What Would The Economic Impact Be If Everyone Installed Solar Panels?
Updated: Jul 13, 2020
What is the economic impact if everyone install solar panel on their roof? originally appeared on Quora: the knowledge sharing network where compelling questions are answered by people with unique insights.
Answer by Paul Mainwood, worked at McKinsey & Company, on Quora:
What is the economic impact if everyone install solar panel on their roof? Before starting, we should recognize that everyone in the world going up and installing solar panels on their roof is not the most cost-effective approach to energy generation. It costs a lot to get all those panels all the way up those ladders, and residential roof-tops are usually not the best locations. It’s far more efficient to design and install proper utility-scale solar farms.
But let’s say you’ve stuck all those panels up on the roof anyway. Fine. We can work with that.
Now there are two ways one might interpret this question:
What would be the economic impact if everyone installed solar panels on their roof, and everyone associated with power generation and distribution were forced to undergo a radical prefrontal lobotomy?
What would be the economic impact if everyone installed solar panels on their roof, and at least a few people associated with power generation and distribution were allowed to retain some portions of their brains?
The answer to the first question (as several have already stated) is that there would be huge glut of cheap/zero-cost electricity during the middle of the day, followed by a shortage as the sun went down. So, the plateau in the middle of the graph below would be overserved, but the peak on the right would not be (at least not in winter, when it is highest).
Since everyone just has had a radical prefrontal lobotomy, they are unable to think about this situation and would be forced to keep online almost all the legacy fossil fuel capacity to service this peak (as well as a smaller amount to get through the night, when electricity consumption is generally low). The market would still do its work, so the grid would pay very little for the cheap oversupply during the day, while paying a great deal to the fossil providers in the evening, where demand was almost outstripping supply.
Now, even in this scenario, the carbon footprint of electricity would go down significantly, for the extra fossil capacity would be standing idle most of the time. Even to the recently lobotomised it would be clear that it was only worth burning fossil fuels for a few hours a day. But we would be left with an expensive and inefficient generation mix, with a great deal of overcapacity spending a lot of time lying around.
But now let’s consider the second scenario, where we skip at least some of this brain surgery.
In this situation there is a massive amount of cheap electricity during the day but this drops off in the evening, just as the evening peak ramps up. If our grid planners are capable of agency and reasonable thought, they might give some thought to the gigantic arbitrage opportunity that just opened up.
That is, if there was a way of storing the cheap electricity for just a few hours, it could be released to serve the evening peak, and then more slowly through the night. And given any kind of functioning market for electricity, there will be a strong financial incentive to do this, in the form of a grid that was prepared to pay a great deal for electricity delivered during the early evening.
So, does this electricity storage technology exist? Yes, there’s lots. But right now, most of it is pretty expensive. Here are the major competitors vs. their applications, with a reasonable* methodology on looking at the cost of storage.
Assuming no subsidies, the critical question is whether this storage capacity can be brought in cheaper than the fossil “peaker” plants that our first scenario kept around in order to serve the evening peaks.
Given that the cheapest type (gas peakers) can supply electricity at $165–220 per MWh, there are only two types of storage that are clearly winners right now: compressed air and pumped hydro.
Both of these are great solutions if they’re available, but this depends a lot on the local landscape and geology. Pumped hydro needs mountains and reservoirs, whereas compressed air needs the underground caverns like those created from salt extraction.** So, now it depends on which country you are in. If you are in Norway, you’re done: convert your hydroelectricity capacity to allow pumped hydro, build a bit more and you can transfer your daytime peak. If you are somewhere with plenty of salt caverns just begging to be pumped full of compressed air, get going on that. You’re done.
But let’s say you’re somewhere unlucky where you’ve neither of these. Or at least not enough capacity to cover that evening peak.
What happens now will depend on some of the following factors:
What is the rest of your energy mix, and how well does it serve the evening peak?
What are your goals vis-a-vis low carbon, and are you prepared to subsidise to achieve them?
Do you have long-range power grid interconnectors to areas where either the sun is out at meaningfully different times, or where they have a lot of storage capacity?
Are you prepared to ask people to change their consumption habits (e.g., by time-of-day pricing)?
What year are we in?
Depending on the answers to the first four questions, you will come up with a different mix of solutions that will probably involve some transfer from other regions, some peaking generation, some peak flattening, plus a variety of other storage approaches, subsidized or not.
But the last one is crucial. Are we talking about today? Or are we talking about five years time (after all, we’ve got to give everyone enough time to install panels on their roof, right?)
It matters because battery technologies are on a cost dive that will swiftly make the numbers on the chart above obsolete.
The other technologies may well be able to reduce their costs over time, but batteries are doing it year after year, on a regular downward trend that we can recognize as a classic technology “learning curve” — every time you double capacity installed, the price per unit goes down in a regular, almost law-like manner.
Lithium-ion in particular are on a pattern that appears consistent with a 22% learning factor (double capacity, and unit price reduces by around 22%). In the last few years, this has resulted in price movements like this (NB: the numbers on the vertical axis of this graph should not be compared to the numbers on the last graph, they are measuring very different things). The percentage drop is the thing that matters:
Other battery technologies may be going down even faster (Lazard think that Lead acid may be poised to go down fastest of all). But let’s stick to Li-I for simplicity and just think about what this magnitude of price movement means.
It means that in just a few years (let’s say five, to be generous) that storage will cost-effectively be able to transfer electricity generated the peak solar capacity in the middle of the day to the evening, and overnight. That’s a lot of batteries. But so long as that experience curve keeps going, I can’t see why it will not happen — it can serve the capacity and it will be the cheapest option. Arguments about whether these batteries should be placed in the home (so-called “behind the meter”) or planned at utility scale in the grid can be had another time: they depend on the sort of network you want to have and thus the incentives you set.
So, in summary — what would happen if you got everyone to install roof-top solar? Three points.
Solar’s the right path, but please don’t stick it all on roof-tops; that’s not the most cost-efficient approach, look at utility-scale solar farms. Even in 2016, and with subsidies stripped out, they’re often cheaper than fossil.
If you do this today (2016), you will have a glut of cheap electricity during the day with evening peak demand to sort out as well as overnight capacity. There are a variety of approaches to this, depending on your aims and on your geography, but the price-optimal solution might well involve keeping a substantial amount of fossil peaking capacity around.
This situation is temporary. If you actually did install roof-top solar everywhere, the incentive to establish cost-effective storage would accelerate even the current astonishing plunges in the price of storage technologies — this peaking capacity would likely only be around for a few years before being replaced.
Subsidies could change all of this, but in their absence, something like this would happen. Market forces. Market forces and time.
[*] Methodology detailed in the source: https://www.lazard.com/media/438…
[**] There are other versions which are location independent, but these are likely to be higher cost than shown here.
This question originally appeared on Quora. Ask a question, get a great answer. Learn from experts and access insider knowledge. You can follow Quora on Twitter, Facebook, and Google+. More questions: