The Quarter in Context
NiSource closed the second quarter of 2026 with $1,289.5 million of customer revenue, a 3.4% year-over-year increase, a steady regulated growth profile rather than a breakout quarter. Total operating revenues of $1,342.4 million compared with $1,283.0 million in the same quarter of 2025, an increase of roughly 4.6%. Operating income of $228.1 million was 13.2% below the comparable quarter of 2025, even as revenue rose, because interest expense climbed to $199.2 million from $139.1 million. The quality of reported earnings was dragged down by a $362.2 million depreciation and amortization line, up sharply from $286.7 million a year ago, which signals an asset base that is growing faster than the income statement can currently absorb.
The cash-flow statement is the more revealing page. Operating cash flow of $680.4 million in the quarter was overwhelmed by $1,097.4 million of capital expenditure, producing a derived standalone-quarter accounting free-cash-flow proxy of roughly negative $417 million. The standalone-quarter figure is calculated by subtracting the prior cumulative filing and is therefore an author calculation from SEC facts rather than a directly reported quarterly line item. For the first six months of 2026, capital expenditures of $1,902.6 million outran the $1,122.7 million of operating cash flow, leaving the utility dependent on long-term debt issuance of $1,239.7 million and minority-interest contributions of $124.5 million to keep building. The quarter is therefore best read as a heavy-investment phase, not a steady-state earnings phase.
Business Model and Profit Engine
Who pays, and what they buy, sits at the foundation of the profit engine. Customers pay NiSource for two distinct regulated services bundled under one holding company. The Columbia Operations segment delivers natural-gas distribution to residential and commercial customers in Kentucky, Maryland, Ohio, Pennsylvania and Virginia through local distribution companies, while NIPSCO Operations delivers regulated electric service and regulated gas service in northern Indiana. Those customers are buying safe, reliable delivery of energy under tariffs approved by state public utility commissions, with approved return on equity ranging from 9.60% at Columbia of Ohio to 10.00% at Columbia of Pennsylvania, and approved incremental revenue from recent rate cases running from $7.8 million at Columbia of Maryland to $257.0 million at NIPSCO Electric. Revenue is overwhelmingly recurring, rate-based, and collected under approved tariffs rather than priced in a competitive market.
The revenue engine runs through three layers. Layer one is the rate-regulated allowed return on a constructed rate base: the utility earns an authorized return on debt and equity that finance property, plant and equipment of $40.3 billion gross at June 30, 2026. Layer two is the recovery of prudently incurred operating costs, including operation and maintenance of $474.9 million in the quarter, through base rates and periodic true-ups. Layer three is the cost-recovery tracker layer, where gas-cost-adjustment clauses pass commodity costs through to gas customers, fuel-adjustment clauses do the same for electric customers, and trackers such as FMCA and TDSIC in Indiana and SAVE in Virginia allow abbreviated proceedings to true up capital programs. Weather and revenue normalization adjustments in Maryland, Virginia, Pennsylvania and Kentucky stabilize the residential and small-commercial revenue stream against temperature swings.
How revenue becomes profit and cash is best understood by separating fixed from variable costs. Cost of energy of $193.6 million in the quarter, down from $261.8 million in the year-ago quarter, is the most variable line and largely offset by customer revenue through the trackers, leaving operation and maintenance of $474.9 million plus depreciation and amortization of $362.2 million as the controllable cost base. A gross-profit figure is not separately disclosed in the filings, so the pass-through nature of cost of energy is author inference from the regulatory-mechanism language rather than a quantified gross-margin statement. Operating income of $228.1 million divided by total operating revenues of $1,342.4 million produces an operating margin of approximately 17.0% in the quarter, with the same ratio producing 17.8% on a six-month basis.
A newer engine is taking shape alongside the regulated core. NIPSCO has executed capacity agreements and engineering contracts tied to data-center customers, and the filings explicitly identify an ADS Contract with Amazon Data Services and an Alphabet Contract, supported by a portfolio of generation and transmission Contract Assets and Pool Resource Assets. The filings flag that realization depends on construction on time and on budget, ability to recover investments under data-center contracts, ability to maintain investment-grade credit ratings while financing this build, and any decision by ADS or Alphabet to terminate or reduce committed capacity. Management has framed the strategy as regulated, but the filings repeatedly warn that the data-center load is incremental to the traditional rate-regulated engine and requires separate execution.
Pricing power and unit economics are constrained by design rather than by competitive pressure. Because rates are set in regulatory proceedings, the utility has no commercial pricing discretion, but it has structural protection against margin compression through trackers and weather adjustments. Operating leverage is positive in the sense that incremental rate-base investment earns an authorized return, but it is also capital-intensive: every additional dollar of approved capex requires supporting debt and equity financing until construction enters rate base. Working capital is dominated by accounts receivable of $846.5 million net at June 30, 2026, gas storage of $173.6 million, materials and supplies of $212.4 million, regulatory assets of $400.0 million current and $2,360.8 million non-current, against accounts payable of $1,135.6 million, creating a working-capital deficit that the filings disclose but do not flag as a concern.
Observable conditions that would weaken the profit engine include any ADS or Alphabet termination or capacity reduction, a downgrade below investment grade, failure of the EPA proposal to repeal the 2024 carbon-pollution standards for fossil-fuel-fired generation to finalize and remove the regulatory drag on planned gas generation that the filings estimate at roughly $675 million of contemplated additional cost, or any state commission disallowing recovery of data-center-related capex. On the positive side, successful execution of the data-center build, completion of approved rate cases at requested revenue levels, and a return to positive free cash flow would confirm that the embedded expectation is being earned.
Earnings Quality and Cash Conversion
Where the issuer reports cash flow and capital expenditure cumulatively, the standalone-quarter cash-flow figures are derived by subtracting the preceding cumulative filing. They are calculations from SEC facts rather than separately reported quarterly line items. Reported net income attributable to NiSource was $45.5 million for the quarter, a 55.5% drop from $102.2 million in the same quarter of 2025, even though revenue was higher. Diluted earnings per share fell to $0.09 from $0.22, and the year-to-date figure of $1.15 is running slightly behind the $1.22 reported in the first six months of 2025. The pattern is consistent with a utility whose reported earnings are being compressed by interest expense and depreciation, not by operating performance: operating income actually held up well at the six-month level, $1,047.3 million in 2026 versus $1,022.3 million in 2025.
Cash conversion is the real concern. Trailing-twelve-month operating cash flow of $2,303.2 million is comfortably positive and up 37.3% versus the prior twelve months, which signals healthy working capital and rate recovery. Capital expenditure over the same trailing window was $3,389.4 million, up 66.7%, producing a trailing-twelve-month accounting free-cash-flow proxy of negative $1,086.2 million. Because the filings do not separate maintenance from growth capital spending, that figure is best read as a rough proxy of how much operating cash the utility must supplement with external financing each year, not as a measure of maintenance capex. Dividends paid were $144.3 million in the quarter and $293.3 million year-to-date, comfortably covered by operating cash flow on a year-to-date basis but not by the accounting free-cash-flow proxy.
The balance sheet preserves execution capacity. Long-term debt excluding current portion stood at $15,612.6 million, with $1,098.4 million of current portion and $692.0 million of short-term borrowings at June 30, 2026, against $70.0 million of cash and $25.4 million of restricted cash. A current-debt line of $19.7 million exists at December 31, 2025, but it is excluded from current quarter calculations because the snapshot predates the latest balance-sheet date. Total available liquidity was $2,098.0 million, and the debt-to-capitalization ratio of 52.6% sat well below the 70.0% revolver covenant. Credit ratings of BBB+ at S&P, Baa2 at Moody's and BBB at Fitch, all with stable outlooks, leave room to issue the long-term debt the filings confirm was raised in May 2026.
Accounting free cash flow remains -$1.1 billion. In a separate, more conservative owner-earnings sensitivity, treating TTM stock-based compensation of $0.1 billion as an economic dilution cost produces approximately -$1.1 billion. This is an analytical approximation, not a company-reported measure, and it treats compensation paid in shares as an economic cost even though the cash-flow reconciliation adds it back.
Business Quality, Moat, and Industry Position
On the available evidence, the company is best described as an important participant, not a demonstrated core supply-chain bottleneck. This classification describes strategic position only; it is not independent proof of a moat or valuation upside. NiSource's economic moat is a regulated monopoly franchise. Local distribution companies in five states and a vertically integrated electric and gas utility in northern Indiana operate under certificates of public convenience and necessity, and approved rate cases set the return on equity the company can earn on its rate base. Replacement-program riders such as the Ohio CEP and Kentucky SMRP, combined with weather and revenue normalization adjustments in Maryland, Virginia and Pennsylvania, give the utility a recurring, predictable revenue stream that is largely insulated from commodity price swings.
The moat, however, is geographically narrow. NiSource serves a contiguous Midwestern and Mid-Atlantic footprint, with no diversified utility platforms and limited geographic optionality. The data-center strategy is a deliberate attempt to broaden the load profile, but it concentrates customer exposure: the ADS Contract and the Alphabet Contract, plus any future data-center counterparties, are large individual counterparties whose performance the filings flag as a material risk. Supplier-side, the filings disclose long-term contracts for equipment and engineering, procurement and construction, but they do not name the suppliers, leaving supply-chain resilience partially unverifiable from the available evidence.
Regulatory and environmental tailwinds are mixed. Approved ROEs in the 9.60% to 10.00% range and pending rate cases in Virginia and Kentucky imply continued rate-base growth if approved. The EPA's proposal to repeal 2024 carbon-pollution standards for fossil-fuel-fired generation could remove regulatory drag on planned gas generation, with management estimating roughly $675 million of avoided cost contemplated through the 2024 NIPSCO integrated resource plan. Offsetting this, Maryland building-performance standards and a possible Clean Heat Standard could erode the gas-distribution customer base over time, and the filings explicitly state that NiSource cannot yet predict the final impact of those policies. The moat is therefore durable in the regulated core, but its future shape is being contested by climate policy on both ends.
Management and Capital Allocation
Management has chosen a capital-allocation policy that prioritizes growth over near-term cash return. Capital expenditures of $1.9 billion in the first six months of 2026, and $3.4 billion on a trailing-twelve-month basis, dwarf the $293.3 million of common dividends paid year-to-date and dwarf stock-based compensation of $50.9 million. Long-term debt issuance of $1,239.7 million in the first six months, plus minority-interest contributions from Blackstone affiliates of $124.5 million at the NIPSCO and GenCo vehicles, funded the gap. Equity issuance was minimal, with only $7.0 million of net proceeds from common stock and modest activity under employee stock programs. The company is not buying back shares.
The capital-allocation framework is internally consistent with a regulated utility in build-out mode. Dividend per share of $0.30 in the quarter and $0.90 year-to-date imply a payout that absorbs a meaningful share of net income but is comfortably below operating cash flow on a year-to-date basis. The October 2025 GenCo minority-interest transaction with Blackstone and the earlier NIPSCO Holdings II transaction give the parent a template for raising partial project equity without diluting common shareholders, while letting the asset-level lenders and JV partners share specific risks. Incentives appear aligned with regulated build-out: a substantial portion of executive compensation is delivered in restricted stock units and performance share units under the 2020 Omnibus Incentive Plan, and the Compensation Recoupment Policy referenced in the 10-K exhibits addresses clawback risk.
Incentives, however, are tilted toward size and visible execution. Management's case for the data-center strategy depends on realized return on the Contract Assets and Pool Resource Assets, which the filings warn is subject to construction risk, performance risk and regulatory recovery risk. The same filings note a $153.9 million collateral requirement that would be triggered by a downgrade below investment grade, which links management's execution to credit-rating preservation. Until free-cash-flow generation turns positive and data-center capex stops absorbing operating cash, capital allocation is best judged as execution-dependent and reversible rather than as a steady shareholder-return program.
Valuation and What the Market Already Expects
At $41.07 per share and a $19.7 billion market capitalization as of September 1, 2026, the dated price implies the market is paying for a utility whose value will be unlocked over several years of build-out. The cash-flow yield at the consolidated level is negative: trailing-twelve-month operating cash flow of $2.30 billion against trailing capital expenditure of $3.39 billion implies the market is underwriting continued external financing of roughly $1 billion per year. Against reported diluted EPS of $1.57 on a trailing basis, a $41.07 price implies a price-to-earnings multiple in the mid-twenties, but that multiple is depressed by the heavy depreciation and interest expense flowing through the income statement rather than by depressed operating performance, and because the denominator is a true trailing-twelve-month EPS figure the multiple is mechanically defined rather than a quarter run-rate sensitivity.
The balance sheet gives the market some comfort. Long-term debt of $15.6 billion against $9.57 billion of NiSource stockholders' equity implies a book leverage that, while substantial, is consistent with a regulated utility whose rate base is also debt-financed. The debt-to-capitalization ratio of 52.6% against a 70.0% covenant gives the company runway to issue more debt, and the BBB+ to BBB credit ratings with stable outlooks leave that capacity unimpaired. What the market is implicitly underwriting is the data-center contracts, the regulatory recovery of the elevated capex, and a stable cost of capital during the multi-year construction period, not current free-cash flow.
This is a qualitative discussion of dated price versus dated market capitalization and valuation burden only. No scenario intrinsic value, terminal multiple, exit multiple or audited fair value is asserted here. With current free-cash-flow generation negative, any reverse-discounted-cash-flow discussion must rely on assumptions about future cash-flow normalization rather than on compounding from the present base.
Ten-Year Cash Flow to Fair Market Value
Direct compounding from the current trailing-twelve-month accounting free-cash-flow proxy of negative $1,086.2 million is not meaningful, because a negative base produces no defensible single-scenario intrinsic value. A bear, base and bull normalization framework is therefore more useful, and every input below is an author assumption rather than a stated company forecast.
In the bear case, revenue growth slows to low single digits, the data-center load does not ramp, and the accounting free-cash-flow proxy remains modestly negative through 2030 before recovering to roughly $200 million by 2035, leaving the dated market capitalization to imply continued reliance on debt and minority-interest financing with no margin of safety. In the base case, approved rate-case ROEs of 9.60% to 10.00% hold, the data-center contracts proceed, and the accounting free-cash-flow proxy turns positive in 2028 at roughly $300 million, then expands to roughly $800 million by 2035, with dividend per share compounding modestly. In the bull case, the EPA repeal of 2024 GHG standards finalizes, the data-center load ramps ahead of schedule, and the accounting free-cash-flow proxy reaches roughly $1.5 billion by 2030 and $2.5 billion by 2035.
Using illustrative discount rates of 8% to 10% on terminal-year free cash flow, the dated market capitalization of $19.7 billion would require a terminal-year accounting free-cash-flow proxy in a roughly $1.6 billion to $2.0 billion range under the base case to support a typical regulated-utility total-return profile, a level materially above trailing-twelve-month results that would take several years of capex normalization and rate-case approvals to reach. That is the core qualitative observation about what the dated price already implies, and the gap between today's negative cash conversion and the implied terminal-year normalization is the central risk that the conclusion depends on.
Five Value-Investing Lenses
Benjamin Graham Lens
The Graham downside floor is mixed. The company has positive stockholders' equity of $9.57 billion, current liabilities of $4.41 billion are well covered by current assets of $2.20 billion only when combined with operating cash flow, and long-term debt of $15.6 billion is the principal balance-sheet exposure. The debt-to-capitalization ratio of 52.6% sits well below the 70.0% revolver covenant, but the company is structurally dependent on continued debt issuance to fund the capex program. A famous monopoly franchise and a declining share price are not, on their own, evidence of cheapness; the trailing price-to-earnings ratio is elevated because depreciation and interest have compressed reported earnings, not because the price is low. Without a defensible normalized free cash flow, the Graham margin-of-safety test cannot be passed with confidence.
Benjamin Franklin Lens
The Franklin lens asks whether capital is being deployed with thrift and discipline. NiSource is committing nearly $3.4 billion of trailing capital expenditure against $2.3 billion of trailing operating cash flow, financing the gap with long-term debt and minority-interest capital rather than equity dilution, which preserves per-share economics but raises the cost-of-capital risk. The opportunity cost of those dollars is the dividend, which on $41.07 per share and roughly $1.20 annual dividend per share implies a cash yield near 2.9% before growth. Thrift would suggest the dividend payout be held flat or grown modestly while the capex program is in flight, which is consistent with what management is doing. Discipline would suggest caution on the data-center strategy until the data-center cash flows are visible. The filings do not yet show that discipline has been earned.
Warren Buffett Lens
Buffett would focus on owner economics and capital allocation. The accounting free-cash-flow proxy is negative on a trailing basis, so there is no positive owner-earnings figure to anchor valuation today. Long-term debt of $15.6 billion, current portion of $1.1 billion, and short-term borrowings of $0.7 billion against $70.0 million of cash imply that the dated market capitalization of $19.7 billion is supported by equity and substantial debt, with enterprise value far above equity value. Capital allocation has been directed at growth and dividend continuity rather than at share repurchases, which is appropriate for a regulated utility but means common shareholders are relying on the regulated build-out to compound per-share value. Until free cash flow turns positive and the data-center strategy produces visible earnings, the Buffett lens cannot confirm an attractive entry point.
Charlie Munger Lens
Munger would invert the question and look for the thesis-breaker. The most likely wrong assumption is that ADS and Alphabet will not terminate or reduce committed capacity under their data-center contracts, and the filings explicitly warn that any such termination is a material risk. A second wrong assumption is that PHMSA pipeline-safety regulations will not impose material new costs; the filings state that the company cannot predict the final impact of changing pipeline-safety regulations. A third wrong assumption is that Maryland's building-performance standards and possible Clean Heat Standard will not erode the gas-distribution customer base over time. Incentive-wise, management compensation is tied to long-term incentive plans that reward execution and growth, which is appropriate, but the absence of any share repurchase program means executives are not signaling undervaluation with their own capital. The thesis would break quickly if any of these risks materialized.
Li Lu Lens
Li Lu would ask whether this is inside the circle of competence and whether the durable advantage is intact. The regulated-utility core is well inside any traditional value-investor's competence: rate-base growth, approved ROEs, and recurring revenue are textbook durability drivers. Culture appears constructive, with a safety-management system framework, weather normalization adjustments, and approved rate cases across multiple jurisdictions. The data-center strategy, however, is a different business that depends on construction execution, regulatory recovery, and the creditworthiness of a small number of large counterparties, and the filings flag that NiSource's Net Zero Goal by 2040 is subject to assumptions that did not include data-center load. Permanent-loss risk today is low for the regulated core but elevated for the data-center build, and the cleanest Li Lu framing is to monitor the company as a regulated utility whose circle of competence is being stretched by a new growth bet.
Risks, Disconfirming Evidence, and What Would Prove This Thesis Wrong
Three observable thesis breakers stand out. First, any termination or material reduction of committed capacity by ADS or Alphabet under the Current Data Center Contracts would invalidate the primary growth narrative; the filings flag this as an explicit forward-looking-statement risk. Second, a downgrade below investment grade would trigger roughly $153.9 million of additional collateral requirements under ratings-trigger agreements and could materially raise the cost of debt funding the capex program. Third, adverse outcomes in the pending Columbia of Virginia and Columbia of Kentucky rate cases, or disallowance of data-center-related capex by any state commission, would weaken the rate-base growth story that supports the dated market capitalization.
Other disconfirming evidence to monitor includes changes in PHMSA pipeline-safety regulations that impose material new costs, Maryland building-performance standards or a Clean Heat Standard that accelerates gas-customer attrition, and any decision by the EPA not to finalize repeal of the 2024 carbon-pollution standards, which would preserve the roughly $675 million of contemplated cost in the 2024 NIPSCO integrated resource plan. On the positive side, successful execution of the data-center build, completion of approved rate cases at the requested revenue levels, and a return to positive free cash flow would confirm the embedded expectation.
Investment Conclusion
NiSource is a regulated utility whose current dated price reflects a multi-year build-out tied to hyperscale data-center customers, not current free cash flow. The balance sheet and credit ratings give the company room to execute, and the regulatory franchise is durable, but the trailing-twelve-month accounting free-cash-flow proxy is structurally negative. The qualitative implication is that any attractive entry point depends on the embedded expectation that the data-center strategy ramps on time and on budget, that data-center capex is recovered in rates, and that investment-grade credit is preserved. Without those conditions, the dated price does not yet offer the kind of margin of safety that a value investor would require. The investment is best framed as worth watching, with the conclusion conditional on explicit cash-flow normalization assumptions.
*This article is for educational and informational purposes only and does not constitute investment advice, an offer to buy or sell any security, or a recommendation to take any specific action. *
Share this article
