The Williams Companies: A Long-Term Owner's Perspective

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 an existing pie, or creating an entirely new market?5/10

  A

  Conclusion: the ceiling is not low, but Williams is "enlarging an existing big pie" rather than creating an entirely new market; it shares in the structural upswing of U.S. natural gas demand and does not open up a new demand category itself. By Baillie Gifford's "five-fold in ten years" yardstick, this means its upside looks more like "hitching a ride on high-quality infrastructure" than a disruptive exponential expansion.

  First, how big the pie itself is. According to company disclosure, Williams's assets carry roughly one-third of all U.S. natural gas, centered on interstate pipelines such as Transco. Its "market" is the transportation, gathering, processing, and storage of U.S. natural gas: a mature pool that has existed for decades and is measured in the trillions of cubic feet annually. The demand-side tailwind is real: the U.S. Energy Information Administration (EIA) Short-Term Energy Outlook expects Lower-48 U.S. natural gas production to keep growing in 2026 versus 2025, and expects total U.S. electricity use to set successive records in 2026 and 2027; the International Energy Agency (IEA) data-center electricity report expects global data-center electricity use to nearly double to about 945TWh by 2030 in the base case. These will all pull demand for generation gas and supporting pipelines, and the report's description is consistent with the primary sources.

  But we must honestly distinguish "enlarging the pie" from "creating a new market." Williams has not invented a new energy form or a new demand scenario; it channels the incremental gas demand from LNG exports, gas-fired power, industry, and data centers more into its own existing scarce corridors. In other words, AI and data centers are new markets created by others, and Williams is one of the "toll-road contractors" for this new demand. This is an excellent position, but in essence it is absorbing rather than creating.

  The hard constraints on the ceiling must also be spelled out. First, regulated by FERC, new interstate facilities require a Certificate of Public Convenience and Necessity, and the high entry threshold is both a moat and a speed limit on expansion; second, the report cites a May 2026 Reuters report noting that clean-energy generation's share in the ERCOT region exceeded fossil energy for the first time, showing that natural gas in the power system carries "peaking/mid-load" duty rather than unlimited baseload growth, which sets a ceiling on the long-term growth multiple.

  Overall, this is an existing market that is "large and steady, and still structurally growing"; Williams's share and positioning are high-quality, but it shares in industry growth rather than creating a new track itself. The market ceiling is enough to support steady compounding, but hard to alone support a Baillie Gifford-style five-fold imagination, consistent with the report's tone of "a good business, but with bounded growth imagination."

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

  A

  Conclusion: doubling revenue over the next five years is almost impossible, and on this measure Williams clearly falls short. Growth is driven mainly by "volume + rates" (throughput from capacity expansion, and net rates from re-contracting and expansion), with almost no "new-business"-style jump. Placed under Baillie Gifford's hard threshold of "doubling revenue in five years," the answer is a clear no.

  First, do the math with the company's own guidance. At its 2026 analyst day Williams laid out a medium-term framework of roughly 10% compound adjusted-EBITDA growth and 5%–7% dividend growth through 2030 (see the Williams 2025 results and analyst-day press release). A 10% EBITDA CAGR means roughly 60% cumulative growth over five years, clearly short of a "double" (which needs about 15%/year). More importantly, for a natural gas midstream company, revenue is heavily disturbed by commodity prices and derivative fair values, and the company itself watches adjusted EBITDA and AFFO/share rather than total revenue—and these cleaner measures likewise point to "steady compounding" rather than "doubling."

  Break down the sources of growth, and it is volume and price, not new businesses. According to the Williams Q4 and full-year 2025 results 8-K, 2025 Transmission, Power & Gulf segment adjusted EBITDA reached $3.720 billion (versus $3.307 billion in 2024), with the increment coming mainly from higher Transco net rates and expansion projects; the same filing shows 2025 adjusted EBITDA of $7.750 billion, and the company guided 2026 to $8.05–8.35 billion—a single-digit-percentage step, delivered by projects coming online one by one (volume) and rates/re-contracting (price).

  Driving this growth curve is a round of high-intensity capital spending, not an asset-light demand explosion. The same 8-K shows 2026 growth-capex guidance of $6.1–6.7 billion (note: the "$7.0–7.6 billion" written in the report body does not match the company's primary guidance; use $6.1–6.7 billion), with maintenance capex of $850–950 million. In other words, for every dollar of EBITDA growth, Williams must first spend real money building pipelines, compressor stations, and processing facilities—a classic case of "capital for growth," which keeps its growth rate naturally constrained by project cadence and return rates in the mid-single to low-double digits, structurally ruling out a five-year revenue double.

  So the honest answer to this question is "no": Williams's growth is real, visible, and supported jointly by volume and price, but the magnitude is steady compounding rather than a double, and it depends heavily on continued large capex. It clearly loses points on this hardest Baillie Gifford growth threshold—consistent with the report's judgment of "good asset quality, but bounded growth imagination."

  Jun 11, 2026
• QFive years from now, what will take the baton as the next growth engine? Does this "second curve" exist today?4/10

  A

  Conclusion: Williams's "second curve" already exists today and is visible—it is the handoff from traditional gathering and processing to "gas-fired power / data-center power + LNG-export support," and the company has named a segment Transmission, Power & Gulf precisely to signal this intent. But this curve is an extension of the same business model (toll infrastructure), not an independent new engine that can change the company's nature.

  First, it does indeed exist and is not a pipe dream. According to the Williams Q4 and full-year 2025 results 8-K, of 2025 adjusted EBITDA of $7.750 billion, the Transmission, Power & Gulf segment alone contributed $3.720 billion (about 48%), with the increment coming clearly from Transco expansion and higher net rates; at its analyst day the company listed power innovation as a key contributor to double-digit EBITDA growth through 2030, backed by the demand pull described in the IEA data-center electricity report, where global data-center electricity use nearly doubles by 2030. This means the baton-holder is not "some future R&D," but an already revenue-generating and largest active segment that is accelerating.

  But we must honestly assess the quality of this "second curve." In the Baillie Gifford context, an ideal second curve is something that lets the company jump beyond its original ceiling and build a new growth pole (for example, a platform company growing an independent second business from a first). Williams's "power/data-center/LNG support" is in essence still "building toll corridors for new destinations of natural gas molecules"—its business model, regulatory framework, and return mode are entirely consistent with the main business. It is a high-quality extension of the main curve that will lengthen the growth duration, but it will not turn Williams into a different kind of business. It is a plus, but not a magnitude-changing second engine.

  Supporting this curve is a round of real-money project spending. The same 8-K shows the company's 2026 growth-capex guidance of $6.1–6.7 billion (the "$7.0–7.6 billion" written in the report body does not match the primary guidance; use $6.1–6.7 billion), on top of the roughly $23 billion project backlog at about a 5x build multiple disclosed at the analyst day (see the Williams 2025 results and analyst-day press release). This shows the second curve is "built by spending money," and its realization depends heavily on projects coming online on schedule and reaching their return rates—also the core uncertainty the report repeatedly emphasizes.

  The risk must also be spelled out: one demand premise of this curve (gas-fired generation's marginal position) is not unconditionally secure. The report cites a May 2026 Reuters report noting that clean-energy generation's share in ERCOT has already surpassed fossil for the first time. If new power load is increasingly absorbed by wind, solar, storage, and nuclear, the slope of this power/data-center baton curve would be dampened.

  Overall: the second curve genuinely exists, is right in front of us, and is the current largest segment accelerating—better than many companies whose "second curve is still in the PowerPoint"; but it is an extension of the main business rather than a fresh start, and its magnitude is still constrained by the same infrastructure economics. It can lengthen the compounding duration, but can hardly single-handedly support a Baillie Gifford-style five-fold leap.

  Jun 11, 2026
• QWhat is its core competitive advantage? Over the next three to five years, will this moat widen or narrow?6/10

  A

  Conclusion: the core competitive advantage is "irreplaceable interstate-pipeline positioning + FERC regulatory entry barriers + high switching costs," a real and strong moat; over the next three to five years it will most likely "widen slightly," but the widening is gradual, coming from demand concentrating into scarce corridors rather than from the moat itself changing in kind. This is the most solid item across Williams's whole scorecard.

  The moat's foundation is assets and regulation, not brand. The report accurately points out that Williams's strongest moat lies in network position, irreplaceable assets, regulatory permits, scale, and low substitutability—Transco connects America's most important supply areas, consumption areas, and LNG and generation load centers. Replicating an interstate network of the same class requires first clearing FERC's Certificate of Public Convenience and Necessity, environmental approvals, land coordination, and years of construction, which naturally forms an extremely high entry barrier. Once a customer is connected to the existing system, switching or building an alternative pipeline is costly and slow, so switching costs are strong. This kind of advantage does not rely on user preference; it is cycle-resistant and substitution-resistant, one of the hardest kinds in infrastructure.

  Whether it "widens or narrows" should be judged by evidence, not emotion. The widening direction has data support: according to the Williams Q4 and full-year 2025 results 8-K, 2025 was the company's 13th consecutive year of adjusted-EBITDA growth, and the Transmission, Power & Gulf segment contributed $3.720 billion of EBITDA on higher Transco net rates and expansion—showing that, in a landscape where new interstate approvals grow harder and demand concentrates more into existing corridors, the scarcity and bargaining power of the installed network are rising. Looking ahead, the EIA Short-Term Energy Outlook expects U.S. electricity use to set successive records and production to keep growing, which will further lift Transco's utilization and expansion opportunities. These are all signals of the moat widening at the margin.

  But we must separate "once strong" from "stronger in the future." The consecutive-growth record above proves the moat has not been eroded to date; what truly determines its width over the next three to five years is two forward-looking things: one, whether the roughly $23 billion project backlog at about a 5x build multiple (see the Williams analyst-day press release) is delivered at its return rate—if built, it physically widens the moat; if returns fall short, it merely enlarges the asset base; two, whether natural gas's marginal position in the power system holds—the report cites a May 2026 Reuters report that clean-energy generation's share in ERCOT has already surpassed fossil, and if gas only handles peaking long term, the moat's "growth premium" would narrow, even though the "not-replaceable" chassis remains.

  Pricing power should be noted briefly: it is not consumer-style free price-setting, but a "mechanism-based" return improvement driven by contracts, rates, and expansion, and 2025 and Q1 2026 results already reflect the increment from higher Transco net rates. This is a steady, understated kind of pricing power.

  Overall judgment: the moat is strong and, in the base case, widens slightly over three to five years, mainly from demand concentrating into scarce corridors + new projects landing; but the realization of that widening depends on project returns and gas's marginal position, and there is a tail risk of the growth aspect being narrowed by the evolving energy mix. The chassis is extremely hard; the sustainability of the growth premium is the real variable—consistent with the report's conclusion of "the moat is stable and slightly widening."

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

  A

  Conclusion: Williams's "self-reinvention DNA" is limited and mild—it is not a company that survives through violent transformation, but one that keeps extending the same toll-infrastructure through stable operations; the upside is that it does not need disruptive reinvention, the downside is that if the core demand logic really collapses, it lacks the ability to switch tracks quickly. In handling mistakes and bad news, its style is "institutionalized, conservative, and fully disclosed," rather than founder-style aggressive course-correction.

  First, respond to the implicit premise of "what if the core business is disrupted." Natural gas midstream is "disrupted" not by a better competitor appearing, but by a shift in end-demand structure—the power system marginalizing gas. The report cites a May 2026 Reuters report that clean-energy generation's share in ERCOT has already surpassed fossil, precisely this kind of chronic erosion. Facing such risk, Williams's response is not to "reinvent into a new-energy company," but to extend its assets along natural gas's new destinations (LNG exports, data-center power supply)—in essence still the same toll business. In other words, its "reinvention DNA" is the adaptability to graft the same capability onto new demand, not the regenerative power to start over from scratch. This is reasonable in a heavy-asset regulated industry, but it truly lacks the DNA of "setting up a new operation after the core is disrupted," and that must be honestly acknowledged.

  Historically, it went through a real "life-or-death moment" and survived, showing the resilience of its chassis. In 2015 Williams had a famous, ultimately-collapsed M&A tug-of-war with Energy Transfer, which produced long-running litigation; 2023 earnings even included a $534 million Energy Transfer litigation judgment (see the discussion of adjustment items in the Williams Q4 and full-year 2025 results 8-K). It was not dragged down by that turmoil, but returned to focusing on its natural gas infrastructure main business—an error-correction path of "contracting the focus, repairing the balance sheet," defensive rather than aggressive.

  In handling mistakes and bad news, what is visible is "institutionalization and transparency." Per the same 8-K, the company explicitly excludes unrealized derivative gains/losses, asset impairments (about $212 million of impairment/write-offs recognized in 2025), litigation judgments, and other one-time items from its adjusted figures, disclosing noise separately from sustainable earnings; the report also judges that "no clear red flags of financial fraud or aggressive accounting are seen, and disclosure is adequate." Add the hard governance constraints—per the Williams 2026 annual meeting proxy statement (DEF 14A), the company prohibits directors, executives, and employees from hedging, pledging, short-selling, or trading derivatives on company stock, elects directors annually, and has independent committees review compensation. This is an error-correction mechanism that "relies on process, not heroes": steady and predictable, but it also means that when facing truly paradigm-level bad news, the reaction will lean slow and conservative.

  The management handoff also confirms this style. The report notes that current CEO Chad J. Zamarin took the baton from predecessor Alan Armstrong on July 1, 2025 in a planned succession—a formal, pre-announced handover rather than emergency firefighting, a plus for a long-term capital-intensive company, but likewise reflecting "smooth continuity" rather than "bold reinvention."

  Overall: Williams has the adaptability to extend its assets to new demand, the resilience to weather a major upheaval, and an institutionalized, transparent error-correction culture, all of which make it quite steady under the premise of "not needing disruptive reinvention." But it lacks the aggressive DNA Baillie Gifford prefers—"regenerating a new company even if the core is disrupted"; its safety comes from being hard to disrupt, rather than from being able to be reborn after disruption.

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

  A

  Conclusion: management's long-term vision and system design are pluses, but "interests deeply tied to the company" clearly falls short of the Baillie Gifford standard—this is a mature infrastructure company led by professional managers and bound by compensation terms, without founder-style high-proportion own-capital co-ownership. It is willing to sacrifice current free cash flow for long-term investment (high capex, 52 consecutive years of dividends), but this is a function of the business model, not a personal bet.

  First, on long-term vision: this item is real. The report notes that current CEO Chad J. Zamarin took the baton from predecessor Alan Armstrong on July 1, 2025 in a planned succession, a formal handover pre-announced by the board rather than emergency firefighting—for a capital-intensive company whose projects routinely span years, a smooth handoff is itself an expression of long-termism. The company has also paid cash dividends for 52 consecutive years (see the Williams Q4 and full-year 2025 results 8-K), and such decades-spanning capital discipline shows the operating team runs the business on a "marathon" basis rather than a "quarterly" one.

  But on "deep interest alignment," the evidence points to "not deep." Per the readable figures in the Williams 2026 annual meeting proxy statement (DEF 14A), directors and all executives together beneficially own less than 1% of total shares, and the data cited by the report as of early 2025 is about 5.83 million shares; Zamarin, who had not yet been elevated to CEO at the time, personally held about 519,000 shares, a very small proportion relative to both his compensation and the company's size. Against Baillie Gifford's preferred paradigm of "a founder with a heavy stake, sharing the company's fate with shareholders," this is the "compensation-bound" governance typical of mature American utility/infrastructure companies—the incentives exist, but not the kind of own-capital co-ownership that makes you slap the table. This is Williams's hard weakness on this dimension, and it must be recorded honestly.

  That said, the "quality" of the compensation design is a clear plus and deserves a symmetrical assessment. Per the same DEF 14A, the company's long-term incentives are tied chiefly to ROCE (return on capital employed) and AFFO per share (adjusted funds from operations per share), with a relative-TSR modifier added; annual cash incentives look at adjusted EBITDA, controllable costs, and safety and methane reduction. More importantly, there are hard share-ownership guidelines: the CEO must hold stock worth at least 6x annual salary and other executives 3x annual salary, and until those thresholds are met they must retain 50% of vested awards, net of tax. Putting "return on capital" and "per-share cash-generating ability" into long-term incentives is far more rational than simply pursuing scale and revenue—this lowers the agency risk of "growing for the sake of growing" and partly compensates for the low own-ownership.

  On "willing to sacrifice today's profit for five-to-ten years out"—the answer is yes, but attribute it correctly. Per the above 8-K, the company's 2026 growth-capex guidance is $6.1–6.7 billion (note: the "$7.0–7.6 billion" written in the report body does not match the company's primary guidance; use $6.1–6.7 billion), which will materially depress traditional free cash flow: 2025 operating cash flow was $5.898 billion, but total capex was $4.893 billion, leaving reported free cash flow of only about $1 billion. Management is indeed sacrificing near-term cash freedom to buy 2030 growth. But this is more a structural choice of midstream infrastructure's "capital for growth," layered with compensation-term incentives (ROCE/AFFO), than a founder's all-in bet made on personal conviction.

  Overall: long-term vision (present) + rational compensation design (present) + willingness to invest for the long term (present), but deep personal-interest alignment with the company (weak). This is a "trustworthy professional-manager system," not Baillie Gifford's most-favored "founder-style heavy shared stake." Credible, but the thickness of the alignment is not enough to score high on this dimension—consistent with the report's judgment of "sound system design, but ownership not high."

  Jun 11, 2026
• QIf it disappeared tomorrow, how much would customers miss it? Is its way of growing sustainable, without harming society and regulators?6/10

  A

  Conclusion: if Williams disappeared tomorrow, customers would "miss it enormously"—its indispensability is extremely high, with almost no substitute in the short term; and its way of growing generally does not harm society and sits in the gray zone of the energy transition, with neutral-to-positive regulatory sustainability, though not without tension. This is a very strong item for Williams.

  First, "how much customers would miss it"—this concerns indispensability, the first layer of this question. According to company disclosure, Williams's assets carry roughly one-third of all U.S. natural gas, and the core Transco connects the main supply areas, consumption areas, and LNG and generation load centers. If this network went down, the power plants, industrial users, LDCs/utilities, and LNG exporters that depend on it for gas could not find an equal-capacity substitute corridor in the short term—building a same-class interstate pipeline requires first clearing FERC's Certificate of Public Convenience and Necessity, environmental approvals, and years of construction. The report accurately rates both switching cost and channel advantage as "strong." In other words, this is not "customers like it," but "customers cannot do without it"—its disappearance would directly threaten regional power and gas supply security, so the degree of missing it approaches an infrastructure-level rigid need. This indispensability is the root of stable unit economics (repeatable, contracted fees).

  Next, "whether the way of growing is sustainable and does not depend on harming society and regulators"—this is the second layer, needing a two-sided examination. On the social dimension, natural gas is generally seen in the energy transition as a "transition energy / peaking mainstay," and is emission-reducing relative to coal; the company also includes safety and methane reduction in its annual incentives (see the Williams 2026 annual meeting proxy statement DEF 14A). Its growth relies on building compliant toll corridors and absorbing LNG and data-center power demand, not on harming consumers or exploiting regulatory loopholes. On the regulatory dimension, the FERC certificate regime is both a moat and an expression of "social license"—the company expands within a regulatory framework, which is itself evidence of sustainability rather than its opposite.

  But the tension must be honestly pointed out; we cannot only talk about tailwinds. First, FERC certificates and environmental approvals are double-edged: expansion projects can easily be affected by procedural delays and litigation, and the report already lists regulatory/permitting as one source of permanent capital loss. Second, long-term social license depends on natural gas's positioning in the low-carbon path—the report cites a May 2026 Reuters report that clean-energy generation's share in ERCOT has already surpassed fossil. If policy and social preferences accelerate toward wind, solar, storage, and nuclear, the "long-term social necessity" narrative for natural gas infrastructure would come under pressure, and regulators' tolerance for new pipelines could fall. This is not "banned tomorrow," but "long-term social license needs continuous maintenance."

  Putting the two layers together: indispensability (extremely strong, no short-term substitute) + a way of growing that does not harm society and expands compliantly within a regulatory framework (neutral-to-positive sustainability), layered with the tension that "long-term social license must be continually earned as the energy transition proceeds." The customers' "degree of missing it" is undoubtedly very high; the social and regulatory sustainability of the growth generally holds, only not as unconditionally as a consumer rigid need—consistent with the report's judgment of "tailwind demand logic, but pressure from the evolving energy mix."

  Jun 11, 2026
• QWhat are this business's unit economics (gross margin, incremental returns)? Do they get better or worse as it scales? Where does the money earned go?5/10

  A

  Conclusion: the unit economics are "high-quality but heavy-asset"—high cash margins, contracted, repeatable; as it scales, the unit economics generally hold or even improve slightly (higher net rates, expansion leverage), but this is by no means a "lighter-as-it-grows" consumer franchise; the money earned goes mainly to growth projects and dividends, with restrained buybacks. The incremental return rate (how much EBITDA each dollar of capex buys) is the lifeblood of this business.

  First, look at the foundation of the unit economics, using a first-hand cash basis rather than GAAP. According to the Williams Q4 and full-year 2025 results 8-K, in 2025 the company delivered adjusted EBITDA of $7.750 billion, operating cash flow of $5.898 billion, and available funds from operations (AFFO) of $5.858 billion, with AFFO-basis dividend coverage of 2.40x. This set of figures shows it is a true cash-flow business: repeatable fees, high margins, solid cash conversion. Even rarer, "per-share" improves in step—the same filing discloses a five-year adjusted-EBITDA CAGR of 9%, adjusted-EPS CAGR of 14%, and AFFO/share CAGR of 10%, on which the report correctly judges it is "not fake growth where revenue appears to rise while per-share cash does not."

  Do the unit economics get better or worse as it scales? The answer is "holds and slightly improves, but with a heavy-asset ceiling." The better side: large pipeline networks have marginal expansion advantages, expanding an existing corridor is cheaper than building new, and in 2025 the Transmission, Power & Gulf segment lifted EBITDA to $3.720 billion on higher Transco net rates—showing scale brings pricing and utilization leverage. But the worse/limited side must also be stated: this is heavy assets, and the larger the asset base, the higher the absolute maintenance capex and maintenance costs (2026 maintenance-capex guidance of $850–950 million); it does not have software's "marginal cost trending to zero" property. So the unit economics are "steady with a slight scale dividend," not exponential improvement.

  The incremental return rate is what truly needs watching, and that is precisely the uncertain item. The report honestly marks "whether return on capital is excellent" as uncertain, and lists "ROIC/ROCE" as one of the metrics to track most. The reason is that the company is entering a round of high-intensity growth capex: per the above 8-K, 2026 growth-capex guidance is $6.1–6.7 billion (note: the "$7.0–7.6 billion" written in the report body does not match the company's primary guidance; use $6.1–6.7 billion), and the analyst day also disclosed a roughly $23 billion project backlog at about a 5x build multiple (see the Williams analyst-day press release). A roughly 5x build multiple means, in theory, every $5 invested buys $1 of EBITDA (about a 20% incremental EBITDA return), which sounds good—but this is a planning figure, and the real incremental return must be verified after projects come online using ROCE and AFFO/share. If incremental returns hit target, the unit economics stay excellent; if not, it merely enlarges the asset base and dilutes per-share value.

  Where does the money earned go? The path is clear and fairly rational. Per the same 8-K and historical disclosures: the first destination is growth projects (the above $6.1–6.7 billion); the second is dividends—2025 common-stock dividends paid were about $2.442 billion, the 2026 annualized rate was raised to $2.10 per share, with 52 consecutive years of dividends; buybacks are extremely restrained, with essentially no significant buybacks in 2025 or 2024 and only about $130 million in 2023. In other words, management has not used buybacks to "dress up per-share metrics," but has invested capital into projects and dividends—for an infrastructure company, this is usually preferable to chasing EPS financial engineering, and the report holds the same view.

  Overall: high-quality unit economics (high cash margins, contracted, per-share growth in step), holding or slightly improving with scale but constrained by heavy assets, and rational capital destinations (projects + dividends, restrained buybacks). The only real suspense is whether this round of large capex can deliver the promised roughly 5x build multiple's incremental return—the key that separates a "good business" from a "good investment," consistent with the report's tone of "strong cash flow, but return on capital yet to be verified."

  Jun 11, 2026
• QFor it to rise five-fold in ten years, what conditions must hold simultaneously? Are they realistic? What does today's share price imply?2/10

  A

  Conclusion: for Williams to rise five-fold in ten years, a chain of low-probability conditions must hold at once, and the realism is on the low side; what today's roughly $72 share price implies is not a "five-fold expectation"—on the contrary, it implies the market has already priced in the quality premium of "high-quality, scarce, growth-certain." In other words, the share price currently pre-pays for "steady delivery," not "explosion"; expecting five-fold amounts to demanding a toll-infrastructure company run a tech-stock curve.

  First, anchor today's facts. According to stockanalysis's WMB quote page, as of the June 10, 2026 close WMB was about $72.26, market cap about $88.4 billion, trailing P/E about 31.7x, dividend yield about 2.9% (annualized dividend $2.10). (Note: the report body's "share price about $78.47, market cap about $96.2 billion, P/E 34.4x, dividend yield 2.7%" corresponds to an earlier point in time; for the current period, use about $72, about $88.4 billion, about 31.7x, about 2.9%.) A five-fold gain in ten years means the share price reaching about $360 and market cap the order of about $440 billion, corresponding to about 17.5%/year compound appreciation.

  What conditions must hold "simultaneously" for a five-fold gain in ten years—stress-test them one by one:

  First, earnings must compound at about 17%/year for ten years. But the framework the company itself gave at the analyst day is roughly 10% compound adjusted-EBITDA growth and 5%–7% dividend growth through 2030 (see the Williams analyst-day press release), of which the part supported by FID'd/contracted projects is about 8%–9%. 10% is already management's fairly aggressive framework, a wide gap from the ~17% needed for five-fold, and the physical and regulatory constraints of infrastructure make it hard to systematically exceed.

  Second, the valuation cannot compress, and ideally should expand. But today's roughly 31.7x P/E and roughly 15x–16x EV/EBITDA are already among the most expensive of its peers: per stockanalysis's KMI data, Kinder Morgan's EV/EBITDA is about 12–14x and market cap about $74.7 billion; Enterprise Products (EPD) has a market cap of about $80–82 billion and a dividend yield of about 5.8%. For a five-fold gain, the market must not only avoid converging WMB's premium back to the peer average, but also grant an even higher multiple—the opposite direction to "mean reversion's" gravity.

  Third, the incremental returns of a round of large capex must not miss at any point. Per the Williams Q4 and full-year 2025 results 8-K, 2026 growth-capex guidance is $6.1–6.7 billion (note: the "$7.0–7.6 billion" written in the report body does not match the primary guidance; use $6.1–6.7 billion), on top of a roughly $23 billion project backlog at about a 5x build multiple. A five-fold gain in ten years requires these projects to compound and reinvest with almost zero delay and zero return shortfall—an extremely high execution assumption.

  Fourth, the demand premise must keep beating expectations. The EIA and IEA do point to rising electricity and data-center demand, but the report also cites a May 2026 Reuters report that clean-energy generation's share in ERCOT has already surpassed fossil—gas's long-term marginal position risks being eroded by wind, solar, storage, and nuclear. A five-fold gain needs this demand curve not only to hold but to exceed management's expectations.

  Multiply the four together: about 17% compound growth × (a high valuation still not compressing) × (zero misses on large-capex returns) × (demand continually beating expectations), and the joint probability of all holding at once is very low. So a five-fold gain in ten years is unrealistic for Williams; its reasonable profile is "steady compounding + dividends," and the report's neutral annualized return of 7%–9% and optimistic 10%–12% ranges cross-validate with the company's fundamentals and primary guidance, and are far below the slope needed for five-fold.

  So what does today's share price imply? It implies that "the market already believes it can steadily deliver the roughly 10% growth framework and is willing to pay a quality premium for that certainty." The roughly 31.7x P/E, the roughly 2.9% dividend yield below peers, and the EV/EBITDA above KMI/EPD all show that the good news has been substantially pre-paid. This is precisely the report's core judgment that "the current price looks more like it is pre-paying for the good news of the next few years, and the margin of safety is insufficient"—buying here earns the money of "delivery," with almost no room for "re-rating upside," let alone five-fold.

  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 would become the "narrative inflection point"?3/10

  A

  Conclusion: for Williams this question must be asked in reverse—the market has not "yet to realize" how good it is, but has "already fully realized it, even slightly premium-priced" it. It is neither misunderstood nor looked down upon; if anything is not fully priced, it is two "can't-see-far" tails: one, the extent to which LNG/data-center power lengthens gas demand duration; two, the compounding thickness of AFFO/share after the large capex's incremental returns are delivered. The real "narrative inflection point" is therefore two-way, and the downside probability is no smaller than the upside.

  First, prove that "the market has already realized it," which is where it differs most from most Baillie Gifford targets. According to stockanalysis's WMB quote page, as of June 10, 2026 WMB was about $72.26, market cap about $88.4 billion, trailing P/E about 31.7x, dividend yield about 2.9%. Comparing horizontally with peers: per stockanalysis's KMI data, Kinder Morgan's EV/EBITDA is about 12–14x and market cap about $74.7 billion; Enterprise Products (EPD) has a market cap of about $80–82 billion and a dividend yield of about 5.8%. WMB's EV/EBITDA (about 15x–16x) is clearly higher than both, yet its dividend yield is lower—there is only one explanation: the market not only understands Transco's scarcity and respects its asset quality, but is also willing to pay a quality premium for "purer natural gas exposure + better project visibility + a long record of per-share growth." "Can't understand it / looks down on it" does not hold here.

  So what remains not fully priced in the "can't-see-far" area? Two longer-dated, hard-to-model things. First, demand duration: the IEA data-center electricity report expects data-center electricity use to nearly double by 2030, and the EIA expects U.S. electricity use to set successive records—if gas holds "mid-load mainstay" rather than merely peaking in this round of power expansion, Williams's growth duration would be longer than currently priced. Second, per-share compounding thickness: per the Williams Q4 and full-year 2025 results 8-K, if the roughly $23 billion project backlog at about a 5x build multiple is delivered at its return rate throughout and converted into AFFO/share, the market may be underestimating the compounding effect of "high-quality reinvestment." These two are potential upside cognitive gaps—but both are prospects that "need time to prove," not clear cards undervalued today.

  The symmetrical risk side must be stated clearly: the downside "can't-see-far" exists too. The report cites a May 2026 Reuters report that clean-energy generation's share in ERCOT has already surpassed fossil—if the market someday re-rates "gas only handles peaking long term," today's roughly 15x EV/EBITDA premium would compress faster than imagined. Meanwhile, per the above 8-K, 2026 growth-capex guidance is $6.1–6.7 billion (note: the "$7.0–7.6 billion" written in the report body does not match the primary guidance; use $6.1–6.7 billion), and during the high-capex period any project delay or return shortfall would shake the "growth certainty" premise the market has priced in.

  What would become the "narrative inflection point"—laid out both ways:

  • Upside inflection: long-term contracts from data centers/LNG significantly beat expectations, re-rating gas demand duration from "transition energy" to "structural long-term mainstay"; or the large-capex projects come online early with ROCE/AFFO/share clearly beating guidance, proving the roughly 5x build multiple's reinvestment can keep compounding.
  • Downside inflection: AFFO/share growth consecutively below dividend growth (financial policy starting to decouple from fundamentals); net debt/EBITDA breaking 4.5x long-term with no downward path (the company's 2026 leverage midpoint guidance is about 4.0x); or natural gas's marginal position in the power system significantly weakening, triggering mean reversion of the quality premium.

  Overall: Williams's peculiarity is that it is not "a cheap good company overlooked by the market," but "a good company fully recognized by the market and already paid a premium for." The narrative inflection point is two-way, and at the current price the realism of a downside re-rating (premium compression) is no lower than an upside re-rating. This is fully consistent with the report's judgment that "the quality is already partly reflected in the price, the margin of safety is insufficient, and the most fragile point is whether the market will keep granting the high-quality premium"—for a long-term owner, the conclusion is not "bad," but "good, but no longer cheap."

  Jun 11, 2026