Key considerations for serving large loads

Securing sufficient power supply and managing costs are key in large-load negotiations. Early coordination with the G&T cooperative or other upstream suppliers helps align priorities and broaden supply options. Growing grid constraints and extended interconnection timelines have made power availability the No. 1 consideration for siting new data centers — diminishing the relative weight of other criteria, such as fiber network availability and tax breaks. Effective large load supply management requires a high level of coordination between retail and wholesale suppliers to ensure that service risks are not simply transferred upstream on the grid.

As electricity demand continues to surge, the increase in interconnection requests from very large commercial and industrial clients is unprecedented. These customers frequently possess greater resources than the cooperatives that supply them and may even provide their own power generation. The rise of these large, sophisticated buyers has decisively shifted the balance of power, raising questions whether attitudes toward service obligations need to adapt.

When attracting large load customers, it's important to broadly assess their impact on local economic development. Data centers and similar high-energy businesses can boost the local economy by increasing tax revenue, providing well-paying jobs, creating new opportunities for area businesses and potentially upgrading public infrastructure. At the same time, communities need to weigh the considerable stress these facilities place on both power grids and water resources, as well as the risk that increased costs could be distributed across the population. Currently, ensuring a sufficient energy supply is the main factor in choosing locations for data centers, but as energy access improves, factors like water availability and growing community resistance are likely to become increasingly significant considerations.

Making sure that large loads cover their electricity service costs not only protects cooperative members, it also secures community interests. When large-load customer facilities closed or moved offshore after growing rapidly in the 1950s, 60s and 70s, electric co-ops struggled with stranded assets, leading to higher rates for remaining members and negative local impacts. Using strong contract terms and cost recovery mechanisms helps cooperatives avoid these financial risks.

Utilities face a surge in interconnection requests from large loads like data centers, but most of these projects will probably not get built. In the absence of a standard process to identify viable projects and firm proposals, utilities evaluate end-user contracts, require site control and permits, review engineering studies, and demand upfront payments for system studies and infrastructure purchases as well as other financial guarantees to avoid stranded investment and cost shifting.

Most electric cooperatives establish a minimum size and/or load factor threshold that triggers the need for special treatment when serving these new large load members. The exact definition of a large load is very system-dependent and seems to range from 25 to 150 MW. The determinate is defining the level of load that will trigger the need for substantial system investment.

Energy + Environmental Economics (E3) has coined the term “transformational loads” as these new members break down the traditional rate dynamics, requiring their own classification for cost allocation.

Evaluate the organizational structure and the “people equation” for serving large loads. Consider distributing responsibilities amongst a wider swath of specialized staff to better match the specific interconnection phase requirements with the skillsets required to get the job done. Establish a single point of contact within the organization to manage the relationship, guiding the new member past the initial intake on through facility energization and beyond. Ensure adequate staffing for the new workload as large load projects cannot bend or flex to other co-op workloads. Keep in mind that the factors impacting large load member siting within the community occur outside the co-op walls, so be prepared for increased community engagement, with skillset(s) in local economic development critical for that engagement.

As power supply aligns with demand in the coming years, its impact on site selection will lessen, while water availability and community opposition will become more significant. These factors may lead to development in areas with abundant resources and fewer local objections. There are important trade-offs for encouraging large-load development within a community that are oftentimes not well understood, necessitating even greater outreach and engagement by the co-op.

The importance of the intake process is to better assess member needs, the feasibility and readiness of the project as well as prepare upfront financial commitments (application fee, CIAC, milestone payments, etc.). Simplify and standardize the customer information intake or application process (i.e., consider using a fillable PDF) gathering all pertinent information about the new customer’s requirements including business details, site location and plan, evidence of site control, obtaining necessary regulatory approvals and permits, details of load ramp requirements and a breakdown of the total MW consumption by function (IT equipment, cooling, lighting, etc.). Potential operational concerns, such as voltage or frequency thresholds that could trigger customer-initiated demand shedding, can be identified early and plans for back-up supply systems can be better understood (see NERC Data Center Information Collection Questionnaire).

A completed application is often accompanied by a fee to initiate a formal cost-of-service study. This is an important initial vetting step to advance a project, demonstrating member financial commitment. While the actual costs incurred by a cooperative for developing a cost-of-service study will vary, these fees are typically charged on a flat rate basis with size and complexity of the project determining the actual member payment.

Electric cooperatives face significant financial risk during the period after a data center’s initial intake request and before the start of actual energy service. To help mitigate this exposure, co-ops are increasingly adding contractual safeguards and requiring upfront payments.

As a preliminary step, co-ops can utilize a Letter of Intent (LOI) to ensure clarity and mutual understanding of responsibilities and expectations upfront, facilitating a smooth and mutually beneficial partnership. While non-binding, these documents establish the framework for the more formal Letter of Agreement (LOA) to follow. The LOA serves as the primary “bridging” mechanism for cost recovery leading to the point at which an executed Energy Services Agreement (ESA) would take effect. LOAs protect utilities by securing upfront payments, clarifying responsibilities and establishing penalties for cancellation or delays — ensuring that only serious, commercially backed projects move forward and that the utility is not exposed to unrecoverable costs before the ESA is in force.

The best LOAs are clear, flexible, and comprehensive — protecting the utility from financial risk, ensuring fair cost allocation, and providing operational certainty for both parties. Thanks in large part to other cooperatives such as NOVEC, REC and Umatilla, data centers are now accustomed to paying for all of their delivery infrastructure upfront through Contribution in Aid of Construction (CIAC) payments specified in the LOA, with lower recurring fixed payments to reflect the operation, maintenance and future replacement costs for the infrastructure. Other key LOA provisions include project milestones, penalties for delays or cancellation, assignment clauses, minimum load requirements, provisions for onsite generation or storage and detailed customer responsibilities, all aimed at protecting the co-op before the ESA takes effect. Some of the relatively important particulars as to who owns, operates and maintains the infrastructure will be set forth in this agreement and then mirrored in the subsequent ESA.

An LOA is typically put in place first to govern the pre-service phase—covering upfront financial commitments, project milestones, and risk mitigation—while the ESA is executed later, once all conditions in the LOA are met and the facility is ready for energization and ongoing service. This staged approach allows co-ops to protect themselves financially and operationally during the high-risk development period.

Another contractual bridge that can supplement the LOA is a separate Construction Development Agreement (CDA). The CDA helps ensure that the cooperative’s obligations match those in its Engineering, Procurement, and Construction (EPC) contract. Since most costs for a cooperative are typically incurred during construction, the CDA outlines exactly what the customer must pay as the co-op takes on expenses with third parties. Often, a large portion of CDA payments—often 50% or 60%—is required at the outset, with the remainder paid out in parts as different project milestones are completed, and a final settlement paid at the end. The CDA requires substantial capital contributions upfront, and additional project risks are secured to guarantee the cooperative’s cost recovery. As with the LOA, the CDA also clearly details how disputes will be resolved and under what conditions the agreement can be terminated.

An executed Energy Service Agreement needs to be in place well in advance of facility energization outlining the terms and conditions under which an electric cooperative will provide energy services. Keep in mind that the LOA serves as a bridging mechanism, requiring a legal contract to bridge to, which is the ESA.

The services specified in the agreement always include electricity delivery but may not necessarily include physical supply. Historically, large commercial and industrial members were offered discounted rates for supply to attract large load members into a service territory. More recently, novel tariff filings introduce an adder to reflect the additional marginal costs for securing supplies. Alternatively, generation may be made available through an unbundled option and memorialized in a separate contract or provided by a third party for retail choice states.

Cooperatives might consider separating power obligations in a stand-alone separate agreement such as a Power Purchase Agreement (PPA). Unlike physical delivery (transmission and distribution), generation can be packaged in a variety of ways including allowing the member to bring their own resources. Because of long lead times for building new generation, these supplies will often be contracted through an agreement executed between the cooperative and the large-load member, a third party and the member or directly between the generator and the member. In fact, historically, many of the large-scale hyperscalers have traditionally procured energy directly from the market through a PPA.

There are benefits to consumer-directed PPAs and bring-your-own-resource programs as they reduce a co-op’s financial risk of warehousing additional costs associated with wholesale supply. However, the location of that member-directed supply is incredibly important, as the member must be in the same grid region as the generating facility. And the ESA must align with the terms of the member’s PPA, if supply is embedded in the agreement.

Physical PPAs also require coordination with grid operators for delivery and can entail significant interconnection costs for the member. An alternative would be a “sleeved PPA” where the cooperative acts as an intermediary between the generator and the customer. Once again, upstream to downstream contract alignment is critically important.

The initial term of the ESA should coincide with the useful life of the infrastructure investment, which is often between 10 and 20 years, with an average duration of roughly 15 years that might include a facility ramp-up period of no more than five years. Previously, most tariffs had contract lengths of three years, but this has now increased.

Example of a specific load ramp schedule:

• In Year 1: 50% contract capacity
• In Year 2: 65% contract capacity
• In Year 3: 80% contract capacity
• In Year 4: 90% contract capacity

A novel alternative to an annual load ramp schedule, introduced in the Tri-State High Impact Load Tariff (HILT) filing, requires the member to submit a load ramp projection and then enforces a narrow tolerance band around those projections. Admittedly, even with rigorous load-factor guarantees, data center growth often occurs in stages: initial build-out of infrastructure, incremental installation of IT racks, and finally the steady state of full rack density. If the ramp-up schedule slips, the co-op must continually reevaluate its resource plan, potentially delaying or accelerating generation commitments to avoid overbuilding or underbuilding. That mismatch between load realization and resource procurement can lead to capacity shortfalls or unnecessary spot purchases at premium prices, both of which translate to higher costs for all members. Consequently, rather than specifying a discrete ramp window, there are relatively strict requirements that the project’s initial load ramp projection cannot exceed a 5% tolerance band for each of successive three-year updates to preserve the integrity of resource planning.

As with all member bills, large-load rate design can be categorized into two major categories — fixed demand payments and volumetric energy payments. Increasingly the delivery service charge specified in the ESA is structured as a fixed demand charge. Data centers, for example, have consistent high demand, so a demand-based rate better matches system cost.

To date, notably absent from most approved large load tariffs are flexibility provisions. Yet, an ESA might also include negotiated service interruption provisions, such as identifying the level of firm load and the maximum number of hours or interruption events as well as compensation for interruptible service.

Early termination or exit fees are also commonly specified if a large-load customer decides to break their contract. For example, Dominion Energy allows customers to reduce their contracted capacity by 20% before triggering an exit fee. The fee is usually calculated based on the remaining term of the agreement and the expected revenue the utility would have earned and reimbursement for any costs incurred.

Maintain organizational awareness of the major customer financial outlay milestones, especially during the construction phase of the infrastructure buildout. Consider retaining an additional outside auditing firm (other than the co-op’s annual auditor) to conduct a quarterly attestation of utility large-load transactions during periods of large investment outlays. There is an obligation to the broader co-op membership to be able to account for the costs associated with serving these high-density loads and to track them, especially if a project unwinds or cancels.

Also, continue that awareness beyond facility energization to ensure fair cost allocation for new large loads. Consider the initial cost-of-service study as a working document that should be revisited based on lessons learned. Traditionally, system-wide studies might be conducted every three to five years. Yet, given rising costs, the cadence might be reconsidered. It is important to question whether the timing of this research is in lockstep with large-load cost drivers.

Electric cooperatives typically charge data centers a flat fee for a cost-of-service study, which generally ranges from tens of thousands to several hundred thousand dollars. The most common fee structures are tiered by the size (capacity) of the large load. The actual costs incurred by a cooperative for developing a cost-of-service study will vary based on several factors, such as the size and complexity of the facility being energized, the depth and breadth of the analysis required to ascertain costs and the geographic location. But generally speaking, the fees are designed to cover all and any costs incurred by the cooperative for developing an in-depth understanding of what is required to provide service to the new facility, which includes all upstream and downstream costs including engineering analysis, infrastructure planning, and financial modeling.

More recently, a number of cooperatives, appear to be charging $150,000 or more especially for very large projects (e.g., 80–200 MW). For projects exceeding 200 MW, fees can reach $250,000. At the more modest end of the spectrum, we’ve included AEP Ohio's 2024 Data Center Tariff (DCT) settlement, approved mid-July 2025, to illustrate a tiered-structure approach:

Upfront payments to cover major equipment and infrastructure costs are commonly required by cooperatives in the form of a Contribution in Aid of Construction (CIAC). A CIAC ensures that the cooperative is not exposed to stranded asset risk in the event the project does not proceed to completion. These upfront payments are commonly used to cover the equipment and construction required for service.

In order to serve large loads, electric cooperatives often invest in new or upgraded substations, large power transformers, and high-capacity distribution feeders. They also purchase advanced switchgear and circuit breakers for safe control and protection, as well as metering and real-time monitoring equipment to track usage and power quality. Additional investments may include integration with backup generation or battery storage systems and robust communication and control infrastructure for remote monitoring and automated system management. These investments add up but are critical for ensuring reliable, high-capacity service.

A CIAC is also used to fund engineering studies and more generalized system upgrades so that the financial burden does not fall on other cooperative members.

Aligning risk and reimbursement with developmental milestones encourages large load growth and ensures members remain engaged. Schedule payments according to specific project milestones—such as completing engineering studies, securing permits or finishing construction phases—to maintain ongoing financial security as the project advances. Project risk is the highest early in development with order commitments for large equipment and construction starting. As members increase their investment, cooperatives reduce their exposure, transitioning payment and collateral in step with designated milestone criteria. Once the project becomes operational and generates revenue, cooperatives should shift their approach to financial risk, using maintenance and replacement needs to guide the timing of milestone payments.

Cooperatives commonly require cash deposits from large load customers to help cover indeterminate costs and other financial risks associated with providing service. The primary purpose is to ensure that the co-op is not left with unrecoverable expenses in the event the project is delayed, canceled, or if the member fails to meet its normal financial obligations outlined in the LOA or ESA.

Deposits can help offset unexpected costs or overruns that may occur during construction. After construction is complete, price fluctuations in acquiring market supply create a major risk for under-collection; this risk can be managed by pre-paying or maintaining sufficient cash reserves.

Cooperatives might also require large load members to provide a letter of credit from a reputable financial institution as a form of collateral. This instrument serves as a guarantee that the new member can meet its financial obligations for electricity service, including unexpected construction costs, milestone payments and other financial risks.

If the member defaults under the terms of the LOA or ESA — such as failing to pay required fees or if the project is delayed or canceled — the cooperative can draw upon the letter of credit to recover unrecoverable expenses. By doing so, the cooperative actively manages the financial uncertainty and volatility that can arise from serving large, energy-intensive members, ensuring they do not face significant financial risk during construction and operational phases.

The consultancy Energy + Environmental Economics (E3) recently developed a credit and collateral framework that co-ops can use to design effective credit and collateral requirements. Well-designed credit and collateral policies not only protect cooperatives and existing membership from default risks, they also support the timely integration of essential infrastructure. When cooperatives achieve this balance, they ensure that large load growth delivers broad benefits to customers, the grid and the economy.

E3 makes a critical observation: Rigid, one-size-fits-all credit policy or a policy that exaggerates the risk of service can exclude legitimate, creditworthy projects that will bring value to the community. On the other hand, poorly designed or overly permissive credit policies can expose cooperatives and their membership to significant financial risks. By enabling optionality, cooperatives introduce flexibility and offers a defined set of acceptable credit instruments for themselves and their large-load customers.

Credit requirements often use several tools—such as CIACs, milestone payments, deposits and letters of credit—to manage risk. Without a clear plan, these tools can overlap and result in overcollateralization, which can restrict customer capital and may delay or end projects. If cooperatives excessively layer these instruments, they risk reducing clarity, obscuring true risk and limiting scalability.

Watch the voltage and frequency ride-through behavior of large loads during disturbances as they can contribute to instability. Ride-through behavior is primarily defined in terms of how long the load remains connected during a given voltage and/or frequency disturbance. However, changes in the load’s real or reactive power consumption during or after the disturbance are important as well (e.g., how much time passes before a disconnected load reconnects, and how quickly it returns to its original consumption). ERCOT has observed many events over the last three years where one or more large loads tripped or instantly reduced consumption during a system fault and system protection operated as designed.

Collaboratively work with large load members to address operational concerns. For example, members might be able to proactively shape a workload’s power profile, which could be more efficient than simply adapting to it. For data centers, large-scale batch-synchronized machine learning training workloads exhibit substantially different power usage patterns. Aligning on solutions to address these unique challenges could prove much more effective than addressing problems after the fact.

NERC established its Large Load Task Force (LLTF) to address the reliability and security challenges posed by the rapid growth of large loads — such as data centers, cryptocurrency mining and hydrogen production — on the bulk power system. The LLTF’s recommendations emphasize the need for utilities and transmission operators to conduct thorough gap analyses of existing reliability standards and operational practices. This includes identifying areas where current engineering requirements, modeling, and monitoring protocols may fall short in managing the unique risks associated with large energy-intensive customers. The task force also calls for improved data collection, real-time monitoring, and enhanced commissioning procedures to ensure that large loads are integrated safely and reliably into the grid.

A key recommendation from the LLTF is the development of clear definitions and classification frameworks for large loads, moving beyond simple peak demand metrics to consider operational characteristics, ramping behavior and potential impacts on system stability. Utilities are urged to establish robust interconnection requirements, including design and performance criteria for steady-state and dynamic modeling, as well as protocols for customer-initiated load shedding and system restoration. The task force highlights the importance of coordination between retail and wholesale suppliers, proactive workload shaping, and the installation of high-speed disturbance monitoring equipment to maintain grid reliability as large loads proliferate.

Looking ahead, NERC’s next steps include working with industry stakeholders to refine reliability standards, develop best practices for large load integration and promote information sharing across the sector. The LLTF encourages utilities to incorporate large loads into both long-term and near-term planning, engage with regulatory bodies to ensure alignment with evolving market conditions, and invest in communication infrastructure to support ongoing operational readiness. By following these recommendations, cooperatives can better manage the risks and opportunities presented by large loads, safeguarding both system reliability and the interests of their membership.