Steel Consuming Market Outlook March 2026: Construction, Housing, and HVAC Trends

The steel-consuming market tied to construction and housing is entering 2026 in a state that defies easy summary. It is neither the broad expansion of 2021 nor the sharp contraction that some analysts feared after a turbulent 2025. Instead, what is unfolding looks more like a selective reset, one where the strength of individual segments matters far more than headline aggregate figures, and where the composition of building activity is shifting in ways that carry meaningful implications for steel producers, distributors, and service centers.

On one side of the ledger, nonresidential construction planning is normalizing. The Dodge Momentum Index, the most closely watched leading indicator for commercial and institutional building, has posted back-to-back monthly declines to open 2026, retreating from the record surge it logged through the second half of last year. On the other side, residential housing construction is finding its footing again. US housing starts rose 7.2% month-on-month to a seasonally adjusted annual rate of 1.487 million in January 2026, up from a downwardly revised 1.387 million in December and well above forecasts of 1.35 million, marking the third consecutive monthly increase and lifting starts to their highest level since February 2025. The driver of that strength is notable: multi-family construction, which is consistently more steel-intensive per unit than detached single-family building, has surged with particular force.

Meanwhile, the HVAC market, a segment deeply intertwined with both construction cycles and residential activity, is experiencing one of its sharpest corrections in a decade. Overall shipments of air conditioning units and heat pumps in 2025 were down by 20% compared to the previous year, impacted by the transition away from R-410A, an uncertain economy affected by tariffs, high interest rates, the termination of federal tax credits, and inventory destocking. That backdrop creates a more nuanced steel demand outlook than either the optimists or the pessimists are currently pricing in.

The core theme of this analysis is that 2026 is a normalization phase, not a collapse. Structural building demand appears resilient in pockets, particularly in multi-family housing, data centers, and energy infrastructure. Mechanical and equipment-related steel demand is in a correction. For anyone supplying steel into construction and housing, understanding which side of that divide their end markets sit on is the most important analytical task of the year.

The Dodge Momentum Index is one of the construction industry's most reliable leading indicators. It is a monthly measure based on the three-month moving value of nonresidential building projects going into planning, shown to lead construction spending for nonresidential buildings by a full year to 18 months. Because it captures the early-stage planning activity of developers and owners, it provides a forward window into construction spending that most other indicators cannot. When the index rises, it generally signals that more projects are moving through the pipeline and that construction activity will accelerate roughly a year to a year and a half down the road. When it falls, it raises the prospect of a slower workload ahead.

What the index does not capture, however, is the absolute level of projects already under construction or deep in the permitting process. Projects that were planned during 2024 and 2025, when the index was surging, are still moving forward regardless of what new planning activity looks like today. That distinction matters enormously for the steel market, because steel demand tracks actual construction activity, not planning intention.

The Dodge Momentum Index declined 7.3% in February 2026 to 250.0 (2000=100) from the downwardly revised January reading of 269.8. Over the month, commercial planning fell 8.9% and institutional planning momentum slowed by 4.0%. That followed an already notable decline the month before. The DMI declined 6.3% in January to 272.7 from the downwardly revised December reading of 291.0, with commercial planning falling 7.2% and institutional planning momentum slowing by 4.4%. Together, the two months represent a cumulative pullback of roughly 14% from December's elevated reading, erasing a substantial portion of the gains accumulated during the strong second half of 2025.

To put the current reading in context, it is worth noting how far the index rose before this pullback began. In 2025 as a whole, the DMI was up 37% from the average reading in 2024, with the commercial portion up 35% and the institutional portion up 43% over the same period. A correction of 14% from that peak, while notable, still leaves the index at historically robust levels. Despite the recent declines, the index still shows 26% annual growth for commercial projects and 34% for institutional projects compared to January 2025, and the overall index is up 29% year over year. In other words, the planning environment is still meaningfully stronger than it was a year ago, even after the recent softness.

The causes of the January and February declines are worth understanding because they reveal which segments are pulling back and which are holding firm. Planning momentum cooled across most commercial and institutional sectors, with data center projects continuing to lead the way, but most nonresidential sectors easing into a more sustainable growth pattern. In February specifically, warehouses were the sole bright spot in the commercial category, with all other commercial and institutional sectors seeing slowing momentum.

The reasons behind the broader pullback are not hard to identify. Elevated risks around costs, labor, and geopolitics are continuing to constrain builder confidence in the near term, according to Sarah Martin, Associate Director of Forecasting at Dodge Construction Network. Recent tariff increases on steel, aluminum, and some copper products, in some cases reaching 50%, are showing up directly in bid prices, with many contractors reporting at least one project in the past year that was canceled or scaled back because updated material quotes pushed the budget beyond what the market was willing to accept. That kind of cost friction is a legitimate deterrent to new planning commitments, even when underlying demand for space is strong.

Critically, the data center sector continues to anchor the index at elevated levels even as other segments pull back. The largest commercial projects to enter planning in February included the $500 million CyrusOne Data Center in Whitney, Texas, the $448 million TX12 Data Center in San Antonio, Texas, and two QTS DFW2 Data Center buildings in Wilmer, Texas each valued at $290 million. These are not marginal projects. They represent billions of dollars in future construction spending and substantial steel demand, since data centers rely on steel for structural framing, server trays, cable racks, mechanical supports, and exterior enclosures, with some estimates suggesting 500 to 8,300 server racks per facility.

Perhaps the most important takeaway from the current Dodge data is what the still-elevated index level implies for the medium-term outlook. The robust planning pipeline suggests an acceleration in construction spending in 2027, according to Dodge's own forecasters. That is a meaningful signal for steel market participants with a planning horizon beyond the next two quarters. While near-term order books may reflect some hesitation from developers, the structural pipeline built up during the 2025 planning surge should translate into genuine construction activity, and therefore genuine steel demand, as those projects move from permits to ground-breaking over the next 18 months.

Residential construction is one of the most direct drivers of steel demand, absorbing rebar for foundations and retaining walls, structural steel and metal decking in multi-family buildings, wire rod for mesh reinforcement, and steel-intensive building components including roofing, mechanical systems, and appliances. The January 2026 housing data, released by the Census Bureau on March 12, offers a picture that is genuinely encouraging in the headline but more complicated in the details.

Privately-owned housing starts in January were at a seasonally adjusted annual rate of 1,487,000, up 7.2% above the revised December estimate of 1,387,000 and 9.5% above the January 2025 rate of 1,358,000. Three consecutive monthly gains mark the most sustained improvement in residential construction momentum since early 2025, and the January reading came in well above the consensus forecast of approximately 1.35 million units, suggesting that the market is outperforming expectations as financing conditions gradually ease and pent-up demand builds.

The composition of that growth carries enormous significance for steel consumption. While single-family starts decreased 2.8% to a 935,000 seasonally adjusted annual rate, the multifamily sector increased 30% to an annualized 552,000 pace. The National Association of Home Builders chairman noted that the single-family market has slowed as builders continue to deal with elevated construction costs while affordability conditions are a cause of concern for many potential home buyers. Single-family affordability remains constrained by the interaction of still-elevated mortgage rates and home prices that have not corrected meaningfully in most major markets, pushing many would-be buyers to the sidelines and keeping builder confidence subdued in that segment.

Multi-family construction, by contrast, is benefiting from strong urban rental demand, a large backlog of projects that received financing before rates peaked, and the structural reality that many markets simply do not have enough rental units. The year-over-year gain in multifamily starts is particularly striking, with multifamily starts up 29.1% month-over-month and 56% year-over-year, representing a powerful demand signal for structural and reinforcing steel in dense urban markets.

The shift toward multi-family housing is not just a quantity story. It is a quality and intensity story that has direct implications for steel consumption per unit built. Mid-rise and high-rise apartment buildings use far more steel per unit than detached single-family homes, for several interconnected reasons. First, structural loads in multi-story buildings require steel framing, columns, and beams that simply do not appear in wood-framed single-family construction. Second, concrete cores and floor plates in taller buildings require substantial quantities of reinforcing bar, wire mesh, and post-tensioning components. Third, multi-family buildings incorporate sophisticated mechanical, electrical, and fire protection systems that have a higher steel content than the simpler systems found in single-family homes.

A concrete example helps illustrate the scale of this effect. The Meta Mesa Data Center in Mesa, Arizona, which consumed approximately 12,000 tons of steel, included structural shapes for beams and columns, joist and deck, rebar in the foundation, and galvanized steel products for miscellaneous use. While a data center is not a residential building, it demonstrates how steel-intensive large-format, multi-story structures are relative to smaller building types. A 300-unit mid-rise apartment building in a major urban market can easily consume several thousand tons of structural and reinforcing steel, a figure that simply does not scale from comparable numbers of single-family homes.

Regionally, construction activity in January climbed sharply in the South, up 11.4% to 810,000, and the Northeast, up 47.4% to 196,000, though it fell in the West by 7.5% and the Midwest by 10.8%. The geographic divergence suggests that the current housing recovery is not uniform, and that steel distributors serving the South and Northeast may see more immediate demand benefits than those concentrated in Western markets.

The positive start data comes with an important caveat from the permit side. Privately-owned housing units authorized by building permits in January were at a seasonally adjusted annual rate of 1,376,000, down 5.4% below the revised December rate of 1,455,000 and 5.8% below the January 2025 rate of 1,460,000. Single-family authorizations were similarly soft, and single-family permits were down 0.9% month-over-month and 11.6% year-over-year to a rate of 873,000.

Since building permits lead housing starts by roughly one to four months, a sustained decline in permit activity will eventually translate into fewer starts. Senior Economist Joel Berner characterized the data as a "mixed bag," noting that building permits paint a pessimistic picture of new construction to come. This is the central tension in the residential construction outlook: current activity is strong and supports near-term steel demand, but the forward-looking signal from permits suggests that builders are not aggressively committing to new projects at the rate needed to sustain the current pace through the back half of 2026.

The reasons for permit caution are well understood. Mortgage rates remain meaningfully above the 3% lows that defined the pandemic-era boom. Construction costs are roughly 44% higher than in 2020, which compresses builder margins and makes it harder to price new homes competitively. And while multifamily permits did show a year-over-year improvement of 8.9%, the month-over-month decline of 13.4% suggests that even the most resilient segment of the housing market is not immune to near-term headwinds.

The HVAC market occupies an important but sometimes overlooked position in the steel-consuming ecosystem. HVAC equipment cabinets, heat exchangers, refrigerant coils, structural supports, ductwork connections, and associated infrastructure all incorporate steel, primarily flat-rolled and coated sheet products. In aggregate, HVAC manufacturing represents a meaningful slice of steel demand tied to the construction and housing cycle, and its performance in 2025 was by any measure a significant negative surprise.

US combined shipments of central air conditioners and air-source heat pumps in 2025 totaled just over 7.7 million units, a 20% decline from roughly 9.7 million in 2024, according to the latest AHRI data, representing the fewest combined units shipped since 2016. That single statistic captures the magnitude of the correction. Seven years of cumulative market growth were effectively unwound in a single year, driven by a confluence of factors that are worth examining individually because they determine how quickly a recovery might materialize.

The most significant structural factor was the refrigerant transition. The January 2025 cutoff for manufacturing R-410A systems prompted massive pre-buying throughout the second half of 2024, as distributors and contractors stocked up on equipment before the transition to lower-global-warming-potential refrigerants like R-454B. Carrier CEO Dave Gitlin attributed much of the 2025 downturn to excess channel inventory left over from the pre-buy of equipment due to the refrigerant transition, which was more than anticipated. That inventory overhang suppressed new orders throughout most of 2025 as the distribution channel worked through its elevated stock levels.

Beyond inventory, the market faced genuine demand-side headwinds. High interest rates suppressed new construction activity, reducing installations in new buildings. Energy efficiency regulation changes removed some of the urgency that had previously driven early replacement cycles. And consumer caution, particularly among existing homeowners managing tight household budgets, led many to defer non-emergency system replacements. The consequences were severe: shipments of air conditioning and heat pump units fell between 26% and nearly 50% year over year during the critical summer months, with September's 49% plunge in air conditioning shipments marking the low point of the season.

Major HVAC manufacturers were candid in their assessments of the damage. Carrier closed 2025 with a nearly 40% decline in residential HVAC business, while Lennox reported overall revenue declined 11% in the fourth quarter, including a 21% decrease in Home Comfort Solutions revenue. These are not minor fluctuations. They represent the kind of structural demand reset that typically takes multiple quarters to work through, and Carrier's leadership was notably cautious about the pace of recovery. Gitlin noted that the residential US market consists of about 130 million installed HVAC systems, with a typical annual replacement rate of around 6%, or roughly 8 million units, and that while a return to the norm in 2026 would be tremendous, he did not believe the market would get back to those levels in just one year.

There is a silver lining buried in the data, however. Air conditioning units outsold heat pumps for the full year 2025, but heat pumps outsold air conditioning units each month from September to December, suggesting that the longer-term electrification trend toward heat pumps remains intact even within a broadly weak market. And the commercial HVAC segment is a genuine bright spot: Carrier reported that its data center revenue doubled to $1 billion in 2025 and that its backlog stretches into 2028, while Trane reported commercial HVAC bookings up 22% from Q4 2024, with overall backlog of $7.2 billion, more than 90% of which is commercial. The same data center and energy infrastructure boom that is sustaining the Dodge Momentum Index at elevated levels is also driving robust commercial HVAC demand, providing at least a partial offset to the residential weakness.

For steel consumption, the HVAC correction matters in two ways. First, lower production volumes directly reduce demand for flat-rolled and coated steel used in unit housings, heat exchanger cabinets, and ancillary components. The flat-rolled steel market, already navigating mixed signals from automotive and appliance demand, loses another meaningful demand anchor when HVAC shipments fall 20% in a single year. Second, and more subtly, soft HVAC installation activity can signal reduced construction fit-out work broadly, affecting demand for pipe, ductwork, fasteners, and related steel-intensive materials that accompany full system installations.

The partial recovery path, as construction and housing starts gradually improve and channel inventory normalizes, should begin to support HVAC demand in the latter part of 2026. But the base effect of the refrigerant transition is unlikely to fully reverse within a single year, and builders of new multi-family projects that are breaking ground today will not be installing mechanical systems for another 12 to 24 months. The recovery in HVAC-related steel demand is real but lagged.

Looking across nonresidential construction, residential housing, and HVAC equipment simultaneously, a clear pattern emerges. Steel demand tied directly to building structures, foundations, and framing is in considerably better shape than steel demand tied to mechanical equipment and fit-out. That divergence has practical implications for which steel products and which consuming industries are likely to see the most pressure versus the most resilience in 2026.

The structural side of the equation benefits from several concurrent tailwinds. The multi-family housing surge is generating real demand for rebar, structural shapes, and metal decking. The data center construction boom, which continues to anchor the Dodge Momentum Index despite the broader planning pullback, is consuming steel at a rate that few other building categories can match per project. US steel demand is forecast to grow approximately 1.8% in 2026, extending the recovery from 2025, supported by federal infrastructure spending, manufacturing reshoring, and energy investments. And the energy sector, which lies somewhat outside the core scope of this analysis but intersects with it meaningfully, is adding substantial demand for structural steel through LNG terminals, pipeline expansions, and renewable energy infrastructure.

The equipment side faces more headwinds. HVAC shipments are at multi-year lows, appliance demand remains subdued, and the general commercial construction softening seen in the Dodge data implies less near-term demand for the mechanical systems that accompany office, retail, and institutional projects. The AGC's 2026 Construction Hiring and Business Outlook found that contractors reported significantly fewer opportunities to bid on projects compared with last year, while also grappling with rising material costs, labor shortages, and uncertainty tied to tariffs and immigration enforcement, with far more mixed expectations than in 2025.

No analysis of steel-consuming construction markets in 2026 is complete without acknowledging the cost environment. Steel prices themselves are not falling sharply, which creates a complicated dynamic. Nationally, nonresidential construction costs rose 1.05% over the past quarter and 7.35% over the previous two years, with key increases including steel framing erection at 3.8%, plumbing systems at 3.3%, and HVAC at 1.9%. These cost pressures are not catastrophic, but they are real, and they are being compounded by tariff-related uncertainty that makes it difficult for developers and contractors to lock in budgets with confidence.

Rebar prices are forecast to fall 1.8% in 2026 while structural shapes are expected to decline 6.1%, according to S&P Global Market Intelligence estimates, which would offer some relief to construction budgets even as broader material costs remain elevated relative to pre-pandemic baselines. For steel mills and service centers, moderating prices combined with uncertain volume growth creates a challenging margin environment, particularly in segments tied to the softer parts of the nonresidential market.

One factor that does not always receive sufficient attention in steel demand analyses is the construction labor market. Associated Builders and Contractors estimate that the industry will need about 349,000 net new workers in 2026, down from more than 500,000 in recent years due to more modest construction growth, with most new worker demand stemming from retirements. While this is an improvement over the acute shortages of prior years, skilled labor constraints remain a genuine binding constraint in markets like data center construction, where the convergence of complex electrical, mechanical, and structural work creates extreme demand for specialized tradespeople.

Labor availability can slow the translation of planning activity into actual construction starts, which matters for the timing of steel demand even when the underlying project pipeline is healthy. A project that enters planning in early 2026 may not break ground until late 2026 or early 2027 not because financing or demand is absent, but because the labor force to build it is already committed elsewhere. That dynamic reinforces the view that the robust Dodge Momentum Index pipeline will translate into real construction activity, but on a timeline that may extend further than the 12-to-18-month historical average.

One of the most important structural shifts reshaping steel demand in construction is the scale of data center investment. Unlike previous technology booms that were relatively light on physical infrastructure, the current wave of AI-driven data center construction is extraordinarily materials-intensive. Projects such as the Meta Mesa Data Center in Mesa, Arizona, are partially operational, with Meta investing about $1 billion in the facility and consuming an estimated 12,000 tons of steel in structural shapes for beams and columns, joist and deck, rebar in the foundation, and galvanized steel products for miscellaneous use.

The scale of the data center pipeline in early 2026 is staggering. Amazon has announced a $15 billion expansion of its New Carlisle, Indiana facility, described as the state's largest project. Multiple hyperscale facilities from Google, Microsoft, and Meta are in various stages of planning and construction across the Sun Belt and Mid-Atlantic. Carrier's data center HVAC backlog stretches into 2028, which is perhaps the clearest single signal that this is not a short-cycle phenomenon but a multi-year structural demand driver that will continue absorbing steel at high volumes well into the back half of the decade.

For steel suppliers, data centers represent a fundamentally different customer profile than traditional commercial construction. Projects are large, concentrated, and frequently sponsored by investment-grade technology companies with strong balance sheets and long-term commitments. That reduces the credit and project-cancellation risk that is always present in developer-driven commercial construction. It also creates more predictable demand for specific steel products, particularly structural shapes, rebar, and galvanized sheet, enabling more efficient supply chain planning.

The construction and housing steel-consuming market in early 2026 is best understood as a normalization rather than a downturn. The Dodge Momentum Index has given back some of its exceptional 2025 gains, but the index is still up 29% year over year, and the overall planning pipeline remains robust enough to support an acceleration in construction spending in 2027. Residential housing construction is on a meaningful upswing, with three consecutive monthly gains in starts and a powerful surge in multi-family projects that represent some of the most steel-intensive building activity in the residential sector. And even the weak HVAC market contains early signals of eventual recovery, anchored by strong commercial and data center demand that is keeping major manufacturers' order books healthy even as the residential business corrects.

For steel market participants, the practical implication is the need to be selective and data-driven. Growth pockets, including multi-family housing in the South and Northeast, data center construction across multiple regions, energy infrastructure, and selected healthcare and institutional projects, are genuine and durable. Softer segments, including traditional office construction, HVAC residential equipment manufacturing, and the replacement market for mechanical systems, require more caution. As presented at the Tampa Steel Conference in early 2026, steel demand will remain uneven, driven by regional and sector-specific volatility. That unevenness is not a temporary disruption but a structural feature of where we are in the construction cycle.

Three questions will determine whether this normalization eventually transitions into renewed broad expansion or settles into a slower but sustainable plateau. First, will housing permits stabilize and recover, providing the forward momentum needed to sustain the current improvement in residential construction into late 2026 and 2027? Second, can the Dodge Momentum Index find a floor as cost and geopolitical pressures ease, allowing the exceptional project pipeline built in 2025 to translate cleanly into ground-breakings? And third, when will HVAC shipments bottom out and begin to align with the improving construction fundamentals that are gradually taking shape? The steel market's transition from normalization to expansion depends on the answers unfolding in the right sequence.

Call to Action: If you are active in construction, steel supply, or HVAC manufacturing, the strategic question for 2026 is not whether the market is strong or weak in aggregate but where within it you are positioned. Are your end markets concentrated in the growing segments of multi-family housing, data centers, and energy infrastructure? Or are they weighted toward the more challenged areas of traditional commercial building and residential equipment? Mapping your exposure to that distinction, and adjusting your strategies accordingly, is the most important analytical work you can do in the months ahead.

Sources

^{Disclaimer
The content provided in this article is for general informational purposes only and does not constitute financial, legal, or professional advice. Readers should seek consultation with qualified professionals before making any financial, investment, or legal decisions. We disclaim any liability for losses, damages, or adverse outcomes resulting from decisions made based on the information presented herein.}