It took the Vogtle nuclear plant in Waynesboro, Georgia, roughly fifteen years and $35 billion to add two new reactors to the American grid, a project that ran years behind schedule and well over its original budget. Private investors still poured $40 billion into United States nuclear startups in 2025 alone, more money in a single year than it took to build Vogtle over a decade and a half. Nuclear is, admittedly, an imperfect test case on its own. The ticket sizes involved are unusually large even by the standards of the physical economy, and the technology carries a well-documented history of cost overruns that pressures investor returns regardless of how the financing is structured. But strip away nuclear’s specific baggage and the underlying pattern holds across capital-intensive industries more broadly. The money exists. What does not exist, reliably, is a way to get that money to a company at the moment it needs it, in the size it needs, on terms that match what the company is actually doing and the risk it actually carries.
The Vogtle nuclear plant in Waynesboro, Georgia took roughly fifteen years and $35 billion to add two new reactors.
That gap has a name in energy transition circles: the missing middle, though the term undersells how many different pools of capital actually fail to reach these companies. It is not simply a gap between venture capital and private equity. Venture capital is built to fund software-style bets, where a handful of outsized winners return an entire fund, a model built around light balance sheets and fast scaling that sits close to the opposite of what building physical infrastructure requires. Infrastructure and private equity capital sit at the other extreme, wanting a de-risked asset with predictable, already-contracted cash flow before they will commit, which is precisely what a first-of-a-kind physical technology cannot yet offer. The gap runs wider than just those two pools, though. Institutional capital, the pension funds, insurers, and endowments that control the largest pools of long-duration money in the world, typically requires an investment-grade credit rating or a public listing before it can participate at all, and public equity and debt markets are effectively closed for the same reason: neither status is available to a company still proving out a novel physical technology at commercial scale.
I recently spoke with Sanjeev Krishnan, co-founder and managing partner of S2G Investments, whose firm closed a billion-dollar fund earlier this year built specifically around this financing stage. What makes the missing middle framing useful is not that it is original to any one firm, but that it names a mismatch every one of these capital pools would recognize: not a shortage of capital in aggregate, but a shortage of capital shaped correctly for what these particular companies are actually doing.
That distinction, between novel and established infrastructure, matters more than the venture-versus-private-equity framing suggests on its own. Conventional wind and solar farms using proven technology, or mature transmission lines, do not generally face a missing middle problem at all, because infrastructure capital flows readily once the underlying technology risk has already been retired and the cash flows can be contracted. The gap is specific to first-of-a-kind physical technology: the kind still moving down its own cost and development curve, still more expensive than it will eventually be once it reaches scale, and still carrying a level of technology risk that a conventional infrastructure investor is not set up to underwrite. Nuclear sits at the extreme end of that problem, since large reactor projects combine early-stage development risk with an unusually large ticket size and a documented history of overruns that punishes returns independent of how the capital is structured. The same dynamic shows up in less dramatic form across new battery chemistries, novel industrial processes, and other capital-intensive technologies still early in their development curve. A company that has proven one of those technologies, signed its first customers, and needs $50 million to build a second manufacturing line does not fit into venture, does not meet a private equity buyer’s threshold for a de-risked asset, and has neither the credit rating nor the public listing that institutional and public market capital require. Francis O’Sullivan and Gokul Raghavan, energy investors at S2G Investments, gave the gap its name in a September 2023 white paper, and it has since become one of the more widely cited diagnoses of what is actually wrong with climate capital markets.
The Council on Foreign Relations put a number on the problem in an April 2026 report. O’Sullivan and Raghavan estimated that $270 billion was raised in the United States and Europe between 2017 and 2022 to support the shift to lower-carbon energy. Of that total, $120 billion went to early and mid-stage venture capital, funding what the Council on Foreign Relations called “early and risky technical innovation.” Another $100 billion went to infrastructure funds built to deploy fully de-risked, proven technology such as wind farms and transmission lines. That leaves $55 billion, barely a fifth of the total, for the stage in between: companies too far along for a venture fund’s power-law bets but not yet bankable enough for infrastructure money. The Council on Foreign Relations’ own estimate of the gap’s total size lands at $100 billion to $200 billion, a range wide enough to reflect how hard the thing is to measure and specific enough to make clear it is not small.
Of the $270 billion raised for the energy transition between 2017 and 2022, just $55 billion, one-fifth of the total, went to late-stage and growth-stage companies: the missing middle.
Two forces are widening the gap rather than closing it. The first is sheer scale. The International Energy Agency’s Net Zero Roadmap puts global clean energy investment at $1.8 trillion in 2023, a figure that needs to climb to roughly $4.5 trillion a year by the early 2030s to stay aligned with a 1.5 degree Celsius warming pathway. That is not a rounding error away from current spending; it approaches a tripling within a decade. The same analysis found that technologies not yet commercially available account for around 35 percent of the emissions cuts the world needs by 2050, down from an estimate of nearly half just two years earlier as some of those technologies matured. Progress is real. It is also evidence that a meaningful share of the transition still depends on unproven technology finding a way to scale, which is precisely the category of company the missing middle strands.
The second force is where the capital that does exist is actually flowing. Climate-focused funds raised a record $92 billion across 179 vehicles in 2025, according to Sightline Climate’s Dry Powder and New Funds report. Seventy-seven percent of that total concentrated among the largest infrastructure players, firms comfortable underwriting utility-scale solar, wind, and battery projects because the risk is already well understood. Early-stage venture’s share of total climate capital fell from around 20 percent in 2021 to under 8 percent in 2025. Only 39 percent of climate-focused venture funds that tried to raise a new vehicle in 2025 actually closed one, against 73 percent of infrastructure funds and 60 percent of growth funds. Within what venture capital does remain, the skew continues: three-quarters of new capital raised by climate-focused funds in 2025 was earmarked for energy investments specifically, leaving transportation, industrial technology, and agriculture increasingly cut off. The capital stack is not simply missing a middle. It is actively sorting itself toward the two ends and away from it.
Public policy tried to build a bridge across this exact gap and then, in large part, stopped trying. The 2021 Infrastructure Investment and Jobs Act established the Office of Clean Energy Demonstrations inside the Department of Energy, giving it more than $20 billion to invest in first-of-a-kind projects, complementing an expanded Loan Programs Office. The 2022 Inflation Reduction Act added further tax incentives and clean procurement funding on top. It was, by the Council on Foreign Relations’ account, the largest-ever policy effort to support emerging energy technologies through exactly this stage. Implementation lagged: the EFI Foundation found that only 47 percent of the Department of Energy’s newly appropriated funding had been obligated by early 2025, and roughly 5 percent had actually reached recipients. Then the second Trump administration eliminated the Office of Clean Energy Demonstrations entirely and canceled a large share of the associated grants and loans. Whatever the merits of any individual project, the episode illustrates a structural fragility: a financing gap this size cannot depend entirely on a policy commitment that can be reversed inside a single election cycle.
The kind of first-of-a-kind project the Office of Clean Energy Demonstrations was built to fund.
Some of the response is now coming from capital markets themselves, in forms that were not built for this stage but are being adapted to it. The global sustainable debt market, tracked by the Climate Bonds Initiative, closed 2025 with $6.8 trillion in cumulative aligned issuance, more than $4 trillion of it in green bonds alone, after a year in which annual issuance topped $1 trillion for the third year running. That scale is proof capital markets can absorb climate risk once an asset is standardized enough to rate and trade. It is also, by its own success, a demonstration of the problem: bond markets work for assets with predictable cash flows and established structures, which is exactly what a missing middle company does not yet have.
A newer channel is opening at the intersection of debt markets and this specific stage. Private credit managers, a category that barely touched climate infrastructure a decade ago, have begun structuring products aimed directly at the missing middle: Lightrock, Nest, and Galvanize among the firms New Private Markets identified in June 2026 as building debt vehicles for exactly this gap. It is an acknowledgment that equity alone, whether venture or growth stage, was never going to close a gap this large on its own, and that a company past the proof-of-concept stage with real revenue, even modest revenue, is often better served by debt that does not require diluting a founder’s ownership than by another priced equity round.
Blended finance is the more deliberate attempt to bridge it, and its results are instructive about scale. The International Finance Corporation, the World Bank Group’s private-sector arm, estimates that emerging economies need close to $3 trillion a year by 2030 to adapt to a changing climate. Between fiscal year 2018 and fiscal year 2025, the International Finance Corporation committed just over $4.4 billion in blended concessional finance across more than 500 projects, catalyzing an average of $8 in commercial funding for every $1 of concessional capital deployed. An eight-to-one leverage ratio is a genuinely strong result for a de-risking tool. Multiplied out, it is still a fraction of what the institution’s own estimate says the need requires annually. The tools that exist work. They are not yet sized for the problem they are meant to solve.
The Climate Policy Initiative’s Global Landscape of Climate Finance report put a figure on what happens if that mismatch persists: economic losses reaching 15 percent of global gross domestic product by 2050 under 2 degrees Celsius of warming, and 30 percent by 2100 under 3 degrees. Private climate finance did cross $1 trillion for the first time in 2023, according to the same report, edging past public investment for the first time on record. That is real momentum. It has not yet translated into capital shaped for the specific, awkward stage of a company that has outgrown its venture backers and is not yet ready to sell control to a private equity buyer.
What the last three years of data make clear is that the missing middle is not a temporary dislocation waiting for interest rates to fall or an initial public offering window to reopen. It is a structural mismatch between how capital markets are organized, around software-era return assumptions in venture capital and credit-rated, already-proven asset assumptions everywhere else, and what capital-intensive, physical, first-of-a-kind projects actually require to scale. Closing it will take more than one additional venture fund or one additional infrastructure vehicle. It will take financial structures built for the specific economics of building something physical, the kind of structures this sector has so far been better at diagnosing than delivering.
Listen to the full conversation with Sanjeev Krishnan on the SRI360 podcast, where he discusses S2G Investments’ approach to closing the missing middle across food, energy, and ocean systems.
For more interviews and insights on sustainable, responsible, and impact investing, visit the SRI360 podcast archive.
