This group brings together the best thinkers on energy and climate. Join us for smart, insightful posts and conversations about where the energy industry is and where it is going.


Net Metering for Rooftop Solar: How to Fix the Problems

Full Spectrum: Energy Analysis and Commentary with Jesse Jenkins

Summary: Net metering policies are effective at supporting solar power adoption but can threaten the financial stability of electricity distribution companies and result in cross-subsidies between electricity users once solar penetration grows. The solution is to align network charges with the real drivers of network costs.

A widely adopted method to encourage distributed solar photovoltaics (PV) may cause real problems once solar adoption reaches high levels, according to a new paper by researchers from the Instituto de Investigación Tecnológica (IIT) at the Universidad Pontificia Comillas in Madrid and the Delft University of Technology (TU Delft).

Under “net metering” policies adopted across most of the United States, Spain, and other jurisdictions, owners of solar can offset their charges for electricity consumption by feeding excess energy back into the grid. In some cases, net metering can reduce electricity bills to close to zero.

Net metering has proven effective at incentivizing solar adoption, but as solar penetration rises, these policies also give rise to new problems.

Electricity distribution companies typically recover most of their costs through volumetric charges per kilowatt-hour of energy delivered across their networks. Yet while charges are based on energy delivered, most distribution network costs arise from fixed investments in wires, transformers, and other equipment sized to meet peak demands. To be clear, this mismatch between (mostly) fixed costs and (mostly) volumetric charges predates solar adoption and can cause inefficiencies and cross-subsidies on its own. Yet when volumetric network charges are combined with net-metering policies, the results can threaten the cost-recovery of distribution utilities and give rise to cross-subsidies between network users. 

According to the paper, a typical Spanish residence with a 2 kilowatt solar system could offset 69% of their volumetric charges if consumption and production are netted out daily, 77% if credits can be saved across a month, and 99% for a full year rolling credit. If solar penetration rises to 20%, that could threaten 8-20% of a distribution company’s revenue. If utilities raise rates to cover these shortfalls, customers without solar end up paying more while solar owners continue to offset most of their bills through net-metering, resulting in cross-subsidies between electricity users.

The solution, according to the researchers, is to align network charges with the real drivers of network costs. That would include a blend of capacity charges based on a network user’s contracted or measured peak in electricity consumption and production and smaller volumetric charges for electricity consumed from or injected into the grid. To support solar PV adoption, implicit net-metering subsidies should be replaced by explicit supports for solar production or investment.

With solar penetration growing rapidly in many countries, the time is now to fix these issues and keep both solar growth and utility revenues healthy.   

Publication: “The economic effect of electricity net-metering with solar PV: Consequences for network cost recovery, cross subsidies and policy objectives,” Energy Policy 75 (2014): 244-254.

Authors: Cherrelle Eid is pursuing an Erasmus Mundus Joint Ph.D. degree in Sustainable Energy Technologies and Strategies and is a visiting PhD researcher at the Institute for Research in Technology (IIT). Dr. Javier Reneses Guillén is Associate Research Professor at IIT and Dr. Pablo Frías Marína is a Professor of Electrical Engineering, both at the Universidad Pontificia Comillas in Madrid. Rudi Hakvoortis Associate Professor at the faculty Technology, Policy and Management of Delft University of Technology.

Note: This is article is part of an ongoing series of concise summaries of interesting and important conclusions from new research and peer-reviewed journal articles. This series at Full Spectrum is written in partnership with Observatorio de las Ideasa Spanish-language publication which finds and summarizes important, cutting-edge ideas for policy makers, business leaders, and others on key topics like energy, health care, economics, and more.

Conversation Starters:

  • As solar penetrations rise, when should regulators prioritize a “fix” to net metering?
  • What would be the best way to incentivize solar while preserving the financial viability of network utilities?
  • Can we avoid a solar vs utilities war? Or with new competition, is this a fight we can’t avoid?
  • Solar is the first distributed energy resource reaching significant scales. But batteries, microgenerators, fuel cells, active demand response, and electric vehicles are gaining ground as well. How do we adapt regulation of utilities to facilitate adoption of these distributed resources and encourage their efficient siting and use?
Jesse Jenkins's picture

Thank Jesse for the Post!

Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.


Bas Gresnigt's picture
Bas Gresnigt on Jan 20, 2015 10:36 am GMT

… consider such grid-defections toward fossil fuel to be quite environmentally disgracefu…

The customer replaces his 110% efficient boiler with the 140% efficient micro-CHP. Good evironmental improvement.

It becomes a great improvement if you consider that the utility uses a power plant with only 65% efficiency, and that 5% is lost in the grid. So the netto efficiency of the utility is only 60%.

So government should stimulate these micro-CHP developments.

~30% of Dutch electricity is produced using customer CHP’s. No problems with reliability, though we are with ~25min/a out less than the Germans. But those have more distributed generation.

Bas Gresnigt's picture
Bas Gresnigt on Jan 20, 2015 10:39 am GMT

So our Dutch national energy plan excludes nuclear.

Engineer- Poet's picture
Engineer- Poet on Jan 21, 2015 12:53 am GMT

There are ~25 countries which generate >80% of their electricity via renewable and do not use nuclear.

Give at least 3 examples.  Rank them by fraction of wind and solar generation, and list their fraction of wind and biomass-fired generation.

There is only one country, France, which generates >80% ‘carbon free’ using nuclear in addition to wind, etc.

For France, you aren’t honest if you don’t put hydro in second place and everything else as the “etc.”.  Not that I ever expect you to rise to that standard.  Every interaction with you is an exercise in finding and calling out deliberate falsehoods.  That is why you were asked to leave Atomic Insights.

France is busy to implement its new law, which brings the share of nuclear down, from ~70% towards 50% in 2025.

It’s doable, with enough coal.  The Socialist and Green parties should be international pariahs for embarking on a course that’s so damaging to the earth’s climate, and domestic pariahs for going on a course which will result in hectokilodeaths if not megadeaths in France.

China is expanding wind and solar each much faster than nuclear.

It is very easy to expand intermittent generation at low levels of penetration.  I note that every time I ask you to name me a grid, anywhere, that has wind and solar at fractions on the order of France’s penetration of nuclear, you dodge by referring to “renewables” which include hydro.  No one questions whether a grid can run completely on hydro; e.g. Quebec has a sufficiently large hydro resource to do it quite nicely.  We are questioning whether wind and solar, which are the only renewables which can be expanded readily, can run a grid.  Your evasions prove you don’t believe your own claims.

Wind [in China] already bypassed nuclear.

And it is still not expanding fast enough to stop the expansion of coal.  Ironically, what is likely to stop the expansion of coal mining in China is the lack of water.  Wind can expand until it reaches 10% or so of generation, many times the 1.5% it provided in 2011 or the 2% or so last year.  But to replace coal, non-combustion generation needs to reach 80% or more.  No source with a sub-20% capacity factor like Chinese wind can possibly do that.

Wikipedia has some interesting quotes on these matters:

“China will need to add a substantial amount of coal-fired power capacity by 2020 in line with its expanding economy, and the idea is to bring some of the capacity earlier than necessary in order to facilitate the wind-power transmission,” according to Shi Pengfei, vice president of the Chinese Wind Power Association.[35] Shi is also concerned about the high cost of wind power, which makes the industry dependent on the government’s willingness to subsidize renewable power. “It isn’t that wind power is showing signs of over-heating. It has already overheated.”[36]

So not only has Chinese wind been expanded too fast, it needs coal to be viable.  This makes three renewable “success” stories you’ve promoted that are actually relying on coal, Bas.  That’s what your “renewables” are the way to.

Bas Gresnigt's picture
Bas Gresnigt on Jan 21, 2015 1:37 pm GMT

Give at least 3 examples
Check the wikipedia page than you have them all, incl. some details!

France: Sorry I didn’t check the share of hydro. I remembered that they embarked on a great wind program, which initially was denounced by Brussels, so they had to revise.
Found Wind remarkable as France is well suited for solar compared to us (NL) or Germany.

a grid …  wind and solar … on the order of France’s penetration of nuclear
Wind alone in Denmark is now ~40%, will pass 50% in 2020, and increase further. So you can assume that Danish wind will surpass nuclear in France, which would be 50% in 2025.

China: has enough hydro to compensate the variability of wind and sun.

Joris van Dorp's picture
Joris van Dorp on Jan 22, 2015 10:42 am GMT

The only thing I’d note: not every thread here has to turn into a chat about nuclear power’s relative safety or cost!!”

I agree, but clearly it is Gresnigt who is forcing everybody to respond to his constant repetition of anti-nuclear nonsense that was debunked multiple times already on TEC in other threads.

If you (like me) want to have original, constructive, informative discussion, then I believe you must consider banning Gresnigt, or at least giving him a warning about his constant deliberate (mis)use of flawed, unscientific sources and his constant repetition of statements and viewpoint that are already debunked at length, multiple times. Gresnigt is causing confusion and misunderstanding with almost every comment he makes, and I think it is quite obvious that this is exactly his intention! People like myself have been on Gresnigt’s (and his alternate identities) tail for years, and we will keep on his tail because his propaganda (particularly his deadly fear-mongering about radiation) is known to seriously injure people through causing morbidly radiophobic depression, suicide and unnecessary abortions.

Joris van Dorp's picture
Joris van Dorp on Jan 22, 2015 12:06 pm GMT

While this WHO report predicts increasd cancer rates (up to 7%) for children in the concerned area despite speedy evacuations.”

This is a relative increase in risk only for leukemia (which is a very rare form of cancer).

WHO predicts about a 1% absolute increase in all cancer risk at most, and this only in the most exposed group of the population. This risk is probably too low to be detectable against the baseline cancer risk of about 40% in Japan.

This 1% increased absolute risk is based on conservative assumptions about the actual received radiation dose, and already looks to be about two times too high. So the actual absolute predicted increase in cancer risk in the most exposed group is likely to be no more than 0.5%.

Assuming all of the 300.000 Japanese initially evacuated due to radiation fear are part of ‘the most exposed group’ (which they obviously are not) then the absolute number of attributable cancer cases is at most 1500 people. Since about 50% of people who get cancer also die from cancer, the predicted ultimate number of deaths due to Fukushima according to the methodology employed by the WHO is at most 750.

Since these calculations all rest on application of the controversial LNT hypothesis for cancer risk, the actual number of deaths due to Fukushima could be zero (or even less than zero, due to a possible small reduction of cancer risk due to radiation hormesisof low-dose radiation)

For comparison, in Germany about 3000 people die every year due to air polution from coal power generation.

This shows that the Fukushima incident implies that coal should be replaced by nuclear power if health is the issue. Even if a Fukushima incident occurred every three months, the maximum predicted total number of deaths would be less than the routine deaths caused by coal air pollution in Germany alone!

Bas Gresnigt's picture
Bas Gresnigt on Jan 22, 2015 2:44 pm GMT

… predicted ultimate number of deaths due to Fukushima according to the methodology employed by the WHO is at most 750.

We can agree on that.
So judgements such as that of the IAEA; ‘Fukushima has (neither will have) no death toll due to radiation’,
can be eliminated.

That would bring IAEA regarding Fukushima, more in line with IAEA’s new statement regarding the death toll of Chernobyl:”It is impossible to assess reliably, with any precision, numbers of fatal cancers caused by radiation exposure due to the Chernobyl accident….”.

Though this new statement increases uncertainty, doubt and fear; it avoids having to admit that their previous stated death numbers (<4000) were ridiculous under-estimations.

German scientists disagree with your last paragraph. So for Germany all nuclear out and then coal out. It cannot afford to do both at same time (citing federal min. Gabriel).

Joris van Dorp's picture
Joris van Dorp on Jan 23, 2015 9:30 am GMT

So judgements such as that of the IAEA; ‘Fukushima has (neither will have) no death toll due to radiation’,

can be eliminated.”

No it cannot be eliminated. The fact is that the expected death toll will be between 0 and 750. This specifically includes the event that there will be zero deaths.

The fact is also that the 750 additional deaths (assuming that this prediction is correct, for which there is no scientific evidence whatsoever) will not be detectable, because there will be about 60.000 baseline cancer deaths having nothing to do with the Fukushima incident among this population in the same period.

Furthermore, there is an even more sinister cancer threat at work here: Stress causes cancer. Depression and resulting alcohol abuse causes cancer. Moving from clean rural areas around Fukushima to crowded polluted cities in order to escape ‘radiation’, causes cancer. The fear-mongering propaganda about cancer risk spouted by people like you is itself a cause of cancer. Just like shouting ‘Fire!’ in a movie theater is a criminal offense, I believe your malafide radiation fear-mongering should be a criminal offense as well.

Joris van Dorp's picture
Joris van Dorp on Jan 23, 2015 10:35 am GMT

“And showed the DNA damage that normal NPP’s cause to people living in their surroundings (which damage health and IQ of next generations).”

Sickening nonsense! Again!

Nuclear power plants expose the surrounding public to at most 0.01 mSv per year, which is less than one hundredth of the normal natural background radiation dose.

Even if this minuscule increase in radiation exposure could cause significant DNA damage in humans, it would be impossible to detect, because the natural background radiation in different places where people live varies by more than two orders of magnitude, compared to this 0.01 mSv/year!

If we choose to assume that such a tiny dose of radiation causes significant DNA damage, then we should not fly airplanes, not go sunbathing, not live in elevated areas, not live in rocky areas, never accept nuclear medicine, not eat bananas, etc, etc, all of which cause much more than 0.01 mSv of radiation exposure.

If 0.01 mSv/year of radiation is a real concern – as you imply- then the whole population of Germany should be evacuated to the Netherlands immediately, because the natural background radiation in Germany is 3 mSv compared to only 2 mSv in the Netherlands! This evacuation would provide Germans with more than one hundred times the radiation exposure reduction as compared to moving away from their nuclear power plants!


Bas Gresnigt's picture
Bas Gresnigt on Jan 23, 2015 11:15 am GMT

You realize the Japanese were lucky?

Fukushima winds took 97% of the radiation to the ocean.
Despite that luck and fast action; up to 750 deaths.

Bas Gresnigt's picture
Bas Gresnigt on Jan 23, 2015 2:02 pm GMT

The raw data from the population registers is at slide 16. So check yourself.
With p-values down to 0.0001 little chance that it is by accident. Especially since other research found similar.
So the issue is: What caused it?

Few options:
1- The explanation by Kusmierz (p.27): Escaping fast neutrons*) that interact outside the NPP with 40Ar (Argon) and generate radioactive 41Ar with half life of near 2hours. That is enough to be transported by the wind along the measured distances while radioactive.

2- Unnoticed or accidental escapes, such as those that occurred at Indian Point, Vermont Yankee, and others.
3- Combination.

*) Fast neutrons can easily pass thick (steel) walls. Remember the neutron bomb, designed to kill tank crews without damaging roads, bridges, buildings, etc.

Paul O's picture
Paul O on Jan 23, 2015 4:04 pm GMT

I won’t bother responding to Gresnight (below), he has an agenda.

A greenie as I use the term is no more a slur than the word Hippie is. It is someone who viscerally rejects a solution to Global Warming that doesn’t preponderantly or exclusvely depend on (natural) Solar or Wind Power.

Greenies have been openly hailling the “Death Spiral” of the Grid for months or even years, it is selective amnesia, to say ” The grid is not being ‘destroyed’ . . . that’s just hyperbolic whining.” 

I did miss you suggestion of the use of Nuclear Power, and for that I do apologize. It is not clear then, that the term greenie actually fits you sir.


Joris van Dorp's picture
Joris van Dorp on Jan 23, 2015 4:19 pm GMT

Hagen Scherb is a certifiable kook. His ‘research’ that you point to is sheer nonsense, as detailed in the following document. His fear-mongering drivel isn’t worth the paper it is printed on. This has been pointed out to you many times, by the way. Why do you think you can just ignore that?


Even if we put this aspect aside, the arguments provided by Scherb and Voigt are not convincing and do not provide reasons of concern for public health. Most of the trends identified by their study disappear when larger periods of time are considered, and their dose–effect estimation is based on one observation in one country not reproduced in any of the other 33 European countries. The only clear and progressive increase of sex ratios shown in this study is that in Russia between 1980 and 2000. However, Scherb and Voigt do not provide any scientific proof of a stepwise increase in 1987, and they did not attempt to compare this with similar trends observed in the same period of time in many Asian countries albeit with much higher amplitude.

Using this scant evidence, Scherb and Voigt go far beyond a sole effect on the sex ratio at birth. They make the extraordinary claim that “the internationally established radiation risk concept based on average absorbed dose is in error at three to four orders of magnitude or, more likely, it is conceptually wrong” and that there are at least “one million missing children.” If this were true, there would be dramatic consequences.

For example, the use of ionizing radiations in diagnostic and therapeutic medicine would need to be immediately examined and probably drastically reduced. If the authors’ claim was correct, many effects should already be observable in patients. Therefore, continuing the generally accepted practice of applying the precautionary principle and radiation protection policies recommended by the ICRP is still reasonable and coherent with the current state of good science.


Engineer- Poet's picture
Engineer- Poet on Jan 23, 2015 6:46 pm GMT

750 deaths is a calculation using the Linear No-Threshold (LNT) model.  This model is known to be false and is further known to have been adopted due to corruption of the BEAR I genetics committee and scientific misconduct by its chair and others.

The expected rate of fatalities from the radiation exposures from the Fukushima meltdowns, using models of actual dose-response curves, is zero.

Bas Gresnigt's picture
Bas Gresnigt on Jan 23, 2015 9:16 pm GMT

In my post here you find summary of the original article,
the comment, and
the defense of the authors.

Bas Gresnigt's picture
Bas Gresnigt on Jan 23, 2015 9:25 pm GMT

It is mainstream science. No real studies supporting threshold ideas.

Engineer- Poet's picture
Engineer- Poet on Jan 23, 2015 10:32 pm GMT

LNT is politics, not science.  Read the supplemental material.  (Not that I expect Bas to, but everyone else should have a look at the shenanigans behind the adoption of LNT.)

Bas Gresnigt's picture
Bas Gresnigt on Jan 24, 2015 4:02 pm GMT

No support for threshold. Calabrese only suggests that NAS BEIR acted filthy.
If true, others had protested.

I’ve put the comment here at the new nuclear blog (more appropriate).

Mark Heinicke's picture
Mark Heinicke on Jan 25, 2015 1:40 am GMT

SolarCity:  thumbs down on grid defection.

Can we assume that SolarCity folks know something about the value of grid defection?  Here are some views:

got to that from here:

starting from here:

So fixed fees are a clumsy mechanism.  Agreed.  A still clumsier mechanism is going to home storage; even those who can afford it may–especially without subsidies–come to rue the day.  It costs money, space, and monitoring, that appeal to renewable energy extremists and hobbyists.  The cost of a turnkey installation will rival the costs of the panels themselves.    

Spec Lawyer's picture
Spec Lawyer on Jan 25, 2015 5:49 pm GMT

So I assume that you feel every nuclear power proponent should have to install many weeks of storage at their homes to cover the weeks of downtime for nuclear refueling too, right?

Of course not, you think that only the green energy sources should not share in the advantages of a grid while the existing technologies get to.  *rollseyes*

Spec Lawyer's picture
Spec Lawyer on Jan 25, 2015 6:07 pm GMT

Amazing.  I’m accused of “cooked statistics” when my post doesn’t include a single number in it.  Thou doth protest too much.  I think we can all see where the lack of honesty here resides.  

Engineer- Poet's picture
Engineer- Poet on Jan 26, 2015 3:21 am GMT

So, one-third (and soon, one-quarter) of light-water reactors refuel in spring and fall, when demand is at its minimum.

Who’d need storage to cover an outage of a portion when there’s 2/3 to 3/4 of full capacity to cover for the fraction that’s briefly out of service?

Spec Lawyer's picture
Spec Lawyer on Jan 26, 2015 3:49 am GMT

Solar generates power during the day when demand is highest.  Who’d need storage when solar power is generate duing the daytime when demand for electricity is highest.  And solar PV generates a lot of power on those sunny days when people are all cranking up their air conditioners.


You seem to want to allow the old technologies to benefit from the grid but not allow the new technologies to benefit from the grid.   Seems like unfair and irrational hatred of the new kid on the block.  

Joris van Dorp's picture
Joris van Dorp on Jan 26, 2015 8:58 am GMT

Bas, I’ve read their defense, and it contains more lies and half-truths typical for the pathological anti-nuke pseudoscientific kook community. For instance, Scherb and Voigt state that childhood leukemia is elevated in the vicinity of French nuclear power plants and they reference a study from 2012. However, when I read the particular study they reference, it does not conclude that leukemia is elevated due to the NPP’s, In fact, the study shows that there is no relation between (minute) increase in radiation exposure due to French NPP’s and the incidence of leukemia at all, which is to be expected.

Scherb and Voigt write: “Recently, it has been shown that in France also, childhood leukemia has doubled around nuclear power plants (Sermage-Faure et al. 2012).”

… whereas the Sermage-Faure et al. 2012 study actually states:

Conversely, using the DBGZ resulted in OR and SIR close to one in all of the dose categories. There was no increase in AL incidence over 1990–2001 and over the entire 1990–2007 period. The results suggest a possible excess risk of AL in the close vicinity of French NPPs in 2002–2007. The absence of any association with the DBGZ may indicate that the association is not explained by NPP gaseous discharges.

Scherb and Voigt and pathological liars and a disgrace to the the scientific profession. In my opinion, their baseless antinuclear fear-mongering and lies are equivalent to man-slaughter on a global scale and therefore constitute a crime against humanity. Your support for their ‘research’ sickens me.

Bas Gresnigt's picture
Bas Gresnigt on Jan 26, 2015 9:15 am GMT

Even in present markets these easy to install combined battery-inverters for home solar are already a success in Germany. Management done by the inverter part of the unit.

And expectations are that the costs of this type of combinations will fall much further…

Engineer- Poet's picture
Engineer- Poet on Jan 26, 2015 2:42 pm GMT

Solar generates power during the day when demand is highest.

Except when it’s cloudy, you mean.  And except in places with cold winters (like New York) where total energy demand peaks in winter, when the sun is notably weak and scarce.  (A lot of those cold places are very heavily populated.)

Who’d need storage when solar power is generate duing the daytime when demand for electricity is highest.

You need storage even in sunny Texas and Arizona, because demand peaks in the evening around the dinner hour.  You can address elements like the A/C peak using ice storage to shift demand as far as the wee hours of the previous night (when wind is alleged to peak), but there’s a surprising amount of resistance to this obvious solution.

You seem to want to allow the old technologies to benefit from the grid but not allow the new technologies to benefit from the grid.

If you’re truly a spec lawyer, you should bone up on the specs of the grid.  Its requirements include second-by-second balancing of supply and demand, adequate reactive power for voltage support, and a host of other things.  Half or more of the grid’s costs are fixed, incurred just to be there to supply power when needed.  The “new technologies” currently do little or nothing to actually supply the grid everything it needs, and the fee structures shift the costs they impose onto other consumers.

Seems like unfair and irrational hatred of the new kid on the block.

‘Tain’t unfair or irrational to want everyone to have to play by the same set of rules.  We’d have a very different grid, both generators and loads, if we simply aimed for minimum total cost.  What we probably would not have is a lot of unreliable generators on it, with everything else forced to work around their idiosyncrasies.

Mark Heslep's picture
Mark Heslep on Jan 26, 2015 8:26 pm GMT

“That implies savings for the grid (thinner cables) “

Line amp rating is set by the max load the grid must deliver, not average load, i.e. monthly kWh. Intermittent PV can’t reduce, say, the 7pm maximum load on the line.

Mark Heslep's picture
Mark Heslep on Jan 26, 2015 10:07 pm GMT


There are ~25 countries which generate >80% of their electricity via renewable and do not use nuclear.”

As likely most of the posters on TEC are aware, most all of those countries obtain high “renewable” shares by means of hydro and a little biomass;  quite a few of them are more than 90% total generation by hydro, e.g. the like of Norway, Paraquay (100%), Tajikistan, Mozambique (100%), Zambia (100%), Albania (100%), Congo, Costa Rica, Bhutan (100%), etc.  These hydro resources have largely been in place for years, and they have limits to expansion.   I see no useful point about the future feasibility of new renewable energy, or nuclear power, by drawing attention to these examples.


Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »