Sempra: A Long-Term Owner's Research Report

Hunting ten-year five-baggers among great growth stocks — pressing the upside question: "Can it get much bigger?"

• QHow high is its market ceiling? Is it enlarging a slice of an existing pie, or creating an entirely new market?5/10

  A

  The ceiling is not high, and in essence it is "thickening an existing pie bounded by legal limits" rather than creating a new market. This point is key to Baillie Gifford's "5x in ten years" paradigm: the market size of a regulated utility is naturally boxed in by service territory, population, and regulated rates, and there is no internet-style exponential expansion.

  First look at how big the pie itself is and its nature. Sempra is a regulated energy-infrastructure holding platform, and within its three segments the ones that truly determine long-term value are the regulated grids/pipeline networks: California's SDG&E and SoCalGas, and Texas's Oncor. The report discloses that in 2025 the weighted-average regulated rate bases of SDG&E and SoCalGas were about $18.019 billion and $13.985 billion respectively, and Oncor about $31.5 billion — this is the accounting measure of the "pie." Its growth does not come from grabbing users, but from investing capital to form a larger rate base and then earning money at the allowed return rate approved by regulators. In other words, the ceiling equals "how much capital you are allowed to invest within the service territory, and how much return regulators let you earn," not a TAM that can be infinitely scaled.

  So is this pie thickening? Yes, and it is thickening faster than for traditional utilities. Oncor has announced a 2026–2030 base capital plan of about $47.5 billion (including about $9 billion in 2026 and about $10 billion in 2027), and at the parent level a 2026–2030 five-year capital plan of about $65 billion (official wording: "approximately $65 billion, revised up from the prior version's $5.6 billion"), of which more than 95% goes to Texas and California regulated utilities, and it has also identified about $9 billion of incremental potential Oncor capex. The drivers are real: Texas industrial reshoring, population inflows, and data-center load are enlarging the denominator of "how much capital this region allows you to invest."

  But we must be honest: this is "enlarging an existing pie," not "creating a new market." It is unlike a platform company opening up a pool of demand that did not previously exist — grids and gas distribution are century-old legacy infrastructure, and what Sempra does is invest more intensively within its own statutory-monopoly map, converting more capital "compliantly" into rate base. Texas LNG (Port Arthur) does carry a bit of "new-business" flavor, but the report makes clear this slice is volatile, and the 45% stake in SI Partners has already been priced at $10 billion for sale to KKR; after closing, Sempra will retain only a 25% interest and lose control, so it is increasingly not a "second growth pole" that public shareholders can fully enjoy, but is instead being actively shrunk.

  Conclusion: the ceiling is "moderate, visible, but capped by hard regulatory and geographic boundaries." It can support steady mid-single-digit rate-base growth, but it lacks the feature Baillie Gifford most loves — "conjuring up a new market ten times as large out of thin air." For growth investing, this is both its downside protection and its upside constraint.

  Jun 11, 2026
• QOver the next five years, can its revenue at least double? Is growth driven mainly by volume, price, or new business?4/10

  A

  Almost impossible. Doubling revenue over the next five years is unrealistic for Sempra; it is a mid-single-digit-growth regulated utility, not a fast-growth stock; and the drivers are mainly "volume" (capital investment expanding the balance sheet + load growth) rather than "price" or disruptive "new business." This question is crucial, because one of Baillie Gifford's hard thresholds is "double revenue in five years," and Sempra clearly falls short on this line.

  First, use first-hand data to pin down the scale. Sempra had 2025 total revenue of $13.702 billion (2024 was $13.185 billion) — doubling in five years would mean 2030 revenue needs to reach about $27.4 billion, an annualized rate of about 15%. But management's own guidance is in EPS terms rather than revenue: the company affirmed 2026 adjusted EPS guidance of $4.80–5.30, gave $5.10–5.70 for 2027, and an outlook of $6.70–7.50 for 2030. From 2025 adjusted EPS of $4.69 to a 2030 midpoint of about $7.1, that is an annualized EPS growth of about 8.7% over five years — this is the growth pace the company itself anchors to, and even profit does not reach "double in five years" (doubling requires ~15%/year), let alone revenue.

  Now look at the "composition" of growth. The vast majority is "volume": investing capital into the grid to form a larger rate base, then earning the regulated allowed return. The report discloses that 2025 PP&E capex reached $10.612 billion, and the five-year capital plan is about $65 billion (about $12.7 billion in 2026 alone); Oncor plans to invest about $9 billion in 2026 alone and add about 13,400 new customer connections. This is the source of "volume" — more grid assets, more load (industrial reshoring / data centers).

  There is a "price" component, but it is regulated, time-lagged, and not free pricing. On the Texas side, the PUCT approved Oncor's rate case on April 17, 2026, raising the annual revenue requirement by about $560 million (+8.7%), with authorized ROE of 9.75% and new rates effective June 1. This is a genuine price increase, but a single rate case contributes only a single-digit percentage of revenue increment, and requires waiting for a rate cycle that comes only once every few years — far from enough to support a doubling.

  What about "new business"? LNG (Port Arthur) is the only part with new-business elasticity, but the report repeatedly emphasizes it is volatile and carries high execution risk, and the 45% of SI Partners has already been priced at $10 billion for sale to KKR; after closing Sempra loses control and deconsolidates, so this slice's revenue contribution to public shareholders is being actively "sold down" rather than grown.

  In a word: growth is "volume-driven mainly, price secondary, new business shrinking," with a pace of mid-to-high single-digit EPS and slower revenue, making a five-year doubling unrealistic. This is precisely where it fails Baillie Gifford's growth threshold — a good business, but not a fast-growth business.

  Jun 11, 2026
• QFive years from now, what will take over as the next growth engine? Does this "second curve" exist today?5/10

  A

  Sempra's "second curve" is the Texas grid (Oncor); it already exists today, and it is exactly the direction the company is actively betting on — but honestly, it looks more like "an extension and acceleration of the same main curve" than a brand-new-in-nature growth pole. Baillie Gifford asks "what takes over five years from now," expecting a growth stock to have a new engine that can re-create a fresh space for imagination; Sempra's answer is to push its bet even more fully back onto its steadiest old core business.

  First, its "actively chosen successor." The report clearly shows management is undertaking a clearly directed capital reallocation: pulling firepower out of the volatile infrastructure/LNG platform and concentrating it in Texas and California regulated utilities. The evidence is a 2026–2030 five-year capital plan of about $65 billion, with over 95% going to Texas and California regulated utilities, of which Texas's Oncor is the core — Oncor's own 2026–2030 base capital plan is about $47.5 billion, up about 32% from the prior five-year plan, and the company has separately identified about $9 billion of incremental potential capex mainly supporting Oncor grid expansion. This is the hard evidence that "the next engine already exists today": it is not a slide-deck concept but capital that has already obtained PUCT rate approval and is being deployed.

  The demand driving it is also a real structural variable: Texas industrial reshoring, population inflows, and especially data-center load. Oncor disclosed in Q1 2026 that it added about 13,400 new customer connections in the single quarter and plans to invest about $9 billion in 2026; the report also states bluntly that "the Texas grid is not short of demand; it may in fact benefit from industrial reshoring, population migration, and data-center load." In this sense, Oncor is indeed a "second growth pole" with more upside than California gas.

  But we must apply a Baillie-Gifford-style honesty discount. First, in nature this curve is "doing the old business more densely and faster," not opening up a new market — it is still regulated transmission-distribution, still capped by the ceiling of rate base × allowed return, lacking the "brand-new market, non-linear volume ramp" feature Baillie Gifford prefers. Second, the LNG (Port Arthur) that truly carries a "new-business" flavor could have been the second curve, but the report makes clear the company is actively shrinking it: the 45% stake in SI Partners is priced at $10 billion for sale to KKR, and after closing Sempra retains only a 25% interest and loses control, so for public shareholders this slice is being "sold down" rather than nurtured into a successor. Third, California gas (SoCalGas) is the opposite — eroded over the long run by electrification and decarbonization policy, and the report judges its moat "no longer widens as a matter of course as it did ten years ago," making it even less likely to take over.

  Conclusion: the successor exists and is the highest-quality one (the Texas grid), and management is indeed sacrificing the short term for it, concentrating on the regulated return pool — this deserves credit. But it is a "certain but not sexy" handoff that can sustain mid-to-high single-digit growth yet cannot support a brand-new narrative that would re-rate the valuation. For the Baillie Gifford paradigm, this is a solid score, not an exciting one.

  Jun 11, 2026
• QWhat is its core competitive advantage? Will this moat widen or narrow over the next three to five years?5/10

  A

  The core moat is license/regulatory barriers plus the scarcity of regionally monopolistic network assets; over the next three to five years it is "evolving on a split path" — the Texas grid side is widening, the California gas side is slowly narrowing, and the whole is roughly stable. This is precisely why Sempra cannot be treated as a monolithic "no-brainer utility."

  First, clarify the moat's sources. Sempra's advantage is not brand, not network effects, and certainly not data, but the fact that "no one can replicate a grid/pipeline network": customers cannot swap out the transmission-distribution or gas-distribution network on their own, and a new entrant rebuilding a network needs both enormous capital and regulatory permission plus a long-term social license. The report therefore positions the moat as the overlay of three things: license/regulatory barriers, customer switching costs, and scale plus network-asset density. The yardstick to measure it is the regulated rate base: the report discloses that in 2025 the weighted-average rate bases of SDG&E and SoCalGas were about $18.019 billion and $13.985 billion respectively, and Oncor about $31.5 billion — the larger and denser the network, the deeper the moat.

  Why say the Texas side is "widening"? Because widening a moat requires two things in place at once: regulators allowing you to invest more capital, and allowing you to earn a reasonable return. Texas has both. The PUCT approved Oncor's rate case on April 17, 2026, with an authorized capital structure of 56.5% debt / 43.5% equity, authorized ROE of 9.75%, authorized cost of debt of 4.94%, an annual revenue-requirement increase of about $560 million (+8.7%), and new rates effective June 1; at the same time Oncor announced a 2026–2030 base capital plan of about $47.5 billion, up about 32% from the prior version. Investable capital is rising, the return mechanism is clear, and demand (industrial reshoring / data centers) is real — this is the textbook form of a widening moat.

  Why say the California gas side is "narrowing"? Because the same regulatory moat can both widen and be eroded in reverse by regulation. The report cites the 10-K explicitly: gas demand at SoCalGas/SDG&E is affected by the Southern California economy, alternative energy, consumer preferences, energy-efficiency programs, environmental regulations, and California policies that support electrification and renewable generation; the company also specifically flags that building gas bans, long-term gas-infrastructure planning, and decarbonization goals could harm asset utilization and returns. A more concrete cost has already occurred — the 2025 income statement includes a $651 million regulatory impairment / disallowed recovery, direct evidence that "a regulatory moat can also bite back." This part of the moat will not collapse immediately, but the era of "widening as a matter of course" is over.

  There is also a layer of "verifiability" discount to apply. How strong the moat is partly depends on whether shareholders can see it clearly. Sempra is somewhat weak here: Oncor is equity-method accounted (important to shareholders yet not fully consolidated into revenue/EBITDA), and SI Partners is in a transition of a 45% stake priced at $10 billion for sale to KKR and deconsolidation after closing, while the year-end 2025 balance sheet still carries $31.024 billion in held-for-sale assets and $11.704 billion in held-for-sale liabilities. The assets are very good, but the structural complexity makes "how deep the moat is" harder to verify precisely from the outside.

  Conclusion: the moat's strength is real and on the strong side (the report gives 4/5, which I agree with), but it is "dynamically split" — Texas widening, California gas slowly eroding, the whole stable — plus a verifiability discount from the complex accounting structure. Over the next three to five years, what to watch most is whether the pace of Texas widening can continue to outweigh California's policy erosion.

  Jun 11, 2026
• QIf its core business is disrupted, does it have the DNA to reinvent itself? How does it treat mistakes and bad news?4/10

  A

  For a regulated grid, "being disrupted" is not a one-night collapse but a slow policy erosion; Sempra's "self-reinvention DNA" for coping with this erosion is structural asset restructuring (selling the weak, buying the strong) rather than technological reinvention — its handling of bad news is transparent disclosure, but it hardly amounts to a bold self-revolution. For a utility, this question needs rephrasing: its core business is almost impossible to "disrupt head-on" by a startup, and the real disruption comes from the long-term de-valuation of a certain asset class (California gas) by regulation / the energy transition.

  First, where the "self-reinvention DNA" shows up. Facing California's long-term gas policy risk, Sempra's answer is not a technological gamble but a proactive capital-allocation shift: pulling firepower away from volatile, policy-disadvantaged assets and concentrating it in the more regulated, more stable-return Texas/California grids. The hardest evidence is selling the 45% stake in SI Partners to KKR for $10 billion and selling the Mexican Ecogas gas-distribution network for about $500 million, both expected to close in Q2 to Q3 of 2026, while raising the five-year capital plan to about $65 billion, over 95% directed to Texas and California regulated utilities. This is a kind of "portfolio-level self-reinvention" — continuously converging the company toward the most disruption-resistant direction by buying and selling assets. It is genuinely effective, but in essence a financial-engineering shift, not the regenerative ability of "starting over after the core business is disrupted." Frankly, a grid company neither needs nor is good at the latter.

  Now "how it treats mistakes and bad news" — this is exactly where the company is relatively commendable: it tends to put bad news on the table all at once and transparently, rather than hiding it. The most typical case is that in 2025 it recognized a $651 million regulatory impairment / disallowed recovery, directly compressing full-year GAAP profit from $2.817 billion in 2024 to $1.796 billion, and GAAP diluted EPS from $4.42 to $2.75. The company did not use adjustment items to mask this, but itemized it in the GAAP↔adjusted bridge table (regulatory impairment, Mexican currency and inflation positions, derivatives, Port Arthur LNG interest-rate swaps, held-for-sale taxes, etc.). I agree with the report's judgment on this: the company is "not obviously manipulating accounting," and bad-news disclosure is clear — but what investors must guard against is the other side, the complexity of "looking very stable while actually carrying many project and financial disturbances."

  But three honesty discounts apply. First, for assets like California gas that are "slowly disrupted by policy," the company's handling is more about acknowledging and isolating risk than reshaping it into new growth — the report's sell conditions state plainly that "if California significantly accelerates de-assetization and de-recovery, it would mark SoCalGas's value multiple lower," showing management itself holds no card to reverse this erosion curve. Second, transparent disclosure ≠ a proactive error-correction culture; a utility's "reinvention" window is constrained by rate cycles and the regulatory cadence, so the speed of self-revolution is naturally slow. Third, safety and operations are the most lethal source of "bad news" for this kind of company (California wildfires especially), and the report notes the company supports California SB254 to strengthen the wildfire fund and that SDG&E has won reliability awards for years running — a positive signal, but this is more soundness at the compliance and operating level than a post-disruption regenerative DNA.

  Conclusion: it has the pragmatic DNA of "portfolio self-reinvention" (selling the weak, buying the strong, concentrating on the regulated return pool), and its attitude toward mistakes and bad news is honest disclosure with no hidden landmines — a plus; but it lacks the technological regenerative power Baillie Gifford prefers, the kind that "rises from the ashes even when the core is disrupted," and facing the chronic disruption of California gas it can manage but not reverse it.

  Jun 11, 2026
• QDoes management (especially the founder) have a long-term vision and interests deeply tied to the company? Are they willing to sacrifice current profit for five to ten years out?4/10

  A

  Management has a long-term vision, fairly strict governance and ownership requirements, and highly long-term-oriented compensation; interest alignment is "qualified and on the strong side" — but this is a regulated utility run by professional managers, without founder-style deep alignment, and its move to "sacrifice current profit for the future" shows up more in capital-allocation discipline than in actively depressing earnings to bet on the long term. Baillie Gifford places particular weight on founder long-termism; on this dimension Sempra must honestly be marked down a notch.

  First, the hard evidence on interest alignment, which is genuinely solid. The report discloses that the CEO ownership guideline is 8x base salary, and as of March 20, 2026, Chairman and CEO Jeffrey W. Martin's holdings (per company rules) were about nearly 37x his base salary, far exceeding the requirement, with other named executives also meeting or exceeding it; the most recent say-on-pay shareholder support was about 95%; the company also disclosed clawback, anti-hedging, and anti-pledging policies — the governance framework has no obvious red flags. This shows executives' personal wealth is indeed bound to the stock price to a considerable degree, not a "collect a fixed salary and walk away" mindset.

  Compensation design is also clearly long-term-oriented. The report discloses that of the CEO's target total compensation about 91% is at-risk pay and about 74% is long-term equity incentive; the annual bonus is tied mainly to ABP Earnings (weighted 80%), safety metrics (12%), and responsible business practices (8%); long-term incentives combine relative TSR, EPS growth, and options/restricted stock. For a regulated-utility holding company, this design at least shows management is not optimizing purely for short-term EPS — a genuine signal of long-term vision.

  "Are they willing to sacrifice the present for five to ten years out?" — the answer is "willing at the capital-allocation level, limited at the profit-sacrifice level." On the positive side, management is indeed doing several long-focused, short-term-unflattering things: concentrating capital in the more-stable-but-short-term-unsexy Texas/California regulated businesses (a five-year plan of about $65 billion, over 95% into Texas and California, raised up from the prior version's ~$5.6 billion/year pace); proactively selling 45% of SI Partners (priced at $10 billion to KKR) and Mexican assets to release capital and reduce future reliance on external equity financing, even though this means giving up control of the LNG platform and some upside. This is a real trade-off of "sacrificing short-term scale/elasticity for long-term quality." But on the other side, it has not, like the founders Baillie Gifford loves, "actively depressed current profit to invest heavily in the long term" — it still affirms 2026 adjusted EPS of $4.80–5.30 and gives clear year-by-year growth guidance to $6.70–7.50 for 2030, which is essentially "steady-delivery type" rather than "sacrifice-the-present, gamble-on-the-future type."

  Two discounts must apply. First, there is no founder. This is a professional-manager / engineering-and-regulation team, aligned by institutional design (ownership guidelines + long-term equity) rather than the founder's kind of intrinsic long-termism where "the company is my life's work" — in the Baillie Gifford paradigm this is a structural deduction. Second, capital allocation is "right in direction but not exceptional." The report puts it plainly: this is a highly capital-intensive enterprise that needs continuous financing, with weighted-average shares rising from about 630 million in 2023 to about 652.7 million in 2025, and 2024 common-stock issuance bringing in about $1.219 billion in cash while buybacks were tiny — it looks more like doing "balance-sheet management" than "big buybacks when undervalued to compound per-share value for shareholders." Rational, but not a top-tier capital allocator.

  Conclusion: worth handing your money to them to run utility assets (the report gives management 3/5, which I agree with) — clean governance, ownership alignment in place, long-term vision, transparent bad news. But it only partly satisfies the two traits Baillie Gifford values most — "founder-level deep alignment + actively sacrificing profit for the long term" — so one should not pay a valuation premium for management.

  Jun 11, 2026
• QIf it vanished tomorrow, how much would customers miss it? Is its way of growing sustainable and not reliant on harming society and regulators?6/10

  A

  If Sempra vanished tomorrow, customers would "miss it intensely" — it supplies electricity and natural gas, irreplaceable necessities, so its indispensability is nearly full marks; but its way of growing is split on "social/regulatory sustainability": the Texas grid side is highly sustainable, welcomed by regulators and society, while the California gas side has long-term tension with the social policy direction of decarbonization/electrification. This question must be split into two layers, "indispensability" and "sustainability," and examined separately.

  First, indispensability — this is Sempra's strongest side. Its customers are not "contestable traffic" but energy demand that must be supplied within a statutory service territory. The report discloses that in 2025 SDG&E served electric customers covering a population of about 3.6 million and gas customers covering a population of about 3.3 million, SoCalGas had about 6.21 million gas meters, and Oncor is the regulated transmission-distribution utility for north-central, eastern, western, and panhandle Texas. These customers cannot swap out the grid or gas-distribution network on their own. So the answer to "how much would you miss it if it vanished tomorrow" is nearly extreme: once electricity and gas supply is cut, society is instantly paralyzed — this is the highest-certainty necessity, and indispensability can be scored close to full marks. This is also the foundation of its defensiveness.

  But the real test point of this Baillie Gifford question is the second layer: "is the way of growing sustainable and not reliant on harming society and regulators." Here Sempra is split, and must be discussed in two parts.

  The Texas grid side — highly sustainable, and even "actively needed by society and regulators." Its growth comes from meeting real load (industrial reshoring, population inflows, data centers), and it earns returns within the regulatory framework through public rate proceedings: the PUCT approved Oncor's rate case on April 17, 2026, raising the annual revenue requirement by about $560 million (+8.7%), with authorized ROE of 9.75% and new rates effective June 1, corresponding to an increase of about $7 per month for residential customers. This is growth where "society wants it to invest more and regulators approve it to invest" — strongly sustainable, and not reliant on harming consumers or exploiting regulatory loopholes.

  The California gas side — sustainability in question, and this is exactly the most fragile part of its way of growing. The report cites the 10-K explicitly: gas demand at SoCalGas/SDG&E is affected by California energy policies that support electrification and renewable generation, as well as building gas bans, decarbonization goals, and long-term gas-infrastructure planning. In other words, California's social and regulatory direction is itself "gradually reducing reliance on gas infrastructure," while the return on this Sempra asset depends on continuing to invest and being allowed to recover. When the direction of social policy is opposite to the direction of your asset utilization, "growth" may be judged as something that should no longer be recovered — that $651 million regulatory impairment / disallowed recovery in 2025 is the concrete cost of this tension. This is not Sempra "harming society"; it is society's priorities changing, no longer paying for certain gas investments.

  One utility-specific dimension of "social license" should also be added: safety (California wildfires especially). The report notes the company supports California SB254 to strengthen the wildfire fund, and that SDG&E has won reliability awards for years running — this shows it is working to maintain its "social license," because once a utility has a major safety incident, regulatory and social license can tighten instantly. This part is currently positive.

  Conclusion: indispensability near full marks (necessity, statutory monopoly, irreplaceable); growth sustainability "strong in Texas, weak in California gas." Overall it does not grow by harming society or arbitraging regulation, but it must actively manage the long-term tension between California gas and decarbonization policy — which is also why the report lists it as one of the biggest uncertainties. For the Baillie Gifford paradigm, this is a combination of "an extremely stable floor + one structural policy risk."

  Jun 11, 2026
• QWhat are this business's unit economics (gross margin, incremental returns)? Does it improve or worsen as it scales? Where is the money it earns spent?5/10

  A

  The unit economics are the typical look of a "regulated model": book margins are not low, but incremental returns are capped by "allowed ROE," and every slice of growth swallows enormous capital — so scaling up does not make unit economics significantly better, and almost all the money earned is reinvested back into the grid rather than returned to shareholders. This is the most fundamental divide between Sempra and the "high incremental returns, low reinvestment pressure" model Baillie Gifford prefers.

  First, the margin layer. Sempra had 2025 total revenue of $13.702 billion and GAAP net income of $1.796 billion, a net margin of about 13.1% — but that year was depressed by a $651 million regulatory impairment / disallowed recovery and interest expense rising from $1.049 billion to $1.532 billion; excluding significant items, 2025 adjusted profit was about $3.066 billion and adjusted EPS $4.69 (2024 was $4.65), so the normalized net margin is actually around 20%. So "earning power" itself is not weak — the problem is not gross margin but the ceiling on incremental returns and the reinvestment intensity.

  Now "incremental returns" — the hard constraint of a regulated business. The return on new capital is not set by the market but locked by the regulatory-authorized ROE. The Texas side is clearest: the PUCT approved Oncor's authorized ROE of 9.75%, a capital structure of 56.5% debt / 43.5% equity, and an authorized cost of debt of 4.94%. This means that for every $1 invested in rate base, the ceiling on the equity portion's return is roughly about 9.75% — even the best execution can hardly break through this regulatory cap. This is a different species from the "incremental capital can earn 30%+ returns" businesses Baillie Gifford loves: here incremental returns are "reasonable but capped" high single digits.

  "Does it improve or worsen as it scales?" — roughly flat, even slightly dragged down, not a scale dividend. The evidence is that book return on capital is declining: the report discloses ROE fell from about 10.9% in 2023 to about 9.4% in 2024 and then about 5.7% in 2025 (2025 distorted by impairment and held-for-sale reclassification). Even on a normalized basis, the regulated model has no internet-style leverage of "the larger the scale, the lower the marginal cost" — every additional region served and every additional stretch of grid built requires matching capital and regulatory approval, so unit economics are basically linear and do not improve structurally with size. Its benefit is stability and repeatability, not increasing returns to scale.

  "Where is the money it earns spent?" — this is the most crucial and most honesty-demanding point: almost all of it is reinvested back into the grid, and free cash flow is negative for the long term. The report discloses 2025 operating cash flow of $4.565 billion, but PP&E capex as high as $10.612 billion (about $12.627 billion including investment capex), traditional free cash flow deeply negative, and this is a structural feature of the past five years rather than a one-off. In other words, the cash the unit economics generate not only has no surplus to distribute to shareholders, but needs external financing to fill the gap — 2024 common-stock issuance brought in about $1.219 billion in cash, and the weighted share count rose from about 630 million to about 652.7 million in 2025. Where does the money go? Into Texas/California rate-base expansion (a five-year capital plan of about $65 billion, over 95% into regulated utilities), plus an annualized dividend of about $2.63/share. For a regulated growth stock this logic is self-consistent (capital → larger rate base → future allowed returns), but it means "the more it grows, the more capital it needs" rather than "the more it grows, the more cash it has."

  Conclusion: unit economics are sound and repeatable, with a normalized margin of about 20%, but incremental returns are capped by the regulatory ROE, scale brings no increasing-returns dividend, and the cash generated is entirely swallowed back by capex, with free cash flow negative for the long term. This is a "good but not light" business — it earns money steadily, yet lacks the "high incremental returns + self-funded snowballing" feature Baillie Gifford values most.

  Jun 11, 2026
• QFor it to 5x in ten years, what conditions must hold simultaneously? Are these conditions realistic? What expectations does today's stock price imply?2/10

  A

  Almost unrealistic. For Sempra to 5x in ten years requires about 17.5% annualized price return, yet it is a regulated utility whose own management guides only to mid-to-high single-digit EPS growth, with a dividend of about 2.9% and a current valuation that is on the expensive side — the required conditions either exceed the business's physical ceiling or need a large valuation expansion, and multiple low-probability events must hold at once. This is the question in the Baillie Gifford ten where Sempra should least be forced into a growth narrative, and I give a straight negative judgment.

  First, translate "5x" into hard metrics. 5x in ten years = annualized total return of about 17.46%. For a stock, long-term total return ≈ per-share earnings growth + dividend yield ± valuation change. Sempra's current dividend yield is about 2.9% (annualized dividend of $2.63/share, current price about $91). That is, earnings growth alone must contribute about 14.5%/year — and the company's own guidance is far below that.

  What growth pace does the company anchor to? Sempra affirmed 2026 adjusted EPS of $4.80–5.30, gave $5.10–5.70 for 2027, and an outlook of $6.70–7.50 for 2030. From 2025 adjusted EPS of $4.69 to a 2030 midpoint of about $7.1, that is an annualized EPS rate of about 8.7%. Even generously assuming this pace continues into a second five years, EPS growth is only 7%–9%/year — plus about 2.9% dividend, with valuation unchanged, long-term total return is roughly 10%–12%/year, corresponding to about 2.6–3.1x over ten years, a structural gap from 5x. This matches the report's judgment: the report puts the neutral long-term annualized return at the current price at only about 5%–8%.

  So what conditions must hold at once for "5x"? At least four, and none is easy: ① EPS growth jumping from about 8.7% and sustaining above 14% long-term — but the regulated model's incremental returns are capped by the authorized ROE of 9.75%, so structurally it is hard to keep running out mid-teens growth; ② massive capex delivered at high quality — the about $65 billion over five years, about $12.7 billion in 2026 alone must all convert into rate base and per-share earnings, not just enlarge the balance sheet; ③ financing without dilution — but the report already shows weighted shares rising from about 630 million to about 652.7 million, with free cash flow negative for the long term and growth relying on external financing, so per-share growth is easily swallowed by dilution; ④ the valuation still has to expand further — but at a trailing P/E of about 31x and a forward P/E of about 17x it is already not cheap, leaving limited room for multiple expansion to contribute a few more percentage points of annualized return. Hitting all four is a stacking of low probabilities.

  "What expectations does today's stock price imply?" — it implies a "normal-to-slightly-expensive price for high-quality delivery of regulated growth," not an undervaluation. The report's three-method valuation gives: conservative intrinsic value about $50–65, fair about $75–90, optimistic about $105–125; the current about $91 sits at "the upper edge of the fair range, slightly above." Cross-sectionally, SRE's forward P/E of about 17x is higher than NextEra's about 22x is an exception, but relative to Duke's about 19x and an industry average of about 17x it is not cheap, and the trailing end of about 31x is clearly elevated. In other words, the market has already priced in part of the good thing that "Texas growth will be delivered" — you are not buying unrecognized growth at a discount, but paying a normal-to-high price for growth that has already been seen. This means that even if the business delivers on schedule, the triple dividend that 5x requires — "cheap entry + fast growth + further valuation expansion" — has already had its most critical first component spent by today's price.

  Conclusion: 5x in ten years is unrealistic for Sempra — it would need to force a business that is capped by the regulatory ROE, with negative free cash flow and a currently expensive valuation, into a fast-growth stock at 17%+ annualized, requiring multiple low-probability conditions to hold at once. A more reasonable expectation is a neutral annualized return of about 5%–8% and optimistic about 8%–11% (the report's judgment, which I agree with). It wins on defensiveness, not on the odds of a 5x.

  Jun 11, 2026
• QWhy hasn't the market realized all this yet? Is it that they can't understand it, look down on it, or can't see far enough? What will become the "narrative inflection point"?3/10

  A

  The most honest answer is: the market has basically "understood" Sempra, and there is no severely mispriced perception gap waiting to be discovered — its current price of about $91 has already priced in the good thing of Texas growth to a level that is reasonable-to-slightly-expensive. So this question for Sempra should be answered in reverse: rather than asking "why hasn't the market realized it," it is better to acknowledge that "the market has realized it and has accordingly given a normal-to-high valuation." This does not match the growth-stock picture Baillie Gifford expects of "the market can't see far, leaving mispricing," and forcing the fit would mislead.

  Why say the market has basically understood it? Check three layers one by one.

  "Can't understand it" — partly true, but does not constitute systematic undervaluation. Sempra's financials are indeed complex: Oncor is equity-method accounted (important to shareholders yet not fully in consolidated revenue/EBITDA), 45% of SI Partners is priced at $10 billion for sale to KKR and deconsolidates after closing, and year-end 2025 carries $31.024 billion in held-for-sale assets and $11.704 billion in held-for-sale liabilities. The report also states bluntly that P/FCF and EV/EBITDA are "almost meaningless / not clean enough" here. But this complexity brings valuation "two-way noise" — it can lead people to overvalue as well as undervalue, rather than a systematic discount. In fact, a trailing P/E of about 31x and a forward P/E of about 17x show the market is assigning a high rather than a low multiple, indicating that complexity has not beaten it down into a "cheap cigar butt."

  "Look down on it" — not the case. It is not a shunned, obscure stock. The company clearly defines its mission as "building America's leading utility growth platform," and raised the five-year capital plan to about $65 billion, over 95% into Texas and California, with a 2030 EPS outlook of $6.70–7.50; sell-side and institutional coverage is ample. It is precisely because the market "thinks highly of" its Texas story that it is willing to give it a relative valuation above Duke's about 19x and the industry average of about 17x.

  "Can't see far enough" — this is the only place where disagreement may exist, but the direction is neutral. Bulls would say the market has not yet fully priced in Texas load (data centers / industrial reshoring) and Oncor's $47.5 billion capital plan; bears would say the market has not yet fully demanded a discount for California gas policy risk and long-term dilution. The report's judgment is balanced and key: its best part (Texas/California regulated growth) the market has partly seen, and its worst part (gas policy risk, financing dilution, project volatility) the market has not yet demanded enough of a discount for — that is, the perception gap is a two-way hedge, with no clearly one-sided undervaluation direction.

  So "what will become the narrative inflection point"? Precisely because it is not a mispriced target today, a future re-rating is more likely to be triggered by specific events than by "the market suddenly understanding." I would watch four kinds of inflection points: ① asset rotation landing — the SI Partners and Mexican Ecogas transactions are expected to close in Q2 to Q3 of 2026; if they close smoothly and clearly reduce reliance on external equity financing, the narrative may shift from "complex, needs constant financing" to "purer, more regulated," a potential positive inflection point; ② Texas load delivery — if data-center/industrial load keeps pushing up Oncor's rate base and earnings, the market may revise long-term growth upward, and Oncor has already identified about $9 billion of incremental capex as an observation window; ③ California gas policy — if California significantly accelerates building gas bans / de-assetization and, following the $651 million regulatory impairment in 2025, more unrecoverable assets appear, that would be a negative inflection point; ④ dilution and per-share value — if the common-share count keeps rising faster than per-share earnings growth and dividend coverage does not improve, the negative narrative of "growth being swallowed by financing" would strengthen.

  Conclusion: Sempra is not a hidden growth stock "mispriced because the market can't understand it / can't see far" — it is already priced reasonably-to-slightly-expensive, with a two-way-hedged perception gap and no clear undervaluation direction. The real inflection points come from trackable events like asset rotation, Texas load, and California policy, not from a market awakening in perception. For the Baillie Gifford paradigm, this precisely shows it lacks the upside fuel of "mispricing + narrative re-rating," and is better suited to the high-quality watch list, waiting for a cheaper price, rather than betting on a narrative inflection.

  Jun 11, 2026