Memorandum by Professor Andrew Bain
I am a semi-retired professional economist.
During my career I have been a Professor of Economics at several
Scottish universities and from 1984-90 was chief economist at
Midland Bank. I have been involved with a number of government
bodies, including the Wilson Committee on the City, and I served
on the Board of Scottish Enterprise for eight years. I am currently
a Member of the Competition Appeal Tribunal. I participated in
the Royal Society of Edinburgh's Inquiry in Energy, though I resigned
before the Report was completed, and I have given evidence at
Public Inquiries into the proposed Beauly-Denny transmission line
and the proposed Beinn Mhor wind farm in Lewis.
1. INTRODUCTION
1.1 I welcome this opportunity to submit
evidence to the House of Lords Select Committee on Economic Affairs
in connection with its inquiry into The Economics of Renewable
Energy. My particular concern is with the costs of electricity
transmission and their impact on the location of renewable generation.
It is important to ensure that the incentives and constraints
facing developers provide the correct locational incentives and
do not lead to electricity consumers incurring unnecessarily high
costs for renewable energy. Consideration of transmission costs
led the RSE Committee to express its "grave doubts about
the overall economic rationale for large-scale wind turbines in
locations remote from the consumer."[1]
1.2 In a properly functioning system electricity
consumers would pay the same for energy, regardless of the location
of the generator. By setting renewables targets, imposing renewables
obligations on distributors, and granting variable entitlements
to renewables obligation certificates (ROCs), the UK and Scottish
Governments have fragmented the energy market, so that energy
generated from renewables receives a premium which varies according
to the source of the energy. As a result consumers will pay more
for wave or tidal energy than for energy from wind, and more for
energy from off-shore than from on-shore wind. Nevertheless, within
each category the cost to the consumer should be independent of
the location of the generator, with any differences in production
and transport costs being reflected in the profits of the generating
companies.
2. TRANSMISSION
CHARGES
2.1 For this to occur the costs necessarily
incurred in transmitting energy from a generator in, say, Central
Scotland to a distribution company in, say, the London area would
be charged to the generator. That is not what happens in practice.
There are two categories of transmission cost: the capital and
operating costs of the grid, and the transmission losses associated
with transmission through the grid. Charges to generators for
the use of the grid are determined using National Grid's Transmission
Network Use of System (TNUoS) Model. No account is taken at present
of transmission losses.
2.2 Transmission losses are materially greater
when the generator is distant from the consumer, and are likely
to be particularly large for generators in the Western and Northern
Islands of Scotland. For example, generation losses on the proposed
interconnector to transmit energy from the Western Isles as far
as Beauly, west of Inverness, are said to amount some 5% of output,
and for energy generated in the Western Isles there would be further
significant losses during transmission from Beauly to the south
of England. Failure to take account of these losses means that
generators are receiving payment for significantly more energy
than they are delivering to consumers—a cost that is borne
by electricity consumers generally. Devising a robust method for
allocating transmission losses to generators, and imposing charges
correspondingly, should now have a high priority.
2.3 The TNUoS model used at present is not
an appropriate model for allocating the capital and operating
costs of the grid to intermittent generators. It is a peak demand
pricing model, in which the charges for grid access reflect the
use made of the grid at peak demand periods. This presupposes
that the grid configuration is determined by the configuration
of generation and demand at that peak demand. For intermittent
renewables generation this is patently not the case—what
is required is a model that takes account of peak supply.
2.4 To illustrate. The scale of wind generation
in Scotland is geared to exporting energy from Scotland, and the
peak transmission requirement from intermittent generators in
Scotland will occur when demand in Scotland is low but wind conditions
are favourable. For example, on a warm and windy summer night
Scotland's two nuclear power station plus its conventional stations
operating at minimum practical loads already provide more generation
than is required to meet Scotland's demand, with the result that
some energy has to be exported through the Scotland-England interconnector.
To provide sufficient capacity to accommodate the proposed high
level of wind-energy from Scotland the interconnector will have
to be substantially reinforced, as will major sections of the
grid to the north and south of it. This has absolutely nothing
to do with meeting peak UK demand. Indeed it is not unlikely that
peak UK demand will coincide with severe winter conditions, when
an anticyclone with still air predominates over the UK and wind
turbines make very little contribution to supply.
2.5 There are other weaknesses in the TNUoS
model. It allocates less than 30% of the costs of the grid network
to generators, with the rest being payable by distribution companies.
There is no evident rationale for this division, though it may
reflect the fact that there is a high degree of redundancy built
into the system to ensure that the supply of energy to consumers
is secure in the event of unavailability of sections of the grid.
This redundancy is of much greater benefit to consumers than to
generators.[2]
The TNUoS model also assumes that the grid has a 40-year life
and that capital charges can be geared to that period. Wind farms
are currently assumed to have a 25-year life, and may well be
superseded by other technologies after that period. If account
was taken of the "stranded asset" risk TNUoS charges
for wind farms would be higher.
2.6 Transmission charges may also be adjusted
by government direction under Section 185 of the Energy Act 2004.
This gives the government power to adjust the transmission charges
for generators in particular areas if it seems likely that renewable
generation would otherwise be discouraged. The government has
decided that the additional costs of transmission from the Scottish
Islands should be halved. There is no evidence that industrial-scale
wind turbines in eg the Western Isles will be needed in order
to meet the Scottish Government's 50% target for renewable generation
(see below), so there is no compelling case for encouraging wind
farms in Lewis. Nevertheless, if a new connector for the Western
Isles is constructed and generators pay for only half of the additional
costs, consumers throughout the UK will end up paying more for
wind-energy from these Islands than from other regions.
2.7 The upshot is that the present method
for determining transmission charges needs a radical overhaul
to ensure that the financial incentives facing wind farm developers
encourage development where the costs, including transmission
costs, to the consumer will be lowest.
3. SCOTTISH RENEWABLES
POLICY AND
THE NEED
FOR INVESTMENT
IN THE
GRID
3.1 Regardless of the financial incentives
for renewables development, actual development is frequently constrained
by planning requirements and by access to the grid. These are
driven by a mixture of economic and political considerations.
3.2 It is generally accepted that within
a European context conditions in the north of Scotland, and in
particular in the Western and Northern Isles, are uniquely favourable
to renewables generation from wind and marine sources. This has
led many to conclude that there is a very strong case for developing
wind and marine generation in these areas. Statements such as
"Strategically it make sense to harvest renewable energy
where the resource is greatest and operational efficiencies can
be maximised"[3]
are treated as axiomatic by the Scottish government and by regional
authorities in Scotland. The need to transport the energy to the
market, and the costs incurred in doing so, are overlooked. Far
from it being axiomatic that renewable energy should be harvested
where the resource is greatest, it is appropriate to harvest energy
in these areas only if the operational efficiencies are sufficient
to compete with other renewable energy sources and, in addition,
more than compensate for any higher transmission costs.
3.3 This "axiom" underlies much
of Scottish policy towards renewables. It is prayed in aid in
support of the Scottish Government's target for renewables generation
equal to 50% of Scotland's electricity consumption in 2020. It
is reflected in continuing political pressure to relieve generators
in the North of Scotland and the Islands of part or all of the
transmission costs, as evidenced by, for example, Section 185
of the Energy Act 2004, statements from Scotland's First Minister
that the TNUoS charges are unfair, and a recent consultancy report
commissioned by the Highlands and Islands Councils and Highlands
and Islands Enterprise recommending greater "socialisation"
of grid costs.[4]
3.4 It is also reflected in the Scottish
Government's draft National Planning Framework 2 (NPF2)[5].
This includes amongst its "national infrastructure priorities"
proposed grid reinforcements to facilitate the attainment of the
target and promote marine development. The proposals emanate from
the Scottish grid companies, Scottish Hydro Electric Transmission
Limited (SHETL) and Scottish Power Transmission Limited (SPTL),
in response to applications for connections from potential wind
farms. The total capacity in these applications, many of which
will be turned down on planning grounds, greatly exceeds the amount
required to attain the 50% target. There is no evidence that the
grid priorities set out in NPF2 have had regard to the likelihood
of planning consents for the proposed developments, to the capacity
of the existing grid to cope with increased levels of intermittent
generation, to the economic viability of wind farms if the transmission
costs had to be borne by the generators, or to the likelihood
that wave and tidal development will ever prove to be economic.
Moreover, the Scottish Government's 50% target is predicated on
much of the energy being exported to the Midlands and South of
England. To the extent that generators do not pay for all the
necessary grid reinforcement consumers throughout the UK, not
only in Scotland, will have to bear the cost.
3.5 NPF2 recognises that individual proposals
for strengthening the grid will not be implemented without first
being approved by Ofgem. Before approving investment in the grid
Ofgem conducts an economic assessment, with the object of ensuring
that the investment is economically justified. So the fact that
a particular proposal for new grid is identified as a national
priority by the Scottish Government does not necessarily mean
that it will happen. But it does mean that there is strong political
and industrial pressure on Ofgem to approve the proposal, particularly
when Ofgem's objectives include not only the promotion of efficiency
and economy in supply and transmission but also to contribute
to the achievement of sustainable development[6].
If Ofgem turns down a Scottish "national priority" on
the strength of an unfavourable economic assessment it will inevitably
be accused both of failing to contribute to sustainable development
and of thwarting the policies of the democratically-elected Scottish
Government. There is danger that this may put more strain on the
regulator than it can reasonably be expected to bear, with consequences
for the rigour of its economic assessment.
4. OFGEM'S
ECONOMIC ASSESSMENT
METHODOLOGY
4.1 When so much depends on Ofgem it is
essential that its methodology for assessing the economics of
grid reinforcement is sound. Unfortunately, this cannot be relied
on: the methodology adopted by Ofgem in 2004 for identifying its
baseline programme of grid investment was defective[7].
Its assessment of the proposed Beauly-Denny grid line provides
an example.
4.2 Ofgem approved the Beauly-Denny grid
reinforcement as part of its baseline programme on the strength
of an economic analysis in which it compared the present value
of the costs of constraining off energy—primarily the cost
of compensating generators—if the grid was not reinforced
with the present value of the costs of reinforcing the grid to
accommodate additional wind farms in the north of Scotland. It
found that for the likely level of wind farm development the constraint
costs exceeded the cost of the line, and that the reinforcement
was therefore justified.
4.3 Ofgem's methodology was defective in
several respects. It assumed that the same level of security in
the system was required for intermittent generators as for conventional
generators[8].
That meant the capacity of the existing grid was taken as some
500MW less than it could actually transmit[9],
ie the existing grid system between Beauly and central Scotland
was capable of transmitting another 500MW before constraint costs
would have to be incurred.
4.4 No allowance was made for the rather
obvious possibility of storing energy by operating hydro stations
in a way that complements wind power and reduces the need to constrain
off wind-energy in peak generating conditions[10],
and the scope for using the pumped storage facility at Foyers
to add to the demand load at wind power peaks, and thus avoid
constraining off, was underestimated; no account at all was taken
of the possibility of operating the Cruachan pumped storage facility
(in the SPTL area) to provide temporary storage for peak generation
from wind farms along the southern part of the proposed line.
4.5 No account was taken of the cost of
the further substantial grid reinforcement to the south of Denny
that would be required to enable the energy to be transmitted
to southern England.
4.6 Taken together these omissions fatally
undermined Ofgem's conclusion that the Beauly-Denny proposed grid
line was economically justified[11].
4.7 Apart from these errors of implementation,
Ofgem's methodology is inherently unsound because it effectively
double-counts the ROCs associated with constrained-off energy.
This is because Ofgem assumes that the wind farm developments
will be in place, and then calculates the constraint costs, including
the cost of the ROCs. So, the income from ROCs is taken into account,
first by the developer in deciding whether to proceed with the
development and then again by Ofgem in respect of any energy that
has to be constrained off the system when considering the case
for grid reinforcement. A proper economic assessment of the whole
project, namely constructing a wind farm together with any associated
grid reinforcement that will be needed, would take the total capital
cost and compare it with the total expected net revenue from the
development, counting the ROCs only once.
4.8 My conclusion is that experience to
date does not give grounds for confidence that Ofgem will ensure
that only those grid reinforcements that are justified by economic
fundamentals receive authorisation. It is in the financial interest
of the grid companies to maximise the size of the grid, because
once authorised they can reasonably expect to earn profits on
it over its assumed 40-year life. The political pressures will
always weigh in favour of allowing proposals to proceed. The present
arrangements put too much weight on the proposals from the companies,
with too little independent objective scrutiny, backed by political
muscle, of the need for reinforcing the grid. The outcome is likely
to be that electricity consumers in future will be saddled with
unnecessarily high grid costs.
13 June 2008
1 Royal Society of Edinburgh Inquiry into Energy Issues
in Scotland, Final Report, June 2006, p.73 Back
2
The TNUoS model is intended to ensure that the differential
between generators whose energy is transmitted on average over
great distances and those that are close to their markets reflects
the difference in the grid costs, but the absolute amounts of
the charges are not geared to the absolute costs involved. Back
3
Beinn Mhor Power Public Local Inquiry, Scottish Government Reference
IECIEC/3134, Precognition of Calum Iain Maciver, Head of Economic
Development of Comhairle nan Eilan Siar, paragraph 69(5). Back
4
Xero Energy, Grid Connection of the Scottish Islands-A Strategic
Viewpoint, 20 June 2007. Back
5
National Planning Framework for Scotland 2: Discussion Draft,
paragraphs 226-9 Back
6
Electricity Act 1989, Section 3A (a) and (ba), as amended. Back
7
Ofgem, Transmission investment for renewable generation-final
proposals, December 2004. Back
8
SHETL now recognise that a lower level of security-a "non-firm
connector"-is appropriate for wind-energy from eg the Western
Isles. Back
9
Subject to some relatively minor capital spending for "operational
intertripping" to protect the rest of the system if a fault
arose. Back
10
SHETL now recognises that this is feasible. Back
11
Ofgem did not give evidence to the Beauly-Denny Public Inquiry. Back
| 
