Co-Client Information Sharing

Co-Client Information Sharing

Rule 1.6(a) contains no carve-out for joint representation. A lawyer may not reveal information relating to the representation of a client unless the client gives informed consent, the disclosure is impliedly authorized in order to carry out the representation, or paragraph (b) permits it.¹ When one lawyer acts for two or more clients in a single matter, each is a client, and each holds that protection. On the Rule's text alone, what one co-client tells the lawyer remains protected unless a gateway opens, even though the other co-client is a client too.

Two pressures pull against that silence. Co-clients chose one lawyer because they share an objective, and each takes part in decisions about the shared matter, so a fact one of them holds may be exactly what the other needs. One commentary states the point strongly: because both clients are entitled to be kept fully informed, the lawyer may not keep a secret on behalf of one client that would be contrary to the interest of the other.² Against that stands the client who confided in reliance on the silence every client expects. The ethics committees of New York and Florida sided with that client and concluded that disclosure to a co-client is prohibited.³

The answer comes chiefly from what the co-clients agreed. An advance agreement is informed consent applied to a joint setting, so the standard from earlier in this chapter governs: the clients must hear the risks and the alternatives before they agree. Where no one addressed the subject, the inquiry shifts to what a joint representation reasonably involves, which is the question implied authorization asks. Both client-based gateways therefore turn on the same facts in this setting, and a decision on concrete facts, A. v. B., shows how they interact.

Disclosure Agreements Among Co-Clients

Informed consent given in advance is the cleanest gateway for co-clients, because it settles how information will move before anyone holds a secret. In A. v. B., the New Jersey Supreme Court called the arrangement a “disclosure agreement” and described two forms it can take. The co-clients can agree that any confidential information concerning one of them, whether it came from that client or from another source, will be shared with the other. Or they can agree that unilateral confidences and other confidential information will be kept confidential by the lawyer. Either way, the court said, a prior agreement clarifies the expectations of the clients and the lawyer and diminishes the need for future litigation.⁴

Either form binds the lawyer. The court read the Restatement (Third) of the Law Governing Lawyers § 112 comment l to say that a prior, explicit agreement between the lawyer and the co-clients controls whether the lawyer discloses one co-client's confidential information to another. Co-clients “may explicitly agree to share information,” they “can also explicitly agree that the lawyer is not to share certain information ... with one or more other co-clients,” and “[a] lawyer must honor such agreements.”⁵

The agreement carries that force because it is informed consent. Under Model Rule 1.0(e), informed consent is “the agreement by a person to a proposed course of conduct after the lawyer has communicated adequate information and explanation about the material risks of and reasonably available alternatives to the proposed course of conduct.”⁶ In a joint representation, the risk is that what one client says may reach the other, and the alternative is a rule that keeps certain confidences. A signature on a form the lawyer never explained does not meet that definition.

Timing explains why the agreement belongs at the start of the engagement. A commentary on multiple representation advises consulting the clients in every case, discussing how confidential information will be treated, and explaining that when two or more persons jointly consult the same lawyer, neither may invoke the attorney-client privilege to prevent the lawyer from disclosing communications to the other.⁷ Waiting is costly. If a conflict or confidentiality issue arises in a dual representation without written consent obtained at the start, the lawyer may be forced to withdraw, and must withdraw from both clients if the lawyer cannot use independent judgment for each or cannot assure each that the client's confidences will not be breached. The lawyer may try to obtain consent when the problem arises, but the duties of loyalty and confidentiality often make that nearly impossible.⁸

A secondary-source example from immigration practice shows the keep-confidences form in use. In a dual representation of an employer and a non-citizen employee on a visa petition, the lawyer and both clients could agree that the scope of the representation for the employee is limited to obtaining the visa and that the lawyer will not disclose confidential information given by either of them unless required by law. The lawyer would then explain that the employee will not see some information, such as prevailing wage determinations.⁹ The same example shows why a lawyer wants the terms settled early. If, in discussing the employee's ultimate career goals, the lawyer learns that the employee plans to open an engineering firm of his own, the employee's interest is adverse to the employer's, and the lawyer faces a separate conflict of interest.¹⁰

Without an agreement, confidentiality and communication collide, and the authorities split on which gives way. One standard the court recounted leaves the choice to the lawyer: after considering all relevant circumstances, the lawyer has discretion to inform the affected co-client of the specific communication if, in the lawyer's reasonable judgment, the immediacy and magnitude of the risk to that co-client outweigh the communicating client's interest in continued secrecy.¹¹ The Florida committee ruled the other way. It reasoned that the lawyer's duty of confidentiality takes precedence over the duty to communicate all relevant information to a client, concluded that the lawyer had no discretion to reveal the information, and added that joint clients do not necessarily expect everything one communicates to be shared with the other.¹²

The New York committee's reasoning shows what tipped the balance. It concluded that the lawyer may not disclose the communicating client's statement to the co-client, resting on the absence of prior consent by the clients to the sharing of all confidential communications and on the fact that the client “specifically in advance designated his communication as confidential, and the lawyer did not demur.”¹³ The lawyer's silence counted against disclosure. A lawyer who cannot or will not keep one client's confidence from the other should say so before the client speaks.

The same call to a lawyer with no agreement at all would turn the answer toward silence, as the New York committee found. What moves the result is whether the clients settled the rule before the confidence existed. In A. v. B., the court noted two features that set the case apart from the New York and Florida opinions: the husband and wife had signed an agreement suggesting their intent to share all information with each other, and the firm learned of the husband's paternity from a third party rather than from him.¹⁴

The opening conversation should cover privilege as well. When two or more persons jointly consult the same lawyer on a matter of common interest, neither may invoke the attorney-client privilege to prevent the lawyer from disclosing communications to the other, and neither may waive the other's privilege.¹⁵ The communications remain privileged against third persons, so the exception matters only in a later controversy between the two clients. The privilege rule says what the clients can keep from each other if they become adversaries. A disclosure agreement says what the lawyer may and may not do while the representation continues.

A. v. B. and the Waiver Letters

A. v. B. tests Rule 1.6(a) where it is hardest to apply: one co-client holds a secret that bears on the other's decisions. A law firm representing a husband and wife in their estate planning learned that the husband had a child he had told no one about, and the firm wanted to tell the wife.¹⁸

The firm learned of the child through its separate representation of the child's mother. A lawyer in the firm's family law department wrote to the husband about her paternity claim. He neither objected to the firm's representing her nor alerted the firm to the conflict; he retained Fox Rothschild instead. After initially denying paternity, he agreed to voluntary DNA testing, which showed that he is the father. Negotiations over child support failed, and the mother sued.¹⁹ The firm then wrote to the husband that it believed it had to tell his wife that his illegitimate child existed, though not who the child was, and that her current estate plan might send part of her assets through her husband to that child. It suggested that he tell her himself and said that if he did not, the firm would. Restraints imposed by the Appellate Division kept the firm from speaking.²⁰

Both spouses were clients, and each invoked a different duty. The firm relied on Rule 1.4(b), which requires a lawyer to “explain a matter to the extent reasonably necessary to permit the client to make informed decisions,” and which imposes a “continuing duty to inform the client of any information important to him.” The husband objected under Rule 1.6(a), under which a lawyer “shall not reveal information relating to the representation of a client unless the client consents.”²¹ The court held that the firm may inform the wife of the existence of the husband's illegitimate child. Because the firm wished to make the disclosure, the court did not decide whether a lawyer's obligation to disclose is discretionary or mandatory. It authorized disclosure of the child's existence but not the child's identity, which matches what the firm itself had proposed, and it noted that this limit would not contravene a New Jersey statute, N.J.S.A. 9:17-42.²²

The court set its case against the New York and Florida opinions and pointed to three differences. The first is how the firm learned the fact. The husband never told the firm about the child; he concealed the child from his wife and from the firm alike, and the firm found out only through the mother's paternity action. The court reasoned that his expectation of nondisclosure might therefore be less than if he had communicated the information to the firm in confidence.²³ The inference is a modest one. Rule 1.6(a) protects information relating to the representation, and its text does not limit that protection by source, so a third-party source opens no gateway by itself. It weakens the client's claim that he reasonably expected secrecy, and expectation is the measure for any implied authorization or agreement among co-clients. A client who confides extends a trust the lawyer can betray; a client whose secret surfaces in someone else's lawsuit extended none.

The second difference is the waiver the spouses signed when they engaged the firm, which provided that “information provided by one client could become available to the other.”²⁴ The court described it as an agreement suggesting their intent to share all information with each other, unlike the agreements, if any, of the co-clients in the New York and Florida opinions.²⁵ The waiver resembles a disclosure agreement but says less than one. Under the Restatement language the court quoted, it is a prior, explicit agreement about sharing that controls.²⁶ A warning that information could become available does not promise that it will be shared, and it does not say what the firm may do with a particular secret. Its force lies in the expectations it creates: spouses who sign it weaken any expectation that a fact touching both of their plans will stay with whichever of them knows it.

The third difference is local to New Jersey. New Jersey's rule permitted disclosure “to rectify the consequences of a client's ... fraudulent act in the furtherance of which the lawyer's services had been used,” and the court named the absence of that exception from the New York and Florida rules as the most significant difference between those opinions and its own case.²⁷ It treated the husband's failure to disclose the child as “a fraud on his wife” and cited the exception in allowing the disclosure.²⁸ That exception is a feature of New Jersey's rule text. A lawyer elsewhere must ask whether the local rule contains a comparable provision, and the first two differences cannot stand in for it.

Taken together, the three differences show what the decision does and does not hold. It does not hold that signed waiver letters alone open a Rule 1.6(a) gateway. By the court's own ranking, the fraud exception was the most significant difference, while the source of the information and the waiver shaped the husband's expectations. Read through Rule 1.6(a), those two features support an argument from the co-clients' reasonable expectations, which is an implied-authorization argument, though the court did not label its reasoning by gateway. The contrast with New York is sharp. There the client had designated his communication confidential in advance, the lawyer did not demur, and the clients had not agreed to share all confidential communications. The A. v. B. husband confided nothing, the spouses had signed an acknowledgment about shared information, and New Jersey's rule contained a fraud exception.

Students most often skip the second step. Signed conflict waivers are not consent to every disclosure. A waiver that says only that information could become available supports at most an inference that the co-clients expected their information to reach each other. When a stem adds a specific secret that one client asked the lawyer to keep, read what the letters actually say and how the lawyer learned the secret.