I. NATURE OF THE CASE

The Michigan Constitution confers enormous responsibility and authority on the governing boards of public universities: our Constitution grants to boards of public universities the “supervision of the institution and the control and direction of all expenditures from the institution’s funds.” Const 1963, art 8, § 6. In furtherance of this constitutional mandate, our Legislature similarly invests university boards with this significant oversight role. MCL 390.551 et seq.1

Consistent with its constitutional and statutory role, the Board of Regents (Board) of Eastern Michigan University (University) investigated expenditures for the president’s residence at the University, and, as part of its investigation, the Board, through one of its members, Jan Brandon, asked an immediate subordinate of the then-president of the University, Vice President of Finance Patrick Doyle, for his written opinion of the president’s role in this project. In furtherance of its investigation, the Board also sought the assistance of an outside-certified public accounting firm, and asked De-loitte & Touche, LLP (Deloitte), to conduct a comprehensive audit relating to the expenditures for the president’s residence. Deloitte ultimately issued a “voluminous and exhaustive”2 report on the subject, which the Board made public and gave to the press. Upon receiving a Freedom of Information Act (FOIA), *188MCL 15.231 et seq., request from the Ann Arbor News3 for documents relating to the president’s residence, the University, through its FOIA director, cited the “frank communications” exemption and identified, but declined to disclose, the Doyle-to-Brandon letter. Herald filed suit and asked the court to order disclosure and argued that the public had the right to know the contents of the Doyle letter. The Board responded that the Doyle letter clearly falls within the frank communications exemption because the public interest in fostering candid appraisals by subordinates of their supervisors at the highest level of the University administration is necessary to the Board’s effective investigative and oversight role. The trial court reviewed the disputed letter in camera, balanced the public interests in disclosure versus nondisclosure and, in a written opinion, concluded that the frank communication exemption under these facts “clearly outweighs the public interest in disclosure.”4 Because our Supreme Court has ruled that we are to grant deference to trial courts, which have the difficult task of balancing the public interests under the FOIA, because our Supreme Court has specifically held that we are to uphold a trial court’s “balancing” judgment unless the trial court committed clear error, and because we find that the trial court did not clearly err in its ruling, we affirm the trial court’s holding.

II. FACTS AND PROCEDURAL HISTORY

As part of the Board’s investigation into alleged overexpenditures for the president’s residence, in the summer of 2003, Jan Brandon, a member of the Board, requested a letter from University Vice President of *189Finance Patrick Doyle regarding the construction of the president’s house. In particular, Brandon desired to learn more about the University president’s role in the construction project. There was a controversy regarding construction costs, and the Board needed information to aid it in determining the appropriate course of action. Doyle’s letter, dated September 3, 2003, contained his candid appraisal of the conduct of the president regarding the construction.

On September 10, 2003, Herald sent the Board an FOIA request for documents relating to the construction of the president’s residence. Citing MCL 15.243(l)(m), the Board’s FOIA coordinator provided the following written explanation for the Board’s refusal to provide a copy of the Doyle letter in response to Herald’s FOIA request:

Please be advised that [EMU] has identified one other document which may be within the scope of your September 10,2003 [FOIA] request. The document is a September 3, 2003 letter from Patrick Doyle to EMU Regent Jan Brandon. Pursuant to [MCL 15.243(l)(m)] of the Michigan FOIA, EMU is denying your request for this letter as the letter is a communication/note within the public body EMU of an advisory nature covering other than purely factual material and preliminary to a final agency decision. Further, EMU has determined that in this particular instance the public interest in encouraging frank communications between officials and employees of EMU clearly outweighs the public interest in disclosure.

Thereafter, Herald brought this suit and asked the trial court to review the Doyle letter in camera and order its disclosure. Herald claimed, among other things, that the claimed public interest in encouraging frank communications between public officials and employees did not clearly outweigh the public interest in disclosure *190because “the Doyle letter speaks to critical issues involving the President’s financial accountability and his management style.”

In its response to Herald’s motion, the Board indicated that the Doyle letter was requested by Regent Brandon “to assist her in determining the appropriate course of action for [the Board] to take during the early stages of the controversy,” and that the letter was “used as part of the deliberative process that [the Board] engaged in, through its individual members, to determine its course of action in the University House matter.”

In light of these facts, the Board argued that the Doyle letter should be considered exempt from disclosure under MCL 15.243(l)(m) because it was an advisory communication from a subordinate regarding a superior, prehminary to a “final determination of action” by the Board, and the public interest in encouraging frank communication between officials and employees of the University clearly outweighed the public interest in disclosure.5 The Board also argued that its publication of “a voluminous and exhaustive report on the investigation into the University House controversy,” prepared by an independent auditing firm, Deloitte, weighed against disclosure of the Doyle letter. The Board asserted that all the facts had been released and were part of the public record, but that the opinions and personal views of Doyle, which were part of the deliberative process of the Board, should be protected from disclosure.

*191The trial court held a hearing, reviewed the Doyle letter in camera, denied Herald’s motion to compel disclosure of the Doyle letter and granted summary disposition in favor of the Board, and held that the letter fell within the FOIA exemption provided by MCL 15.243(l)(m). The trial court stated:

In the opinion of the Court, Defendant has sufficiently articulated a particularized justification for exemption under [MCL 15.243(l)(m)]. Based on its in camera review of the letter, the Court finds that: (1) the contents are of an advisory nature and cover other than purely factual materials; (2) the communication was made between officials and/or employees of public bodies; and (3) the communication was preliminary to a final agency determination of policy or action.

Although the document contains some “factual material,” it is primarily a summary of events from Doyle’s perspective. Any factual material contained in the letter is not easily severable. Doyle clearly exercised judgment in selecting the factual material, evaluating its relative significance, and using it to facilitate the impact of his opinions. See, Montrose Chemical Corp v Train, 491 F2d 63 (DC Cir, 1974) (Federal Court held that two factual summaries of evidence developed at a hearing before the Administrator of the EPA were exempt under a parallel provision of the federal FOIA). Further, under recent persuasive Michigan authority, a court may determine that a particular document that contains “substantially more opinion than fact” falls within the exemption. Barbier v. Basso, 2000 WL 33521028 [2000 Mich App LEXIS 2560].

The trial court further ruled that the letter was exempt from disclosure under “the parameters set forth in Herald Co, Inc v Ann Arbor Pub Schools”6 and made the following findings:

*192(1) The letter contains substantially more opinion than fact, and the factual material is not easily severable from the overwhelming majority of the comments: Doyle’s views concerning the President’s involvement with the University House project.

(2) The letter is preliminary to a final determination of policy or action. The communication was between officials of public bodies. The letter concerns [the Board’s] investigation and ultimate determination of what action, if any, would be taken regarding the University House controversy.

(3) The public interest in encouraging frank communications within the public body or between public bodies clearly outweighs the public interest in disclosure. [Herald’s] specific need for the letter, apparently to “shed light on the reasons why a respected public official resigned in the wake of [the University] being caught misleading the public as to the true cost of the President’s house,” or the public’s general interest in disclosure, is outweighed by [the Board’s] interest in maintaining the quality of its deliberative and decision-making process.

(4) [The Board] conducted an investigation and recently published a “voluminous and exhaustive report” concerning its findings regarding the University House project, a copy of which was furnished to [Herald].

This Court denied Herald’s motion for peremptory reversal, but granted its motion for immediate consideration and ordered this appeal to be expedited. This Court also directed the Board to file a copy of the Doyle letter with this Court and the Clerk to “suppress the letter from public view upon receipt.”7

III. STANDARD OF REVIEW

Our Supreme Court’s decision in Federated Publications,8 provides the rule of law and the rationale for the *193appropriate level of deference we are to give to trial courts that conduct the difficult and fact-sensitive balancing tests under the FOIA. In an opinion authored by Justice MARKMAN, our Supreme Court observed that the standard of review for FOIA cases is not contained in the legislation itself, but in “our case law.”9 Specifically, the Court held that:

Exemptions involving discretionary determinations, such as application of the instant exemption requiring a circuit court to engage in a balancing of public interests, should be reviewed under a deferential standard. We therefore hold that the clearly erroneous standard of review applies to the application of exemptions requiring determinations of a discretionary nature. A finding is “clearly erroneous” if, after reviewing the entire evidence, the reviewing court is left with a definite and firm conviction that a mistake has been made. [Federated Publications, supra at 106-107 (emphasis added).]

Our Supreme Court in Federated Publications emphasized that as trial courts carry out the “public interest balancing,” each case, with its special facts, will implicate “differing public interest considerations.”10 Equally important, our Supreme Court ruled that “in undertaking this balancing, however, the circuit court must consider the fact that the inclusion of a record within an exemptible class ... implies some degree of public interest in the non-disclosure of such a record.”11 The Court further observed:

That is, some attribute of these records has prompted the Legislature to designate them as subject to disclosure only upon a finding that the public interest in disclosure predominates. [Id].[12]

*194In other words, our Supreme Court in Federated reasoned that although the FOIA’s disclosure policy serves the public interest in good governance, our Legislature made clear in the same legislation that the public interest in good governance may also be served by the nondisclosure policy illustrated by specific ex-emptible classes of records:

[I]n performing the requisite balancing of public interests, the circuit court should remain cognizant of the special consideration that the Legislature has accorded an exemptible class of records. [Id. at 110.]

Accordingly, the relevant inquiry under Federated Publications is whether the trial court’s ruling constitutes clear error.

IV ANALYSIS

Under federal and state freedom of information acts (FOIAs), the public has a broad right to inspect government documents, and the general policy promoted is one of “full disclosure.” Swickard v Wayne Co Medical Examiner, 438 Mich 536, 543; 475 NW2d 304 (1991). This right to review documents under FOIAs promotes the public interest in good government.13 Yet, our Legislature clearly determined that there are certain circumstances where revealing information would undermine rather than further good governance.14 Hence, the public’s right to view government documents is conditional, and FOIAs contain specific exemptions that qualify, and in certain cases, override the right to disclosure.

*195A. THE PURPOSE OF THE FRANK COMMUNICATIONS EXEMPTION

The quality of a governmental decision is only as good as the information that informs it, and, accordingly, it is widely recognized that the public has a strong interest in promoting frank communications between government officials, as evidenced by numerous federal and state laws that contain exemptions for information falling into this category.15

One example is the federal FOIA, which contains a broad exemption for “inter-agency or intra-agency memorandums or letters which would not be available by law to a party other than an agency in litigation with the agency....” 5 USC 552(b)(5). The United States Supreme Court articulated the reason for the frank communications exemption:

That Congress had the Government’s executive privilege specifically in mind in adopting Exemption 5 is clear. The precise contours of the privilege in the context of this case are less clear, but may be gleaned from expressions of legislative purpose and the prior case law. The cases uniformly rest the privilege on the policy of protecting the “decision making processes of government agencies,” .... The point, plainly made in the Senate Report, is that the “frank discussion of legal or policy matters” in writing might be inhibited if the discussion were made public; and that the “decisions” and “policies formulated” would be the poorer as a result. As a lower court has pointed out, “there are enough incentives as it is for playing it safe and listing with the wind,” and as we have said in an analogous context, “[h]uman experience teaches that those who expect public dissemination of their remarks may well temper candor with a concern for appearances... to the *196detriment of the decisionmaking process.” [Nat’l Labor Relations Bd v Sears, Roebuck & Co, 421 US 132, 150-151; 95 S Ct 1504; 44 L Ed 2d 29 (1975) (citations omitted).]

State courts, have expressed similar reasoning. The “deliberative process” exemption to New York’s FOIA “was enacted to foster open and candid discussion among public officials and to protect uninhibited recommendations, made within the family, from being scrutinized by those affected and by the public.” In the Matter of Shaw.16 In Shaw, the plaintiff, a high school referee, sued to obtain rating reports that had been compiled on him by high school coaches. The court held that the reports fell within the exemption and, thus, did not have to be disclosed:

It is not only preferable but imperative that the individual ratings remain private because disclosure would be extremely detrimental to the public interest. A public dissemination of the ratings would temper an honest and free evaluation with fear of reprisals and animosity and deter a proper decision.

In the instant case the rating process provides useful advisory opinions which would become meaningless or nonexistent if the cloak of confidentiality were to be removed. The coaches and officials would hesitate to participate in any rating process which would be made public and any rating madé under such circumstances would reflect more concern for its public acceptance than for its actual truth. The inevitable result would be an interference with the true sportsmanship of scholastic events and a detrimental impact upon the public’s interest and participation in public high school functions. The potential harm to the public interest far outweighs any possible benefit to the single participant. If disclosure is more harmful to the public than nondisclosure is harmful to the person seeking the information, the scales of justice must tip toward *197nondisclosure. Public welfare is more important than public knowledge. [Shaw, supra at 261-262 (citations omitted; emphasis added).]

B. THE MICHIGAN FRANK COMMUNICATIONS EXEMPTION

Michigan also recognizes that the public has a strong interest in promoting frank communications between government officials.

The Michigan Legislature determined that the public’s interest in promoting frank communications necessary to the proper functioning of government may, at times, outweigh the disclosure policy of the FOIA, and thus included a specific exemption in the FOIA for

[c] ommunications and notes within a public body or between public bodies of an advisory nature to the extent that they cover other than purely factual materials and are preliminary to a final agency determination of policy or action. This exemption does not apply unless the public body shows that in the particular instance the public interest in encouraging frank communication between officials and employees of public bodies clearly outweighs the public interest in disclosure.... [MCL 15.243(l)(m).]

This exemption explicitly recognizes that there are special cases in which nondisclosure better serves the public’s interest in good governance. The exemption forces courts to view the big picture and ask whether the public interest in the disclosure of a particular piece of information may be clearly outweighed by certain decision-making realities in which the disclosure would ultimately frustrate the goal of good governance.17

We note also that Michigan’s frank communications exemption is narrower than the federal exemption. The *198federal exemption contains an implicit presumption that the value of promoting frank communications is such that it outweighs the public’s right to know. However, the Michigan exemption is more limited: in order to prevent disclosure, the government must not only show that disclosure would inhibit frank communications, it must articulate why the promotion of frank communications, “in the particular instance,” “clearly” outweighs the public’s right to know.

Therefore, to conduct its analysis under MCL 15.243(l)(m), the trial court will ask and answer these questions: (1) did the public body show that the requested document covers “other than purely factual materials”; (2) did the public body show that the document is “preliminary to a final agency determination of policy or action”; and (3) did the public body “establish that the public interest in encouraging frank communications within the public body or between public bodies clearly outweighs the public interest in disclosure.” Ann Arbor Pub Schools, supra at 274, quoting in part MCL 15.243(l)(m). Herald concedes the first and second points, but challenges the University’s position and the trial court’s ruling on the third point.

C. THE “CLEARLY OUTWEIGHS” STANDARD

1. MICHIGAN

In McCartney,18 this Court balanced the applicable public interests and applied the “clearly outweighs” standard. In McCartney, the plaintiff sought the release *199of memoranda prepared by the Attorney General’s staff regarding the Governor’s negotiations with Indian tribes over casino rights. The defendant argued, among other things, that the memoranda were protected by the frank communications exemption. The Court agreed, and specifically affirmed the following argument:

“The large number of assistants and divisions, the diverse location of the divisions, the vast number of matters under consideration at any given moment, the pressure of court imposed deadlines, and the need to fully consider and evaluate various concerns make it absolutely essential that the Department of Attorney General utilize written memoranda as a means of communication to assist in decision making.

“The release to the public of the internal memoranda of the type at issue in this case would discourage the preparation of such memoranda and would impact negatively on the quality of the department’s decision-making process with detrimental effect on the legal services provided to state agencies as well as on the public’s interest.” [Id. at 734-735.]

This Court, in Favors v Dep’t of Corrections, 192 Mich App 131; 480 NW2d 604 (1991), also applied the clearly outweighs standard. The plaintiff, an inmate, sought to obtain a review form, which was used to determine disciplinary credits. The form contained a sheet used to record the committee’s comments, which were then used to make a final decision. This Court noted:

The comment sheet is designed to allow the committee members to state their candid impressions regarding the inmate’s eligibility for disciplinary credits. Release of this information conceivably could discourage frank appraisals by the committee and, thus, inhibit accurate assessment of an inmate’s merit or lack thereof. [Id. at 135.]

*200This Court held that the public interest in nondisclosure clearly outweighed the interest in disclosure:

[T]he public interest in encouraging frank communications within the Department of Corrections clearly outweighs the public interest in disclosure of these worksheet forms. The public has a clear interest in encouraging the members of disciplinary credit committees within the department to communicate frankly with a warden with regard to the issue of inmate disciplinary credit, an issue that affects the length of an inmate’s incarceration. The public has a far greater interest in insuring that these evaluations are accurate than in knowing the reasons behind the evaluations. [Id. at 136 (emphasis added).]

When, as here, the public body makes the proper showing that good governance is better served by nondisclosure than by disclosure, it will not be required to release the information. To make the proper showing, the public body must show that the information falls within the frank communications exemption and that nondisclosure clearly outweighs the public’s interest in disclosure.

McCartney demonstrates how and why this balancing of public interests may favor nondisclosure. The goal of the communications in McCartney was the provision of accurate legal advice, undeniably a matter of great importance. Likewise, the nature of the communications, legal advice, is a sensitive subject that normally requires confidentiality. Because the communications in McCartney were of a type generally recognized as requiring confidentiality and were directed toward an important goal, the public interest in nondisclosure greatly outweighed the interest in disclosure. Favors also shows how the specific nature of a communication can justify nondisclosure. If the committee members knew that the inmates would view their comments, they would understandably be less candid in their *201appraisal of the inmates. Furthermore, their candid comments were invaluable to the warden’s final determination: the warden could not be expected to keep track of and evaluate every inmate himself, thus he relied on the candid comments of the committee members.

2. CALIFORNIA

Another jurisdiction that uses a “clearly outweighs” standard is California. The California FOIA contains a provision the allows a public body to withhold disclosure of a document if “on the facts of the particular case the public interest served by not making the record public clearly outweighs the public interest served by disclosure of the record.” West’s Ann Cal Gov Code 6255. The court analyzed this frank communications exemption in Times Mirror Co.19 The plaintiff sought to obtain copies of the Governor’s appointment schedules. The Governor argued that disclosure would interfere with his decision-making process because “disclosure of the records in question, which identify where, when and with whom he has met, would inhibit access to the broad spectrum of persons and viewpoints which he requires to govern effectively.” Id. at 1339. The California Supreme Court first noted that the public had a strong interest in the disclosure of the schedules. “In politics, access is power in its purest form. Entrance to the executive office is the passport to influence in the decisions of government. The public’s interest extends not only to the individual they elect as Governor, but to the individuals their Governor selects as advisors.” Id. at 1344. The court also noted that public exposure could expand, rather than limit, the variety of people the *202Governor met with. Id. at 1345. With the goal of promoting good government, the court ultimately concluded:

The answer to these arguments is not that they lack substance, but pragmatism. The deliberative process privilege is grounded in the unromantic reality of politics; it rests on the understanding that if the public and the Governor were entitled to precisely the same information, neither would likely receive it. Politics is an ecumenical affair; it embraces persons and groups of every conceivable interest: public and private; popular and unpopular; Republican and Democratic and every partisan stripe in between; left, right and center. To disclose every private meeting or association of the Governor and expect the decision making process to function effectively, is to deny human nature and contrary to common sense and experience. [Id. (emphasis added).]

Thus, the court held that “the public interest in nondisclosure clearly outweighs the public interest in disclosure.” Id.

3. APPLICATION TO THE DOYLE LETTER

Because the goal of both the FOIA and its exemptions is good government, not disclosure for disclosure’s sake, our Legislature, by placing the frank communications exemption within the FOIA, made the policy judgment that “public welfare is more important than public knowledge.”20 That is, the public has a far greater interest in ensuring that boards of public universities provide effective oversight of the administration’s expenditure of public funds than knowing the opinions of one administrator about another. The Board needed more than cold and dry data to do its job, it needed the unvarnished candid opinion of insiders to make policy *203judgments and, particularly, to conduct sensitive investigations of top administrators. And, when a high-level administrator is asked to give his opinion of the highest ranking official in the administration, the president, his immediate superior, whose favor he needs for job security, the insider may be naturally reluctant to trust the outsider and to trust the confidentiality of the communication. Also, not unimportantly, the outside board member, in assessing the advisability of conducting further and more exhaustive investigations into alleged over-expenditures for the president’s residence, must assess the reliability, credibility, and validity of such communications. In other words, these frank communications are essential to an outside board’s ability to discharge its vital constitutional oversight function on behalf of the public. There is a substantial risk that these vital sources of candid opinions would dry up were insiders justifiably fearful that their candid appraisals would make front-page headlines. This is especially true where, as here, the Board is investigating potential misconduct of a high-ranking official and seeks the insight of other high-ranking officials who work for and side-by-side with the target of the investigation. The natural human tendency to “circle the wagons” or “play it safe,” coupled with apprehension of retaliation if the written opinion is made public, would, we fear, deprive the Board of an important perspective:

The point, plainly made in the Senate Report, is that the “frank discussion of legal or policy matters” in writing might be inhibited if the discussion were made public; and that the “decisions” and “policies formulated” would be the poorer as a result. As a lower court has pointed out, “there are enough incentives as it is for playing it safe and listing with the wind,” and as we have said in an analogous context, “[h]uman experience teaches that those who expect public dissemination of their remarks may well tern-*204per candor with a concern for appearances ... to the detriment of the decisionmaking process.” [Sears, Roebuck & Co, supra at 150 (internal citations omitted).]

To make Doyle’s letter public would likely hurt, not advance, the public interest. It would, in this context, kill the goose that laid the golden egg, because, to paraphrase the California Supreme Court, if the public and the Board are entitled to receive exactly the same information, then neither would likely receive it. See Times Mirror Co, supra at 1345.

Also important to our decision is the uncontroverted fact that the Board acted in fulfillment, not in derogation, of its constitutional role. That is, the Board investigated and reported to the public, it did not conceal and sweep the issue under the rug.21 Had this been a case in which the president himself concealed documents to hide his alleged misconduct, with the complicity of the Board, then the balancing of public policy interests and the calculus of decision making would clearly weigh in favor of disclosure. But, where, as here, a board needs insiders’ opinions to investigate other insiders to protect the use of public funds and, where that board honorably discharges its obligations, the public interest in nondisclosure clearly predominates. Indeed, this factual scenario strikes us as the prototype the Legislature had in mind when it adopted the frank communications exemption in the FOIA. The express recognition by the Legislature of the need for candor and its vital role in internal decision making and *205internal investigations22 gave birth to the frank communications exemption, and, were we to hold this exemption inapplicable under these facts, this may very well sound the death knell of this vital tool for board members to discharge their oversight roles for the benefit of the public.

D. THE "CLEARLY ERRONEOUS” STANDARD OF REVIEW

Because we agree with the trial court that the public interest in protecting frank communications clearly outweighs the interest in disclosure, a fortiori, we conclude that the trial court did not commit clear error by so ruling. And because our Supreme Court instructs us to use the clearly erroneous standard when we review a trial court’s balancing judgment, we hold that the trial court did not clearly err in ruling that the public interest in nondisclosure predominates here. Indeed, the clearly erroneous standard was adopted by our Supreme Court to provide deference to trial courts that engage in precisely the type of balancing of public interests conducted here. Federated Publications, supra at 105-107. There is often a delicate balance between the public interest in disclosure and the public interest in nondisclosure. The trial court must make a careful appraisal of the special circumstances and all relevant facts to ensure that the correct balance is struck.23 Because the trial court is in a better position to hear testimony and review documents in camera and appraise the multiple factors that influence this balance, *206its determination should be accorded the appropriate deference reflected in a “clearly erroneous” standard of review. Id. at 107.24

The United States Supreme Court has given the following description of the application of the clearly erroneous standard of review:

Although the meaning of the phrase “clearly erroneous” is not immediately apparent, certain general principles governing the exercise of the appellate court’s power to overturn findings of a district court may be derived from our cases. The foremost of these principles, as the Fourth Circuit itself recognized, is that “[a] finding is ‘clearly erroneous’ when although there is evidence to support it, the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed.” This standard plainly does not entitle a reviewing court to reverse the finding of the trier of fact simply because it is convinced that it would have decided the case differently. The reviewing court oversteps the bounds of its duty under [F R Civ P] 52(a) if it undertakes to duplicate the role of the lower court. “In applying the clearly erroneous standard to the findings of a district court sitting without a jury, appellate courts must constantly have in mind that their function is not to decide factual issues de novo.” If the district court’s account of the evidence is plausible in light of the record viewed in its entirety, the court of appeals may not reverse it even though convinced that had it been sitting as the trier of fact, it would have weighed the evidence differently. Where there are two permissible views of the evidence, the fact finder’s choice between them cannot be clearly erroneous. [Anderson v Bessemer City, 470 US 564, 573-574; 105 S Ct 1504; 84 L Ed 2d 518 (1985) (citations omitted; emphasis added).]

Also, in colorful language adopted from the United States Court of Appeals for the Seventh Circuit, the *207Michigan Supreme Court has stated: “To be clearly erroneous, a decision must strike us as more than just maybe or probably wrong; it must... strike us as wrong with the force of a five-week old, unrefrigerated dead fish.” People v Cheatham, 453 Mich 1, 30 n 23; 551 NW2d 355 (1996), quoting Parts & Electric Motors, Inc v Sterling Electric, Inc, 866 F2d 228, 233 (CA 7, 1988).

V RESPONSE TO THE DISSENT

Among the many misstatements, misapprehensions, and mischaracterizations contained in the dissent, the most glaring flaw in the dissent’s reasoning is the dissent’s failure to properly apply the principles regarding the standard of review, enunciated by our Supreme Court in Federated Publications, to the trial court’s role in the balancing of public interests required by MCL 15.243(l)(m). While inaccurately accusing the majority of ignoring the “clearly outweighs” standard to determine when disclosure prevails over nondisclosure, the dissent ignores our Supreme Court’s express review limitations articulated in Federated Publications. That is, our Supreme Court in Federated made it abundantly clear that, not simply in that case, but in any case in which a trial court makes “discretionary determinations” involving “balancing of public interests,” we are not to disturb the trial court’s findings simply because we may disagree (as the dissent clearly does). Rather, we may overrule the trial court only when the trial court “clearly” errs. The dissent overstates the clearly outweighs standard under the FOIA beyond its intended meaning to accomplish the dissent’s purpose of overruling the trial court because it disagrees with the trial court. At the same time, to accomplish the dissent’s purposes here, the dissent relegates our Supreme Court’s mandated “clearly erroneous” standard to *208something much less than our Supreme Court intended. In doing so, the dissent falsely accuses the majority of positing a balance between disclosure for disclosure’s sake and good government. This is simply wrong. Rather, the majority simply makes the observation that it was not we, but our Legislature, that, by creating the frank communications exemption, determined that good governance in limited cases may be better served by nondisclosure than by disclosure in order to encourage the very kind of successful investigation that we witness here. Moreover, the dissent mistakenly accuses the majority of conflating the clearly erroneous standard with the abuse of discretion standard. The simple answer is that we do not conflate or confuse the two standards. Instead, we simply note the concrete fact that it was not we, but our Legislature, that determined that there are clear exceptions to disclosure when nondisclosure clearly outweighs the public interest in disclosure. And, equally important and compelling to our analysis is our Supreme Court’s holding and teaching in Federated that “exemptions involving discretionary determinations ... requiring a circuit court to engage in a balancing of public interests, should be reviewed under a deferential standard.”25 It is this admonition that the dissent ignores. And, contrary to the dissent’s hyperbolic accusations, we do not invent this standard of review. Rather, our Supreme Court simply articulated the appropriate standard of review in Federated. Simply because the balancing here requires the trial court to find that one interest “clearly” outweighs another does not render meaningless the obvious proposition that the trial court’s job — weighing one interest against another in light of all the facts of the particular case — remains one of conducting a balancing *209test. That the frank communications exemption carries with it a “clearly outweighs” mandate, which is obvious, nonetheless leaves the trial court with the discretionary job of weighing public interests and leaves us, as a reviewing court, with the obligation to review the trial court’s ruling using what Federated tells us is a “deferential standard.” As our Supreme Court makes clear in Federated, “some attribute of these records,” here records that fall within the category of frank communications, prompted our Legislature to give them “special consideration” — to make them subject to special treatment (unlike public records falling outside any exemptible class) as an “exemptible class of records.” We observe that, throughout the dissent, the dissent prefers to minimize the “clearly” in the clearly erroneous standard of review and to inflate the “clearly” in the clearly outweighs of the FOIA to effectuate the dissent’s objectives.

Moreover, the dissent, again inaccurately and unfairly, accuses the trial court and the majority of balancing the public interests and reviewing the trial court’s balancing decision, respectively, contrary to the legislative mandate, by ignoring the language, “in the particular instance.” To support this unfair characterization, the dissent accuses the majority of speculating about facts (which we do not) while the dissent itself speculates about the meaning of some of Doyle’s statements in his letter (speculation that is, in our view, naive).

Again, the dissent is simply wrong. The trial court and this Court each makes its respective ruling with the particular facts of this case at the center of the analysis. Indeed, in its opinion, the trial court said that defendant articulated “a particularized justification.” Further, the trial court specifically details its reasoning and *210its basis for its holding “in this particular instance.” Significantly, we conduct our review of the trial court’s ruling with special emphasis on this particular instance. Unlike the dissent, we cannot and do not speculate on: (1) why Doyle wrote what he did; (2) when he wrote the letter; (3) whether Doyle is credible to the Board in his opinions; (4) how the Board may have judged his credibility, reliability, or sincerity; or (5) what the Board may have known about the relationship between Doyle and the University president and how this affected its decision regarding further investigations. This is for the constitutionally mandated board to sort out, not for us. The Michigan Constitution gives the Board, not judges, the very difficult job of protecting the public interests by ensuring that public funds are properly spent. And here, there is no question that the Board was able to discharge its duty in no small part because of its ability to obtain the opinions and assessments of insiders about other insiders, a perspective that the Board may not have obtained absent the frank communications exception. The management of this very sensitive mix of an outside board, insiders’ opinions about other insiders, and the weighing of motivations and credibility in a delicate balancing of investigations is the constitutional charge of the Board, not judges. It is this delicate balancing of interests that creates the unique “particular instance” here that informed the trial court’s well-reasoned, correct, and most certainly not “clearly erroneous” decision under the frank communications exemption.

VI. CONCLUSION

In balancing the public interests, the trial court determined that the Board’s important, constitutional oversight function and investigative role, and thus, the *211public interest in good government, would be better served by nondisclosure rather than disclosure of the Doyle letter. In so finding, the trial court did not clearly err.

For all the foregoing reasons, we hold that the trial court properly granted summary disposition in favor of defendant.

Affirmed.

(dissenting). This case arises from the construction of a new official residence, University House, for the president of Eastern Michigan University (the University). Apparently, there was considerable public controversy regarding the expenditures associated with University House, and ultimately the president resigned, perhaps because of this controversy. In any event, it is clear from the record that University Regent Jan Brandon wrote a letter to Patrick Doyle, the University’s vice president for finance, asking Mr. Doyle to address several questions relating directly or indirectly to the construction of University House. On September 3, 2003, Mr. Doyle responded by letter, and it is this communication (the Doyle letter) that is at issue here. Plaintiff Herald Company1 sought to obtain a copy of the Doyle letter, and the University denied the request on the basis of the “frank communications” exception2 of the Freedom of Information Act (FOIA).3 The Herald Company sued, and the trial court upheld the University’s denial, as does the majority here.

I respectfully dissent from the majority’s opinion. In construing the frank communications exemption of the *212FOIA the majority has posited a false choice between “good government” on the one hand and “disclosure for disclosure’s sake” on the other. The FOIA contains no such choice but, by reading it into the statute, the majority assures that the contents of the Doyle letter will remain secret. In the process, the majority ignores the concept of accountability that is so essential to the process of governing. It disregards the requirement in the frank communications exemption that the public body must show in the particular instance the public interest in encouraging frank communication between officials and employees of public bodies clearly outweighs the public interest in disclosure. It articulates what amounts to an abuse of discretion standard for appellate court review of FOIA cases. It speculates about what may occur in the future under the guise of construing the frank communications exemption while ignoring facts that are, in my view, outcome determinative in the particular circumstances of this case. And finally, relying on a New York case, it reaches the amazing conclusion that “public welfare is more important than public knowledge.” In the process, the majority overlooks the fundamental proposition that in a democracy public knowledge is essential to public welfare and ignores the explicit public policy statement in the FOIA that “[t]he people shall be informed so that they may folly participate in the democratic process.”4 For these reasons, I dissent.

I. OVERVIEW: ACCOUNTABILITY AND THE PROCESS OF GOVERNING

Chess is a game of complete information.5 In a chess game, each player looks at the board and sees the *213same information and that information is all that is available. By contrast, poker is a game of partial information. In a stud poker game, for example, all players have some information that they share equally — that is, knowledge of the cards that have been dealt face up — but each player also has some information unique only to that player — that is, knowledge of the cards that are in that player’s hand.

The game of poker is more analogous to real life than is the game of chess, which may account for poker’s significantly greater popularity. As individuals within a larger society, we rarely have exactly the same information and almost never do we have all the information that exists. The decisions that we make, therefore, may depend as much on past experience, on intuition, on context, and on our own value systems as they do on factual information.

The process of governing is a real life exercise and, while it is most certainly not a game, it is an exercise characterized by partial information. Rarely do individual citizens have the same information about governmental decisions. Almost never do such citizens have all the information that exists. In part, this is inevitable. Although the direct democracy of the town meeting still exists in a few areas, we now largely function within a representative form of government in which elected and appointed officials make decisions on our behalf without our participation and, indeed, often without our knowledge.

Nonetheless, as citizens we must be able to hold our elected and appointed officials accountable for the decisions that they make on our behalf. Accountability, in turn, depends on information; we cannot make an informed judgment about whether a decision of a gov-*214eminent official was the correct one without having at least some information about that decision. In 1976, the Michigan Legislature took a decisive step toward regularizing the access that citizens have to information about governmental decision-making and, thereby, toward ensuring accountability by elected and appointed governmental officials. That step was the passage of the FOIA.

The first section of the FOIA spells out a policy that would appear to be premised upon the concept of perfect information:

It is the public policy of this state that all persons, except those persons incarcerated in state or local correctional facilities, are entitled to full and complete information regarding the affairs of government and the official acts of those who represent them as public officials and public employees, consistent with this act. The people shall be informed so that they may fully participate in the democratic process.[6]

The mechanisms of the FOIA, however, do not actually result in the provision of full and complete information in all instances. Section 136 7 of the FOIA currently contains twenty-five discrete exemptions from the broad sweep of the act. The inclusion of such exemptions reflects a wholly realistic series of policy decisions by the Legislature that, sometimes, full disclosure would not advance the process of governing. Court after court, however, has said that these exemptions are to be construed narrowly.8

*215Further, there can be no question that the concept of accountability is central to both the broad policy and the implementing mechanisms of the FOIA.9 The FOIA, then, is a pro-disclosure statute that by its enactment sought to expand access to information in the hands of government officials. Thereby it allows the citizens of this state to hold those officials accountable for the decisions that they make on our behalf. While the Legislature did not, and could not, provide for complete access to information, it did significantly shift the balance away from restricted access to open access in all but a limited number of instances. The Legislature therefore necessarily made the decision that disclosure, except in a few instances, facilitates the process of governing because it incorporates the concept of accountability.

This was a deliberate, reasoned policy choice and one to which we in the judiciary should, in the process of judicial review, defer. In my view, the majority here *216exhibits no such deference. Rather, the majority substitutes its own view of proper policy — that the process of governing would be hindered in the context of the “frank communications” exemption by providing access to the Doyle letter — on grounds that are suspect at best when the actual language of that exemption is examined.

II. THE FRANK COMMUNICATIONS EXEMPTION

Section 13(l)(m)10 of the FOIA is the “frank communications” exemption. The frank communications exemption contains, first, a description of the public documents that are to be exempted and, second, a requirement for a necessary showing for the exemption to apply. The description of the public documents to be exempted provides that such documents must be (1) communications and notes within a public body or between public bodies, (2) other than purely factual materials, and (3) preliminary to a final agency determination of policy or action. The trial court found, and I agree, that the Doyle letter at issue here met each of these three prongs.

The “necessary showing” requirement is, however, another matter. Section 13(l)(m) states that “[tjhis exemption does not apply unless the public body shows that in the particular instance the public interest in encouraging frank communication between officials and employees of public bodies clearly outweighs the public interest in disclosure.” (Emphasis supplied.) Thus, the public body claiming exemption must show with particularity how the public interest in encouraging frank communications clearly outweighs the overall public interest in disclosure.

*217It is within the context of this language that I find the majority’s reliance on the “public interest balancing” mentioned in Federated Publications11 to be, at the very least, interesting. It is clear from the case law, including Federated Publications, that applying the FOIA, of necessity, requires the balancing of the interest in disclosure and the interest in nondisclosure. However, in the frank communications exemption, the Legislature, in a manner of speaking, put its thumb on the scale. The Legislature placed the burden squarely on the public body to show that the interest in nondisclosure clearly outweighs the interest in disclosure. In addition, the Legislature provided that this showing must be made in the particular instance. Thus, in the frank communications exception the competing interests in nondisclosure and disclosure do not stand on equal footing. Rather, the Legislature has weighted the balance in favor of disclosure.

It follows that it is not enough to state that there is a public interest in the nondisclosure of communications and notes within a public body or between public bodies that contain other than purely factual materials and that are preliminary to a final agency determination of policy or action. The Legislature has already made such a determination and it is a given. Merely repeating that given advances the analysis not at all. The issue here is whether the interest in nondisclosure clearly outweighs the competing interest in disclosure in this particular instance. In my view, the majority skirts this issue, in the process conflating two considerably different standards of review.

*218III. STANDARD OF REVIEW12

The majority states in its section on the standard of review that the applicable standard is whether the trial court’s ruling constitutes clear error. Curiously, later in its analysis the majority revisits the standard of review. In its later analysis, the majority refers to Federated Publications to bolster its position that “the clearly erroneous standard was adopted by our Supreme Court to provide deference to trial courts that engage in precisely the type of balancing of public interests conducted here.” Ante at 205.

This is simply inaccurate, factually and logically. Federated Publications did not deal at all with the frank communications exemption nor with its explicit “clearly outweighs” standard. Rather, Federated Publications dealt with the FOIA exemption applicable to personnel records of a law enforcement agency.13 Therefore, Federated Publications did not deal at all with “precisely the type of balancing of public interests conducted here.” It dealt with a wholly different “equal footing” balancing scheme applicable to another, and wholly distinct, exemption in which the Legislature had not weighted the scales in favor of disclosure. As articulated in Federated Publications, and subject of course to the broad policy bias in favor of disclosure and to the narrow scope of the exemptions to disclosure in the FOIA, the interest in disclosure and the interest in nondisclosure in the law enforcement exception stand on something akin to equal footing. There is no such equal footing standard in the frank communications *219exception. That exemption has its own distinct and discrete “clearly outweighs” standard.

More broadly, there are three general categories of appellate review: de novo, clear error, and abuse of discretion. Federated Publications discussed the first two of these categories. It noted that the Supreme Court had, in some instances and without elaboration, applied a standard of review de novo to FOIA cases.14 However, Federated Publications limited review de novo to applications of FOIA exemptions involving legal determinations.15 In a footnote, the majority here propounds the theory that it would be an inefficient use of judicial resources to require appellate courts to review every FOIA challenge de novo. While I generally agree, I do note that Federated Publications appears to stand for the proposition that review de novo is required with respect to the applications of FOIA exemptions involving legal determinations.

Federated Publications does hold, squarely, that “the clearly erroneous standard of review applies to the application of exemptions requiring determinations of a discretionary nature” and that “[a] finding is clearly erroneous if, after reviewing the entire evidence, the reviewing court is left with the definite and firm conviction that a mistake has been made.”16

The majority, however, is not satisfied with this reasonably straightforward standard and quotes People v Cheatham,17 to the effect that to be clearly erroneous a *220decision must “strike us as wrong with the force of a five-week old, unrefrigerated dead fish.” Cheatam was a criminal case and, in writing it, Justice BOYLE noted that “[credibility is crucial in determining a defendant’s level of comprehension, and the trial judge is in the best position to make this assessment.”18 Credibility is, generally, not at issue in FOIA cases and most certainly not an issue in this FOIA case; the trial court here made its decision after a review in camera of the Doyle letter in which credibility determinations played no part. The majority does not explain why an admittedly colorful illustration of the clearly erroneous standard in a footnote in a criminal case that quotes a federal Circuit Court of Appeals decision in another circuit is of any assistance in understanding the clearly erroneous standard in a Michigan FOIA case that involves no credibility determinations whatsoever.

Beyond that, however, is the fact that the majority has in essence conflated the clearly erroneous standard with the abuse of discretion standard. Federated Publications did not discuss the abuse of discretion standard and, clearly, it has no application to FOIA cases. At its core, the abuse of discretion standard recognizes that in some circumstances a trial court is in a better position to make certain factual determinations and is therefore to be accorded considerable deference as “an acknowledgment of the trial court’s extensive knowledge of the facts and that court’s direct familiarity with the circumstances . .. .”19 The majority here seizes upon the word “deference,” and states that because of the trial court’s ability to “hear testimony and review documents in camera and appraise the multiple factors that influence *221this balance,” the trial court’s determination should be accorded “great deference.”

There were no credibility determinations involved in the trial court’s decision here. While the trial court reviewed the Doyle letter in camera, so have we. If there were other “multiple factors” that influenced the trial court’s balancing process, those factors are not discern-able from the trial court’s opinion or from the record in this case. By conflating the clearly erroneous standard with the abuse of discretion standard and, in essence, applying the latter, the majority has made the trial court’s decision virtually unreviewable. This is a far cry from a standard that requires us, in order to reverse, to review the entire evidence and come to a definite and firm conviction that the trial court has made a mistake. The deference that is due the trial court’s decision is the deference that flows from a careful review of the evidence and from a reasoned analysis of that decision, no more and no less. I suggest that it is this review that we should be conducting in this case. I further suggest that this is not the review that the majority has conducted.

IV THE “PARTICULAR INSTANCE” OF THIS CASE

A. THE MAJORITY’S VIEW

The majority addresses the particularized circumstances of this case in one very specific instance and then in a series of very broad statements. Specifically, the majority notes that the University’s Board of Regents honorably discharged its obligations. Presumably, the majority here refers to the undisputed fact that the University ultimately released a comprehensive report by the independent auditing firm that investigated the University House controversy. I agree that the University acted responsibly and in good faith in releasing this *222report. Were this the only factor bearing on this case, I would be inclined to affirm the trial court’s decision. Of course, this is not the only factor involved here. (I do note, however, that the situation here is not precisely the same as in the federal case of Montrose Chemical Corp of California v Train,20 a decision on which the majority relies. In Montrose, the court was faced with a situation in which all the facts concerning the matter at issue were in the public record and, therefore, the document that was being withheld was, to a considerable extent, redundant. Here, a review in camera of the Doyle letter plainly discloses that all the facts are not in the public record.)

The majority then offers a series of generalized policy statements in support of its view. (For example, “The natural human tendency to ‘circle the wagons’ or ‘play it safe,’ coupled with apprehension of retaliation if the written opinion is made public would, we fear, deprive the Board of an important perspective .... ” Ante at 203.) Ostensibly, these statements are related to the situation that the University’s Board of Regents faced here. However, these generalized concerns do not actually relate to the particular circumstances of this case; in fact, they express an overall view on proper public policy not with respect to this instance, but to future instances. But speculation about what may occur in the future is not our task when construing the frank communications exemption of the FOIA. By the language of that exemption, our task is to confine our inquiry to the “particular instance” of this case. If we limit our inquiry to the facts as they exist here, then I am at a loss to understand how the public interest in *223encouraging frank communications “clearly outweighs” the public interest in disclosure.

B. THE FACTS AS THEY EXIST HERE

The majority acknowledges, in passing, that the matter here involves the “administration’s expenditure of public funds.” To me, this fact is central to our consideration of this case. We are not dealing here, as we were in McCartney v Attorney General,21with legal memoranda that the Attorney General’s staff prepared regarding the Governor’s negotiations with Indian tribes over casino rights. We are not dealing here, as we were in Favors v Dep’t of Corrections,22 with a comment sheet used by a Department of Corrections disciplinary credit committee to determine whether to recommend the award of disciplinary credits. Rather, we are dealing with the direct expenditure of public funds — derived, we may reasonably assume, from a combination of taxpayer dollars and tuition payments — by the president of a major university for the construction of a residence in which he would live. Further, we are dealing with a situation in which there were allegations, confirmed at least in part by the University’s report, that these expenditures were extravagant and inappropriate. Thus, the question of the president’s accountability, not just to the University’s Board of Regents, but also to the taxpaying public, for these expenditures is at the core of this case.

The majority’s opinion keeps the Doyle letter, a document that was highly critical of the president, hidden from public view. It posits, in my view, a false *224choice between “good government” on the one hand and “disclosure for disclosure’s sake” on the other. There is no provision in the FOIA for disclosure for disclosure’s sake. Rather, there is the broad policy decision by a fully cognizant Legislature that disclosure, because it fosters accountability, facilitates good government. To hide the contents of the Doyle letter behind the facade of a Manichean choice between “good government” and the disclosure of arguably extravagant and inappropriate expenditures of public funds by a public official is not only to run from reality, it is to obscure the very existence of that reality.

The second fact central to our consideration of this case is that it is apparent that Mr. Doyle had decided to retire well before he wrote his letter to Regent Brandon and, as the trial court noted in its opinion, Mr. Doyle resigned several days after he wrote that letter. The majority’s concern that a high level administrator such as Mr. Doyle might be “naturally reluctant” to give his candid opinion of the “highest ranking official in the administration, the president, his immediate superior, whose favor he needs for job security,” ante at 203, is thus absolutely unfounded. Mr. Doyle could have no fears about his future job security, or about the president’s “favor,” because he had already decided to retire. Further, he had made that decision known to the president months before he penned his letter to Regent Brandon.23

In my view these facts determine the outcome in this case, for they exemplify precisely the sort of circum*225stances the Legislature commanded us to consider in the particular instance of an exemption claimed under the frank communications exemption to the FOIA. The majority avoids this conclusion by turning to case law from other states. It places heavy reliance on the California case of Times Mirror Co v Sacramento Co Superior Court,24 In that case, the Times Mirror sought disclosure of the Governor’s appointment schedules. The California Supreme Court ultimately denied that disclosure, stating: “The deliberative process privilege is grounded in the unromantic reality of politics; it rests on the understanding that if the public and the Governor were entitled to precisely the same information, neither would be likely to receive it.”25

I first note that the issue of access to a Governor’s appointment schedule simply could not arise in Michigan as the definition of a “public body” “does not include the governor or lieutenant governor, the executive office of the governor or lieutenant governor, or employees thereof.”26 Second, the California court fell into the same error as the majority here when it expounded its own view of proper public policy, which was based on speculation about what might happen in the future, while ignoring the language that the legislature had actually enacted.

The New York decision in In the Matter of Shaw27 exhibits the same hubris. At issue were rating reports of a high school referee that had been compiled by high school coaches. There, the court stated that “[a] public dissemination of the ratings would temper an honest *226and free evaluation with fear of reprisals and animosity and deter a proper decision.”28 The court reached the amazing conclusion that “[pjublic welfare is more important than public knowledge.”29 Remarkably, the majority here cites In the Matter of Shaw, a New York case, for the proposition that the Michigan Legislature had made the policy judgment that public welfare is more important than public knowledge. How a decision construing a New York statute can shed any light whatsoever on the intent of the Michigan Legislature in enacting the FOIA completely eludes me. My puzzlement is increased by the fact that neither the Michigan Legislature nor, to my knowledge, any court construing the FOIA in Michigan has ever reached the astounding conclusion that the public knowledge of the functioning of its government is overridden by the incantation of the phrase “public welfare,” a phrase that both the New York court and the majority here leave conveniently undefined. If this is the law in Michigan, then the FOIA is simply a dead letter.

V THE MAJORITY’S RESPONSE TO THIS DISSENT

The majority responds to my dissent in this case and I will do similarly, briefly. The majority’s response commences with the charge that there are “many misstatements, misapprehensions, and mischaracter-izations contained in the dissent....” Ante at 207. Such alliterative ruffles and flourishes neither require nor deserve a response. The majority’s view and my own are set out in the language of our respective opinions and I am content to let the chips fall where they may.

*227More substantively, the majority circles around the question of the standard of review at some length, with frequent references to Federated Publications.30 The majority view appears to have two components. The first is that Federated Publications articulates a “clearly erroneous” standard of review. I agree. Indeed, I say exactly that in the body of this dissent. I also point out, however, that Federated Publications dealt with the FOIA exemption applicable to personnel records of a law enforcement agency and not to the frank communications exemption at issue here. As the majority appears to concede, the frank communications exemption has its own “clearly outweighs” standard. Unless the specific language of the frank communications exemption is to be rendered entirely nugatory, this “clearly outweighs” standard, along with the requirement to take into account the “particular instance” of a case involving the frank communications exemption, must be part of the public interest balancing that Federated Publications requires.

The second component of the majority’s view appears to stem from the rather common-sense observation in Federated Publications that “[i]n contrast with the universe of public records that are non-exemptible, the Legislature has specifically designated these classes of records as exemptible.”31 Of course, the fact that the Legislature designated a class of records as exemptible does not end the inquiry. As the Supreme Court said, “[W]e emphasize that these records are merely exemptible and not exempt, and that exemption is not automatic.”32 And, I suggest, even when taking into account the Supreme Court’s following comment that a review*228ing court should remain “cognizant of the special consideration that the Legislature has accorded an exemptible class of records,”33 that special consideration can be overridden by a conclusion that the records should be made public when, as here, the public interest in encouraging frank communication does not clearly outweigh the public interest in disclosure.

In short, I do not see the conflict in emphasis on which the majority seizes. To me, the process is rather simple. Under Federated Publications, we are to review a lower court’s decision under a “clearly erroneous” standard. Under the language of the frank communications exemption, that review necessarily involves a special inquiry into whether the public interest in encouraging frank communications “clearly outweighs” the public interest in disclosure. The second inquiry is just as important as the first and neither can be disregarded. Indeed, in my view at least, the two inquiries constitute a seamless whole.34

In this regard, the majority states that I disagree with the trial court’s findings. Indeed, I do. But I do not simply disagree. After reviewing the entire evidence, I am left with the definite and firm conviction that the trial court made a mistake. As set out below, that mistake was in ignoring the special “clearly outweighs” standard contained in the frank communications exemption and thereby ignoring the fact that, with re*229spect to this particular exemption, the Legislature has made a policy decision that tilts the balance in favor of disclosure.35

VI. CONCLUSION

In its conclusion, the majority states:

In balancing the public interests, the trial court determined that the Board’s important, constitutional oversight function and investigative role, and thus, the public interest in good government, would be better served by nondisclosure rather than disclosure of the Doyle letter. In so finding, the trial court did not clearly err. [Ante at 210-211.]

I see nothing in the trial court’s opinion referring to the “important, constitutional oversight function and investigative role” of the University’s Board of Regents. However, I do agree that the trial court found in essence that nondisclosure of the Doyle letter would better serve the public than would disclosure. And it is for that precise reason that the trial court’s decision was clearly erroneous.

In its opinion, the trial court reached a general conclusion: “The public interest in encouraging frank *230communication within the public body or between the public bodies clearly outweighs the public interest in disclosure.” The trial court apparently recognized, however, that such a general conclusion, standing alone, could not carry the day. The trial court therefore went on to say:

Plaintiffs specific need for the letter, apparently to “shed light on the reasons why a highly respected public official resigned in the wake of EMU being caught misleading the public as to the true cost of the President’s house,” or the public’s general interest in disclosure, is outweighed by Defendant’s interest in maintaining the quality of its deliberative and decision-making process.

Obviously, the trial court was aware of the “clearly outweighs” standard. However, when analyzing the particular instance of the Herald Company’s FOIA request, it ignored that standard. Rather, the trial court simply balanced the interest in nondisclosure against the interest in disclosure and came down on the University’s side. In so doing, the trial court failed to recognize that, under the FOIA’s frank communications exemption, the interest in nondisclosure and the interest in disclosure do not stand on equal footing. With respect to this particular exemption, the Legislature has weighted the scales in favor of disclosure. Ignoring this legislative policy decision is the very definition of clear error.

The majority commits the same error. It states that, “[w]hen, as here, the public body makes the proper showing that good governance is better served by nondisclosure rather than by disclosure, it will not be required to release the information.” Ante at 200. Like the trial court, the majority is obviously aware of the “clearly outweighs” standard. Indeed, it quotes that standard in its very next sentence. Like the trial court, *231however, it then simply ignores that standard. Like the trial court, it balances the supposed harm that may flow from disclosure against the supposed good that may flow from nondisclosure, in the future as a policy matter, without regard to the legislatively imposed mandate that requires consideration of the particularized instance of this case. Like the trial court, it overlooks the concept of accountability that is at the core of the FOIA. Like the trial court, therefore, it clearly errs.

In my view, this error is profound. The majority reaches the astounding conclusion that in Michigan the “public welfare” — defined without regard to the particular circumstances of this case — is more important than public knowledge. If this is the law of this state, then the Legislature’s broad policy decisions in the FOIA and its carefully tuned implementing mechanisms are without meaning. In the process, a narrowly tailored exemption from the broad sweep of the act will have swallowed the overall rule. Within the context of the frank communications exception, this consigns our citizens to the receipt of only that information that the public body determines it is safe, according to its definition of the public welfare, to release. I cannot agree that this is the result the Legislature intended. I would reverse and remand.