In South-East Europe, gas does not merely influence electricity prices through marginal cost. It multiplies volatility by interacting with structural congestion in the power grid. When gas tightness coincides with constrained transmission, the price impact is no longer incremental; it becomes discontinuous. This interaction explains why electricity markets across the region experience abrupt price separations and extreme spikes even in years without headline fuel crises. Seasonal system assessments by ENTSO-E highlight adequacy envelopes, but market outcomes show that congestion turns gas stress into amplified price events.
The mechanism is straightforward but underappreciated. Gas marginality raises the cost of the last available megawatt. Congestion determines where that megawatt can be delivered. When interconnectors bind, higher marginal costs cannot be arbitraged away, and prices decouple sharply across borders. In South-East Europe, where grid reinforcement has lagged generation and demand changes, this interaction has become a dominant driver of volatility.
The regional grid remains characterised by a small number of critical corridors. North–south flows linking Hungary with Serbia and onward to the southern Balkans, as well as east–west links between Romania and Bulgaria, carry a disproportionate share of regional balancing flows. When gas tightness pushes gas-fired units to the margin during winter peaks, these corridors saturate rapidly. Once saturated, even small changes in gas availability can trigger outsized power price separation.
Quantitatively, the effect is dramatic. Under unconstrained conditions, a €20/MWh increase in gas-driven marginal cost might translate into €20–30/MWh higher electricity prices across the region. Under congestion, the same marginal cost increase can result in €70–120/MWh price spreads between adjacent bidding zones within hours. This is not a failure of market coupling; it is the market correctly pricing physical limits.
Recent winter stress events illustrate the point. During cold spells affecting Serbia and Bulgaria, peak prices in constrained zones exceeded €250–300/MWh, while neighbouring markets with residual transfer capacity cleared €120–160/MWh. Intraday repricing was violent, with spreads of €50–100/MWh emerging late in the day as gas nominations tightened and interconnectors reached security limits. These outcomes occurred even without exceptional gas price moves, underscoring that congestion—not fuel price level—was the multiplier.
For traders, congestion transforms gas exposure into locational optionality. The value is no longer in predicting average price direction, but in identifying when gas stress and grid stress align. Corridors that appear liquid under normal conditions become binary under stress: either they flow at full capacity or not at all. Positions that monetise this binary behaviour—through spreads, options, or intraday flexibility—capture the majority of volatility-adjusted returns.
Congestion also explains why correlations collapse during stress. Power prices that move together 90% of the time can decouple completely in the remaining 10%—the very periods that dominate P&L. Traders relying on historical correlations or cross-hedges underestimate tail risk. In SEE markets, congestion ensures that tail outcomes are local, not regional.
Industrial electricity buyers experience the same mechanism as abrupt cost divergence between sites. A plant in Serbia may face peak prices above €300/MWh while a facility across the border in Hungary sees half that level on the same day. For multi-site operators, this creates unexpected cost dispersion that cannot be explained by fuel prices alone. Procurement strategies that assume regional convergence fail when congestion binds.
This has direct implications for contracting. Fixed-price electricity agreements often embed an assumption of functional interconnection—that suppliers can source power from elsewhere when local prices spike. When corridors are saturated, that assumption breaks. Suppliers either pass through congestion-driven costs or embed risk premia ex ante. Buyers who do not explicitly address locational risk in contracts remain exposed to the most expensive outcomes.
Gas tightness increases the frequency with which congestion matters. As coal exits accelerate in Romania and Bulgaria and hydro flexibility tightens during dry winters, gas becomes marginal in more hours. Each of those hours is a test of grid capacity. Without reinforcement, congestion events become more frequent, and volatility increases even if average demand growth is modest.
The economic scale of congestion rents illustrates the stakes. On key SEE corridors, annual congestion rents have reached €30–70 million in recent years, concentrated in a handful of winter weeks. These rents represent the market’s valuation of constrained flexibility. They are paid by consumers and captured by those positioned to exploit spreads, rather than being systematically reinvested into grid reinforcement at sufficient speed.
For traders, congestion-driven gas volatility favours speed and optionality over size. Intraday liquidity, fast response, and granular locational exposure matter more than large directional positions. The most profitable hours are often the last few of the trading day, when updated gas nominations, weather forecasts, and grid constraints collide.
For industrial buyers, the lesson is that location is risk. Electricity procurement cannot be optimised purely on price; it must account for grid exposure. Plants located behind constrained corridors face structurally higher volatility. Mitigation strategies include peak caps, locational hedges, on-site flexibility, or demand response agreements. Paying €3–8/MWh more on average to reduce locational exposure can prevent €40–80/MWh spikes during congestion-driven events.
Carbon convergence will intensify this interaction. As dispatchable coal capacity exits faster than grid reinforcement progresses, gas marginality and congestion will coincide more frequently. Even as renewable capacity grows, its output variability does not alleviate corridor saturation during winter peaks. Without accelerated grid investment—often requiring €0.8–1.2 million per kilometre for new 400 kV lines—the volatility multiplier will persist.
The unified conclusion is clear. In South-East Europe, gas does not merely set marginal prices; it activates congestion, transforming manageable fuel stress into extreme power outcomes. Traders who understand this treat congestion as a feature, not a bug, and position accordingly. Industrial buyers who recognise it redesign procurement around locational risk rather than average price levels.
As long as gas remains the marginal fuel during critical hours and grid reinforcement lags system needs, congestion will remain the amplifier through which gas tightness becomes volatility. The next major price dislocation will not require a gas crisis—only a cold day, a constrained corridor, and a system with no slack left.