Justice MARTINEZ delivered the Opinion of the Court.

We granted certiorari in this case to consider whether the governmental deliberative process privilege exists in Colorado. We hold that such a privilege does exist. We hold further that materials falling within the ambit of the deliberative process privilege are not subject to disclosure in the context of a request for public records under the Colorado open records laws, §§ 24-72-201 to — 309, 7 C.R.S. (1998) (the "open records laws"). Accordingly, we reverse the judgment of the court of appeals in White v. City of Colorado Springs, 950 P.2d 637 (Colo.App. 1997).

I.

Pursuant to the open records laws, respondent David White requested copies of certain materials in the possession of the Community Services Department of the City of Colorado Springs. The requested materials included a report generated by an outside consultant, Dr. Don Warrick, (the "Warrick Report") at the request of the head of the Community Services Department, Carla Hartsell. The report contained the results of an investigation of the Industrial Training Division, an entity under the supervision of the Community Services Department. The report was related to an internal evaluation of the Industrial Training Division.

Hartsell, as custodian of the Warrick Report, denied inspection of the report, asserting that the report was privileged under the governmental deliberative process privilege. Hartsell, however, did release a copy of the consultant contract between the City and Dr. Warrick which revealed the fees paid for his services. Pursuant to section 24-72-204(5), 7 C.R.S. (1998), White applied to the District Court of El Paso County for an order directing Hartsell and the City (the "Petitioners") to show cause why they should not permit inspection of the Warrick Report. The Petitioners maintained that inspection of the report was properly denied under section 24-72-204 (3)(a)(IV), 7 C.R.S. (1998), because the report was "privileged information" within the meaning of the statute. White asserted that the deliberative process privilege does not exist in Colorado, and thus the Petitioners had no basis to deny the inspection request.

After a hearing and in camera review of the Warrick Report, the trial court agreed with the Petitioners and discharged the order to show cause. The trial court found that the open records laws excepted information from inspection that is protected by the deliberative process privilege. The trial court found further that the Warrick Report was protected by the privilege because: (1) the report was "predecisional," (2) the report "contains information that is candid and personal from employees of the Industrial Training Division," and (3) "public disclosure of the report would chill honest and frank communications in the future."

The court of appeals reversed. See White, 950 P.2d at 639. The court acknowledged that federal authority recognizes a deliberative process privilege, but found "no corollary authority in Colorado law." Id. The court also noted that "[m]ost of the federal authority arises under a provision of the Freedom of Information Act [the "FOIA"], 5 U.S.C. § 552 (b)(5) (1994)." Id. Because the court found that the open records laws do not include a provision similar to the FOIA § 552 (b)(5), the court was further convinced that Colorado courts need not recognize the deliberative process privilege.

The court of appeals also looked to recent amendments to the open records laws in which the General Assembly exempted from the definition of public records "`work product prepared for elected officials.'" White, 950 P.2d at 639 (quoting § 24-72-202 (6)(b)(II), 10B C.R.S. (1996 Supp.)); see also § 24-72-202(6.5), 10B C.R.S. (1996 Supp.). The court concluded that these amendments amounted to creation of "a limited statutory deliberative process privilege." White, 950 P.2d at 639. Consequently, the court declined to recognize an "expansive" common law deliberative process privilege "where, as here, the General Assembly initially declined to create any such privilege at the time [the open records laws were] adopted and later created a limited statutory deliberative process privilege which does not cover the document in question." Id. The court of appeals remanded the case for entry of an order requiring the Petitioners to allow inspection of the Warrick Report.

Upon certiorari review by this court, the Petitioners claim that the deliberative process privilege is a common law evidentiary privilege that exists independently of either the open records laws or the FOIA. Thus, the Petitioners contend that the fact that the open records laws do not contain language identical to the FOIA is not dispositive of whether the privilege exists in Colorado. The Petitioners assert that the policies supporting the privilege in the context of the federal government are equally applicable to state government. Further, the Petitioners allege that the deliberative process privilege, and the policies which support its existence, have already been recognized under a different name by this court in Martinelli v. District Court, 199 Colo. 163, 612 P.2d 1083 (1980). Finally, the Petitioners maintain that the trial court correctly found that the Warrick Report qualifies for protection under the deliberative process privilege.

II.

Essential to the question of whether the deliberative process privilege exists in Colorado is an understanding of the origin and purposes of the privilege. Thus, we will first discuss the development and rationale of the deliberative process privilege. Next, we will examine the role of the privilege in Colorado.

A.

The deliberative process privilege is unique to the government. See Coastal States Gas Corp. v. Department of Energy, 617 F.2d 854, 866 (D.C. Cir. 1980). It is a widely recognized confidentiality privilege asserted by government officials. See Capital Info. Group v. Alaska, 923 P.2d 29, 33 (Alaska 1996) (citing Natalie A. Finkelman, Note, Evidence and Constitutional Law, 61 Temp. L.Rev. 1015, 1033 (1988)). The privilege rests on the ground that public disclosure of certain communications would deter the open exchange of opinions and recommendations between government officials, and it is intended to protect the government's decisionmaking process, its consultative functions, and the quality of its decisions. See id.

According to some commentators, the deliberative process privilege originated in the eighteenth and nineteenth centuries within the concept of the English "crown privilege." See Russell L. Weaver James T.R. Jones, The Deliberative Process Privilege, 54 Mo. L.Rev. 279, 283 (1989) (hereinafter, "Weaver Jones, The Deliberative Process Privilege"). Early American cases recognized a privilege derived, to some degree, from the crown privilege. See id. at 284 n. 29. Although these cases did not recognize the deliberative process privilege by name, the cases "did protect materials of the sort which the deliberative process privilege now encompasses." Id. at 285. Those early cases that are generally considered critical to the development of the privilege in this country, and to our discussion, include Morgan v. United States, 304 U.S. 1, 18, 58 S.Ct. 773, 82 L.Ed. 1129 (1938), Kaiser Aluminum Chemical Corp. v. United States, 141 Ct.Cl. 38, 157 F. Supp. 939 (Ct.Cl. 1958), and Carl Zeiss Stiftung v. V.E.B. Carl Zeiss, Jena, 40 F.R.D. 318 (D.D.C. 1966). See, e.g., NLRB v. Sears, Roebuck Co., 421 U.S. 132, 149-50, 95 S.Ct. 1504, 44 L.Ed.2d 29 (1975); Capital Info. Group, 923 P.2d at 34; Weaver Jones, The Deliberative Process Privilege, at 286-88.

The Court in Morgan encountered a challenge to the method used by the Secretary of Agriculture to set stockyard rates. The Court held, "[I]t was not the function of the court to probe the mental processes of the Secretary in reaching his conclusions." Morgan, 304 U.S. at 18, 58 S.Ct. 773. This protection for the mental processes of government decisionmakers was built upon in Kaiser. That case involved a discovery request for production of a "confidential intra-office advisory opinion" of the kind "that every head of an agency must rely upon for aid in determining a course of action." Kaiser, 157 F. Supp. at 945-46. Drawing an analogy to the "mental processes" rule described in Morgan, the Kaiser court held that the document in question belonged to the class of governmental documents privileged from inspection as "part of the administrative reasoning process." Kaiser, 157 F. Supp. at 946.

In Carl Zeiss Stiftung, the court recognized a "well-established" evidentiary privilege for "intra-governmental documents reflecting advisory opinions, recommendations and deliberations comprising part of a process by which governmental decisions and policies are formulated." 40 F.R.D. at 324. The court found support for this privilege in the "policy of frank expression and discussion among those upon whom rests the responsibility for making the determinations that enable government to operate." Id. The court held that the privilege also rests upon the rule, discussed in Morgan, that the judiciary is not authorized to probe the mental processes of an executive or administrative officer. See id. at 325. The evidentiary privilege for intra-governmental opinions and deliberations, the court found, is necessary to the integrity of the administrative process. See id. at 326.

In the cases discussed above, the courts termed the privilege at issue the "executive" privilege. Traditionally, the executive privilege has at least two components or branches: one based in the constitution and the other based in the common law. See, e.g., Sears, 421 U.S. at 150-51, 95 S.Ct. 1504 (noting that executive privilege has both constitutional and common law content); Vaughn v. Rosen, 523 F.2d 1136, 1146 (D.C. Cir. 1975) ("Vaughn II"); Capital Info. Group, 923 P.2d at 34; Hamilton v. Verdow, 287 Md. 544, 414 A.2d 914, 924 (Md. 1980); Walker Jones, The Deliberative Process Privilege, at 288-90. The constitutionally based privilege is also known as the "state secrets" or "presidential communications" privilege, and is based on the principle of separation of powers. See United States v. Nixon, 418 U.S. 683, 703, 94 S.Ct. 3090, 41 L.Ed.2d 1039 (1974); United States v. Reynolds, 345 U.S. 1, 6-8, 73 S.Ct. 528, 97 L.Ed. 727 (1953); In re Sealed Case, 121 F.3d 729, 736 742 (D.C. Cir. 1997); see also Walker Jones, The Deliberative Process Privilege, at 279 (noting that deliberative process privilege "is a branch of executive privilege" distinct from the "presidential version" of executive privilege); Note, Discovery of Government Documents and the Official Information Privilege, 76 Colum. L.Rev. 142, 142 (1976) (hereinafter, "Note, Discovery of Government Documents") (discussing constitutional basis of "state secrets privilege").

As the United States Court of Appeals for the District of Columbia observed, in sorting the different facets of the executive privilege, the protection for the deliberative processes of government officials generally comes from "the common sense-common law privilege, i.e., the recognition that the Government cannot operate in a fish bowl." Vaughn II, 523 F.2d at 1146. "By `common sense-common law privilege,'" the court explained, "we mean what is usually referred to as `executive privilege,' shorn of any constitutional overtones of separation of powers." Id.; see also Sears, 421 U.S. at 150-51 n. 17, 95 S.Ct. 1504 (noting that common law executive privilege protecting "the decision making processes of government agencies" is distinct from constitutionally based executive privilege discussed in Nixon).

Hence, the deliberative process privilege has often been referred to as the common law executive privilege. Indeed, our review of those cases which discuss the common law privilege protecting the deliberative processes of government reveals that courts have employed several names to describe the privilege. The following names have been used interchangeably (if not always precisely): "executive privilege," "deliberative process privilege," "governmental privilege," and "official information privilege." See, e.g., In re Sealed Case, 121 F.3d at 736 (noting that "the most frequent form of executive privilege raised in the judicial arena is the deliberative process privilege"); Tax Analysts v. Internal Revenue Serv., 117 F.3d 607, 616 (D.C. Cir. 1997) (noting that deliberative process privilege is "a variant of executive privilege"); Wolfe v. Department of Health Human Servs., 839 F.2d 768, 773 (D.C. Cir. 1988) (stating that "[t]he common law discovery privilege at issue is the executive or deliberative process privilege"); Taxation with Representation Fund v. Internal Revenue Serv., 646 F.2d 666, 676-77 (D.C. Cir. 1981) (noting that deliberative process privilege is "sometimes called the `executive' or `governmental' privilege"); Rodgers v. Hyatt, 91 F.R.D. 399, 405 (D.Colo. 1980) (using "governmental" and "deliberative" privilege synonymously); Capital Info. Group, 923 P.2d at 34 (holding that "executive privilege . . . encompasses what other commentators have called the deliberative process privilege" and considering the terms synonymous); Times Mirror Co. v. Superior Court, 53 Cal.3d 1325, 283 Cal.Rptr. 893, 813 P.2d 240, 248 n. 10 (Cal. 1991) (noting that "[t]he terms `executive privilege' and `deliberative process privilege' refer to the same concept" and using terms interchangeably); Guy v. Judicial Nominating Comm'n, 659 A.2d 777, 782 (Del.Super.Ct. 1995) (noting "executive privilege" also referred to as "official information privilege" or "deliberative process privilege"); Hamilton, 414 A.2d at 920 (noting that "executive privilege" also referred to as "official information privilege"); Michigan Council of Trout Unlimited v. Department of Military Affairs, 213 Mich. App. 203, 539 N.W.2d 745, 751 (Mich.Ct.App. 1995) (noting that deliberative process privilege "is sometimes called the `executive privilege'"); 26A Charles A. Wright Kenneth A. Graham, Jr., Federal Practice and Procedure § 5680, at 125 (1977) (stating deliberative process privilege has been known by variety of names including "executive privilege"); Weaver Jones, The Deliberative Process Privilege, at 279 (calling deliberative process privilege a "branch of the executive privilege").

Although the privilege protecting the deliberative processes of government originated as a common law privilege, the privilege is most commonly encountered, at the federal level, in FOIA litigation. See In re Sealed Case, 121 F.3d at 737; Wolfe, 839 F.2d at 773. This stems from the fact that Congress incorporated the deliberative process privilege into the FOIA by way of "Exemption 5," 5 U.S.C. § 552 (b)(5) (1994). See, e.g., Sears, 421 U.S. at 150, 95 S.Ct. 1504; Environmental Protection Agency v. Mink, 410 U.S. 73, 86, 93 S.Ct. 827, 35 L.Ed.2d 119 (1973); Schell v. United States Dep't of Health Human Servs., 843 F.2d 933, 939 (6th Cir. 1988); Mead Data Cent., Inc. v. United States Dep't of Air Force, 566 F.2d 242, 256 (D.C. Cir. 1977); Vaughn II, 523 F.2d at 1146. Exemption 5 also incorporates other common law evidentiary privileges such as the attorney-client privilege and the attorney work product privilege. See Sears, 421 U.S. at 154, 95 S.Ct. 1504; Schell, 843 F.2d at 939; Mead Data Cent., 566 F.2d at 252. The "ultimate goal that Exemption 5 seeks to achieve is to prevent the quality of agency decisionmaking from deteriorating as a result of public exposure." Schell, 843 F.2d at 939; see also Sears, 421 U.S. at 151, 95 S.Ct. 1504.

In addition to the federal authorities, several state courts have recognized the privilege protecting the deliberative processes of government. See, e.g., Capital Info. Group v. Alaska, 923 P.2d 29, 33-34 (Alaska 1996); Times Mirror Co. v. Superior Court, 53 Cal.3d 1325, 283 Cal.Rptr. 893, 813 P.2d 240, 248-51 (Cal. 1991); Hamilton v. Verdow, 287 Md. 544, 414 A.2d 914, 924 (Md. 1980); Ostoin v. Waterford Township Police Dep't, 189 Mich. App. 334, 471 N.W.2d 666, 668 (Mich. Ct.App. 1991); Nero v. Hyland, 76 N.J. 213, 386 A.2d 846, 853 (N.J. 1978); State ex rel. Attorney Gen. v. First Judicial Dist. Court, 96 N.M. 254, 629 P.2d 330, 333-34 (N.M. 1981); Dorchester Master Ltd. Partnership v. Cabot Pipeline Corp., 137 Misc.2d 442, 521 N.Y.S.2d 209, 210-11 (N.Y.Sup.Ct. 1987); Killington, Ltd. v. Lash, 153 Vt. 628, 572 A.2d 1368, 1373-74 (Vt. 1990). These courts uniformly find support for the existence and scope of the privilege in the federal decisions discussing the executive or deliberative process privilege (e.g., Kaiser and Sears).

B.

This court has never explicitly recognized the deliberative process privilege by that name. In Martinelli, however, we accepted the common law evidentiary privilege "variously referred to as the `official information,' `governmental,' or `executive' privilege." 199 Colo. at 169, 612 P.2d at 1088. This privilege, we explained, "sanctions the nondisclosure by governmental agencies or instrumentalities of information, files, reports, or memoranda maintained by those agencies or instrumentalities, when `disclosure would be harmful to the public interest.'" Id. (quoting United States v. O'Neill, 81 F.R.D. 664, 666 (E.D.Pa. 1979)). We hold that the official information privilege recognized in Martinelli encompasses what is often termed the "deliberative process privilege."

That the terms "official information privilege," "executive privilege," "deliberative process privilege," and "governmental privilege" are used interchangeably and refer to the same concept is illustrated further by examining the authorities relied upon in Martinelli to support the existence of the privilege. One such authority observes, "The origins of the common law official information privilege can be traced to Kaiser Aluminum Chemical Corp. v. United States." Note, Discovery of Government Documents, at 156; see Martinelli, 199 Colo. at 169, 612 P.2d at 1088. " Kaiser is widely regarded as critical to the development of the deliberative process privilege." Walker Jones, The Deliberative Process Privilege, at 287 n. 38.

Another authority relied upon in Martinelli makes clear that the privilege for "official information" protects precisely the class of documents shielded by the common law executive or deliberative process privilege:

[T]here remains the question whether the department or agency head may properly invoke privilege as to materials in his official custody claimed by him not to be secrets of state but something less. The answer to that question is yes. . . . The privilege is not automatic; it is not held to exist merely because the document or other information happens to be in governmental custody. On the other hand, it has been held applicable to interdepartmental communications, reports of agents and subordinates, advisory recommendations and departmental records and files.

8 John H. Wigmore, Evidence § 2378, at 805 n. 21 (McNaughton rev. 1961) (citing Kaiser); see Martinelli, 199 Colo. at 169, 612 P.2d at 1088.

In Martinelli, we also looked to the case of Wood v. Breier, 54 F.R.D. 7 (E.D.Wis. 1972), as support for the official information privilege. See Martinelli, 199 Colo. at 169, 612 P.2d at 1088. The court in Wood dealt with "what is commonly termed executive privilege," and noted that the "privilege is generally recognized for `intra- and inter-agency advisory opinions and recommendations submitted for consideration in the performance of decision- and policy-making functions.'" Wood, 54 F.R.D. at 12 (quoting Freeman v. Seligson, 405 F.2d 1326, 1339 (D.C. Cir. 1968)).

Finally, we note that our discussion of the privilege at issue in Martinelli parallels the traditional distinction between the constitutional and common law components of the executive privilege. After acknowledging that we were addressing some form of executive privilege, we took care to "specifically exclude" from our discussion that form of the privilege "which relates to diplomatic or military secrets." Martinelli, 199 Colo. at 169 n. 1, 612 P.2d at 1088 n. 1. Our explanation of the limits of our holding further demonstrates that Martinelli concerned the common law executive privilege (i.e., the deliberative process privilege).

Thus, not only has the evidentiary privilege recognized in Martinelli been sometimes called the deliberative process privilege, but it derives from the same common law origins and encompasses the same policy concerns as the deliberative process privilege. Accordingly, we conclude that the deliberative process privilege is part of the common law of Colorado.

III.

In light of our recognition of the deliberative process privilege as part of the common law of Colorado, we next address the substantive requirements for the protection of the privilege. We will then discuss the appropriate procedure for asserting the deliberative process privilege.

A.

The deliberative process privilege is a qualified privilege. See, e.g., In re Sealed Case, 121 F.3d at 737. It applies in a particular instance if the purposes of the privilege are thereby served. The primary purpose of the privilege is to protect the frank exchange of ideas and opinions critical to the government's decisionmaking process where disclosure would discourage such discussion in the future:

[The privilege] serves to assure that subordinates within an agency will feel free to provide the decisionmaker with their uninhibited opinions and recommendations without fear of later being subject to public ridicule or criticism; to protect against premature disclosure of proposed policies before they have been finally formulated or adopted; and to protect against confusing the issues and misleading the public by dissemination of documents suggesting reasons and rationales for a course of action which were not in fact the ultimate reasons for the agency's action.

Coastal States, 617 F.2d at 866; see also Jordan v. Department of Justice, 591 F.2d 753, 774 (D.C. Cir. 1978).

Thus, a key question in a deliberative process privilege case is whether disclosure of the material would expose an agency's decisionmaking process in such a way as to discourage discussion within the agency and thereby undermine the agency's ability to perform its functions. See Schell, 843 F.2d at 940; Dudman Communications Corp. v. Department of Air Force, 815 F.2d 1565, 1568 (D.C. Cir. 1987); see also Martinelli, 199 Colo. at 171, 612 P.2d at 1089 (holding that whether privilege applies depends upon "the degree to which governmental self-evaluation and consequent program improvement will be chilled by disclosure"). In light of the purposes of the privilege, it protects only material that is both predecisional (i.e., generated before the adoption of an agency policy or decision) and deliberative (i.e., reflective of the give-and-take of the consultative process). See, e.g., In re Sealed Case, 121 F.3d at 737; Coastal States, 617 F.2d at 866.

Predecisional documents are protected because the "quality of a particular agency decision will clearly be affected by the communications received by the decisionmaker on the subject of the decision prior to the time the decision is made." Sears, 421 U.S. at 151, 95 S.Ct. 1504. Moreover, the public has only a marginal interest in the disclosure of "reasons supporting a policy which an agency has rejected, or with reasons which might have supplied, but did not supply, the basis for a policy which was actually adopted on a different ground." Id. at 152, 95 S.Ct. 1504; see also Taxation with Representation Fund, 646 F.2d at 677.

Conversely, postdecisional documents, communications made after the decision and designed to explain it, are not protected by the privilege. This distinction is supported by two considerations. First, the quality of a decision will not be affected by forced disclosure of communications occurring after the decision is finally reached. See Sears, 421 U.S. at 151, 95 S.Ct. 1504. Second, the public has a strong interest in the disclosure of reasons that do supply the basis for an agency policy actually adopted. See id. at 152, 95 S.Ct. 1504; Taxation with Representation Fund, 646 F.2d at 678. "These reasons, if expressed within the agency, constitute the `working law' of the agency," and as such should be disclosed to the public. Sears, 421 U.S. at 152, 95 S.Ct. 1504.

Material that is predecisional in nature normally retains its protection even after the decision is made. See May v. Department of Air Force, 777 F.2d 1012, 1014-15 (5th Cir. 1985); United States v. Capitol Serv., Inc., 89 F.R.D. 578, 583 (E.D.Wis. 1981); Walker Jones, The Deliberative Process Privilege, at 292-93. Because an agency's final decision is not protected under the privilege, however, predecisional material can lose its protected status if the decisionmaker incorporates the material by reference, or expressly adopts it, in the final decision. See Sears, 421 U.S. at 161, 95 S.Ct. 1504; Bristol-Meyers Co. v. Federal Trade Comm'n, 598 F.2d 18, 23 (D.C. Cir. 1978); Walker Jones, The Deliberative Process Privilege, at 293.

The predecisional/postdecisional dichotomy is a useful starting point in determining which documents are privileged. Of course, not all predecisional material is privileged; the material must also be part of the deliberative process by which a decision is made. See Taxation with Representation Fund, 646 F.2d at 678; Mead Data Cent., 566 F.2d at 257. The material must reflect the "give-and-take of the consultative process." Coastal States, 617 F.2d at 866. In determining whether material is deliberative, courts have distinguished between "advisory materials which truly reflect the deliberative or policymaking processes of an agency" and "purely factual, investigative material" which is not protected. Mink, 410 U.S. at 89, 93 S.Ct. 827; see In re Sealed Case, 121 F.3d at 737; Schell, 843 F.2d at 940; Wolfe, 839 F.2d at 774; see also Martinelli, 199 Colo. at 171, 612 P.2d at 1089 (holding that whether privilege applies depends upon whether information sought is factual data or evaluative summary).

While the fact/opinion distinction is helpful in determining what is privileged, courts should not "mechanically apply" the test in every case. Wolfe, 839 F.2d at 774; see also Schell, 843 F.2d at 940 (noting that fact/opinion distinction "should not serve as a talismanic incantation in every case"). The deliberative process privilege protects factual material that is "so inextricably intertwined with the deliberative sections of the documents that its disclosure would inevitably reveal the government's deliberations." In re Sealed Case, 121 F.3d at 737; see Mink, 410 U.S. at 87-91, 93 S.Ct. 827; Wolfe, 839 F.2d at 774.

Courts have also looked to other considerations in assessing whether material is predecisional and deliberative. The function and significance of the document in the agency's decisionmaking process are relevant. See Taxation with Representation Fund, 646 F.2d at 678; Coastal States, 617 F.2d at 868. Documents representing the ideas and theories that go into the making of policy, which are privileged, should be distinguished from "binding agency opinions and interpretations" that are "retained and referred to as precedent" and constitute the policy itself. Sterling Drug, Inc. v. Federal Trade Comm'n, 450 F.2d 698, 708 (D.C. Cir. 1971); see Coastal States, 617 F.2d at 859-60.

Furthermore, courts examine the identity and decisionmaking authority of the office or person issuing the material. See Taxation with Representation Fund, 646 F.2d at 679; Coastal States, 617 F.2d at 868. A document from a subordinate to a superior official is more likely to be predecisional, "while a document moving in the opposite direction is more likely to contain instructions to staff explaining the reasons for a decision already made." Coastal States, 617 F.2d at 868; see Taxation with Representation Fund, 646 F.2d at 679-81; Walker Jones, The Deliberative Process Privilege, at 291-92.

Finally, in addition to assessing whether the material is predecisional and deliberative, and in order to determine if disclosure of the material is likely to adversely affect the purposes of the privilege, courts inquire whether "the document is so candid or personal in nature that public disclosure is likely in the future to stifle honest and frank communication within the agency." Coastal States, 617 F.2d at 866; see also Van Aire Skyport Corp. v. Federal Aviation Admin., 733 F. Supp. 316, 320 (D.Colo. 1990). As a consequence, the deliberative process privilege typically covers recommendations, advisory opinions, draft documents, proposals, suggestions, and other subjective documents that reflect the personal opinions of the writer rather than the policy of the agency. See, e.g., Sears, 421 U.S. at 150, 95 S.Ct. 1504; Taxation with Representation Fund, 646 F.2d at 677; Coastal States, 617 F.2d at 866.

B.

The initial burden of proof falls upon the government entity asserting the deliberative process privilege. See Coastal States, 617 F.2d at 868; Mead Data Cent., 566 F.2d at 251; Vaughn II, 523 F.2d at 1146; Walker Jones, The Deliberative Process Privilege, at 300. Because the government has an obvious advantage over the party seeking the information, the requirements for assertion of the privilege are rather technical. The procedural requirements are intended to (1) assure that the party's interest in the information is not "submerged beneath governmental obfuscation and mischaracterization," and (2) allow the courts to effectively and efficiently evaluate the nature of the disputed documents. Vaughn v. Rosen, 484 F.2d 820, 826 (D.C. Cir. 1973) ( "Vaughn I"); see also Mead Data Cent., 566 F.2d at 251; Walker Jones, The Deliberative Process Privilege, at 303.

The government cannot meet these requirements by conclusory and generalized allegations of privilege. Most courts require that the assertion of the privilege be made in the form of a " Vaughn index," named for the decision which first imposed the requirement. See Vaughn I, 484 F.2d at 826-27; Walker Jones, The Deliberative Process Privilege, at 301. The Vaughn index requires a specific and detailed assertion of the privilege, although the index need not be so detailed that it compromises the purposes served by the privilege. See Church of Scientology v. United States Dep't of Army, 611 F.2d 738, 742 (9th Cir. 1979); Vaughn I, 484 F.2d at 826.

The Vaughn index should provide a specific description of each document claimed to be privileged. See Mead Data Cent., 566 F.2d at 251; Vaughn I, 484 F.2d at 826-27; Walker Jones, The Deliberative Process Privilege, at 301. Typically, the description should provide each document's author, recipient, and subject matter. See, e.g., Arthur Andersen Co. v. Internal Revenue Serv., 679 F.2d 254, 258 (D.C. Cir. 1982). Further, the index should explain why each document qualifies for the privilege, including descriptions of the deliberative process to which the document is related and the role played by document in that process. See, e.g., Mead Data Cent., 566 F.2d at 250-52; Walker Jones, The Deliberative Process Privilege, at 301-03. The government should also discuss, in the form of an affidavit, why disclosure of each document would be harmful. See Exxon Corp. v. Department of Energy, 91 F.R.D. 26, 43-44 (N.D.Tex. 1981); Smith v. Federal Trade Comm'n, 403 F. Supp. 1000, 1016-17 (D.Del. 1975); Walker Jones, The Deliberative Process Privilege, at 302. Finally, and especially in the case of a large document, the government should distinguish between those portions of the document that are disclosable (such as purely factual data) and those that are allegedly privileged. See Vaughn I, 484 F.2d at 827. The index should subdivide the document into manageable portions cross-referenced to the relevant parts of the government's justification for the privilege. See id.

Properly constructed, the Vaughn index should provide litigants with fundamental information about the allegedly privileged material, and provide them with a meaningful opportunity to challenge the government's claims. See Arthur Andersen, 679 F.2d at 258; Vaughn I, 484 F.2d at 825-26. Additionally, a proper Vaughn index is an aid to reviewing courts. An in camera inspection of the disputed material need not automatically follow upon the claim of privilege. See Kaiser, 157 F. Supp. at 947-48; First Judicial Dist. Court, 629 P.2d at 334; Walker Jones, The Deliberative Process Privilege, at 312-13. Where the government has met the procedural requirements for assertion of the deliberative process privilege, the privilege presumptively applies. See Kaiser, 157 F. Supp. at 947; Capital Info. Group, 923 P.2d at 37; Hamilton, 414 A.2d at 926-27; Killington, 572 A.2d at 1375; Walker Jones, The Deliberative Process Privilege, at 312-13. In such a case, in camera inspection of the documents is neither mandatory nor (where many documents are involved) practical. See Arthur Andersen, 679 F.2d at 258. The trial court should honor the claim of privilege unless the party seeking discovery makes a preliminary showing that the material may not be privileged or that there is some necessity for its production. See Kaiser, 157 F. Supp. at 947; Capital Info. Group, 923 P.2d at 37; Hamilton, 414 A.2d at 926-27; First Judicial Dist. Court, 629 P.2d at 334.

To reiterate, the deliberative process privilege is a qualified privilege. It may be overcome upon a showing that the discoverant's interests in disclosure of the materials is greater than the government's interests in their confidentiality. See, e.g., Martinelli, 199 Colo. at 170, 612 P.2d at 1088. The determination of need must be made flexibly on a case-by-case, ad hoc basis. See In re Sealed Case, 121 F.3d at 737. Factors relevant to this balancing include: the relevance of the evidence, whether there is reason to believe the documents may shed light on government misconduct, whether the information sought is available from other sources and can be obtained without compromising the government's deliberative processes, and the importance of the material to the discoverant's case. See In re Sealed Case, 121 F.3d at 737-38; Martinelli, 199 Colo. at 171, 612 P.2d at 1089; Walker Jones, The Deliberative Process Privilege, at 318-20.

In most cases, where the party seeking discovery makes a sufficient showing of need, the court should grant an in camera inspection. See Hamilton, 414 A.2d at 927; First Judicial Dist. Court, 629 P.2d at 334; Killington, 572 A.2d at 1376; Walker Jones, The Deliberative Process Privilege, at 312-13. Depending upon the circumstances, the court should use the inspection "to determine whether the material is privileged, to sever privileged from non-privileged material if severability is feasible, and to weigh the government's need for confidentiality against the litigant's need for production." Hamilton, 414 A.2d at 927.

IV.

We next determine that the deliberative process privilege may be asserted in response to an open records request. Finally, we apply the privilege to the facts before us.

A.

In the instant case, respondent White requested inspection of the Warrick Report pursuant to the open records laws. The Petitioners do not dispute the fact that the Warrick Report is a "public record" under the open records laws. See § 24-72-202 (6)(a)(I), 7 C.R.S. (1998). The Petitioners contend, however, that material protected by the deliberative process privilege is not subject to public inspection because section 24-72-204 (3)(a)(IV) exempts "privileged information." We agree.

In enacting the open records laws, the General Assembly "did not intend that the open records laws would supplant discovery practice in civil litigation." Martinelli, 199 Colo. at 177, 612 P.2d at 1093. The exemption for "privileged information" existed in the original draft of the open records legislation. See Colorado Legislative Council, Research Publication No. 126, Open Public Records for Colorado, at xvii (1967). This category of records "extends to various types of government-held information essentially private in nature." Id. The exemption for "privileged information" in section 24-72-204 (3)(a)(IV), therefore, includes various common law evidentiary privileges, such as the attorney-client privilege and the attorney work product privilege. See Denver Post Corp. v. University of Colo., 739 P.2d 874, 880 (Colo.App. 1987).

One of the most frequently asserted common law privileges for government-held information is the deliberative process privilege. See discussion supra Part II. Federal courts have found that Exemption 5 of the FOIA incorporates the attorney-client privilege, the attorney work product privilege, and the deliberative process privilege. The court of appeals has specifically found that section 24-72-204 (3)(a)(IV) incorporates the attorney-client privilege and the attorney work product privilege. See Denver Post, 739 P.2d at 880. Neither the FOIA nor the open records laws refer to these common law privileges by name, rather they protect material insulated by these privileges by general reference to discovery principles. Because the common law deliberative process privilege exists apart from the open records laws, these statutes need not identify it in detail in order to preserve the privilege in the public records context. See, e.g., Killington, 572 A.2d at 1375. We hold that the open records laws incorporate the deliberative process privilege in section 24-72-204 (3)(a)(IV).

Furthermore, we are not persuaded that recent amendments to the open records laws have abrogated this common law privilege. See White, 950 P.2d at 639. In 1996, the General Assembly amended the definition of public records to exclude various materials related to elected officials. See ch. 271, sec. 4, § 24-72-202, 1996 Colo. Sess. Laws 1479, 1481-82. These materials include "work product" prepared for elected officials, certain correspondence without a connection to the exercise of an official's functions, and confidential communications between constituents and elected officials. See id.

Statutes are not presumed to alter the common law unless they expressly, or by necessary implication, provide for such alteration. See Vaughan v. McMinn, 945 P.2d 404, 408 (Colo. 1997); Robinson v. Kerr, 144 Colo. 48, 52, 355 P.2d 117, 119-20 (1960). A statute is merely cumulative of the common law if the legislature intended not to interfere with preexisting rights, but to give additional relief. See Vaughan, 945 P.2d at 408. In the 1996 amendments to the open records laws, the General Assembly did not manifest an intent to alter the common law deliberative process privilege, either expressly or by clear implication. The exemptions from the definition of a public record created by the amendments are neither coextensive nor necessarily inconsistent with the deliberative process privilege. The amendments protect from public inspection certain materials, such as correspondence from constituents, that would not automatically be protected by the deliberative process privilege.

Moreover, nothing in the legislative history of these amendments suggests that the General Assembly contemplated the deliberative process privilege or the official information privilege (as it was termed in Martinelli). Instead, the legislative history demonstrates that the 1996 amendments were intended to extend the traditional notions of the attorney-client and attorney work product privileges to cover certain legislative materials. See Hearing on S. 96-212 Before the Senate Committee on Business Affairs Labor, 60th Gen. Assembly, 2nd Reg. Sess. (Mar. 11, 1996); Hearing on S. 96-212 Before the House Committee on State, Veterans, Military Affairs, 60th Gen. Assembly, 2nd Reg. Sess. (Apr. 9, 1996). See generally White, 950 P.2d at 639 ("the General Assembly [in the 1996 amendments] intended to limit, not expand, access to public documents"). Thus, the General Assembly has not expressed a clear intent to abrogate the deliberative process privilege; indeed, the history and effect of the 1996 amendments evidence an intent to expand, rather than to narrow, common law privileges with respect to legislative material. Accordingly, we find that the deliberative process privilege remains viable and applicable to a public records request.

As it does in the discovery context, the government entity asserting the privilege has the initial burden of proof in response to a public records request. The government must compile a proper Vaughn index with supporting affidavits. See, e.g., Mead Data Cent., 566 F.2d at 250-51 (applying Vaughn requirements in FOIA context). In contrast to the discovery context, however, the need of the party requesting disclosure is not relevant to a request for public records. See Sears, 421 U.S. at 149 n. 16, 95 S.Ct. 1504; In re Sealed Case, 121 F.3d at 737 n. 5; Mead Data Cent., 566 F.2d at 252 n. 14. The particular purpose for which one seeks the public record is not relevant in determining whether disclosure is required because the open records laws only require disclosure of materials which would be routinely disclosed in discovery. We find support for this conclusion in the mandatory language of the open records laws, which provides that "[t]he custodian shall deny the right of inspection" of privileged information. § 24-72-204 (3)(a) (emphasis added); cf. § 24-72-204 (2)(a), 7 C.R.S. (1998) (stating that "[t]he custodian may deny the right of inspection" of certain other records) (emphasis added). In the context of the FOIA, the Supreme Court has explained that:

[t]he ability of a private litigant to override a privilege claim set up by the Government, with respect to an otherwise disclosable document, may itself turn on the extent of the litigant's need in the context of the facts of his particular case; or on the nature of the case. However, it is not sensible to construe the [FOIA] to require disclosure of any document which would be disclosed in the hypothetical litigation in which the private party's claim is the most compelling. . . . Exemption 5 was intended to permit disclosure of those intra-agency memoranda which would "routinely be disclosed" in private litigation.

Sears, 421 U.S. at 149 n. 16, 95 S.Ct. 1504 (citations omitted).

Therefore, once the government has met its burden of proof by satisfying the procedural requirements, the privileged material is beyond public inspection.

B.

In the present case, the Petitioners asserted that the Warrick Report is protected by the deliberative process privilege, and is therefore exempt from inspection under section 24-72-204 (3)(a)(IV). The trial court responded to this assertion of privilege by taking testimony on the issue of the government's need for confidentiality and by conducting an in camera review of the Warrick Report. The transcript of the hearing on the order to show cause, as well as the text of the court's order, make clear that the court applied the proper legal standards in assessing the deliberative process privilege claim. The court found that (1) the report was "predecisional," (2) the report "contains information that is candid and personal from employees of the Industrial Training Division," and (3) "public disclosure of the report would chill honest and frank communications in the future." Accordingly, the court concluded that the report was protected by the deliberative process privilege.

Because we had not articulated the procedural requirements for assertion of the privilege at the time the Petitioners made their claim, we will not evaluate the claim for strict compliance with these requirements. Under these circumstances, we find that the trial court's in camera review was more than an adequate substitute for an evaluation of a Vaughn index. Like the trial court, we have conducted our own examination of the Warrick Report. We subsequently review the trial court's conclusions pursuant to an abuse of discretion standard. Cf. Bond v. District Court, 682 P.2d 33, 40 (Colo. 1984) (holding that, so long as trial court applies correct legal standard, "[d]iscovery rulings are ordinarily within the discretion of the trial court"). See generally In re Sealed Case, 121 F.3d at 740 (noting that appellate court ordinarily reviews trial court's ruling that privilege applies to documents "only for arbitrariness or abuse of discretion").

Our review of the Warrick Report confirms that it contains an evaluation of the working environment and policies of the Industrial Training Division. The report contains observations on the current atmosphere and suggestions on how to improve it. Thus, the report qualifies as predecisional in that it was designed to guide the Petitioners in developing strategies to improve the division. The report is also deliberative. The report is largely composed of employees' opinions as to the strengths and weaknesses of the Industrial Training Division and its administrator. The role of these opinions and observations was to assist the decisionmaking process rather than to serve as an expression of a final agency decision.

We also note that neither Dr. Warrick, the author of the report, nor the employees interviewed by him possessed the authority to promulgate final decisions for the agency. They could only offer suggestions to the true decisionmakers, and did so in the report. Furthermore, in order to promote frank and open discussion during an agency's deliberative process, the deliberative process privilege protects opinions and recommendations to a government agency by outside consultants so long as such opinions and recommendations are obtained during the agency's deliberative predecisional process. See Wu v. National Endowmnent for Humanities, 460 F.2d 1030, 1032 (5th Cir. 1972); Soucie, 448 F.2d at 1078 n. 44; Walker Jones, The Deliberative Process Privilege, at 300. Finally, the Warrick Report contains the employees' candid and personal views about the division and its administrator. As the trial court found, the knowledge that these views may be disclosed publicly is likely to discourage such frankness in the future.

Respondent White also argues that the Warrick Report does not qualify for the protection of the deliberative process privilege because (1) the Petitioners have not identified a specific policy or decision to which the report is connected and (2) the report contains the results of a survey and therefore contains disclosable factual material. Hence, White asserts that the trial court erred in finding that the report came within the ambit of the privilege. We disagree.

First of all, the government need not be able to point to a specific decision or policy in connection with which the material was prepared in order for the material to be considered predecisional. See discussion supra Part III.A. Although it is unnecessary to point to a specific decision or policy, we note that petitioner Hartsell's affidavit to the court identified "the evaluation and assessment of the Industrial Training Division and its Administrator in order to identify any current problems within the Industrial Training Division and to develop strategies to improve the division," as a specific deliberative process to which the Warrick Report was related.

Secondly, while it is true that the report contains the results of a survey of employees it does not follow that this material is necessarily factual in nature. As explained above the report is composed almost entirely of the employees' personal opinions. These contents can hardly be defined as purely factual material. We therefore reject White's arguments. We conclude that the trial court did not abuse its discretion in finding that the cause the report was protected by the deliberative process privilege.

V.

We hold that the deliberative process privilege is part of the common law of this state. We also find that the open records laws exempt material that comes within the scope of the privilege from public inspection. Because the court of appeals found that the privilege does not exist in Colorado, we reverse the judgment of that court.