Why a year's working day total starts with 260, 261 or 262 weekdays

Every year has 260, 261 or 262 weekdays. Nothing else is possible. Fifty-two weeks account for 364 days and exactly 260 weekdays; the one or two days left over decide the rest, and they are decided by what weekday 1 January falls on and whether February has 29 days.

That is the entire calculation. Everything people argue about, closures, leave, fixed payroll divisors, sits on top of it and comes from somewhere other than the calendar.

The weekday January 1 falls on is the only variable that matters here.

It is old enough to have its own notation: the dominical letter encodes exactly that and identified a year's shape centuries before anyone needed a payroll divisor.

What to take away

The arithmetic for nine years

Read the pattern. A common year starting on any weekday has 261, since its one leftover day is a weekday. A common year starting on Saturday or Sunday has 260.

A leap year has two leftover days. It can reach 262 when both are weekdays, or fall to 260 when the pair is Saturday and Sunday, as in 2028.

A leap year is divisible by four, with two exceptions: divisible by 100 is not, unless also by 400.

The rule and its history explain why 2000 had February 29 and 1900 did not.

From the ceiling to the days you can use

The weekday count is where planning starts, not where it ends. Build downwards in named rows, and take each figure from the body that owns it:

The first row is universal, but every row after it belongs to a place, sector and employer. The answer for one is not the answer for another, and two organizations on the same street will not agree on row two, and should not.

Which list governs a deadline, and why a national list is often wrong, is set out under adding business days. The same choice applies to a yearly count.

Hours

Multiply working days by the length of the working day, and be explicit about which length. At eight hours, 261 days is 2,088 hours. At seven and a half, it is 1,957.5. The difference between the two is more than sixteen working days' worth of hours, and neither figure is wrong; they answer for different contracts.

Many systems use a fixed annual hours figure or a fixed number of working days as a divisor for every year, rather than the actual count. That constant is a policy choice inside the system, chosen for stability, and the only reliable place to read it is the system's own configuration or the policy that set it.

Reproducing the count

For a year in A2:

=[NETWORKDAYS](/working-days-per-year-excel)(DATE(A2,1,1), DATE(A2,12,31), Closures)

Both endpoints count, which is correct for a whole year. Omit Closures to get the weekday ceiling.

For a workweek other than Saturday and Sunday, NETWORKDAYS.INTL takes a seven-character string for Monday through Sunday, with 1 marking a rest day. A six-day week, for instance, gives a wholly different ceiling, and no shortcut links the two.

By hand: 260, plus one if 1 January is a weekday, plus one more in a leap year if 31 December is also a weekday. Then subtract weekday closures.

Where a yearly count is the wrong tool

A year is an average of twelve unequal months. Individual months carry between 20 and 23 weekdays, so any per-month figure derived by dividing the year by twelve will be wrong for every month; the monthly counts have their own rule and their own table.

A yearly count hides sequence. Two closures in one week differ from two six months apart when you plan a deadline chain.

A chain counted backward from a fixed end must be built link by link, as set out under subtracting business days.

The payroll divisor is a policy choice

A fixed divisor, whether it is a number of days a month or a number of hours a year, is not derived from the calendar. It is a decision to smooth a figure that genuinely varies, and it is a defensible one as long as everyone knows that is what it is.

Two consequences follow. First, a divisor and a calendar count will disagree in most years. That is not an error to reconcile away. It is the smoothing working as intended.

Second, the direction is predictable. A divisor set from a long-run average will overstate capacity in years with 260 weekdays and understate it in years with 262.

What causes trouble is using the divisor in one place and the calendar count in another, in the same calculation. A rate derived from a smoothed divisor and applied to an actual count of days does not give back the amount you started from, and the shortfall looks like a rounding problem rather than a mixed method.

Write down which of the two any given figure used. That one habit turns an unexplained variance into a labeled one.

Common questions

Why do two consecutive years have different counts when neither is a leap year?

They do not, unless one of them starts on a weekend. Consecutive common years differ only when the starting weekday crosses from Friday to Saturday, or from Sunday to Monday.

Is 260 a safe number to assume?

It is the minimum, so a plan built on it never overstates capacity. It also understates it by a day or two in most years, and by more once you realize that the number you actually wanted was working days after closures, which is smaller still.

Does a leap year always add a working day?

No. It adds a calendar day, a working day only if it falls on a weekday.

2028 gains a day and loses a weekday versus 2027. Its extra day is a Saturday, and its start moved to Saturday too.

Where do I get the closures for my count?

From the body whose days you are counting: the court, the agency, the bank, the employer. Not from a general calendar page and not from here. The endpoint rules that decide whether the first and last day of any span are both in are covered under counting between two dates.