The Illusion of Growth: 15 Months of “Real” Nonresidential Decline

 

If you feel like you're running up a down escalator, you aren't imagining it. While mainstream headlines often fixate on nominal dollar records in the broader economy, those of us in preconstruction see a different math.

Yesterday's Census Bureau release for December 2025 spending data (finally out after the federal delays) tells a story of a sector that isn't just stalled; it's effectively shrinking when adjusted for the cost of doing business. The Census Bureau noted a 0.3% increase year-over-year in nonresidential spending. When adjusted for escalation, that figure becomes a -3.8% decrease in real spending year-over-year.

How long has this been trending downward?

We have now hit a somber milestone: overall nonresidential spending, when measured in real dollars, has declined year-over-year for 15 consecutive months. This marks every month since October 2024. In 2023, the industry saw a massive surge that outpaced inflation. But since late 2024, the script has flipped. Even as the checks being written stay large, the physical volume of actual put-in-place construction is contracting. We are paying more to build less.

What sectors are driving the decline?

The sharpest declines are concentrated where you might expect them: manufacturing is down 15% year-over-year in real terms, office is off 17.1%, and warehousing has contracted 6.8%. These are all sectors that were either policy-fueled or pandemic-distorted and are now correcting hard. But what makes this cycle particularly concerning is that the traditionally more resilient sectors aren't providing the offset they once did. Private healthcare is down 7.6%, public education is down 5.4%, and commercial is off 3.8%. Even lodging, which held up through much of the post-pandemic period, has declined 5.1%. Infrastructure is the lone bright spot, essentially flat at +0.5%, which while not a recovery, is at least not a drag.

I thought data centers were booming, isn’t that helping?

If there is a floor under this market, it's made of servers. Spending on data centers grew by nearly 24% year-over-year, but they represent only 3.6% of all nonresidential construction spending and 2% of all construction spending. This isn't a "Construction Boom", it's a "Data Center Boom" happening inside a broader nonresidential recession. For those of us in preconstruction, this matters beyond the headline. The market rate for MEP trades is being set by Northern Virginia and Phoenix, regardless of what's happening in your local healthcare or commercial market. A sector representing 3.6% of spending is dictating the capacity and pricing environment for the other 96.4%.

What should we think about moving forward?

While a -3.8% real decline sounds daunting, it marks the beginning of what I call the "Precision Era" of preconstruction, and counterintuitively it is good news for the experts.

We must retire "2023 Logic." That was a market where volume covered a multitude of sins: loose estimates, thin procurement strategies, and speculative scopes all floated on a rising tide. That tide has gone out. What replaces it is a market that rewards the surgical over the speculative. The projects advancing in this environment are higher quality, better funded, and more mission-critical. Owners who are spending in a real-dollar contraction have conviction behind their capital. That means the preconstruction team's job isn't just budget management; it's feasibility defense. The firms that will win in this cycle are the ones that can demonstrate, with real economic precision, that a project pencils out, that the supply chain is de-risk­ed, and that the GMP reflects the market as it actually is, not as the nominal headlines suggest it is.

The escalator might be moving down, but for those of us who have spent years building our preconstruction prowess, this is where the work gets meaningful.

Let's get to work.