Davy v. Central Intelligence Agency 550 F.3d 1155

Opinion for the Court by Circuit Judge ROGERS.

Concurring opinion by Circuit Judge TATEL.

Dissenting opinion by Senior Circuit Judge RANDOLPH.

ROGERS, Circuit Judge:

This is the second time William Davy has appealed the denial of his request for an award of attorney’s fees and costs under the Freedom of Information Act (“FOIA”). Davy first appealed the district court’s finding that he was ineligible, and this court reversed, holding that he was eligible as a prevailing party and remanding the case for the district court to determine whether Davy was entitled to fees upon applying a familiar four-factor test. Davy v. CIA (“Davy I”), 456 F.3d 162 (D.C.Cir.2006). On remand the district court again denied fees. This court must reverse and remand again. Because the district court’s findings on some factors are unsupported by the record, and the record indicates that Davy is the quintessential requestor of government information envisioned by FOIA, he is entitled to an award of fees and costs, and upon remand the district court shall enter an appropriate order.

I.

The details of Davy’s two FOIA requests are set forth in Davy I. Suffice it to say, in 1999, six years after Davy, acting pro se, filed his first FOIA request, the agency responded by refusing disclosure, stating that it could neither confirm nor deny the existence of responsive records due to national security reasons, citing FOIA exemptions (1) and (3). 456 F.3d at 163. Davy obtained a lawyer but no relief by administrative appeal and filed suit against the agency. The district court dismissed his complaint with leave to amend on the ground that it was based on a FOIA request made in 1993 and so was untimely. *1158Davy then filed a second FOIA request in November 2000, renewing his initial request and seeking additional documents. In December 2000, having received no response from the agency, Davy amended his complaint to focus on his second FOIA request. On May 4, 2001, the district court entered an order adopting the parties’ agreement on a schedule for the agency to produce documents pursuant to Davy’s second FOIA request. Thereafter the agency produced on schedule some documents but not others and moved for summary judgment. Davy also moved for summary judgment. After the agency filed a superseding motion for summary judgment, the district court granted the agency’s motion.

Davy thereafter timely filed a motion for attorney’s fees under 5 U.S.C. § 552(a)(4)(E),1 which the district court denied. On appeal, this court held that Davy had substantially prevailed and was therefore eligible for fees and remanded the case so that the district court could, in the first instance, apply a four-factor test for determining entitlement. Davy I, 456 F.3d at 166-67. Davy now appeals the district court’s finding on remand that he was not entitled to an award of fees. Our review of the district court’s application of the four-factor test is for abuse of discretion. Tax Analysts v. U.S. Dep’t of Justice, 965 F.2d 1092, 1094 (D.C.Cir.1992); see generally Kickapoo Tribe v. Babbitt, 43 F.3d 1491, 1497 (D.C.Cir.1995).

II.

This court, drawing on the Senate and House Committee reports for FOIA and its amendments,2 explained long ago that the provision for attorney’s fees “was not enacted to provide a reward for any litigant who successfully forces the government to disclose information it wished to withhold,” but instead “had a more limited purpose — to remove the incentive for administrative resistance to disclosure requests based not on the merits of exemption claims, but on the knowledge that many FOIA plaintiffs do not have the financial resources or economic incentives to pursue their requests through expensive litigation.” Nationwide Bldg. Maint., Inc. v. Sampson, 559 F.2d 704, 711 (D.C.Cir.1977) (citing S.Rep. No. 93-854, at 17). The court embraced the view that a distinction is to be drawn between the plaintiff who seeks to advance his private commercial interests and thus needs no incentive to file suit, and a newsman who seeks information to be used in a publication or the public interest group seeking information to further a project benefit-ting the general public. Id. at 712-13 (quoting S.Rep. No. 93-854, at 19). The court observed in conclusion that:

The touchstone of a court’s discretionary decision under section 552(a)(4)(E) must be whether an award of attorney fees is necessary to implement the FOIA. A grudging application of this provision, which would dissuade those who have been denied information from invoking their right to judicial review, would be clearly contrary to congressional intent.

Id. at 715; see also LaSalle Extension Univ. v. FTC, 627 F.2d 481, 484 (D.C.Cir.1980).

*1159With this understanding, the court has directed the district court to consider at least four criteria in determining whether a substantially prevailing FOIA litigant is entitled to attorney’s fees: (1) the public benefit derived from the case; (2) the commercial benefit to the plaintiff; (3) the nature of the plaintiffs interest in the records; and (4) the reasonableness of the agency’s withholding of the requested documents. Tax Analysts, 965 F.2d at 1093-94; see also S.Rep. No. 93-854 at 19. No one factor is dispositive, although the court will not assess fees when the agency has demonstrated that it had a lawful right to withhold disclosure. See Chesapeake Bay Found. v. USDA (“Chesapeake I”), 11 F.3d 211, 216 (D.C.Cir.1993), abrogated in part on other grounds by Buckhannon Bd. & Care Home, Inc. v. W. Va. Dep’t of Health & Human Resources, 532 U.S. 598, 601-02, 121 S.Ct. 1835, 149 L.Ed.2d 855 (2001).

The first factor assesses “the public benefit derived from the case,” Tax Analysts, 965 F.2d at 1093, and requires consideration of both the effect of the litigation for which fees are requested and the potential public value of the information sought, see Chesapeake Bay Found. v. USDA (“Chesapeake II), 108 F.3d 375, 377 (D.C.Cir.1997); Cotton v. Heyman, 63 F.3d 1115, 1120 (D.C.Cir.1995); Tax Analysts, 965 F.2d at 1093-94. The district court found that “Davy’s FOIA request and subsequent litigation were intended to compel disclosure of information relating to the activities of a government agency (the CIA) in relation to a significant historical event,” and thus that this factor favors Davy. There can be little question that this factor favors Davy.

The information Davy requested — about individuals allegedly involved in President Kennedy’s assassination — serves a public benefit. See, e.g., Allen v. CIA, 636 F.2d 1287, 1300 (D.C.Cir.1980), overruled on other grounds by Founding Church of Scientology v. Smith, 721 F.2d 828, 830 (D.C.Cir.1983). At least one of the requested documents was not previously available to the public, and the agency did not challenge Davy’s description of the released documents as providing “important new information bearing on the controversy over former [District Attorney Jim] Garrison’s contention that the CIA was involved” in the assassination plot. Davy Decl. ¶ 2. Nothing in the record indicates that the releases, which occurred only after the May 4, 2001 order of the district court, were not a fruit of Davy’s litigation; despite Davy’s second FOIA request, the agency did not turn over any documents to him until after he filed suit. As this court stated in Davy I, it was the district court’s disclosure-schedule order that “provide[d] Davy with the precise relief his request sought.” 456 F.3d at 165.

The agency’s position — that the district court erred by failing to focus on the value of the litigation — presents a variation on its position, rejected in Davy I, that Davy did not “substantially prevail” in his litigation and so was not eligible for fees. Davy I, 456 F.3d at 166. Because nothing in the record indicates that Davy would have received the information without filing suit, the district court’s consideration of the value of the information sought necessarily entailed consideration of the value of the litigation that led to the disclosure of that information. The cases on which the agency relies are inapposite, involving a pre-litigation offer of release in Chesapeake II, 108 F.3d at 377, or litigation that produced only faster disclosure of publicly available information in Tax Analysts, 965 F.2d at 1094, or establishment of a legal precedent defining “public interest” in Cotton, 63 F.3d at 1120. The fact that some of the material turned over to Davy concerns an *1160event of national importance and is newly released is a key distinction between this case and the litigation at issue in Tax Analysts.

Although the district court’s determination that the first factor weighed in Davy’s favor was not an abuse of discretion, we reach a different conclusion regarding its determination of the other factors. The second and third factors, which are often considered together, assess whether a plaintiff has “sufficient private incentive to seek disclosure” without attorney’s fees. See Tax Analysts, 965 F.2d at 1095. The second factor considers the commercial benefit to the plaintiff, while the third factor considers the plaintiffs interest in the records. Applying these factors, the district court found that because the requested documents were used to research a book that was later published, albeit with “presumably ... limited commercial success,” “Davy’s interest in the records was clearly commercial.” These findings are based on inappropriate considerations and clearly erroneous findings of fact.

First, the mere intention to publish a book does not necessarily mean that the nature of the plaintiffs interest is “purely commercial.” See S. Rep. No. 93-854, at 19. Surely every journalist or scholar may hope to earn a living plying his or her trade, but that alone cannot be sufficient to preclude an award of attorney’s fees under FOIA. “If newspapers and television news shows had to show the absence of commercial interests before they could win attorney’s fees in FOIA cases, very few, if any, would ever prevail.” Tax Analysts, 965 F.2d at 1096. Yet their activities often aim to ferret out and make public worthwhile, previously unknown government information — precisely the activity that FOIA’s fees provision seeks to promote. Furthermore, a conclusion that using information obtained under FOIA in connection with research for purposes of writing a book necessarily makes fees unavailable is inconsistent with the distinction that underlies this court’s analysis of the relevant factors. Cf. Nationwide, 559 F.2d at 713 (discussing Goldstein v. Levi, 415 F.Supp. 303, 305 (D.D.C.1976)).

Essentially, the first three factors assist a court in distinguishing between reques-ters who seek documents for public informational purposes and those who seek documents for private advantage. The former engage in the kind of endeavor for which a public subsidy makes some sense, and they typically need the fee incentive to pursue litigation; the latter cannot deserve a subsidy as they benefit only themselves and typically need no incentive to litigate. Thus, on the one hand, the court has reversed an award of attorney’s fees where the plaintiff was an attorney requesting information in preparation of private litigation for a client, Cotton, 63 F.3d at 1120, and upheld a denial of fees where a nonprofit organization was reprinting federal tax decisions in a newsletter sent to paying subscribers, Tax Analysts, 965 F.2d at 1093. Although Tax Analysts had a news interest, it simply sought “efficient, easy access” to information that was already public, id. at 1095, in order to make it available to its subscribers sooner, and the court concluded that such a subscriber organization did “not need the attorney’s fees incentive” to pursue litigation, id. at 1096. On the other hand, the court has long recognized that “news interests,” regardless of private incentive, generally “should not be considered commercial interests” for purposes of the second factor, Tax Analysts, 965 F.2d at 1096; Fenster v. Brown, 617 F.2d 740, 742 n. 4 (D.C.Cir.1979) (quoting S.Rep. No. 93-854, at 19), and that “a court would generally award fees if the complainant’s interest in the *1161information sought was scholarly or journalistic or public-interest oriented, [unless] ... his interest was of a frivolous or purely commercial nature,” Fenster, 617 F.2d at 742 n. 4 (quoting S.Rep. No. 93-854, at 19); Tax Analysts, 965 F.2d at 1096.

Second, in finding that Davy’s interest was “purely commercial, the district court relied exclusively on the fact that Davy is the author of Let Justice Be Done, a book about the investigation by New Orleans District Attorney Jim Garrison and the trial of Clay Shaw for conspiracy to assassinate President Kennedy.” Yet this book was published in 1999, prior to the release of documents by the agency. This alone suffices to show that the district court abused its discretion. But even if the district court had been correct about the book, such scholarly interests are not “clearly commercial” under this circuit’s precedents. Davy’s FOIA requests for information related to the agency’s QKEN-CHANT and ZRCliff projects, which were based on his interest in the agency’s alleged involvement in the assassination, Davy I, 456 F.3d at 163, reflect more of a scholarly than a disqualifying commercial interest. The record indicates not only that Davy has also published magazine articles on the assassination but that some of the information released to him under FOIA had not previously been made public. There is no suggestion in the record that his requests were frivolous. In fact, Davy’s unchallenged declaration makes clear the substantive value of the released documents and the agency has not meaningfully argued otherwise even on appeal.

Additionally, Davy states in his sworn declarations that his “primary motivation was to obtain records which would shed light on [the Garrison] investigation, ... and present an accurate historical record regarding it,” and that his book made a “miniscule” amount of money. Davy Supp. Deck ¶ 4. Contrary to the district court’s speculation and the agency’s suggestion that because his book, out of print since 2004, was for sale on Amazon.com he must be commercially profiting, Davy stated that he received no money from such second-hand sales. Davy Supp. Deck ¶¶ 4-5. These are not the circumstances indicative of private, self-interested motivation or commercial pecuniary benefit that the court has recognized “will be sufficient to insure the vindication of the rights given in the FOIA.” Fenster, 617 F.2d at 743 n. 4 (quoting S.Rep. No. 93-854, at 19).

To the extent that Davy has a scholarly interest in publishing publicly valuable information in a book, his interest is at most “quasi-commercial,” Campbell v. U.S. Dep’t of Justice, 164 F.3d 20, 36 (D.C.Cir.1998). This alone is not disqualifying as nothing in the record would suggest that his private commercial interest outweighs his scholarly interest, much less the public value in providing him an incentive to ferret out and publish this information. See, e.g., Aronson v. U.S. Dep’t of Hous. & Urban Dev., 866 F.2d 1, 3 (1st Cir.1989). As the court observed in interpreting the FOIA provision providing for waiver or reduction of copying fees, 5 U.S.C. § 552(a)(4)(A)(iii), “Congress did not intend for scholars (or journalists and public interest groups) to forego compensation when acting within the scope of their professional roles.” Campbell, 164 F.3d at 35-36; see also Nat’l Treas. Employees Union v. Griffin, 811 F.2d 644, 649 (D.C.Cir.1987). Furthermore, Davy sought the information not for its republication value standing alone as in Tax Analysts, but in relation to a larger work addressing an historical issue of national importance.

Davy was thus much like a journalist who “gathers information of potential interest to a segment of the public, uses [his] *1162editorial skills to turn the raw materials into a distinct work, and distributes that work to an audience,” Tax Analysts, 965 F.2d at 1095, and as such is among those whom Congress intended to be favorably treated under FOIA’s fee provision, id. at 1096. He is at least the quintessential “average person,” Cuneo v. Rumsfeld, 553 F.2d 1360, 1363-64 (D.C.Cir.1977), abrogated on other grounds by Kay v. Ehrler, 499 U.S. 432, 438, 111 S.Ct. 1435, 113 L.Ed.2d 486 (1991), requesting information under FOIA about what the government was up to that he intends to share with the public as part of his scholarship or “news” gathering role rather than merely to promote his private commercial interests. For the reasons discussed and because the record provides no basis to doubt that his purpose in filing the FOIA request and pursuing litigation was to increase the public fund of knowledge about a matter of public concern, the district court abused its discretion in determinating that the second and third factors weighed against Davy.3

The fourth factor considers whether the agency’s opposition to disclosure “had a reasonable basis in law,” Tax Analysts, 965 F.2d at 1096, and whether the agency “had not been recalcitrant in its opposition to a valid claim or otherwise engaged in obdurate behavior,” LaSalle Extension, 627 F.2d at 486; see Fenster, 617 F.2d at 744. “If the Government’s position is correct as a matter of law, that will be dispositive. If the Government’s position is founded on a colorable basis in law, that will be weighed along with other relevant considerations in the entitlement calculus.” Chesapeake I, 11 F.3d at 216; see Cotton, 63 F.3d at 1117; Nationwide, 559 F.2d at 712 n. 34. The district court found that “there is no basis to conclude that the [a]gency unreasonably withheld these requested documents.” But this *1163analysis mistakenly shifts the burden to the requester. The question is not whether Davy has affirmatively shown that the agency was unreasonable, but rather whether the agency has shown that it had any colorable or reasonable basis for not disclosing the material until after Davy filed suit.

The agency did not reach an agreement to disclose the requested documents until March 2001, after Davy filed his lawsuit and four months after he filed his second FOIA request. Davy observes on appeal that the agreement coincided with the filing date for meet-and-confer statements under Local Rule 16 of the district court, and that it took more than a year for the agency to process and release a moderate quantity of documents. That aside, although the agency invoked FOIA Exemptions 1 and 3 when it finally responded to Davy’s first FOIA request, it provided no such legal basis in response to Davy’s second FOIA request. Failing to explain the basis for deferring its response to his second request until after he filed suit is exactly the kind of behavior the fee provision was enacted to combat. For the agency to receive the benefit of the fourth factor it must present at least a “colorable basis in law” for its failure to respond to Davy’s second request, and it has not done so. See Nationwide, 559 F.2d at 712 n. 34. It is not enough to say that “once the [ajgency faced a justiciable FOIA claim, it offered no resistance,” Appellee’s Br. at 19, because the agency did not disclose the documents until after Davy had pursued litigation, including filing a cross-motion for summary judgment and negotiating a release schedule. Even on appeal the agency does not suggest that the “information disclosed after initial resistance was nonetheless exempt from the FOIA,” or that “it had a reasonable basis in law for resisting disclosure.” Nationwide, 559 F.2d at 712 n. 34. If the government could defeat an award of fees by citing a lack of resistance after the requester files a lawsuit to obtain requested documents, then it could force plaintiffs to bear the costs of litigation. Absent evidence that the agency had a reasonable basis for failing to respond to Davy’s second request, the district court abused its discretion in determining that the fourth factor weighed in the agency’s favor.

Accordingly, because the record reflects that he is the type of requester Congress contemplated when it sought “to lower the ‘often ... insurmountable barriers presented by court costs and attorney fees to the average person requesting information under the FOIA,’ ” Tax Analysts, 965 F.2d at 1095 (quoting Cuneo, 553 F.2d at 1363-64), and because no factor weighs in the agency’s fayor, a balancing of the factors can only support the conclusion that Davy is entitled to an award of attorney’s fees. Accordingly, we reverse and remand the case only for the district court to enter an appropriate order awarding fees and costs as to all matters on which Davy prevailed.

TATEL, Circuit Judge,

concurring:

Agreeing completely with Judge Rogers’s application of our long-standing test for FOIA fee eligibility, I join her opinion in full. I write only to clarify a single point: that William Davy presents a paradigmatic case for the award of attorney’s fees even if we step back from the particulars of the test the dissent so maligns and focus instead on FOIA’s purposes. While recognizing the test as binding precedent, the dissent nonetheless casts aspersions on Davy’s case and subjects it to newly minted standards inconsistent with the very purpose of FOIA’s fees provision. Indeed, because the barriers the dissent would erect appear insuperable, I read it not as a dissent from this Court’s opinion, but from *1164Congress’s decision to provide fees to prevailing FOIA requesters at all.

Begin with the first factor, which asks whether the FOIA requester pursued the litigation in the public interest. Maj. Op. 1159. The purpose of this inquiry is obvious: Congress meant to incentivize the pursuit of public informational interests, not the mining of government records for private advantage. E.g., Tax Analysts v. U.S. Dep’t of Justice, 965 F.2d 1092, 1095 (D.C.Cir.1992). But the dissent would ignore the purpose of the document request and ensuing litigation, focusing instead on whether the particular records released would be interesting to the public or would instead be greeted with “a yawn.” Dis. Op. 1167. As the facts of this very case demonstrate, however, assessing the content of the specific documents disclosed rather than the reasons they were requested makes little sense. Before suing, re-questers in Davy’s position have no idea what documents responsive to their FOIA requests might contain because the agency has told them nothing — here, it never even gave Davy a Vaughn index. In fact, Davy knew only that his request implicated matters of such enormous national concern that the CIA at first expressly refused to either confirm or deny the existence of responsive records. So Davy went to court seeking more information, exactly as Congress intended. Indeed, Congress created a fees incentive precisely so that people in Davy’s situation would sue where, as here, the agency digs in. Maj. Op. 1158; see also Nationwide Bldg. Maint., Inc. v. Sampson, 559 F.2d 704, 715 (D.C.Cir.1977) (“A grudging application of this [fees] provision, which would dissuade those who have been denied information from invoking their right to judicial review, would be clearly contrary to congressional intent.”). The dissent would force requesters to bear the risk that the revealed documents might ultimately be boring, but since no one in Davy’s position can know before suing what the requested documents say or even whether they exist, the dissent’s rule would in fact chill all FOIA suits, preventing the discovery of important and unimportant content alike. So long as the information sought was of a type the public might want to know, we should consider the objective underlying this element of our test met. Davy’s suit sought records regarding the assassination of an American president; we need know nothing more to find that the first factor favors him.

The dissent disagrees with this view of both the law and the facts. As to the law, the dissent argues that a FOIA request’s purpose is irrelevant, Dis. Op. 1167-68, citing our standard in Cotton v. Heyman, 63 F.3d 1115 (D.C.Cir.1995), which asks whether a request “is likely to add to the fund of information that citizens may use in making vital political choices,” id. at 1120 (internal quotation marks omitted). This argument founders on Cotton itself, which in fact applied its standard by looking to the purpose behind the FOIA request:

In this case, no evidence exists that the release of the two non-exempt documents will contribute to the public’s ability to make vital political choices. Plaintiff does not even argue this point. Rather, she sought these documents for the sole purpose of facilitating her employment discrimination suit.

63 F.3d at 1120 (emphasis added). As to the facts, the dissent disagrees with my representation of Davy’s request as one seeking records related to the assassination of President Kennedy. Dis. Op. 1167-68. Of course, this isn’t my representation, it’s the district court’s, see Davy v. CIA, 496 F.Supp.2d 36, 38 (D.D.C.2007) (“Davy’s FOIA request and subsequent litigation were intended to compel disclosure *1165of information relating to the activities of a government agency (the CIA) in relation to a significant historical event .... ”), and the dissent never so much as mentions the deference we owe that finding. But even so, Davy’s request clearly did relate to the Kennedy assassination; as the dissent itself observes, “[m]uch of what Davy obtained was already in the public domain, released under the John F. Kennedy Assassination Records Collection Act.” Dis. Op. 1167-68 (emphasis added).

This brings us to the second and third factors, which inquire into the nature of the requester’s interest in the records and whether the requester can be expected to benefit commercially from obtaining the documents. Maj. Op. 1159-60. The reason for these inquiries is obvious as well: Congress did not intend to subsidize those who stand to profit from pursuing litigation and so need no subsidy. See, e.g., Cuneo v. Rumsfeld, 553 F.2d 1360, 1367 (D.C.Cir.1977) (“If the potential for private commercial benefit was sufficient incentive to encourage [plaintiff] to pursue his FOIA claim, the policy objectives of section 552(a)(4)(E) would be met and it would not be improper for the trial court to deny his request for attorney fees.”), abrogated on other grounds by Kay v. Ehrler, 499 U.S. 432, 438, 111 S.Ct. 1435, 113 L.Ed.2d 486 (1991). The dissent’s dismissive aside about whether Davy should qualify as a “journalist” under these elements of the test is thus a red herring. Nothing in the record suggests that Davy’s persistent interest in the Kennedy assassination was ever likely to earn him any pecuniary gain. To the contrary, uncontroverted record evidence demonstrates that absent the promise of fees, Davy lacked both the incentive and the ability to pursue his request through litigation. These factors thus favor him regardless of his scholarly or journalistic credentials.

That said, it is worth pausing to consider why we ask whether a requester pursues information in a journalistic capacity. Scholars, authors, and journalists straddle the incentive inquiry framed by the first three factors of the entitlement test. Always searching for information that the public will want to consume, journalists must surely be thought of as pursuing records in the public interest. At the same time, because they have a strong profit motive in that pursuit, they need fees less than most. Echoing the judgment of Congress as expressed in the legislative history, we have long resolved the tension by considering scholarly or journalistic interests to be public rather than private. See Maj. Op. 1160, 1161-62 (collecting authorities). The dissent faults that judgment, but in reality, it cannot itself decide which side of the debate to join. In one breath it decries awarding fees to large media organizations that profit from obtaining and reselling information, Dis. Op. 1166 & n. 1; in the next it faults Davy for his inability to prove that he had already revealed the information he obtained to the public, hypothesizing that the records went straight from some government file cabinet to Davy’s closet, id. at 1167. Of course, the only kinds of FOIA requesters who can prove that they almost immediately circulate the information they obtain belong to “the group that is in the business of profiting from the information when it winds up in their newspapers and magazines and TV shows.” Id. at 1166 n. 1. So in the dissent’s world no one can obtain fees: journalists circulate their information too successfully and so are excluded, while independent scholars such as Davy fail to circulate their information quickly or widely enough.

This Catch-22 is completely at odds with the intent of Congress’s fees provision. Scholars like Davy often lack resources and need more time to research their *1166work. No single FOIA request is likely to produce the smoking gun that independently verifies their complicated hypothesis or grabs the public’s attention. That the requester wisely waits to evaluate and synthesize released records before broadcasting information to the public in no way undermines the fact that the records were sought in the public interest by an individual without adequate ability or incentive to sue.

A brief word on the entitlement test’s final element, which serves a different purpose from the first three. Asking whether the government had a reasonable basis for withholding documents, this fourth factor is intended to disincentivize requesters from complaining about reasonable with-holdings while incentivizing the government to promptly turn over — before litigation is required — any documents that it ought not withhold. That purpose will be ill-served if the government can prevail on this factor by saying nothing and forcing the requester to sue, only then to offer “no resistance,” Appellant’s Br. 19, as it did here. As we explained in Nationwide, 559 F.2d at 710, “[i]f the government could avoid liability for fees merely by conceding the cases before final judgment, the impact of the fee provision would be greatly reduced.” Were that so, “[t]he government would remain free to assert boilerplate defenses, and private parties who served the public interest by enforcing the Act’s mandates would be deprived of compensation for the undertaking.” Id. That is why we consider plaintiffs to have substantially prevailed if they win a stipulated release, and that is why the government’s argument is essentially foreclosed by Davy’s victory in his first appeal. See Davy v. CIA, 456 F.3d 162 (D.C.Cir.2006).

In short, our four-factor test is a heuristic, a somewhat crude mechanism for testing whether fees in a particular case are consistent with the purposes for which Congress subsidized FOIA litigation. We hardly need such a divining rod for Davy, whose entitlement to fees is clear as day. The dissent accuses us of rendering the test “more senseless,” Dis. Op. 1166, yet it is the dissent’s requirements that would divorce the test from the ends Congress intended FOIA fees to serve.

RANDOLPH, Senior Circuit Judge,

dissenting:

Precedent forces the majority to apply a longstanding test for determining whether to award attorney’s fees. It is time to recognize that this test is a legal relic. It is derived not from the statute but from statements in committee reports, it is inconsistent with now-settled FOIA law that the identity of the requester is irrelevant, see, e.g., NLRB v. Sears, Roebuck & Co., 421 U.S. 132, 149, 95 S.Ct. 1504, 44 L.Ed.2d 29 (1975); Sterling Drug, Inc. v. FTC, 450 F.2d 698, 705 (D.C.Cir.1971), and it draws an irrational line between news organizations and other commercial and non-commercial businesses and individuals.1 Although stare decisis commands us to use the test, it does not command us to render it even more senseless than it already is.

The majority holds that Davy should be treated as a journalist and is entitled to attorney’s fees because he provided a public benefit by gathering valuable information through this lawsuit. This is unsupported and unsupportable. Davy wrote a book a few years before the government complied with his FOIA request. I do not know if that makes him a “journalist.” I *1167cannot see why that should matter in any event. Davy provided nothing to show that the information he sought and received was valuable or important. The majority is incorrect in stating that the government conceded otherwise. See Gov’t Br. at 11. Much of what Davy obtained was already in the public domain, released under the John F. Kennedy Assassination Records Collection Act of 1992, Pub.L. No. 102-526, 106 Stat. 3443. No one can say why the relatively few newly released documents Davy obtained benefited the public in any way. Davy asserts that new information came to light, but this consisted of the names of people who had obtained a clearance for classified material or the code names of already-known people and enterprises. See Davy Decl. ¶¶ 1-3. This is the kind of data the populace would greet with a yawn. It surely does not amount to “information that citizens may use in making vital political choices.” Cotton v. Heyman, 63 F.3d 1115, 1120 (D.C.Cir.1995). Davy may think the public profited from his efforts, but he has never said why.

Even if his documents amounted to anything, Davy failed to show that “the public” — whoever that might be — was somehow better off as a result of his FOIA request. Davy obtained the documents in 2001, two years after his book was published. He submitted no evidence that he showed the documents to anyone else (other than his lawyer and the court) or that he posted them online or that he published anything about them or that he plans to do so in the future. For all we know the documents are gathering dust in the corner of his closet. Tax Analysts stressed that the very small circulation of a publication was a reason for denying fees, see Tax Analysts v. Dep’t of Justice, 965 F.2d 1092, 1094 (D.C.Cir.1992), and several of our cases have rejected fee requests when the requested information was not widely disseminated to the public, see Cotton, 63 F.3d at 1120; Fenster v. Brown, 617 F.2d 740, 744 (D.C.Cir.1979).

In short, Davy did not even come close to satisfying his burden of showing that his lawsuit produced something of value. See Hensley v. Eckerhart, 461 U.S. 424, 437, 103 S.Ct. 1933, 76 L.Ed.2d 40 (1983); Anderson v. Soc’y of Health & Human Servs., 80 F.3d 1500, 1504 (10th Cir.1996). For that reason the public should not have to foot the bill for his litigation costs.

I will end with a few words about the concurring opinion. Judge Tatel says that what matters in terms of public benefit are “the reasons [the documents] were requested.” Op. of Judge Tatel 1. He caps this off by telling us that Davy “sought records regarding the assassination of an American president.” Id. at 2. Judge Ta-tel’s first proposition misstates the law; his second misstates the facts. As I have already said, the law of the circuit is clear: the public benefit criterion favors awarding fees only when “the complainant’s victory is likely to add to the fund of information that citizens may use in making vital political choices.” Cotton, 63 F.3d at 1120. Nothing in that formulation turns on the requester’s motives in seeking the documents. Rather, a court must “evaluate the specific documents at issue in the case at hand,” id., and determine whether the public actually benefited from the FOIA litigation, Chesapeake Bay Found., Inc. v. Dep’t of Agric., 108 F.3d 375, 377 (D.C.Cir.1997). As to the facts, Davy did not seek records relating to the assassination of President Kennedy, as Judge Tatel represents. Davy requested files relating to a program involving background checks and a CIA-operated airline. There is nothing to connect any newly released information to the Kennedy assassination. Maybe Davy imagined some connection. But it had not occurred to me that the taxpayers *1168ought to be subsidizing someone who is pursuing a figment.