Southern Company (SO)

Total return at 5 years: 2.6%/year = -1.3% appreciation + 3.9% dividend. The target price ($82) is ex-dividend; the $17 in dividends collected over 5 years are added separately.

The methods disagree: one places the value today above the price ($87) and the other below.

Today — expensive, no margin of safety: at $87 trades ~8.7% above its value discounted to today (~$80); the expected return does not even reach the risk-free rate (4.5%).

At 5 years — Preserva valor: the target price ($82) plus dividends yield just enough to preserve nominal capital, below the required 4% floor.

The bridge: the return at 5 years falls below the risk-free rate (4.5%) — which is why there is not even a discount to today's value. To require a 15% annual return, it would need to be bought at ~$52.

Educational / informational. Does not constitute investment advice.

The inverse approach: instead of projecting growth to obtain the value, the market price ($87) is taken as given and it solves for what annual owner-earnings growth would need to hold for 5 years for the present value —at the method's rate (4.5%, no-growth terminal)— to equal that price. It is the disconfirmation test: the expectations the price already pays for, contrasted against the method's projection.

The market discounts less growth (4.3%/year) than we project (6.5%/year) → if the base case holds, there is margin: perception is more pessimistic than the estimated reality.

That growth implies ~$7.9 bn of owner earnings in year 5 (vs ~$8.7 bn of our base case). It recalculates if the DCF assumptions are edited.

A regulated utility is valued on P/E over adjusted earnings (equity-level): the value sits in the regulated base (rate base), which grows with capex at an allowed ROE. GAAP can be depressed by the mark-to-market of hedges → the adjusted figure is used. It returns capital via dividend (a dividend aristocrat); it may issue some equity to fund capex. Total return adds the dividend collected along the way. In edit mode, the metric, shares, dividend, and exit multiple can be adjusted.

A utility is valued on P/E over net income attributable to common (Southern Company does not declare a separate 'adjusted' measure in this extraction, unlike a hybrid utility with an unregulated renewables arm). The value is anchored in the regulated rate base of the three electric subsidiaries (Alabama, Georgia and Mississippi Power) plus Southern Company Gas: Georgia Power, the largest subsidiary, has a retail ROE point of 10.50% (approved band 9.50%-11.90%, equity ratio 56%) and earned within that band in 2023-2025 per the 10-K. Year-0 is the rolling year ended Jun 30, 2026: net income attributable to common $4,657M (EPS ~$4.08), built from levels on fiscal year 2025 ($4,341M) plus the first half of 2026 ($2,531M) less the prior-year half ($2,214M), with the quarter read from the 10-Q's XBRL instance because the SEC API only reaches the first quarter. It is reported earnings, not the measure that excludes items: the company's own reconciliation runs the other way —$2,775M versus $2,531M in the half— meaning what year-0 carries are charges, not income that does not repeat. Against that year-0 the path starts at +7.0%, below the +8.7% year-over-year realized: of the 31 cents of per-share improvement in the half, 9 come from an effective tax rate that fell to 14.1% from 21.1% and 9 from the allowance for equity funds used during construction, and neither repeats at that magnitude as a growth driver. From there it decelerates to +5.0% by year 5. The revenue in the table is the subtotal of revenue from contracts with customers ($28,942M in FY2025), which is not to be confused with the consolidated total operating revenue line ($29,553M). Shares grow ~2.0%/year (actual equity issuance funding the construction program, which nearly doubled in two years on data center demand: $8,955M FY24 → $14,139M TTM). The rate is kept even though the count rose 4.6% between June 2025 and June 2026: the jump is concentrated in the second quarter of 2026 —a placement under the equity distribution agreement that added some 23 million shares and US$2,100 million to paid-in capital— and a continuous placement leaves no usable revealed rate; the two- and three-year windows give 2.5% and 1.8%, which bracket the model's 2.0%. Year-0's dividend was not re-anchored: the $3.04 is the current annualized rate and the second-quarter 2026 payment was already $0.76. The construction-program series is cash capex from the statement of cash flows, not the 10-K's US$78,100 million five-year plan, which includes long-term service agreement and nuclear fuel commitments. The multiple compresses toward 16.5x (within the band of a regulated utility, 15-18x, §4). Debt ($75,582M) and TTM interest ($3,223M) were verified directly against the SEC EDGAR XBRL because the extraction's factsPack did not carry them.

Today's multiple compresses on its own going forward as the metric per share grows. The exit multiple at 3 years is higher than the terminal at 5 years —at 3 years there is more growth still ahead—. Total return adds the dividend collected; the required return is applied to the base scenario.

They are valued separately, with their own rationale, and are not incorporated into the base or the verdict (they are excess return). When assigning them value — in Editmode —, the total with optionalities updates live, without moving the base.

The verdict, the base CAGR, and the margin of safety are always calculated on the base; optionalities do not alter them (with optionalities at $0 they do not move).

Each card fixes a required annual return and answers: if the business is worth $82 in 5 years, what is the maximum that can be paid today to obtain that return? Since it now trades at $87, the margin of safety is how much cheaper the market is than that maximum. The three thresholds: 4% covers inflation (the floor), 10% is the long-term average return, and 15% is the level of a great investment.

With a target price of $82 in 5 years plus $17 of dividends collected (the dividend adds to the return, not to the price) and a required return of 4.5% annually, the maximum to pay today is $80. Against the current market price ($87), the margin of safety is -8.7% (trades above the maximum → a premium is paid) and the total return at that price would be 2.6% annually.

• Multiple vs. the peer group. 21× TTM net income attributable — above the band for a regulated utility (15-18×).
• Growth vs. multiple. The multiple compresses sharply 21×→16.5× as the path normalizes after the FY2025 noise and converges to the band of a regulated utility.
• Margin of safety. Base case -1%/year of price appreciation (+3% with dividend) falls short of the average return (~10%); the discount today is small or nil at the group's highest multiple.

Each margin over sales, year by year: historical (solid line) → projection (dotted).

• ✕ ROIC exceeds the cost of capital (~10%)
• ✕ High incremental ROIC (compounding)
• ✓ CFROIC backs up the ROIC (100%, cash vs. accruals)
• ✓ Healthy balance sheet (low corporate debt)
• ✓ Durable competitive moat (multiple advantages)

• Regulated return (earned vs. authorized ROE). Georgia Power earned within its authorized band (9.50%-11.90%, 10.50% fixed point) in 2023-2025 per the 10-K — solid execution within the regulatory framework.
• Value creation (consolidated ROE vs. the 10% bar). ~11.8% GAAP (~13.6% on tangible equity) vs. the 10% bar — the method's 'good' category (10-15%), close to the 'excellent' threshold (≥15% on tangible equity).
• Earnings quality. The EPS decline from FY2024 to FY2025 (−1.8%) is explained by ~$223M of net non-recurring charges declared with exact amounts (Vogtle, Southern Power, Nicor Gas) — transparent, not a deterioration of the regulated business.

Annual levels from the official filing (10-K); the % over each bar is the year-over-year (YoY) growth vs the prior year.

Southern Company's real engine is the regulated rate base of its three electric utilities and Southern Company Gas × the ROE authorized by jurisdiction (Georgia Power: fixed point 10.50%, band 9.50%-11.90%, with general base rates frozen 2026-2028 under the ARP extension), financed by a construction program that nearly doubled in two years (from $8,955M in FY2024 to $14,139M TTM) and whose five-year estimate (2026-2030, US$78,100 million) rises ~73% versus the prior three-year window, to serve data-center and artificial-intelligence demand (~9 GW of large-load contracts signed since 2023, ~13 GW of resources certified by the Georgia PSC, of which ~9 GW at Georgia Power carry a certified cost of US$19,500 million). It is financed with debt + equity issuance (real diluted-share dilution ~2%/year in the recent trajectory, with US$2,000 million of Equity Units issued in November 2025 whose conversion to shares is already contractually set for December 2028, consistent with dividends + capex + debt maturities exceeding operating cash flow over the 2026-2028 three-year window); the small unregulated segment (Southern Power, ~7.5% of revenue) operates under long-term PPA contracts with high-credit-quality counterparties, not under cost-of-service tariff.

The % are the annual (year-over-year) growth: each year —historical and projected— vs the prior one; the TTM (trailing 12m) vs the TTM of a year ago, to avoid overlapping windows. The historicals are exact figures from the official filings; the projected years come from the year-by-year model (the intermediate years without their own series are interpolated between the anchors). The projected columns (+1y…+5y) are 12-month windows counted from the TTM close (30-jun-2026): the projection starts from the most recently reported data, not the fiscal year. The projected base is realistic and unbiased — the risk discount is applied at the end, via the required return. The rationale for each metric is in the (i).

• Real, not nominal. The growth lives in the physical rate base (plants, transmission, pipelines built), not in monetary revenue or an easy comparison.
• Growth engine. Systemwide construction program nearly doubled in two years + a five-year estimate (2026-2030, US$78,100M) ~73% higher than the prior three-year window, driven by ~13 GW of resources certified by the Georgia PSC, of which ~9 GW at Georgia Power carry a certified cost of US$19,500M, and data-center demand.
• Sustainability. The certified growth is well founded through 2030. The risk is execution (the Vogtle precedent) and whether projected data-center demand holds up beyond what is already contracted.

• Relevant in 10 years (recurrence). Electricity and natural gas are about as recurring a demand as exists, and data-center/AI demand structurally accelerates it. Maximum predictability of regulated service.
• Pricing power. Set by the regulator (authorized ROE on the rate base, e.g. Georgia Power 10.50%) — not market pricing power, but a guaranteed return within the regulatory framework.
• Moat and its direction. Wide moat (regulatory in four utilities + contractual in Southern Power), stable with point widening from data center interconnection position — though Georgia Power's next general rate case (2029-2031) already foresees downward pressure of at least US$556M/year on that incremental revenue.
• Cyclicality / sensitivity. Not very cyclical (stable regulated demand) but sensitive to rates (high leverage) and to execution/recovery risk on a construction program that has nearly doubled in two years.

Not all debt is equal: only the structural needs refinancing; the rest is operational (self-liquidating).

Structural debt is what is exposed to the contraction phase of the cycle; operational debt (leases, matched funding) self-liquidates with the business.

• ✓Regulated return (earned vs allowed ROE)Earns 11.8% vs allowed 10.5%
• !Value creation (allowed ROE − 10% bar)+0.5pp
• ✓Leverage (net debt / EBITDA)5.3x
• ✓Predictability (% regulated + rate-base growth)93% regulated

A traffic-light interpreted by the method (not generic): float (negative WC) adds up, capex is judged by incremental ROIC (malinvestment test), and a lender is not subjected to corporate solvency. The (i) shows the derivation of each number.

• Leverage (regulated lens). High by design ($75,582M of verified debt, ~5.32× net debt/EBITDA), backed by a regulated, investment-grade rate base and a staggered maturity schedule (2026-2030: US$6,211M · US$3,407M · US$5,464M · US$1,993M · US$4,103M), with the peak in 2026 and no disproportionate spike — judged against utility thresholds, not corporate solvency.
• Credit quality. The 10-K doesn't give Southern Company's current rating letter, but it does give a directional signal: Moody's revised Southern Company's outlook from stable to negative on 23-Sep-2025, and Fitch revised Georgia Power's outlook from positive to stable on 22-Aug-2025 — a deterioration in credit perception, not a default.
• Negative cash flow by design. FCF is negative (~−$3,488M TTM) because the capex builds out rate base (which then earns the authorized ROE) — this is regulated investment, not cash stress.
• Funding of the construction program. Funded with debt + equity issuance (dilution ~2%/year, with US$2,000 million of Equity Units issued in November 2025 whose conversion to shares is already set for December 2028); the 10-K itself anticipates that dividends + capex + debt maturities will exceed operating cash flow over the 2026-2028 three-year window — it depends on continued, well-priced access to capital markets for a program that nearly doubled in two years.

Who owns the shares — the alignment and whether there is a controlling shareholder.

• Skin in the game. No founder control; the largest holders per the DEF 14A proxy are passive institutions (BlackRock 6.9%, State Street 5.7%). Director/officer ownership was not anchored in this extraction.
• Capital allocation. Deploys capital almost entirely to the regulated rate base (TTM capex $14,139M) + the dividend (75% payout on FY2025 declared diluted EPS, $2.94/$3.92 — at the usual high end for a regulated utility, 65-75%). No buybacks since 2017; no material M&A — organic growth via the certified program, funded in part with Equity Units whose conversion to shares is already contractually set for December 2028.
• Integrity / candor. The 10-K discloses the FY2025 non-recurring charges (Vogtle, Southern Power, Nicor Gas) and their after-tax effects in detail with exact amounts, without burying them in the reported result under accounting rules.

• It is not cheap: it trades at 21× TTM earnings, above the band for a regulated utility (15-18×).
• The re-rating has an identifiable cause, it is not a pricing error: Georgia Power's direct exposure to data-center demand (~9 GW of large-load contracts since 2023, ~13 GW of resources certified by the Georgia PSC, of which ~9 GW of Georgia Power's own carry a certified cost of US$19,500 million) is real and verifiable in the 10-K, not a rumor.
• It's a case of an expensive business with a genuine reason (§9): the market is paying for rate-base growth that is indeed certified and underway, but at 21× it is already paying for a large share of that growth up front — with no margin if execution stumbles.

No identifiable missing buyers or motivated sellers: the price reflects genuine and well-founded enthusiasm for data-center demand, not an evident gap between perception and reality. The base return (+3% total) is modest precisely because the entry multiple is already high — the opportunity, if any, lies in a correction of the multiple toward the band of a regulated utility (15-18×), not in the business today.

The annual return (CAGR at 5 years) in each scenario, with the total period return below — the margin of safety made visual: upside range wide, downside range narrow.

Even in the bear scenario, the return holds at -3%/year (-12% total): the margin of safety protects the downside. The bull (+6%/year, +32% total) exceeds it comfortably — a favorable asymmetry, with a narrow downside range and a wide upside range.

• Regulatory recovery of data-center capex. The central disconfirmer: Georgia Power already committed, as a condition of the December 2025 Certification Stipulation, to a downward pressure of at least US$556 million/year (levelized) on incremental large-load revenue in its next general rate case (2029-2031); the 10-K also acknowledges the risk of cost overruns, delays, and of the projected demand not materializing — with its own precedent at Plant Vogtle.
• Elevated entry multiple. At 21× TTM earnings, any growth disappointment hits the multiple first, not just earnings — little room for error.
• Regulatory prudence disallowances. It has already happened twice at Nicor Gas: US$58 million in 2023 and US$63 million in 2025 — a precedent that not all capex is automatically recovered in rates; Southern Company Gas's goodwill (~US$5,000 million, practically all of the group's consolidated goodwill) remains exposed if the regulatory proceedings on the future of natural gas underway in some jurisdictions advance.
• Nuclear risk. Alabama Power and Georgia Power operate/have an interest in nuclear plants (Farley, Hatch, Vogtle), subject to environmental, safety, and regulatory (NRC) risks, including the funding of decommissioning trusts whose return depends on market volatility.
• Customer concentration and credit perception. Southern Power's three largest customers concentrate ~25% of its 2025 revenue, and the group's credit perception started to deteriorate: Moody's downgraded Southern Company outlook from stable to negative in September 2025, Fitch downgraded Georgia Power from positive to stable in August 2025 — early signals that nuance the reading that data center demand is upside without risk counterpart.

Each thinker's analytical framework applied to our data.

• Buffett / Graham Quality + margin of safety

  A wide moat and ROE 12% above the 10% bar, but the price sits above value (no margin) → great business, expensive.

• Peter Lynch Growth at a reasonable price (GARP)

  A stalwart growing 7% at a PEG of 3.3 → expensive for its growth.

• Joel Greenblatt Cheap and high-return (Magic Formula)

  Not applicable — the Magic Formula excludes financials and regulated businesses (EBIT/EV does not capture the operating leverage).

• Howard Marks Perception vs reality + cycle

  The price discounts 4% vs our 7%: perception is more pessimistic than reality, with an identified source of the discount.

• Seth Klarman Capital protection (bear scenario)

  Bear-scenario floor -7%/yr, bull-scenario ceiling +2%/yr over 5y: reasonable asymmetry, without an ample cushion.

• Pat Dorsey Moat strength (Five Rules)

  A wide moat, stable; sources: intangibles, efficient scale, cost advantage, switching costs → passes the Five Rules.

• Aswath Damodaran Expectations implied by the price

  Justifying the price requires discounting 4%, within what we project (7%) — the story squares with the numbers.