The Confidentiality Fetish

The problem with attorney-client privilege

Last year the former Oklahoma governor Frank Keating resigned as head of a commission appointed to look into the sexual-abuse crisis in the Catholic Church, because some bishops were refusing to provide information the commission requested. Why would the Church, having appointed such a panel, deny it access to relevant information? Because, Keating explained, the uncooperative bishops had "turned to their lawyers when they should have looked into their hearts." Although most Catholics favor "transparency" in this crisis, another commission member explained, "the attorneys for a diocese do not think that way."

This is an all too familiar refrain in American public life. The captain of the Staten Island ferry that crashed into a pier in October of last year, killing ten people, wouldn't tell investigators where he had been at the time of the crash—his lawyer wouldn't let him. At about the same time, in Palo Alto, California, a teenage girl involved in a bizarre hit-and-run accident that killed a small child received a surprisingly harsh sentence. The victim's family had urged severity after repeatedly expressing bitterness at the girl's failure to apologize to them. Her lawyers had advised her not to. After he was forced to admit his involvement with Monica Lewinsky, Bill Clinton earned further opprobrium by refusing—most likely on the recommendation of his lawyers—to concede that he had earlier lied about Monica during the Paula Jones case.

Even people with little reason to fear liability turn to lawyers to vet their public statements. Anita Hill, herself a lawyer, appeared at Clarence Thomas's confirmation hearings with a battery of lawyers—even though her role was simply to give a straightforward account of events she had witnessed, with no prospect of ensuing litigation. Sherron Watkins, the Enron accountant who confronted Kenneth Lay face-to-face at a time when many of her peers were afraid to do so, often spoke to the media through a lawyer after she became famous.

Lawyers, in short, have carved out a role for themselves as the privileged keepers of much information that is important to the public interest. Historically, lawyers have liked to think of themselves as defenders of individual liberty against an overbearing state, primarily through traditional advocacy—that is, persuasively asserting a client's rights. Today, however, lawyers' typical efforts to mediate between clients and the state rely less on advocacy and more on information control. This is a disturbing development; lawyers have brought to their new role as information guardians a powerful predisposition toward needless secrecy that suppresses and distorts information about many matters of public importance.

Never say anything to the police without talking to a lawyer first. Never volunteer anything in a deposition. Never try to explain or apologize when you are in an accident. Don't talk to anyone about your civil case. These canonical precepts serve to maximize the profession's influence. Lawyers are excessively prone to advise confidentiality for the same reason that surgeons are excessively prone to advise surgery. Confidentiality puts a premium on services and protections that lawyers are distinctively qualified to provide.

Dependence on lawyers often impedes responsible or humane behavior. It also has tangible costs for some clients. The Palo Alto teenager would probably have been better off if she had been more candid about her wrongdoing. So, for that matter, would Bill Clinton. Candor might have created more sympathy for the teenager at her sentencing hearing, and it might have eased the political pressure on the President.

In a world that increasingly seeks transparency in government and business, the legal profession stands out for its frankly uncompromising commitment to opacity. Supreme Court Justice Louis Brandeis's principle that "sunlight" is the "best of disinfectants" is a major influence on contemporary public policy nearly everywhere except in his own profession.

The bar's commitment to confidentiality is not just an ideology; it is also a marketing strategy. In the bar's competition with other professions, confidentiality is often a more important advantage than legal expertise. Many accountants can give better advice than most lawyers about what belongs on a tax return; and one needs an engineer to draft some reports that are required by environmental laws. But lawyers can give clients something that other professionals, with the exception of doctors and priests, cannot: strong confidentiality rights. Although the legal system routinely requires accountants, bankers, and business consultants to disclose ostensibly private communications with their clients, attorney-client privilege protects most communication between lawyers and their clients from involuntary revelation.

This principle got a boost from the Supreme Court's 1981 decision in Upjohn v. United States. Upjohn involved an investigation by a lawyer at the pharmaceutical company of an allegation that middle managers had bribed foreign government officials. When the Internal Revenue Service subpoenaed the lawyer's files, the Court ruled that they were privileged. In the years since Upjohn was decided, corporate investigations have become a big business, and lawyers have acquired a virtual monopoly over it.

Tobacco-company lawyers further tested the limits of confidentiality when they persuaded their clients to give them responsibility for much of the companies' research on the health effects of smoking. The lawyers reasoned that since health effects were central to many product-liability suits against the companies, the research facilitated legal advice, and hence should be privileged. If the results looked good, the studies were shifted away from the lawyers and made public. If they looked bad, the lawyers invoked privilege. Some courts rejected this strategy as an abuse of privilege, but it is hard to distinguish from the more routine practice, encouraged by Upjohn, of using lawyers to investigate allegations of wrongdoing.

A recent move to tighten financial-disclosure requirements for public companies—the Sarbanes-Oxley Act, passed by Congress in reaction to the Enron scandal and other corporate infractions—has increased the demand for lawyers even more. Although the reporting that Sarbanes-Oxley calls for is financial, lawyers, rather than accountants, determine what is to be disclosed. Anxious about the expanded liability the statute imposes, executives would rather talk to lawyers than to accountants or consultants, because lawyers confer greater confidentiality. When board meetings address sensitive issues these days, the directors frequently ask nonlegal advisers to leave the room.

The bar is currently fighting a proposal by the Securities and Exchange Commission that would require a corporation to report to the commission if one of its lawyers withdraws because the corporation persists in conduct the lawyer has advised is illegal. The bar argues that such a "noisy withdrawal," by signaling the lawyer's knowledge of wrongdoing, would effectively disclose the communications that prompted it. Noisy withdrawal is a long-standing requirement for accountants. But lawyers have so far been successful in blocking its application to themselves.

Of course, most business and professional relations involve confidentiality. In most dealings, however, the law requires that confidentiality yield to stronger public or third-party interests. Yet the bar often insists that attorney-client confidentiality be preserved at the expense of even the weightiest competing interests. Suppose—to take a case studied in law-school ethics classes—that a now deceased client once confessed to his lawyer that he had committed a crime of which someone else has just been convicted. May the lawyer disclose the confession in order to save the wrongfully convicted person? The stakes for the convicted person are enormous; the stakes for the deceased client are nil. Nevertheless, the Arizona Supreme Court once held in a similar case that disclosure is prohibited. The American Bar Association recently revised its position on confidentiality to permit disclosure in situations like this; but most jurisdictions still follow the rule applied in the Arizona case.

For the bar, the rationale for confidentiality is that without it, people would avoid lawyers or withhold information from them. By itself, of course, such a consequence would not hurt anyone but lawyers. Whether there is a public interest in encouraging consultation with lawyers depends on a further contention: that it tends to induce compliance with the law. Assured of confidentiality, the argument goes, clients will freely disclose their plans to their lawyers, who can then counsel them not to pursue plans that would violate the law. Without the assurance of confidentiality, many of these discussions would never take place, and lawyers would lose the opportunity to direct their clients along the straight and narrow.

It seems unlikely, however, that the kind of advice clients seek only when assured of confidentiality has socially desirable effects. One problem with the claim that legal advice promotes compliance with the law is ambiguity in the meaning of "compliance." Some legal advice is designed to help clients get around a law by casting their affairs in a way that technically conforms to it but defeats its purposes. Many lawyers insist that they have a duty to exploit loopholes in the interests of their clients whenever possible. Vinson & Elkins, a firm that helped Enron craft bizarre business structures for the sole purpose of giving an unrealistically favorable appearance to the company's financial statements, has responded to criticism by saying that as long as what the client wanted could be accomplished within the law, the law firm was not responsible for any bad consequences. Society may be obliged to permit lawful though obnoxious conduct (although many people dispute that this was even lawful), but it is not obliged to encourage it with the shield of confidentiality. If limiting confidentiality reduces Enron-style "compliance," so much the better.

Let's assume, however, that "compliance" means respecting both the letter and the spirit of the law. Is strong confidentiality likely to further it? Someone who feels committed to obeying the law in any case should not need an assurance of confidentiality to disclose his plans. The bar's argument assumes that the client has some disposition to disobey it. So what does a lawyer say to induce him to comply? The lawyer can advise the client about the likelihood that wrongful behavior will be prosecuted, and explain what the penalty for getting caught might be—but there is no guarantee that this information will encourage compliance. In certain areas of practice both the likelihood of prosecution and the size of the penalties are so small relative to the prospective gains from wrongful behavior that full knowledge of the law often makes a client more inclined to violate it. For instance, low audit rates and low penalties for a number of abuses make tax evasion a rational strategy for many clients. (The Internal Revenue Service recently went to court to force several law firms, including the Dallas firm of Jenkins & Gilchrist, to disclose their tax-shelter client lists, quite understandably suspecting that the firms' legal advice increased the propensity of their clients to engage in practices the IRS considers illegal.)

Cutting back confidentiality might leave lawyers with fewer opportunities for encouraging their clients to comply, but they would have more success with the opportunities they did get. If lawyers had to disclose illegal behavior through noisy withdrawal, as the SEC proposes, they would have much more leverage with those clients who did consult them.

The net effect of confidentiality, therefore, is probably to reduce compliance with the law. This truth is implicit in laws such as those requiring therapists to report possible child abuse by their clients. Knowledge of this requirement may prevent some potential abusers from seeking professional help in time—but policymakers have consistently assumed that disclosure laws prevent and stop abuse much more often. Indeed, the bar's own confidentiality rules allow disclosures that would prevent criminal acts likely to cause "death or substantial bodily injury." If confidentiality really promoted compliance, then the rules protecting it would be strengthened, not relaxed, when the stakes were high.

The rationale for strong confidentiality evaporates completely in those areas of corporate practice covered by regulations like the SEC's noisy-withdrawal proposal. For one thing, corporate executives do not have the luxury of withholding information from lawyers, because doing so would deprive them of the right to invoke the "advice of counsel" defense in a criminal case or the "business judgment" defense in a civil case. Moreover, under even the strictest confidentiality rules a corporate manager cannot assume that everything he tells his company's lawyer will remain private—because the "client" in the corporate context is the organization, not the managers who work for it. Corporations can waive confidentiality for their managers' statements, and after wrongdoing is discovered they often do. What frequently happens is that the managers are fired, and the board waives confidentiality for their statements to counsel in the hope of inducing prosecutors to be lenient toward the corporation.

In 1995 Bernard Nussbaum had a suggestive exchange with Senator Richard Shelby in the course of a Senate Whitewater Committee investigation into the circumstances that forced Nussbaum to resign as White House counsel. One of the episodes that led to his resignation was his refusal to allow the Justice Department to search the office of Vincent Foster, the deputy White House counsel, after Foster was found shot to death in a nearby Virginia park. Asserting a lawyer's prerogative to control information, Nussbaum insisted that he review all the documents in Foster's office first, in order to decide whether any of them should be kept confidential. The Justice lawyers were looking for evidence pertinent to Foster's death, and, in fact, a suicide note was eventually discovered in his office.

Months later, when Nussbaum appeared before the committee, he professed bewilderment at the fuss his obstruction had provoked. Virtually any lawyer would have done as he did, he said, with unfortunate plausibility. But when he proceeded to argue that any client would want such behavior, he got into trouble with Senator Shelby.

shelby: Why didn't you say [to Deputy Attorney General Philip] Heymann, come on down and we'll look this over together?

nussbaum: Senator, did you trust the Justice Department?

shelby: On stuff like that, I certainly do.

nussbaum: Would you let them into your office, your counsel's office, to look at your personal stuff?

shelby: Absolutely. I have nothing to hide. They can come over tomorrow or today.

Lawyers are prone to forget that confidentiality often comes at a price—the implication that the person involved has something to hide. It now seems clear that Foster committed suicide; but there was much confusion at the time, and Nussbaum's assertion of confidentiality fueled suspicion of a cover-up.

Another indication that lawyers underestimate the costs of confidentiality—for both clients and society—comes from a recent experiment at the Veterans Affairs Medical Center in Lexington, Kentucky. After suffering large losses in two malpractice cases, the hospital responded in a daringly novel manner. Instead of trying to tighten control over information relevant to potential claims, it adopted a policy of routine voluntary disclosure. When a mistake was discovered, the hospital informed the patient or the family, apologized, and recommended consulting a lawyer. It made information available to the patient immediately without request, and when a malpractice claim was made, it responded promptly with what it considered a reasonable offer.

The hospital's premise was that malpractice claims are fueled less by financial motives than by anger and distrust. Disclosure and apology go a long way toward assuaging such feelings. (Jennifer Robbennolt, a psychologist at the University of Missouri, has found support for this idea in experimental simulations. Subjects asked to decide whether to accept specific settlement offers in hypothetical cases were most likely to accept when the defendant was described as open and apologetic.) And a claim can be resolved at lower cost when the defendant makes a convincing demonstration of good faith.

Of course, openness has a price too. Many patients who would never have discovered mistakes are told of them. Thus it's not surprising that the Lexington hospital sees a higher rate of claims than most VA hospitals. But its average payment per claim, and the total cost of claims made against it, are among the lowest recorded by any VA hospital. "The attorneys around here in Lexington used to think we were crazy," Ginny Hamm, then the hospital's lawyer, told a reporter for The Philadelphia Inquirer in 1999. But the hospital considers its program a resounding success, and many other institutions are reconsidering their practices as a result.

There are limits on how far such private initiatives can prudently go. Optimal transparency would probably require much deeper changes, including making our courts more efficient and scaling back our harsh and intrusive liability system. Still, as the Lexington hospital's experience suggests, the system leaves more room for transparency than lawyers typically realize—or tell their clients.