A rooftop solar array on a home in Weymouth. (Photo by Senthil Balasubramanian via Wikimedia Commons/US Department of Energy)

Quick question: How do you think your electric utility makes its money?

If you said, “by selling electricity,” you’re in good company — and you’re also wrong. The three big utilities that serve most of Massachusetts — Eversource, National Grid, and Unitil — don’t make their money on electricity. They buy it on your behalf and bill you what it costs them. No markup, no profit.

They earn their profit on the wires, poles, substations, and transmission lines that deliver the power. State regulators let them earn a guaranteed return — currently around 9 to 10 percent a year — on the money they sink into that equipment, for its entire decades-long life. The industry term is “rate base,” and the bigger the grid gets, the more the utilities’ shareholders earn. A large share of your delivery charge is that guaranteed profit.

Why does this matter? Because a study now sitting in front of the Massachusetts Department of Public Utilities could be used to cut what homeowners are paid for their solar power, and its conclusion only holds up if you don’t notice where utilities actually make their money. Once you do, it’s no basis for cutting anyone’s solar credits.

Most homes with rooftop solar are “net metered.” On a sunny afternoon, your panels often make more power than your house needs. That extra electricity flows onto the grid for a neighbor to use, and your meter runs backward, crediting you for what you send out. At night or on cloudy days, you draw power from the grid, which counts against your credit. At month’s end, you pay only for your net usage — what you took minus what you sent back.

Net metering is what makes home solar financially workable for ordinary families, and it’s the thing some parties now want to scale back.

Last month, MIT economist Christopher Knittel filed a study with the DPU, through MIT’s Center for Energy and Environmental Policy Research, concluding that rooftop solar pushes electricity rates up while large utility-scale solar farms push them down. He didn’t formally recommend cutting net metering credits. He didn’t have to. The study is now in the official record, and the parties who want net-metering credits reduced are already pointing to it.

Before acting on a study, regulators should ask whether it measured the right thing. This study asked a fair question: Which is the better deal for ratepayers — rooftop solar, or big solar farms? But it left out the one thing that decides the answer — the grid costs those solar farms trigger.

A utility-scale solar farm in Central Massachusetts has to connect to the grid that carries power to homes. The developer pays to reach the nearest line — but those lines often can’t carry the extra power, so the utility upgrades them. It adds those upgrades to its rate base and earns that guaranteed 9 to 10 percent on them for decades. That cost is real and large, and lands on your bill — but it never appears in the MIT study’s cost of utility-scale solar.

Rooftop solar creates little of that added grid cost, because the power is used right where it’s made — by the home itself, or by a neighbor next door when there’s extra — instead of crossing the grid from a distant solar farm. It needs little or no new utility equipment. And when it does need an upgrade — usually a new transformer, which runs $5,000 to $10,000 — the homeowner pays for it, not the utility. Either way, those costs never enter the utility’s rate base, so they earn it no guaranteed profit.

So the MIT study isn’t comparing apples to apples. It counts every penny of the net-metering credit against rooftop solar, but leaves the delivery equipment — where the utility earns its money — out of the cost of the big solar farms. That’s not a finding about solar; it’s an accounting choice that favors the utilities’ business model.

A solar home still leans on the grid for power at night and through short winter days — no one disputes that. Knittel’s claim is that residential solar causes a cost shift among ratepayers. Because solar owners buy far less grid electricity overall, they pay far less toward the delivery charges — including the utility’s guaranteed profit — bundled into every kilowatt-hour, so the burden shifts onto everyone else. A home that all but zeroes out its bill can sidestep those charges almost entirely.

There’s something to that — but less than it seems. Start with the picture the argument depends on — the solar home that erases its whole bill. In 20 years of installing these systems, I’ve found those homes to be the exception. Roofs are only so big, and most can only fit enough panels to cover part of a home’s use. The typical solar owner still buys plenty of grid power and pays full delivery charges, guaranteed profit and all. The same logic that faults residential solar would indict anyone who uses less electricity: Switch to LED bulbs or hang your laundry instead of running the dryer, and you too buy fewer kilowatt-hours and pay less toward fixed costs. We don’t call efficiency a subsidy.

What a solar home is actually worth to the grid — the peak demand it shaves, the grid upgrades it defers, the power it feeds the house next door — is exactly what the MIT study leaves out. When Lawrence Berkeley National Laboratory measured rooftop solar’s real effect on other customers’ rates, it found the impact negligible for years to come. Whether solar owners pay their fair share is a fair question — but this study can’t answer it, because it never counts what utility-scale solar costs the grid.

A wide body of independent research — including work by the US Department of Energy’s national laboratories and the New England-based Acadia Center — keeps reaching the same conclusion: Rooftop solar is worth it to the grid. Most independent and regulator-commissioned studies land there; the ones that don’t were typically commissioned by utilities. As the Brookings Institution put it in 2016, after reviewing studies from regulators in at least 10 states: “Far from a net cost, net metering is in most cases a net benefit — for the utility and for non-solar rate-payers.”

Back in 2015, a Massachusetts task force on net metering recommended a comprehensive value-of-solar study to quantify these avoided infrastructure costs. Ten years later, that study still hasn’t been done. In their recent report, the MIT researchers join the call to finally conduct it — but in the same breath, they essentially make the case for the DPU to change net metering before it is done.

None of this means net metering can never change; how we pay for solar should evolve with the grid. But any change has to count what utilities collect on the big solar farms that rooftop solar is measured against. Otherwise, regulators aren’t weighing two fair options; they’re setting one’s full price against the other’s discounted price and drawing a conclusion based on the difference between them.

The DPU should not change net metering until that missing number is part of the calculus. Leaving it out tilts the scale toward the very business model regulators are supposed to keep in check, and away from the ratepayers they’re supposed to protect.

Mark Durrenberger is the founder and president of New England Clean Energy, a solar and heat pump company based in Hudson. He has been installing residential and small commercial solar systems in Massachusetts for more than 20 years.