Printer Friendly

The power to privilege.

A new and startling development has recently occurred in the law of delegation: Congress has for the first time expressly delegated to an administrative agency the power to write rules of privilege. Privileges abound in federal law, but until now, they have been defined either by statute or judicial opinion. The type of law that Congress has now authorized agencies to create--the regulatory evidentiary privilege--is a true novelty in our legal system.

This Article is the first to grapple with the implications of migrating the power to write rules of privilege from Congress and the courts, on the one hand, to the executive branch, on the other. It begins by describing an underappreciated aspect of the administrative state: the law of privilege is becoming increasingly important to the functioning of administrative agencies. As a result, administrative agencies are actively pursuing control over the law of evidentiary privilege to further their substantive mandates.

Granting agencies that sought-after control through a privilege delegation will imperil key federal and state regulatory and governance interests. First, privilege delegations will reduce agency accountability. A delegated authority to write privileges that enables an agency to shield its own communications from disclosure will allow the agency to insulate itself from external review and oversight. Second, privilege delegations will erode state interests in allowing litigants and the public broad access to information. Agencies promulgating regulatory evidentiary privileges are likely to displace state laws that would permit disclosure to a greater extent than would be the case if Congress and the courts retained the privilege pen. Third, privilege delegations threaten to undercut state sovereignty. When Congress authorizes federal agencies to privilege the communications of state officials, it obstructs the capacity of the states to monitor state agents and thereby produces a type of harm akin to prohibited congressional commandeering of state governance.

After establishing the risks attendant to privilege delegations, this Article offers some design principles that should govern the institution chosen to draft any new set of privileges that may be invoked by executive branch agencies and explains that the existing judicial rulemaking system fits well with these principles. Finally, this Article explains why this innovation in delegation provides a unique opportunity to test prevailing scholarly models of why and to whom Congress chooses to delegate. When it delegates the power to privilege to an agency, Congress is substituting a new delegate--a politically accountable executive agency--for an old delegate--the politically unaccountable federal courts. Accounts of delegation grounded in party competition have greater explanatory power for this swapping of delegates than alternative accounts.
     A. Common Law Privileges
     B. Statutory Privileges
     C. The Executive Rebuffed
     D. A Word on Some Mechanics
     A. Agency Enforcement
     B. Agency Coordination
     A. Accountability
     B. Preemption
     C. Commandeering
     D. A Summary


In 2010, Congress enacted sweeping reform of the health insurance market, cutting through decades of deadlock with a mammoth piece of legislation. Embedded within the Patient Protection and Affordable Care Act (ACA) (1) was a historically unprecedented provision. This provision augmented federal authority in novel ways. It threatened to encroach upon long-recognized state prerogatives. It placed federal agencies in an unfamiliar and intrusive regulatory role. And it entirely escaped scholarly and public attention.

The provision is not the "individual mandate" targeted by the Commerce Clause challenges to the ACA. (2) Rather, it is an amendment to the Employee Retirement Income Security Act (ERISA) made by section 6607 of the Act. The new provision authorizes the Secretary of Labor to promulgate regulations that "provide[] an evidentiary privilege for, and provide[] for the confidentiality of communications between or among" a host of federal and state entities, including the Treasury Department, the Department of Justice, state attorneys general, and an association of state insurance regulators with no official governmental status whatsoever. (3) The Department of Labor is also authorized to privilege communications between "[a]ny other Federal or State authority," as long as--in the Secretary's determination--the extension of the privilege is "appropriate" for the purposes of enforcing ERISA's employee benefit provisions. (4) The power to privilege entrusted to the Department of Labor is simple and startling: it is a wholesale delegation of the authority to craft regulatory evidentiary privileges covering communications between dozens of federal, state, and private entities.

In the ongoing cacophony of debate surrounding the ACA, section 6607 has been overlooked. (5) Yet this provision could (eventually (6)) prove to have a more sustained impact on public law than the individual mandate--which, when all was said and done, turned out to be merely a poorly phrased tax provision. (7) The type of law contemplated by section 6607--the regulatory evidentiary privilege--is a true novelty. Privileges abound in federal law, but they are defined either by statute or judicial opinion. Section 6607 bestows on federal regulators a power that they have never before held: the power to write rules of privilege from the ground up.

Many within the federal bureaucracy will no doubt welcome this innovation in delegation as long overdue. Equipped with the power to privilege communications between an agency and regulated entities, an agency could more easily induce regulated parties to cooperate with its investigations. An agency could also more comfortably coordinate its activities with other agencies, state entities, or private parties, if it could shield from disclosure its communications with these other entities through promulgating regulatory evidentiary privileges. In these and other contexts, a delegated power to write privileges could be valuable indeed.

Given the influence that federal agencies wield over the shape of federal legislation, it was perhaps only a matter of time before Congress enacted an express delegation of the power to promulgate privileges. Section 6607 is the first such delegation, but it probably will not be the last. Now is the time--before more such delegations are enacted--to think through the implications of migrating the power to write rules of evidentiary privilege from Congress and the courts, on the one hand, to the executive branch, on the other.

This contribution is valuable because no scholarly literature probes the intersection of the law of delegation and the law of privilege. Scholarship on delegation ignores privilege law, despite the fact that privilege law rests on a sweeping delegation of interpretive authority to courts. (8) And scholarship on privilege law starts from the widely shared initial premise that Congress and the courts, as opposed to executive agencies, will control the substance of federal evidentiary privileges. (9) As section 6607 shows, this premise is faulty. This law opens up the prospect of federal administrative agencies crafting new evidentiary privileges through the rulemaking process. This Article is the first to grapple with the ramifications of that scenario--a scenario that is no longer theoretical.

Delegating the power to write rules of privilege to an executive agency poses three risks. The primary threat is to agency accountability. Authorizing an agency to write rules to protect its own communications from disclosure is an invitation to mischief. Executive agencies resist compliance with open government laws, and they overutilize the mechanisms already available to them for shielding their own information. If an agency can write regulations to shield its own information and communications from exposure to the public or to adversaries in litigation, the transparency and accountability of government will decrease.

Authorizing agencies to write evidentiary privileges will cause trouble even if the resulting regulations apply only to the communications of parties outside the agency, such as private parties or state entities. The chief concern here arises from the likelihood that agencies will want to give their new regulatory privileges broad preemptive effect. Many substantive state laws are designed to ensure either public access to government information (e.g., open government laws) or litigant access to private information (e.g., rules of discovery). As much literature on administrative federalism suggests, federal agencies will likely prioritize achieving their substantive mandates over the federalism harms of preempting such state rules. (10) The result is likely to be a greater volume of privileges with a concomitantly greater degree of displacement of state law than if Congress and the courts retained the privilege pen.

A distinct harm to federalism--and one that may be more disturbing--is the prospect that a federal regulatory evidentiary privilege might shield the communications of state agents, such as state attorneys general or state insurance regulators, from disclosure to the public or to their state-level principals, such as governors or state legislatures. By obstructing the ability of the states to monitor their agents, privilege delegations could imperil not only state regulatory interests but also state governance interests.

After canvassing some necessary background, the discussion below elaborates on these problems with delegations of the power to privilege. It then leverages this critique to generate some institutional design principles that should govern the body charged with creating any new set of privileges that might be necessary to satisfy the needs of federal administrative agencies. The existing judicial rulemaking process fits well with these principles, and it could easily be charged with drafting new proposed privileges for Congress's consideration.

Of course, that is not the option that Congress chose when it enacted section 6607. The Article next turns to the broader puzzle of what can be learned from Congress's decision to undertake such a dramatic innovation in delegation. Section 6607 offers a unique opportunity to test the prevailing scholarly models that seek to explain why and to whom Congress chooses to delegate. That is because Congress did not merely select a delegate in section 6607; it swapped in a new delegate--a politically accountable executive agency--for an old delegate--politically unaccountable federal courts. This delegation swap, I suggest, is best understood by accounts of delegation that emphasize partisanship as a causal factor in the structuring of the administrative state rather than by alternative accounts of congressional choice of delegate.

The Article proceeds in five parts. Part I demonstrates the novelty of delegations of the power to privilege to the executive branch and walks through the legal mechanics of how such delegations would apply in federal or state cases raising federal or state claims. Part II describes how privilege law plays a central role in shaping the capacity of agencies to enforce the law and to coordinate their activities with other governmental and private actors. Part III explains why delegating the power to privilege to executive agencies will undercut agency accountability and erode important state regulatory and governance interests. Part IV lays out design principles that should govern the institutional process responsible for creating any new evidentiary privileges applicable to administrative agencies. Finally, Part V examines section 6607 in light of various competing accounts of why and to whom Congress chooses to delegate.


Where do privileges come from? (11) Answering this question requires a brief overview of the general statutory structure governing the federal rules of judicial procedure and of how this structure treats rules of evidentiary privilege.

The Rules Enabling Act gives the Supreme Court the power to write rules of procedure and evidence that govern in the federal courts. (12) Ordinarily, rules written by the Court take effect if Congress does not veto them within a certain time period. (13) The rules of evidentiary privilege are, however, carved out from this process. (14) Rules of evidentiary privilege take effect only if enacted by Congress; they do not become law by virtue of congressional inaction. (15)

At first blush, Congress's decision to place privilege law in this special category might seem to reflect a choice by Congress to retain more control over privilege law than it was choosing to retain over other procedural and evidentiary rules. (16) In fact, however, Congress delegated substantial power over this body of law to the federal courts. (17) In Federal Rule of Evidence 501, Congress provided that "[t]he common law--as interpreted by United States courts in the light of reason and experience--governs a claim of privilege," unless otherwise provided by the Constitution, a federal statute, or rule prescribed by the Supreme Court. (18) By thus letting common law decisionmaking by federal courts set the content of federal privilege law, Congress was able to avoid the difficult task of drafting a set of statutory privilege rules that would please the many powerful interest groups with a stake in the shape of federal privilege law. (19)

The upshot of Rule 501 is that evidentiary privileges in federal law today have two basic sources: (1) the common law as expounded by the federal courts and (2) the single material exception to Rule 501--federal statutory law. The other exceptions--for constitutional privileges and for privileges prescribed by rule--effectively collapse into these two categories. Constitutional evidentiary privileges are developed through common law decisionmaking by federal courts. (20) And because 28 U.S.C. [section] 2074 makes Supreme Court rules concerning evidentiary privilege ineffective unless approved by statute, the "rules prescribed by the Supreme Court" proviso in Rule 501 means, in effect, a privilege enacted by federal statute. Thus, evidentiary privileges are either statutory--i.e., created directly by Congress by positive law--or common law--i.e., developed by federal courts through precedential decisionmaking.

Notably absent is a corpus of privilege law created directly by the executive. Though the executive may claim privilege, it has historically not held any power to proclaim privilege. And, as explained in further detail below, where the executive has attempted to assert that it has been delegated the power to proclaim rules of evidentiary privilege, the courts and Congress have rebuffed those efforts.

A. Common Law Privileges

Working under the auspices of Rule 501, the federal judiciary has developed a robust body of privileges. (21) These include several privileges that are specifically or exclusively available to the executive branch, (22) as well as many privileges that are available to all litigants, including the executive branch. Either way, federal case law sets the metes and bounds of the privilege.

Consider the deliberative process privilege. Through common law decisionmaking, federal courts have defined this privilege to contain certain elements. The privilege covers "documents reflecting advisory opinions, recommendations and deliberations comprising part of a process by which governmental decisions and policies are formulated," (23) as long as the communications are predecisional. (24) Communications between an agency and certain sorts of extra-agency parties may be privileged, (25) while other communications may not be. (26) The privilege is qualified in the sense that a court may disregard it if the government's litigation adversary shows a strong need, (27) and it is limited in the sense that "if the government can segregate and disclose non-privileged factual information within a document, it must." (28) In all events, the government bears the burden of establishing the elements of the privilege. (29)

Other privileges have generated similar lines of cases. The goal here is not to provide an exhaustive catalog of these doctrines. Rather, the point is a simple one: federal decisional case law supplies the structure of many of the evidentiary privileges recognized by federal law.

B. Statutory Privileges

Statutes affecting confidentiality, secrecy, and privilege crop up in random contexts scattered throughout the U.S. Code. (30) For example, federal statutes provide protection for national security, defense, and diplomatic secrets; for the confidentiality of "required reports" submitted to federal agencies; and for the protection of government files on private individuals, such as tax returns. (31) But most of these statutes merely require confidentiality and do not create true evidentiary privileges. (32) For example, in Jicarilla Apache Nation v. United States, the Court of Federal Claims held that a provision of the Indian Mineral Development Act requiring the Department of Interior to hold mineral development information as "privileged proprietary information" of the affected Indian tribe did not create a discovery or evidentiary privilege but rather a requirement of confidentiality. (33) On that basis, the court ordered the Department of the Interior to produce the information to its litigation adversaries. (34)

Statutes that actually create privileges are very rare: "only a handful of statutes ... can be said to clearly fall within the exception in Rule (501)." (35) These stand-alone privileges were enacted by Congress to further specific policy purposes. A rare example of a statutory privilege was discussed in Baldrige v. Shapiro, a case that concerned statutory provisions that exempted from disclosure certain information collected by the Bureau of the Census. (36) The Supreme Court held that "[n]o discretion [was] provided to the Census Bureau on whether or not to disclose the information referred to in [those provisions]." (37) In evaluating whether these sections of the statute created a privilege, the Court reasoned that they "embody explicit congressional intent to preclude all disclosure of raw census data reported by or on behalf of individuals" and concluded that "[t]his strong policy of nondisclosure indicates that Congress intended the confidentiality provisions to constitute a 'privilege' within the meaning of the Federal Rules." (38) In other words, Congress had created a true privilege in the Baldrige statute.

The Supreme Court considered another stand-alone federal statutory privilege in Pierce County v. Guillen, (39) In that case, a federal statute, 23 U.S.C. [section] 409, shielded materials collected by state public works departments or agencies from discovery or admission into evidence in cases concerning accident sites or hazardous road conditions. (40) The Court held that the statute fell within the scope of Congress's Commerce Clause power, precluding the plaintiff from obtaining the materials he sought in his state court action. (41)

Other examples of statutory privileges address similarly diverse circumstances. (42) As the original Advisory Committee that drafted Rule 501 put it, "privileges created by act of Congress.... do not assume the form of broad principles; they are the product of resolving particular problems in particular terms." (43) The point here is not to catalog these privileges; it is only to show that Congress does, on occasion, involve itself in expressly articulating rules of privilege and that such privileges supplement those developed by the federal courts under Rule 501.

C. The Executive Rebuffed

In contrast to the courts and to Congress, the executive branch's power to privilege has never been on firm footing. For decades, some took the position that the Housekeeping Act authorized executive agencies to promulgate regulations limiting access to information in court. (44) Enacted in (1789), this prosaically titled statute gave executive officers of the federal government the authority to set up offices and maintain government files. (45) In 1900, however, the Housekeeping Act assumed a new importance after the Court's decision in Boske v. Comingore. (46) In Boske, a state tax collector sought to obtain federal tax records from a federal tax collector to use against an alcohol distillery. (47) Treasury regulations forbade the federal tax collector from producing those records to a state court. (48) On appeal from a contempt citation issued by the state court, the Supreme Court held that the federal tax collector could not be held in contempt because he had merely complied with valid Treasury regulations. (49) The Court held that the Secretary of the Treasury could require that all decisions concerning the use of department papers be reserved for his own determination; the opinion did not address whether the Secretary himself was authorized to resist subpoena. (50) Nonetheless, many executive branch officials construed Boske as authority to issue regulations that "privileged" information. (51) By the mid-twentieth century, agencies were routinely using regulations promulgated pursuant to the Housekeeping Act to decline to produce information in response to a subpoena or court order. (52)

The Supreme Court's 1951 decision in United States ex rel. Touhy v. Ragen53 offered oblique support for the executive's position. The facts of Touhy were quite similar to those in Boske. In Touhy, an executive branch official--an FBI agent--had refused to comply with a subpoena seeking records in his control because a Department of Justice regulation restricted their production. (54) The district court held the agent in contempt. (55) On appeal, the circuit court reversed, holding that the regulation was authorized by the Housekeeping Act and that the regulation "confers upon the Department of Justice the privilege of refusing to produce unless there has been a waiver of such privilege," which there had not been. (56) The Supreme Court affirmed, holding that that the trial court could not cite a subordinate executive department official for contempt when he had no option but to comply with the Department of Justice regulation. (57) The Court refused to address whether a department head possessed the authority to refuse a court's order to produce government papers in his possession. (58)

Following Touhy, executive agencies wrote scores of regulations limiting disclosure of government information in response to subpoenas. Federal agencies routinely relied upon these regulations, usually described as "Touhy regulations," to avoid complying with federal discovery requests. (59) Challenging the invocation of Touhy regulations was an arduous enough endeavor that "Touhy temporarily created a privilege-in-effect." (60)

From its inception, this practice faced sharp criticism. The Federal Rules of Civil Procedure, adopted in 1934, had made the government subject to ordinary discovery procedures. (61) Courts began to insist more frequently that "the judiciary, and not the executive branch, possessed the ultimate authority to evaluate privilege claims" (62) and began to reject Touhy regulations that would otherwise block discovery requests aimed at the government.

Congress, too, was far from indifferent. In 1958, the House of Representatives determined that the Housekeeping Act had become a "convenient blanket to hide anything Congress may have neglected or refused to include under specific secrecy laws." (63) To address this problem, Congress amended the Housekeeping Act in 1958 to state specifically that the act "does not authorize withholding information from the public or limiting the availability of records to the public." (64) Notwithstanding this express curtailment of executive authority, Touhy regulations limiting disclosure of government records continued to proliferate after 1958--with the difference that agencies promulgated those regulations in reliance on other statutory sources of authority. Ironically, the other statutes that the executive relied upon were often enacted specifically in order to augment, not contract, government openness. (65) Courts and commentators have given these repackaged Touhy claims short shrift. (66) After the 1958 amendment, "[w]ith near unanimity.... those courts considering the issue have concluded that, when the United States is a party to the litigation, the reach of disclosure-limiting Touhy regulations ends at the courthouse doors." (67) The consensus view became that the "housekeeping privilege," if it ever existed, was defunct. (68)

In the modern era, sporadic claims have surfaced that one statute or another confers upon an agency the authority to privilege materials by promulgating regulations. These claims have not met with much success. (69) Courts have insisted that absent some clear indication of congressional intent, courts, and not agencies, must retain control over discovery because the power to determine what information the executive branch can withhold must remain subject to external checks. (70)

But although courts have rebuffed agency efforts to claim the power to write rules of privilege, one must take care to be precise about the nature of the rebuff. The rare court that has considered the question has not ruled out the possibility that agencies can hold the power to privilege by regulation. Instead, the judicial stance appears to have been more modest: it is that courts will not find that Congress has delegated the power to agencies without a crystal-clear congressional statement. (71)

The language of section 6607 passes any threshold of clarity that a court might reasonably apply. As a delegation, it is technically perfect; it grants, in haec verba, the power to promulgate evidentiary privileges to an administrative agency. (72) Before studying this innovation through a normative lens, (73) it is first necessary to cover some mechanics regarding the applicability of a regulatory evidentiary privilege in federal or state proceedings raising federal or state claims. The next Section tackles that task.

D. A Word on Some Mechanics

Section 6607, in clear terms, authorizes an agency to promulgate regulatory evidentiary privileges. What legal effect would such privileges have in federal and state court? (74)

It is useful to begin by examining the situation that will surely make up the lion's share of potential applications of a federal regulatory evidentiary privilege: the case in which a litigant seeks to invoke a federal regulatory evidentiary privilege in federal court with respect to a federal claim. An example would be a Freedom of Information Act (FOIA) lawsuit seeking disclosure of intra-agency materials. In such a suit, the federal common law of privilege incorporated via Exemption 5 of FOIA would normally govern whether a document was exempt from the Act. (75) If a newly coined federal regulatory evidentiary privilege applied to the materials not protected by the federal common law of privilege, the agency might seek to use the regulatory evidentiary privilege to fend off disclosure on the grounds that the federal regulatory evidentiary privilege, and not federal common law, controlled the privileged status of the sought-after records. For instance, the Secretary of Labor might promulgate a regulatory evidentiary privilege that shielded inter-agency communications that occurred after a policy decision was reached, rather than before a policy decision was reached, which is the time period spanned by the extant deliberative process privilege. (76)

How would a court assess such a claim? The starting point for the analysis is Federal Rule of Evidence 501. In federal civil cases on federal claims, the rule provides that "[t]he common law--as interpreted by United States courts in the light of reason and experience--governs a claim of privilege" unless a federal statute or other enumerated authority provides otherwise. (77) A statute explicitly authorizing a federal agency to promulgate an evidentiary privilege (such as section 6607) would likely qualify for this exception. As an influential treatise explains, "[w]here Congress creates a statutory privilege, then explicitly authorizes making of regulations governing scope and procedural incidents of privilege," the exception in Rule 501 for privileges provided by "Acts of Congress" should include "administrative regulations purporting to create a privilege." (78) A regulatory evidentiary privilege would, in other words, supersede the otherwise applicable federal common law of privilege where a statute expressly authorizes such a regulation to be made.

The next question is how a federal regulatory evidentiary privilege would apply in state court proceedings on state law claims. A state FOLA or sunshine act lawsuit could easily pose this question, as could a tort lawsuit in state court that sought discovery of communications between a regulated entity and its federal regulator. In the event that a federal regulatory evidentiary privilege more generously shielded materials than applicable state law, what law would a state court apply?

The starting point for analyzing this issue is the Supremacy Clause. (79) Because of the supremacy of federal law, federal regulations can preempt state law. (80) In addition, federal statutes regarding privileges can preempt contrary state law even in state court proceedings. (81) Taken together, these propositions suggest that a federal regulatory evidentiary privilege could, in principle, be written to govern claims of privilege in a state court proceeding on a state law cause of action. (82) Indeed, for reasons described in more detail below, if a federal regulatory evidentiary privilege did not have force in state court proceedings with respect to state law claims, it would be pretty much worthless; a parallel state suit could force disclosure of the information purportedly "shielded" by the federal regulatory evidentiary privilege. (83)

The next question is how a federal regulatory evidentiary privilege would apply to a state law claim in federal court. The starting point for analyzing this question will be how the language of the regulatory evidentiary privilege treats the state law proviso of Federal Rule of Evidence 501. That rule provides that in a federal civil case, the state law of privilege will govern "a claim or defense for which state law supplies the rule of decision." (84) But this proviso can be superseded by subsequent clear language. So, for example, the recently adopted Federal Rule of Evidence 502, which addresses waiver of attorney-client privilege, stipulates that it applies "notwithstanding Rule 501 ... even if state law provides the rule of decision." (85) A federal regulatory evidentiary privilege could likewise be drafted in a way that would supersede Rule 501's state law proviso. Of course, it is true that the Federal Rules of Evidence are a statute like any other enacted by Congress and therefore cannot be repealed or superseded by administrative regulation. (86) But when Congress enacts an express delegation of the authority to write rules of privilege to an agency, as it has here, it has effectively authorized that agency to write regulations that would amend the rules of evidence. Any other reading of the delegation would fatally weaken it: if a federal regulatory evidentiary privilege did not govern the privilege applicable to state law claims in federal court, then parties could force the disclosure of information shielded by the federal regulatory evidentiary privilege in routine diversity suits by relying on the (less protective) state privilege law. So long as a federal regulatory evidentiary privilege is drafted with sufficient clarity to achieve this result, there is no reason why it could not govern in federal suits where state law supplies the rule of decision.

The final scenario--and the one that is least likely to arise due to the likelihood of removal--is a federal regulatory evidentiary privilege's application in a state court suit raising a federal claim. If the federal regulatory evidentiary privilege is expressly made applicable in state proceedings, the governing law will be federal because of the Supremacy Clause. (87) Indeed, in this context it is likely that a state court would apply the federal regulatory evidentiary privilege even if the language of that privilege left some ambiguity as to its scope. This is because of the so-called reverse-Erie doctrine, (88) which governs the extent to which federal law is applicable in state courts. In this framework, a state court will apply federal substantive law. (89) Rules of privilege are generally regarded as substantive for conflict of laws purposes (90)--therefore, in jurisdictions that apply the law of the state with the "most significant relationship" to the dispute, the law of the forum with the most significant relationship to a privileged communication determines the existence and scope of the privilege. Even if this categorization were contestable, however, it is reasonable to expect that a state court would apply the federal law of privilege because in cases of doubt states operating under the reverse-Erie doctrine tend to defer to federal law. (91)

To fans of the two-by-two matrix, the foregoing discussion can be summarized in the following table:
A Federal Regulatory Evidentiary Privilege ("FREP") Would Apply

                    In Federal Court            In State Court

To Federal Claims   If the FREP meets FRE       Under the Supremacy
                    501's exception for "Acts   Clause and reverse-
                    of Congress."               Erie doctrine.

To State Claims     If the FREP is drafted to   Under the Supremacy
                    trump FRE 501's state       Clause.
                    law proviso.

Due to the novelty of privilege delegations, these questions have never been squarely presented in precisely the form they will be now. (92) And, as noted throughout, much will turn on the precise wording of a given regulatory evidentiary privilege and of a given delegation of the power to privilege. But in principle there is no obstacle to the results just summarized. Assuming the privilege and its authorizing delegation are drafted with sufficient clarity, a regulatory evidentiary privilege could govern claims of privilege in both state and federal court and on both state and federal claims. Before exploring the pathologies of such a delegation, the next part elaborates why obtaining such a delegation would be important to an executive agency.


Information is the "lifeblood" of modern government. (93) One of the fonts of legitimacy of the administrative state is its claim to expertise, which is to say informed decisionmaking. (94) For that reason, doctrines or practices that regulate the administrative state's ability to gather and disseminate information affect not only its functioning as a practical matter but also, at a deeper level, its capacity to claim legitimacy for the fruits of its decisionmaking processes.

This context suggests the first of two reasons why an agency would seek to gain control over the substantive law of evidentiary privileges. Fourth Amendment doctrine forms only a soft check on administrative information gathering from regulated entities. (95) It thus falls to privilege law to set hard limits on what information agencies may or may not procure from regulated entities. Conversely, open government laws such as FOIA create a default presumption that government information ought to be accessible to the public, while the Federal Rules of Civil Procedure similarly create a default presumption that relevant government information ought to be accessible to the government's litigation adversaries. It therefore also falls to privilege law to set the hard limits on what information the agency can be compelled to disclose to the public or to its opponents in court. The landscape of privilege law thus has a powerful effect on the functioning of administrative agencies.

Once the lens is broadened to take inter-agency interactions into account, another reason for the importance of privilege law comes into view. Privilege law determines how easily an agency can communicate with other state and federal agencies and with private parties. Absent a privilege shielding it, extramural communications by an agency may result in the waiver of any privileges that protect what is communicated and result in the potential disclosure of that information via discovery in litigation or through the operation of state or federal open government laws. The landscape of privilege law thus affects the capacity of agencies to communicate and coordinate with each other and with private parties.

Enforcement and coordination are both powerful administrative imperatives. Agencies thus are amply incentivized to influence aspects of privilege law. As explained below, obtaining a delegation of the power to privilege is a natural next item on the agency agenda.

A. Agency Enforcement

For an agency charged with enforcing federal law, one aspect of privilege law--whether and how privilege is waived--is critically important. (96) Like prosecutors, agencies frequently compel (or induce (97)) production of attorney-client privileged material from regulated parties under investigation. Agencies want to be able to control the consequences of that production--that is, they want regulated parties to be able to "selectively waive" privilege as to the agency without waiving the privilege as to third parties.

Why? As Professors Broun and Capra have explained, the answer is simple. Selective waiver has the potential to
   encourage targets of [an agency] investigation to cooperate more
   fully with the agency. The same encouragement would exist with
   regard to any agency investigation. Not only would selective waiver
   benefit the agency, it would relieve the target companies, which
   could comply fully with agency requests without the fear that their
   privileged documents would be used in private litigation. (98)

In short, a selective waiver power would facilitate an agency's efforts to enforce the law by encouraging regulated parties to cooperate fully with agency investigations. (99)

The difficulty is that most courts do not allow selective waiver as to government agencies. (100) As one court reasoned, permitting selective waiver would transform the oldest of common law privileges--the attorney-client privilege--into "merely another brush on an attorney's palette, utilized and manipulated to gain tactical or strategic advantage." (101) As a result, to assert that they have selective waiver power, agencies need to be able to point to a statute or a regulation that would authorize regulated parties to selectively waive privilege as to the agency. But very few statutes provide for selective waiver, (102) and these statutes apply to very limited and specialized contexts. (103) Outside these contexts, "a muddled patchwork of common law rules [applies to]... documents produced to other regulators," even those regulators cooperating with the regulators designated in these statutes. (104)

Other agencies that believe they would benefit from having selective waiver power have been actively seeking it. The most aggressive in its pursuit has been the Securities and Exchange Commission (SEC). (105) It is worthwhile to narrate the SEC's efforts to obtain selective waiver power in some detail because this story reflects both how important the power to privilege is to administrative agencies and also how valued a straightforward delegation of that power would be.

The SEC's campaign to secure selective waiver power dates back at least to the 1980s. (106) In 1984, the Commission supported enactment of a proposed amendment to the Securities and Exchange Act of 1934 that would have established selective waiver for any documents produced to the agency. (107) The amendment was referred to a House committee, which took no action. (108)

In 2002, the Commission proposed a regulation in which it simply gave itself selective waiver authority. (109) The regulation would have permitted selective waiver as to the SEC when the SEC and an issuer entered into a confidentiality agreement. (110) In the initial public comment period for the rule, the Commission took the position that Congress's general delegation of rulemaking power to the SEC authorized the agency to adopt a selective waiver rule.111 In its final rule, however, the agency dropped the selective waiver provision. It reasoned that courts were unlikely to accept the notion that Congress had implicitly delegated the power to write a selective waiver rule to the SEC. (112)

The agency's next move was to return to Congress and ask it to enact legislation that would give the SEC selective waiver power. (113) The bill, titled the Securities Fraud Deterrence and Investor Restitution Act of 2003, (114) would have amended the Securities Exchange Act of 1934 to authorize regulated parties to share information with the Commission without waiving work-product protection or attorney-client privilege over that material as to any third party. (115) As the SEC's director of enforcement testified to Congress, that provision would "help the Commission gather evidence in a more efficient manner by eliminating a strong disincentive to parties under investigation to voluntarily produce to the Commission important information." (116) He further explained that "[m]ore expeditious investigations could lead to more prompt enforcement actions, with a greater likelihood of recovery of assets to return to investors." (117) Despite the SEC's urgings, however, Congress again failed to enact a selective waiver law.

The SEC next tried to enlist an unusual ally: state courts. This was an uphill fight because virtually no state recognized selective waiver. (118) Nonetheless, "[i]n an attempt to influence the jurisprudence over selective waiver, the SEC has appeared as amicus [curiae] in a number of state court cases urging that defendants who produced materials to the SEC did not waive work-product privilege." (119) State courts were not, however, particularly receptive to the SEC's litigation campaign. (120) Moreover, even the agency's sporadic successes in court were of little use because they left the law nationwide in an unsatisfying state of nonuniformity. This patchwork privilege regime does not accomplish the agency's goal, which is to provide peace of mind to cooperating regulated entities. (121)

Finally, the SEC has also tried its hand at lobbying the federal judicial rulemaking process. In 2006 and 2007, the Advisory Committee on Evidence Rules met to discuss a new proposed rule of evidence, Rule 502, which would address waiver of attorney-client privilege. An early draft of the proposed Rule 502 included a broad selective waiver provision that preserved privilege over any disclosure in a federal or state proceeding "made to a federal public office or agency in the exercise of its regulatory, investigative, or enforcement authority." (122) The SEC, (123) as well as the Commodity Futures Trading Commission (CFTC),* (124) not only urged the Advisory Committee to include this "essential" government selective waiver provision in the final rule but also to amend the rule so that it would have preemptive effect over contrary state law. (125) The Advisory Committee was less confident of the provision's merits. It sent the government selective waiver provision to Congress in brackets to indicate that the Committee had no position on whether the provision ought to be adopted. (126) Ultimately, the government selective waiver provision was dropped entirely from the final rule enacted by Congress. (127)

This sequence of efforts by the SEC suggests that at least some agencies can take an ongoing active interest in influencing the development of privilege law in a direction that will aid their enforcement missions. It also suggests that agencies may encounter significant resistance when they pursue this goal using ordinary methods. Despite years of effort in lobbying Congress, judicial rulemakers, and state courts, the SEC has yet to achieve its ultimate objective of procuring a selective waiver power, let alone one that has preemptive effect on contrary state law. (128)

One can readily see how eagerly the Commission would welcome a delegation of the power to write rules of privilege via regulation. With such a delegation in hand, the Commission could cut directly to its endpoint and promulgate a regulatory evidentiary privilege shielding from disclosure any materials produced by regulated entities to the agency. Perhaps the most surprising thing about section 6607 is that when Congress chose to delegate the power to privilege to an agency, the delegate it selected was the Department of Labor, not the SEC.

B. Agency Coordination

A second reason that privilege law is increasingly important to agency enforcement is the steady rise of regulatory overlap, or of situations where multiple state and/or federal regulators are tending the same pot. (129) These are areas that Professors Freeman and Rossi have dubbed "shared regulatory space." (130) In such contexts, various government agencies at both the state and federal levels must coordinate their enforcement roles and regulatory agendas. (131)

It is not easy to maneuver in these shared regulatory spaces: "With the accretion of federal regulatory authority, the potential for conflicts between agencies, separately empowered by distinct statutory regimes, necessarily grows." (132) It is no longer adequate for one federal agency--say, EPA--to unilaterally make some goal--say, controlling the greenhouse gases emitted by automobiles--a regulatory priority. (133) Rather, EPA must consult other federal agencies with power in the relevant space, such as the National Highway Traffic Safety Administration (NHTSA). (134) State agencies may need or want a say as well. (135) As Professors DeShazo and Freeman explain, "[o]nce one peels back the skin of administrative decisionmaking, one finds not lone agencies making isolated decisions in a cocoon of bureaucratic insularity, but collections of agencies intervening in each other's decisionmaking processes, sometimes quite formally and sometimes less so." (136)

Inter-agency coordination at the federal tier is but one piece of a larger mosaic--a mosaic that reveals the hybridization of the administrative form. In the core agency functions of rulemaking, enforcement, and information gathering, new collaborative forms of regulation and governance are constantly developing. Sometimes Congress directs the creation of these hybrid forms through statute; (137) at other times, the hybrid forms come into being at the direction of the President. (138) Some hybrid regulatory forms involve private organizations (or "marketized bureaucracy"), (139) while others include organizations of public actors that are not quite private. (140) Others use state governments to act as surrogates or agents of the federal government. (141) Still others involve foreign governmental entities working in concert with U.S. federal entities. (142)

This world of hybrid administrative forms poses many new and exciting questions. One critical but underappreciated aspect of this world is the way that it is shaped by privilege law. (143) Consider the situation faced by the enforcement agencies that want selective waiver authority over information produced to them. As outlined above, enforcement agencies want to be able to receive information without forcing waiver of the regulated party's privilege. (144) But they also want to be able to transmit or share information with other regulators without waiving the regulated party's privilege shielding that information.

This poses a problem once one considers the sheer number of regulators that potentially have an interest in privileged material produced to one regulator. Consider, for example, the sphere of banking regulation. A conservative list of agencies involved with enforcing banking law would include the Federal Reserve Board, the Office of the Comptroller of the Currency (OCC), the Federal Deposit Insurance Corporation, the SEC, the CFTC, Office of Foreign Assets Control, Financial Crimes Enforcement Network, the Internal Revenue Service (IRS), and state and foreign bank supervisors. (145) "All of these regulators routinely cooperate... in the investigation of, and imposition of penalty and remedial provisions upon, financial institutions that have committed or are suspected of committing infractions." (146)

There are a scant handful of statutory provisions that expressly authorize agencies to share information without potentially losing privilege over the communicated information, but these statutes are far from comprehensive. For example,
   under 12 U.S.C. [section] 1821(t), federal banking agencies and a
   few related federal agencies, such as the [Farm Credit
   Administration] and the Federal Housing Finance Agency, may share
   privileged information without waiving privilege. This provision
   does not, however, extend to state bank supervisors, federal or
   state prosecutors, the IRS, the SEC, the CFTC and many others.

Another recently enacted statute authorizes sharing of information between federal and state agencies that have mortgage oversight authority without the loss of privilege. (148) A third statutory provision, enacted in 2010 as part of Dodd-Frank, permits the SEC to share information without loss of privilege with, inter alia, "any agency," "any self-regulatory organization," and "any State securities or law enforcement authority." (149) This provision further permits "[f]ederal agencies, State securities and law enforcement authorities, [and] self-regulatory organizations" to transfer privileged information to the SEC without loss of privilege. (150) The Public Company Accounting Oversight Board (PCAOB) has the authority to share certain information without loss of privilege with the SEC, the Attorney General, and various other self-regulatory organizations and federal and state regulators. (151) A final example is 12 U.S.C. [section] 1828(b), which authorizes sharing of data pertaining to antitrust review of transactions without loss of federal or state privilege among the OCC, Office of Thrift Supervision, Federal Deposit Insurance Corporation, the Federal Reserve, the Attorney General, and the Federal Trade Commission. (152)

Without some such statutory safe harbor, however, agencies pool information at the peril of exposing the materials they share to the public gaze--or at least to the gaze of an adversary in litigation. Positive enactments, such as open government laws or rules of discovery, may make documents produced to a regulator vulnerable to exposure once they are shared with another regulator. (153) If a single cloak of privilege securely covered all of these entities--all for one and one for all--then regulated parties could cooperate with requests for privileged information with any of them without running the risk that subsequent sharing of that information among the cooperating pool of regulators would strip away the privilege.(154)

Regulated parties may be less than thrilled at this prospect. When, for example, the Consumer Financial Protection Bureau (CFPB) announced the (quite dubious) position that it could share privileged information provided to it by regulated parties with state agencies--including state prosecutors--without waiving the attorney-client privilege shielding the material, (155) it set off alarm bells in some quarters. (156) Viewed from the perspective of the regulated party, it is clearly less than ideal for one's immediate regulator to have an unfettered ability to share privileged information demanded by that regulator with potentially adverse parties, such as federal or state prosecutors, (157) or with other entities who might disclose that information to private plaintiffs. (158)

To an agency charged with enforcing federal law, however, there is only value in having broad leeway over the sharing of privileged information produced to it. An agency's ideal position would be to have not only the threshold authority to share privileged information that has been produced to it, but also the further assurance that this sharing of information will not result in the loss of any privilege protecting the information it shares. A delegation of the power to write privileges would give an agency the broadest degree of latitude on this question, particularly if the delegation specifies--as does section 6607--that any communications that are covered by a new regulatory evidentiary privilege "shall not waive any privilege otherwise available to ... any person who provided the information that is communicated." (159)

The imperative of agency coordination also points to a separate and perhaps more fundamental reason why a privilege delegation would be a valuable tool for an administrative agency. Entities, including agencies, must converse if they are going to coordinate. If the nature of this crosstalk matters--as some scholars have argued that it should (160)--then should agencies have discretion over whether to cloak these communications from external scrutiny? Delegations of the power to privilege will play a critical role in determining whether such crosstalk will be accessible to the public and to litigants.

The ACA demonstrates the importance of this question. The ACA fundamentally overhauled the American system of health insurance by placing new regulations on the pricing, benefits, coverage, and issuance of insurance plans. (161) The ACA enlists states and state officials to create exchanges and to enforce the ACA's restrictions against insurance plans. (162) In addition, the Act assigns to the National Association of Insurance Commissioners (NAIC) the responsibility to determine whether the amount that health insurers are spending on health care is adequate or whether they must issue rebates to their policyholders. (163) Finally, the ACA requires consultation between federal agencies, the NAIC, and state officials "on a variety of issues central to the ACA's implementation." (164) In essence, the ACA links together federal agencies, state officials, and the NAIC into a hybrid superenforcement structure charged with implementing and enforcing the provisions at the heart of the act.

The agencies and entities that constitute this hybrid superenforcement structure will naturally value the ability to communicate regarding future actions without risking disclosure of their crosstalk. The discussion below will explore further whether or not allowing privilege over such communications is desirable. (165) For now, however, the point is simple. The incentive of an agency within the hybrid structure is to obtain the maximum amount of latitude over what is privileged--that is, the broadest possible authority to communicate with public or private entities of the agency's choosing, as well as the ability to exercise discretion within that zone as circumstances warrant. Put differently, an agency's ideal position is to have both the authority to communicate within a privileged channel and also the further authority to determine to whom it will open privileged channels of communication.

The most direct way for an agency to obtain that leeway is to secure a broad delegation of the right to promulgate rules of evidentiary privilege governing communications between or among a permissively specified list of entities. This delegation sounds strikingly similar to the actual language of section 6607. (166) This provision is proof positive not only that agencies want a broad power to specify the parameters of privilege but also that agency efforts to obtain such a power can eventually succeed.
COPYRIGHT 2015 University of Pennsylvania, Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2015 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Introduction through II. Administering Privilege, p. 487-524
Author:Sohon, Mila
Publication:University of Pennsylvania Law Review
Date:Jan 1, 2015
Previous Article:Foreign sovereign immunity and comparative institutional competence.
Next Article:The power to privilege.

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |