The main paper is more interesting, though the commentary paper is ... ETA unconvincing (edited to avoid bothering the libel lawyers.)
The paper recognises that wind is a mature technology, and close to its thermodynamic limit. So cost can't fall much further. Tick.
Yes, PV cells can still be reasonably expected to fall considerably in price. Tick. Though I don't think we can be quite as certain of it as they suggest, because PV has some key differences from the benchmarks they are using to project the falling cost. The present methods of making PV cells are mostly close to their technological limits. Breaking that limit, to keep the cost falling by orders of magnitude, requires some quite different technology. But this was not true for many of the things they are benchmarking it against. For example, with either silicon chips or windmills early in their cost reduction phases. We stuck with the same technology, and refined it, because it was capable of being refined. We had landed on the right method, and knew there was huge room for refinement. We didn't need a totally new technology to get past some thermodynamic limit of the present technology. Cost does keep reducing. But it is no the same cost reduction process. We have not found the best method yet, that we only have to refine. So I can reasonably believe that PV cells can get a lot cheaper. But it requires technological breakthroughs that are harder than some of the other things they compare it to.
So, let's suppose we can reduce the cost of PV cells by an order of magnitude. Just to be clear, this is crucial to their projections. If this doesn't happen, they don't have any other generation technology that they think can get a lot cheaper. They have sensibly stayed away from wave power and all those other things that have stubbornly refused to fall in cost despite the invesments.
So we now have very cheap solar cells. We now have available a huge quantity of electricity harvestable at low cost. But its only there when the light is bright enough. They recognise this, and put in place systems to store the electricity, so it can be used when it is wanted, not when it is generated. Tick.
What is their technology for doing this? Well, they have two, green hydrogen (or equivalent chemical carrier that PV can make), and gridscale batteries. All sorts of other energy storage mechanisms have been proposed, but they just don't show the potential for really large reductions in cost. They are usually close to thermodynamic limits even in the prototype. So tick, they have identified the two best chances, so far as we can know.
The batteries are like the PV cells. The ones we currently have are close to their thermodynamic limits. If we are to achieve order of magnitude cost improvements, we need technology changes different from just making the present technology work a lot better. Yes, I can believe that batteries can get a lot cheaper. But, like PV cells, it is not at the level of certainty of technologies where we have landed on the right method.
Green hydrogen. Well I also believe that hydrogen or some other chemical store is likely the best method of storing energy at a seasonal level. What we need is an electrolyser that is cheap enough we only run it during those relatively short periods of time when we have a large excess of solar energy. That is probably at most 25% of the time. Maybe in some low latitude desert a bit more. Half the time it is dark. During the day, the sun is low in the sky near sunrise and sunset, and in the winter if you have one. So if you have excess solar power 25% of the time, you are doing well.
So then we have the cost of electrolysing, or some other electrically run process to make a chemical with high energy storage density. We don't have much of a cost reduction record for these. I was aware that cost was just not reducing very much for much of the last couple of decades. But according to their chart, there is now some serious rate of cost reduction beginning from around mid decade. I will give them the credit of having the widest range of uncertainty around the future trend of this, including not reducing very much further at all. And we need to do this on an industrial scale. Unlike the other technologies above which are largely small things vastly replicated. I suppose you can take the vastly replicated approach to this too, though somehow it seems less likely. Again, the general problem is that we do need to think of some rather different ways of doing it, as the present electrolysers are close to thermodynamic limit.
So they have something I can roughly believe. The way is far from smooth. And they recognise it is not totally certain these cost reductions will happen.
What they haven't addressed though, is the huge supporting infrastructure that is required to make this happen. Where are all these PV cells to generate a Europe quantity of electricy going to go? I don't think we have land to afford of suitable insolation to do that within Europe itself. Wind turbines are not going to get vastly cheaper, it's solar we need, and it needs to go somewhere sunny and non-agricultural like the Sahara desert. And then we need huge power lines. All those electrolysers will have to go somewhere there is water, and, by present technology, fresh water. So, even if what they say comes to take place, there are some additional enormous infrastructural requirements that are not within their paper.
This paper is futurology, and as we know, making predictions is very difficult, especially about the future, so futurologists have to resign themselves to a high error rate. My sister gave me Yuval Noah Harari's 21 Lessons for the 21st Century for Christmas, a book of futurology. Whilst I have expressed my own criticism of the weak points in this paper, it is much more plausible than Harari.