QSEPs, Rates, and Taxes

Lots of recent discussion on taxes and customer solar energy systems and utility rates.

I am NOT a tax attorney and I do NOT offer tax advice, but I do have an opinion on the issues. Here is some of that opinion.

Alliance for Solar Choice (TASC) recently hired the law firm of Skadden Arps to write a memo outlining potential tax issues associated with Arizona Public Service’s recent proposal to gut net energy metering. Nobody who likes solar likes what APS is proposing, and no Commission should support it.

TASC and Skadden wanted to point out that an improperly structured Feed-In Tariff (FIT) or other Buy-All-Sell-All (BASA) tariff arrangement could cause problems.

Here is the Skadden memo -
TASC Arizona Tax Memo on FITs

My concern is that the law firm committed what philosophers would call a “category error” or a “category mistake” -
Oxford definition here.

The memo correctly points out that if a tariff is structured so that all the solar energy generated on the premises is SOLD to the utility, the tax benefits for the customer may be lost and the payment from the utility may be taxable.

I have raising this issue for years! In fact, I carefully considered how to avoid it when I crafted the Austin Energy Value of Solar Tariff (VOST).

Here is where the lawyers made their mistake, I think: In the background section of their memo, they said that a VOST is a form of solar tariff that works by having the customer sell all their generation to the utility. That is simply not true.

Here is how I explained it to Herman Trabish at GreenTech Media - and here is
the piece he wrote.

Herman had written to me:

“i have a [snip name] presentation from i think last year about the AE VOST that looks an awful lot like buy-all, sell-all. does this mean AE's VOS is taxable?
1.Customer pays for total consumption at applicable residential rate, plus applicable charges & adders as any other residential customer
2. Austin Energy credits customer for solar production at value
calculated using “Value of Solar” (2012 VOS is $0.128/kWh)”

I wrote back:

On Wed, Aug 21, 2013 at 3:23 AM, RabagoEnergy - Karl wrote:


I don't see what you see, exactly the opposite, in fact. It's late. I can't sleep, so here goes.

1. Customer always gets charged or PAYS for consumption. It would still be net metering with customer charges, standby charges, partial service charges etc. Even with mechanical meters, the meter was registering use, just spinning backward faster when sun was shining bright, slower as use approached generation level, forward when use was exceeded. But as memo says, this all matters NOT because purchase is separate and distinct. So don't get distracted by that.

The first test, according to my non-tax lawyer reading of the Skadden memo, is whether more that 80% of generation is used by the customer, as an indicator of the ultimate test of whether the transaction is a SALE to the utility in a separate transaction. Put otherwise, the IRS is reasonable - in its effort to determine whether you should get this tax benefit, they ultimately want to know if this is generation for sale or generation for consumption, and they offer a rule of thumb test at 80% as an indicator.

So ....

2. Note that VOS is a CREDIT (not a payment) for production. AE has been careful in that regard. She did NOT say "you sell all your solar generation to us," or "we buy all your output."

Again, AE is not taking energy output in a buy arrangement. Customer is not selling their power into the market or to the utility, only receiving a bill credit, that never becomes cash, for the quantity of their generation, whether exported or not. There is no title transfer for the energy, no sale arrangement. No power purchase agreement. No term limitation as with FIT arrangements.

These structural indicators, in addition to the plain language of the tariff, below, go to the ultimate test: whether it is generation (primarily) for sale or generation (primarily) for consumption. We tried hard to make it NOT a sale of all output (in fact, writing a sales agreement or FIT with the clauses normally appurtenant would have been more complex and was clearly not our goal - see below) and I think we accomplished that. And we structured the compensation and incentives so as to not encourage systems that would grossly over-generate and trigger the 80% test. Of course, I am not offering a tax opinion, only reading one.

The category error that I perceived in the memo is that just because one could conceive of a rate design that PAID at a calculated VOS level in a SALE OF ALL OUTPUT arrangement, one should characterize ALL rate designs that use a calculated VOS as being BASA arrangements. (I just realized another reason to hate the BASA term - in monopoly or single provider setting, you always "buy all" from the utility/retailer - that doesn't indicate one thing or another. The only thing that is important is that you "sell all" because that's what makes you someone in the business of generating and could lead you to lose one tax benefit.)

In reality, however, there is so far only one VOS rate out there. And it was NOT structured or languaged as a sale under any reasonable interpretation that I can think of. So use of a VOS calculation or name is not BY ITSELF (lawyers might say "ipso facto"), determinative of whether a rate is a sale or is a credit. Because of this logic error, my problem with the memo is the categorization of VOS ipso facto as BASA in the Background section, even if that characterization was limited in the fine print of the footnote. I ascribe no motive to that characterization, but only invite more careful research and some clarification.

Here is a question for the real tax lawyers - if a customer signs a separate contract (regardless of the contract name) to sell even one unit of energy to the utility and that contract transfers title to the energy, etc., what is the tax treatment? Would it still depend on the outcome of 80% test? From my read of the Skadden memo, I would say it does not depend on the 80% test, but rather on the character of the transaction - and the "sale" character would be dispositive. I know what you are going to say, can the tax problem be avoided merely by the name you chose for your transaction? Of course not, that would another error. So, in addition to the name, a reasonable person (and we hope, the IRS) looks at the structure and character of the transaction created by the tariff or rate - Is it for more than 20% of the generator's net output or likely to encourage such excess generation as to exceed that threshold? Does it resemble a credit mechanism or is it more like a sales agreement in terms and language? Does it speak to title, ownership? Does the operational and compensation arrangement, in the end, appear to be intended to be source of additional income for the generation facility owner or is it more about compensation in the form of credit for self-generation to meet on-site consumption needs?

I am leaving aside the question of whether the form of the tariff is irrelevant to the IRS in situations where a net metering customer generates more 1.2x of their consumption level - the "separately metered pole barn with solar on top" example. But I do note that the reason many advocate for aggregated net metering may be to avert this potential problem. The rural customer wants consumption at the barn added to consumption at the house and milking shed before the test of "generation for consumption or generation for sale" is applied. And if I am right, that only bolsters my argument that it is the sale to the utility that is dispositive, not the name of the credit mechanism.

That is, does aggregated net metering cause a problem where solar is connected to just one of several meters on the property, but the output exceeds the load at that one meter all the time? I would say no problem where aggregated net metering is allowed, where all the meters belong to the same contiguous single-owner site, where the total production of the solar system is not SOLD to the utility, and where the total consumption on the site is equal to at least 80% of the total generation. Because all those point to "generation for consumption" not "generation for sale."

But you COULD have a problem if those conditions were missing or not satisfied! The summing of credits done in aggregated net metering is like the summing done in the Austin Energy VOS - that is, as a net billing arrangement.

Want to see the difference?

Here is a link to
a real FIT

Here is the Austin Energy tariff sheet actual language -




This rider applies to any customer receiving residential electric service who owns and operates an on-site solar photovoltaic system with a capacity of 20 kW or less that is interconnected with Austin Energy's electric distribution system.

Monthly Charges

Billable kWh under this rate schedule shall be based on the customer’s total energy consumption during the billing month, including energy delivered by Austin Energy’s electric system and energy consumed from an on-site solar system. All non-kWh-based charges under this rate schedule shall remain unaffected by the application of this rider.

Solar Credit

For each billing month the customer
shall receive a non-refundable credit equal to the metered kWh output of the customer's photovoltaic system, times the current Value-of-Solar Factor plus any carry-over credit from the previous billing month. The Value-of-Solar Factor shall initially be $0.128 per kWh, and shall be administratively adjusted annually, beginning with each year’s January billing month, based upon the marginal cost of displaced energy, avoided capital costs, line loss savings, and environmental benefits.

Any amount of solar credit in excess of the customer's total charges for electric service under the residential rate schedule shall be carried forward and applied to the customer's next electric bill. The customer's carry-over credit, if any, shall be reset to zero in the first billing month of each calendar year.

Disco of the Future?

PEPCO has recently filed another application to increase rates for electric distribution service in the District of Columbia. The Grid 2.0 Working Group and the Washington, D.C. Chapter of the Sierra Club sponsored my appearance in the case as an expert witness.

In the case, I had a chance to take issue with the “maintain and harden” approach proposed by the utility, and propose instead the creation of an independently-run project aimed at a strategic investment plan for optimizing the use of grid-supporting resources. Our case is essentially that the mere repetition of the past is not appropriate to the challenges of the future, and that a holistic, systems-based approach is in order.

Find my testimony at the District of Columbia PSC website,

I would love your feedback!