Employee vs. Independent Contractor in 2026: The Tests Every Startup Must Pass Before Classification

Why Classification Got Harder in 2024 — And Where Things Stand Now

For most of the 2010s, the federal standard for determining whether a worker was an employee or an independent contractor was unsettled but at least predictable in its instability. That changed sharply in January 2024, when the U.S. Department of Labor issued a final rule (RIN 1235-AA43) reinstating a six-factor economic reality test — effective March 11, 2024 — that replaced the Trump administration's 2021 two-factor rule. The 2021 rule had made contractor status easier to establish by centering the analysis on just two factors: control over the work and opportunity for profit or loss. The 2024 rule restored a totality-of-the-circumstances framework that weighs all six economic reality factors, making contractor status considerably harder to defend.

The 2024 rule's reign as enforcement policy was short. In May 2025, the DOL issued Field Assistance Bulletin 2025-1, directing its Wage and Hour Division to stop enforcing the 2024 rule and revert to the pre-2024 Fact Sheet #13 framework. That bulletin created a bifurcated landscape that still governs today: federal agency enforcement follows the older standard, but the 2024 six-factor rule remains technically in effect — and available to plaintiffs' counsel in private FLSA litigation and collective actions. A startup that passes muster under Fact Sheet #13 is not necessarily safe from a class action invoking the 2024 test.

The Trump DOL moved to formalize the rollback in February 2026, publishing a proposed rule (RIN 1235-AA46) to officially rescind the 2024 rule and restore a framework centered on the same two factors — control and opportunity for profit or loss — that defined the 2021 rule. The comment period closed in spring 2026. As of mid-2026, the rule is not finalized, which means the dual-track risk described above is not going away soon.

On the labor relations side, a parallel rollback has already landed. The NLRB's October 2023 joint employer rule — which would have broadly expanded the circumstances under which a business could be held liable for the labor practices of its contractors and staffing agencies — was vacated by a federal court and formally withdrawn by the NLRB in February 2026. The narrower 2020 standard now applies, requiring substantial direct and immediate control over essential terms of employment before joint employer liability attaches. For startups using staffing agencies or contractor platforms, that withdrawal is meaningful relief — but it does not touch FLSA classification risk at all.

The DOL's Six-Factor Economic Reality Test

The 2024 rule restored a framework courts and the DOL applied for decades before the 2021 two-factor shortcut: the economic reality test. It asks a single overarching question — economically dependent on the hiring party, or in business for yourself? Six factors inform the answer, and no single factor is automatically determinative. The analysis is a totality-of-the-circumstances inquiry, which means a contractor who looks good on four factors but terrible on two can still be reclassified. Startups that treat any one factor as a trump card are building a false sense of security.

Agency enforcement has reverted to the pre-2024 Fact Sheet #13 framework under FAB 2025-1, so the DOL's Wage and Hour Division is not currently auditing against the six-factor test. But the 2024 rule remains in effect as law — and plaintiffs in private FLSA collective actions can invoke it. Each factor below is worth understanding on its own terms, because each is a potential litigation theory.

Factor 1: Opportunity for Profit or Loss

This factor focuses on managerial skill, not output volume. A developer who earns more by billing more hours is not exercising the kind of entrepreneurial control this factor targets — they're just working more. The question is whether the contractor's earnings go up or down based on their own business decisions: how they price engagements, whether they take on multiple clients, whether they invest in their own capabilities. Fixed hourly or per-job rates, without the ability to negotiate or walk away, point toward employee status on this factor.

Factor 2: Investments

The investments factor compares the relative capital commitment of the worker versus the business — but type of investment matters more than dollar amounts. Hardware, software licenses, or platform accounts that the company requires the contractor to use do not count toward contractor status, because those aren't entrepreneurial investments; they're conditions of engagement. Contractors who make their own capital decisions — buying equipment they use across multiple clients, maintaining their own toolchain — fare better here. A game artist who works on a company-issued machine using company-licensed software is essentially operating on the employer's infrastructure.

Factor 3: Permanence of the Relationship

Continuous, indefinite, or exclusive working arrangements look like employment. Project-based or sporadic engagements look like contracting. A software engineer who has worked for one startup for eighteen months without a defined end date is exposed on this factor regardless of what the contract says. Single-client exclusivity amplifies the problem — exclusivity is a hallmark of employment, not independent contracting — and longer tenure makes it harder to argue the engagement is project-scoped rather than ongoing.

Factor 4: Degree of Control

Not all control is equal under the 2024 rule. Control exercised for legal compliance — requiring contractors to follow OSHA standards, data security protocols, or non-discrimination policies — generally does not indicate employee status. Control exercised for operational convenience does. Mandatory Slack availability windows, required attendance at daily sprint standups, specified work hours, and approval chains for how work is performed are behavioral controls that push this factor toward employee status. If a startup manages how the work gets done, not just what needs to be delivered, that distinction matters.

Factor 5: Integral to the Business

This is the factor that most frequently surprises tech founders. If the contractor's work is central to the company's core product or revenue stream, it is likely integral to the business — and integral work is characteristic of employment. Contract engineers building the primary codebase of a SaaS product, game artists creating shipped assets, or EdTech developers building the content delivery platform are not peripheral contributors. They are doing the thing the company sells. The more central the work is to what the company offers customers, the harder it is to defend contractor status on this factor.

Factor 6: Skill and Initiative

Technical expertise alone does not make someone a contractor. A highly skilled engineer who works exclusively for one company, at that company's direction, on that company's roadmap is not automatically a contractor because of their specialization. The relevant question is whether they deploy that skill through their own business-like initiative — deciding which clients to pursue, how to structure their services, what to charge, and how to differentiate their offering in the market. A contractor who is highly skilled but functions as a directed member of the team is still, under this factor, exhibiting employee behavior.

The IRS Common Law Test — And Why It's a Different Problem

The IRS doesn't use the DOL's economic reality test. It applies a three-category common law framework that asks distinct questions — and reaches its own conclusions, independently of whatever the Department of Labor might say about the same worker. The categories are behavioral control, financial control, and type of relationship. A worker can clear all three DOL factors as a contractor and still fail every IRS category as an employee.

Behavioral control is where most tech and game studio arrangements break down. The IRS isn't asking whether you control the outcome — it's asking whether you control the method and means of getting there. Providing a contractor with onboarding documentation on how to perform their tasks, requiring them to use internal tools in a prescribed sequence, or scheduling mandatory workflow training all register as employee indicators. The right to control — not whether you actually exercise it — is what the IRS weighs.

Financial control looks at whether the worker operates like an independent business. IRS Publication 15-A (2026) identifies the key markers: payment method (hourly or salaried wages suggest employment; project-based payment suggests contracting), whether the worker can realize profit or loss from the engagement, and whether they carry unreimbursed business expenses. A contractor who invoices per deliverable, absorbs their own equipment costs, and works for multiple clients simultaneously looks very different from one who receives a consistent monthly retainer with all tools provided.

The third category — type of relationship — examines written contracts, employee benefits, and whether the parties intended a continuing relationship. The IRS also looks at whether the work is central to your business — whether the relationship is ongoing and the services are at the core of what you offer. A long-term engagement covering work integral to your product, even if papered as a contractor relationship, tilts toward employment under this prong. Note that these are IRS concepts, not the DOL's enumerated permanence and integral-work factors — the two tests ask similar questions but evaluate them differently.

The practical upshot: running only a DOL checklist before classifying workers leaves a gap. The DOL and IRS analyses are parallel tracks, not one test with two names. An EdTech platform that correctly identifies its curriculum designers as contractors under the economic reality test — perhaps because they set their own hours and work for other clients — can still face IRS reclassification if those same designers received company training on how to structure lesson content using proprietary internal frameworks.

What Texas Adds: The TWC Standard

Federal tests are not the only classification gauntlet a Texas startup must run. The Texas Workforce Commission applies its own direction-or-control standard under the Texas Unemployment Compensation Act — and it begins from the opposite presumption: the worker is an employee until you prove otherwise. The burden of rebuttal falls entirely on the company.

The TWC's twenty-factor analysis, adapted from older IRS guidance, turns on one especially dangerous principle: the right to control a worker's methods is sufficient to establish employment status — actual exercise of that control is not required. A startup that contracts for a developer's output but reserves the right to dictate how the work gets done is exposed under the TWC standard even if it never picks up the phone to issue a single instruction. Standard consulting agreements that include scope-of-work oversight language, deliverable approval rights, or revision mandates can quietly preserve that reserved-right problem.

Texas did carve out a narrow exception for "marketplace contractors" under 40 T.A.C. § 815.134(b), introduced in 2019 with the gig economy in mind. The carve-out applies only to UI tax liability and requires the platform to satisfy a precise three-part definitional test plus nine specific work-relationship criteria. Product companies and studios with direct contractor relationships — the typical startup model — will not qualify. The exception was designed for platform-based marketplace operators that match independent workers to customers through a digital platform — not for a game studio that retains specific freelance animators for defined project work.

The tax math sharpens the risk. For 2026, Texas UI taxes apply to the first $9,000 of each employee's wages, at rates starting at 2.70% for new employers and rising with experience ratings. A misclassification finding across even a modest contractor workforce — say, ten contractors at $80,000 each — produces immediate retroactive liability on the full $9,000 taxable wage base per worker, plus potential penalties and interest. Classification gets expensive fast.

Where Misclassification Risk Is Highest for Startups

Not all contractor arrangements carry equal legal exposure. The DOL's enforcement guidance identifies three structural patterns that trigger heightened scrutiny under both the Fact Sheet #13 framework and the six-factor economic reality test: workers who perform services exclusively or primarily for one client, workers who follow a set schedule or are paid hourly, and workers who rely on company-provided tools and equipment. When two or more of these overlap — which is common in startup workforces — the risk compounds.

Games, EdTech, and SaaS: The Highest-Exposure Sectors

Game studios routinely engage contract artists, composers, and musicians on project-based terms — but when those workers are assigned tasks through internal project management systems, attend required standups, and use company-licensed software, the practical arrangement looks less like a freelance engagement and more like employment. EdTech operators face similar exposure with contracted tutors, curriculum developers, and data annotators. The December 2024 Scale AI class action is the clearest recent example: workers performing AI data labeling who lacked control over their assignments, payment rates, task subject matter, or deadlines — and were required to follow company policies and use specific software — sued on the theory that the economic reality of their work made them employees.

SaaS companies are not insulated from this risk simply because their contractors are engineers rather than gig workers. A nationwide technology customer-support company learned this when the DOL found that 22,000 service workers classified as independent contractors were actually employees — the company paid a $3 million settlement. The common thread: the work was repetitive, company-directed, and performed under ongoing supervision. High-volume contractor workforces doing structured, company-defined tasks at scale are the defining risk pattern in tech misclassification enforcement.

Startups using contractors on core engineering — particularly those where the contractor works exclusively on the company's product, follows internal sprint cycles, and uses company accounts and infrastructure — are in materially the same position as that customer-support company.

Exposure and What to Do About It

The financial stakes of misclassification are concrete and computable — which means you can model your exposure before a regulator does it for you. Under the FLSA, a startup found to have willfully misclassified workers faces up to three years of back wages; non-willful violations carry a two-year lookback. Historically, liquidated damages doubled the bill by matching the back-pay amount dollar-for-dollar. That calculus shifted in June 2025: DOL Field Assistance Bulletin 2025-3 directs DOL investigators not to seek liquidated damages in pre-litigation administrative settlements — reserving that remedy for court actions. Private plaintiffs bringing FLSA collective actions still can seek liquidated damages in court, so the policy change narrows DOL administrative exposure, not litigation exposure.

The IRS calculates its own tab independently. Under Section 3509(a) — which applies when the employer at least filed the required Form 1099 — the aggregate effective rate runs roughly 10.68% on wages up to the Social Security wage base and 3.24% on wages above it. Section 3509(b) applies if the employer failed to file 1099s, and it roughly doubles both rates. File your 1099s. On top of those rates, willful misclassification can trigger additional FLSA penalties of 20% of wages plus 100% of both employer and employee FICA — and potentially criminal fines up to $1,000 per misclassified worker, with imprisonment possible in the most egregious cases.

The IRS offers a structured off-ramp for startups willing to act before an audit begins: the Voluntary Classification Settlement Program (VCSP). Accepted applicants reclassify workers prospectively as employees and pay only 10% of the Section 3509(a) liability for the most recent tax year — with no lookback audit for prior years. For a company carrying significant contractor headcount, that discount is substantial.

Actionable Next Steps

• Run a classification audit now. Map every contractor relationship against the economic reality test (federal) and the TWC direction-or-control test (Texas). Flag any worker where multiple factors point toward employment.
• Verify your 1099 filing history. Section 3509(b) rates — roughly double — apply if required 1099s weren't filed. Pull records for the past three years before deciding whether VCSP is available to you.
• Model your exposure before a regulator does. Use the three-year willful lookback as your worst-case FLSA scenario and the Section 3509(a) rates as your IRS baseline. If the number is material, reclassification is a business decision, not just a legal one.
• Evaluate VCSP eligibility immediately. If you've maintained consistent contractor treatment and clean 1099 records for three years — and you're not currently under audit — VCSP lets you correct course at 10% of the reduced Section 3509(a) rate for one year — a fraction of what a full audit assessment would cost — with no lookback. The window closes the moment an audit opens.
• Document the classification decision for every new engagement. A written classification memo citing the applicable test factors — completed at the time of engagement, not retroactively — is your first line of defense in an audit and demonstrates good-faith non-willfulness if a dispute arises.
• Review your contractor agreements for control indicators. Behavioral control clauses (set hours, required tools, exclusivity, non-competes) undermine independent contractor status regardless of what the contract is titled. Agreements should reflect economic reality, not manufacture it.