The Henry Hub spot price averaged USD 2.19 per MMBtu across 2024 and USD 4.13 per MMBtu in January 2025, with a daily high of USD 9.86 per MMBtu on 17 January 2025. Those are the figures the US Energy Information Administration published in its Short-Term Energy Outlook of 11 February 2025, which forecast an average of USD 3.79 per MMBtu for 2025. For a plant in Mexico buying an index-linked molecule, the fuel line roughly doubled year on year in a single winter month.
That matters to a specific reader. If your plant runs boilers, direct-fired ovens, kilns or onsite gas generation, the January move went straight through to conversion cost, and it did so before any hedge, contract or efficiency measure could react. The question in front of a plant director now is not whether gas prices are a risk. It is which peso of efficiency capital removes the most gas cost.
The answer is narrower than most efficiency lists suggest. Of the measures normally recommended to Mexican plants, only combustion and steam work, heat recovery and fuel switching touch the gas bill. Power factor correction and rooftop solar change the electricity bill and leave gas burn untouched. Gas-fired cogeneration raises gas consumption rather than cutting it. Sorting the list on that basis is the whole of the analysis below.
Why a Texas cold snap lands on a Mexican plant
Mexico's industrial gas supply is physically an extension of the Texas market. The EIA reported on 15 August 2023 that US pipeline exports to Mexico averaged 5.9 Bcf per day in the first half of that year and set a monthly record of 6.8 Bcf per day in June 2023. El Financiero reported on 4 June 2024 that imports from the United States reached 5.894 Bcf per day in the first quarter of 2024, the highest quarterly volume on record, while Pemex output kept falling. The direction of travel has been one way for a decade. This is how Mexico's natural gas supply is structured, and it is also the risk of relying on US natural gas.
The consequence is not abstract. In February 2021, freezing wellheads and pipelines in Texas cut supply into northern Mexico. Al Jazeera reported on 16 February 2021 that 4.8 million customers lost power across Chihuahua, Coahuila, Nuevo León, Tamaulipas, Durango and Zacatecas. Plants in those states did not lose gas because of a price. They lost it because the molecules were not there, and because upstream demand had first call on them.
Price and availability are therefore two separate exposures with two separate remedies. A hedge protects the first. Only fuel diversity, onsite storage where it is permitted, or a firm transport position protects the second.
The contract position decides the exposure
Two contract terms determine how much of a winter event a plant absorbs, and most finance teams have never seen either written down.
The first is firm versus interruptible transport. A plant holding interruptible capacity can be curtailed when the system is constrained, which is precisely when it most needs the gas. Firm capacity costs more and is worth what it costs only if the plant cannot run without heat.
The second is index versus fixed pricing. An index-linked molecule passes through every move in the reference price. A fixed or collared price transfers that volatility to the counterparty at a premium. The transport component sits on top and is regulated separately. Maximum tariffs for the Sistrangas system are set by resolution and published in the Diario Oficial de la Federación, and the 2024 schedule was published on 13 February 2024. The transport element of delivered cost is therefore knowable in advance in a way the molecule price is not.
One more term deserves attention before any efficiency project is approved. Where gas is bought with a minimum take, or where firm transport capacity is reserved and paid for whether or not it is used, cutting consumption does not cut the bill in proportion. Check the take-or-pay floor before modeling the saving, because it sets a limit on what efficiency can recover this year.
Where the gas actually burns
Gas leaves an industrial site through three routes, and only two of them are worth attacking with efficiency capital.
Steam generation is the first and usually the largest. It is also the most recoverable, because the losses are in combustion tuning, condensate return, distribution insulation and failed traps rather than in the process itself.
Direct-fired process heat is the second. Cement kilns, steel reheat furnaces, glass tanks and ceramic kilns burn gas at flame temperature into the product. Efficiency work here is real but bounded, and electrification is not available at the temperatures involved. Food and beverage plants sit in a different position, because most of their gas goes to steam and hot water, where recovery measures work well and heat pumps become an option at the lower temperature end.
Onsite generation is the third. A plant that makes its own electricity from gas has converted part of its power bill into gas exposure, which is why cogeneration belongs on the gas side of the ledger rather than the savings side.
What cuts gas, and what does not
The table below separates the measures by what they actually change. It is the sorting step most efficiency proposals skip.
| Measure | What it changes | Effect on the gas bill |
|---|---|---|
| Steam trap survey and repair | Gas burn in the boiler house | Direct reduction, lowest capital |
| Combustion tuning and excess air control | Boiler thermal efficiency | Direct reduction, low capital |
| Condensate return and insulation | Feedwater temperature and distribution loss | Direct reduction, low to medium capital |
| Waste heat recovery | Heat that currently goes up the stack | Direct reduction, medium to high capital |
| Boiler replacement | Base thermal efficiency | Direct reduction, high capital |
| Biogas or biomass switching | The fuel, not the demand | Reduces gas volume, adds supply and permitting risk |
| Gas-fired cogeneration | Purchased electricity | Raises gas burn, lowers the power bill |
| Rooftop or onsite solar | Purchased electricity | None, unless the site self-generates from gas |
| Power factor correction | A CFE surcharge | None |
| Energy management system | Visibility and control | Indirect, enables the measures above |
None of this makes the electricity measures bad investments. It makes them the wrong answer to a gas price question, and presenting them as gas measures is how efficiency programs lose credibility with a finance team.
Boiler and steam work comes first
Steam trap failure is the clearest case of money burning with nobody watching. The US Department of Energy's Steam Tip Sheet Number 1, dated January 2012, states that in systems not maintained for three to five years, 15% to 30% of installed traps may have failed. A trap stuck open passes steam continuously, and a plant with a hundred traps and no survey program is likely carrying several dozen of them.
The rest of the boiler house work runs in the same order. Measure stack temperature and excess oxygen on every burner, because excess air carries heat out of the stack and the correction is a tuning job rather than a capital project. Measure the condensate return rate, because every tonne of condensate not returned is replaced with cold makeup water that has to be heated from ambient. Then look at what leaves the stack. Heat recovery systems in industrial plants capture that energy into feedwater, combustion air or a process stream, at a capital cost that only makes sense once the free measures are exhausted.
Fuel switching to biogas or biomass sits after all of this, and the constraint is rarely the boiler. It is the fuel supply contract, the storage and handling area, the emissions permit and the burner modification. Ask for the fuel supply agreement before the boiler quotation.
The electricity measures, kept in their place
Power factor correction belongs in a plant's efficiency program. It just does not belong in a gas strategy. Under the CFE tariff schedule, a power factor below 0.90 attracts a surcharge calculated as three fifths of the shortfall ratio applied to the bill. At a power factor of 0.85 that works out at roughly 3.5% of the billed amount, at 0.80 roughly 7.5%, and at 0.75 roughly 12%. Capacitor banks are cheap and the payback is usually short, which is why the measure survives on every list. It removes an electricity surcharge and nothing else.
An energy management system earns its place for a different reason. Without submetering, a plant knows its total gas consumption from the invoice and nothing about where it went. A gas meter on each boiler and a steam flow meter on the main header turn a monthly invoice into a burner-by-burner picture, and that picture is what makes the measures above provable to a CFO. Buy the measurement before you buy the equipment.
Where to spend the next peso on gas
The order is not a matter of taste. Start with a steam trap survey and combustion tuning, because both are maintenance spend rather than capital and both act on the gas bill immediately. Move next to condensate return and insulation. Only then look at heat recovery and boiler replacement, and require the take-or-pay floor in the gas contract to be stated in the business case, since it caps what any of this recovers in the current contract year.
Run the contract review in parallel, not afterwards. Confirming whether the site holds firm or interruptible transport, and whether the molecule is indexed or fixed, costs nothing and changes the risk picture more than most capital projects do. A plant that discovers it is on interruptible transport in the middle of a January cold snap has discovered it too late.
Find out what share of your gas burn is recoverable
Mexico Energy Partners will review twelve months of gas invoices with the contracted transport arrangement, the boiler schedule with ratings and ages, and the steam or hot water distribution layout. The output is the gigajoules consumed per unit of output, the share of gas burn addressable by combustion tuning, condensate return and steam trap repair, and the capital and simple payback for each measure. You supply the invoices, the contract and the equipment list. We return the assessment and the assumptions behind it. This is an analysis of your own data, not a promise of savings.
Start the process on the request for quote page, or send the files to info@mexicoenergypartners.com.
Sources
- US Energy Information Administration, Short-Term Energy Outlook, 11 February 2025. Henry Hub spot price averages for 2024 and January 2025, the 17 January 2025 daily high and the 2025 forecast.
- US Energy Information Administration, Today in Energy, US natural gas exports to Mexico set a monthly record in June, 15 August 2023. Pipeline export volumes to Mexico.
- El Financiero, México depende más de gas natural de EU, importaciones suben 7% en 1T2024, 4 June 2024. First quarter 2024 import volume and Pemex production trend.
- Al Jazeera, Texas freeze leaves millions in northern Mexico without power, 16 February 2021. Customers affected and states involved.
- Centro Nacional de Control del Gas Natural and Diario Oficial de la Federación, maximum Sistrangas transport tariffs for 2024, published 13 February 2024.
- US Department of Energy, Advanced Manufacturing Office, Steam Tip Sheet Number 1, Inspect and Repair Steam Traps, January 2012. Failed trap share in unmaintained systems.
- Comisión Federal de Electricidad, published tariff schedule, low power factor surcharge, schedule in force February 2025.