Every year, Japan generates a large amount of solar power and then deliberately refuses to take it.
The usual reaction is that this is a waste. I buy fuel for a living in the Japanese energy industry, and I don't think it is — at least, not in the way most people mean.
You paid for that electricity, and it never got generated
Start with the scale.
In FY2024 (April 2024 to March 2025), curtailed renewable output across Japan came to about 2.42 TWh. The previous fiscal year it was roughly 1.76 TWh, so it grew about 1.4x in a single year.
For a household using 400 kWh a month, that's roughly a year of consumption for 500,000 homes.
Here's the part that makes it personal. Those solar farms are underwritten by a surcharge that appears on every electricity bill in Japan — the renewable energy levy. For FY2026 it's ¥4.18 per kWh (roughly $0.028).
So the structure is:
We collectively pay to buy this electricity, and then it is never produced.
Stated that way, irritation is the reasonable response. I'm still going to defend it.
If you don't throw it away, the lights go out
Electricity has an awkward property that rarely makes it into public conversation: you can't store it.
More precisely, storing it requires expensive equipment. Left alone, power has to be consumed the instant it's generated.
So a grid operator's job is to keep supply and demand matched continuously — in Japan this rule is called dōji dōryō, "simultaneous equal volume."
This is where intuition breaks. Electricity isn't only dangerous when there's too little. Too much is dangerous too.
When the balance breaks, frequency drifts. Generators trip off the grid to protect themselves. Other generators pick up the load, drift further, and trip in turn. The end state is a wide-area blackout.
Curtailment is the operation that stops this one step before it starts.
We're not throwing power away because the system is broken. We're throwing it away because the system is working.
Before you say "just add batteries"
Most people reach the same place at this point: store it in batteries.
As a technical statement, that's correct. My objection is about cost.
Who buys enough storage to catch 2.42 TWh?
Grid-scale batteries are not a cheap purchase. If government subsidies cover them, the source is tax revenue. If they're recovered as transmission network investment, they land in wheeling charges — which land in electricity rates.
Either path ends on the same bill.
So: to avoid curtailing 6%, how much are you willing to raise electricity prices? Saying "what a waste" without doing that comparison is a little dangerous.
There is money on the other side of "waste," too.
Kyushu curtails 6%. Tokyo curtails zero.
The regional split is where the real problem shows up.
| Region | Curtailment rate | Why |
|---|---|---|
| Kyushu (main island) | ~6.2% in FY2024, ~6.1% in FY2025 | Very high solar penetration, limited transmission capacity to send surplus elsewhere |
| Tokyo area | 0% as of June 2025 | Demand is large enough to absorb midday solar outright |
Kyushu adopted solar faster than anywhere else in Japan, and the transmission lines out of the region have finite capacity. Tokyo simply has enough load to swallow it.
Which means: under one national scheme, only generators in Kyushu have their revenue cut.
I won't defend that part. Curtailment itself may be unavoidable. Concentrating the losses on one region's operators is a separate problem, and a worse one.
So what percentage is acceptable?
Honestly: I don't have an answer.
Is 6% tolerable and 10% the point where you should invest? I can't justify where that line goes.
Even working inside the industry, I rarely meet anyone who can state this number clearly. I've come to think that's because it isn't a technical question. It's a values question.
The less you curtail, the more electricity costs. The cheaper you keep electricity, the more you curtail.
Someone has to draw the line somewhere.
And in Japan, we prepay for the right to vote on that line every month, in the form of the levy on our bills.
Where would you draw it?
I'd genuinely like to hear how this looks from other grids. California, South Australia, Germany, Chile — the curtailment numbers differ, but I suspect the argument doesn't.
I work in fuel procurement in the Japanese energy industry and write about what actually moves energy prices, mostly in Japanese at note.
Related, if you read Japanese:
Top comments (1)
For context on the regional split: Kyushu has roughly the highest solar penetration in Japan but limited transmission capacity out of the region, while Tokyo has enough industrial demand to absorb midday solar entirely.
If you're on a grid that curtails — where does your operator draw the line, and who pays for it?