EFG Files Testimony Recommending Good Practices for Large Load Forecasting

Last week, the Kentucky Public Service Commission wrapped up the hearing on a request from Kentucky Utilities Company/Louisville Gas & Electric's ("KU/LG&E") request for approval of two 645 MW 1x1 combined cycle gas plants, one 400 MW battery energy storage system, and installation of a selective catalytic reduction ("SCR") system on Ghent Unit 2 – a total of $3.7 billion worth of capital investment. EFG Senior Consultant Chelsea Hotaling submitted testimony in the case recommending that the Commission deny approval of the proposed new generation resources and the Ghent 2 SCR on the basis that need for these generators is contingent on demand from prospective customers, none of whom have entered into service contracts or otherwise signaled any financial commitment to the Companies.

Indeed, this highlights two double standards often at play when considering what generation to build and what new load is likely to materialize. First, utilities, including KU/LG&E, often require new large load customers to enter into contracts that require a reimbursable commitment that will cover interconnection facilities costs if the customers fails to take service from the utility. These contracts must be signed before the construction of those facilities commences. These contracts may also require some level of collateral to guarantee these payments in the event that the customer does not take service. This is in stark contrast to how investment in generation is often treated when a utility asks its regulator for Commission approval to construct new generation facilities on behalf of prospective large load customers who have not agreed to cover new generator costs. The standard here is backward – where more money is at stake, no commitment is required.

Second, there is a double standard with respect to how the risks of prospective large load customers are evaluated. Whereas generating resources submitting responses to an all-source request for proposals ("RFP") are not all assumed to materialize and are often evaluated for non-cost factors that will influence project success, the same is not true of new loads. Some of the non-cost factors from an AES Indiana RFP are given in the table, below. Nearly identical factors could be evaluated for new loads rather than assuming that any new load request has the same probability of success.

Given the current environment for supply side generators, in particular the market for new thermal resources, there is little room for error in load forecast projections for new large load customers. In this particular case, the procurement of just one extra CCGT would cost customers $1.4 billion dollars of capital investment alone, without considering the rate of return or the ongoing fuel and maintenance expenses over the lifetime of that unit. Given the level of capital needed to serve these large load customers, there should be a high level of scrutiny applied to new large load requests.

Chelsea's testimony also included recommendations related to tracking prospective customer data and raising interconnection barriers to help identify the risks associated with prospective customers and to ensure customer evaluation is focused on those customers that are not speculative. Chelsea's testimony specifically recommended that KU/LG&E be directed to file quarterly reports with the Commission that follow the process Georgia Power has with submitting reports to the Georgia Public Service Commission. This information includes data points such as the announced project load, projected load ramp including load and timing, changes to project status since the filing of the last report such as updates to announced load, updates to load ramp, project stage (new project, existing process and undergoing transmission study, contract status), and reason for project loss, if known. Chelsea also suggested that the Commission direct KU/LG&E to collect additional information on prospective customers including site control, construction progress, permit status, whether or not data center developers have a tenant in place for the proposed site, and the number of projects that the developer has experience with.

When customers have not made a commitment to take service, there exists material risk that those customers will cancel their project and/or pursue alternative locations in other utility service jurisdictions. When there are low barriers for entry into a utility's load queue, i.e. no costs or time limits for how long the customer can be in the queue, that tips the balance towards a likelihood that the prospective customer will enter multiple queues. When there is no or relatively little cost for customers to be in numerous load queues, the likelihood of those customers entering numerous queues at the same time is high since they bear relatively little risk and potentially significant benefit in holding that queue position. The table below outlines recommendations developed by Elevate Energy Consulting and GridLab on improving interconnection requirements to ensure credible new customer requests are in the prospective customer queue.

While there are multiple potential benefits associated with bringing new customers to utility service territories, especially potential economic development impacts, caution still needs to be exercised in the evaluation of those prospective customers to ensure that existing ratepayers are not burdened by the significant capital investment needed for new supply generation if that load does not materialize.