Nisource Inc NI

Thesis

NiSource (NI) is a pure-play regulated utility: ~70% of operating earnings come from natural-gas distribution across Indiana, Ohio, Pennsylvania, Virginia, Kentucky, and Maryland; ~30% from NIPSCO, the integrated electric utility serving Northern Indiana. The compounding engine is the regulatory compact — the company invests capital into approved rate base, regulators set rates that allow recovery of operating costs plus a return on equity (typically 9.5–10%), and earnings grow with rate base. NiSource is guiding to roughly $19.4B of capex through 2029 and 6–8% rate-base CAGR, juiced by NIPSCO's coal retirements (Schahfer retired, Michigan City in flight), renewable replacement generation, gas-system modernization, and — newly — large-load data-center customers in Indiana driving incremental generation needs.

The ground-truth scorecard is sober: composite 54, 10-year average ROIC of 9.35%, ROIIC of 22.19% (decent, though lumpy and likely flattered by recent rate cases), share count up 3.89% over 10 years (utility equity issuance is part of the model), and FCF conversion of -108% — i.e., the business consumes more cash than it generates because growth capex dwarfs depreciation. PE TTM is 29.68 versus a 10-year average of 29.87, and reverse-DCF implies 5.43% growth — exactly in the band the company guides to.

The IV math is unforgiving. iv_low $26.63, iv_base $36.32, iv_high $53.79; market price $48.08 implies a px/iv of 1.3238. You are paying 32% above the central case and only 11% below even the bull case. Buy below $36; trim above $54.

Moat

Regulated utilities derive their moat from a legal monopoly granted by state public utility commissions (PUCs), which Damodaran characterizes as 'a mixed blessing' — the franchise is secure, but the regulator simultaneously caps the return [4]. Below I walk the five moat types as they apply to NI.

1. Cost advantages. NIPSCO is a vertically integrated electric utility serving roughly 500,000 meters in Northern Indiana; the gas LDC serves roughly 3.4 million customers across six states. The cost moat is structural: it is uneconomic to run a parallel set of gas mains or transmission and distribution wires down the same street. Total system replacement cost for NI's network would run into tens of billions, dwarfing the $19B 5-year plan. A $10B + 5-year competitor entry, the brief's stress test, is essentially impossible — even Berkshire Hathaway Energy, the deepest-pocketed strategic in the space, expands by acquisition of incumbent franchises, not greenfield duplication. Buffett describes this dynamic in 2013 [6]: MidAmerican's value comes from 'huge investment in very long-lived, regulated assets' and 'recession-resistant earnings, which result from these companies exclusively offering an essential service.' That paragraph could be lifted verbatim onto NI.

2. Pricing power. Pricing power is real but bounded. NiSource cannot raise rates unilaterally; rate cases must be filed at each state PUC and litigated against consumer advocates. What it can do is recover prudently incurred capex (rate base) at an authorized ROE typically 9.5–10%, plus pass through fuel and gas costs. Trackers and riders (e.g., Indiana's TDSIC for gas, federally mandated CapEx) shorten regulatory lag. This is the engine, but Damodaran's caveat is decisive [4]: 'that entity usually preserves the right to control the prices charged and margins earned through regulation.' You earn the cost of capital plus a sliver — not Coca-Cola economics.

3. Switching costs. Effectively infinite for monopoly service. A residential gas customer in South Bend cannot 'switch' to a competing pipe in the street; they can only switch fuel (electrify with a heat pump) or move. Heat-pump electrification is the long-term threat to gas LDC volumes (see Inversion). Switching costs against the franchise: zero. Switching costs against the molecule: meaningful and rising.

4. Network effects. Weak. The grid has scale benefits but not a true Metcalfe-style network effect — adding a customer doesn't raise the value of the service for other customers. What scale does buy is regulatory and financing efficiency: NI's investment-grade credit (BBB+ at parent, A- at NIPSCO) lowers WACC, which compounds inside rate base.

5. Intangibles. Brand: irrelevant — Buffett's 2004 line that 'we wait in vain for I'd like a National Indemnity policy' [2] applies double here; no customer specifies their gas LDC. The genuine intangible is the regulatory franchise itself plus institutional relationships with the IURC, PUCO, PA PUC, etc. These are not transferable and are slow to erode.

Competitor stress test. Spend $10B over 5 years to take share from NI? You cannot. You can only buy NI itself. The relevant 'competitor' is a state legislature reshaping the regulatory compact (e.g., performance-based rates, decoupling rewrites) or federal/state policy accelerating gas demand destruction. Both are slow-moving and partially handicappable.

Erosion risk. Two genuine threats. (a) Gas demand in residential/commercial decarbonizes via heat pumps; cities like New York and several Massachusetts municipalities have begun banning new gas hookups. NI's footprint is in slower-decarbonizing red/purple states, which delays but does not negate the trend. (b) Regulatory ROE compression in a low-real-rate-with-political-pressure regime; authorized ROEs have drifted from ~10.5% a decade ago toward ~9.5%.

Moat verdict: NARROW. The franchise is durable but structurally rate-limited; the volume base is in slow secular decline on the gas side. This is a toll road where the toll is set by a committee that is allowed by law to take most of the surplus.

Management & Capital Allocation

NiSource is run by CEO Lloyd Yates (since early 2023, formerly of Duke Energy) with CFO Shawn Anderson. The five capital-allocation choices for a regulated utility are unusual: most of the discretion is removed by the regulatory model, which essentially demands one strategy — invest into approved rate base, finance the gap with a balanced mix of debt, retained earnings, and equity, pay a meaningful dividend.

1. Reinvest in the business. The dominant choice. NiSource's $19.4B 2024–2029 capital plan implies investment of roughly 2.5x annual operating cash flow, financed externally. The signature projects are the NIPSCO generation transition (retiring coal at Schahfer/Michigan City, replacing with wind, solar + storage, and selectively gas peaking under the Genco JV structure with controllable affiliates), gas system modernization (a multi-decade replacement of cast-iron and bare-steel mains under TDSIC and similar riders), and load-growth capex tied to large data-center customers in Indiana (Microsoft, Amazon, Google announcements landing 2024–2025). ROIIC of 22.19% is encouraging but I discount it: utility incremental returns track rate-base growth × authorized ROE, ~9.5%; the 22% figure is a 5-year window flattered by rate-case true-ups, deferred-tax and DAC dynamics, and deconsolidation of the NIPSCO minority sale to BIP. Real durable incremental ROIC is closer to 9–10%.

2. Acquisitions. NI is largely done with M&A. The Columbia Pipeline Group spinoff (2015) cleaned up the portfolio. The 2023 sale of a 19.9% minority interest in NIPSCO to Blackstone Infrastructure Partners for $2.15B was a clever financing transaction — non-dilutive to NI shareholders at a premium-to-book valuation, and BIP added a follow-on tranche in 2024. Mild positive for management grade.

3. Debt. Net-debt/EBITDA reads -0.04 in the scorer's metrics, which I treat as a data error — actual leverage is materially higher. Utilities run at 5–6x net debt/EBITDA; NI is in that range. Interest coverage is null in the scorecard but is in the 3–4x range based on disclosed interest expense. Investment-grade and well-laddered.

4. Buybacks. None of consequence. Share count has risen 3.89% over 10 years, consistent with utility equity issuance to fund growth capex while preserving the credit rating. Per Buffett, you want managers buying when stock < intrinsic. NI sells equity instead — appropriate when stock trades at 1.32x base IV (today), inappropriate at 0.7x. Management has shown discipline by using the BIP minority sale and ATM programs rather than block deals at depressed prices.

5. Dividends. Current dividend ~$1.10/share, ~2.3% yield, 60–65% payout target. Steady annual increases. Aligns with utility-sector norm.

Communication quality. Investor day disclosures are detailed: rate-base bridges, capex by segment, regulatory case calendar, financing plan. Management does not over-promise; the 6–8% EPS CAGR target has been roughly delivered. Disclosure on data-center load growth (a 2024–2025 development) is appropriately cautious — they are signing tariff structures intended to protect existing customers from the cost of incremental generation needed for hyperscalers, but the load forecasts could prove either too high or too low and the regulatory treatment is novel.

Negatives. The 2018 Merrimack Valley gas explosions (Columbia Gas of Massachusetts subsidiary) were a catastrophic safety failure that ultimately required divesting the Massachusetts business and a $143M federal criminal fine. Current management was not in seat for that, but it remains a permanent reminder that this business model has fat-tail safety risk.

Capital allocator: B. Limited discretion, executed competently. The BIP minority transaction was clever; the equity-funded growth pipeline is appropriate at scale; safety culture must continue to improve. There is no Buffett-style asymmetry to harvest.

Industry Structure

Regulated electric and gas utilities are a cleaner Five-Forces analysis than almost any other industry, because the regulator dampens or eliminates most competitive dynamics by design.

1. Threat of new entrants — VERY LOW. A new entrant cannot obtain a Certificate of Public Convenience and Necessity in NI's service territories without effectively replacing the incumbent — a political and economic non-starter. Greenfield duplication of pipes/wires is uneconomic. Damodaran [4] flags the offset: the regulator extracts most of the rents in exchange. So while the moat is high, the prize behind it is small.

2. Bargaining power of buyers — LOW for residential, RISING for large industrial / data center. Residential and small-commercial customers have no choice. Large industrial loads can self-generate or relocate, and hyperscaler data centers in Indiana are negotiating bespoke tariffs with substantial influence over rate design (rate riders, capacity charges, locational marginal pricing pass-throughs). The data-center cohort is, in effect, a new powerful buyer class with non-trivial leverage with the IURC. Net: average buyer power low; marginal buyer power rising.

3. Bargaining power of suppliers — MEDIUM. Key inputs: natural gas (passed through), coal (declining), wind/solar/battery equipment (capital cost), turbine OEMs (GE, Siemens — supply-constrained right now and pricing it), labor (union, especially line workers — tightening), and capital itself (debt and equity markets). Pass-through mechanics insulate fuel costs but capital cost increases are recovered with regulatory lag.

4. Threat of substitutes — MODERATE AND RISING. Two distinct substitution stories. (a) Electric: distributed solar + battery is the relevant edge, but Indiana's lower retail rates and modest insolation make rooftop solar payback long; substitution risk is bounded over a 10-year horizon. (b) Gas: heat-pump electrification is the existential question. Cold-climate heat pumps now work at -20°F, IRA tax credits subsidize the swap, several states are mandating it on new construction. NI's gas territories (IN/OH/PA/VA/KY/MD) are mostly slower-decarbonizing jurisdictions, so the trajectory is decades not years — but the trajectory is unmistakable. Volumes will eventually decline; the question is whether rate-base recovery via fixed monthly charges and depreciation outruns the volume decline. This is the central long-term question.

5. Rivalry — LOW. Competitors do not compete for NI's monopoly customers. They compete for capital in the equity market and for talent, and they compete for the favor of state regulators (i.e., the optics of being the 'good utility' to invest with). Rivalry pressure expresses itself as cost-of-capital pressure — you do not want to be the utility a PUC is angry at.

Value pool location and trajectory. The value pool sits in the regulator's hand: authorized ROE × rate base, minus the political haircut. The pool grows in dollars (rate-base growth at 6–8% per year) but shrinks slightly per dollar (authorized ROEs have drifted from 10.5% a decade ago toward 9.5% today). Net: value pool growing at GDP+, dollar value of EPS likely 6–8% annual. Free cash to equity is structurally negative in growth phases — earnings are reinvested back into the rate base, with the gap funded by debt and stock issuance.

Specific to NI. The gas-LDC mix is 70% of earnings — slightly worse positioning than a pure-play electric in the energy transition, partially offset by a relatively benign regulatory geography (Indiana especially is constructive) and the data-center tailwind on the NIPSCO side, which provides a rare growth lever a pure gas utility lacks.

Industry Verdict: Average. The industry is durable, predictable, and impossible to disrupt in a 5-year window — but structurally returns are capped at the cost of capital plus a thin spread, and the gas half faces multi-decade demand erosion.

Inversion (Bear Case)

I am the short-seller. I will not hedge.

1. The single event that kills this. A serious safety incident on the gas distribution system. NiSource has done this once in living memory — the 2018 Merrimack Valley over-pressurization explosions that killed one person, injured 22, leveled homes across three Massachusetts towns, and ultimately forced divestiture of the entire Columbia Gas of Massachusetts business plus a $143M federal criminal fine. The capital-allocation pipeline now includes vastly more aged cast-iron and bare-steel main replacement, and the next event — at NIPSCO, in Ohio, in Pennsylvania — is a question of operational discipline and luck across millions of touchpoints. One incident with multiple fatalities in a major rate jurisdiction can: (a) trigger PUC backlash that suppresses authorized ROE for years, (b) accelerate municipal gas-ban politics, (c) impose multi-billion-dollar settlements, and (d) permanently alter the cost of capital. The Pacific Gas & Electric template is in the mind of every utility creditor.

2. Why the moat is narrower than bulls think. Bulls think 'monopoly franchise' = wide moat. Damodaran [4] is the corrective: the regulator owns the surplus. The moat protects you from competition, not from the regulator. In the last decade, authorized ROEs have drifted from ~10.5% to ~9.5%, capital structures have been pushed toward higher equity weight (which lowers blended return), and consumer-advocate intervenors have grown more aggressive on prudence disallowances (renewables build cost overruns, fuel-clause adjustments). The moat against competitors is wide. The moat against the regulator's pencil is narrow and getting narrower. ROIIC of 22% will not persist; the steady-state number is ~10% pre-tax, ~7–8% after-tax, which is barely above NI's cost of capital.

3. Why management is worse than it appears. Lloyd Yates is competent but new (since 2023). The Merrimack Valley culture failure occurred under prior management but the culture is built over decades, not quarters. Fewer than 5 years after the largest gas-utility safety event in modern American history, NI is accelerating capex into legacy gas systems. This is necessary work, but it is high-risk work. Equity issuance to fund growth (share count up 3.89% over 10 years) means management is choosing dilution over returning capital — appropriate at fair value, value-destructive at a premium. Today the stock trades at 1.32x base IV; selling stock here is fine, but the bull case requires management to keep selling stock at increasingly fair-to-rich prices, and history tells us that issuance windows close exactly when the company most needs them.

4. What bulls are extrapolating that won't hold. Bulls extrapolate three things. (a) Data-center load growth at hyperscaler-announced trajectories — but data-center load forecasts have a long history of overstatement (the 2000s 'all video moves to streaming' build-out left stranded fiber); even if the load arrives, the regulatory treatment is novel and the IURC may force protective rate design that limits NI's upside (i.e., hyperscalers pay incremental cost, with limited margin for shareholders). (b) Gas demand stability — bulls assume the heat-pump transition is decades away in NI's footprint. The IRA's heat-pump tax credit is ~$2,000/install, federal weatherization grants stack on, and several Ohio cities are already studying gas-line moratoria. The first 5% of demand decline is recoverable via fixed charges; demand declines beyond ~15% trigger a death-spiral dynamic in which fixed costs spread over a shrinking volume base. (c) Stable cost of capital — bulls assume long real rates stay low; if the term premium normalizes to historical levels, both the multiple and the realized spread on rate base compress simultaneously.

5. Valuation trap. PE TTM 29.68, 10-year average 29.87, reverse-DCF growth 5.43%. The 10-year average being similar to current PE looks comforting until you observe that the 10-year window covers an era of zero-rate utility multiple inflation. Pre-2014, regulated utilities traded at 14–17x earnings, not 27–30x. The IRR math at 30x for a 7% grower with a 2.3% dividend is roughly: 7% earnings growth + 2.3% yield = ~9.3% nominal, before any multiple change. A re-rating to 18x — completely consistent with prior decades — is a (29.68/18 - 1) = 39% drawdown, partially offset by ~5–6 years of compounded earnings growth. Net 5-year IRR if regime change occurs: low single digits, possibly negative. iv_low is $26.63; iv_base is $36.32. Either of those is at least 24% below current price. The market is pricing this as a 1.32x base-IV asset and the expected return from current levels in a non-bull-case world is meaningfully negative.

If I am right, the stock could be worth $27–$32 within 3 years.

Lollapalooza Bias Check

Biases I notice in myself analyzing NI right now:

Authority / canon bias. Buffett owns regulated utilities (BHE, MidAmerican, BNSF) and his 2013 letter [6] is unambiguously favorable about the model: 'recession-resistant earnings, which result from these companies exclusively offering an essential service.' My instinct is to treat any regulated utility as a default-good business because Buffett does. The corrective: BHE owns its utilities at book and finances them at sub-investment-grade-equivalent advantage; a public-market investor at 1.32x base IV is in a fundamentally different position than Buffett buying MidAmerican in 2000 at fair value with permanent capital.

Recency bias on data centers. The 2024–2025 data-center narrative is everywhere — every Indiana/Ohio/Pennsylvania utility has announced large-load tariffs, every sell-side note frames data centers as a multi-year tailwind. Recency bias makes me weight the announced load forecasts as if they will all materialize and all benefit shareholders. Both assumptions are weak: load forecasts will partially materialize, and regulatory protections for existing customers will limit shareholder upside on the marginal load.

Anchoring on the 10-year average PE. PE TTM 29.68 vs PE 10y avg 29.87 looks like 'fairly valued at history.' I am anchored on the wrong window. Pre-ZIRP utility multiples were materially lower; the 10-year average reflects a regime that may not return.

Confirmation bias toward 'narrow moat / pass.' Once I form a view that this is a Hold/Avoid, I will tend to weight evidence supporting it. The check: ROIIC of 22.19% is the strongest pro-bull data point; if real (not flattered by transactions and rate-case timing), it changes the picture. I have argued it down to ~10% steady-state — am I sure? Reasonably so for a regulated utility, but I should hold the possibility open that data-center capex is genuinely higher-return than legacy capex.

Social proof / yield-investor crowd. Utilities have a stable dedicated holder base (income funds, retirement portfolios). The presence of these holders props up multiples. I should not mistake a stable holder base for a reason to own the stock at any price; it is a reason the stock will not crater absent a specific catalyst, which is different from a reason to buy.

Deprival super-reaction (the FOMO version). NI just made a Microsoft / hyperscaler announcement a few quarters ago. The bias whispers: if I do not own this, I will miss the big data-center utility re-rating. The corrective: the re-rating is partially in the price already (px/iv 1.32x), and the upside if the bull case materializes is a re-rating to maybe iv_high $53.79 — 12% above today. The downside if the bull case fails is iv_base $36.32 or worse — 24%+ below today. Asymmetry argues for waiting.

Net bias check: my tendencies all bias me toward owning this. The math does not support it at this price.

10-Year Outlook

Will NiSource look fundamentally similar in 2036?

Same business model? Yes. Regulated electric and gas utility under state PUC oversight, growing rate base, paying a dividend, issuing equity periodically to fund capex. The model is 100+ years old; a 10-year extrapolation is among the highest-confidence forecasts available in public equity.

Customer base larger? Modestly larger on the electric side (Indiana population growth + data-center load), modestly smaller in customer count on the gas side as electrification slowly advances. Total meters roughly flat to up 5%. Total kWh and Dth volumes diverge: kWh up materially with data-center load, Dth slowly down on residential gas. Net rate base substantially larger — likely 80–100% above 2025 levels at 6–8% CAGR.

Profit per customer higher? Probably yes in nominal terms (rate-base growth × ROE × inflation-adjusted), uncertain in real terms because authorized ROE may compress further and political pressure on rates is the binding constraint.

Moat wider? No. The regulatory franchise is durable but not deepening. The genuine risk is that 10 years from now, gas-LDC moats are narrower because (a) more municipalities ban new hookups, (b) heat-pump electrification has eroded 10–20% of residential gas volume, and (c) revenue requirements increasingly recover stranded cost rather than productive investment, attracting consumer-advocate hostility.

Single biggest threat. Either (a) a major safety incident, or (b) a faster-than-expected gas-demand decline that creates a 'death spiral' dynamic in 1–2 jurisdictions and pushes regulators to disallow recovery on stranded gas assets. Probability-weighted, I believe (b) is the larger long-run risk.

Confidence in 10-year picture. I can predict the shape of NiSource in 2036 better than I can predict the shape of almost any technology, consumer, or financial business. The franchise is intact, the model works, the volumes drift slowly. What I cannot predict with high confidence is the multiple at which the market will value it — that depends on long real rates, energy-transition policy, and the survival of the gas-LDC narrative. I can predict the business; I cannot predict the price. For a regulated utility this is an acceptable confidence level.

CONFIDENCE: medium