The greening rate

How much turnover your covered population needs just to hold its demographic cost flat, and what the shortfall costs per employee per year.

Fringe Theory / Tool

Every covered population needs a certain amount of turnover just to hold its age structure still. Set your workforce plan below, find out whether it clears the bar, and see what the drift costs before medical trend touches anything. Pin a scenario to compare it against a second plan.

Read the argument behind this tool.

Average age of covered employees 42 Enrolled employees, not total headcount. The covered group usually runs older, since younger workers waive coverage more often. The median American worker is about 42.

Headcount plan, per year 0% Net growth or contraction, whatever the reason: expansion, layoffs, attrition you choose not to backfill, roles handed to AI.

Annual turnover, all separations 13% Voluntary plus involuntary, on an annual basis. Watch the units: the quits rate you see quoted, at or below 2.0 percent, is monthly, which annualizes to roughly 23 percent nationally and about 13 percent in information-sector work. Quits also exclude layoffs, so they understate total separations. Use your own number.

Who's leaving

The same turnover rate does very different things depending on who it takes. Early career is the national pattern, since median tenure is shortest for the youngest workers, and it is the hardest case for holding age flat.

Share of hires early in career 50% Early career means ages 25 to 34, weighted toward the younger end. Everyone else arrives at 38.

Age people leave for good 67 Defaults to 67, the Social Security normal retirement age for everyone born after 1959. Surveys put the actual average exit closer to 62, so drag it down if your people leave when their health or their savings decide.

Age cost curve

The 3:1 curve is the federal rating floor, compressed by law. Actuarial guidelines put the true allowed-cost spread near 4.2:1, so the regulatory setting understates the drift.

Horizon

Every figure below is measured at this horizon. Five years is about the length of a contract or a planning cycle, and it is where the annual drift runs hardest.

Employees on the plan Scales the people and dollar figures. The percentages do not depend on it.

Annual plan cost per employee Employer plus employee share, medical only.

Fully loaded cost per hire Recruiting, signing, relocation, training, ramp. Recruiting alone averages about $4,700; loaded it runs far higher.

Your greening rate

—

Average age

—

Headcount

—

Cost per employee

—

Total plan cost

—

Hiring spend, per year

—

Demographic trend, annualized

—

Cost per employee Total plan cost Pinned scenario

Both lines start at 100, meaning today's level, before medical trend. When they split, the plan is shrinking and aging at the same time.

Year	Employees	Avg age	Cost / employee	Total plan	Hiring spend

What this is, and isn't

This is arithmetic, not a study. Give the model an age curve, a retirement age, and a churn rate, and the equilibrium age structure is determined. Nobody hired at 28 can be 55 within ten years, and no amount of hiring changes that. That makes the greening rate unarguable, since it is your own census restated rather than a claim about the world, but it also means the model is not discovering anything new about your workforce. It is showing you what you already implied when you set the inputs.

And this prices average cost by age, generally speaking, not large claims. The curves behind it, the ACA default and the steeper actuarial version, describe how allowed cost rises with age on average. They say nothing about the incidence of the claims that actually break a renewal, which tends to be lumpier and to rise faster with age than a smooth curve suggests. Treat what is here as the floor of the exposure, not the whole of it.

One caution on the annualized figure. Most renewals already re-rate your census, so what you see here is generally the demographic share of the trend you were quoted rather than a separate cost to add on top of it. Ask your carrier or your actuary whether their trend is census-adjusted. If it is, this number tells you how much of the renewal is medicine and how much is your own workforce. If it is not, it belongs on top.

If you run your own numbers and something looks wrong, I want to hear it. The model is simple enough to argue with, which is the point.

Demographic effect only; carrier medical trend sits on top of everything shown here. The regulatory cost curve is the federal default standard age curve at 45 CFR 147.102; the actuarial setting rescales it to a 4.2:1 spread, the allowed-cost range in Milliman's Health Cost Guidelines, using the Commonwealth Fund's rate-band scaling method. Both track average allowed cost by age and are not claims projections. Below-retirement separation rates are proportional to the inverse of median tenure by age from the BLS Employee Tenure release, which is the "mostly early career" setting; the other two settings flatten or reverse that shape while holding your stated turnover rate constant. Turnover context is the BLS Job Openings and Labor Turnover Survey; actual retirement age reflects Gallup and EBRI 2026 survey findings; average recruiting cost per hire is from SHRM benchmarking. Fringe Theory is independent and unaffiliated. Views expressed are my own and do not represent those of my employer. Nothing here is legal, tax, medical, or investment advice.