
CREG Resolution 101 127 of 2026 introduced a temporary framework to address the energy supply risk associated with the 2026–2027 El Niño event. Its immediate objective is to preserve hydro reservoir levels by increasing thermal generation. The measure temporarily suspended, in full, the Statute for Situations of Supply Shortage Risk that had been in force since 2014, until either its definitive amendment is adopted or 31 May 2027, whichever occurs first.
The mechanism was activated through CREG Circular 343 of 5 September 2026 and began to apply to system operation on 7 September.
The Resolution goes beyond setting a minimum level of thermal generation to preserve reservoir levels. It also introduces reference values for hydro generators’ bids and rules that affect spot-price formation and the remuneration of thermal generation dispatched out of merit. Its effects therefore extend to the allocation of reliability costs, contractual hedging and the economic signals received by the market.
A two-pronged intervention
The partial reservoir level — defined as the energy equivalent of the usable storage volume of the reservoirs in the National Interconnected System (SIN), excluding those in the Bogotá reservoir chain — is compared with the ISO-Thermal threshold, corresponding to the ISO-GT curve associated with the minimum thermal generation level established by CREG. This comparison is one of three indicators used to determine system conditions. The other two compare the Peak-Period Spot Price with the Activation Scarcity Price and the seven-day moving average of thermal generation with the thermal generation level associated with the applicable ISO-GT curve.
The combination of these indicators determines whether the system is under monitoring, surveillance or alert conditions. An alert is triggered when the partial reservoir level falls below the ISO-Thermal threshold, the price indicator remains low and the seven-day average thermal generation is below the level associated with the relevant ISO-GT curve. This was the condition that led to activation of the mechanism in September 2026.
Once the risk period is activated, the National Dispatch Center (CND) must schedule, in both dispatch and redispatch, a minimum level of thermal generation associated with the applicable curve. The objective is straightforward: to displace hydro generation and preserve water for the months ahead.
At the same time, reservoir-based hydro plants continue to submit bids, but must do so taking into account a reference curve that provides a regulatory approximation of the opportunity cost of stored water: the lower the reservoir level, the higher the value of the stored energy; as the reservoir approaches spilling conditions, that value declines. The generator’s discretion is therefore constrained, and bids that deviate by more than ±10% from the reference criteria must be reported by the CND to the Superintendence of Public Utilities.
Mandatory thermal dispatch does not, however, mean that all such generation will clear in the Ideal Dispatch. When a thermal plant is dispatched to preserve water but remains out of merit, its generation gives rise to a positive reconciliation payment, with the resulting cost passed on to demand through constraint charges.
CREG expressly anticipated this effect. CREG Document 901 431 of 2026 explains that the two components of the mechanism operate in different directions: minimum thermal dispatch preserves reservoir levels but increases constraint costs, while the reference curve for hydro bids contains spot-price formation. In the scenarios analysed, the Commission estimated constraint costs of approximately COP 68 to COP 126/kWh.
In other words, containing upward pressure on the spot price does not eliminate the cost of the thermal generation required. Part of that cost may instead appear through constraint charges.
2024: the immediate precedent
The 2024 experience helps explain the current design. In April of that year, before the conditions for activating the Statute had been met, the Ministry of Mines and Energy issued Resolution 40116 of 2024, establishing a minimum daily thermal generation reference. If economic dispatch fell short of that level, the CND would introduce an additional constraint, and out-of-merit thermal generation would be settled as security generation. Spot-price formation, however, remained subject to the ordinary market rules.
Several months later, through CREG Circular 072 of 2024, the Commission activated the Statute established under CREG Resolutions 026 and 155 of 2014. XM reported that the mechanism operated from 30 September to 20 November and relied on the Energy Sold and Stored mechanism (EVE): hydro energy was remunerated so that it would remain stored in reservoirs, while the bids of the participating plants were adjusted accordingly.
CREG Document 901 431 of 2026 identifies several shortcomings in the 2024 experience. The targeted level of thermal generation did not fully materialise in actual system operation, while the EVE mechanism imposed additional costs on demand through constraints and demand not backed by Firm Energy Obligations (OEF), while at the same time reducing the enforceability of the OEFs of the plants benefiting from the mechanism.
Resolution 101 127 reinstates minimum thermal generation, eliminates the EVE mechanism and retains a regulatory reference for hydro bids. Rather than an entirely new framework, it is a reformulation of the measures tested in 2024.
The cost shifts channel — and allocation
The distinction between spot prices and constraint charges has a direct distributional consequence. Contractual hedging reduces, to varying degrees, retailers’ exposure to the spot price. It does not, however, hedge them against constraint charges, which are borne by demand as a whole.
If the required thermal generation were to clear in merit and push up the spot price, the higher cost would fall primarily on those with greater exposure to the spot market. When part of that generation remains out of merit and is remunerated through constraint charges, the cost is spread across a broader base, including market participants that had managed their spot-price exposure through contracts.
The measure therefore has a redistributive effect, even though CREG does not state redistribution as an objective.
CREG Document 901 431 of 2026 makes another design consideration explicit: the financial sustainability of the market. The Commission warns that prolonged periods of high spot prices can place liquidity pressure on retailers with limited contractual coverage and potentially escalate into a systemic risk. Avoiding extended periods of high and sustained spot prices therefore forms part of the design.
The measure thus seeks to balance two risks: an energy supply risk — preserving water — and a financial risk — avoiding prolonged pressure on market participants with greater spot-market exposure. From that perspective, it cannot be viewed simply as a reduction in total cost, but rather as a change in the profile and allocation of that cost.
Where does this leave the Reliability Charge?
The debate goes deeper because the Colombian power market already has a permanent reliability mechanism.
Under CREG Resolution 071 of 2006, Firm Energy Obligations (OEF) become enforceable when the Spot Price exceeds the Activation Scarcity Price (PEA). Resolution 101 127 may intervene earlier: its alert condition combines reservoir levels below the threshold, thermal generation below the required level and a low price indicator. In other words, the mechanism may be triggered when an energy supply risk is already present but the price signal that makes the OEFs enforceable has not yet materialised.
This timing difference explains the preventive rationale behind the intervention, but it does not eliminate the design question raised by the coexistence of the two mechanisms. The Resolution itself recalls that, as early as 2014, the Superintendence of Industry and Commerce recommended reviewing the Reliability Charge in order to minimise potential duplication among mechanisms designed to protect system security, and that in 2026 the CND and several participants in the consultation process again called for a comprehensive review.
CREG Resolution 101 127 corrects one specific interference created by the previous framework by eliminating EVE and preserving the OEFs. A structural question nevertheless remains: if the preventive mechanism contains the price signal and remunerates part of the required generation through constraint charges, does it complement the Reliability Charge, or does it partially substitute the signal intended to activate a mechanism for which demand is already paying?
Early results and long-term signals
The initial data are consistent with the architecture of the measure. Public information from XM shows a reduction in the spot price and an increase in constraint costs following activation. At the same time, XM reported that aggregate hydro reserves in the SIN remained relatively stable, while monthly water inflows stood at 73.31% of the historical average.
It is still too early to assess the overall outcome of the mechanism and, in particular, to attribute changes in reservoir levels causally to the Resolution.
The debate also extends beyond the immediate impact on tariffs. Drawing on the 2023–2024 experience, Fernando Barrera Rey had previously argued that repeated market interventions during scarcity episodes can alter the expectations and incentives of a market designed to respond to critical hydrological conditions. He frames this as a moral-hazard or “self-fulfilling prophecy” problem: if market participants expect a new intervention once scarcity deepens, they may incorporate that expectation into their decisions.
That argument connects directly with investment signals. If a material share of the cost of scarcity is remunerated outside the spot price, the relationship between the physical condition of the system and the economic signal observed by the market becomes less direct.
This does not mean that the Resolution eliminates the investment signal or that, by itself, it will reduce investment. It may, however, make contractual hedging less complete against the cost of critical system conditions and add regulatory risk to projects with long investment horizons. That risk may translate into higher risk premiums, a greater need for contracted or regulated revenues, and less reliance on the spot price as a foundation for the investment case.
The underlying issue
CREG Resolution 101 127 responds to a concrete need: increasing thermal generation today in order to preserve sufficient hydro reserves for the next dry season.
But the regulatory decision goes beyond system operation. By combining a reference mechanism that contains spot-price formation with minimum thermal dispatch that may generate positive reconciliations, part of the cost of reliability may be shifted to constraint charges. This changes how the cost is allocated, alters the scope of contractual hedging and reopens the debate over the interaction between the temporary mechanism and the Reliability Charge.
The question is not whether the system should preserve water in anticipation of El Niño. It is how to do so without weakening hedging mechanisms, blurring the role of the Reliability Charge or undermining the economic signals that should help ensure the market is better prepared when the next scarcity event occurs.