Pumped storage is returning to the centre of SEE energy planning because the region’s renewable ambitions are now colliding with grid reality. Solar and wind pipelines are growing, but without long-duration flexibility, the power system faces curtailment, price cannibalisation, balancing pressure and rising dependence on imports during low renewable output. That is why projects such as Serbia’s Đerdap 3, North Macedonia’s Čebren and Serbia’s Bistrica are again becoming strategic infrastructure rather than legacy hydropower ideas.
The Serbian-Romanian Danube system is the most important regional case. Romania is assessing Serbia’s Đerdap 3, also known in Romanian as Porțile de Fier 3 or Iron Gates 3. The project would be located on the Serbian side of the Danube, upstream of Đerdap 1, and Bucharest is reviewing economic, power-system and environmental impacts. Romania has indicated openness, while Hidroelectrica could potentially participate with a 50% stake once Serbia provides the necessary documentation.
The regional scale is already visible. Serbia and Romania share Đerdap 1, with 2,300 MW, and Đerdap 2, with 520 MW. A third storage-based facility would sit inside one of the Balkans’ most important hydro corridors. For Serbia, it would support renewable integration and system security. For Romania, it could contribute to balancing a rapidly changing generation mix. For the region, it could become a Danube flexibility asset.
North Macedonia’s Čebren is another example of the same logic. The government is close to deciding to develop the pumped storage hydropower plant independently through state-owned ESM, after repeated failed efforts to secure a private partner. The project is estimated at €1.5bn including interest, with government calculations pointing to an internal rate of return above 10%. It has been considered in the 333–458 MW range, with annual output of 1–1.2 TWh.
Serbia’s Bistrica adds a further layer. The project is designed with two 170 MW reversible turbine-generator sets and two 170 MW conventional units on the Uvac and Lim rivers, with works expected to begin in 2028.
The investor message is that pumped storage is becoming the grid’s insurance policy. Batteries can capture short-duration spreads and ancillary revenues, but pumped hydro offers depth, duration and system inertia. In SEE, where grids are weaker, interconnectors are constrained and RES growth is uneven, storage will define the bankability of the next renewable cycle.
