Periods Measured in Years, and Their Edges
Long periods are counted the easy way and evidenced the hard way. Two years ends on the same date two years later; establishing which date it started on, when nobody wrote anything down at the time, is where these questions are actually decided.
The rule in short
A period expressed in years runs to the corresponding date in the later year, with the twenty-ninth of February as the single recurring complication. The arithmetic is straightforward. What causes difficulty is evidence: long periods are usually limitation periods, and the trigger date sits years in the past with no contemporaneous record of it.
Nobody miscounts two years. What they cannot do, five years afterwards, is prove which day the two years began on, because at the time it was just a day when something happened.
The arithmetic
The anniversary date ends the period. Two years from the tenth of a month ends on the tenth of that month two years later, without counting days.
The first day is still excluded. The convention about the trigger day applies to long periods exactly as it does to short ones.
Leap days are the one complication. A period of one year from the twenty-ninth of February has no corresponding date in three years out of four.
The usual answer is the twenty-eighth. Most systems treat such a period as ending on the last day of February, though the position should be checked.
Closures still adjust the end. An anniversary falling on a weekend or holiday moves in the ordinary way, per weekends, holidays and closures.
Years and days are different units. Seven hundred and thirty days is not two years across a leap period, and the rule chose one of the two deliberately.
An anniversary is easy to diarize. Unlike a computed day count, a period of years produces a date that can be entered the moment the trigger is known.
Where long periods appear
Limitation periods on claims. Most are expressed in years, and their starting point is a legal question rather than a calendar one, per when a claim accrues.
Statutes of repose. Outer limits measured in years from a fixed act, which run regardless of discovery and do not extend.
Retention obligations. Requirements to keep records for a number of years, which are counted from the record rather than from the event.
Residence and presence requirements. Conditions requiring a person to have been somewhere for a period, counted in years and evidenced day by day.
Enforcement windows. Periods within which a judgment or an order can be acted on, frequently measured in years and frequently renewable.
Contractual terms and renewals. Anniversary dates in agreements, where the notice required to renew or end the term is itself pegged to that anniversary.
Warranty and guarantee periods. Consumer and commercial guarantees run in years from delivery or installation, and the trigger is whichever the document names.
| Period from 29 February | Common answer | Note |
|---|---|---|
| 1 year | 28 February | No corresponding date |
| 4 years | 29 February | Leap year again |
| 2 years from 15 June | 15 June | Straightforward anniversary |
| 730 days from 15 June | Depends on leap day | Counted in days |
| Anniversary on a Sunday | Next open day | Closure adjustment |
Proving where a long period started
Nobody records an ordinary day. The trigger for a limitation period is usually a day that mattered to nobody at the time, and the record is whatever happens to survive.
Contemporaneous documents fix it. A dated letter, an invoice, an appointment record or a message thread from the period is worth more than any recollection.
Institutional records outlast personal ones. Bank, employer, medical and utility records survive on retention schedules and can be requested.
Reconstruct from surrounding events. Anchoring the date to something dated, such as a birthday or a move, narrows it where nothing records it directly.
Ranges are sometimes enough. Where the period is long and the claim is early, an approximate date can be sufficient if it cannot fall outside the window.
Record it now for next time. Where a period is running, writing down the trigger and the evidence for it costs nothing today and cannot be done at all later.
Ask the holder before they dispose. Institutional records are destroyed on schedule, so a request made inside the retention window succeeds where the same request afterwards does not.
Long periods are lost to evidence rather than to arithmetic. A single dated file note recording when the period started, and what document establishes it, is trivial to make at the time and impossible to make three years later.
What can move a long period
Tolling and suspension. Long periods pause for the same reasons short ones do, and the pauses are covered in equitable tolling.
Discovery rules. A period of years can start much later than the conduct, which is the practical effect of a discovery provision.
Acknowledgment and part payment. In some systems an acknowledgment of a debt restarts the period entirely, which is a substantial difference.
Minority and incapacity. A period may not run at all while a claimant is under age or incapable, and it then runs in full from that point.
Contractual shortening. An agreement can cut a statutory period down, subject to limits on how far.
Repose limits ignore all of it. An outer limit measured from an act is not moved by discovery, tolling or anything else, which is the point of it.
Handling it in practice
Diarize the anniversary early. A period of years disappears from view precisely because it is long, and a reminder well before the date is worth more than the date itself.
Fix the trigger in writing at the start. The single most useful step, and it takes one line in a file note.
Gather the evidence while it exists. Records that establish the start date are easiest to obtain in the first year and hardest in the last.
Decide the leap-day question in advance. Where a period runs from the twenty-ninth of February, the end date should be settled rather than discovered.
Do not rely on the whole period. A limitation period is an outer limit rather than a schedule, and evidence decays throughout it.
Check whether anything paused it. Before assuming a period has expired, the pauses that apply are worth confirming.
Periods of years end on the anniversary, and the only recurring complication in the arithmetic is the leap day, which has a settled answer worth confirming once.
The difficulty in long periods is evidential. Their triggers are ordinary days that nobody documented, and the question years later is not how to count but where to count from.
Contemporaneous records are what fix a distant start date, and institutional records outlast personal ones because they are kept on schedules rather than by intention.
Long periods move for the same reasons short ones do, with the exception of repose limits, which are designed to be immovable and are therefore the ones to check first.
The practical discipline is to record the trigger in writing when the period begins, to diarize the anniversary long before it arrives, and to treat the full period as an outer limit rather than a plan.
Points to carry away
- A period of years ends on the anniversary of the trigger.
- The leap day needs a decided answer rather than an assumption.
- The arithmetic is easy; the start date is the argument.
- Contemporaneous records are what fix a distant trigger.
- Long periods still exclude the day of the act.
Questions readers ask
How is a period of years counted?
To the corresponding date in the later year. Two years from the tenth of March ends on the tenth of March two years later, without any counting of days, and the day the triggering event occurred is excluded in the ordinary way. The single recurring complication is a period running from the twenty-ninth of February, which has no corresponding date in most years; the usual answer is that it ends on the twenty-eighth, though the computation provision for the forum should be checked.
Why are long periods harder than short ones?
Because the arithmetic is trivial and the evidence is not. A limitation period usually starts on a day that mattered to nobody at the time, so there is no contemporaneous record of it, and the question years later becomes what establishes the start date rather than how to count from it. Dated documents from the period, institutional records that survive on retention schedules, and anchoring to other dated events are what answer it.
Should a claim wait until the end of a limitation period?
Rarely. A limitation period is an outer limit rather than a schedule, and everything that makes a claim provable decays across it: witnesses move, memories reshape, records reach the end of their retention and are destroyed. Filing well inside the period also removes the risk that the start date turns out to be earlier than assumed, which is a real risk precisely because long periods begin on days nobody recorded.
Sources
- Legal Information Institute — Statute of Limitationslaw.cornell.edu
- Federal Rules of Civil Procedure — Rule 6, Computing and Extending Timelaw.cornell.edu
- 1 U.S.C. 1 — Rules of Constructionlaw.cornell.edu
- Legal Information Institute — Statute of Reposelaw.cornell.edu
- Legal Information Institute — Tollinglaw.cornell.edu
- National Archives — Records Management Schedulesarchives.gov
Urban Justice Docket is a publication, not a law firm. This article states general rules and cites its sources; it is not advice about any particular case, and the law differs by state and changes over time.
More in Counting the Days
When the Clock Actually Starts Running
Every time limit has a trigger, and the trigger is defined by the rule rather than by common sense. Service, issue, receipt, demand and completion are all different events, and a period counted from the wrong one is wrong by however many days separate them. Finding the trigger in the text is the first step in any deadline question.
What a Month Means in a Deadline
A period expressed in months runs to the same day-number in the later month rather than by adding a fixed number of days. Where the later month is shorter and has no corresponding date, the period generally ends on its last day. Periods expressed in days are counted in days regardless of how many months they span, and mixing the two conventions is a common source of error.
Weekends, Holidays and Closures
Where the last day of a computed period falls on a weekend, a holiday or a day the filing office is inaccessible, the period generally extends to the next day it is open. The extension applies to the end of a period rather than to days inside it, and an office being busy, understaffed or closed to visitors is not the same as being inaccessible.