The Core Prohibition

Assisting Client Crime or Fraud

The Core Prohibition

Rule 1.2(d) of the Model Rules of Professional Conduct states the prohibition in a single sentence: “A lawyer shall not counsel a client to engage, or assist a client, in conduct that the lawyer knows is criminal or fraudulent, but a lawyer may discuss the legal consequences of any proposed course of conduct with a client and may counsel or assist a client to make a good faith effort to determine the validity, scope, meaning or application of the law.”¹ The word “but” divides the sentence in two. Before it comes the prohibition, which this section teaches. After it comes a lawyer’s freedom to talk candidly with clients about what the law allows, which a later section develops. Together the halves show a rule aimed at taking part in wrongdoing, not at giving advice about the law.

The prohibition marks the point where the client’s authority over the representation ends. Rule 1.2(a) directs a lawyer to abide by the client’s decisions about the objectives of the representation and to consult the client about the means, but it opens with the words “Subject to paragraphs (c) and (d).”² Paragraph (d) is the prohibition, so the client’s control over objectives stops where crime or fraud begins. The reason lies in what a lawyer contributes to any undertaking: legal skill, access to courts and agencies, and the legitimacy other people presume in whatever a lawyer touches. A client may direct those assets toward lawful ends, because the lawyer acts as the client’s agent for them. A lawyer who knowingly lends them to a crime or fraud shares in the wrong, and the case for deferring to the client no longer applies.

A violation requires three elements at once. The first is a forbidden act: counseling or assisting, which together reach the lawyer who recommends a wrong and the lawyer who merely helps carry it out. The second is a mental state: knowledge, which shields a lawyer whom a client uses unwittingly. The third is a subject: conduct that is criminal or fraudulent, terms that take their content from law outside the Rules, so the ethics rule never writes its own definition of wrongdoing. The sections below take the three in that order.

Counseling and Assisting

Rule 1.2(d) names two ways to take part in a client’s wrong, and its grammar keeps them apart: a lawyer “shall not counsel a client to engage, or assist a client, in conduct that the lawyer knows is criminal or fraudulent.”³ To counsel is to recommend, urge, or advise the wrongful conduct, which makes the lawyer the source of the idea or the push that moves the client to act. To assist is to furnish services that carry out conduct the client has already settled on, which makes the lawyer the instrument of someone else’s decision. Either verb satisfies the act element. A lawyer who proposes nothing but implements everything therefore stands where the lawyer who drew the plan stands, once knowledge and a criminal or fraudulent subject are added.

The rule has two verbs because the two contributions work differently. A client who hears a lawyer call a scheme workable acts partly on the strength of the lawyer’s professional judgment, so counseling puts the lawyer’s authority behind the decision itself. A client who has already decided needs something else: the skill to finish the scheme and the legitimacy that follows a lawyer’s signature, closing, or court filing. The lawyer supplies both whether or not the idea was the lawyer’s. Because the harm comes from the contribution, the identity of the scheme’s author does not decide the question. Assistance also hides more easily, since it can be assembled from tasks that each look routine.

Posting bail, transferring real estate, and preparing papers for a purchase are services lawyers perform for lawful clients every day, so none of them announces a violation. What turns them into assistance is what the lawyer knows the combination serves.

The services in the two scenes are the same kind of work. The answer changes because of what the second lawyer knows the work is for, so the lawfulness of a task taken alone settles nothing. The question is whether the lawyer, with that knowledge, is furnishing the machinery of a wrong.

The same reasoning reaches a lawyer who puts false material into a filing, and the example separates the two verbs.

The rule stops short of holding a lawyer answerable for everything a client later does with the lawyer’s work. The comment to Rule 1.2 requires a lawyer to give an honest opinion about the actual consequences likely to result from a client’s conduct, and it adds that a client’s use of advice in a criminal or fraudulent course of action “does not, of itself, make a lawyer a party to the course of action.”⁴ If every misused opinion counted as assistance, lawyers would have reason to hold back candid analysis, and clients would decide with less information. A lawyer who accurately explains the law governing a transaction, and who never learns that the client folded the advice into a fraud, has informed a choice and furthered nothing. As one opinion puts it, there is a “critical distinction between presenting an analysis of legal aspects of questionable conduct and recommending the means by which a crime or fraud might be committed with impunity.”⁵ The working test therefore asks whether the lawyer furthered the wrong. Neither the origin of the idea nor the client’s later use of advice can answer that question.

In SEC v. Fehn, the Ninth Circuit faced a securities lawyer whose client, Wheeler, adamantly refused to disclose an SEC investigation and the contingent liabilities stemming from past violations in the company’s quarterly reports.⁶ The omission was the client’s choice. Fehn’s part was to review and personally alter a draft Form 10-Q prepared by a non-lawyer employee, to leave editing notations on later reports prepared by his firm, and to submit the forms to the SEC without properly advising the client of the material omissions.⁷ The court held that this editing and submission lent “substantial assistance” to the client’s securities violation. It rejected his claim of good faith because his efforts were not “reasonable” in light of well-established disclosure requirements.⁸ Fehn was an SEC enforcement action, so it applied the securities laws’ own standard rather than Rule 1.2(d), and the ethics rule’s demand that the lawyer actually know is the subject of the next section. What the case shows is the shape of assistance: the client decided, and the lawyer implemented by editing and filing.

The Knowledge Requirement

Rule 1.2(d) reaches only conduct “that the lawyer knows is criminal or fraudulent,” and that verb separates the lawyer who has joined a client’s wrong from the lawyer who merely stood near it.¹⁰ The Rule says “knows,” not “suspects” or “should have realized,” so a violation requires that the lawyer actually understood the conduct to be criminal or fraudulent. Two reasons support the standard. It protects the lawyer whom a client uses without the lawyer’s awareness, who would otherwise be disciplined for failing to detect what the client concealed. And it keeps counsel available to clients whose situations are only aggressive or uncertain, because a lawyer who could be disciplined on a hunch would turn those clients away, and they most need someone to say where the legal boundary lies.

Actual awareness does not require that the client announce the wrongdoing. If knowledge could be proved only by an admission, a lawyer could escape the Rule by never asking questions and never reading closely the papers in front of the lawyer. A factfinder may therefore infer knowledge from the surrounding circumstances, and the lawyer’s denial is one piece of evidence weighed against the rest. re Zang shows the method. It is an Arizona Supreme Court disciplinary decision under the Model Code of Professional Responsibility, the Model Rules’ predecessor, whose counterpart provision barred a lawyer from “[c]ounsel[ing] or assist[ing] his client in conduct that the lawyer knows to be illegal or fraudulent.”¹¹ The knowledge question is therefore the one the Model Rules pose today.

Zang represented Betty Daniels after an automobile accident. He obtained a settlement draft from her own insurer, State Farm, covering medical bills and $1,400 in property damage, and then settled with the other driver’s insurer, Equitable, for $3,900. When State Farm asserted its subrogation claim (its right to be repaid from the client’s recovery for the property damage it had already covered), Zang refused it and insisted that the Equitable money paid only for bodily injury.¹² The disciplinary bodies rejected that account and found that he knew otherwise. The court upheld the finding on circumstantial grounds. Equitable’s check stated that it was issued in payment of any and all claims, the release Zang signed covered both personal injuries and property damage, and he never explained why neither he nor Equitable struck the property damage language from the form.¹³ His experience counted too. He had negotiated and participated in many similar settlements, and the court found it hard to believe that he both failed to understand the plain language of the check and negligently failed to modify the release.¹⁴ The finding rested on the documents he handled and the experience he brought to them.

The inference in Zang ran from facts the lawyer actually held, and a hunch supplies none.

Knowledge also works as a clock, because the moment it arrives fixes what the lawyer must do. Before an engagement the lawyer can simply refuse it. Once work is under way, the only way to stop assisting may be to leave. The first move in any problem is therefore to place on a timeline the moment, if any, when the lawyer came to know.

The middle case catches lawyers who began in good faith. The comment to the Texas version of the rule states it directly: a lawyer “may not continue assisting a client in conduct that the lawyer originally supposes is legally proper but then discovers is criminal or fraudulent,” and “[w]ithdrawal from the representation, therefore, may be required.”¹⁷ Knowledge is measured at the time of each act. Work done before the lawyer knew does not become assistance after the fact, but every act done afterward is done with knowledge.

The word “knows” is not universal across the states. Florida’s version of the rule prohibits assisting a client in conduct that the lawyer “knows or reasonably should know” is criminal or fraudulent.¹⁸ The added phrase makes the standard partly objective, so a lawyer who ignores obvious warning signs can be reached without proof of actual awareness, which the Model Rule would require. Texas adds a separate duty of its own: when a lawyer has confidential information “clearly establishing” that a client is likely to commit a criminal or fraudulent act likely to cause substantial injury to the financial interests or property of another, the lawyer must promptly make reasonable efforts under the circumstances to dissuade the client.¹⁹ That threshold differs from knowledge under Rule 1.2(d), and the later section on warning and dissuading takes it up. An exam answer should state the rule as the Model Rules frame it, with “knows” as the standard, and flag a variation when the problem names a jurisdiction that departs from it.

At the far end of the timeline sits the lawyer who never learns. A commentary on Rule 4.1(b), which turns on the same idea of assisting, observes that a client who appropriates a lawyer’s services for a fraud the lawyer never discovers has given the lawyer nothing to stop, and that the oblivious lawyer lacks the scienter to be held responsible later as an accomplice.²⁰ That protection depends on the facts, because a factfinder remains free to disbelieve a disclaimer of knowledge, as Zang’s own denial was disbelieved.

Immigration practice supplies a quieter version of the problem, because only the client knows the client’s own intent, and intent can change. A student visa applicant who otherwise qualifies may be ineligible if the applicant has a present intent to immigrate, so a client who mentions a wish to stay permanently puts the lawyer in a delicate spot.²¹ A lawyer who explains the legal effect of each possible intent is not urging a course of conduct or assisting a fraud. The lawyer describes two legal alternatives and their consequences, and the client decides.²² Timing matters, though. The key is to convey that information before the client describes his intent in detail, because a detailed statement of a fixed intent, followed by a request to attest the opposite, could foreclose the lawyer from assisting further.²³

Criminal or Fraudulent Conduct

Rule 1.2(d) forbids assisting conduct “that the lawyer knows is criminal or fraudulent,” yet the rule never says what makes conduct criminal or fraudulent.²⁵ It borrows both words. Whether a client’s plan is a crime depends on the penal law that governs it. Whether it is fraud depends on Rule 1.0(d), which defines fraud as conduct “that is fraudulent under the substantive or procedural law of the applicable jurisdiction and has a purpose to deceive.”²⁶ The definition has two parts: an outside body of law that decides what counts as fraud, and a mental element, the purpose to deceive. In plain terms, “[d]eliberate action that misleads another is fraud, and a lawyer may not counsel or assist a client in such conduct.”²⁷

Each part does a different job. Borrowing the content from outside law keeps the ethics rules from becoming a second criminal code. If Rule 1.2(d) carried its own list of forbidden schemes, the list would need revising whenever a legislature changed a statute, and lawyers would answer to a standard different from the one that binds everyone else. Tying the rule to substantive law means the line a lawyer must respect is the line the client faces, and when that law is unclear, legal analysis can supply an answer. The purpose-to-deceive element keeps aggressive but lawful arrangements outside the rule, because a plan that is merely hard on the other side lacks the deliberate effort to mislead that fraud requires. The same service can therefore be proper for one client and forbidden for another: the lawyer’s task is identical, and only the outside law’s verdict on the client’s conduct differs.

Examinees most often skip the first step. Calling a client’s plan “improper” and moving straight to counseling and assisting leaves the analysis without a foundation, because the rule has no force until some other body of law has classified the conduct as a crime or a fraud. Two settings show that classification at work, and in each the lawyer’s exposure turns on law found outside the Rules.

Asset protection planning is the first. The traditional and still general rule is that one cannot use a trust to shield one’s own assets from one’s creditors, and when the beneficiary is the settlor, creditors have greater rights to reach the beneficiary’s interest in the trust.²⁸ Some states nonetheless recognize asset protection trusts, in which a person transfers property into a trust that the person also benefits from, so the planning is lawful in itself in those places. What separates a lawful plan from a fraud is the aim and timing of the transfer. A transfer made with intent to hinder or defraud a creditor’s claim is actual fraud, recoverable under the widely adopted Uniform Fraudulent Transfer Act, while a transfer made before any creditor’s claim arises is treated differently.²⁹ Rule 1.2(d) is the reason a lawyer must proceed very carefully in creating a trust that would delay, hinder, or defraud a client’s existing or foreseeable creditors, since knowing help with that kind of transfer is help with a fraud.³⁰

The words “existing or foreseeable” carry weight in that caution. A creditor need not have sued for a transfer to be aimed at that creditor, so the lawyer must ask what the client expects to happen and cannot stop at what has been filed. The lawyer must also check the law of the particular jurisdiction, because the Uniform Voidable Transactions Act, adopted in several states, contains a controversial provision that makes certain transfers to an out-of-state asset protection trust voidable even when the transferor has no pending, threatened, or expected claims.³¹ The definition of the wrong can shift from state to state, and the reach of Rule 1.2(d) shifts with it.

The enforcement side of the same line appears in Federal Trade Commission v. Affordable Media, LLC. Denyse and Michael Anderson ran a telemarketing scheme that operated as a Ponzi scheme, defrauded investors of millions of dollars, and sheltered the proceeds in a Cook Islands trust, an offshore self-settled spendthrift trust of which they were co-trustees.³² After the FTC won an injunction ordering them to account for the money and to repatriate it (bring it back to the United States), the district court held them in civil contempt, and the Ninth Circuit affirmed. The court was not convinced that the Andersons had put the money beyond their control, particularly because they remained the trust’s “protectors,” a position that carries significant control over an offshore trust, including the appointment of successor trustees.³³ No lawyer’s discipline was at issue, so the case shows what an actual fraud looks like, not how Rule 1.2(d) applies. The fraud was the Ponzi scheme, which other law defines, and the trust was the device that put its proceeds out of reach. A lawyer who knew the money came from that scheme and built the structure to hold it would be furnishing the machinery of a known fraud, which is the assistance the rule names.

Threats of criminal prosecution show the same dependence on outside law in a setting with no trust at all. A threat by an attorney to institute criminal proceedings in order to aid a client in a civil dispute may make any resulting agreement unenforceable for duress and may also subject the attorney to discipline for professional misconduct.³⁴ Such threats may amount to extortion under state criminal statutes, and Rule 1.2(d) and Rule 4.4 are among the ethical grounds pointed to for condemning them. Some states have modified the Model Rules to prohibit threats of criminal proceedings used to gain an advantage in a civil matter, South Carolina’s Rule 4.5 being an example.³⁵ The Restatement (Second) of Contracts counts threats of criminal prosecution as improper threats for duress purposes, while threats to litigate are not per se improper unless made in bad faith.³⁶

The second letter is dangerous because of what the threat does. A civil claim exposes the other party to a loss that depends on the merits, which the party can contest in court. A threat of prosecution adds a loss imposed by state power that has nothing to do with whether the money is owed, and that is the pressure criminal statutes against extortion exist to reach. The ethics rule supplies the professional standard for the lawyer, and any criminal or civil liability comes from the substantive law itself, a point the closing section of this chapter returns to.

Because the definition of wrongdoing lives in outside law, and outside law is sometimes unsettled, a lawyer needs room to ask what that law means without becoming a participant in whatever the answer turns out to be. The second half of Rule 1.2(d) supplies that room by protecting good-faith efforts to determine the validity, scope, meaning, or application of the law, and the next major section develops it.³⁷