Single-Employee Corporation PHSP Eligibility

A single-employee corporation can qualify for a health services plan of this kind, but the path is narrower than most incorporated professionals assume. The tax authority's own guidance has shifted, tightened, and shifted again over the past two decades, and the structure of the plan itself, not just who it covers, is what survives an audit.

Why the insurance-risk requirement is the hardest obstacle for a one-person corporation

Interpretation Bulletin IT-339R2 sets the definition: a PHSP is an undertaking by one person to indemnify another for an agreed consideration, against a loss from an event whose happening is uncertain. There has to be a reasonable element of risk assumed by the employer or a third-party insurer, whoever's on the hook. Risk is the operative word. Without it, there's no insurance, and without insurance, there's no PHSP.

A cost-plus Health Spending Account for a sole employee-shareholder runs into trouble here. Under a cost-plus arrangement, the plan administrator reimburses the employee for whatever medical expenses actually happened, then bills the corporation for that amount plus an administrative fee, usually in the 8 to 10% range. Nothing is pooled. Nothing is underwritten. The plan pays out what gets claimed, dollar for dollar. No one assumes any risk of paying out more than what's owed, because there's only one person in the plan and the reimbursement always matches the claim exactly.

Once the corporate paperwork is stripped away, what's left is a shareholder using company funds to cover personal medical bills. That might feel like a technicality, but the CRA doesn't treat it as one.

The clearest statement of this problem comes from CRA Technical Interpretation 2014-0521301E5, dated June 25, 2014. The agency concluded that a cost-plus plan for a sole employee-shareholder "would not likely constitute a plan in the nature of insurance," which means it fails the PHSP test at the definitional level, before the question of employee status even comes up. It's a widely referenced ruling on this exact structure, and for good reason: it names the problem.

The comparison to sole proprietors is instructive. A sole proprietor with no employees can't use a self-insured PHSP at all, because there's no risk pooling possible with one person and no employer-employee relationship to speak of. The same logic reaches into a one-person corporation. Incorporating changes the legal relationship on paper, but it doesn't manufacture risk where none exists. Many incorporated professionals assume that once they've incorporated, they no longer need to worry about the PHSP's insurance-risk test. That assumption doesn't hold. The insurance-risk test is a separate hurdle about how the plan is built.

The evolution of CRA guidance on sole shareholder-employees from 2006 to 2022

The paper trail here is genuinely messy, and it's worth walking through each document rather than skipping to a summary, because the direction of travel matters as much as the current position.

Start with CRA Document 2005-0163771E5, dated March 14, 2006. At that point, the agency signaled it would allow a PHSP for a corporation's sole shareholder-employee, provided the person was actively working as an employee and the benefits matched what an arm's-length employee in a similar role would reasonably get. That was the permissive baseline. For years, it's the position a lot of tax advisors pointed to.

Then came the 2014 interpretation mentioned above. CRA Technical Interpretation 2014-0521301E5 didn't just add nuance, it reversed course on the specific mechanics of a cost-plus HSA. The agency found the plan lacked the necessary elements of insurance and would not likely qualify. That's a narrower, tighter reading than 2006 offered, and it applies squarely to the self-insured format most incorporated professionals actually use.

CRA Document 2016-0633741C6, from the May 3, 2016 CALU Roundtable, added a carve-out that sounds helpful until you read the fine print. A plan can qualify for a single plan member who is the company's sole employee and deals at arm's length with the company. Notice the condition: the sole employee can't be the shareholder. That exception helps almost no one running a typical incorporated professional practice, since in that setup the sole employee and the sole shareholder are usually the same person.

CRA Document 2017-0703871C6, dated September 14, 2017, reads more permissively again. It addressed a corporation providing a self-insured health spending account for its only employee, who was also its sole shareholder, and the agency appeared open to allowing such plans under certain circumstances. The document even noted that the sole employee-shareholder would likely be reimbursed the full amount allocated each year.

The most recent word on this comes from the 2022 CALU Roundtable, CRA Document 2022-0928901C6, dated May 3, 2022. The agency confirmed that a self-insured HSA set up for a sole employee-shareholder and family members likely does not qualify as a PHSP, and it restated the same insurance-risk reasoning from 2014.

So where does that leave things? The 2006 permissive stance is effectively superseded for this specific structure. Every document from 2014 forward, including the 2022 roundtable, converges on the same conclusion: a cost-plus or self-insured plan where the sole plan member is also the sole shareholder does not qualify. The 2017 document reads more favorably in isolation, but it doesn't override the position CRA restated five years later. Treating the 2006 letter as current guidance would be a mistake, and treating the question as genuinely open at this point isn't supported by the record.

What the CRA looks for to distinguish shareholders from employees

Even setting aside the insurance-risk test covered earlier, there's a second, separate hurdle: capacity. Benefits have to be received because someone is an employee, not because they're a shareholder. Shareholder benefits get taxed as personal income to the individual and aren't deductible by the corporation, which is exactly the double-taxation outcome a PHSP is supposed to avoid.

The CRA's starting assumption works against the sole owner-operator. Shareholders who can significantly influence business policy, which describes just about every sole shareholder of a professional corporation, are presumed to receive benefits because they own the business, not because they're employed by it. That presumption doesn't go away on its own. It has to be actively rebutted with evidence.

CRA's stated position, as applied in Canadian tax jurisprudence, treats any PHSP benefit received by the sole employee of their own professional corporation as flowing from share ownership rather than employment, and it gets taxed as a shareholder benefit accordingly. That's the default outcome unless the facts point the other way.

What tips the facts the other way? The CRA looks at several things:

• Active, day-to-day engagement in running the business, not passive ownership
• A formal employment contract that names the PHSP as a term of employment
• T4 salary income rather than compensation paid entirely through dividends (TaxTips.ca notes there's no definitive CRA ruling on whether dividends alone disqualify someone, but the conservative move is to draw at least some salary)
• Minute book language confirming the same plan would be offered to any future arm's-length hire, and that the plan can't be changed at the individual's personal discretion
• Benefit levels that match what an arm's-length employee doing similar work at a similarly sized company would reasonably receive

That last point creates its own headache. The CRA allows qualification if someone can show that non-shareholder employees with similar duties at a similarly sized corporation get similar benefits under a similar plan. In practice, finding a genuinely comparable arrangement to point to is hard, particularly for a small professional corporation where there's no obvious peer group to reference.

The 2004 Tax Court decision in Spicy Sports Inc. v. the Queen shows what happens when this goes wrong. The court found that a large payment made from a cost-plus PHSP amounted to a shareholder benefit. It's a concrete illustration of the exact outcome this whole framework is built to prevent.

Passing the capacity test is necessary, but it doesn't resolve the insurance-risk test covered earlier. A plan can clear the employee-capacity bar and still fail because it lacks any real element of risk. The two tests are independent, and a plan needs to clear both.

How adding one genuine employee changes the eligibility picture under the arm's-length employee pathway

Adding a genuine arm's-length employee to the corporation changes the math. It doesn't just add a body to the payroll, it changes the plan's risk profile in a way that speaks directly to the insurance-risk requirement discussed earlier.

The structural basis for this comes from IT Bulletin 529, paragraph 26, sometimes called the "Top Hat" rule. It states that having at least one arm's-length employee is what legally permits a shareholder to use a pay-as-you-claim PHSP. Once there's more than one person in the risk pool, and that other person isn't related to the owner, the plan starts to look like something other than a personal expense account run through a corporation.

Arm's-length, in this context, means genuinely unrelated. A spouse, a child, a sibling, or a parent doesn't count, no matter how much actual work they do for the business. Even a part-time arm's-length employee can satisfy the requirement, so long as the employment relationship is real and not manufactured for tax purposes.

That last qualifier matters. The employee has to receive coverage under the same plan, in a meaningful and equivalent way, not a token inclusion designed to check a box. CRA guidance requires that equivalent coverage extend to a participant with the same status as the arm's-length employee, so the shareholder can't set up a rich plan for themselves and a thin one for the employee just to claim the pathway applies.

A handful of design requirements apply regardless of how many employees are on staff:

• Every full-time T4 employee has to be included in the plan
• Benefit classes need defined, reasonable maximums, capped for every category, with nothing left open-ended
• A cost-plus HSA has to be 100% employer-funded (insured group plans can allow employee cost-sharing, reported under T4 code 85, but that's a different structure)
• The plan can't be altered at any one employee's personal discretion
• Rollover or forfeiture rules have to apply the same way across every benefit class

None of this solves the problem for a professional who genuinely has no arm's-length employees and no near-term plan to hire one. That person is still stuck with the disqualification outlined in the earlier sections, and adding an employee purely to unlock PHSP eligibility, without a real employment relationship behind it, is likely to draw the same scrutiny the plan was designed to avoid. Substance over form governs here, the same as everywhere else in tax law.

Plan structure features the CRA uses to assess whether an arrangement is genuine

Beyond who's covered, the CRA looks hard at how the plan itself is built. A written plan document isn't optional. There has to be something in writing that spells out coverage terms, and a plan with no coverage limitations at all is unlikely to be treated as a PHSP no matter who's enrolled in it.

Benefit classes need to be defined and finite. A plan can set different annual limits for different groups, Executive, Manager, Full-Time, Part-Time, whatever categories fit the business, but each class needs a cap, and that cap has to be reasonable. Unlimited allocations for any class disqualify the whole arrangement. Every covered employee also needs actual access to their plan's details.

The carry-forward rules, as set out in IT-529, are specific and easy to get wrong:

• Claims have to be submitted within a defined window following the benefit year in which the expense happened
• CRA guidance places restrictions on how unused allocations and eligible expenses may be carried forward within a plan
• Carrying expenses back to a prior year isn't allowed
• Moving unused funds between categories, say from a vision allowance into a prescription allowance, isn't allowed either

Both prohibitions exist for the same reason: they preserve whatever element of risk the plan has left. If money could roll forward indefinitely or move freely between categories, the whole thing would function more like a personal savings account with a tax break attached, and it would lose the insurance character that makes it a PHSP.

There's no statutory cap on how much an incorporated business can contribute, but contributions still have to be reasonable given the circumstances. A common industry guideline keeps employer contributions under roughly 25% of an employee's gross annual salary. That's not a legal ceiling; it's a practical norm among plan providers, and it functions as a sanity check rather than a rule.

Adding an insured component alongside a self-funded HSA can also help. Pairing the spending account with genuine insured coverage, something like Major Medical or Travel Insurance, introduces actual underwritten risk into the arrangement, which speaks directly to the insurance-risk concern raised earlier. It's one of the more direct ways to add real substance to a plan that would otherwise look like pure reimbursement.

On the administrative side, a plan administrator typically handles adding and removing employees, and administrative fees running around 8 to 10% of claims are themselves deductible for the employer, a structural detail consistent with a genuine plan arrangement. None of this is just paperwork. The CRA treats plan structure as evidence, using it to distinguish a genuine PHSP from a personal expense account wearing a corporate label.

The double-taxation consequence when a plan fails CRA scrutiny

When a payment to a shareholder doesn't hold up as a qualifying PHSP payment, it doesn't just get denied quietly. It gets reclassified as a taxable shareholder benefit under the Income Tax Act, and that reclassification triggers a specific, costly chain of events.

First, the amount lands in the shareholder's personal income and gets taxed there at their marginal rate. Second, the corporation loses the deduction it originally claimed for the same payment. The same dollars get taxed twice, once going out of the corporation and once landing in the individual's hands, which defeats the entire point of setting up a PHSP to begin with. There's no CPP or EI wrinkle added on top of that, since the payment was never treated as employment income in the first place, but the double taxation alone is enough to erase whatever the plan was supposed to save.

That same 2004 case. v. the Queen decision from 2004 remains the clearest judicial example of this outcome. A large cost-plus PHSP payment got reclassified as a shareholder benefit, and the taxpayer ended up with the exact result a properly structured plan is meant to prevent: tax owed personally, with no corresponding deduction on the corporate side to offset it.