[rsj_hn]

Because there are fixed costs, and you need to pay for all the infrastructure in case it's a cloudy day.

Imagine there is a car that costs $10,000 and works every day, or there is a car that costs $3000 but works only every other day. But you need to drive every day.

Does it make sense to buy both cars and drive the cheaper one on sunny days? No, economically you only buy the reliable car and you don't buy the unreliable car at all. Notice the discontinuity -- until the reliability problem is solved, it makes no economic sense to deploy any solar. This is the economics of fixed costs.

Whereas what you are thinking of are variable costs. So say the unreliable car can be fueled for $1 a day and the reliable one costs $2 dollars a day. In the world of variable costs, it makes sense to use solar up to 100% of sunny capacity.

Whenever there is a battle between fixed and variable costs, what determines the winner is the interest rate. With low enough interest rates, you get to 100% solar on cloudy days. But with high interest rates, you don't go with solar at all because it is too expensive to keep all those fixed assets on standby and pay for twice the capacity you need just because your solar system goes offline.

So it is the relationship between interest rates and energy prices that determine when it makes sense to max out the cheap, unreliable producer and pay the reliable producer to be on standby, or when you only go with the reliable producer.

And interest rates and energy prices differ over time and in different places. There are many individual producers making these binary choices and in aggregate they never end up using 100% of one option or 0% of the other, it is usually something in between. But just because you see that there is something in between doesn't mean that lack of storage capacity is inhibiting the adoption of solar. It is the only thing inhibiting it. On the other hand, if you solve the storage problem, you can go way past 100% usage on sunny days, you can go to 100% total usage.

So because of fixed costs, it's totally wrong to focus on what happens just on sunny days. What prevents the adoption of solar is affordable storage -- and that is inhibiting the adoption of solar right now.

The unwillingness of people to deal with the storage problem, how they handwave it away or just pretend that it's not a blocker in the present is one of the most frustrating aspects of energy discussions. It is all about storage. That is 100% of the problem. Whether renewables succeed or fail is solely determined by the affordability of storage. The storage problem dwarfs every other consideration by a mile.

[TheAlchemist]

I'm sorry I still don't understand why this is the biggest problem now. I get why it's a long term problem - that for sure.

But, your comparisons are about building new stuff. My understanding is that the gas / coal capacity is already there - so it's not a question of adding more, right ? So we don't have those fixed costs - they are already paid. We already have 100% of coverage with those, so adding solar, even without storage, will indeed decrease those variable costs.

Economically speaking, if we say solar is cheaper than gas / coal - what's preventing a rational actor from installing more capacity ? Until we get to 100% solar on sunny days, he is sure to sell 100% of his production if it's cheaper gas / coal no ?

[rsj_hn]

> So we don't have those fixed costs - they are already paid.

No, that's generally not how infrastructure is funded. Infrastructure is funded by bonds that are sold and have to be serviced. So the payment for the fixed costs is spread out over time -- ideally over the lifetime of the asset -- and what happens when you add unreliable source that are cheaper -- like solar -- is that they cannibalize the revenue stream used to service the debt and those plants are taken offline right now, or they have to be subsidized somehow by the taxpayer -- which is why regions that make a big push to solar end up paying high energy prices and need to import a lot of energy. They are not paying high energy prices because solar is expensive, but they end up paying high prices to keep their reliable plants online. Or they require subsidies from the taxpayer. California is one example, but you see this often in Europe as well. California just got a billion dollar Federal subsidy to keep diablo online for a few more years, but the reason it was going to go offline was because it needed massive subsidies to remain operational in a market in which it competes with solar. But at the same time, solar can't replace Diablo without blackouts and water shortages - Diablo supplies 10% of California's energy and is used to pump water, so this is something that requires reliability, but it is losing money every year because it's cannibalized by solar. That's the situation of the household that chooses to buy both cars -- that household has to spend a lot more.

Solar has basically made nuclear uneconomical in the U.S. because nuclear has massive fixed costs but almost no variable costs, so it can't exist in a de-regulated market with solar -- the solar just drives it out of business. At the same time, solar is driving greater use of coal because coal plants have the cheapest fixed costs. So Germany was ditching nuclear and using more coal as a result of deploying all the renewables. If you don't pay extra to maintain your reliable capacity, you get blackouts or pay emergency rates to beg for energy from your neighbors (California has to import 20% of its electricity).

Rational actors -- like the household choosing between the two cars who knows they need to drive every day -- are required to plan for sufficient capacity look at this and say "we're not ready to deploy solar at scale" because they want to avoid the blackouts and price spikes. Politicized actors choose a different route, and you get the energy mess that is California, UK, Germany, etc. And all of that is solely because of storage, and it gets worse the more solar you deploy without storage.

[TheAlchemist]

Well that's fascinating, thank you very much for explanation !

[Things4People]

Wow! Thanks for the great explanation!