Serbia is entering 2026 with a structural problem that looks like an inflation story on the surface but is actually a competitiveness story. The EU’s Carbon Border Adjustment Mechanism is moving from a reporting-only phase into a regime where embedded emissions become a cash cost at the border, and the sectors that sit inside the initial CBAM scope—iron and steel, aluminium, cement, fertilisers and electricity—map directly onto Serbia’s most energy-intensive export segments. The result is that “green electricity” stops being a CSR line item and becomes a procurement instrument that decides who keeps EU customers and who bleeds margin until orders migrate.
Publicly available regional trade mapping shows that CBAM-covered goods represent roughly 11.7% of Serbia’s total export value to the EU, with the exposure concentrated in iron and steel at about 4.5%, electricity at about 4.7%, aluminium at about 2.0%, fertilisers at about 0.5%, and cement at about 0.1%. Those are trade shares, not profit shares. The profit exposure is typically higher because these are high-energy, high-carbon products where carbon cost and electricity cost are directly embedded into unit economics.
The buyers that matter under the CBAM export frame are not households and not even domestic Serbian retail customers. The buyers that matter are EU-based industrial importers and OEM supply chains who will be forced to report and then internalize embedded emissions. They will start re-rating suppliers along a very simple axis: “Can you evidence lower embedded emissions at scale, with credible electricity attributes and stable delivery, without raising price unpredictably?” This is where Serbia’s energy-intensive exporters face a two-sided squeeze. On one side, they are competing with EU producers who are paying ETS prices but also gaining access to decarbonisation subsidies, grid priority, and increasingly lower-carbon electricity mixes. On the other side, they compete with non-EU suppliers who may be cheaper but will also face CBAM costs, meaning the winner is the one who can decarbonise fastest per euro of CAPEX and per MWh of power.
For Serbia’s heavy industry, green electricity is the fastest lever because it can be contracted before process technology is rebuilt. Steel, aluminium, and fertilisers can reduce reported embedded emissions materially by switching the electricity attribute stack, especially where indirect emissions are counted or where customers treat electricity provenance as a gating condition for supplier selection. This is not the same as full decarbonisation—blast furnace steel still has dominant direct emissions—but it changes the commercial conversation with EU buyers immediately because it reduces the “avoidable” part of the footprint and creates a credible transition pathway.
The moment you translate this into investor logic, the question becomes: how much green electricity must be contracted or built to defend CBAM-exposed exports, and what does that cost in CAPEX, grid integration, and return terms?
A practical base case for industrial green power procurement in Serbia is a contracted demand envelope of around 1.5–2.5 TWh per year by 2028–2030, anchored by the biggest electricity buyers in the CBAM export stack—steel, aluminium, and fertilisers—plus a smaller layer from cement and export-oriented fabrication that will be pulled into customer-driven emissions disclosure. An upside case, if EU buyers tighten supplier requirements and if Serbia’s largest exporters pre-emptively lock “green supply” as a competitive shield, is 3.0–4.0 TWh per year.
To supply 2.0 TWh per year with domestic renewables, you are typically looking at a portfolio roughly equivalent to either 650–750 MW of wind at a 32–38% capacity factor, or 1,200–1,400 MW of solar at a 17–19% capacity factor, or a mixed stack that reduces system stress. The least fragile stack in Serbia is not solar-only. A system-credible base case is a mixed build of roughly 400–500 MW wind plus 400–600 MW solar, supported by 100–200 MW / 200–400 MWh of storage and firming contracts. That combination can deliver roughly 1.8–2.6 TWh while keeping curtailment and capture-price erosion within controllable bounds.
On CAPEX, financing-grade 2026-style delivered costs in the region give you a realistic range. Utility-scale solar typically sits at €0.55–0.90 million per MW, rising toward €0.95–1.10 million per MW when grid works are heavy. Onshore wind typically sits at €1.20–1.80 million per MW depending on civil complexity and HV distance. Two-hour battery systems often land at €0.35–0.55 million per MWh fully installed. Under a mixed base case of, say, 450 MW wind plus 500 MW solar plus 150 MW / 300 MWh of batteries, an all-in CAPEX envelope typically lands in the €1.1–1.8 billion range once owner’s costs and interconnection are included. The upside case, scaling toward 700 MW wind plus 800 MW solar plus 200 MW / 400 MWh of batteries, pushes the envelope toward €2.0–3.2 billion, with the spread dominated by grid reinforcement intensity and site complexity rather than equipment alone.
Returns depend on whether industrial offtake is bankable and whether the renewable attributes are recognized in a way that EU buyers accept. In Serbia, the most investable structure for CBAM-facing industry is not pure merchant exposure; it is long-term PPAs with credible settlement, typically 10–15 years, structured either as physical delivery where feasible or as financial PPAs paired with guarantees of origin and balancing services. Under that structure, unlevered project IRRs for wind and solar often sit in the 7–10% band for contracted cash flows. Wind can sit toward the higher end because it has stronger capture-price resilience and system value, while solar can compress faster if it becomes too concentrated.
Curtailment is where CBAM-facing industry starts to care about system integration, not just price per MWh. If an industrial buyer signs a green PPA expecting a fixed volume of renewable electricity attributes and the project is curtailed, the buyer either receives fewer attributes or must true-up in the market, which creates both cost volatility and compliance uncertainty. At portfolio scale, even “modest” curtailment is a material economic leak. If a green supply platform is delivering 2.0 TWh per year, each 1% of curtailment equals 20 GWh of lost eligible volume. At a blended PPA value of €70–90/MWh, that is €1.4–1.8 million of value erosion per percentage point, per year, before you even price in penalties, shape mismatches, or customer renegotiations.
The reason wind behaves differently here is that wind curtailment tends to be more event-driven and localized, with less structural coincidence than solar. Wind’s higher capacity factor means fewer MW are needed per TWh, which reduces connection-node saturation and the “MW pressure” on specific substations. Wind also produces more in hours that are not the universal solar peak, so its capture price and deliverability are more stable as penetration rises. That stability directly improves the probability that industrial buyers receive a usable attribute volume that matches their procurement strategy.
Grid integration constraints are the deal-breaker in Serbia if they are not modeled explicitly. The first constraint is connection-node saturation, which is not evenly distributed; it concentrates around the strongest corridors and substations, meaning marginal MW becomes disproportionately expensive. The second constraint is voltage and reactive power management, which becomes a cost item at scale. The third constraint is ramping and reserve, where solar-heavy build-outs create steep daily ramps that require fast flexibility. The fourth constraint is balancing and settlement: the industrial buyer’s true “cost” of green electricity is not just the PPA strike; it is the PPA strike plus imbalance costs plus shaping costs plus any curtailment-driven replacement costs.
This is where aggregation and virtual balancing become strategic. Industrial buyers under CBAM do not simply want a low price; they want a defensible emissions narrative with predictable cost. A portfolio-level aggregator can firm renewable output blocks, reduce forecast error, minimize imbalance exposure, and use batteries and intraday repositioning to deliver a much higher “usable green volume” than standalone assets could. For CBAM-facing exporters, that can mean the difference between a green procurement program that is stable enough to satisfy EU customers and one that triggers constant exceptions and renegotiations.
The most investor-relevant stress test is what happens if grid upgrades slip by 12–18 months, because that is the failure mode that turns “green transition” into stranded CAPEX. In a realistic portfolio rollout, a 12–18 month grid delay does not delay everything equally; it strands specific nodes. If 300 MW of wind and solar capacity planned for an industrial PPA platform is delayed by 18 months, deferred generation can easily reach 700–1,000 GWh, depending on technology mix. At €70–90/MWh, that is roughly €49–90 million of postponed revenue and attribute delivery, concentrated in early years where debt service and contract milestones are most sensitive. Equity IRR impact in contracted structures commonly compresses by 100–200 basis points in a base case and 150–250 basis points in an upside case if delays coincide with higher grid CAPEX and curtailment-driven true-ups. Wind-heavy mixes tend to degrade more gracefully than solar-heavy mixes because wind’s commissioning can be staged with partial output at dispersed nodes and because its capture prices remain more resilient when delayed.
The strategic conclusion for Serbia’s CBAM-exposed industrial buyers is that green electricity is no longer optional, but it must be procured in a way that is system-realistic. The bankable route is a portfolio approach that emphasizes wind as the stability backbone, uses solar as a volume layer where nodes are strong, embeds storage as insurance rather than decoration, and relies on aggregation to deliver firmed attribute volumes rather than theoretical MWh. In that configuration, Serbia’s exporters can convert CBAM from a margin threat into a commercial differentiator: “We can evidence lower embedded emissions at scale, with reliable renewable attributes, and without price chaos.” If they cannot, the EU buyer will not necessarily stop buying overnight, but they will price the risk into procurement—and Serbia will lose competitiveness quietly, contract by contract.
Elevated by clarion.energy
