Construction Input Costs vs Bid Prices: 8.4% vs 3.5%

Construction input costs vs bid prices is the single comparison that explains most contractors’ year. In its May 2026 release, the Associated General Contractors of America reported the producer price index for inputs to new nonresidential construction up 1.8% for the month and 8.4% year over year, the largest annual jump since the pandemic, while contractors’ bid prices rose only 3.5% year over year.

That is a gap of 4.9 percentage points. Costs moved roughly 2.4 times as fast as the prices contractors were able to charge. Chief economist Ken Simonson noted construction costs were rising at about twice the rate of the consumer price index, which is the reply to any owner who tells you inflation is over.

The gap does not show up as one big loss. It shows up in four places: in the window between bid day and buyout, in the material lines of jobs you already sold, in the margin you thought you had, and in the credit requests you never sent. This post walks each one.

The gap is not a one-month story

One month is noise. Three months in a row is a condition.

Month reported	Inputs to nonresidential construction, year over year	Contractors’ bid prices, year over year	Spread
April 2026	+6.6%	+3.6%	3.0 points
May 2026	+8.4%	+3.5%	4.9 points
June 2026	+7.1%	not reported in the retrievable portion	n/a

April’s release showed inputs up 1.7% for the month and 6.6% year over year against bid prices up 3.6%. AGC economist Macrina Wilkins framed the pricing gap as raising the risk of project delays, which is the honest consequence: when the spread stays open, contractors stop bidding rather than bid at a loss.

June softened at the margin. Inputs fell 0.5% for the month but were still 7.1% above a year earlier. A single down month is not relief.

For scale, the whole of 2025 was quiet by comparison: AGC reported the PPI for materials and services used in nonresidential construction up 3.3% December to December. If you set your escalation allowance off 2025 and have not revisited it, it is running at roughly half of what the input side is doing now.

What is actually driving the input side

The headline is an average. Underneath it, a few categories are doing almost all the work, and whether they are your categories decides whether the 8.4% is your problem or somebody else’s.

AGC’s May 2026 numbers put the year-over-year moves at diesel fuel up 105.9%, aluminum mill shapes up 48.8%, copper and brass mill shapes up 26.8%, truck transportation up 17.3%, and fabricated structural steel up 15.6%. Earlier in the same stretch, AGC reported aluminum mill shapes up 39.1%, steel mill products up 20.9%, structural metal and rebar up 20.0%, and copper and brass mill shapes up 15.1% from February 2025 to February 2026.

You can watch the same moves in the raw BLS series, which is what AGC is quoting. The headline itself is inputs to nonresidential construction, goods, which read 166.9 in January 2026 and 177.5 preliminary in April 2026, a move of about 6.4% in four months. Copper and brass mill shapes ran from 645.990 in January 2025 to 747.384 in January 2026 and 803.275 preliminary in June 2026. Steel mill products went from 315.369 in January 2026 to 361.439 preliminary in June 2026. Plastics pipe, the commodity series that covers PVC pipe, moved from 171.280 in January 2026 to 182.609 preliminary in June 2026. BLS marks recent months preliminary and revises them, so re-pull before you quote a number in a claim.

Two readings fall out of that. If you are a wire and pipe trade, your exposure is worse than the headline. If your material mix is heavier on commodities that have been flat, it is better. Either way the headline is a starting point, not your number.

Where the squeeze shows up, one place at a time

1. The window between bid day and buyout

This is where the gap bites hardest, because it is the only part of the problem that is purely about timing. You priced material on a quote. The award came eleven weeks later. Nothing about the job changed, but the material did.

The mechanical fix is not clever, it is procedural: know your quote validity window, get the buyout done inside it, and treat an expired quote as a repriced quote rather than an inconvenience. If your award cycle is routinely longer than your quotes are good for, that is a structural exposure and no contingency percentage will fix it.

2. The material lines of jobs you already sold

Once the contract is signed at a stipulated sum, index moves become your problem unless the contract says otherwise. This is what escalation clauses exist for, and what AGC has characterized as the number one issue in construction contracts, with the absence of a clause described as a killer clause for a GC on private vertical work.

The standard instrument here is ConsensusDocs 200.1, marketed as the only standard material price escalation clause: it names the specific impacted materials on a project and adjusts the contract price against an agreed objective market index, in both directions. Typical structure pairs a trigger threshold with a cap, and ConsensusDocs commentary cites a sample 3% trigger and points to FAR 52.216-2 as a standard 10% limit.

Read what it costs before you assume it makes you whole. Under 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. A 3% trigger on a year running at 8.4% still leaves the first three points with you.

And it is not free of paperwork. Escalation adjustments are either index-based or cost-based on documented actual cost, and either route requires you to retain initial quotes, bid proposals, purchase orders and invoices. A clause you cannot document is a clause you cannot collect on.

3. The margin you thought you had

Here is the arithmetic, and every number in this section is illustrative. Say a job carries $500,000 of material at bid.

Line	Illustrative figure
Material at bid	$500,000
Cost move at the input rate of 8.4%	$42,000
Recovered at the bid price rate of 3.5%	$17,500
Unrecovered	$24,500

The unrecovered $24,500 is exactly 4.9% of the material base, which is the spread, applied. It comes straight out of job margin. On a job carrying 8% margin, $24,500 is more than half the profit on a $500,000 material package. That is the whole point of tracking the gap rather than the headline: the spread is the loss rate.

Real numbers will differ. Your material mix is not the national mix, your escalation allowance is probably not 3.5%, and part of the input index covers services and fuel you may buy very little of. Run the same three lines on your own job and your own categories.

4. The credits you never asked for

This is the part the national data will never show you, and it is the part you control completely.

An index tells you the market moved. It does not tell you whether your supply house billed you correctly. Those are separate questions and they get conflated constantly, because both of them look like “the price went up” on an invoice. A move from a real commodity run is not a billing error. A duplicate invoice is. So is a quantity that does not match the packing slip, and so is a line priced off the counter default when your account has a contract rate loaded.

In a year where the market genuinely is up 8.4%, the market becomes cover. A price that drifted for no reason blends into a price that rose for a good reason, and nobody checks, because everybody already believes prices are up. That is exactly when price creep is easiest to miss and most expensive to ignore.

Telling market moves from everything else

The test is comparison, and you need two comparisons, not one.

Compare the SKU to itself over time. Same part number, same unit of measure, same branch, four purchases across six months. If your unit price on three-quarter inch copper 90s is up 26% over a year when copper and brass mill shapes are up 26.8%, your distributor is passing through the market and there is nothing to dispute. If it is up 40%, something else is happening and it is a fair question. That question is the entire basis for negotiating with your supply house from documented history.

Compare the invoice to the quote and the PO. Market moves do not explain a line billed at a different price than the quote it was written against, and they do not explain the same invoice arriving twice. Those are quote versus invoice mismatches and duplicate invoices, and they get credited when you catch them promptly.

None of this is an accusation against your distributor. Your supply house is absorbing the same aluminum, copper, steel and diesel moves you are reading about, on its own cost of goods, with its own rebate claims outstanding. The problem is the silence. Nobody called to tell you which of your prices rose and why, and in a fast market the notice gets skipped.

What to actually do this quarter

Reset your escalation allowance off current data, not last year’s. A 2025 allowance built on a 3.3% year is not sized for a 7 to 8% run rate.
Weight the index by your own material mix. Pull your top ten SKUs by annual spend, map them to the categories AGC breaks out, and weight the moves by your dollars. That number is your rate, not the headline.
Shorten the gap between quote and buyout. Every week between them is unpriced exposure, and it is the cheapest thing on this list to fix.
Get an index into the contract where you can. Objective, published, narrow enough to actually cover your scope. If you are picking one, the tradeoffs between a broad construction cost index and a materials-specific PPI are worked through in what the ENR Construction Cost Index measures and misses.
Bring the market data to the owner conversation. The monthly release is a third-party document with a named economist on it, and it argues better than you do. How to assemble it is covered in AGC’s construction inflation data as a bid input.
Audit the invoices anyway. The gap you can recover fastest is not the 4.9 points against the national index. It is the lines on last month’s invoices that were simply wrong, and those get credited on a phone call when they are fresh.

The short version

May 2026: construction input costs up 8.4% year over year, contractors’ bid prices up 3.5%. A 4.9-point spread, costs moving roughly 2.4 times as fast as prices.
April was 6.6% against 3.6%. June inputs fell 0.5% for the month and were still up 7.1% year over year. The condition is persistent, not a spike.
Fuel and metals are doing most of the work. Whether that is your exposure depends on your mix, not on the headline.
The spread applied to your material base is your loss rate. On an illustrative $500,000 material package it is $24,500.
A hot market makes unannounced price drift and billing errors invisible, because everything looks like inflation.
Compare each SKU to its own history, and compare every invoice to its quote and PO. Those two checks separate the market from the mistakes.