Ormat's Record Quarter Came With a Warning Label
A $153.8M EBITDA record built on PJM merchant pricing management says is peaking, a 2028 target quietly trimmed, and 170,000 EGS acres the tape refuses to pay for.
ORA · Earnings Call · 2026-11-05
The beat is real; the quality is contested
Ormat delivered a genuinely strong third quarter. Revenue rose 10.2% to $275 million, gross margin expanded 250 basis points to 28.1%, adjusted EBITDA hit a third-quarter record of $153.8 million (+11.1%), and management raised the midpoint of full-year revenue and EBITDA guidance. On the headline, this is a company executing. And yet the stock sits 35.9% below its June 3 peak of $145.10 and 16.2% lower over the last ninety trading days, having ground down from an August 14 high of $114.76. That is the puzzle worth solving, because the gap between the print and the tape is where the actual story lives. Start with where the growth came from. Energy storage revenue rose 145.2% to $49.9 million at a 62.1% gross margin. Of the increase, $19.6 million came from existing PJM assets — price, not new steel — versus $9.3 million from newly commissioned facilities. So the single biggest driver of the quarter was merchant pricing, and management spent the Q&A volunteering that it does not last:This is not a new caution either. On the August call, Doron Blachar already noted that “the merchant prices, over the last few weeks, we see them becoming a bit more normal as we went toward the end of July and into August” — Doron Blachar, Chief Executive Officer · 2026-08-06. Two consecutive quarters of flagging normalization, and a beat that leaned harder on it anyway. Investors are not disputing the quarter; they are discounting the run-rate.In 2026, it was a very, very high year for merchant prices in PJM... 2026 prices are more than double of the 2025 prices. For 2027, we expect to be a bit lower than the 2026 numbers, but it's merchant.
What stopped being said
For roughly a year, the recurring Ormat pitch was 250 megawatts of hyperscaler and data center PPAs in late-stage negotiation. On the November 2025 call the CEO said: “We are actually in very final negotiations on a couple of PPAs with hyperscalers and in the same magnitude that we have been discussing. We hope to be able to finalize them, sign them and announce them in the next couple of months.” — Doron Blachar, Chief Executive Officer · 2025-11-04 That was four quarters ago. On this call, no conventional hyperscaler PPA was announced, and the topic barely surfaces except as a margin tailwind — "adding more assets to our Google and other data centers portfolio." The keyword history confirms the drift: data centers was the company's single strongest theme in 2024Q2 and again in 2025Q1; its momentum has since swung sharply negative and its rank has slid toward the bottom of the top thirty. The replacement narrative is enhanced geothermal. That substitution — contracted, bankable hyperscaler demand swapped for a technology option — is what the multiple is repricing.The margin arithmetic nobody wants to do
Electricity segment revenue actually fell 2.1% to $163.7 million, with segment gross margin at 22.2% versus 25.4% a year ago. Strip out the Puna impact — a hurricane-related forced outage that cost $5.1 million of revenue, with more well cleanout downtime scheduled for Q4 — and it was roughly 24.5%. Management's 2030 target is 40%. Getting there rests almost entirely on recontracting: blend-and-extend agreements against geothermal PPAs still clearing above $100 per megawatt hour. The CFO was emphatic: “we should start seeing an increase in margin already in Q4, and we expect to see a jump already in 2027” — Assi Ginzburg, Chief Financial Officer · 2026-11-05. Note the word "jump" — this is a company that has been promising a 1-2 point annual margin improvement for several calls running (“we do expect margin to improve towards the end of the year” — Assi Ginzburg, Chief Financial Officer · 2026-08-06, August 2026). Of the two renegotiated contracts, one started in October and the other is expected to reset sometime next year — meaning the 2027 "jump" depends on a negotiation not yet concluded. There is a subtler problem buried in the full-year guide. Revenue at the midpoint grows roughly 19.7% while adjusted EBITDA grows roughly 10.8%. Every incremental dollar of this business is arriving at a lower margin than the base, because the product segment — $310-320 million of the total at an approximately 18% full-year gross margin — is now more than a quarter of the topline. That is a mix problem dressed as a growth story. Leverage tells the rest. Non-current debt has climbed to $2.7 billion, up 178% over two years, against total debt of $3.3 billion and net debt of roughly $2.8 billion — 4.3x adjusted EBITDA. Interest coverage has compressed to 0.8x from 4.0x in 2017, while liabilities have risen to 59.6% of assets. Another $287 million of capex is slated for Q4.The acreage option
Against that, the EGS story got materially bigger in eight weeks. The land position expanded from roughly 70,000 acres to 170,000: “our EGS land position has grown from approximately 70,000 acres to approximately 170,000 acres in the last two months” — Doron Blachar, Chief Executive Officer · 2026-11-05. Estimated capacity potential rose from 3-4 GW at the September Investor Day to 5-6 GW. Interconnection requests are being filed into 2027, with Interconnection explicitly named as the gating item alongside water rights at Dixie Valley. But the economics are unresolved, and management said so more bluntly than any geothermal developer typically does:And the clock is sliding. A year ago the plan was to drill pilot wells "towards the second half or the end of '26"; the prepared remarks now say drilling is planned for late 2026 to early 2027, and in Q&A the CEO said “The pilots are due to start end of 2027” — Doron Blachar, Chief Executive Officer · 2026-11-05 before any Dixie Valley decision. First commercial phase lands 2029-2030; 1 GW of operating EGS is a 2033-2035 ambition. Against a stock that has already de-rated on 2028 numbers trimmed roughly 3% at Investor Day — the trigger for an early-October sell-side downgrade to neutral — asking investors to underwrite 2034 is a hard trade. One broader check: the demand premise underneath all of this — AI-driven power scarcity — is itself being repriced across the tape. AI data centers as a market theme is down roughly 30% over the last thirty days across dozens of decliners. Ormat is not in that basket, but its geothermal pricing power is a derivative of it. So the conclusion is uncomfortable rather than dramatic. Nothing broke this quarter. The beat was real, the guidance went up, and the land grab is a legitimate free option. What changed is that the cheap, contracted growth story got quieter at the exact moment the merchant tailwind got louder and shorter — and the market chose to believe the second half of that sentence.I think in today's pricing and technology, a $100 PPA for an EGS project is not high enough in order for it to be economic. Maybe over time, when the cost to develop EGS projects will continue to go down, it will become economical. Today, we believe we need higher pricing than that.