Flexible and bankable connection agreements allow large electricity users to connect under terms that better align their demand with available system capacity while reducing the risk of cost shifting to other customers. Instead of requiring every new load to receive a fully firm connection from the outset, agreements can make part of the connection conditional on the customer reducing or shifting demand when the system is constrained. Modeling of six potential data center sites in the United States found that combining flexible grid connections with customer-provided capacity could enable full operation three to five years earlier than conventional connection pathways, while maintaining grid power availability for more than 99% of annual hours and requiring on-site resources for approximately 40–70 hours per year. The same framework can determine which network costs should be recovered directly from the connecting customer and which should be shared where upgrades provide wider system benefits. 

Long-term contracts, minimum payment obligations, collateral and exit provisions can give utilities greater confidence that investments made to serve the load will be recovered even if demand develops more slowly than expected. Together, these mechanisms can enable earlier connections, make better use of existing infrastructure and reduce the risk of stranded network investment. Their effectiveness depends on clear curtailment rules, transparent cost allocation and contractual terms that remain sufficiently predictable for utilities, customers and financiers to make long-term investment decisions.  

Main Outcomes of the Solution 

• Affordability: More direct cost allocation and improved asset utilization can reduce the risk of existing customers funding underused or customer-specific network investment.  
• Reliability & Flexibility: Curtailment and demand-response provisions allow large users to reduce or shift consumption during constrained periods, providing operators with additional flexibility to manage load growth.  
• Business Value: Long-term commitments, minimum payments and financial security can increase revenue certainty and improve the bankability of infrastructure required for large-load connections. 

Key implementation and scaling factors 

1. Establish clear curtailment rules, including triggering conditions, duration, notice requirements and operational responsibilities. 
2. Define transparent cost-allocation rules that distinguish customer-specific infrastructure from investments providing wider system benefits. 
3. Align long-term contractual commitments with the scale and timing of infrastructure investment and expected load ramp-up. 
4. Set proportionate minimum-payment, collateral and exit provisions based on customer creditworthiness and investment exposure. 
5. Establish regulatory approval and cost-recovery frameworks for connection, tariff and network-investment arrangements. 
6. Ensure adequate metering, telemetry, communications and integration with operational systems where connections depend on real-time load flexibility. 

Case studies

Google (2026), United States of America  

In March 2026, Google reported that it had integrated 1 GW of data-center demand-response capacity into long-term energy contracts with five U.S. utility partners: Indiana Michigan Power, Tennessee Valley Authority, Entergy Arkansas, Minnesota Power and DTE Energy. Under these arrangements, Google can temporarily reduce or shift portions of machine-learning workloads running in its data centers during specified periods, allowing participating utilities to treat part of the load as flexible rather than assuming that all demand must be served continuously. Google notes that the degree of flexibility varies by location and workload.  

Google states that incorporating demand response into these agreements can support faster connection of new data centers by allowing utilities to manage periods of high demand using existing system resources while additional generation and network capacity is developed. The arrangements illustrate how large-load customers can leverage operational flexibility to enable faster connection arrangements that better reflect available system capacity and bring benefits to the broader grid.

AEP Ohio and Public Utilities Commission of Ohio (2025), United States of America  

In July 2025, the Public Utilities Commission of Ohio adopted a settlement establishing AEP Ohio's dedicated data-center tariff, which became effective on July 23. The tariff introduces stronger financial commitments for new data centers seeking service, including minimum billing requirements, collateral and financial viability requirements, and exit fees where projects are canceled or contractual obligations are not met.  

Minimum billing demand is generally capped at 85% of total contract capacity, while the initial contract term comprises a load-ramp period of up to four years followed by eight years. As of February 12, 2026, AEP Ohio reported that data centers and developers had signed binding contracts covering 5,642 MW under the tariff and provided the associated collateral. The framework links infrastructure commitments more closely to contracted customer demand, reducing the risk that costs associated with unused capacity are shifted to other customers. 

Indiana Michigan Power (2025), United States of America 

In February 2025, the Indiana Utility Regulatory Commission approved modifications to Indiana Michigan Power's industrial power tariff for large-load customers. The provisions apply to customers with contract capacity of at least 70 MW at an individual site or 150 MW on an aggregated basis.  

The framework includes long-term contractual commitments, minimum payment requirements, collateral and protections where customers reduce or terminate service. Collateral requirements are generally linked to 24 months of the customer's maximum monthly non-fuel bill, subject to exemptions based on creditworthiness and liquidity. These measures are designed to reduce the financial exposure associated with infrastructure investments made in anticipation of large new loads by requiring customers to make financial commitments before substantial investment is undertaken. The case illustrates how contractual and tariff design can strengthen the bankability of large-load connections while reducing the risk of costs being transferred to other electricity customers. 

References:

1. A New Milestone for Smart, Affordable Electricity Growth | Google  
2. Data Center Tariff | AEP Ohio  
3. Flexible Data Centers: A Faster, More Affordable Path to Power | Camus Energy, encoord & Princeton ZERO Lab 
4. Indiana Michigan Power Receives Order in Large Load Settlement | Indiana Michigan Power  
5. Large-Load Framework | Indiana Michigan Power