Beyond the headline names, a set of mid-cap infrastructure plays is capturing the real capital flows in clean energy - and most investors haven't noticed yet.
The mainstream narrative around the energy transition looks like this: solar panels, wind turbines, and a few big EV makers. TSLA. ENPH. FSLR. The headline names get the headlines.
That's the wrong story - or at least, an incomplete one. The energy transition is an infrastructure story. Infrastructure, historically, is where the durable money gets made.
While investors piled into solar and EV manufacturers, a smaller set of mid-cap industrial and electrical contractors were quietly locking up the contracts, the backlog, and the pricing power that come with being the people who actually build the thing.
In 2024, global clean energy investment crossed $2 trillion for the first time. According to the IEA World Energy Investment Report, clean energy capex exceeded fossil fuel capex in the same year - the first time that has ever happened. Not a projection. A fact.
The chart below shows the trajectory. From $671B in 2018 to a projected $2.48 trillion in 2025 - a nearly 4× increase in seven years. The compounding nature of this growth is what matters for equity investors: the spending isn't slowing, and the contractors who build this infrastructure are booking record backlogs.
What's notable is what this investment is actually funding. The majority isn't going into panels and turbines - it's going into the grid infrastructure required to move the electricity those sources produce. That's transmission lines, substations, switchgear, battery storage, and the EPC (engineering, procurement, construction) firms that connect everything together.
The United States power grid was largely built in the 1960s and 70s. It was not designed for bidirectional power flows from distributed solar, it was not designed for the data center clusters now consuming gigawatts in single campuses, and it was not designed for an economy where electricity demand is set to grow substantially for the first time in two decades.
The Department of Energy's 2024 Grid Study puts a number on the problem: the US needs 40,000 miles of new and upgraded transmission lines by 2035. To put that in context, the entire existing US high-voltage transmission network spans roughly 160,000 miles - built over more than a century. We need to add 25% of that in a decade.
This is where companies like Quanta Services (PWR) and MYR Group (MYRG) become structurally interesting. They don't make solar panels. They don't manufacture batteries. They build the lines that carry the power - and there is no substitute for what they do.
The most underappreciated constraint in the entire grid buildout is transformers. Large power transformers - the equipment that steps voltage up and down at substations - have lead times of two to three years in the current supply environment. There are fewer than a dozen manufacturers globally. Any new transmission line or substation waits on transformer delivery before it can go live. This isn't speculation; it's already delaying projects.
The single biggest new driver of US electricity demand isn't EVs. It isn't reshoring. It's AI data centers.
A modern hyperscale data center campus consumes 500–1,000 MW. A single large AI training cluster can draw 500 MW on its own. Multiple independent forecasters - the EIA, BloombergNEF, and Goldman Sachs Research - project US data center electricity demand growing by 300–400 TWh by 2030. That's equivalent to adding the annual electricity consumption of a country the size of Poland to the grid.
The cascade effect matters here. Every data center megawatt requires new grid connections, new substations, new transmission capacity to reach it - and often new generation capacity nearby, since utilities are not able to redirect power from existing loads fast enough. This creates end-to-end demand for the entire EPC and grid infrastructure stack.
The investment case in mid-cap infrastructure plays isn't glamorous. There's no narrative about AI or software margins. But the numbers are compelling: multi-year backlogs, contracted revenue visibility, and pricing power that comes from being one of only a few contractors capable of large-scale transmission EPC work.
Below are six names that represent different angles of the same infrastructure trade. These aren't predictions - they're picks-and-shovels plays with real contract exposure to the capex flows described above.
The common thread: each of these companies benefits from a structural supply/demand imbalance in skilled labor and project execution capacity. Large transmission EPC contracts are not easily won by new entrants - the certifications, safety records, and relationships with utilities required take years to build. That's a moat, even if it doesn't look like one on a slide deck.
Grid-scale battery storage has moved from a novelty to a core component of grid architecture in roughly five years. In 2020, annual global installations were 12 GWh. By 2024, they were 338 GWh - an 87% compound annual growth rate over four years. By 2025, estimates put installations at 590 GWh.
The economics drove this. Lithium iron phosphate (LFP) battery costs fell over 80% from 2015 to 2024, per BloombergNEF data. At current prices, four-hour grid storage pencils out as economically rational in most US markets without subsidy - and with the IRA's standalone storage investment tax credit, the economics improve substantially further.
What this means practically: utilities are now procuring storage at scale not as an experiment, but as a standard part of capacity planning. Storage is becoming grid backbone. The companies supplying the inverters, the racking, the balance-of-system components, and the EPC services for storage projects are in the same structural tailwind as transmission.
The energy transition infrastructure trade has real risks. Investors who ignore them will get hurt.
Policy reversal. The Inflation Reduction Act is the primary legislative engine behind the current capex surge. A sustained political effort to claw back IRA credits - unlikely but not impossible - would meaningfully reduce the economics for renewables and storage projects, reducing order flow for infrastructure contractors. Tax credit transferability and direct pay provisions are particularly important to watch.
Permitting. The US permitting process for transmission infrastructure is notoriously slow. Transmission projects frequently take 10–13 years from initial planning to energization. Recent reforms (FERC Order 1920) are a step forward, but regulatory and landowner opposition remains a structural bottleneck. Backlogs for grid connection queue points now stretch 4–6 years in some regions. Projects can be delayed even when capital is committed.
Supply chain - transformers. As noted above, transformer supply is the most acute physical constraint in the current buildout. A sustained shortage could delay project completions and compress revenue recognition for EPC contractors, even with full backlogs. The shortage is improving, but slowly.
Valuation. Some of these mid-cap names have re-rated significantly as the infrastructure trade gained attention. PWR, for example, trades at a premium multiple to its historical range. Buying into a crowded trade after a re-rating is a different risk profile than buying at the start of a cycle. Entry price matters.
The energy transition is a $2 trillion per year capex story with decades of runway. The stocks that capture it most directly aren't always the ones making headlines. Picks-and-shovels infrastructure plays - EPC contractors, electrical constructors, grid equipment specialists - have structural revenue visibility and competitive moats that are easy to underestimate. The risks are real. The trade deserves a position in any portfolio with a multi-year horizon.
Historically the durable returns in an infrastructure build-out accrue to the builders and suppliers rather than the product manufacturers. In the energy transition that points at mid-cap electrical contractors, grid equipment makers and storage integrators with contracted backlog, rather than at solar panel or EV makers competing on price.
Global clean energy investment crossed $2 trillion for the first time in 2024. The IEA’s World Energy Investment Report shows clean energy capital expenditure exceeded fossil fuel capex in the same year — the first time that has happened.
AI data centers add large, concentrated, always-on load to a grid built decades ago. That forces spending on transmission, substations, transformers and storage well ahead of new generation, which is why grid and electrical infrastructure names see demand before renewable generators do.