ENR Construction Cost Index Explained for Contractors

The ENR Construction Cost Index, usually written CCI, is a construction cost index published by Engineering News-Record and it turns up in contracts constantly, usually as the reference index in an escalation clause or a price adjustment provision. Procore’s escalation guidance lists it alongside CPI and PPI as the standard index-based options when an escalation adjustment is not being computed off documented actual cost. That is what it is for: settling, by formula, how much a contract price moves when the market moves.

What it is not is a measure of what happened to your material package. A blended construction cost index is built to describe general construction cost movement over time. Your exposure is concentrated in a handful of commodities that move much faster and much less evenly than any blend. Naming a broad index in a clause and assuming it will make you whole on copper is the single most common way an escalation clause turns out to be worth less than the contractor thought.

One disclosure up front. This post does not reproduce ENR’s published index values or the composition and weights of its basket. Those live behind ENR’s own publication, and every number in a contract exhibit should come from the publisher rather than from a secondary summary. Read the current values and the methodology at the source, then come back to the decision framework below.

What an index actually does in a contract

Strip the jargon and an index-based escalation clause is a formula with four parts: a named index, a baseline reading at a stated date, a settlement reading at a stated date, and a rule for converting the change between them into dollars.

ConsensusDocs 200.1, marketed as the only standard material price escalation clause, is the clearest published example of the shape. It names the specific impacted materials on a project and adjusts the contract price against an agreed objective market index, in both directions. That last phrase matters: a real escalation clause is symmetric, and owners are far more willing to sign one that gives money back when the index falls.

Around that core sit two more parameters. ConsensusDocs commentary describes the standard pairing of a trigger threshold with a cap, citing a sample 3% trigger and pointing to FAR 52.216-2 as a standard 10% limit. Construction Dive’s summary of the field describes three structural variants: day-one recovery, time-delayed recovery where the first stretch of the job is at the contractor’s risk, and threshold-based risk sharing above a stated percentage.

So the index is one of five decisions. It is the one people argue about, and it is not always the one that costs the most. A 3% trigger in a year where inputs are running high leaves the first three points with you regardless of which index you named.

Why any general construction cost index will miss your materials

Here is the structural problem, and it applies to every blended index, not just ENR’s.

A general construction cost index is an average across the things construction consumes. Your job is not an average. In a market where categories move together, the difference does not matter much. In a market where they scatter, it matters enormously, and this is a scattered market.

Look at how far apart the categories have been running. In its May 2026 release AGC reported the producer price index for inputs to new nonresidential construction up 8.4% year over year, with diesel fuel up 105.9%, aluminum mill shapes up 48.8%, copper and brass mill shapes up 26.8%, and fabricated structural steel up 15.6%. One headline number, and underneath it a spread from single digits to triple digits. How to read those monthly releases and weight them to your own mix is covered in AGC’s construction inflation data as a bid input, and the widening gap between input costs and what contractors can charge is the subject of input costs up 8.4% against bid prices up 3.5%.

If your material package is heavily copper and aluminum, an index that settled at the blended rate recovered a fraction of what you actually absorbed. If your package is light on metals, the same index would have overpaid you and the owner would have noticed. Neither outcome is the index being wrong. It is the index answering a different question than the one your job asked.

Three specific mismatches to look for:

Labor weight versus material weight. A general construction cost index reflects the cost of building things, which includes what it costs to hire people to build them. A material escalation clause is trying to cover material. When those two components move at different speeds, an index that blends them tracks neither cleanly. If you are a trade contractor asking for escalation on a pipe package, ask whether the index you are being offered is even mostly about material.

Basket composition versus your takeoff. Any index prices a fixed basket. If your scope is copper tube and fittings, or sheet metal duct, or PVC, the relevant question is whether the commodities in the index are the commodities in your takeoff, in roughly the proportions your takeoff carries. Reputation has nothing to do with it.

Geography and timing. National indexes average regional markets. Published indexes also lag, and monthly series get revised. Both facts have to be handled in the settlement date language or you will be arguing about which reading applies.

The realistic alternatives, side by side

The choice is not “ENR or nothing.” Here is the actual menu, and the tradeoffs are real in both directions.

Option	What it tracks	Where you get it	Best fit	Main weakness
A general construction cost index (such as ENR’s CCI)	Broad construction cost movement over time	The publisher, on the publisher’s terms	Whole-project escalation on a GC contract; owners recognize the name	Blends categories and includes components your scope may not carry
Inputs to new nonresidential construction PPI	Goods purchased as inputs to nonresidential construction	BLS, series WPUIP2312001, free and public	Whole-project material escalation where you want a materials-weighted number	Still an average; includes fuel and categories you may not buy
A commodity-level PPI	One material family	BLS, for example copper and brass mill shapes, steel mill products, sheet metal products, plastics pipe	Trade contracts with a concentrated material exposure	Narrow enough that a scope change can strand the clause
An industry PPI as a backstop	One manufacturing industry	BLS, for example air-conditioning and heating equipment manufacturing	A whole product category where the commodity series is too narrow for your scope	Industry output prices, not the exact product you buy
Cost-based adjustment on documented actual cost	Your actual invoices	Your own records	The most accurate recovery available	Requires clean documentation and invites line-by-line audit

ConsensusDocs commentary explicitly names BLS monthly publications as an objectively verifiable index source, which is the point in favor of the middle three rows: they are free, public, and nobody has to buy a subscription to check your math at settlement.

A note on the last two rows. The industry index is not the only option for HVAC equipment and it is usually not the best one. BLS publishes commodity series for unitary air-conditioners, except air source heat pumps (WPU114802) back to December 1977 and heat pumps (WPU11480734) back to December 2009. Name the one that matches what you install, keep PCU333415333415 as the backstop where a scope spans both, and note that WPU114802 publishes about a month behind most PPI series, so a clause tied to it needs a lag allowance in the settlement date language. The real problem to avoid is naming an index that does not measure the thing you buy. Copper wire and cable has that problem: WPU10260314 is one series titled simply copper wire and cable, and it will not isolate the THHN or MC on your job. Plastic pipe is another case where the commodity series does exist: BLS publishes plastics pipe (WPU07210603) and plastics pipe fittings and unions (WPU07210604) as commodity series back to December 2005, so name those rather than the industry index, and know that neither one separates PVC from CPVC from PEX. Know that before you write a series ID into a contract and discover at settlement that it does not measure quite what you assumed.

Questions to answer before you name any index

Work through these with the actual document in front of you. Every one of them has bitten somebody.

What exactly does this index measure, in the publisher’s own words? Not what everyone says it measures. Read the methodology.
Can both parties access it at settlement time, for free or at a cost they have already agreed to bear? A subscription-gated index is fine if both sides can see it. It is a problem when only one side can.
How is the index published, and how often? Monthly, and with what lag?
Does the publisher revise? BLS marks recent months preliminary and revises them. Say in the clause which reading governs: first publication or final revised.
What is the baseline date, and does the contract state a specific published reading rather than “the index at time of bid”? Ambiguity here is what turns settlement into a dispute.
Does the index actually correlate with your scope? Test it. Pull twelve months of your own unit prices on your top five SKUs and compare the trend to the index over the same twelve months. If they diverge badly, you have your answer. Building that history is a separate discipline, and the method is in material price creep.
What is the trigger, what is the cap, and who carries the first slice? These frequently cost more than the index choice.
Is it symmetric? If the index falls, does the owner get money back? Say so explicitly. It is the concession that gets the clause signed.

What the index will never settle

Even a perfectly chosen index answers exactly one question: did the market move, and by how much. It leaves three questions completely open, and all three of them are where money actually goes missing.

Did your supplier’s price to you move with the market, or past it? An index is a national aggregate of transactions. It says nothing about your account, your multiplier, or your branch. If copper and brass mill shapes ran up 26.8% year over year and your fitting prices ran up 40%, the index has not caught that and never will. Only comparing a SKU to its own purchase history does. And if a line got priced off the counter default because an agreement lapsed or a part number got substituted, that is not inflation at all, it is a contract price that did not apply.

Did the invoice match the quote? An escalation clause covers market movement between bid and buy. It does not cover a line billed at a different number than the quote it was written against. That is a quote versus invoice mismatch, and it gets corrected with a credit request, not a change order.

Can you prove the movement hit this job? An index reading proves the market moved. It does not prove your job absorbed it. Procore’s guidance is to retain initial quotes, bid proposals, purchase orders and invoices to support a claim, and that is the real work. Under ConsensusDocs 200.1 the contractor gets no overhead and profit on an equitable adjustment, impacts caused by the contractor or its own subs and suppliers are excluded, and both parties carry a duty to mitigate. An owner reading that language closely will ask you to demonstrate the exclusions do not apply, and only documents do that.

None of this is a knock on your distributor. When metals and fuel are moving at the rates AGC has been reporting, your supply house is passing through cost it did not choose, on its own margin, with its own rebate claims outstanding. The reason to keep your own price history is that a clause settles on evidence, and your invoices are the only evidence that is specifically about your job.

If you are the one drafting the exhibit

Keep it short and mechanical. A workable escalation exhibit names:

The specific materials covered, listed, not described in general terms.
The exact index, including series identifier where one exists, and the publisher.
The baseline reading, by publication month and value.
The settlement rule: which published reading governs, and how revisions are handled.
The trigger percentage and the cap.
Symmetry, stated explicitly.
The documentation you will provide, and the documentation the owner may request.

AGC has characterized price escalation and supply chain disruption as the number one issue in construction contracts, and the absence of an escalation clause as a killer clause for a GC on private vertical work. That is the argument for having one at all. The argument for having a well-drafted one is that a vague clause and no clause settle for about the same amount.

The short version

The ENR Construction Cost Index is a general construction cost index and one of the standard named options in index-based escalation clauses. Read its values and methodology at ENR itself; this post deliberately does not reproduce them.
Any blended index averages away the category dispersion that is currently doing all the damage: one recent monthly release ranged from single-digit to triple-digit year-over-year moves under an 8.4% headline.
For material-specific exposure, a BLS input or commodity PPI is usually the better fit, and it is free and public so both sides can check it.
The index is one of five decisions. Trigger, cap, baseline, settlement rule and symmetry often cost more than the index choice.
Test any candidate index against twelve months of your own unit prices before you sign it.
No index catches unannounced price drift on your account or a billing error on your invoice. Only your own purchase history does that.