Can an agency properly invoke the deliberative process privilege to shield internal deliberations over a sham memo requesting that another agency take action, knowing that the recipient agency will use the request to hide the real reason for its contemplated action?  Earlier this year, the D.C. Circuit answered in the affirmative. Campaign Legal Center v. DOJ, 34 F.4th 14 (D.C. Cir. 2022). This is the third in a series of three posts addressing Campaign Legal Center. Having discussed the D.C. Circuit’s dismissive treatment of a potential “government misconduct” exception argument (Part I), and then laid out the caselaw regarding that exception (Part II), this third post will provide my own analysis of the issue at the heart of the case.

Analysis

Is the Government Misconduct Exception Applicable in FOIA Cases?

A governmental misconduct exception certainly exists, and is needed, in civil litigation where the motive or intent of a government actor is legitimately put into question, see, Subpoena Duces Tecum Served on the Office of the Comptroller of the Currency, 145 F.3d 1422, 1424-25 &  n.2 (D.C. Cir. 1998).  In that context, protective orders can be used to limit the dissemination of materials that will not ultimately prove admissible at trial, Fed R. Civ. P. 26(c)(1).  As the Supreme Court has noted, no such option is available under FOIA.  NARA v. Favish, 541 U.S. 157, 174 (2004).

That said, once the documentary evidence is in the trial record, it is ordinary accessible to the general public without any particularized showing of need, Nixon v. Warner Communications, Inc., 435 U.S. 589, 597-98 (1978),[1] much like FOIA requesters need have no particular reason to obtain government records.    Moreover, even in some “government misconduct” cases in which the courts have focused on a party’s particularized need for government documents, the courts have noted the heightened broader public interest in the disclosure of documents regarding alleged government misconduct. See note 9, infra (discussing such comments in In re Franklin National Bank Securities Litigation and Bank of Dearborn v. Saxon).

The misconduct exception is analogous to the attorney-client privilege’s crime/fraud exception.  RESTATEMENT (THIRD) OF THE LAW GOVERNING LAWYERS §82, §93 (2000); 24 CHARLES ALAN WRIGHT, ET. AL., FEDERAL PRACTICE & PROCEDURE: EVIDENCE §5501 (1st ed.)(available on westlaw)(FED. PRAC. & PROC.: EVID.); Tom Lininger, No Privilege To Pollute: Expanding The Crime-Fraud Exception To The Attorney-Client Privilege, 105 MINN. L. REV. 113 (2020). Attorney-client communications are protected for similar reasons as the deliberative proves privilege — ensuring candid communications, in particular, client candor critical to providing legal representation.  Yet the privilege does not cover entering into an attorney-client relationship or engaging in particular communications aimed at furthering an ongoing or future crime or fraud.  RESTATEMENT (THIRD) OF THE LAW GOVERNING LAWYERS §82; 24 FED. PRAC. & PROC.: EVID. §5501 (text accompanying notes 51-54.2). Were a FOIA requester to show that a deliberation between agency officials and agency attorneys was in furtherance or a crime or a fraud, presumably the agency could not invoke the attorney-client privilege to withhold the documents exchanged as a part of that deliberation.  

FOIA and similar public records statutes target actual and potential corruption, i.e., official abuse of power, inter alia.  “Open government has been recognized as the best insurance that government is being conducted in the public interest.”  S. Rep. No. 93-854, 1 (May 16, 1974) (report accompanying S. 2543) (Comm. on the Judiciary).[2]  This is, of course, typified by Justice Brandeis’ famous aphorism describing sunlight as the best disinfectant.[3] See, e.g., Brigham Daniels, Mark Buntaine, Tanner Bangerter, Testing Transparency, 114 NW. U. L. REV. 1263, 1270-71, 1288-89 (2020); Seth Kreimer, Rays Of Sunlight in a Shadow “War”: FOIA, The Abuses of Anti-Terrorism, and the Strategy of Transparency, 11 LEWIS & CLARK L. REV. 1141, 1147 (2007)(“[t]ransparency can potentially discipline an overreaching Executive before, during, or after the fact.”).[4]

And official deception is even more problematic than mere secrecy.  See, Bernard W. Bell, Legislative History Without Legislative Intent: The Public Justification Approach to Statutory Interpretation, 60 OHIO ST. L.J. 1, 20-21 (1999). Indeed, enabling government officials to take actions when unwilling to state the actual reasons for the action defeats a key constraint upon government officials’ pursuit of parochial interests, whether their own or those of others, in derogation of the public interest.[5]  See generally, Publicity, §1.1, STANFORD ENCYCLOPEDIA OF PHILOSOPHY (Jan. 12, 2005, revised Nov. 8, 2021)(entry prepared by Axel Gosseries and Tom Parr).[6]

Nevertheless, recognition of the government misconduct exception in the FOIA context is vulnerable to attack.[7]  Apparently, the exception has rarely been used to breach the deliberative process privilege; it appears to have been successfully asserted in only two reported FOIA cases.[8]  Moreover, even recognition of the exception is rare, with only a Seventh Circuit and a few D.C. Circuit decisions acknowledging it in dicta. 

The recognition of the government misconduct exception in the FOIA context is not sufficiently well-grounded in precedent to deter unsympathetic courts of appeal or the U.S. Supreme Court from declaring that no such exception applies in FOIA cases.  In the past 25 years, the Supreme Court has not been swayed by long-standing and stronger court of appeals precedents with respect to other FOIA doctrines.[9] Granted, in most of those cases the Court relied on the text of the FOIA exemption to overturn, or question, the lower federal court precedent.  And the pre-FOIA cases do not present a strong case for breaching the deliberative process privilege without some particularized showing of need. 

Indeed, the privilege arguably requires some particularized showing of need and ad hoc judicial balancing of a variety of factors before discovery is ordered in civil litigation.[10]  This may suggest that the situations in which an allegation of government misconduct overcomes assertions of the deliberative process privilege are not sufficiently frequent as to qualify as “routine.” However, it is not clear that the type of statutorily-mandated showing beyond relevance needed to breach a privilege like the attorney work-product privilege is necessary with respect to breaching the deliberative process privilege when government officials’ motivations are genuinely at issue. As noted earlier (in Part II), the absence of a requirement for such a mandatory showing beyond relevant may be sufficient to show that documents protected by privilege are routinely or normally available in civil litigation in certain circumstances.

Moreover, unsympathetic courts of appeal and the Supreme Court may have concerns about the difficulty of cabining such an exception in a manner that would avoid frequent breaches of the deliberative process privilege, and the resulting potentially chilling effect upon government officials’ candor in discussing of even legitimate matters (for fear that a court might someday find the discussion related to government misconduct). See, ICM Registry, LLC v. U.S. Dep’t of Commerce, 538 F. Supp. 2d 130, 133 (D.D.C. 2008).

As the Supreme Court has observed in another context — allegations of government misconduct are “‘easy to allege and hard to disprove,’” Crawford–El v. Britton, 523 U.S. 574, 585 (1998)(quoting Crawford–El v. Britton, 93 F.3d 813, 821 (D.C. Cir. 1996)(en banc), rev’d, Crawford–El v. Britton, 523 U.S. 574 (1998). Thus, in a different context, the Supreme Court has narrowed accessibility to documents implicating privacy interests when FOIA requesters seek such documents citing their desire to unearth potential corruption. NARA v. Favish, 541 U.S. 157 (2004). There, the Court held that where a requester seeks to defeat assertion of the privacy exemption for law enforcement records based on the public interest in uncovering governmental impropriety, “the requester must produce evidence that would warrant a belief by a reasonable person that the alleged Government impropriety might have occurred.”  Id. at 174.  “Bare suspicion” of impropriety is insufficient.  Id. at 174.  The Court emphasized the lack of any “mechanism under FOIA for a protective order” that would limit access to allow “the requester to see whether the information bears out his theory,” but proscribe the information’s “general dissemination” — “if the information is subject to disclosure, it belongs to all”.  Id. at 172, 174.[11]

Do the DOJ Deliberations Over the Sham Letter Qualify As Government Misconduct?

If the “government misconduct exception” can be invoked by a FOIA requester, does its invocation defeat an assertion of the “deliberative process privilege,” to shield records of the internal DOJ deliberations regarding the sham request letter sent to the Department of Commerce?[12]

The circumstances surrounding Secretary Ross’s solicitation of a request from DOJ to add a citizenship question appears to present a strong case for finding both governmental misconduct and the perversion of government power.  The overall scheme appears to have been conceived to further either or both of two goals.  First, it may have been designed to frustrate judicial review, by providing a pretextual, and presumably more legitimate, reason for taking an action that could lead to an undercount of certain segments of the population. Second, it might have been designed to mislead the public into believing that the goal of such action was enhancing enforcement of the Voting Right Act rather than politically harming groups or states that Secretary Ross, Attorney General Sessions, and/or President Trump disfavored politically.  The use of authority to achieve improper political advantage, by reducing their opponents’ political representation while purporting to act impartially, would seem to be particularly appropriate scheme for exposure through public records laws like FOIA.[13]

But the Campaign Legal Center was not seeking records of the conversations between Secretary Ross and Attorney General Sessions.  Rather it sought communications arising out of deliberations within the Civil Rights Division and discussion among officials in the higher echelons of DOJ.  And, of course, a “sincere” discussion of whether a citizenship question would assist in enforcement of the Voting Rights Act would be entirely proper for such DOJ officials. However, we might assume that even lower-level DOJ officials who worked on the sham letter realized that its purpose was to facilitate a Department of Commerce subterfuge, given that officials in the Civil Rights Division had no real interest in any information the addition of the citizenship question would produce. But even assuming the participants’ knowledge of their role in the overall subterfuge, such deliberations might nevertheless satisfy the ICM Registry standard. Even in that context, the discussions of lower-level officials were not such that “merely discussing them was evidence of a serious breach of the responsibilities of representative government.”   ICM Registry, supra, 538 F. Supp. 2d at 133.

Moreover, the officials participating in the deliberations, whose candor we seek to encourage, may have had little or no control over whether a letter would be provided.  At least at the lower levels, and perhaps even at the higher levels, their deliberations concerned how to ensure that a bad, possibly politically-motivated decision, to request that a citizenship question be added to the census, produced as little damage as possible.  Perhaps it is reasonable to protect the views and statements of individuals in such a position, even if not the views of the higher-level officials who place them in that dilemma.  After all, the motives of the lower-level officials do not appear to be in question — the resistance of DOJ officials to providing such a letter was a reason Secretary Ross had to appeal directly to the Attorney General in the first place. 

But perhaps most importantly however, in this case, the Supreme Court concluded that extra-record discovery, including a deposition of Acting Assistant Attorney General Gore, was appropriate.  Department of Commerce, supra, 139 S. Ct. at 2564, 2574.[14]  This is somewhat unusual, as Justice Roberts himself remarked in the opinion for the Court.  Department of Commerce, supra, 139 S. Ct. at 2576. Thus, there is reason to consider whether the internal DOJ discussions should fall outside the deliberative process privilege.  It would be difficult to fully explore John Gore’s reasons for crafting the letter as he did without seeing the comments he received from those with whom he collaborated within DOJ.

The Potential of the Foreseeable Harm Standard

In 2016, Congress codified the foreseeable harm standard for withholding records that fall within one or more FOIA exemptions.  FOIA Improvement Act of 2016, Pub. L. No. 114-185, 130 Stat. 538.[15]  In deciding whether to withhold even records that fit within a FOIA exception, an agency must assess whether foreseeable harm of the type the exemption was designed to prevent is likely to flow from release its release of the record. Sen. Rpt. 114-4, 7-10 (Feb. 23, 2015).[16]

Indeed, the foreseeable-harm requirement may apply “with special force to deliberative process withholdings,” which the Congress that enacted the FOIA Improvement Act “viewed as posing particular risks of ‘overuse.’” Center for Investigative Reporting v. U.S. Customs and Border Protection, 436 F. Supp. 3d 90, 105 (D.D.C. 2019) (quoting H.R. Rep. No. 114-391, at 10 (2016)); accord, Hall & Associates v. United States Environmental Protection Agency, — F. Supp. 3d —, 2022 WL 4482569 (Sept. 22, 2022); see generally, OIP Guidance: Applying The “Forseeable Harm” Standard Under Exemption Five, FOIA Update Vol. XV, No. 2 (Jan. 1, 1994) (“there is much room for agencies to apply the ‘foreseeable harm’ standard within the realm of the deliberative process privilege under Exemption 5”).

The foreseeable harm standard may offer an approach to incorporate consideration of government misconduct or perversion of government processes into decisions regarding whether deliberative documents must be released.   Courts may be reticent to recognize a broad exception that would nullify the deliberative process privilege whenever a requester can credibly allege either government misconduct or the perversion of governmental processes.  That judicial impulse is reflected in the results of the modest number of reported cases in which the governmental misconduct exception has been discussed.  And, of course, given the balancing tests employed in civil litigation and the availability of protective orders, breach of the deliberative process privilege in the context of non-FOIA litigation alleging governmental misconduct is far more constrained.

The foreseeable harm standard offers a more flexible means to permit the release of communications that arguably should remain unprotected because they discuss matters in which the agency should not be involved or reveal potentially unstated motivations for actions, motivations that could theoretically be wrongful or illegal (or in any event, ones that decision-makers are not willing to expose to public scrutiny).  Agencies could decide to release protected information because agency officials conclude that the discussions involved reflected misconduct or a perversion of governmental processes.  And those judgment calls would be subject to judicial review; which will require agencies to explain the nature of the particular harm it can foresee from releasing particular documents. 

While courts have not fully sketched out the degree of scrutiny to which foreseeable harm determinations will be subjected, it appears that such scrutiny will not be pro forma.  “Perfunctory, sweeping, and undifferentiated declaration[s] that release of every single record withheld would have an ‘inhibiting effect’ by ‘chill[ing] full and frank discussions’” are not sufficient to defend a foreseeable harm determination.  Reporters Committee for Freedom of the Press v. FBI, 3 F.4th 350, 372 (D.C. Cir. 2021); National Public Radio v. DHS, Dkt. No. 20-2468 (Civ.), 2022 WL 4534730 (D.D.C. 2022); Reporters Committee for Freedom of the Press v. U.S. Customs & Border Protection, 567 F. Supp. 3d 97, 109 (D.D.C. 2021); Center for Investigative Reporting, supra, 436 F.Supp.3d at 106 (“generic, across-the-board articulations of harm that largely repeat statements already found in the Vaughn Index,”); Judicial Watch, Inc. v. DOJ, Dkt, No. 17-0832 (Civ.), 2019 WL 4644029, at *5 (D.D.C. Sept. 24, 2019)(“boilerplate” “nebulous articulations of harm are insufficient”). 

Rather, the agency must specifically and thoughtfully determine whether it “reasonably foresees that disclosure” of each particular record “would harm an interest protected by [the] exemption.”  Reporter’s Committee v. FBI, supra, 3 F.4th at 372.  Thus, to satisfy the foreseeable harm standard, an agency must “identify specific harms to the relevant protected interests that it can reasonably foresee would actually ensue from disclosure of the withheld materials” and “connect the harms in a meaningful way to the information withheld.” Center for Investigative Reporting, supra, 436 F. Supp. 3d at 106 (quoting Judicial Watch v. DOJ, supra, 2019 WL 4644029, at *5). In shouldering that burden, the agency must provide “context or insight into the specific decision-making processes or deliberations at issue, and how they in particular would be harmed by disclosure.” Center for Investigative Reporting, supra, 436 F.Supp.3d at 107 (quoting Judicial Watch v. DOJ, supra, 2019 WL 4644029, at *5)

Granted the appropriateness of the communications fitting within the deliberative process privilege is not one of the many factors that the Office of Information Policy has discussed as appropriate for consideration under the foreseeable harm analysis.  OIP Guidance: Applying The “Forseeable Harm” Standard Under Exemption Five, supra.[17]  And to date, no court has suggested that the propriety of the deliberation is a relevant factor in the foreseeable harm calculus (even though they have, as described above, considered the propriety of the government’s deliberations when the government asserts the privilege in civil litigation).  The factors focus on whether disclosure will chill frank discussions, not, in a sense, whether the discussion is one that should be chilled, or at least revealed to the public as improper.

In conducting such review, courts could consider agencies’ exercises of discretion in a fact-sensitive manner that might sometimes lead courts to conclude that the nature of the deliberations were such, in terms of government misconduct or perversion of government, that release of the records would have little implications for the candor of appropriate discussions involving agency action.

Conclusion

Ultimately, in Campaign Legal Center, there was a potential argument that government conduct or perversion of government processes could justify release of some or all of the communications between officials at DOJ regarding the crafting of the letter requesting adding the citizenship question.  The government misconduct exception is ill-defined, and has never been recognized by the U.S. Supreme Court in the FOIA context.  Nevertheless, when a court concludes that extra-record discovery is appropriate because of the government’s bad faith in setting forth the basis of its decision, precluding invocation of the deliberative process privilege in related FOIA litigation should certainly be strongly considered.  In any event, the propriety of the action being discussed (including providing another agency with a pretext to use in hiding the basis for its actions), should at least be considered in determining whether releasing the relevant communications will cause any real harm of the type the deliberative process privilege seeks to protect.

---

[1] In Nixon v. Warner, the Court observed:

“It is clear that the courts of this country recognize a general right to inspect and copy public records and document, including judicial records and documents. . . . American decisions generally do not condition enforcement of this right on a proprietary interest in the document or upon a need for it as evidence in a lawsuit. The interest necessary to support the issuance of a writ compelling access has been found, for example, in the citizen’s desire to keep a watchful eye on the workings of public agencies, and in a newspaper publisher’s intention to publish information concerning the operation of government.”

435 U.S. at 597-98.

Granted, like the right to access granted by FOIA, the common law right of access to judicial documents is not absolute. Id. at 598.

[2] On the linked version of the report page 1 is identified as page 153.

[3] Justice Brandeis famously wrote that “[s]unlight is said to be the best of disinfectants.”  LOUIS D. BRANDEIS, OTHER PEOPLE’S MONEY AND HOW THE BANKERS USE IT 92 (1914).

[4]  See H.R. 89-1497, 6 (May 9, 1966) (page 27 in the linked reprint)(“[h]istorically, Government agencies whose mistakes cannot bear public scrutiny have found ‘good cause’ for secrecy.”); accord, H.R. Rpt. 114–391, 11 (Jan. 7, 2021)(quoting the 1966 House Report regarding hiding government mistakes and observing that the deliberative process privilege has replaced the “good cause” exception as “the legal vehicle by which agencies continue to withhold information about government operations”).

[5] Of course, allowing the government to withhold deliberative documents may not defeat FOIA’s purpose of enabling the public to “see what the government was up to.” U.S. v. Reporters Committee for Freedom of the Press, 489 U.S. 749, 772-73 (1989) (quoting EPA v. Mink, 410 U.S. 73, 105 (1973)(Douglas, J., dissenting), and Henry Steele Commager, The Defeat of America, THE NEW YORK REVIEW OF BOOKS, Oct. 5, 1972, at 7). This phrase could be viewed as focusing on the actions government officials take, not the discussions they engage in when considering that actions it will take.  The fact that the deliberative process privilege already prevents agencies from shielding their actions, including their ultimate decisions on a course of action, NLRB v. Sears Roebuck, 421 U.S. 132, 150-53 (1975), and factual matter related to those decisions, EPA v. Mink, 410 U.S. 73, 87-91 (1973), may go a long way toward making ensuring that the public is informed about the agency’s actions, even if not its deliberations.

[6] The entry discusses Immanuel Kant’s axiom that “[a]ll actions relating to the right of other human beings are wrong if their maxim is incompatible with publicity.”  Publicity, supra §1.1.. It notes in passing David Lubban’s statement that “the best way to make sure that officials formulate policies that could withstand publicity is by increasing the likelihood that policies will withstand publicity.” Id..  Interesting, the entry discusses the optimal confidentiality of FOI law and the justifiability of Glomar responses to FOIA requests (i.e., neither confirming or denying the existence of documents. Id. at §§1.4.4, 1.4.2. Harvard Political Scientist Dennis F. Thompson has discussed Kant’s axiom as well. POLITICAL ETHICS AND PUBLIC OFFICE 116-22 (Harvard Univ. Press 1987).

[7]  Indeed, the Department of Justice, in its Guide to FOIA, notes the D.C. District Court’s recent “expression of skepticism” regarding the applicability of the government misconduct exception in the FOIA context.  DOJ Guide to FOIA, Exemption 5, at 44 (posted Aug. 26, 2019).  The Guide cites Judicial Watch, Inc. v. Dep’t of State, 235 F. Supp. 3d 310, 313-14 (D.D.C. 2017), and Judicial Watch, Inc. v. Dep’t of State, 241 F. Supp. 3d 174, 183 (D.D.C. 2017) (finding that “the only applicable Circuit authority militates against recognizing a government misconduct exception in a FOIA case”).  Separately it cites Environmental Tech International for the proposition that “the privilege is not limited to deliberations connected solely to agency activities that are specifically authorized by Congress.” DOJ Guide to FOIA, Exemption 5, supra, at 27 & n.133.

[8] Tax Reform Research Group v. Internal Revenue Service, 419 F.Supp. 419 F. Supp. 415, 426 (D.D.C. 1976); Alexander v. FBI, 186 F.R.D. 154, 164 (D.D.C. 1999).

[9] Food Marketing Inst. v. Argus Leader Media, 139 S. Ct. 2356 (2019)(overturning longstanding exemption 4 doctrine); Milner v. Department of the Navy, 562 U.S. 562 (2011)(eliminating the longstanding high-2 exemption); Department of Interior v. Klamath Water Users Protective Assn, 532 U.S. 1, 12 (2001)(questioning the consultant’s corollary to the deliberative process privilege).

[10] In re Franklin National Bank Securities Litigation, 478 F. Supp. 577, 583 (E.D.N.Y. 1979)(setting out five factor test including consideration of the relevance of the information the government seeks to withhold and the availability of other evidence); Bank of Dearborn v. Saxon, 244 F. Supp. 394, 403 (E.D. Mich. 1965), aff’d sub nom., Bank of Dearborn v. Manufacturers Nat’l Bank, 377 F.2d 496 (6th Cir. 1967).  In Bank of Dearborn, the District Judge observed that the competing assertions of the deliberative process privilege and the government misconduct exception “cannot be resolved in the abstract.” Rather, “[t]he merits of the particular matter before the court must be considered, the necessity of disclosure weighed against the need for privacy in the light of the circumstances disclosed.”  Bank of Dearborn v. Saxon, supra, 244 F. Supp. at 402. 

Indeed, the D.C. Circuit has distinguished the deliberative process privilege from the presidential communications privilege precisely because “balancing is more ad hoc in the context of the deliberative process privilege.”  In re Sealed Case, 121 F.3d 729, 746 (D.C. Cir. 1997).  Granted, the observation was made in a case in which the deliberative process privilege was not at issue.  Nevertheless, as suggested above, many courts appear to conduct such ad hoc balancing even when allegations of government misconduct are raised in an attempt to defeat the deliberative process privilege.

[11] Accord, Sikes v. Department of the Navy, 896 F.3d 1227, 1234-35 (11th Cir. 2018); Yonemoto v. Department of Veteran Affairs, 686 F.3d 681, 689-90 (9th 2011); Maricopa Audubon Society v. U.S. Forest Service, 108 F. 3d 1082, 1088 (9th Cir. 1997).

[12] Bank of Dearborn v. Saxon, supra, 244 F. Supp. at 402 (“[h]ere we have a claim of subterfuge, of sham, of the use of devices bordering on fraud whereby, it is alleged, the Comptroller’s office sought to cloak an illegal act in the habiliments of legal propriety and good faith.”)

[13] Secretary Ross’s action, seeking to diminish the voting power of certain segments of the population, would seem to qualify as an abuse of power.  One definition of “abuse of power” is “the exercise of power by public officials for private or political gain.”  Yasmin Dawood, The Antidomination Model and The Judicial Oversight of Democracy, 96 GEO. L.J. 1411, 1436-37, 1479-80 (2008).  Professor Dawood, in discussing this as the classical view of “abuse of power,” quotes John Locke’s assertion that power is abused when the ruler’s “Commands and Actions are not directed to the preservation of the Properties of his People, but the satisfaction of his own Ambition, Revenge, Covetousness, or any other irregular Passion.” Id. at 2008.  Among other philosophers in this tradition are said to be Thucydides, Plato, Aristotle, Machiavelli and Rousseau. Id.

[14] It stayed the District Court’s order that Secretary Ross sit for a deposition, without explanation.  In Re Department of Commerce, 139 S.Ct. 16 (Oct. 22, 2018). 

[15] This had been the policy of the Clinton and Obama administrations.  Sen. Rpt. 114-4, supra, at 3.  President Barack Obama, Memorandum for the Heads of Executive Departments and Agencies, Subject: Freedom of Information Act (Jan. 21, 2009); Attorney General Janet Reno, Attorney General, Memorandum for Heads of Departments and Agencies, Subject: The Freedom of Information Act (Oct. 4, 1993).

[16] In a sense the foreseeable harm rule is a recognition that the exemptions are defined in a manner that is overly-inclusive, that the rules allowing withholding are broader than the rationale for the rules.  In effect, the foreseeable harm standard requires the Government to assess not only whether the proposed withholding fits into the “rule” defining the exemption, but also the “rule’s” rationale.

[17] The Office of Information Policy advised that the principal considerations are: (a) the nature of the decision involved, (b) the nature of the decision-making process, (c) the status of the decision (i.e., whether or not a final decision has been made), (d) the status of the personnel involved, (e) the potential for process impairment, (f) the age of the information, (g) the sensitivity of individual record portions.