Carry-Over, Expiry, and the Year-End Leave Rush

June is the right month to think about December. Not because anything is urgent yet, but because the year-end leave rush is entirely predictable and entirely preventable, and the window in which you can still prevent it closes some time around September.

The pattern is familiar to anyone who has run a team through a full leave year. For eleven months, nobody thinks about their balance. Then a reminder goes out — or someone mentions it in a standup — and suddenly a third of the company wants the last two weeks of December off, half of them are requesting days they would not otherwise have taken, and the ones who genuinely planned ahead are being told the coverage is full.

That is not a booking problem. It is a visibility problem that surfaces as a booking problem, and the way out is to fix the policy design and the balance display together.

What Carry-Over Actually Costs You

Carry-over rules exist for a reason: people have bad years, projects slip, and it is unfair to punish someone for a leave year in which they genuinely could not get away. But unlimited carry-over creates a liability that grows quietly. Accrued unused leave is generally an obligation — in many jurisdictions it is payable on termination — and it also represents rest your team has not had.

Broadly, employers pick one of three shapes:

Use it or lose it. Nothing carries into the next year. Clean to administer, brutal in a bad year, and in several jurisdictions not permitted for statutory minimum entitlement — check your local rules before adopting it, because the answer differs between countries and, in Canada and the US, between provinces and states.

Capped carry-over. A fixed number of days moves forward — five is a common choice — and the rest expires. This is the most widely used shape because it acknowledges that things go wrong without letting the balance compound forever.

Capped carry-over with an expiry window. The same cap, but the carried days must be used by a stated date — end of Q1 is typical. This is the version that actually changes behaviour, because it converts "I have five extra days somewhere" into "I have five days that disappear on March 31."

The third option is the one worth arguing for. A cap alone still lets the rush happen; it just moves it. A cap with a deadline forces the conversation into January and February, when coverage is much easier to arrange than the last fortnight of December.

The Real Problem Is That Nobody Can See Their Balance

Whatever shape you pick, the policy only works if people can see where they stand without asking anyone. This is where spreadsheet-based tracking quietly fails: the balance exists, but it lives in a file that HR owns, gets updated in batches, and is out of date the moment somebody books a day.

An employee who cannot see their balance behaves in exactly two ways. Either they under-book, because they are not sure how much they have and would rather not be told no — and then arrive at December with a pile of days. Or they over-book, discover after the fact that they were three days short, and now someone has to unwind an approved request.

The fix is that the number should be in front of them at the moment they are deciding. Not in a monthly email, not on request. In our own leave management product, the balance preview updates live inside the request form — you see what the request will leave you with before you submit it. That single piece of feedback removes most of the guesswork that produces the December pile-up.

If you are still on a spreadsheet, the cheapest version of this is a shared, always-current balance view that employees can open themselves. It is not as good, but it is enormously better than a number only HR can see.

Draw Down Carried Days First

A small mechanical decision with an outsized effect: when someone books leave, which bucket does it come out of?

If new-year entitlement is consumed first, carried days sit untouched until they expire, and you have recreated the problem your expiry window was meant to solve. If carried days are consumed first, every request in January and February quietly eats into the balance that was going to be a problem, and by the time the expiry date arrives most of it is gone.

Draw down carried days first. It requires no communication, no reminders, and no policy enforcement — it just works while people take their normal leave.

Give the Year-End a Shape

Even with good rules, December is structurally hard: it contains public holidays, a natural slowdown, and school breaks, so demand really is concentrated. Three things help.

Publish the coverage constraint before people book, not after. If a team of six needs two people available in the last week of December, say so in October. Approving on a first-come basis without stating the rule looks arbitrary from the outside, even when it isn't.

Set an internal booking deadline for the busy window. "Requests for the final two weeks of December should be in by 15 November" gives you one decision point with the full picture instead of twenty decisions made blind. It also gives the people who lose out enough notice to plan something else.

Do a mid-year balance check. Around the halfway mark, look at who has taken less than a third of their entitlement and talk to them. It is almost never about the leave — it is a workload problem, a coverage gap, or a manager who has never visibly taken time off themselves. All three are worth knowing about in June and expensive to discover in December.

Watch the Edges

Two categories of employee break naive carry-over arithmetic and are worth handling deliberately.

Mid-year joiners. Someone who starts in September has a pro-rated entitlement, and applying a full-year carry-over cap to a partial-year balance produces a number that makes no sense to anyone. Pro-rate the entitlement, then apply the cap to what remains.

People on long-term leave. Parental leave, extended sick leave, and sabbaticals all interact with accrual and carry-over, and the rules vary considerably by jurisdiction. This is one of the few areas where reading the actual statute — or asking someone who has — is worth the hour.

The Policy Is Only Half of It

You can write a sensible carry-over rule in an afternoon. Making it work takes two more things: employees who can see their balance whenever they think about it, and managers who plan coverage before demand concentrates rather than after.

Get those in place by early autumn and December stops being a scramble. It becomes what it should be — a slow month where most of the team happens to be off, because they booked it in August when there was still room.

If you want to see where your team currently stands, our PTO tracking view breaks balances down by employee and leave type, with carry-forward caps and expiry windows configured per type. Worth ten minutes in June.