The Spread Narrows. The Gap Remains.

The Bureau of Labor Statistics released June 2026 Producer Price Index data this week. The headline for construction: nonresidential input costs decelerated to 7.1% y-y, down 130 basis points from May's 8.4% print. The chart below tracks two lines. One tracks wholesale resource costs; the other tracks final bid pricing. In June, outputs held at 3.5%. The spread between those two figures is 360 basis points.

The gap is narrower than last month, but has not closed.

What is driving the deceleration in inputs?

America may run on Dunkin’, but construction runs on diesel. June's CPI report showed energy costs pulling headline inflation lower, and construction inputs followed. Diesel fell sharply in June, and fuel-sensitive categories moved with it. That is the cooldown visible in the chart, not a broad retreat in materials pricing.

When energy leads a deceleration, relief is real but it is also the most reversible and volatile component of the index.

Did anything move in the other direction?

Yes, but the story is not a uniform march upward. It is a market deeply bifurcated by project type and material category.

Base bulk commodities that drove significant escalation over the last three years have stalled or reversed. Cement registered -0.6% y-y in June. Gypsum building products came in at -1.1%. If your project is a standard warehouse shell or a drywall-heavy commercial build-out, those supply chains are providing real, measurable relief.

That is precisely where the read gets complicated. While cement and gypsum flatline or decline, the sub-components of steel tell a different story than the fabricated aggregate suggests. Joists and rebar are up 19.6% y-y and added another 3.1% in June alone. Pipes and tubes are up 9.0% y-y. Stainless steel is up 6.4%. Plastics and resins, critical to mechanical and process systems, are up 16.2% y-y, though June showed a 2.4% monthly pullback. These categories carry disproportionate budgetary weight in complex healthcare, lab, or data center projects relative to a general commercial project.

The fabricated structural steel aggregate tells a different story at 1% y-y in June, a sharp deceleration from 7.9% in May. That figure warrants scrutiny rather than comfort. As documented in my post on structural steel lead times, import suppression and domestic mill order book concentration are driving 36-to-40-week fabrication windows that do not yet fully appear in transaction-based price indexes. The aggregate may be lagging conditions on the ground rather than contradicting them.

Interior finishing categories show similar persistence. Paint and coatings are up 5.9% y-y. Carpet and rugs held at 3.5%, matching the output index exactly while sitting entirely on the input side of the ledger.

The aggregate deceleration in the headline input figure is being flattered by the drop in diesel and a cooling in base bulk commodities. For owners and estimators working on technically complex or interior-intensive projects, the categories that determine whether a budget holds are still climbing.

So is the spread still a problem?

360 basis points is still a structural problem. Last month I characterized 490 basis points as a spread in which the difference between inputs and outputs must land somewhere, and it lands on margins. That has not changed. The arithmetic is slightly less severe. The direction has not reversed.

Contractors are still absorbing more cost than they are passing through. That gap has now persisted for 14 straight months. Competitive pressure outside the data center sector continues to keep firms active in bid rooms even as input costs run above bid pricing. The margin compression thesis remains intact.

What should we do with this data?

The deceleration in inputs is worth noting, but it does not change preconstruction planning. Three things follow from this month's data.

First, do not use the narrowing spread as a signal to reduce escalation assumptions. A 360 basis point gap between what contractors pay and what they charge is still a gap, and the categories driving it on the materials side did not participate in the June cooldown.

Second, treat the energy component as a separate variable. Diesel and fuel-sensitive costs are the mechanism behind this month's input deceleration. They do not belong in the same escalation model as structural steel or process piping. One is mean-reverting over short cycles while the other reflects capacity and tariff dynamics that move on a different timeline entirely.

Third, keep the spread itself as the primary variable to watch. The direction of travel matters more than any single monthly print. Inputs moving from 8.4 to 7.1% is movement in the right direction. It becomes meaningful when outputs begin to close from the other side, and there is no evidence of that yet at 3.5%.

Where does this leave the broader picture?

One data point is not a trend. June's deceleration in nonresidential construction inputs is a real and measurable development. It is also almost entirely explained by a single volatile category that has reversed faster than it fell in prior cycles. The materials that carry the most weight in a healthcare or life sciences budget did not participate in that relief.

The spread between inputs and outputs has persisted for 14 consecutive months. Until outputs move or the materials-driven portion of inputs decelerates alongside energy, the correct read on this data is that conditions improved at the margin in June, not that they corrected.

The next PPI release is scheduled for August 13.

Series referenced: PPI for Nonresidential Construction Inputs; PPI for Nonresidential Construction Outputs; PPI for Carpet and Rug Mills: Carpet and Rugs [PCU3141103141100]; PPI for Paint and Coatings Manufacturing [PCU3255132551]; PPI for Cement [WPU1322]; PPI for Gypsum Building Products [WPU13710102]. Sub-component steel and plastics data drawn from the BLS PPI Detailed Report, June 2026. Source: U.S. Bureau of Labor Statistics, Producer Price Index News Release, June 2026, released July 15, 2026.