Elektroprivreda Srbije has adopted a conservative electricity demand outlook for the period 2026–2028, projecting that total national electricity consumption will increase by approximately 1 percent per year over the three-year horizon. This assumption, embedded in the company’s medium-term business plan, reflects expectations of steady but moderate economic growth, gradual electrification of end-use sectors, and incremental efficiency gains across industry and households.
At the same time, the plan anticipates a parallel decline of roughly 1 percent per year in electricity production from EPS’s own generation assets, primarily driven by ageing thermal capacity, tighter environmental constraints, and the gradual retirement or reduced utilisation of lignite-fired units. The combined effect of rising demand and falling internal supply results in a widening structural gap that becomes most visible by 2028, when the shortfall between projected consumption and EPS production is estimated at around 6.5 percent of total national demand.
In absolute terms, Serbia’s annual electricity consumption currently fluctuates around 33–35 terawatt-hours, depending on hydrology and industrial activity. A sustained 1 percent annual increase would lift total demand by roughly 1.0–1.1 TWh over the three-year period. Simultaneously, a 1 percent annual decline in EPS generation would reduce internal output by a similar magnitude. By 2028, this arithmetic implies a net supply gap of approximately 2.0–2.3 TWh per year, which EPS planning documents assume will be covered through market imports rather than accelerated domestic capacity expansion.
This approach effectively formalises Serbia’s transition from a system that oscillates between self-sufficiency and export surplus into one that structurally relies on regional electricity markets during normal hydrological years. While such reliance is manageable under stable market conditions, it exposes the system to price volatility during periods of regional scarcity, particularly winter peaks or drought-affected summers.
Public debate around the EPS plan intensified after senior political leadership questioned whether a 1 percent annual demand growth assumption adequately captures Serbia’s evolving consumption profile. Particular attention has been drawn to data centres, digital infrastructure, and high-load industrial facilities, whose electricity demand characteristics differ sharply from traditional residential or manufacturing users. These consumers typically require continuous, high-availability power with limited tolerance for interruption, effectively behaving as base-load demand rather than flexible consumption.
Estimates presented in public discussions suggest that a single advanced industrial or digital facility could require 500–600 megawatts of firm capacity, equivalent to the output of two large thermal units at the Nikola Tesla A power plant. Even a small number of such investments would materially alter Serbia’s demand trajectory, pushing annual consumption growth above the 1 percent baseline assumed in EPS planning.
From a system-planning perspective, this raises a critical question: whether Serbia’s power strategy is being calibrated to historical demand patterns rather than to the structure of future growth. Electrification of transport, gradual replacement of fossil fuels in heating, and the expansion of digital services all introduce upside risk to demand projections. Under a scenario in which electricity demand grows at 1.5–2.0 percent annually, rather than 1 percent, the import requirement by 2028 would expand toward 3.0–4.0 TWh per year, materially increasing exposure to regional wholesale price cycles.
EPS currently operates an installed capacity base exceeding 7,300 megawatts, dominated by lignite-fired thermal plants and large hydropower stations, with smaller contributions from wind, solar, and gas-fired units. However, installed capacity does not translate directly into firm, dispatchable output. Thermal availability is constrained by maintenance cycles and environmental limits, while hydropower output remains highly dependent on hydrological conditions. As a result, effective firm capacity during peak periods is materially lower than nameplate figures suggest.
To address these constraints, EPS has outlined a substantial investment programme for the 2026–2028 period, with total planned capital expenditure of approximately 422.6 billion dinars, equivalent to roughly €3.6 billion. The investment envelope is split between maintaining and modernising existing thermal and mining infrastructure, reinforcing transmission and distribution assets, and expanding renewable generation capacity. Approximately half of the planned investment is earmarked for renewable projects, primarily solar and wind, complemented by smaller hydropower upgrades.
While these investments are significant in nominal terms, their timing and system impact are critical. Solar and wind additions improve energy balance on an annual basis but do not automatically resolve peak capacity or system adequacy challenges without parallel investment in flexibility, storage, or dispatchable backup. Under current plans, renewables primarily offset energy deficits rather than eliminate the need for imports during peak demand periods.
A proposed alternative planning approach would explicitly separate energy balance from capacity adequacy. Under such a framework, Serbia could accept a moderate level of annual energy imports while prioritising domestic investments that secure peak capacity and grid stability. This could include gas-fired peaking units, large-scale battery storage, or hybrid renewable-storage projects designed to provide firm capacity rather than energy alone.
Financially, the current EPS strategy assumes that imports covering a 6.5 percent supply gap remain economically manageable. However, sensitivity analysis suggests that each €10 per megawatt-hour increase in average import prices would translate into an additional €20–25 million in annual system costs at a 2.0–2.5 TWh import volume. Under tighter regional market conditions, these costs could rise materially, feeding through to end-user tariffs or EPS balance-sheet pressure.
In this context, the EPS plan reflects a cautious, risk-averse approach that prioritises balance-sheet stability and incremental transition over aggressive capacity expansion. Whether this approach remains viable will depend on how accurately demand growth unfolds relative to assumptions, and on Serbia’s ability to secure predictable, competitively priced electricity imports during periods of system stress.
The period from 2026 to 2028 thus represents a critical inflection point. Decisions taken now on capacity, flexibility, and demand forecasting will determine whether Serbia manages a controlled transition toward a more import-integrated power system, or whether it faces recurrent exposure to market volatility at precisely the moment when electricity becomes the backbone of industrial and digital growth.
