In Re Franklin Nat. Bank Securities Lit., 478 F. Supp. 577 (E.D.N.Y. 1979)

WEINSTEIN, District Judge.

The Federal Deposit Insurance Corporation (FDIC) seeks to compel the Office of the Comptroller of the Currency (OCC) to produce Examination Reports resulting from OCC's periodic examinations of Franklin National Bank and various OCC summaries and analyses of these reports. Federal Rules of Civil Procedure, Rules 34 and 37(a). A claim of official information privilege is upheld with respect to the summaries and analyses of the reports, but denied with respect to the reports themselves.

This complex litigation followed the financial collapse of the Franklin National Bank (FNB). See generally In re Franklin Nat'l. Bank Securities Litigation, 73 F.R.D. 25 (E.D.N.Y.1976); H.R.Rep. No. 94-1669 (Adequacy of the Office of the Comptroller of the Currency's Supervision of Franklin National Bank), 94th Cong., 2d Sess. (1976); Frank Wille, "The FDIC and Franklin National Bank: A Report to the Congress and all FDIC-Insured Banks," presented before the 81st Annual Convention of the Savings Banks Association of New York State, Boca Raton, Florida, November 23, 1974. A brief outline of the primary claims raised in the consolidated actions will help set this discovery motion in context.

Various Bankers Blanket Bonds insured both FNB and its parent holding company, Franklin New York Corporation (FNYC), against any loss through dishonest or fraudulent acts of employees of either corporation. Both the FDIC, in its capacity as the receiver of FNB, and the Trustee in Bankruptcy of FNYC, have filed suit seeking recovery on those bonds. See Federal Deposit Insurance Corp. v. National Surety Corp., 425 F. Supp. 200 (E.D.N.Y.1977). Concurrently, both the FDIC and the Trustee have filed complaints against various officers and directors of FNB and FNYC. The FDIC and the Trustee have also sued Ernst & Ernst, the auditors of FNB and FNYC, claiming negligent auditing procedures. Overlapping claims have been made by a class of stockholders of FNYC. See In re Franklin National Bank, 381 F. Supp. 1390 (E.D.N.Y.1974), aff'd. in part, rev'd. in part, rem'd., 574 F.2d 662 (2d Cir. 1978), clarified, 599 F.2d 1109, (2d Cir. 1979). Finally, the primary defendants have all asserted third party claims against the OCC, *579 the FDIC and the Federal Reserve Board, based upon their regulation of FNB. See In re Franklin National Bank Securities Litigation, 445 F. Supp. 723 (E.D.N.Y.1978), supplemental opinion 449 F. Supp. 574 (E.D. N.Y.1978). In addition to these and other civil actions now before us, criminal actions are pending against various officers and directors of FNB in the United States District Court for the Southern District of New York.

The primary documents at issue are the reports produced during the OCC's periodic examination of Franklin National Bank. Pursuant to statutory directive, the OCC examined FNB fifteen times from 1963 to 1974. 12 U.S.C. § 481.

Each examination is supervised by an examiner in charge and conducted by a team of national bank examiners and Assistant National Bank Examiners. The examiners seek to "determine the condition and performance of banks, the quality of their operations, the capacity of management, and compliance with Federal laws." H.R.Rep. No. 94-1669 (Adequacy of the Office of the Comptroller of the Currency's Supervision of Franklin National Bank), 94th Cong., 2d Sess. 10 (1976). On bank premises the examiners inspect bank documents, interview bank employees, officers and directors, and observe bank operations; the examination focuses upon the quality of the bank's assets, the sufficiency of internal controls, the adequacy of capital structure and the soundness of management's policies.

Reports of Examination collect and evaluate the findings. Each report contains two sections, the open section ("white pages") and the closed section ("yellow pages"). The open section compiles the basic data of the report; it includes, for example, a Consolidated Statement of Condition, the examiner's comments and observations, and financial data supporting the figures on the consolidated income statement. The only evaluative or analytical material in the open section is found in the "Page 2 Comments," the examiner's "Comments on Matters Requiring Attention." The closed confidential section evaluates these basic findings and data; it includes "the examiner's evaluation of the condition, management, earnings, capital, internal controls, ownership and future prospects of the bank." H.R.Rep. No. 94-1669, (Adequacy of the Office of the Comptroller of the Currency's Supervision of Franklin National Bank), 94th Cong., 2d Sess. 11 (1976).

The other documents covered by OCC's claim of official information privilege are primarily summaries and analyses of the Examination Reports prepared by various officials within the OCC.

After the Examination Reports had been prepared by the examiners who had visited the premises of FNB, an Assistant Chief National Bank Examiner at the Washington office would prepare a summarizing review memorandum. This Assistant Chief never actually conducted visits or interviews, but simply highlighted and analyzed trends for intra-office purposes. Beginning in 1967, Larry T. Gerzema prepared these intra-office FNB summaries.

After the failure of FNB, John E. Shockey, the Comptroller's Chief Counsel, requested that John W. Shockey, the senior in-house National Bank Examiner, review the OCC's past monitoring of the bank's condition. The memoranda prepared by John W. Shockey analyze the Examination Reports of FNB and suggest improvements in the agency's examination and reporting procedures. They were intended solely for the internal use of the OCC.

A subcommittee of the House of Representative's Committee on Government Operations, the Rosenthal Committee, conducted hearings investigating the adequacy of the OCC's regulation of FNB. In preparation for the Comptroller's testimony the *580 OCC staff prepared memoranda containing drafts for the Comptroller's opening statement, suggesting materials for the Comptroller to review before testifying and proposing answers to possible questions from the committee.

The OCC has, for the most part, made the open sections of the Examination Reports fully available to the other parties. The sole exception relevant to the issues addressed here is the redaction of some of the "Page 2 Comments," containing the examiner's "Comments on Matters Requiring Attention." Although the OCC has also produced the confidential sections of the Reports, they have been made available in stringently censored form. Eliminated are the examiner's evaluations, interpretations and opinions. The OCC has also refused to produce either the "Shockey to Shockey Memoranda" or the "Rosenthal Committee Preparation Memoranda" and has heavily cut the "Gerzema Review Memoranda."

Rule 509 of the Proposed Rules of Evidence promulgated by the Supreme Court summarized the existing case law defining the scope of the official information privilege. Although Congress eliminated Rule 509, together with the other specific rules of privilege, it never rejected the concept it embodied. See, e. g., Hearings Before the Comm. on the Judiciary, United States Senate, 93rd Cong., 2d Sess. on Federal Rules of Evidence 49-50 (1974). As Congressman Hungate, Chairman of the House Judiciary Subcommittee, put it explaining the Conference Report on the floor, the intent was "to leave the Federal Law of Privileges where we found it." Cong.Rec. H 12254 (daily ed. Dec. 18, 1974). Thus, under Rule 501 of the Federal Rules of Evidence, providing that privileges "shall be governed by the principles of the common law as they may be interpreted by the courts of the United States in the light of reason and experience," Proposed Rule 509 "remains a useful guide and standard." United States v. Fatico, 441 F. Supp. 1285, 1298 (E.D.N.Y. 1977), rev'd. on other grounds, 579 F.2d 707 (2d Cir. 1978).

Under subsection (b) of Proposed Rule 509 the "government has a privilege to refuse to give evidence . . . upon a showing of reasonable likelihood of danger that the evidence will disclose . . . official information as defined in this rule." Subsection (a) (2) defines intragovernmental opinions as one type of official information:

information within the custody or control of a department or agency of the government the disclosure of which is shown to be contrary to the public interest and which consists of: (A) intragovernmental opinions or recommendations submitted for consideration in the performance of decisional or policy-making functions.

. . .

The documents at issue here are claimed to be privileged solely because they are such intragovernmental communications. No claim has been made that discovery must be denied because the documents contain privileged military, e. g., United States v. Reynolds, 345 U.S. 1, 73 S. Ct. 528, 97 L. Ed. 727 (1953); Totten v. United States, 92 U.S. 105, 23 L. Ed. 605 (1875), or diplomatic information, e. g., Republic of China v. National Union Fire Insurance Co., 142 F. Supp. 551 (D.Md.1956); United States v. Burr, 25 Fed. Cas. p. 30, No. 14692d (C.C.D.Va.1807). See generally Zagel, The State Secrets Privilege, 50 Minn.L.Rev. 875 (1966); Note, National Security and the Amended Freedom of Information Act, 85 Yale L.J. 401 (1976); Note, National Security and the Public's Right to Know: A New Role for the Courts Under the Freedom of Information Act, 123 U.Pa.L.Rev. 1438 (1975). Because demands for the production of reports prepared by bank regulatory bodies cannot be made under the Freedom of Information Act, 5 U.S.C. § 552(b) (8), the Act's exemption for "inter-agency or intra-agency memorandums," 5 U.S.C. § 552(b) (5), is not implicated here. See Denny v. Carey, 78 F.R.D. 370 (E.D.Pa.1978).

The primary rationale for the intragovernmental opinion privilege is that effective *581 and efficient governmental decision making requires a free flow of ideas among government officials and that inhibitions will result if officials know that their communications may be revealed to outsiders. The theory is not unlike that supporting the attorney-client privilege, see generally McCormick, Evidence §§ 87-97 (2d Ed. 1972); 8 Wigmore, Evidence §§ 2290-2329 (McNaughton rev. 1961), the psychotherapist-patient privilege, see generally Slovenko, Psychotherapy, Confidentiality and Privileged Communication (1966); Slovenko, Psychiatry and a Second Look at the Medical Privilege, 6 Wayne L.Rev. 175 (1960); Note, Confidential Communications to a Psychotherapist: A New Testimonial Privilege, Nw.U.L.Rev. 384 (1952), or the husband-wife privilege, see generally 8 Wigmore §§ 2332-2341 (McNaughton rev. 1961).

The assumption is that "government, no less than the citizen, needs open but protected channels for the kind of plain talk that is essential to the quality of its functioning." Carl Zeiss Stiftung v. V. E. B. Carl Zeiss, Jena, 40 F.R.D. 318, 325 (D.D.C. 1966), aff'd. on opinion below, 128 U.S.App. D.C. 10, 384 F.2d 979, cert. denied, 389 U.S. 952, 88 S. Ct. 334, 19 L. Ed. 2d 361 (1967). See Kaiser Aluminum & Chemical Corp. v. United States, 157 F. Supp. 939, 946, 141 Ct.Cl. 38 (1958). Without frank, confidential discussion between subordinates and their superiors, "crucial decisions may be made with insufficient knowledge or inadequate advice." Note, Discovery of Government Documents and the Official Information Privilege, 76 Colum.L.Rev. 142, 143 (1976). See Soucie v. David, 145 U.S.App. D.C. 144, 161-62, 448 F.2d 1067, 1080-81 (1971) (concurring opinion) ("otherwise the advice received and the exchange of views may not be as frank and honest as the public good requires").

A subsidiary theory supporting the official information privilege relies upon a related, though still distinct principle: "The judiciary . . . is not authorized `to probe the mental processes' of an executive or administrative officer." Carl Zeiss Stiftung v. V. E. B. Carl Zeiss, Jena, 40 F.R.D. 318, 325 (D.D.C.1966), aff'd. on opinion below, 128 U.S.App.D.C. 10, 384 F.2d 979, cert. denied, 389 U.S. 952, 88 S. Ct. 334, 19 L. Ed. 2d 361 (1967). See United States v. Morgan, 313 U.S. 409, 61 S. Ct. 999, 85 L. Ed. 1429 (1941). Cf. Montrose Chemical Corporation of California v. Train, 160 U.S.App. D.C. 270, 277, 491 F.2d 63, 70 (1974) (making similar point concerning exemption 5 of the Freedom of Information Act); United States v. Schipani, 289 F. Supp. 43, aff'd., 414 F.2d 1262 (2d Cir. 1969), cert. denied, 397 U.S. 922, 90 S. Ct. 902, 25 L. Ed. 2d 102 (1970) (effect on burden of proof of inappropriateness of court's probing reasons for executive decisions). This principle forbids judicial investigation into "the methods by which a decision is reached, the matters considered, [and] the contributing influences or the role played by the work of others." Carl Zeiss Stiftung at 326. Unlike the first ground, which rests upon pragmatic considerations of efficiency and effectiveness, this second justification draws upon constitutional considerations, derived from the doctrine of separation of powers. See, e. g., United States v. Nixon, 418 U.S. 683, 94 S. Ct. 3090, 41 L. Ed. 2d 1039 (1974); Nixon v. Sampson, 389 F. Supp. 107, 150 n. 112 (D.C.D.C.1975); Berger, How the Privilege for Governmental Information Met its Watergate, 25 Case W.Res.L.Rev. 747 (1975); Cox, Executive Privilege, 122 U.Pa. L.Rev. 1383 (1974); Owens, The Establishment of a Doctrine: Executive Privilege After United States v. Nixon, 4 Texas So.U. L.Rev. 22 (1976); Bishop, The Executive's Right of Privacy: An Unresolved Constitutional Question, 66 Yale L.J. 477 (1957).

Implicit in the two theories upon which the official information privilege rests is an important limitation upon its scope. The privilege protects only expressions of opinion or recommendations in intragovernmental documents; it does not protect purely factual material. See, e. g., Environmental Protection Agency v. Mink, 410 U.S. 73, 93 S. Ct. 827, 35 L. Ed. 2d 119 (1973); Black v. Sheraton Corp. of America, 184 U.S.App.D.C. 46, 56-57, 564 F.2d 531, 541-42 (1977); Boeing Airplane Co. v. *582 Coggeshall, 108 U.S.App.D.C. 106, 112-113, 280 F.2d 654, 660-661 (1960). Cf. General Services Administration v. Benson, 415 F.2d 878 (9th Cir. 1969); Consumers Union of United States, Inc. v. Veterans Administration, 301 F. Supp. 796, 806 (1969), dismissed as moot after full disclosure by the government, 436 F.2d 1363 (2d Cir. 1971); Long Island Railroad Co. v. United States, 318 F. Supp. 490, 499 n. 9 (E.D.N.Y.1970); Farrell v. Piedmont Aviation, Inc., 50 F.R.D. 385, 386 (W.D.N.C.1969) (making same distinction under the Freedom of Information Act). Disclosure of facts would neither hinder the free flow of advice in government decision making nor involve improper judicial interference with that process. Only the protection of opinions, recommendations and analyses furthers these two goals. Of course, as noted below, the line between "fact" and "opinion," here as elsewhere in the law, is neither distinct nor fixed.

Even when the government asserts the official information privilege with respect to opinions or recommendations, nondisclosure is not guaranteed. The official information privilege is a qualified privilege; it is not absolute. "[T]he privilege is a discretionary one that depends upon ad hoc considerations of competing policy claims." United States v. Article of Drug Consisting of 30 Individually Contained Jars More or Less, 43 F.R.D. 181, 190 (D.Del.1967). See Timken Roller Bearing Co. v. United States, 38 F.R.D. 57, 64 (N.D. Ohio 1964). Strong competing interests must be weighed against the government's interest in nondisclosure. Foremost is the interest of the litigants, and ultimately of society, in accurate judicial fact finding. See, e. g., Bank of Dearborn v. Saxon, 244 F. Supp. 394, 401-403 (E.D.Mich.1965), aff'd, 377 F.2d 496 (6th Cir. 1967) ("the real public interest under such circumstances is not the agency's interest in its administration but the citizen's interest in due process"); United States v. Article of Drug Consisting of 30 Individually Cartoned Jars More or Less, 43 F.R.D. 181, 190 (D.Del.1967); Note, Discovery of Government Documents and the Official Information Privilege, 76 Colum.L. Rev. 142, 143 (1976). Thus "the potential value to the private litigant of the requested production" must be carefully weighed. United States v. Beatrice Foods Company, 52 F.R.D. 14, 20 (D.Minn.1971).

There is in addition, in some circumstances, a public interest in opening for scrutiny the government's decision making process. As Justice Brennan has recently observed, a paradox inheres in the privilege's rationale: "so as to enable the government more effectively to implement the will of the people, the people are kept in ignorance of the workings of their government." Herbert v. Lando, 441 U.S. 153, 195, 99 S. Ct. 1635, 1659, 60 L. Ed. 2d 115 (1979) (dissenting opinion). Government documents are protected from discovery so that the public will benefit from more effective government; when the public's interest in effective government would be furthered by disclosure, the justification for the privilege is attenuated. Thus, for example, where the documents sought may shed light on alleged government malfeasance, the privilege is denied. See, e. g., Bank of Dearborn v. Saxon, 244 F. Supp. 394, 401-03 (E.D.Mich.1965), aff'd., 377 F.2d 496 (6th Cir. 1967); Note, Discovery of Government Documents and the Official Information Privilege, 76 Colum.L. Rev. 142, 143-44 (1976).

Given this clash of strong competing interests, the official information privilege usually requires examination of the documents in camera. See, e. g., Carl Zeiss Stiftung v. V. E. B. Carl Zeiss, Jena, 40 F.R.D. 318, 331 (D.D.C.1966), aff'd. on opinion below, 128 U.S.App.D.C. 10, 384 F.2d 979, cert. denied, 389 U.S. 952, 88 S. Ct. 334, 19 L. Ed. 2d 361 (1967); Note, Discovery of Government Documents and the Official Information Privilege, 76 Colum.L.Rev. 142, 168-70 (1976). The Supreme Court's Proposed Rule 509(c) provided for in camera inspection; usually only after such an examination can the court determine whether the government's interest in nondisclosure outweighs the interests of the litigants and public in disclosure. See, e. g., Jabara v. Kelley, 75 F.R.D. 475 (E.D.Mich.1977); United States v. Article of Drug Consisting of 30 Individually Cartoned Jars More or *583 Less, 43 F.R.D. 181, 190 (D.Del.1967); United States v. Beatrice Foods Company, 52 F.R.D. 14, 20 (D.Minn.1971).

In this balancing of competing interests, some of the factors that assume significance are (i) the relevance of the evidence sought to be protected; (ii) the availability of other evidence, see, e. g., Carl Zeiss Stiftung v. V. E. B. Carl Zeiss, Jena, 40 F.R.D. 318, 331 (D.D.C.1966), aff'd. on opinion below, 128 U.S.App.D.C. 10, 384 F.2d 979, cert. denied, 389 U.S. 952, 88 S. Ct. 334, 19 L. Ed. 2d 361 (1967); (iii) the "seriousness" of the litigation and the issues involved, see, e. g., Freeman v. Seligson, 132 U.S.App.D.C. 56, 60, 405 F.2d 1326, 1340 (D.C.Cir.1968); (iv) the role of the government in the litigation, see, e. g., Carl Zeiss Stiftung, 40 F.R.D. at 329; Bank of Dearborn v. Saxon, 244 F. Supp. 394, 401-03 (E.D.Mich.1965), aff'd., 377 F.2d 496 (6th Cir. 1967); and (v) the possibility of future timidity by government employees who will be forced to recognize that their secrets are violable.

The government contends that the redacted portions of the Examination Reports the material from the "Page 2 Comments" of the Open Section and the material from the Confidential Sectionis protected by the official information privilege because (i) the material consists of opinions and recommendations made by the examiners, and (ii) the interest of the government in nondisclosure outweighs the interest of the litigants and the public in disclosure. Having examined the documents in camera, the court finds neither contention convincing.

The first component of the government's argument, the claim that the redacted material consists of just that brand of intragovernmental expressions of opinion that the official information privilege seeks to protect, is initially appealing. The standard Examination Report forms on which the examiner presents his findings do suggest that the redacted material consists of "opinions" qualitatively different from the "facts" tabulated in the open section. "Page 2 Comments" are entered on a form headed "Examiner's Comments on Matters Requiring Attention." The form instructs the examiner to "comment on items in previous report of examination which have not been satisfactorily adjusted." Closed or confidential sections are contained on forms entitled "Confidential Memorandum to the Comptroller of the Currency." The initial page of these forms lists questions calling for the examiners comments on precisely specified problem areas: whether the examiner "regard[s] the management as safe;" whether "the directors exercise reasonable and independent supervision over the bank's affairs;" whether the bank has violated statutes forbidding "unsafe or unsound practices;" and whether the examiner "consider[s] the bank solvent." Subsequent pages, entitled "General Remarks," allow the examiner to comment on the broader topics of: (i) "Conditions of the Bank;" (ii) "Management;" (iii) "Earnings;" (iv) "Capital;" (v) "Internal Control and Audit Procedures;" (vi) "Ownership;" and (vii) "Future Prospects." The examiner is instructed to begin "each heading with a descriptive word covering the over-all condition such as Excellent, Good, Fair or Poor." This format, it is urged, indicates that the material presented by the examiner constitutes opinions, analyses, evaluations and recommendations qualitatively different from the facts and statistics contained in the open sections.

Examination of the examiner's actual comments, however, reveals that no such crisp distinction can be drawn between "opinion" and "fact." The wisdom of this observation has already been firmly acknowledged in the context of other areas of the law. For example, both courts and commentators sharply criticized the distinction drawn between "facts" and "conclusions" by the requirements of Code pleading *584 rules. See, e. g., 2A Moore, Federal Practice, ¶ 8.12-13 (1978); Wheaton, Manner of Stating Cause of Action, 20 Cornell L.Q. 185, 209 (1935); Cook, "Facts" and "Statements of Fact," 4 U. of Chi.L.Rev. 233 (1937); Cook, Statements of Fact in Pleading Under the Code, 21 Colum.L.Rev. 416 (1921). The problem, as Justice Holmes observed, was that statements of conclusion often only collect and summarize statements of fact: such a "statement may be called a conclusion, but it is a conclusion of fact, just as the statement that a certain liquid was beer is a conclusion of fact from certain impressions of taste, smell and sight." Southern Ry. Co. v. King, 217 U.S. 524, 538, 30 S. Ct. 594, 54 L. Ed. 868 (1910) (Holmes, J., dissenting). See Note, 38 Cornell L.Q. 79, 81 (1952) ("`conclusions,' `facts,' and `evidence' differ only by degrees.").

The same observation centered prominently in criticism of the old evidentiary "opinion rule," allowing lay witnesses to testify to facts but not opinions. Wigmore, for example, questioned the possibility of clearly distinguishing the two: "As soon as we come to analyze and define these terms . . . the distinction vanishes, . . . a flux ensues." 7 Wigmore § 1919 (Chadbourne rev. 1978). Moore also acknowledged "the illusory quality of such a fact-opinion distinction." 11 Moore, Federal Practice § 701.02 (1976). The critical point bearing on the issue before us is not simply the philosophical insight that statements usually contain both objective and subjective components, see, e. g., Securities and Exchange Commission v. Texas Gulf Sulphur Co., 446 F.2d 1301, 1305 (2d Cir.), cert. denied, 404 U.S. 1005, 92 S. Ct. 561, 30 L. Ed. 2d 558 (1971), rehearing denied, 404 U.S. 1064, 92 S. Ct. 733, 30 L. Ed. 2d 753 (1972) ("[a]ll human perception includes elements of subjective conclusions"), but rather the practical experience that opinions often represent a summary of statements of fact. The "lay witness uses his opinion as a shorthand rendition of a set of collective facts otherwise difficult to state." United States v. Milne, 487 F.2d 1232, 1235 (5th Cir. 1973).

Federal Rule of Evidence 701 modifies the old opinion rule and allows a lay witness to testify "in the form of opinions or inferences" when "those opinions or inferences . . . are (a) rationally based on the perception of the witness and (b) helpful to a clear understanding of his testimony on the determination of a fact in issue." Fed. R.Evid. 701. This new formulation recognizes that statements sounding like opinions can contain significant factual content. As the Advisory Committee on Federal Rules of Evidence noted in support of Rule 701, "witnesses often find difficulty in expressing themselves in language which is not that of an opinion or conclusion." Advisory Committee Notes, Rule 701, Federal Rules of Evidence.

Perusal of the bank examiner's comments in the confidential sections of the examination reports and in the "Page 2 Comments" suggests that these observations apply to the distinction drawn between "fact" and "opinion" by the official information privilege. Most of the statements that the government characterizes as expressions of opinion could, as well, be characterized as "shorthand" statements of fact.

Consider, for example, the examiner's response, in the November 14, 1973 report, to the question "Do the directors exercise reasonable and independent supervision over the bank's affairs or is the management left largely to the active officers?" The examiner commented: "Directors exercised reasonable and independent supervision, however, daily supervision is left largely to active management." The government redacted the observation, characterizing it as an opinion protected by the official information privilege. Yet, although this statement has the conclusory ring of an opinion, it is nonetheless a statement with predominant factual content. The bank examiner visited the bank premises, observed its operations, and interviewed the directors and officers. This comment that the director's supervision was "reasonable and independent" collects and summarizes his numerous observations; it is, in this respect, a shorthand statement of fact, far easier to express than a catalog of each individual observation *585 that evidenced the "reasonable and independent" supervision.

This same analysis applies to the examiner's comments in the portion of the confidential section entitled "General Remarks." Under the heading "Condition of Bank," in the November 14, 1973 examination report, the examiner commented: "Extremely poor. Bank's liquidity position is considered hazardous by this examiner. . . . The bank would be hard pressed to meet a substantial decline in deposits and its position would be even more precarious should it lose access to the Federal funds market." Again, this statement undoubtedly expresses an opinion, but it also contains a significant factual component. The comment summarizes the examiner's study of the financial records and discussions with bank personnel. It is a shorthand statement condensing many individual observations and impressions. Although it is true that the financial data underlying the statement is collected in the open section of the report, the examiner's comment is still significant because it draws upon his first hand impressions, his contemporaneous "feel" for the bank's condition. This factual component is lacking in the raw financial statistics.

In this sense, then, the examiner's comments redacted by the government are not opinions, but are factual statements by a first hand observer of the bank; they should not be protected from disclosure by the official information privilege. To be sure, some of the examiner's comments contain less of this factual component than others, and thus perhaps deserve protection by the privilege. These few exceptions, however, are both trivial and harmless and do not justify the effort necessary to weed them out. Moreover, cutting a phrase here and there would make the document difficult to read and almost impossible to evaluate fairly. Partial "suppression will distort the tenor of the document." Moore-McCormack Lines, Inc. v. I.T.O. Corp., 508 F.2d 945, 949 (4th Cir. 1974).

As the second component of its argument, the government insists that this material contains just that brand of blunt, candid discussion that the privilege seeks to foster; disclosure would chill such communication, frustrating effective future bank examinations by the OCC. This line of argument, however, misconstrues the policy behind the official information privilege; the privilege's protection does not extend as far as the government contends. The privilege does not protect all intragovernmental communications; it protects only those issued in the "deliberative or policy making process." Environmental Protection Agency v. Mink, 410 U.S. 73, 90, 93 S. Ct. 287, 35 L. Ed. 2d 119 (1973).

The primary aim is to avoid disruption of the "consultative functions of government." Kaiser Aluminum & Chemical Corp. v. United States, 157 F. Supp. 939, 946, 141 Ct.Cl. 38 (Ct.Cl.1953). As the preceding discussion of the factual content of the redacted material has suggested, although the tone of the examiner's comments is often blunt and chatty, the content is primarily reportorial and expository, not deliberative. The Examination Report contains the raw data of the bank examination, not the discussions, deliberations and consultations "comprising part of a process by which governmental decisions and policies are formulated." Carl Zeiss Stiftung v. V. E. B. Carl Zeiss, Jena, 40 F.R.D. 318, 324 (D.D.C.1966), aff'd. on opinion below, 128 U.S.App.D.C. 10, 384 F.2d 979 (D.C.Cir.), cert. denied, 389 U.S. 952, 88 S. Ct. 334, 19 L. Ed. 2d 361 (1967). Disclosure of such investigatory reports and factual compilations has been held not to jeopardize the consultative function that the official information privilege seeks to protect. See, e. g., Environmental Protection Agency v. Mink, 410 U.S. 73, 88-89, 93 S. Ct. 827, 836, 35 L. Ed. 2d 119 (1973); Machin v. Zuckert, 114 U.S.App.D.C. 335, 316 F.2d 336, cert. denied, 375 U.S. 896, 84 S. Ct. 172, 11 L. Ed. 2d 124 (1963); Boeing Airplane Co. v. Coggeshall, 108 U.S.App.D.C. 106, 112-113, 280 F.2d 654, 660-661 (1960). As a general rule the lower the level of the official making the observation and the lower the level of abstraction in the writing, the less the need for the privilege.

*586 New policies adopted by the OCC, redefining the contents of the open and confidential sections of the examination reports, lend support to this analysis. Under current practice, much of the material formerly shrouded in the confidential section is now uncovered in the open section. See Oversight Hearings into the Effectiveness of Federal Bank Regulation, Hearings before the Subcomm. on Commerce, Consumers and Monetary Affairs of the House Comm. on Government Operations, 94th Cong., 2d Sess. 184-86 (1976) (Statement of Hon. James E. Smith, Comptroller of the Currency). This policy change suggests that the OCC itself does not consider that revelation of much of the material in the confidential section it seeks to hide would disrupt future deliberative processes of the agency.

The government also points to a second disruption of effective bank regulation that might result from disclosure of the redacted material: a chilling of the relationship between bank examiners and bank officials. The possibility of disclosure of material from the confidential section, the government contends, might inhibit bank officials from candidly discussing their bank's financial condition; a bank examiner's ability to detect potential problems would thus be impaired. See U. S. v. Provident National Bank, 41 F.R.D. 209 (E.D.Pa.1966). The government exaggerates the danger. The official information privilege requires a fresh balancing of the competing interests in each case where the privilege is asserted. Holding that the unique circumstances of this case require the disclosure of the confidential section of the FNB Examination Reports does not insure that all future reports will be similarly disclosed. Moreover, the new policy of the OCC redefining the open and confidential sections suggests, again, that the OCC itself does not consider the danger attendant upon disclosure as great as it now claims.

One final interest in nondisclosure must be considered. In some circumstances, revelation of the confidential section might breed public misunderstanding of an examiner's comments, unduly undermining confidence in a bank. See United States v. Provident National Bank, 41 F.R.D. 209, 210 (E.D.Pa.1966). This consideration has little bearing here. Franklin National Bank has been closed for five years; any danger of panic among depositors has long since become academic. Privileges properly enforced during a critical period of government control and investigation "will expire upon the lapse of an unreasonable length of time." Brown v. Thompson, 430 F.2d 1214, 1215 (5th Cir. 1971). See also Frankenhauser v. Rizzo, 59 F.R.D. 339, 345 (E.D.Pa. 1973) (no need for privilege two years after end of investigation). But cf. Black v. Sheraton Corp. of America, 184 U.S.App. D.C. 46, 61, 564 F.2d 531, 546 (1977) (effect on discouragement of candor in future investigations).

Even were the government's case for shielding its documents more persuasive, the interests of the litigants and the public in disclosure are much more compelling. Accurate judicial factfinding is predominant. This factor is powerful in a situation like that presented here, where no satisfactory alternative source of information exists. See Carl Zeiss Stiftung v. V.E.B. Carl Zeiss, Jena, 40 F.R.D. 318, 325 (D.D.C.1966), aff'd. on opinion below, 128 U.S.App.D.C. 10, 384 F.2d 979, cert. denied, 389 U.S. 952, 88 S. Ct. 334, 19 L. Ed. 2d 361 (1967). The Examination Reports provide a unique and objective contemporaneous chronicle of the financial decline of Franklin National Bank; no satisfactory substitute exists. The pendency of criminal proceedings in the Southern District of New York underscores the indispensibility of the Examination Report's account of the past; the principle figures in that prosecution assert their Fifth Amendment privilege against testifying. The Examination Reports thus offer an essential alternative eyewitness view.

The information in the reports is relevant to numerous issues in the litigation; the litigant's claim of need is concrete, not abstract. Ernst & Ernst, the auditors of FNB, for example, urge that access to the *587 examiner's evaluation of FNB's financial condition will assist in defending the actions against them for negligently conducting their audits. Where the examiner's investigation paralleled that of Ernst & Ernst, the comments in the confidential section will constitute critical evidence. If the examiner failed to detect weakness in, for example, the loan portfolios, the internal controls in the foreign exchange department, or the competence of management, that will offer relevant evidence in determining whether Ernst & Ernst should be held negligent for failing to detect such weakness.

The claims filed against the outside directors of FNB offer a second useful example of the relevance of the information in the Examination Reports. Numerous parties have alleged that these outside directors negligently managed the bank. The confidential sections of the examination reports contain much information bearing on this issue: the examiner's observations on the competency of the bank's officers and employees, the degree of supervision exercised by the directors, and the sufficiency of internal accounting controls, to suggest just a few, would all be helpful in resolving these claims against the outside directors.

Finally, as a third example, consider the relevance of the confidential sections to the third party claims against the OCC alleging its negligent regulation and management of FNB. In the resolution of these claims, the knowledge of OCC officials of the financial condition of the bank is directly in issue. The examiner's comments are more than relevant here; they are crucial. Evaluation of the OCC's regulatory activities will rest, in part, upon the propriety of those activities in light of the OCC's knowledge of the bank's position. Moreover, an element of unfairness would enter if the government could further its defense against these claims by concealing relevant evidence behind the screen of government privilege. See Carl Zeiss Stiftung v. V.E.B. Carl Zeiss, Jena, 40 F.R.D. 318 (D.D.C.1966), aff'd. on opinion below, 128 U.S.App.D.C. 10, 384 F.2d 979, cert. denied, 389 U.S. 952, 88 S. Ct. 334, 19 L. Ed. 2d 361 (1967).

Aside from the importance of the material in the confidential sections of the Examination Reports to a just and accurate adjudication of the facts, an additional consideration of the public interest reduces the force of the claim for nondisclosure. Nine months before its collapse, the Franklin National Bank stood as the nation's twentieth largest bank. Its fall far surpasses the significance of the failure of more provincial banks; it threatened not only the economic health of the locality, but of the nation, and even of the international economic community. See The FDIC and Franklin National Bank: A Report to the Congress and all FDIC-Insured Banks, presented by Frank Willie before the 81st Annual Convention of the Savings Bank Association of New York (Boca Raton, Florida, November 23, 1974); H.R.Rep. No. 1669 (Adequacy of the Office of the Comptroller of the Currency's Supervision of Franklin National Bank, House Committee on Government Operations) 94th Cong., 2d Sess. (1976). The inability of the bank regulatory system to detect and prevent that failure requires thorough exploration. A strong public interest thus demands that the confidential sections be disclosed to fully air the story behind the bank's collapse and to help evaluate the adequacy of existing bank examination procedures. Unlike the usual case, where the application of the official information privilege would help provide more effective government, here it would only frustrate that goal. Cf. Bank of Dearborn v. Saxon, 244 F. Supp. 394, 401-03 (E.D.Mich.1965), aff'd, 377 F.2d 496 (6th Cir. 1967); Carl Zeiss Stiftung v. V.E.B. Carl Zeiss, Jena, 40 F.R.D. 318, 329 (D.D.C. 1966), aff'd. on opinion below, 128 U.S.App. D.C. 10, 384 F.2d 979, cert. denied, 389 U.S. 952, 88 S. Ct. 334, 19 L. Ed. 2d 361 (1967); Note, Discovery of Government Documents and the Official Information Privilege, 76 Colum.L.Rev. 142, 143-44 (1976) (suggesting a similar exception for cases of suspected government "malfeasance").

The summaries and analyses of the Examination Reports prepared within the *588 OCCthe Gerzema Review Memoranda, the Shockey to Shockey Memoranda, and the Rosenthal Committee Preparation Memorandapresent a different balance of considerations than the Examination Reports themselves. Here, the interests of the litigants and the public in revelation do not outweigh the interests of the government in nondisclosure.

The interest in just and accurate judicial factfinding, whether viewed from the narrow perspective of the litigants or the broader perspective of the public interest, has little bearing. These summaries and analyses contain no "new" facts; any information found in them is in the Examination Reports themselves. None of the authors of these documents actually visited the bank or conducted interviews; they relied solely upon the Examination Reports. Denying the parties access to these documents would not significantly deprive them of relevant evidence. In these circumstances, where "[t]he objective facts . . . are otherwise available," the goal of accurate fact finding is not impaired and the argument for disclosure is not compelling. Kaiser Aluminum & Chemical Corp. v. United States, 157 F. Supp. 939, 946, 141 Ct.Cl. 38 (1958).

Although disclosure would arguably further the public interest in fully exploring the reasons for the bank's demise and the failure of the bank regulatory system, this interest is outweighed by the primary policy supporting the official information privilege, the interest in preserving the free flow of communication between government officials. Unlike the Examination Reports, the summaries and analyses constitute part of the "consultative functions of government" exercised by officials at a policy making level that the privilege seeks to protect. Kaiser Aluminum & Chemical Corp. v. United States, 157 F. Supp. 939, 946, 141 Ct.Cl. 38 (1953). Although none of the documents constitutes an exhaustive examination of alternatives, prepared at the final stage of policy deliberation preceding agency action, the protection of the official information privilege is not so narrowly confined; it extends to a broad spectrum of "intragovernmental documents reflecting advisory opinions, recommendations and deliberations comprising part of a process by which government decisions and policies are formulated." Carl Zeiss Stiftung v. V.E.B. Carl Zeiss, Jena, 40 F.R.D. 318 (D.D.C.1966), aff'd. on opinion below, 128 U.S.App.D.C. 10, 384 F.2d 979 cert. denied, 389 U.S. 952, 88 S. Ct. 334, 19, L.Ed.2d 361 (1967). Both the Shockey to Shockey Memoranda and the Rosenthal Committee Preparation Memoranda are essentially introspective documents evaluations of the adequacy of OCC procedures prepared by high-level subordinates for the Comptroller and his Chief Counsel. The value of such self-examination depends upon the "open, frank discussion between subordinate and chief" that the official information privilege tries to foster. Kaiser Aluminum & Chemical Corp. v. United States, 157 F. Supp. 939, 946, 141 Ct.Cl. 38 (1953). Although the Gerzema Review Memoranda only summarize the Examination Reports on Franklin National Bank and highlight problem areas for review by senior officials, this very review and selection constitutes a step in the agency's deliberative process, albeit only an initial step. Separation of the important fact from the unimportant is basic to any deliberation. Cf. Montrose Chemical Corporation of California v. Train, 160 U.S.App. D.C. 270, 278, 491 F.2d 63, 71 (1974) (for purposes of the Freedom of Information Act: "The work of the assistants in separating the wheat from the chaff is surely just as much part of the deliberative process as is the later milling by running the grist through the mind of the administrator."). Thus with respect to the Gerzema Review Memoranda as well, the policy of encouraging free and open communication between subordinate and chief is strongly implicated.

The United States' assertion of the official information privilege is upheld with respect to the Shockey to Shockey Memoranda, the Rosenthal Committee Preparation Memoranda and the redacted portions *589 of the Gerzema Review Memoranda. The assertion of the privilege is denied with respect to the confidential sections and "Page 2 Comments" of the Examination Reports. These documents must be produced, subject to the protective order agreed to by the parties.

SO ORDERED.