Federal Prosecution Policy and the Attorney-Client Privilege

Testimony Crime and Justice

Federal Prosecution Policy and the Attorney-Client Privilege

February 11, 2005 22 min read
Paul Rosenzweig
Paul Rosenzweig
Visiting Fellow, The Heritage Foundation
Visiting Fellow, The Heritage Foundation

Testimony before the American Bar Association in the Task Force on the Attorney-Client Privilege

Good morning Mr. Chairman and Members of the Committee. Thank you for the opportunity to testify before you today on the topic of the attorney-client privilege in the context of the Federal Sentencing Guidelines, the Department of Justice's prosecutorial policy, and similar policies adopted by the Securities and Exchange Commission.

For the record, I am a Senior Legal Research Fellow in the Center for Legal and Judicial Studies at The Heritage Foundation, a nonpartisan research and educational organization. I am also an Adjunct Professor of Law at George Mason University where I teach Criminal Procedure and an advanced seminar on White Collar and Corporate Crime. I am a graduate of the University of Chicago Law School and a former law clerk to Judge Anderson of the U.S. Court of Appeals for the Eleventh Circuit. For much of the first 15 years of my career I served as a prosecutor in the Department of Justice and elsewhere, prosecuting white-collar offenses. During the two years immediately prior to joining The Heritage Foundation, I was in private practice representing principally white-collar criminal defendants and I continue to represent several clients in a small practice.


I should note that portions of my testimony are substantially derived from a law review essay which I authored: "Truth, Privileges, Perjury, and the Criminal Law," 7 Tex. Rev. Law & Pol. 513 (2002).

* * * * *


First Principles: "Why A Privilege?" - When confronted with a conundrum of the sort presented by the Department of Justice's policy on attorney-client privilege waiver, it is useful, in my view, to return to first principles. So, to begin at the beginning: A testimonial privilege is the right of an individual to refuse a command to testify truthfully. There are many such privileges in our system. Wives cannot be compelled to testify against their husbands; priest cannot be compelled to disclose what they learn in confessional.


These testimonial privileges obviously obstruct the search for truth. Simply put, if a witness with relevant information is empowered to refuse to testify, that diminishes the quantum of information available -- to the prosecutor, to the jury, to the courts, and ultimately to the public. As a general matter our legal system is founded on the premise that it serves to determine the truth of historical events. We litigate cases and have trials in order to find out what has happened . . . what the facts are. 


Yet even as we construct a truth-seeking system, we erect obstacles to its operation. The law of privilege is in conscious derogation of the truth. It is as if we say, we can seek the truth but we choose not to. What would motivate such a choice? Plainly, we see in the various privileges some other, superior value.


The attorney-client privilege has its historical origin in England as early as the 1500s. It stems not from any professional obligation but rather from a broader "code of gentlemen" (recalling that this rule stems from a time when only men were subject to public legal obligations). All gentlemen, including attorneys, were obliged not to disclose anything told to them in confidence. This rule was also congenial with the law, which prevailed in England well into the 19th century, that parties to the litigation themselves were incompetent to testify, whether it was on their own behalf or if called by their adversaries. It was thought that as an interested party the potential for untruthful testimony was too great. And if so for the party, then so too, of course, for all of his confidants, whether attorneys or mere close friends.


In the United States, however, the attorney-client privilege has never been the product of social convention. As early as the time of the Revolution our courts repudiated the code of the gentlemen, reasoning, as McCormick (one of the original crafters of our evidentiary rules) has said, "that the need for the ascertainment of truth for the ends of justice loomed larger than the pledge of secrecy."[1]


Thus, the attorney-client privilege in modern America is not a moral or social imperative. Rather it is an avowedly instrumental doctrine. We say that the confidentiality of the attorney-client privilege facilitates an individual's ability to obtain effective legal advice and services. We presume that untutored clients need lawyers to help them understand the law -- a fair assumption in today's complex world. We also assume that to obtain counsel clients need to disclose facts to their counsel sufficient to allow counsel to advise them effectively and adequately -- again, a fair assumption.


Finally, we assume that clients would be unwilling to make full disclosure of facts to an attorney if the lawyer could be required to testify as to those facts.[2] This too seems a reasonable assumption but it is worth examining somewhat. There are some commentators, for example, who wonder: Do many clients actually think about confidentiality? Won't they make disclosures anyway because they need a lawyer's assistance, whatever the price? Some have gone further - they think that the attorney client privilege is just one of those cases where a group gives itself a monopolistic economic advantage.[3] Because information told to lawyers is secret and information told to accountants is not, maybe the client will choose to talk to the lawyer.


Though there is some doubt as to the necessity of client confidentiality generally, there is one area where the attorney-client privilege almost certainly is necessary because it serves a quasi-constitutional function. In case of criminal prosecution - a situation where the client can assert a Fifth Amendment privilege against self-incrimination -- the attorney-client privilege is a necessary corollary to the privilege. It would not do to tell a putative defendant that his Fifth Amendment privilege is of no value because the same information can be obtained from his attorney. So, at least in the criminal context, the underpinnings of the attorney-client privilege have a fairly strong foundation since the privilege is in service of a constitutional principle. In this context, the privilege is something more than operational - it implements a legal norm grounded, in part, in the natural law judgment that no person should be compelled to be a witness against himself.


The Value of Confidentiality -- But given the broader applicability of the confidentiality rule it is worth asking: Do we really want so stringent an approach to client confidentiality? Do the benefits of the rule, serving broad societal interests, outweigh its costs? One implication of the Department of Justice's recent policy actions is that it has institutionally concluded that the answer is "no" - where there is no constitutional necessity, the benefits do not outweigh the costs.


This judgment is at odds with a widely shared consensus within the legal community. Consider one extreme case that demonstrates just how seriously the legal system takes the idea that privileges may justly impede the search for truth and that values of confidentiality outweigh the need for justice in individual cases. 


William Macumber was on trial for first-degree murder.[4] In his defense at trial he offered to prove that another person had confessed to the murder for which he was being tried. The third party had confessed he was the murderer to his two attorneys. After the confessed murderer had died, the two lawyers stepped forward and were willing to testify at Macumber's trial about the confessions. The trial court, on its own motion, excluded the attorneys' testimony.


On appeal, the Arizona Supreme Court affirmed. Even though the two attorneys had gotten an informal opinion from the Committee on Ethics of the state bar that the attorney client privilege did not prevent them from testifying,[5] the court held that the dead client's lawyers could not answer questions about confidential communications they had with the client prior to his death. According to the court, the privilege survived the death of the client and could only be abrogated in situations where doing so would further the dead client's interest. Here, of course, the interest that would be furthered by the attorney's testimony was Macumber's, not the dead client's. As a result, the two lawyers were not permitted to testify.


The decision is remarkable. Potentially probative evidence that bore on the guilt or innocence of an alleged murderer who faced a potential penalty of life imprisonment (imagine if this were a capital case!) was excluded, even though the client who had relied on the promise of confidentiality was dead. The only persuasive argument in favor of such a rule has to be something of the following form: "If we void the privilege after death, other clients in later cases will be aware of the rule and are likely to be dissuaded from being candid with their lawyers fearing posthumous disclosure of their own confidences. This, in turn, will have the effect of rendering the adversarial process less effective in discerning the truth for a larger number of cases because lawyers will not be able to effectively advocate on behalf of clients without full knowledge of their confidences. Hence, even in the case of a dead client whose statements might save an innocent man from life imprisonment the system-wide values are too great to permit the privilege to be breached."


Note what is happening here -- the constitutional right of the accused to present his defense is rendered subservient to the reputational interest of the dead client in keeping his disclosures private. For after all, the only personal interest the dead client can possibly have is the lingering desire not to have his reputation tarnished after his death. While concededly substantial - everyone wants to be well thought of after his passing - it is not unreasonable to think that the reputational interest is of less significance than the interest in a just resolution of a murder charge. Yet we preserve the confidentiality. Plainly something more is going on here and it can only be the concern for the inherent value of confidentiality in other cases - not the immediate case pending. [6]


To be sure, some might see the whole case as an example of the attorney-client privilege carried to absurd lengths. But the Macumber vision of the law is deeply embedded in our legal structure. The Supreme Court has since confirmed the Arizona court's understanding of the privilege, holding that an attorney may refuse to answer questions in a federal criminal investigation based upon the attorney-client privilege, even after the death of his client.[7]


The Nature of the Corporate Privilege - Given how strongly we protect the privilege in individual criminal cases (as in Macumber), what possibly can explain the hostility to the privilege in the corporate context that appears to animate much of contemporary white-collar criminal enforcement policy?

First, consider what distinguishes the corporation from the individual: In the corporate setting, the attorney-client privilege is unique in that it is the entity to whom the privilege attaches and not the individual employees who communicate with the attorney. Similarly, the decision whether to waive the privilege belongs to the corporation, and not to its employees. 

One might have thought that categorically this would lead to the conclusion that corporations have no attorney-client privilege at all. Corporations have less in the way of individual privacy interests - they have very substantial interests in confidentiality for their proprietary information but as creatures of public creation when their stock is traded in public markets, their business is, to a very real degree, an open book. And the confidences being disclosed to an attorney by corporate employees, while germane to the corporation's interest are, ultimately, all the confidences of particular individuals. The corporation's confidences have no independent existence apart from the individuals who hold them. Despite the differences between corporations and individuals, even when the attorney's client is a Fortune 500 corporation an attorney-client privilege exists.[8]  

Similarly, the corporation's reputational interests, while significant, are of a decidedly different character than that of individuals. Here, too, the corporation's reputation is really nothing more than the accumulation of the reputations of its product, and its stockholders, directors and employees. When we think, today, of Enron, for example, we think of Kenneth Lay's acts - not some numinous concept of Enron unattached to any particular individual.

The most significant distinction, however, is that corporations don't have a Fifth Amendment privilege -- the Fifth Amendment is a personal right.[9] This presents a fascinating conundrum of contemporary criminal law - how did the non-constitutional common law attorney-client privilege come to be deemed of greater value to a corporation than a constitutionally mandated privilege against self-incrimination? 

One can posit two answers. The first, and most likely, is one of historical accident - at the time corporations were deemed to not possess a privilege, part of the reasoning was that the lack of a privilege was a natural consequence of the fact that corporations could not be prosecuted. It was only a few years later, however, that the Court determined that corporations could be prosecuted[10] - and it has never resolved the seeming inconsistency. And so, in the criminal context the corporate attorney client privilege does not (unlike the individual privilege) act in service of an underlying constitutional principle. 

The other possible answer is one of necessity -- if corporations had a Fifth Amendment privilege the prosecution of regulatory offenses would be effectively impossible. Since the corporation is holder of the regulatory records that contain the details of the offense, if the privilege were extended to corporations allowing them to contest disclosure it would deprive regulatory prosecutors of the necessary evidence of the offense (and, indeed, often of the corpus delecti). 

Mandatory Disclosure Polices -- Thus, whether by virtue of necessity or historical accident the only current justification for the privilege in the corporate world is its perceived utilitarian value of fostering communications between the corporation's employee's and its counsel.[11] And framed thus, the proper structure for examining DOJ's policies becomes clear.


The Task Force is examining two related developments - the Sentencing Commission's decision to require a corporation to provide privileged material to the prosecution to show "thorough" cooperation; and the closely related DOJ policy and practice of requiring targets of criminal investigations or prosecution in certain circumstances to provide privileged material as a condition of leniency.


If we leave aside certain limited cases where such disclosure is obviously necessary (as when the corporation seeks lenity on the ground that it was following attorney advice that has since proven to be in error) it is clear that policy choices made by the Commission and the Department rest fundamentally on a rejection of the utilitarian premise - that is, they rest on the view that the long run benefits of confidentiality do not outweigh the near-term benefits of disclosure.


There are three reasons to substantially doubt the assessment of those who reject the utilitarian calculus. First, of course, as detailed earlier, it rejects the lessons of history. Yet one has the sense that social polices develop out of tradition for reasons of usefulness. We should be exceedingly reluctant to whole-heartedly reject the traditional, time-honored rules without any assessment of their historical antecedents. 


In doing this, the new policies are, more or less, recapitulating the arguments over the Fifth Amendment privilege for corporations. We denied corporations that privilege historically because they could not be prosecuted and we continue that rule now because granting them the privilege would render regulatory prosecutions more difficult. It was, if you will, a pure power grab by the government. Since the demise of the Fifth Amendment privilege corporations have found a way of resurrecting similar protections for the products of their internal investigations through the attorney-client privilege, cloaking some of the same analysis in the common law privilege. Here, too, the government now brings its greater power and monopoly on the lawful use of force to bear and, seeing another obstacle to its success, works to eradicate that as well. 


We ought to recognize the underlying hostility to the corporate form that this new conception reflects - in no other context would we accept the idea of criminal liability without the protections of Constitutional provisions - yet for the corporation, that has been the rule for nearly a century, and the new policies are merely an expansion of that anti-commercial rhetoric.


Second the new policies disregard second-order effects. To be sure, in the near-term they will advance prosecutorial interests by giving governmental authorities easier access to corporate information developed through internal investigations. But the natural consequence of routine use of this investigative tool will necessarily be that corporations will decrease their use of internal investigations, or, if they do conduct such investigations, the cautions that the investigating attorneys are obliged to give the corporation's employees will create a disincentive for full disclosure. I have, for example, already heard one senior corporate lawyer opine in a corporate bar meeting that it is legal malpractice to conduct an internal investigation - while perhaps an extreme view it accurately reflects the disincentives that are being developed. 


And of course, from the broader societal perspective that is precisely the wrong answer. We want corporations to be self-regulating to the extent possible. We want to encourage and foster introspection and self-correction. And good business practices would normally advance those interests. Now, however, with the prospect of an adverse investigative result, the salutary self-review instinct is tempered by fear of governmental disclosure. Indeed as one corporate general counsel reported in an informal survey conducted by the ABA's White Collar Crime Section: "I can no longer tell employees that their remarks will be held in confidence. And when I indicate that the company might well have to waive the attorney-client privilege, employees clam up and I really can't get to the bottom of the problem."[12]


[As an aside, this hesitancy will only be exacerbated if the Department also aggressively pursues its new policy (first used in the prosecution of Computer Associates executives) of bringing obstruction of justice charges against corporate officers who allegedly lie to internal corporate investigators.[13] First, it will surely dissuade corporate officials from cooperating with internal investigations. It is bad enough when the attorney must warn the officer: "information you give to me may be shared with the government." Imagine the reaction when they add: "and if you lie to me, the government may charge you with obstruction of justice." Second, if it becomes routine for corporate investigations to be used by the government, and if corporate officers are obliged to cooperate with the investigation as a condition of continued employment, then it is likely that corporate officers, like their blue-collar compatriots, will have derivative Fifth Amendment Garrityrights against self-incrimination that will allow them to refuse to cooperate or preclude the introduction of their statements as substantive evidence.[14]]


Indeed, given the increasing frequency of disclosure to the government of internal investigations, I am slowly becoming persuaded that the corporate attorney's ethical obligation to clarify his role in the internal investigation may grow into an ethical obligation to make disclosures of about the nature of likely subsequent corporate and government action. Ethical rules already impose disclosure obligations on the corporate attorney, so that the employee is not confused as to who represents him. [15] As corporate internal investigations become, more frequently, proxies for governmental investigations the nature of this ethical disclosure may well be required to expand.


Third, the new policies are a mistake given the broader context of the criminalization of productive corporate activity. The types of white-collar offenses at issue here typically involve the criminalization of conduct that, in most instances, is not inherently wrongful in the same way that fraud and bribery are. Rather, we have seen a growth in the category of "public welfare offenses" - a category first created with modest penalties and now increasingly felonized. Second, and of special significance in weighing moral culpability, the statutes involve offenses where the mental element (or mens rea requirement) is substantially diminished, if not eliminated. For example, we now punish as strict liability offenses the taking of migratory birds - even if done utterly by accident. Third, this type of white-collar offense increasingly involves criminal prosecutions of managerial officers for, in effect, vicarious liability. The growth in this form of white-collar criminal offenses is what Professor John Coffee has called the "technicalization" of crime. As a result, for this category of white-collar offenses, the criminal law is increasingly being used interchangeably with civil remedies.

Consider: In 1999, the ABA Task Force on the Federalization of Criminal Law noted that there were now more than 3,500 federal criminal offenses. More recent studies by Professor John Baker of Louisiana State University put the number at over 4,000.[16] Those offenses incorporate either directly or by reference prohibitions contained in more than 10,000 separate regulations. Remarkably, nobody knows the exact number either of criminal statutes or criminal regulations. They are so diverse and so widely scattered throughout the federal code that they are literally uncollectable. I am told that, when it was recently asked to undertake the project, the Congressional Research Service said that the task was virtually impossible. This, too, breeds disrespect for the law and disaffection from the judicial system: When those who make the laws cannot themselves identify all the laws they have made, it borders on the arbitrary and capricious to allow prosecutors to select from among those laws and to criminalize conduct that, in the eyes of other prosecutors, might warrant only civil sanctions.

Where the law is so uncertain, one would think that the uncertainty would counsel hesitancy in the derogation of a long-standing legal rule. Instead, the uncertainty, which brings with it a broad arrogation of prosecutorial discretion, has had the opposite effect. Now prosecutors may comfortably choose the target first, confident that they will find a crime (any crime) to charge later. And the routine abrogation of the attorney-client privilege is just another means of achieving that end goal - for the internal investigations will often disclose conduct that is potentially criminalizable. And now, given the breadth of the law, the mere threat of criminal sanctions is more than ample to coerce changes in corporate behavior.


Coercing the Individual - One further aspect of the Department's policy evolving policy merits particular mention because of its especially pernicious potential and as an example of the coercive effect of broad criminal liability. I refer here to the apparently new practice of conditioning corporate lenity on the corporation's abrogation of its contractual obligation to provide for a defense for its employees. 


Recently, for example, the Department told KPMG LLP that if it continued to pay the legal costs for the defense of one of its employees, Jeffery Eischeid, that it would be deemed not to be fully cooperative. The government said, in essence, that because Eischeid was not cooperating by acknowledging his own personal guilt, the company was fostering that non-cooperation (and being uncooperative itself) by continuing to fund Eischeid's defense. This despite the fact that KPMG had contractual and statutory obligations to pay Eischeid's criminal defense costs.[17] KPMG has (at the behest of the government) conditioned its payments on the willingness of the employees to cooperate.


To the extent that this practice becomes commonplace it goes beyond a disagreement with the utilitarian calculus that has animated much of this discussion. For the government's actions will have the practical effect of denying corporate officers access to the counsel of their choice, artificially limiting their options based upon a greatly reduced financial ability to pay. To be sure, because the government is addressing its demands to the corporation rather than the individual it avoids directly trenching on the underlying Constitutional right to counsel. But it is equally clear that by taking steps of this form the government is acting indirectly to achieve what it cannot achieve directly and thus trenches perilously close to coercing a private party, as a condition of lenity, to assist it in denying another private individual his Constitutionally-protected right of access to an attorney.


* * * * *

Mr. Chairman, the new policies are deeply troubling. They are troubling because they reflect a short-term utilitarian calculus in disregard for historical antecedents that go back as much as 500 years. But more importantly, they are troubling because, in my judgment, they reflect nothing more than a deep-seated animus - an animus for corporations; an animus for defense counsel; and an animus most of all for any procedural obstacles that frustrate the government's objectives. But those procedural "obstacles" are not obstacles at all - rather they are systemic checks on governmental power that exist for the benefit of all citizens. I commend the Task Force for its attention to this problem and urge it to squarely call for the government to forego its new policies in the interest of greater long-term social benefits for everyone.


Mr. Chairman, thank you for the opportunity to testify before the Committee. I look forward to answering any questions you might have.

The Heritage Foundation is a public policy, research, and educational organization recognized as exempt under section 501(c)(3) of the Internal Revenue Code. It is privately supported and receives no funds from any government at any level, nor does it perform any government or other contract work.

The Heritage Foundation is the most broadly supported think tank in the United States. During 2013, it had nearly 600,000 individual, foundation, and corporate supporters representing every state in the U.S. Its 2013 income came from the following sources:

Individuals 80%

Foundations 17%

Corporations 3%

The top five corporate givers provided The Heritage Foundation with 2% of its 2013 income. The Heritage Foundation’s books are audited annually by the national accounting firm of McGladrey, LLP.

Members of The Heritage Foundation staff testify as individuals discussing their own independent research. The views expressed are their own and do not reflect an institutional position for The Heritage Foundation or its board of trustees.



[1] McCormick on Evidence § 87, at 314 (4th ed. 1992).

[2] The same question recurs in other privilege contexts where the premise of the privilege is the need for confidential communications. Would one fearing eternal damnation refrain from confession because of the fear of criminal prosecution? By contrast we assume that one seeking medical assistance is so motivated by self-interest to be truthful that his statements to a doctor may be used as substantive evidence. E.g. Fed. R. Evid. 803(4).

[3] Fishel, Lawyers and Confidentiality, 65 U.Chi.L.Rev. 1 (1998).

[4] State v. Macumber, 544 P.2d 1084 (Ariz. 1976), app. after remand, 582 P.2d 162 (Ariz. 1978).

[5] Id. at 1087 (Holohan, J. specially concurring).

[6] Id. at 1088.

[7] Swidler & Berlin v. United States, 524 U.S. 399 (1998).

[8] Upjohn Co. v. United States, 449 U.S. 383 (1981).

[9] Hale v. Henkel, 201 U.S. 43 (1906).

[10] New York Central & Hudson River RR Co. v. United States, 212 U.S. 481 (1909).

[11] Of course extraneous considerations, such as adverse publicity, may make it more difficult for a corporation to assert its privilege than for an individual. Enron's former attorney's have testified truthfully and fully about advice they gave Enron - but only because Enron has authorized them to do so and waived the attorney-client privilege, presumably for reasons of public relations. E.g.Statement of Christian G. Yoder, Esq. Before  the Senate Committee on Commerce, Science and Transportation (May 15, 2002), [available at /static/reportimages/28414CB15C4E51E7F97F12C53118FCD9.pdf].

[12] Quoted in Letter from Sheldon Krantz, Esq. to Asst. Attorney General Michael Chertoff (Jan. 24, 2003).

[13] See Alex Berenson, "Case Expands Type of Lies Prosecutors Will Pursue," NY Times (Apr. 9, 2004).

[14] See Garrity v. New Jersey, 385 U.S. 493 (1967) (choice between cooperation and termination renders statements involuntary and therefore in violation of self-incrimination privilege).

[15] See, e.g., D.C. Bar Legal Ethics Op. 269 (1997).

[16] See John S. Baker, Jr., "Measuring the Explosive Growth of Federal Crime Legislation," (The Federalist Society for Law and Public Policy, 2004).

[17] See Laurie P. Cohen, "In the Crossfire: Prosecutor's Tough New Tactics Turn Firms Against Employees," W.St. J. (June 4, 2004).


Paul Rosenzweig
Paul Rosenzweig

Visiting Fellow, The Heritage Foundation