Two weeks in the dark: the machine behind NIPSCO's slow recovery

Two weeks in the dark: the machine behind NIPSCO's slow recovery

Two weeks after a violent line of storms swept across northwestern Indiana, around 10,000 homes served by Northern Indiana Public Service Company (NIPSCO) were still dark. Most were in Gary. On August 24th Indiana's governor, Mike Braun, directed the Office of Utility Consumer Counselor to petition the state's regulatory commission to investigate the utility — not only for restoring power slowly but for how it spent the money that ratepayers had handed over for reliability. "NIPSCO has failed to keep its end of the bargain," he said.

The storm deserves its due. A derecho, one of the strongest thunderstorm complexes of the summer, drove gusts near 100mph across Gary. At its peak more than 370,000 customers lost power — the largest outage event in the company's history and, by NIPSCO's own account, roughly three-fifths of its electric customers. The storm broke 1,230 distribution poles, against 127 in the four previous major storms combined. Crews came from out of state, and the National Guard was deployed. A proposed class action, filed in Porter County, accuses the company of failing to manage vegetation and inspect its network, and seeks punitive damages.

NIPSCO's defence is fair as far as it goes: it says spending on vegetation management has more than doubled since 2016, that no maintenance programme survives a 100mph wind arriving on every circuit at once, and that most of the remaining customers would be back by August 25th. All of that may be true and yet miss the point. The question the governor has put to regulators is narrower and more consequential for the shares than the storm's damage bill. It asks whether money that Indiana approved through a device called a tracker, intended for reliability, was "used for their intended purposes". Understanding why that question bites requires a look at how a regulated utility earns its keep.

A utility is a monopoly granted by the state, at prices the state approves. In exchange for a decent chance of profit it promises to keep its patch of grid safe and dependable. Indiana tried to strengthen that bargain in 2013 with the Transmission, Distribution and Storage Improvement Charge, a tracker created by the legislature. Under it a utility files a seven-year improvement plan with the regulator and then collects 80 per cent of the approved spending automatically, month after month, through a line item on its customers' bills; the remaining fifth is deferred to the next full rate case. The statute's declared purpose was, explicitly, maintenance: safety, reliability and modernisation.

The trouble is what the tracker pays for. It rewards capital — new poles, transformers, substations, smart meters — because those enter rate base, on which the utility earns a regulated return, plus depreciation. It adds almost nothing for the recurring labour of trimming trees away from wires, an operating expense that creates no rate base and therefore no extra profit. The seam has been visible since the beginning. In 2018 the state supreme court ruled that only specific, pre-approved projects could be billed through the tracker, warning that billions of dollars were funnelling through it. The legislature overturned that ruling in 2019, relaxing approval into what critics call a rubber stamp and a blank cheque. And at this very moment NIPSCO stands before the same supreme court arguing that $1.64bn of cost overruns on its electric-infrastructure plan should be recoverable through the tracker — over the objection of industrial customers led by US Steel, who say the mechanism lets the utility dodge the prudence review meant to protect customers from unjustified spending. A further $741m gas-tracker plan is in the pipeline.

None of this need threaten the dividend today. Storm restoration costs are recovered in rates, with a lag and a return, and one storm's bill is a rounding error beside 2026 adjusted earnings guidance of $2.02–2.07 a share, spread over a company worth about $23bn. The danger is to the growth story that justifies the yield. NiSource, NIPSCO's parent, plans to grow its rate base by 9–11 per cent a year to 2033 on a $28.6bn capital programme, and guides long-run earnings growth of around 9–10 per cent. Rate-base growth of that kind is not income; it is an agreement with regulators to be paid later. A governor prepared to describe the utility as a broken promise is a bad omen for the pace and generosity of future approvals — and if the investigation finds tracker money misspent, disallowances and clawbacks would land first on the customers who funded the programme under scrutiny, and eventually on the shareholder who expected it to keep compounding.

So far the market has filed the episode under weather. On the evening the governor acted, NiSource's shares traded around $41 — roughly twenty times guided earnings, on a dividend of $1.20 a year that yields about three per cent. That is a utility priced as though its growth plan will sail through its regulators, from whose mercy every dollar of that plan is borrowed. What happens next is unusually legible. The commission will pore over NIPSCO's vegetation records and its tracker accounts. The supreme court must rule on the $1.64bn case, setting the standard for all five of Indiana's investor-owned utilities at once. And the tracker's own statute requires a full base-rate case when the seven-year plan ends — the moment when a disallowance, or a political grudge, becomes an actual bill.

The deeper lesson is about the design, not the delinquent. Indiana wrote "maintenance" into the tracker's charter in 2013 and then removed most of the machinery that might have verified that maintenance occurred. A regulator that cannot see what its own tariff money buys is the weak link in the machine — and the machine, not the storm, is what moves the share price. Every investor in a regulated utility is ultimately buying the same thing: the continuing credibility of a bargain with government. Gary spent two weeks discovering what it looks like when that bargain has to be tested, and how slowly the answer arrives.