Decision brief · World
Independence changed more than the seller
The Baltic case separates a commercial exit, a change in shared operation and the services needed to make that change workable.
- The trading break came before the operating breakRussian commercial electricity imports stopped in May 2022. The Baltic systems remained synchronously connected until February 2025.
- Independence required a new partnershipThe Baltic grids joined Continental Europe through Poland, with supporting infrastructure and reserve capabilities.
- An installed battery and a purchased service are different thingsLithuania’s 200 MW system supported a separate balancing service of up to 40 MW. That ceiling does not measure actual energy delivery.
In May 2022, the Baltic states stopped importing Russian electricity. Almost three years later, they disconnected their power systems from the Russian synchronous area. Both events were described through the language of independence. They removed different dependencies.
That distinction changes how the transition should be judged. A country can stop buying electricity from a supplier while retaining an operating relationship with its system. It can also leave that operating relationship while continuing to import electricity from elsewhere. A single domestic-generation percentage cannot tell these stories apart.
The useful case therefore runs beyond the moment of disconnection. It connects the two exits with a more modest-looking Lithuanian arrangement: a temporary balancing service from a publicly owned battery operator. Together, these documents show that the freedom to leave an old partnership depends on what replaces its practical functions.
The commercial exit left an operating relationship behind
Latvia’s transmission operator AST announced that Russian electricity imports into the Baltics were suspended from 22 May 2022. Its statement made an explicit distinction: commercial imports had stopped, but parallel synchronous operation with the Russian system continued. Technical transit and balancing flows remained part of that relationship.
Trade answers who sells electricity. Synchronous operation concerns how connected systems run together and maintain frequency. Ending a commercial transaction does not automatically redesign those operating arrangements. Treating the 2022 event as a complete physical separation would erase the remaining task.
Poland provided a route into a different operating partnership
On 8 February 2025, Estonia, Latvia and Lithuania disconnected from the Russian IPS/UPS system and entered isolated operation for a test. On 9 February, they synchronised with Continental Europe. ENTSO-E confirmed the successful connection and the permanent disconnection of the electricity interconnections with Russia and Belarus.
ENTSO-E’s account identifies Poland’s PSE as the physical connection to Continental Europe and a central coordinator of the transition. It also describes infrastructure upgrades in Poland and the Baltic states. The resulting arrangement shares frequency management with Continental European partners, replacing the previous reliance on the Russian and Belarusian systems.
There was also a trade-off inside the new relationship. Litgrid’s 2025 half-year report, page 9 says trade with Poland decreased after synchronisation because LitPol Link became more restricted for commercial use to support safe system operation. Changing the operating partnership did not automatically increase the capacity available to traders.
The political implication is more precise than a claim of complete self-sufficiency. The Baltics changed the group with which they had to cooperate. That gives the exit geopolitical meaning without pretending that an interconnected electricity system becomes a solitary national machine. Poland’s position also entails operating responsibilities; physical centrality alone does not prove the size of any diplomatic concession it can obtain.
This connects with the distinction between cooperation and control in Poland’s rearmament debate. Shared infrastructure or industrial relationships can increase a country’s freedom of action. Whether they do so depends on the particular obligations and capabilities being shared.
A reserve asset is different from the service bought from it
Lithuania supplies a concrete example of the capabilities behind the transition. Energy Cells, owned through the state-owned EPSO-G group, commissioned a storage system rated at 200 MW and 200 MWh to provide an isolated-operation reserve. Those two ratings describe different physical quantities.
Three numbers that answer different questions
| Figure | What it describes | What it cannot establish |
|---|---|---|
| 200 MW | Installed power rating of the storage system | Continuous delivery of all that power |
| 200 MWh | Installed energy capacity | The battery’s charge at a particular moment |
| Up to 40 MW | Reported temporary balancing-service capacity | Actual balancing energy delivered in MWh |
The service ceiling is 40 ÷ 200 = 20% of the installed power rating. That is a comparison between two power figures. It is not a utilisation rate, a measure of energy dispatched or proof that the remaining 160 MW was simultaneously available for another task. Those claims would require operating data and service constraints.
The same attention to units matters in reading a thermal battery’s storage duration, although that is a different technology and question. Here the distinction that matters politically is between owning an asset and securing a particular service from it. A large project announcement cannot substitute for the terms under which the operator may actually use it.
A temporary public service can leave room for a market
Energy Cells’ 2025 integrated annual report, pages 20–23 describes up to 40 MW of balancing capacity supplied after synchronisation. The service uses aFRR, an automatically activated frequency-restoration reserve. Under the reported 2025 arrangement, its balancing-energy bid came at the end of the auction and its batteries were activated when other participants could not supply the service. The report also describes conditional continuation into 2026, ending when sufficient market provision became available or at year-end. This describes the reported arrangement, not its live status in September.
The design suggests a useful distinction in energy sovereignty debates. Public ownership and a competitive supply of operating services need not be mutually exclusive. A reserve provider can cover a gap while commercial alternatives develop. The crucial questions concern when that provider is used and what evidence would allow its temporary role to shrink.
Imports continued after the Russian synchronous relationship ended
The strongest counterexample to equating independence with producing everything at home comes from Lithuania’s own figures. In February 2026, Litgrid reported preliminary data for 2025: domestic power plants covered 74% of electricity demand, with 26% covered by imports. These are annual Lithuanian figures, not a Baltic-wide result or an hourly adequacy test.
Those imports do not undo the change in synchronous partnership. They show why the supplier relationship, operating relationship and contingency arrangements must be evaluated separately. Conversely, even a much higher annual generation share would not, by itself, demonstrate that sufficient power and reserves were available during every difficult hour.
Broader accounts of Europe’s supply-chain realignment need this kind of sector-specific qualification. A border crossing can carry exposure to an unwanted dependency or access to a chosen alternative. The volume crossing it does not resolve which relationship matters.
The Baltic transition is therefore a case of replacing an operating dependency through deeper cooperation elsewhere. Its lesson for another proposed energy exit is concrete: identify the function that the old arrangement still performs, the partner or asset that will replace it, and the conditions under which the replacement service is available. Until those questions are answered, a promise to stop buying from someone describes only part of the work.
Operator statements and the Energy Cells report
Primary documents checked on 14 September 2026. This is a document-based worked case with a checked 40/200 power comparison, not a physical grid experiment. No outage probability, live dispatch, price saving or September 2026 service availability was measured. The 74/26 split is explicitly preliminary and limited to Lithuania in 2025.
- AST — As of Sunday, Russian electricity no longer enters the European market, 22 May 2022
- Litgrid — Baltic power systems disconnected from Russian IPS/UPS, 8 February 2025
- ENTSO-E — Successful Baltic synchronisation, 9 February 2025
- Energy Cells — Storage system commissioned at 200 MW
- Energy Cells — Integrated Annual Report 2025
- Litgrid — Interim report for the six months ended 30 June 2025, management report page 9
- Litgrid — Record Breakthrough in Renewable Electricity Generation, 9 February 2026
