David Roodman

How I Helped Double the Limit on Rooftop Solar in My “State” and Maybe You Can Too

Guest post by David Roodman

On August 14, there appeared online a fresh issue of an obscure publication of the government of my home territory, the District of Columbia. Within its dour pages, the DC Register announced that net-metering customers, which includes most of us with rooftop solar, would eventually be allowed to install systems twice as large as before. When I saw the news, I let out a small “whoop” of exultation, for it culminated my multi-year effort as a private citizen toward just this goal. I’m going to tell you the story of that effort—not, I sincerely believe, for my own glorification, but in the hope that it will multiply the modest good I have done. Perhaps I will inspire you to pursue the same reform where you live, or pursue other changes in the investigatory style that worked for me.

The spark

My story begins almost exactly three years ago, when within a month two of my solar panels emailed me to tell me they’d gone offline—as it were. I realized that in seven years, I’d had ten failures among my 30 Enphase microinverters—the boxes on the backs of my solar panels that converted from DC to wall-plug AC. Sensing a problem, I scoured the Internet. I learned that defects in my generation of microinverter were a known thing, and that Enphase was spending enough money replacing them under warranty that the whole company was in danger of going offline—as it were. That scared me: my microinverters might keep failing, with no one to fix them.

But then I discovered that my original installation contract contained a buyback clause. So I decided to exercise it, forcing the installer to take back my panels and inverters at a fraction of the original price. Then I prepared to get new panels that were more efficient and didn’t come with those faulty inverters. Happily, panel prices had fallen a lot since my first installation. (Enphase has since obtained new financing and launched a microinverter replacement program.)

As I spoke anew with installers, I noticed that they disagreed about how large my new solar system could be. One said my projected electricity production could not exceed 100% of my family’s historical usage. Another said 120% could be OK. So I probed. What is the size limit and where is it written? The installers didn’t really know. Frontline customer service at my utility, Pepco, confirmed 100%, but could not cite legal chapter and verse. I asked to talk to people higher up. I cold-called the regulatory agency of which I had till then been only dimly aware, the Public Service Commission (PSC). After repeated calls and explanations that did not check out, I got one that rang true. DC law defines a net-metering customer-generator (that’s me, and probably you if you have solar) as one whose system “is intended primarily to offset all or part of the customer’s own electricity requirements.”

Hmph, I thought. So the law says my panels should be intended primarily for own use. My utility, by capping expected production at 100% of consumption, is interpreting the law as intended exclusively? Why couldn’t I install a system that generated 150% of the power my house uses? Two-thirds of that—100% out of that 150%—would be for own use, which seemed like “intended primarily.”

That was when I felt the fire to challenge the limit. A few factors motivated me. First was the feeling that we have to fight for climate progress on every front. Second was the confidence I’d gained over decades as a professional researcher: policy-relevant analysis is one of my superpowers. And third, I’ve come to appreciate, was the sense of agency that I have acquired as a quantitatively oriented, educated, straight, white male. Nothing in my life has taught me that I should flinch at the danger of challenging the system.

Doing my homework

The upcoming Hollywood film inspired by my story portrays its next chapter through a video montage with a stirring soundtrack. Played by Matt Damon, I’m cradling a phone on one shoulder while furiously jotting on a notepad, like Woodward and Bernstein. Then I run up the steps to a library entrance, like Rocky at the art museum, pull musty law books from the shelves, and intensely thumb the pages…

OK, maybe not. Really, most of my calls were not returned. And I did the rest of my research on my computer. The point is, I learned enough that now, in the retelling, I need to skim the surface. (But see the end of this post for bonus facts.)

I learned that that vague phrase, “intended primarily” came into DC law through a bill the city council passed in 1999 to allow net metering—to allow, that is, for customers to sometimes produce more power than they consume, causing their electric meters to run backward. I read through the lawyerly back-and-forth that played out over five years between DC government agencies, environmental groups, Pepco (the regulated power company), and the PSC (the regulator) as the PSC decided how to translate the law into regulations. I learned that the PSC ultimately ducked the challenge of turning “intended primarily” into a number. I learned that Pepco had filled that regulatory vacuum by imposing its own limit: one’s expected solar production should not exceed 100% of historical usage. And I noticed that Pepco told consumers that government rules mandated this ceiling. It didn’t seem that way to me.

Looking outside of DC, I learned that at least 18 states have the same “intended primarily” phrase in law. It seems as if California came first, in 1995, and the rest copied it. Yet states have diverged in how they quantify “intended primarily.” California’s regulator was the strictest, imposing a 100% interpretation. Other states have derived no usage-based limit from that phrase, which I confirmed for Vermont and Wyoming by calling utilities in those states and asking questions as if I lived there. I think other states, such as Ohio, were like DC: the regulators were silent and the utilities imposed 100%. (But it looks like Ohio just officially went to 120%.)

Most striking for me was the situation in Maryland. For some DC residents, Maryland was just across the street. On either side of Eastern Ave., residents contracted with the same legal entity, Pepco, for their electricity service, which was delivered over the same physical electrical grid, under the same “intended primarily” legal language. Yet Marylanders were allowed to install twice as much solar. For the Maryland regulator had taken a stand on “intended primary” and defined it as up to 200% of usage. The rationale was that if a customer’s production is 190% or 199% of consumption—edging toward 200%—then consumption is still a bit more than half of production, and that’s “primarily.”

Picking a fight

Having sized up the situation, I coordinated with my new solar contractor to pick a fight. In January 2018, we applied for permission to install 7.92 kilowatts of panels, enough to get me to about 120%. As expected, Pepco promptly said no. I then filed an informal complaint with the regulator, the PSC. In its written response, Pepco asserted that its formula for converting kilowatts of capacity (the maximum power a system can generate at any moment) to kilowatt-hours of output (how much energy it will produce in a year) was “very liberal.” So while their 100% limit looked on paper like a strict “intended exclusively” standard, really it was looser. In particular, Pepco used the rule of thumb that in the DC region, each kilowatt of capacity puts out 1200 kilowatt-hours of electricity over the course of a year.

Since Pepco backed its assertion with no data, I collected my own. Using Google Maps, I found houses near mine with panels atop, went and knocked on doors, told my story, and asked if the owners would share their data. I also extracted information for some homes online, including from pvoutput.org. My unscientific sample of 15 houses in Washington, DC, suggested that Pepco’s 1200 rule of thumb is merely a little liberal. The houses I got data for produced 1292 kilowatt-hours each year for each kilowatt of capacity, which is only 8% above Pepco’s 1200. So maybe Pepco’s limit looked like 100% on paper and worked like 108% in practice: still pretty strict. Pepco never rebutted me with its own data.

In April 2018 came the meeting to resolve the informal complaint. The PSC employee in charge quickly saw that this was no routine billing dispute. She encouraged me to file a formal complaint.

So I did that, and earned myself a pre-hearing meeting the next month, with a PSC officer presiding, a Pepco lawyer opposing, and a free lawyer from the Office of the People’s Counsel at my side. The movie version will be shot in an oak-paneled courtroom with a gavel-pounding judge. In real life, we gathered in small, windowless conference room, where the proceedings mixed cordiality with flashes of anger. The hearing officer voiced skepticism about my case, and the Pepco lawyer, in her element, suggested I’d be more successful if I filed a comment in an ongoing rulemaking process. In other words, the writing on the drab conference room wall was that I should shift from changing how the current rule was interpreted to changing how it was written. I think Pepco was especially concerned that if I won my complaint about the current interpretation, then hundreds of DC solar owners could sue the utility for having wrongly restricted their installations. But an official rule change would validate the historical practice as the old normal.

So, while awaiting the hearing officer’s decision after our meeting, in early June, I filed a comment with the PSC. In the hope of getting taken seriously, I copied the professional submissions I’d been reading, in font, margins, and citation style. I wrote much of what I’ve just explained to you. Through a DC-area Google group run by DC SUN, I encouraged others to file supportive comments, and some did.

In July, the hearing officer handed down his decision. There would be no more proceedings. He had found a precedent in favor of Pepco’s 100% limit. In 2015, the PSC had stated that net-metering customers are “limited by law and [PSC] rules to installing generation that is forecast to provide no more than 100 percent of annual historic use.” It was a weak precedent in my view since it implied that in some other text, the PSC had declared that “intended primarily” = 100%—and as far as I know there is no such text. But a weak precedent is still a precedent. After some reflection and consultation with my free lawyer, I decided that I probably couldn’t overturn the rejection of my formal complaint.

The working group

Time passed. In late October I thought to scrounge through the PSC’s document archives and discovered that a month earlier the agency had gently endorsed my June comment in an official order. “It is particularly important to consider Mr. Roodman’s proposal because it would harmonize our practice with Maryland’s and raising the generation threshold could further the District of Columbia’s express policy goals of increasing renewable generation from sources within the District of Columbia.” The PSC directed that the issue be added to the agenda of a working group it had already created to deal with DC’s difficulties with community solar.

Working groups, it turned out, are open to anyone, and are announced on the PSC web site. The PSC uses the groups to gather the views of concerned parties and seek consensus. No one had sent me the memo about this one! It reminded me of the what the Vogons said just before they destroyed the earth: “There’s no point in acting surprised about it. All the planning charts and demolition orders have been on display at your local planning department in Alpha Centauri for 50 of your Earth years….What do you mean you’ve never been to Alpha Centauri? Oh, for heaven’s sake, mankind, it’s only four light years away….if you can’t be bothered to take an interest in local affairs, that’s your own lookout.”

Though I was late to learn about the orders on display at my local planning department, I was not too late. I hadn’t missed the first meeting. In the next year and change, I convened with representatives from Pepco, the PSC, other government agencies, and the solar industry. I was the only person there in a private capacity. We met seven times in person and once virtually. As you can see, the PSC did not shepherd us along with Manhattan Project–like speed.

The biggest surprise for me in the meetings was that the regulator appeared more skeptical of raising the net metering limit than the regulated. Perhaps Pepco saw that raising the legal limit would not extend physical limits by a single photon. For example, I doubt I can get much past 125% on my roof with today’s technology, even though I work hard to conserve energy and have a pretty good roof for solar. Whereas I suppose the PSC worried about unintended consequences such as destabilizing the electricity distribution system and shifting the costs of the system onto non–solar customers. In the third meeting, the PSC staff proposed raising the limit from a de facto 100% to an official 120%. This they cast as compatible with a continuing, longer-term rise. But that nuance disappeared in the meeting minutes: it looked like they just wanted to go to 120% and stop there. Meanwhile, the PSC staff commissioned a survey of what other states do, and cited it as tending to support such a modest increase. But they seemed reluctant to share the report with me.

At that point I got a bit combative. I filed a Freedom of Information Act request for the report, which the PSC staff promptly and politely fulfilled. I suggested in email to the working group that the PSC would look tin-eared if it made such a minimal adjustment in the face of a climate crisis, the city’s ambitious new climate law, and Maryland’s 200% example. I encouraged other solar enthusiasts to join the next meeting to push for more aggressive change.

If cautious, the PSC staff were also conscientious. They listened to the working group consensus. At the next meeting, a representative of DC’s Department of Environment proposed raising the limit to 200% over five years. The utility, Pepco, did not oppose. On the spot, I decided it was time to discard my somewhat adversarial stance, join with others, and shut up. I endorsed the proposal. I think I’m inept at working in groups, but I sense that if others advocate what you favor while you stay silent, that’s a victory.

Accepting the crystallizing consensus for a steady rise toward 200%, the PSC staff focused in the following months on the question of price. If you’re pushing more power into the grid than you’re pulling out of it, how much should you get paid for the excess? Many solar owners gravitate to the self-serving answer: we should get paid the full retail price. I appreciate, however, that the local power system does me a huge amount of good even when my house produces more power than it consumes. The grid is there for me at night, it’s there for me if my panels break, and it smoothly delivers my excess to my neighbors. I worry that if I get paid too much for my excess then I’ll be underpaying for the distribution system and shifting its costs onto other people, including some who are poorer than me.

And I know that the economics of electricity are complex. Whole books are devoted to power pricing. The true cost and value of each kilowatt-hour of electricity depends on when and where it is generated, how far ahead its production is promised, and other factors. This complexity preordains that any practical pricing rule for ordinary consumers will be simplistic. Sometimes we’ll pay or get paid more for our electricity than it’s worth, sometimes less. Both can happen in the same day.

The rough-and-ready compromise that DC chose, like many states, is to pay the wholesale or “generation” rate for each year’s net excess. Under this arrangement, if you merely reduce your annual usage by installing panels, it’s the same as if you get a more efficient refrigerator: your bills fall according to the retail price. But if you go farther and generate a surplus for the year, you get paid for it at a somewhat lower price. That works for me. In DC, wholesale and retail prices don’t differ that much: roughly 7¢ versus 9¢ per kilowatt-hour. Meanwhile, DC solar owners get an additional (and extraordinary) 40¢ per kilowatt-hour through solar renewable energy credits, and we receive that for all the power we’re estimated to generate, not just the excess we push into the grid.

The gears of the regulatory machine ground slowly. But they also ground surely. The consensus proposal bubbled up to the actual three-person Public Service Commission. On August 5, over video chat, the PSC approved the rule change in a matter of seconds. It went into effect a couple weeks later, upon publication in the DC Register.

DC’s net metering limit now stands at 120% of usage. It will go to 140% on January 1, 2021, then to 200% over three more years. The rule allows the PSC to pause the increases if there are signs of trouble, like a local transformer being overloaded by all the excess power generated nearby. But I doubt that will happen. Maryland has been at 200% for years without difficulties. And Pepco already had the power to block any net-metered installations it thinks could cause such problems.

And it’s not like most customers can jump right to 200%. Roofs are only so big. However, as solar panels continue to gain efficiency, more DC residents should be able to take advantage of the rising legal ceiling when they put solar on their roofs.

Can you do what I did?

People elsewhere may be able to copy this success in DC. As I mentioned, I found 18 states with “intended primarily” laws like DC’s: AR, CA, CT, HI, IL, KS, LA, MD, MO, MT, NV, NH, OH, SC, UT, VT, VA, and WY. Maybe I missed some. To my knowledge, most have not formalized a number such as 100% or 200%. It appears that in some states, only absolute limits such as 10 kilowatts apply, while in others, utilities have filled the regulatory void by imposing 100%.

That said, I am keenly aware that if my utility had chosen to fight this change, the story could have ended differently. The regulator might well have deferred to whatever rationales or rationalizations the utility invoked. Victory might then have required bare-knuckled political combat, which I think I’m not suited for. And the activists who are good at it might have reckoned, with limited resources and many potential ways to make change, that raising the net metering ceiling was not the hill to die on (as actually was the case in DC).

Thus while I’m glad that my drive to analyze—to study history, gather data, hear out arguments, and formulate my own—did the trick in this case, I appreciate that more combative and less nuanced advocacy is often what is needed to win the day. There is strength in a diversity of approaches.

That brings me back to the question: Can you do what I did? The answer depends not only on the characteristics of your state. It also depends on the characteristics of you. I’ve long believed that whatever is most unusual and extreme about a person—ability to listen, capacity for vision, etc.—is both a strength and a weakness, depending on context. The art and the challenge of life is finding a station where what makes you different is more often a strength. For me, that happened here: I was able to contribute by being who I am.

If you like to investigate, analyze, and explain then I think you too can contribute to change in your community in the way I did, by participating as a citizen in a public process. You can focus on net metering limits or on something else, such as infrastructure for bicycling or electric vehicles. Voting is not the only way to participate in and vivify democratic governance.

Bonus facts

You can stop reading now. But if you’re a solar policy geek like me, maybe you’d like to hear more of what I learned. Here are some themes I skipped in the story I just told.

How I skirted my utility’s 100% limit even as I fought it. Sharing my story in my local solar email group brought a side benefit: an experienced installer explained how I could stay within Pepco’s 100% limit on paper while exceeding it in practice. The key is that Pepco looks at the capacity not of one’s panels, but one’s inverter, the thing that converts DC to AC, because that is what connects to Pepco’s distribution system. So I ended up getting a 6 kW inverter and 7.92 kW of panels. Oversizing the panels cost me less power than you might think. True, on perfect solar days, the panels produce more than the inverter can handle, so I guess it shuts some of them down. The graph of my output then looks like the flat-topped Devil’s Tower in Close Encounters (check out June 12, 2020 here). But in the hours that are not ideal—such as when the sun is low enough that my chimney is casting some panels into shadow—my inverter is more efficient than a larger one would be because it is operating closer to its design capacity. A 6 kW inverter is more efficient than a 7.6 kW inverter when receiving 5 kW. Those efficiency gains are small per hour, but such hours are common. To persuade my installer to go along with this scheme, I digitized a SolarEdge inverter performance curve, downloaded several years of 5-minute, panel-level production data for my old Enphase-based system, and crunched the numbers. Sure enough, my family is now at about 120%.

Federal tax credit. The federal government lets you claim a credit against income tax for residential solar investment, currently 26% of the cost. On the one hand, that’s great for solar. On the other hand, in 2013, the IRS issued guidance that the credit can only be claimed “for the portion of the solar electric property expenditure that relates to the electricity generated for use in the taxpayer’s home”—that is, the part that gets you to 100%. On the third hand, “the taxpayer may be able to claim the § 48 credit for a portion of the solar electric property expenditure if the requirements of § 48 are satisfied.” The “§ 48” refers to the credit for commercial rather than residential solar investment. I’m NOT A LAWYER OR TAX EXPERT, but on my reading, the requirements for the residential and commercial credits are essentially identical. And declaring a solar business on the side for income tax purposes isn’t as hard as it might sound. For example, reporting self-employment (contract) income on your personal income taxes doesn’t require formally starting a business.

FERC jurisdiction and “sale for resale.” Until DC’s rule change a few weeks ago, DC residents were not entitled to cash payment for any excess power they generated. Credit balances could accumulate on our bills, but Pepco and the other power suppliers were not obliged to cash out these balances by mailing us checks. Many sent checks anyway. But as a matter of law, the price guaranteed to the customer was zero.

That price did not come about by chance. Back in 2001, the Federal Energy Regulatory Commission (FERC) issued an order on a matter involving the MidAmerican Energy Company. FERC wrote, “There may be, over the course of the billing period, either a net sale from the individual to the utility, or a net purchase by the individual from the utility. When there is a net sale to a utility, and the individual’s generation is not a [Qualifying Facility], the individual would need to comply with the requirements of the Federal Power Act.” (A Qualifying Facility is a construct under Federal law meant to embrace non-utility generators, but conceived before residential solar was common.) The FERC quote can be read to say that if I run a surplus even in one month, I fall under federal rather than state regulation. The rationale is that I would then pushing power into the regional grid, and that’s interstate commerce, which the Constitution prevents any state from regulating.

This precedent is the reason why the PSC, when it was implementing DC’s net metering law back in 2003, was persuaded not to require cash payment for net-metering customers’ surpluses. In the legal language of electricity regulation, no “sale for resale” would then occur. Net-metering customers producing excess would just be donating it to the grid, and I guess there’s no Constitutional ban on state regulation of interstate charity(!). This is also one reason why, when this question again arose in our working group of how to pay for excess production, I did not push hard for a better deal for people who produce more than they consume.

As it happened, after DC decided in 2005 not to guarantee payment for excesses, many states did guarantee such payment, usually at a price lower than retail. So when we in DC revisited the issue in 2019, the dominant thinking seemed to be that we’d be safe if we stayed with the pack. We settled on pricing excesses at the wholesale and guaranteeing cash-out.

Events soon transpired to validate that consensus, and in dramatic fashion.

For years, some experts, notably David Raskin, have argued forcefully that FERC must make good on the logic of MidAmerican: it must wrest control of rooftop solar from the states. Others disagree (thisthis). The issue came to a head this spring when the opaquely funded New England Ratepayers Association (NERA) filed a fast-track petition pressing FERC to assert control over net metering. The petition aroused a storm of opposition, in the form of thousands of comments from citizens, state agencies, members of Congress, and environmental groups. Maybe you heard about it. On July 16, FERC dismissed the petition, arguing that it was not obliged to deal with the ambiguity now, especially since NERA had not identified anyone who was harmed by the status quo. As a matter of substance, the issue is not resolved. As a matter of politics, maybe it is. I’d guess that FERC will never want to enter the business of regulating millions of cantankerous solar owners like me.

JOIN THE MOVEMENT Sign up for updates from Vote Solar on solar progress and energy justice across the U.S.