Excusable, Compensable, and Concurrent Delay

Every delay on a construction project eventually gets sorted into a few buckets. Is the contractor excused, meaning they get more time and don't owe liquidated damages for it? Is it compensable, meaning they also get paid for the extra time? Or is it the contractor's own problem? And when two delays overlap, what then? This post lays out the categories, shows how they combine, and works a concurrency example on a small schedule so you can see why the critical path decides most of it.

A caution up front: these terms are defined by your contract and by the law where the project sits, and both vary. What follows is the common framework schedulers work within, not legal advice. Read your contract and talk to counsel before you rely on any of it for a claim.

The basic categories

Non-excusable delay

The contractor caused it, or it was a risk the contractor took on. Late mobilization, a sub that didn't show, rework from a failed inspection on defective work, a fabricator that missed its own delivery promise. The contractor gets no time and no money, and if the delay pushes past the contract completion date, the owner may assess liquidated or actual damages.

Excusable, non-compensable delay

Neither party caused it, and the contract allocates the time risk to the owner but the cost risk to the contractor. The classic examples are unusually severe weather, strikes beyond the contractor's control, acts of government, and similar events many contracts list in a force majeure or delay clause. The contractor gets a time extension, so no damages for that period, but absorbs its own extended overhead.

Weather has its own documentation burden, which is covered in Documenting Weather Delays So They Hold Up.

Excusable, compensable delay

The owner, or someone the owner is responsible for, caused it. Late owner-furnished equipment, design changes, slow responses to RFIs and submittals beyond the contract's review periods, a differing site condition, an owner-directed suspension. The contractor gets time and money: typically extended general conditions, field overhead, and sometimes home office overhead, depending on the contract.

The matrix

Note the word "usually." Contracts move these lines. Some contracts make certain owner delays non-compensable through a no-damages-for-delay clause. Some don't grant weather days at all unless the weather exceeded a stated baseline. The table is a starting point, not a ruling.

It has to hit the critical path

A delay to an activity is not a delay to the project. For a time extension, the delaying event generally has to push the project completion date or a contract milestone. That means it has to affect the critical path, or affect an activity enough to eat through its float and become critical.

If an owner change delays the parking lot striping by two weeks and striping has six weeks of float, the activity was delayed but the project was not. No time extension, in most frameworks. There may still be cost if the change itself added work, but that's priced in the change order, not as delay.

How float gets treated is often written into the contract, and it changes this analysis a lot. See Who Owns the Float?.

Concurrent delay

Concurrent delay is when two or more delays, from different responsible parties, affect the project completion during the same period. The common example: the owner is late delivering a piece of owner-furnished equipment, and at the same time the contractor's crew isn't ready to install it because the contractor's own prerequisite work is behind.

The usual outcome in many frameworks is that the contractor gets time but not money for the concurrent period. Neither party can recover from the other for delay both caused. The contractor isn't charged damages, but doesn't collect extended overhead either. Some contracts and jurisdictions apportion differently, and some define concurrency narrowly, so check yours.

The hard part is proving the delays were actually concurrent, meaning both were on the critical path at the same time. That requires a schedule.

A worked concurrency example

Here's a small network for the back end of a project. The convention is 0-based: ES of the first activity is 0, EF = ES + duration, and a successor's ES is the latest EF of its predecessors (finish-to-start, no lag). Durations are in working days.

As planned, with the contract finish set equal to the planned early finish of day 35, so the critical path has zero float:

ID  Dur  ES  EF  LS  LF  TF
A   10    0  10   0  10   0
B   15    0  15   7  22   7
C   12   10  22  10  22   0
D    5   22  27  22  27   0
E    8   27  35  27  35   0

The critical path is A, C, D, E. The switchgear delivery has 7 days of float: it isn't needed until day 22.

What actually happens

• The switchgear arrives on day 30 instead of day 15. That's 15 days late, owner-caused.
• Electrical rough-in finishes on day 28 instead of 22. That's 6 days late, contractor-caused (a short-crewed sub).

Set and terminate switchgear can't start until both are done. Its actual start is day 30, the later of the two. It finishes day 35, and commissioning finishes day 43. The project is 8 days late.

Sorting it out

Walk through it period by period.

• Through day 22, nothing is late on the critical path yet. The switchgear is late against its own early date, but it still has float until day 22.
• From day 22 to day 28, both are holding up D. Rough-in isn't done (contractor) and the switchgear isn't there (owner). Those 6 days are concurrent.
• From day 28 to day 30, rough-in is done. Only the switchgear is holding things up. Those 2 days are owner delay alone.

So under the common framework: 8 days of time extension total, 6 of them concurrent (time, no money), 2 of them owner-caused (time and money). The contractor owes no damages for any of the 8.

Notice what the switchgear's float did. The switchgear was 15 days late, but only 8 days hit the project, because 7 days were absorbed by float. Someone arguing for 15 days of owner delay is ignoring that. And notice the rough-in slip of 6 days would have caused 6 days of delay on its own, but in this case it was fully overlapped by the switchgear.

Real projects are messier. Delays start and stop, the critical path shifts between updates, and progress on one path changes float on another. That's why forensic analysts break the job into windows and analyze each one; see As-Planned, As-Built, and Windows: Forensic Schedule Analysis Methods.

Pacing

Pacing is when one party deliberately slows its work because a delay by the other party has created float on its path. In the example above, if the contractor knew the switchgear would be 15 days late and chose to run the rough-in with a smaller crew, the contractor might argue the rough-in slip wasn't a delay at all, but a reasonable response. Owners tend to argue the opposite. Pacing arguments go much better when the contractor documented the decision at the time, in writing, rather than discovering it during the claim.

What helps you later

• Monthly updates that are accepted and accurate. You can't do period-by-period analysis without them.
• A responsible-party code on activities, so owner, contractor, and third-party work can be filtered.
• Notice letters sent when the delay starts, not when it ends. See Delay Notices: What to Include and When to Send Them.
• Daily reports that record what was holding up work, not just who was on site.
• A time impact analysis done close to the event, rather than a reconstruction done years later.

Key takeaways

• Non-excusable: no time, no money. Excusable non-compensable: time only. Excusable compensable: time and money.
• A delay has to hit the critical path, or eat through an activity's float, to extend the project.
• Concurrent delay commonly means time without money for the overlapping period, but contracts and jurisdictions differ.
• Float absorbs part of a late activity's slip; the project delay is often smaller than the activity delay.
• Period-by-period analysis on accepted updates is what turns these categories into actual days.