The News-Press v. U. S. Dept. of Homeland Security

489 F.3d 1173, 35 Media L. Rep. (BNA) 2289, 2007 U.S. App. LEXIS 14817

• Court: Court of Appeals for the Eleventh Circuit
• Decided: June 22, 2007
• Docket: 05-16771, 06-13306
• Status: Published

Opinion

MARCUS, Circuit Judge:

These consolidated appeals arise out of an unprecedented storm season in which four hurricanes — Hurricanes Charley, Frances, Ivan, and Jeanne — hit Florida within one six-week period during 2004. In response, the Federal Emergency Management Agency (“FEMA”) disbursed $1.2 billion in individual disaster assistance to more than 605,500 Floridians, and also paid out claims to tens of thousands of individuals whose structures were insured under FEMA’s National Flood Insurance Program. After questions were raised concerning how individual disaster assistance was disbursed in one Florida county following one of the hurricanes, the plaintiff newspapers collectively requested, under the Freedom of Information Act, 5 U.S.C. § 552 (“FOIA”), disbursement data for all four 2004 hurricanes plus an additional 27 disasters dating back some ten years. FEMA redacted the names and addresses from the data it provided, on the grounds that disclosing this information “would constitute a clearly unwarranted invasion of personal privacy” within the meaning of Exemption 6 of the FOIA, 5 U.S.C. § 552(b)(6). In News-Press v. U.S. Department of Homeland Security, the United States District Court for the Middle District of Florida held that disclosure of both the names and the addresses was exempt under Exemption 6. In Sum-Sentinel v. U.S. Department of Homeland Security, the United States District Court for the Southern District of Florida reached nearly the opposite conclusion, holding that FOIA requires FEMA to disclose the addresses, although not the names.

At issue today is whether FEMA has established that the names and addresses of 1.3 million individuals who applied for aid or made insurance claims after one of 31 federally-declared disasters are exempt from disclosure under the FOIA, on the grounds that releasing this information “would constitute a clearly unwarranted invasion of personal privacy” within the meaning of Exemption 6. After thorough review, we conclude that the addresses are not exempt under Exemption 6 because FEMA has failed to meet its heavy burden of showing a “clearly unwarranted invasion of personal privacy.” In light of FEMA’s awesome statutory responsibility to prepare the nation for, and respond to, all national incidents, including natural disasters and terrorist attacks, there is a powerful public interest in learning whether, and how well, it has met this responsibility. Plainly, disclosure of the addresses will help the public answer this question by shedding light on whether FEMA has been a good steward of billions of taxpayer dollars in the wake of several natural disasters across the country, and we cannot find any privacy interests here that even begin to outweigh this public interest. However, because there is only minimal additional public interest in disclosing the names, we conclude that they are exempt under Exemption 6.

I. Background

The following facts are culled from both summary judgment records and are undisputed. In the aftermath of the September 11, 2001 terrorist attacks, Congress created a Cabinet-level Department of Homeland Security (“DHS”) to serve as an umbrella organization for twenty-two federal departments. Homeland Security Act of 2002, Pub.L. No. 107-296, 116 Stat. 2135 (2002). On January 10, 2003, President George W. Bush nominated Michael D. Brown (“Brown”) to serve as the DHS’s first Under Secretary of Emergency Preparedness and Response. On March 1, 2003, the Federal Emergency Management Agency (“FEMA”) — whose mission is to “lead the effort to prepare the nation for all potential disasters and to manage the federal response and recovery efforts following any national incident — whether natural or man-made,” http://www.fema. gov/library/view Record.do?id=1756; see also Homeland Security Act § 502, 116 Stat. at 2212 — was formally folded into the DHS, and Brown assumed the position of Director of FEMA.

A. The 2004 Florida Hurricane Season

Within six weeks during August and September of 2004, portions of Florida sustained extensive damage from Hurri canes Charley, Frances, Ivan, and Jeanne — the first time since 1886 that a state has been struck by four hurricanes in a single year. After each hurricane, President Bush issued a major disaster declaration pursuant to the Robert T. Stafford Disaster Relief and Emergency Assistance Act, 42 U.S.C. § 5121 et seq. (“the Stafford Act”), and directed FEMA to provide federal disaster assistance to the affected areas. At that time, FEMA’s response to the four 2004 hurricanes comprised the largest mobilization of emergency response and disaster recovery resources in FEMA history, exceeding even the agency’s operational response to the attacks of September 11, 2001.

As a result of the 2004 hurricanes, FEMA received over 33,000 claims from Floridians under the National Flood Insurance Program (“NFIP”). Under this program, homeowners, renters, and business owners in certain designated flood-prone areas are eligible to purchase federal flood insurance for their building structures and their contents.

FEMA also paid out $1.2 billion in aid to more than 605,500 Floridians under the Individuals and Households Program (“IHP”), which provides financial assistance (that is not repaid by the recipient) and direct services to individuals and households who seek to “prevent, mitigate, or overcome a disaster-related hardship, injury or adverse condition” when those needs cannot be met in any other way. ■ 44 C.F.R. § 206.111; see also 42 U.S.C. § 5174(a)(1). There are two components of the IHP — Housing Assistance (“HA”) and Other Needs Assistance (“ONA”)— and roughly half of the IHP aid was disbursed under each component. Under the HA component, FEMA compensates individuals for temporary housing, the repair or replacement of damaged housing, or for “hazard mitigation measures” to their residences to reduce the likelihood of damage in the future. Individuals may also receive direct assistance in the form of temporary FEMA housing. Finally, 95,000 Floridians qualified for Expedited Assistance (“EA”), a form of Housing Assistance under which they were immediately given, without inspection, funds in the equivalent of one month’s fair market rent to be used toward their disaster-related housing needs. Under the ONA component, FEMA compensates individuals for funeral expenses, medical and dental costs, the repair and replacement of personal property, transportation expenses, moving and storage expenses, and other expenses, such as generators, deemed by FEMA to constitute a necessary expense or serious need.

With the exception of EA, all other IHP aid is disbursed only after FEMA’s contract inspectors verify the accuracy of claims, including the existence of disaster-related damage to real and personal property, as well as ownership or occupancy. Inspectors using portable data devices upload their findings to FEMA’s processing system, the National Emergency Management Information System (“NEMIS”), and in more than 90% of cases, an award is made on the basis of that inspection information. 1 Inspectors determine the amount of personal property loss under the ONA component using the Generic Room Concept: rather than comparing the state of household belongings after a disaster to the state Of those belongings before the disaster, FEMA inspectors compare the post-disaster state of a household’s belongings to the belongings in a hypothetical room FEMA has predetermined constitutes an “average” American kitchen, bathroom, living room, or bedroom. 2 The value of a full, hypothetical room ranges from $862 (for a bathroom) to $2,495 (for a bedroom). Inspectors make ONA awards by categorizing rooms in one of three ways. If the inspector determines that all items in a room must be replaced, the household receives the full value of an average room. If some of the room’s contents must be replaced but others can be repaired, however, the household receives only a portion of the value of a full room. Finally, if all of a room’s items can be repaired, then the household receives a still smaller portion of the value of a full room.

Damage to other personal property, such as appliances, clothing, and automobiles, is assessed differently. For example, inspectors categorize damaged automobiles as “destroyed,” “repairable,” or “cosmetic,” and award the replacement cost rather than market value; inspectors must confirm that damage is disaster-related, but are not required to note how they confirmed this, or even the type of damage sustained. Similarly, inspectors are generally permitted to award money for losses to personal property that is not present at the time of inspection, so long as the damage or loss can be reasonably verified in some other way. Inspectors note such verbal representations as “PP Verbal”s, but are not required to document how they verified the loss, or even what the lost item was.

Of the $1.2 billion paid in IHP aid to Floridians as a result of the 2004 hurricane season, $31 million went to residents of Miami-Dade County for damage resulting from Hurricane Frances alone, including $13 million in Housing Assistance (over $1 million of which was disbursed to 1,400 residents as Expedited Assistance) and just under $18 million in Other Needs Assistance (the majority of which was disbursed to compensate personal property losses).

By October of 2004, soon after the hurricane season ended, the media had begun to question why Miami-Dade County, which in fact suffered only tropical storm-force sustained winds and no substantial rainfall accumulation, had been declared eligible for disaster relief as a result of Hurricane Frances, the eye of which made landfall some 100 miles to the north of the county, 3 and why its residents apparently required $31 million in IHP assistance. Counties that suffered direct hits, the media reports claimed, received less aid. Reports also surfaced that since 2003, the number of National Flood Insurance Program-covered properties that have flooded repeatedly had more than doubled; for example, one North Miami property had reportedly flooded 17 times but remained NFIP-eligible, and without increased premiums. The media’s concern was soon shared by federal, state, and local officials, as well as by the public at large.

We recount in some detail both the various allegations that FEMA’s disbursement of IHP aid in Miami-Dade (and quite possibly in other Florida counties during 2004 and indeed in other states in other years after other disasters) was plagued with fraud, waste, and abuse, as well as FEMA’s responses to these allegations. We do so for several reasons. First, the weight of the discernible public interest in learning whether FEMA has been a good steward turns, in part, on how many tax dollars were spent in IHP aid following various disasters, as well as how much of that money may have been disbursed under conditions of fraud, waste, or abuse. Similarly, the weight of the public interest also turns in some measure on whether whatever problems occurred in Miami-Dade after Hurricane Frances were likely confined to that occasion and location, or whether they may instead be an indication of more systemic problems in FEMA’s individual assistance program that may have affected many more Americans and billions more tax dollars. Finally, the weight of the public interest turns, in part, on whether these questions have already been answered fully, or whether, instead, important questions remain that the names or addresses may help resolve.

In a prepared statement distributed to the press, Daniel Craig, FEMA’s Director of Recovery (“Craig”), downplayed the possibility that either applicant fraud or agency error contributed to FEMA’s distribution of aid to Floridians. “[W]hen you consider the magnitude of the recovery effort,” he said, “our process handled the applications very well.” Prepared Statement of FEMA Director of Recovery Daniel Craig Regarding Florida Disaster Assistance 3 (Jan. 10, 2005). “[TJhere’s currently no evidence of widespread fraud, waste or abuse of FEMA’s disaster assistance programs in Florida.” Id. at 1. “We’ve found that the majority of concerns raised regarding assistance provided to individuals in Florida have logical explanations and are not representative of widespread fraud.” Id. at 2 (emphasis in original).

Despite FEMA’s attempts to downplay any systemic problems in its distribution process, various internal and external investigations into FEMA’s response to Hurricane Frances in Miami-Dade County soon began. In January of 2005, both the Office of Inspector General (“OIG”) of the Department of Homeland Security and the United States Senate Committee on Homeland Security and Governmental Affairs (“Senate Committee”) announced they were opening investigations. Meanwhile, on March 2, 2005, the United States Attorney for the Southern District of Florida announced the indictments of fourteen Miami-Dade defendants on charges that each fraudulently claimed thousands of dollars — sometimes tens of thousands of dollars — in damage to his or her personal property from Hurricane Frances. In some cases, the indictments charged that damage to the defendant’s personal property had been sustained prior to Hurricane Frances. In at least three cases, the indictments charged that the defendant claimed losses from a hurricane-damaged property where he had not resided at the time of the hurricane.

On May 18, 2005, the OIG released the results of its internal audit into FEMA’s response to the 2004 hurricane season. See Dep’t of Homeland Sec., Office of the Inspector Gen., Audit of FEMA’s Individuals and Households Program in Miami-Dade County, Florida, for Hurricane Frances (2005) [hereinafter OIG Audit Report]. 4 The audit was overseen by Richard Skinner, then Acting Inspector General (“Skinner”). On the same day, the Senate Committee held hearings pursuant to its own investigation, in which it heard testimony from Brown and Skinner concerning the OIG Audit Report.

The purpose of the OIG audit was “to determine whether FEMA had sufficient evidence to support [Miami-Dade] county’s eligibility for IHP assistance and whether adequate program controls existed to ensure that funds were provided only to eligible applicants, for eligible expenses.” Id. at 3. Importantly, the audit was limited to (1) only 3% of (2) IHP awards (3) disbursed to residents of Miami-Dade County (4) who claimed damage from Hurricane Frances. In addition, the audit did not “evaluate the controls in NEMIS or the validity and reliability of its data.” Id. at 7. Nor did the audit attempt to determine the extent of potential fraud. Id. Yet despite the limited scope of the audit, the report concluded that the errors it identified were likely systemic:

The policies, procedures, and guidelines used in Miami-Dade County for the IHP were also used throughout the State of Florida, casting doubt about the appropriateness of IHP awards made to individuals and households in other counties of the state as a result of the four hurricanes, particularly those counties that had only marginal damage. Further ..., most of the procedures were used for disasters in other states making the conditions and recommendations broadly applicable to FEMA’s implementation of the IHP nationwide.

Id. at 4.

The report found “shortcomings” at both “key control points” in the IHP award process — the disaster declaration process and the verification of losses reported by individuals. Id. at 3. Specifically, the report found that FEMA designated Miami-Dade County eligible for IHP assistance without a proper preliminary damage assessment, that claims were not properly verified, that guidelines for making awards were generally lacking, that oversight of inspections was deficient, and that funds disbursed were not based on actual losses. The report concluded that while Miami-Dade “residents obviously sustained some degree of damage,” “such damages did not necessarily warrant federal assistance,” and “the inclusion of Miami-Dade County in the amended declaration was questionable.” Id. at 10-11.

The report made sixteen recommendations for improvement. For instance, the report questioned FEMA’s use of the Ge neric Room Concept which, as FEMA itself was forced to admit, almost by definition results in applicants receiving federal funds for items they did not own at the time of the disaster. The report concluded that since this procedure “may permit funding to repair or replace items not damaged or destroyed by a major disaster,” it is “inconsistent with the Stafford Act and is potentially wasteful.” Id. at 13-14. The report was also critical of FEMA inspectors reporting damage based only on the verbal assurances of applicants. Id. at 15 & n. 19. Similarly, the report found that FEMA inspectors failed to document how vehicle damage and losses were disaster-related, and that the generic replacement amount of $6,500 often far exceeded the blue book value of the car. Id. at 16-17. Finally, the report found that thousands of applicants who received money to be used as rental assistance remained in their allegedly unsafe homes. Id. at 25.

Before finalizing its report, the OIG presented FEMA with a draft and allowed FEMA to submit a formal response. 5 FEMA’s response, signed by Brown, began oddly. FEMA expressed its “grati-fi[cation] that the report affirms the absence of widespread or systemic Recovery program fraud, waste or abuse in the state, and conclusively establishes that no special treatment was afforded to Miami-Dade County.” Id. at 43. 6 As noted, however, the report expressly stated that it had not attempted to determine the extent of fraud, and Skinner himself would later object to this characterization of the audit report.

However, FEMA then went on to “take exception to many of the individual conclusions contained” in the report, id., such that the OIG later determined that six of the report’s sixteen recommendations were “unresolved,” meaning that FEMA and the OIG either did not agree that a problem existed, or disagreed on the proper solution. Generally speaking, FEMA rejected any suggestion that Miami-Dade County should not have been included in the Hurricane Frances declaration, calling the OIG’s conclusions to the contrary “inaccurate and misplaced.” Id. at 50.

Although the strongest sustained winds Miami-Dade County experienced were 47 miles per hour — which, as measured by the Saffir-Simpson scale, 7 constitute only tropical storm-force winds — FEMA argued that “the Saffir-Simpson scale is predicated on sustained winds, and does not fully account for the impact of wind gusts that may reach hurricane force, wind-driven rain, and high-velocity tornadic winds that commonly occur in the outer bands of hurricanes.” Id. at 49. Moreover, FEMA said, “the affected areas of Miami-Dade County were predominantly low-income neighborhoods that contained much of the State’s oldest housing stock, and were not built to more recent State and local building codes.” Thus, “homes in Miami-Dade County were far more susceptible to damage.” Id.

FEMA also objected to the OIG’s extrapolation from its Miami-Dade response to conclusions about FEMA’s overall disaster relief procedures and policies. “The scope of the audit is extremely narrow (Miami-Dade County), yet the audit’s conclusions are overly broad.” As a result, FEMA said, “many” of the OIG’s “program-wide conclusions ... are, at best, misleading.” Id. at 45. FEMA argued that “the extraordinary nature of the challenging 2004 hurricane season,” which involved FEMA working at “several times above our standard operating capacity,” made FEMA’s response during this time unique, so that “the OIG expectation of error-free execution and a seamless trail of decision-supporting documentation is both unrealistic and inappropriate.” Id.

At the May 18, 2005 Senate Committee hearings, Director Brown continued to vigorously defend FEMA’s response to Hurricane Frances in Florida as well as its general policies and procedures. He reiterated that he “strongly disagreefd] with any objection to the inclusion of Miami-Dade County in the Hurricane Frances declaration,” calling the OIG’s conclusions to the contrary “inexplicable.” See FEMA’s Response to the 200Jp Florida Hurricanes: A Disaster for Taxpayers?: Hearings Before the S. Comm, on Homeland Sec. & Gov’tl Affairs, 109th Cong. (2005) [hereinafter Senate Hearings] 8 (written statement of Brown at 14). And although Brown conceded that “there was some assistance given incorrectly — perhaps through errors in data entry, inspections, and even through fraudulent claims” — he remained “proud of how few errors have surfaced out of the hundreds of thousands of inspections conducted.” Id. at 11. 9 Brown said it was “clear that many of those early concerns [regarding Miami-Dade County] are misguided,” and blamed the media for causing “[perspective ... to have been lost in the public discussion.” Id. at 8. He warned that “[mjedia portrayals can be dramatic and compelling, but they can also be inaccurate or incomplete. They should not be considered the only starting point for inquiries or reviews of policies and procedures as they often can be, despite good intentions, misleading, misguided, or flawed.” Id. at 12.

In July of 2005, the Senate Committee released its preliminary findings, which largely tracked those of the OIG. See Senate Comm, on Homeland Sec. & Governmental Affairs, Senators Collins & Lieberman Release Findings & Recommendations to Improve Safeguards in FEMA’s Disaster Relief Program (July 10, 2005) [hereinafter Collins & Lieberman Press Release]. 10 Like the OIG, the Committee’s investigation found “serious shortcomings at key stages of FEMA’s program” — specifically, in FEMA’s designation of counties as eligible for relief and in its administration of the IHP — that “allowed taxpayer dollars to be wasted.” Id. The Committee found fourteen problems in FEMA’s administration of the IHP and identified nineteen “[n]ecessary [¡Improvements” to “ensure fairness, accountability, and transparency in the administration of the IHP program.” Id. Like the OIG, the Committee concluded that “[bjecause the policies, procedures, and guidelines used in Florida were for the most part used throughout the nation, we are deeply concerned about the appropriateness of IHP awards nationwide.” Id. at 3.

By June 28, 2005, FEMA had initiated 6,579 recoupment actions to recover more than $27 million as the result of duplicate payments or overpayments to Floridians during 2004. 11

B. Procedural History

While the OIG and the Senate Committee investigated FEMA’s response to the 2004 Florida hurricanes, various Florida media outlets tried to do the same by requesting various FEMA documents under the FOIA, 5 U.S.C. § 552, which requires that “each [federal] agency, upon any request for records ... shall make the records promptly available to any person,” id. § 552(a)(3)(A), subject to certain statutory exemptions, see id. § 552(b).

In News-Press v. United States Department of Homeland Security, No. 2:05-CV-102-FTM-29DNF, 2005 WL 2921952 (M.D.Fla. Nov.4, 2005), three news organi zations — The News-Press and Pensacola News Journal, both divisions of Multimedia Holdings Corporation, and Cape Publications, publisher of Florida Today (collectively, “News”) — submitted several FOIA requests seeking, in pertinent part: (1) NEMIS data pertaining to IHP awards from the four 2004 Florida hurricanes, including applicants’ names and addresses; and (2) spreadsheet data reflecting NFIP claims for the same hurricanes, including addresses of the flooded properties.

In Sum-Sentinel Co. v. United States Department of Homeland Security, 431 F.Supp.2d 1258 (S.D.Fla.2006), the South Florida Sum-Sentinel (“Sun”) requested, in pertinent part, the same NEMIS data from the four 2004 Florida hurricanes, as well as from 27 additional disasters in various states spanning back to 1998, including hurricanes, tropical storms, tornadoes, and wildfires. In both cases, FEMA released voluminous amounts of respon sive data, but withheld nearly 1.3 million IHP applicants’ names and addresses and 33,000 NFIP claimants’ addresses, 12 citing the Privacy Act, 5 U.S.C. § 552a, and FOIA Exemption 6, the combination of which requires agencies to withhold “personnel and medical files and similar files the disclosure of which would constitute a clearly unwarranted invasion of personal privacy.” 5 U.S.C. § 552(b)(6). After exhausting their administrative appeals, News and Sun sued FEMA and its parent agency, the Department of Homeland Security, in the United States District Courts for the Middle and Southern Districts of Florida, respectively, seeking to compel disclosure of the names and addresses. The newspapers conceded that names and addresses constitute “similar files” for purposes of Exemption 6, so the parties’ dispute was limited to whether their disclosure “would constitute a clearly unwarranted invasion of personal privacy.” Following oral arguments on the parties’ cross motions for summary judgment, each district court rendered its decision.

In News, the district court held that disclosure of the names and addresses “would constitute a clearly unwarranted invasion of personal privacy,” and thus that this information was properly withheld under Exemption 6. Balancing what it deemed to be “substantial” privacy interests in withholding the names and addresses against “the extent to which disclosure ... would contribute to the public understanding of the operations and activities of FEMA,” News, 2005 WL 2921952, at *18, the court concluded that the former “substantially outweighs” the latter, id. at *19. The court acknowledged the “significant, legitimate public interest in the activities and operations of FEMA, both with regard to the 2004 Florida Hurricanes and generally,” and held that “knowing whether FEMA has adequately performed its statutory duties is certainly cognizable under FOIA.” Id. at *18. However, the court found that “[disclosure of the names and addresses would say little more about FEMA” beyond what was, or could be, known about FEMA from the documents that FEMA had already released. Id.

About five months later, the district court in Sun reached nearly the opposite conclusion. Although the court conceded that disclosing the addresses “[cjlearly ... raises a substantial privacy interest” by making it possible to link the addresses to the already released personal information, 431 F.Supp.2d at 1268, it nevertheless held that “the release of these addresses ... is uniquely important under the facts of this case,” id. at 1273. The court concluded that “there is a substantial and legitimate public interest in FEMA’s handling of disaster assistance in the wake of recent hurricanes,” id. at 1269, and that “the addresses will provide critical information that is currently lacking from the public debate,” id. at 1270 n. 7. However, the court found that disclosing the names, which the court said would involve the same “significant” privacy interest as disclosing the addresses, would be “clearly unwarranted” because the public interest in the names is only minimal.

News appeals the News court’s decision that the names and addresses of IHP applicants and the addresses of NFIP claimants in the four 2004 Florida hurricanes were properly withheld under Exemption 6. FEMA, in turn, appeals the Sun court’s decision ordering disclosure of the addresses of IHP applicants in the four Florida hurricanes as well as 27 additional disasters nationwide. (Sun does not appeal the district court’s- decision that the names were properly withheld under Exemption 6.) We consolidated the two appeals.

II. Discussion

A. Standard of Review

The parties dispute the proper standard of appellate review in these cases. 13 Sun, which succeeded bn its relevant claims in the district court, urges this Court to review that entire decision deferentially, for clear error only. Both FEMA and News, by contrast, argue that we must review these cases de novo. In even moderately close cases, the standard of review may be dispositive of an appellate court’s decision. And in these cases, where two different district courts reached nearly opposite conclusions, the standard of review would appear to be particularly important: after all, under clear error review, but not de novo review, it would be possible to affirm both courts, yielding the anomalous result that disclosure of the names and addresses would constitute a clearly unwarranted invasion of personal privacy in the Middle District of Florida but not in the Southern District.

However, due to a peculiarity of the FOIA, it is largely — although not wholly — irrelevant to the practical outcome of these cases whether we review the district court decisions de novo or for clear error. Even assuming arguendo that clear error review is the proper standard, we would easily affirm the Sun decision under this standard. And once FEMA is required to disclose records to one member of the public, the FOIA requires it to release the same records to any other citizen who requests them. See, e.g., Nat’l Archives & Records Admin, v. Favish, 541 U.S. 157, 172, 124 S.Ct. 1570, 158 L.Ed.2d 319 (2004) (“As a general rule, if the information is subject to disclosure, it belongs to all.”). Thus, even if we also affirmed News under clear error review, the plaintiffs in that case could simply rely on the decision in Sun to gain access to nearly all of the information they seek. However, only the IHP addresses are at issue in the Sun appeal, whereas News also seeks the IHP names and the NFIP addresses. Therefore, we cannot avoid deciding the proper standard of review in these cases.

This Court has set out two lines of cases governing the standard of review of district court decisions, on summary judgment, in FOIA cases, one providing for clear error review and the other calling for de novo review. In Stephenson v. Internal Revenue Serv., 629 F.2d 1140 (5th Cir.1980), 14 the former Fifth Circuit held in binding precedent that “[a]n appellate court has two duties in reviewing determinations under FOIA. (1) We must determine whether the district court had an adequate factual basis for the decision rendered and (2) whether upon this basis the decision reached was clearly erroneous.” Id. at 1144 (footnote omitted); see also Chilivis v. Sec. & Exch. Comm’n, 673 F.2d 1205, 1210 (11th Cir.1982); Qurrie v. Internal Revenue Serv., 704 F.2d 523, 528 (11th Cir.1983). In each of those cases, there was a factual dispute between the parties as to the very nature of the withheld documents, and thus as to whether they even fell within the applicable exemption. And for the most part, determining the factual nature of the withheld documents was dis-positive of those plaintiffs’ FOIA claims. We therefore reviewed the district courts’ decisions for clear error.

In today’s cases, by sharp contrast, the parties do not dispute the nature of the withheld information, which all agree consists of names and addresses of IHP applicants and NFIP claimants. 15 The parties do not even dispute whether the names and addresses constitute “similar files” for purposes of Exemption 6 — a mixed question of fact and law. In fact, at oral argument, counsel for Sun — the only party that has suggested that the Stephenson standard of review might be applicable here— conceded that there are no factual disputes of any kind in this case. 16

In Cochran v. United States, 770 F.2d 949 (11th Cir.1985), we squarely held that where, as here, “the facts of the case are undisputed and the only issue is the proper balance under FOIA exemption six, the ‘clearly erroneous’ standard employed in Chilivis and Stephenson is inappropriate.” Id. at 956 n. 8 (citations omitted). Although we noted in Cochran that “[w]e need not determine whether the proper standard of review of the district court’s application of the balancing test is de novo or abuse of discretion, since it would have no effect on the result in the present case,” id. at 955-56 n. 8, the abuse of discretion standard was only potentially at issue in that case because the plaintiff had brought a so-called “reverse FOIA” case under the Privacy Act, 5 U.S.C. § 552a. 17 No one argues that abuse of discretion applies to these cases, nor could they. 18 Thus, the only remaining standard of review Cochran leaves open is de novo review. Indeed, in Times Publishing Co. v. United States Department of Commerce, 236 F.3d 1286 (11th Cir.2001), we held, in an Exemption 3 19 case involving “cross-motions for summary judgment in the district court based upon the undisputed record,” that “[w]e review the district court’s grant of summary judgment de novo.” Id. at 1288 & n. 1. Similarly, in Office of the Capital Collateral Counsel v. Department of Justice, 331 F.3d 799 (11th Cir.2003), we again squarely held, citing Cochran and Times, that in an Exemption 6 case, like this one, where the facts were not in dispute, where the plaintiff agreed the information constituted “similar files,” and where the issue on appeal was “limited to the legal application of FOIA exemption 6, [that] the Chili-vis clear error standard does not apply,” and “appellate review is de novo.” Id. at 802 & n. 4. 20

B. The Privacy Act and the Freedom of Information Act

In denying the plaintiffs’ requests for names and addresses, FEMA said it was prevented from disclosing that information by both the Privacy Act, 5 U.S.C. § 552a, and by FOIA Exemption 6. The Privacy Act provides that:

No agency shall disclose any record which is contained in a system of records by any means of communication to any person, or to another agency, except pursuant to a written request by, or with the prior written consent of, the individual to whom the record pertains, unless disclosure of the record would be ... required under section 552 of this title [the FOIA].

5 U.S.C. § 552a(b)(2). The newspapers have admittedly not secured the written permission of the applicants to disclose their names or addresses. Thus, FEMA may not disclose this information unless such disclosure is required by the FOIA.

The FOIA, in turn, generally requires federal agencies to disclose records in their possession upon request, see id. § 552(a)(3)(A), but permits agencies to withhold records if one of several exemptions applies, such as Exemption 6’s exemption for “personnel and medical files and similar files the disclosure of which would constitute a clearly unwarranted invasion of personal privacy,” id. § 552(b)(6). The net effect of the interaction between the two statutes is that where the FOIA requires disclosure, the Privacy Act will not stand in its way, but where the FOIA would permit withholding under an exemption, the Privacy Act makes such withholding mandatory upon the agency. Thus, as both district courts correctly held, the dis-positive question in these cases is whether disclosure of the names or addresses “would constitute a clearly unwarranted invasion of personal privacy” under FOIA Exemption 6. 21 See Cochran, 770 F.2d at 954-55 (explaining the “clearly estab lished” relationship between the Privacy Act and FOIA Exemption 6).

C. The Freedom of Information Act and Exemption 6

The FOIA was expressly intended to avoid the pitfalls of its predecessor, § 3 of the Administrative Procedure Act of 1946, Pub.L. No. 79-404, 5 U.S.C. § 1002 (1964) (“APA”). Although § 3 provided that matters of official record be made available to the public, disclosure was subject to several qualifications. Requesters had to show that they were “properly and directly concerned” with the information, and agencies could in any case refuse to disclose records pertaining to “any function of the United States requiring secrecy in the public interest” or which the agency deemed “confidential for good cause found.” Nor did § 3 provide a remedy for wrongful withholding of records. As the Senate Report to the FOIA said, § 3 was “full of loopholes which allow agencies to deny legitimate information to the public. It has been shown innumerable times that withheld information is often withheld only to cover up embarrassing mistakes or irregularities and justified by [§ 3’s vague exceptions].” S.Rep. No. 88-1219, at 8 (1964). The House similarly commented that “[government agencies whose mistakes cannot bear public scrutiny have found ‘good cause’ for secrecy.” H.R.Rep. No. 89-1497, at 27 (1966), U.S. Code Cong. & Admin. News 1966, pp. 2418, 2423. Thus, § 3 “was generally recognized as falling far short of its disclosure goals and came to be looked upon more as a withholding statute than a disclosure statute.” Envtl. Prot. Agency v. Mink, 410 U.S. 73, 79, 93 S.Ct. 827, 35 L.Ed.2d 119 (1973) (citing S.Rep. No. 89-813, at 5 (1965); H.R.Rep. No. 89-1497, at 5-6, U.S. Code Cong. & Admin. News 1966, at 2422-23).

In supplanting § 3 with the FOIA, Congress created “a broad disclosure statute which evidences a strong public policy in favor of public access to information in the possession of federal agencies.” Cochran, 770 F.2d at 954 (quotation marks omitted). “Without question, the Act is broadly conceived. It seeks to permit access to official information long shielded unnecessarily from public view and attempts to create a judicially enforceable public right to secure such information from possibly unwilling official hands.” Mink, 410 U.S. at 80, 93 S.Ct. 827. “In enacting the FOIA ..., Congress sought to open agency action to the light of public scrutiny. Congress did so by requiring agencies to adhere to ‘a general philosophy of full agency disclosure.’ ” U.S. Dep’t of Justice v. Tax Analysts, 492 U.S. 136, 142, 109 S.Ct. 2841, 106 L.Ed.2d 112 (1989) (quoting S.Rep. No. 89-813, at 3) (other quotation marks and citations omitted). “The basic purpose of FOIA is to ensure an informed citizenry, vital to the functioning of a democratic society, needed to check against corruption and to hold the governors accountable to the governed.” NLRB v. Robbins Tire & Rubber Co., 437 U.S. 214, 242, 98 S.Ct. 2311, 57 L.Ed.2d 159 (1978); see also Nat’l Archives & Records Admin, v. Favish, 541 U.S. 157, 171-72, 124 S.Ct. 1570, 158 L.Ed.2d 319 (2004) (“FOIA is often explained as a means for citizens to know ‘what the Government is up to.’ This phrase should not be dismissed as a convenient formalism. It defines a structural necessity in a real democracy.” (citation omitted)). Fittingly, President Johnson signed the FOIA into law on July 4, 1966, and it became effective on July 4, 1967. Like its overall goal of broad disclosure, the specific

provisions of the Freedom of Information Act stand in sharp relief against those of § 3. The Act eliminates the “properly and directly concerned” test of access, stating repeatedly that official information shall be made available “to the public,” “for public inspection.” ... Aggrieved citizens are given a speedy remedy in district courts, where “the court shall determine the matter de novo and the burden is on the agency to sustain its action.” 5 U.S.C. § 552(a)(3).

Mink, 410 U.S. at 79, 93 S.Ct. 827.

However, “Congress realized that legitimate governmental and private interests could be harmed by release of certain types of information,” Fed. Bureau of Investigation v. Abramson, 456 U.S. 615, 621, 102 S.Ct. 2054, 72 L.Ed.2d 376 (1982), and provided for certain exceptions to the rule of broad disclosure. Nevertheless, “these limited exemptions do not obscure the basic policy that disclosure, not secrecy, is the dominant objective of the Act.” Dep’t of the Air Force v. Rose, 425 U.S. 352, 361, 96 S.Ct. 1592, 48 L.Ed.2d 11 (1976). Because the net effect of the FOIA, with its exemptions, is to “ ‘place[ ] emphasis on the fullest responsible disclosure,’ ” Mink, 410 U.S. at 80, 93 S.Ct. 827 (quoting S.Rep. No. 89-813, at 3), the Supreme Court has “repeatedly stated that the policy of the Act requires that the disclosure requirements be construed broadly, the exemptions narrowly,” Rose, 425 U.S. at 366, 96 S.Ct. 1592 (quotation marks, alteration, and citations omitted); see also Tax Analysts, 492 U.S. at 151, 109 S.Ct. 2841 (“Consistent with the Act’s goal of broad disclosure, these exemptions have been consistently given a narrow compass.”); U.S. Dep’t of Justice v. Julian, 486 U.S. 1, 8, 108 S.Ct. 1606, 100 L.Ed.2d 1 (1988) (“FOIA exemptions are to be narrowly construed.”); Abramson, 456 U.S. at 630, 102 S.Ct. 2054 (same).

D. Application of Exemption 6 to the Withheld Information

We now apply Exemption 6 to each category of withheld information at issue in these appeals—the addresses of the households that received IHP aid, the names of the IHP recipients, and the addresses of households that made flood insurance claims under the NFIP. We do so keeping in mind that FEMA bears the burden of justifying its action whether it withholds entire records or, as here, portions of records. See U.S. Dep’t of State v. Ray, 502 U.S. 164, 173, 112 S.Ct. 541, 116 L.Ed.2d 526 (citing 5 U.S.C. § 552(a)(4)(B)).

1. IHP Addresses

a. Public Interest

The Supreme Court has explained that, “as a general rule, when documents are within FOIA’s disclosure provisions, citizens should not be required to explain why they seek the information. A person requesting the information needs no preconceived idea of the uses the data might serve.” Favish, 541 U.S. at 172, 124 S.Ct. 1570. “When disclosure touches upon certain areas defined in the exemptions, however, the statute recognizes limitations that compete with the general interest in disclosure, and that, in appropriate cases, can overcome it .... To effect this balance and to give practical meaning to the exemption, the usual rule that the citizen need not offer a reason for requesting the information must be inapplicable.” Id. Instead, the requester must indicate how disclosing the withheld information “would serve the core purpose of the FOIA, which is contributing significantly to public understanding of the operations or activities of the government.” U.S. Dep’t of Def. v. Fed. Labor Relations Auth., 510 U.S. 487, 495, 114 S.Ct. 1006, 127 L.Ed.2d 325 (1994) (quotation marks and emphasis omitted).

Here, the newspapers say that the public has a powerful interest in knowing whether FEMA appropriately handled disaster relief claims, and that the addresses where alleged damage occurred are necessary to determine whether aid was in fact disbursed to those who suffered disaster-related damage. We easily conclude, as did both district courts, that the asserted interest in learning whether FEMA is a good steward of (sometimes several billions of) taxpayer dollars in the wake of natural and other disasters is one which goes to “the core purpose of the FOIA, which is contributing significantly to public understanding of the operations or activities of the government.” Id. at 495, 114 S.Ct. 1006; see also U.S. Dep’t of Justice v. Reporters Comm. for Freedom of the Press, 489 U.S. 749, 773, 109 S.Ct. 1468, 103 L.Ed.2d 774 (1989) (“[FOIA] focuses on the citizens’ right to be informed about what their government is up to. Official information that sheds light on an agency’s performance of its statutory duties falls squarely within that statutory purpose.” (quotation marks omitted)). Moreover, although courts “generally accord Government records and official conduct a presumption of legitimacy,” Ray, 502 U.S. at 179, 112 S.Ct. 541, the newspapers have put forth ample evidence that FEMA’s response to Hurricane Frances in Miami-Dade County may have been plagued with fraud, waste, or abuse.

The newspapers have identified a substantial public interest cognizable under FOIA that would be served by disclosing the addresses. FEMA now bears the “burden of demonstrating that the disclosure of the [IHP data already released] adequately served the statutory purpose and that the release of the information identifying the particular [IHP addresses] would constitute a clearly unwarranted invasion of ... privacy.” Id. at 175, 112 S.Ct. 541. FEMA says, and the district court in News agreed, that disclosing the addresses would not serve the admittedly legitimate interest of evaluating the adequacy of FEMA’s aid disbursement under the IHP. The district court held that “[r]elease of the ... address[es] will shed little additional light on FEMA’s conduct ____ [Disclosure of the addresses will say something about FEMA, at least to the extent they reveal the addresses at which physical damage was claimed to have occurred. Nonetheless, ... this carries little weight in the overall scheme of things in the balancing process.” News, 2005 WL 2921952, at *18. The district court held that the most important public benefit to come from the addresses would also “tip[ ] the scale towards the privacy invasion side.” Id. The newspapers indicated that they might have occasion to attempt to contact some of the aid recipients in order to learn more about FEMA’s operations. The district court determined that any new information thereby learned about FEMA’s disaster response was more than offset by the privacy invasion that would result. Id. at *18-19. We disagree.

In focusing on the possibility of learning more about FEMA through aid recipient interviews, the district court gave inadequate weight to the substantial light that would be shed on FEMA’s activities directly from the release of the addresses. As the newspapers have explained, they plan to superimpose the path of each of 31 disasters on a street-level map showing the specific households where damage was alleged to have occurred, and where FEMA dollars were disbursed. Any “outliers”— homes outside the path of the disaster that nevertheless received FEMA aid — plainly would raise red flags. It is true that the newspapers have indicated that, depending on their initial findings, they may attempt to contact the residents of some outlier homes. But it is worth noting that, as Acting Inspector General Skinner opined, the inspection data was often so vague and conelusory that in order to verify the appropriateness of an award, the OIG itself was often forced to interview recipients. See Senate Hearings (oral testimony of Skinner at 50). 22 In any case, we consider the privacy implications of contacting IHP aid recipients below. For now it is enough to conclude that the red flags raised by the release of the addresses (or the absence of such flags) themselves constitute valuable information.

FEMA responds that the addresses are not necessary to determine whether FEMA has been a good steward, for two reasons. First, FEMA says that the OIG and the Senate have already undertaken thorough investigations into FEMA’s method of IHP aid disbursement. Second, FEMA says that the IHP data it has already released, broken down by zip code, is sufficient. Neither argument is persuasive.

As for the previous investigations, FEMA rejected several of the OIG’s findings, and disputed the significance of the OIG audit report. FEMA claimed that the audit “affirms the absence of widespread or systemic Recovery program fraud, waste or abuse in the state, and conclusively establishes that no special treatment was afforded to Miami-Dade County.” OIG Audit Report at 43. But Skinner testified that this was “not at all” a fair reflection of the audit’s conclusions. Senate Hearings (oral testimony of Skinner at 36). He noted that the audit did indeed reveal “some very serious systemic weaknesses” in the IHP; that “the purpose of the audit was not to identify fraud, waste, and abuse per se”; that the investigation remains ongoing and it was “premature” to conclude that there was no widespread fraud, waste, or abuse; and that much fraud is de minimis and is thus difficult to prosecute. Id. at 36-37.

Moreover, in the same breath that it declared that the OIG report “eonfirm[ed] the fundamental soundness of FEMA’s time-tested policies, procedures, and guidelines,” OIG Audit Report at 48, FEMA “[took] exception to many of the individual conclusions contained” in the OIG report, id. at 43, and six of the OIG’s recommendations remained unresolved. FEMA cannot fairly claim here that the OIG and Senate investigations conclusively determined the extent of FEMA’s problems, especially when it vigorously disputed much of those investigations’ findings before Congress and in its press releases to the public.

But even if FEMA agreed with the OIG and the Senate about the meaning of the findings, the newspapers seek to evaluate FEMA’s conduct going well beyond FEMA’s response to Hurricane Frances in Miami-Dade County. As the Sun court recognized, and as FEMA itself has emphasized to Congress, “[t]he scope of the OIG audit is extremely narrow.” Id. at 45 (emphasis added). Again, that audit was limited to only 3% of IHP aid recipients in one Florida county following one hurricane, and the Senate investigation appears to have largely tracked the scope of the OIG audit. By comparison, News seeks to investigate FEMA’s response across Flori da for all four 2004 hurricanes, and Sun seeks to investigate an additional 27 disasters. The results of the OIG and Senate investigations thus cannot be said to have resolved the question of whether FEMA appropriately disbursed aid in other areas, following other disasters.

Nor can the public simply extrapolate from the OIG and Senate investigations conclusions about the appropriateness of FEMA’s disaster response in general. Although both the OIG and the Senate Committee expressed grave concerns that the problems they uncovered in FEMA’s Miami-Dade response were not limited to that disaster, FEMA protested that the OIG report reached “overly broad” conclusions based on an “extremely narrow” inquiry, and thus that those conclusions were, “at best, misleading.” Id. Brown told the Senate Committee that “[t]he extrapolation of things that were found in Miami-Dade to other areas of the state, particularly areas of the state that were particularly hard hit, I think does draw incorrect ... conclusions.” Senate Hearings (oral testimony of Brown at 65). He explained that in hard-hit areas where damage is severe and thus obvious, the inspectors’ job of verifying damage is relatively easy. “It’s more difficult for an inspector to make a determination of what has really occurred in those marginal areas .... Particularly when you’re making those discerning kinds of judgments in housing stock that is old and decrepit, and ... is certainly substandard.” Id.

Brown also argued that it would be inappropriate to conclude, from the fact that in Miami-Dade the Generic Room Concept yielded awards for personal property that the recipient never owned, that this method similarly overcompensates victims in other areas. “[I]n most areas, it is safe to assume that in the destroyed home that you see, that is, the typical middle-class home, it’s easy to make the assumption that, yes, there is that property in the structure.” Id. at 66. After FEMA so vigorously disputed the prior investigations’ findings and urged that no extrapolations be made from them, we are not disposed to hear FEMA tell us that these same investigations are sufficient to allow the public to satisfy its interest in knowing whether, in general, FEMA appropriately distributes disaster relief.

For similar reasons, we reject FEMA’s second argument that the addresses are unnecessary to evaluate FEMA’s performance of its statutory duties. FEMA says that the IHP aid disbursement data it has already provided the newspapers, which is broken down by zip code, is sufficient. It is important to recall, however, that Sun requested IHP data not only from hurricanes, tropical storms, and wildfires, but also from tornados, which routinely destroy one home while leaving the home next to it intact. FEMA does not begin to explain why disclosing the zip code in which an IHP recipient resided would be sufficient to assess the appropriateness of that disbursement in such a context.

As for IHP data from hurricanes and similar disasters, the newspapers and the district court in Sun say that zip codes, which can cover a large piece of geography in less densely populated areas, are still too indiscriminate an area to allow the newspapers to determine whether it was appropriate for FEMA to have disbursed aid to any given home within a zip code. Indeed, they point out that even in smaller zip codes, houses on one street may be damaged — say because they border a river and suffer flood damage — while houses on the next street over may suffer little or no damage. They find support for these contentions from an unlikely source: FEMA itself. When the media first began raising questions about the appropriateness of FEMA aid to Miami-Dade County, since the path of the storm fell some 100 miles to the north, top FEMA officials criticized the media, and emphasized that the existence of damage cannot always be inferred from the path and strength of the storm, but that each home must instead be evaluated independently.

Thus, for example, FEMA Director of Recovery Daniel Craig issued a press release stressing that the appropriate amount of aid can be assessed only on a home-by-home basis:

While each application is subject to the same eligibility criteria, it’s important to caution against making comparisons of damage assistance provided across individuals, communities or counties. There are many factors that determine whether a citizen is eligible for FEMA assistance, and two homes next to each other may have different eligibility because of these factors. Did they have insurance? Were they under-insured? Was there wind damage or flood damage? Did the roof leak or basement flood? Is the applicant a renter or homeowner? Do they live in a mobile home? Is it a primary residence? Did the applicant lose electricity? Do they have special medical needs? All of these factors make a difference. In a disaster, every state, every county and every home is different.

Prepared Statement from FEMA Director of Recovery Daniel Craig Regarding Florida Disaster Assistance 2 (Jan. 10, 2005) (emphasis added). Brown similarly specifically blamed the existence of “faulty results and incorrect conclusions” on

[ejarly press reports that engaged in county-by-county comparisons of total outlays .... In addition to levels of damage, many factors influence the distribution of IHP assistance, including the population, the proportion of insured applicants in counties affected by disasters, and income levels .... [Sjtrict comparisons of totals between counties, as opposed to individuals, does not take into consideration the multitude of other factors, such as insurance and income levels, which can preclude registrants from receiving FEMA aid.

Senate Hearings (written statement of Brown at 8-9) (emphasis added).

Similarly, when the Senate Committee questioned Brown about his decision to require inspection companies to perform twice the number of daily inspections as they were required to perform under their contract, thus forcing those companies to hire new, untrained inspectors, Brown defended that decision this way:

I can do one of two things. I can either stop all inspections, such as was done in the 1994 Northridge earthquake, and just pay money out based on zip codes, or I can ramp up, work with the contractors, do everything I can trying to be a good steward of the taxpayer dollars and get eyes on every claim. My objective was to get eyes on every claim made, and not pay things out by zip code. So when you’re doing 885,000 inspections, there are going to be errors. I want to clean those up. But I still believe I made the right decision in terms of the taxpayers and the disaster victims of continuing to get aid out to them, but not do it on a blanket basis, like was done in 1994, or a zip code basis.

Id. at 78 (emphasis added). “Again,” Brown reiterated, “the choice was ramp up the inspections, try to get as many out there so I’ve got eyes on every claim, or just do the blanket zip code. I refuse to do the latter.” Id. at 81-82 (emphasis added). The Committee “agree[d] with [Brown] that [he] made the right decision in not doing the zip code approach.” Id. at 82 (statement of Sen. Collins).

If it would not constitute good stewardship of taxpayer dollars simply to make decisions about disaster aid based on zip code, then neither can zip codes be seen as an altogether accurate or complete way for the public to evaluate FEMA’s distribution of aid. We note also that in Sun, the district court asked whether it would satisfy Sun’s needs if FEMA provided the locations by something more specific than a zip code but less specific than a street address — say, a nine-digit zip code, which essentially narrows down a location by street, though not by house. Sun’s counsel responded that she believed that FEMA was incapable of breaking down the information by anything other than zip codes or street addresses. If this was incorrect, FEMA’s counsel did not correct the record. See Sun, Tr. at 52. Given FEMA’s own vigorous arguments regarding the inappropriateness of making aid decisions by zip code and the need to consider individual, house-by-house factors, we conclude that faced with a choice of disclosing the aid information by zip code or by street address, the latter must prevail.

In short, the public interest in determining whether FEMA has been a proper steward of billions of taxpayer dollars is undeniable and powerful. FEMA’s responses to the various investigations of its disbursement in Miami-Dade county following Hurricane Frances have produced more questions than answers. The addresses, however, will go a long way in resolving the factual disputes that exist between FEMA, on the one hand, and the OIG and the Senate Committee, on the other. Thus, we readily find that the addresses would further a powerful public interest.

b. Privacy Interests

Having concluded that the addresses would serve a substantial and legitimate public interest cognizable under the FOIA, we turn to the countervailing privacy interests that FEMA has identified to determine whether FEMA has met its burden of demonstrating that these privacy interests are so weighty that, despite the substantial public interest involved, disclosing the addresses “would constitute a clearly unwarranted invasion of personal privacy.”

“Congresses] primary purpose in enacting Exemption 6 was to protect individuals from the injury and embarrassment that can result from the unnecessary disclosure of personal information.” U.S. Dep’t of State v. Wash. Post Co., 456 U.S. 595, 599, 102 S.Ct. 1957, 72 L.Ed.2d 358 (1982) (emphasis added). “[I]t is quite clear from the committee reports that the primary concern of Congress in drafting Exemption 6 was to provide for the confidentiality of personal matters in such files as those maintained by the Department of Health, Education, and Welfare, the Selective Service, and the Veterans’ Administration.” Rose, 425 U.S. at 375 n. 14, 96 S.Ct. 1592 (citing S.Rep. No. 89-813, at 9 (1965); H.R.Rep. No. 89-1497, at 11 (1966), U.S. Code Cong. & Admin. News 1966, at 2428) (emphasis added) (partially quoted in Wash. Post Co., 456 U.S. at 599, 102 S.Ct. 1957).

To achieve this end, Congress established, in Exemption 6, the following two-tier test: “personal information in government agency files is exempt from mandatory disclosure only if: (1) the information was within personnel, medical, or similar files; and (2) a balancing of individual privacy interests against the public interest in disclosure reveals that disclosure of the information” would constitute a clearly unwarranted invasion of personal privacy. Cochran, 770 F.2d at 955. The Supreme Court has made clear that “similar files” has a “broad, rather than a narrow, meaning,” Wash. Post Co., 456 U.S. at 600, 102 S.Ct. 1957, and includes any “detailed Government records on an individual which can be identified as applying to that individual,” id. at 602, 102 S.Ct. 1957 (quoting H.R.Rep. No. 89-1497, at 11, U.S. Code Cong. & Admin. News 1966, at 2428). The records at issue here fall within this broad definition, as the newspapers have conceded.

Instead, the crux of Exemption 6 is its second prong, which asks whether disclosure “would constitute a clearly unwarranted invasion of personal privacy.” See id. at 600; S.Rep. No. 89-813, at 9 (“[T]he scope of the exemption is held within bounds by the use of the limitation of ‘a clearly unwarranted invasion of personal privacy.’ ”); H.R.Rep. No. 89-1497, at 11, U.S. Code Cong. & Admin. News 1966, at 2425 (same). Courts reviewing the legislative history of the FOIA have concluded that Congress’s use of the “clearly unwarranted” language “was a considered and significant determination,” Rose, 425 U.S. at 378 n. 16, 96 S.Ct. 1592, and “the expression of a carefully considered congressional policy favoring disclosure,” Getman v. NLRB, 450 F.2d 670, 674 n. 11 (D.C.Cir.1971). In fact, during hearings on the bill, various agencies strenuously urged deletion of either the modifier “clearly” or the entire phrase “clearly unwarranted,” so that agencies would have been permitted to withhold information whenever its disclosure would result in any invasion of privacy. See Hearings on S. 1160 Before the Subcomm. on Admin. Practice & Procedure of the Senate Comm, on the Judiciary, 89th Cong. 36 (1965) (statement of Edwin Rains, Assistant Gen. Counsel, Treasury Dep’t); id. at 491 (statement of William Feldesman, NLRB Solicitor); Hearings on H.R. 5012 before a Subcomm. of the House Comm, on Gov’t Operations, 89th Cong. 56, 230 (1965) (statement of Fred Burton Smith, Acting Gen. Counsel, Treasury Dep’t); id. at 257 (testimony of William Feldesman, NLRB Solicitor); see also Hearings on S. 1160, at 417 (testimony of the Gen. Counsel, Dep’t of Def.) (objecting to “heavy” burden of showing a clearly unwarranted invasion of personal privacy); cf. Hearings on H.R. 5012, at 151 (testimony of Clark R. Molenhoff, Vice Chairman, Sigma Delta Chi Comm, for Advancement of Freedom of Info.) (urging retention of “clearly unwarranted”). As the Supreme Court noted, however, “[t]he terms objected to were nevertheless retained, as a ‘proper balance,’ H.R.Rep. No. 1497, p. 11, U.S. Code Cong. & Admin. News 1966, at 2428, to keep the ‘scope of the exemption ... within bounds,’ S.Rep. No. 813, p. 9.” Rose, 425 U.S. at 378 n. 16, 96 S.Ct. 1592.

Moreover, this legislative history “stands in marked contrast” to the record surrounding Exemption 7(C), which covers investigatory files compiled for law enforcement purposes. Id. As the Supreme Court has explained, Exemption 7(C) was once drafted to match Exemption 6: agencies had to show that disclosure of law enforcement records “would” constitute a “clearly unwarranted” invasion of personal privacy:

Exemption 7 was amended to exempt investigatory files compiled for law enforcement purposes only to the extent that their production would “constitute a clearly unwarranted invasion of personal privacy” or meet one of several other conditions. In response to a Presidential request to delete “clearly unwarranted” from the amendment in the interests of personal privacy, the Conference Committee dropped the “clearly,” and the bill was enacted as reported by the conference committee.

Id. (citations omitted). On October 27, 1986, Exemption 7 was again amended as part of the bipartisan Anti-Drug Abuse Act of 1986. Congress further reduced the burden on agencies withholding records under this exemption by replacing the onerous requirement that they show that disclosure “would” constitute an unwarranted invasion of personal privacy with the far lesser burden of showing that disclosure “could reasonably be expected to” do so. These two differences between the exemptions, and Exemption 6’s comparative narrowness, are thus “no mere accident in drafting.” Nat'l Archives & Records Admin, v. Favish, 541 U.S. 157, 165-66, 124 S.Ct. 1570, 158 L.Ed.2d 319 (2004); see also U.S. Dep’t of Justice v. Reporters’ Comm, for Freedom of the Press, 489 U.S. 749, 756 n. 9, 109 S.Ct. 1468, 103 L.Ed.2d 774 (1989) (“[T]he move from the “would constitute’ standard to the ‘could reasonably be expected to constitute’ standard represents a considered congressional effort ‘to ease considerably a Federal law enforcement agency’s burden in invoking [Exemption 7].’” (quoting 132 Cong. Rec. 31425 (1986))). What Congress was willing to yield with respect to Exemption 7 it has steadfastly refused to yield as to Exemption 6.

The federal courts, including this one, have therefore generally concluded that an agency’s burden under Exemption 6 of showing that disclosure “would constitute a clearly unwarranted invasion of personal privacy” is an onerous one. See, e.g., Cochran, 770 F.2d at 955 (“If the balance [between an individual’s right to privacy and the public’s right to know] is equal the court should tilt the balance in favor of disclosure.”); Stern v. Fed. Bureau of Investigation, 737 F.2d 84, 91 (D.C.Cir.1984) (Exemption 6’s language “requirefs] a balance tilted emphatically in favor of disclosure”); Kurzon v. Dep’t of Health & Human Servs., 649 F.2d 65, 67 (1st Cir.1981) (“By restricting the reach of exemption 6 to cases where the invasion of privacy ... is not only unwarranted but clearly so, Congress has erected an imposing barrier to nondisclosure under this exemption.” (citing K. Davis, Administrative Law Treatise § 5:38 (2d ed.1978))).

The district court in News held that disclosure of the addresses where disaster damage was alleged to have occurred “would constitute a clearly unwarranted invasion of personal privacy.” We remain unpersuaded. As a threshold matter, the legislative histories behind the FOIA and the Privacy Act show that Congress did not intend either names or addresses to automatically be withheld, even when they could be linked with other information about those individuals. Between 1973 and 1977, numerous bills were introduced that would have amended the FOIA (or established an independent law) by either prohibiting or limiting the sale or distribution by federal agencies of lists of names and addresses, including names and addresses of individuals registered with, or required to provide information to, an agency. Agencies would have been permitted to release such lists only if specifically authorized to do so by statute or by their statutory function, or if the recipient certified that it would not use the list for commercial or other solicitation. See H.R.s 855, 889, 1779, 2578, 3995, 4468, 5434, 6838, 6839, 6840, and 8086, 93d Cong. (1973); H.R. 12558, 93d Cong. (1974); H.R.s 662, 721, 869, and 1464, 94th Cong. (1975); H.R. 1048, 95th Cong. (1977). None of these bills survived committee. Moreover, the Privacy Act prohibits federal agencies from selling or renting an individual’s name and address, but specifically cautions that this provision “shall not be construed to require the withholding of names and addresses otherwise permitted to be made public.” 5 U.S.C. § 552a(n).

Similarly, the federal courts have held that while names and addresses qualify as potentially protectable “similar files” under Exemption 6, the release of a list of names and other identifying information does not inherently and always constitute a “clearly unwarranted” invasion of personal privacy. Instead, “whether disclosure of a list of names is a significant or a de minimis threat depends upon the characteristic(s) revealed by virtue of being on the particular list, and the consequences likely to ensue.” Ray, 502 U.S. at 176 n. 12, 112 S.Ct. 541 (internal quotation marks omitted).

The district court in News gave five reasons why the disclosure of these particular addresses would be clearly unwarranted. First, the court observed that release of the addresses “will enable others to link the great deal of highly personal information already disclosed by FEMA to particular individuals.” 2005 WL 2921952, at *16. As both district courts accurately acknowledged, once addresses are disclosed, it would be fairly simple for anyone so inclined to determine, through public records, the residents of those addresses at the time of the disasters. Those individuals could then be linked to the information FEMA has already disclosed about their IHP awards.

This makes it important to understand precisely what information FEMA has already released about IHP applicants that could be linked to them if their addresses are disclosed. In News, FEMA introduced into the record five pages of NEM-IS data constituting what FEMA referred to as a “representative sample” of the IHP information FEMA has already released. For each entry, the spreadsheet includes the following fields: a disaster number assigned to the relevant hurricane or other disaster; a nine-digit registration ID assigned to the applicant; damage type, which in the sample was either “Hail/Rain/ Wind Driven Rain,” “Tornado/Wind,” or, in one case, “Other”; item description, which in each case in the sample was “Clothing”; and item quantity, which ranged .in the sample from one to seven. There is also a category that appears to indicate the level of damage of each item, which in every case in the sample was “Replace.” Another category appears to indicate whether each item was covered by insurance; in the sample, the ratio was about seven “Uninsured” items to every “Insured” item of clothing. A final category apparently indicates the amount of the FEMA award for that item; each item 23 of clothing in the representative sample was assigned a generic amount — either $822.23, $822.70, $833.5, $844.81 or $853.99 — and those found to have had multiple items of clothing damaged received multiples of one of these amounts.

FEMA says that it provided News and Sun with NEMIS data broken down by individual applicant, including disaster number and registration ID, as well as the following inspection information: line item description, quantity recorded, insurance status for line item, damage level, item amount, and damage type. These categories seem to be reflected in the “representative sample” FEMA submitted into evidence. However, FEMA says that it released the following additional information about each IHP recipient: zip code, county, category of assistance, assistance status, assistance type, assistance status detail, eligibility date, approved date, eligible amount, determination type, and ownership status. And FEMA says that it released the following additional inspection information: Small Business Loan (“SBA”) status, 24 water level, cause of damage, personal property description, clothing, miscellaneous item description, generic room, essential tool item description, and real property damage item description.

Assuming FEMA did release this greater number of categories of data, the only categories that bear any remote resemblance to information one might find in a medical or personnel record are “SBA status,” “ownership status” (presumably referring to whether the IHP recipient owns or rents her home), and “insurance status for line item.” We discuss the implications of this information below, in our consideration of stigma. For now, it is enough to observe that there is no evidence in the record to support the proposition, suggested in FEMA’s briefs, that detailed descriptions of applicants’ personal property have been released. As the newspapers have pointed out, because FEMA awards money for repair and replacement of damaged personal property based on predetermined, generic amounts, as the sample data indicates, property is described in the broadest and most generic of terms— “clothing,” not “Gucci heels” or “Keds”; “television,” not “high definition plasma” or “black and white set from 1974” — and is assigned an equally generic monetary value that cannot indicate anything about the specific kind or quality of property the IHP recipient once possessed. Indeed, counsel for FEMA conceded as much at oral argument.

Second, the district court in News held that the individuals would suffer “public embarrassment or stigma” as recipients of government assistance. Id. at *16. But, as FEMA’s counsel conceded at oral argument, there is no “means test” for receiving IHP aid from FEMA. That is, unlike many government benefits programs, such as welfare, Medicaid, and unemployment, one need not fall below a certain annual income level to qualify for disaster assistance. Indeed, outside the context of this litigation, FEMA has gone to some lengths to disabuse citizens of the notion that FEMA aid is a type of welfare. See, e.g., FEMA, Common Misunderstandings May Cause Some Victims To Miss Disaster Assistance (Sept. 27, 2004) [hereinafter Common Misunderstandings Press Release ] (It is “[n]ot [t]rue” that applicants “have to be poor to qualify for disaster assistance ____ Federal and state disaster assistance programs may be available to those who suffered damage, regardless of income. The programs are not ‘welfare.’ The kinds of help provided depend on the applicant’s circumstances and unmet disaster-related needs.”). 25

However, both an IHP applicant’s insurance status and her SBA status may be relevant, although not dispositive, in determining whether she will receive funds from FEMA. As for SBA status, it does not appear that this status is expressly indicated in the NEMIS data. Instead, FEMA argues that since IHP recipients must necessarily have been turned down for an SBA loan, identification as an IHP recipient is tantamount to identification as having been turned down for an SBA loan which, in turn, suggests the individual is poor or has bad credit. Our response to this argument is twofold. First, those applying for Housing Assistance (“HA”) un der the IHP need not apply for an SBA loan, and it is a violation of the Stafford Act, 42 U.S.C. 5174(a)(2), for FEMA to require otherwise. See McWaters v. FEMA 408 F.Supp.2d 221, 232 (E.D.La.2005) (finding that FEMA, through “mis-communication or inartful communication,” caused some applicants to believe that an SBA loan application is a necessary prerequisite to receiving any HA aid and granting a preliminary injunction preventing FEMA from continuing to communicate the same); McWaters v. FEMA 436 F.Supp.2d 802 (making preliminary injunction permanent upon finding that FEMA violated the court’s prior order by issuing a February 13, 2006 press release with the “confusing and incorrect” headline “SBA Loan Application Necessary for Assistance”). 26

Second, while it is true that an Other Needs Assistance (“ONA”) applicant must first apply for an SBA loan, the applicant will be eligible for IHP aid either if she is denied an SBA loan or if she claims that that loan does not meet all of her covered needs and expenses. See 44 C.F.R. § 206.119(a). Even in the former case, FEMA has elsewhere stressed that being turned down for an SBA loan is not tantamount to being turned down for a regular bank loan. According to FEMA, it is “[n]ot [t]rue” that applicants “have to be turned down by [their] bank before [they] can apply for a disaster loan,” as the SBA “has its own criteria for determining each loan applicant’s eligibility.” Common Misunderstandings Press Release. Because a recipient of ONA may have been denied an SBA loan or may simply have represented to FEMA (accurately or not) that ,such a loan was insufficient to cover her needs, being identified as an ONA aid recipient is not tantamount to being identified as one who was turned down for an SBA loan, much less one who was turned down for a regular bank loan.

As for insurance, FEMA’s regulations allow insured individuals to receive IHP aid if their claim is denied or if the insurance proceeds are less than the maximum amount of assistance FEMA can provide 27 and are insufficient to meet the applicant’s covered needs. 28 See 44 C.F.R. § 206.113(a)(2), (4). Insured individuals may also receive IHP aid if their insurance proceeds are delayed, subject to the individual’s obligation to repay such aid if she later receives it from her insurance company. See id. § 206.113(a)(3). In fact, in response to the OIG’s finding that insured applicants were awarded financial assistance even in cases where FEMA regulations prohibited it, see OIG Audit Report at 23-24, FEMA said that “over 20 years of experience in previous disasters” suggested “that these multiple, back-to-back storms would cause additional delays in aid delivery to the public, not just from FEMA and other Federal agencies, but also from State and local authorities, private insurers, and voluntary agencies,” id. at 46. Thus, FEMA said, in providing IHP aid to insured applicants, it had simply responded to “credible indications that area residents were likely to face assis tance delays due to multiple, back-to-back storms, delayed insurance settlements, and the limited number of insurance adjusters and available building contractors.” Id. at 55-56. Indeed, FEMA has elsewhere highlighted that insured individuals are eligible for FEMA assistance: “Insurance is your main source for money to put your life back in order after a disaster. But there are many things that insurance does not cover. That is where federal disaster programs may be able to help.” Common Misunderstandings Press Release.

If the addresses are released, some IHP recipients may be identifiable as having lacked insurance to cover a specific damaged item, such as clothing. And although the record is not clear, it also appears that some individuals may be identifiable as home renters rather than owners. We acknowledge that some IHP recipients may feel some stigma if these facts become known to others. 29 However, the “ [legislative history of [Exemption 6] disfavors privacy claims by those who receive a governmental benefit.” 2 James T. O’Reilly, Federal Information Disclosure § 16:53 (3d ed.2000). The Senate Report accompanying the FOIA expressly stated that “health, welfare, and selective service records are highly personal to the person involved, yet facts concerning the award of a pension or benefit should be disclosed to the public.” S.Rep. No. 89-813, at 9 (1965) (emphasis added). The House Report similarly observed that Exemption 6 was “intended to cover detailed Government records on an individual which can be identified as applying to that individual and not the facts concerning the award of a pension or benefit or the compilation of unidentified statistical information from personal records.” H.R.Rep. No. 89-1497, at 11, U.S. Code Cong. & Admin. News 1966, at 2428 (1966) (emphasis added). Although the House Report also said that “[t]he public has a need to know, for example, the details of an agency opinion or statement of policy on an income tax matter, but there is no need to identify the individuals involved in a tax matter if the identification has no bearing or effect on the general public,” id. at 8 (emphasis added), these addresses do have a bearing — and a crucial one at that — on the public’s ability to assess FEMA’s performance of its statutory duties. Cf. Reporters’ Comm., 489 U.S. at 773-74, 109 S.Ct. 1468 (“The deletions were unquestionably appropriate because the names of the particular cadets were irrelevant to the inquiry into the way the Air Force Academy administered its Honor Code; leaving the identifying material in the summaries would therefore have been a ‘clearly unwarranted’ invasion of individual privacy.” (discussing Rose, 425 U.S. 352, 96 S.Ct. 1592, 48 L.Ed.2d 11)).

In any case, FEMA’s counsel conceded at oral argument that it is “essentially right” that the already-released IHP data does not even remotely resemble the kind of information Congress intended Exemption 6 to protect — that is, “personal information” whose release would cause “injury and embarrassment.” Wash. Post Co., 456 U.S. at 599, 102 S.Ct. 1957. Indeed, counsel for FEMA conceded that FEMA does not regard stigma as a “crucial” factor on the privacy side of the calculus at all.

Third, the district court in News determined that disclosure would create “a reasonable danger of identity theft, not to mention actual theft.” 2005 WL 2921952, at *17. As we noted earlier, however, there is no evidence whatsoever in the record to support this conclusion as to identity theft, as FEMA’s counsel has conceded.

As for “actual theft,” the district court worried that if thieves knew that residents of a certain address received a generic amount of money to replace, say, a television, they would find these locations profitable to burgle. But, IHP aid recipients are not required to spend money they receive replacing a particular item; they may well invest that money elsewhere or simply put it in the bank. Moreover, it has been several years since these individuals applied for FEMA aid; the 2004 Florida hurricane season is nearly three years old, and some of the other disasters implicated in Sun’s FOIA request are nearly ten years old. The “new” items the News court was concerned about are almost certainly no longer tempting to thieves, if they ever were. In short, we think that thieves will not find the IHP addresses to be useful at all.

Fourth, the court held that News’s intention to make direct contact with at least some recipients “magnifies the importance of the personal privacy interest at stake” by rendering the individuals “the target of unsolicited and perhaps unwanted contact.” Id. However, individuals are under no obligation to speak to reporters, and on balance, the modest annoyance of a “no comment” is simply the price we pay for living in a society marked by freedom of information laws, freedom of the press, and publicly-funded disaster assistance.

FEMA also argues that IHP recipients had a reasonable expectation of privacy in the information they provided to the agency, since they are provided with “FEMA’s privacy policy, which explains that the Privacy Act governs the disclosure of individually identifiable information of this sort.” In the first place, FEMA does not have the power to promise that it will not disclose that which the FOIA requires it to disclose. But in fact, FEMA’s privacy policy 30 cautions that an applicant’s information “may be shared with [her] bank, insurance company, or other assistance providers to ensure there is no duplication of benefits,” as well as with “state and local governmental agencies to help reduce future disaster losses,” and nowhere promises that FEMA will not disclose this information to still other sources. Moreover, the policy refers applicants to FEMA’s Individual Assistance Privacy Impact Assessment, 31 which states (at page 2) that the individual assistance data is subject to a variety of “legal requirements,” including the FOIA. Finally, as FEMA’s Privacy Impact Assessment for the EP & R/FEMA Privacy Act “Disaster Recovery Assistance Files” System of Records 32 indicates (at § 2.1), “[i]ndividu-als always have the right not to apply for Federal disaster assistance.”

Finally, the district court in News held that since FEMA must make available to any requestor what it makes available to the newspapers, disclosure would allow “commercial advertisers or solicitors” to bother the disaster victims with offers of “special goods, services, and causes likely to appeal to” them. Id. Again, because it has been several years since the disasters, it seems unlikely that building contractors and the like would find soliciting these individuals very profitable, and there is no record evidence to support this supposition. In any case, to the extent that IHP aid recipients experience slightly more commercial solicitation than the average American, this, too, is a modest intrusion, at most. 33

In order to affirm withholding the addresses, we would have to find that the privacy interests against disclosure are greater than the public interest in disclosure. See Cochran, 770 F.2d at 955 (“If the balance is equal the court should tilt the balance in favor of disclosure”). This we cannot do. Quite simply, the disclosure of the addresses serves a powerful public interest, and the privacy interests extant cannot be said even to rival this public interest, let alone exceed it, so that disclosure would constitute a “clearly unwarranted” invasion of personal privacy. On this record we do not find the balancing calculus to be particularly hard.

2. Names of IHP Award Recipients

News also appeals the district court’s decision not to require FEMA to disclose the names of IHP aid recipients. Although we conclude that FEMA must disclose the addresses where damage was alleged to have occurred, disclosure of the names of the IHP recipients “would constitute a clearly unwarranted invasion” of those individuals’ personal privacy. The newspapers articulate only one central need for the IHP names. Of the several indictments involving Floridians who allegedly defrauded FEMA following the 2004 hurricane season, at least three involve individuals who allegedly applied for IHP aid using someone else’s address. Because those homes did suffer damage, knowing that aid was disbursed there would not suggest fraud. But if the recipients’ names were also disclosed, then News theoretically could use public records to cross-reference those names with the disaster addresses to determine whether those individuals had any legitimate connection to that property, or whether they instead defrauded FEMA.

As the Sun court noted, withholding the names would “substantially reduce[ ]” the potential for negative secondary effects of disclosing the addresses. Although we have previously acknowledged that it is possible to derive names from addresses through public records, we see no reason to enable this process with a ready-made list of names, absent some compelling public interest. And we cannot say that the public interest News has articulated is terribly strong. As the district court in Sun explained, “[wjhereas the addresses go[] to the heart of whether FEMA improperly disbursed funds to property that sustained no damage, the names of disaster claimants are not as probative. In [the vast majority of] cases where the name and address[] accurately reflect the property where the disaster claimant resides, the name of the disaster claimant would provide no further insight into the operations of FEMA.” 431 F.Supp.2d at 1271. As for those cases, presumably relatively few, where a recipient provided someone else’s disaster-struck address, the recipient’s name would say more about her actions than FEMA’s. In any case, the names are not necessary to determine the extent of fraud against FEMA: News can contact the legitimate residents of the homes where FEMA aid went to confirm that they did, in fact, apply for and receive aid. Accordingly, we conclude that the convenience to News of a ready list of names from which to research the extent of fraud against FEMA is outweighed by the increased privacy risks to those individuals of having the same ready list of names and addresses available to commercial solicitors, members of the press seeking quotes, and others, and that disclosure of the IHP recipients’ names would therefore constitute a clearly unwarranted invasion of personal privacy.

3. Addresses of NFIP Claimants

Finally, News appeals the district court’s decision not to require FEMA to disclose the addresses of Florida residences that were the subject of NFIP claims in 2004. News claims (and FEMA does not dispute) that one North Miami building has flooded seventeen times but remains eligible for NFIP insurance coverage without increased premiums, and that in general, in two years preceding this lawsuit, the number of NFIP-insured properties that have flooded more than once more than doubled — from 5,844 buildings with $285 million in losses in 2003 to 12,177 properties with $692 million in losses in 2005. News argues that without access to the NFIP addresses, it cannot begin to evaluate this trend and the concomitant cost to taxpayers of the federal government’s decision to continue insuring flood-prone property. We agree that this constitutes a substantial public interest, and that the NFIP addresses would serve that purpose.

Against this public interest in disclosure we weigh the privacy interests against disclosure. The NFIP addresses have received comparably little attention from the parties. FEMA provided News with spreadsheets entitled “2004 Florida Loss Report by County/Community/Date of Loss” and “Florida Premiums for Policies In Force Report by County/Community” but withheld the addresses of the claimants. No representative sample of this released data is part of the record, and FEMA has not begun to explain why being identifiable as someone who purchased federal flood insurance would constitute any invasion of privacy, much less a clearly unwarranted one. We, therefore, readily conclude that FEMA has failed to meet its burden and must disclose the NFIP addresses.

III. Conclusion

The public interest in evaluating the appropriateness of FEMA’s response to disasters is not only precisely the kind of public interest that meets the FOIA’s core purpose of shedding light on what the government is up to; the magnitude of this public interest is potentially enormous. “The critical nature of [disaster] assistance makes reports of waste, mismanagement and outright fraud particularly disturbing. We cannot sweep such allegations under the rug; we must face them head-on to preserve public confidence in this critical program.” Sen. Joseph I. Lieberman, Senators Collins and Lieberman: Investigation Reveals Waste, Mismanagement and Fraud in FEMA’s Disaster Aid Program in Miami-Dade (May 18, 2005), available at http://lieberman.senate.gov/ newsroom/release.cfm?id=237837. “The tradition of Americans helping Americans in the aftermath of a disaster .... will be jeopardized if Americans come to feel their tax dollars are not being spent fairly, efficiently — and with accountability.” Senate Hearings (written statement of Sen. Lieberman at 2). Nor is ensuring that FEMA properly spends taxpayer money only of concern to Floridians and residents of other hurricane-ravaged states. As Senator Bill Nelson of Florida told the Senate Committee, it is also of concern “to Californians, who live on fault lines, and Washingtonians, who live in the shadows of active volcanoes; rural Americans, who live near rivers that swell; and city-dwellers, who live in metropolitan areas that could be targeted by terrorists.” Senate Hearings (written statement of Sen. Bill Nelson at 2). Although we acknowledge the privacy interests at stake, given the enormous public interest involved, we cannot say that FEMA has come close to meeting its heavy burden of showing that the privacy interests are of such magnitude that disclosure of the IHP and NFIP addresses would constitute a clearly unwarranted invasion of personal privacy under Exemption 6.

The judgment of the district court in Sun is AFFIRMED. The judgment of the district court in News is REVERSED and REMANDED for further proceedings consistent with this opinion.

1 . NEMIS automatically determines eligibility in over 90% of cases based on business rules encoded in its software. FEMA caseworkers determine eligibility in the remaining cases.

2 . For example, the average kitchen, according to FEMA, consists of 25 items: dinnerware, glassware, and flatware for 8; a set of mixing bowls; a set of pots and pans with lids; a set of dining linens; 4 sets of dish towels/pot holders; a 7-piece knife set; a cooking spoon; a meat fork; a spatula; a whisk; miscellaneous cooking utensils; a dish rack/drainer; a coffee maker; a handheld mixer; a 2-slot toaster; a blender; an electric can opener; a fire extinguisher; a mop and bucket; a broom; a trash can; a 2’x4’ area rug; and a 3'x4' mini blind set. The average living room consists of 9 items: an upholstered 8’ sofa, loveseat, and chair; a coffee table; two end tables; two lamps; a clock; a 5'x8' area rug; and a 4’x5' mini blind set. The average bedroom consists of 11 items: two twin beds, standard pillows, sheet sets, blankets and bedspreads; two four-drawer chests; two nightstands; two lamps; an 18”x48” mirror; a 5’x8’area rug; and a 4'x5’ mini blind set. The average bathroom consists of 11 items: two towel racks; four sets of personal brushes/combs; four sets of personal hygiene items at $50 each; a shower rod and curtain; a tub mat; a laundry hamper; a toilet paper holder; a storage cabinet; a 3-piece rug set; and a 3'x4' mini blind set.

3 . In the days before Frances made landfall, based on its anticipated path, then-Florida Governor Jeb Bush submitted a disaster declaration request to FEMA requesting that all 67 Florida counties be declared eligible for public assistance and that 18 counties, including Miami-Dade, be declared eligible for individual assistance, including IHP aid. President Bush declared Frances a major disaster and authorized FEMA to provide public assistance to all Florida counties,' but IHP aid to only five counties, not including Miami-Dade. President Bush did, however, authorize FEMA to designate other counties eligible for IHP aid subject to FEMA’s completion of a Preliminary Damage Assessment ("PDA”) for such counties. Within 24 hours after the hurricane’s impact and without performing a PDA, FEMA amended the declaration to include for IHP eligibility Miami-Dade and the other 12 counties that the Governor had initially requested but that were excluded in the President's declaration. Once a county is declared eligible for IHP relief, any resident of that county may apply for IHP aid.

4 . The OIG Audit Report is available at http:// www.dhs.gov/xoig/assets/mgmtrpts/OIG_05-20_May05.pdf.

5 . FEMA’s response appears at pages 43-57 of the OIG Audit Report as Appendix H: Management Comments.

6 . The same week that FEMA responded to the draft audit report, it issued two public statements. The first, issued by Brown, noted that "Inspector General audits and congressional oversight are not uncommon events in the wake of a major disaster, and while every disaster sadly comes with some level of fraud and abuse, I am pleased by the report’s findings verifying our own initial conclusions of nothing widespread.” See http://www.fema. gov/news/newsrelease.fema?id= 17427. The second was a press release entitled "Hurricane Season 2005: Building on Success,” which touted FEMA’s response to the "unprecedented” 2004 hurricane season as having involved the delivery of aid "more quickly and more efficiently than ever before,” and outlined ways in which techniques that evolved during the 2004 season would be implemented in 2005 and beyond. See http:// www.fema.gov/news/newsrelease.fema?id= 17324.

7 ."The Saffir-Simpson Hurricane Scale is a 1-5 rating based on the hurricane's present intensity. This is used to give an estimate of the potential property damage and flooding expected along the coast from a hurricane landfall. Wind speed is the determining factor in the scale, as storm surge values are highly dependent on the slope of the continental shelf in the landfall region .... [A]ll winds are [measured] using the U.S. 1-minute average.” OIG Audit Report at 32. A tropical storm involves winds ranging from 39 mph to 73 mph. A category I hurricane involves winds from 74 mph to 95 mph. Id. at 10 n. 10.

8 . The Senate Hearings are available at http:// www.senate.gov/~govt-afl7mdex.cfm? Fuseaction=Hearings.Detail&HearingID= 235.

9 . FEMA's own quality control inspections in Miami-Dade County showed what the Senate Committee called "alarming” error rates of 37% on personal property inspections, 18.5% on unsafe home determinations, 16% on furnishings, 16% on clothing, and 11.5% on will-mgness to relocate. For instance, the quality control report found instances in which thousands of dollars were paid to recipients whose homes showed no damage; in one case, money was provided to repair a dryer in a home where no dryer existed. Brown, however, defended those rates as "commendable." Senate Hearings (oral testimony of Brown at 74-76).

10 . The Committee interviewed over 40 witnesses and reviewed over 50,000 pages of documents related to FEMA’s response to the 2004 Florida hurricane season. See Collins & Lieberman Press Release, available at http:// www.senate.gov/~govt-aff/index.cfmPFuse Action=PressReleases.Detail&Affiliation=R& PressRelease_id=1042&Month=7&Year= 2005.

11 . In some cases, FEMA payments duplicated payments from private insurance companies, and in other cases FEMA payments duplicated themselves. In still other cases, damage was not disaster-related, applicants received rental assistance to escape habitable homes, applicants had already received assistance from a fellow household member, or applicants received assistance legitimately but were overpaid. See News-Press v. United States Department of Homeland Security, No. 2:05-cv-102-FTM-29DNF (M.D.Fla. Nov. 4, 2005), Decl. of Jeff Cull, R45 ¶¶ 4-5 (describing recoupment data received from FEMA).

12 . FEMA says it withheld from News approximately 605,500 IHP names and addresses and 33,000 NFIP addresses related to the four 2004 Florida hurricanes. See News, Def's. Mot. for Summ. J. at 16. FEMA says it withheld from Sun an additional 684,866 names and addresses related to the other 27 disasters. See Sun, Berl Jones Decl. at 3.

13 . The FOIA clearly provides that a district court’s review of an agency’s decision to withhold information is de novo, see 5 U.S.C. § 552(a)(4)(B), but is silent as to the proper standard of appellate review.

14 . The Eleventh Circuit, in Bonner v. City of Prichard, 661 F.2d 1206 (11th Cir.1981) (en banc), adopted as precedent the decisions of the former Fifth Circuit decided prior to October 1, 1981.

15 . Thus, in both cases, the parties agreed that neither a so-called Vaughn index describing the withheld names and addresses, see Vaughn v. Rosen, 484 F.2d 820, 827-28 (D.C.Cir.1973); Ely v. Fed. Bureau of Investigation, 781 F.2d 1487, 1493-94 (11th Cir.1986), nor an in camera inspection of them was necessary to provide the district court with an adequate factual basis on which to determine the applicability of Exemption 6.

16 . The parties agreed that the closest thing to a disputed fact in these cases is the News court’s finding that disclosure of the names or addresses could lead to identity theft. However, counsel for FEMA conceded at oral argument that there is no basis in the record to support this finding, such as evidence that FEMA has already released applicants’ social security numbers, mothers' maiden names, or other data which, when paired with the disclosed names or addresses, could predictably lead to identity theft.

17 . That is, the plaintiff argued that the agency violated the Privacy Act by releasing personal information about him that, he said, fell within FOIA Exemption 6.

18 . While courts generally review challenges to agency action for an abuse of discretion, it is clear that where, as here, an action is brought under the FOIA, there can be no question of review for abuse of discretion. See Currie, 704 F.2d at 526-28 (rejecting agency's argument that its decision to withhold tax returns under FOIA Exemption 3 should be reviewed for abuse of discretion under the Administrative Procedure Act, 5 U.S.C. §§ 701 et seq., where plaintiff had brought an action to compel disclosure under the FOIA).

19 . Exemption 3 covers records "specifically exempted from disclosure by statute (other than section 552b of this title), provided that such statute (A) requires that the matters be withheld from the public in such a manner as to leave no discretion on the issue, or (B) establishes particular criteria for withholding or refers to particular types of matters to be withheld.’’ 5 U.S.C. § 552(b)(3).

20 . To the extent that any of our cases decided after Cochran could be read as suggesting that a district court's balancing of interests under Exemption 6 is reviewed for clear error, see, e.g., O'Kane v. U.S. Customs Serv., 169 F.3d 1308, 1309-10 (11th Cir.1999) (per curiam) (stating that we review a district court’s grant of summary judgment de novo but its FOIA "determinations” for clear error, and holding, in that Exemption 6 case, that the district court "did not clearly err in determining that an individual’s interest in his or her home address outweighs the 'public interest' [plaintiff] asserts”), we are bound by Cochran’s plain holding that in such cases, "the 'clearly erroneous’ standard employed in Chilivis and Stephenson is inappropriate.” Cochran, 770 F.2d at 956 n. 8. See Cohen v. Office Depot, Inc., 204 F.3d 1069, 1072 (11th Cir.2000) ("[W]here two prior panel decisions conflict we are bound to follow the oldest one.”).

21 . The FOIA, 5 U.S.C. § 552, provides that: "(b) This section does not apply to matters that are ... (6) personnel and medical files and similar files the disclosure of which would constitute a clearly unwarranted invasion of personal privacy .... ”

22 . Thus, for example, one applicant received $6,500 to replace a car that the inspector indicated had been destroyed by electrical fire. The inspection report did not indicate how a hurricane could have caused an electrical fire or how the inspector verified the damage. Skinner testified that only by calling the aid recipient did the OIG discover that the car had allegedly been towed prior to the inspection and that the award was based on the owner's verbal representations. "[T]hat was not reflected in the inspector’s report. We obtained that information by, in fact, talking to the ... individual.” Senate Hearings (oral testimony of Skinner at 50).

23 . Judging from the amounts awarded, we assume that each ''item'' of clothing was in fact a unit akin to something like a full wardrobe.

24 . The U.S. Small Business Administration provides low-interest disaster loans to homeowners, renters, and owners of non-farm businesses of any size that sustained uninsured or underinsured damage or loss to real or personal property.

25 . The press release is available at http:// www.fema.gov/news/newsrelease.fema?id= 14333.

26 . See www.fema.gov/news/newsrelease. fema?id=23562 (last visited June 18, 2007).

27 . Currently $28,200, as adjusted. See 42 U.S.C. § 5174(h); 44 C.F.R. § 206.110(b).

28 . According to FEMA's counsel, FEMA’s disaster relief program "makes no provision for [insurance] deductibles as such,” and FEMA’s website says that FEMA does not cover deductibles per se. See Frequently Asked Questions, http://www.fema.gov/assistance/dafaq. shtm#18. However, the unmet needs of insured individuals are covered, see id.., so that if an insured individual still has unpaid losses after her insurance settlement, which she likely would if she had a hefty deductible, then presumably FEMA could pay for those losses.

29 . We note, however, that it is highly unlikely that the newspapers will publish a list of many IHP recipients indicating after each name .whether a particular individual rents her home, had some uninsured property, or was possibly turned down for an SBA loan. Nevertheless, once the addresses are disclosed, this information would be available for anyone sufficiently disposed to seek it out.

30 . Available at http://www.fema.gov/help/ privacy_registration.shtm.

31 . Available at http://www.fema.gov/pdf/help/ privacy.pdf.

32 . Available at http://www.fema.gov/pdi/help/ part_ad.pdf.

33 . FEMA also relies on several cases, most of which barred the release of names and addresses under Exemption 6, for the proposition that there is a privacy interest in names and addresses, especially where they are coupled with additional information. We do not deny that a privacy interest against disclosure may exist, although it is not very substantial here, but unlike each of the privacy interests detailed in these cases, the privacy interest here is dwarfed by the powerful public interest in disclosure. See U.S. DOD v. FLRA, 510 U.S. 487, 497, 114 S.Ct. 1006, 127 L.Ed.2d 325 (1994) (public interest is “negligible, at best”); FLRA v. U.S. DOD, 977 F.2d 545, 548 (11th Cir.1992) ("no FOIA-related public interest”); FLRA v. U.S. Dep’t of Treasury, 884 F.2d 1446, 1452 (D.C.Cir.1989) (public interest is "only modestly distinguishable” from that in the court's prior decision in Homer, where the public interest was "absolute zero”); Painting & Drywall Work Pres. Fund, Inc. v. Dep’t of Hous. & Urban Dev., 936 F.2d 1300, 1303 (D.C.Cir.1991) ("no obvious public interest”); Am. Fed’n of Gov't Employees, Local 1923 v. United States, 712 F.2d 931, 932 (4th Cir.1983) (per curiam) ("[A]ny benefits flowing from disclosure of the [addresses] would inure primarily to the union, in a proprietary sense, rather than to the public at large.”); Lepelletier v. FDIC, 164 F.3d 37, 47 (D.C.Cir.1999) ("no clearly discernible public interest”); Forest Guardians v. U.S. FEMA, 410 F.3d 1214, 1219-20 (10th Cir.2005) (declining to quantify the privacy interest in NFIP data where the public interest was “nonexistent"); Comm, on Masonic Homes of R.W. Grand Lodge, F. & A.M. of Pa. v. NLRB, 556 F.2d 214, 221 (3d Cir.1977) ("no significant public interest”); Wine Hobby USA, Inc. v. U.S. IRS, 502 F.2d 133, 137 (3d Cir.1974) (“no direct or indirect public interest”); FLRA v. U.S. DOD, 984 F.2d 370, 375 (10th Cir.1993) ("[E]ven a 'minimal' privacy interest in an employee's name and home address outweighs a nonexistent public interest ....”); Hopkins v. U.S. Dep’t of Hous. & Urban Dev., 929 F.2d 81, 88 (2d Cir.1991) (disclosure "would shed no light on HUD’s performance in enforcing the prevailing wage laws”); Sheet Metal Workers Int’l Ass’n, Local No. 9 v. U.S. Air Force, 63 F.3d 994, 998 (10th Cir.1995) (only additional public interest in employee names was "attenuated,” since it would be achieved, if at all, derivatively, through direct contact with the employees (quotation marks omitted)); Painting Indus, of Haw. Market Recovery Fund v. U.S. Dep’t of Air Force, 26 F.3d 1479, 1485 (9th Cir.1994) (same); Nat'l Ass’n of Retired Fed. Employees v. Horner, 879 F.2d 873, 878-79 (D.C.Cir.1989) (disclosure of names and addresses of individuals on government annuity rolls, indicating that they were either retired or disabled and the recipient of monthly government annuity checks, involves only a "modest” privacy interest, despite expectation of a "barrage of solicitations” by mail, phone, and at home, and disclosure must be barred only because there is “no public interest in” their disclosure); Minnis v. U.S. Dep’t of Agric., 737 F.2d 784, 787 (9th Cir.1984) (disclosure "would not further [a cognizable FOIA] objective” and the public interest in disclosure was therefore "negligible”); Heights Cmty. Cong. v. Veterans Admin., 732 F.2d 526, 530, 527 (6th Cir.1984) (district court was not clearly erroneous in barring release of addresses of veterans who received federal loans where community group stated a public interest "in merely 'monitoring' the operation of a federal program, without more,” to determine if "lenders and realtors [not the government] were manipulating the VA loan program so as to steer white and black veterans into specific areas of” the city, where it was likely that any lender or realtor accused of steering would "interrogat[e]” the veterans and where the community group could instead solicit participation in its investigation from veterans); Aronson v. U.S. Dep’t of Hous. & Urban Dev., 822 F.2d 182, 186-88 (1st Cir.1987) (requiring disclosure of names and addresses of individuals "owed a substantial sum of money” by HUD despite an expectation they “may become ... target[s] for those who would like to secure a share of that sum by means scrupulous or otherwise,” where HUD had failed to locate the individuals after one year, such that disclosure would serve the "quite substantial” public interest "in the revelation and consequent correction of an inability of HUD to disburse funds to their rightful owners”).