Customer Charges in the Restructured World:
Historical, Policy, and Technical Issues
adapted from a presentation
to the National Association of
Regulatory Utility Commissioners
Energy Resources and Environment Committee,
July 20, 1999
by William B. Marcus, Principal Economist, JBS Energy, Inc. (email@example.com)
Eugene P. Coyle, Ph.D., Economic Counsel (firstname.lastname@example.org)
This presentation identifies a new trend to raise both customer costs and customer charges under restructuring – disproportionately
raising regulated distribution rates to small customers, and charging a higher percentage of these rates in fixed charges. It recalls the past history of this debate and identifies policy issues related to the
new rise of customer costs and customer charges.
The Trend to Higher Customer Charges, Customer Costs, and Rising Residential Distribution Charges
In the last few months, utilities across North America have made a number of proposals to raise residential customer charges, or to impose
them where none existed before. The most extreme cases that are in Nevada, where Nevada Power proposed residential customer charges in the range of $40 to $55 per month before any electricity is used, and
Sierra Pacific looks moderate by comparison with proposed charges of about $20 per month. Charges in excess of $20 were also proposed in Alberta, Canada. Even in areas that have not had significant
electric residential electric customer charges for years, such as California, utilities are proposing new customer charges.
Along with raising customer charges, utilities are also changing cost allocation to increase the amount of costs
that are considered to be customer-related, thereby raising regulated distribution rates for residential and other small customers. In many cases, they are calling more of the distribution system
customer-related than they used to. For the first time in nearly 20 years, Pacific Gas and Electric is asking the California PUC to deem part of the primary distribution system to be customer related. As
a result, PG&E proposes a 40% industrial rate reduction from 1996 levels when stranded costs are paid off. Residential customers (including rate reduction bonds) will see decreases of only 15% from 1996
In other places, utilities have been using creative accounting to place expenses in the Customer Service and Information category rather
than in more appropriate places in the uniform system of accounts. It is an even more common practice to hide marketing and major account representative services for large customers in accounts where they are
considered non-competitive corporate overhead costs. As a result, these costs are neither directly assigned to and paid by large customers, nor are they subjected to competition.
In sum, restructuring the electric industry is bringing about an attempt to shift allocation of the remaining regulated costs to raise residential rates as a whole and to raise fixed charges for residential customers.
The 25 Years of Historical Perspective That Are Being Conveniently Forgotten
Prior to 1973 there had been a lot written about how to allocate costs. But there had NOT been much adversarial exploration of these
assertions about cost allocation.
The California Commission, which had one of the largest staffs in the nation, and which had a major input in
the NARUC Cost Allocation Manual, really didn't challenge the assertions put forward by the utilities, and didn't do cost allocation studies – nor, to our knowledge, did any other Commissions.
That first NARUC manual was basically a reporting of how the utilities proposed to do cost allocation. Cost allocation was arcane, complicated, and could not be easily done without computer technology that was
expensive and cumbersome. Even today, cost allocation models are often far more complex than they need to be because of their initial construction in the days before cheap spreadsheet technology.
There were two eras of change in the cost of service assumptions underlying cost allocation and customer charges. Between roughly
1973-1980, a small number of economists were challenging the utilities myths about cost allocation. And I'm going to focus on "customer costs" to the exclusion of other important fights.
In defense of large customer charges (and to some extent, an alternative version, the minimum bill) the utilities had developed some
arguments, which largely proved to be mythical. The task was to divide the distribution system into “demand” and “customer” portions.
The myths started with the proposition that everything that wasn't clearly "demand related" was "customer
cost." Then two myths came into play, the "Phantom system" -- or "minimum system" a system designed to keep from falling down from its own weight. The second was "the Zero
Intercept method" -- the idea that plotting the cost of various-sized components as dots on a graph, e. g. poles of 30 feet, 40 feet, 50 feet in height, and then drawing a line connecting the dots and extending
it to zero-sized poles gave you a legitimate cost for a zero-sized pole.
The cost of running a system beside a golf course or empty lots – or empty neighborhoods was simply divided
into the number of customers. But simply dividing everything by the number of customers was arithmetic, not analysis.
To be brief, analysts were able to show was that the tales and myths that had long ruled rate design and cost allocation didn't afford a
successful defense of large customer charges. In many states the assault by the consumer economists carried the day.
As part of the debunking of this myth, George Sterzinger wrote an influential paper showing that the customer
charge resulted in double collecting from the smallest customers, particularly if a minimum system method was used to calculate demand costs, because the minimum system could carry a large portion of the residential
The second round of customer cost analysis, in the early 1980s, involved an examination of several other issues. With a deeper
analysis of distribution cost allocation, some other very interesting things turned up.
First, the dollars of investment in transformers depends, on the price of fuel. If fuel costs were high, it
made sense to invest in a lower-loss transformer. There are two kinds of transformer losses -- core and copper. Core losses are energy losses from a transformer if energized, even if no customers are
taking power on that circuit. “Copper” losses are the additional losses if customers are taking power that increase with the square of the load. Transformer capital spending can be
substituted for fuel (and generation capital). By buying a lower loss transformer, at a higher dollar outlay, fuel is saved. So even some costs that were previously "demand" could now be seen
to be "energy" related -- even further removed from "customer cost."
We also learned, in this era of massive data requests, that even service drops were sized depending on the load at
the house. Houses with more demand had bigger and more expensive service lines. While this sounds like an obvious truism, it has implications for the use of customer charges in rate design. If it
is claimed that some part of the system is a "customer cost" to be legitimately collected in a "customer charge", then if each customer is asked to pay the same dollar amount, then small users within a customer class are going to be subsidizing larger ones.
Third, we discovered in the early 1980s that the size of the customer is positively correlated in a non-linear way with the dollars in the
local distribution system. What this discovery meant was that the bigger the residential customer, the bigger the distribution investment -- in an accelerating way. In other words, the dollars of
investment went up faster than the size of the customer.
Finally, we learned that the myths about customer size and capacity factors, and customer size and coincidence factors were wrong as well.
The smallest customers used more kWhs per kW of demand than did the largest, CONTRARY to what everybody "knew." And the smallest customers were really off the system peak, and off the class
peak as well. So using the traditional allocators for demand costs, CP and NCP, was unfair.
Thus, if you take the total distribution investment, and divide by the number of customers, to get the customer
component of costs, as is proposed in Nevada, the small customers are going to get an unfair allocation.
The Motivation of Utilities to Support Fixed Charges in a Restructured World
Now we come to the present. Why are customer charges coming back into vogue, even though distribution
allegedly remains regulated and not subject to significant competition?
In sum, electric utilities with high customer charges can fend off competition from energy efficiency and distributed generation, position
themselves well against gas companies, ensure revenue stability, increase profits of affiliated marketers and generators, and (if they don’t go all the way to 100% distribution customer charges) increase
profits of the regulated company under performance-based ratemaking (PBR).
Each of these benefits of customer charges to the utility is discussed in more detail below.
The old competition was energy efficiency investments instead of purchasing kilowatt-hours. The new competition is distributed
generation (DG). The demand-side management (DSM) boom of a few years ago increased the utilities’ role in energy efficiency. The restructured world is reducing the utilities’ role in
DSM. Now that it is harder for utilities to control DSM, it is in their interest to throw up roadblocks to efficiency. In terms of the new competition, high customer charges take control over DG away
from customers and giving it to utilities.
Some utilities are open about it. Sierra Pacific Power in a California PUC filing said that an advantage of
charging all distribution prices in fixed costs is that it removes incentives from customers to install DG.
Just as the cost of photovoltaics is coming down in applications such as PV roof tiles, a large reduction in energy charges financed by
rising customer charges will block it. Is this the market at work or a scared monopolist?
The third form of competition is between a utility as a default provider (which can make profit from providing default service) and
alternative sellers. High customer charges can reduce the leeway that retailers have to add value for residential customers by helping to control the efficiency with which they use energy.
In addition to making life difficult for energy efficiency and DG, high customer charges may help electric
utilities in their old war against gas space and water heating. If electric distribution rates are collected in fixed charges, regardless of whether the customer has electric or gas space and water heating, then an
energy rate that collects only generation and transmission costs becomes attractive in the electric utility’s attempt to grab gas company customers. In essence, this rate design makes it cheaper for the
customer to have the electric utility burn gas or coal relatively inefficiently in a powerplant, transmit that electricity over the whole infrastructure, and deliver it to the customer so that the customer
doesn’t choose to burn gas directly, at greater efficiency, in his or her on-site water heater.
The fallacy of this argument is that large portions of the distribution system (even for a summer peaking utility)
must be built more expensively to serve neighborhoods of new electric heat customers than customers using gas for space and water heating. In essence, small customers, though innocent bystanders, could end up
as casualties in the age-old fight between gas and electric heat by cross-subsidizing the electric utility.
Utilities will collect their money regardless of economic conditions and will not have to worry significantly about demand risk ever again.
Again, utilities are open about this. Nevada Power said that with its proposed distribution rate design
based exclusively on fixed charges, it could bill the customer (or their retailer in a customer choice environment) in advance. No pesky working capital. No risk of collecting the money. No meter
reading costs for the distribution company, since it doesn’t need to read the meter to get its money. But Nevada Power still wants an 11.9% return on equity to compensate for all the risk that it sees of
being in the electric distribution business. Some things never change.
4. Under certain conditions, if demand increases due to high customer charges utilities’ profit increases.
If the utility isn’t going all the way to 100% fixed cost recovery, increased demand can bump up profits. If the utility is
raising its customer charge but still collecting some costs in energy rates and has a PBR structure, its profits will go up if it can design rates that cause customers to use more electricity.
Moreover, if the utility makes profits from default marketing or has a strong affiliate marketing energy, higher sales induced by a
fixed-charge rate design will raise energy-related profits.
Critical Policy Issues
Customer Charges, Marketing, and Discrimination
Assertion: Customer charges were always about marketing electricity and will be in the future.
In most unregulated industries, customer charges are not established separate from service. No one pays a customer charge to enter a supermarket. Only in some rare examples, have customer charges been
used for marketing. For example, nightclubs have had three different means of charging, depending on how renowned its performers were: a cover charge for the most popular performers, a two-drink minimum for
the next rank, and no charge on open-mike night. Customer charges are coming into vogue in long-distance telephone as a means of raising revenue while still encouraging more lucrative large users to switch
Customer charges and their close cousins, declining block rates, were the foundation of the all-electric marketing
binge of the 1960s and 1970s that is returning, ironically, as part of DSM for some utilities.
The issue going forward, then, is control over marketing. The analysis of customer charges is about constraining unfair marketing.
One likely outcome is that the customer charges are going to be used in unfair discrimination among customers. Most states are
supposed to deliver "Just, reasonable, and non-discriminatory rates." That looks to me to be out the window in the future.
Customer charges are one way to discriminate against customers seen to be, if not actually be, unprofitable.
Suppose that margins are small per kWh – then a marketer wants only customers who buy a lot of energy. If one is offering
only a price per kWh, it becomes difficult to discourage or exclude the frugal customers. However, by combining the per kWh price with a large customer charge, the effective cost per kWh cost is raised more to
the small customer than to the large customer. The small customers will elect to shop elsewhere if they can -- i. e. you will be rid of them. If the utility cannot rid itself of them, because they
are “default” customers, it can charge them more for the privilege of staying without affecting the competitive needs of other customers.
Distribution System Efficiency: The Incentives are Going Away
One issue that falls out of the previous discussion is that if it makes sense to substitute capital for energy on
the distribution system to reduce losses, then who is going to do it? In an unbundled world, where the Distribution Company (Disco) is separate from generation and acts only as an owner of wires, why should
the Disco spend money to reduce losses? Even if the Disco is a default provider of generation, those costs are a pass-through providing no incentive to reduce losses. We have lost an incentive for
efficiency in this area through the break-up of the system into its components. With the onset of PBR, with its incentives to reduce capital spending, the picture could get worse. In essence, we need to
be concerned that distribution losses will creep up in the new world when no one is charged with doing anything about them.
Raising Customer Charges, Demand Elasticity, and Energy Efficiency
A key question that must be considered is the extent to which higher customer charges are likely to increase demand.
In an extreme case, such as that presented by Nevada Power, the rate design is being changed from a $5 customer charge and a 6.2 cent/kWh
energy charge to a $40-$55 customer charge and an energy charge of 3.9 cents/kWh. This is a reduction in the energy charge of about 40%.
The proponents of high customer charges often claim not to have considered the elasticity of demand when making their proposals.
To assess the impact, we come to a key technical question, one of those questions about which economists disagree. Is elasticity of
demand based on marginal or average rates or both? Some entities, such as Pacific Gas and Electric, believe that marginal rates affect demand. Others use average rates.
If demand elasticity is based on marginal rates, a 40% reduction in energy charges could ultimately raise
residential consumption in the vicinity of 5% in the near term and 20-30%, all else being equal, in long term.
In thinking about this question, we must examine institutional factors that may lead us toward the high end of the range.
First, a fixed charge rate design makes any remaining efforts to reduce energy consumption that much more difficult and expensive. Any
incentives that are offered to deal with market imperfections must be bigger with high customer charges and lower energy charges. The higher customer charges create a system of dueling incentives.
Therefore, any state that wants to keep public purpose programs to support energy efficiency will either pay more or accomplish less under a rate design with more emphasis on fixed charges.
Second, even larger impacts on demand could result if federal and state appliance and building standards are
relaxed because saving energy is no longer as cost-effective as before.
In other words, a rate design shift will make GHG reduction more burdensome. On a national policy level, removing incentives to
efficiency in rate design is one more step that is essentially moving the U.S. away from a conservation strategy to GHG reduction toward a generation strategy which is more expensive in the long run. It is
thus one more sign of encouragement to the voices that claim that it is too expensive to our economy to stop global climate change.
Even if one supports renewable energy and even if one specifically supports the marketing of green power, the best
way to reduce emissions associated with energy use is to make inexpensive investments so that we will not use the energy in the first place, not to build renewable plants to produce all the energy that we might
possibly consume without making those investments.
The Opposing Economists’ View: We Were All Wrong to Want Energy Efficiency in the First Place
You will hear from some economists that the policy shift is good – that we should be using more energy, and
that high customer charges will remove the market imperfections that are stopping American residential customers from using enough energy to be economically efficient. It is interesting that the California
Energy Commission, which still has an agency mission of supporting energy efficiency and renewable energy, and which took aggressive action to work against customer charges in the early 1980s has made a 180-degree
turn. Its Staff is now among the most forceful exponents of the economic efficiency of higher customer charges, and that Staff position has been adopted by the Commission.
This view essentially assumes that nothing is wrong with energy markets that can’t be cured with a dose of promotional rate design. In
essence, it is assumed that residential customers are the efficient consumers of the economics textbook, and that the only thing preventing rational behavior is electric rate design.
This textbook view assumes that the case for energy efficiency investments and DSM made in the early 1990s was misguided and false.
First, it assumes that consumers have perfect information on the costs of obtaining energy services. In other words, they can perfectly balance choices between kilowatt-hours and more efficient
refrigerators. A subset of this assumption is that there are no split incentives between landlords and tenants because tenants know about the energy use of competing apartments. There are no split
incentives between homebuilders and home buyers, because home buyers understand the future energy use of appliances installed in new homes by builders and can make rational choices between efficient appliances,
upgraded carpets, and buying another house altogether. This view also assumes that access to capital at different costs is not relevant in energy markets. Finally, of course, this view assumes that there
are no external effects from air pollution, water pollution and construction of transmission lines through open space that have not been perfectly internalized. We could be in the best of all possible worlds,
if we would just increase those customer charges. In other words, adherents of this view believe that our entire policy since the energy crisis of 1974 has been misguided.
Line Extension Hookup Charges as Alternatives to Customer Charges
If the issue were truly about reflecting costs, then the most economically efficient means of doing this is to reduce line extension
allowances given to developers of new housing. Local governments generally charge hookup fees for the utilities under their control (water and sewer), and a number of municipal utilities in Northern California
give either no allowances or much smaller allowances to developers than the private utilities.
The days of using allowances as a marketing tool to promote the use of electricity, even if arguably ever
reasonable in the past, are clearly gone in a restructured world. The distribution company only provides wire service. This fundamental fact is likely to lead at least to a significant reduction in
allowances in many places, if they were previously based on other types of system costs, and could provide arguments for their elimination altogether.
If the developer pays for more of the distribution system up front, it will end up being financed by homeowners in house prices at lower
interest than the utility’s rate of return including income taxes. Distribution costs and hence rates will decline, and the costs that will be reduced are the ones that the utilities are most interested
in placing in customer charges.
The major losers are the utilities, which lose some of the most risk-free rate base on their systems, and home
developers, who arguably have been subsidized. But, asking the utilities to collect money up front from developers is a strong alternative to reversing 25 years of history to increase fixed customer charges.
The trend toward increasing customer charges requires a strong but informed response.
It is important, particularly for those who are relatively new to the regulatory process, to understand the history – that these are
not new issues. Many of the fallacies in utility logic were identified and debunked 15 or 20 years ago. We just need to bring them back to the collective memory.
It is important to understand the motivation of utilities to improve their competitive position, improve the certainty of their cash flow,
and discriminate among customers to market electricity.
It is important to regain incentives for distribution system efficiency that have been lost in the unbundling of the vertically integrated
utility and to look at changing the balance between ratepayers, shareholders, and developers through reductions in line extension allowances.
Finally, it is important to understand that the result of a rush to customer charges may result in burgeoning demand, backsliding on energy
efficiency programs and standards, and unthinkingly throwing away the most cost-effective responses to global climate change.