Why multi-site buyers need a portfolio strategy

CENACE’s wholesale market report for the week of June 14 to 20, 2026 put the national average local marginal price at 818.14 pesos per MWh. The Cozumel node cleared at 13,764 pesos per MWh that same week, roughly 17 times the national figure, according to reporting by Factor Energético on June 28, 2026. The congestion sits where it sat a year earlier. CENACE’s report for the week of November 2 to 8, 2025 also showed Cozumel as the system’s worst node. It cleared 11,339.24 pesos per MWh against a national average of 586.20, and Mazatlán cleared negative that week. The Riviera Maya zone averaged nearly six times the system price. Factor Energético’s June 2026 analysis calls the pattern structural and not temporary, because the same congested corridors keep reappearing across two years of comparison.

For a manufacturer running one plant, that is a fact to price into a single contract. For a manufacturer running four or six plants across Querétaro, Guanajuato, Nuevo León, and Coahuila, it is a portfolio problem. That corridor is drawing the bulk of new industrial investment in 2026, and most multi-site buyers in it do not manage power as a portfolio. Mexico attracted 1.72 billion dollars across 13 announced industrial investments in 2026 as of mid-year. Inaugurated projects reached 4.5 billion dollars by late February, roughly two and a half times the pace of the prior year, according to Industry & Energy Magazine’s reporting on June 22, 2026. Volvo Group’s 1 billion dollar expansion in Nuevo León is one entry on that list. Every one of these new sites lands on a different node. The price behavior differs. So does the local congestion pattern.

Multi-site manufacturers have two levers a single-plant operator does not. They can aggregate demand across facilities under common ownership to reach qualified-supply scale, even when individual plants fall short of the threshold alone. They can also structure a supply contract around the real, persistent price dispersion across their nodes instead of accepting a single blended number. On our read, most groups use the first lever, because a broker or supplier raises it early in a sales conversation. They skip the second, because it takes a harder negotiation. The second lever is where the money is.

The rules for aggregating load

Aggregation of load centers to reach qualified-user status has been available for years. Its legal footing firmed up twice in the past decade, most recently in the 2025 reform. The original framework appeared in the Diario Oficial de la Federación and was updated effective March 2, 2017, according to legal analysis published by Garrigues. It allows load centers belonging to one or more companies within the same corporate group to be aggregated, provided the parent company maintains control. Each aggregated center needs at minimum 25 kW of demand and proper metering under CRE or Market Rules standards. The 2017 update extended the mechanism to low-voltage load centers, beyond the medium- and high-voltage sites the original 2016 rule covered.

The Ley del Sector Eléctrico, published March 18, 2025, carried this forward and did not narrow it. Article 73 confirms that a load center, or a group of load centers aggregated to reach the 1,000 kW measured-demand threshold, can register for qualified supply. Legal commentary from Greenberg Traurig on the enacted law describes that explicitly as a group right, not a single-site one. Industry guidance from Enerlogix Solutions frames the 2025 reform as formalizing multi-plant demand aggregation for corporate groups. That closed an ambiguity which had left some multi-site groups unsure whether dispersed, smaller facilities would qualify collectively. The practical gate is common ownership under a single tax identity, the RFC, plus CENACE’s technical criteria for how the aggregated load is measured and represented in the market.

One thing has not changed, and it is easy to miss in a conversation focused on aggregation. CENACE still settles the market node by node. Aggregating your demand for qualified-user registration does not aggregate your price exposure. A supplier serving five plants across five different nodes is still buying five different price profiles to cover your load. The contract terms decide whether that reality is priced transparently or averaged into one number. A blended number quietly favors the supplier at the cheaper nodes and costs you at the expensive ones.

What scattered sites cost

Start with the aggregation benefit, because it is the more straightforward number. A group with four plants each running 400 kW of measured demand, 1.6 MW combined, qualifies individually for qualified supply nowhere. Each site sits well under the 1,000 kW threshold. Aggregated under common ownership, the group clears qualified-user status and can pursue the 15% to 30% discount off CFE basic-supply pricing that Mexico Energy Partners’ procurement work typically finds for buyers making this transition. On a combined delivered power bill in the range of 2 to 3 million dollars a year across four mid-sized sites, that discount is 300,000 to 900,000 dollars annually. Four separate sub-scale plants could not reach that money on their own.

Now the harder number, the one a blended contract can hide. Take a group with one plant near Cozumel and one plant in a less congested zone. Suppose the Cozumel-area plant’s node clears at anything close to the 13,764 pesos per MWh Factor Energético recorded in June 2026, even for a handful of hours a month. A supplier pricing that load on a national or zonal average is either eating a loss it will try to recover elsewhere in the contract, or it has priced the contract to protect itself against exactly that risk. The second is far more common. In that case the group pays an implicit premium across all its sites to cover the exposure at the one bad node. A group that does not know which of its plants sits in a congested corridor cannot tell whether its blended rate is fair, cheap, or quietly loaded against it.

If you cannot name the CENACE node behind each plant in the group, the blended rate on the table cannot be checked, and the node price history that settles it is a request your supplier can fill this week. Talk to an advisor.

The fix is not to avoid nodal exposure. It is to price it explicitly. A contract with node-specific energy pricing, or a defined pass-through formula tied to each site’s actual local marginal price, shows a multi-site buyer which plants carry the congestion cost. That lets the buyer negotiate against the fact directly. The moves available are load-shifting at the exposed site, a narrower margin on the sites that are cheap to serve, and better information about where future sites should not land.

Why 2026 forces the decision

What makes 2026 different from three years ago is where the new industrial investment lands. The nearshoring corridor running through Querétaro, Guanajuato, Jalisco, Nuevo León, Coahuila, and Chihuahua is where automotive and advanced manufacturing capacity is concentrating, per Industry & Energy Magazine’s account of 2026 investment activity. That corridor does not sit on uniform grid infrastructure. Some of these nodes are well served. Others sit behind the same transmission bottlenecks CENACE has flagged as structural.

Executives closest to industrial power procurement already describe a shift toward combined instruments. Alex Vicentico Pérez of ENGIE Mexico has argued that covering multi-site industrial load now takes a combination of power purchase agreements, storage, and self-supply schemes. One instrument no longer covers a whole portfolio. Mexico Business News reported his remarks on June 2, 2026. Daniel García of Atlas Renewable Energy, in the same reporting, frames the constraint on renewable contracting for industrial buyers as access to reliable delivery, not a shortage of interested buyers. Put plainly, where your plant sits on the grid now matters as much as the price you negotiate. Industrial renewable PPAs in Mexico typically run 10 to 15 years, per the same reporting. A contract signed today without node-level exposure priced in locks that blind spot for a decade or more.

Where aggregation stops working

Aggregation stops at the common-control test. A joint venture, a minority-owned plant, or a facility held through a structure that gives the parent no clear control will likely fail that test, even when it shares a brand and a supply chain with the rest of the group. Confirm your corporate structure against CENACE’s control criteria before you assume every site in the portfolio counts toward the threshold. A plan built on an assumed aggregate that CENACE later rejects has to be rebuilt under time pressure.

Node-level pricing transparency also cuts both ways. A group that pushes for site-specific pricing and finds most of its portfolio in favorable nodes has strengthened its negotiating position. A group that finds the opposite has to decide whether to accept that cost, hedge it with onsite generation or storage at the exposed site, or factor it into where the next expansion lands. Asking for transparency does not guarantee a better number. It guarantees an accurate one. Every one of these decisions needs an accurate number.

One scenario would make the whole argument matter less. If CENACE and CFE make real progress on the transmission investment the PLADESE planning process calls for, congestion at nodes like Cozumel and Riviera Maya could ease, and the price dispersion could narrow. That is the outcome to hope for, and not the one to plan around. Factor Energético’s finding that the same corridors have stayed congested across two years of comparison is the more reliable base case for a contract being signed today.

Four steps for your portfolio

Map your portfolio by node, not just by plant. Before any procurement conversation, identify which CENACE node or zone each facility settles against, and pull at least a year of that node’s local marginal price history. Your supplier already has this data. Buyers rarely ask for it. It is the foundation for every decision that follows.

Confirm aggregation eligibility against the control test as well as the ownership test. Verify that every site you intend to include shares common corporate control under a single RFC. Get that confirmed in writing as part of any qualified-user registration, early and not late.

Negotiate node-specific or pass-through pricing instead of a single blended rate across the portfolio. Ask any supplier proposing a uniform contract to show the node-level cost assumptions behind it. A supplier unwilling to break out that pricing is asking you to accept a risk allocation you cannot verify.

Sequence expansion decisions with the node data in hand. A group choosing between two otherwise comparable sites in the nearshoring corridor should treat grid congestion history as a site-selection input on the same footing as labor cost and logistics access. Pricing congestion in after the lease is signed is too late.

Sources

• CENACE weekly wholesale market report, week of June 14 to 20, 2026: national average local marginal price of 818.14 pesos per MWh, Cozumel node at 13,764 pesos per MWh, and identification of Riviera Maya, Cozumel, Malpaso-Tabasco, Papaloapan, and Yucatán Peninsula corridors as structurally congested across a two-year comparison. Factor Energético, June 28, 2026.
• CENACE weekly wholesale market report, week of November 2 to 8, 2025: national average local marginal price of 586.20 pesos per MWh, Cozumel maximum of 11,339.24 pesos per MWh, Mazatlán minimum of negative 964.61 pesos per MWh, and Riviera Maya zone average of 3,370.06 pesos per MWh. CENACE weekly Mercado Eléctrico Mayorista report, November 2025.
• Mexico industrial investment activity in 2026: 1.72 billion dollars across 13 announced projects, 4.5 billion dollars in inaugurated projects by late February 2026, roughly 2.5 times the prior year’s pace, Volvo Group’s 1 billion dollar Nuevo León expansion, and concentration of nearshoring investment in Querétaro, Guanajuato, Jalisco, Nuevo León, Coahuila, and Chihuahua. Industry & Energy Magazine, “Nearshoring impulsa parques industriales del Norte y Bajío,” June 22, 2026.
• Load center aggregation rules for qualified-user status: common corporate group control requirement, 25 kW minimum demand per aggregated center, and the 2017 extension to low-voltage load centers. Garrigues, “México: Actualización de requisitos de agregación de centros de carga para ser considerados usuarios calificados.”
• Ley del Sector Eléctrico, published in the DOF on March 18, 2025, Article 73 basic-supply option and group aggregation to reach the 1,000 kW qualified-user threshold. Greenberg Traurig, “Reformas al Sector Energético, Parte I: Ley del Sector Eléctrico,” February 2025.
• 2025 reform formalizing multi-plant demand aggregation for corporate groups under common RFC and CENACE technical criteria. Enerlogix Solutions, “Usuarios Calificados: Guía Completa para Empresas Industriales en México.”
• Qualified supply discount of 15% to 30% versus CFE basic supply. Mexico Energy Partners procurement analysis.
• Industry commentary on multi-instrument power procurement strategy and typical 10 to 15 year industrial PPA terms in Mexico, including remarks attributed to Alex Vicentico Pérez of ENGIE Mexico and Daniel García of Atlas Renewable Energy. Mexico Business News, “Securing Clean, Resilient Energy for Industrial Operations,” June 2, 2026.