TXNM Energy

IThe business

TXNM Energy is a holding company with two regulated electric utilities and a headquarters in Albuquerque. PNM, which began in 1917, makes and delivers power in New Mexico. TNMP, whose predecessor was organized in 1925, delivers power in Texas and sells none of it.³ Together they serve about 849,000 customers and employ about 1,755 people.⁵³ The company was named PNM Resources until August 2024.⁹

A regulated utility makes money in a way that is unlike most businesses on this site. Nobody sets prices by competing. A commission sets them, and sets them to cover the utility’s costs plus a fair return on the money invested in its network. That invested money is called rate base. The return is a percentage, called the allowed return on equity, applied to the part of rate base that shareholders fund, called the equity ratio. So the earnings of a utility are roughly rate base times equity ratio times allowed return. Grow the first number and the earnings grow, if the regulator lets the spending in. The gap between what a utility spends and when its customers start paying for it is called regulatory lag, and it is where returns leak away.

The two halves differ. TNMP is a pure wires business inside the Texas grid, ERCOT. Texas law lets it recover new investment in the transmission and distribution network through filings made every six months, without a full rate case, and a 2025 law lets it defer costs not yet in rates and book them as an asset.⁵ PNM is a vertically integrated utility: it owns plants, buys power, runs the wires and answers to the New Mexico commission on almost everything, including which plants it may build or close.³ New Mexico law requires 50% renewable energy by 2030, 80% by 2040 and zero carbon by 2045.⁵

What it is spending

The company plans $10.2B of construction spending for 2026 to 2030, of which $1.57B is for 2026 and $8.65B for 2027 to 2030.⁵ Its net utility plant is $9.9B, so the plan adds about a full plant’s worth over five years. In February 2025 the plan was $7.8B for 2025 to 2029, and the company said regulated rate base “grows 12%” under it, with earnings per share growing 7% to 9% a year.¹⁰ The release does not say whether the 12% is a yearly rate; I read it as one, since that is what the capital plan implies E (estimate). The plan now runs a year later and is bigger. Three items stand out: a $595M 345 kV line that PNM was due to file on 30 October 2026, a request for 1,650 MW of new resources for 2029 to 2032 that also asks to leave Four Corners in 2031, and $367M of grid modernization over six years.⁴⁵

The reason is load. In Texas, TNMP set a new system peak in September 2026, and in the first nine months weather-normalized demand-based load rose 3.5%.⁵ In New Mexico, industrial load rose 9.6% and a data center for Meta takes power under a special rate.⁵ One cautionary number sits beside these: TNMP’s data center load fell 24.0% in the first nine months, after rising 70.5% in 2025.⁵³ Demand is real, and lumpy.

The deal

On 18 May 2025 TXNM agreed to be acquired by affiliates of Blackstone Infrastructure for $61.25 a share in cash. The announcement valued the whole company at $11.5B, counting net debt other than securitization bonds, and said the price was a 23.0% premium to the 30-day average price before press reports of a deal on 6 March 2025.⁶ Shareholders approved on 28 August 2025.⁵ The price is paid in cash, with no financing condition in the agreement, and the buyer plans to fund it with equity and keep the debt the company already has.⁵⁶ Everything else on this page has to be read with that in mind: the shares are priced to a takeover, and the business is the thing that matters if the takeover fails.

IIThe moat

The strength of a utility is permission. Nobody else may provide electricity delivery to these customers in these places, a customer cannot switch, and a rival cannot build a second grid. TNMP is the exclusive provider of transmission and distribution in most of its areas, and PNM must serve every customer in its territory even where its franchise has lapsed.³ Demand for power does not vanish in a recession, and the biggest single customer of PNM is 5.4% of its retail revenue.³ That is a durable business. The question for a shareholder is how much of the value of that permission a regulator lets the shareholder keep.

The record on that is mixed, and I can put numbers on it. PNM asked in 2024 for $174.3M more revenue, a 10.45% return on equity and 52.5% equity. It got a settlement for $105.0M, 9.45% and 51.0%, a full point less return and a smaller revenue increase, though the two revenue figures are not built the same way.³ TNMP asked for 10.4% and 47.54% equity and settled at 9.65% and 45%.⁵ The cases were not lost fights, since the parties signed the settlements and no commission rejected them. But in two states, in 2025 and 2026, the return came in about 0.75 to 1.0 point under the ask. I expect that gap to repeat.

The permission also comes with duties. PNM cannot close a plant, build one or buy one without approval. It must plan for 100% zero-carbon power by 2045, and it must serve a load that is growing in places that need new wires.⁵ A commission that approves the spending lets the utility earn on it. A commission that does not leaves the utility with a plant that customers will not pay for.

Two features help. In Texas the periodic filings make the delay between spending and earning short. In New Mexico, the Energy Transition Act lets the company issue securitized bonds to recover the cost of retired coal plants, and it has done so: $325M of those bonds are in its debt.⁵ Both are rules the legislature wrote, and a legislature can rewrite them.

So the permission is wide and the return on it is narrow. I would call the business protected, not rich. What it protects is a return near 9.5% on equity. If my cost of equity is 8.0%, that is a spread of about 1.5 points on each dollar of equity, and the spread is what a shareholder gets for the long wait.E (estimate)

IIIThe competitive advantage

A utility has no competitors in the usual sense, so the advantage that matters is whether it earns its allowed return, and whether it grows in places where the return can be earned. I see three things.

Where it grows

The first is geography. TNMP serves small and mid-sized communities in three separate areas of Texas, one of them in West Texas, where state law asked the grid operator for new reliability plans for the Permian Basin. The company said in February 2026 that TNMP was supporting a 28% increase in system peak demand, and 125 retail electricity providers use its network, the two biggest being 24% and 19% of its revenue.⁷³ A concentration to watch: two customers are 43% of TNMP’s revenue, and if either one failed, TNMP would bill the other.³ Texas has also passed laws, in 2023 and 2025, to encourage investment, and the Texas commission adopted a rule in February 2026 for connecting large loads of 75 MW or more.⁵

How fast it is paid

The second is recovery speed. TNMP’s transmission and distribution trackers, filed each half year, produced interim rate increases of $11.5M and $12.3M in transmission and $25.0M and $1.8M in distribution during 2025.³ In September 2026 it filed its first single combined tracker, asking for $77.9M and $27.0M more.⁴ A utility that can raise rates on new plant within months earns close to its allowed return. PNM does not have that. It waits for a full rate case, and its last one phased in over two steps, July 2025 and April 2026.³ Its transmission, which is on a FERC formula, is better: the formula resets each year on actual costs and projected projects, with a true-up.³

How reliable and how clean

The third is operating quality, which is how a utility earns goodwill with a commission. PNM’s outage duration measure, SAIDI, finished between the pay plan’s target and maximum in 2025, and TNMP’s customer survey did as well, but PNM’s own customer survey only reached the plan’s minimum.⁸ The chief executive said PNM delivered 80% carbon-free energy in 2025.⁷ Hurricane Beryl left about 116,000 TNMP customers without power in July 2024, and TNMP has $54.0M of related costs, of which it has asked to recover $20.5M.³ None of this is unusual for a utility, and I found no regulator’s ranking of either company that I could open, so I can’t say where they stand against peers.

What is missing is the kind of advantage that raises returns above the allowed level. TXNM does not have lower costs than a typical utility that I can measure, and it does not have a technology edge. Its advantage is a good growth footprint, and a recovery mechanism in the faster-growing half. The one place where it could matter most is the one it cannot control, which is the New Mexico commission.

IVManagement

Pat Collawn, 67, joined the company in 2007 and has been a director since 2010. She was chair and chief executive until 30 June 2025, became Executive Chair in July 2025, and the deal announcement says she steps down from that role when the deal closes.⁸⁶ Don Tarry, 55, a utility executive with more than 30 years in the industry, has been chief executive since that date. Henry Monroy became finance chief in May 2025.⁸ The team that agreed the deal is the team that would run the company afterward, which the Blackstone announcement said it wanted.⁶

How they are paid

For 2025 Collawn’s reported pay was $5.22M and Tarry’s was $3.90M.⁸ The annual bonus is 60% on a measure called Incentive EPS, 20% on outage duration and 20% on customer satisfaction. No bonus is paid unless Incentive EPS reaches $2.74.⁸ The long-term plan is 70% performance shares and 30% restricted stock. The performance shares pay on three things over three years: earnings growth, 40%, with 8% a year as target; share return against a utility index, 40%; and the ratio of cash flow to debt, 20%, with 14% as target.⁸

The Incentive EPS for 2025 was $2.77. GAAP earnings were $1.48. The difference includes $0.76 a share of merger costs and $0.45 of pension cost from a gas business sold years ago, both added back.⁸ The company’s own ongoing figure was $2.33.⁷ So the number the bonus uses is 19% higher than the company’s ongoing figure and 87% higher than GAAP. The bonus was capped at 80% of target by the earnings result.⁸ I don’t say the adjustments are wrong. The merger costs are a real one-off. But deal costs are a cost, and pay that excludes them gives executives a free pass on the price of the thing they are doing.

What that pays them to do

It pays them to grow earnings per share by 8% a year, to keep cash flow to debt above 13%, and to beat a utility index. It does not pay on return on capital, and it does not pay on value per share after the equity they issue. In a utility that is a flaw, because capital spending is the way earnings grow, and a plan that rewards growth in earnings can reward spending whether or not it is worth its cost. The plan has one safeguard: the cash flow to debt measure means they cannot borrow without limit. If the deal closes before the plan period ends, the performance shares pay at the higher of target and actual.⁸ That means the deal pays them to close it.

How much of their own money is in

Collawn holds 910,050 shares and rights, worth $53.2M at $58.49 and $55.7M at the deal price. Tarry holds 84,182, worth about $4.9M.⁸ All 13 directors and executives together own 1.07%.⁸ The chief executive’s stake is about 1.3 times his 2025 pay. Collawn’s is a real stake. Both would receive cash at the deal price, so for them the deal and the share price point the same way.

What they did with capital over a cycle

Over the ten years to 2025 the company spent $8.3B on plant and paid $1.1B in dividends, against $5.1B of operating cash. The difference came from debt and new shares.² Net utility plant grew 7.4% a year. Diluted shares grew from 80.1M to 102.4M. The dividend per share grew 6.9% a year, yet GAAP earnings per share were $1.46 in 2016 and $1.48 in 2025. On the company’s Incentive EPS, which I have for five years, earnings rose from $2.45 in 2021 to $2.77 in 2025, 3.1% a year, while net plant grew 8.4% a year.⁸² Plant grew more than twice as fast as the earnings per share. Five-year total return to 2025 was $142 for each $100, against $144 for the peer index the proxy uses.⁸

Two episodes belong here. In October 2020 the company agreed to be acquired by Avangrid for $50.30 a share. The New Mexico commission rejected the merger on 8 December 2021 after a hearing examiner advised against it, and Avangrid ended the agreement on 31 December 2023.¹¹¹² In May 2025 the company agreed a second sale, and again needs the same commission’s approval. Between the two came a second problem. The commission found that the May 2025 sale of $400M of TXNM shares to a Blackstone affiliate, at $50 a share, broke Section 62-6-12 of the Public Utility Act because it took place without the commission’s prior approval.⁵ The hearing examiners ordered the company to show cause in March 2026 and the commission issued a final order on 2 July 2026 with a $100,000 penalty. The 8 million shares were cancelled, $13.3M of dividends were returned, and TXNM repaid the $400M with a term loan, then with shares sold publicly on 31 August 2026.⁵ The penalty was small. The message was that a state regulator reads its statutes strictly where this company is concerned.

My view on management is simple. They run the two utilities competently, as far as I can see in the numbers, and they have not been good at turning growth into earnings per share. They are paid to grow, and the owner pays for it in new shares. That is common among utilities and it is not a mark against them in particular. The mark against them is the regulatory record in New Mexico, and I leave it for the next section, where I cover what I can’t judge.

VFinancials

The ten years need care. The company was PNM Resources until August 2024, and it still carries a pension cost from a gas business it sold years ago.⁹⁸ Operating income was $441M in 2025 and $144M in 2019, and I did not itemize the one-off charges between. Revenue includes fuel and purchased power that customers repay, so it says little. The columns that matter are plant, cash and shares.

Over those ten years net plant grew 7.4% a year and the dividend 6.9%. Depreciation doubled, from $242M to $483M.² Revenue grew 5.3% a year, which is slow, because the rate increases and the pass-through of fuel costs offset each other.

What the year to date shows

The first nine months of 2026 were strong. Revenue was $1,786.8M, up 9.5%. Operating income was $471.4M, up 35.7%. GAAP net earnings to TXNM were $238.4M against $161.2M, and ongoing earnings per share were $2.42 against $1.87.⁵⁴ The third quarter alone was $1.62 against $1.33. PNM contributed $1.24 of the quarter and TNMP $0.50, and the holding company lost $0.12.⁴ The reasons given are the new PNM rates, higher transmission margin, TNMP’s trackers and the Texas deferral law, warm weather and higher load, and lower returns on the nuclear decommissioning and coal-mine reclamation trusts.⁵⁴ The company does not give guidance during the deal.⁴ If the fourth quarter earns around $70M, as I assume E (estimate), ongoing earnings per share for 2026 are about $3.07 on about 111.5M diluted shares, a 32% rise on 2025’s $2.33.

That jump is not growth in the business so much as new rates arriving. The New Mexico order took effect in two steps, July 2025 and April 2026, and the Texas order was issued on 30 July 2026.³⁵ The base is therefore higher for 2027, and what rises from here is rate base times a return, not rates.

What an owner can really take out

Operating cash flow in 2025 was $584M. Depreciation was $483M. The difference, $101M, is a rough measure of cash left after standing still, before any growth C (calculated). Dividends were $163M, and ongoing earnings were $239M.²⁷ So the cash the business made after keeping up the plant did not cover the dividend, and the dividend was paid from debt and new shares. In 2025 the company sold $800M of shares, half of it the Blackstone purchase later unwound.⁷⁵ It is the usual profile of a utility with a big capital plan. Owners are paid in a growing asset that regulators promise to put into rates, not in free cash.

The balance sheet

Debt was $6.0B at 30 September 2026, including $325M of securitization bonds, and common equity was $3.6B.⁵ Equity was 38.7% of TXNM’s total capital, 46.5% at PNM and 50.6% at TNMP.⁵ The difference between the holding company’s figure and the utilities’ figures is the debt that the holding company owes itself, including a $550M convertible note and 7.000% junior subordinated notes issued in December 2025.⁵¹³ Moody’s rates TXNM Baa3 and S&P rates it BBB, with the junior debt a notch below investment grade at Ba1 and BB+.⁵ S&P called the commission’s show-cause order a negative development but said it did not immediately affect ratings.⁵

Operating cash flow in 2025 was 10.4% of year-end debt on my calculation, below the 13% that is the floor for the pay plan’s credit measure. The pay plan uses its own definition, so the two are not the same measure. Interest charges were $214M in the first nine months against operating income of $471M, a cover of 2.2 times. The company has $407.6M of debt maturing through October 2027, and $757M of room on its credit lines.⁵ One lender in each of the three revolving facilities declined to extend in December 2025, so from March 2029 the lines shrink by 8% to 12%.⁵ That is small, but it shows lenders do not all want to go further.

VIValuation

First, how this goes badly wrong

Before any numbers I asked what is most likely to hurt a holder from here, and worked back. Three things can.

• The deal breaks and the regulator stays hostile. The shares are priced to a takeover, and the price in a break is not $58. It is somewhere from $45 to $57 on my estimates below, and lower if New Mexico also turns against the company on its own rate cases. A break also leaves the company with a $10.2B plan, a Baa3 holding company, and a commission that has just fined it. The last time it happened, in 2021, the stock lost 9% in five weeks and was down 21% from its pre-ruling level on the day the deal was formally ended.¹¹¹²¹⁴
• Rates stay where they are. The 10-year Treasury yields 4.95% and the 30-year 5.32%.¹⁵ A utility is priced like a long bond with a regulator attached. At 1.77 times book, the shares assume a cost of equity near 7.3%. Each half point on the discount rate moves my base value by about $5.¹⁵
• A fire or storm that insurance does not cover. TNMP has already had Hurricane Beryl, and both companies say wildfire is a risk in their territories.³ I can’t put a probability on this. For scale, a $500M uninsured loss after tax would cut my base by $3.5 a share, and $1B by $7.1 C (calculated).

None of these needs the company to run its network badly. The first two are about outside forces, the third is weather. That shapes my method. I value the business alone, because that is what is left if the deal fails, and then I put the deal on top as a separate question.

What an owner can really take out

A utility reports earnings, but an owner can only take out what is left after the equity needed to keep growing. In my model the company keeps its equity at 38.5% of rate base. Each year, owner cash is earnings less the new equity needed to keep that ratio. In 2027 my base earnings are $377M, and the equity needed is $452M, so owner cash is −$75M. It is negative through 2028, near zero in 2029, and reaches $371M by 2036 E (estimate). The dividend is paid anyway, and the difference is raised by selling new shares: $290M in 2027 in my base.

So the distance from earnings to owner cash is large. In 2028 my base earnings are $423M and owner cash is −$69M E (estimate). Value in a growing utility comes from what the money earns when it is put into the plant, and from the day it stops growing. In my base, 83% of the value comes after 2036. That is the weak point of any such model, and I show the range for it below.

From the allowed return to the owner’s return

Leverage matters in two ways here. The utilities finance themselves about half and half. On top of that, the holding company borrows to put equity into the utilities, so TXNM’s own equity of $3.6B supports $4.5B of utility equity. That multiplies the return on the shareholders’ equity by 1.24, and it takes interest off the top. The next table builds the return from the parts.

It reconciles to within about 8% of the earnings I expect for 2026, and it gives a return on shareholders’ equity of 9.7%. That is the starting point of my base. It is above the allowed 9.45% and 9.65% because of the leverage at the top, and a half-point cut in allowed returns would take about 0.6 point off it, because the utility equity is 1.24 times mine.

What the price implies

At $58.49 the market values the shares at $6.4B and the whole company, counting debt and the convertible at its cash value, at about $12.3B. That is 1.77 times book value and about 19 times my 2026 earnings. The dividend of $1.69 yields 2.9%.⁵ The deal price is 20 times the same earnings, and the enterprise value at that price is about 1.4 times my proxy for rate base, which is $9.0B at 30 September C (calculated).

Three ways to read it. First, at an 8.0% cost of equity the price needs a 10.6% return on equity every year, for ever, which is more than the commissions allow. Second, it needs 163% of the stated capital plan built. Third, it needs a 7.3% cost of equity on my base cash flows, and the deal price needs 7.2%. That last number is worth stopping on. The company’s own junior subordinated notes, issued in December 2025, pay 7.000%.¹³ An owner’s required return of 7.2% is 0.2 points above the cost of a bond that ranks ahead of the shares. A buyer paying $61.25 for the business as it is would be accepting about the return of a hybrid bond. Blackstone must be counting on something else: cheaper money than a public investor has, more rate base than the plan, a better return than the commissions allow, or a view that the cost of equity for a regulated utility is 7%.

The dividend check says the same. If the dividend grows at 5% a year for ever, and the required return is 8.0%, the price is $59. At 4% it is $44. The company has raised the dividend by 5.2% and 3.7% in the last two Decembers.⁵ The price asks for it to keep growing faster than the economy, for ever.

What I can’t judge

I can’t judge the New Mexico commission. I have not read the staff and intervenor testimony in the case, because it is not in an SEC filing, and I do not know how the commissioners think about a private equity owner, a revised customer package, or a company they have just penalized. I know what the filings say: the commission rejected the Avangrid merger in 2021 after its examiner advised against it, and that PNM’s return in 2025 and TNMP’s in 2026 came in a point or three quarters of a point under what each asked, from the New Mexico and Texas commissions respectively.¹¹³⁵ I also can’t say what the Public Regulation Commission will decide on PNM’s next rate case, how wildfire liability would be treated in New Mexico or Texas, or whether load from data centers will arrive. I therefore show ranges and leave the odds of the deal as a band of 60% to 80%, which is a guess. The commission’s own record is the best evidence I have, and I give it above.

The discount rate

I use 8.0% for a utility’s equity. The ten-year Treasury yields 4.95% and the implied equity risk premium is 4.20%.¹⁵¹⁶ I measured TXNM’s beta against the S&P 500 on 62 monthly returns up to March 2025, before the deal, and got 0.40.¹⁷ That gives 6.6%. I don’t trust it. It is a small sample, and it ends just before the takeover story began. A beta of 0.7 gives 7.9%, and I round to 8.0%. It is also 1.0 point above the company’s hybrid debt. I use 9.0% in the bear case and 7.25% in the bull case, since for a company like this the rate is the largest single thing that can change, and I let it change.

That is lower than the 9.0% I ask on the Enbridge page and the 8.75% on the Constellation Energy page. It is lower for a reason: a regulated utility with an allowed return is a calmer business than a pipeline company or a power producer. But it is a judgement, and the table below shows what it does to the answer.

Three cases

The model has one step per year to 2036. It starts at a proxy for rate base: net utility plant, plus regulatory assets, less deferred income taxes and regulatory liabilities, which was $8.5B at the end of 2025 and $9.0B at 30 September 2026 C (calculated).⁵³ I put the end of 2026 at $9.5B E (estimate). Each year rate base grows with the capital plan, less depreciation at 5.6% and deferred taxes at 1.0%. Equity follows rate base at a fixed 38.5%, and earnings are the return earned on average equity. Owner cash is earnings less the equity needed. I discount it at the cost of equity, add a value for after 2036 as earnings grow at the long-run rate, and divide by 110.1M shares plus the shares the convertible adds. The company has not given a rate base figure, so my proxy is a rough one. It is built from reported numbers, and it ties to the earnings I expect for 2026.

One check on the plan. For 2027 to 2030 the company lists $8.65B of construction spending. In my base 90% of it is built, so spending is $1.8B in 2027 and $2.0B to $2.2B afterward.⁵ That is faster than the $1.57B for 2026. The company has not yet spent its 2026 plan: $903M went on plant in the first nine months.⁵

The weighted value on 25%, 50% and 25% is $51. It is a little above my base because the bull is further above it than the bear is below. The convertible notes add shares only when the price is above their $44.32 conversion price, and the model counts them when it is.¹³ None of the three cases includes the deal.

The deal: spread, break risk and what the price already says

Everything above is the company by itself. The price is mostly something else. It is the price of a promise from Blackstone to pay $61.25 in cash, and the promise depends on a decision that has not been made. A cash takeover leaves a spread: the gap between the market price and the deal price. Here it is $2.76, or 4.7%. A holder who keeps the shares gets dividends too, about $0.85 from two quarterly payments in early 2027 if the rate holds, and a small final dividend at closing.¹⁸⁵ If the deal closes on 30 June 2027, a share bought today returns 6.7%, or 10.4% a year. If closing slips to 31 December 2027 the gain is 8.1%, or 7.0% a year.⁵¹⁹ One-year Treasury bills pay 4.4%.¹⁵ The extra is the market’s payment for the risk of a break.

The New Mexico commission has changed the package once. The first application offered $105.0M of rate credits over four years, $10.0M for the Good Neighbor Fund, $35.0M for economic development and $25.0M for innovation. The revised one, accepted on 16 October 2026, offers $220.0M of rate credits over ten years, $15.0M for the fund, $40.0M for workforce and local projects, and $25.0M for innovation.⁵ That is $300.0M against $175.0M, or about $2.70 a share against $1.60 C (calculated). Texas settled for $45.5M. The price of approval has gone up, and the companies have offered it.⁵ The case also has no legal deadline.⁵ TXNM says it still expects closing in the first half of 2027, a date that slipped from the second half of 2026 in February.¹⁹⁷

The calendar is tight. Hearings end on 1 February 2027. In the Avangrid case, the evidentiary hearing ended on 19 August 2021, the hearing examiner’s recommendation came on 1 November and the commission voted on 8 December, 111 days after the hearing.¹¹ The same pace in 2027 gives 23 May, eight days before 31 May, when either side regains the right to end the agreement. A slower pace would need a new agreement between the parties, or Blackstone could leave and owe the $175.0M reverse fee, which is only about $1.60 a share. On 17 July 2026, after the commission ruled against the company, TXNM gave up half that fee, from $350.0M to $175.0M, as part of the waiver that bars either side from ending the agreement before 31 May 2027.⁵ That bought time. It also made it cheaper for Blackstone to leave.

The Avangrid case also teaches how a spread can mislead. A month before the commission voted in 2021, the shares were 1.1% under the deal price. The market was close to certain. The stock lost 9% by 2 December and the deal was rejected on 8 December. Market confidence in the shares was a view of regulators that turned out to be wrong, and it is no better than that here. The spread today, 4.7%, is wider than the 2021 one, so the market sees more risk. It is still narrow for a deal whose regulator ruled against the company this year.¹¹¹⁴

What odds the price implies

If I know what a share is worth when the deal breaks, the price tells me the odds the market is giving the deal. The answer depends heavily on that break price, and I do not know it. I use three. One is $45, below my base. One is $51, my base case plus the $175M fee after tax. The third is the 5 March 2025 average price of $49.80, which is the $61.25 deal price divided by the 23.0% premium, moved with the utility index since then, a rise of 12.1%, plus the fee, which gives $57.⁶²⁰

At the middle break price, $51, and a 6% discount rate, the price implies an 83% chance of closing. At $45 it is 90%, and at $57 it is 48% C (calculated). So the market says anywhere from about even to almost certain, depending on a number nobody can know. My own guess is 60% to 80%, which gives a share value of $57 to $58 at a break price equal to my weighted value plus the fee. At 70% it is $57.58, against $58.49. The price is about where the odds say, and I have no edge in the odds.

The two sides of the bet are lopsided. If it closes, a holder gains $2.76 on the price and about $1.14 of dividends, $3.90 in all. If it breaks, the loss runs from $1.40 to $26, depending on the case. At my base it is about $7.60 after the fee. That is two parts of loss to one of gain, and the odds have to be about 83% for the expected return to equal 6% a year C (calculated).

The margin of safety

My margin-of-safety price is $41: the point at which my $31 bear value is three quarters of the price paid. It is the same rule as on my other pages. It has nothing to do with the deal. It is the price I would want if the deal did not exist, and it sits 30% under the market, below every close of the past year; the lowest close was $56.40.¹⁴ At $41 the shares would trade at 1.24 times book and the dividend would yield 4.1%. Paying $41, my base cash flows return 8.8% a year and my bear flows 7.7% E (estimate). I set it that way because the main risk to a holder is in the bear case, and in the bear case the cost of equity and the allowed return fall together.

Other uses of the same dollar

The one-year Treasury pays 4.40% and the ten-year 4.95%, with no regulator.¹⁵ Closing the deal on 30 June 2027 would return 10.4% a year, and on 31 December 2027, 7.0%. That is two to five points over Treasuries, and the loss if the deal fails is 2% to 44%. Without the deal, my base returns 7.3% a year at this price and my bear 6.1%. The Enbridge page, which is the nearest regulated business on this site, finds that its price returns 9.9% a year on its base cash flows against a 9.0% requirement. Enbridge is in Canadian dollars and has no takeover at its price. The Constellation Energy page is the other power-demand name here, and it asks 8.75% on a business whose price is set by markets and not by a commission. Neither is a utility like this one, and each figure is on its own basis.

Quality, and price

These are two questions. As a business, TXNM is a decent one: two monopoly franchises, one in a fast-growing part of Texas, regulated returns that keep up with inflation if commissions allow, and a long capital plan. It is not outstanding. It earns about its cost of equity plus a point and a half, its growth has not turned into earnings per share, and it needs outside money every year. As a price, $58.49 is a different thing. Without the deal, it is 1.77 times book, 17% above my base and 14% above my weighted value. With the deal, it is about fair for the odds. A decent business at a price set by someone else is what this is, and I have no view on which way the New Mexico commission will go.

VIIThe case for the price

What most people miss

Three facts cut in the company’s favour and get little attention. First, the offer in New Mexico has risen from $175.0M of customer benefits to $300.0M, with rate credits of $220.0M over ten years instead of $105.0M over four.⁵ The companies are still paying to win. Second, both of the last two rate cases ended in settlements, not rulings: PNM’s 9.45% was unopposed, and every party in Texas signed the 9.65%.³⁵ Third, the equity for 2026 is already raised. The company sold 7.08M shares at $55.935 on 31 August and 2.12M on the market near $58.25, both above my $50 base for the business alone.⁵ If the deal breaks, that money is in the company and was raised above its standalone value.

The bull case

In plain words: a long-term owner pays a full price for two regulated utilities in fast-growing places, and brings the equity they need to build. Sean Klimczak, who leads Blackstone Infrastructure, said at the announcement on 19 May 2025 that Blackstone is a long-term investor with perpetual capital that wants to support economic development in both states.⁶ The company’s side is demand: Don Tarry said on 30 October 2026 that PNM and TNMP both reached new system peaks.⁴ Nine-month earnings per share are up 29%, and five of the six regulatory approvals are in hand.⁴⁵ I found no sell-side view I could verify from a primary source, and I did not check 13F filings.

Against reality

If the commission approves a package Blackstone accepts, and the deal closes by mid-2027, the price is fair: a share is worth $62.39 then, 6.7% above today’s price. That needs the commission to rule before 31 May 2027, which is close to the Avangrid pace of 111 days from hearing to vote.¹¹ Today the hearings have not started, there is no deadline in law, and the company has a $100,000 penalty on its record.¹⁹⁵

It helps to multiply it out. Say the commission approves a package Blackstone accepts, in time: 80%. No appeal or new delay pushes closing beyond 2027: 90%. Blackstone does not walk for any other reason: 95%. Those are my guesses. Multiplied they give 68% E (estimate). That is inside the 60% to 80% I use and gives a value of about $58, close to the price. The price asks for no more than this, and it asks for all of it.

What a 10x would take. Ten times $58.49 is $585, about $64B of market value. At 19 times earnings that is about $3.4B of net income a year, ten times my 2026 estimate and 1.6 times TXNM’s entire 2025 revenue of $2.17B C (calculated).³ It would take about $35B of equity at a 9.7% return, nearly ten times today’s $3.6B.⁵ By 2036 that is 26% a year for ten years; in twenty years it is 12% a year with no loss of the multiple. The shares rose 1.9 times from $30.17 in January 2016 to $58.49.¹⁴ Until the deal closes or breaks, the price cannot move far above $61.25. Ten times is close to impossible for a regulated utility of this size.

Where I come down

The revised New Mexico package and the Texas, FERC and nuclear approvals moved me, and they keep my odds, which are a judgement, at 60% or more. The Avangrid record, the penalty and the timing did not move me, and they stop me at 80%. What would raise my values is a New Mexico order that Blackstone accepts, or a PNM rate case that sets returns at 9.75% or more with the whole plan approved. A cost of equity half a point lower would add about $6.50 a share to my base.

VIIIRisks, and what would change my view

The three ways this goes badly wrong are at the start of Valuation. These are the rest.

• The commission’s decision. It can approve, approve with conditions Blackstone refuses, or refuse. I cannot read it. The commission rejected the Avangrid merger in 2021, found against this company in 2026, and has no deadline.¹¹⁵
• Timing and the fee. The ban on ending the agreement runs to 31 May 2027. After that either side may leave, and the fee Blackstone would owe is $175.0M.⁵
• Money for the plan. The company says it will need new debt and equity to fund $10.2B of spending, and its holding company is Baa3 with junior debt below investment grade.⁵ A fall in ratings raises borrowing costs for the utilities, which regulators may or may not pass on.
• Recovery of what is already spent. The balance sheet carries $928.6M of regulatory assets, costs the company expects to recover in future rates.⁵ If a commission disallows them, they are written off.
• Four Corners and the coal tail. PNM asked in May 2026 to leave Four Corners in 2031 and to replace the power with 1,650 MW of new resources.⁵ The plant is on Navajo Nation land, and the owners’ right to operate it there runs to July 2041.³ Exit costs and reclamation go through the same commission.
• Nuclear. PNM owns shares of three Palo Verde units, which ran at 92.3% availability in 2025.³ An outage or a decommissioning cost increase falls on the same customers and regulators.
• Load. TNMP’s data center load fell 24.0% in nine months after rising 70.5% in 2025.⁵³ Two customers are 43% of TNMP’s revenue.³ New Mexico industrial load is growing, partly from one large customer.⁵
• Affordability and politics. A plan that raises rate base 12% a year raises bills. A commission under political pressure over bills may slow approvals. The 10-K lists higher insurance premiums for wildfire risk among the year’s cost increases, $2.6M of them.³
• The convertible. $549.8M of 5.75% notes convert at $44.32. At the deal price they would be worth about $760M in cash, $210M over their face value, which the buyer inherits on top of the $61.25.¹³