Redefining how we take our
electricity supplies. The complexities of allocating fixed costs.
And the need
to recognise environmental costs through carbon pricing.
It is hard to understate the importance of retail tariffs[1]
for the efficient financing and operation of public utilities, and especially
in the power sector. Tariffs represent the pricing and charging structure
through which most consumers are supplied. They influence how consumers use
electricity. Tariffs revenues underpin the returns necessary to pay for utility
investment.
To meet objectives of both equity and economic efficiency, it
is generally accepted that prices should accurately reflect costs. This provides a means to coordinate consumer choices,
on how much and how they use power, with utility decisions on how they manage,
operate and invest in their networks and in their sources of generation.
Purchase, as opposed to sales, tariffs set the terms on which small scale
producers can sell into the network and are an important influence on the
development of decentralised power production in particular. However, interpretation of how best to reflect
costs is, as we shall discover, quite complex and requires careful analysis and
judgement.
In the new world of low carbon energy, three important
trends will change the way in which electricity is produced and delivered, the
shape of future tariffs, and the nature of the service relationship between
utilities and households (and business). These trends are Decarbonisation, Decentralisation and Digitalisation
(the three D’s), and they impact on tariffs.
Decarbonisation of the energy sector is widely perceived as
requiring much greater use of non-fossil electricity to substitute for current
fuels in heating and transport. But it also changes the cost structure and
operational characteristics of generation technology. It substantially reduces
the importance of variable (fuel) costs, which are what largely underpin the
design and operation of today’s markets, and raises the importance of capital
costs. Low carbon technologies (both nuclear and renewables) are inherently
less flexible in adjusting to fluctuations in consumer demand. This raises the
importance of managing consumption patterns, and hence of tariffs and price
signals, within a coordinating price mechanism for balancing supply and demand
in real time.
Decentralisation is implied by the growing significance of
small scale producers (also sometimes known as prosumers), by the smaller scale
of many renewable technologies, and the increasing importance attaching to management
of local network constraints, as electricity plays a large and increasing role
in overall decarbonisation strategies. Tariffs, especially those on which small
consumers can sell to a public network, should be designed to make an efficient
connection between small operators and larger local or national grids.
Finally, digital technologies permit much more complex and sophisticated
information and control systems. These can help maintain stable and balanced
power systems, enable more sophisticated tariffs and consumer choices, and permit
more efficient management of consumer requirements. They are therefore part of
the solution.
We need to review many questions on retail tariffs, and establish
general principles for development of retail tariffs and retail supply in a low
carbon future. These proposals provide a new paradigm within which sector
policies can evolve, and a benchmark against which options should be judged.
The Long Run Marginal Cost (LRMC) Approach
It is widely understood that for both renewable energy (and
nuclear power) the marginal cost of generation, defined as the cost of an
additional unit of production from existing generation assets, production or
from low carbon technologies is close to zero. Setting retail prices at this
zero short run marginal cost (SRMC) is clearly not a viable basis for pricing,
and making the consumer’s marginal electricity consumption free at the point of
use has the potential to create an unlimited demand that cannot be satisfied.
There is however a well-established cost reflective benchmark
for approaches to electricity tariffs based on long run marginal cost (LRMC)
principles. This is intended to ensure that consumers pay the full incremental
costs (at least of generation), including capital costs, that they impose on
the power system “Marginal cost (as LRMC) is an engineering estimate of the
effect upon the future time stream of outlays of a postulated change in the
future time stream of output.”[2]
This implies that the real costs of meeting very different
types of load can be very different. High load factor (eg continuous or
baseload) loads or those that are well matched to patterns of production will
require lower capacity requirements per unit of energy supplied. Low load
factor loads, or loads concentrated in winter when generation is cheaper in
summer, will be more expensive solar power. This makes the calculation of LRMC,
and hence what different kinds of consumption might pay, subject to careful
analysis and calculation. “There are as
many marginal costs as there are conceivable postulated changes.” [3]
The importance of digital technologies is that they allow us
to think about this kind of cost reflectivity in a much more granular way, and reflect
the very different costs implied by different kinds of consumption such as the
traditional household applications, electricity for heat pumps, or the charging
of electric vehicles. Each of these can have a very distinct load profile, with
a very different servicing requirement for the consumer.
Reliability requirements, the differentiated nature of
consumer needs, and supplier managed load
The importance of capacity costs also brings into sharp
relief the fact that the standard of supply reliability is itself a very
important driver of costs. A high standard of reliability, defined as a very
low probability of failure to meet the maximum, unconstrained, instantaneous,
aggregate demand of all consumers, implies a need for very substantial spare
capacity margins. These may be needed to cater for daily and seasonal peak
loads, for generator downtime (eg breakdowns) and for weather related
fluctuations in renewables output.
However not all consumption requirements need the same level
of instant access and reliability. We quite reasonably expect our power need
for lighting, or for a television programme, to be met instantaneously. We are
likely to have a very different approach to, for example, overnight charging of
an electric vehicle battery with perhaps 50 kWh of energy, and be largely
indifferent to when it is delivered, eg overnight or even over two or more
days. It makes sense to permit the supplier to choose the timing of delivery,
within clearly defined parameters, in order to match generation availability
and any network constraints. Other loads, such as laundry, or domestic water
heating, will also have their own requirements, which the consumer can choose,
in relaxing the requirement for instantaneous delivery of power.
What we expect to see in a low carbon future, therefore, is
consumers being able to make a selection from a menu of tariffs, with different
supply arrangements and prices in each case:
·
Some supplies, eg for lighting circuits, taken
at a premium price, with the highest level of guaranteed reliability.
·
Some consumers choosing a lower reliability
standard, at least for some of their needs, with a lower price.
·
Some large loads, such as vehicle battery
charging or heating, provided on the basis that the supplier manages the timing
of energy delivery.
Future systems will place a high premium on pro-active and
effective management, based on innovative tariffs and a redefined approach to
retail supply, of the use of electricity for electric vehicle charging and
domestic heating applications.
Allocation of fixed costs
Generation costs are however only a part of the story. A
substantial proportion of total power sector costs reside in high voltage
transmission, and even more in the local distribution networks. As with many
networks (including road and rail) the marginal cost of accommodating extra
throughput (the extra car or train) is, at least in uncongested networks, very
low. But the fixed cost still needs to be recovered. How best to do it poses
some very difficult questions in terms of reconciling considerations of equity
and income distribution, on the one hand, and the efficient allocation of
economic resources on the other.
Current UK practice for smaller retail consumers, for
example, is simply to average most fixed costs over all units of energy sold.
This seems fair, and prima facie results in those who consume most (and might
broadly also be those with higher incomes) paying the most towards the fixed
costs. However this distorts the economic message, that the actual marginal
cost is much lower. When policies for a low carbon economy include persuading
consumers to use large amounts of extra
electricity for heating (eg with heat pumps), this becomes a very
serious obstacle. For a household consumer, a higher fixed charge in the
tariff, and a lower unit energy charge, transforms the choice between using the
low carbon solution (electric heat pumps) and traditional fossil fuels (gas or
oil).
Another problem arises with purchase tariffs. These provide
an incentive to small scale producers that should, in ideal world, result in
consumers installing their own generation when this is “efficient” and results
in a reduction in total societal costs. However, if the kWh rate in the
purchase tariff is overstated by including an allocation of fixed cost, it will
result in too much own generation. There will be no saving in fixed cost and,
while the individual consumer with own generation may benefit, a larger share
of fixed network costs will be picked up by others.
There are potential answers to this question that not argued
in detail here, since they take us deeper into complex policy, political and
administrative questions than is appropriate for a short article. Possibilities
include the recovery of fixed costs
through property taxes, and approaches in which fixed costs are recovered with
differentiation according to the use to which power is put, for example with a
higher fixed cost levy on EV charging (a premium use of electricity) than for
heating which is in competition with gas.
Reflecting the substantial environmental and climate
costs of CO2 emissions
In the transition to a low carbon economy the case for more
realistic levels of carbon taxation, as an incentive to invest in low carbon
generation assets, and to minimise the share of fossil fuel in both consumption
and production, is overwhelming. However this is not current policy in many
countries. The UK currently has a particularly perverse approach in that the
burden of renewables innovation policy is loaded on to electricity but not on to other fuels, notably gas. A
major plank of low carbon policy is to encourage the use of electricity for
heating, through the medium of heat pumps, and to substitute for gas. But the
impact of current policies imposes a discriminatory tax on electricity, raising
prices and reducing any incentive for consumers to switch from gas. A well
constructed carbon tax, by contrast, would increase the cost of gas, restore a
level playing field, and tilt the balance of running cost comparisons towards the
electric technology of heat pumps. Perversely, recovering the cost of
innovation support through the power sector hampers progress towards a low
carbon economy.
[1]
What we usually mean by a tariff is a set of prices that are published in
advance, are open to all buyers (or for purchase tariffs, sellers) complying
with a given set of conditions. They contrast with bilateral trading
arrangements, and with “market” structures involving multiple buyers and
sellers. They provide the standard route through which most consumers, and
certainly smaller consumers, obtain their supplies of energy, water, and many
communications services. It is quite normal for a supplier to offer a number of
alternative tariff structures, between which consumers can choose an option
that most closely reflects their needs. They can follow either simple one-part or
two-part formats, or have more complex structures.
[2] Ralph
Turvey, one of the pioneers, with Boiteux, of LRMC theory in electricity. Turvey,
R., What are marginal costs and how to estimate them? University of Bath, 2000.
