By Christopher Combs, Chief Investment Officer
Silicon Valley Capital Partners
April 24, 2026
A quiet but meaningful change is taking shape inside the U.S. economy. After years in which investors treated manufacturing as cyclical, fragile, and secondary to the consumer, the industrial side of the economy is beginning to look more strategic, more durable, and more central to the next phase of growth.
This is not a simple factory rebound. It is broader than that. The United States is moving into a capital-investment cycle driven by supply-chain security, reshoring, defense replenishment, electrification, grid modernization, factory automation, and the enormous physical infrastructure required by artificial intelligence.
The latest data does not show an economy free of stress. It shows something more interesting: manufacturing and business activity are expanding even as input costs rise. That combination suggests the economy is not merely recovering from weakness. It may be building new productive capacity.
Three signals are especially important: the Kansas City Fed’s April survey, the national S&P Global business activity data, and the data-center capital expenditure boom.
1. The Kansas City Fed Survey Shows Manufacturing Is Still Expanding, But the Cost of Expansion Is Rising
The Kansas City Fed’s April manufacturing survey showed continued expansion across the central U.S. factory base. The composite manufacturing index came in at 10, slightly below March’s 11, but still positive. That matters because a positive reading signals that activity is still growing, even if momentum cooled modestly.
The details were more important than the headline. New orders remained positive, though they eased to 12 from 15. Most major subindexes were still in expansion territory, with weakness concentrated in export orders and employee headcount. That tells us the factory economy is not booming uniformly, but it is also not rolling over.
The forward-looking data was more encouraging. The six-month expectations index rose to 18 from 16, while expected production over the next six months rose to 31 from 26. Manufacturers may be dealing with uneven demand today, but they are still planning for stronger production tomorrow.
The clearest warning signal was inflation. Prices paid for raw materials jumped to 63 from 37, a sharp move that suggests the industrial supply chain is again facing rising input pressure. Transportation, energy, raw materials, and inventory replacement are becoming more expensive.
This is the first key insight: the manufacturing renaissance is real, but it is not deflationary in the short run. Rebuilding domestic capacity requires steel, copper, power, labor, equipment, land, logistics, and capital. Supply-side expansion can improve productivity over time, but the early phase often feels inflationary because everyone is competing for the same inputs.
That is exactly what the Kansas City Fed data appears to show: steady activity, better expectations, but rising costs.
2. U.S. Business Activity Is Accelerating, With Manufacturing Leading the Move
The national data points in the same direction. S&P Global’s flash U.S. composite PMI rose to 52.0 in April from 50.3 in March, showing that business activity improved and moved further into expansion.
The most important part of the report was manufacturing. The flash manufacturing PMI rose to 54.0 from 52.3, beating expectations and reaching its strongest level in nearly four years. Services also improved, rising to 51.3 from 49.8, but the manufacturing acceleration was the more important macro signal.
This matters because manufacturing has been one of the most pressured parts of the economy during the higher-rate period. If factories are strengthening despite elevated financing costs and higher input prices, then something structural may be happening beneath the surface.
Companies are not simply responding to ordinary consumer demand. They are preparing for a world where supply chains are less reliable, inventories matter more, defense and infrastructure spending are rising, and domestic production capacity has strategic value.
But here again, the expansion comes with inflation pressure. S&P Global’s measure of prices charged by manufacturers and service providers rose to the highest level since mid-2022. Output prices rose to 59.9 from 58.1, while input prices rose to 62.6 from 60.9.
That is the central tension of the current economy. Growth is improving, but it is not yet the clean productivity boom investors would prefer. It is a supply-side rebuild, and rebuilds are expensive before they become efficient.
Employment data also looked mixed. Composite employment improved slightly, but manufacturing employment slipped below the expansion line. That may sound negative, but it also reflects a defining feature of the new industrial cycle: companies are trying to increase output through automation, technology, and higher utilization before adding labor aggressively.
The old manufacturing cycle was labor-heavy. The new one is capital-heavy. It depends on machinery, software, robotics, energy systems, semiconductors, grid infrastructure, and advanced logistics. That makes it less visible in payroll data but potentially more powerful for productivity and margins over time.
3. Data Centers Are Turning Artificial Intelligence Into an Industrial Supercycle
The third signal is the most important for the long-term thesis: the data-center buildout.
Artificial intelligence is often discussed as a software revolution, but its near-term economic impact is deeply physical. AI requires massive data centers. Data centers require land, electricity, cooling, fiber, semiconductors, memory, storage, copper, steel, concrete, transformers, generators, substations, switchgear, and industrial engineering.
In other words, AI has turned the digital economy into an industrial demand shock.
According to the data-center capital expenditure forecast you provided, hyperscalers including Amazon, Alphabet, Meta, Microsoft, and Oracle are projected to spend approximately $700 billion in aggregate capital expenditure in 2026, more than 70% above 2025 levels. Consensus estimates cited in the same material put aggregate cloud capital spending at roughly $762 billion in 2026, up about 64% year over year.
The U.S. Big Four hyperscalers, Google, Microsoft, Amazon, and Meta, are projected to spend $495 billion in 2026, up approximately 35% year over year. The forecast also suggests spending may grow another 20% in 2027 before moderating into the later part of the decade.
The cumulative numbers are extraordinary. The forecast implies more than $4 trillion of data-center spending from 2026 through 2030, with some estimates pointing to nearly $7 trillion of global hyperscaler and colocation infrastructure investment between 2025 and 2030, including about $800 billion for electrical and mechanical equipment.
That is not just a technology cycle. It is a power cycle. It is a construction cycle. It is an electrical equipment cycle. It is a cooling cycle. It is a semiconductor cycle. It is a grid-modernization cycle.
For U.S. manufacturing, this matters because the AI buildout creates persistent demand for the physical inputs needed to build, connect, power, and cool the next generation of computing infrastructure. Every new data-center campus creates downstream orders for transformers, switchgear, backup generation, turbines, cables, servers, racks, networking equipment, and energy infrastructure.
This is why the data-center boom may become the backbone of the U.S. supply-side expansion. It converts technology demand into domestic capital investment. It forces utilities to expand capacity. It gives industrial suppliers multi-year order visibility. And it creates a strategic reason for the United States to build more of the AI value chain at home.
The Real Story: America Is Rebuilding Capacity
The deeper story is not that the economy is suddenly overheating. It is that America is rebuilding capacity after decades of underinvestment in the physical side of growth.
For much of the post-globalization era, the U.S. optimized for low-cost production, lean inventories, offshore supply chains, and asset-light business models. That model created enormous efficiency, but it also exposed vulnerabilities: fragile logistics, concentrated semiconductor production, insufficient grid capacity, aging infrastructure, and dependence on foreign manufacturing for critical inputs.
The new cycle is different. Companies and policymakers are now prioritizing resilience, redundancy, domestic capacity, and strategic control. That means more factories, more grid investment, more industrial equipment, more energy infrastructure, and more capital expenditure.
This is why the latest surveys matter. The Kansas City Fed data shows that regional manufacturing remains expansionary even as costs rise. The S&P Global PMI shows that national business activity is improving, with manufacturing leading. The data-center forecast shows that one of the largest investment waves in modern history is just beginning to flow through the industrial economy.
The three points are connected. Stronger manufacturing activity is not happening in isolation. It is being pulled forward by the need to rebuild supply chains and support new infrastructure demand. Rising input costs are not random noise. They are partly the price of trying to expand physical capacity quickly. Data centers are not just cloud infrastructure. They are the new anchor tenant for the industrial economy.
Why This Matters for Investors
For investors, the implication is important. The winners of the next cycle may not be limited to the obvious technology platforms. The AI boom may also reward the companies that provide the physical picks and shovels of the digital age.
That includes electrical equipment manufacturers, power-generation suppliers, grid technology companies, cooling providers, engineering and construction firms, semiconductor equipment companies, industrial automation leaders, specialty materials suppliers, and select utilities with credible growth plans.
The market often prices AI as if the value will accrue only to chipmakers and cloud platforms. But if the buildout is as large as projected, the second-order beneficiaries may be broader and more durable. The industrial economy could become one of the most important transmission mechanisms through which AI spending reaches the real economy.
At the same time, investors should not ignore the risks. Rising input costs can compress margins. Higher prices can keep inflation sticky. Power shortages can delay projects. Permitting bottlenecks can slow construction. Supply-chain shortages in transformers, switchgear, memory, chips, or cooling equipment can create volatility. And if hyperscalers eventually slow spending, the most overextended suppliers could face a downcycle.
But the larger point remains: this does not look like a short restocking cycle. It looks like a multi-year capacity cycle.
The Bottom Line
The U.S. manufacturing renaissance is not a nostalgic return to the past. It is a modern supply-side expansion built around technology, energy, infrastructure, and strategic resilience.
The Kansas City Fed survey shows that factory activity remains positive and expectations are improving, even as raw-material costs surge. The national PMI data shows business activity accelerating, with manufacturing leading the move. The data-center buildout shows that artificial intelligence is creating a physical infrastructure boom that could drive industrial demand for years.
The economy is not just consuming. It is building.
And that may be the most important macroeconomic shift of all.
Sources
- Federal Reserve Bank of Kansas City, Tenth District Manufacturing Survey, April 2026.
- User-provided Bloomberg-style source notes: “US April Kansas City Fed Index Falls to 10; Est. 10,” NEWS, April 23, 2026.
- User-provided Bloomberg-style source notes: “Central U.S. Factory Activity Growth Continues in April, Kansas City Fed,” NEWS, April 23, 2026.
- User-provided Bloomberg-style source notes: “Kansas City Fed Services Index Indicates Slower Expansion in April,” NEWS, April 24, 2026.
- User-provided Bloomberg-style source notes: “US Business Activity Expands While Prices Rise Most Since 2022,” NEWS, April 23, 2026.
- User-provided Bloomberg-style source notes: “PMI Details Not Quite as Good as Headline Suggests,” NEWS, April 23, 2026.
- Uploaded data-center capital expenditure forecast, “Data Center Capital Expenditure Forecasts, 2026–2033,” including sell-side and Bloomberg Intelligence research summaries.