Joint Representation

Joint Representation

Joint representation puts both halves of Rule 1.7(a) to work on a single engagement. When one lawyer acts for two or more clients, usually on the same side of a matter, the lawyer must ask whether the representation involves a concurrent conflict, which the rule defines in two ways.¹ The first is a representation of one client that will be directly adverse to another client.² The second is a significant risk that the representation of one or more clients will be materially limited by the lawyer’s responsibilities to another client, a former client, or a third person, or by a personal interest of the lawyer.³ Co-plaintiffs, business founders, spouses planning an estate, an employer and the employee it sponsors for a visa, and several people charged in one crime all fit the pattern.

The rule does not forbid the arrangement, and for good reason. Representing several clients on the same side of a civil case is common and frequently makes good sense,⁴ and joint clients can pool resources and present a uniform front to an adversary.⁵ The risk grows from the same root as the benefit. Whenever more than one client is represented, their interests may diverge at some point,⁶ and a lawyer who owes each client equal loyalty has no side to take when they do.

The two definitions do different work in this setting. At the start, the lawyer stands beside both clients against a third party, so the lawyer’s advocacy is aimed at neither, and the direct-adversity definition has nothing to catch. The analysis usually runs through material limitation, which asks whether the lawyer’s responsibilities to one co-client create a significant risk of narrowing the advice or advocacy owed to the other.⁷ Direct adversity enters later, if the clients collide so sharply that one asserts a claim against the other and the lawyer would have to advocate against a current client.⁸

The subsections that follow take the analysis in order: what a lawyer owes each joint client, how to judge the risk that the clients will diverge, and how the test plays out for civil co-parties, transactional clients, criminal co-defendants, and an organization and its people. Whether a conflict can be waived, and how aggregate settlements are handled, belong to later readings. The question here is only whether a conflict exists.

Duties Owed to Joint Clients

A lawyer who acts for two clients on the same side of a matter opposes neither of them, so the question under Rule 1.7(a)(2) is whether duties to one will materially limit the work for the other.⁹ The answer depends on what each joint client is owed. Comment 31 to Rule 1.7 states it in two parts: the lawyer owes each client an equal duty of loyalty, and each client has the right to be informed of anything bearing on the representation that might affect that client’s interests and to expect the lawyer to use that information to that client’s benefit.¹⁰

Equal loyalty leaves the lawyer with no side to take. With one client, a helpful or harmful fact simply goes to work for that client; with two, the same fact may help one and hurt the other, and neither is entitled to less of the lawyer’s allegiance. The information right enforces the same impartiality. Because the lawyer’s common duty is to represent both clients zealously, all material information relating to the representation must be shared with both, including what one client told the lawyer outside the other’s presence.¹¹ Who supplied a fact does not matter; what matters is whether it bears on the shared matter.

Secrets between joint clients therefore cannot work. Both clients are entitled to be kept fully informed, so the lawyer may not keep a secret for one client that is contrary to the interest of the other.¹² If the lawyer honors a request for silence, the lawyer withholds from one client something that client has a right to know, and may give advice while concealing the very fact that would change it. If the lawyer discloses the fact, the first client’s confidence is betrayed, and equal loyalty forbids preferring either. A duty owed to one client goes unperformed whichever way the lawyer turns, and that is the significant risk of material limitation the rule describes.¹³

Withdrawal is the usual consequence, and it is usually total. If a conflict or confidentiality problem arises and the lawyer did not obtain written consent 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 confidences will not be breached.¹⁴ Asking for consent after the conflict surfaces is the alternative, but the duties of loyalty and confidentiality often make it nearly impossible.¹⁵ Dropping one client and keeping the other rarely helps, because by then the lawyer holds confidences from both.

The attorney-client privilege shields confidential communications between client and lawyer from compelled disclosure, and in a joint representation it works against outsiders but not between the clients. When two or more persons jointly consult the same lawyer, neither may invoke the privilege to prevent the lawyer from disclosing communications to the other.¹⁶ The rule is not an exception to the privilege, because the communications remain privileged as to the outside world; it reflects the sharing duty, since clients who are entitled to everything the lawyer learns can hold no expectation of confidence against each other.¹⁷ If the clients later become adversaries on the matter, neither may invoke the privilege against the other, while the privilege stays robust against strangers to the relationship.¹⁸

The two audiences are the point students most often miss. A joint client who hears that the privilege does not protect communications between the clients should not conclude that the communications are open to the world. The loss is confined to disputes between the joint clients themselves.

Because most clients arrive expecting strict confidentiality, the lawyer should explain at the outset how confidential information will be treated and that neither client may invoke the privilege against the other.¹⁹ The same conversation should cover the possibility of divided loyalty and what will happen if a conflict emerges or the representation must end.²⁰ A client who learns the rule only after confiding has already been hurt by the surprise. Whether clients may agree in advance to a narrower arrangement is a consent question, taught separately.

Two mistakes recur on exams. One treats a joint client’s request for confidentiality as binding on the lawyer. The other reads the joint-client rule to mean that the clients’ communications are open to outsiders.

Assessing Divergence Risk

The test for a joint representation is Rule 1.7(a)(2): whether there is a significant risk that representing one client will be materially limited by the lawyer’s responsibilities to the other.²² Some possibility of differing interests accompanies every simultaneous representation, so the word “significant” does the sorting.²³ Far-fetched possibilities of harm do not create conflicts; but if you can imagine a dishonest, lazy, or careless lawyer failing to carry out a duty owed to one client because of the representation of another, a conflict exists.²⁴ The test therefore weighs how likely divergence is and how much it would limit what the lawyer can do for each client. What changes in the joint setting is the source of the pull, because the competing responsibility is owed to the client sitting beside the one in question.

Comment 23 to Rule 1.7 identifies where divergence tends to surface when one lawyer serves several clients. Its examples include “substantial discrepancy in the clients’ testimony,” “incompatibility in positions in relation to an opposing party,” and “substantially different possibilities of settlement.”²⁵ Each marks a point where duties to one client begin to demand what duties to the other forbid.

Testimony is the plainest divergence, because a lawyer cannot adjust what a witness remembers. If two clients will give different accounts of one event, each client’s best course is to have the lawyer press that client’s version, which usually undercuts the other’s. A lawyer who presses both versions at full strength contradicts the lawyer’s own case, and a lawyer who softens one account to protect the other has limited the representation of the client whose account was softened.

Incompatible positions toward the opposing party show up in the defenses the clients have available. Suppose one lawyer defends a property owner and its manager against a suit over a broken staircase. The owner’s best argument is that the manager ignored written maintenance instructions; the manager’s best argument is that the owner refused to fund repairs. Each client’s best position is the other’s worst, so no course of action serves both. The divergence is built into the defenses from the first day, and the lawyer need not wait for the case to develop to see it.

Settlement exposes divergence because every offer forces a decision. A lawyer asked to advise on an offer must say what is best for each client individually, and when the clients stand to gain or lose different amounts, or value different outcomes, no single recommendation serves both. Equal loyalty gives the lawyer no basis for preferring one.

In the balcony, staircase, and heating examples the divergence is visible at intake. Bottoms v. Stapleton shows the same inquiry reaching the opposite answer. One lawyer represented a limited liability company and its majority shareholder as defendants, and the court focused on the degree of risk that the lawyer would be unable to fulfill duties to both.²⁶ It found the defendants’ interests not adverse in the present posture of the case and no substantial evidence that they would likely become adverse, particularly because one defendant was an LLC and the other held the controlling interest in it.²⁷ It therefore found no significant potential for divergence and no present conflict.²⁸ Nothing in those facts shows a clash of accounts, incompatible positions toward the plaintiff, or different settlement goals.

The rule has two sides, and the two poles above mark them. When the clients’ commonality of interest is high and the likelihood of conflict insignificant, the rule does not demand consent to the shared representation. When there is a significant risk that their interests will diverge, especially if the divergence already exists, even on what appear to be minor or subsidiary points, such that the lawyer’s loyalty would be divided, a present conflict exists, and the lawyer may undertake the representation only if informed consent can be obtained.²⁹

Minor points count because equal loyalty obliges the lawyer to give each client sound advice wherever the clients’ interests differ, and a point that looks small in the file may matter a great deal to the client it affects. A disagreement that has already surfaced is also a fact rather than a forecast, so the likelihood half of the test is met, and the only remaining question is whether the lawyer could recommend the best course for one client on that point without shortchanging the other. Students therefore err in two directions. Some treat the clients’ shared goal as the end of the analysis, which overlooks divergence already present. Others treat any conceivable difference as proof of a conflict, which ignores the requirement that the risk be significant.

Misjudging the risk is costly. If the joint representation fails and the lawyer can no longer use independent judgment for each client or assure each that confidences will be kept, the lawyer must withdraw from both.³⁰ A prudent lawyer therefore consults the clients in every multiple representation, even when the risk looks too small to require consent. The lawyer raises the possibility of divided loyalty, explores the points along which the clients might divide, explains how confidential information will be treated, and considers how a conflict would be reconciled or the representation ended.³¹

Civil Co-Parties

Co-plaintiffs and co-defendants on the same side of a civil case are tested under Rule 1.7(a)(2).³³ They are not opposing each other when the engagement begins, so direct adversity has nothing to catch at the outset.³⁴ The inquiry becomes whether their interests are likely to part ways and whether the lawyer could still serve both if they do. In civil litigation the answer is often favorable. Joint representation in a criminal matter is generally inadvisable, but representing several clients on the same side of a civil case is not only common but frequently makes good sense.³⁵

The lawyer may proceed only if it is apparent that the clients are reasonably congenial and dedicated to a common objective. If the danger that a conflict will develop is significant, the lawyer may not continue.³⁶ Four warning signs show where that condition fails, and any one of them ordinarily means the lawyer may not undertake or continue the representation:³⁷

1. The clients show animosity toward each other, which turns every strategic decision into a dispute the lawyer must referee.
2. There is a significant prospect that the clients will stake out adverse positions or assert claims against each other, which would put the lawyer in the position of advocating against a current client.
3. One client appears likely to seek more favorable treatment from the lawyer than the other receives, which defeats the equal loyalty each is owed.
4. One client wishes to withhold confidential information from the other, which the lawyer cannot honor because both are entitled to be told what bears on their shared matter.

The second sign shows how a joint representation can move from one definition of conflict into the other. While a claim between the clients is only a prospect, the lawyer’s duties to each limit what the lawyer can do for the other, which is material limitation. Once one client’s best course is to assert that claim, the lawyer would be acting as an advocate against a current client, and paragraph (a)(1) applies without any inquiry into likelihood or into whether the matters are related.³⁸

The clients, the accident, and the opposing motorist are identical in both halves of that story, and the one fact that moved is a fact that hands one client a claim against the other. Before the admission, advice that served the driver served the passenger, because both gained from proving the other motorist’s fault. Afterward, the passenger’s best course includes proving the driver’s fault while the driver’s best course includes keeping it quiet, so the duty to inform the passenger and the duty not to use the driver’s statements against her cannot both be performed. Exam answers go wrong in two mirror-image ways here. One treats the shared goal of recovering from the defendant as proof that no conflict can arise, which overlooks claims the co-parties may hold against each other. The other treats the bare chance that co-plaintiffs will someday disagree as proof that a conflict already exists, which ignores the requirement that the risk be significant.

Dropping the driver and keeping the passenger does not solve the problem. The hot-potato doctrine sharply limits a lawyer’s ability to withdraw prematurely from a representation in order to take on the adversary of the now-former client.³⁹ By the time the claim surfaces, the lawyer holds confidences from both clients and owes loyalty to each, so continuing for either would be adverse to the other, and the ordinary result is that the lawyer may not continue.⁴⁰

The car pool presents the second warning sign in a form visible at intake, before anyone has confided anything, so the lawyer must decide at the outset rather than midstream. Finding the conflict establishes only that it exists. Rule 1.7(a) prohibits the representation except as paragraph (b) provides, so the next question is whether paragraph (b) lets the lawyer proceed, a question taught in a separate reading that includes informed consent from the clients.⁴¹ Where the case turns on allocating fault between the co-parties, that route is hard to sustain. An offer that would settle the claims of several co-parties together raises further problems, which a later reading covers.⁴²

Transactional Joint Clients

A lawyer who acts for several clients outside a lawsuit faces the same Rule 1.7(a)(2) question: whether there is a significant risk that duties to one client will materially limit the representation of another.⁴⁴ The setting changes how the risk is weighed. Founders forming a company, spouses preparing wills, and parties documenting a deal they have already struck all want one result, and one lawyer can serve that result without the cost and friction of separate counsel.

Harmony makes the arrangement sensible but not risk-free. Clients who agree today may part ways tomorrow on a term that is still open, and because the rule looks to the risk that interests will diverge at some point, the lawyer must predict as well as observe.⁴⁵ Even where that risk looks too small to require consent at the outset, it is advisable to consult the clients about the chance of divided loyalty, how confidential information will be treated, and what will happen if a conflict emerges.⁴⁶

No adversary sits across the table in a transaction, so the lawyer’s job is to move the clients toward a shared goal. When the clients have settled the terms and ask only for documents, the lawyer never has to press one client’s position against the other’s, and a difference on a peripheral point leaves each client’s options intact. The role has a ceiling, though. Once a term remains open, someone must decide whose interest prevails, and a lawyer who owes equal loyalty to both cannot push either one.

The two couples differ on one fact: whether the terms are settled. That fact decides whether the lawyer’s facilitating role fits or whether the lawyer must advise each client on what is best for that client alone.

Employment-based immigration shows the same pattern in a commercial setting. Dual representation of an employer and a non-citizen employee is possible because both seek the same objective, an immigration benefit that lets the employee work for the employer, and each may be willing to accommodate the other’s interests.⁴⁷ Lawyers in dual representation must still take special care from the outset, because two ethical concerns arise,⁴⁸ and the withdrawal rules show what they are: a conflict of interest and a confidentiality problem.⁴⁹

Elder law adds a threshold question that comes before any conflict analysis: who is the client? The lawyer must decide whom the lawyer represents before testing for a conflict. A relative who drives the client to the office, translates, co-owns property, or pays the bill has a role the lawyer must identify, and a helper is not necessarily a client. Good intentions do not erase the relative’s own interests. If Mother creates a joint bank account with Son to facilitate his handling of her financial affairs, her plan to split her estate between Son and Daughter may be jeopardized by Son’s ability to drain the account for his own benefit.⁵² A lawyer asked to serve Mother and Son together on that estate plan would face a divergence on a term central to the work. A lawyer who represents only Mother, with Son in the room to assist, has no co-client whose interests compete with hers.

Some transactions are poor candidates for joint representation because the terms are open and one side’s gain is the other’s loss.

Spouses dividing property, support, and parenting time contest the same kind of zero-sum questions, and the same analysis applies. When an unavoidable conflict emerges in a transactional matter anyway, the lawyer’s options are narrow, and withdrawal from both clients is the ordinary result if the lawyer cannot use independent judgment for each or protect each one’s confidences.⁵³ Planning at intake is easier than a mid-matter exit, which is why the prediction carries so much weight.

Criminal Co-Defendants

A lawyer who defends two people charged in the same crime stands beside both against the prosecution, so neither is opposed at the outset and the analysis runs through Rule 1.7(a)(2): is there a significant risk that representing one will be materially limited by responsibilities to the other?⁵⁵ Criminal defense is where that risk is greatest, and Comment 23 says so directly: “[t]he potential for conflict of interest in representing multiple defendants in a criminal case is so grave that ordinarily a lawyer should decline to represent more than one codefendant.”⁵⁶ The word “ordinarily” carries weight. The Model Rules do not prohibit the arrangement outright, and an occasional case presents a unified defense that benefits both clients.⁵⁷ The comment sets a strong presumption against joint representation while leaving room for exceptions.

The presumption is strong because a prosecutor can reward one defendant for testimony against another, the defendants’ roles and records seldom match, and each client’s liberty depends on choices that may help one at the other’s expense. Interests that look identical at arraignment can part at nearly every later stage:

1. At plea bargaining, the prosecutor may offer a reduced charge or dismissal to one defendant in exchange for testimony against the other, and the lawyer cannot advise the first client to accept without harming the second.⁵⁸
2. At trial, the defendants may raise conflicting defenses that implicate each other, and a decision about whether either testifies helps one and hurts the other, because the testimony of one highlights the silence of the other.⁵⁹
3. At closing argument, a comparison showing that the evidence is much weaker against one defendant serves that defendant and damages the other.⁶⁰
4. At sentencing, the defendants’ roles in planning and committing the crime, or their backgrounds and prior records, often differ substantially, and the same kind of conflict follows.⁶¹

Each stage is a place where a duty owed to one client demands what the duties owed to the other forbid. A lawyer juggling two defendants may be unable to strike separate plea agreements, unable to determine which theory of the case best supports one client’s defense, and unable even to consider offering one client’s testimony against the other in return for a reduced sentence or immunity.⁶² Those lost options are the limitation, and the likelihood half of the test is met because prosecutors routinely make exactly these offers. The harm is also hard to detect, since a plea discussion that never took place leaves nothing behind for a reviewing court to examine.

The Supreme Court gave that invisibility its sharpest statement in Holloway v. Arkansas. Counsel for several defendants renewed his motion for separate counsel on the ground that if one or two of them testified he could not cross-examine them, because he had received confidential information from them. The trial court denied the motion and told counsel to proceed.⁶³ The Court explained that in a joint representation of conflicting interests the evil lies in what the advocate finds himself compelled to refrain from doing, not only at trial but also in possible pretrial plea negotiations and at sentencing.⁶⁴ An appellate court asked afterward to judge the effect on plea negotiations would be reduced to unguided speculation.⁶⁵

A joint trial is not required for the problem to arise. In Wheat v. United States, attorney Iredale represented both Bravo and the petitioner, and the government asked that Bravo be made available to testify against the petitioner in exchange for a modified position at Bravo’s sentencing.⁶⁶ If Bravo were called, Iredale’s position would become untenable, because ethical proscriptions would forbid him to cross-examine Bravo in any meaningful way, and failing to do so would deny the petitioner effective assistance of counsel.⁶⁷ Joint representation of defendants who will be tried separately is itself a recognized conflict setting, as is representing, in the same or another matter, a likely prosecution witness or the victim of the alleged offense.⁶⁸

When one client becomes a witness against the other, the analysis can also move from the second definition of conflict to the first. A lawyer who must cross-examine a current client is acting as an advocate against that client, and direct adversity under Rule 1.7(a)(1) calls for no showing about likelihood.⁶⁹ The situation is especially disturbing when the lawyer’s other client is a key witness for the prosecution against the lawyer’s criminal defendant client.⁷⁰

None of this makes joint criminal defense invalid per se. Holloway itself recognized that multiple defendants can appropriately share one lawyer in some cases and that certain advantages might follow, quoting Justice Frankfurter’s view that “[j]oint representation is a means of insuring against reciprocal recrimination. A common defense often gives strength against a common attack.”⁷¹ Pooled resources and one lawyer coordinating one strategy can keep a prosecutor from dividing the defendants.⁷²

The usual exam errors run in opposite directions. One treats Comment 23 as an absolute ban and ignores that the Model Rules allow the arrangement in the rare case where interests align. The other reasons from the clients’ shared wish for acquittal that no conflict exists, which overlooks the plea-offer and testimony risks that make the comment’s caution so strong. Whether informed consent can cure the conflict is a paragraph (b) question, and the ABA’s Standards for Criminal Justice advise that each client’s informed written consent be obtained and made on the record before the court, with appropriate inquiries by counsel and the court.⁷³

The Sixth Amendment adds a separate layer, and it should stay separate in your analysis. The constitutional right to effective assistance entitles the defendant to the undivided loyalty of counsel.⁷⁴ When a trial court improperly required joint representation of multiple defendants over their timely objection, Holloway presumed prejudice and made reversal automatic, while a trial judge need not inquire into multiple representation when defense counsel never asks for separate counsel.⁷⁵ Those doctrines govern when a conviction is set aside. The disciplinary question is whether the lawyer may take or continue the representation, and it is answered at the outset by asking whether there is a significant risk of material limitation, with no proof that the defendant was actually harmed.⁷⁶

Organization and Constituent

A concurrent conflict between an organization and one of its people exists only if the lawyer owes duties to both, so the first question is whether the individual is a client at all. An organization acts only through directors, officers, employees, members, shareholders, and other constituents, and a lawyer retained by the entity works with them daily without representing them. Rule 1.13(g) lets the lawyer choose to represent a constituent too, and the constituent then becomes an additional client, subject to the conflict-of-interest limits.⁷⁸

Once the lawyer has taken both, the arrangement is called a dual representation, and the lawyer simultaneously assumes the full range of professional responsibilities to the organization and to the individual.⁷⁹ The lawyer must represent each client loyally and zealously and safeguard each one’s confidential information. The rule permits this, but it carries risk, because the constituent’s interests may become adverse to the organization’s, or the reverse, and the lawyer could then be disqualified from continuing for either.⁸⁰ The size of the risk depends on the purpose of the work done for the constituent and on the nature of the entity.⁸¹

Unrelated work sits at the low-risk end. A lawyer for a corporation who also drafts its chief financial officer’s will is representing the officer in a matter that calls on no duty owed to the corporation, so the potential for conflict is reduced. Reduced risk is still some risk. Both people are current clients, so even if the representations are unrelated, the lawyer may not represent one adversely to the other.⁸² For direct adversity, relatedness is immaterial,⁸³ which is the Earl’s Rooter lesson from earlier in this chapter.

Related work sits at the other end, and the question is how likely it is that the organization’s interests in the matter will diverge from the individual’s. In many instances a united defense of the organization and its members is appropriate, effective, and cost-efficient.⁸⁴ In others, the organization may want at some point to jettison the officer or employee and point the finger at that person as primarily responsible, which makes continued dual representation impossible.⁸⁵ A lawyer defending a company and a manager in the same harassment suit shows the problem. The company’s best argument may be that the manager disregarded company policy, while the manager’s best argument may be that senior management approved the conduct. Each client’s best position is the other’s worst, so a dishonest, lazy, or careless lawyer could easily fail a duty owed to one client because of the representation of the other, and that is a conflict.⁸⁶

Confidentiality makes the divergence harder to manage, because the lawyer owes that duty to both clients. The lawyer may not “interrogat[e] one client to benefit another client” or disclose the individual’s privileged communications to third parties without the individual’s consent.⁸⁷ What the manager tells the lawyer in preparing the defense therefore cannot be turned against the manager to help the company, and what the company reveals cannot be turned against it to help the manager. This is the same mechanism that made secrets between joint clients unworkable earlier in this chapter.

The nature of the entity also moves the risk. Where the organization is a closely held corporation, the interests of the entity and the individual are more likely to run in parallel, so dual representation of a principal is less likely to provoke a conflict.⁸⁸

The most serious case arises when the constituent and the organization become directly adverse in the very matter on which the lawyer is engaged in dual representation, because one has commenced or is about to commence litigation or introduce a cross-claim against the other. That conflict is irresolvable, and the lawyer ordinarily may not continue to represent either client.⁸⁹ Direct adversity under Rule 1.7(a)(1) needs no showing about how likely harm is, since the lawyer would be advocating against a current client.⁹⁰ The lawyer cannot escape by keeping the organization and dropping the individual, because the hot-potato doctrine sharply limits a lawyer’s ability to withdraw prematurely from a representation in order to take on the adversary of the now-former client.⁹¹ A suit with the same lawyer on both sides also illustrates the conflicts clients cannot waive, since the test for non-consentable conflicts asks whether the lawyer is representing parties on opposite sides of the litigation.⁹²

Between the two poles, where no one has yet moved against anyone, the inquiry is whether there is a significant risk of material limitation.⁹³ If the risk is significant but manageable, the lawyer moves to the consent inquiry, taught separately. A lawyer may try to head off withdrawal from the organization by securing its agreement in advance, but Lawrence Fox and Professor Susan Martyn warn that this is “a prospective waiver that can always be challenged later as not sufficiently well informed.”⁹⁴

The sorting turns on step three. A lawyer who sees a united defense and stops there has treated the clients’ present harmony as a guarantee, and a lawyer who sees a cross-claim has no analysis left to run.