The cost of renewables has fallen, but the cost of grid scale energy storage is still quite high. When the cost of the entire energy grid is viewed holistically (including required storage if using renewables) the cost of nuclear is competitive with renewables.
Unless people are suddenly ok with blackouts, energy grid have to be robust. For renewables, that means having enough stores energy for times of decreased production.
I have long considered that those buying their power from wind or solar should have remote relay switches in place that simply cut power when there is no solar or wind produced. Seems exactly what they are buying.
Barely: the majority of energy tabulated as "hydroelectric power" comes from pumped hydroelectric storage. However, computing this as hydroelectric power is very naive:
Option a) the power to pump was generated using fossile fuels. In this case pumped storage is just "greenwashing" energy and doesn't actually produce any.
Option b) the power was generated using renewables. In this case the pumped storage just counts the renewables twice.
This "pumparound" is what makes up the gains in hydroelectric power, not only in Germany, but in most countries: There simply isn't infinite amounts of useable space for hydroelectric power.
If you look at the generated energy, though you will see a constant increase in generated power:
2016: 11,258 MW hydroelectric power, of which 6,806 MW (more than half) are pumped. Generation: 21.5 TWh
2019: 11,022 MW hydroelectric power, of which 6,364 MW (still more than half) are pumped. Generation: 27.88 TWh
A major component of the rise in generation is that pump-around.
From an energy storage PoV they aren't even that effective: Let's say you wanted to supply Munich and surroundings with power for 3 days (~2Mio people). Let's also assume that there are only people living there, so no industry or infrastructure. Let's also assume we have 100% efficiency in our pumped storage system. Let's also assume that everybody lives in two-person housholds (to make the math eaiser). According to destatis, a two person houshold consumed 3221kWh in 2018 (without industry and infrastructure).
Dividing by 356 gives us the daily consumption.
In Joules, this amounts to 3.25710^13 joules.
Dividing by 9.81 m/s^2 and we achieve a value of 3d3.3210^12 kgm.
There are now two ways to handle this:
1. high mass
2. high distance
Let's choose something flashy and pump lake constanze (bodensee), the largest freshwater lake in central europe. We would have to pump this up by almost a quarter of a meter to save this amount of energy. And this is with completely optimistic values: double this for electric cars, then add the same amount for electric heating, then again for industry, then add another at least 20% for inefficiency of the hydroelectric generation. And then you have to consider that this is one German city. Or, maybe in a different view: German renewables only need to lose 1/40 th of their efficiency for you to be in this situation (if you want to go 100% renewable).
One also has to consider that, even if it were possible to add more capacity (which the German hydroelectric association doesn't believe is), one has to consider that hydroelectic is by far the most environmentally destructive of all power generation methods.
Simply the flooding of a single valley is such an utter environmental disaster, that one has to be very sure that this valley absolutely _has_ to become a hydroelectric installation. Additionally one has the construction of the damn itself, as well as the enormous disturbance of every instance of wildlife both up and downstream for decades to centuries (and that ignores the both human and wildelife destruction brought on by a damn break).
I would gladly take the construction of a nuclear powerplant over a hydroelectric one, considering that the biblis reactor alone has a nameplate capacity of almost half the amount of non-pumped hydroelectric power.
Unless people are suddenly ok with blackouts, energy grid have to be robust. For renewables, that means having enough stores energy for times of decreased production.