What does it cost to buy a battery storage plant?
— not the land, but the acquisition price of the project and the plant

Over the past year, eight listed companies have disclosed that they bought battery storage plants — and those are only the ones we could verify. Only three of the disclosures state the acquisition price. Buyers have nothing to line prices up against. In COLUMN 39 in August, we wrote that the price comes as a ladder. This time we counted the rungs. We took the asking prices we have seen over the past nine months, collapsed them to one figure per project, and set the disclosed acquisition prices beside them.

First, the scope. We counted high-voltage (HV) plants of the 2 MW class (1.5–2 MW output, 3.5 hours' duration or more), at the total price for one project including land and the grid connection cost contribution, excluding consumption tax. Extra-high-voltage (EHV) projects vary too much in size, so for those we give only a per-kW band. Projects that won the Long-term Decarbonization Auction (LDA) are out of scope: moving their contractual position requires the prior consent of OCCTO, Japan's cross-regional grid coordinator (Article 32 of the contract terms), and we have seen none offered for sale. This is about grid-scale projects and plants, not second-hand home batteries. If you are selling, net proceeds and the steps are in the second half.

High-voltage 2 MW class (1.5–2 MW output, 3.5 hours or more). Total for one project including land and the grid connection cost contribution, excluding tax. Medians of the asking prices we saw from January to September 2026, collapsed to one figure per project (our observation — not published statistics). There are only nine connected projects, so read that figure as a band. All amounts are in Japanese yen; no currency conversion has been applied.

Three rungs: rights ¥120M, completed handover ¥670M, connected ¥750M

What we counted are the asking prices we saw between January and September 2026: the prices we quoted to buyers, and the prices at which other companies' projects were offered through brokers or developers, collapsed to one figure per project. Offers made to us by sellers, and our own purchase prices, are not included. This is the going rate for the price tags that actually land in front of a buyer.

High-voltage 2 MW class, total for one project including land and the grid connection cost contribution, excluding tax. Our observation (January–September 2026). Project counts collapse several quotes for the same project into one. The three disclosed figures are given as disclosed, including "approx.". The original Miyama disclosure states the capacity as "approx. 8 MW". The Miyama and Shokawa disclosures do not say how the land is treated or whether the price includes tax.

The disclosed acquisition prices sit almost on top of our bands. Shokawa at approx. ¥700 million (approx. 2 MW / 8 MWh, to be acquired in November and connected in December) and Miyama at approx. ¥684 million (approx. 2 MW, acquired with the equipment installed and awaiting connection) sit right beside the pre-connection median of ¥670 million. Kirishima's ¥120 million (the price paid for the project site and the generation rights) is the rights median exactly. Most listed-company disclosures hide the amount behind wording such as "less than 30% of net assets", so these three are the only public figures we could verify — but they are not far from the price tags we see.

There is a gap between the price tag and the price a deal actually closes at. In our observation, closing prices often land a little below the median asking price, within a margin of about 10%. Buyers who bid with that in mind, and sellers who set their price tag with it in mind, get to a deal faster.

The gap between rights and completed handover is mostly construction cost

What is the ¥550 million between ¥120 million for rights and ¥670 million on completed handover? Mostly construction cost. The government study group puts the FY2024 domestic average at ¥68,000/kWh (storage system ¥54,000 plus works ¥14,000; see COLUMN 40), which comes to ¥540 million for 8 MWh. In COLUMN 39 in August we put the hard cost at ¥350–450 million. Among public examples, Kirishima's total development cost is ¥570 million (excluding tax; the disclosure does not say whether it includes the ¥120 million acquisition cost). Add the ¥540 million domestic average to ¥120 million of rights and you get ¥660 million — almost the same as the completed-handover median of ¥670 million.

In other words, the completed-handover price tag is at about the same level as the cost of building the plant at the domestic-average build cost. Most sellers are developers or EPC contractors, and their margin is inside the build cost (the margin on equipment procurement and on the works). It is not a separate layer sitting on top of the price tag. The grid connection cost contribution sits outside the build cost, so the larger the contribution, the less the seller keeps. Plants built with a subsidy need the approval of SII (the body that administers the subsidy), and repayment, if they are transferred within the disposal restriction period set by the statutory useful life, so they rarely come up for sale on completed handover. We treat build cost as unsubsidized.

For a buyer, it means this. Buy the rights for ¥120 million and build it yourself, and you still pay an EPC contractor, whose margin sits in the build cost just the same. The difference lies in how far you can bring the build cost down through your choice of EPC and your equipment procurement — and in whether you carry the risks yourself: the final cost contribution, construction delays, and equipment performance. If you have the team to choose the EPC, manage the schedule and see the plant through to completion, buy rights; if not, buy on completed handover. It is not a decision to make on the price gap alone.

The premium only shows clearly once the plant is connected. Against ¥670 million before connection, connected and operating plants ask ¥750 million: ¥80 million, or about 10%, more. Not one yen of equipment has been added. What has been added is the fact that it is connected, and the disappearance of the wait before it earns in the markets. A buyer who can wait for connection saves ¥80 million by buying before it; a buyer who wants the plant in before its fiscal year-end uses that ¥80 million to buy time. There are only nine connected projects, so read this as a band.

From the seller's side, a plant you build and then sell carries a price tag at about cost, and your margin is taken as the construction margin inside the build cost. The only premium that reliably goes on top of the price tag is the ¥80 million after connection. If you can hold until connection, it is worth holding.

How this differs from the bands we gave in August

In COLUMN 39 in August we wrote: rights ¥150–200 million, completed handover ¥700–800 million, and ¥800 million to ¥1.0 billion with a balancing-market track record. Those were the bands of the offers we had seen at that point. Recounting nine months of listings, one figure per project, the median on every rung sits at the bottom of the August band or a little below it. As for ¥800 million to ¥1.0 billion with a track record, on this count our observations reach only the top of the connected range (¥840 million). These figures are closer to the actual market, so from now on we use the numbers in this article. There are still few observations since 1 September, when the price cap for primary reserve, secondary reserve ① and the composite product fell to ¥10, so it is too early to read any price movement from September onward.

What moves the price is connection, not the service area

We often hear "it's a Tokyo project, so it's expensive" or "Tohoku is cheap." Split the listings by service area and there is almost no difference.

Figure 3 as a table (by service area)

Asking prices on completed handover (connected plants included) — by service area. High-voltage 2 MW class, total per project, excluding tax, our observation

Service area	Median	Range	Observed
Tokyo	¥700M	¥570–860M	43 projects
Kyushu	¥680M	¥500–900M	33 projects
Chubu	¥670M	¥480–840M	47 projects
Kansai	¥670M	¥550–790M	18 projects
Chugoku	¥670M	¥620–730M	7 projects
Tohoku	¥650M	¥560–800M	20 projects
Hokuriku	¥550M	¥550–780M	3 projects

Shikoku and Hokkaido had only one or two projects each and are not shown. The 15 projects whose service area is unknown are not included in the table.

Across the six areas with enough projects to count, from Tokyo at ¥700 million to Tohoku at ¥650 million, the spread is ¥50 million. Hokuriku is lower at ¥550 million, but that is three projects — too few to read as a band. Rights tell the same story: Tokyo ¥120 million, Chubu ¥100 million, Kansai ¥130 million (three projects), Kyushu ¥110 million, Chugoku ¥120 million — all within ¥100–130 million, and the order of the areas does not match the one for completed handover. In COLUMN 35 we showed that annual revenue for the same 2 MW differs widely by area, with Tohoku on top at about ¥290 million. The revenue map and the price-tag map do not line up. For buyers, that means a project in Tohoku or Chugoku earns more for the same price. For sellers, it means there is no reason to accept a discount because of the service area.

What does move the price is connection.

Projects connecting in FY2026 and projects connecting in FY2027 ask the same ¥670 million; once connected, ¥750 million. It is not the fiscal year that changes the rung but whether the plant is connected. Extra-high-voltage is the opposite: there, the unit price moves with the year of connection. Rights listings have a median of ¥22,000/kW (range ¥3,000–41,000/kW, 23 projects); projects connecting in FY2027 and FY2028 ask ¥27,000–31,000/kW, and FY2029 or later ¥16,000/kW. The counts are small (4, 5 and 10 projects respectively), so read these as bands. There are too few extra-high-voltage projects on completed handover to give a band.

Figure 3 as a table (by connection timing)

Asking prices on completed handover — by connection timing. High-voltage 2 MW class, total per project, excluding tax, our observation

Connection timing	Median	Range	Observed
Connected	¥750M	¥660–840M	9 projects
Connecting within FY2026	¥670M	¥500–900M	76 projects
Connecting in FY2027	¥670M	¥480–730M	57 projects
Connecting in FY2028	¥700M	¥570–700M	7 projects

Connection timing is the Japanese fiscal year (April to the following March) of the planned start of operation. There are only nine connected projects, so read that row as a band.

What to check inside the band

Counting the wording attached to the offers, these are the conditions that separated the top of the band from the bottom, in this order. A project that meets the conditions that pulled prices down gives you grounds to bid at the bottom of the band. A project with every condition that pushed prices up sells first, even at the top of the band.

This overlaps with the five factors in our article on land (COLUMN 26JA), but for projects the order changes. Before the land conditions come into it, the price moves on connection, the cost contribution and how far the procedures have got. Once you have seen the price, five checks are enough.

1. Is the price inside the band? Around ¥670 million on completed handover, around ¥750 million once connected. The service area is no reason for the price to differ; what differs by area is revenue.
2. Who paid how much of the deposit and the cost contribution? This is what decides the rung. Look at the receipts and the notice of the final amount in the original.
3. Has proof of land-use rights been submitted? For contract applications from 1 October, it must be submitted within two months of interconnection approval; if it is not, the connection reservation is cancelled (see below).
4. What carries over? The capacity market by consent; the balancing market and JEPX (the wholesale power exchange) need new contracts in an asset deal; the name registered with the transmission and distribution utility (the TSO) needs an enquiry everywhere except Kyushu; subsidized equipment needs approval, and possibly repayment (next section).
5. Do not assume immediate depreciation on used equipment — not until its treatment under the Bold Investment Promotion Tax Incentive can be confirmed from published sources (see the tax section).

The technical items are in The technical checks before you buy (COLUMN 37); the sequence and the documents are in How a battery storage plant changes hands (COLUMN 46).

Four ways to buy: what carries over, and whose consent you need

There are four ways to buy: buy the rights, buy on completed handover, buy an operating plant as assets, or buy the equity of the company that holds the project (the SPC). Below is what moves and whose consent it needs, as far as we could confirm from published rules. Of the ten TSOs, Kyushu Electric Power Transmission and Distribution is the only one whose name-change form we could confirm is published. For the others, you enquire with the TSO project by project.

The rules behind the table, by type of deal:

In COLUMN 39 in August, we wrote that the qualifications and positions that pass by rule are almost limited to the capacity market, and that the balancing market has no express provision. To add to that: Article 26 of the balancing-market model contract does provide for succession on a merger or business transfer, with written notice and the counterparty's approval, and Article 19 of the JEPX Trading Member Rules provides for succession on a merger or company split. Neither, however, covers an asset deal in which you buy only the equipment. If you buy an operating plant as an asset deal, budget on the basis that you will sign new balancing-market and JEPX contracts yourself — the conclusion from August stands. The ¥80 million premium on a connected plant pays for the fact that it is connected, not for market contracts coming with it.

Tax for the buyer, net proceeds for the seller

On the buyer's side, two tax questions bear on the price. The first is immediate depreciation. The SME Management Enhancement Tax Incentive and the SME Investment Promotion Tax Incentive apply only to new equipment, and the electricity business is not among the designated businesses (COLUMN 45). Under the Bold Investment Promotion Tax Incentive, open for applications since 31 July, how equipment bought on completed handover or as an operating plant is treated as "used" cannot yet be confirmed from published material. If you are pricing on the assumption of immediate depreciation, confirm this first through the confirmation application. The second is subsidies. Transferring subsidized equipment within the disposal restriction period (the statutory useful life) requires SII's prior approval, and the subsidy may have to be repaid (Article 22 of the Act on Regulation of Execution of Budget Pertaining to Subsidies, etc.; SII application guidelines). That is why the first thing to ask a seller is whether there is a grant decision notice.

The seller's net proceeds bear on the negotiation too. An individual who sells land held for five years or less pays the short-term rate — twice the long-term one — so some sellers want to wait out the five years. For a corporate seller the gain is taxable income, so they care which side of their fiscal year-end the deal settles on. The table shows only rates and categories. How they apply differs from case to case, and this article is not tax advice. Please confirm the treatment with your tax adviser.

Sources: National Tax Agency Tax Answer Nos. 6201, 3105, 3208, 3211, 1463, 5759, 7140 and 7108; the Basic Circular on the Stamp Tax Act; NTA Q&A cases. For fiscal years beginning on or after 1 April 2026, the Special Defense Corporate Tax is added (4% of the base corporation tax — corporation tax before certain tax credits — less a basic deduction of ¥5 million a year). Individual income tax carries the 2.1% Special Income Tax for Reconstruction on top.

Who is buying

Between October 2025 and September 2026, the entities that announced acquisitions of, or investments in, grid-scale battery storage plants were, as far as we could confirm: six listed operating companies (JALCO Holdings, Tamagawa Holdings, J-Holdings, Glome Holdings, Correc Holdings and AD Works Group), two power companies (eREX and Energy Power), and two funds (Kan-denchi Fund No. 1 of Kansai Electric Power, Kinden and MUFG; and SBI Money Plaza). JALCO alone announced four purchases at different rungs within a year: a completed plant, land and rights, a TK interest in an SPC, and an operating plant. The motives stated in the disclosures were "to start the business early" (Tamagawa) and "a stable revenue opportunity of a highly infrastructural nature" (Glome); none of the disclosures whose full text we could check gave immediate depreciation as a motive.

Entities that announced acquisitions or investments between October 2025 and September 2026 (timely disclosures and press releases; some confirmed through reprints). Rungs as stated in the disclosures. Not an exhaustive count.

While buyers multiply, the supply of projects is narrowing. Since 1 August 2026, the number of connection studies a single operator may apply for has been capped (Hokkaido 5, Tohoku 6, Tokyo 11, Chubu 7, Hokuriku 5, Kansai 12, Chugoku 5, Shikoku 5, Kyushu 8). As new responses become harder to obtain, projects that already hold one become scarcer. For buyers, that means the projects inside the band with every condition in place go first. It comes down to when you submit your letter of intent.

From the seller's side — documents, deadlines, and whether we buy or broker

Now the seller's side. The price is the same band buyers are paying, and deals close a little below the median. A seller can do two things: show on paper the conditions that put a project at the top of the band, and not let the deadlines pass.

Nearly half the projects that reach us come with none of the documents — response, application receipt, deposit, cost contribution, land, equipment. The most common form is an asking price and nothing else. That is where the buyer's desk review stops. Before you put a project on the market, assemble these seven.

1. The connection study response (a PDF of the original). The response date is where its one-year validity starts.
2. Your copy of the contract application, and the receipt for the deposit. If you have got this far, the project moves up a rung.
3. The notice of the final grid connection cost contribution. Is it still indicative, or final? Projects with a final figure price at the top of the band.
4. The land registry, or the lease. If leased, check in the contract whether a transfer needs the landowner's consent.
5. The equipment models. The PCS and EMS models, and their JC-STAR registration status. For high-voltage, this becomes a requirement for connection contract applications from April 2027 onward (COLUMN 14).
6. If the plant is operating, its record of clears and revenue. The aggregator's monthly statements are enough.
7. If it received a subsidy, the grant decision notice. Approval is needed before a transfer.

The deadlines are published. A project left sitting loses its value when they pass.

Open the table of deadlines and sources

Deadline	If you miss it	Basis
One year from the connection study response	The response lapses, and a new study is needed before a contract application. The study fee is ¥200,000 + tax per point of receipt per study (¥220,000 at TEPCO Power Grid), and a response takes about three months in principle	OCCTO Network Operation Guidelines, Article 89(1)(vi); each TSO's application guidance
Deposit at contract application 10% of the indicative cost contribution (from 1 April 2026)	Forfeited if you withdraw for your own reasons. Applies to contract applications received from April 2026	ANRE Next-Generation Power Grid WG, 6th meeting, Material 3 (24 December 2025) and 7th meeting, Material 1-1 (9 February 2026); OCCTO notice (18 March 2026); TEPCO Power Grid notice (1 April 2026)
Paying the cost contribution in instalments	At least 50% of the total in the first instalment	Same as above
Two months from interconnection approval proof of land-use rights	If it is not submitted, the connection reservation is cancelled. Applies to contract applications from 1 October 2026. On 22 July, METI approved amendments to the wheeling service terms of all ten TSOs (including making land documents a requirement at the connection study application). Tokyo, Hokuriku and Shikoku have published the 1 October effective date; for the others, check each TSO's notice	Same WG (10 June 2026); METI news release
Cap on connection studies per operator from 1 August 2026	Applications above the cap are not accepted. Hokkaido 5, Tohoku 6, Tokyo 11, Chubu 7, Hokuriku 5, Kansai 12, Chugoku 5, Shikoku 5, Kyushu 8	Same WG; notices from TEPCO Power Grid and Hokkaido Electric Power Network
Connection contract applications from April 2027 (high-voltage)	JC-STAR ★1 becomes a connection requirement. Publication of the revised guidelines could not be confirmed as of the end of September 2026	Next-Generation Power Grid WG (9 February 2026)

Once a year has passed since the response, the project disappears — scarcity and all. Decide among the three options — develop it yourself, sell the development rights, or withdraw (You have a connection study response but cannot proceed) — counting from the response date.

Whether we buy a project ourselves or introduce it to a buyer depends on the project, and there are no fees until a deal closes. Send us five things — the location (prefecture and service area), the output, how you want to sell, the stage of the connection study response, and the status of the cost contribution — and our first reply will tell you whether we are likely to buy, whether we would introduce a buyer instead, and which documents are missing (contact us).

Frequently asked questions

Primary sources consulted, and related articles