NextFin News - Colombia imported a record 31% of its monthly natural gas supply in July 2026, the clearest signal yet that the country's decades-long run as a self-sufficient gas producer is over. Domestic production has fallen from 1,070 million cubic feet per day in 2022 to 673 million cubic feet per day in the first eight months of 2026, proved reserves have collapsed by 39% in three years, and a strengthening El Niño is now forcing the power grid to burn imported gas at a pace the system was never built to sustain. The twist: total gas demand is essentially flat, so every cubic foot of imports is replacing domestic supply that is not coming back.
The Deficit Is Structural, Not Cyclical
Colombia lost gas self-sufficiency in 2024, and the deficit has widened since. Average annual marketed production dropped from 1,070 million cubic feet per day (MMcf/d) in 2022 to 772 MMcf/d in 2025, and fell further to 673 MMcf/d between January and August 2026, according to Colombia's gas market administrator BEC. The fields that once carried the system — Cusiana, Cupiagua, Pauto Sur, and the offshore Guajira fields — have been declining since 2022 with no large discovery developed to replace them. Proved reserves fell 39% to 1.7 trillion cubic feet between 2022 and 2025, the National Hydrocarbons Agency reported.
Into that void stepped LNG. Regasified import volumes rose from just 5 MMcf/d in 2021 to 174 MMcf/d in 2025, and imports reached a record 31% of monthly supply in July 2026, BEC data show. Through August 2026, imports accounted for 28% of total supply, Promigas said.
The timing is the problem. Total gas demand has averaged 865 billion British thermal units per day (BBtu/d) so far in 2026, only 0.9% above the 2025 annual average of 858 BBtu/d. Demand is not the driver; the supply base is shrinking underneath a flat demand curve, and imports are filling the gap.
That gap is about to widen. On June 11, NOAA announced that El Niño had formed in the tropical Pacific. By October, the Climate Prediction Center had upgraded its outlook. In its October advisory, the center said:
"there's a greater than 83% chance of a strong to very-strong event through the first quarter of 2027."
El Niño brings drier weather to northern South America from December through February, draining the reservoirs that supply roughly two-thirds of Colombia's electricity and forcing thermal plants — mostly gas-fired — to run harder.
The switch is already visible. In 2025, a normal year, hydropower supplied 81% of electricity, natural gas 8%, and coal 4%. By July 2026, hydropower's share had already fallen to 66% while coal and natural gas each supplied 12%. During the last strong El Niño, thermal generation reached as much as 58% of electricity demand, compared with about 13% under normal conditions. If the current event matches that intensity, gas demand from the power sector could nearly double again — on top of a domestic supply base that has shrunk by more than a third in four years.
Why Imports Are the New Marginal Fuel, Not a Stopgap
The first-order story is simple arithmetic: production fell, demand held, imports filled the gap. The second-order story is that the import share is now structurally embedded in the dispatch order of the power system.
Colombia's electricity grid is a hydro-dominant machine with a thermal backstop. In wet years the backstop barely runs and gas imports are marginal. In dry years the backstop becomes the marginal price-setter, and every additional megawatt-hour of thermal generation requires gas that no longer exists domestically. That is the transmission channel: a weather shock in the Pacific becomes a call on Atlantic LNG cargoes, and the wholesale price of electricity in Bogotá becomes partly a function of global LNG markets.
This is not a cyclical dip that mean-reverts on its own. A cyclical gas shortfall would rest on a short-term driver — a dry year, a pipeline outage, a temporary demand spike — with a demonstrated path back to the prior equilibrium. Colombia has the opposite: four consecutive years of production decline, a 39% destruction of proved reserves in three years, and capital that is still being allocated to crude rather than gas. Reserves are the inventory of future production; when the inventory shrinks faster than output, the output decline is not an accident, it is a schedule.
The evidence floor for a structural call is threefold. Production has declined every year since 2022 (1,070 to 772 to 673 MMcf/d), a three-point trend, not a single bad print. Reserves fell 39% to 1.7 trillion cubic feet between 2022 and 2025 — a permanent balance-sheet change, not a weather event. And Ecopetrol's 2026 investment plan directs COP 14 trillion to exploration and production, of which only 11% is earmarked for gas, with 8 to 10 exploratory wells planned; crude output is expected to rise, "offsetting natural gas field declines," the company said.
Against that, the cyclical leg is real and it is El Niño. Dry-season gas demand will spike and then recede when the rains return. But the receding demand will land on a smaller domestic base than the year before, so the import share ratchets higher each cycle. That is the hallmark of a structural shift wearing a cyclical mask: the wave goes out, but the waterline does not come back.
What the Switch Costs
The cost of the transition does not land evenly. BBVA Research warned in April 2026 that drought conditions would require greater use of thermal generation, which is more costly and gas-dependent, raising wholesale electricity prices. Energy-intensive industries — cement, steel, chemicals — would face higher production costs, part of which would pass through to final prices.
The power generators' association Acolgen estimates Colombia could face firm-energy deficits of 2.3% in 2026 and 4.4% in 2027. A firm-energy deficit is not a dry-year inconvenience; it is a signal that the system cannot guarantee supply to all contracted customers, which in practice means rationing risk and price spikes in the spot market.
The bill has a price tag. Corficolombiana estimates each hour of power rationing would cost 5.6 billion pesos (about US$1.75 million) in lost output, and that thermal plants' gas demand grew 85% year on year in August 2026. During the previous El Niño, thermal plants' gas use nearly tripled in 2024. A scenario prepared by Russell Bedford shows a household with a monthly electricity bill of US$54 could see it rise to US$71 if consumption increases by 20% and generation costs climb. A broader Fedesarrollo estimate puts a 1990s-style rationing scenario at about US$8.7 billion, or 1.5 percentage points of GDP.
The International Energy Agency noted that during the 2023-2024 El Niño, Colombia's reservoir levels fell to less than 30% of their average by April 2024 and natural gas demand in the power sector almost doubled. If the current event matches that intensity, the same mechanism repeats on a weaker domestic supply base — meaning a larger import call, at global LNG prices, to keep the lights on.
Nine Terminals Chasing a Shrinking Market
Colombia's import dependence is running ahead of its import infrastructure — and may soon overshoot it. The country has only one operating LNG import terminal, SPEC LNG near Cartagena on the Caribbean coast, in service since 2016 and owned 51% by Promigas and 49% by Royal Vopak. Its expansion is due online in 2026, as early as the end of October, Promigas has said.
Behind it, a wave of projects is queued. The Buenaventura LNG terminal on the Pacific coast targets startup by November 2026. The Puerto Bahia and Ballenas LNG terminals could begin operations in 2027. Wood Mackenzie estimates Colombia's LNG import capacity will rise from 450 million cubic feet per day at the end of 2025 to about 1,300 million cubic feet per day by 2028 — nearly a threefold increase.
But the build-out may be chasing a ceiling. At least nine major LNG import terminal projects and expansions are planned to begin operations between 2026 and 2030. If all were developed, Colombia would have the capacity to import around 2,500 MMcf/d by 2030 — more than double what the government projects gas demand to be at that time, according to an energy research group. Regasification terminals are fixed, capital-intensive assets that must be paid for whether or not cargoes arrive. If demand falls short, Colombia could be left with stranded import capacity and take-or-pay contracts at the worst possible moment. The infrastructure build is a bet that the decline is permanent; if it is wrong, the bill comes due in the late 2020s.
The Adversarial Case: Demand Destruction Caps Volumes, Not Dependence
The strongest argument against a straight-line rise in import volumes is that Colombia is already substituting away from gas. Industrial gas consumption fell 28% in the first four months of 2026, with roughly half of that demand shifting to liquefied petroleum gas and a fifth to coal; some users have turned to sugar-cane bagasse and fuel oil. That is demand destruction, and it works against the import thesis as directly as any new supply.
The counter-thesis, then, is this: Colombia's loss of self-sufficiency is structural, but its import volumes face a ceiling. High LNG prices push industry off the gas grid; regasification capacity overshoots demand; and the import share stabilizes well below the worst-case El Niño spikes. In that world, the 31% July print is a weather-driven peak, not a trend line.
There is weight behind it. A sector report on gas supply and El Niño warned:
"if a new strong El Niño requires a similar increase, thermal demand could absorb a significant portion of available supply, place pressure on regasification infrastructure and make supply dependent on the timely entry of projects that are not yet ready."
The same logic cuts both ways: if regasification projects are not ready, volumes are capped by infrastructure; if they are all ready, volumes are capped by demand.
The answer to the counter-thesis is that substitution caps volumes but not dependence. Even if industrial users flee gas, the power sector cannot switch fuels — a gas-fired turbine cannot burn LPG or bagasse. And as long as hydropower remains the backbone of the grid, every dry season forces a marginal call on thermal generation, and therefore on imports. The volume may be capped; the dependence is not.
Supply Is Real, but It Is Dated
The strongest argument against permanent import dependence sits offshore in the Sirius-2 discovery. Petrobras and Ecopetrol confirmed more than 6 trillion cubic feet of gas in place — a find large enough to triple Colombia's reserves — and the project could produce 13 million cubic meters a day, nearly half of current domestic demand, for a decade.
But the timeline is the constraint. Petrobras expects to confirm commercial viability by 2027, and first production would follow three years after licensing — putting gas on the market no earlier than 2030, and only if the project passes its investment decision. Ecopetrol has said offshore production is not expected before 2029. A reserve that arrives in 2030 does not cover a deficit that peaks in 2026 and 2027.
A nearer-term alternative is Venezuela. Colombia is weighing gas imports from its neighbor as the sanctions outlook shifts, which could deliver volumes through pipeline infrastructure sooner than any offshore project. But that option substitutes one form of dependence for another and carries political risk that no commercial contract can price away.
Outlook: Three Scenarios for the Rest of the Decade
The base case: Colombia's gas import share stays elevated through 2027, driven by the structural production decline and amplified by El Niño. The regasification build-out will keep pace with demand, but it will not reverse the direction of trade — Colombia is a net gas importer for the rest of the decade. The loss of self-sufficiency is structural; the import volume is cyclical-plus-structural, capped by demand destruction but floored by the power sector's inability to switch fuels.
Upside case: the 2026-2027 El Niño weakens faster than forecast, reservoirs recover, and gas-fired generation falls back toward its 2025 share of 8%. Industrial users continue shifting to LPG and coal, holding total import volumes flat even as the import share of a shrinking domestic base stays elevated.
Downside case: a strong-to-very-strong El Niño persists through the first quarter of 2027 as NOAA forecasts, reservoirs drop toward 2024 lows, and firm-energy deficits materialize. In that scenario gas demand from the power sector doubles, import terminals run at full capacity, and wholesale electricity prices spike — with inflationary pass-through to industry and households.
What to watch: monthly import share from BEC — if it falls back below 20% while production stabilizes, the structural call weakens; SPEC LNG expansion and Buenaventura startup dates — slippage would tighten the 2027 balance, while on-time delivery would add capacity into a potentially soft market; Sirius-2 FID and licensing progress — an accelerated timeline would extend the structural call's expiration date; industrial gas demand — a continued 20%+ year-over-year decline would confirm the demand-destruction ceiling; and Venezuela import negotiations — a signed deal would cap the import-price risk but add political exposure.
The falsifying signal: if Colombia's marketed production stabilizes above 772 MMcf/d for four consecutive quarters while the import share falls back below 20%, the permanent-dependence thesis is wrong. Absent that, the 31% print is not a peak — it is the new floor.
Colombia built its modern economy on exporting energy; the next decade will be defined by how cheaply it can import what it used to produce.
Explore more exclusive insights at nextfin.ai.
