The divergence between Serbia and Romania in the 2025–2028 period marks one of the most consequential structural shifts in South-East Europe’s power system. While both countries entered the decade with comparable roles as regional anchors—large thermal fleets, significant hydro assets, and strong cross-border interconnections—their trajectories have separated sharply as Romania accelerates coal retirements and Serbia maintains dispatchable baseload continuity. This divergence is now visible not only in national adequacy metrics but in cross-border flows, price formation, and the risk profile of the entire Balkan–Central European interface, as consistently reflected in seasonal system assessments by ENTSO-E.
Romania’s power system is undergoing a structural contraction in dispatchable capacity. By early 2026, approximately 1.7 GW of lignite-fired generation is scheduled to exit the system, concentrated primarily in Oltenia. These units historically provided baseload energy and, more importantly, winter reliability during cold spells when hydro inflows weaken and wind output is uncertain. Their removal compresses reserve margins at precisely the moment when electrification, heat-pump deployment, and industrial demand recovery are pushing peak loads upward. Romania’s peak winter demand now regularly approaches 9.0–9.5 GW, while dependable dispatchable capacity is shrinking faster than replacement assets are commissioned.
Serbia, by contrast, enters the same horizon with a largely intact thermal base operated by Elektroprivreda Srbije, exceeding 4.4 GW of installed lignite capacity. While ageing, these units continue to deliver dependable output with marginal production costs in the €25–35/MWh cash OPEX range, excluding carbon. Hydro capacity above 3.0 GW provides additional peak modulation. The result is a system that can still meet winter peaks of 7.5–8.0 GW without structural import dependence. This asymmetry between Romania’s tightening margins and Serbia’s relative surplus is now reshaping regional power flows in ways that were far less pronounced even five years ago.
Historically, Romania functioned as a net exporter toward Hungary and, intermittently, toward Serbia and Bulgaria, particularly during periods of strong hydro output on the Danube and in the Carpathians. As coal units retire, this pattern becomes increasingly episodic. During normal hydrology and high renewable output, Romania can still export. During cold, dry, or wind-poor conditions, however, the country transitions rapidly into a net importer. This switch places pressure on north-south corridors linking Romania with Hungary and Serbia, and it alters the economic logic of regional dispatch.
Serbia’s role in this new configuration is subtle but decisive. When Romania’s system tightens, Serbia is no longer merely another Balkan neighbour competing for imports; it is one of the few adjacent systems capable of sustaining domestic balance without drawing heavily on the same constrained corridors. In practical terms, this means that Serbia often avoids importing at precisely the moments when Romania, Bulgaria, or Hungary need capacity most. The absence of Serbian import demand during stress periods effectively frees cross-border capacity for Romanian balancing, reducing the probability of cascading congestion and price spikes.
Price dynamics reflect this shift. Romanian day-ahead prices have become increasingly volatile, with sharper upward spikes during winter stress events. Serbia’s prices, while rising in absolute terms, remain more anchored to lignite marginal costs during such periods. The spread between Romanian and Serbian wholesale prices during cold spells has therefore widened, occasionally exceeding €20–30/MWh. These spreads are not structural arbitrage opportunities in the classical sense, as transmission constraints limit full convergence, but they signal a deeper reallocation of scarcity rents within the region.
For investors, this reallocation matters. Romania’s system is moving toward a profile where scarcity pricing becomes more frequent, increasing revenue potential for flexibility assets such as storage and gas peakers, but also increasing risk for energy-intensive consumers. Serbia’s system, conversely, monetises reliability indirectly by dampening volatility rather than amplifying it. This difference has implications for industrial location decisions, long-term power purchase agreements, and grid-service investments across the region.
The transmission interface between Serbia and Romania becomes a focal point in this context. Multiple 400 kV interconnections link the two systems, forming part of a broader east-west and north-south flow network. While technical transfer capacity is substantial, commercial capacity is often constrained by N-1 security requirements and internal bottlenecks on both sides of the border. As Romanian coal exits accelerate, these constraints become binding more often, particularly during simultaneous stress in Hungary and Bulgaria. Serbia’s ability to remain internally balanced reduces the probability that it will exacerbate these bottlenecks, but it also highlights the value of targeted grid reinforcements.
Indicative CAPEX for strengthening the Serbia–Romania interface, whether through new lines, series compensation, or substation upgrades, typically falls in the €200–400 million range for projects of regional significance. Such investments are increasingly justified not by national adequacy alone, but by avoided congestion costs and reduced price volatility across multiple markets. In effect, Romania’s coal retirements increase the system value of every additional megawatt of dependable transfer capacity from Serbia.
The divergence also exposes contrasting carbon risk profiles. Romania internalises EU ETS costs fully, pushing coal out of the merit order and accelerating retirements. Serbia, outside the EU ETS, retains lignite’s cost advantage in the short term. However, as CBAM mechanisms mature and regional market coupling deepens, Serbian exports will increasingly face implicit carbon pricing. This creates a temporal arbitrage: Serbia’s lignite fleet is most valuable to the region precisely in the window before carbon convergence fully materialises. For investors and policymakers, this suggests a finite period during which Serbia can leverage its baseload strength to finance transition investments without undermining adequacy.
Operational risk further differentiates the two systems. Romania’s transition increases dependence on variable renewables and gas, exposing the system to correlated weather and fuel risks. Serbia’s risk profile is more concentrated around asset reliability and fuel logistics. Sustaining lignite output requires continuous mining CAPEX and disciplined maintenance. Annual O&M and sustaining CAPEX across Serbia’s thermal and mining complex are estimated at €300–450 million, a figure that directly underwrites not only domestic adequacy but regional stability. Any failure in this domain would reverberate more strongly now that Romania’s buffer capacity is gone.
Hydropower complicates the picture. Both countries rely on the Danube, yet hydrological conditions increasingly exhibit correlated stress during winter cold spells, when inflows are low and reservoirs are managed conservatively. Romania’s reduced thermal flexibility under such conditions amplifies its vulnerability, while Serbia’s retained baseload provides a counterweight. This asymmetry reinforces Serbia’s role as a stabiliser during precisely those events when regional risk is highest.
From a strategic perspective, the Serbia–Romania divergence is not merely a transitional phase; it is a preview of how uneven decarbonisation reshapes regional power systems. Systems that exit coal rapidly without equivalent dispatchable replacements become import-sensitive and price-volatile. Systems that retain baseload longer assume a stabilising role but accumulate carbon exposure. The regional equilibrium depends on how these opposing trajectories interact over time.
For Serbia, the implication is clear. Its relative advantage is not permanent, but it is material in the near to medium term. Converting this advantage into durable value requires deliberate action: reinforcing cross-border grids, investing in flexibility that can eventually replace lignite’s system services, and structuring regional contracts that monetise reliability. For Romania, the challenge is to replace lost baseload not just in energy terms, but in adequacy and system-service terms, without importing instability from neighbours.
As ENTSO-E’s seasonal assessments continue to highlight Romania’s tightening margins and Serbia’s relative resilience, the regional map of power flows is being redrawn. The centre of gravity is shifting southward and eastward, toward systems that still control dispatchable capacity. In this evolving landscape, Serbia is no longer simply adjacent to Romania’s transition; it is one of the key variables that determines whether that transition unfolds smoothly or under persistent stress.