Time for 2026 updates to my annual “opinions about solar” thread, now with storage.
-
1. It’s the end of the beginning of the age of cheap solar. 2026 is likely to be the first down year for PV build, despite module prices of 11-12 US cents per Watt in normal markets, not substantially up from the lowest levels ever (9 cents in 2025 before a Chinese export tax rebate was cancelled).
2. Solar was about 8.8% of global electricity generation in 2025 (from BNEF’s generation stats, which stick to published figures, ie no selfconsumption or offgrid solar or fossil). It’s much higher in places, eg 18% in Germany (shown as chart). This is causing fewer problems than once feared.
-
2. Solar was about 8.8% of global electricity generation in 2025 (from BNEF’s generation stats, which stick to published figures, ie no selfconsumption or offgrid solar or fossil). It’s much higher in places, eg 18% in Germany (shown as chart). This is causing fewer problems than once feared.
3. Solar is struggling because it’s running into saturation effects in major markets, where there’s already plenty of power in sunny hours, and adding more solar panels generates the most power in the same hours. The power stack in Germany in a sunny week now looks like this.
-
3. Solar is struggling because it’s running into saturation effects in major markets, where there’s already plenty of power in sunny hours, and adding more solar panels generates the most power in the same hours. The power stack in Germany in a sunny week now looks like this.
4. This leads to negative feedback mechanisms: low wholesale power prices in sunny hours, which in forward-looking jurisdictions are passed on to consumers. For example, Australia’s Solar Sharer scheme obliges many utilities to offer a price plan with 3 hours of free power around midday.
-
4. This leads to negative feedback mechanisms: low wholesale power prices in sunny hours, which in forward-looking jurisdictions are passed on to consumers. For example, Australia’s Solar Sharer scheme obliges many utilities to offer a price plan with 3 hours of free power around midday.
5. Solar is still the cheapest source of bulk electricity in many countries, and the quickest to deploy, and now you couldn't stop it if you wanted to. The limits to PV build in most places that aren’t yet running into saturation effects are grid access and permitting.
-
5. Solar is still the cheapest source of bulk electricity in many countries, and the quickest to deploy, and now you couldn't stop it if you wanted to. The limits to PV build in most places that aren’t yet running into saturation effects are grid access and permitting.
6. In a way, it’s reassuring that solar is saturating, as it means solar is actually displacing fossil fuel rather than power demand increasing vertically. But it means the beginning of a new phase: the age of cheap solar and batteries. This too will pass, but it’s just getting started now.
-
6. In a way, it’s reassuring that solar is saturating, as it means solar is actually displacing fossil fuel rather than power demand increasing vertically. But it means the beginning of a new phase: the age of cheap solar and batteries. This too will pass, but it’s just getting started now.
7. Batteries are going to grow like crazy for the next five years or so and BNEF’s forecasting team are being cowards about it. Batteries are still expensive compared with no batteries, but they are easy and quick to build, and can sometimes replace hard-to-organise investments like transmission.
-
7. Batteries are going to grow like crazy for the next five years or so and BNEF’s forecasting team are being cowards about it. Batteries are still expensive compared with no batteries, but they are easy and quick to build, and can sometimes replace hard-to-organise investments like transmission.
8. The consensus answer to intermittency is now just capitalism and batteries. However, batteries reduce peak evening prices much faster than they increase sun-hour prices, so they cannibalize their returns before they increase solar returns. As we already see in Australia.
(Paywall)
https://www.bnef.com/insights/40165 -
8. The consensus answer to intermittency is now just capitalism and batteries. However, batteries reduce peak evening prices much faster than they increase sun-hour prices, so they cannibalize their returns before they increase solar returns. As we already see in Australia.
(Paywall)
https://www.bnef.com/insights/401659. Batteries still do not solve the seasonal problem. A battery of any type that only charges and discharges a few times per year is extremely expensive, even if the capex is low!
-
9. Batteries still do not solve the seasonal problem. A battery of any type that only charges and discharges a few times per year is extremely expensive, even if the capex is low!
10. We oughtta be building more wind, especially in seasonal climates like Europe. Solar will get built anyway, it’s cheap, but wind blows at night and in the winter. We’re going to need something to charge the batteries in winter, especially with rising heatpump use.
-
10. We oughtta be building more wind, especially in seasonal climates like Europe. Solar will get built anyway, it’s cheap, but wind blows at night and in the winter. We’re going to need something to charge the batteries in winter, especially with rising heatpump use.
11. How mismatched is European seasonal solar supply and electricity demand? Well, here is the situation for my fully electrified house in Switzerland with 13.2kW of solar, a heatpump and an EV. I would have needed about 122kW of PV to be self-sufficient across December.
-
11. How mismatched is European seasonal solar supply and electricity demand? Well, here is the situation for my fully electrified house in Switzerland with 13.2kW of solar, a heatpump and an EV. I would have needed about 122kW of PV to be self-sufficient across December.
12. Sometimes the story is the dog that doesn’t bark in the nighttime. European grids would once have struggled in a 2026-level heatwave, but they had only a few problems and high evening prices. CAISO and Ercot, which also have lots of batteries, had good reliability years despite record loads.
-
12. Sometimes the story is the dog that doesn’t bark in the nighttime. European grids would once have struggled in a 2026-level heatwave, but they had only a few problems and high evening prices. CAISO and Ercot, which also have lots of batteries, had good reliability years despite record loads.
13. Europeans shouldn't feel guilty about using electricity for airconditioning (or reversible heatpumps), it already mostly comes from the sun, especially if you pre-cool. This is good news for hot countries as well. Heat is deadly.
-
13. Europeans shouldn't feel guilty about using electricity for airconditioning (or reversible heatpumps), it already mostly comes from the sun, especially if you pre-cool. This is good news for hot countries as well. Heat is deadly.
14. Data center power demand increase is significant, but localised, and probably not as high as clean power developers would like. BNEF estimates that data center electricity use will more than triple by 2035, reaching 5.4% of global demand, though it makes up 23% in the US PJM market, 18% in Malaysia and 15% in the UK.
-
14. Data center power demand increase is significant, but localised, and probably not as high as clean power developers would like. BNEF estimates that data center electricity use will more than triple by 2035, reaching 5.4% of global demand, though it makes up 23% in the US PJM market, 18% in Malaysia and 15% in the UK.
15. Data center / Big Tech firms are by far the largest corporate buyers of clean power under power purchase agreements worldwide. They are also signing expensive deals for clean firm power, like nuclear and solar-and-storage, seeking to meet hourly matching targets. Non-paywall:
-
15. Data center / Big Tech firms are by far the largest corporate buyers of clean power under power purchase agreements worldwide. They are also signing expensive deals for clean firm power, like nuclear and solar-and-storage, seeking to meet hourly matching targets. Non-paywall:
16. It is relatively easy to run a baseload power load, like a data center, *mostly* on solar and batteries. Especially somewhere like Saudi Arabia where the solar resource is quite constant all year. However, to get to the reliability needed, you need something else – either gas backup, or some bonkers long-duration storage tech.
-
16. It is relatively easy to run a baseload power load, like a data center, *mostly* on solar and batteries. Especially somewhere like Saudi Arabia where the solar resource is quite constant all year. However, to get to the reliability needed, you need something else – either gas backup, or some bonkers long-duration storage tech.
17. One good thing about data centers is that they drive some investment into the scale-up of bonkers long-duration storage tech, eg Form Energy’s iron-air batteries (link). They care much less about the cost of power than about speed-to-power, which honestly is a new term for me this year.
https://www.canarymedia.com/articles/batteries/gigantic-form-energy-battery-google-minnesota
-
17. One good thing about data centers is that they drive some investment into the scale-up of bonkers long-duration storage tech, eg Form Energy’s iron-air batteries (link). They care much less about the cost of power than about speed-to-power, which honestly is a new term for me this year.
https://www.canarymedia.com/articles/batteries/gigantic-form-energy-battery-google-minnesota
18. Lithium is going to continue to be the main technology for batteries for the foreseeable future, although China does have some policy support for sodium-ion.
-
18. Lithium is going to continue to be the main technology for batteries for the foreseeable future, although China does have some policy support for sodium-ion.
19. After extensive complaints from me, BNEF now defines long-duration storage correctly: not in hours, but as technologies where “the cost and scale of energy capacity (GWh) are largely decoupled from those of power capacity (GW)".
It's where you have a big tank of some cheap storage medium, basically.
-
19. After extensive complaints from me, BNEF now defines long-duration storage correctly: not in hours, but as technologies where “the cost and scale of energy capacity (GWh) are largely decoupled from those of power capacity (GW)".
It's where you have a big tank of some cheap storage medium, basically.
20. Long-duration storage deployment by BNEF’s definition was just 2GW / 9.6GWh in 2025, mostly in China. The main technologies getting built so far are compressed air and flow vanadium redox.
-
20. Long-duration storage deployment by BNEF’s definition was just 2GW / 9.6GWh in 2025, mostly in China. The main technologies getting built so far are compressed air and flow vanadium redox.
21. The Hormuz crisis is having relatively little direct effect on solar panel sales, because solar doesn’t usually replace liquid fuel. The crisis is boosting EV sales, particularly in Europe where they reached 36% of car sales in Q2 2026 (China’s figure was 60%, though the car market overall slowed down). This does eventually support solar.
