A Fundamental Question of Electric Rate Design

Branko Terzic & R. Bruce Williamson

When I hear the term “volumetric rates,” the question I ask as a regulator and economist is simply, “Whose volumetric rates?” The term presumes some degree of usage or consumption measurement, but can be used to label entirely different outcomes in rate reform in practice depending upon (1) whether and (2) how much fixed charges are tucked in. It is easier to define volumetric pricing or rates by what they are not. They are neither wholly fixed access charges (related only to obtaining entry to call on capacity or supply requirements, without usage or consumption) nor are they wholly measured usage charges per unit of consumption (without recovering the fixed costs of necessary capacity to supply).

The former, in purest form, is an access charge: you pay each time to get in the door. That may be monthly, or each time you arrive, or in days gone by, for example, each time you mailed a letter. In the early days of telephony in Europe, the telephone systems were run by the National Post systems. You could mail a letter with a single stamp that would travel anywhere in that National Post system, from next door or to an address a thousand kilometers away. It is striking to think that instead of a sending a letter at a fixed postal stamp rate at that moment, you could step into an adjoining National Post telephone booth in the same postal lobby to place a telephone call at a purely varying rate determined by the call distance, time, etc.

Of course, there is better recognition of fixed costs today than in the days of the early subsidized telephone and post systems of many European countries. For the purest access charge, it is a fair question if in fact the access price is really hiding the Disneyland “rides are free” (i.e., all variable charges for your expected usage of rides are already tucked into the ticket price).

For the other extreme of entirely (or nearly so) usage sensitivity in volumetric rates, there is  the example of local measured service of telephony, where there are up to four pricing margins for each local call, including the time of day (night, shoulder, and peak) and day of week (weekdays vs weekends) of the call, how often you placed calls (frequency), what distance your call was transmitted, and how long (duration) your call occupied the telephone circuit and kept other users from making a call. Because most circuit switch-equipped telephone exchanges up to the mid 1990s had finite capacity to handle calls, if all circuits were in use the marginal caller’s attempt to place a call was blocked or dropped. On the other hand, if the caller obtained a circuit, there was a non-zero usage cost to the system on several margins that could be priced and passed on to the caller. For the user this provided a great deal of pricing information to help the caller with making voluntary optimizing tradeoffs. I could shorten call duration, I could call less often, and so on.  The problem in the US was that more than 90% of telephone users hated the idea that “the meter was running” on every call they made, so the idea was dropped. In Europe, many countries did not abandon their measured systems until later. For many European systems, lifting the usage disincentive built into detailed usage pricing meant a great deal of new system capacity would be needed. Investing in circuit based switching remained very expensive until IP-based packet switching came along in the 1990s.  We will return to this idea of tradeoffs and repression of usage with usage pricing in a moment.  I’ll note here that in practice in the US, a trivial fixed monthly access charge was used to cover de minimus system costs. For example, I paid $2.25 a month access to an optional four part local measured service tariff when the flat rate access charge was at least $14.50 a month.

Notably, none of these ways of costing access and usage for the user reach the economist’s Holy Grail of marginal prices. This may be changing. In older systems, telephone, electric, gas, water, the cost of measurement with such granularity was cost prohibitive, and certainly not known in real time.  Today, real time pricing that could reflect the marginal cost of usage and supply and instantaneously signal the user has the possibility of optimizing the delicate balance of user demand vs the available supply. There are at least three caveats however. First, the devices that produce the real time information and relay that information to the user (or the user’s automatic devices) are only now reaching the right availability/price points for adoption by suppliers and users. Second, there is the matter of user’s acceptance or compliance with the pricing information given. Some usage is non-negotiable, and price will not change our behavior. Third, and most serious, usage priced at marginal cost, even in real time, will not recover all system costs of providing the needed capacity and distributional infrastructure without charging for access.

How much should be charged for access? Without going into the economics, the allocation of these necessary system costs, because they are largely fixed costs in any at-scale utility system, are typically averaged across classes of similar users in the user base. There are then equitable adjustments made on a regular basis in these averaged system costs to calculate the prices (rates) that would be necessary to recover those costs. And with something for the sine qua non of access included in the cost of serving end users, we are back to the starting point in this conversation: “Whose volumetric rates?”

I need to suggest the current “why” behind the arguments for one or another volumetric rate in today’s regulatory environment.  It is not conservation, or the end of oil reserves, or the imminent depletion of everything, unless one is a Neo-Malthusian. Rather, the fundamental reason why advocates now press for rate reform and some kind of “volumetric rate” is to force greater price responsiveness into customer behaviors. To advocates, customers aren’t being flexible enough and adapting to the requirements of this different kind of grid. Recall my earlier comment about tradeoffs becoming a part of customer behavior in response to measured service prices?  Today, in order to make modernized electric grids work at anything like the reliability and resilience level of our traditional grid means obtaining, through one means or another, the customers’ “demand flexibility.” That is, customers must trim their usage back to whatever the new grid requires whenever the new grid requires it.  The problem as advocates see it is that customers are not price responsive enough to significantly alter their usage behaviors whenever and wherever the new grid needs customer compliance.

Hmm. The lessons from experiments with measured service – widely overlooked by today’s generation of electricity advocates – is that when measured service in the form of a highly usage-loaded volumetric rate is optional, only a small percentage of all users find it beneficial enough for their lifestyles or businesses to bother to subscribe, even with advertising and promotion. When a stiff volumetric rate is mandatory, users are forced to subscribe to receive service at all. In the first case, advocates won’t get the needed demand flexibility to support the new grid dependent upon generation and storage that doesn’t work all the time. In the second case, advocates get forced customer compliance, but they also get customer annoyance. Over time, that discontent with the policy aims of the new utility can undermine the whole net-zero crusade in many unforeseen ways.