Clean Power

Published on October 20th, 2014 | by John Farrell


The Future of Solar Economics and Policy

October 20th, 2014 by  

Solar power is at a unique place in history. It’s growing rapidly, its price is falling precipitously. Within the next 10 years, it will compete favorably with utilities for electricity sales, on price, and without subsidies.

Given its rapid ascent, it might seem silly to talk about change.  But the continued expansion of distributed solar power may rely on modifying a bedrock of distributed solar policy.

The Past

The solar past is all about net metering (and generous federal, state and utility incentives), a policy allowing smaller scale arrays (usually 1 megawatt or smaller) to connect to the grid at low cost, and for that solar energy to be credited to the producer’s electric bill as though it were a comparable amount of energy conservation. If a solar array produces 100 kilowatt-hours (kWh) in a month and the customer uses 300 kWh, then the customer’s bill is for the “net,” 200 kWh.

Why Net Metering Is Awesome

  • Typically reduces or eliminates extraneous fees on producing local energy
  • Standardized tariff – same deal for everyone
  • Easy to understand compensation
  • Tax free energy production, because it is “paid” as an energy credit, not cash
  • Customer doesn’t need a battery because net metering is an accounting policy, not an electrical engineering one

Why Net Metering Isn’t Everything

  • In most cases, you can’t produce more than you consume. If you’re a commercial warehouse with roof space for 2 megawatts of solar but very little on-site demand, it’s not economical to fill the roof with solar panels.
  • Compensation was – historically – much less than the actual value of solar to the utility, its customers, and society.
  • The customer may have a perverse incentive to increase electricity consumption if they are producing lots of on-site energy, because the price paid for excess energy is much lower than for energy used to offset on-site use.

The following chart explains how net metering laid the financial foundation for solar PV projects, but that incentives had to make up the difference. It also illustrates how, based on the preliminary estimates of the value of solar from Minnesota‘s new policy, utility compensation for solar energy produced under net metering was likely far less than the actual value of that energy (unless the utility also offered rebates or incentives).

future of solar economics and policy - net metering solar leasing vost.004

The Present

By the end of 2013, the U.S. had installed 13,000 megawatts (MW) of solar PV systems, and net metering contributed to a huge portion. The cost of installing solar had fallen 60% in five years and, in certain parts of the country, the cost of a solar array averaged over 25 years of energy production (called the “levelized cost”) had dropped below the price many customers paid to the utility per kWh. The following chart illustrates, with the levelized cost data for Missouri (about average for the U.S., but much above the cost in sunny areas like California or Colorado).

installed solar capacity and cost u.s..001

The growth in solar power and falling prices have led to a new dynamic in solar economics. For the first time in many places, solar electricity from the rooftop is cheaper than utility-provided power – without subsidies!  And in particularly sunny places, the levelized cost of solar may even be below the “value of solar,” meaning that solar energy producers (if paid this value) could make a return on investment just on these merits.  This “present” phenomenon will take place in different regions of the country at different times, but will happen everywhere within 5-7 years.

future of solar economics and policy - net metering solar leasing vost.006

What’s important to note is that there’s a convergence: the retail electricity price, the value of solar, and the levelized cost of energy from solar panels are all relatively close.

It also means that by historical accident, net metering rates now and in the near future are very close to what solar energy is worth to the grid. And, it may mean that no incentives (local, state or federal) are needed to finance a solar project.

The Future

With a repeated nod toward the success of net metering and other incentives at building a growing solar market (at least in some regions of the U.S.), the future portends significant political problems if the policy prescription remains unchanged. The three key economic drivers of solar (the net metering rate, the value of solar, and the cost of solar) that have recently converged will begin to diverge rapidly. The following graphic illustrates the phenomenon that first strikes states with high electricity prices and abundant sunshine (already), but eventually spreads everywhere (in the next 10-15 years):

future of solar economics and policy - net metering solar leasing vost.007

What does this divergence mean? Net metering will become increasingly lucrative for solar energy producers. For example, if Germany used net metering instead of a feed-in tariff policy, customers with solar arrays would be paid $0.30 per kWh for solar energy that cost less than $0.13 per kWh to produce, a 130% profit margin!

Big profit margins for solar producers is a novelty, but it’s bad policy. First, it’s economically inefficient to pay so much more than is necessary to shift the grid to solar energy. Arguably, a 13% profit margin rather than 130% could move enormous numbers of utility customers to local solar energy. Second, when using net energy metering policy and paying a retail energy rate that is higher than the value of solar, it means that electricity customers as a whole are paying more for solar energy than its worth (assuming that the calculated value of solar is accurate). Finally, it’s inequitable. More than 75% of residents don’t own a suitable sunny rooftop for solar energy. If solar producers are getting over-sized profits, it comes out of the pockets of those who can’t afford to go solar.

But the economics are secondary to the political implosion that results from poor policy. Americans overwhelmingly love solar energy because it represents self-reliance, clean energy, and local power. It means monopoly utilities (which they dislike) lose and they win.  It’s a political punch that unites the Sierra Club and Tea Party in Georgia. And it’s gone in a hurry if the evidence suggests that solar is perpetuating an economic system of winners and losers.

Already, utilities are fighting back against net metering and other distributed solar policies on the suggestion that the problems illustrated above are happening now (see the map of states where distributed renewable energy is under fire below).  They aren’t.  Utilities fighting now are fighting for a 20th century model of centralized control and comfortable monopoly profits. But in a decade they’ll have the truth on their side and not just innumerable lawyers and lobbyists.

battlegrounds over net metering and distributed generation 2014-0521

So, what needs to be done differently?

Fixing the Future

No matter the changes, it’s important that the future distributed solar policy embrace the same principle as net metering: democratic access to the means of producing local energy with reasonable and equitable compensation.


  1. The “value of solar” – change the form of compensation for on-site energy production from a per-kWh credit to a credit based on the true value of solar, as Minnesota has legalized. That state’s policy still uses the net metering framework (bill credits, matching production to onsite consumption) but changes the compensation from the retail energy rate to the transparent, calculated value of solar. Presuming the price is accurate, it can be enough to finance solar projects and hold non-participants harmless. (Note: some have suggested that net metering and value of solar could be combined, with retail rate compensation for power offsetting local use and value of solar used for excess production.  I’m not sure that this addresses the economic inefficiency issue mentioned earlier, unless net metering rates were used for power actually used on-site and not just on a net basis…)

A Net Metering and Value of Solar Comparison

The following charts show how revenue from net metering and revenue from value of solar compare based on an actual 2.5 kilowatt (kW) solar array installed in Minneapolis. For a 2.5 kW solar array installed in 2014, the total revenue over 25 years is about $1500 more for net metering. For a solar array installed in 2020, the gap jumps to $5000. If the value of solar rate is an accurate representation of the actual value of solar to the grid, electric customers, and society, then the gap represents – from an economist’s perspective – over-payment to solar energy producers.

Note: while the value of solar rate will change year to year for newly installed projects, it is locked in on a 25-year contract when a 1-MW or smaller project comes online. For more on Minnesota’s value of solar, read this report.

net metering v vos estimate example minnesota 2014


net metering v vos estimate example minnesota 2020

  1. A feed-in tariff or CLEAN program – go a step further than Minnesota (and join Vermont, Hawaii, and various utilities like Consumers Energy in MI, Palo Alto in CA, and others) and use a policy that completely separates energy consumption from production. Set a price to pay for solar energy that grows the market or based on its value (both have been tried).  Producers still buy all their energy from the utility, but they get paid for all their production (and likely pay taxes on it – after all, it’s income).


  1. Something else entirely. Maybe 10 years from now battery storage is so cheap that everyone who has solar has one, and only excess power generation is sold to the grid at the value of solar. Or perhaps there’s community shared storage.

None of these ideas is the perfect fit for every state or every utility. And there are no doubt two other good policy concepts for each one I’ve listed here. But the future of solar energy will require a different approach than the past. And we’d best start thinking about what that future will be.

[slideshare id=33339479&doc=futureofsolarpolicyilsr2014-0409-140409160032-phpapp01]

The Future of Solar Economics and Policy from John Farrell


This article originally posted at For timely updates, follow John Farrell on Twitter or get the Democratic Energy weekly update.

Check out our new 93-page EV report, based on over 2,000 surveys collected from EV drivers in 49 of 50 US states, 26 European countries, and 9 Canadian provinces.

Tags: , , , , , , , , ,

About the Author

directs the Democratic Energy program at ILSR and he focuses on energy policy developments that best expand the benefits of local ownership and dispersed generation of renewable energy. His seminal paper, Democratizing the Electricity System, describes how to blast the roadblocks to distributed renewable energy generation, and how such small-scale renewable energy projects are the key to the biggest strides in renewable energy development.   Farrell also authored the landmark report Energy Self-Reliant States, which serves as the definitive energy atlas for the United States, detailing the state-by-state renewable electricity generation potential. Farrell regularly provides discussion and analysis of distributed renewable energy policy on his blog, Energy Self-Reliant States (, and articles are regularly syndicated on Grist and Renewable Energy World.   John Farrell can also be found on Twitter @johnffarrell, or at

  • Calamity_Jean

    No, I hadn’t seen the website, but I’ve seen verbal descriptions of the method. It’s sort of like pumped storage, with rocks instead of water. Still wouldn’t work in some of the very flat Great Plains states.

    There’s also this:
    and this:

    Storage will be unnecessary until the fraction of grid power that is supplied by renewables gets to be much larger than it is now in most places. Most of the US is nowhere close to needing storage for renewable electricity.

  • Shiggity

    I’m already thinking about the *next* paradigm shift, solar has already beaten centralized fossil fuel power generation.

  • spec9

    Lots of people in the USA wanted feed-in tariffs for solar PV, but we got net-metering instead which was much more kind to the utilities here who lobbied hard against feed-in tariffs. As a result, the solar biz grew much faster in Europe. And now that net-metering is a great deal for consumers, the utilities are now trying to kill it off. Don’t let them. They made their choice, they need to live with it.

  • Adrian

    Not considered: The decreasing cost of renewables will bend (is bending) the curve of future retail electricity prices. A steady climb of retail and net-metered prices is not at all certain.

    • GCO

      I definitely expect e.g. increasingly cheaper wind to push wholesale prices down; this is already happening in Germany.
      Retail rates include transmission, utilities costs (and profits, where applicable) and various taxes, which may rise or remain significant; e.g. Germany again.

  • Matt

    I do find it interesting that net-metering only becomes a “problem”, when utilities started to pay roof top PV close the the amount it is worth. As long as they paid nothing or less than PV is worth all was good. Note that the net meter “issues” go away if you move everyone to TOD pricing. To get there likely have to split generation, distribution, long transmission. Since all generators need to get less when TOD price is low. Which of course makes 20 year PPAs more complex. But does provide space in the market for storage. Also in local distribution, you charge a fee to the load point not the draw point. Charge less (small, zero, negative?) where power is needed; more where too much power is generated. Now there is no problem with a factory producing much more than it uses. TOD would change during the year (seasonal), and need to be tweaked as time went on (since production/use would change). Same with power load fee.

  • JamesWimberley

    CLEAN tariffs aren’t the same as German FITs, which as the name implies reimburse only the power supplied to the grid. Self-consumption is omitted. So current solar FITs at 40% of retail still provide a substantial incentive, increasingly driven by the high value of self-consumption. One manageable problem is that the scheme creates too strong incentives to install small behind-the-meter storage batteries, less efficient than the larger-scale solutions available to utilities.

    On the main point, I agree with John: net metering is unsustainable politically in the long run. Given the pace of change in solar, this coukd mean in 5 years.

    • Larmion

      Actually, such behind the meter storage would be welcomed by many utilties. Countries with a large share of nuclear power (France and Belgium spring to mind) often had the rather annoying problem of having to reduce output during off-peak hours or use expensive backup during peak hours – both options that reduce margins sharply.

      Any system that reduces the delta between peak and through of the daily demand cycle is a massive boon to utilities as well as consumers. It would have the double effect of making utilities more accepting of solar power as well as increasing the viability of new nuclear, thereby accelerating the uptake of two wonderful sources of low carbon energy.

      • Hans

        Behind the meter storage is inefficient from a technical and environmental point of view. Storage always means an energy loss, and thus should be avoided when possible.

        Simple example: you have a pv system, but you are at work during the day. In the case of a behind the meter storage, your batteries will store as much as possible so you can use it in the evening. But maybe your neighbour is at home and using the clothes drier. If he/she could use your solar electricity production directly it would mean less losses.

        In Germany batteries for self-usage maximisation already raise concerns by the grid operators. These systems operate according to a simple algorithm:
        -As soon as the own production is higher than the self-demand start filling up the battery,
        – if the battery is full dump the overproduction to the grid.
        This causes extra steep ramps in the PV power supply, which are difficult for the grid operator to handle.

        For the best utilisation of PV power the charging and de-charging of the batteries should be serving the entire electricity system, rather than one household. Ideally, this could be done by variable feed-in tariffs that reflect the instantaneous spot market prices.

        However, in the real world we are dealing with constant feed-in tariffs and obstructing utilities. And if behind the meter storage can make PV systems feasible in areas with low feed-in tariffs, I say power to the system owner. When the electricity market changes in a way that would stimulate holistic use of private batteries, some new charging/discharging software can always be retrofitted.

        • Larmion

          Storage is obviously an inefficiency, but so is forcing underutilization of nuclear plants or wind turbines during times of excess supply or (more likely) ramping up fossil power plants during periods of excess demand. And that’s to say nothing of using pumped hydro, the only other currently viable method of large scale storage. That has an efficiency of 70% plus transmission loss as compared to 90%+ efficiency and no transmission loss for local storage.

          The simple reality is that is impossible to utilize multiple direct substitutes with maximum efficiency when one or more of them are variable (a simple consequence of the mathemical impossibility of maximizing multiple variables in a single system).

          That is not in itself a problem, but it is something that has to be factored into the design of a power grid. Countries with a large stock of nuclear power, for example, benefit greatly from home storage because it is uneconomical and unecological to run an existing nuclear power plant at less than 100% capacity (in this case, you optimize for maximum usage of the nuclear assets). Consumption of solar energy should occur at peak times as much as possible, even if the sun doesn’t shine at peak (as tends to happen during the evening peak).

          On the other hand, a country with lots of flexible natural gas or hydro shouldn’t invest in local storage. Why bother with that when your alternatives maximize their profit by running only during times of peak demand?

          Of course, in a world where utilities (and centralized power generation in general) are entirely gone and feed in tariffs are the norm, you are absolutely right. But we are still years away from that and it’s not even clear if that is the most ecological and economical scenario. So until then, the needs of distributed and centralized generation need to be balanced as best as possible. Home storage is one way to achieve that, but other alternatives do of course exist.

          • GCO

            Home storage is ill-suited to address most of the issues you raised, like absorbing excess production elsewhere on the grid (also, transmissions losses would obviously apply here too; not that it matters much anyway).

            What you’re advocating for is larger-scale, utility-managed storage instead.

      • Calamity_Jean

        OTOH, solar coincides with peak demand rather neatly, especially if some of the solar panels face east and west rather than south.

        There’s some increase over the overnight demand in the hour or two before dawn, and a slightly bigger increase in demand from sunset to bedtime that will need to be filled from backup power or storage.

    • spec9

      Net metering is fine. Especially since many places already charge a fixed monthly fee if you do net metering. Right now, I pay $5/month fixed fee. If the utility starts losing money then maybe we can discuss raising it. But for now, things are fine.

  • Michael G

    Interesting issues keep getting raised and never answered.

    If those who have sunny roofs can not only get very low cost energy but sell it to others, isn’t this making the divide between owners and renters greater?

    If the cost of energy rises, isn’t living in a sunny area a benefit, thereby accelerating demographic shifts to the south where water is in shortage?

    Monterey, CA (roughly N-S middle of CA) is on the same parallel as Algiers. From the southern border of VA to Miami and Monterey to San Diego the US is on the same latitudes as much of the Sahara. We could have a very unequal division of energy production.

    It is amazing that Solar is cost effective in Minnesota at all.

    • Larmion

      The first point is valid, but can be dealt with. Take a look at Europa: Germany has a greater per capita solar capacity than neighbours like France or Belgium, yet far lower rates of home ownership. The key is providing a proper incentive to the owners of rental properties.

      You can’t exactly transfer German solutions to the US (for a start, Germans consider debt the root of all evil while Americans lease everything from solar panels to cars), but it does show that it’s not an insurmountable obstacle. The US is the home of financial engineering, put that talent to good use! 😉

      As for the second point: electricity is a relatively minor factor in the overall cost of living. People move in search of well paid jobs and affordable homes and both have been plentiful in the Southwest for some reason. I strongly doubt that something as insignificant as solar power can induce people to move away (and don’t forget that sunnier climates also mean a higher consumption of electricity due to cooling needs, something that can easily shift the balance).

      But most important of all: the South has wonderful solar resources, but coastal areas and much of the Interior have extremely good wind and hydro resources. And so far, very few if any studies predict solar overtaking either anytime soon on cost or installed capacity.

  • Will E

    Today the price of Solar is 6.5 US dollar cents, see auction Georgia, US
    stop talking about the future, or 10 years from now
    My solar rooftop KWH price is 4 cent. TODAY. EVERYDAY.

Back to Top ↑