In the aftermath of a devastating winter storm, can we take lessons from Texas?

Photo by Pixabay on Pexels.com

It is never fun to see our fellow Americans suffer, whether it’s from pandemic diseases or weather disasters. Our hearts go out to the residents of Texas who suffered without electricity and heat for days, some of them also without safe drinking water, and a few of them even dying from exposure, fires or carbon monoxide poisoning as they tried to keep warm. 

On the other hand, picking apart the preposterous excuses from Texas leaders seeking to avoid responsibility for the fully preventable power outages and the misery that accompanied them—well, that’s another matter. And it’s made so much easier by those leaders’ insistence on trying to score political points instead of admitting that at least some of the blame rests on their shoulders. 

Take Governor Greg Abbott, who went on Fox News to blame liberals for the debacle. Ignoring his state’s failure to plan for climate change and invest in power grid winterization, he told talk show host Sean Hannity the problem was actually the portion of the state’s electricity supply that comes from wind and solar. “This shows how the Green New Deal would be a deadly deal for the United States of America. Our wind and our solar got shut down, and they were collectively more than 10 percent of our power grid, and that thrust Texas into a situation where it was lacking power on a statewide basis.”

No one in Abbott’s echo chamber pointed out that a) solar actually did just fine, b) states like Iowa and South Dakota, with much worse winter weather, rely much more heavily on wind power than Texas does, yet there are no stories about their turbines seizing up and their grids collapsing, and c) if a shortage of ten percent shuts your grid down, you have way more problems than you can blame on the Green New Deal. In fact, the biggest factor in the grid failure was some 28,000 megawatts of coal, nuclear and gas power that went offline, as the Electric Reliability Council of Texas reported

For his part, former Governor Rick Perry preferred swaggering to problem-solving, saying in a blog post, “Texans would be without electricity for longer than three days to keep the federal government out of their business.” This seems to have been written at about the same time Governor Abbott was asking the federal government for disaster relief

And then there was Ted Cruz. I’m not referring to the farce of his skipping out on the post-storm misery to fly to Cancun, then pinning it on his daughters before high-tailing it home to make a show of handing out relief supplies. That incident just reminds us that no matter how deep our divisions, Americans can always find unity in our collective loathing of Ted Cruz.  

No, in this case I want to point to a pair of tweets from Cruz, almost exactly two years apart. February 13, 2019: “Success of TX energy is no accident: it was built over many years on principles of free enterprise & low regulation w more jobs & opportunities as the constant goal. We work to export this recipe for success t more & more states so that all Americans enjoy the same prosperity.” 

And here he is on February 22 of this year, reacting to news that free enterprise and low regulation had produced $5,000 electric bills for some customers in the aftermath of the storm: “This is WRONG. No power company should get a windfall because of a natural disaster, and Texans shouldn’t get hammered by ridiculous rate increases for last week’s energy debacle. State and local regulators should act swiftly to prevent this injustice.”

Luckily for us, lots of other people have been more interested in understanding what happened and preventing it from happening again than in trying to duck blame and score political points. The real story, it turns out, is simple at its core: “low regulation” meant the Texas grid and power providers did not adequately prepare for winter storms that climate change is making worse than they used to be. And because the Texas grid is cut off from the rest of the country (a feature, not a bug, to cowboy politicians), when the crisis hit there was no way to import power from other states that were better prepared.

Let’s take a closer look at what went wrong, how it could have been avoided, and what lessons it offers for the rest of us. 

The setup: an isolated grid with “free enterprise and low regulation”

The grid that serves Texas is uniquely isolated, which also gives it a unique vulnerability. The Electric Reliability Council of Texas serves most of the state, and no other states. Texans are proud of that (or were before this month), because it means there is no role for federal regulators like FERC. It also means that when power ran out, ERCOT couldn’t just import it from parts of the country with a surplus. Of course, states near Texas also suffered in the storm, so there may not have been a lot of surplus power to be had. It is worth noting, though, that the border city of El Paso fared better than the rest of Texas because it is not part of ERCOT but part of a larger regional transmission organization (RTO) serving several southwestern states.

Another feature of ERCOT is the low regulation that Ted Cruz celebrated. ERCOT keeps it simple for power generators. They get paid for the power they produce. Other RTOs have what is called a “capacity market” to reward generating plants just for being available to run when called on, and they penalize participants who fail to perform. ERCOT does neither. With a reserve capacity of only about ten percent and no way to guarantee generators would be available when needed, ERCOT had set itself up for trouble.

If generators had faced penalties for nonperformance, they could have—and almost certainly would have—spent the money needed to prepare their facilities for colder-than-usual weather. Winterization is a normal cost of doing business for a power provider in a northern state, but Texas winters are usually warm enough not to require it. If you won’t be penalized for not winterizing, you have little incentive to do it when you’re competing on cost with other power sellers. 

ERCOT was vulnerable for another reason. Demand for power in Texas is usually higher in summer, with air conditioners running, than it is in the state’s typically mild winters, so ERCOT plans for that. But in cold weather, gas-fired power plants face competition for fuel, when some of the gas supply goes for heating buildings. This month, when gas wells and pipelines also froze up, there simply wasn’t enough fuel to go around. ERCOT’s overreliance on gas proved to be a liability much greater than the smaller amount of renewable energy on the grid. 

The last important feature of the Texas system is retail competition. Electricity customers in ERCOT can choose among dozens of power providers. Some providers keep rates constant; others offer a variable rate that just passes through the wholesale cost of power, with only a small monthly fee added. When wholesale rates are low, the consumer saves money on a plan like that. But regulators didn’t insist on any safeguard to protect customers against the possibility of wholesale prices spiking to astronomical levels due to a power shortage. That’s exactly what happened in the aftermath of this month’s storm. 

That $5,000 power bill Cruz criticized? That’s unfettered free-market supply-and-demand at work. It’s a feature, not a bug. If you don’t like that feature, Senator Cruz, maybe low regulation isn’t for you. Helping consumers avoid power bills in the thousands of dollars would have been easy, but it would have required a little bit of regulation. 

The storm; or how nature takes no interest in political posturing

Well before this storm hit, ERCOT was fully aware of the vulnerabilities of its particular brand of laissez-faire operations. Ten years ago, in the wake of another winter storm, Texas operators were warned of the dire consequences that could ensue if they did not require generators to winterize operations. 

But, they didn’t, and this chart from the U.S. Energy Information Agency shows what happened to generation as a result. Before the storm, you can see natural gas and coal plants running less when high winds produce plenty of cheaper wind power, then cranking up when wind speeds drop. As the week goes on, power supply from natural gas plants increases to meet higher demand from colder weather, while other generation holds steady. Then suddenly you see every category of energy resource except solar drop in output, as critical components of some generating units freeze up and the units fall offline, while fuel supplies also dwindle. Some wind generation falls off, but so does coal, nuclear, and—especially—natural gas, just as they are all needed most.   

The storm was, to be sure, one of the worst winter storms ERCOT had ever faced. And the situation could have been worse. If operators had not proactively cut power to customers, demand in excess of supply would have damaged grid infrastructure so severely that large swaths of the population would have been without power for weeks or months. (Let us now praise faceless bureaucrats, for they just saved Texas.)

So it was bad, and could have been worse. Why didn’t Texas prepare for it, even after being warned? I have one theory. People who cling to simplistic notions that global warming “should” produce only warmer winters have a tiresome habit of pointing to cold weather as evidence that climate change isn’t real, but I think they also take secret comfort in the idea that if the planet is warming, extreme cold weather events will become less common, with less need to prepare for them. If your political philosophy requires you to see regulation as an evil, your own willful misunderstanding of climate science might provide all the excuse you’re looking for not to act. 

Could it happen here? 

Bad weather can happen anywhere, and it’s always safer not to gloat. That said, several features distinguish ERCOT from PJM, and Texas from Virginia. As noted before, PJM has a capacity market that rewards even otherwise-uneconomic generators for hanging around being ready to produce at short notice, and those generators are penalized if they don’t perform when needed. As a result, we are much less likely to see the kind of power shortage and price spikes that Texans experienced. (Not that PJM is without flaws. Its capacity market unnecessarily discriminates against wind and solar, its policies are making the integration of renewable energy harder than it ought to be, and it has incentivized such an oversupply of gas generation that consumers are paying higher prices for the inefficiency. But that’s another story.)  

Virginia also features monopoly power companies rather than retail choice. There is plenty of disagreement as to whether that is good or bad for consumers. The monopoly model requires strong regulation to ensure captive consumers aren’t being overcharged, and are being offered the products they want—like renewable energy. Critics (and I’m among them) have argued that Virginia isn’t doing enough on this front. 

On the other hand, the retail choice model depends on consumers being well informed, and also requires regulators to scrutinize the tactics of power providers and punish the ones who take advantage of unwary consumers. So, ironically, a deregulated electricity market requires strong regulation to protect participants. Strong regulation could have prevented Texas providers from offering residential customers a tariff based on wholesale prices, with risks that residents couldn’t easily understand or mitigate against.   

Texas was also more vulnerable to disruption because power generators were not required to winterize their plants or penalized for not doing so. Sure, a winterized plant would have turned a hefty profit in this storm, but in a more average winter, the extra cost would not have paid off. The option not to winterize isn’t a good one in PJM. As a result, when the power does go out in PJM, the problem is inevitably in the delivery infrastructure, not the generation.

Virginia’s system of vertically-integrated utilities means our utilities own their electric generation as well as the power lines. They can charge customers for building and maintaining those generating facilities, so they have less incentive to skimp on weatherization. That increases the reliability of those facilities. But even if several power plants in Virginia were to fail all at once, we could still draw power from more than 1,200 facilities across PJM, or even from the larger Eastern Interconnection. By design, Texas does not have that option.

One distinction between ERCOT and PJM that doesn’t make a difference, in spite of Governor Abbott’s claims, is the greater percentage of wind in ERCOT than in PJM. Wind actually makes up 23% of generation in ERCOT, more than perhaps Abbott wanted to admit, given that most of it came online under his watch. In PJM, wind makes up only about 3%. If Abbott were correct that wind turbines can’t handle winter weather, that would be a reason for more northern grids like PJM to avoid wind. But of course, Abbott’s claim is political wishful thinking divorced from reality. Wind turbines operate just fine in the much colder winters of Iowa, the Dakotas, Canada—heck, even in the frigid and stormy North Sea, where offshore wind ramps up production in winter

As for solar, you could see from the chart that it was not affected by the cold weather. Texas residents who were lucky enough to have both rooftop solar and batteries spent the aftermath of the storm bragging about never losing power. That’s a compelling argument not just for more solar in the generation mix, but for more distributed generation in particular, including solar microgrids and resilience hubs to help communities weather future storms. 

In the wake of this month’s storm, the independent Electric Power Research Institute (EPRI) analyzed what went wrong and issued recommendations for Texas grid operators. Among the unsurprising recommendations: ERCOT should do better planning for resource adequacy and increase its interconnections to other power systems so it does not have to go it alone. 

I would add one more recommendation: keep your ideology out of it. You can’t deliver reliable power that is also reasonably priced without robust regulation. If leaders refuse to learn from this winter, they’ll simply set up Mother Nature for another opportunity to mess with Texas.   

A version of this article appeared in the Virginia Mercury on February 25, 2021.

Northern Virginia governments look at major renewable energy energy purchase

If the Northern Virginia Regional Commission has its way, local jurisdictions will buy power from a solar or wind farm in the near future. Photo credit Christoffer Reimer.

The Northern Virginia Regional Commission has selected a consultant to assist area localities in a joint procurement of renewable energy. Participating cities and counties will be able to aggregate their demand to get the kind of economies of scale that have allowed corporations like Amazon and Facebook to lower their energy cost by investing in large solar and wind farms.

On November 17, NVRC announced that Customer First Renewables, a national expert in matching large energy users with renewable energy projects, will serve as its technical advisor. Customer First Renewables and NVRC plan to issue a Request for Proposals (RFP) to identify one or more “shovel-ready,” large scale projects to present to NVRC’s thirteen member governments. Those local governments will individually decide whether to participate in the group purchase.

According to a Request for Qualifications that NVRC issued in its search for a consultant, the project(s) to be selected must be greater than 10 megawatts in size and result in new renewable energy capacity added to the grid. Preference will go to projects located in Virginia or in the regional grid that serves Virginia, known as PJM.

NVRC will act as a central contact and facilitator, but it will be up to the participating localities to negotiate a contract. NVRC Director Robert Lazaro said discussions with representatives of area localities indicate the interest is there for a major renewable energy buy like this. Arlington, Alexandria, Manassas Park, and Falls Church are among the jurisdictions most interested, with others possibly joining as they learn more.

“We see this as a way to green the grid, save money, and assist the solar industry in Virginia,” said Lazaro.

Niels Crone, Senior Vice President for Business Development at Customer First Renewables, said his company was excited to be involved in the procurement effort. “We are delighted to work with NVRC to help Northern Virginia jurisdictions get affordable, large-scale renewable energy,” he said.

NVRC is following an ambitious timeline. A project workshop for local government staff is scheduled for December 11, and a Request for Proposals (RFP) is due January 2, 2018.

How does a group purchase work?

Amazon and other large corporations have become major drivers of new wind and solar projects in PJM, including several large solar farms in Virginia. The steadily-tumbling costs of wind and solar make it possible for the companies to green their energy supply while lowering their overall energy costs using innovative financing approaches. Not all of these would work in Virginia, but one that does is the wholesale, or “virtual” power purchase agreement.

A virtual PPA allows a customer to buy and sell energy in the wholesale market, avoiding potential obstacles such as a utility’s monopoly on the retail sale of electricity.

Local governments in Virginia buy electricity from Dominion Energy Virginia at retail under a contract negotiated by the Virginia Energy Purchasing Governmental Association (VEPGA). That contract makes Dominion their only electricity supplier, and Dominion currently does not offer wind or solar as an option. A virtual PPA would not change this relationship; Dominion will continue to supply localities with electricity from its (decidedly un-green) power plants.

A virtual PPA would, however, let participating localities contract for the output of a renewable energy project, with the electricity sold into the wholesale market rather than delivered to the localities. Given the right project, the price for the electricity in the wholesale market could exceed the price paid to the project owner under the PPA, allowing the localities to pocket the difference—indirectly lowering their energy costs. The localities would also receive an additional benefit in the form of renewable energy certificates generated by the projects, demonstrating they have legally “greened” their energy supply.

There is some financial risk involved, since the PPA price is fixed, while wholesale prices fluctuate. Part of Customer First Renewables’ job will be to find the best project economics with the least risk. Corporate buyers, universities and large institutions around the country have used this approach successfully to lower their energy costs and meet their sustainability goals.

As members of the Metropolitan Washington Council of Governments (MWCOG), Northern Virginia localities are committed to reducing greenhouse gas emissions to 80 percent below 2005 levels by 2050, with an interim goal of 20 percent by 2020. MWCOG’s 2017-2020 Regional Climate and Energy Action Plan (available here) also sets a target of meeting 20 percent of the region’s electric consumption from renewable sources by 2020.

As the report notes, however, “There needs to be an immense undertaking to meet the 2020 and 2050 goals.” Here’s hoping Northern Virginia is ready to do its share.

Virginia wind and solar policy, 2015 update

where are the renewables 1

[Note: Although this is a terrific article, it is now a bit dated.  You can find the 2017 update to the Virginia Wind and Solar Policy Guide here.]

The past year has seen a lot of activity on wind and solar in the Old Dominion, and yet Virginia lags further than ever behind neighboring states in installations to date. Why? And more importantly, what can we do about it?

I’ll try to answer these questions as briefly as possible in this third annual update of Virginia renewable energy law and policy. But yes, this is a long post. If you’re the kind of person who only reads executive summaries or prefers the elevator pitch to the full Ted Talk, let me try this:

Virginia’s utility model is built on monopoly control and large, centralized generating systems, and this model does not serve 21st century needs and technologies. The free market solution is to open Virginia’s electricity market to competition and lower the barriers to customer-sited wind and solar generation.

Virginia is further than ever behind

2015 wind and solar table copy

Virginia still has no utility scale wind or solar projects and very little in the way of customer-owned and other distributed generation. The 2015 legislative session improved prospects for solar at the utility scale, but utility interest in wind remains low. Meanwhile, barriers to the rapid adoption of customer-owned generation remain firmly in place.

Virginia utilities won’t sell wind or solar to customers (and they won’t let anyone else do it either)

With one very narrow exception for commercial customers, Virginia residents can’t pick up the phone and call their utility to buy electricity generated by wind and solar farms. Worse, they can’t even buy renewable energy elsewhere.

This wasn’t supposed to happen. Section 56-577(A)(6) of the Virginia code allows utilities to offer “green power” programs, and if they don’t, customers are supposed to be able to go elsewhere for it. (See the section on third-party-owned systems for what happened when one customer tried to go elsewhere.)

Ideally, a utility would use money from voluntary green power programs to build or buy renewable energy for these customers. However, Virginia utilities have not done this, except in very tiny amounts. Instead, utilities pay brokers to buy renewable energy certificates (RECs) on behalf of the participants. Participation by consumers is voluntary. Participants sign up and agree to be billed extra on their power bills for the service. Meanwhile, they still run their homes and businesses on regular “brown” power.

In Dominion’s case, these RECs meet a recognized national standard, and some of them originate with wind turbines, but they primarily represent power produced and consumed out of state, and thus have no effect on the power mix in Virginia. For a fuller discussion of the Dominion Green Power Program, see What’s wrong with Dominion’s Green Power Program.

In the case of Appalachian Power, the RECs come from an 80 MW hydroelectric dam in West Virginia. No wind, and no solar.

The State Corporation Commission ruled that REC-based programs like these do not qualify as selling renewable energy, so under the terms of §56-577(A)(6), customers are permitted to turn to other licensed suppliers of electric energy “to purchase electric energy provided 100 percent from renewable energy.” Unfortunately (and in this English major’s opinion, wrongly), Virginia utilities claim that the statute’s words mean that not only must another licensed supplier provide 100% renewable energy, it must also supply 100% of the customer’s demand. Obviously, the owner of a wind farm or solar facility cannot do that; the customer will need to draw from the grid part of the time. Ergo, say the utilities, a customer cannot go elsewhere. Checkmate!

The SCC may rule on this interpretation some day, but there is still another problem with the statute: under its terms, customers are allowed to turn to other electric suppliers only if their own utility doesn’t offer a qualifying program. So if the SCC sides with the English majors on this one, Dominion could (and surely would) gin up a variation of its Green Power Program consisting of true renewable energy. It would still not have to offer Virginia-based wind and solar—crappy biomass and old hydro would do, so long as it was actual energy “bundled” with the RECs. Nor would it have to offer a competitive price.

Really, the statute doesn’t ask much. It’s astonishing the utilities haven’t taken steps already to close that loophole. But surely they’re ready, and that’s enough to scare off any would-be competitors.

Earlier this year Dominion seemed poised to offer customers a program to sell electricity from solar panels, which would have qualified. Notwithstanding its name, however, the “Dominion Community Solar” program is not an offer to sell electricity generated from solar energy, and seems likely to attract customers only to the extent they are deceived into believing it is something it is not.

For customers to have real energy choice in Virginia, the GA has to change the terms of §56-577(A)(6). Let people buy wind and solar from any willing seller, whether it be their utilities or the private market. Utilities will benefit by customers taking on their job of lowering Virginia’s carbon emissions. Virginians will benefit from cleaner air, new clean energy jobs, and a stronger grid.

Virginia’s Renewable Portfolio Standard (RPS) is a miserable sham

Many advocates focus on an RPS as a vehicle for inducing demand. In Virginia, that’s a mistake. Virginia has only a voluntary RPS, which means utilities have the option of participating but don’t have to. On the other hand, it costs them nothing to do it, because any costs they incur in meeting the goals can be charged to ratepayers. Until a few years ago, utilities even got to collect bonus money as a reward for virtue, until it became clear that there was nothing very virtuous going on.

Merely making our RPS mandatory rather than voluntary would do nothing for wind and solar in Virginia without a complete overhaul. Most important, the statute takes a kitchen-sink approach to what counts as renewable energy, so meeting it requires no new investment and no wind or solar.

The targets are also modest to a fault. Although nominally promising 15% renewables by 2025, the statute sets a 2007 baseline and contains a sleight-of-hand in the definitions section by which the target is applied only to energy not produced by nuclear plants. The combined result is an effective 2025 target of about 7%.

The RPS is as impotent in practice as it is in theory. In the case of Dominion Virginia Power, the RPS has been met largely with old hydro projects built prior to World War II, trash incinerators, and wood burning, plus a small amount of landfill gas and—a Virginia peculiarity—RECs representing R&D rather than electric generation.

There appears to be no appetite in the General Assembly for making the RPS mandatory, and even efforts to improve the voluntary goals have failed in the face of utility opposition. The utilities have offered no arguments why the goals should not be limited to new, high-value, in-state renewable projects, other than that it would cost more to meet them than to buy junk RECs.

But with the GA hostile to a mandatory RPS and too many parties with vested interests in keeping the kitchen-sink approach going, it is hard to imagine our RPS becoming transformed into a useful tool to incentivize wind and solar.

That doesn’t mean there is no role for legislatively-mandated wind and solar. But it will be easier to pass a bill with a simple, straightforward mandate for buying or building a certain number of megawatts than it would be to repair a hopelessly broken RPS.

Customer-owned generation: for most, the only game in town

Given the lack of wind or solar options from utilities, people who want renewable energy generally have to build it themselves. A federal 30% tax credit makes it cost-effective for those with cash or access to low-cost financing. The credit is available until the end of 2016 (when it falls to 10% for commercial but goes away entirely for residential).

This year the GA passed legislation enabling Property Assessed Clean Energy (PACE) loans for commercial customers. This should help bring low-cost financing to energy efficiency and renewable energy projects at the commercial level. That would make it the year’s most helpful piece of legislation from the standpoint of customer-owned generation.

Now that some barriers to residential PACE have been removed at the federal level, we hope the legislature will extend the law to let localities offer PACE loan programs to homeowners in the near future.

Virginia offers no cash incentives or tax credits for wind or solar. The Virginia legislature passed a bill in 2014 that would offer an incentive, initially as a tax credit and then as a grant program, but it did not receive funding, and the same bill, reintroduced in 2015, died in a subcommittee. North Carolina’s tax credit for solar is widely credited with making that state a solar leader, and it could have the same effect here. With solar panel prices continuing their breathtaking descent, utility and commercial-scale solar probably won’t need that kind of help for long, so a modest program of three-to-five years duration would suffice to catalyze the market. Residential solar would benefit from longer-lasting support.

The lack of a true RPS in Virginia means Virginia utilities generally will not buy solar renewable energy certificates (SRECs) from customers. SRECs generated here can sometimes be sold to utilities in other states (as of now only Pennsylvania) or to brokers who sell to voluntary purchasers.

Limits to net metering hamper growth

Section 56-594 of the Virginia code allows utility customers with wind and solar projects to net energy meter. System owners get credit from their utility for surplus electricity that’s fed into the grid at times of high output. That offsets the grid power they draw on when their systems are producing less than they need. Their monthly bills reflect only the net energy they draw from the grid.

If a system produces more than the customer uses in a month, the credits roll over to the next month. However, at the end of the year, the customer will be paid for any excess credits only by entering a power purchase agreement with the utility. This will likely be for a price that represents the utility’s “avoided cost” of about 4.5 cents, rather than the retail rate, which for homeowners is closer to 11 cents. Given the current cost of installing solar, this effectively stops people from installing larger systems than they can use themselves.

Legislation passed in 2015 makes it less likely that new solar owners will have any surplus. At Dominion’s insistence, the definition of “eligible customer-generator” was amended to limit system sizes to no larger than needed to meet the customers demand, based on the previous 12 months of billing history. The SCC is currently writing regulations that should address issues of new construction as well as questions arising from other new language in the law.

This limitation is crazy, no? If customers want to install more clean, renewable energy than they need and sell the surplus electricity into the grid at the wholesale power price, why would you stop them from performing this service to society? And what were Dominion lobbyists thinking, since it is clearly in their company’s interest to buy peak power at a cut-rate price? We can only speculate that the primal fear of customers with solar must be stronger even than the smell of money.

Virginia law also does not allow system owners to share the electricity with other consumers through community net metering or solar gardens. Several bills that would have permitted this were introduced in the 2013 and 2014 sessions but defeated due to utility opposition. Community net metering remains one of the solar industry’s highest priorities as a way to open the market to people who can’t own solar facilities themselves. It would also spur the market for community wind.

In August of this year, Dominion received permission from the SCC to begin a program the company is calling “Dominion Community Solar.” Reading the fine print, however, makes it apparent that participants will not actually buy solar power. They will pay a significant premium on their electric bills to fund construction of a solar installation, but the electricity generated will be sold to other people rather than credited to the participants.

Under a bill introduced by Delegate Randy Minchew (R-Leesburg) and passed in 2013, owners of Virginia farms with more than one electric meter are permitted to attribute the electricity produced by a system that serves one meter (say, on a barn) to other meters on the property (the farmhouse and other outbuildings). This is referred to as “agricultural net metering.” The law took effect July 1, 2014 for investor-owned utilities (Dominion and Appalachian Power) and July 1, 2015 for the cooperatives.

Standby charges hobble the market for larger home systems and electric cars

Dominion Power and Appalachian Power are at the forefront of a national pushback against policies like net metering that facilitate customer-owned generation.

The current system capacity limit for net-metered solar installations is 1 MW for commercial, 20 kW for residential. However, for residential systems between 10 kW and 20 kW, a utility is allowed to apply to the State Corporation Commission to impose a “standby” charge on those customers.

Seizing the opportunity, Dominion won the right to impose a standby charge of up to about $60 per month on these larger systems, eviscerating the market for them just as electric cars were increasing interest in larger systems. (SCC case PUE- 2011-00088.) Legislative efforts to roll back the standby charges were unsuccessful, and more recently, Appalachian Power instituted even more extreme standby charges. (PUE-2014-00026.)

The standby charges supposedly represent the extra costs to the grid for transmission and distribution. In the summer of 2013, in a filing with the SCC (PUE-2012-00064, Virginia Electric and Power Company’s Net Metering Generation Impacts Report), Dominion claimed it could also justify standby charges for its generation costs, and indicated it expected to seek them after a year of operating its Solar Purchase Program (see discussion below). As far as I can tell, it hasn’t carried out this threat yet, and it would likely need legislation to do so.

A bit of good news for residential solar: homeowner association bans on solar are largely a thing of the past

Homeowner association (HOA) bans and restrictions on solar systems have been a problem for residential solar. In the 2014 session, the legislature nullified bans as contrary to public policy. The law contains an exception for bans that are recorded in the land deeds, but this is said to be highly unusual; most bans are simply written into HOA covenants. In April of 2015 the Virginia Attorney issued an opinion letter confirming that unrecorded HOA bans on solar are no longer legal.

Even where HOAs cannot ban solar installations, they can impose “reasonable restrictions concerning the size, place and manner of placement.” This language is undefined. The Maryland-DC-Virginia Solar Energy Industries Association has published a guide for HOAs on this topic.

Third-party ownership of renewable energy facilities could open the market, but Virginia utilities won’t step aside

One of the primary drivers of solar installations in other states has been third-party ownership of the systems, including third-party power purchase agreements (PPAs), under which the customer pays only for the power produced by the system. For customers that pay no taxes, including non-profit entities like churches and colleges, this is especially important because they can’t use the 30% federal tax credit to reduce the cost of the system if they purchase it directly. Under a PPA, the system owner can take the tax credit and pass along the savings in the form of a lower electricity price.

In 2011, when Washington & Lee University attempted to use a PPA to finance a solar array on its campus, Dominion Virginia Power issued cease and desist letters to the university and its Staunton-based solar provider, Secure Futures LLC. Dominion claimed the arrangement violated its monopoly on power sales within its territory, under that same §56-577(A)(6) we previously discussed. Secure Futures and the university thought that even if what was really just a financing arrangement somehow fell afoul of Dominion’s monopoly, surely they were covered by the exception available to customers whose own utilities do not offer 100% renewable energy.

Yet the threat of prolonged and costly litigation was too much. The parties scuttled the PPA contract, though the solar installation was able to proceed using a different financial arrangement.

After a long and very public fight in the legislature and the press, in 2013 Dominion and the solar industry negotiated a compromise that specifically allows customers in Dominion territory to use third-party PPAs to install solar or wind projects under a pilot program capped at 50 MW. Projects must have a minimum size of 50 kW, unless the customer is a tax-exempt entity, in which case there is no minimum. Projects can be as large as 1 MW. The SCC is supposed to review the program every two years beginning in 2015 and has authority to make changes to it.

Appalachian Power and the electric cooperatives declined to participate in the PPA deal-making, so the legal uncertainty about PPAs continues in their territories. In June of this year, Appalachian Power proposed an alternative to PPAs that does not offer anything like a viable solution. The matter is before the SCC. The case is No. PUE-2015-00040. An evidentiary hearing is scheduled for September 29, 2015.

Meanwhile, Secure Futures has developed a third-party-ownership business model that it says works like a PPA for tax purposes but does not include the sale of electricity, and therefore should not trigger a challenge from Appalachian Power or other utilities. Currently Secure Futures is the only solar provider offering this option, which it calls a Customer Self-Generation Agreement.

Tax exemption for third-party owned solar may prove a market driver

In 2014 the General Assembly passed a law exempting solar generating equipment “owned or operated by a business” from state and local taxation for installations up to 20 MW. The law now classifies solar equipment as “pollution abatement equipment.” Note that this applies only to the equipment, not to the buildings or land underlying the installation, so real estate taxes aren’t affected.

The law was a response to a problem that local “machinery and tools” taxes were mostly so high as to make third-party PPAs uneconomic in Virginia. In a state where solar was already on the margin, the tax could be a deal-breaker.

The 20 MW cap was included at the request of the Virginia Municipal League and the Virginia Association of Counties, and it seemed at the time like such a high cap as to be irrelevant. However, with solar now becoming increasingly attractive economically, Virginia’s tax exemption is turning out to be a draw for solar developers. We are told Amazon’s 80 MW solar farm will proceed in four stages, indicating a desire to work around the cap—and suggesting that the tax exemption may have been a factor in the choice of Virginia as the project’s location.

Dominion “Solar Partnership” Program suggests distributed solar might be better left to the private sector

In 2011, the General Assembly passed a law allowing Dominion to build up to 30 MW of solar energy on leased property, such as roof space on a college or commercial establishment. The SCC approved $80 million of spending, to be partially offset by selling the RECs (meaning the solar energy would not be used to meet Virginia’s RPS goals). The program has resulted in several commercial-scale projects on university campuses and corporate buildings. Unfortunately, it has also been plagued by delays and over-spending.

The program was supposed to proceed in two phases, with 10 MW in place by the end of 2013, and another 20 MW by December 31, 2015. However, the program got off to a very slow start. In August of 2014 the company acknowledged it was behind schedule and would likely not achieve more than 13 or 14 MW of the 30 MW authorized before it ran out of money. On May 7, 2015 Dominion filed a notice with the SCC that it needed to extend the phase 2 end date to December 31, 2016, and confirmed that it would install less than 20 MW altogether.

Dominion’s Solar Purchase Program: bad for sellers, bad for buyers, and not popular with anyone

The same legislation that enabled the Community Solar initiative also allowed Dominion to establish “an alternative to net metering” as part of the demonstration program. The alternative turned out to be a buy-all, sell-all deal for up to 3 MW of customer-owned solar. As approved by the SCC, the program allows owners of small solar systems on homes and businesses to sell the power and the associated RECs to Dominion at 15 cents/kWh, while buying regular grid power at retail for their own use. Dominion then sells the power to the Green Power Program at an enormous markup.

I’ve ripped this program from the perspective of the Green Power Program buyers, but the program is also a bad deal for most sellers. Some installers who have looked at it say it’s not worth the hassle given the costs involved and the likelihood that the payments represent taxable income to the homeowner. There is also a possibility that selling the electricity may make homeowners ineligible for the 30% federal tax credit on the purchase of their system. Sellers beware.

And then there’s the problem that selling the solar power means you aren’t powering your home or business with solar—which is the whole point of installing it, right?

Dominion’s Renewable Generation tariff for large users of energy finds no takers; Amazon votes with its feet

Currently renewable energy projects are subject to a size limit of 1 MW. These limitations constrain universities, corporations, data centers, and other large users of energy that might want to run on wind or solar. On top of this, the utilities’ interpretation of Virginia law prohibits a developer from building a wind farm or a solar array and selling the power directly to users under a power purchase agreement.

In 2013, Dominion Power rolled out a Renewable Generation Tariff (PUE-2012-00142) to allow customers to buy larger amounts of renewable power from providers, with the utility acting as a go-between and collecting a monthly administrative fee.

From the start the program appeared flawed, cumbersome and bureaucratic, and as far as we know there have been no takers. Amazon Web Services chose to contract directly with a developer for the 80 MW solar farm it announced this year (avoiding Dominion’s monopoly restrictions by selling the electricity directly into the PJM market).

2015 marks Dominion’s foray into utility-scale solar

Late in 2014, Dominion signaled an interest in building utility-scale solar in Virginia. In 2015, at the utility’s behest, two bills promoted the construction of utility-scale solar by declaring it in the public interest for utilities to build solar energy projects of at least 1 MW, and up to an aggregate of 500 MW. At the solar industry’s urging, the bill was amended to allow utilities the alternative of entering into PPAs for solar power prior to purchasing the generation facilities at a later date, an option with significant tax advantages.

Dominion’s first solar project is expected to be a 20 MW solar farm in Remington, Virginia. The proposal is before the SCC (PUE-2015-00006). Dominion proposes to build and operate the facility itself, which will earn it a return on investment but give up tax advantages that would save money for ratepayers.

On July 17, Dominion issued a Request for Proposals for third party bidders to develop up to 20 MW of additional projects. The RFP came with an absurdly short deadline, surely limiting the number of good responses, but developers are nonetheless hopeful the results will be strong enough to convince Dominion to follow it with a larger request.

2015 will be another year without a wind farm, but there is hope

No Virginia utility is actively moving forward with a wind farm on land. For the past few years, Dominion Power’s website has listed 248 MW of land-based wind in Virginia as under development, without any noticeable progress. There has been a lot of press about the current standoff in Tazewell County, where supervisors are blocking Dominion’s proposed wind farm. Yet Dominion’s advocacy for its project feels perfunctory. The company has signaled it prefers solar, and its 2015 IRP dismisses wind as too costly. On the other hand, Appalachian Power’s IRP suggests an interest in wind as a low-cost renewable resource that could help it meet the Clean Power Plan.

With no utility buyers, Virginia has not been a friendly place for independent wind developers. In previous years a few wind farm proposals made it to the permitting stage before being abandoned, including in Highland County and on Poor Mountain near Roanoke.

As of 2015, however, Apex Clean Energy is in the development stages for a wind farm of up to 80 MW in Botetourt County. No customer has been announced, but the company believes the project can produce electricity at a competitive price.

As for Virginia’s great offshore wind resource, the perception that offshore wind energy will be costly continues to hold back progress. In 2013 Dominion won the federal auction for the right to develop about 2000 MW of offshore wind power, and the lease terms call for the company to file construction plans within five years. The federal government’s timeline leads to wind turbines being built off Virginia Beach around 2020. As I’ve discussed elsewhere, Dominion is something less than committed to seeing the process through. This puts advocates in the legislature and in the business and environmental communities in the odd position of being keener on a development than the developer is.

Meanwhile, however, Dominion is part of a Department of Energy-funded team designing a pilot project of two 6-MW offshore wind test turbines, originally scheduled for installation in 2017. This year Dominion declared it was taking a “step back” when the sole bid for the contract came in way too high. Stakeholders have been meeting this summer to help chart a path forward.

Will a Solar Development Authority help?

One of the MacAuliffe Administration’s initiatives this year was a bill to establish the Virginia Solar Development Authority. The Authority is explicitly tasked with helping utilities find financing for solar projects; there is no similar language about supporting customer-owned solar. The Authority is supposed to identify barriers to solar, but isn’t given any tools to remove them. The Authority has not been given funding. And members have not been named yet. Meanwhile, the clock is ticking on that December 31, 2016 expiration of the 30% federal tax credit.

The Clean Power Plan: better to switch than fight

On August 3, 2015, EPA issued the final rule known as the Clean Power Plan. Under the rule, states with existing fossil-fuel generating plants must develop plans to reduce total carbon pollution from power plants. In Virginia, the task will fall to the Department of Environmental Quality.

While Virginia’s goals under the plan are modest, the rule means the state, utilities and the SCC must for the first time take carbon emissions into account in their planning. The EPA has signaled a strong interest in seeing wind and solar deployed as solutions.

Some legislators have succumbed to partisan pressure to attack the Clean Power Plan, using talking points provided by fossil fuel front groups. Not only does this do a disservice to Virginians already suffering the effects of climate change, it’s bad economic policy. EPA’s analysis shows Virginia is already on track to meet or come close to our Clean Power Plan goals. Wasting time fighting the plan, or mandating that utilities keep outdated coal plants open, makes far less sense than using the plan as a catalyst to begin an efficient and cost-effective energy transition.

The transition need not even happen fast, as EPA’s numbers suggest that all we need to do is keep our total carbon emissions from increasing over time. Energy efficiency has a huge role to play in achieving this, but so would a requirement that utilities meet any increases in electrical demand with wind and solar. Freeing up the private market will go a long way towards achieving that goal. And of course, when customers install solar “behind the meter,” it keeps electric demand from growing.

The Department of Environmental Quality will be holding “listening sessions” this fall to take public comment prior to developing a state implementation plan under the rule.

 

Apex moves forward with Rocky Forge wind farm as the Clean Power Plan makes Virginia utilities look harder at renewables

Wind turbines in the Poconos, Pennsylvania. Photo credit Mitchazenia/Wikimedia Commons.

Wind turbines in the Poconos, Pennsylvania. Photo credit Mitchazenia/Wikimedia Commons.

It had begun to look like no one would ever build a wind farm on land in Virginia. Appalachian Power Company (APCo) hasn’t shown interest since the State Corporation Commission bounced its proposal for West Virginia wind farms several years ago. Just this past November, Dominion Resources let it be known the company saw no future in land-based wind. One after the other, wind development companies put their Virginia plans on hold, citing permitting issues, anti-wind local ordinances, and—especially—a challenging policy environment.

But interest in Virginia wind never went away, and now Charlottesville-based Apex Clean Energy is pushing ahead with plans for up to 25 turbines on a tract of private land in Botetourt County, 30 miles north of Roanoke. Although development is still in the early stages, the company expects construction to take place in 2017, with electricity flowing that same year.

Apex has years of experience developing wind farms across the country, but this would be its first venture in its home state. The timing seems good; the EPA Clean Power Plan will make renewable energy more valuable to utilities and state officials, and wind energy costs have grown more competitive every year. And while previous wind farm proposals in Virginia have run into opposition from landowners and others, Botetourt County officials unanimously passed a wind ordinance that will allow the project to move forward, with public backing that included an endorsement from the Roanoke Group of the Sierra Club.

Yet anyone who has followed the fates of previous wind farm proposals has to wonder whether Apex can succeed where others have failed. With that in mind, I talked with Apex’s Tyson Utt, Director of Development for the Mid-Atlantic, to gage just how likely we are to see turbines up and running two years from now.

Utt explained that the project is still in the design phase, so a lot of the pieces still have to fall into place. Studies are ongoing to determine the optimal size, type and number of turbines. The project could be as large as 80 megawatts (MW), enough to power up to 20,000 homes, and would represent an investment of up to $150 million. A transmission line crosses the site, and Apex is working with Dominion to ensure grid access.

Apex has not lined up a buyer for the electricity at this stage. Utt said options would include a power purchase agreement (PPA) or sale of the completed project to a utility such as Dominion or APCo. Other possibilities include striking a deal with a corporation that wants to buy wind energy, as Apex has done with Ikea in Illinois and Texas.

Recent events suggest the utilities could be persuaded to take a close look. APCo’s 2015 Integrated Resource Plan (IRP) lists wind energy as a low-cost option for complying with the Clean Power Plan. And Dominion, in spite of all-but-dismissing wind in its own IRP, is still pushing aggressively for the right to put turbines on land it owns in Tazewell County.

Apex is not alone in thinking this year could be a turning point for wind energy in our region. Just over the border in eastern North Carolina, the Spanish wind company Iberdrola will hold a groundbreaking ceremony this week on a $600 million, 102-turbine wind farm near Elizabeth City. That project has been in the works since 2011 and was once thought dead after utilities including Dominion and Duke Energy turned down opportunities to buy the power. There has been no word yet on who will buy the power from Iberdrola.*

Making the money work

The wind industry has been buffeted by the stop-start history of the federal Production Tax Credit (PTC). With the credit, the industry boomed. With each expiration, it tanked. Today most observers doubt it will be reauthorized. This isn’t fatal in parts of the country where flat land means low development costs. Wind remains the least-cost energy option in many states. But building wind farms in mountainous areas of the east is a more expensive proposition. (Consider the logistics of hauling hundred-foot-long turbine blades up winding mountain roads.)

So almost my first question to Utt was how he thought Rocky Forge could produce power at a competitive price. Utt acknowledged the challenge posed by the loss of the PTC but insisted that even in Virginia, wind power can be competitive so long as there is some mechanism that levels the playing field with fossil fuels. If it’s not the PTC, he said, perhaps it will be Master Limited Partnerships, which currently offer tax advantages for development of oil and gas but not for wind and solar. Sales of Renewable Energy Certificates will also help bridge the money gap.

With Rocky Forge still in the early stages, and no nearby projects of its own to compare it to, Apex doesn’t yet know where the cost per kilowatt-hour will fall. But bottom line, said Utt, “We think we can be competitive with gas plants.”

These days, of course, solar energy dominates the news, with solar prices tumbling at a breathtaking rate. (Just this month we learned that First Solar Inc. has contracted to sell solar electricity to Nevada Power for 3.87 cents per kilowatt-hour, a new low price record for solar.)

Apex develops solar projects, too, said Utt. But wind and solar “are different,” and both will have roles to play under the Clean Power Plan, which he described as “a game-changer.”

“Millions of dollars in local economic benefit”

Clean energy is popular, but local economic benefits often carry more weight with county officials. Utt said the project will provide “millions of dollars in local economic benefit through tax revenues and local spending on goods and services over the 30 year life of the project.” It will also “create up to 100 full-time equivalent construction jobs and 5 to 10 long-term local operations jobs.”

It surely helps that Apex is itself based in Charlottesville, making it a known quantity. Utt said Apex “has a track record of hiring wind turbine technicians from local wind technician programs similar to the program at nearby Dabney Lancaster. At Dabney Lancaster, several local residents have completed the wind technician program,” but they have to seek jobs in other states.  “We would like to see those jobs stay in Virginia.”

For Utt, the jobs question is personal. “I was born and raised in Virginia and wanted to get into wind, and I had to leave the state,” he told me. “I spend most of my time driving to Maryland or North Carolina. We are a Virginia-based company and want to get this industry going here. We have a hundred-some people in Charlottesville, most of them working on projects in other states. We want this to set a precedent for other projects in the state.”

Birds, bats and neighbors

Public acceptance of wind energy can’t be taken for granted in Virginia, but the Rocky Forge site may be as good as it gets here. Much of the area where the turbines will go has been previously cleared, and the land is privately owned. The nearest home is a mile and a half away, and a high-voltage transmission line already crosses the property. No bald eagle nests have been found within a four-mile buffer area, and Utt said the company has had biologists on site every two weeks to study wildlife issues.

Nonetheless, a handful of opponents showed up at the county supervisors’ meeting, with one speaker reportedly comparing Apex building a wind farm to ISIS taking over the Middle East. (A certain level of anti-wind hysteria seems to be endemic to Roanoke. Just a few years ago the Roanoke Tea Party web site warned that renewable energy was part of a United Nations plot to make us all live sustainably, as un-American a concept as could be imagined.)

More seriously, opponents cite concerns about birds and bats. Studies have shown that wind turbines are a relatively minor cause of bird deaths compared to the other ways we humans kill birds (windows, wires, vehicles, pesticides and letting Kitty out the door), but bat mortality is a real concern in the Appalachian Mountains. Utt said he felt the wind industry has learned a great deal about building turbines in bat areas in recent years. Apex will include mitigation measures in its operating plan, such as shutting down the turbines at low wind speeds and during key migration times.

Apex’s proactive approach to wildlife issues, and its early engagement with local residents going back many months, helped it win over local officials and environmental activists. Dan Crawford, the chair of the Roanoke Group of the Sierra Club, invited Apex employees to give a presentation about the project in early May, and the group ended up endorsing the proposal.

The Sierra Club had supported a previous effort to build a wind farm on Poor Mountain, which stalled in 2012 when developer Invenergy gave up on Virginia. The Sierra Club supports appropriately-sited wind farms as part of America’s transition from fossil fuels to clean energy. Crawford says he is hopeful now that the Apex project will move forward.

“Like a dance floor, someone has to be first. Rocky Forge will open the door for future wind power development in Virginia and the Allegheny Mountains of the Southeast.”


 

*Update: Later on July 13, the buyer was revealed to be Amazon Web Services. Anybody notice a trend?