoup monies lost due to fraud, and (3) deterring fraud against the Government. See, e.g., False Claims Amendment Act of 1986, S.Rep. No. 99-345, at 7 (1986), reprinted in 1986 U.S.C.C.A.N. 5266, 5272 (“In addition to detection, investigative and litigative problems which permit fraud to go unaddressed, perhaps the most serious problem plaguing effective enforcement is a lack of resources on the part of Federal enforcement agencies.”); id. at 8 (“[a]n additional problem ... exists when large, profitable corporations are the subject of a fraud investigation and able to devote many times the manpower and resources available to the Government.”); id. (“The Committee believes that the amendments ... which allow and encourage assistance from the private citizenry can make a significant impact on bolstering the Government’s fraud enforcement effort.”); id. at 23-24 (amending the FCA’s qui tarn provision in order “to encourage more private enforcement suits”). The Ninth Circuit’s analysis in Hall focuses primarily on the first interest: providing insiders with incentives to come forward with information about fraud against the Government. See Hall, 104 F.3d at 233. Most of the cases applying the Green/Hall framework have likewise focused on these incentives. See, e.g., United States ex rel. Longhi v. Lithium Power Tech., Inc., 481 F.Supp.2d 815, 818-21 (S.D.Tex.2007).
The problem with the Hall exception is that it undermines one of the other public interests under the FCA: the Government’s compensatory interest in recouping funds lost due to fraud. Presumably, the Government has two potential reasons for declining to pursue a civil fraud claim independently when it knows of the allegations: (1) the suit lacks merit, or (2) the Government lacks the resources to be an active party to the litigation. The FCA is designed to help remedy the latter scenario:
[PJerhaps the most serious problem plaguing effective enforcement is a lack of resources on the part of Federal enforcement agencies. Unlike most other types of crimes or abuses, fraud against the Federal Government can be policed by only one body — the Federal Government. State and local law enforcement are normally without jurisdiction where Federal funds are involved.
Taking into consideration the vast amounts of Federal dollars devoted to various complex and highly regulated assistance and procurement programs, Federal auditors, investigators, and attorneys are forced to make ‘screening’ decisions based on resource factors. Allegations that perhaps could develop into *1175very significant cases are often left unaddressed at the outset due to a judgment that devoting scarce resources to a questionable case may not be efficient. And with current budgetary constraints, it is unlikely that the Government’s corps of individuals assigned to anti-fraud enforcement will substantially increase.
S.Rep. No. 99-345, at 7; see also Ridenour v. Kaiser-Hill Co., L.L.C., 397 F.3d 925, 931 n. 9 (10th Cir.2005) (“The increased incentives to private individuals have proven effective at helping the Government recover substantial sums of fraudulently obtained payments.”); Green, 59 F.3d at 968 (“The relator’s right to recovery exists solely as a mechanism for deterring fraud and returning funds to the federal treasury.”); cf. United States v. Health Possibilities, 207 F.3d 335, 343 n. 6 (6th Cir.2000) (“There is absolutely no statutory authority for the proposition that simply because the government decides not to expend the resources to proceed with an action itself, it thereby authorizes the relator to settle the government’s claims in whatever manner he wishes. Indeed, such a construction would only force the government to unnecessarily intervene in qui tam cases and thereby frustrate the efficacy of the qui tam framework.” (citations omitted)); United States ex rel. DeCarlo v. Kiewit/AFC Enter., Inc., 937 F.Supp. 1039, 1047 (S.D.N.Y.1996) (“[T]he government’s decision not to participate in the qui tam action is not equivalent to a consent to voluntarily dismiss a claim against a defendant with prejudice. Non-intervention does not necessarily signal governmental disinterest in an action, as it is entitled to most of the proceeds even if it opts not to intervene.”). By allowing contractors to buy off relators whose resources would otherwise be available to pursue fraud claims on behalf of the Government, the exception from Hall disregards the Government’s compensatory interest in recouping funds lost due to fraud.
As I see it, a better rule would be that prefiling releases are unenforceable under the FCA unless the Government — and more specifically, the Attorney General— has provided its express written consent.1 An express consent rule is consistent with the FCA’s post-filing settlement requirements. See 31 U.S.C. § 3730(b)(1); Ridenour, 397 F.3d at 931 n. 8. Such a rule also accounts for the Government’s compensatory interest in recouping funds lost due to fraud. If the Government wants the relator to have the opportunity to bring suit on the Government’s behalf to recoup the lost funds, then the Government can simply withhold its consent.
An express consent rule also decreases the risk that the relator and contractor will bargain away the Government’s claim. See Health Possibilities, 207 F.3d at 340-41 (concluding that “the power to veto a privately negotiated settlement of public claims is a critical aspect of the government’s ability to protect the public interest in qui tam litigation,” and that without such power “the public interest would be largely beholden to the private relator”); Searcy v. Philips Elec. N. Am. Corp., 117 F.3d 154, 160 (5th Cir.1997) (noting that the express consent provision of § 3730(b)(1) allows the government to reduce the risk that “relators [will] manipulate settlements in ways that unfairly enrich them and reduce benefits to the *1176government”); Green, 59 F.3d at 966 (noting that “a rational relator” has a financial incentive “to accept a substantially smaller amount to settle the claim immediately than to preserve the right to eventually file a qui tarn action in which the government would retain the lion’s share of the proceeds”). Admittedly, even the Hall exception provides some protection from parties bargaining away the Government’s claims, because the Government must know about a claim for Hall to apply — and the Government can always bring the claim later, if it so desires. See Hall, 104 F.3d at 233. Nevertheless, there is still a risk under Hall that the Government will not bring a meritorious claim simply because the Government lacks the resources to do so. The parties will likely know this, and the contractor will pay the relator more money to reflect the probability that resource constraints will preclude the Government from bringing the claim independently. An express consent rule helps prevent this problem by allowing the Government to veto the release of relators who would otherwise pursue meritorious FCA claims with their own resources.
Moreover, if we require the Government’s express consent, then contractors will be forced to disclose enough information to satisfy the Government. This will help the Government investigate the alleged fraud, and it will ensure that the Government is satisfied with its investigation before a relator can be released. Along these same lines, an express consent rule eliminates a very difficult task for the courts: attempting to determine, based on the conflicting evidence and testimony of two adverse parties (the contractor and the relator), whether a third party (the Government) is satisfied with its knowledge and investigation of the alleged fraud. In the case before us, as in any case which involves a representation that a party will “tell all,” it is very difficult to determine whether all relevant information was in fact disclosed by Lockheed and whether, as a consequence, the extent of the alleged fraud was fully investigated by the Government. From my distant vantage point in this case, the exchange of information and the extent of the Government’s investigation appears superficial.
I fully recognize that requiring express consent will likely deter more settlements than under the Hall framework. If the Government withholds its express consent, then the parties will have to value the settlement to include the probability of the relator filing a subsequent qui tam action. If that probability is high, then the value of settling, from the contractor’s perspective, will be correspondingly low. This may well result in fewer settlements, which will impinge on the parties’ interest in resolving disputes. As the Ninth Circuit has pointed out, though, the settlements at most risk of falling apart in this context are those in which the contractor is most likely to have committed fraud. Green, 59 F.3d at 969. This, in my view, is not necessarily a bad result.
I also recognize, relatedly, that an express consent requirement may make it more difficult for parties to craft an agreeable prefiling release. Most likely, the Government will not be bound by the pre-filing release between the relator and the contractor, so the Government will still be able to bring its own claim for the alleged fraud. See Hall, 104 F.3d at 233 (“The government, of course, was not a party to the release, and is therefore not barred by it from pursuing a claim against Tele-dyne.”). This is one way in which prefiling consent will differ from post-filing consent: in the post-filing settlement context, the FCA claim is usually dismissed with prejudice, and the Government is barred from pursuing it further. The contractor has several potential solutions to this problem, *1177however. Most obviously, the contractor can attempt to include the Government in the prefiling settlement and release. This will actually further the FCA’s purpose, because the Government will only agree to be a party to the release if it has adequately investigated the alleged fraud and if the settlement provides the Government with adequate compensation. Also, in situations where the Government has provided its express consent but is not a party to the release, the contractor can discount the value of the settlement by the risk that the Government will bring its own claim for the same fraud. In short, an express consent rule protects and furthers the interests at stake under the FCA, while at the same time providing the parties with some flexibility to account for the circumstances of each individual case.
Because I favor adoption of an express consent rule, I must in turn conclude, since the Government never consented to Rit-chie’s release, that we must reverse the district court’s grant of summary judgment in favor of Lockheed. That is, I would conclude that the release does not bar Ritchie’s subsequent qui tam action under the FCA.
Costs
Because I would reverse the district court’s grant of summary judgment in favor of Lockheed, I would also, necessarily, reverse the district court’s award of costs to Lockheed pursuant to Rule 54(d).
. The Green court actually hinted at a consent requirement as a potential prerequisite for enforcing a prefiling release. Cf. Green, 59 F.3d at 965 ("[P]ermitting a prefiling release when the government has neither been informed of, nor consented to, the release would ... frustrate one of the central objectives of the Act.” (emphasis added)). Green did not explicitly require such consent, though, and the Hall court chose to import a broader exception.