18 11월 2026
The Backlash That Pays for the Grid
- Permitting reform
- BAAJA
- Data center power costs
- Transmission lines
- Transformer shortage
Supercycle® Research
Why making data centers pay for the grid is the most positive policy development for grid spending in two decades
By David Fetherstonhaugh, Chief Investment Officer, VistaShares | November 18, 2026
The Bipartisan American Affordability and Jobs Act is a bill that would make it faster to approve new power lines and would require data centers to pay for the grid built to serve them. The Senate is close to passing it, and we think it is the most positive policy development for grid spending in two decades, because it removes the two things that have held that spending back, which are slow permits and the fight over who pays.
핵심 요점
- What it is. A bipartisan bill, known as BAAJA, that speeds up federal approval of energy projects, lets federal regulators approve large power lines that cross state borders, and makes every data center pay the full cost of the grid built to serve it.
- Where it stands. The Senate voted 70 to 27 on November 16 to end debate, and a final vote could come as soon as November 18 before the bill moves to the House and then to a President who has endorsed it.
- Why it is positive for grid spending. One estimate has data centers contributing $55 billion toward the grid over ten years, and independent modeling suggests long-distance line building would rise to 1.9 to 2.6 terawatt-miles a year from 0.7, on top of equipment order books that already stretch beyond three years.
What the bill does, and why it matters for the grid
BAAJA matters for the power grid because it shortens and simplifies the federal approvals that energy projects need and because it changes who pays for new grid, by requiring data centers to cover the full cost of what is built to serve them. The bill was written by two Republican and two Democratic senators, runs to nearly 500 pages and would be the largest change to federal permitting since 2005, and most of what has been written about it concerns its changes to environmental law. The Senate voted 70 to 27 on Monday to end debate, a final vote could come as soon as November 18, and the bill then goes to the House and to a President who endorsed it on November 8. Our focus is the electricity provisions, which have drawn the least argument and which decide how the grid gets planned, approved and paid for.
To see why those provisions matter, it helps to remember that the grid has been the slowest part of the energy system to grow. Electricity demand was nearly flat for most of this century, so there was little reason to build, and the rules for approving a new line were never updated for a time when demand is rising quickly. A high-voltage line that crosses state borders now takes about ten years to finish, partly because any state along the route can refuse it and partly because nobody has settled who should pay for it. The country added only 0.7 terawatt-miles of such line a year over the past decade.
“You don’t have affordability unless you have reliable and affordable energy.”
A House Republican from a power-exporting state, remarks on the bill, November 17, 2026
How the backlash becomes the funding
The lawmaker quoted above represents a state that produces more power than it uses, and the point explains why a bill this large has support in both parties. A state with power to sell and a state that wants to buy it both need a line built between them, and neither gets one while approval rests with every state in between. The bill deals with this by letting federal energy regulators approve large interstate lines, by requiring neighboring grid regions to plan those lines together within a year, and by setting a national method for sharing the cost among the customers who benefit. Each of these removes a reason a line could be delayed, but the provision that matters most to the investment case is the one about data centers, because it solves a problem that had started to threaten the whole theme.
That problem was political, and it grew as data centers arrived, because the cost of the grid built to serve them was spread across every customer in the same way that the cost of new highway lanes for the heaviest trucks is spread across everyone who drives. Households saw it on their power bills, this month’s elections turned on the cost of living, and voters in several states turned down new data centers. Because a utility’s spending plan has to be approved by state regulators who answer to those households, the anger was becoming the largest single obstacle to grid investment. The bill reverses the arrangement by requiring every data center, whatever its size, to pay the full cost of the grid built to serve it, so that the trucks pay for their own lanes. This is the point we think most commentary has missed, which is that the bill takes the public anger that was the main risk to grid spending and turns it into the mechanism that pays for it.
What that could be worth
| 기간 | New interregional transmission (TW-miles per year) |
|---|---|
| 2015–2024, actual | 0.7 |
| Through 2040, current policy | 0.9 – 1.9 |
| Through 2040, with the bill | 1.9 – 2.6 |
Exhibit 1 shows the most widely cited estimate of what follows. The industry measures long-distance transmission in terawatt-miles, which is the amount of power a line can carry multiplied by the distance it carries it, and one independent model finds that with the bill in place the country would add between 1.9 and 2.6 of them a year through 2040, compared with 0.7 over the past decade. Even the most cautious case in that range matches the best outcome the same model expects without the bill, and a second model built on different assumptions reaches a similar result, with total transmission capacity about 30% above its baseline by 2041 and the lines that link regions more than doubling. Our own arithmetic on those ranges is that the bill would add roughly 29 to 39 terawatt-miles by 2040, against 14 to 29 without it and about 7 in the whole of the past decade, so that at the midpoints the difference is close to 13 terawatt-miles, or nearly two decades of building at the old pace.
The money to pay for this comes largely from the new rule. One estimate has data centers contributing $55 billion toward the grid over ten years, about $35 billion more than they would under today’s rules, and because that money no longer comes from households, the same modelers expect the average family’s power bill to be $50 to $78 a year lower by 2040.
A theme that was not waiting for Congress
What makes this attractive from an investor’s point of view is that electrification did not need the bill to be a strong theme. The average wait for a transformer more than doubled between 2021 and 2024 and the largest units now take over three years, as Exhibit 2 shows, prices for power transformers are up about 77% since 2019, at least one of the largest turbine makers is sold out beyond 2030, and the Korean manufacturers that hold strong positions in the U.S. market are adding factory capacity against American orders they have already won. The bill’s benefits would begin to reach the grid around 2028 and build through the 2030s, which means they would arrive just as today’s backlog is being worked through and would carry the order book into the following decade.
| Year and unit class | Average transformer lead time (weeks) |
|---|---|
| 2021, all units | ~50 |
| 2024, all units | 120 |
| 2025, power units | 128 |
| 2026, largest units | 160+ |
Put simply, the bill adds years to a cycle that is already under way, and it does so at every step of the supply chain, although through a different provision and on a different timetable at each one.
Along the chain
| Part of the chain and timing | How the bill helps |
|---|---|
| Line upgrades and grid technology Earliest, 2027–28 | Upgrades in existing corridors skip environmental review, and utilities must adopt newer wire and controls with costs recovered in rates. |
| Transformers and high-voltage equipment Backlog past 2030 | Federal approval of large interstate lines adds demand for the largest transformers, which only a few makers in the U.S., Europe and Korea build. |
| Line builders and engineering firms Through the 2030s | A 150-day limit on lawsuits, and permits that stay in force while flaws are fixed, make an approved project far likelier to be built on time. |
| Utilities that own transmission From 2027 | National cost sharing, with data centers paying their own way, allows larger investment plans without raising household bills. |
| Power plants and on-site generation Near term | Grouped connection studies at a fixed cost, and states able to require data centers to bring their own supply, support turbine demand. |
Will it become law
Whether the bill passes before this Congress ends on January 3 is a fair question, and the people who study it for a living disagree. Before the first Senate vote on November 9, one Washington policy firm put the odds at 30% and one brokerage at 58%, while prediction markets implied about 20%, and the brokerage added that seventy or more Senate votes would lift its estimate to between 65% and 70%, a mark the Senate has since cleared twice. Support outside Congress is unusually broad, running from the country’s largest bank and the largest technology companies to labor unions, and a recent national poll found that 79% of voters support the bill, a figure that fell only to 66% after they had heard the arguments against it.
A multi-decade buildout, with or without the bill
Our one reservation is about how far the bill reaches, because although a bipartisan law of this size is welcome and clearly a step in the right direction, it leaves a great deal of power with state and local governments, which in our view still have more say than the national interest warrants. States keep the first review of any new line before federal regulators can step in, local authorities keep control of zoning for data centers and power plants, and the towns and counties that voted against new projects this month can go on doing so. The practical result is that the benefits will not be spread evenly across the country and will instead flow to the states and communities that already welcome investment, where projects can move as quickly as the new federal rules allow.
We are constructive because we see the bill as one chapter in a much longer story, in which electricity demand is growing again after two decades of standing still and the grid that has to carry it was built for a smaller economy. Rebuilding that grid will take decades, which is why the independent models run to 2040 and beyond. If the bill passes, the work gets faster and better funded, and analysts who follow the Korean equipment makers call it the most important near-term catalyst for their demand. If it stalls, the buildout continues on the strength of the order books shown in Exhibit 2, with no slowdown in projects reported ahead of the election. Because the building will vary from one state to the next, we think the best exposure is to the companies that supply every region, since they benefit wherever the lines and plants are built.
What we are watching
| When | What |
|---|---|
| This week | The Senate’s final vote, where the margin matters as much as the result because a count near 70 gives members of both parties in the House room to follow. |
| December | House action, possibly attached to the funding bill due December 11, since a bill that is not signed by January 3 expires and has to start again in a new Congress. |
| +90 days | The federal policy statement on data-center costs, which is due 90 days after enactment and will be the first rule that tells utilities how to bill. |
What does this mean to clients?
- Congress is close to passing a bill that makes data centers pay for the power lines built to serve them, so that the cost no longer lands on everyone’s electric bill.
- Anger over power bills was the biggest threat to grid spending, and this bill turns that anger into the thing that pays for the grid, which is why both parties support it.
- The companies that build and equip the grid already have years of orders in hand, and passage could nearly triple the pace of long-distance line building, so portfolios with exposure to grid infrastructure are well placed to benefit from a buildout that now looks both longer and better funded.
출처
- Rhodium Group, transmission modeling of the bill (November 2026).
- Resources for the Future, transmission modeling of the bill (October 2026).
- Wood Mackenzie transformer lead-time surveys, as reported by NIAC and Reuters.
- Bipartisan American Affordability and Jobs Act, bill text as amended November 7, 2026.
정보 제공 목적으로만 제공됩니다. 본 자료는 작성자의 견해를 반영하며, 투자, 법률 또는 세무 자문이 아니고, 어떠한 증권의 매수 또는 매도에 대한 추천이나 제안도 아닙니다. 언급된 기업은 예시를 위한 것이며 추천이 아닙니다. 미래예측 진술 및 기업 가이던스는 변경될 수 있습니다. 과거 성과가 미래 결과를 나타내는 것은 아닙니다. 모든 투자에는 원금 손실 가능성을 포함한 위험이 따릅니다.
Legislative status is as of the morning of November 18, 2026. The Bipartisan American Affordability and Jobs Act has not been enacted. It may be amended, may not pass either chamber, may not be signed, and if enacted may be implemented differently than described. Nothing in this note is a prediction of any vote or a view on the merits of the bill’s environmental provisions. Third-party research is cited for context; VistaShares is not affiliated with the organizations cited and has not independently verified their estimates.
투자자는 투자하기 전에 투자 목표, 위험, 수수료 및 비용을 신중히 고려해야 합니다. 이 정보 및 펀드에 관한 기타 정보가 포함된 투자설명서 또는 간이투자설명서는 (844) 875-2288로 전화하시거나 www.VistaShares.com을 방문하여 받으실 수 있습니다. 투자하기 전에 투자설명서 또는 간이투자설명서를 주의 깊게 읽으십시오.
Supercycle® is a registered trademark of VistaShares. Foreside Fund Services, LLC, 판매회사.