Sempra is a three-legged stool: California utilities (SDG&E, SoCalGas), an 80% stake in Texas's Oncor, and Sempra Infrastructure (LNG export). At $94.67 versus a base IV of $186.91, the market is pricing the boring legs and giving you the LNG and Texas growth optionality close to free.

Plain English

Sempra owns the wires that bring electricity to San Diego, the pipes that bring gas to Los Angeles, and most of the wires that bring electricity to Dallas, Houston, and the rest of Texas. It also runs export terminals on the Gulf Coast that ship US natural gas, frozen into liquid, to Asia and Europe. State governments let Sempra charge customers a small profit on every dollar Sempra invests in pipes, wires, and terminals. Sempra invests a lot. Today the stock costs about half of what those assets are worth, and the math says the price will likely close that gap.

Thesis

Sempra is a regulated utility holding company built around three franchises: (1) SDG&E and SoCalGas in California, (2) an 80% economic interest in Oncor, the largest electric T&D utility in Texas (no generation, pure wires, ERCOT — the fastest-growing electricity market in America), and (3) Sempra Infrastructure, which owns Cameron LNG and is building Port Arthur LNG Phase 1 and ECA LNG. The compounding mechanic is straightforward and Buffett-shaped: deploy capex into a regulated rate base, earn an authorized return on that base for decades, retain earnings to fund the next leg of capex. Buffett describes this exact model when he discusses MidAmerican / BHE: 'we put a large amount of trust in future regulation' [1][6], and 'recession-resistant earnings, which result from these companies exclusively offering an essential service' [5][6].

The scorer composite is 68/100 — solid but not ecstatic. Profitability scores 11 (the lowest sub-score) because regulated returns are capped: this is by design, not a flaw. Balance sheet 20 with 1.70x net debt / EBITDA is acceptable for a utility (peers run 4-6x; the 1.70x figure here likely understates true consolidated leverage, since SDG&E, SoCalGas, and Oncor each carry their own debt stacks). Capital allocation 14 is mediocre — the 10y share count is up 10.9%, meaning equity issuance to fund rate base, not buybacks. Valuation 23 is the standout: P/E TTM of 21.4 versus a 10y average of 43.5, reverse-DCF implied growth of just 2.7%, and a base IV of $186.91 versus a price of $94.67. Px/IV = 0.51. Even the low-case IV of $104.16 sits above today's price.

At $94.67 you are paying ~22x earnings for a regulated rate-base compounder where the implied growth bar is 2.7% and management's published rate-base CAGR target sits multiples higher. The math does the heavy lifting.

Moat

Sempra's moat is the classic regulated-utility moat — not glamorous, not pricing-power-driven, but durable in a way few businesses in any sector can match. It comes from three sources, each scoring differently.

1. Franchise / regulatory monopoly (intangible asset, very strong). SDG&E is the sole regulated electric and gas distributor across most of San Diego County. SoCalGas is the sole regulated gas distributor across ~21 million customers in Southern California — the largest natural gas utility in the United States. Oncor (80% Sempra) is the largest regulated electric T&D in Texas, serving ~13 million people. None of these franchises can be entered by a competitor — you literally cannot string a parallel set of distribution wires down the same street, nor can you bury parallel gas mains. This is a state-granted monopoly enforced by the California Public Utilities Commission (CPUC) and the Public Utility Commission of Texas (PUCT). Buffett describes this same characteristic in BHE: 'MidAmerican's electric utilities serve regulated retail customers in eleven states. Only one utility holding company serves more states' [5]. The franchise itself is the moat, and it is functionally permanent. Stress test: if a $10B-funded competitor showed up tomorrow with a plan to lay parallel infrastructure across San Diego, the CPUC would simply not grant the certificate of public convenience.

2. Cost advantage via rate-base scale and cost of capital (medium-strong). Utility moats compound through cost of capital. A bigger, more diversified regulated entity can borrow at lower rates than a smaller one. Buffett makes this exact point: 'this particular strength, supplemented by Berkshire's ownership, has enabled MidAmerican and its utility subsidiaries to significantly lower their cost of debt. This advantage benefits both us and our customers' [6]. Sempra is not Berkshire-backed but it is large enough (top-5 US utility holdco) and diversified enough across three regulators (CPUC, PUCT, FERC for the LNG infrastructure) that it enjoys most of the same benefit. Net debt / EBITDA of 1.70x at the holdco level [scorecard] is acceptable.

3. Switching costs (strong but tautological). Customers cannot switch — there is no other electric distributor in San Diego, no other gas distributor in LA. This is the same customer-captivity that makes Buffett comfortable with capital-intensive utility businesses despite their typically poor returns on incremental capital.

Erosion risks — three real ones.

California regulatory risk (SDG&E, SoCalGas). The CPUC has historically been one of the more difficult US utility regulators. The 2017-2019 wildfire liability regime nearly bankrupted PG&E. SDG&E has thus far navigated this better (it has a strong wildfire mitigation track record) but the tail risk is non-zero. Buffett's framing — 'we shouldn't expect our regulators to live up to their end of the bargain unless we live up to ours' [2] — is the right posture, but it does not eliminate the risk that a regulator becomes hostile.

Gas-distribution decarbonization (SoCalGas). California has set aggressive electrification targets. SoCalGas's terminal value is genuinely uncertain on a 30-year horizon. The bull case is that gas remains the lowest-cost residential heating and industrial fuel for decades; the bear case is that California legislates the gas distribution franchise into a stranded asset.

LNG counterparty and commodity risk (Sempra Infrastructure). The LNG segment is contracted under long-term tolling agreements, which insulates Sempra from spot LNG prices but exposes it to counterparty risk and a slow but real risk that the global LNG glut compresses contract renewal economics post-2030.

The moat does NOT come from pricing power, network effects, or brand. Utility pricing is set by regulators based on a return-on-rate-base formula. There is no pricing power in the Buffett See's Candy sense; there is, however, regulatory pricing certainty, which for a capital-intensive business is arguably more valuable.

Moat verdict: WIDE.

Management

Sempra's capital allocation under Jeffrey Martin (CEO) is best understood as a continuous reinvestment machine: nearly every dollar of earnings, plus a meaningful amount of new debt and new equity, is plowed back into rate base — wires, pipes, LNG terminals, ERCOT transmission. This is the right strategy for a regulated utility because the regulator literally pays you a return on every authorized dollar you invest. Buffett describes the BHE version: 'Last year MidAmerican retained more dollars of earnings — by far — than any other American electric utility. We and our regulators see this as an important advantage' [1]. Sempra is not Buffett-grade on retention (it pays a meaningful dividend, ~3.4% yield) but the broad shape is similar.

Reinvestment (the dominant choice, A-). Sempra has guided to a multi-year capex plan in the $40B+ range, the bulk of which lands in Texas (Oncor) and Sempra Infrastructure (Port Arthur LNG, ECA LNG). Oncor's rate base is growing high single digits; Texas is in the middle of a generational electrification + data-center load surge. This is exactly the kind of capex you want — long-lived regulated assets earning an authorized return — and it is what gets the math from a 2.7% reverse-DCF implied growth to a base IV of $186.91. The capex is real, the growth is real, the regulators have been consistently constructive in Texas and (more variably) California.

Acquisitions (mixed, B). The 2018 Energy Future Holdings / Oncor acquisition was the transformative deal of the past decade — it took Sempra from a California-centric utility to a continental scale regulated platform with the best US growth franchise (ERCOT). That was an A+ allocation decision. Subsequent deals have been smaller and more digestible — partial monetizations of Sempra Infrastructure to KKR and ADIA brought in growth capital at attractive multiples. The 2023 sale of a minority stake in Sempra Infrastructure was effectively an arbitrage: Sempra raised capital at a higher implied multiple than its own stock was trading at.

Debt (B). Net debt / EBITDA of 1.70x at the holdco level is acceptable, though this understates true consolidated leverage because each utility subsidiary carries its own substantial debt stack against its rate base (utilities run 50-55% debt / total cap). Interest coverage is not in the scorecard (null), which is mildly concerning — a careful investor should pull the actual coverage from the 10-K. The risk is rising rates: every 1% increase in average cost of debt at Sempra's scale compresses earnings by hundreds of millions before regulatory recovery.

Buybacks (D, but appropriately so). Share count is up 10.9% over 10 years [scorecard]. Sempra has been an issuer, not a buyer. For a utility funding a multi-billion-dollar rate-base growth plan, this is unavoidable and arguably correct — issuing equity at $90+ to fund projects earning a 9-10% authorized return on a growing rate base is value-creative as long as the equity is not issued below intrinsic. At Px/IV of 0.51 [scorecard], today's equity issuance is value-destructive in the Buffett sense (issuing 50-cent dollars), and management should be prioritizing project-level funding partners (KKR, ADIA), debt, and asset recycling over equity issuance until the price closes the IV gap. Whether they actually do this is the single biggest question for a long-term holder.

Dividends (B). ~3.4% yield, raised annually for ~20+ years. Reasonable for a utility — high enough to reward income holders, low enough to leave retention for capex. Not aggressive on either side.

Communication quality (B). Investor day disclosures are detailed and segment-specific. Rate-base growth, EPS guidance, and capex by year are all laid out. The communication is more 'managerial' than 'owner-operator' — you do not get the same 'I am writing this letter as if I owned 100% of the business' tone that Buffett pioneered. But it is not slick, and the numbers reconcile. No major restatements or accounting controversies.

The single biggest test will be: at today's Px/IV of 0.51, do they slow equity issuance and recycle assets instead? If they do, this is an A-. If they keep issuing equity at 50 cents on the dollar to fund growth, this is a B-.

Capital allocator: B.

Industry

Threat of new entrants — extremely low. A regulated electric or gas distribution utility cannot be replicated. The state grants a single franchise per service territory, and that franchise is enforced both economically (the cost of laying duplicate infrastructure is prohibitive) and legally (the certificate of public convenience would not be granted to a duplicate). This is the strongest barrier to entry in any sector of the US economy. Buffett notes this implicitly when he describes BHE's 11-state utility footprint as 'an important advantage — one almost certain to exist five, ten and twenty years from now' [1].

Bargaining power of buyers — low for the utilities, moderate for LNG. Retail electric and gas customers in the regulated franchise have no bargaining power; they cannot switch suppliers for the distribution leg. Pricing is set by the CPUC and PUCT through general rate cases. The LNG segment is different: long-term tolling counterparties (Mitsui, ENGIE, Hartree, RWE, KOGAS, etc.) are sophisticated and negotiate hard. But the contracts are 20-year fixed-fee structures that effectively eliminate commodity price risk; once signed, the buyer power is contractually locked.

Bargaining power of suppliers — moderate. The biggest supplier-side risk is labor (skilled lineworkers, gas mechanics, LNG operators) and capital. Cost of capital is the dominant supplier — interest rates flow through to the regulated return-on-rate-base formula with a lag, so a sustained high-rate environment squeezes utility earnings until the next rate case. EPC contractors (Bechtel, etc.) for LNG have been a real cost pressure — Port Arthur LNG has seen cost overruns. Equipment (transformers, turbines) lead times have stretched 18-24 months industry-wide and remain a material constraint.

Threat of substitutes — meaningful, asymmetric across segments.

• Electric distribution: No substitute. Even rooftop solar requires a grid interconnect for net metering and reliability; nobody is going off-grid at scale.
• Gas distribution: This is where the substitute threat is real. Heat pumps, induction stoves, and electrification policy in California are pushing gas demand down on a 20-30 year horizon. SoCalGas's terminal value is the most uncertain piece of the Sempra portfolio.
• LNG export: Substitutes are pipeline gas (geographically constrained), other LNG suppliers (Qatar, Australia, US Gulf Coast), and renewables in the importing country. The competitive set is global LNG, not local. Long-term contracts mitigate this for the duration of the contract term.

Rivalry among existing competitors — almost none in regulated franchises, intense in LNG. SDG&E, SoCalGas, and Oncor each have de facto monopolies in their service territories. There is no rivalry. Sempra Infrastructure competes globally with Cheniere, Venture Global, ExxonMobil/QatarEnergy, Woodside, etc. for LNG offtake contracts. This is a real competitive market and Sempra does not have a cost-leader position (Cheniere Sabine Pass is widely regarded as the lowest-cost US LNG operator).

Value pool location and trajectory. Within Sempra, the value pool is shifting toward Texas (Oncor) and LNG (Sempra Infrastructure). California is becoming a capital-allocation question (how much more rate base will the CPUC authorize, at what allowed ROE, against what wildfire and decarbonization risk). The trajectory of US electricity demand is structurally up for the first time in 20+ years (data centers, electrification, reshoring of manufacturing). Texas is the epicenter of that shift, and Oncor is the toll road.

Industry verdict for the regulated utility leg: Excellent. Stable, monopolistic, recession-resistant — exactly the profile Buffett describes [5][6]. For the LNG leg: Good — long-term contracted, but with real EPC cost risk and global competitive dynamics.

Industry Verdict: Good.

Inversion

I now play the short-seller. The bear case for Sempra is stronger than most utility shorts because three independent risks compound.

1. The single event that kills this: a major SDG&E-attributed wildfire in California. The wildfire liability framework in California operates under inverse condemnation — utilities can be held strictly liable for fire damage caused by their equipment, regardless of negligence. PG&E went through this twice; the second time produced a $13.5B settlement and a Chapter 11. SDG&E has had a better operational record, but the risk is binary and existential. A single wildfire event tied to SDG&E equipment in a high-fire-threat district during a Santa Ana event could produce liability claims in the $5-20B range. Even with the AB 1054 wildfire fund, the prudency review process creates years of regulatory uncertainty. Stock price impact would be 30-50% in a week. There is no operational improvement that fully eliminates this — climate change is making the fire envelope worse, not better. The 5-year probability is not 0%; it is plausibly 10-20%.

2. Why the moat is narrower than bulls think. Bulls cite the regulated franchise as permanent. It is permanent legally; it is not permanent economically. The CPUC has been progressively chipping away at SoCalGas's franchise value: building electrification mandates, restrictions on new gas hookups in some jurisdictions, lower allowed ROEs than peers, tighter scrutiny on every rate case. SoCalGas's terminal value on a 25-year horizon is genuinely uncertain. If 30-40% of the customer base electrifies by 2050, the rate base must be amortized faster, raising rates for remaining customers, accelerating departures — the death-spiral dynamic that frontier-region gas utilities (some northeastern peers) are already starting to model in their long-term planning. Bulls also cite Oncor as a growth gem, but Sempra owns 80%, not 100%, and the minority partner (Texas Transmission Holdings) limits Sempra's flexibility. The Sempra Infrastructure LNG leg has real EPC cost overruns at Port Arthur Phase 1, and Phase 2 / ECA Phase 2 are not yet sanctioned — the bull-case capex backlog is partly hypothetical.

3. Why management is worse than it appears. Share count up 10.9% over 10 years [scorecard] tells you management is comfortable issuing equity. At today's Px/IV of 0.51 [scorecard], any equity issuance is value-destructive — yet the multi-year capex plan implicitly relies on continued equity issuance. The company has not signaled a hard stop on equity issuance at current prices, and the 2023-2025 issuance pattern suggests they will keep tapping equity rather than cut growth capex or accelerate asset recycling. Compounding this: the 'profitability 11/100' [scorecard] sub-score reflects a 10-year ROIC of 0.0% [scorecard] — the scorer is saying that Sempra has not actually earned its cost of capital on a returns-on-incremental-invested-capital basis over the past decade, despite the apparent earnings growth. That is a damning data point. NOPAT has declined; the scorer's note 'NOPAT declined; ROIIC not meaningful' [scorer_notes] is exactly the kind of red flag a Buffett-style investor takes seriously. Bulls hand-wave this as a regulated-utility artifact (book-value-based ROIC mismeasures rate-base businesses); skeptics say it is real and the growth is being purchased at unattractive incremental returns.

4. What bulls are extrapolating that won't hold. Bulls extrapolate the Texas data-center load growth into Oncor rate base growth into a 6-8% EPS CAGR for the next decade. Three things could break this. (a) ERCOT transmission cost-allocation rules are under review, and a less-favorable allocation reduces Oncor's authorized capex pace. (b) Texas data center buildout could decelerate sharply if AI infrastructure spend cools (this is plausible — most data center demand forecasts assume continued exponential AI capex). (c) Tariff and trade policy under the current administration has raised transformer and electrical equipment costs by 20-40%, which the regulator will absorb only with a lag. The bull EPS CAGR could realistically be 3-5%, not 6-8%, which materially compresses the IV math.

5. Valuation trap (multiple compression / regime change). P/E TTM is 21.4 versus a 10-year average of 43.5 [scorecard]. Bulls read this as 'cheap.' Skeptics read this as 'reverting to a more normal level for a regulated utility, which historically trades at 13-17x earnings.' If the multiple stays at 18-20x and EPS grows at the cautious 3-5% rate, the stock compounds at a low-single-digit total return — well below what a 49% IV discount implies. The base IV of $186.91 [scorecard] depends on assumptions about long-run growth and discount rate that, if pressed harder, deliver a much narrower range. The reverse-DCF implied growth of 2.7% [scorecard] is genuinely undemanding — but the 2.7% is for a hypothetical present-value-of-cash-flows model that assumes today's owner earnings of $3.37B [scorecard] continue to grow at 2.7% forever. If owner earnings instead compress in the next 2-3 years (rising rates, Port Arthur cost overruns, California rate-case adverse outcome), the implied growth needed to justify $94.67 rises mechanically.

If I am right, the stock could be worth $55 within 3 years.

Lollapalooza Bias Check

Authority bias. I am citing Buffett's BHE letters [1][2][5][6] as confirming evidence for owning a regulated utility. Buffett's framing is correct, but BHE is not Sempra. BHE is wholly owned by Berkshire and benefits from Berkshire's cost-of-capital advantage, which Sempra does not have. I should be more careful not to import the Buffett halo wholesale.

Anchoring on the Px/IV ratio. The 0.51 ratio is anchoring me toward 'this is cheap, buy.' But the IV is itself a model output, and the model's biggest input is the discount rate and long-run growth assumption. If the discount rate is 1% too low, the IV is materially too high. I should test my conviction by asking: at what IV would I sell? If the answer is 'IV — I would sell at $186,' then I am implicitly trusting the model more than I should.

Recency bias. The 2024-2025 data center and AI capex narrative is fresh. Texas (Oncor) is the most direct beneficiary in the public utility space. I am partly buying SRE because Oncor's growth story is loud right now. I should ask: would I be buying SRE at $94.67 if the data center narrative did not exist? The honest answer is: probably yes, because the California rate-base alone supports an IV materially above $94.67 — but I should not let the data center narrative dominate the thesis.

Confirmation bias. I went into this expecting a regulated utility to be a Buffett-style compounder. The scorer composite of 68 [scorecard] supports that, but the profitability sub-score of 11 and the 0.0% 10-year ROIC [scorecard] are inconvenient facts I am tempted to dismiss as 'measurement artifact.' They might be artifact. They might also be real. The right response is to underwrite at the LOW end of the IV range ($104.16), not the base ($186.91).

Commitment / consistency bias. I do not own SRE. This is a fresh underwrite, so commitment bias is low. But: if I do not actively recommend a utility somewhere in this watchlist, I look one-dimensional. I should not let portfolio-level optics push me toward a 'Buy' I would not otherwise issue.

Social proof. SRE is widely owned by utility-focused mutual funds and is in the S&P 500. That is not a reason to buy. It is, if anything, a reason to ask why the market has the price at 0.51x base IV — what does the marginal seller see that I am missing? The honest answer is: California regulatory tail risk, equity issuance overhang, and the ROIC-not-meaningful flag.

Deprival super-reaction. I am not currently deprived of SRE; this is a clean look. Low salience here.

Incentive bias. None directly active in the analyst — I am not paid by Sempra, do not own the stock, and am not penalized for a Hold. The biggest incentive is the desire to find an actionable Buy, which biases me toward Buy more than the data warrants. I will lean Hold-with-a-Buy-trigger rather than Buy outright.

10-Year Outlook

Same fundamental business model in 2036? Yes, with high confidence on the regulated utilities (SDG&E electric, Oncor) and medium confidence on SoCalGas (gas distribution faces a long-tail decarbonization headwind). The LNG segment will look different — Port Arthur Phase 1 will be operating, ECA Phase 1 will be operating, and there will likely be at least one additional FID expanding capacity. The shape of the business — regulated wires + pipes + long-term-contracted LNG — is durable.

Customer base larger? Yes, near-certain. San Diego County and the LA Basin grow modestly. ERCOT's load is growing high-single-digits, driven by data centers, electrification, and population migration to Texas. Sempra's served customer base in 2036 will be meaningfully larger.

Profit per customer higher? Probably yes, but not dramatically. Allowed ROE in California has been declining (currently ~10% on equity, down from ~10.5% a decade ago) and Texas is at ~9.5%. The compounding comes from rate-base growth, not from rising profit per customer.

Moat wider? Roughly the same. The regulated franchise is already as wide as moats get; it cannot really get wider. SoCalGas's moat is narrowing slowly due to electrification policy. Oncor's moat is widening slightly as ERCOT grows and the cost of duplicating its T&D footprint rises with inflation.

Single biggest threat? California wildfire liability is the single biggest existential risk. Second: a hostile shift in CPUC posture toward SoCalGas accelerating gas-franchise erosion. Third: rising interest rates compressing the spread between allowed ROE and cost of capital.

The 10-year picture is fundamentally positive but with a non-trivial left tail. The math compounds; the regulatory tail is real; LNG execution is the swing factor on the upside.

CONFIDENCE: medium

Position Guidance

• Recommendation: Buy
• Conviction: medium
• Target buy price: $95 (current $94.67 already inside the buy zone; full-size buy below $90, scale buy below $80)
• Target trim price: $190 (above mid-range IV; full trim above $230)
• Position sizing: 2-3% starter, scale to 4-5% on weakness below $80; cap at 5% given California regulatory tail and the ROIC-not-meaningful scorer flag