445 F.3d 1147
UNITED STATES of America, Plaintiff, and
Patricia Haight, ex rel. and in Defense of Animals, Plaintiff-Appellant,
v.
CATHOLIC HEALTHCARE WEST; Catholic Healthcare West Arizona; Michael Berens, Defendants-Appellees.
No. 03-16937.
United States Court of Appeals, Ninth Circuit.
Argued and Submitted September 13, 2005.
Filed April 19, 2006.
Jeremy L. Friedman, Law Office of Jeremy L. Friedman, Oakland, CA, for the appellants.
Lawrence A. Kasten, Lewis and Roca LLP, Phoenix, AZ, for the appellees.
Appeal from the United States District Court for the District of Arizona; Frederick J. Martone, District Judge, Presiding. D.C. No. CV-01-02253-FJM.
Before: B. FLETCHER, JOHN R. GIBSON,* and BERZON, Circuit Judges.
B. FLETCHER, Circuit Judge:
Dr. Patricia Haight and the organization In Defense of Animals (collectively "Relators") appeal the dismissal of their complaint for lack of subject matter jurisdiction under the False Claims Act's "public disclosure" jurisdictional bar, 31 U.S.C. § 3730(e)(4). We reverse.
I.
Relators brought this qui tam action under the False Claims Act ("FCA"), 31 U.S.C. §§ 3729-33, against Catholic Healthcare West and Dr. Michael Berens (collectively "Defendants"). The Relators allege that Berens submitted a fraudulent grant application to the National Institutes of Health ("NIH"). Based on this application, he was awarded over $700,000 in grant money.
Defendant Berens is a research scientist at Barrow Neurological Institute, a division of Catholic Healthcare West. Berens is the primary researcher on a project that uses beagle dogs to research glioma, a form of malignant brain tumor. The study is designed to develop a large animal model for studying glioma and then extrapolate any knowledge gained to treat glioma in humans. Berens attempted to develop the model by injecting glioma tumor cells into the flanks of gestating beagle puppies. This procedure had to be done during gestation to prevent the immune systems of the puppies from rejecting the gliomal cells and was designed to cause tumors to develop after the puppies were born, which could then be transplanted into their brains. Initially, Berens' research was privately funded. Later he sought public funding. Berens' first grant application to the NIH was rejected, but his second application was approved, resulting in an award of over $700,000 in grant money.
Relator Haight is an experimental psychologist and the Southwest Regional Director of In Defense of Animals. Haight performed extensive research on Berens' study and uncovered the fraud alleged in the complaint. Haight began investigating Berens' research at the request of a student and community members who were protesting animal research conducted at Barrow and Arizona State University ("ASU"). She first identified the funding source for Berens' study — the National Institute of Neurological Disorders and Strokes, an institute within the larger NIH — and requested documents relating to the study pursuant to the Freedom of Information Act ("FOIA"). She was subsequently informed that she could obtain the documents from her FOIA request directly from Barrow, where Berens' study was being performed, so she walked down personally to obtain them. The documents she obtained because of her FOIA request included the 1997 grant application and continuation forms, and a paper and paper abstract by Berens.
Haight then requested documents from ASU, where the project had formerly been housed, including animal care records and minutes from the Institutional Animal Care and Use Committee ("IACUC") meetings, where animal research projects are vetted. In response to this request, Haight received protocols submitted to ASU by Berens starting in 1990, animal care records, IACUC minutes, and correspondence between Berens and IACUC concerning Berens' withdrawal of the project from ASU in 1997.
Becoming concerned about the efficacy of the project, Haight arranged to tour the ASU facility where the dogs had been housed and the Veterans Administration hospital to which the dogs were transferred when Berens' project ceased to have the support of ASU. She also interviewed numerous individuals who provided information about Berens' research, including Edward Castaneda, Chairman of the IACUC at ASU; Ted Brandon, Director of ASU's Animal Care Program; Neal DeNardo, ASU Veterinarian; Ronald Barr, Assistant Vice Provost for Research at ASU and IACUC member; Tol Chesko, IACUC member; Joan McGregor, ASU faculty member; and graduate student Clare Rhodes. In conjunction with her tour of the VA facility, Haight interviewed Jedd Nelson, Director of Animal Care. Haight contacted Geoffrey Pilkington, Professor of Experimental Neuro-oncology at King's College of London, who was listed as a consultant to Berens' project on the NIH application. Haight also met with Joan Rankin Shapiro, Vice President for Research at Barrow Neurological Institute.
In the course of these discussions and tours, Haight learned what she believes to be the truth about Berens' research. First, she learned from DeNardo and Brandon that Berens' research has produced little, if any, scientific results. Many of the pups died, were aborted, or were born with congenital defects, and those that were born healthy did not develop the expected subcutaneous tumors. This assessment of the project was supported by Shapiro, who told Haight that throughout the ten years that the project had been ongoing, only three pups had developed tumors, and one of those tumors was so small that it could not be seen until after the pup was killed. Second, she learned that Berens' competency to perform the complicated surgery on the prenatal pups was questionable and that the researchers could not identify gross anatomical landmarks when attempting to inject the pups' flanks with gliomal cells. Shapiro stated that because the uterus of the beagle is murky, it became necessary to "inject everything we think we can see" when attempting to implant the cells in the pups' flanks. Third, she learned that while Berens had officially withdrawn his project from ASU, he had done so under pressure from the university's IACUC to terminate the project because of its failure to produce meaningful results after ten years of attempts. Fourth, she learned that none of the dogs temporarily housed at the VA Hospital after the project was terminated at ASU had developed subcutaneous tumors. Fifth, she discovered from Dr. Pilkington that he was surprised to have been listed as a consultant for the project and that he had not, as stated in the grant application, agreed to provide any dogs for the project.
The Relators challenge an array of allegedly fraudulent or misleading statements. The Relators' allegations are as follows:
1. Relators challenge the indications of success stated in the grant application, specifically that:
a. "successful allotolerant pups had been achieved in four separate litters";
b. "4 out of 5 pups in one litter developed tumors";
c. "one in four litters yielded at least one pup with a subcutaneous tumor";
d. "there were subcutaneous tumors in up to 80% of the pups."
2. Relators challenge the assertions in the grant application that the researchers have "achieved the technical and surgical skills necessary for realizing the desired outcome," that they can identify gross anatomical landmarks, and that the cells were embedded in the desired locations.
3. Relators challenge the projections in the grant application as to the number of tumors the researchers would create.
4. Relators challenge the statement in the grant application that "neonatal pups cannot be induced to accept allogenic cells," thus necessitating the approach taken by Berens.
5. Relators challenge the assertions in the grant application as to Pilkington's involvement as a consultant and provider of control dogs (i.e. dogs with naturally occurring brain tumors).
With respect to each allegation, the false or misleading statement was contained in the grant application obtained via Haight's FOIA request.
Before Relators filed their complaint, several news articles were published on the Berens' controversial animal experiments. See, e.g., Scott Bewick, Researcher's Beagle Testing Unnecessary, Protesters Say, ASU STATE PRESS, Sept. 28, 2000, available in part at http://studentmedia.vpsa.asu.edu/ webarchives/; James Hibberd, Screwing the Pooch, PHOENIX NEW TIMES, Jan. 4, 2001, at 14-24, available at http:// phoenixnewtimes.com/Issues/2001-01-04/news/feature.html. In addition, In Defense of Animals issued several press and news releases relating to the Berens study. Relators argue that what they allege to be the truth about Berens' study was discovered through Haight's independent investigation.
Haight and In Defense of Animals filed their complaint on behalf of the United States under the qui tam provisions of the FCA in the Northern District of California. The United States declined to intervene in the action. Based on Defendants' motion, the suit was transferred to Arizona.
After transfer, the Relators filed an amended complaint, and the Defendants filed a motion to dismiss for failure to state a claim. After the motion to dismiss was denied, the Defendants filed an Answer, and the parties initiated discovery. Then the Defendants filed a second motion to dismiss, now based on lack of subject matter jurisdiction under 31 U.S.C. § 3730(e)(4). After argument the district court granted the Defendants' motion, holding that a response to a FOIA request is a public disclosure via an enumerated source and that accordingly Relators' suit was barred under 31 U.S.C. § 3730(e)(4). The district court also awarded costs to the Defendants. The Relators appeal both the district court's order granting the motion to dismiss and the order awarding costs.
We review de novo the district court's dismissal for lack of subject matter jurisdiction. United States ex rel. Hochman v. Nackman, 145 F.3d 1069, 1072 (9th Cir. 1998).
II.
The False Claims Act imposes liability on those who defraud the government. 31 U.S.C. § 3729. It encourages the uncovering of such fraud by permitting private persons to bring qui tam actions on behalf of the government. Id. § 3730(b); see also United States ex rel. Campbell v. Redding Med. Ctr., 421 F.3d 817, 823 (9th Cir.2005) ("`[T]he Committee's overall intent in amending the qui tam section of the False Claims Act is to encourage more private enforcement suits.'" (quoting Sen. Rep. No. 99-345, at 23-24 (1986))). Such qui tam relators then share in any recovery obtained on the government's behalf. 31 U.S.C. § 3730(d). At the same time, the FCA discourages opportunistic qui tam relators by depriving the courts of subject matter jurisdiction in actions where the fraud allegations were publicly disclosed via a source listed in the provision, unless the relator was the original source of the allegations. The jurisdictional bar provision reads:
No court shall have jurisdiction over an action under this section based upon the public disclosure of allegations or transactions in a criminal, civil, or administrative hearing, in a congressional, administrative, or Government [General] Accounting Office report, hearing, audit, or investigation, or from the news media, unless the action is brought by the Attorney General or the person bringing the action is an original source of the information.
Id. § 3730(e)(4)(A).
Whether an action is barred under this provision is a two-step inquiry. First, it must be determined whether the allegations were publicly disclosed via a source enumerated in § 3730(e)(4)(A). If so, then it must be determined whether the relator was the original source of those allegations. United States ex rel. Wang v. FMC Corp., 975 F.2d 1412, 1416 (9th Cir.1992) ("Where there has been no `public disclosure' within the meaning of section 3730(e)(4)(A), there is no need for a qui tam plaintiff to show that he is the `original source' of the information."). "Original source" is defined in § 3730(e)(4)(B) as "an individual who has direct and independent knowledge of the information on which the allegations are based and has voluntarily provided the information to the Government." 31 U.S.C. § 3730(e)(4)(B). Because we conclude that the allegations underlying the complaint were not publicly disclosed within the meaning of § 3730(e)(4)(A), we need not proceed to the second step.
A.
Determining whether the allegations underlying a fraud claim have been publicly disclosed under § 3730(e)(4)(A) itself requires two inquiries. First, to constitute a public disclosure, the fraud must have been disclosed in one or more1 of the sources specified under the statute: 1) a criminal, civil, or administrative hearing;
2) a congressional, administrative, or [General] Accounting Office report, hearing, audit or investigation; or
3) the news media.
Id. § 3730(e)(4)(A). If there has been such a disclosure, the content of that disclosure must consist of the "allegations or transactions" giving rise to Relators' claim. A-1 Ambulance Serv., 202 F.3d at 1243. "The substance of the disclosure . . . need not contain an explicit `allegation' of fraud, so long as the material elements of the allegedly fraudulent `transaction' are disclosed in the public domain." Foundation Aiding the Elderly, 265 F.3d at 1014.
If X + Y = Z, Z represents the allegation of fraud and X and Y represent its essential elements. In order to disclose the fraudulent transaction publicly, the combination of X and Y must be revealed, from which readers or listeners may infer Z, i.e., the conclusion that fraud has been committed.
Id. at 1015 (quoting United States ex rel. Springfield Terminal Ry. Co. v. Quinn, 14 F.3d 645, 654 (D.C.Cir.1994)). In a fraud case, X and Y stand for "a misrepresented state of facts and a true state of facts." Id. at 1015 (internal quotation marks and citation omitted). Thus, for the suit to be barred, Defendants must show that its essential elements, both the alleged truth and the allegedly fraudulent statements, were publicly disclosed via an enumerated source.
The district court determined that the allegations underlying the fraud claim were publicly disclosed because the alleged fraud occurred in a grant application obtained by Relators pursuant to a FOIA request, and Relators' claims were disclosed in press releases and various news articles. We conclude that the district court was incorrect in holding that the response to the FOIA request was a public disclosure via an enumerated source within the meaning of § 3730(e)(4)(A). Because the misrepresented state of facts, "X," was not publicly disclosed via an enumerated source, we need not reach the questions of whether the true state of facts, "Y," was publicly disclosed via an enumerated source or whether Relators were the "original source" of that true state of facts.
B.
Whether a response to a FOIA request is a public disclosure in a form prescribed by the jurisdictional provision is an open question in the Ninth Circuit.2 Appellees contend that we should hold that all documents obtained via FOIA request are public disclosures for the purposes of the jurisdictional statute. In support, they cite United States ex rel. Mistick PBT v. Housing Authority, 186 F.3d 376, 383 (3d Cir.1999), in which the Third Circuit concluded that a response to a FOIA request is a publicly disclosed "administrative report" or "administrative investigation" so as to invoke the jurisdictional bar. See also United States ex rel. Reagan v. E. Tex. Med. Ctr. Reg'l Healthcare Sys., 384 F.3d 168, 176 (5th Cir.2004); United States ex rel. Burns v. A.D. Roe Co., 186 F.3d 717, 723-24 (6th Cir.1999). We disagree: a response to a FOIA request is not necessarily a report or investigation, although it can be, if it is from one of the sources enumerated in the statute. Since the document obtained via FOIA request here was not from one of the sources enumerated in the statute, the FOIA response does not trigger the jurisdictional bar.
In interpreting the jurisdictional bar, we "look first to the plain language of the statute, construing the provisions of the entire law, including its object and policy, to ascertain the intent of Congress." United States ex rel. Lujan v. Hughes Aircraft Co., 243 F.3d 1181, 1187 (9th Cir. 2001) (quoting United States v. Hockings, 129 F.3d 1069, 1071 (9th Cir.1997)). If the statutory text is ambiguous, we look to the legislative history. Id.
Interpreting "report" or "investigation" as listed in the jurisdictional bar to include any document obtained in response to a FOIA request would stretch the meaning of those terms too broadly. FOIA requires only that federal agencies search their records for those that are responsive to the request. 5 U.S.C. § 552(a)(3)(D) ("`[S]earch' means to review, manually or by automated means, agency records for the purpose of locating those records which are responsive to a request."). In essence, a FOIA request is a "mechanism for duplicating records that are in the possession of the federal government and that are not otherwise excludable from members of the public." Mistick, 186 F.3d at 393 (Becker, C.J., dissenting). In contrast, reports and investigations generally involve independent work product. "Report" denotes a document that includes an analysis of findings; "investigation" implies independent governmental leg-work. Moreover, the FCA's jurisdictional bar groups "report" and "investigation" with a series of other enumerated sources that each involve extensive governmental work product and involvement.3 Because responding to a FOIA request requires little more than duplication, labeling any response to a FOIA request a "report" or "investigation" would ignore the way in which each of the enumerated sources involves governmental work product.
The legislative history of the jurisdictional provisions of the False Claims Act lends support to the interpretation that a response to a FOIA request is not, standing alone, an administrative report or investigation. See United States ex rel. Springfield Terminal Ry. Co. v. Quinn, 14 F.3d 645, 649-51 (D.C.Cir.1994) (providing a more detailed history); United States ex rel. Stinson, Lyons, Gerlin & Bustamante, P.A. v. Prudential Ins. Co., 944 F.2d 1149, 1152-54 (3d Cir.1991), (same). The original version of the False Claims act, enacted in 1863, permitted qui tam suits to be filed regardless of how the relator came by the information that underlay the fraud claim, even if the government was already aware of that information. In 1943, the Supreme Court interpreted this broad statutory language to permit qui tam suits even when the allegations in the complaint were copied by the relator directly from criminal indictments in public court files, clearly within the government's cognizance. See United States ex rel. Marcus v. Hess, 317 U.S. 537, 546-548, 63 S. Ct. 379, 87 L. Ed. 443 (1943). Congress responded quickly to this interpretation of the statute by swinging far in the other direction. The 1943 amendments to the FCA prohibited qui tam suits "based on evidence or information the Government had when the action was brought," 31 U.S.C. § 3730(b)(4) (1982) (superceded), even when that information was obtained by the relator through substantial independent effort and provided to the government in anticipation of the suit. See, e.g., United States ex rel. Wisconsin v. Dean, 729 F.2d 1100 (7th Cir.1984) (barring jurisdiction over a qui tam suit filed by the state of Wisconsin alleging Medicaid fraud the state had uncovered and reported to the federal government as required by law).
In 1986, Congress passed further amendments, acknowledging that less restrictive jurisdictional provisions were required to fight the ever-growing problem of fraud. The current language, then, was designed to create a scheme in which "the qui tam suit provision[s] operate somewhere between the almost unrestrained permissiveness represented by the Marcus decision, and the restrictiveness of the post-1943 cases, which precluded suit even by the original sources." Stinson, 944 F.2d at 1154 (citation omitted); see also United States ex rel. Devlin v. California, 84 F.3d 358, 362 (9th Cir.1996) (explaining that the 1986 amendments to the FCA represent "the golden mean between adequate incentives for whistle-blowing insiders with genuinely valuable information and discouragement of opportunistic plaintiffs who have no significant information to contribute of their own" (internal quotation marks and citation omitted)). The current provisions seek to balance the primary fraud-detection function of the FCA while "ensur[ing] that no qui tam relator could profit from information that had become part of the public domain." Mistick, 186 F.3d at 391 (Becker, C.J., dissenting). Thus, the jurisdictional bar provisions "must be analyzed in the context of these twin goals of rejecting suits which the government is capable of pursuing itself, while promoting those which the government is not equipped to bring on its own." Springfield Terminal, 14 F.3d at 651.
Interpreting the jurisdictional provision to bar qui tam suits where the allegations are based on otherwise private information obtained via a FOIA request would be out of step with Congress's intentions in amending the jurisdictional provision of the FCA. In order to prevent opportunistic qui tam relators from profiting from information that was already part of the public domain, Congress sought to bar suits in which the government could already be expected to be on notice of the fraud. It determined that the government could be expected to be aware of information derived from the enumerated sources, because the information originated from the government or involved governmental work product.4 By limiting the enumerated sources to that narrow list, however, Congress also sought to capitalize on the independent efforts of prospective qui tam relators who call information to the attention of the government. As stated by the D.C. Circuit:
[T]he entire qui tam regime is premised on the idea that the government's knowledge of misrepresented claims against the federal fisc (without knowledge that they are misrepresented) does not in itself translate into effective enforcement of the laws against fraud.
Springfield Terminal, 14 F.3d at 656. While the government can be expected to be on notice of fraud when the allegations are contained in a public disclosure such as an administrative or congressional hearing, when responding to a FOIA request, the government need not assimilate the information contained in the requested documents. The duplication of FOIA-requested documents does not require the degree of familiarity and cognizance that the drafting of a report or the conducting of an investigation would. Accordingly, prohibiting qui tam relators from basing their allegations on any information obtained in a FOIA response would damage the fraud-detection purpose of the FCA while failing to serve its twin goal of preventing opportunism.5
The specific facts of this case illustrate how ill-fitting the labels of "report" or "investigation" can be for responses to FOIA requests. Haight ultimately obtained the documents 4398 that she had requested by FOIA — saliently Dr. Berens' successful grant application containing the alleged misrepresentations — by walking to Barrow and obtaining them herself. In this instance, the FOIA response consisted only of alerting Haight to the location of the documents relating to Dr. Berens' research. Far from putting any work product into those documents, in this instance the FOIA response did not even involve duplication. Responding to Haight's FOIA request did not position the government to detect or respond to the alleged fraud.
Haight, on the other hand, put in substantial time and effort into uncovering the allegations the Relators make in their complaint. In addition to filing a FOIA request, Haight requested documents, including animal care records and minutes from IACUC meetings, from Arizona State University; she toured the two facilities where dogs associated with Berens' research were housed; she contacted and interviewed at least nine people who had knowledge about the Berens' project; she also corresponded with a scientist overseas who was listed as a consultant to Berens' research. Moreover, she combed the documents that she received through her FOIA inquiry and from ASU in an attempt to learn about Berens' research. Ultimately, she uncovered what Relators allege to be serious misrepresentations about the project that resulted in an award of over $700,000 in federal funds. Relators did not behave opportunistically. To the contrary, Haight performed precisely the sort of investigative work that the qui tam provisions of the FCA encourage in order to promote detection of fraud against the government.
We hold that whether a document obtained via FOIA request should invoke the jurisdictional bar should be determined by reference to the nature of that document itself. If the document obtained via FOIA request is a public disclosure of a "criminal, civil, or administrative hearing, . . . a congressional, administrative, or [General] Accounting Office report, hearing, audit, or investigation, or [is] from the news media," then the jurisdictional bar is applicable. If, as was the case here, the document obtained via FOIA does not itself qualify as an enumerated source, its disclosure in response to the FOIA request does not make it so.
In the present case, the allegations and transactions in the complaint were based on information found solely in the grant application Haight obtained in response to the FOIA request. Because the grant application was not publicly disclosed via an enumerated source, Relators' suit is not barred by § 3730(e)(4)(A).
C.
Because we hold that the misrepresented state of facts, "X," was not publicly disclosed via an enumerated source, we need not, and do not, reach the questions of whether the true state of facts, "Y," was publicly disclosed via an enumerated source, or whether Relators were the "original source" of that true state of facts.
III.
We REVERSE and REMAND.
Notes:
The Honorable John R. Gibson, Senior United States Circuit Judge for the Eighth Circuit, sitting by designation
Contrary to Appellants' assertion, we assume that the elements of the fraud allegation need not have been made public in a single documentSee, e.g., United States ex rel. Found. Aiding the Elderly v. Horizon West Inc., 265 F.3d 1011 (9th Cir.2001), amended by, 275 F.3d 1189 (9th Cir.2001), reprinted as amended, 2001 U.S.App. LEXIS 27363, *7-11 (pin cites and subsequent citations are to the reprinted opinion) (analyzing several documents to determine whether the allegations and transactions were publicly disclosed); A-1 Ambulance Serv., Inc. v. California, 202 F.3d 1238 (9th Cir.2000) (finding a complaint barred by the jurisdictional provision based on the disclosures in several administrative hearings and the documents filed in conjunction with those hearings).
This court addressed a related question inUnited States ex rel. Schumer v. Hughes Aircraft Co., 63 F.3d 1512, 1520 (9th Cir.1995), vacated on other grounds, 520 U.S. 939, 117 S. Ct. 1871, 138 L. Ed. 2d 135 (1997), which the district court cited in support of its decision. Schumer's holding is not on point, however. Schumer holds only that material in a government file that is potentially accessible to the public through a FOIA request is not publicly disclosed. Id. at 1519-20. Only in dicta does the court discuss information that is disclosed through FOIA, saying: "In the FOIA context, information cannot be deemed disclosed until a member of the public requests the information and receives it from the government. Only then is the information actually, rather than theoretically or potentially, available to the public." Id. at 1520. Furthermore, Schumer was vacated by the Supreme Court on grounds unrelated to the public disclosure analysis. See Hughes Aircraft Co. v. United States ex rel. Schumer, 520 U.S. 939, 952, 117 S. Ct. 1871, 138 L. Ed. 2d 135 (1997). It is thus only persuasive authority and is not binding precedent. Orhorhaghe v. INS, 38 F.3d 488, 493 n. 4 (9th Cir.1994). The Schumer opinion does not supply the answer as to whether information received through a FOIA request is publicly disclosed.
The enumerated sources are "criminal, civil, or administrative hearing[s], ... congressional, administrative, or Government [General] Accounting Office report[s], hearing[s], audit[s], or investigation[s], or . . . the news media." 31 U.S.C. § 3730(e)(4)(A). With the exception of the news media, each of these enumerated sources is a form of work product originating with the government. The contents of those sources is expected to be well-known to the government. Likewise, the reports of the news media can be presumed to be known by the government
The sole exception to this generalization about the enumerated sources is the news media. The same principle applies in this instance, however. While information in the news media does not necessarily originate from the government, it is so public that the government can be presumed to be on notice of it
Moreover, holding that a FOIA response is necessarily a "report" or "investigation" would deter individuals who suspect fraud from investigating it. FOIA requests are one of the simplest vehicles by which interested citizens can uncover possible fraud against the government. If information obtained pursuant to FOIA requests could never form the basis of aqui tam action, prospective relators would have to invest substantially more energy into uncovering the suspected fraud through other means. Moreover, the government is not harmed by permitting claims based on information obtained pursuant to FOIA request from going forward; it still has an opportunity to intervene and take over the suit. See 31 U.S.C. § 3730(b)(2). Permitting claims based on otherwise private information obtained via a FOIA request to go forward also has the happy effect of encouraging private citizens with suspicions of fraud to take the most expeditious route toward uncovering information related to that fraud and hastening recovery for the government.