How can climate change be solved without the use of government? originally appeared on Quora: the knowledge sharing network where compelling questions are answered by people with unique insights.
We already have a solution to climate change, and it doesn’t rely on government. Something like it is going to happen, the only question is how fast. The solution comes in three parts:
None of these require the government to intervene. All we need is existing technology, market forces, and time. Our hope is that the time we need isn’t long enough for us to cook the planet with fossil fuels.
Let’s go through these pieces one by one.
1. Solar PV will beat any other power generation method on cost by 2030 (yes, I do mean unsubsidized cost):
Solar photovoltaics are already way cheaper than you think they are. Even in cloudy, northern countries like Germany and Denmark, they’re at grid parity. In sunnier counties like India they’re the cheapest power source available (providing the lowest bids in the Andra Pradesh, Telengana, Punjab and Madhya Pradesh state auctions in 2014 and 15, beating local and imported coal). And in ideal places like Chile and Nigeria, they come in way, way below the next-cheapest options, even if you take away subsidies and match them vs. local power stations burning the plentiful local coal.
What are unsubsidized costs looking like versus other generation technologies? In 2015, they looked like this:
Source:– grey diamonds = 2017 estimates.
Look at the “utility scale” lines: PV at scale is cheaper than fossil fuels (and even for residential/commercial it’s comparable with grid costs). Let’s emphasize again: this is unsubsidized levelized cost.
But this isn’t the astonishing part: the astonishing part is the rate at which PV cost is dropping.
While this pattern has been known for a while, it has recently been recognized and named as “Swanson’s Law”: with every doubling of solar installation, we get a price reduction per watt of approximately 20%.
Swanson’s law is just a specific example of the familiar exponential “learning curve” as studied by T. P. Wright in the 1930s, and later commercialized by BCG. And while a 20% version is a pretty high factor, it’s not in any way unprecedented. So why is this so exciting?
It’s exciting because of how often the volumes are doubling: every 2.2 years, give or take. Put another way, around 90% of the solar PV generating electricity today was installed since 2010.
A 20% cost-learning factor, with 2.2 year doubling time has such a powerful effect on prices that it’s hard for humans to grasp. If it continues, then by 2030, solar PV will be around a third of the cost it is today. There will be nothing to compete with solar PV in almost any country in the world; subsidized or unsubsidized. Not coal, not gas, not nuclear, not anything.*
What is more, PV solar even at its current efficiency levels is easily capable of supplying the world’s long-term energy demands (best-guess, around 30TW) and can do so using less than 1% of the land area we have available.
So why do we have this idea that solar has a punitive cost? Mostly because it’s only been in the last few years that Swanson’s law has pushed prices into the strike zone for grid parity. Even in 2010, people felt comfortable dismissing PV as far too pricey for all but niche applications, simply it because it was about twice the price it is today. And people don’t like changing their beliefs. Even now, those who do accept Swanson’s law search around for reasons why this breakneck cost reduction cannot possibly continue. Computer industry veterans will be reminded of the reams of analysis predicting that Moore’s law would break down in every year since 1965.
It seems that being close to the effect doesn’t help: the experts keep on getting overtaken by the pace at which solar gets cheaper. Let’s look at the IEA (International Energy Authority) projections for energy cost. In 2010 they published their ten year projections, saying that in places without much sun like the UK, then in 2020 you might get an all-in cost of $0.21 per KWh, rising to around $0.105 per KWh in the very sunniest countries.
In 2015, the UK held a competitive auction, and obtained bits of $0.12 per KWh. In sunnier countries they were getting bids of $0.06. That is, five years earlier than the IEA’s dates, prices were already 40% lower than the IEA estimates. Oops.
In the meantime, the IEA noticed the plummeting cost of solar was overtaking their predictions, so in 2014 they revised their estimates down a bit. This time they went out further, and boldly projected that in 2050, there was a best case scenario in which solar PV could produce a cost as low as $0.05 per KWh.
But in May 2016, Dubai accepted a bid just under $0.03 per KWh, and Chile accepted one for $0.029 in August. These were good sites, and they benefited from policy, but they were 40% lower than the IEA estimate, and delivered … er … 34 years early. Big oops.
This keeps on happening. Solar keeps getting cheaper, and keeps doing so faster than anyone is prepared to write down.
A lot of people will be thinking (or shouting): “But solar only works when the sun’s shining! What damn good is it to have cheap electricity only during the day, and when it’s not cloudy! And what about winter, when energy needs are the highest?”
These people are right. That’s what the other two bits of the solution are about.
2. A cost-effective energy storage solution to solve short-term intermittency (hours to days):
There are many options for short-term energy storage, including compressed air, flow batteries, pumped hydro, and even esoteric technologies like super-capacitors. But the most likely storage option in the near term is good old lithium-based batteries.
Why do I say this? It’s because of their price: lithium-ion cells are racing down a learning curve similar to the one that PV is on. And we’re seeing a slope of about the same steepness (20% cheaper every time you double capacity). It may be even faster: it’s been estimated at 22%, and with new large-scale production put in place by Panasonic and Tesla, we might see it go even faster than that.
There may be better technologies on the drawing board (lithium air comes to mind as the lurking gasoline-killer), but with current prices of $100–200 per kwh today, and a learning curve of 20%+, it’s going to be damn hard for even a superior tech to catch up with lithium-ion. So here’s the second part of the vision: generation is provided by solar at prices cheaper than anything else, with short-term storage provided mostly by lithium based batteries to take account of night-time and cloudy days (and yes, there’s enough lithium to make, and enough space to place these battery packs where they’re needed).
3 A cost-effective energy storage solution to solve long-term intermittency (weeks and months) i.e. winter
The trouble is, if we’re anywhere away from the equator, we need to get through the winter. And running cables from countries near the equator doesn’t work.** Wind helps (especially if in the UK, Germany or Nordics) but there are still months at a time where you’ll see a deficit. This will be made up in the summer months, where there will be a massive glut of low-priced electricity from PV, so what we’re looking for is some way to store a few hundred terrawatt hours of energy for several months.
This is much harder than short-term energy storage. In my view, there’s only one reasonable way of doing energy of this scale for months at a time and it goes by the name of Power to Gas (P2G).
When you have a massive surplus of PV power (summer), you run electrolyzers to produce hydrogen from water (you probably did this experiment in high school chemistry). We once thought that you could then store and use this hydrogen directly in fuel cells and turbines. But the hydrogen solution is a lot harder than it looks. Hydrogen has an awful energy storage capacity by volume: it escapes and erodes anything that hasn’t been purpose-built to hold it, and needs to be kept under high pressure. Fuel cells aren’t there yet either. The investment to make a full hydrogen infrastructure will be astronomical.
Any decent civilization would get together, overcome these technical and financial obstacles, and make the hydrogen economy work. But we’ve got a set of second-rate civilizations and are trying to do it without government help, so we need a solution that doesn’t incur such vast infrastructure costs.
What will work?
It turns out that there’s a very simple process (the Sabatier reaction) that combines hydrogen with carbon dioxide to make two ingredients: water and methane — natural gas. And we already have all the infrastructure to store a lot of energy in the form natural gas. Even better, we have the efficient combined cycle gas turbines, which can turn this gas back into electricity on demand.
So now start looking at the interesting companies (e.g., Germany’s) who have commercialized the power-to-gas process. All these companies need is a source of CO2 (from e.g., concrete manufacture, or fossil fuel generation) a load of electricity and a bit of water. With cheap PV, we’ll have all three of those ingredients.
This is the way we go. We take our surplus electricity in summer (produced at zero cost by our solar PV) and use it to make natural gas to store for the winter. We store and recycle the CO2 given off when we burn the natural gas in our CCG turbines, and plunge it back into the process again.
To emphasize, this long-term storage solution isn’t anywhere close to commercial cost levels. Every step of this cycle needs to be a lot cheaper than it is right now to be supportable without government subsidy, a carbon tax or other interventions. But no-one’s even started going down the learning curve on these stages.
(Looking further into the future: the limiting factor will be CO2 supply. We will be making concrete using processes that don’t generate any CO2, and we certainly don’t want to burn any new fossil fuels. Where will we find our sources of CO2 then? The obvious place to look is the air around us, and capture atmospheric CO2. This capture is pretty hard, but with processes being implemented right now by companies such asin Switzerland, we’re on the path. We use our cheap solar again to pull CO2 out of the atmosphere and into the P2G: this neatly manages to make a bit of a dent on the too-high levels of atmospheric CO2 by storing it our energy reservoir of natural gas. It’ll be 2030+ before this sort of thing becomes commercial, but it’s a nice bonus nonetheless.)
The neat thing about all this is that there are very few barriers in our way. We are using existing technologies and existing infrastructure (electricity and gas transmission are already there). All we need is for the learning curves to keep going,*** which will drive the installation volumes to keep coming, and market forces will do the rest.
In theory, government could speed all this up if it got involved. It could get us down those learning curves faster by supporting installations (e.g., by investing directly, adding subsidies or imposing carbon taxes). This would really speed the technologies to their way to widespread adoption.
But we don’t need government. What we need is private finance to start investing in large scale solar PV, in large-scale battery manufacture, and in power-to-gas technologies. We need some tech billionaires to think about future generations, look at the technologies in front of them and make some big bets with their cash piles; we need pension funds to start investing in smaller bets than they’re used to, and we really need the big energy companies to get their heads out of their barrels of oil, realise that they’re going to get clobbered by solar in the next 10–20 years, and need to get on board with it.
Fortunately some of these things are happening, if only in a rather scatter-gun way. But we need more, and quickly, because although something like this future will happen just on by market forces, we as humans need it to happen fast: we need this switch to solar to happen before climate change causes irreparable damage.
Notes and sources:
[*] None of these older technologies are undergoing a experience curve similar to solar. Fossil fuels aren’t because they maxed out the capacity increases years ago; nuclear isn’t because it doesn’t seem to have an experience curve, probably due to increasing safety concerns and the fact that new plants are installed so infrequently that new crews are used for every new one and have no chance to learn. Wind is the only generation tech that is undergoing a similar curve, which will likely provide a great complement to solar, especially for northern regions of Europe and the US. But it’s also intermittent, so the other two parts of the solution will be needed even when large-scale wind is added to the mix.
[**], who knows a shed-load more than me about all this, discusses in comments that I may be being over-pessimistic about the potential for long-range electrical transport to cover local intermittency. If so, the need for short and long-term storage is reduced, and we can get this future easier, cheaper and faster than described. For the purpose of this answer, I will assume that new large capacity electrical transport requires government help, so is out of scope.
[***] Will the solar and battery learning curves keep going? Work by Doyne Farmer and others suggests they will. Once a tech gets on a reliable learning curve it tends to keep going down it, until saturation volumes are reached.
Sources:, , , ; ; most indebted to Goodall (2016) The Switch
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: