As Serbia’s electricity sector approaches the 2026–2032 timeframe, it is poised to undergo significant transformations driven by external market dynamics, increased renewable energy integration in neighboring EU markets, and inherent grid constraints. The focus has shifted from merely assessing the need for dispatchable capacity to determining the optimal form of dispatchability that minimizes fiscal risks, curtailment exposure, and overall system costs amid declining baseload economics.
The current framework of Serbia’s dispatchable energy supply is heavily reliant on aging lignite power plants managed by EPS, supplemented by hydropower sources that are becoming increasingly volatile year-on-year. Despite the marginal role of gas-fired capacity, changes in cross-border pricing dynamics have already begun to influence dispatch strategies. Electricity imports from Hungary, Romania, and Croatia are increasingly dictating marginal pricing during peak solar generation periods, while scarcity pricing predominantly arises during evening demand spikes and winter peak times. This evolving price structure is critical for evaluating asset performance.
A preliminary stress test for Serbia’s power system covering the 2026–2032 period indicates a dramatic reduction in full-load hour expectations for thermal assets. Historically operational for 6,000 to 6,500 hours annually, lignite units are now projected to operate between 3,500 and 4,200 hours due to increasing competition from regional solar capacity. In contrast, new combined-cycle gas turbines (CCGTs) may struggle to achieve more than 2,000 to 2,500 hours under merchant conditions, which would hinder their ability to recover capital expenditures without long-term contracts or capacity payments. Battery storage technology offers a different approach by capitalizing on market volatility rather than sheer volume, performing optimally where traditional thermal assets falter.
The financial implications of these developments are stark. Constructing a modern CCGT with a capacity of 400–500 MW in Serbia could require an investment ranging from €350 million to €450 million upfront, not accounting for necessary gas infrastructure upgrades or fuel hedging expenses. Even under favorable market conditions, annual earnings before interest, taxes, depreciation, and amortization (EBITDA) would likely be unstable and subject to significant downside risk as solar generation in Hungary and Romania approaches a threshold of 30% to 35% of total output—a point that is nearing realization. The risk landscape is further complicated by potential indirect carbon pricing impacts on Serbian electricity prices through imports.
While coal reserve strategies may initially appear more cost-effective, they conceal underlying inefficiencies. Maintaining lignite units in either cold or strategic reserve circumvents immediate capital expenditure but reallocates costs towards maintenance and staffing requirements alongside forced outages. The estimated annual reserve cost for Serbia’s older lignite facilities could reach €70 to €90 per kW per year, translating into an annual financial burden of approximately €140 million to €180 million for a 2 GW reserve fleet. Notably, these costs do not enhance operational flexibility; they merely ensure availability with extended start-up times and inadequate ramping capabilities—characteristics that do not align with the evolving demands of the power system.
In contrast, battery energy storage systems (BESS) present a compelling alternative. A proposed utility-scale battery installation with a capacity of 1 GW and 4 GWh would necessitate an initial investment between €700 million and €850 million at current market rates but could decrease to between €550 million and €650 million by the late 2020s. Although the upfront capital requirement appears higher than that for gas peakers or coal reserves, the overall system value derived from batteries is significantly greater. These systems can effectively capture intraday price fluctuations, diminish reliance on imports during high-demand periods, reduce reserve margins, and postpone necessary grid enhancements. Over a lifespan of 15 years, the net cost per avoided MWh during scarcity conditions is lower than that associated with both gas peakers and coal reserves.
Grid limitations further exacerbate these issues. Serbia’s north-south transmission bottlenecks and restricted dynamic line ratings hinder its ability to capitalize on low-cost imports or export excess hydropower generation. In this context, strategically positioned batteries outperform centralized thermal plants by minimizing redispatch costs at congested points within the grid. Conversely, gas peakers situated away from these congestion zones lose their operational flexibility while maintaining coal reserves fails to alleviate any congestion challenges.
From a fiscal perspective, the hierarchy of options is evident: coal reserves contribute ongoing costs without flexibility benefits; gas peakers risk becoming stranded assets within a decade; whereas battery storage requires upfront capital but stabilizes operational expenditures while mitigating exposure to volatile import prices. Under conservative projections, Serbia’s optimal approach for minimizing system costs leans towards accelerating battery storage deployment while gradually reducing reliance on older lignite reserves rather than investing in new baseload gas generation.
This analysis underscores a crucial point: Serbia’s evolving grid dynamics favor speed and flexibility over traditional fuel throughput metrics. Future capacity remuneration mechanisms must therefore prioritize technology-neutral frameworks weighted towards flexibility; otherwise, they risk entrenching public investments in diminishing asset classes.
