Contractor Tenure Limits: Does the 18-Month Rule Still Make Sense?

Last updated: September 12, 2026

By Robert Ardell, Co-Founder and Strategic Advisor, KORE1

No federal or state law caps contractor tenure at 18 months. The rule is a private risk-management convention, and the court case that inspired it was lost over benefit plan language and IRS control tests, not over calendar time.

Somebody at your company can probably recite the tenure cap from memory. Eighteen months, usually. Twelve at some shops, twenty-four at the more relaxed ones. Ask that same person which statute it comes from and the room gets quiet.

No such statute exists. Not federally, not in any state code, and not in the IRS guidance most people assume it hides in.

The eighteen-month rule is inherited furniture. It arrived in the early 2000s, it made sense against the legal backdrop of that moment, and most companies have never reopened the question since. Then it calcified. Meanwhile the backdrop moved twice in 2026 alone. Our contract staffing desk gets asked about this every few weeks, almost always by a manager who has just been told they have to release someone productive because a date arrived.

Where the Eighteen-Month Rule Actually Came From

Blame Redmond, mostly.

Through the 1990s Microsoft ran a large population of long-term temporary workers, internally nicknamed permatemps, who sat alongside badge employees doing similar work for years. The IRS examined the arrangement and concluded many of them met the common-law definition of employee. A class action followed. In Vizcaino v. Microsoft Corp., the Ninth Circuit ruled in 1997 that these were common-law employees, and that signing a contractor agreement had not waived their claim on the company benefit plans. The stock purchase plan included. Every one of them had signed that agreement. The signatures bought Microsoft nothing. Microsoft settled in December 2000 for $97 million.

Here is the part that got lost in transmission. Microsoft’s own response was not an eighteen-month rule. It capped temporary assignments at 364 days and required a 100-day break before a worker could return. Somewhere between that policy and the trade press, 364 days drifted into twelve months, twelve months drifted into eighteen, and a specific company’s specific reaction to a specific tax examination hardened into an industry-wide number that nobody can source.

The lesson everyone took away was about duration. The actual holding was about control and paperwork. Different problem entirely.

What the Law Actually Says About Contractor Tenure Limits

A contractor tenure limit is a company policy that caps how long an individual contract worker may stay on a single assignment before being rotated off, usually with a mandatory break before they can return. It is a voluntary internal control. No statute, regulation, or agency guidance requires one.

The IRS worker classification guidance sets out three categories of evidence: behavioral control, financial control, and the type of relationship between the parties. Read it end to end and you will not find a maximum engagement length, because there isn’t one. Duration is one fact among many that can suggest an ongoing employment relationship. It is not a threshold, and clearing it buys you nothing on its own.

Which means a company can rotate a contractor off at seventeen months and still lose a classification argument badly, if that contractor was managed like an employee the whole time. Somebody else can sit in a genuinely independent, well-documented contract role for three years without creating the exposure the cap was built to prevent.

Two Federal Standards Moved in 2026, Both Toward the Employer

This is the part that makes 2026 a reasonable year to reopen the policy. Two standards shifted. Neither went the direction the cap assumes.

The National Labor Relations Board’s 2023 joint-employer rule would have swept in companies that held indirect or unexercised control over a staffing firm’s workers. It never took effect. A federal judge in the Eastern District of Texas vacated it in March 2024, and in February 2026 the Board formally removed the 2023 language from its regulations, restoring the narrower prior standard that turns on substantial direct and immediate control.

Separately, the Department of Labor issued a notice of proposed rulemaking on February 26, 2026 to rescind the 2024 independent contractor rule and restore an economic reality test weighted toward two core factors, control over the work and opportunity for profit or loss. The comment period closed April 28, 2026. No final rule has been issued yet, so this one is a direction of travel rather than a settled fact, and it is worth watching rather than planning around.

Notice the pattern. It repeats. The two standards that loosened are the ones a tenure cap was never really addressing. The two that did not move are the ones it still doesn’t address.

What the Cap Costs You

Rotation is not free, and the invoice arrives in places nobody attributes back to the policy. Nobody budgets for it.

A contractor who has been in your environment for a year knows which service owns the flaky nightly job, which director actually approves schema changes, and which of the four dashboards anyone trusts. Try finding any of that in a wiki. When the calendar removes them, it leaves with them, and the replacement spends six to ten weeks rebuilding it while billing the same rate. We have watched teams rotate off a senior integration contractor in month eighteen and then pay a premium to bring back a less experienced replacement into the same seat forty days later, because the work had not ended. Only the clock had.

Then there is the search itself. Across our desk the average time-to-hire on an IT contract role runs 17 days, which is quick, and it is still 17 days during which the seat is empty and the sprint slips. Teams feel it.

The quieter cost is who accepts your work. Strong contractors talk. A shop known for hard eighteen-month caps and no conversion path gets a thinner top of funnel over time, because the people with options take assignments where staying is possible. That shows up two years later as a sourcing problem nobody connects to a policy written in 2004.

The Risks a Calendar Doesn’t Touch

If you keep only one section, keep this one. Read it twice. These are the exposures the eighteen-month rule is imagined to solve, and what actually resolves each:

• Benefit plan claims. This is the real Vizcaino risk. The Ninth Circuit was unimpressed by contractual disclaimers that told workers they were contractors. What holds up is plan document language keyed to payroll records, excluding anyone not carried as a common-law employee on your books even if a court or agency later reclassifies them. That is a half-day of work for ERISA counsel, and it survives a reclassification that a tenure cap would not have prevented.
• Who actually directs this person day to day? Work assignment, hours, the performance conversation, the laptop, the training. A worker rotated off at month seventeen who was managed exactly like a badge employee for all seventeen has generated the same evidence as one who stayed three years.
• ACA coverage obligations generally sit with the staffing firm as common-law employer, since the firm holds the EIN and aggregates hours across client sites. Generally is doing real work in that sentence. The determination is made on the facts, so a client that has taken over effective control of a worker can find the analysis pointed back at them.
• The boring one outperforms the cap. Statements of work that describe deliverables rather than a role, plus records showing the engagement was scoped, extended deliberately, and reviewed.
• Who you engage through, and how. A W-2 contractor placed through a staffing firm, an incorporated consultant on a corp-to-corp arrangement, and a 1099 individual carry meaningfully different profiles. We wrote up the differences in our breakdown of W-2, C2C, and 1099 engagement models.

Every one of those is about how the relationship is run and papered. Not one of them is about the calendar. That is the whole problem with using a calendar to solve them.

A Better Policy Than a Hard Cap

Replace the automatic termination with an automatic review. Same trigger date, different consequence. That’s the whole change.

1. Set a review trigger at twelve months, not a termination date. The clock’s job is to force a decision, not to make it. Nothing about month twelve should surprise anyone.
2. At the trigger, answer one question honestly. Is this ongoing work that belongs to a permanent seat, or scoped work that genuinely ends? Most teams already know. They just haven’t been asked in a setting where the answer has consequences.
3. If it’s ongoing work, convert. Run the numbers. Put the conversion fee against the cost of rotating and re-ramping, and check it against your actual comp bands before you assume conversion is the expensive option. Our contract-to-hire guide walks through how the fee schedules typically work, and the salary benchmark assistant is a faster way to sanity-check the band than three tabs of aggregator averages.
4. If it’s scoped work, write the scope down properly and extend against it. An extension tied to a defined deliverable is a stronger fact than an extension tied to nothing.
5. Audit control quarterly instead of annually. The test is short. If your manager sets the work, on your systems, in your standups, with no end in sight, then you have a permanent role wearing a contractor label, and no rotation schedule fixes that.
6. Get the plan documents right once. Then stop rebuilding policy around a case from 1997.

Some companies genuinely should keep a hard cap. Federal contractors with flow-down obligations, organizations under a collective bargaining agreement that speaks to contingent tenure, and anyone whose own client contracts impose one. If a customer contract says eighteen months, the number is eighteen months, and the analysis above is interesting but not actionable. Know which situation you’re in.

Things Hiring Managers Ask Us About This

Is there any law that caps contractor tenure at 18 months?

Congress never wrote one, and neither did any state legislature. The IRS classification guidance contains no time threshold either. The number is a private convention that spread after the Microsoft permatemp litigation, and it has been repeated ever since without anyone rechecking it.

So can we just leave a contractor in place for five years?

Legally, there’s no clock that runs out. Practically, a five-year contractor in a permanent seat is usually a signal that the role should have been converted, and the longer it runs the harder the facts look if anyone examines the relationship. The exposure comes from how the work is controlled, not from the anniversary.

Does rotating someone off and rehiring them after 90 days actually reset anything?

Not reliably, and this one causes real damage. A break in service does not erase the prior relationship from a classification analysis, and if the same person returns to the same seat doing the same work under the same manager, the gap reads as a formality. You paid the re-ramp cost for a defense that may not hold.

Our MSP requires the cap, so is it even negotiable?

Ask where the requirement originates. Sometimes it’s a genuine client contract flow-down, in which case you comply. Often it’s the program’s own default from years back, and defaults are negotiable, particularly on individually scoped engagements outside the standard requisition flow.

What’s the tell that a contractor should have been converted months ago?

They’re in your planning meetings. When somebody is being consulted on next quarter’s roadmap, mentoring your junior engineers, and holding institutional knowledge nobody else has, the market already converted them. Your paperwork just hasn’t caught up.

How do the 2026 federal changes affect what we should do right now?

They lower the temperature rather than change your homework. The narrowed joint-employer standard reduces one category of exposure, and the DOL proposal is not final, so the sound move is to fix plan documents and control practices, which matter under every version of these rules.

How We Handle This at KORE1

We place contract, contract-to-hire, and direct hire talent across eight verticals and more than 30 U.S. metros, and our recruiters average 15-plus years in the work. That last number matters here, because a recruiter who has watched a few of these cycles will tell a client the seat is permanent before the client is ready to hear it.

Our bias should be on the table. We bill on contract engagements, so a hard tenure cap that forces rotation generates more searches for us, not fewer. We still argue against the automatic ones, because clients who convert the right people keep teams together, and our 92% twelve-month retention rate depends on placements that were set up to last rather than placements that were set up to expire. Churn is good for a quarter and bad for a decade.

Carrying a tenure policy nobody has revisited since it was written? Worth an hour. For broader context on where contract hiring is heading this year, our 2026 contract labor market outlook covers the supply and rate picture. When you want a second read on a specific engagement, talk to a KORE1 recruiter and we’ll walk the facts with you.