The Classification Line: When the Same Worker Is Both an Employee and a Contractor

A delivery driver works for a platform in New Jersey. Under the IRS common-law test she may be a contractor. Under the federal wage and hour standard the answer depends on which version of a rule that has changed three times in three years applies. Under New Jersey's ABC test she is almost certainly an employee. All three answers can be correct simultaneously, because they answer different legal questions asked by different authorities for different purposes. That's not a loophole or an oversight — it's the actual architecture, and it means a business can be fully compliant federally and substantially exposed at the state level. This report examines the tests, the shifting federal standard, and what the contested boundary costs the workers standing on it.

The core thesis

Our thesis is that worker classification is best understood not as a question with a right answer but as a boundary that several legal systems draw independently, in different places, for reasons specific to each. The confusion in public discussion comes from treating it as a single factual determination — is this person an employee? — when the operative question is always an employee for what purpose, under which statute, enforced by whom?

Three consequences follow. First, a business cannot achieve classification certainty, only defensible positions under each applicable framework — which is why the compliance answer is documentation and structure rather than a single correct label.

Second, federal movement matters less than it appears. The federal wage and hour standard has shifted repeatedly since 2024, and each shift generates headlines. But it doesn't change state tests, doesn't change the tax standard, and doesn't change the standard applied in private litigation the same way it changes enforcement policy. A business in a strict ABC state that relaxes its practices because the federal standard loosened has misread which authority actually governs its exposure.

Third — and this is where it connects to the rest of this desk's work — the classification determines far more than employment protections. It determines how income appears, whether taxes are withheld, whether the earnings are verifiable, and therefore whether the worker can be underwritten by conventional means. The contested boundary is also a boundary in the financial system, and workers on the contractor side face the documentation and volatility problems our income volatility analysis describes.

The question is never "is this person an employee." It's "an employee for what purpose, under which statute, enforced by whom" — and the answers routinely differ.

Three tests, three answers

	IRS common-law	Federal wage and hour	State ABC
Governs	Tax withholding and employment taxes	Minimum wage, overtime, and related federal protections	State wage, unemployment, and workers compensation law
Core question	Who controls how the work is done	Whether the worker is economically dependent on the business or in business for themselves	Whether all three statutory conditions are met
Structure	Multifactor, weighing behavioral control, financial control, and relationship type	Multifactor totality of circumstances	Three conditions, all required
Presumption	Neutral, fact-dependent	Neutral, fact-dependent	Employee unless the business proves otherwise
Difficulty for the business	Moderate	Moderate and shifting	Substantial

The structural difference that matters most: multifactor tests weigh, ABC tests gate. Under a multifactor analysis, a business can be strong on several factors and weak on others and still prevail on the totality. Under an ABC test, failing any one of three conditions ends the inquiry — and the business bears the burden of establishing all three.

Two further points that get lost. The federal wage and hour standard does not redefine contractor status under the tax code, the labor relations framework, or other federal statutes — a rule changing the wage and hour test leaves those untouched. And a business operating in multiple states faces multiple state answers simultaneously, which is the fragmentation our regulatory map documents appearing in employment law.

The ABC test and why B decides everything

The ABC test presumes employment unless the hiring entity establishes all three:

• A — Freedom from control. The worker is free from the hiring entity's control and direction in performing the work, in fact and under the contract.
• B — Outside the usual course of business. The work performed is outside the usual course of the hiring entity's business.
• C — Independently established trade. The worker is customarily engaged in an independently established trade, occupation, or business of the same nature as the work performed.

Prong B is where most cases are decided, and understanding why explains most classification litigation.

Consider the standard illustration: a retail store hiring a plumber to fix a bathroom passes B easily — plumbing is not what the store does. The same store engaging someone to work the sales floor fails B, because selling is precisely the store's usual course of business. No amount of contractual language, scheduling flexibility, or worker preference cures a B failure. If the person does the thing the company sells, the analysis generally ends there.

This is why ABC states are so consequential for platform and gig models. A company whose business is arranging deliveries and whose workers make deliveries has a structural problem with prong B that documentation cannot solve — which is the analytical core of the litigation and legislation in this area over the past decade.

Two qualifications worth stating. Exemptions are extensive and specific. States adopting ABC tests have carved out numerous occupations and relationship types, and those carve-outs are continually revised — California, for instance, has amended its exemption structure repeatedly, with further modifications effective at the start of 2026. And some states apply ABC only for certain purposes, such as unemployment insurance, while using a different test elsewhere — so "an ABC state" can mean several different things.

The federal standard in motion

The sequence, because the direction has reversed more than once and the current position is genuinely unsettled:

Date	Development
January 10, 2024	Department of Labor issues a final rule adopting a multifactor economic reality test; effective March 11, 2024. The rule explicitly declines to adopt an ABC test, relying instead on the long-standing totality-of-circumstances analysis used by courts
2024 onward	The rule becomes the subject of multiple legal challenges
May 2025	The Department issues field guidance directing investigators to stop applying the 2024 rule in enforcement, and to analyze status under an earlier framework instead
Through 2025–26	The 2024 rule remains relevant to private litigation over status under the federal wage statute even as the Department has ceased applying it in its own enforcement
February 26, 2026	The Department publishes a Notice of Proposed Rulemaking to revise the classification standard across the federal wage, family leave, and agricultural worker statutes; published in the Federal Register February 27
April 28, 2026	Comment period closes; no final rule or effective date at the time of writing

Two features deserve emphasis. The Department expressly rejected the ABC test in this rulemaking cycle, characterizing it as too restrictive and incompatible with the interpretation of the federal wage statute — which means the federal-state divergence is deliberate rather than accidental, and unlikely to close from the federal side.

And enforcement policy and private litigation have diverged. A business can face a claim brought privately under a framework the enforcing agency has stopped applying. This is an unusual and uncomfortable position, and it's the specific reason "the federal rule changed" is not a safe basis for changing practice.

Where states have gone the other way

While the federal standard has moved toward flexibility, several states have moved decisively the other direction — and the state movement is where the enforcement risk concentrates.

New Jersey is the clearest recent example. The state labor department adopted regulations codifying its ABC test in May 2026, scheduled to take effect October 1, 2026. The significance is procedural rather than substantive: the test already applied through case law, and codifying it into the administrative code converts a body of scattered precedent into an explicit playbook for auditors. Administrative penalties run up to defined per-worker amounts that escalate for repeat offenses, and the state retains authority to issue stop-work orders halting operations until liabilities are satisfied — which is the enforcement tool with the most operational consequence, because it doesn't wait for litigation.

California maintains an ABC test with an extensive and frequently amended exemption structure, alongside a separate voter-approved framework for certain app-based transportation and delivery work that was upheld by the state's supreme court in 2024. The result is a layered regime where the applicable test depends on the occupation.

Massachusetts applies one of the strictest formulations in the country.

The practical instruction that follows: federal relaxation does not reduce state exposure, and in the states with the most workers it barely matters at all. A business assessing its position by reading federal developments is assessing the wrong jurisdiction. This is the same lesson the preemption fights in our medical debt analysis and our earned wage access report keep producing — federal and state movement in opposite directions, with the stricter one governing conduct.

The penalty stack

Misclassification exposure is expensive because it accumulates across several bodies of law at once, and the liability generally attaches regardless of intent and regardless of whether the worker agreed.

What can stack:

• Unpaid employment taxes — the employer share of payroll taxes plus amounts that should have been withheld, with penalties and interest. This connects to the trust fund exposure our payroll guide describes, which can reach responsible individuals personally.
• Information reporting penalties for filing the wrong forms.
• Unemployment insurance contributions, retroactively.
• Back wages and overtime, with potential liquidated damages doubling the amount in some circumstances.
• Workers compensation premiums, plus exposure for any injury that occurred while the worker was uninsured — which is the single largest catastrophic risk in the stack.
• Benefit plan liabilities, including retroactive eligibility claims.
• Health coverage exposure where reclassification pushes headcount past applicable thresholds.
• State civil penalties, which in aggressive states run to substantial per-worker amounts and escalate for repeat findings.
• Stop-work orders where authorized.
• Private litigation, including collective and representative actions.

Two features make this worse than the list suggests. It's retroactive, typically across multiple years, so a classification decision made once compounds every pay period until corrected. And it's discovered in clusters — an unemployment claim by one worker, or a workers compensation injury, triggers an audit that examines everyone similarly situated. The exposure is rarely one worker; it's a category.

What the classification is worth

The financial gap explains why the boundary is contested rather than simply clarified.

What a business avoids by classifying someone as a contractor: the employer share of payroll taxes, unemployment insurance contributions, workers compensation premiums, benefits, paid leave obligations, overtime exposure, and the administrative cost of payroll — which together add a meaningful percentage on top of the cash compensation.

What the worker bears instead: self-employment tax covering both halves of the payroll tax, no employer benefit contribution, no unemployment eligibility in most circumstances, no workers compensation coverage, no overtime protection, and the quarterly estimated tax obligations our gig tax guide covers.

The honest complication, and the reason this isn't a simple exploitation story: many workers genuinely prefer contractor status, for schedule control, multiple clients, tax treatment of business expenses, and autonomy. That preference is real and the legal tests largely don't credit it — under an ABC test, mutual preference is irrelevant if prong B fails.

Which produces the genuine policy tension. A test strict enough to prevent misclassification of workers who should be employees also reclassifies workers who wanted independence — and the responses to that tension have been occupational exemptions and sector-specific frameworks, which is how the current patchwork got built. Nobody designed this; it accumulated as the resolution of that tension case by case.

The credit access consequence

The dimension this desk is best positioned to add, and one absent from most classification coverage.

Classification determines how income appears to a lender, and the difference is substantial:

• An employee has withheld taxes, regular deposits from a consistent payer, and payroll records verifiable through the source connections our verification guide describes. Income is easy to establish.
• A contractor has gross receipts from varying payers, no withholding, and qualifying income assessed net of business expenses — typically averaged across two years. The tension our tax guide identifies applies directly: expenses that reduce the tax bill also reduce verifiable income.

The consequences compound:

• Mortgage qualification is harder, requiring longer history and more documentation, per our readiness guide.
• Income volatility raises the buffer a household needs while making it harder to accumulate — the dynamic in our savings analysis.
• No unemployment insurance means an income interruption has no institutional cushion, which is what drives the products in our paycheck advance report.
• Cash flow underwriting helps and is the most direct remedy — observing deposits over time captures contractor income in a way document-based verification doesn't, per our cash flow guide.

Stated plainly: the classification line is also a line in the credit system. Workers on the contractor side face measurably harder access to credit, housing, and financial products — not because they earn less, but because the financial system's verification infrastructure was built around payroll.

What businesses and workers should do

For businesses engaging contractors:

1. Identify every applicable test — federal wage and hour, tax, and the law of every state where workers perform.
2. Assume the strictest applicable test governs your practice, since compliance with the loosest protects you from nothing.
3. Examine prong B honestly. If the worker performs your core service, no contract fixes it — and that finding should drive the decision rather than be argued around.
4. Match documentation to reality. A contract describing independence while practice shows control is worse than no contract, because it evidences awareness of the standard.
5. Audit periodically. Relationships drift toward employment over time as a contractor becomes integrated, so a defensible classification at engagement can become indefensible without anyone deciding anything.
6. Get counsel in ABC states and for any significant contractor population — this is a domain where the cost of advice is trivial against the exposure.
7. Consider correcting proactively rather than waiting for an audit, since voluntary correction programs exist in some contexts and the arithmetic of retroactive liability worsens every period.

For workers:

• Understand what you're not getting — no unemployment eligibility, no workers compensation, no overtime, no employer tax contribution.
• Price it in. Contractor compensation should exceed equivalent employee wages by enough to cover self-employment tax and the absent benefits, and frequently doesn't.
• Set aside for taxes from the first payment, since nothing is withheld.
• Keep records that establish income for lending purposes, and route everything through a bank account so it's verifiable.
• Know that agreement doesn't settle it. Signing a contractor agreement doesn't make you one if the legal test says otherwise, and you may have rights regardless.
• Build the credit file deliberately, because the verification disadvantage is real and a documented file partially offsets it.

Scenarios and what we're watching

Scenario	Shape of the world	Signposts
Base case — divergence persists	Federal standard loosens while strict states hold or tighten; multistate businesses comply with the strictest; the patchwork deepens	Final federal rule; state enactments and regulations; enforcement volumes
Third-category case	Sector-specific or portable-benefit frameworks expand, creating intermediate status with some protections without full employment	State portable benefit legislation; sector carve-outs; platform-specific frameworks
Consolidation case	Litigation or legislation narrows the gap between tests, reducing the possibility of contradictory simultaneous answers	Appellate decisions; federal preemption arguments; test convergence

What we're watching: the final federal rule and its treatment in private litigation, since enforcement policy and litigable standard have diverged; state codification and enforcement, particularly stop-work authority, which is the tool that changes behavior fastest; occupational exemption revisions, which is where the genuine policy work happens quietly; and portable benefit experiments, which are the only proposals that address the underlying tension rather than relitigating the boundary.

A worker can be an employee and a contractor at the same time, in the same job, on the same day. Until the tests converge, that's not a paradox to resolve — it's the operating condition.

Frequently asked questions

---

This report is for general information only and does not constitute legal or tax advice. Worker classification standards differ by statute and jurisdiction, the federal position described is subject to pending rulemaking and litigation, and state requirements vary substantially; consult qualified employment counsel and a tax professional about specific classifications.