What the 2023 participant-count change actually did to first-time audits

The immediate effect of the 2023 counting change was widely reported and easy to picture: count only participants with account balances, and plans carrying long tails of eligible non-participants fall out of large-plan filing status. A one-time contraction in the audited population.

The second-order effect got much less attention, and it is the one that matters if your business depends on finding plans that are about to need an audit for the first time. The change did not just make the pool smaller. It changed where new entrants come from, and it drained the source most people are still watching.

The old inflow had three tributaries

Before 2023, a plan arrived at its first audit through one of three routes:

1. Brand new plans that started large — a new 401(k) at an employer with a few hundred staff.
2. Growing plans that crossed 100 participants through headcount growth.
3. Eligibility drift — plans that crossed because the eligible population grew, or because an auto-enrollment feature or a plan-design change swept a large group into eligibility, regardless of whether any of them ever funded an account.

The third route was substantial, and it was the most predictable one. Eligible headcount is a function of employment and plan design. It rises steadily, it rarely reverses, and a plan approaching the line through eligibility drift was about as reliable a leading indicator as this market offered.

The 2023 change deleted that route entirely. Eligible participants with no balance no longer count, so eligibility drift no longer moves a plan toward an audit at all.

What remains in the count-driven path

For entries caused by the participant-count rules, two principal routes remain, and they behave very differently from the one that disappeared. This is not a complete census of first audits: a small pension plan can also require an IQPA report when it cannot satisfy the audit waiver, and a plan in the election band may choose the large-plan filing category.

Brand new plans

These are genuinely new filings with no history to read. They are findable, but not by watching a trend line — a plan that did not exist last year has no trajectory. They show up as a first Form 5500, and the relevant test is the first-year rule: end-of-year participants with account balances against the 100-participant threshold, with no election available because there is no prior return to carry.

Plans crossing 120

Above 120 the election is unavailable and large-plan filing is forced. For a pension plan with no other applicable exception, that generally brings the IQPA-report requirement with it. Crossings here are driven by growth in accounts with balances, not eligible headcount alone.

The watch band is farther from forced-large than it looks

Here is the part that runs against instinct. A great deal of prospecting attention goes to plans sitting in the 80–99 range, on the theory that they are one good year away from an audit.

They are mostly not. Consider what has to happen for a plan at 90 funded balances to arrive at a first audit:

• It has to grow past 120, not past 100, because between 100 and 120 it can elect to carry its prior small category.
• It has to do that on funded balances, which grow more slowly than headcount.
• And its filing and audit history still have to be reviewed; participant count alone cannot establish that the next IQPA report would be the first.

Even from the top of that band, the plan must move from 99 to more than 120 beginning-of-year accounts with balances before large-plan filing is forced. That is at least 22 additional funded accounts, not a one-participant move across a 100-person line. The watch band can still be useful, but it is not a near-term audit conclusion by itself.

Why the shape matters more than the size

If the inflow were simply smaller but the same shape, the response would be to work the same list harder. It is not the same shape, and three things follow.

Trend-watching on participant counts got much weaker

The signal that used to work — eligible headcount climbing toward 100 — no longer measures the population used for this filing-category test. A plan whose eligible population is growing while its funded balances are flat is not moving toward the participant-count trigger, and a trend model that has not been rebased can point in the wrong direction.

Filing history became a first-class input

Because the election exists, the same current count means different things depending on what the plan filed last year. A count without a filing history attached cannot answer the question at all in the 80–120 range, and that range is where most of the interesting plans sit. This is a big part of why we treat a plan as a series of connected filings rather than as a row with a current value.

There is also a population moving the other way

The mirror of a count-driven first audit is a plan moving out of forced large-plan filing. Plans that filed large and now count below 100 participants with account balances can default to small, while those inside the 80–120 band may decline the option to carry the prior large category. Whether an audit still applies requires the separate waiver and exception analysis. The direction matters to an audit practice, adviser, TPA, and sponsor even without an unsupported claim about the population’s relative size.

What is actually findable in public data

None of this requires private information. Every input is in the filings:

• Beginning-of-year participants with account balances, per plan year.
• What category the prior return was actually filed under — which is what determines whether an election is even available.
• Whether an independent auditor’s report was attached, and whose name is on it.
• Whether the auditor changed, and in which year.

The awkward part is that these live across several filing years and, for the attachment questions, inside documents rather than in the form’s data fields. A single-year extract cannot answer “is this a first audit” because first-ness is a statement about history. That is the connection problem, and it is most of what the product does.

The short version

• The 2023 change removed eligibility drift as a route into large-plan filing, which was the most predictable route.
• The remaining count-driven routes are new plans and continuing plans crossing 120 on beginning-of-year accounts with balances.
• The 80–99 watch band is not one participant away from forced large-plan filing; the election remains available through 120.
• Downward moves require the same discipline: filing category and audit-waiver eligibility are related, but they are not the same conclusion.