What Cold Storage Operators Miss About Demand Charges

What most cold storage operators miss about demand charges is not on the meter. It is on the tariff sheet. The charge is calculated from billing determinants that a facility inherited when it signed up for a rate schedule, and those determinants, particularly the ratchet clause, can keep a single bad afternoon on the invoice for the better part of a year. Very few operations teams have read the schedule their site is billed under.

This is a document problem before it is an engineering problem. A refrigerated warehouse can run a disciplined operation, hit every temperature target, and still overpay for electricity for years, because nobody in the building has looked at how the number is constructed. The bill arrives, it gets paid, and the determinants that produced it are never examined.

What is actually on the bill

Commercial and industrial electricity bills are assembled from a small number of standard components. A primer on rate design for cost-reflective tariffs published by the National Association of Regulatory Utility Commissioners describes the structure as built around three core rate elements: an energy charge, a demand charge, and a customer charge.

The distinction the primer draws between the first two is the one that matters most in a refrigerated facility. It defines a demand charge as a charge typically levied against a customer based on its maximum demand in each billing period, calculated on a per kilowatt basis, and states plainly that it is not directly related to a customer's volume of use.

That last clause deserves a slow read. A determinant unrelated to volume of use will not respond to a project that reduces volume of use. Efficiency work lowers the energy charge. It touches the demand charge only if it happens to change what the facility was drawing during the specific interval that set the peak, which is a coincidence rather than a design.

The customer charge, described in the same primer as a fixed monthly charge intended to recover costs that increase with each additional customer such as meters, billing systems, and customer support, is the one component nobody can influence. It is worth identifying on the bill simply so it stops being confused with the two that can be influenced.

The ratchet: one bad afternoon, eleven more invoices

The mechanism that surprises operators most is the ratchet, and it is the reason a single event can outlive the month it happened in. Under a ratchet clause, the demand a facility is billed for is not simply the demand it used. It is the greater of what it used and some percentage of a historical peak.

A 2025 analysis of large-load tariffs from RMI examining how utilities structure minimum billing demand gives a concrete example, noting that one utility's industrial power tariff sets the minimum monthly billing demand as the greater of 80 percent of the capacity contracted by the customer, or 80 percent of the customer's highest monthly demand over the prior eleven months. The same analysis found a minimum monthly billing demand provision in 24 of 65 state-level tariffs it reviewed.

Guides to energy management for cold storage describe peak demand as the largest controllable cost at a refrigerated site, which is a narrower claim than saying it is the largest cost. Controllable is the operative word, and a ratchet clause is what determines how long the consequences of failing to control it persist. Under an eleven-month lookback, an event in July is still being billed the following spring.

The practical consequence is that the value of avoiding one peak is not one month of savings. It is the difference between two billed demand figures multiplied across every invoice the lookback window touches. Facilities that evaluate peak management on a single month's arithmetic consistently understate what it is worth.

How the floor is set

RMI's 2025 review of large-load tariffs reports that the three most common ways these tariffs set a minimum billing demand are a percentage of contracted capacity, commonly 75 to 90 percent, the customer's historical peak, or a fixed floor.

Sitting on the wrong rate schedule

The second thing operators miss is that the schedule itself was a choice, usually made once, often by someone no longer at the company, under load conditions that no longer exist. A facility that added freezer capacity, converted to different equipment, or shifted its receiving pattern may have crossed a threshold that makes a different schedule cheaper. Nothing in the billing process flags this.

Utilities and regulators are actively reworking these structures, which means the option set changes underneath facilities that are not watching. A 2024 EPRI framework on demand flexibility tariff design for large commercial and industrial customers sets out to analyze pricing structures and billing components for cost recovery, evaluating flat rates, inclining blocks, time-of-use, real-time pricing, and subscription pricing, alongside billing components including fixed charges, energy charges, and demand charges.

The same project describes analyzing tariff sheets filed with state public utility commissions and tracking structural trends over time. That is a reasonable description of what an operator should be doing for its own sites at some interval, and almost none do. A rate schedule review is not an engineering project. It is a reading exercise with a spreadsheet, and it is usually the cheapest energy work available.

Time-of-use period definitions belong in the same review. Two schedules with similar headline rates can differ sharply in when their peak windows fall, and a facility whose receiving pattern conflicts with one utility's afternoon window may sit comfortably inside another's. The operational schedule is often easier to reconcile with the tariff than anyone assumed, because nobody had compared the two documents.

The share of spend that gets no advocate

Part of the reason this goes unexamined is that electricity does not dominate the cost sheet the way operators sometimes assume. The Global Cold Chain Alliance's Cold Chain Index for the fourth quarter of 2024, built with the Agribusiness, Food, and Consumer Economics Research Center at Texas A&M University, reports that labor was the largest share of refrigerated warehouse expenses at 40 percent of the total, rent and lease represented 39 percent, and electric power accounted for 9 percent.

Nine percent is large enough to matter and small enough to lose an argument for attention against labor and occupancy. It is also moving. The same index reports that electric power expenses grew by 2.3 percent nationwide in the fourth quarter of 2024 compared with the same quarter a year earlier, while overall refrigerated warehouse expenses rose 4.69 percent.

The asymmetry worth noticing is that the two larger categories are largely structural. Lease terms are negotiated on multi-year cycles and labor costs track a market. The electricity line contains a component that responds to operating decisions made inside the building, which makes it disproportionately worth understanding relative to its share.

Who actually owns the bill

In most refrigerated operations the answer is nobody, and this is the organizational fact underneath every technical one. The invoice arrives in accounts payable, which validates the amount against the contract and pays it. Operations sees a monthly cost figure in a report. The billing determinants sit in between, visible to a department with no ability to influence them and invisible to the department that could.

The fix is procedural rather than technical. Somebody in operations should receive the full bill, not a summary, and should be able to say what the billed demand was, what set it, whether a ratchet applied, and which schedule the site is on. That is four questions. In most facilities the first attempt to answer them takes several weeks, which is itself the finding.

The rate design principles the NARUC primer lists, sufficiency, fairness, efficiency, customer acceptability, and bill stability, describe what regulators are balancing when they approve these structures. Operators do not get to argue with the balance. They do get to understand it, and understanding it is what separates a facility that manages its demand charge from one that merely pays it.

None of this requires new equipment, a capital request, or a consultant. It requires reading a document the facility already receives and a tariff sheet the utility publishes. That is an unusually low barrier for a cost line that compounds across an eleven-month window when it goes unwatched.