Tax In Action

The Overtime Deduction Just Made the Department of Labor’s Definition of Employee Your Problem

Earmark Team · May 8, 2026 ·

The gig economy has exploded over the past decade. From Fiverr to Uber, from seasonal warehouse workers to freelance accountants, the line between employee and independent contractor has become increasingly blurred. California alone spent years in legal battles over worker classification, with court cases dragging on and state laws changing back and forth.

However, a single worker can legally be an “employee” under one federal law and an “independent contractor” under another for the same work, at the same time. And thanks to the One Big, Beautiful Bill Act, this distinction directly impacts your tax practice.

In Episode 24 of Tax in Action, Jeremy Wells, EA, CPA, tackles this complexity head-on in the first part of a two-part series on worker classification and misclassification. He breaks down exactly how the IRS distinguishes between employees and independent contractors and why tax professionals cannot ignore definitions that come from outside the Internal Revenue Code.

Payroll Taxes Are at the Heart of This Discussion

As Jeremy emphasizes early in the episode, “the tax consequences can be significant for both the employer and the worker.” The gig economy creates opportunities for flexible work, but also leaves workers without employment benefits, fair labor protections, and payroll tax matching.

Payroll taxes are “really the most important aspect of this discussion from a tax perspective,” Jeremy explains. It comes down to who’s responsible for the payroll tax or self-employment tax that results from the money earned.

The stakes are high. Misclassifying a worker can lead to both the employer and worker facing tax liabilities that compound quickly. Get it right, and everyone knows where they stand with FICA, FUTA, and federal income tax withholding.

One Word, Multiple Federal Definitions

For most of our careers, we’ve operated within the comfortable boundaries of Title 26, the Internal Revenue Code. If someone mentioned the Fair Labor Standards Act (FLSA), we knew that was the labor lawyers’ territory. Not anymore.

Jeremy explains that “employee” means different things in different contexts across federal law. There’s a well-established principle that a term should have the same meaning within a single title of the U.S. Code, but it can mean something entirely different when you cross from one title to another.

The Department of Labor uses what it calls the “economic reality test” to determine employee status under the FLSA. This test examines six factors:

Opportunity for profit or loss based on managerial skill
Investments by both parties
Permanence of the relationship
Nature and degree of control
Whether the work is integral to the employer’s business
The worker’s skill and initiative

The key question for the DOL is economic dependence. As Jeremy notes from the DOL’s Fact Sheet 13, “If the economic realities show that the worker is economically dependent on the employer for work, then the worker is an employee.”

The critical distinction is that the DOL explicitly states, “employment under the FLSA is not determined by technical concepts or common law standards of control. It is broader than the common law standard often applied to determine employment status under other federal laws.”

The 2025 Change

Why does this matter for tax professionals? The One Big, Beautiful Bill Act created a new deduction for overtime pay, but it specifically references FLSA Section 7, which deals with employees entitled to overtime compensation.

“An employee who is covered under FLSA Section 7 may qualify for a deduction for part of the overtime payment that the worker earned,” Jeremy explains, highlighting the significance.

This creates an unprecedented situation because “a worker can be considered an employee under FLSA and therefore eligible for potentially deductible overtime, yet not considered an employee for federal employment tax purposes.”

The IRS recognized this gap. In Notice 2025-69, the agency provides guidance on “how employers should report overtime paid to workers who are covered under FLSA Section 7 but are not employees for payroll tax purposes and so won’t receive a W-2.”

The IRS Control Standard: Three Categories That Drive Every Decision

So how does the IRS actually decide who’s an employee? It starts with IRC Section 3121(d), which provides four statutory definitions: common law employees, and corporate officers, certain statutory employees, certain statutory nonemployees.

For most situations, we’re dealing with the common law employee definition. That definition hinges on the common law “right to control” standard, which comes from Supreme Court precedent.

The standard boils down to one question: Does the employer retain the right to direct and control the means and details of the work?

“It’s less about whether the employer actually does control the worker, and more about whether the employer retains the right to control the worker,” Jeremy says, emphasizing a crucial distinction.

An independent contractor, by contrast, is “typically subject to control only as to the desired result, not the means or the methods of doing the work.”

The Evolution from 20 Factors to Three Categories

Courts have spent roughly half a century developing this definition. Key cases include Weber v. Commissioner (1994), Professional and Executive Leasing, Inc. v. Commissioner (Ninth Circuit, 1988), and Simpson v. Commissioner (Tax Court, 1975).

In 1987, the IRS and Social Security Administration compiled 20 factors from court precedents and published them in Revenue Ruling 87-41. Then in 1996, the IRS reorganized these into three categories of evidence in an examiner training manual. Jeremy stresses these are “categories of evidence. They are not themselves legal tests.”

Behavioral Control: The Details and Means of Performance

This category examines whether the employer has “the right to direct or control the details and means by which the worker performs the required services.”

Key indicators include:

Instructions: Jeremy uses a simple example: “If I hire a worker and tell that worker, ‘I need you to produce a widget for me,’ and I don’t tell them anything more than that, then I have given that worker essentially no instruction.” That leans toward independent contractor. But if you specify the tools, timeline, location, and step-by-step process, that leans toward employee.
Evaluation: Monitoring how work is performed (not just the final result) indicates greater control.
Training: Required, periodic, or ongoing training on methods and procedures suggests employment.
Uniforms and branding: These can indicate employment, but Jeremy notes modern realities. “Customer security concerns have led some of these companies to insist that their workers dress up in their uniforms, and have their logos displayed even though they’re classified as independent contractors.”

Jeremy adds a nuance particularly relevant for professionals: “Instructions imposed by the business merely to ensure compliance with customer orders or governmental or governing body regulations may indicate weaker control than more stringent guidelines imposed directly by the business.”

Financial Control: The Economic Aspects

This category looks at “the right to direct or control the economic and business aspects of the worker’s activities.”

Important factors include:

Significant investment: Who provides equipment and pays for large expenditures? Jeremy notes everything is relative. “I run an accounting firm. The biggest equipment expense we have is computers. That’s nothing compared to buying large equipment for a factory.”
Business expenses: “Choosing to incur unreimbursed expenses typically indicates that the worker has the right to direct and control the financial aspects of the business operations.”
Market availability: Can the worker seek other business opportunities? Jeremy emphasizes a critical distinction from the DOL test, citing Nationwide Mutual Insurance Co. v. Darden (Supreme Court, 1992): “The question here is whether the worker has the right to direct and control business-related means and details of the worker’s performance, not whether the worker is economically dependent.”
Method of payment: Guaranteed salary or hourly wages typically indicate employment, though Jeremy notes “plenty of independent contractors, especially freelancers and firms as well, bill for time.”

Relationship of the Parties: Intent Concerning Control

This category examines how both parties perceive their relationship.

Written agreements: These help establish intent, but Jeremy warns, “Just because something’s in writing doesn’t necessarily make it so. We still have to look at the substance of the relationship.”
Incorporation: If a worker operates through a legitimate entity that “follows corporate formalities and has at least one non-tax business purpose,” that generally supports independent contractor status.
Employee benefits: Certain benefits, such as tax-qualified retirement plans, 403(b) annuities, and cafeteria plans, can only be provided to employees. Benefits paid to contractors can often uncover a worker misclassification case. Jeremy is clear: “If we see any of these kinds of benefits, then by definition, we have an employee.”

The S Corporation Officer Trap

Jeremy saves one of his strongest warnings for corporate officers. “Corporate officers are generally considered employees, especially if they are providing services to the corporation.”

For S corporations, this is critical. “An officer of an S corporation that provides services to that corporation is an employee, meaning that individual needs to be paid wages.”

The only exception requires meeting both conditions: the officer provides minor or no services AND is not entitled to receive any pay, directly or indirectly.

Jeremy calls out a common but problematic practice. “One way some tax professionals try to use two wrongs to make a right is issuing a 1099-NEC from the S corporation to that individual. Two wrongs don’t make a right.”

“Even though they both go into Social Security and Medicare, paying self-employment tax is different from paying FICA.” The tax liabilities remain; you’ve just created documentation of the misclassification.

Interestingly, Jeremy notes that one person can legitimately receive both a W-2 and 1099 from the same corporation. “You can have an individual working as an officer for a corporation and as a director for a corporation. That individual’s wages earned as an officer would be reported as wages on a form W-2, and then that individual’s pay as a director would be paid as compensation to a non-employee.”

Most Workers Live on a Spectrum

Jeremy brings us back to practical reality. “In the real world it’s a spectrum. On one end of that spectrum is a pure independent contractor where the employer just says, this is what we want you to do. Now go do it. On the other end, we have an employee where the employer tells the employee exactly how to do every single step.”

Most workers fall somewhere in between. As tax professionals, Jeremy explains, “we might have to make a determination of which end of that spectrum does this worker lean toward more?”

What Comes Next

This episode is part one of a two-part series. In part two, Jeremy will cover what happens when we have a misclassification and what workers and employers can do about that misclassification.

For now, the practical takeaways are:

Learn the DOL’s economic reality test. The overtime deduction depends on it.
Review IRS Notice 2025-69 for guidance on FLSA-covered workers who aren’t employees for tax purposes.
Use the three categories of evidence as your analytical framework, remembering the underlying legal test is the control standard.
Audit your S corporation clients. Officers providing services must be on payroll.
Document substance over labels in all worker relationships.

Listen to the full episode of Tax in Action to hear Jeremy walk through the complete analysis, including all the court cases and regulatory citations that inform these critical classification decisions.

These S Corp Election Mistakes Create Years of IRS Problems

Earmark Team · March 23, 2026 ·

A sole proprietor registers a brand-new LLC, reuses the EIN from their old payroll account, files Form 2553 with an effective date of January 1 (months before the entity even existed) and waits for the IRS to bless the election. What they get back instead is a mess: a new EIN they didn’t ask for, returns filed under the wrong number, and IRS notices piling up about unfiled 1120-S returns. It’s the kind of procedural train wreck that Jeremy Wells, EA, CPA, sees regularly in practice, and it’s entirely preventable.

In this episode of Tax in Action, Jeremy breaks down the S corporation election from start to finish, including the eligibility requirements, the precise mechanics of Form 2553, the framework for late election relief under Rev. Proc. 2013-30, and the analytical rigor required before recommending the election in the first place. The episode is a direct response to the flood of oversimplified S corp content circulating online, much of it from influencers who reduce a major business decision to a single rule of thumb about income thresholds.

In reality, the S corporation election decision is loaded with procedural traps and downstream implications that demand careful analysis. Tax professionals who understand the mechanics of making the election and the full range of factors that determine whether it’s actually beneficial serve their clients far better than those chasing shortcuts.

The episode walks through the procedural mechanics of making the election, the common mistakes that derail it, how late election relief actually works, and what practitioners consistently get wrong about it. Finally, Jeremy digs into why the decision to elect S demands analysis that goes well beyond self-employment tax savings, including ownership structure, balance sheet consequences, QBI deduction impacts, and state and local taxes that can wipe out any benefit entirely.

Getting the election right: The procedural traps that create lasting problems

Before evaluating whether the S election makes sense for a client, you need to know how to actually make it correctly. The requirements look straightforward on paper. In practice, several mistakes can cause problems that last years.

Eligibility

The entity must be a domestic corporation or domestic eligible entity under IRC §1361(b)(1). It can have no more than 100 shareholders, although Jeremy notes he’s never worked with an S corp that came anywhere close to that limit. The overwhelming majority have one, two, maybe three shareholders.

Shareholders must generally be individuals, though certain estates, trusts, and organizations can qualify. Jeremy warns that including an S corp interest in an estate plan can be tricky. There’s a serious risk of inadvertently terminating the election when a trust or estate steps into a deceased shareholder’s place. No shareholder can be a nonresident alien. Basically, shareholders need Social Security numbers.

The eligibility requirement that actually blows elections in practice is the single-class-of-stock rule. An S corporation cannot have shareholders with differential rights to distributions. Voting differences are fine—you can have voting and non-voting shares. However, you can’t have distribution waterfalls, preferred stock arrangements, or any structure in which some owners receive distributions before others. That’s partnership territory. Jeremy points out this issue has been litigated repeatedly in tax court and district courts, with businesses forced to choose between their own governing documents and tax law. The S election usually loses.

The check-the-box shortcut most practitioners still get wrong

Jeremy emphasizes this requirement because many practitioners misunderstand it. An LLC electing S does not file Form 8832 separately. When an LLC files Form 2553, it triggers two simultaneous deemed elections. First, classification as an association (which defaults to C corporation status), and then S corporation treatment. Both happen instantaneously. “Do not file Form 8832 to elect a C corporation first, and then file the 2553. Just file the 2553 to elect S,” Jeremy says.

Filing both confuses the situation and makes a mess. The only time you file Form 8832 for an S corporation is when the entity is revoking its S election and wants to revert to its default classification as a disregarded entity or partnership. Jeremy covers that process in episode 15, Breaking Up with Your S Corp Part Two.

Timing matters, and there’s no extension

Taxpayers must file the election by the 15th day of the third month of the taxable year to be effective for the current year. That’s March 15 for calendar-year taxpayers. Miss that date, and the IRS treats the election as effective for the following year. An election effective January 1, 2026, must be filed by March 15, 2026. File it on March 16, and you’re looking at a January 1, 2027, effective date unless you file for late relief.

Form 2553 details that trip people up

Jeremy identifies several issues drawn directly from situations his firm has handled:

EIN confusion. Electing S does not require a new EIN for an existing entity. That’s Treasury Regulation §301.6109-1(h)(1). But when a sole proprietor forms a new LLC to elect S, that new entity needs its own EIN. You cannot reuse the sole proprietor’s old payroll EIN. Jeremy describes exactly what happens when practitioners try. The IRS accepts the election but assigns a new EIN. The practitioner then files 1120-S returns under the old number. A couple of years later, the IRS sends notices about unfiled returns because nothing was filed under the EIN the IRS actually assigned.
Effective date before the entity exists. The S election effective date cannot precede the entity’s incorporation or registration date. If they formed the LLC in June, the effective date cannot be January 1. Jeremy notes that so many people made this mistake that the IRS printed a caution directly on Form 2553 itself.
Wet ink signatures only. Every signature on Form 2553, including the officer’s on page one and each shareholder’s consent on page two, must be wet ink. No e-signatures. Jeremy acknowledges it’s annoying, but his firm has a workaround: provide the form through a secure portal, instruct the client to print, sign, and scan it back using the portal’s smartphone scanner.
The shareholder consent grid. Page two requires each shareholder’s name, address, tax ID, shares owned, acquisition date, tax year end, and signature, all under a statement that reads “under penalties of perjury.” That language matters, especially for late elections, where shareholders also declare they’ve reported income consistently with S corp status for all affected years.

Even when practitioners know these rules, sometimes the deadline slips. The question then becomes whether late relief is available and whether practitioners should even pursue it.

Late election relief

Late election relief is one of the most discussed (and most misunderstood) aspects of S corp elections. Jeremy sees widespread misconceptions about the process of making a late election, and about whether practitioners should make it in the first place. Before you file anything late, you need to understand the legal framework and the IRS requirements.

The statutory authority starts with IRC §1362(b)(5), which allows the Secretary of the Treasury to treat a late election as timely when the entity has reasonable cause for missing the deadline. Treasury Regulation §301.9100-1 lets the Commissioner grant reasonable extensions for regulatory and statutory elections, and §301.9100-3 extends that to entity classification elections provided the taxpayer shows they acted reasonably and in good faith, and that granting relief won’t prejudice the government’s interests.

Over the years, the IRS issued various revenue procedures for different types of late elections. Rev. Proc. 2013-30 consolidated them into a single document that now governs late S elections, along with electing small business trusts (ESBTs), qualified subchapter S trusts (QSSTs), qualified subchapter S subsidiaries (QSubs), and late corporate classification elections.

The four requirements you must satisfy

Section 4.02 of Rev. Proc. 2013-30 lays out four criteria, and Jeremy stresses taxpayers must meet all four:

The entity intended to be classified as an S corporation as of the effective date. Jeremy calls this “the most important to really nail down.”You can prove intent through board meeting minutes, corporate resolutions, communications with a tax advisor—anything that demonstrates the entity wanted S corp status even though it didn’t file the paperwork on time. The problem is most small business owners don’t keep these records. If your client doesn’t have formal documentation, look for email exchanges with advisors, meeting notes, or other evidence that the intent existed before the deadline passed.
Request relief within three years and 75 days of the effective date. That gives you roughly three years, one month, and 15 days. This is the general window, although there is one exception, which Jeremy covers later.
The only reason the entity doesn’t qualify as an S corporation is the untimely filing. Everything else, including eligibility, ownership structure, and a single class of stock, must be in order. If there’s an underlying eligibility problem, late relief won’t fix it.
Reasonable cause for the failure, plus diligent action to correct the mistake. Jeremy notes the most common explanation is straightforward: owners simply didn’t understand the paperwork or deadlines until a tax professional advised them. There are no strict criteria for what constitutes reasonable cause, and Jeremy has seen various approaches, some successful, some not. The key is being honest and specific about what happened.

The procedural mechanics

You still use Form 2553 to request relief, but with modifications. Print “FILED PURSUANT TO REV. PROC. 2013-30” in all caps at the top of page one. Most tax software has a checkbox that handles this automatically. Include a reasonable cause statement either on the form itself (there’s blank space on the bottom half of page one) or on an attached separate sheet.

If the S corporation has filed all its 1120-S returns for tax years between the effective date and the current year, attach the completed Form 2553 to the current year’s 1120-S, as long as the taxpayer files that return within the three-year-and-75-day window. If there are delinquent 1120-S returns, file them all simultaneously. Jeremy admits this makes him uncomfortable. “I don’t feel good doing that. I don’t like filing that many returns on top of one another.” But he’s done it, and it can work.

His firm’s practice is to fax Form 2553 directly to the applicable IRS service center and attach a PDF to the e-filed return. “It can’t hurt to do it both ways,” he says. Just remember, filing Form 7004 to extend the 1120-S does not extend the deadline for the election itself. There is no mechanism to extend Form 2553.

The exception to the time limit

The three-year-and-75-day window doesn’t apply if all of the following are true:

The entity and all shareholders reported income consistent with S corp status for the year the election should have been made and every year after
At least six months have elapsed since the entity filed its return for the first year it intended to be an S corp
The IRS never notified the corporation or any shareholder of a problem within those six months.

Jeremy stresses this last point. Always make sure clients check their physical mailboxes regularly, because the IRS corresponds about S elections exclusively by mail.

If the entity can’t satisfy Rev. Proc. 2013-30’s requirements, the only remaining option is requesting a private letter ruling from the IRS. PLRs can get expensive, and they’re the last resort rather than a routine tool.

Both the officer signing Form 2553 and each consenting shareholder declare under penalties of perjury that the election is true, correct, and complete. For late elections, shareholders also declare they’ve reported income consistently with S corp status for all affected years. This is a sworn statement the IRS takes seriously.

When the S election is (and isn’t) the right call

This is where the internet’s favorite rule of thumb falls apart. The self-employment tax savings that dominate most S corp conversations are just one variable in a multi-factor analysis. Jeremy identifies several factors that can offset or even eliminate those savings.

Stop relying on rules of thumb

The typical logic goes that if you make more than a certain amount (usually some middle five-figure number), you should elect S corporation status. Jeremy calls these rules of thumb “very dangerous” because they omit critical nuance. Yes, the typical purpose of an S corporation is to replace a larger self-employment tax burden with a smaller payroll tax burden. But that single calculation ignores everything else that changes when you make the election.

Review the ownership structure and the operating agreement

S corporations don’t have the flexibility of partnerships. They don’t allow special allocations, differential distribution rights, or waterfalls. The practical problem is that LLC operating agreements are almost always written from a subchapter K (partnership) perspective, not subchapter S. The partnership language baked into those documents won’t translate well for an S corporation. It can set up owners to inadvertently terminate the election.

Jeremy taught a two-hour webinar for the New York State Society of Enrolled Agents on reviewing LLC operating agreements for non-attorneys. He strongly recommends that practitioners request and review operating agreements before recommending any S election. If you’re not already doing this, start.

Examine the balance sheet before you recommend anything

Unlike partnerships, S corporation shareholders don’t get basis for corporate debt, only for bona fide shareholder loans to the corporation. Personal guarantees don’t count. There are no recourse-versus-non-recourse debt considerations like you’d find in a partnership.

Transferring liabilities in excess of assets into the S corporation as part of the §351 exchange—the corporate transfer that happens when an LLC makes that deemed corporate election—can trigger a taxable event. Appreciated fixed assets, especially real estate, create built-in gains issues, and there’s no §754 inside basis step-up available. Jeremy published a detailed post that walks through corporate transfer accounting.

Don’t ignore what happens to the QBI deduction

Owner wages paid by an S corporation are deductible for the business, which reduces qualified business income. That reduction shrinks the §199A qualified business income deduction. Jeremy has seen cases where the QBI reduction offsets most and sometimes nearly all of the self-employment tax savings. “Essentially, it’s a wash.”

The Election Is Easy. The Decision Isn’t

The mechanics of filing Form 2553 may seem straightforward, but the decision to elect S corporation status rarely is. As Jeremy makes clear, the real work is understanding eligibility rules, avoiding procedural traps, and evaluating whether the election actually improves the client’s overall tax picture.

For a deeper walkthrough of the rules, real-world mistakes practitioners make, and the analytical framework Jeremy uses to evaluate S elections, listen to the full episode of Tax in Action.

Why Your Service Business Client Shouldn’t Have Cost of Goods Sold on Their Tax Return

Earmark Team · February 28, 2026 ·

You’re reviewing a new client’s prior-year returns when something catches your eye. The business is a consulting firm—pure services, no inventory to speak of—yet there’s cost of goods sold on Schedule C. You pull up the financial statements and find “cost of services” listed separately from other expenses. The previous preparer apparently decided consistency was the goal and carried the figure straight over to the tax return.

It’s a mistake Jeremy Wells sees all the time. In fact, he’s seen so many tax returns with this exact error that he devoted an entire episode of Tax in Action to breaking down what cost of goods sold really means for tax purposes and why getting it wrong matters more than you might think.

“I’ve seen a lot of tax returns prepared for new clients coming into my firm, where the returns were either self-prepared or prepared by another firm that reported cost of goods sold for a particular business when I knew that that business should not have reported cost of goods sold,” Jeremy explains.

You might think it all reduces taxable income anyway, so what difference does it make where the numbers land? But that reasoning misses something fundamental about what cost of goods sold actually represents in the tax code.

Only Three Types of Businesses Get Cost of Goods Sold

Treasury Regulation 1.61-3(a) tells us that only three types of businesses calculate gross income using cost of goods sold:

Manufacturing: businesses that produce goods from raw materials
Merchandising: businesses that purchase finished goods for resale
Mining: businesses that extract natural resources

If your client isn’t in one of these three categories, they don’t have cost of goods sold for tax purposes.

“No other kind of business has that formula described in terms of gross income,” Jeremy emphasizes. “Only businesses in those three categories: manufacturing, merchandising, and mining.”

This trips up practitioners because every business has what Jeremy calls “direct costs”—the expenses they must pay to generate revenue. He uses his own firm as an example. They use ProConnect tax software with a pay-per-return model, buying individual credits to file or print each client’s return. These are clearly direct costs related to serving specific clients.

But those software credits are ordinary business expenses, not cost of goods sold. Jeremy’s firm provides services, not merchandise. They don’t manufacture anything. They’re not mining. So despite having clear, traceable direct costs for each client, they don’t report cost of goods sold on their tax return.

The confusion gets worse with modern businesses that blur traditional categories. A business coach might sell one-on-one coaching (a service) while also selling digital products or online courses (potentially merchandise). A content creator might offer consulting while also selling physical products. Each revenue stream needs its own analysis.

“I’ve even had some pushback from new clients when we prepare that first tax return, where the prior returns had cost of goods sold reported, the return I prepared doesn’t, and the taxpayer actually notices and questions that,” Jeremy says.

Understanding which businesses qualify is just the start. The real insight comes from understanding why this classification matters so much.

Cost of Goods Sold Isn’t a Deduction—It’s Income Itself

Every tax professional knows the phrase, “expenses are deductible due to ‘legislative grace.’” Congress decides what deductions you can take. They can expand them, limit them, or take them away entirely.

But cost of goods sold works differently.

IRC Section 61 defines gross income as “income from whatever source derived.” For those three special categories of businesses, the regulations specify that gross income equals gross receipts minus cost of goods sold. This happens before you even think about Section 162 ordinary and necessary business expenses.

“Cost of goods sold is actually part of the definition of gross income when it comes to tax,” Jeremy explains. “It’s not just a special kind of expense.”

The courts have interpreted this to mean that cost of goods sold represents a “return of capital” rather than a tax deduction. When a store buys inventory for $50 and sells it for $100, that first $50 isn’t income; it’s just getting back the money they invested. The income is only the $50 profit.

This has real implications for what costs belong in the calculation. The basic formula is:

Beginning inventory + purchases of inventory + production costs (direct labor, freight)

– ending inventory

= cost of goods sold

Selling, general, and administrative expenses never belong in cost of goods sold, no matter how essential they are to running the business. These are always ordinary expenses.

The courts don’t care what you call things. In Atkinson v. Commissioner, a taxpayer tried to classify operating expenses as cost of goods sold. The Tax Court rejected this because the costs weren’t directly tied to inventory. As Jeremy notes, “Economic reality controls over labels used on tax returns or financial statements.”

For most businesses, this distinction is about accuracy. But there’s one area where understanding the difference between cost of goods sold and deductions becomes absolutely critical.

When Getting It Wrong Can Cost Millions: The Cannabis Example

IRC Section 280E is tough on cannabis businesses as it allows no deductions or credits for businesses trafficking in Schedule I or II controlled substances. Since marijuana remains Schedule I under federal law, dispensaries can’t deduct rent, utilities, salaries (except those directly tied to inventory), or any other ordinary business expense.

Their taxable income essentially equals their gross income. Except for cost of goods sold.

“Section 280E doesn’t disallow cost of goods sold,” Jeremy explains. “Because cost of goods sold is not an ordinary deduction; it is a reduction of gross income.”

This distinction became the center of Californians Helping to Alleviate Medical Problems (CHAMP) v. Commissioner, a 2007 Tax Court case that Jeremy calls “a really good illustration of why this concept is important.”

CHAMP operated both a medical marijuana dispensary and provided caregiving services for patients. Same business, two revenue streams, completely different tax treatment.

The IRS looked at the business and said it was trafficking in marijuana, Section 280E applies, no deductions allowed. CHAMP argued that costs of acquiring marijuana inventory were cost of goods sold that reduced gross income.

The Tax Court partially agreed. They allowed cost of goods sold, but only for costs directly tied to acquiring marijuana inventory. The dispensary’s operating costs were disallowed under 280E. It also disallowed the caregiving service costs and since caregiving is a service, those costs couldn’t be cost of goods sold anyway.

“The Tax Court allowed cost of goods sold, but only for the inventory-producing activity, only for the merchandising part of the business,” Jeremy clarifies. “Not for the caregiving services.”

This case shows having inventory isn’t enough to sweep all your costs into cost of goods sold. When a business has multiple activities, you have to analyze each one separately. And courts always look at economic substance over whatever labels you use.

Common Mistakes and How to Fix Them

Jeremy shares a frustration many practitioners face: clients who report the same inventory year after year or give suspiciously round numbers.

“We ask for their ending inventory and we get the same number as last year’s ending inventory, or we get round numbers,” he says. A restaurant claiming exactly $1,000 in beverage inventory while doing millions in revenue? “I seriously doubt that it’s an accurate reflection of their inventory.”

For sole proprietors and single-member LLCs, cost of goods sold goes on Schedule C, Part III. Corporations and S-corporations use Form 1125-A. Both forms walk through the same calculation: beginning inventory, plus purchases and production costs, minus ending inventory.

The key is educating clients about proper inventory counts and valuation. This matters for accuracy and for defending the numbers if the IRS asks questions. The substantiation requirements are the same as for any business expense. You must prove costs were incurred, properly classified, and correctly valued.

Jeremy also mentions that inventory valuation methods matter. Businesses can use First-In-First-Out (FIFO), Last-In-First-Out (LIFO), or average cost methods. But consistency is necessary. You can’t switch methods year to year just to get better results.

One final note on the future: if marijuana gets removed from Schedule I, Section 280E would no longer apply to cannabis businesses. But Jeremy cautions this would likely only affect future years. “It’s very unlikely that something like that would happen” retroactively, he explains. For now, the distinction between cost of goods sold and ordinary expenses is critical for every cannabis business.

The Bottom Line for Tax Professionals

If you take away nothing else from this episode, remember cost of goods sold belongs only to manufacturing, merchandising, and mining businesses. A consulting firm with “cost of services” is ordinary expenses. Same for the coaching business tracking direct costs.

“Just because the financial statements report cost of goods sold or cost of sales or cost of services doesn’t mean the tax return should or even can have cost of goods sold,” Jeremy emphasizes.

This isn’t about matching financial statements to tax returns. Cost of goods sold represents a return of capital invested in inventory, not just another expense category. When you get this right, you properly calculate income.

For most clients, fixing this means moving numbers from cost of goods sold to ordinary expenses. When they ask why their return looks different, you now have the framework to explain why accuracy matters more than consistency with an incorrect approach.

For cannabis clients, the stakes are much higher. Under Section 280E, properly identifying cost of goods sold might be the difference between staying in business and closing doors.

Whether you prepare returns for a local retailer or advise a multi-state dispensary, you should understand what cost of goods sold really means, know which businesses qualify, and report costs where they belong based on substance, not convenience.

To dive deeper into the regulations, court cases, and practical examples, listen to the full Tax in Action episode. Jeremy walks through each concept step by step, giving you the technical foundation to turn confusion into competency.

This $600,000 Lesson Proves You Can’t Outsource Your Filing Deadline

Earmark Team · February 17, 2026 ·

Wayne Lee, a Florida surgeon, hired CPA Kevin Walsh to prepare and file his tax returns. From 2014 through 2016, Wayne provided Kevin with all necessary documents and signed Form 8879 e-file authorizations each year. As far as Wayne knew, Kevin was doing exactly what he’d been hired to do: preparing and filing returns, each showing mid-six-figure tax liabilities but also substantial refunds due.

Wayne discovered the truth on December 5, 2018, when an IRS agent showed up at his office. Kevin had never filed any of those returns. He’d never told Wayne about a software issue that supposedly prevented e-filing. The IRS notices had piled up at an incorrect address that Kevin promised but failed to update. By the time Wayne discovered the problem, the three-year statute of limitations had expired on his $288,000 refund from 2014. Instead of rolling that refund forward as planned, Wayne ended up paying $289,000 to the IRS in 2019 to settle unpaid liabilities, penalties, and interest.

Wayne sued Kevin (settling out of court) and the federal government for a refund. The government won.

In a recent episode of Tax in Action, host Jeremy Wells, CPA, EA, uses Wayne’s case to explore a principle the Supreme Court established decades ago: you can delegate tax return preparation, but you can’t delegate responsibility for filing deadlines. Understanding this distinction and the penalties that follow when taxpayers miss deadlines is crucial for tax professionals and their clients.

How Failure to File and Failure to Pay Penalties Work

The penalties Wayne faced were calculated additions to tax that accrued interest and turned a potentially manageable balance into a serious financial burden.

The failure to file penalty under IRC Section 6651(a)(1) is the more severe of the two delinquency penalties. It’s 5% of the net amount due for each month or fraction of a month the return is late. That “fraction of a month” language carries real weight. File your return one day late, and you owe a full month’s penalty. The penalty maxes out at 25% of the net amount due, unless fraud is involved. In fraud cases, the penalty jumps to 15% per month with a 75% ceiling.

“The net amount due means the tax liability shown on the return less any withholding credits, estimated payments, or any other payments made on or before the due date,” Wells explains, emphasizing a critical point. This definition, from IRC Section 6651(b)(1), means even if you can’t file on time, making payments reduces the base amount for penalty calculations.

The failure to pay penalty under IRC Section 6651(a)(2) is gentler at 0.5% per month, but it comes with its own trap. As Wells explains, “An extension of time to file is never an extension of time to pay.” Treasury Regulation 1.6081-4(c) makes this explicit. That six-month extension gives you time to finish the paperwork, not time to find the money.

When both penalties apply, as in the case of an unfiled return with an unpaid balance, they don’t stack. The failure to file penalty gets reduced by the failure to pay amount, keeping the combined rate at 5% per month. But both count as “additions to tax,” meaning interest accrues on the penalties themselves, compounding the total amount owed over time.

The math drives the strategy. As Wells puts it, “Why pay 5% when you only have to pay 0.5%?” Always file on time, even with a balance due. Always request extensions if you need more preparation time. And make estimated payments before April 15th to reduce the net amount due that serves as the penalty base.

Partnerships and S Corporations Face Different Rules

Pass-through entities don’t pay income tax directly, so there’s no failure to pay penalty. But their failure to file penalties can be devastating.

Under IRC Sections 6698 (partnerships) and 6699 (S corporations), the penalty is an inflation-indexed amount per partner or shareholder, per month the return is late. For 2025, that’s $245 per partner or shareholder, increasing to $255 in 2026 and $260 in 2027.

Wells stresses the multiplication effect. “By definition, a partnership has at least two partners. So if you have a late 1065, then at a minimum, the penalty will be doubled.” A small partnership with four partners and a three-month late filing would owe nearly $3,000 in penalties for 2025.

The penalty amount is determined by the year the return should be filed, not the tax year of the return. So if you file a 2024 partnership return late in 2025, you use the 2025 penalty amount.

Relying on Your Tax Professional Won’t Save You

Wayne’s defense seemed reasonable. He hired a licensed CPA, signed e-file authorizations, provided all documents, and was repeatedly assured his returns were being handled. Surely that demonstrates ordinary business care?

The courts said no, following precedent from United States v. Boyle (1985). In that case, the Supreme Court reversed an appeals court that sided with a taxpayer whose attorney missed an estate tax filing deadline. The Supreme Court held that taxpayers cannot delegate filing and payment deadlines to professionals.

When Wayne’s case reached the 11th Circuit in 2023, his attorneys argued that e-filing changes this dynamic. If only the professional can electronically submit returns, doesn’t that shift responsibility? The court rejected this argument entirely. “E-filing does not change the taxpayer’s duty,” they ruled.

So what actually qualifies as reasonable cause? The IRS outlines acceptable categories in Internal Revenue Manual 20.1.1.3.2:

Death or serious illness of the taxpayer, immediate family member, or key employee within an organization (not an outside professional)
Fire, casualty, or natural disaster directly causing non-compliance
Legitimate inability to obtain essential records after reasonable efforts (poor recordkeeping doesn’t count)
Reliance on erroneous written IRS advice specifically addressing your situation
Ignorance of the law, but only if you had no prior filing requirement, tried to learn the law, and the issue was genuinely complex

Wells notes a crucial limitation: if you’re “demonstrably competent enough to carry on other transactions,” like hiring and working with a tax professional, “reasonable cause won’t work.”

The message is clear: tax professionals can prepare returns and provide advice, but the legal duty to ensure filing and payment remains with the taxpayer.

First-Time Abatement Is Your Best Shot at Relief

While reasonable cause rarely succeeds, taxpayers have another option that’s often more accessible: First-Time Abatement (FTA).

The IRS evaluates penalty relief requests in a specific order, outlined in IRM Chapter 20:

Correction of IRS error
Statutory and regulatory exceptions
Administrative waivers (including FTA)
Reasonable cause

Notice that FTA comes before reasonable cause, even though reasonable cause is statutory. The National Taxpayer Advocate has criticized this ordering, but for taxpayers, it creates opportunity.

FTA covers failure to file, failure to pay, and failure to deposit penalties, but not the underpayment penalty. To qualify, you need:

Clean penalty history for three prior years (no unreversed penalties)
All required returns filed
Tax paid or payment arrangement in place

A critical change occurred in March 2023. “If you have a taxpayer with a relatively small penalty and you don’t ask for first time abatement, and then the following year they get an even bigger penalty, first time abatement is not available because it should have been used on that first penalty,” Wells explains.

The old strategy of “saving” FTA for a potentially larger future penalty no longer works. Use it when you’re eligible, or lose it.

Timing matters for requesting FTA. You can’t abate a penalty before it’s assessed, so wait for the notice—typically a CP14 or CP162. Get Form 8821 or 2848 authorization from clients and check the box to have notices forwarded to your office. When the notice arrives, respond immediately or call the Practitioner Priority Service for faster resolution.

Good news is coming: the National Taxpayer Advocate announced that starting with tax year 2025 returns, the IRS will begin automatically applying FTA to eligible penalties. Until then, it remains a manual process.

Protecting Yourself and Your Clients

“Penalty abatement is really less about proving innocence and more about understanding the timing, the procedures and how the IRS operates,” Wells says to summarize the episode’s overarching message.

For tax professionals, this means building clear communication into every client engagement. Clients must understand that while you prepare returns, they’re still responsible for confirming filing and payment. This isn’t about avoiding responsibility; it’s about accurately representing how tax law works.

Wayne Lee did everything a reasonable person would do, yet still lost nearly $600,000. His case is a reminder that in tax compliance, good intentions and professional help aren’t enough. The responsibility to file and pay on time ultimately rests with the taxpayer, and they can’t delegate that duty.

Listen to the full episode of Tax in Action for Jeremy’s’ complete analysis of IRC Section 6651, including detailed penalty calculations and step-by-step guidance for requesting abatement.

Your Client’s Non-Cash Donation Documentation Is Probably Missing These Critical Details

Earmark Team · February 9, 2026 ·

A married couple donated used goods worth $6,760 to charity. They filed Form 8283 with their 2019 tax return. The IRS examined the return, the case ended up in Tax Court, and the entire deduction was disallowed.

What makes this case remarkable is that neither the IRS nor the Tax Court ever claimed the donation didn’t happen. Both sides agreed the contribution was real and made to a qualifying organization. The taxpayers lost anyway.

In this episode of Tax in Action, host Jeremy Wells breaks down the Besaw v. Commissioner case to reveal what went wrong. It comes down to something every tax practitioner has seen.

“When you help people with tax returns, and you ask them to upload their documents around tax time, and you get a slew of these blank Goodwill receipts,” Wells explains, “the taxpayer’s expecting some sort of tax benefit from all that. And you’re looking at that thinking, I can’t do anything with this.”

When Good Donations Go Bad

The timeline of the Besaw case reads like a warning every tax practitioner should remember.

In 2019, the Besaws filed their joint federal income tax return claiming $6,760 in non-cash charitable contributions. They attached Form 8283, the correct form for reporting non-cash donations of $500 or more. So far, so good.

But the form was basically empty. The Besaws left out the donation dates, the fair market values of the donated property, their cost basis in that property, and the method they used to arrive at the deductible amount.

“Really, all of the important information,” Wells notes.

The IRS examined the return in 2022 and requested proof for those charitable contributions. In response, the Besaws submitted what Wells calls a “reconstructed record,” essentially taking what was on their incomplete Form 8283 and using it to piece together documentation of their contributions.

The IRS said that wasn’t good enough. They denied the deduction and sent a notice of deficiency. The case wound up in Tax Court.

Contemporaneous Is The Magic Word

The fatal flaw in the Besaw case comes down to one word that appears throughout tax law: contemporaneous.

“The word contemporaneous comes up in other places in tax law as well,” Wells explains. “If you drive to meet a client or meet a vendor, you’re supposed to keep a contemporaneous written mileage log. That means you’re supposed to keep that throughout the year. You’re not supposed to look back at the prior year and try to reconstruct that from memory.”

The same principle applies to charitable contribution records. But Section 170 gives a specific definition. Documentation must be received by the earlier of the date you file your return or the due date (including extensions).

The Besaws created their records in 2022, long after filing their 2019 return and while the IRS was already examining it. That timing made all the difference.

What makes this case really striking is that nobody disputed that the donation actually happened. “At no point did the IRS or the Tax Court claim the donation may never have happened,” Wells emphasizes. “Everyone agreed that the donation happened. The issue was whether it was properly documented.”

This flows from a principle Wells returns to often. “Deductions are a matter of legislative grace.” That phrase appears throughout Tax Court decisions. It means the deduction doesn’t come free. You can’t just say you made a contribution and expect the benefit. The law sets rules for how taxpayers claim and prove deductions. Fail to meet those requirements, and the deduction disappears, even if the donation was real.

The Four Levels of Documentation Requirements

Understanding what documentation you need depends on how much you’re donating. The rules get stricter as the amounts go up.

Level 1: Non-Cash Contributions Under $250

For donations under $250, Treasury Regulation 1.170A-13(b)(1) requires a receipt showing:

The name of the charity
A description of the property
The date and location of the contribution

This is where those blank receipts from donation drop-offs become a problem. “A lot of times those slips of paper wind up in our client portal without much more than whatever is preprinted on them by the organization, which is usually just the name of the organization and its location,” Wells observes. “There’s no other information there.”

The receipt doesn’t need to include a value—that’s the taxpayer’s job. But it must describe what the taxpayer gave.

Level 2: Non-Cash Contributions Over $250

Cross the $250 threshold, and you need a contemporaneous written acknowledgment from the organization. This must include:

A description of the property
A statement of whether any goods or services were provided in exchange, and a value or good-faith estimate of those goods or services

“This is actually where the taxpayer failed in this Tax Court case,” Wells points out, “because the charity receipts contained no descriptions of the donated items.”

Remember, the acknowledgment must be received by the earlier of the filing date or the due date, including extensions. Miss that window, and you can’t fix it later.

Level 3: Total Non-Cash Contributions Over $500

When total non-cash contributions exceed $500, you must file Form 8283. Treasury Regulation 1.170A-13(b)(3) says you need to report:

The date you acquired the property
How you acquired it (i.e., purchase, gift, or inheritance)
Your cost or basis in the property
The fair market value at the time of donation
The method used to determine the fair market value

“The taxpayer in this Tax Court case left all these blank,” Wells notes. “And that was problematic for him.”

Level 4: Single Items or Groups Over $5,000

When any single item or group of similar items exceeds $5,000, IRC Section 170(f)(11)(C) requires a qualified appraisal. The taxpayer, the charity, and the appraiser all must sign Section B of Form 8283.

Wells has seen how this plays out. “I’ve had situations where taxpayers wanted to claim a deduction in excess of $5,000, and once we explain the appraisal requirement, either they didn’t want to go through with the appraisal, or they adjusted their valuation of that contribution to an amount below $5,000.”

What You Need vs. What You File

The charity doesn’t provide valuations on receipts. “It’s up to the taxpayer to report and to calculate a fair market value of those contributed items,” Wells explains. “It’s not up to the organization to put that together.”

But descriptions must be detailed enough to identify the items and estimate value. You can’t write “trunk full of stuff.” But you don’t need to list every single piece of clothing either.

Beyond what gets filed with the return, Treasury Regulation 1.170A-13 requires taxpayers to keep:

A detailed list of donated property
Receipts showing dates, locations, and descriptions
Fair market value worksheets showing how they calculated values
Any appraisals, if required

This is what the IRS requested in 2022, and what the Besaws couldn’t produce in its original form. Form 8283 is only part of the package.

Dealing with Those Blank Goodwill Receipts

“When you see those blank Goodwill receipts pop up in your client portal, how will you address those?” Wells asks. It’s a question every practitioner faces each tax season.

When receipts say something like “household goods, used clothing $1,000,” is that enough? Based on the Besaw case and the regulations, “Probably not,” Wells says. “We need a little bit more information than that.”

When non-cash charitable contributions appear, Wells recommends asking these questions:

What exactly did you donate? Get actual descriptions, not just “clothing”
When and where did they donate? Dates and locations matter
Is there a receipt or written acknowledgment? If not, do they have their own records?
Does the record describe the property adequately? “Stuff” won’t cut it
Did they receive anything in return? This affects the deductible amount
Do total non-cash contributions exceed $500? This triggers Form 8283
Does any single item exceed $5,000? This may require an appraisal
How did they determine the value? They need an actual method

Getting Real About Fair Market Value

Treasury Regulation 1.170A-1(c)(2) defines fair market value as “the price a willing, knowledgeable buyer would pay a willing, knowledgeable seller under no compulsion.”

For clothing and household goods, this usually means thrift shop value. Wells recommends a practical test. “If you didn’t know anything about this t-shirt other than you saw it hanging on a rack in a thrift shop and you liked it, what price would you reasonably pay for that?”

One common problem is sentimental value. “Just because an item holds some sentimental value for you or your family doesn’t necessarily mean it has a higher fair market value,” Wells cautions. “Sentimental value doesn’t really translate into actual market value.”

There’s also a quality requirement. Under Section 170(f)(16), clothing and household goods must be in “at least good used condition” to qualify. Worn-out or broken items are not deductible at all.

Four Myths That Create Problems

The Besaw case debunks several common assumptions:

Myth 1: A signed receipt is enough

In reality, the receipt must describe the property and include other details.

Myth 2: You can fix missing details later

This killed the Besaws’ deduction. Documentation must be contemporaneous.

Myth 3: Fair market value is optional for small donations

Form 8283 requires fair market value for every non-cash donation over $500.

Myth 4: If the donation happened, the IRS should allow it

The Besaw case proves otherwise. Both sides agreed that the donation occurred. But the Tax Court still disallowed the deduction.

The Bottom Line

The Besaw case demonstrates that, in tax law, documentation beats facts. A legitimate charitable donation means nothing without proper contemporaneous records.

Tax practitioners must remember:

Contemporaneous means by the filing date or due date (including extensions)
Different amounts trigger different requirements
Form 8283 isn’t the whole story. Taxpayers must keep detailed records
Those blank Goodwill receipts are audit risks waiting to happen

“This Tax Court case illustrates that substantiation is everything,” Wells concludes. “It really is in a lot of areas of tax law, and especially here in terms of charitable contributions.”

The charitable deduction can be powerful, but it comes with serious technical requirements. Listen to the full episode and make sure your clients understand these rules before they drop off that next load at Goodwill, and definitely before they try to claim the deduction.