Interviwe published in Paristech review
ParisTech Review – Let’s adopt the consumer’s point of view. Is it profitable, today, to invest in a Powerwall?
Vincent Champain – The answer is far from being straightforward. With Frédéric Benqué and Alexis de Kouchkovsky, we performed some calculations for theObservatoire du long terme (Observatory of the long term).
Theoretically, electricity consumers with solar panels may want to buy a PowerWall to avoid using the grid: they will implement a battery in their setup in order to cover the gap between the power generated by solar panels and their own consumption.
In practice, though, it’s hard to achieve an sustainable economic balance. The domestic consumption of a 1,600 ft2 house is about 60 kWh per day. The battery will need to cover the lack of sunshine – let’s say approximately 3 days of consumption. After taking into account losses, this represents 225 kWh or 32 batteries of 7 kWh each. A single 7 kWh battery is barely enough to power a refrigerator during three days!
This type of facility costs at least $170,000, including the cost of the solar panels and installation. To pay off the initial investment of $170,000 over 10 years, you would need to save $20,000 a year, i.e. several times the average electric bill of a house of 1,600 sq2. Even by reducing the safety margin to one day – meaning that the consumer is willing to suffer power shortages many times a year after days of low sunlight – the investment is still far from being profitable. Overall, except in cases where the grid connection is really expensive (for example, on an island) or for mobile use (powering a rave party in the middle of a desert), this solution has therefore no economic interest.
Vincent Schachter – I think one needs to distinguish between Elon Musk’s decision to gain a strategic position in the medium term, and the reality of technology and storage market in the residential market in 2015. The batteries offered by Tesla today are not revolutionary: they combine the lithium-ion technology developed by Panasonic with the assembly and low-cost production expertise of Tesla. They are designed to efficiently power electric vehicles rather than houses. The advertised prices (430 $/kWh for 7 kWh packs and 350$/kWh for the 10 kWh packs) do not include the power electronics nor the installation costs, which could double the invoice for the end customer. Finally, value propositions related to the PowerWall and PowerPack are not ready yet: residential backup (powering a home in the case of network shortages) is, for the time being, cheaper with a diesel generator. And current tariff structures (“Net metering”) reward as much, if not more, those who resell their solar power to the grid than those who store it for self-use.
Despite all, advertised prices are significantly more aggressive than those of competitors and it is clever move to aim for economies of scale for stationary production by using the capacity already planned for mobile batteries (the famous “gigafactory”). Last but not least, when Elon Musk speaks, investors listen: his self-fulfilling position could seriously speed up the adoption of individual electrical storage solutions.
ParisTech Review – Little is said about it, but just like the development of shale gas, the boom of residential storage in the United States is largely subsidized. Is it still very far from profitability?
Vincent Schaechter – Subsidies for batteries have a similar purpose than the announcement of Elon Musk: to prime the pump and bring prices down quickly by taking advantage of economies of scale. Governments that fund this initial effort hope to reap benefits by encouraging a local champion and eventually developing specialization, with the famous reward of green jobs, or simply by supporting the development of renewable technologies. This is what motivated the Global Warming Solutions Act (also called AB 32) or the Self Generation Incentive Program in California, where Tesla is established. At the federal level, the energy policy of Obama’s stimulus package but also the Renewable Energy Investment Tax Credit are both worth mentioning.
While it is true that residential storage is not profitable without subsidies today, technology is evolving quickly and regulatory and tariff structures may follow.
From the technological point of view, Li-ion batteries are quite mature and battery costs continue to drop thanks to economies of scale, mainly due to the increasing volumes of the production of batteries for vehicles. This is the challenge of Tesla. Other promising approaches have been developed in recent years in laboratories by altering the chemistry of electrodes and electrolytes, the architecture and the electronic management of the package: these changes introduce higher densities of mass energy and above all, lower kWh costs than the Li-Ion. Lithium-sulfur, metal-air, lithium-air, magnesium-ion, molten metals… all are new avenues for research. But to overthrow the dominant Li-ion technology, these promising news solutions will need to pass the difficult and costly stage of mass production. Meanwhile, R&D needs funding: this is what public authorities are doing, as well as venture capitalists and some industrialists, in the order of increasing technological maturity.
ParisTech Review – Back to profitability. Can one possibly imagine that in a near future, batteries will replace the power grid as we know it today?
Vincent Schachter – Not necessarily. The figures mentioned above are interesting for the understanding of orders of magnitude, but they deal with an extreme case that isn’t the aim of the industry today. The ultimate value isn’t to disconnect completely from the grid, but only in case a power failure – a real issue in some US states. And especially to save on the electricity bill by storing solar energy for use when it is needed and when it is cheaper than electricity from the grid. It is therefore less about substituting than complementing. And here’s when change of model steps in. The maths become: when does reducing the electricity bill exceed the cost of the storage solution? If the economics don’t work today, they could in a five years time.
Vincent Champain – Two conditions would need to be met for this to work. First of all, batteries need to become the least expensive solution to provide that kind of flexibility. For the time being, gas or hydraulic power plants are far more interesting. Second, this type of storage would need to be implemented in the least expensive way. It will be probably cheaper to implement storage at a district or regional scale, rather than multiply small facilities that are expensive to maintain or implement. In other words, it will maybe be interesting some day to use batteries instead of the grid, but in a shared form, at the scale of a small neighborhood rather than in an individual form, like the PowerWall.
Vincent Schachter – Maybe, but cheaper for whom and when? I don’t trust arguments based on the calculation of a global optimum, they tend to overlook the interplay of actors and the dynamics of adoption. Take the example of computing: on theory, pooling storage and computing capabilities has been the best solution for a long time now. This didn’t prevent the computer revolution in the 80s to trigger the massive production of personal computers. It took 30 years for technology, infrastructure (the Internet) and users to see the growth of cloud computing. Back to energy: to me, it doesn’t seem that the interests of electricity distributors or end consumer will necessarily push towards the theoretical optimum for an entire region or country.
ParisTech Review – Moreover, the equilibrium point will probably not be reached at the same rate everywhere. What will make the difference?
Vincent Champain – It is indeed an essential point, which is frequently misunderstood: solar energy and battery storage are already the best solutions in some defined areas (for example, in isolated and sunny locations) but not everywhere. We already know how to produce solar energy at a very low price (6 cents per kWh) in the desert, and according to Philippe Boisseau (Total), PV is accessible without subsidies in dozens of countries worldwide. But that doesn’t mean that this technology is interesting in the North of France. Thinking in terms of averages and making definitive judgments on technology is a mistake: the question must be examined within a specific context. It depends on both the availability of the resource (wind or sun), the cost of connecting to the grid and the cost of available alternatives. In the medium term, we can hope that the price of carbon and the end of both explicit or implicit subsidies to fossil fuels will also become relevant.
Vincent Schachter – I totally agree, profitability very much depends on the context. However, underlying trends can be seen: for example, the cost of solar electricity will continue to drop (the cost of PV modules has been divided by 5 in 5 years!) and consequently, that of the “balance of system.” The cost of storage technologies will also drop drastically.
Besides, a viable business model is also necessary in order to introduce a product in a given market. Thus, the incremental cost between that of producing solar electricity (including batteries) and the price of grid electricity at different times of the day should enable consumers to reduce their bills. Besides, consumers can choose between storing and reselling to the network: the better the resale tariff, the less interesting the storage. Regulation plays an essential role: it can promote self-consumption and/or storage of solar energy or play against it, by prohibiting or canceling their economic value outright.
ParisTech Review – The network’s costs appear to play an important role. We saw this in Europe with water supply: a decline in consumption often results in a rate increase because fixed costs don’t vary. Let’s imagine that Elon Musk achieves his goal and that thanks to subsidies, a significant proportion of Californians use the PowerWall: what will happen to Californian electricity producers?
Vincent Champain – Elon Musk’s bet is that the cost of “flexibility” will drop, for two main reasons. First, battery prices will fall, although they still need to be divided by two or three for the bet to be successful. Second, some customers will obtain storage capacities (either because they bought an electric car or because they bought a PowerWall) and this capacity offer will help lower prices. If this is the case, the means of production which “consume” flexibility (solar, wind energy) will become more interesting. Conversely, other “producers” of flexibility (advanced gas turbines) will lose competitiveness if new competitors are able to lower their costs. There will be winners and losers, as every time a more efficient technology replaces older ones.
However, we haven’t yet reached the point where this type of changes can happen and economic conditions are still far from being met. But it is a possibility that we cannot totally rule out… Note that this option is often feared by the incumbent players because they will bear, by default, the “stranded costs” – i.e. the losses related to the drop of production capacity by the development of new technologies. But as the reportTransition par l’innovation (“Transition through innovation,” available on http://tti.onews.eu) points out, the tariff systems could improve the distribution of these stranded costs in order to turn incumbents in allies of change, rather than opponents. It would also help avoid a two-tier system where those who can invest in new technology will pay less than the others, who will stay connected to the grid and bear alone these “stranded costs.”
ParisTech Review – Will advances in batteries and the possible large-scale dissemination of the Tesla model promote photovoltaics at the expense, for example, of concentrated solar plants? Will any effects be felt in the choice of R&D and investment?
Vincent Schachter – Today, PV promotes the development of batteries, until one day the opposite will hopefully become true! I find it striking to see how the maturity of a technology, solar PV, accelerates the development of another, storage. Following a dramatic decline since 2010, photovoltaic electricity is now competitive without subsidies in over thirty countries, comparing the simple costs of production (LCOE). The same cannot be said of concentrated solar power plants…
PV is growing rapidly (+40 GW in 2014) and its penetration rate into the grid is increasing constantly. Flexibility (i.e. controlled fluctuations) becomes crucial as a response to the intermittence (uncontrolled fluctuations) of solar and wind power. It’s the defensive stand against decentralized renewable energy, the natural reflex for electricity producers: intermittent electricity on the outskirts of a grid primarily designed for centralized production could end up destabilizing the whole system. And if customers use less electricity from the grid, who will pay for the infrastructure? However, decentralized renewable production can also stabilize the grid or avoid new investment, as long as electricity is usable at the right time and the right place: it’s the positive vision of renewable energy through synergy with flexibility. Storage is one of the pillars of this flexibility, the other being the control of demand: the NEST thermostat or the optimal use of machines in a factory. A necessary condition in order to exploit this flexibility is to increase network intelligence – all the talk about smart grids is becoming a reality!
Many power producers and network operators have taken full measure of the changes ahead, and are thinking about how to support or anticipate them.
ParisTech Review – The most striking part of Tesla’s storytelling is the importance it gives to the idea of individual autonomy. If alternatives to large centralized networks need to be found, is it not time to develop forms of pooling?
Vincent Schachter – Evidently, from an ecological point of view and considering the emissions of greenhouse gases, shared storage is more virtuous than individual storage. But when considering the diffusion of these innovations from an economic point of view, it becomes a completely different story. Early adopter residential consumer buy a Nest thermostat because it’s a cool gadget. Certainly not because of its cost-effectiveness or environmental value. The natural point of entry for Silicon Valley companies are objects or consumer services…
Vincent Champain – The problem of this storytelling, quite similar to Jeremy Rifkin’s, is that it isn’t an economic approach (which aims to develop the most effective solutions, globally) but a more of a “romantic” dream, one that offers a pleasant vision, that only describes one side of the reality – often not the least expensive one. Some of us can afford it, but not those for whom energy costs represent a significant part of their budget. If the general interest, experimentation should be left to those who can afford it, while making choices that ensure the lowest possible costs to the poorest households.
The model of “electric autarky” is a beautiful dream but the hard reality is that, although the model will be decentralized in the coming years, it will probably never become “every man for himself.” Even if storage costs were to fall sharply, the law of large numbers makes it is generally more cost-effective to pool the management of supply/demand gaps at the level of a household, a neighborhood or even an entire grid.
ParisTech Review – Overall, today, wouldn’t it be more reasonable for a pragmatic environmentalist to connect to the grid?
Vincent Schachter – Once again, I’m not sure that ecology, nor even economy, are the primary motivations of early adopters who decide to buy a Tesla battery or a Nest thermostat… Future consumers will probably buy based on more objective criteria, but the pooling of offer/demand will certainly be more advanced than it is today thanks to the decentralized generation, storage, aggregation and optimization of energy flows based on real-time algorithms.
These flows will transit through the grid. But who will orchestrate them and who will extract their value? Today’s electricity producers? Producers of renewable electricity? Grid managers? Big digital players? Each of these actors has jurisdiction over some bricks of the future value chain: beyond simplistic oppositions, the final answer will probably emerge from a game of partnerships and reconstructions.
Vincent Champain – I agree with what Vincent said about the motivations of early adopters: from ecological point of view, it is a waste. Allow me a little provocation: wouldn’t it be much better not to buy anything and plant trees in the Amazon forest? We would further reduce CO2 emissions… and for much cheaper!
If we think in the longer term, however, we can be less critical: it’s difficult to imagine the precisely where technological progress will lead us in ten, twenty or fifty years. The only way to find out is to explore these new avenues. Simply put, the bet on the long term will rather be reserved for advanced users that are willing to lose some money and change their consumption patterns in order to allow science to take a leap forward. And we may well be grateful for this approach, while recommending caution to those who cannot afford to take this kind of risk.