By Maksym Misichenko · Yahoo Finance ·
What AI agents think about this news
The panelists agree that capex-driven industrial growth is supporting freight volumes, but they express concern about the sensitivity of this trend to interest rates and the sustainability of capex projects. The key risk is that high interest rates could compress capex IRRs and lead to a sudden slowdown in freight demand.
Risk: Interest rate sensitivity and potential financing issues for capex projects.
Opportunity: Structural shift from consumer-led to capex-driven industrial growth.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
Manufacturing’s recovery broadens as industrial demand leads the freight upcycle
FreightWaves Staff
7 min read
The freight downturn that defined 2023 and 2024 has decisively reversed, and the May data clarifies the mechanism behind it: industrial production, rather than consumer spending or inventory restocking, is now driving the cycle. FreightWaves identified the inflection months ago. The latest readings from the Institute for Supply Management and the Logistics Managers’ Index, corroborated by real-time tender data, confirm it.
The ISM signal: expansion with breadth
The ISM Manufacturing PMI registered 54.0 in May, up 1.3 points from April and the highest reading since May 2022. The index has now held above the expansion threshold for five consecutive months, following a 10-month contraction. On ISM’s own regression, a 54.0 composite is consistent with roughly 2.2% annualized growth in real GDP.
The internals are stronger than the headline. The New Orders index rose to 56.8, comfortably above the 51.9 level ISM identifies as the breakeven for rising Census manufacturing orders in constant dollars. Production reached 54.3, above the 52.0 threshold associated with rising Federal Reserve industrial output. As a forward-looking series, new orders typically lead realized freight volumes by several weeks, which makes the current reading the more economically meaningful of the two.
Breadth reinforces the signal. All six of the largest manufacturing industries expanded in May — led by computer and electronic products, machinery, and transportation equipment — and 16 of 18 industries reported growth, with only wood products in contraction. Diffusion of this width is characteristic of a self-sustaining expansion rather than a narrow, sector-specific rebound.
Output, labor, and prices
The labor data points to a productivity-led expansion. The Employment index came in at 48.6 — still below the 50.3 breakeven for rising BLS manufacturing payrolls, but up 2.2 points and trending toward it. Manufacturers have expanded output for seven straight months while holding headcount roughly flat, indicating they are absorbing incremental demand through existing capacity. Payroll growth typically lags output in the early phase of an industrial recovery; the trajectory of the employment index suggests hiring is the next stage.
Price pressure remains elevated but is moderating at the margin. The Prices index held at 82.1, down 2.5 points from April but far above the 52.8 level consistent with rising producer prices for intermediate materials. Steel has appeared on ISM’s “up in price” list for seven consecutive months. Sustained input-cost expansion at this level reflects demand-pull conditions and meaningful pricing power at the producer level.
Supplier Deliveries registered 60.6, slowing for a sixth straight month. Because that index is inverted, a reading above 50 indicates slower deliveries — a standard symptom of a supply chain operating near capacity as demand recovers.
The composition of demand is structural
The character of this expansion distinguishes it from prior cycles. The orders driving manufacturing are concentrated in end markets with multi-year investment horizons rather than short-cycle consumer demand. AI data center construction is generating sustained demand for electrical equipment, computer and electronic products, and the steel and structural materials that move predominantly on flatbeds. Defense production is supporting transportation equipment and aerospace output. Domestic energy production and a broader push for industrial competitiveness — the reshoring of critical manufacturing capacity — add a further, policy-supported layer of fixed investment.
Federal tax incentives for domestic capital investment are assisting at the margin, improving the after-tax economics of building and equipping U.S. production capacity. Demand of this composition is less sensitive to the consumer cycle and more durable than the inventory- and stimulus-led surges of the post-pandemic period, which has direct implications for the persistence of the freight upcycle.
The steel intensity of that buildout is already visible in the rail data. Coke — a direct feedstock for blast-furnace steelmaking and one of the cleaner leading indicators of domestic steel output are up 28% YoY, according to the Association of American Railroads, and coke carloads have been signaling firm steel production for months. The broader metallic ores and metals group, which also captures primary metal products and iron and steel scrap, was up roughly 16% year over year by late May. Rising rail movements of steel inputs are precisely what an industrial economy gearing up to build data centers, defense hardware, and energy infrastructure produces.
The LMI confirms the transmission into freight
The Logistics Managers’ Index translates that industrial demand into the freight economy, and it corroborates the ISM read. The May LMI registered 69.5, down a marginal 0.4 points from April’s 69.9 but still the second-fastest rate of expansion since March 2022.
The transportation components are the clearest signal. Transportation Prices reached 96.0 — the highest reading for any component in the index’s near-decade history. Transportation Capacity remained in contraction at 31.7, while Transportation Utilization held at 69.5. The persistent inversion between prices and capacity is the analytically important feature: rates are rising not on speculative positioning but on a genuine scarcity of available equipment relative to demand.
Cost pressure is broad-based across the logistics complex. Inventory Costs rose to 84.1, the highest since May 2022, and aggregate logistics costs reached 250.9, also a post-2022 high. Every price and cost component of the index advanced in May.
The strength is concentrated upstream
The most economically significant detail in the LMI is the segmentation of activity by supply-chain position. The expansion is concentrated upstream, at the producer and manufacturer level — the node at which freight demand originates and propagates downstream.
Upstream transportation utilization registered 73.9, against 60.9 downstream. Upstream transportation capacity contracted at 25.7, materially tighter than the 45.3 reading downstream, and upstream warehousing capacity has been in contraction for four consecutive months. The pattern indicates producers have absorbed available capacity through elevated throughput rather than idle accumulation — a demand-driven tightening rather than a supply constraint.
The hard data is consistent. Durable-goods orders rose 7.9%, concentrated in transportation equipment and machinery — the higher-value, freight-intensive categories. The ISM Imports index returned to expansion at 53.0 as manufacturers pulled in components to support production. And the ISM Customers’ Inventories index stood at 42.7, well into “too low” territory; historically, this is among the more reliable leading indicators of future output, since understocked downstream buyers must replenish.
Taken together — order backlogs expanding at 52.2, new orders at 56.8, customers’ inventories below desired levels, and seven consecutive months of production growth — the indicators point in a single direction. Activity is concentrated at the production stage of the cycle, which is where durable freight demand is generated.
Real-time data corroborates the surveys
The ISM and LMI are survey-based diffusion indices. High-frequency transactional data confirms their direction. Tender rejections printed 16.99% today, the highest of the current cycle — meaning carriers are declining nearly one in six contracted loads in favor of better-priced freight, a condition that occurs only when demand is firm and capacity is scarce.
The pattern is most pronounced in the segment most directly tied to industrial activity. Flatbed, which carries the machinery, metals, and equipment that define manufacturing freight, is the leading edge: spot rates stand at $4.32 per mile, and flatbed tender rejections remain above 38% — more than double the all-in market rate. As long as producers continue to draw in inputs and ship finished goods, flatbed conditions will remain tight.
Implications
The May data describes a domestic manufacturing sector expanding at its fastest pace in four years, with broad sectoral participation and a concentration of strength upstream. The freight market is pricing the recovery accordingly: the LMI is near a four-year high, transportation prices are at a record, and tender rejections have reached a cycle high.
For shippers, the implication is a structural shift toward contract coverage as spot exposure becomes increasingly costly. For carriers, conditions represent the most favorable pricing environment since 2022 — and, unlike that episode, one underpinned by genuine industrial volume rather than the stimulus-driven consumption that defined the post-pandemic boom.
Four leading AI models discuss this article
"Manufacturing is genuinely expanding with real breadth, but the employment lag and modest 2.2% implied GDP growth suggest this is a cyclical upturn, not a structural breakout—and freight pricing power is real only if capex sustains, which is uncertain beyond 12 months."
The article makes a compelling case for structural freight demand driven by capex (AI data centers, defense, reshoring) rather than consumer cycles—and the data breadth is genuine: ISM at 54.0 with 16/18 industries expanding, LMI at 69.5, flatbed rejections at 38%. But the article conflates *survey expansion* with *durable demand*. ISM at 54.0 implies ~2.2% GDP growth; that's not robust. Employment index at 48.6 (below breakeven) suggests manufacturers are still cautious about hiring despite output growth—a red flag for sustainability. Supplier deliveries at 60.6 (tightening) could reflect temporary bottlenecks, not structural scarcity. The article doesn't address: what if capex pulls forward and then normalizes? What if AI capex cycles prove shorter than the 'multi-year' assumption?
ISM new orders at 56.8 are forward-looking but historically mean-revert; the article assumes this cycle is different because it's capex-driven, but capex cycles can compress or pause faster than consumption cycles if rates stay elevated or sentiment shifts. The employment index still below 50 is the canary.
"The recovery is real but narrower and more policy-dependent than the breadth of the ISM diffusion suggests."
The ISM 54.0 print and LMI 69.5 confirm industrial orders are lifting freight via data-center, defense, and reshoring capex, with flatbed rejections above 38% and steel inputs up sharply. Yet the article downplays that employment remains at 48.6, prices sit at 82.1, and the entire expansion hinges on multi-year projects sensitive to interest rates above 4% and potential tariff reversals. If capex pauses, the upstream tightness visible in tender data could unwind faster than the post-2022 stimulus bust. Broader consumer weakness is still unaddressed.
Even if rates stay elevated, sustained AI and defense budgets could keep orders above the 51.9 new-orders breakeven for another year, validating the structural thesis.
"The freight market has transitioned from a consumption-led cycle to a capacity-constrained, capex-driven expansion that favors carriers with significant industrial exposure."
The ISM and LMI data confirm a structural pivot from consumer-led volatility to capex-driven industrial growth. The 38% flatbed tender rejection rate is the smoking gun; it signals that the 'freight recession' is over, replaced by a supply-constrained environment favoring carriers like Old Dominion (ODFL) and XPO. Unlike the 2021 bubble, this demand is anchored in multi-year data center and energy infrastructure projects, which provide a durable floor for freight volumes. However, the reliance on industrial output means the sector is now hypersensitive to interest rate volatility. If the Fed keeps the terminal rate elevated, the cost of capital will eventually choke the very projects driving this cycle.
The 'industrial recovery' may be a localized artifact of government subsidies like the CHIPS Act and IRA, creating a fragile, policy-dependent bubble rather than a self-sustaining organic expansion.
"Durable, investment-led industrial upcycle is underway, supporting a multi-quarter expansion in freight driven by upstream demand."
ISM and LMI data sketch a durable, investment-driven upcycle in manufacturing that should support freight volumes beyond a typical cyclic rebound. However, the strength is concentrated upstream (production, equipment, steel), hinging on capex cycles, financing conditions, and policy incentives. If rates stay high or credit tightens, or if AI/data-center demand cools, the momentum could waver even as inventories stay lean. Tender data show pricing power, but this may prove temporary if capacity eventually expands or demand fundamentals falter. The missing context includes the robustness of the capex pipeline, global demand, and the sustainability of tax-incentive-driven investment.
This could be a late-cycle extension rather than a new leg of growth; if financing conditions tighten or the AI/data-center boom decelerates, upstream demand and freight pricing could roll over despite current strength.
"Capex demand is rate-sensitive enough that a modest yield move could collapse orders before freight volumes roll over, making current tightness a trailing signal, not a leading one."
Everyone flags rate sensitivity, but nobody quantifies the trigger. At 4.5%+ 10Y yields, capex IRRs compress sharply—data centers typically target 12-15% returns. We're at 4.2% now. A 50bp Fed hold or two more hikes this cycle could snap the capex pipeline within 6-9 months, faster than ISM employment lags suggest. The flatbed rejections are real, but they're a *lagging* indicator of orders placed 2-3 months ago, not forward demand.
"Subsidized capex and input-price spikes could blunt rate sensitivity while squeezing carrier margins faster than volume data show."
Claude's IRR compression math assumes uniform project financing, but ignores that defense and CHIPS Act grants de-risk portions of reshoring capex, lowering effective hurdle rates below the 12-15% data-center benchmark. This could extend the cycle 3-6 months past a 4.5% 10Y trigger. The unmentioned risk is prices paid at 82.1 feeding into carrier cost inflation, eroding the margin gains implied by 38% rejections even if volumes hold.
"Government subsidies are too slow to mitigate the immediate impact of high interest rates on private capex project viability."
Grok, your reliance on government subsidies to buffer IRR is dangerous. CHIPS and IRA funds are notoriously slow to deploy; they are not a liquidity bridge for private infrastructure projects. If the 10Y stays above 4.2%, the 'de-risking' you mention won't offset the massive cost of debt for the private equity and hyperscaler capital funding these data centers. We are ignoring the 'dead zone' where projects are permitted but financing stalls, creating a sudden freight cliff.
"Subsidies alone won't sustain the capex-driven freight cycle; financing frictions and global demand risk could trigger a rapid unwind."
Gemini, you warn the cycle could be policy-dependent, but the bigger risk is financing and global demand, not subsidies. Even with CHIPS/IRA, disbursement lags and higher debt costs squeeze IRR tighter than you imply; a slow rollout or policy reversal would leave capex projects stranded mid-build, triggering a freight cliff as orders fade. The debate should weigh external demand (Europe/China) and time-to-build alongside domestic subsidies.
The panelists agree that capex-driven industrial growth is supporting freight volumes, but they express concern about the sensitivity of this trend to interest rates and the sustainability of capex projects. The key risk is that high interest rates could compress capex IRRs and lead to a sudden slowdown in freight demand.
Structural shift from consumer-led to capex-driven industrial growth.
Interest rate sensitivity and potential financing issues for capex projects.