Published on March 25th, 2014 | by Guest Contributor


Germany Solar Storage Sector Set To Skyrocket

March 25th, 2014 by  

Originally published on RenewEconomy.
By Sophie Vorrath

Sales of residential solar storage systems in Germany are tipped to boom, with new figures projecting roughly 20-fold growth over the next four years.

A study by research outfit EuPD, commissioned by investment body Germany Trade & Invest (GTI), has predicted a nationwide market of 100,000 units by 2018, up from 6,000 last year, citing falling PV prices as the “most significant reason” for this “monumental” commercial growth.

As we wrote last last year, Germany’s “energiewende” has been a big contributor to the 80 per cent fall in the price of solar modules, and the government is now looking at ways to bring down the cost of the next piece in the puzzle of its renewable energy transition – battery storage.

In the first half of 2013, it kicked off a program to finance the introduction of battery storage into homes and small business – a move it has described as absolutely essential for the European nation to successfully achieve, and move beyond, 40 per cent renewable penetration.

Within six months of the program’s launch, 1,900 homes and small businesses had put their hands up for government loans and grants to install battery storage with solar systems at their homes. By November, around €32 million in loans had been allocated and €5 million in grants, about 10 per cent of the sums allocated in the initial phase of the program.

Earlier this month, another EuPD report found that more than two-thirds of German PV installers are now offering energy storage options to their customers, with British and Italian counterparts set to follow suit.

The study also found that Germany had the biggest uptake of household energy storage systems due to the fact that a percentage of storage system costs were paid as a direct subsidy to consumers.

“A battery is the next logical investment for owners of solar power installations and whose systems are coming to the end of their 20 year contract lifetime,” said Tobias Rothacher, senior manager for renewable energies at GTI.

“Most of these systems will still produce electricity even after their 20 year feed-in tariff period.”

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

is many, many people. We publish a number of guest posts from experts in a large variety of fields. This is our contributor account for those special people. :D

  • Jouni Valkonen

    This is good reason why Feed-in-tariff is already outdated method for supporting solar power. Today solar incentives should be directed into maximum utilization of self-consumption.

    • Hans

      Maximum self consumption does not give the best results for the grid, the macro-economy and the environment. A simple example of what can go wrong:

      It is a semi-sunny summer day with a large electricity demand. A PV system owner is away at work. If his/her system is optimised at self-consumption it will store the electricity in a battery to be used in the evening with a 20%-30% loss. For the grid it would have been better if the system would have fed in during the midday demand peak (in Germany in summer the midday demand peak is higher than the evening peak). The 20%-30% storage loss now have to be compensated from another source, probably conventional, thus more pollution.

      The ideal free market solution would be: Conventional power production plants would have to pay for the damage they do (external costs). Renewables would than not have any problem competing on a free market with prices that will vary during the day. Besides political courage this requires power production and transmission to be in separate companies, in order fro utilities not to sabotage the renewables

      • Hans

        I forgot something:
        With this ideal free-market scenario storage would only be used when it is beneficial for the power production-transmission system as a whole.

        • Jouni Valkonen

          You forgot also something that so called “free market” cannot work with renewable energy, because feeding solar electricity into grid will lead into surplus production of electricity when solar is available. Therefore solar producers cannot get any significant money from feeding electricity into grid, because the spot price of electricity plummets the more the more there are solar power produced.

          The problem is that the marginal cost of solar power is zero, so solar really cannot survive at free markets.

          Therefore on theoretical free electricity markets only self-consumption of solar makes sense, because it substitutes the need for grid power.

          Of course electricity markets are not “free”, but utility scale solar producers can do purchasing agreements with grid operators, where grid operators agree to buy solar electricity at fixed price.

          • Hans

            You are taking an extreme case and generalise it. The power production of solar electricity varies over time, but not in the binary full-power-or-nothing kind of way that you seem to think. There will indeed be times when the solar power supply is larger than the demand. In this case it will make sense to store the electricity for self consumption or to sell to the grid later on (for example if you are on holiday).

            However there will be large periods when solar only covers a part of the demand. In that case the market price is determined by the most expensive of the power producers that are needed to cover demand (Merit order). If this last needed supplier is a fossil power plant this price would be fuel costs + external costs.

            So during periods that variable renewables do not cover the whole demand it can make sense to feed into the grid and get some money for 100% of your production, than saving money on buying electricity later on, especially since you will have 20%-30% storage losses. Furthermore, it could be that there is a lot of wind during the evening and market prices during the evening are lower than during the day.

            So in an ideal fee market the best result is obtained by a combination of storage for self use, storage for selling to the market, and selling directly to the grid. In an ideal free-market with ideal market players (with of course both do not exist in reality) the invisible hand of the market will cause the owner of PV systems with storage to behave in such a way that the optimal result for the market as a whole is achieved.

          • Jouni Valkonen

            you, forgot that the marginal cost solar is zero. This means that solar producers will sell their solar electricity always at zero cost on “free markets”, if we have the absurd law of supply and demand for determining prices.

            But this difficult to understand, because people in general do not understand what the term marginal cost means or they do not understand the law of supply and demand.

          • Bob_Wallace

            No, it means that solar producers can sell for close to zero at times of low demand.

            Utility solar has an O&M cost of about 1c/kWh. And owners want to make something on their investment.

            Solar (and wind) can underbid fuel based generation but they won’t sell for zero (unless they are making something from the PTC).

          • Jouni Valkonen

            What I meant was, that pure solar power needs some kind of PPA (Power Purchase Agreement), because the marginal cost of solar is zero. Alternatively solar needs a fixed price that is set by government (i.e. FIT).

            Of course small amounts of solar can do good in the grid, if the market price of electricity is determined by the marginal cost of fuel based peak load energy generators, such as natural gas generators.

          • Hans

            You do not seem to understand how electricity markets work. Let me explain:

            At some point in time there is a certain demand. Utilities will of course first look at the cheapest suppliers to cover this demand. This would be variable renewables with their marginal cost of zero. However, most of the time these will not cover the whole of the demand. So the utilities turn to the next cheapest supplier, for example coal power plants, if this still does not cover demand they will go to the even more expensive power plants, for example oil or gas power plants. And so on until the demand is met.

            But the cheap suppliers are not stupid, they know that the utility will use their supply as long as they are not more expensive as the last needed power plant. So they can raise their prices to this level and still sell their electricity.

            So even with marginal costs of zero renewables can still sell their electricity at a positive price during a large portion of the time.

          • Jouni Valkonen

            The point is that this market scheme works well when there is less than 5 % solar production in the grid. If we aim to produce conservatively about 50 % of total electricity demand from solar, free markets will lead into market failure and market failure will kill any profitability of solar power.

            And of course we are not satisfied with 50 %. We must go for the 100 % renewables ASAP! And if we add wind power into mix, at least it does not make free markets more efficient!

          • Bob_Wallace

            If the market begins to be over saturated with solar then installation rates will fall.

          • Jouni Valkonen

            that is untrue, because we need electricity 24/7.

          • Bob_Wallace

            I don’t understand what you are trying to say.

          • Hans

            Jouni you have to explain where you get the 5% and 50% from, it seems overly pessimistic to me.

            Let me try to establish some common ground for this interesting discussion:

            – The power production of wind turbines and photovoltaics is variable, but not in the on-off sort of way of coal or nuclear power plants. Most of the time* they will produce some fraction of their nominal power.
            -Market prices are determined by merit order. The marginal costs of the last needed supplier will determine the price. If this last supplier is a conventional power plant these should be fuel costs PLUS external cost.
            – Market prices will go to zero if the the demand is completely met by variable renewables.
            -If this zero-price situation happens during short intervals a few times a year it does not automatically mean a PV installation cannot cover its fixed costs. It could be that at other periods a lot of power can be sold at a good price. One should look at the price times delivered power integrated over at least a year to get the complete picture.
            – Zero price situations will happen more often and will last longer with more PV on the grid. However it should be noticed that currently in Germany and Denmark coal and nuclear also contribute to these situations.
            – How often zero price situations occur depends also on how well the demand and production patterns match. If these match poorly zero price situation will happen more often. If they match well, for example in arid regions PV production and demand match very well, they will happen less often. So the market penetration where variable renewables loose their profitability depends very much on the local circumstances, you cannot give a general number.
            Let’s call this the max. profitable market penetration or MPMP.

            All considerations above are without storage, let us consider it now:

            – I think I established that maximum self-usage (your original point was that we should strive for this) does not give the most efficient result from a macro-economic and environmental perspective.
            – Storage can reduce the number and duration of zero price events, by shifting power production to from zero price situations to positive price situations. It can thus make MPMP larger.
            – the marginal costs of battery storage are not zero. With each loading and unloading cycle batteries loose capacity and thus some of their value. This means that at times that when some of the power is delivered from batteries the prices are not zero.
            – Until the MPMP is reached a free-market can work, especially if external costs are taken into account and (hidden) subsidies on conventional and nuclear are removed.
            -Where the value of MPMP depends on many factors. It cannot be determined with a back of the envelope calculation. I think it would be more than enough to fill a PhD. thesis. My gut feeling is that your 5% is way to pessimistic.
            -Beyond the MPMP we would indeed need some alternative mechanism. It would still need variable pricing to reward efficient usage of storage and demand side management. To be honest I am not quit sure how you should make this work. A fixed FiT** does not do this trick.

            *Special consideration for photovoltaics: during the night, i.e. half of the time, they don’t produce anything at all.

            **FiTs have been very useful to stimulate technology development, to bring down costs, and to compensate for hidden subsidies and neglected external costs of the conentional power plants.

          • Jouni Valkonen

            Solar economics works of course with storage, but the point is that if we have cheap grid storage technology (right now we do not have that), it is more affordable to increase self-use of solar power. In Sunny Australia people are already seriously planning to go off-grid:


            My guess is that in 2020’s solar off-grid solutions are the cheapest form of power for households, at least on latitudes that are below Alps.

            Why solar power does prefer self-use is that power transmission costs are rather significant but localized solar production and storage is not significantly more expensive than centralized solar power production and grid storage.

          • Hans

            You do not really react on the points I make. You do not even take the effort to reveal the source of your numbers.

            You just repeat yourself.

            So let me repeat myself again:

            If FiTs are lower than the consumer price maximum self-usage may be optimal for the PV owner. However, on a macro-economic scale it is wasteful. Why store electricity with a 20-30% loss when you neighbour can use it directly without a loss?

            We need a pricing mechanism that rewards the efficient use of storage, transmission and demand side management. This means prices will have to vary over time, and maybe even spatially, depending on supply and demand.

            Battery storage does have marginal cost, as I showed before.

          • Jouni Valkonen

            Apparently you did not read the link that was discussing about 500 household going off-grid. Among 500 households, there are plenty of neighbors.

            But yet again, you are here proposing artificial princing system (i.e. a “Smart Grid”) for solar power that it would make profitable for household to sell electricity to the grid when there is high demand for electricity. But I can assure you that this kind of setup is economically impossible.

            The reason is simply that decentralized solar power + storage is cheaper than centralized solar power + storage + transmission costs. Therefore in solar economy, where ever solar insolation just allows, solar off-grid solutions are significantly cheaper than relying on grid power. Obviously Northern and snowy countries cannot rely on solar power year around.

            There is other reason, that you do not consider enough that solar power is cannibalistic by its nature. When you have good solar yeald, everybody else has too. This is also the reason, why solar panels should be directed into East and West rather than in South. It is more important to optimize the morning and evening solar production. In the noon, there is in any case huge overproduction of solar power.

          • Bob_Wallace

            “The reason is simply that decentralized solar power + storage is cheaper than centralized solar power + storage + transmission costs.”

            I don’t think that’s true in most cases.

            In the US utility scale solar is average half per installed watt what rooftop is averaging. Utility scale will probably always be cheaper simply due to economies of scale. Large projects can buy panels by the shipping container and get competitive bids from suppliers. They have lower labor costs as there is less time lost moving from job to job.

            The same is likely to be true for storage. Utilities will be able to use some forms of storage such as pump-up and flow batteries which won’t be available at end-user level. And, again, purchasing scale will come into play. Utilities will be buying in shipping container quantities as opposed to ‘half a pickup’ quantities.

            Transmission/distribution won’t be a significant cost since the grid is already in place.

          • Jouni Valkonen

            Bob, things are not that simple. E.g. in germany and australia things are very different and the cost difference between generation cost and retail price is HUGE.

            Also you forgot that installing roof-top panels creates local jobs. This means that that they are cost neutral for the local economy as a whole. And therefore it makes sense to reduce installation costs of roof-top panels from taxation, because they are not real costs for the local economy. In western countries there are plenty of people underemployed, so it is good to get them work that supports local economy.

            The difficulty with costs is that some costs are harmful for the local economy where as other costs are beneficial for the local economy.

          • Bob_Wallace

            Jouni, economics of scale work the same in Europe as they do in the US. Those who purchase in the largest volumes get the best price.

            Europe has great renewable resources. But they aren’t equally spread over all countries. Austria has limited offshore wind.

            What is cheaper, with the storage we have, is to use electricity directly from the generator as much as possible. It’s much cheaper 5 cent wind at night than to store 5 cent solar for 20 cents for night use.

          • Hans

            My line about the neighbours was just a simple example to explain the principle. The thing is: the more ‘neighbours’ you include in your system, the less need for expensive storage.

            “But yet again, you are here proposing artificial princing system”

            Here you are setting up a strawman argument. I called for a free market to determine the prices, however with a fair price for externalities. On the other hand you called for FiTs with the value determined by the government.

            “(i.e. a “Smart Grid”)”
            ‘Smart grid’ has become a catch-all phrase with a wide range of meanings. A free market set-up would require some information exchange about prices and units delivered. I would call this a slightly-less-dumb grid.

            “But I can assure you that this kind of setup is economically impossible. ”
            Argument from self-proclaimed authority. Not very strong.

            “The reason is simply that decentralized solar power + storage is cheaper
            than centralized solar power + storage + transmission costs. ”

            You state this as a self-evident truth, but it isn’t. With transmission you need less storage, because both supply and demand are smoothed out. Currently transmission is mostly cheaper than storage, especially since there is already a lot of transmission there. Furthermore, less storage means less losses. You keep ignoring this.

            “Obviously Northern and snowy countries cannot rely on solar power year around.”

            That is why a grid is so useful, you can get wind power, hydropower and conventional during the night or when solar irradiance is low. Having batteries that can store half a year of production is extremely expensive.

            “There is other reason, that you do not consider enough that solar power is cannibalistic by its nature.”

            I addressed this in detail. You seem not to be willing or not to able to understand my arguments. We even don’t disagree that much. Our difference is more quantitative than qualitative. You assume that zero-price events will make solar energy unprofitable even at low penetration levels. I have given arguments why I think that it will not happen so soon, and that storage can increase the penetration level even further. Most importantly I said that we will not be able to determine this achievable penetration level just by a back of the envelope calculation.

          • Jouni Valkonen

            Look at the German situation. There there is about 6 % solar power penetration and there you can already see that when sunny day happens on week-end, the price of solar plummets into zero.

            Imagine what disaster it would be for the price stability if there was 12 % solar penetration! First of all, transmission grid would just fail to handle peak loads and grid would just collapse.

            Only way to get around this is to increase storage, but then again localized storage is not significantly more expensive than centralized storage. At least not with German or Australian transmission costs.

          • Hans

            Jouni, Germany is far away from the ideal free market situation we discussed earlier. External costs are not taken into account*, and the consumer price and the FiTs do not vary over time.

            If external costs where taken into account base load power plants would quickly be priced out of the market.** If you know the German situation so well, you might also know that during the zero price events there was still a lot of base load power on the grid, especially from coal and lignite power plants. Without this polluting base load power the zero price events would probably not have happened at all.

            Because the FiTs and the consumer price are not related to the spot-market price. Hence, there is no intensive to shift demand or supply.

            As a said before: The occurrence of zero price events does not automatically mean renewable power plants cannot be run in a profitable way. You will need to look at the product of price and supplied power, integrated over time. Determining the penetration level where solar power systems loose their profitability cannot thus be determined by the qualitative kind of reasoning that we do here, neither can it be done with a back-of-the envelope calculation.

            Furthermore, you cannot automatically translate the German situation to the rest of the world. In Germany the match between the demand pattern and supply pattern of PV is relatively poor. For example in California this match will be better, leading to much higher penetration levels.

            Short term price stability (in contrast to long term price stability) is not a goal in itself. To the contrary, a variable price could stimulate suppliers to produce when demand is there, and consumers to consume when the supply is there.

            Regarding storage you completely missed my point. It is not about who owns it, or if it is central or de-central. My main point is about how you manage the storage. If you use storage to maximise the self usage of a single household, company or a small cooperative you will need more storage and waste more energy in the loading and unloading cycle than in the case that the storage is used to optimise the grid as a whole.

            I never said transmission is cheap, but storage is just even less cheap. It would be great if storage would be cheaper than transmission, in that case Germany could avoid a lot of cumbersome political processes, the lengthy planning procedures, and slow downs by court cases. But unfortunately we are not at this point (yet).

            Anyway, an optimal grid structure would be some mixture of transmission, central*** and de-central storage.

            I get the impression that you see this issue very black and white. Maybe it is time to open your mind to the nuance and complexities of the issue.

            * there is a dysfunctional European market for CO2 rights, but the price per ton CO2 is only a fraction of the damage done by it. Fine dust and other pollutants are not taken into account at all.
            **Even without a real price for external costs this is already starting to happen. German power producer RWE is making big losses and its shares have lost two thirds of their value.
            *** The most efficient form of storage is pumped-up hydro, which is mostly large scale and thus central.

          • Bob_Wallace

            ” Market prices will go to zero if the the demand is completely met by variable renewables.”

            Let’s expand this a bit. Market prices go to zero or even negative once in a while because demand is low and renewables, with zero fuel cost and production subsidies, can afford to sell at zero or very close to zero. What generally happens is that thermal plants which can’t easily shut down sell their power for zero or pay the grid to take their power in order to force the renewables to curtail.

            In the long run those subsidies for renewables will expire. Thermal plants will fade away. Renewables won’t sell for zero.

          • Bob_Wallace

            It’s even better for renewables.

            What you are describing in “merit order pricing”. The way it works is that when that last bit of electricity is purchased the price paid is called the “settling price”.

            All suppliers, even wind and solar that bid in close to zero, get the settling price, not what they bid. Wind and solar can bid in low in order to be sure to get to sell, then sit back and let the other players raise the price they will get.

            The other common way to sell electricity is with power purchase agreements (PPAs) which are long term sales agreements with a fixed price.

            Over 2011 and 2012 wind PPA in the US averaged 4 cents per kWh. Wind farms obviously are making money at 4c, otherwise they wouldn’t be selling for that price.

            Solar PPAs are starting to hit 5c/kWh.

            (All energy is subsidized.)

          • Hans

            I know the term merit order pricing, I was trying to explain it in layman terms.

  • JamesWimberley

    Germany seems to have changed the funding model from the long-term FIT guarantees that supported solar panels to short-term and capped grants and loans. It’s optimistic to think that a much less generous programme will have equally dramatic effects.

  • Chris Marshalk

    Again, why isn’t this happening in Australia?

  • Graphite Gus

    Got graphite?

Back to Top ↑