The CFA Code & the Seven Standards: Vignette Lab

Branch: Ethics, Standards & GIPS (Finance-Plus ring) · Node ES1.01 · ~14 focused hours · Mastery gate ≥ 85%

An analyst reads a vignette: a portfolio manager buys 10,000 shares for herself an hour before entering a 200,000-share client order in the same name. Ask most people what happened and they say "that seems bad." Ask them which rule it breaks, whether it breaks it, and what one fact proves it, and the confident answer usually evaporates. That gap (between a vague sense that something is off and a precise, defensible citation) is exactly what this module closes. Ethics vignettes on the CFA exams, and in real compliance work, are graded on precision, not vibes: you must name the specific Standard (a Roman numeral and a letter), state plainly whether it was violated, and point to the one fact in the vignette that decides the case. "It felt wrong" earns zero points. "Standard VI(B), Priority of Transactions, violated, his personal trade ran ahead of and benefited from a client order he knew was imminent, for a ₹22,000 unfair advantage" earns full marks, because it is falsifiable, sourced, and complete.

This Ring is additive institutional/CFA breadth, it presents the toolkit and the plumbing alongside, not in place of, the program's process-over-P&L, concentration-over-diversification philosophy. Ethics is the one place in the Ring where that sentence needs the least translation: an analyst who has internalized this program's spine, do the work yourself, verify every number, never substitute a story for a fact, already has most of the temperament this module needs. What you are missing, and what this module supplies, is the formal vocabulary: the twenty-three lettered sub-standards that the CFA Institute's Standards of Practice Handbook uses to name conduct precisely, so that "something felt off" becomes "Standard III(C), Suitability, violated, because the product's risk rating exceeds the client's documented tolerance by six points on a ten-point scale."

The one-sentence version. Every ethics vignette is an equation with three unknowns (which Standard, violated or not, what fact proves it) and this module drills you, vignette after vignette, until solving that equation is as automatic as reading a balance sheet.

Where it sits. This is the first of two Ethics, Standards & GIPS modules. This module owns the CFA Institute Code of Ethics and all seven Standards of Professional Conduct, the personal-conduct rulebook that governs you as an analyst, wherever in the world you practice. Its sibling, ES1.02, owns the Global Investment Performance Standards (GIPS) in full and the SEC/SEBI/FCA regulatory and fiduciary map, the institutional rulebook that governs the firms and jurisdictions you practice inside. The seam between them is drawn precisely in §1 below and held throughout: this module never re-derives GIPS's composite-construction machinery (that is AA1.03's job) or the full GIPS standard (that is ES1.02's), and those modules never re-argue why cherry-picking a composite is a professional-conduct violation (that argument is this module's, and only this module's). The third sibling, ES1.03, owns the Asset Manager Code, the ethical decision-making framework and the conduct regimes that sit at firm level, which is where you go once you can already name the personal-conduct rule.

Suggested hour budget. Core content (read with the Handbook open) ~5h · worked examples ~2.5h · practice set ~3h · applied mini-project ~2h · flashcards and mastery check ~1.5h.

Case Lab: Twelve Standards, Twelve Vignettes (CL-ES.1) is this module's timed capstone drill, twelve fresh vignettes spanning all seven Standards, scored against a seeded key. Open it only after you have worked every vignette below by hand; it is built to feel like a formality once this module's method is installed, not a first encounter with the format.

Learning objectives

By the end you can:

1. State the six components of the CFA Institute Code of Ethics from memory, and explain, in operational terms (not slogans) what "place the integrity of the profession and client interests above your own" actually requires when a real conflict arises.
2. State all seven broad Standards of Professional Conduct and their twenty-three lettered sub-standards, and explain the difference between conduct that is merely unwise and conduct that actually crosses a specific, citable line.
3. Diagnose any vignette using a repeatable four-step method, identify the actor and the action, list the candidate Standards, isolate the single deciding fact the vignette planted, and state the CFA-defended verdict in one sentence, and run that method under time pressure.
4. Apply Standard II(A), Material Nonpublic Information, correctly, including the mosaic-theory boundary that separates a legitimate synthesis of public data from a genuine MNPI violation, and explain why mere possession without acting is not itself a breach.
5. Apply Standard I(C), Misrepresentation, to guaranteed-return language, spliced or blended track records, and uncredited use of others' work, and distinguish all three from properly labeled hypothetical or back-tested performance, which is permitted.
6. Apply Standard III(C), Suitability, by testing a specific recommendation against a client's documented risk profile and constraints, and distinguish a merely aggressive recommendation from one that is affirmatively unsuitable.
7. Tell apart the program's four "conflict-flavored" Standards (I(B) Independence and Objectivity, IV(B) Additional Compensation Arrangements, VI(A) Avoid or Disclose Conflicts, and VI(C) Referral Fees) and route a vignette to the one it is actually testing, not the one that merely sounds closest.
8. Apply Standard V(C), Record Retention, including the CFA Institute's recommended minimum retention period and the governing rule when local law or firm policy requires longer.
9. Distinguish Standard III(A), Loyalty, Prudence, and Care (owed to clients) from Standard IV(A), Loyalty (owed to your employer), two different duties that share a name and are the single most confused pair in the whole framework.
10. Read a performance-presentation vignette and correctly cite Standard III(D) as the conduct rule being tested, while handing the underlying composite-construction mechanics to AA1.03 and the full GIPS standard to ES1.02, never re-deriving either.
11. (Productivity objective: R10 duality.) Use an AI assistant to draft and drill additional practice vignettes fast, while treating every AI-generated Standard citation and verdict as a hypothesis to check against the Standards of Practice Handbook, never as the answer itself.

The duality, stated once (R10). The gated skill is items 1–10: naming the Standard, calling the violation, and proving it with the vignette's own fact, cold, under a clock. The productivity payoff is item 11, an assistant is a tireless vignette-writing partner. But hold this module's own warning in mind while you use one: an ethics vignette is engineered to make the tempting first-glance answer wrong, and a language model pattern-matches on surface features ("a gift was mentioned → must be Standard I(B)") exactly the way a rushed candidate does. This is one of the highest AI-over-trust-risk regions in the corpus for precisely that reason, the failure is not a fabricated number you can trace to a filing, but a plausible-sounding wrong citation that reads as confidently as a correct one. No tool earns you the gate; only the check does.

Prerequisites & connections

Builds on. No new technical machinery, this module needs no accounting, no valuation math, no statistics. What it does require is everything the program has already built into you under the banner of the program's practical ethics thread, which this module now gives a formal name to. M1.02 taught you to read a serial pattern of "exceptional" cost reclassification as a governance red flag, explicitly flagging it as belonging to "the ethics thread", that instinct (a management that manages optics manages everything) is exactly what Standard I(C) formalizes. M2.08 built the three-duties framework (management integrity, gatekeeper independence, regulatory response) and, more importantly, the language protocol, state facts with the finder attributed, mark allegations as allegations, reserve "fraud" for adjudications, and describe your own analysis as unreconciled evidence rather than a verdict you have not earned. You will use that exact discipline every time you write up a real-world governance concern professionally; this module's Standards are the formal reason the discipline is mandatory, not optional. M6.04 gave you Munger's incentive-caused bias, "show me the incentive and I'll show you the outcome", which is the diagnostic lens behind half the conflicts vignettes below: whose incentives shaped this recommendation, and which way do they push? E11.04 applied the same "does value leak to insiders" question to a company's governance from the outside; this module applies it to your own conduct from the inside: two lenses on one discipline. M10.04 closed the program with the ethics of managing other people's money (fiduciary duty, suitability, honest reporting) as a capstone; this module is where those obligations get their formal Standard numbers and their vignette-defensible form. M0.02 built the SEBI/SEC regulator map you will need every time a vignette turns on which jurisdiction's law is stricter (Standard I(A)), this module does not re-teach who SEBI or the SEC are; it teaches the rule that tells you which one binds when a global firm policy conflicts with local law.

Feeds forward. ES1.02 owns the Global Investment Performance Standards in full, every provision, the required disclosures, verification, and the SEC/SEBI/FCA regulatory and fiduciary map. This module carries only Standard III(D)'s conduct rule (misleading performance communication is a professional violation, independent of whether a firm even claims GIPS compliance); the moment a vignette asks what GIPS itself requires a composite to disclose, stop and go to ES1.02. AA1.03 owns the performance-measurement mechanics and GIPS composite-construction basics, the arithmetic of a composite, the anti-cherry-picking provision, how to audit a compliant presentation line by line. Worked Example 6 below sits exactly on that seam and names it explicitly. PW1.01 owns the actual craft of writing a private-client Investment Policy Statement and running the suitability process; PW1.01 itself says explicitly that "the duty itself, and its formal defense in a vignette, is ES1.01's territory", this is that territory. EC1.02 and other Ring modules that touch disclosure, governance, or fiduciary language all point back here rather than re-arguing the underlying Standard.

Core teaching content

1. The grading discipline, and the Code of Ethics

Every vignette below, and every one you will ever sit for, is answered the same three-part way: name the Standard (a Roman numeral for the broad Standard, a letter for the specific sub-standard ("Standard III(C)," never just "suitability"), state the verdict (violated or not violated) a vignette is not required to contain a violation, and a meaningful fraction of the ones below do not), and give the one deciding fact, the specific sentence, number, or timing detail the vignette planted that makes the verdict provable rather than a guess. Miss any one of the three and you have not answered the question; you have described a feeling about it.

Before the seven Standards, there is the Code of Ethics, six commitments every CFA Institute member and candidate makes, and the frame every Standard exists to enforce. Members and candidates must:

1. Act with integrity, competence, diligence, respect, and in an ethical manner with the public, clients, prospective clients, employers, employees, colleagues, and other market participants.
2. Place the integrity of the investment profession and the interests of clients above their own personal interests.
3. Use reasonable care and independent professional judgment when conducting analysis, making recommendations, taking investment actions, and engaging in other professional activities.
4. Practice and encourage others to practice in a professional and ethical manner that reflects credit on themselves and the profession.
5. Promote the integrity and viability of the global capital markets for the ultimate benefit of society.
6. Maintain and improve professional competence, and strive to maintain and improve the competence of others in the profession.

(As stated in the current CFA Institute Code of Ethics and Standards of Practice Handbook, the wording is periodically refined by CFA Institute; check the current text at cfainstitute.org/en/ethics-standards before quoting it verbatim in any real deliverable.) Component 2 is the one nearly every vignette below tests in some form: whenever a scenario shows you a person's own interest, a fee, a gift, a job offer, a fear of an awkward client conversation, pulling against the client's or the market's, the Code has already told you which way the pull should have been resisted.

The four-step method you will run on every vignette from here forward: (1) actor and action, who did what, acting in what capacity; (2) candidate Standards, usually one or two are genuinely live, the rest are surface-level distractors; (3) the deciding fact, the specific number, disclosure, or timing detail that resolves the ambiguity, which the vignette always contains; (4) the verdict, in one sentence. Standard, violated or not, deciding fact. Section 9 below drills this method directly; Sections 2 through 8 first give you the full rulebook to apply it to.

The GIPS seam, stated once for the whole module. Standard III(D) below governs your conduct around performance communication, you may not misrepresent it, whoever you are, whether or not your firm claims GIPS compliance. The mechanics of how a compliant composite is built, weighted, and disclosed belong to AA1.03; the full GIPS standard itself, every provision, verification, the advertising guidelines: belongs to ES1.02. Every time this module touches performance presentation, it will point at those two rather than reproduce their content.

2. Standard I: Professionalism

Standard I has five parts, and it is the Standard that governs who you are as a professional, independent of any specific client relationship.

I(A) Knowledge of the Law. You must understand and comply with all applicable laws, rules, and regulations (including CFA Institute's own Code and Standards) and, where they conflict, follow the more strict. This is the rule that decides every "our global policy says X, but local law says Y" vignette: it does not matter that a firm's manual is more permissive; a member must follow whichever of firm policy, local law, or the Code imposes the higher standard. The Standard also requires you to disassociate from any ongoing violation you become aware of (stepping away from the activity, or escalating it) even if you did not cause it and even if simply staying silent would be the path of least resistance. M0.02's SEBI/SEC map is exactly what you consult to know which local regulator's rule is in play; this Standard is what tells you it wins the conflict.

I(B) Independence and Objectivity. You must use reasonable care and judgment to achieve and maintain independence and objectivity, and must not offer, solicit, or accept any gift, benefit, compensation, or consideration that reasonably could be expected to compromise your own or another's independence and objectivity. In practice this covers issuer-paid travel and hospitality, sell-side and buy-side pressure, credit-rating-agency pressure, and any quid-pro-quo tied to a rating or a recommendation. Firms typically operationalize this with a de minimis threshold, a dollar value below which a gift needs no special disclosure, and above which it needs written pre-clearance or disclosure to the employer. A gift below the threshold, properly logged, is not a violation; a gift above it, taken without disclosure, right before a rating decision, is the paradigm case.

I(C) Misrepresentation. You must not knowingly make any misrepresentation relating to investment analysis, recommendations, actions, or other professional activities, this reaches guaranteed-return language (no investment product other than a genuinely guaranteed instrument, such as certain insured deposits, may be described as "guaranteed"), performance records spliced from unrelated prior track records, plagiarism (using another's work or ideas without attribution), and overstated credentials. It does not prohibit showing hypothetical or back-tested performance, only unlabeled or misleadingly labeled versions of it. A clearly disclosed "hypothetical, back-tested, not actual client performance" chart is permitted; the same numbers presented as a live track record are not.

I(D) Misconduct. You must not engage in any professional conduct involving dishonesty, fraud, or deceit, or commit any act that reflects adversely on your professional reputation, integrity, or competence. This reaches personal conduct too, a personal-tax-return fraud conviction reflects on integrity and trustworthiness even though it has nothing to do with your investment work, but it is not a catch-all for every embarrassing personal event. A personal bankruptcy triggered by a medical emergency, absent any dishonesty, is misfortune, not misconduct; the Standard requires an element of deceit or a reflection on integrity, not merely a bad year.

I(E) Competence. You must act with and maintain the competence necessary to fulfil your professional responsibilities. This is the fifth sub-standard under Professionalism, added in the revision of the Code and Standards effective 1 January 2024. Verify the current text and effective date in the CFA Institute Standards of Practice Handbook at cfainstitute.org before citing it in a real deliverable, because the sub-standard list moves between editions.

I(E) asks a question that sits one step earlier than every other rule in this Standard. I(A) tells you which rule binds when two conflict, I(B) tells you to stay independent of the people who would like a particular answer, I(C) tells you not to misdescribe your work and I(D) tells you not to be dishonest. I(E) asks whether you were equipped to do the work at all. It is a duty of capability rather than of honesty, and it is the only sub-standard that can be breached by someone who was scrupulously truthful throughout.

Operationally the duty has four moving parts. Assess your own competence against the specific assignment before you accept it, not after the work is delivered. Where the gap is real, take one of the three honest routes: decline the assignment, get supervision or a review from someone who does have the skill, or acquire the skill before you start. Keep competence current, because products, market structure and regulation move underneath a skill that was adequate three years ago. And, if you supervise, do not hand work to someone you already know is not equipped to do it, which is where I(E) and IV(C) touch.

The line the Standard draws is between a gap that is named and a gap that is hidden by silence. An analyst who says "I have never priced a callable structure and I want the desk's rates specialist to review my tree before this goes out" has complied, even if the first draft is wrong, because the assignment was routed to someone who could catch the error. The same analyst who signs an option-adjusted spread onto a published note without ever having built a calibrated tree, and without telling anyone that, has breached I(E) the moment the note is published, and nothing about that verdict depends on whether the number happened to land close.

Two neighbours are easy to reach for by mistake. Standard V(A), Diligence and Reasonable Basis, is about the work product: was the analysis actually supported by adequate research? I(E) is about the actor: was this person equipped to produce it? The routing question that separates them is this. Suppose a fully competent analyst had done exactly this much work and no more; would the recommendation still lack a reasonable basis? If the answer is yes, the vignette is testing V(A), because the deficiency is in the work. If the work would have been sound in competent hands and the deficiency is that the person took on something they could not do, the vignette is testing I(E). Both can fire together, and on the callable-structure fact pattern above they do.

Standard I(D), Misconduct, is the other near neighbour, and the routing question there is shorter: was anyone deceived? I(D) requires dishonesty, fraud or deceit, or an act that reflects adversely on your integrity. I(E) requires none of that. An analyst who is genuinely, transparently out of their depth and says so has no I(D) problem at all and may still have an I(E) problem if they proceed anyway. Add a false claim of expertise on top and you now have I(C) Misrepresentation on the claim itself, I(E) on the conduct, and I(D) as well once the deception is the point rather than a side effect.

3. Standard II: Integrity of Capital Markets

Standard II has two parts, and unlike Standard I it is not about your own conduct alone, it is about protecting the fairness of the market for everyone in it.

II(A) Material Nonpublic Information. If you possess information that is both material (a reasonable investor would want to know it before making a decision, or it would likely affect a security's price) and nonpublic (not yet disseminated to the marketplace generally), you must not act or cause others to act on it. Two things trip up almost everyone learning this Standard for the first time. First, the trigger is acting, not merely knowing, an analyst who receives an unsolicited confidential tip and simply does nothing (no trade, no tip to anyone else, waits for public disclosure) has not violated II(A), because possession alone is not the breach. Second, mosaic theory is the load-bearing exception that makes equity research possible at all: an analyst who combines many individually public, individually immaterial data points, store-visit counts, import/export statistics, supplier disclosures, into a striking, non-obvious conclusion has not created MNPI, however privileged the conclusion feels, because every input was public. The test is never "does this conclusion feel like an edge"; it is "was any single input to it actually material and nonpublic."

II(B) Market Manipulation. You must not engage in practices that distort prices or artificially inflate trading volume with the intent to mislead market participants. This takes two forms: information-based manipulation (spreading false or misleading information (rumors, fabricated news) to move a price) and transaction-based manipulation (trades or orders designed to deceive, such as wash trades between accounts you control, or trading specifically to create a false appearance of activity). The intent to deceive is what makes it manipulation; a genuinely well-researched, aggressively bullish public report is protected opinion, however wrong it later turns out to be.

4. Standard III: Duties to Clients

Standard III has five parts and governs the relationship you owe the people whose money you manage or advise on, the richest, most heavily tested Standard in the framework.

III(A) Loyalty, Prudence, and Care. You must act for the benefit of your clients, place their interests before your employer's or your own, and exercise prudence, judgment, and care. Two subtleties matter: first, you must correctly identify who the client actually is, for a pooled vehicle, the "client" is the fund's investors as a whole, not any one of them individually, which matters when allocating scarce opportunities across accounts. Second, soft-dollar arrangements (paying for research or services with client brokerage commissions) must benefit the client whose commissions were spent, directing a client's commission dollars toward a benefit that flows to you personally, and not to that client, is a Loyalty breach regardless of whether any single trade was executed badly.

III(B) Fair Dealing. You must deal fairly and objectively with all clients when providing investment analysis, recommendations, or taking investment action. "Fairly" does not mean "identically" (different service tiers and fee levels are permitted) but it does mean no client or client group may be systematically favored in the timing or manner of disseminating a recommendation, or in the allocation of a scarce opportunity such as an oversubscribed offering, at the direct expense of another.

III(C) Suitability. When you have an advisory relationship, you must determine that an investment is suitable given the client's documented investment objectives, risk tolerance, time horizon, and any other constraints in the Investment Policy Statement (PW1.01 owns the craft of writing that document); when managing to a specific mandate, you must ensure the portfolio is consistent with that mandate. The test is never "would a reasonable investor find this attractive", it is "does this fit the specific client's own documented profile." A product that is merely aggressive is not automatically unsuitable; a product that directly contradicts a documented constraint, a leveraged instrument for a capital-preservation mandate, an illiquid lock-in for a client with a near-term liquidity need: is.

III(D) Performance Presentation. You must make reasonable efforts to ensure that performance information you communicate is fair, accurate, and complete. This is the Standard that makes excluding a badly performing discretionary account from a composite (precisely because it performed badly) a professional violation, independent of whether the firm claims GIPS compliance at all: AA1.03 already teaches you why GIPS mandates including every actual discretionary account (its anti-cherry-picking provision) and how to audit a composite's arithmetic; Worked Example 6 below is that identical failure seen through the conduct lens, and the full GIPS standard that formalizes the fix is ES1.02's.

III(E) Preservation of Confidentiality. You must keep confidential information about current, former, and prospective clients confidential unless disclosure is required by law, the client consents, or the information concerns illegal activity by the client. This duty survives the end of the relationship, a former client's holdings and reasons for leaving remain confidential exactly as they were the day before the relationship ended, and using them to impress a new prospect is a violation with no statute of limitations.

5. Standard IV: Duties to Employers

Standard IV has three parts and governs what you owe the firm that employs you, a different, and frequently confused, loyalty from Standard III(A)'s duty to clients.

IV(A) Loyalty. You must not harm your employer, deprive it of your skills, misuse its resources, or misappropriate its confidential information or clients, but you are permitted to make ordinary preparations to compete, including researching a new venture and signing an offer letter elsewhere. The line the Standard actually draws is between preparing to leave (permitted) and actively soliciting your current employer's clients, or misappropriating its proprietary work product, while still drawing its paycheck (prohibited), the same fact pattern, timed a few weeks differently, sits on opposite sides of the line. A narrow whistleblower exception exists where reporting a genuine legal or ethical violation is itself required conduct.

IV(B) Additional Compensation Arrangements. You must not accept gifts, benefits, compensation, or consideration from a third party that competes with, or could reasonably be expected to create a conflict with, your employer's interest, unless you obtain written consent from all parties involved, including your employer. This is distinct from Standard I(B): I(B) is about the independence of your research and recommendations; IV(B) is about a separate income stream your employer does not know about and has not consented to, whatever it is used for.

IV(C) Responsibilities of Supervisors. If you have supervisory authority, you must make reasonable efforts to detect and prevent violations by those you supervise, establishing and enforcing a reasonable compliance system, and acting on red flags once you see them. A supervisor can violate this Standard through inaction alone: ignoring repeated internal escalations about a subordinate's suspicious trading is itself the breach, with no personal trade required on the supervisor's part.

7. Standard VI: Conflicts of Interest

Standard VI has three parts, and (together with I(B) and IV(B) above) completes the set of Standards that most often get confused with one another because they all involve some version of "a conflict existed." Section 10 below untangles all four explicitly; here is what each one actually is.

VI(B) Priority of Transactions. Client and employer transactions must take priority over your own personal transactions, and you must avoid actions that could result in personal benefit at a client's expense. The violation is the opportunity to disadvantage a client, not proof of actual harm, a personal trade placed just ahead of a client order in the same direction is a violation the instant it happens, whether or not the client's fill price was measurably worse.

VI(C) Referral Fees. You must disclose to your employer, clients, and prospective clients any compensation, consideration, or benefit received from, or paid to, others for the recommendation of products or services, cash or in kind, and regardless of size. Where no compensation changes hands at all, a professional courtesy referral to an unaffiliated specialist, disclosed as such, there is no Referral Fees issue to disclose, because the Standard is triggered by compensation, not by the mere act of referring.

{
  "skill": "Researching whether a specific local regulatory requirement is stricter than a firm's global policy (Standard I(A))",
  "use": "Have a chat or research assistant identify the specific rule, regulator, and citation for a jurisdiction's requirement on a named topic (e.g., mandatory risk disclosures for retail research distribution), so you know exactly which document to open and read yourself before concluding local law is or is not stricter.",
  "tools": ["Chat assistants - Claude, ChatGPT, Gemini", "Research/browsing tools - a research assistant that can surface the regulator's own page", "SEBI/SEC/FINRA official sites as the primary sources to open next"],
  "prompt": "I need to know whether <JURISDICTION>'s securities regulator imposes a specific requirement on <TOPIC> that a global firm policy might not cover. Do not state the rule as fact - instead, name the specific regulation, the regulator, and the exact page or circular where I can read the current requirement myself, and flag if this is an area that changes frequently.",
  "verify": "Open the regulator's own current text yourself (sebi.gov.in, sec.gov, finra.org) rather than accepting the assistant's paraphrase - a paraphrased regulation is exactly the kind of plausible-sounding, hard-to-spot-check output AI0.01 warns about. Match the effective date against your scenario's date; recompute nothing numeric is usually needed here, but confirm any cited threshold or percentage against the primary text. Cite the specific regulation and section, not the assistant's summary of it (Guardrail 1-5).",
  "diy": "You must already know the I(A) rule itself - follow the stricter of firm policy, local law, and the Code - before any research matters; the assistant finds the document, it does not decide which one binds. That decision is yours and is exactly what this Standard tests.",
  "market": "IN"
}

8. Standard VII: Responsibilities as a CFA Institute Member or Candidate

Standard VII has two parts and is the narrowest Standard, applying specifically to your relationship with CFA Institute itself, its exam program, and its credential.

VII(A) Conduct as Participants in CFA Institute Programs. You must not engage in conduct that compromises the reputation or integrity of CFA Institute or the CFA designation, cheating on an exam, disclosing confidential exam content, or violating any testing policy all fall here, whether or not the underlying conduct has anything to do with actual investment work.

VII(B) Reference to CFA Institute, the CFA Designation, and the CFA Program. When referencing CFA Institute, the charter, or candidacy in the CFA Program, you must not misrepresent or exaggerate the meaning of the designation or your progress toward it. A Level II candidate who has not sat for Level III may accurately describe having passed Levels I and II, but may not attach "CFA" to their name, even hedged with "Levels I & II passed, Level III pending" on the same line, because that construction implies a credential not yet earned. The Standard also prohibits implying that CFA Institute endorses any product, service, or firm, the designation certifies a body of knowledge and an ethical commitment; it certifies nothing about any specific investment process's merit, and CFA Institute endorses no one's product.

9. The four-step method, applied

Section 1 gave you the method in outline; here is how it actually runs on a scenario, before the Worked Examples put it to full use.

Step 1: actor and action. Read past the color (the city, the dinner, the mood) to the bare mechanics: who did what, acting in what capacity, as an employee? a personal investor? a CFA candidate? The capacity matters, because the same act (accepting money from a third party) is a completely different Standard depending on whether the actor is acting as an adviser to a client (VI-territory) or as an employee toward their firm (IV-territory).

Step 2: candidate Standards. List every Standard the fact pattern could plausibly implicate, usually two or three. Most vignettes plant one genuinely correct answer and one or two surface-level lookalikes (Section 10 catalogs the recurring lookalike pairs). Do not commit yet.

Step 3, the deciding fact. Every well-built vignette contains exactly one fact that resolves the ambiguity: a threshold crossed or not crossed, a disclosure made or omitted, a timing detail, a documented client constraint. Find it before you write a verdict, if you cannot point to it, you have not actually solved the vignette, you have guessed.

Step 4, the verdict, in one sentence. State the Standard (Roman numeral, letter), the call (violated / not violated), and the deciding fact, together, in one sentence. If your answer needs a paragraph of hedging to make sense, you have not found Step 3's fact yet: go back.

10. Where vignettes get confused, and how to break the tie

Four pairs (or larger groups) of Standards share enough surface resemblance that they are the most common source of a wrong citation, even from someone who knows every Standard's definition cold. Learn these pairings explicitly; the Worked Examples and Practice Set below are built to drill exactly these discriminations.

• "A conflict existed", which of four Standards? I(B) Independence and Objectivity governs whether something (a gift, a relationship) compromised the objectivity of your research or recommendations. IV(B) Additional Compensation Arrangements governs an undisclosed income stream from a third party that your employer has not consented to. VI(A) Avoid or Disclose Conflicts is the broadest catch-all, any interest that could reasonably color your duties to employer or client, disclosed or not. VI(C) Referral Fees is narrow and specific, compensation tied specifically to referring a client to a product or service. Ask: is money changing hands for a referral specifically (VI(C))? Is it an undisclosed side income your employer doesn't know about (IV(B))? Is it something that could bias your research itself (I(B))? If none of the narrower three fit cleanly, it is VI(A)'s broad disclosure duty.
• "Loyalty", to whom? III(A) is owed to clients; IV(A) is owed to your employer. A vignette about a portfolio manager favoring one client account over another for personal gain is III(A); a vignette about an employee soliciting the firm's clients before resigning is IV(A). The word "loyalty" alone never tells you which, the object of the sentence does.
• A bad recommendation. Suitability or Loyalty? If the recommendation fails a documented client-specific constraint (risk score, liquidity need, tax status), it is III(C) Suitability. If it instead reflects the adviser or manager's own economic interest overriding the client's (a higher-commission product recommended over an equivalent lower-cost one with no suitability difference), the primary Standard is III(A) Loyalty even though the fact pattern can look superficially like a suitability question.
• A striking research conclusion. Diligence or MNPI? If every input was public and the conclusion is the analyst's own synthesis, however sharp, it is at most a V(A) diligence question (was the synthesis actually well-supported?), never II(A). II(A) requires that at least one input actually be material and nonpublic; a brilliant public-data conclusion is never itself an MNPI violation, no matter how much it resembles inside information in hindsight.

{
  "skill": "Drilling additional practice vignettes fast, once the seven Standards are memorized",
  "use": "Have an assistant generate NEW short vignettes targeting a specific Standard or sub-standard you want more reps on (e.g., 'five fresh III(C) suitability vignettes, three violations and two not'), then grade your own written answer against the assistant's suggested answer as a second opinion, not a verdict.",
  "tools": ["Chat assistants - Claude, ChatGPT, Gemini", "The Standards of Practice Handbook itself, open in a second window, as the actual authority"],
  "prompt": "Write me three short (80-120 word) synthetic ethics vignettes testing Standard <X>, with entirely invented people and firms labeled as fictional. Make two of them violations and one NOT a violation, and do not tell me which is which. After I answer, tell me which Standard and sub-letter you intended and the deciding fact - then I will check your intended answer myself before accepting it.",
  "verify": "Treat the assistant's own 'intended answer' as a draft, not a ruling - an assistant will sometimes write a vignette whose planted facts do not actually support the verdict it thinks it wrote, exactly the kind of confident-but-wrong output this module's own opening example warns about. Check any citation the assistant gives against the actual Standard definition in Sections 2-8 above (or the Handbook itself); if the assistant's deciding fact does not match what you were taught, the vignette - not you - is wrong, and you should discard it rather than second-guess a correct answer. Never let an assistant-generated vignette substitute for CL-ES.1, whose key is fixed and pre-verified.",
  "diy": "You must already be able to run the four-step method (Section 9) on a vignette someone else wrote before you can safely judge whether an AI-authored vignette's own claimed answer is even correct. This is drill for speed and volume, not a substitute for the reps in Worked Examples and the Practice Set below.",
  "market": "US"
}

Modern Data Analysis, screening for conflicts and priority-of-transaction breaches at scale. Everything in Sections 7 and 9 is written as a single vignette a human reads and reasons through, but a real compliance desk runs the equivalent check across thousands of trades and disclosures every day, and it does that in code, not by reading. A personal-account dealing (PAD) screen is a join, in pandas terms: match each employee's personal trade blotter against the firm's client and proprietary trade blotters on the same ticker within a window (say, ±2 trading days), and flag any personal trade in the same direction, ahead of, a subsequently placed larger client order, precisely Worked Example 9's VI(B) fact pattern, run automatically over an entire firm's trading population instead of one vignette at a time. A parallel gift-and-entertainment screen matches every logged gift or hospitality entry against the firm's de minimis threshold and its pre-clearance record, flagging anything over-threshold without a matching approval. Section 2's I(B) rule, mechanized. Two traps bite in production that a vignette never shows you. First, timestamp comparisons across systems, a personal-trading system logging in local exchange time against a client order-management system logging in UTC will silently misorder two trades that actually happened in the opposite sequence from what the raw timestamps suggest, manufacturing a false front-running flag (or hiding a real one); always normalize both sides to the same timezone before comparing, and treat any comparison across two different source systems as unverified until you have. Second, fuzzy name-matching across a personal-holdings disclosure form and a client-account registry (to catch an undisclosed VI(A) conflict, such as a covered company's shares held personally under a slightly different legal-entity name) produces false negatives silently, a screen that only does exact string matching will simply miss the very case it exists to catch. The by-hand skill this module gates, correctly naming the Standard on one vignette, is exactly what tells you whether the screen's output is even asking the right question; the quant-code region's data-analysis branch is where you would actually build and vectorize a screen like this at scale.

Common mistakes & how experts think differently

1. Answering with a feeling instead of a citation. "That seems unethical" scores zero on any real assessment of this material. The expert's reflex is immediate and mechanical: name the Standard, call the verdict, state the deciding fact, in that order, every time, before any discussion of how the scenario "feels."
2. Assuming every vignette contains a violation. Vignettes are deliberately calibrated to include clean scenarios, a properly disclosed, under-threshold gift; a mosaic-theory conclusion built entirely from public data; a referral made with no compensation at all. An expert's first move is never "which Standard did this violate" but "did this violate anything at all", and roughly a quarter of the scenarios in this module answer "no."
3. Confusing the four "conflict" Standards. I(B), IV(B), VI(A), and VI(C) all involve some version of "a conflict of interest existed," and a rushed answer reaches for whichever one it remembers first. The expert asks the three routing questions in Section 10 before committing: research-objectivity, undisclosed-third-party-income, or specifically-a-referral-fee, and defaults to VI(A)'s broad disclosure duty only once the narrower three are all ruled out.
4. Confusing III(A) Loyalty to clients with IV(A) Loyalty to the employer. The word is identical; the object of the duty is not. The expert's habit is to ask, out loud, "loyal to whom, in this sentence" before citing either.
5. Treating "aggressive" as synonymous with "unsuitable." Standard III(C) is tested against the client's documented profile, not against a generic notion of prudence. An expert always asks "what does this specific client's IPS actually say" before calling a recommendation unsuitable, and will call a genuinely aggressive recommendation suitable if the client's own documented risk tolerance and horizon support it.
6. Believing mere possession of MNPI is itself the violation. Standard II(A) is triggered by acting (trading, or causing someone else to trade), an analyst who receives an unsolicited tip and does nothing with it has not violated the Standard. The expert's habit on any MNPI vignette is to check specifically for a trade, an order, or a tip to someone else, not merely for the presence of a juicy fact.
7. Over-crediting a specific, well-researched public-data conclusion as inside information. Mosaic theory exists precisely because sharp analysis of public data is supposed to feel like an edge, that is the product of good research, not evidence of a violation. The expert checks every input to a striking conclusion, not the conclusion's impressiveness.
8. Assuming Priority of Transactions requires proof of client harm. Standard VI(B) is violated the moment a personal trade takes an opportunity ahead of a client's, whether or not the client's own execution was measurably worse. The expert never asks "was the client actually hurt", only "did the personal trade have priority over the client's."
9. Treating record retention as a firm-only, IT-department concern. Standard V(C) is a personal professional obligation on the member or candidate, not merely a corporate policy line item, and the seven-year CFA Institute floor (or a longer local-law requirement) applies regardless of what any single firm's retention software happens to do by default.
10. Reading GIPS non-compliance as a defense to a performance-presentation charge. "We never claimed GIPS compliance" is not a defense to Standard III(D), the conduct duty to present performance fairly and completely binds every member and candidate regardless of whether their firm has adopted GIPS at all. The expert keeps these two facts, the conduct rule (III(D), mine) and the voluntary technical standard (GIPS, ES1.02's): permanently distinct, exactly as AA1.03 and ES1.02 keep them distinct from their own side of the seam.
11. Trusting an AI assistant's confident Standard citation without checking the deciding fact yourself. A model will fluently produce "Standard I(B), violated" for any scenario that merely mentions a gift, whether or not the gift was under the firm's threshold and properly disclosed, the same surface pattern-matching a rushed human candidate falls into, delivered with identical confidence whichever way it lands. The expert treats an assistant's vignette answer exactly as R10 requires: a hypothesis to check against the actual Standard's text and the vignette's own planted fact, never the verdict itself.

Worked examples

Worked example 1: Standard I(A) and I(D): the law that binds, and the misconduct that isn't

Vignette (a). Highland Wealth Advisors' global compliance manual permits describing a hybrid annuity-linked product's charges to retail clients only as carrying "a competitive fee structure," with no specific number required. India's applicable retail-distribution rules require disclosing the exact embedded commission in percentage terms to any retail buyer. Highland's Mumbai distribution desk is instructed to use the global template (omitting the specific percentage) because "it's firm policy." The omitted figure, if disclosed, would have shown a 2.75% embedded trail commission.

Solution. Step 1. The actor is Highland's Mumbai distribution team, acting under an explicit instruction from global compliance. Step 2. Candidate Standard: I(A) Knowledge of the Law (a policy-versus-local-law conflict) versus I(C) Misrepresentation (an omission rather than an affirmative false statement). Step 3, the deciding fact. Local law imposes a specific, stricter disclosure requirement than the global manual provides; Standard I(A) requires following the stricter of the two, with "it's firm policy" carrying no exculpatory weight whatsoever. Step 4. Standard I(A), violated, local law's specific-percentage disclosure requirement is stricter than the global template's vague language, and the Standard requires the stricter rule regardless of internal policy.

omitted_pct, typical_upper = 2.75, 1.5
print(round(omitted_pct/typical_upper, 2))   # 1.83

For context only (not the deciding fact itself): the omitted 2.75% commission runs 1.83× the roughly 1.0–1.5% range typical for comparable retail products, a further reason the specific-disclosure requirement exists, though the violation turns on the disclosure rule itself, not on whether the fee happened to be high.

Vignette (b). Two years after joining a different firm, an analyst's personal income-tax return from years earlier is found, in a court proceeding, to have involved deliberate underreporting of freelance income, a matter wholly unrelated to any investment recommendation he has ever made.

Solution. Step 1. The actor is the analyst, acting in a purely personal capacity, on conduct unconnected to any client work. Step 2. Candidate Standard: I(D) Misconduct, the only question is whether personal dishonesty counts. Step 3, the deciding fact. The tax matter involved deliberate, adjudicated deceit (underreporting income), not mere misfortune. Step 4. Standard I(D), violated, I(D) explicitly reaches personal conduct involving dishonesty or fraud that reflects adversely on integrity and trustworthiness, even where it has no connection to professional investment activity. Contrast, not violated: had the same analyst instead filed for personal bankruptcy after an uninsured medical emergency, with no dishonesty anywhere in the record, that would not violate I(D), financial misfortune alone is not misconduct; the Standard requires an element of deceit or fraud, which bankruptcy from bad luck does not carry.

Worked example 2: Standard I(B) and I(C): the gift that crosses the line, and the one that doesn't

Vignette (a). An equity analyst at Cobalt Ridge Capital accepts a $4,500 all-expenses-paid weekend at a covered company's flagship resort property, two weeks before reiterating a "Buy" rating on that company. Cobalt Ridge's own gift policy sets a $200 de minimis threshold requiring written pre-clearance above it; she sought none.

Solution. Step 1. The analyst, acting in her research capacity, accepted an issuer-paid benefit shortly before a rating action. Step 2. Candidate Standard: I(B) Independence and Objectivity. Step 3, the deciding fact. The benefit's value relative to the firm's own threshold, taken with no pre-clearance:

gift, threshold = 4500, 200
print(round(gift/threshold, 1))   # 22.5

$4,500 is 22.5× the firm's own $200 threshold, accepted without the required pre-clearance, immediately ahead of a rating decision. Step 4. Standard I(B), violated, the size of the benefit relative to the firm's own de minimis line, combined with the absence of pre-clearance, is exactly the inducement I(B) exists to prevent; whether her rating was actually swayed is irrelevant, because accepting a benefit capable of compromising independence is itself the breach.

Vignette (b): contrast. A different analyst at the same firm receives a branded $150 desk item at an investor conference from a company she covers, logs it in the firm's gift register the same day, and takes no rating action for the following quarter.

Solution. Step 3, the deciding fact: $150 sits below the $200 threshold ($150/$200 = 0.75, under 1.0), and it was properly logged. Step 4. Standard I(B), not violated, a modest, disclosed gift under the firm's own threshold, with no immediate rating action attached, is exactly the kind of benefit the de minimis system is built to permit without further process.

Vignette (c). Brightleaf Capital's marketing brochure for its "Momentum Growth" strategy states: "Investors are guaranteed an 8% annual return." The strategy's actual realized annual returns over the last five years were −12%, +22%, +5%, +18%, and −3%.

Solution. Step 3, the deciding fact:

rets = [-0.12, 0.22, 0.05, 0.18, -0.03]
g = 1.0
for r in rets: g *= (1+r)
print(round((g**(1/5)-1)*100, 2), sum(1 for r in rets if r < 0))   # 5.23  2

The strategy's actual five-year annualized return is 5.23%, not 8%, and the five-year window includes two negative years, a portfolio that has lost money twice in five years cannot honestly be called "guaranteed" at any positive figure. Step 4. Standard I(C), violated, "guaranteed" is a claim reserved for instruments that genuinely cannot lose value; an equity strategy with two down years in its own disclosed history is definitionally not guaranteed, and stating otherwise is misrepresentation regardless of what the long-run average happens to be. Contrast, not violated: had Brightleaf instead shown the identical 5.23% figure captioned "hypothetical back-tested performance, not an actual client track record, not a guarantee of future results," that disclosure would cure the Standard, I(C) prohibits unlabeled or falsely labeled hypothetical performance, not hypothetical performance itself.

Worked example 3: Standard II(A): the tip and the mosaic

Vignette (a). At a private dinner in Bengaluru, an analyst covering Zenith Consumer Products Ltd (synthetic) overhears the company's supply-chain head confirm, in specific terms, that the company's largest customer has just cancelled a ₹340 crore order, information not yet disclosed to the market. The analyst sells short the next morning.

Solution. Step 1. The analyst, in her research capacity, received a confidential fact and traded on it before public disclosure. Step 2. Candidate Standard: II(A) Material Nonpublic Information. Step 3, the deciding fact:

cancelled, qtr_rev = 340, 2850
print(round(cancelled/qtr_rev*100, 1))   # 11.9

A ₹340 crore cancellation against ₹2,850 crore of quarterly revenue is 11.9% of a single quarter's top line (comfortably material) and it was confirmed by an insider, specifically and in definite terms, before any public disclosure. Step 4. Standard II(A), violated, the information was both material (an 11.9%-of-quarter revenue hit) and nonpublic, and trading on it the next morning is the textbook breach, regardless of the fact that it was overheard socially rather than solicited.

Vignette (b): contrast. A different analyst estimates the same quarter's likely revenue decline for an unrelated company using two entirely public proxies: a public foot-traffic index down 9% and public component-import data down 14%.

Solution. Step 3, the deciding fact:

p1, p2 = 0.09, 0.14
print(round((p1+p2)/2*100, 1))   # 11.5

Averaging the two public data points suggests an 11.5% estimated decline, a conclusion that feels privileged but is built entirely from data anyone could pull. Step 4. Standard II(A), not violated, this is mosaic theory in its clean form: every input was public, and combining public, individually immaterial data points into a sharp non-obvious conclusion is permitted, and is what good research is supposed to look like.

Worked example 4: Standard II(B): the rumor that moved the price

Posting under an anonymous account, a trader buys 5,000 shares of a thinly traded small-cap at $2.00, then posts a fabricated claim of an imminent buyout offer on a public retail-investing forum, and sells his position after the price responds.

Solution. Step 1. The trader bought first, fabricated information second, and sold into the resulting price move. Step 2. Candidate Standard: II(B) Market Manipulation (information-based). Step 3 (the deciding fact) the price moved on the false post, and he profited from having positioned ahead of his own rumor:

shares, buy_px, sell_px = 5000, 2.00, 2.80
print(shares*(sell_px-buy_px))   # 4000.0

The sequence, buy, fabricate, sell into the resulting rise from $2.00 to $2.80: nets $4,000. Step 4. Standard II(B), violated, deliberately spreading false information with the intent to move a price and profit from the move is information-based market manipulation; the specific $4,000 figure is the proof of intent and effect, not merely a coincidental gain.

Worked example 5: Standard III(B) and III(C): unequal access, and a mismatch that isn't aggressive, it's wrong

Vignette (a). Fenwick & Vale Advisors emails a material research-report upgrade to its largest institutional clients 30 minutes before its scheduled release to all other clients. In that window, the stock moves from $60.00 to $61.80, and one institutional client buys 10,000 shares at an average price of $60.20 before the report becomes generally available.

Solution. Step 1. The firm itself staggered dissemination by client tier. Step 2. Candidate Standard: III(B) Fair Dealing. Step 3, the deciding fact:

shares, early_avg, later_px = 10000, 60.20, 61.80
print(shares*(later_px-early_avg), round((later_px/60.00-1)*100, 2))   # 16000.0  3.0

A 30-minute early-access window, during which the stock moved 3.00% and a favored client captured a $16,000 benefit unavailable to anyone else, is a direct, quantifiable transfer of value created purely by unequal timing of dissemination. Step 4. Standard III(B), violated. Fair Dealing does not require identical treatment of every client, but it does forbid systematically privileging one tier's access to a specific recommendation change at the direct expense of everyone else's opportunity to act on the same information at the same time.

Vignette (b). Vayu Capital Advisors recommends a complex, illiquid structured note with embedded leverage (a ₹45,00,000 position) to a retired client whose written Investment Policy Statement caps any single illiquid/alternative holding at 15% of a ₹75,00,000 total portfolio and states a primary objective of capital preservation with monthly income.

Solution. Step 3, the deciding fact:

note, net_worth, cap = 45_00_000, 75_00_000, 0.15
share = note/net_worth
print(round(share*100, 1), round(share/cap, 2))   # 60.0  4.0

₹45,00,000 against a ₹75,00,000 portfolio is 60.0% of total net worth in a single illiquid, leveraged instrument: 4.0× the client's own documented 15% cap, for a client whose IPS explicitly calls for capital preservation, not leveraged exposure. Step 4. Standard III(C), violated, this is not merely an aggressive recommendation; it directly contradicts two separate documented constraints (the 15% illiquid-holdings cap and the capital-preservation objective) at four times the client's own stated limit, which is precisely the affirmative-mismatch standard III(C) is built to catch.

Worked example 6: Standard III(D): the excluded account, and the seam to AA1.03 and ES1.02

Alderbrook Wealth Management's "Growth Equity Composite" fact sheet reports a trailing-year return of 11.424%, calculated across six discretionary accounts totaling $87 million. A seventh discretionary account, $6 million, was terminated by its client after a −16.5% year and excluded from the composite; every actual discretionary account under management must be included.

Solution. Step 1. The firm itself constructed and published the composite figure. Step 2. Candidate Standard: III(D) Performance Presentation. Step 3, the deciding fact:

kept = [(18e6,.132), (12e6,.118), (25e6,.101), (8e6,.145), (14e6,.108), (10e6,.095)]
excluded = (6e6, -0.165)
kept_assets = sum(a for a,_ in kept)
all_assets = kept_assets + excluded[0]
reported = sum(a*r for a,r in kept)/kept_assets
true = sum(a*r for a,r in kept+[excluded])/all_assets
print(round(reported*100,3), round(true*100,3), round((reported-true)*10000,1))
# 11.424  9.623  180.2

Including the terminated account (as every actual discretionary account must be) drops the true asset-weighted return from the reported 11.424% to 9.623%, an inflation of 180.2 basis points created solely by excluding the one account that had a bad year. Step 4. Standard III(D), violated, this is textbook survivorship-bias performance manipulation, provable to the exact basis point, and the violation exists independent of any GIPS claim: AA1.03's §10 already teaches you why GIPS mandates including every fee-paying discretionary account (the anti-cherry-picking provision) and how to audit a composite's arithmetic; this is that identical failure, viewed as a personal-conduct breach that binds Alderbrook's staff whether or not the firm claims GIPS compliance at all. Any question about what GIPS itself requires a compliant presentation to disclose belongs to ES1.02, not here.

Worked example 7: Standard IV(A) and IV(B): preparing to leave versus soliciting while still paid, and the side income nobody approved

Vignette (a). Two months before her planned resignation from Marlstone Asset Management, a relationship manager signs an offer letter with a competitor, then begins privately contacting Marlstone's largest clients to gauge their willingness to move their accounts with her, while still drawing a Marlstone paycheck.

Solution. Step 3, the deciding fact. Signing an offer letter elsewhere is an ordinary, permitted preparation to compete; actively soliciting the current employer's clients while still employed is what crosses the line, the same person, the same week, engaged in both a permitted act and a prohibited one. Step 4. Standard IV(A), violated, soliciting Marlstone's clients before resigning misappropriates the firm's client relationships for personal benefit while she remains its employee; the offer letter itself is not the problem.

Vignette (b). While employed full-time at Redcliff Partners covering fintech names, an analyst separately agrees to advise an early-stage fintech startup for $3,000 per month plus a 2% equity stake, without informing Redcliff.

Solution. Step 3, the deciding fact:

monthly = 3000
print(monthly*12)   # 36000

$36,000 a year in cash, plus an equity stake, from a company operating in the exact sector he covers professionally, arranged with no consent from Redcliff at all. Step 4. Standard IV(B), violated, this is a third-party compensation arrangement that plainly could conflict with his employer's interests (he now has a personal financial stake in a company competing for the attention of the same sector he is paid to analyze objectively), and IV(B) requires written consent from all parties, including the employer, before accepting it, consent his employer was never even asked for.

Worked example 8: Standard V(A) and V(C): borrowed diligence, and a file destroyed too soon

Vignette (a). An analyst issues a "Buy" recommendation built entirely on a third-party research report's discounted-cash-flow valuation, which assumed a discount rate of 7% against comparable-company data supporting 9.5–10.5%, without independently checking that assumption. The underlying cash flow is $50 million, growing at 3% in perpetuity.

Solution. Step 3, the deciding fact:

fcf, g, r_correct, r_wrong = 50.0, 0.03, 0.10, 0.07
v_correct, v_wrong = fcf/(r_correct-g), fcf/(r_wrong-g)
print(round(v_correct,2), round(v_wrong,2), round((v_wrong/v_correct-1)*100,1))
# 714.29  1250.0  75.0

At the market-consistent 10% discount rate the business is worth $714.29 million; at the third party's unchecked 7% rate it is worth $1,250.00 million, a 75.0% overstatement created by a single unchecked assumption, three points below the range the analyst's own comparable-company data would have supported. Step 4. Standard V(A), violated, relying on someone else's work is permitted only after reasonable due diligence into its quality; adopting a discount rate three points below the analyst's own comparable set, without checking it, is not diligence borrowed, it is diligence skipped.

Vignette (b). A research file's most recent supporting analysis is dated 14 November 2019. On 1 June 2026, the analyst deletes the file as "old."

Solution. Step 3, the deciding fact:

import datetime
created = datetime.date(2019,11,14)
required_through = datetime.date(created.year+7, created.month, created.day)
deletion = datetime.date(2026,6,1)
print(required_through, (required_through-deletion).days)   # 2026-11-14  166

CFA Institute's recommended seven-year minimum, running from a 14 November 2019 creation date, does not expire until 14 November 2026, the 1 June 2026 deletion is 166 days early. Step 4. Standard V(C), violated, the file was destroyed before the retention floor was reached, regardless of any informal internal sense that a six-and-a-half-year-old file is "old."

Worked example 9: Standard VI(B) and VI(C): the trade that jumped the queue, and the fee nobody was told about

Vignette (a). A portfolio manager at Devgiri Capital (synthetic) buys 4,000 shares of a small-cap name for his personal account at ₹850.00, then, within the hour, enters a large client buy order in the same name that moves the average client execution price to ₹855.50.

Solution. Step 3, the deciding fact:

personal_shares, personal_px, client_avg = 4000, 850.00, 855.50
print(personal_shares*(client_avg-personal_px))   # 22000.0

His personal trade, filled at ₹850.00 an hour before his own client order pushed the average client price to ₹855.50, hands him a ₹22,000 advantage created entirely by the sequencing. Step 4. Standard VI(B), violated, client transactions must be given priority over a manager's own personal trading in the same name; the ₹22,000 figure proves the advantage existed, but the violation would stand even at ₹0, because the breach is the priority given to his own order, not the size of the gain.

Vignette (b). A wealth manager at Pinebrook Securities refers clients to a specific mortgage broker and receives an undisclosed 0.25% fee on every referred client's loan amount. Over the year, referred loans total $18,000,000.

Solution. Step 3, the deciding fact:

loans, rate = 18_000_000, 0.0025
print(loans*rate)   # 45000.0

$45,000 in compensation, tied directly to referrals, disclosed to neither the affected clients nor his employer. Step 4. Standard VI(C), violated, any compensation received for recommending or referring a service must be disclosed, in full, to both the client and the employer; the specific $45,000 figure is not what makes it a violation (a disclosed fee of any size would be permitted), the absence of disclosure is the entire violation.

Practice set

Work every vignette fully (Standard, verdict, deciding fact) before reading its solution. Roughly a quarter of the vignettes below are not violations; do not assume otherwise. All entities are explicitly synthetic.

P1 (guided). Standard, verdict, fact. An analyst at Kaveri Wealth Partners (synthetic) inherits a 4% stake in a small manufacturer, and is assigned coverage of that manufacturer's sector eight months later. She asks compliance in writing to recuse her from the name; the firm declines, because she is its only analyst covering the sector. She then states the holding, in one plain sentence, in the opening paragraph of every note she writes on the company, and repeats it in each note rather than relying on the onboarding file.

Solution. Standard VI(A), not violated. Two facts decide it, and neither is "she disclosed". First, avoidance was genuinely considered and pursued: she asked to be recused and was refused for a reason outside her control, which is what makes disclosure the correct fallback rather than a toll paid to keep a conflict she could have declined. Second, the disclosure passes all three tests: it is prominent (opening paragraph, not an appendix), in plain language, and effectively communicated, because it is repeated in each note rather than left in a document the client last saw at onboarding. Change one fact and the verdict flips. Had she bought the stake after taking coverage, avoidance was available and unused; had she logged it once with compliance and never repeated it, the disclosure would fail the effectiveness test even though it was made.

P2 (guided). Standard, verdict, fact. An analyst at Thornwood Investment Group personally owns 1,500 shares (bought at $30.00 each) of a company he covers, and does not mention this in any research note recommending the stock.

Solution. Standard VI(A), violated. The deciding fact: a $45,000 personal position (1,500 × $30.00 = $45,000) in a covered name, disclosed nowhere, is exactly the kind of interest that could reasonably be expected to color a recommendation, the duty to disclose is triggered by the undisclosed position itself, independent of whether the recommendation was, in fact, distorted by it.

P3 (guided). Standard, verdict, fact. A CFA candidate who has passed Levels I and II accurately states on her résumé: "Passed CFA Program Levels I and II; Level III candidate."

Solution. Standard VII(B), not violated. The deciding fact: this is an accurate description of candidacy status, worded as progress rather than as an implied credential: VII(B) prohibits attaching "CFA" to a name before it is earned, not accurately describing exam progress in plain language.

P4. Standard, verdict, fact. Brightleaf Capital's client brochure states: "Our disciplined process has been formally endorsed by CFA Institute as a superior approach to risk management."

Solution. Standard VII(B), violated. The deciding fact: CFA Institute certifies a body of knowledge and an ethical commitment through the charter, it does not endorse any specific firm's process, product, or investment approach, and claiming otherwise misrepresents what the designation and the institution actually stand behind.

P5. Standard, verdict, fact. A junior analyst at Marlstone Asset Management, after resigning, emails himself his former employer's proprietary industry-screening model (built entirely on Marlstone's internal work over several years) the night before his last day, and begins using it at his new firm.

Solution. Standard IV(A), violated. The deciding fact: the screening model is Marlstone's proprietary work product, taken without consent immediately before departure, misappropriating an employer's confidential materials is squarely prohibited, distinct from (and worse than) the merely-permitted act of preparing to compete.

P6. Standard, verdict, fact. A former client of Fenwick & Vale Advisors leaves after a poor year. At a dinner party, her former adviser mentions to a prospective new client exactly what the former client's portfolio held and why she left, as a way of demonstrating his experience.

Solution. Standard III(E), violated. The deciding fact: the client relationship had ended, but III(E)'s confidentiality duty survives its end, using a former client's specific holdings and reasons for leaving to impress a prospect is a breach with no expiration date attached to the relationship's termination.

P7. Standard, verdict, fact. A supervisor at Cobalt Ridge Capital receives six separate internal compliance escalations over four months flagging a subordinate trader's suspicious pattern of trades just ahead of large client orders, and takes no action on any of them.

Solution. Standard IV(C), violated. The deciding fact: six escalations over four months (a rate of 1.5 per month: 6/4 = 1.5) with zero action taken is a failure of reasonable supervision on its own; IV(C) can be violated by inaction alone, with no personal trade by the supervisor required.

P8. Standard, verdict, fact. A financial adviser at Highland Wealth Advisors sends every client, unchanged, the identical quarterly letter describing "our disciplined value-investing process" for two years after quietly shifting the actual portfolio to a momentum-based, high-turnover strategy.

Solution. Standard V(B), violated. The deciding fact: the firm's actual investment process materially changed, and clients were never told: V(B) specifically requires disclosing the basic characteristics of the investment process and any material change to it, which a static, outdated template letter fails to do the moment the underlying process diverges from what it describes. This vignette tests only the older half of V(B). The second half, the services-and-costs disclosure owed at the outset of the relationship, is drilled separately at P19, and a shift to a high-turnover strategy usually raises the client's total cost as well, so on these facts a real reviewer would check both halves rather than stopping at the first.

P9. Standard, verdict, fact. An analyst learns, from a public regulatory filing posted that morning, that a company has announced an unexpected executive departure, and issues a same-day note revising his rating downward.

Solution. Standard II(A), not violated. The deciding fact: the information came from a public regulatory filing, already disseminated to the market, there is no nonpublic element at all, so II(A) is not even implicated regardless of how quickly the analyst acted on it.

P10. Standard, verdict, fact. A wealth manager at Vayu Capital Advisors recommends a rank-and-file client, age 61 and eighteen months from retirement, increase his 401(k)-equivalent allocation to employer stock from 10% to 40%, with no documented change in the client's risk tolerance or objectives.

Solution. Standard III(C), violated. The deciding fact:

new_alloc, cap = 0.40, 0.15
print(round(new_alloc/cap, 2))   # 2.67

A 40% single-stock concentration is 2.67× a typical documented 15% single-position cap for a client eighteen months from retirement with no stated change in circumstances, a recommendation that increases concentration risk sharply for a client whose own documented profile does not support it.

P11. Standard, verdict, fact. Anmol Distributors Pvt Ltd (synthetic), a mutual fund distributor, discloses its standard trail commission in a footnote of every client statement, but does not separately disclose an additional ₹3,00,000 incentive trip awarded by the fund house for meeting a sales target, against ₹18,00,000 of the distributor's total annual compensation.

Solution. Standard VI(C), violated. The deciding fact:

trip, comp = 3_00_000, 18_00_000
print(round(trip/comp*100, 2))   # 16.67

The incentive trip is worth 16.67% of the distributor's total annual compensation, a material, non-cash benefit tied to a sales target, disclosed nowhere; VI(C) requires disclosing all compensation and consideration connected to recommending a product, cash or in kind, not merely the standard commission line.

P12. Standard, verdict, fact. A different distributor refers a client to an unaffiliated, fee-only tax preparer purely as a professional courtesy, discloses the referral relationship verbally to the client, and receives no compensation of any kind from the preparer.

Solution. Standard VI(C), not violated. The deciding fact: no compensation, consideration, or benefit changed hands at all. VI(C)'s duty is triggered by receiving or paying compensation for a referral, and a referral made for free, disclosed as a courtesy, has nothing to disclose beyond the relationship itself, which was disclosed.

P13. Standard, verdict, fact. A portfolio manager at Redcliff Partners directs $2,000,000 of annual client brokerage commissions to a specific broker who, in exchange, pays $180,000 a year of the manager's personal office rent, a benefit that flows to the manager, not to the clients whose commissions were spent.

Solution. Standard III(A), violated. The deciding fact:

rent, commissions = 180_000, 2_000_000
print(round(rent/commissions*100, 1))   # 9.0

9.0% of the clients' own commission dollars are being converted into a personal benefit for the manager rather than a benefit for the clients whose trading generated them, soft-dollar arrangements must benefit the client whose commissions were spent, and this one benefits only the manager.

P14. Standard, verdict, fact. An analyst independently builds a valuation model, but before publishing asks a colleague in a different sector to sanity-check her key assumptions and flags any that look unsupported; she revises two assumptions based on that review before publishing.

Solution. Standard V(A), not violated. The deciding fact: the underlying work and conclusions remain her own, supported by her own research, and a peer sanity-check that leads to revised (not blindly adopted) assumptions is exactly the due-diligence process V(A) rewards, not a delegation of the analysis itself.

P15. Standard, verdict, fact. A firm's electronic-communications archive fails to capture business-related messages sent through a personal messaging app for a 45-day stretch, during which an estimated 220 client-instruction messages went unarchived.

Solution. Standard V(C), violated. The deciding fact: 220 client-instruction messages over 45 days are exactly the kind of record V(C) requires the firm and its members to retain and be able to reconstruct, a systems gap that prevents capture is not a defense, because the retention duty is to actually have the records, not merely to have intended to.

P16. Standard, verdict, fact. A CFA Level I candidate, mid-exam, memorizes and later posts several specific exam questions on an online study forum.

Solution. Standard VII(A), violated. The deciding fact: disclosing specific confidential exam content compromises the integrity of the CFA Program's testing process directly: VII(A) governs exactly this conduct, independent of whether the candidate ultimately passes or fails the exam itself.

P17. Standard, verdict, fact. A relationship manager at Malhar Securities (synthetic), still six months from any planned departure and with no competing offer in hand, mentions to a colleague over lunch, in general terms, that she has occasionally wondered what it would be like to run her own advisory practice one day.

Solution. Standard IV(A), not violated. The deciding fact: no client was contacted, no firm resource was used, and no concrete competing arrangement exists yet, an idle, general thought shared privately with a colleague is neither solicitation of clients nor misappropriation of anything; IV(A) is not triggered by merely thinking about, or mentioning, a hypothetical future.

P18. Standard, verdict, fact. A sell-side analyst at Pinebrook Securities issues a "Buy" note citing a covered company's own press release claiming "32% year-over-year revenue growth." The company's subsequently filed financial statements show revenue rising from $250mm to $268mm, a 7.2% increase, with the press release figure having folded in a large one-time acquisition's revenue as though it were organic growth, which the analyst never reconciled against the filing before publishing.

Solution. Standard V(A), violated. The deciding fact:

old_rev, new_rev, claimed = 250e6, 268e6, 0.32
actual_growth = new_rev/old_rev - 1
implied_rev = old_rev*(1+claimed)
print(round(actual_growth*100, 1), round(implied_rev - new_rev, 0))
# 7.2  16000000.0

The filed statements show 7.2% actual growth against a claimed 32% (a gap worth $16,000,000 of implied revenue that was never actually reported) and basic diligence (reconciling a press release against the company's own subsequently filed numbers before publishing) would have caught the gap before, not after, the note went out.

P19. Standard, verdict, fact. Sable Creek Advisors (synthetic) hands every new US client a fourteen-page onboarding pack describing its investment process, its research sources, its rebalancing rule and its benchmark, and reissues the pack whenever the process changes. Nowhere in the pack, and nowhere in the first four quarterly statements, does the firm say how it is paid. The advisory fee of 1.10% of assets, deducted quarterly, and the 0.34% of underlying fund expenses appear together for the first time in a footnote of the annual tax package, on an $850,000 account.

Solution. Standard V(B), violated. The deciding fact:

adv, fund, aum = 0.0110, 0.0034, 850_000
print(round((adv+fund)*aum, 2), round(adv/4*aum, 2))   # 12240.0 2337.5

The client paid $12,240 in the first year, including four quarterly advisory deductions of $2,337.50 each, before being told in plain terms what he was being charged. The process half of V(B) is satisfied impeccably here, which is exactly the trap. The services-and-costs half requires disclosure of the nature of the services and information about their cost to the client at the outset, in plain language and prominently, and a footnote in a tax package delivered after the fourth deduction is none of those three. Specific amounts were never required; a stated basis-point range in the onboarding pack would have discharged the duty completely.

P20. Standard, verdict, fact. An equity analyst at Torrance Hill Research (synthetic), who has never built a calibrated binomial interest-rate tree, is asked to publish an option-adjusted spread on a callable agency structure. She says nothing about the gap, seeks no review, runs a spreadsheet a departed colleague left behind whose calibration step she does not understand, and publishes the resulting OAS under her own name.

Solution. Standard I(E), violated, and Standard V(A) violated alongside it. The deciding fact for I(E) is that she took on work she was not equipped to do and told nobody, so none of the three honest routes (decline, get review, acquire the skill) was taken. V(A) fires separately on the routing question from section 2: a fully competent analyst who had done exactly this much work, running an inherited model without checking its calibration, would also have published an unsupported number, so the work product is independently deficient. Cite both, and say which fact carries which: the hidden competence gap is I(E), the unchecked model is V(A).

P21. Standard, verdict, fact. Same analyst, same assignment, different conduct. She writes to her head of research stating that she has never built a calibrated tree, asks the rates desk to review the calibration before anything is published, and the desk's specialist checks the tree and signs off. Her first draft used the wrong volatility input; the reviewer caught it and the published number is correct.

Solution. Standard I(E), not violated. The deciding fact: the gap was named and the work was routed to someone who could catch the error, which is one of the three routes the Standard permits. I(E) requires you to assess and maintain competence honestly, not to be right on a first draft. The reviewer catching a wrong volatility input is evidence that the control worked, and a vignette that offers you a caught error is usually testing whether you will mistake the catch for the breach.

Applied mini-project

Build and defend a six-vignette compliance memo.

You are the newly appointed junior compliance reviewer at a synthetic firm of your own invention. Write a short memo containing six original vignettes (all entities explicitly labeled synthetic), each 80–150 words, spanning at least five of the seven broad Standards, with at least one India-flavored (₹) and one US-flavored ($) scenario, and at least one vignette that is genuinely not a violation. For each vignette, write the full CFA-defended answer: Standard (Roman numeral and letter), verdict, and the one deciding fact, showing any arithmetic in a short Python snippet, exactly as the Worked Examples above do.

Do this, in order:

1. Draft. Write all six vignettes first, without solving them, the way an exam writer would, plant exactly one deciding fact in each.
2. Solve cold. Set the memo aside for at least an hour (or overnight), then solve all six from scratch as if you had never written them, using the four-step method.
3. Self-audit. For each, check: did you actually name a specific lettered sub-standard (not just a Roman numeral)? Did your stated "deciding fact" genuinely resolve the case, or did you write a paragraph of hedging instead? Is any numeric claim recomputed and correct?
4. Swap-test (recommended). If you can, trade memos with a study partner and grade each other's answers against your own intended key, a vignette that a colleague solves differently from how you intended is usually a vignette with an ambiguous or missing deciding fact, not a wrong colleague.

Rubric, self-score out of 10, pass at ≥8 with no zero on any item.

• 2 pts: all six vignettes correctly labeled synthetic, with at least one India (₹) and one US ($) scenario present.
• 2 pts: at least five of the seven broad Standards represented across the six vignettes.
• 2 pts: at least one vignette is genuinely, defensibly not a violation.
• 2 pts (every answer states Standard + letter, verdict, and deciding fact in one sentence) no paragraph-length hedging.
• 2 pts: every numeric claim is shown recomputed (a short Python snippet or hand arithmetic), not merely asserted.

Lab: once you score ≥8/10 on your own memo, open CL-ES.1 (Twelve Standards, Twelve Vignettes) and run it closed-book, under its stated ~110-minute budget, scoring against its seeded key. Passing CL-ES.1 at its own gate (≥5/6) alongside this mini-project's ≥8/10 is what this module considers complete.

Reading & resources

Everything below is an anchor to learn from and verify against, never a substitute for reading the primary Standards text itself before you rely on it in real work.

The primary source (read this first, and keep returning to it):

• CFA Institute, Code of Ethics and Standards of Professional Conduct, the seven Standards' actual current text, freely published. [Free] [Beginner–Intermediate] (cfainstitute.org/en/ethics-standards, the wording is periodically refined; always quote the current text, not this module's paraphrase, in real work.)
• CFA Institute, Standards of Practice Handbook (current edition), the full guidance, application examples, and procedures for compliance behind each Standard; this module's own worked examples are built in its style but are not a substitute for its case-by-case guidance. [Paid, print/PDF via CFA Institute] [Intermediate–Advanced] (check the current edition number and any provision changes at cfainstitute.org before citing a specific section.)

Applied practice (free, official):

• CFA Institute's Ethics case studies / Ethical Decision-Making Framework materials, worked, real-world-inspired scenarios in the Institute's own voice, useful for a second style of vignette beyond this module's. [Free] [Intermediate]
• CFA Program Ethical and Professional Standards readings (Level I), the exam-facing treatment of everything in this module, at the Level I candidate's depth. [Paid, part of CFA Program enrollment] [Beginner–Intermediate]

The regulatory and fiduciary map (own resources, not re-taught here):

• ES1.02: GIPS in full and the SEC/SEBI/FCA regulatory and fiduciary map; go there for anything about what a regulator or GIPS itself requires, rather than what personal conduct the Code and Standards require.
• SEBI (sebi.gov.in) and the SEC (sec.gov), the two regulators whose rules this module's Standard I(A) tells you to weigh against firm policy and the Code; M0.02 is where you learned to navigate both.

Do this, not just read: take any three real (not synthetic) news stories about an investment-industry enforcement action or resignation you can find, and, using only what the reporting states as fact, never treating an allegation as a finding (M2.08's language protocol), write, from scratch, the Standard-and-verdict citation you believe applies, then compare your citation against any official finding once one exists. Ten minutes of doing this with a live story teaches the discipline faster than another synthetic vignette.

Teach it back & journal

Feynman prompt. Explain to a friend who has never heard of the CFA Standards why "that seems unethical" is a useless answer on an ethics vignette, using this module's own opening example, the portfolio manager who traded personally an hour before a large client order. Walk them through all four steps out loud: name the actor and action, name the two or three Standards that could plausibly apply, find the one number in the story that actually decides it (the ₹22,000, or simply the fact that his trade had priority at all), and state the one-sentence verdict. Then explain why "but the client wasn't really hurt" would not save him, using the idea that Standard VI(B)'s violation is the opportunity his personal trade took, not proof of measurable harm. If your friend can, unprompted, ask you "wait, but is this actually a violation, or is this one of the clean ones?" before you tell them the answer, you have installed the module's real lesson: half of thinking like an analyst here is knowing that not every vignette is guilty.

Journal prompt. Look back over your own recent professional or academic work, an assignment, an internship task, a group project, a recommendation you made to a friend about their own money, and find one moment where two interests genuinely pulled in different directions: yours and someone else's, your firm's and a client's, what was easy to say and what was fully accurate. Write two paragraphs. First, describe the moment plainly, the way M2.08's language protocol requires: what happened, stated as fact, with no more certainty than you actually have. Second, name which Standard (if you were bound by one) that moment would have tested, state honestly whether you resolved it the way the Standard would require, and (if you did not) what the one-sentence CFA-defended verdict on your own choice would have been. The discipline this module teaches is not really about memorizing twenty-three lettered rules; it is about being willing to run your own conduct through the same cold, precise test you just ran on twelve synthetic strangers.

This module's flashcards and mastery quiz are wired into the app: see the node's Quiz and Reviews.