The Definitive Guide to Tax Research & Opinion Letters

A plain-English, taxpayer-focused guide to written tax advice — the confidence levels, what counts as authority, how a properly built opinion shields you from penalties, and how the national tax representation firm of Mike Habib, EA can help

Almost every guide about taxes is written for people who already have a problem. This one is different. It is written for the taxpayer standing at a fork in the road — before the return is filed, before the transaction closes, before the position is taken — who wants to know whether the tax treatment they are counting on will actually hold up. That is the work of tax research, and its work product is the opinion letter: a written analysis that tells you, honestly and with citations, how likely your position is to survive if the IRS challenges it, and that — if it is built correctly — can stand between you and a penalty later.

It is written for the person about to do something with real tax consequences: the business owner structuring a sale or a reorganization; the investor with a large, unusual, or aggressive deduction; the professional weighing a worker-classification decision; the crypto holder or real estate investor in a gray area; the taxpayer whose CPA said “I think that works, but I am not certain”; the executive facing an equity event; the company deciding whether a position must be disclosed. It explains what an opinion letter actually is and what it is not; the history — including the shelter era that produced the rules and the 2014 reform that rewrote them; the confidence levels and what the percentages really mean; the surprisingly short, exhaustive list of what legally counts as “authority”; how an opinion shields you from the accuracy-related penalty and the specific conditions that reliance must satisfy; the anatomy of a real opinion; how opinions fail; and how these questions play out in examination and appeals. It closes with two decades of practitioner lessons and anonymized case studies from the files of Mike Habib, EA.

One idea frames everything that follows, and it inverts how most people think about tax advice: a tax opinion is not a shield you buy after the fact — it is a record you build before it. The penalty defense that a good opinion provides depends on facts that exist at the moment the position is taken: what the law actually said, what authorities existed, what you told your adviser, what your adviser concluded, and whether you actually relied on it in good faith. None of that can be manufactured later. The taxpayer who commissions real research before filing owns a defense; the taxpayer who goes looking for a supporting letter after an audit notice arrives owns a piece of paper. That timing difference is the single most important thing in this guide.

Part One: What a Tax Opinion Letter Actually Is

Q: In plain English, what is a tax opinion letter?

A tax opinion letter is a written analysis, prepared by a qualified tax professional, that states how the tax law applies to a specific set of facts and expresses a level of confidence in that conclusion. It is not a guess, and it is not a marketing brochure. A real opinion does four things: it states the facts it relies on; it identifies the legal authorities that govern; it analyzes those authorities against the facts, honestly weighing what cuts against the position as well as what supports it; and it reaches a conclusion at a stated confidence level — “more likely than not,” “substantial authority,” and so on. The confidence level is not decoration; it is the operative content, because it is what determines the opinion’s legal effect.

Tax research is the work behind it. Before anyone can write a defensible opinion, someone has to actually find the law — the Code section, the regulations, the rulings, the cases — and read it against your facts. That is unglamorous, and it is where the value lives. An opinion letter without genuine research behind it is worse than no opinion at all, because it creates a false sense of security about a position that has never actually been tested against the authorities.

Q: What is the difference between a memo, an opinion, and a “comfort letter”?

These terms get used loosely, and the distinctions matter. A research memorandum is typically an internal work product: it analyzes an issue, often for the adviser’s own file or for the client’s decision-making, and it may be candid about uncertainty. A formal opinion letter is addressed to the client, states a confidence level, and is intended to be relied upon — it is written knowing it may one day be shown to the IRS. A so-called “comfort letter” is generally a vague reassurance with no real analysis, and it is the most dangerous document in this family, because it looks like protection and provides none.

Here is what surprises most people, and it comes straight from the rules: the label does not control. Under Circular 230 §10.37, the professional standards governing written tax advice apply to all written advice on a federal tax matter — a formal opinion on letterhead, a research memo, and an email all fall under the same standard. As practitioners often put it, when a writing interprets the law and applies it to facts, it expresses an opinion, whether it is on letterhead or a napkin. What varies is not whether the standards apply, but how thorough the advice must be, which depends on the scope of the engagement and the specificity of the advice sought. So the question is never “is this an opinion?” — it is “was this advice built to the standard the situation required?”

Q: Why would I ever pay for one? What does it actually buy me?

Three things, and only the first is obvious. The obvious one is knowledge: you find out, before you commit, whether the treatment you are counting on is solid, shaky, or wrong — which sometimes means the opinion talks you out of a position, and that is a successful engagement, not a failed one. The second is penalty protection: a properly built opinion, genuinely relied upon, can establish substantial authority or support the reasonable-cause-and-good-faith defense that defeats the accuracy-related penalty under IRC §6662 — a penalty equal to 20% of the underpayment. On a $500,000 adjustment, that is $100,000, and it is often the entire fight. The third is leverage in examination: a contemporaneous, well-cited analysis handed to an examiner changes the conversation from “justify yourself” to “respond to this.” Auditors and Appeals Officers are moved by authority, and a taxpayer who arrives with the authorities already marshaled is in a fundamentally different posture than one improvising.

And there is a fourth, quieter benefit: the discipline of the process itself. Building a real opinion forces the facts to be pinned down, the assumptions to be surfaced, and the weak points to be confronted before they become surprises. A great many bad tax positions die quietly on the research desk, which is exactly where they should die — long before they cost anyone a penalty.

Part Two: The History — How the Rules for Written Tax Advice Were Written in Blood

Q: Where did the rules for tax opinions come from?

Every rule governing written tax advice today exists because of an abuse it was designed to stop, and knowing that history tells you why the rules look the way they do. For most of the twentieth century, tax opinions were a quiet professional practice: a lawyer or accountant researched a question, wrote up the answer, and the client relied on it. The system worked on professional judgment and reputation. What broke it was the tax shelter era.

Beginning in the 1990s and accelerating into the early 2000s, an industry emerged around mass-marketed tax shelters — engineered transactions with little or no economic purpose beyond generating tax losses. And the fuel that made them sellable was the opinion letter. Promoters would obtain opinions from prominent firms concluding that a shelter “should” work, and then use those opinions as both a sales tool and a penalty shield: buy the shelter, get the opinion, and if the IRS challenges it, point to the opinion as reasonable cause. Opinions were sold by the pound, sometimes with the conclusion effectively pre-ordered and the facts assumed rather than verified. Congressional investigations and a wave of litigation exposed the practice, several prominent firms faced severe consequences, and the credibility of the tax opinion as an institution was badly damaged.

Treasury’s response, in 2004, was the “covered opinion” regime — new Circular 230 §10.35, an intricate set of rules imposing detailed requirements on opinions that fell within defined categories, particularly those touching tax-avoidance transactions. The rules were rigorous, and they were also, in practice, a disaster of a different kind. They were so complex and so burdensome that practitioners responded not by writing better opinions but by disclaiming everything: the era of the ubiquitous “Circular 230 disclaimer” appended to every email, memo, and cocktail-napkin scribble in America was born, warning that the writing could not be relied upon for penalty protection. The disclaimers appeared on communications containing no tax advice at all. They protected no one and informed no one. Treasury itself concluded the rules had increased burden “without necessarily increasing the quality of the tax advice that the client received.”

The correction came in 2014. In Treasury Decision 9668, effective June 12, 2014, Treasury and the IRS eliminated the covered opinion rules of §10.35 entirely and replaced them with a single, principles-based standard for all written tax advice in §10.37. The reformed §10.35 now addresses competence. The new §10.37 requires that a practitioner base written advice on reasonable factual and legal assumptions, consider all relevant facts the practitioner knows or reasonably should know, use reasonable efforts to identify and ascertain the relevant facts, rely on the advice of others only where that reliance is reasonable and in good faith, and — critically — not take into account the possibility that a return will not be audited or that an issue will not be raised. The government evaluates compliance under a reasonable-practitioner standard considering all facts and circumstances. Treasury expected the change to end the era of the reflexive disclaimer, and the IRS said as much in the preamble.

The lesson of that history is worth stating plainly, because it is the reason this guide exists. The rules were built to distinguish real analysis from purchased conclusions. A genuine opinion — one where the facts were actually verified, the contrary authorities actually weighed, and the conclusion actually followed from the analysis — has always been valuable and remains so. A conclusion-first document dressed up as an opinion was the problem then and is worthless now. The regulatory architecture is essentially an elaborate machine for telling those two things apart.

Q: What law governs tax opinions and the penalties they protect against?

Part Three: The Confidence Ladder — What the Levels Actually Mean

Q: What do “more likely than not” and “substantial authority” actually mean?

Tax opinions speak in a specialized vocabulary of confidence, and each rung of the ladder has a distinct legal consequence. Getting these straight is the single most useful thing a non-specialist can learn about opinions, because the words are not interchangeable and the difference between two of them can be a six-figure penalty.

• Will. The highest level — essentially certainty, conventionally understood as roughly 95% or better. Rare, and reserved for positions where the law is unambiguous.
• Should. A high level of confidence, conventionally around 70% or more. The position is clearly supported, though not beyond argument.
• More likely than not (MLTN). The pivotal level, and the one the regulations actually define: a greater than 50% likelihood that the position would be upheld if challenged. This is the threshold required for penalty protection on certain tax shelter items and in various other contexts, and it is the level most sophisticated planning opinions target.
• Substantial authority. An objective standard, defined by regulation as less stringent than more-likely-than-not but more stringent than reasonable basis. It is satisfied when the weight of authorities supporting the treatment is substantial in relation to the weight of authorities supporting contrary treatment. Practitioners conventionally place it around 40%, but note carefully: the regulations assign it no percentage. It is a weighing test, not a math test. Substantial authority on an undisclosed position removes the item from the substantial-understatement penalty calculation entirely.
• Reasonable basis. The regulations call it “a relatively high standard of tax reporting” — significantly higher than not-frivolous, and not satisfied by a position that is merely arguable or a merely colorable claim. Conventionally placed around 20%. A position with reasonable basis can escape the penalty if it is adequately disclosed (see Part Five).
• Not frivolous. The floor — a position that is not patently improper. This is not a level anyone should be operating at, and it does not protect you from anything.

Two warnings about the percentages. First, as noted, only the MLTN threshold (>50%) is actually stated in the regulations; the others are professional conventions that give practitioners a shared vocabulary, and no examiner is going to be persuaded by an arithmetic claim of “42% confidence.” Second, and more importantly, the regulation is explicit that the possibility a return will not be audited — or that the issue will not be raised if it is — is irrelevant to whether these standards are met. The audit lottery is not a tax strategy, and any adviser who factors it in is violating the standards.

Part Four: What Actually Counts as “Authority” — The List That Surprises People

Q: Can my adviser rely on a tax treatise, or an article, or another CPA’s opinion?

No — and this is one of the most consequential and least understood rules in the entire subject. Treasury Regulation §1.6662-4(d)(3)(iii) provides an exhaustive list of what may be treated as “authority” for determining whether substantial authority exists. The regulation says “only the following are authority,” and the list is:

• The Internal Revenue Code and other statutory provisions;
• Proposed, temporary, and final regulations construing those statutes;
• Revenue rulings and revenue procedures;
• Tax treaties, the regulations under them, and official explanations of them;
• Court cases;
• Congressional intent as reflected in committee reports, joint explanatory statements of managers in conference reports, and floor statements made before enactment by a bill’s managers;
• General Explanations of tax legislation prepared by the Joint Committee on Taxation (the “Blue Book”);
• Private letter rulings and technical advice memoranda issued after October 31, 1976;
• Actions on decisions and general counsel memoranda issued after March 12, 1981;
• IRS information or press releases, and notices, announcements, and other administrative pronouncements published in the Internal Revenue Bulletin.

Now the part that surprises people. Legal treatises, law review and journal articles, tax services, and the opinions of other tax professionals are NOT authority for this purpose. They may be enormously useful research tools — they will lead you to the authorities — but they cannot themselves supply substantial authority. This means an opinion that leans on a well-regarded treatise and a couple of practitioner articles, without grounding itself in the Code, regulations, rulings, and cases, is not built on authority at all in the eyes of the regulation. It is a common and quietly fatal defect.

Q: How is the weight of an authority determined?

The regulation is specific about weighing, and the rules reward precision over volume. All authorities relevant to the treatment must be considered — including the ones that cut against you; ignoring contrary authority does not make it go away, and an opinion that fails to address it is not credible. Weight depends on relevance, persuasiveness, and source: an authority that is materially distinguishable on its facts carries little or no weight, and one that merely states a conclusion is less persuasive than one that cogently relates the law to the facts. The type of document matters — a revenue ruling gets more weight than a private letter ruling on the same issue. Age matters — an older PLR, TAM, GCM, or action on decision is accorded less weight than a recent one, and any such document more than ten years old is generally given very little weight. And an authority ceases to be authority if it is overruled or modified by a body with power to do so. Notably, the taxpayer’s residence is not taken into account in determining substantial authority, and a well-reasoned construction of the statute can itself supply substantial authority even where other authority is sparse.

The practical upshot is that citation-counting is not analysis. Ten weak or distinguishable citations do not outweigh one squarely adverse revenue ruling. A real opinion weighs; a bad one merely lists. And because the standard is objective — the taxpayer’s subjective belief that substantial authority exists is expressly irrelevant — the question is always what a fair weighing of the actual authorities shows, not how confident anyone feels.

Part Five: The Penalty Shield — How an Opinion Actually Protects You

Q: What penalty are we actually defending against?

Principally the accuracy-related penalty under IRC §6662: an addition to tax equal to 20% of the underpayment attributable to, among other things, negligence, disregard of rules or regulations, or a substantial understatement of income tax. For an individual, an understatement is “substantial” when it exceeds the greater of 10% of the tax required to be shown on the return or $5,000. Twenty percent sounds modest until you put a number on it: a $486,000 adjustment carries a $97,200 accuracy penalty on top of the tax and the interest. In many examinations, the tax is conceded and the penalty is the entire remaining fight — which is precisely why the research that supports the position is worth commissioning before the return is filed, not after the notice arrives.

Q: How does an opinion defeat that penalty? There are two doors.

Door one: substantial authority. Under §6662(d), the understatement is reduced by any portion attributable to an item for which there was substantial authority. This is an objective test — it does not depend on your state of mind at all. If the authorities, fairly weighed, substantially support your treatment, the item simply comes out of the substantial-understatement calculation. A properly built opinion is the vehicle for establishing this, because it is the document that actually performs the weighing.

Door two: adequate disclosure plus reasonable basis. If your position does not reach substantial authority but does have a reasonable basis, you can still take the item out of the understatement by adequately disclosing it — typically on Form 8275, “Disclosure Statement,” or Form 8275-R where the position is contrary to a regulation. Disclosure is a genuine strategic tool and one of the most underused in tax practice: it trades a small increase in audit visibility for meaningful penalty protection on a position you believe in but cannot fully defend. Note the tradeoff honestly — disclosure does not make a wrong position right, and it does not protect a position lacking even reasonable basis.

And behind both doors: reasonable cause and good faith. IRC §6664(c) provides that no accuracy-related penalty applies to any portion of an underpayment for which the taxpayer had reasonable cause and acted in good faith. Reliance on professional advice is the classic route to this defense — but it is not automatic, and the conditions are strict.

Q: What does reliance on an opinion actually require?

This is where most failed reliance defenses die, and the requirements are well settled. Drawing on Treasury Regulation §1.6664-4 and the Tax Court’s framework (articulated in Neonatology Associates, P.A. v. Commissioner and applied continually since), reliance on professional advice excuses a penalty only if three things are true:

• 1. The adviser was a competent professional with sufficient expertise. Not a promoter, not a salesperson, not someone with no real knowledge of the area. Crucially, a taxpayer generally cannot reasonably rely on advice from someone who is promoting the very transaction — the promoter’s opinion is the paradigm case of unreasonable reliance, and this principle killed reliance defenses across the entire shelter era.
• 2. The taxpayer provided necessary and accurate information. You cannot feed your adviser incomplete or false facts and then hide behind the resulting opinion. If the opinion assumes facts that are not true, or was built without facts you possessed, the reliance fails — and this is why a real opinion pins the facts down in writing and asks you to confirm them.
• 3. The taxpayer actually relied, in good faith, on the adviser’s judgment. Genuine reliance, not decoration. A taxpayer who had already decided, who shopped for a favorable conclusion, or who obtained the opinion to paper a decision already made, has not relied in good faith — and courts are quite willing to say so.

Notice what this three-part test implies about timing, independence, and honesty. It implies the opinion must come before the decision, from someone with no stake in the transaction, based on facts you actually gave and that are actually true. That is a demanding standard, and it is demanding on purpose — it is the mechanism the law uses to separate genuine professional reliance from purchased cover. It is also, encouragingly, an entirely achievable standard for a taxpayer who simply does the thing honestly: get real advice, from a real professional, before you act, having told the truth.

Part Six: The Anatomy of a Real Opinion Letter

Q: What does a properly built opinion actually contain?

There is no government form for a tax opinion — Circular 230 §10.37 deliberately does not mandate a rigid structure, leaving the depth to depend on the scope of the engagement and the specificity of the advice sought. But two decades of writing and defending these documents produces a reliable anatomy, and every element earns its place because it corresponds to something that can be attacked later:

• The statement of facts. Detailed, specific, and confirmed by the client. This is the foundation, and it is the most common failure point: an opinion resting on facts that turn out to be wrong or incomplete protects nobody, because reliance condition two collapses.
• Assumptions and representations. Stated explicitly, and — per §10.37 — they must be reasonable. An opinion may not assume away the very issue in dispute, and unreasonable assumptions are the classic tell of a purchased conclusion.
• The issues presented. Framed precisely. Vague questions produce useless answers.
• The authorities. Drawn from the exhaustive §1.6662-4(d)(3)(iii) list — Code, regulations, rulings, cases, and the rest — not from treatises and articles, which are research tools rather than authority.
• The analysis — including the contrary authorities. The heart of the document. It weighs the supporting and opposing authorities honestly, distinguishes what can be distinguished, and concedes what must be conceded. An opinion that never mentions a contrary authority is not a strong opinion; it is an incomplete one, and an examiner will notice.
• The conclusion, at a stated confidence level. The operative sentence: “we conclude that it is more likely than not that…” Vague conclusions provide vague protection.
• Limitations and scope. What the opinion covers, what it does not, what it relies on, and what would change the analysis. Honest limitations strengthen an opinion; they show the document is a real analysis rather than a blanket blessing.

And one negative rule worth stating: the opinion may not consider the likelihood of audit. Circular 230 §10.37 expressly forbids taking into account the possibility that a return will not be audited or that an issue will not be raised, and the substantial-authority regulation says the same. An adviser who tells you “this is aggressive, but they will never catch it” has not given you an opinion; they have given you a reason to find a different adviser.

Part Seven: Worked Examples — Real Numbers, Start to Finish

Composites built from typical fact patterns. They illustrate method; every situation differs, and nothing here is advice on your facts.

Example 1: The research that changed the plan (and saved the penalty by preventing it)

A business owner planned a transaction expected to generate roughly a $1.2 million deduction, on the strength of a promoter’s assurance that it “should” work. Before filing, a real research engagement was commissioned. The analysis found the supporting material consisted largely of practitioner articles and the promoter’s own opinion — neither of which is authority under §1.6662-4(d)(3)(iii) — while a revenue ruling and two circuit decisions cut squarely against the structure as designed. The honest conclusion was that the position did not reach substantial authority, and might not reach reasonable basis. The client did not take the position. Outcome: no deduction, no adjustment, and no penalty — and the roughly $240,000 accuracy penalty that a $1.2 million adjustment would have carried never came into existence. This is the least visible and most valuable kind of engagement: the one where the research talks you out of it.

Example 2: The substantial-authority opinion that killed the penalty

A taxpayer took an aggressive but genuinely supportable position on the characterization of a large payment — roughly $486,000 of income treated in a way the IRS later challenged. Before filing, a full research memorandum had been prepared: the Code section, the regulations, three cases (including one squarely adverse, which the memo addressed and distinguished on its facts), and a revenue procedure, weighed honestly to a conclusion of substantial authority. On examination, the IRS proposed the full $486,000 adjustment plus a 20% accuracy penalty of $97,200. The contemporaneous memorandum was produced. The examiner sustained a portion of the adjustment, but the penalty was conceded in full — because the item was supported by substantial authority and, independently, because reliance on the contemporaneous professional analysis established reasonable cause and good faith. Outcome: the tax fight narrowed, and $97,200 of penalty eliminated by a document that existed before the return was filed.

Example 3: Disclosure as strategy — the Form 8275 decision

A taxpayer faced a genuinely uncertain deduction of about $150,000. The research concluded there was a reasonable basis and a decent argument, but candidly, not substantial authority — the authorities were thin and the leading case was distinguishable in both directions. Two paths existed: abandon the position, or take it and disclose it on Form 8275. The client chose disclosure. The IRS examined and disallowed the deduction. Because the item had been adequately disclosed and had a reasonable basis, it was excluded from the substantial-understatement calculation, and no accuracy penalty applied to it. Outcome: the taxpayer lost the deduction and paid the tax and interest — but paid no penalty, having made an informed, documented, disclosed decision rather than a hidden bet. That is what a good adviser means by “taking a position responsibly.”

Part Eight: Rejected Cases — Why Reliance Defenses Fail

Q: When does an opinion NOT protect you?

Opinions fail as penalty shields for a recognizable set of reasons, nearly all of which map back to the three reliance conditions or to defects in the document itself:

• It came from the promoter. The single most reliable way to lose a reliance defense. An adviser who is selling, marketing, or has a financial stake in the transaction is not an independent professional, and courts have consistently held reliance on such advice unreasonable. If the person recommending the deal also wrote the opinion blessing it, you have no shield.
• The facts in it were wrong or incomplete. Reliance condition two fails. If the opinion assumed facts that were not true, or was written without material facts you knew, it protects nothing — no matter how elegant the legal analysis.
• It was obtained after the decision — or after the notice. You cannot rely in good faith on advice you did not have when you acted. An opinion procured to paper an already-made decision, or worse, commissioned once an examination began, cannot establish that you relied on it in taking the position.
• It assumed away the issue. An opinion that assumes the very fact in dispute (that the transaction had economic substance, that the entity was bona fide, that the valuation was correct) has analyzed nothing. Circular 230 §10.37 requires reasonable assumptions, and this is what unreasonable looks like.
• It ignored contrary authority. An analysis that never confronts the adverse revenue ruling or the on-point case is not a weighing, and substantial authority is a weighing test. Examiners find the omitted authority quickly, and the omission itself damages credibility.
• It relied on non-authority. Treatises, articles, and other practitioners’ opinions are not authority under §1.6662-4(d)(3)(iii). An opinion built on them is not built on authority.
• The taxpayer had reason to doubt it. Good faith has content. A taxpayer sophisticated enough to know the deal was too good to be true, who ignored red flags or contrary advice from another professional, may find good faith hard to establish.

The unifying theme is that the law is trying to reward genuine professional reliance and punish purchased cover, and it is fairly good at telling them apart. Which is encouraging, in a way: the taxpayer who does this honestly — real adviser, true facts, before the decision — is protected precisely because the doctrine is built to protect exactly that person.

Part Nine: Examination and Appeals — Where the Research Pays Off

Q: How does my opinion actually get used if I am audited?

It becomes the spine of your defense, and the way it is deployed matters. In examination, a contemporaneous research memorandum handed to the Revenue Agent reframes the engagement: instead of the taxpayer scrambling to justify a position under time pressure, the agent is presented with a completed analysis — the facts, the authorities, the weighing — and must respond to it. Agents are not free to disregard authority, and a well-cited memo often narrows the issues immediately or resolves them outright. It also, importantly, establishes the contemporaneous record: it proves the analysis existed before the return was filed, which is the fact on which the entire reasonable-cause defense depends.

If the agent proposes an adjustment and a penalty anyway, the opinion becomes even more valuable in the Independent Office of Appeals, because Appeals evaluates cases on the hazards of litigation — the realistic probability the government would lose if the case were tried. That is precisely the question a good opinion already answers, with citations. A memorandum concluding there is substantial authority, weighing the contrary cases honestly, is a hazards argument in ready-made form, and penalties are among the most commonly conceded items in Appeals for exactly this reason. It is worth adding a procedural point that has erased a great many penalties on its own: under IRC §6751(b), most penalties require timely written supervisory approval before assessment, and a defect there is a pure legal defense that requires no opinion at all. Checking for it is standard practice. This series’ companion guides on IRS audit representation and the IRS appeals process cover those forums in depth.

Part Ten: Special Situations and Strategy Notes

Q: What if my transaction is a “reportable transaction”?

Then disclosure is not optional and confidence levels will not save you. The reportable-transaction regime under IRC §6011 and its regulations requires taxpayers to disclose participation in defined categories of transactions — listed transactions, transactions of interest, confidential transactions, transactions with contractual protection, loss transactions, and others — on Form 8886, with material advisers subject to their own disclosure and list-maintenance obligations under IRC §§6111 and 6112. The penalties for failing to disclose a reportable transaction under IRC §6707A are severe and are imposed for the non-disclosure itself, independent of whether the underlying position turns out to be correct. A strong opinion does not excuse a failure to disclose. The first question in any unusual, marketed, or loss-generating structure is therefore not “will it work?” but “must it be disclosed?” — and an adviser who never raises that question is not protecting you.

Q: Does the opinion protect my tax preparer too, or just me?

They are related but separate regimes, and understanding the difference explains a dynamic you may have felt without naming. Under IRC §6694, a return preparer faces their own penalties for an unreasonable position: generally, a preparer needs substantial authority for an undisclosed position, or reasonable basis for a position that is adequately disclosed (with a higher standard for tax shelters and reportable transactions). This is why a preparer may decline to sign a return taking a position you want, or may insist on disclosure — they are managing their own exposure, and their standard is not identical to yours. Good research serves both parties: it tells you whether you are protected, and it tells your preparer whether they can sign. Where the taxpayer and preparer standards diverge, disclosure on Form 8275 is frequently the bridge that lets a defensible-but-uncertain position be taken responsibly by everyone.

Q: Do the same rules apply to my California taxes?

California has a parallel architecture, and taxpayers routinely forget the state side until it bites. The Franchise Tax Board imposes its own accuracy-related penalty largely conforming to the federal model, and California has its own reportable- and listed-transaction disclosure regime with its own substantial penalties, along with a noneconomic substance transaction understatement penalty that is notably harsh. California also maintains its own standards for what will excuse a penalty. Because California conforms to much of federal law but not all of it — and because the FTB is entitled to reach its own conclusions — a federal opinion does not automatically resolve the state question, and a position that is solid federally can encounter a different answer in Sacramento. Any research engagement for a California taxpayer with a significant position should address the state consequences explicitly rather than assuming conformity. This series’ companion guide on California FTB matters covers the state agency in depth.

Q: What about a Private Letter Ruling — should I just ask the IRS directly?

Sometimes, and it is worth knowing the option exists. A Private Letter Ruling is a written determination issued by the IRS to a specific taxpayer, applying the law to that taxpayer’s specific facts on a prospective transaction. Its great virtue is certainty: a PLR binds the IRS with respect to the taxpayer who obtained it, on the facts presented. Its costs are real too: a user fee, a submission process, a wait of months, the requirement to lay out your transaction to the IRS before doing it, and the fact that some issues are simply not ruled on. A PLR obtained by someone else is not precedent and may not be cited as such by another taxpayer — though, under the substantial-authority regulation, PLRs and TAMs issued after October 31, 1976 do count as authority for weighing purposes, with less weight than a revenue ruling and very little if more than ten years old. The practical calculus: for a large, unusual, prospective transaction where certainty is worth more than speed, a ruling can be the right tool. For most planning questions, a well-built opinion delivers most of the protection at a fraction of the cost and time.

Q: How long should I keep the opinion, and what should I do with it?

Keep it as long as the years it supports remain open — which, given that the assessment statute runs three years from filing (six for a substantial omission of income, and indefinitely for an unfiled return or fraud), generally means at least six or seven years, and longer for positions with continuing effects such as basis, carryovers, or depreciation, where the position lives on for as long as the attribute does. Store it with the engagement letter, the facts you confirmed, and the return it supports, because those documents together are the reliance defense. And do not treat it as static: an opinion speaks as of its date, and a subsequent revenue ruling, regulation, or circuit decision can undermine it. For a position taken year after year, the analysis should be revisited periodically — a conclusion that was solid in one year can quietly become indefensible after an adverse development, and continuing to take the position without checking is exactly the kind of thing that turns a reasonable-cause defense into a negligence finding.

Part Ten-B: Lessons from 500+ IRS & State Cases — What Two Decades of Research and Opinion Work Actually Teaches

Everything to this point could, in principle, be assembled from the Code, the regulations, Circular 230, and the case law. What follows cannot. In our experience representing taxpayers for more than 20 years — writing research memoranda and opinions, and then, on the other side of the desk, defending positions in examination and Appeals where somebody else’s opinion was the only thing standing between a client and a penalty — the same patterns repeat with such regularity that they function as rules. These observations come from casework: from memoranda that held up under audit, from reliance defenses that failed for entirely predictable reasons, and from the uncomfortable conversations in which a taxpayer learns that the impressive-looking letter they paid for protects nothing. They are not from AI summaries or public IRS documents, and they are shared because the difference between a real opinion and an expensive decoration is invisible to most taxpayers until it is tested.

Ten mistakes taxpayers make before hiring representation

1. Relying on the promoter’s opinion. The most expensive mistake in this entire field, and the most common. The person selling you the structure cannot be the person blessing it — that reliance is unreasonable as a matter of settled law, and it has cost taxpayers more in penalties than every other error on this list combined.
2. Getting the opinion after the decision. An opinion obtained to paper a choice already made cannot establish good-faith reliance in taking the position. Sequence is substance.
3. Shopping for a conclusion. Calling advisers until one says yes does not produce an opinion; it produces evidence of bad faith, and it is often discoverable.
4. Withholding inconvenient facts. The fact you did not mention is the fact that voids the opinion. Every failed reliance defense we have seen from the inside had a fact problem at its center.
5. Confusing a comfort letter with an opinion. A two-paragraph reassurance with no facts, no authorities, and no stated confidence level is not a shield. Taxpayers routinely cannot tell the difference until an examiner explains it to them.
6. Assuming a treatise or an article is authority. They are not, under §1.6662-4(d)(3)(iii). An opinion resting on them is resting on nothing that counts.
7. Ignoring the disclosure question. Taxpayers obsess over whether a position will win and never ask whether it must be disclosed — and §6707A penalties for non-disclosure of a reportable transaction land regardless of who was right on the merits.
8. Believing the audit lottery is a strategy. The regulations expressly exclude the likelihood of audit from every confidence standard. Any adviser who counts on not being caught is disqualifying their own advice.
9. Never revisiting a standing position. A position taken every year on an opinion written six years ago, through an intervening adverse circuit decision, is no longer supported — and continuing it looks less like reliance and more like disregard.
10. Skipping the research to save the fee. The arithmetic is unkind: a research engagement costs a fraction of the 20% accuracy penalty on a mid-six-figure adjustment. Taxpayers economize on the analysis and then pay for the penalty, the interest, and the representation anyway.

Q: What Revenue Agents actually ask when a position is challenged — and what they are really testing

When an examiner meets a significant or unusual position, the questions follow a pattern that mirrors the legal standards precisely, and knowing them tells you exactly what a good opinion must be prepared to answer. What is your authority for this treatment? — the direct question, and the one that goes nowhere pleasant if the answer is a treatise or a feeling. When did you get this advice, and from whom? — testing the timing and the independence, the first and third reliance conditions. What facts did you give your adviser, and are they accurate? — testing the second condition, and the place most defenses break. Did your adviser have a financial interest in the transaction? — the promoter question, asked directly. Did anyone tell you this might not work? — probing good faith. And, quietly: does the analysis address the authority that cuts against you, or only the authority that helps?

What the agent is really testing is whether this was genuine professional reliance or purchased cover — the same distinction the entire regulatory architecture exists to draw. In our experience, the single most decisive factor is whether the memorandum is contemporaneous and honest. An examiner encountering a dated, well-cited analysis that candidly weighs an adverse case and still reaches a supportable conclusion is looking at a taxpayer who behaved reasonably, and penalties tend not to survive that impression. An examiner encountering a conclusion-first letter that ignores the obvious contrary ruling is looking at something else entirely, and they know it within minutes. Agents are far better at spotting the difference than taxpayers assume, because they see both kinds constantly. The honest memo is not just more ethical — it is more effective, which is a happy convergence.

Q: Why penalty defenses fail — the file-level anatomy

• The adviser who wrote the opinion was promoting or selling the transaction, so reliance was unreasonable as a matter of law and the analysis was never reached.
• The opinion rested on assumed facts that examination disproved — the assumption was doing the work the analysis should have done.
• The document was procured after the return was filed, sometimes after the examination opened, and could not establish reliance at the time of the position.
• The analysis cited only supporting authority and never addressed the on-point contrary revenue ruling, so it failed the weighing that substantial authority requires.
• The position was a reportable transaction that went undisclosed, and the §6707A penalty applied regardless of the merits — a penalty no opinion can cure.

The inverse of each failure is a practice standard: independent adviser, verified facts, contemporaneous analysis, honest weighing of contrary authority, and disclosure where disclosure is required. Do those five things and the penalty defense is usually available. Skip any one of them and it usually is not.

Q: How tax research and opinion practice has changed over the past decade

A practitioner writing opinions in the mid-2010s would recognize the standards but not the environment. The 2014 reform reset the framework: with the covered opinion rules gone and one §10.37 standard governing all written advice, the reflexive disclaimer disappeared and the emphasis shifted from formalistic compliance to substance — reasonable assumptions, real facts, honest analysis. Enforcement of disclosure intensified: the reportable-transaction regime has expanded through listing notices and transactions of interest, and §6707A penalties for non-disclosure have become a first-order risk rather than a footnote. The §6751(b) supervisory-approval defense matured into a routine and powerful procedural check on penalties, reshaping how penalty cases are defended. Information reporting and data matching have narrowed the factual gray zones that once absorbed aggressive positions, so the questions that reach a research desk today are more genuinely legal and less about what the IRS will see. And new substantive frontiers — digital assets, complex passthrough regimes, evolving worker-classification law, state-level economic nexus — have created large areas where authority is genuinely thin and a well-reasoned construction of the statute is doing real work. Net of ten years: less formalism, more substance, higher stakes on disclosure, and a growing premium on advisers who can actually research an open question rather than recite a settled one.

Part Ten-C: Anonymized Case Studies — Process and Outcome

Drawn from actual representation matters handled by the firm. No names, no identifying details, no confidential information; figures are rounded and certain facts generalized to protect client identity. They demonstrate process and outcome — never a promise of results, because every position turns on its own facts and authorities.

Case study: the $486,000 adjustment where the memo killed a $97,200 penalty

Client took a significant and genuinely contestable position; before the return was filed, we prepared a full research memorandum — Code, regulations, revenue procedure, and three cases including one squarely adverse, which we addressed and distinguished on the facts — concluding there was substantial authority for the treatment. On examination the IRS proposed an adjustment of approximately $486,000 and a 20% accuracy-related penalty of $97,200. We produced the contemporaneous memorandum. The examination sustained part of the adjustment on the merits, but the penalty was conceded in full, on both grounds: substantial authority under §6662(d), and reasonable cause and good faith under §6664(c) based on contemporaneous reliance on a competent, independent professional analysis. Outcome: the tax dispute narrowed and $97,200 of penalty eliminated — by a document that existed before the return was ever filed.

Case study: the promoter’s opinion that protected no one

Client came to us already under examination, holding an impressive-looking opinion letter obtained from the firm that had marketed the structure to them. The letter concluded the transaction “should” work. It also assumed the very economic substance that was in dispute, cited no contrary authority, and had been provided as part of the sales process. We advised candidly that the reliance defense was very likely unavailable — the adviser was a promoter, and reliance on a promoter’s opinion is unreasonable as a matter of settled law — and redirected the defense to the merits and to a §6751(b) supervisory-approval review, which ultimately produced meaningful relief on the penalty. Outcome: a partial recovery, but the hard lesson is the point of the study: the client had paid substantial money for a document they believed was a shield, and it was not one. The time to learn that is before the transaction, not during the audit.

Case study: the research that talked a client out of a $1.2 million deduction

Client was presented with a structure promising roughly a $1.2 million deduction and asked us to review it before committing. The research found the supporting material was practitioner articles and the promoter’s own analysis — neither of which is authority under §1.6662-4(d)(3)(iii) — while a revenue ruling and two circuit decisions were squarely contrary. We concluded the position did not reach substantial authority and quite possibly not reasonable basis, and we said so. The client declined the transaction. Outcome: no deduction, no adjustment, no penalty, no examination — and roughly $240,000 of accuracy-related penalty exposure that simply never came into existence. The best tax controversy is the one that never happens, and the least celebrated engagement is the one that prevents it.

Case study: disclosure that cost the deduction and saved the penalty

Client wished to take an uncertain deduction of roughly $150,000. Our analysis was candid: reasonable basis, yes; substantial authority, no. We laid out the three options — abandon, take it undisclosed and accept penalty exposure, or take it with adequate disclosure on Form 8275. The client chose disclosure. The IRS examined and disallowed the deduction in full. Because the item was adequately disclosed and had a reasonable basis, it was excluded from the substantial-understatement computation and no accuracy-related penalty attached to it. Outcome: tax and interest paid, penalty avoided entirely — an informed, documented, disclosed decision rather than a hidden bet, which is precisely what taking a position responsibly looks like.

Case study: the standing position that outlived its opinion

Client had taken the same position for six consecutive years on the strength of a research memorandum written in year one. In year four, a circuit decision had come down squarely against the treatment; nobody revisited the analysis, and the position continued unchanged. On examination of years five and six, the reasonable-cause defense was substantially weaker: an opinion speaks as of its date, and continuing a position through a known adverse development looks less like reliance and more like disregard of the law. We rebuilt the analysis, conceded what had to be conceded, and negotiated the penalty down in Appeals on the strength of the original good-faith adoption and the genuine ambiguity that had existed at the outset. Outcome: partial penalty relief, and a standing engagement to revisit the position annually. The lesson generalizes: research is not a monument, it is a living record, and the law moves underneath standing positions.

Part Eleven: Bad Opinion Help — How to Tell a Shield From a Decoration

Q: How do I know if the opinion I am being offered is real?

This is the most practically useful question in the guide, because a taxpayer holding a worthless opinion usually feels perfectly safe — that is the whole problem. The document looks impressive. It is on letterhead. It uses the vocabulary. And it will not survive contact with an examiner. The warning signs are specific, and once you know them they are not subtle:

• It came from the person selling you the deal. Stop here. This is disqualifying. Reliance on a promoter’s opinion is unreasonable as a matter of settled law, and no amount of quality in the document cures it.
• It has no confidence level, or a mushy one. A real opinion says “more likely than not” or “substantial authority” and means it. “We believe this is a supportable position” is not a confidence level; it is a hedge.
• It never mentions a single authority that cuts against you. Substantial authority is a weighing standard. An analysis with no counterweight has not weighed anything, and an examiner will locate the contrary ruling in an afternoon.
• It assumes the thing in dispute. If the opinion assumes economic substance, or a valuation, or a business purpose, and those are the issues, the opinion has assumed away the case. Circular 230 §10.37 requires reasonable assumptions.
• Its authority is treatises, articles, and “the opinion of national tax counsel.” None of those is authority under Treas. Reg. §1.6662-4(d)(3)(iii). It is a bibliography, not a foundation.
• Someone mentions the odds of being audited. Expressly forbidden as a consideration, and an instant signal that you are not receiving professional advice.
• It arrived after you had already decided. Then it is documentation, not reliance, and the third condition of the defense fails.
• Nobody asked you hard questions about your facts. A real engagement is uncomfortable. If the process was frictionless and nobody probed the inconvenient details, the facts were not verified — and the facts are the foundation.

The contrast worth stating plainly: a real opinion is independent, contemporaneous, built on verified facts, grounded in actual authority, honest about what cuts against you, explicit about its confidence level and its limits, and entirely willing to tell you no. That last quality is the tell. An adviser who has never told a client that a position does not work is not an adviser; they are a vendor. The value of genuine research is precisely that it can come back negative — and the taxpayers best protected are the ones who hired someone willing to say so.

How Mike Habib, a Federally Licensed Enrolled Agent Helps

As a federally licensed Enrolled Agent admitted to practice before the Internal Revenue Service under Treasury Department Circular 230, Mike Habib is authorized to advise and represent taxpayers in all 50 states — and that dual capacity is exactly what makes the research work valuable. Mike does not write opinions in the abstract; he writes them having spent two decades on the other side of the table, in examinations and in the Independent Office of Appeals, watching which documents hold up and which collapse. A memorandum written by someone who has personally defended positions in front of Revenue Agents is built differently from one written by someone who has never had to.

Mike Habib, EA brings a combination that is genuinely uncommon in this work: two decades of hands-on federal and California tax controversy experience layered on a corporate finance career as a former Controller at Xerox Corporation and Director of Finance at AEG. Tax research at the level that matters is a rigorous analytical exercise — finding the authorities that actually count, weighing the ones that cut against you, pinning the facts down before they shift, and reaching a conclusion you are prepared to defend. Clients get an adviser who reads the authorities the way an examiner will, and who says no when the answer is no.

What the engagement actually looks like at Mike Habib, EA:

• The facts pinned down first. A rigorous factual development — the uncomfortable questions asked, the documents reviewed, the assumptions surfaced and tested — because a wrong or incomplete fact voids the entire defense, and this is where most failed reliance defenses were lost.
• Real research, in real authority. The Code, the regulations, rulings, procedures, and cases — the authorities that actually count under Treas. Reg. §1.6662-4(d)(3)(iii) — not a bibliography of treatises and articles that carry no weight when it matters.
• The contrary authority confronted, not hidden. Every adverse ruling and on-point case addressed and either distinguished or conceded — because substantial authority is a weighing test, and an opinion that only cites support has weighed nothing.
• A stated confidence level you can act on. Will, should, more likely than not, substantial authority, reasonable basis — named explicitly, with the consequences of each explained, so you know exactly what protection you are buying and what risk you are keeping.
• The disclosure question answered. Whether the position must be disclosed as a reportable transaction (Form 8886), and whether adequate disclosure on Form 8275 is the right strategic choice — because a §6707A penalty lands regardless of who was right on the merits, and no opinion cures it.
• An honest no when the answer is no. The most valuable thing a research engagement can produce is the conclusion that a position does not work — delivered before you take it, while it still costs you nothing.
• And the defense, if it ever comes. Because Mike handles the examination and the appeal as well, the memorandum is written from the outset to be the document that wins the audit — and if the position is challenged, the person who wrote it is the person who defends it.
• Direct, personal work from start to finish. Mike personally handles every engagement — no junior staff hand-offs, no research farmed out. When you call, you reach him.

The firm provides tax research and written opinions for individuals, business owners, investors, and companies nationwide — all 50 states and Americans abroad — on planning and transaction questions, business structuring and reorganizations, worker classification, real estate and passthrough issues, equity and compensation events, digital assets, residency and multi-state questions, and California matters where the FTB, CDTFA, or EDD consequences run alongside the federal ones. And because the same person handles the controversy work, the research is built for the fight it may one day have to survive. The companion guides in this series cover that side in depth — IRS audit representation, the IRS appeals process, penalty abatement, and the California agencies.

Part Twelve: Rapid-Fire FAQs — Straight Answers to the Questions Taxpayers Ask

Your Next Step

If you have read this far, you understand the thing that most taxpayers learn only in an examination room: that a tax opinion is not a product you buy to feel safe, but a record you build to be safe — and that the difference is decided by facts that are fixed long before anyone challenges you. Who advised you, and were they independent. What you told them, and was it true. When you got the advice, and did you actually rely on it. Whether the analysis rested on real authority and honestly confronted what cut against it. Get those right and you own a genuine defense — one that can eliminate a 20% penalty and reframe an entire audit. Get them wrong and you own an expensive piece of paper that will not survive its first serious reader. What no guide can do is answer your question, on your facts, with the authorities that actually govern.

That is where Mike Habib, EA starts every research engagement. Call 562-204-6700 or toll-free 1-877-788-2937, or visit myirstaxrelief.com, to discuss your position or transaction. You will speak directly with Mike — a federally licensed Enrolled Agent with 20+ years of representation experience and a corporate finance background as a former Controller and Director of Finance — not a salesperson working a script, and not someone with a stake in the transaction you are asking about. Engagements are quoted as a transparent flat fee for the defined scope of the question, so you know the full investment before work begins: no hourly meters running while the research unfolds, no surprise invoices, and a fraction of what large national firms charge for work handled by rotating junior staff. You will get real research, in real authority, with an honest confidence level — and, when that is the right answer, an honest no, delivered while it still costs you nothing.