Golden Valley Electric Asks Alaska Regulators to Delete Its Demand Ratchet and Bill Commercial Demand by Load Factor

Golden Valley Electric Association has asked the Regulatory Commission of Alaska to price commercial demand three different ways depending on how evenly a building consumed power during the month. Under the Fairbanks-area cooperative’s pending general rate case, Large General Service Primary demand would be billed on a tier structure running from $32.12 to $36.12 per kilowatt, with the applicable rate set by whether the customer’s monthly load factor falls below 60 percent, lands between 60 and 80 percent, or exceeds 80 percent. The same kilowatt of billed demand carries a different price in each band.

The proceeding can run to March 1, 2027.

The ratchet. A demand ratchet sets billed demand at no less than a fixed share of the highest demand recorded in the preceding eleven months. GVEA’s current provision uses a 70 percent floor, which means a single interval of elevated draw propagates through most of a year of bills. The filing removes that provision entirely from General Service classes 2(P), 2(S) and 3, and prices each month on its own.

The tiers. Large General Service Secondary runs $29.77 to $33.77 per kilowatt. Large General Service Primary runs $32.12 to $36.12. Industrial Service runs $37.38 to $41.38. Each class carries a $4.00 per kilowatt spread between its top and bottom tier. GVEA has not published the intermediate rate for the 60 to 80 percent band in its filing summary; only the endpoints of each range are on the public record.

The increases. Requested class increases range from 7.0 percent for General Service 3 to 17.8 percent for General Service 2(P). The classes carrying demand meters are the classes absorbing the larger end of that range.

The arithmetic of a tier crossing. A customer that crosses a load-factor threshold does not merely stop paying for the kilowatts it removed. It reprices every kilowatt it still bills. The following example is illustrative and assumes a midpoint of $34.12 for the unpublished middle Primary tier; the figure is a modeled estimate, not a filed rate.

Take a customer billing 500 kilowatts of demand against an average demand of 290 kilowatts. Load factor is 58 percent, the applicable charge is $36.12, and the demand line is $18,060.

Shave the peak to 480 kilowatts. Average demand holds near 290, so load factor reaches roughly 60.4 percent and the assumed middle tier applies. The demand line falls to approximately $16,378. Twenty kilowatts of peak reduction produced about $1,682 of savings, or roughly $84 per kilowatt removed.

Under a flat $36.12 charge, the same twenty kilowatts would have saved $722.

The effect runs in both directions. A customer already operating above 80 percent load factor pays $32.12 with no tier left to capture, and the same twenty kilowatts saves $642. On the published endpoints alone, the gap between the best and worst case for an identical physical reduction is wide, and the incentive concentrates at the thresholds rather than spreading evenly across the range.

Why storage moves the ratio twice. Load factor is average demand divided by peak demand. Discharging into the peak lowers the denominator. Charging during off-peak hours raises consumption and therefore the numerator, provided the charging window does not set a new peak. Both halves of a battery cycle push the ratio upward, including the round-trip losses, which register as additional off-peak consumption.

That is a different optimization target than the one commercial storage controls are generally tuned for. Flat demand charges reward the largest achievable reduction in a single interval. Load-factor tiers reward reduction up to a threshold and then stop paying for it until the next threshold.

What monthly reset changes. A ratchet makes storage performance sticky in both directions. A controller that clips the annual peak locks in a lower floor for eleven subsequent bills. A controller that misses one interval raises the floor for the same eleven bills. Deleting the ratchet collapses that into twelve independent exposures.

For anyone writing a savings guarantee against this tariff, the risk changes shape rather than size: smaller individual misses, more of them, and no residual credit carried forward from a single well-managed summer.

The same ratio, measured on the wires. California codified a related concept in September. SB 905, signed alongside SB 913, establishes a grid utilization metric applied to distribution grid segments, exposing which circuits have headroom during off-peak hours, and empowers the California Public Utilities Commission to mandate year-over-year utilization improvements. The statute names customer batteries, EV chargers and smart thermostats as the resources expected to shift consumption timing to meet those targets. California is the first state to codify the metric, which the Utilize Coalition published in late September.

California measures utilization on the distribution system. GVEA proposes to price a customer-level version of the same ratio on the bill. Neither mechanism pays for the absence of a peak in isolation; both pay for the relationship between peak and average.

What the deployment data shows. Lawrence Berkeley National Laboratory’s distributed solar and storage dataset, covering roughly 5.3 million systems and an estimated 93 percent of the 2025 distributed market, puts non-residential storage attachment at 11 percent in 2025 against 7 percent in 2024. California non-residential attachment reached 18 percent from 12 percent. All other states combined doubled from 4 percent to 8 percent. Residential attachment reached 37 percent nationally and 74 percent in California.

Those figures establish that commercial storage is growing off a small base outside California. They do not isolate which rate feature is driving the decision, and no published measurement yet separates demand charge design from the other variables.

What travels and what does not. GVEA is a cooperative filing before a single state commission in a service territory outside any regional transmission organization. Whether the design spreads is unresolved. Cooperatives copy rate structures from one another more readily than investor-owned utilities do, and the filing arrives while the underlying concept of measuring utilization rather than peak is being written into statute elsewhere.

The narrower point is settled. If the commission approves the structure, a commercial battery in GVEA territory will be paid for a load shape rather than for the absence of a peak. A shape appears on every monthly bill. An avoided annual peak appears once.

Sources