Eversource Has Cleared Almost Every Overhang — Except the One It Just Admitted Might Have to Be Booked

Revolution Wind reached COD, Aquarion is sold, credit metrics sit ~300bp above thresholds. So why did the FERC refund language quietly change from 'not booking' to 'GAAP may require us to reflect'?

ES · Earnings Call · 2026-11-05

The overhang that quietly closed

For four straight calls, Eversource carried Revolution Wind like a bad debt. It was the company's No. 4 keyword in Q4 2025 (Revolution Wind), a contingent liability that management was asked about on every Q&A, and as recently as July the CEO was doing yardage math on it. Thursday morning it ended, off someone else's press release: “As you may have heard on Ørsted's earnings call this morning, Revolution Wind has reached its commercial operations date. Construction is now complete and the wind farm is generating electricity.” — Joseph Nolan, Chief Executive Officer · 2026-11-05 Here is the tell: the phrase is essentially gone from the curated keyword set this quarter. So is the word that dominated the credit conversation for a year — Eversource has stopped explaining the wind farm and started explaining the rate case. Not entirely clean, though: there are contractual obligations with the buyer to "close out and settle up with them in 2027," against which management says it holds an "adequate reserve." Watch that line item.

The rate case is now the entire company

The CL&P rate case is the No. 1 keyword this quarter by a wide margin, and the framing is unusually candid about what's being asked. Sixty-six percent of the revenue deficiency is capital. Seventeen percent is storm restoration and vegetation management. Only roughly $50 million, or 12%, is O&M — against nearly $100 million had Eversource grown O&M with inflation since its last order. That's the pitch: “we've got a lot of wood to chop before we get to a position where we would be able to talk settlement” — Joseph Nolan, Chief Executive Officer · 2026-11-05. The second-order point is the one sell-side tends to underwrite too fast. A final decision is expected June 2027 (draft in May). That is precisely why 2028 — not 2027 — is the inflection year in the guidance. Alongside it sits investment in Connecticut: $4 billion of distribution infrastructure over five years, contingent on getting a PBR mechanism with inflation indexation and performance metrics. The smart meter story is the proof-of-concept — 500,000 installed in Massachusetts, one-third of the way through — while Connecticut AMI sits inside the same docket with a cost estimate still being refreshed.

The grenade in the room

Narrowed guidance and better credit metrics are the headline. The language shift on the FERC refund is the story. In May, the position was unambiguous:

we feel we have a strong legal position that supports us not booking anything additional until we have further determination and clarity on the retroactive piece going back to 2014.
John M. Moreira, Executive Vice President and Chief Financial Officer · 2026-05-07

Thursday, on Slide 10, it became this:

Generally Accepted Accounting Principles may require us to reflect the final refund amount in our financial statements before the case is concluded.
John Moreira, Executive Vice President and Chief Financial Officer · 2026-11-05

That is not a hedge — it is an accounting event risk. The long dated refund requires recalculating 154 months of load settlement data retained by ISO New England; management concedes it will be "challenging" to hit FERC's May 2027 deadline, while the D.C. Circuit's schedule (final briefs April 2, decision late summer or early fall 2027) lands *after* that deadline. Read that sequencing carefully: the legal vindication, if it comes, arrives after the obligation starts being paid. Asked directly whether it could become a probable charge, the CFO confirmed the framework rather than the outcome: “when any entity starts paying off its obligations, you have to assess what the -- is it probable that, that's going to be sustained” — John Moreira, Executive Vice President and Chief Financial Officer · 2026-11-05. He leans on Moody's and Fitch having signalled they'd treat it as a onetime, nonrecurring item. Prospective ROE is the nearer catalyst: the Section 205 record is complete, a decision is due on a rate effective November 30, with 11.39% proposed and 10.79% as an alternative using a Bloomberg beta.

What the numbers and the tape actually say

Credit is genuinely repaired: FFO-to-debt of 15.4% (S&P) and 16.1% (Moody's) as of June 30, roughly 300 basis points of cushion against downgrade thresholds, with a stated target of holding 100bp above. That's the FFO to debt ratio arc that took the company from 11.5% in early 2025 to the mid-teens. But the income statement is thinner than the credit story. Net income of $56M in the June quarter, with net margin at 1.9%, and Q3 recurring EPS of $0.92 versus $1.19 a year ago, dragged by the 9.57% transmission ROE, a ~$40M CL&P storm-cost disallowance, the absence of Aquarion earnings and a higher tax rate. The company nevertheless narrowed 2026 to $4.62–$4.70 and raised the midpoint to $4.66. Interest coverage of 1.5x, down from above 5x in 2010 against $26.6B of long-term debt explains why the $1.1 billion Connecticut/New Hampshire securitization — roughly $700M in the door mid-2027, New Hampshire in early 2028 — is treated as an earnings driver, not just a balance-sheet cleanup: “$1.1 billion of storm costs through the securitization, both Connecticut and New Hampshire... having $1.1 billion of interest cost -- of debt eliminated, you do the math, it's a pretty sizable reduction in our interest costs.” — John Moreira, Executive Vice President and Chief Financial Officer · 2026-11-05 Equity needs are pushed to the back half of the plan, $800M–$1.1B through 2030, with nothing needed this year thanks to $1.7B of Aquarion net proceeds.

The theme Eversource conspicuously isn't riding

Across the recent reporter set, utilities are tripping over themselves to talk about large load and data-center demand. Eversource has gone the other way, and then gone quiet. In May the CEO was openly hostile — “I am really not interested in data centers coming here; it is of no value to our residential customers and would only drive up the price of energy” — Joseph R. Nolan, Chairman, President and Chief Executive Officer · 2026-05-07 — and this quarter the topic simply doesn't appear. The data center keyword has now fallen out of Eversource's curated set for two consecutive quarters. That is a deliberate position, not an oversight: management frames New England's problem as supply scarcity and affordability, monetized through transmission (its $700M share of the ISO-NE $2.2B joint award with Avangrid) and a ~$50M DOE SPARK grant for the DREAM initiative rather than through hyperscaler load. So the setup into 2027 is unusually legible: a flat stock (down 0.4% over 90 days, 4.6% below its July peak, and still 27% below its 2020 high) at 18.7x earnings, waiting on a June 2027 rate decision, a November 30 ROE ruling, and an accounting judgment it spent all year telling investors it wouldn't have to make.