Idaho Power's authorised return on equity is 9.6%
IDACORP closed on Thursday at $135.45. Issue a share at that price, then put the proceeds into the rate base, and it earns $13.00 a year. Last year, the company earned $5.90 a share.
Every new share brings in more than twice its own current earnings. That is why this stock trades at 23.0x and why the equity raising ahead of it is not the problem people assume it is. I want that on the table before anything else, because the arithmetic is not close. I nearly wrote something that ignored it.
Authorised ROE of 9.6% from the 2025 Idaho general rate case; FY2025 diluted EPS $5.90 from the Form 10-K via SEC XBRL; closing price 10 September 2026
I rate IDACORP (IDA) a Hold. The accretion is there. The question just depends on when it arrives. This week I found dates.
How Much Equity, And Why It Does Not Frighten Me
Idaho Power will spend $6.3 to $7.2 billion on plants between 2026 and 2030. It is disclosed in the 2025 annual report.
Operating cash was $602 million last year, against $1,179 million of capital spending. The year prior, $594 million against $1,009 million. Let the operating cash grow with the rate base. Generously, call it $3.5 billion across five years. Remove the dividends of $188 million and rising, so $1.05 billion. That leaves about $2.5 billion of internal cash against the programme.
Almost $4 billion comes from the outside. Holding the capital structure near its current split, $1.9 to $2.4 billion of that is just equity, against a market capitalisation of $7.83 billion. What does that mean? That is 57,842,043 shares on the Q2 cover, as of 24 July, at Thursday's close.
24-31% dilution over five years. The 2,995,502 shares already in forward sales agreements, about $406 million, are just a fifth of it.
IDACORP Forms 10-K, 2023 to 2025
I run those shares through the first calculation. The dilution pays for itself better than twice over. That doesn't sound like a company being forced to sell itself cheap. More around the idea that it is the regulated utility model working exactly as designed: raise equity above book, put it in the rate base, earn the allowed return, repeat.
So the bear case cannot be the dilution. It narrows down to something that stops the second half of that sentence from happening.
The Bridge
Here is the mechanism that carries the return across the gap. Also the reason a small accounting line matters more than its size.
Between the day money is spent and the day the asset enters the rate base, it earns nothing in cash. What it really earns instead is the allowance for funds used during construction, which capitalises into the asset and is recovered later. AFUDC is that bridge. It is how the 9.6% stays whole while the plant is still a hole in the ground.
You can see that bridge carrying the weight. First-half diluted EPS was $3.00 against $2.87. 13 cents of growth. Over the same two halves, the equity AFUDC credit went from $29.608 million to $38.060 million, while diluted shares went from 54,249 thousand to 56,767. Per share, 54.6 cents to 67.0 cents. The credit rose 12.4 cents while earnings rose 13 cents.
Trust me, I am not calling that a red flag. The accrual rate has been falling, from 4.70% of the construction balance in 2021 to 3.59% in 2025. It is what a company with 30.8% of its net plant sitting in construction looks like.
The point is what it tells you about exposure. Deny the accrual and the wait becomes a cost rather than a timing difference.
Someone Has Recommended Denying It
On 31 July, the Idaho Public Utilities Commission Staff filed comments in the certificate case for two new gas plants. Staff agrees the capacity deficits are real, agrees the load assumptions are reasonable, and recommends granting both certificates.
Then, on the 222 MW South Hills plant, which began as the company's own 60 MW battery bid: "This complete bid replacement occurred after the bid deadline had passed; after the bid projects had been scored and ranked by the Company and the independent evaluator; and after the FSL had been approved by the OPUC... the Company effectively bypassed the entire RFP selection process."
On the 430 MW Peregrine plant: "Staff believes the Company awarded itself the self-build Peregrine project outside of any RFP process, in direct contravention of the Commission's orders."
The staff literally recommends denying the construction allowance on both and capping recoverable cost.
Idaho Power does not dispute the dates. Only the standards, arguing the procurement ran under Oregon's rules "as required by the Idaho Commission at the time" and that the Idaho-specific order took effect on 2 January 2026, after the process closed. It calls the criticism "unreasonable and unsupported by the facts."
A serious argument, with ONE point it does not reach.
In Order No. 36958, served 10 March 2026 on the Bennett gas project, Idaho Power asked this same Commission to approve accruing on future projects "at the time the project has been deemed viable by the Company." The ordering paragraph: that request "is denied, and instead will be determined on a case-by-case basis where specific facts and circumstances can be evaluated."
The reply of 14 August, on South Hills and Peregrine. It was appropriate to begin "once South Hills and Peregrine had been deemed viable and expenditures associated with the resource procurements have been incurred."
The Commission refused to bless that trigger in advance and said such requests would be decided case by case, on the facts. This is that case, and the justification the company brings to it is the same trigger. And Bennett earned its accrual for a stated reason: the project "was selected through the Company's 2028 Request for Proposals bid evaluation process, after the Oregon Public Utility Commission determined that the Company had conducted the procurement in a fair and competitive manner consistent with applicable bidding rules." A clean procurement is what bought it, and a clean procurement is what Staff says is missing here.
That record closed on the 14th of August. So I pulled the docket again this morning. No order. 28 days, yet the only filing since is the irrigators asking to be paid for their own participation.
There is a sharper way to see this delay. The Commission's own agenda has it. Decision meetings run most Thursdays. The last published agenda is 3 September, and it carries fifteen items, four of which are under " Fully Submitted Matters," which is where a case sits once the record has closed and the Commissioners are deliberating. South Hills and Peregrine is not among them. Its record had closed three weeks earlier, and it had not made the list. No agenda has been published since that meeting. I will be fair about what that proves: the agenda states that fully submitted matters are listed "as a courtesy" and "may be privately deliberated," so absence is not proof of inaction. It is evidence that the case is not being moved.
Switch over, look at what that same Commission did move. On 8 September it issued Order No. 37159, a Notice of Modified Procedure, in IPC-E-26-19, the energy services agreement to serve Chobani. Micron intervened in that case on 7 August. So a new large-load service agreement is being advanced under modified procedure while the case asking whether the last two resources were procured properly sits without an order. In that stalled case, Staff had recommended the Commission delay signing new large loads.
IPUC dockets IPC-E-26-04 and IPC-E-26-19, and the decision-meeting agenda of 3 September 2026. All retrieved 10 September 2026
Let's size the accrual fight honestly. Amounts are redacted, but the rate is disclosed at 7.2%. At a generous $300 million of spend across both projects, eighteen months of denied accrual is about $32 million, or 56 cents a share, once, non-cash. At a likelier $150 to $200 million, 28 to 37 cents. 5-10% of one year's earnings.
Thirty cents is not the thesis. What it tells you is that the bridge is now something a regulator is willing to argue about, at the exact moment the company needs the longest bridge it has ever built.
And The Wait Just Got Longer
The next general rate case is not simply "sometime." It sits behind another proceeding, by the company's own agreement.
In the stipulation that settled the 2025 rate case, approved in Order No. 36892, Idaho Power agreed to "initiate a single-issue case related to CCOS methodology for the Commission's consideration in advance of filing a Notice of Intent for its next GRC..." CCOS is class cost-of-service, the rules deciding which customer class pays for which asset.
That case is IPC-E-26-07. On 11 August, Staff put a schedule in front of the Commissioners. Company testimony: 23 September. Staff and intervenor testimony: 25 November. Public comment closes 14 December. All-party testimony 13 January. Technical hearing, with witnesses, the week of 18 January 2027.
Pay attention to that sequence. The methodology fight does not reach a hearing until week 3 of January 2027. The notice of intent comes after that. The rate case comes after the notice. And through all of it the company spends roughly $1.2 billion a year while authorised rate base sits at the $4.9 billion set in 2025.
IPUC Decision Memorandum, 11 August 2026, Case No. IPC-E-26-07 (Staff-recommended schedule); Order No. 36892; IPC-E-26-23 application, 19 August 2026
There are two qualifications I would state here. That schedule is what Staff recommended: the memo ends by asking the Commissioners whether they wish to issue a scheduling order, and I could find none on the docket this morning. And utilities are not mute between rate cases. Idaho Power has a power cost adjustment and a fixed cost adjustment that move money annually. Neither of them resets the rate base.
Eight intervenors are in that queue arguing over who pays: Micron, the Industrial Customers of Idaho Power, the irrigators, the City of Boise, a Bayer affiliate, the Northwest Energy Coalition, Clean Energy Opportunities for Idaho, and the Department of Energy for the federal agencies.
They are all arguing about one customer, well, mostly. Staff computed from the company's own resource plan that the large-load class goes from 258 average megawatts in 2026 to 875 in 2030. That is an increase of 617 against total system growth of 705. 88% of all load growth. The Micron fab is around 445 of that, on Staff’s own estimate of 495 MW at an assumed 90 percent load factor. 72% of the class. The company doesn't contest it: "The Company does not dispute that large loads are contributing to the capacity needs."
Both sides of that. Staff's 495 MW figure is a minimum monthly billing demand under an approved contract, so Micron pays whether it takes the power or not. This argument has lost before. In Bennett, it was dismissed in a sentence. But not all of the class is contracted. The irrigators named the Diode Ventures "Gemstone Technology Park" as a load where "no executed energy service agreement or binding contractual safeguards existed." And "Staff asked the Company to perform a reliability assessment without the Micron FAB load, but the Company declined."
What You Are Paying For The Wait
IDA trades at 23.0 times last year's $5.90. Xcel is 22.0 on $3.42, OGE 20.1 on $2.32, Pinnacle West 19.1 on $5.05, Portland General 17.5 on $2.77, each diluted EPS straight from the 2025 Form 10-K and each priced at the same close.
Forms 10-K for 2025 (diluted EPS via SEC XBRL); closing prices 10 September 2026
Okay, so that comparison does have a limit. Xcel is a multi-state utility many times IDACORP's size. None of these four faces the same single-customer concentration. They are the closest regulated electrics I can put on an identical basis, not a matched set. Read it as a range, NOT a verdict.
Where I Could Be Wrong
The funding estimate is mine. Faster operating cash, capex at the low end, or a heavier debt mix all shrink the equity need materially.
The share count is as of 24 July against a 10 September price. Shares outstanding have been rising, so the true market cap is probably a little higher and my dilution percentage a little lower.
There is a filing I nearly missed that cuts against me. On 19 August, the company filed IPC-E-26-23, proposing depreciation rates effective 1 January 2028, up in seven of nine functions, transmission 1.96 to 2.21% and distribution 2.22 to 2.45, driven by more negative net salvage. Higher depreciation returns more cash through rates and shrinks the external funding need. The dollar total sits in an exhibit I did not open.
Staff recommending is not the Commission ordering. That applies to the schedule as much as to the procurement finding, and I am reasoning from one comparable case.
And 9.6% might not survive the next rate case in either direction. Everything above assumes it does.
So
The dilution is accretive, better than two to one. The premium multiple is defensible on that basis alone. Neither of those is in dispute here.
However, what is in dispute is the interval. The company spends $1.2 billion a year into a rate base that cannot reset until a methodology case reaches a hearing in January 2027 at the earliest, with a rate case behind that. The mechanism protecting the return across that interval is the construction allowance. And in the one docket where the Commission is being asked to bless spending right now, its own staff has recommended denying that allowance and said the procurement was bypassed, and 28 days after the record closed, we still await an order.
Hold rather than sell, with the first number being the reason. A business converting new equity into 2.2 times its own earnings does not get sold short over a schedule. It gets bought later.
Company testimony 23 September. Q3 retail volume in late October, where growth above 3% means the load is arriving. Staff and intervenor testimony 25 November. And the certificate order, whenever it comes.
I am wrong if that order lands clean and the hearing moves earlier.
Hold. $135.45