FTE Calculator

Ten part-timers are ten people and five FTE

A team of ten people working 20 hours a week is a headcount of ten and 5.0 full-time equivalents. Both numbers describe the same team, and both are correct. They answer different questions, and the trouble starts when one is used to answer the other's.

Headcount is about people. Ten people need ten one-to-ones, ten laptops, ten onboardings and ten seats. FTE is about capacity and cost: those ten people deliver five full weeks of work and cost roughly what five full-time salaries cost. A manager who plans a schedule from the FTE figure ends up with half the calendar they needed. A finance team that budgets from headcount books twice the salary they will actually pay.

Almost every rule you will meet picks one of the two deliberately. Grant reports and capacity models want FTE. Apprenticeship and disability quotas in Brazil, union thresholds, and most reporting obligations count people. The US employer mandate uses a hybrid that is neither, which is the subject of half this page. Read the rule before picking the number, not after.

The divisor is a choice, and it belongs on screen

FTE is a division. The numerator is the hours a person worked; the denominator is what your organisation calls a full schedule for the same window. Everything interesting happens in the denominator.

• United States: 40 hours a week, 2,080 hours a year.
• Brazil: 44 hours a week under the CLT, 2,288 hours a year.
• France: 35 hours a week since the Aubry laws, about 1,820 a year.
• United Kingdom: commonly 37.5 contracted hours a week.
• The ACA: 30 hours a week, or 130 in a month — not a working week anywhere, just the line above which the statute calls someone full time.

The same payroll of 400 weekly hours is 10.0 FTE against a 40-hour divisor, 9.09 against 44 and 11.43 against 35. That is a swing of more than two full positions on identical facts. A calculator with 40 baked into its code does not fail when you are outside the US; it answers confidently with the wrong country's assumption, which is worse.

Watch the monthly conversion too. A month is 52 divided by 12, or 4.33 weeks, so a 40-hour week is 173.33 hours a month and not 160. Treating a month as four weeks understates every part-timer by about 8 percent and quietly moves anyone near a threshold to the wrong side of it.

Overtime does not create 1.2 of a person

Someone who works 50 hours in a 40-hour week is 1.0 FTE. Not 1.25. FTE counts how many full-time positions the payroll occupies, and one person can only occupy one position no matter how many hours they put into it.

This is the most common way a total comes out too high. Sum the hours of a busy month, divide by the full-time figure, and every hour of overtime turns into fractional colleagues who do not exist. On a team of forty where a third of the staff run five hours over, that is 13 people and 65 extra hours: about 1.6 phantom FTE — enough to cross a reporting line or to justify a headcount freeze that was never needed.

The cap is on by default in this tool, and the result panel shows what the overtime would have added if it had been counted. There is one honest use for the uncapped number: capacity planning, where the question is how much work was actually delivered rather than how many positions exist. For anything that triggers an obligation, cap at 1.0.

The 50-FTE test in the ACA is a different calculation

Under section 4980H, an employer with 50 or more full-time employees including full-time equivalents is an Applicable Large Employer and must offer affordable health coverage the following year. The count is not the FTE figure from the top of this page. It has its own procedure, and the procedure has four steps that people skip.

1. Work one calendar month at a time. Twelve separate calculations, one per month of the prior calendar year.
2. Count full-time employees as one each. Anyone with 130 hours of service or more in that month is one employee — 130 hours and 260 hours both count as exactly one.
3. Convert everyone else, with a ceiling. Add up the hours of service of all non-full-time employees for that month, counting no more than 120 hours for any single person, then divide the total by 120.
4. Average the twelve monthly totals and round down. The fraction is dropped, not rounded to the nearest whole number. An average of 49.99 is 49, and 49 is not an Applicable Large Employer.

Notice what that 120 divisor does. A full month at a 40-hour week is 173.33 hours, so dividing part-time hours by 120 credits them as though a full month were only 120 hours — around 27.7 hours a week. The ACA figure therefore runs higher than a true FTE for the same people. Press "fill every month from the roster" above and the eleven-person default team reads 6.75 by the ordinary calculation and 8.78 by the ACA one. The two numbers are not comparable and neither is wrong: the statute asks for the second, and a budget needs the first.

Two details are worth pinning down. Hours of service include hours for which payment is due, so paid holiday, paid leave and paid vacation count even though nobody worked them. And there is a seasonal exception: if you exceeded the threshold for 120 days or fewer in the year and the excess was seasonal workers, you are not an Applicable Large Employer despite the arithmetic.

Note also the dead zone the rule creates between 120 and 130 hours. An employee at 129 hours in a month is not full time, and their hours are still trimmed to 120 — nine hours that count for nothing on either side of the line.

A worked example where the shortcut changes who is covered

The shortcut is to sum the year's part-time hours and divide once by 1,440, which is 120 times twelve. Averaging is linear, so that shortcut agrees with the monthly method exactly as long as nobody crosses 120 hours in a month. The monthly ceiling is the only thing that separates them — and when it bites, it decides the outcome.

Take an employer with 42 full-time employees all year and a summer crew of 20 people who each work 200 hours in June, July and August and nothing in the other nine months.

The gap is entirely the ceiling: each summer worker contributes 120 hours a month under the rule, not the 200 they worked, so 600 hours of individual effort enter the formula as 360. Three of those seasonal hours in every five simply do not count. The shortcut keeps them, crosses 50, and tells an employer with no obligation that they must offer coverage to everyone. That is the panel loaded by default in the calculator above; clear it and enter your own months.

A last wrinkle in the same example: three individual months land at 62, comfortably over the threshold, while the year averages 47. Monthly peaks do not create the obligation. It is the average of the twelve that decides, which is also why the seasonal exception exists.

Half an FTE is half a fully loaded cost, not half a salary

Once a budget is written in FTE, a role that appears as 0.5 is half of everything that role costs the organisation, not half of the number on the offer letter. Employer taxes, insurance, retirement contributions, equipment and software licences all scale with the fraction.

They do not all scale cleanly, which is where fractional budgeting goes wrong. Payroll taxes and benefits generally do scale. A laptop, a desk and most per-seat software licences do not: a 0.5 FTE analyst uses a whole seat of the analytics tool. Budgets built by halving a salary and forgetting the fixed per-person costs come in low every time, and the gap is widest for expensive senior roles where the load is heaviest.

The cost field above takes the fully loaded annual cost of one full-time equivalent and multiplies it by your total FTE. If you enter a bare salary you will get a bare-salary answer, which is usually 25 to 40 percent under the real figure in the US and more than that in countries with heavier statutory charges.

Where FTE numbers quietly go wrong

• Contractors and agency staff. They consume capacity but are usually outside the employee count. Whether they belong in your FTE depends on whether you are answering a capacity question or a legal one — and mixing the two in one spreadsheet is how a number becomes indefensible.
• Employees on leave. Someone on unpaid leave contributes to headcount but few or no hours of service, so their FTE contribution falls to nearly zero while the person very much still works for you. Paid leave is the opposite: full hours, no output.
• Mid-period hires and leavers. A person who starts on the 20th is one headcount and a fraction of an FTE for that period. Counting them whole is the second most common source of an inflated total after overtime.
• Different definitions for different programmes. The ACA test, PPP loan forgiveness, federal grant reporting and your own capacity model all use different divisors and different caps. An FTE figure without the rule it was computed under is not a figure, it is a rumour.
• Ratios per FTE. Revenue per FTE and cases per FTE are only comparable between organisations that use the same divisor. Comparing a 44-hour country against a 35-hour one on revenue per FTE flatters the shorter week by about a quarter before anyone does any work.

What this calculator does, and what it does not

It takes a list rather than a single total, because FTE only matters when schedules differ and a pre-summed total is exactly where the per-person cap disappears. It shows headcount and FTE at equal weight. It puts the divisor on screen with presets for the common national weeks and lets you count in hours per week, hours in a period, or hours in a year. It applies the 1.0 cap by default and tells you what the overtime would have added. It runs the ACA test the way the regulation describes it, month by month, and prints the annual shortcut beside the correct answer so a disagreement is visible rather than silent.

It does not know your jurisdiction, your collective agreements or your plan year, and it is arithmetic rather than legal or tax advice. Every figure stays in your browser: nothing is uploaded, nothing is stored and there is no account, which matters when the input is a staff list with hours attached.