Cepa Report - Analysis of Emr Policy Options For Northern Ireland 2

1
LLLC1RICI1Y MARKL1 RLIORM

ANALYSIS OI POLICY OP1IONS IOR DL1I AND NIAUR

October 20JJ

Iinal report

IOR PUBLICA1ION

Submitted by:

Cambridge Lconomic Policy Associates Ltd

P k u l 5
c P N l

IMPOR1AN1 NO1ICL
1his report has been commissioned by the Department o Lnterprise 1rade and Inestment and the
Northern Ireland Authority or Utility Regulation. Ioweer, the iews expressed are those o CLPA
alone. CLPA accepts no liability or use o this report or any inormation contained therein by any
third party. CLPA also accepts no liability or the use o data rom third parties which has not been
independently eriied.
All rights resered by Cambridge Lconomic Policy Associates Ltd.

CON1LN1S
Lxecutive Summary ................................................................................................................... i
J. Introduction ....................................................................................................................... J
1.1. Context .............................................................................................................................................. 1
1.2. 1erms o reerence .......................................................................................................................... 1
1.3. Structure o report ........................................................................................................................... 3
Part A: Context, issues and options .......................................................................................... 4
2. Context and issues ............................................................................................................. 4
2.1. 1he Northern Ireland context ....................................................................................................... 4
2.2. 1he Great Britain context...............................................................................................................
2.3. 1he Republic o Ireland context ................................................................................................... 9
2.4. Subsidy design context .................................................................................................................. 10
2.5. Operation o the RLlI1 .............................................................................................................. 14
2.6. Operation o the Single Llectricity Market ,SLM, ................................................................... 14
2.. Issues ............................................................................................................................................... 18
2.8. Summary ......................................................................................................................................... 20
3. Options ............................................................................................................................. 2J
3.1. Options or structuring subsidy ................................................................................................... 21
3.2. lixed lI1 ........................................................................................................................................ 22
3.3. RLlI1 arrangement ...................................................................................................................... 23
3.4. CD lI1 .......................................................................................................................................... 25
3.5. Premium lI1 .................................................................................................................................. 29
3.6. RO ,or other quantity based instrument, ................................................................................... 34
3.. Combinations o options .............................................................................................................. 35
3.8. Options or unding subsidy ........................................................................................................ 36
3.9. 1reatment o small-scale renewables .......................................................................................... 38
4. Lvaluation of options ....................................................................................................... 39
4.1. Laluation o options .................................................................................................................... 39
4.2. Questions and issues to be considered in Part B ...................................................................... 45
PAR1 B: ANALYSIS OI POLICY OP1IONS ...................................................................... 48
S. Modelling framework and assumptions .......................................................................... 49

5.1. Model structure .............................................................................................................................. 49
5.2. Modelling LMR options ............................................................................................................... 51
5.3. Model inputs ................................................................................................................................... 56
5.4. Summary ......................................................................................................................................... 61
6. Baseline ............................................................................................................................ 62
6.1. Oeriew ......................................................................................................................................... 62
6.2. Baseline assumptions ..................................................................................................................... 62
6.3. Results - inestment and plant mix ............................................................................................ 63
6.4. Results - wholesale prices ............................................................................................................ 65
6.5. Results - renewable generation support payments ................................................................... 65
6.6. Results - cost beneit analysis ...................................................................................................... 66
6.. Summary ......................................................................................................................................... 1
7. Carbon Price Support ....................................................................................................... 72
8. Analysis of II1 options ................................................................................................... 73
8.1. Oeriew ......................................................................................................................................... 3
8.2. Premium lI1 policy ...................................................................................................................... 3
8.3. CD lI1 policy .............................................................................................................................. 6
8.4. Results - inestment and plant mix ............................................................................................
8.5. Consumer cost comparison.......................................................................................................... 9
8.6. Summary ......................................................................................................................................... 85
9. Small-scale generation ..................................................................................................... 87
9.1. Small-scale lI1 design .................................................................................................................. 8
9.2. Generation mix .............................................................................................................................. 88
9.3. Consumer costs .............................................................................................................................. 89
9.4. Implementation issues................................................................................................................... 90
9.5. Conclusions on small-scale generation ....................................................................................... 90
J0. Sensitivities................................................................................................................... 9J
10.1. Continuation o the NIRO beyond 201 ................................................................................... 91
10.2. Low-gas price sensitiity ............................................................................................................... 92
10.3. Low marine , oshore wind deployment .................................................................................. 94
10.4. Summary ......................................................................................................................................... 95
JJ. Conclusions ...................................................................................................................... 96
Appendix A: Operation of Renewable energy Ieed-in 1ariff (RLII1) ................................. 98

Appendix B: Premium II1 in the SLM ................................................................................ J07
Appendix C: Wholesale electricity price projections ............................................................. JJ3
Appendix D: Generation capacity assumptions .................................................................... J20
Appendix L: Generation cost assumptions ........................................................................... J24

i

LXLCU1IVL SUMMARY
L.J Context: Llectricity Market Reform
1he UK Goernment published its new Lnergy \hite Paper
1
in July ollowing its Llectricity Market
Reorm ,LMR, consultation process. 1he \hite Paper sets out a package o reorms to the United
Kingdom ,UK, and Great Britain`s ,GB, electricity policy, including a new system o long-term
contacts in the orm o leed-in 1aris ,lI1s, with Contracts or Dierence ,CD, which will
replace the existing support mechanism or renewable electricity generation in GB ,the Renewables
Obligation ,RO,,.
2

1he \hite paper also highlights that it is likely an organisation ,or organisations, at arm`s length
rom Goernment will negotiate and administer the CD lI1s or GB. A decision on the roles and
responsibilities o Goernment and those o the deliery institution ,s,, as well as more detail on the
unctions, contracting strategy and planning cycle o the CD lI1 policy, are expected to be set out
around the turn o the year.
L.2 Purpose
Publication o the \hite Paper means that changes will be needed to the current renewable
electricity support scheme or Northern Ireland ,the NIRO,, which were not explicitly set out within
the \hite Paper. CLPA has been commissioned by DL1I and NIAUR to look at the implications
or Northern Ireland o DLCC`s LMR package, ocusing on the implications o implementing a
CD lI1 or Premium leed-in 1ari ,PlI1, in the Single Llectricity Market ,SLM,. \e hae
assessed a range o policy options through a combination o quantitatie modelling and qualitatie
analysis.
L.3 Options to be considered
1he high-leel options or Northern Ireland that we hae been asked to consider are as ollows:
1. Continue with the NIRO as the sole support scheme or renewables indeinitely, een i the
Lngland and \ales RO closes to new applications rom April 201.
2. Continue with the NIRO indeinitely and introduce another support scheme alongside it.
3. Close the NIRO to new applications at the same time as the Lngland and \ales RO, and
introduce another support scheme to replace it.
4. Close the NIRO to new applications as soon as a new support scheme can be introduced to
replace it.

1
DLCC ,2011,: Planning our electric uture: a \hite Paper or secure, aordable and low-carbon electricity`
2
1he RO will close to new accreditations on 31 March 201 and the Goernment will grandather RO support or all
technologies at the rate applicable on 31 March 201.

ii

1hese options inole both ,i, the .trvctvre o the support arrangements in terms o the mechanism
proided - particularly as regards whether the DLCC proposals or GB can be paralleled or
Northern Ireland - and the institutional arrangements needed to support it, and ,ii, the issue o rbo
pa,. or the gien leel o support - whether Northern Ireland electricity consumers alone, or
whether such costs are socialised across the whole o the UK. Alternaties to the DLCC proposals
include the introduction o a PlI1 or a structure o CD lI1 based more on that operating in the
Republic o Ireland rather than on the structure being applied within GB
3
.
Ioweer, een on an initial ealuation, Option ,1, looks unattractie, both rom an administratie
easibility and cost perspectie. Moreoer, Option ,2, also looks unattractie, as it combines the
downsides o running the NIRO indeinitely with the costs o running a new support scheme.
\e would argue that the dierence between options ,3, and ,4, is one o timing. 1he ambition to
create a single LU electricity market
4
is the major drier here. A single market would link UK,
Ireland, lrance electricity markets by the end o 2014. Len i this slips by a year or two, it might
be expected to be irmly in place by 201.
\e would see the issue as being one o whether a new scheme should be introduced to a similar
timetable as proposed by DLCC ,2014, or a less ambitious timetable aligned with the SLM`s market
integration project ,we return to the issue o policy timing below,.
Oer and aboe this, there is the more detailed issue o the speciic eatures and structure o the new
support mechanism and the interaction with the SLM.
L.4 Design features of different support mechanisms
A starting point or understanding the dierences between support mechanism is to understand the
speciic eatures o their design, the risks this gies rise to and how these risks might be mitigated.
1he appropriateness o a particular mechanisms can be drien by both varet reqvirevevt. as well as
the requirements o the t,pe. of tecbvoog, being supported.
1able L1 illustrates some o the speciic support mechanism options, which include some o
DLCC`s most recent proposals, as well as other mechanisms typically obsered, including the
RLlI1
5
, which operates within the Republic o Ireland.

3
Note that the .trvctvre o the support mechanism is a dierent issue to its ere.
4
Linked to the internal electricity market directie 2009,2,LC.
5
Renewable Lnergy leed in 1ari.

iii

1abe 1: Mecbavi.v ae.igv featvre. ava re.vtivg ri..
Design features
and resulting
risks to
generators
Iixed II1
(RLII1)
Classic CfD II1
(GB possibly
for Nuclear
baseload)
Metered CfD II1
(GB for
intermittent
wind)
Premium II1
(GB no longer
recommended)
Price lixed
lixed ,depending
on design
structure,
lixed ,depending
on design
structure,
Variable
Volume llexible lixed olume llexible llexible
Payment Agreed strike price Agreed strike price Agreed strike price
\holesale price
plus subsidy
Market olume
and trading risks
Out o the market In the market In the market In the market
Balancing risk
,post nomination,
None \es \es \es
ovrce: CP.
As well as showing the economic characteristics o each support mechanism ,e.g. lI1 contract price
and olume,, 1able L1 also illustrates the risks arising rom the generators` potential role in the
market` in terms o whether they bear market olume and balancing risks. Both are important, as
they inluence market operation and inestors` perception o market risk.
6

Seeral o these risks already exist under the existing NIRO, which or a price ,a discount to the
reenues that they receie rom suppliers, generators can mitigate through entering into negotiated
power purchase agreements ,PPA, or contractual arrangement with energy suppliers or another
intermediary who might manage these risks or them ,through the pool,.

6
Balancing risk - depending on market structure - can aect the generators` ability to sell power and the price ,reenue,
realised in the market. 1hese risks depend on the structure o balancing and power sales in the wholesale electricity
market structure in question. Balancing risks can be addressed dierently according to support mechanism design
and,or changes to market rules.

i

1he RLlI1 - which rom the point o iew o the generator can be characterised as a lixed lI1 -
allocates market and balancing risks to suppliers, through a standardised PPA, with the supplier
receiing a contracted allowance to manage them.
In this case, suppliers are required to manage

market reenues receied across trading periods and the olatile nature o renewable output. 1he
lixed lI1 proides a ixed predictable reenue stream or renewable generators.
O the new options, it is the PlI1 which most closely resembles the existing NIRO. It could
operate in any uture market structure by being delinked rom the wholesale electricity price.
Inestors are comortable with a ixed reenue stream which they know they can receie alongside
wholesale market reenues ,which are more risky,.
Ioweer, an eectie CD lI1 design structure should ,in theory, be perceied more aourably by
inestors as it proides a more predictable reenue stream or generators. In comparison, under a
PlI1 policy generators bear wholesale electricity price risk, as currently under the NIRO. By
proiding a more predicable reenue stream, a CD lI1 could help to reduce risk and long term
consumer costs through a slightly lower cost o capital.
8

L.S: Implications of different market arrangements for the support mechanism
1he proposals being deeloped by DLCC are to deal with the speciic requirements o the GB
market. 1hus, whilst haing the same support mechanism structure as GB appears attractie, the
practical implications o applying it within the SLM need to be taken into account. In ealuating the
practicability o dierent support mechanism or Northern Ireland it is also essential that both
cvrrevt and fvtvre market arrangements are taken into account.
At the moment, the SLM is quite dierent to the British Llectricity 1rading and 1ransmission
Arrangements ,BL11A, operating in GB, which is a much less liquid, so called net market`, with
most electricity being sold through bi-lateral contracts ,PPAs, between generators and suppliers ,see
1able L2 oerlea,. 1he SLM in contrast is a gross mandatory market with a separate energy
component and a capacity payment, central dispatch and intermittent generation ,such as wind,
treated with priorit, ai.patcb.
9

Under the SLM 1rading and Settlement Code a renewable generator can sell electricity either
directly into the pool or enter into a PPA with a supplier or intermediary to sell the energy on its
behal. \hile a generator ,participating, in the SLM does not ace balancing risks in the same way as
a GB generator, it will still need to manage its output and the requirements rom participating in the
market. Market risks and requirements rom participating in the market may dier by renewable
generation technology.

Suppliers receie a balancing payment - 15 o the large wind category tari - to coer the cost o managing the short
term ariable production o wind energy.
8
See CLPA ,2011,: Note on the impacts o the CD lI1 support package on costs and aailability o capital`
9
See NIAUR & CLR, July 2009, Privcipe. of Di.patcb ava tbe De.igv of tbe Maret cbeave iv tbe 1raaivg c ettevevt Coae -
. Cov.vtatiov Paper

1abe 2: Covpari.ov of M ava 11. rboe.ae varet ae.igv
Issue SLM BL11A
Market structure Gross mandatory pool Bilateral contracts, net-market
Payment structure Separate energy and capacity market Lnergy only market
Dispatch Central ,priority, dispatch Sel-dispatch
Balancing Central Balancing market
ovrce: CP.
1he market dierences between GB and the SLM are more important in deeloping a lixed lI1 or
the CD lI1 approach than in the case o the NIRO or a PlI1, as a lI1 CD approach, in
particular, is much more embedded within the electricity market, rather than being essentially an
additional income stream that is added to that o the wholesale reenues.
1o begin with, this means that the reerence price or a CD lI1 policy will need to be determined
dierently within the SLM to what it is in GB. Currently, the reerence price proposed or GB ,in
the case o intermittent generation, is a day-ahead price sourced rom the best representation o day-
ahead prices at the time the CD lI1 is allocated. 1he aim o the proposed reerence price in GB is
to try and keep generators within the already illiquid GB wholesale market. In contrast, in the
current market a CD lI1 in the SLM would need to reerence ,in some orm, the single System
Marginal Price ,SMP, or the SLM.
In itsel, such a dierent reerence price would not appear to be an insurmountable hurdle. \hereas,
a concern with introducing a CD lI1 within the GB market is the basis risk that this gies rise to,
generators within Northern Ireland might not ace such a basis risk with the SLM. 1he central SMP
and the balancing arrangements or the market, mean that in the SLM, there may not be the
additional uncertainty through the introduction o basis risk which, or example, might be
introduced through a CD lI1 policy based on an aerage` wholesale traded price index as has
been proposed in GB.
10

lrom this perspectie, a CD lI1 policy may actually it better within the SLM than it does in the
BL11A, at least rom those considering inesting in generation in Northern Ireland. Ioweer, at a
more detailed leel there may still be risks that we hae not currently identiied. Indeed, these might
only be identiied when new CD lI1 contracts are deeloped or generators in Northern Ireland,
which cannot be based entirely on those deeloped or GB, due to the dierent market
arrangements that need to be taken account o.
11

12

10
Generators in the SLM would receie wholesale energy payments while an ex-post outturn SMP would proide a clear
reerence or settlement o the CD lI1.
11
lor example, the limited number o market participants who we contacted during the course o our project
highlighted uture issues around compensation or constraints and curtailment o wind generation and the appropriate
contract olume ,should payment be based on metered olume or irm olume ,i.e. aailability,, \e hae not been able
to explore the materiality o these issues urther within the scope o our study.

L.6: Subsidy payment flow
A major component o the reorms in GB is that the subsidy low is in
geverator, as opposed to tbrovgb tbe .vppier
Republic. 1he desired beneits o the proposals
enter the GB market directly, rather than through a
\hile in the context o GB, it remains to be seen w
be achieed in the absence o additional changes ,
clear that this approach will work where
PPA in order to be bankable, the contractual interace o a CD lI1 policy req
consideration in Northern Ireland.
igvre 1: 1 approacbe. - C CfD approa

GB CfD proposal
1here are certain attractions o a structure simila
suppliers to the generators, and remuneration o a
incurred by the generators are ixed rather than ne
renewable generators a ixed payment through PPAs,
enorced contractually, and being applicable to all
we return to below,.
13

12
See LirGrid , SONi All Island 1SO lacilitation o
operational implications associated with high shares o wind power in the All Island
13
1he model is also consistent with how RLlI1 operate
nomination o wholesale power through the SLM.
o the reorms in GB is that the subsidy low is intended to
tbrovgb tbe .vppier as with the RO,NIRO and indeed the RLlI1
he desired beneits o the proposals are to help independent, project-inanced generators
enter the GB market directly, rather than through a PPA with suppliers.
\hile in the context o GB, it remains to be seen whether the desired beneits o the proposals will
be achieed in the absence o additional changes ,een with subsidy lowing to the generator
clear that this approach will work where an independent renewable generation project require
, the contractual interace o a CD lI1 policy req
consideration in Northern Ireland. ligure L1 sets out the options.
C CfD approacb or R1
RLII1

1here are certain attractions o a structure similar to RLlI1 where the subsidy lows through the
suppliers to the generators, and remuneration o arious balancing and market participation costs
incurred by the generators are ixed rather than negotiated. Such an arrangement would proide
renewable generators a ixed payment through PPAs, and has the beneits o simplicity, being
enorced contractually, and being applicable to all sizes o generation including small

See LirGrid , SONi All Island 1SO lacilitation o Renewables Studies` or challenges o managing the
d with high shares o wind power in the All Island power balance.
1he model is also consistent with how RLlI1 operates in the Republic and would be consistent with supp

i
tended to be airect, to tbe
as with the RO,NIRO and indeed the RLlI1 in the
inanced generators
hether the desired beneits o the proposals will
een with subsidy lowing to the generator it is not
an independent renewable generation project requires a
, the contractual interace o a CD lI1 policy requires similar
r to RLlI1 where the subsidy lows through the
rious balancing and market participation costs
Such an arrangement would proide
has the beneits o simplicity, being
sizes o generation including small-scale ,a point

Renewables Studies` or challenges o managing the technical and

s in the Republic and would be consistent with supplier

ii

Under this approach, Northern Ireland`s policy makers might be responsible or setting oerall
policy and administering the scheme ,as currently under the NIRO, while suppliers would be
responsible or managing renewable output in the market and the settlement ,cash-out`, o
wholesale and subsidy payments across trading periods. Institutionally the approach is also simpler
,see Section L8 below, although the scheme would still require administration.
1he alternatie would be to ollow a similar route to that proposed in GB with goernment
responsible or negotiating, managing and monitoring the CD lI1s and the subsidy paid directly to
Northern Ireland`s generators. 1his model would allow renewable generators to sell directly into the
pool and receie subsidy payments directly. 1his could proide more commercial lexibility than
generators haing to contract with registered electricity suppliers in Northern Ireland who then act
as intermediaries or the generators in the market.
14

\hile a renewable generator in GB generally requires a PPA to dispatch, renewable generators in the
SLM ,through priority dispatch rules, hae a guaranteed route to market. 1he issue or Northern
Ireland`s generators ,although compensation or constraints and curtailment may in uture mean
olumes become more o an issue depending on the design o support mechanism, would be the
price realised or their power. A payment route airect, to tbe geverator would thereore seem to address
price risk and allow renewable generators to continue to operate within the market.
Ioweer, compared to a RLlI1 approach, a direct payment route requires a more complicated
contractual structure. 1his might deter inestors. In some ways it is also an aaravtage rather than
disadantage that a subsidy payment low tbrovgb .vppier. means intermediary agents manage
renewable output in the market, rather than eery generator in Northern Ireland haing to interace
with the market and manage these actiities directly. In particular, with substantial numbers o new
,predominantly wind, generators granted access to the SLM, transmission is expected to become
increasingly constrained. 1he resulting compensation or constraints and curtailment may make
market reenues more uncertain and costly or renewable generators. Part o the rationale or
retaining RLlI1`s 15 per cent balancing payment within the SLM has also been to remunerate
suppliers or managing ex-post correction o payments under the R-actor correction process which
can take oer a year.
1he implication is a direct payment route to generators could lead to high leels o administration
and complexity. Generators would need to manage cash payments on metered output across the
year`s trading period rather than receiing a ixed payment stream.
Clearly there are arguments in aour o either approach. Northern Ireland`s policy makers need to
consider the uture role they enisage or renewable generators in the SLM, and how 1rading and
Settlement rules acilitate generators contracting inancially through intermediary structures, rather
than directly interacing with the market.

14
A risk with the RLlI1 approach ,payment to suppliers, is generators ability to achiee a air price or their power is
diminished with only the small number o suppliers to contract with in Northern Ireland.

iii

L.7: Market coupling
1here are urther complications, because o changes to the SLM to meet Luropean electricity
market integration goals, speciically in terms o the need or market coupling and it would seem, a
market design more along the lines o the BL11A than the SLM.
I the expectation is that the GB market and the SLM will be more strongly coupled in uture - in
some ways, orming a single market - haing a broadly similar .trvctvre or a support mechanism
across the whole o the British Isles is likely to make sense. Indeed, the SLM Committee is
currently considering a number o options or integrating the SLM with its neighbouring electricity
markets.
15

lrom market structure perspectie, this might support delaying a moe to a new scheme until uture
market arrangements become clearer. Against this, there is also an argument that with GB adopting
a CD lI1 rom 2014 ,and RLlI1 in place in the Republic o Ireland, inestment could be
undermined in Northern Ireland, where there is delay in implementing a new ,and potentially
improed, support scheme.
1here are pros and cons either way, clearly there may be alue ,rom both an inestor and policy-
maker perspectie, in delaying a inal decision on rbev to vore to a new scheme until the requirement
or SLM redesign ,i any, is clearer ,expected to be early 2012,. In the interim, more detailed analysis
is required to examine how a CD lI1 might be designed or the SLM beore and ater integration
with neighbouring markets.
L.8 Administration arrangements
\hilst as regards markets, there may be arguments on both sides as to whether to pursue the
proposed GB approach, or that o the RLlI1, we beliee that these are less ambiguous when it
comes to the institutional administration requirements o adopting the GB arrangements in
Northern Ireland.
Under the likely GB deliery model, an arm`s length organisation is likely to be responsible or
negotiating, managing and monitoring the CD lI1s. 1he arm`s length body may also be contract
counterparty to CD lI1s. I this were to be applied in Northern Ireland, either a standalone
parallel entity would need to be established or the SLM, or the GB entity would need to be the
counterparty to contracts operating in the SLM.
1he irst would be likely to be costly, whereas the latter would require an understanding o the SLM
as well as BL11A, so could also be administratiely more complex than the operation o the NIRO
and potentially expensie too. In comparison, under a RLlI1 arrangement, the suppliers would be
responsible or managing both wholesale and subsidy lows through the PPA - as in the Republic o
Ireland - which would seem administratiely simpler.

15
Including day ahead coupling using the inancial contracts market and the potential long term redesign o the SLM to
meet Luropean single market requirements.

ix

L.9 Iunding
1he \hite Paper makes no clear decision on how CD lI1 costs will ultimately be unded except
that unding will be rom electricity consumers. One option would be to use the same approach as
the RO, which is unded by imposing an obigatiov on suppliers, who then recoer the cost o this
rom their customers. 1he alternatie is to und the lI1 through an explicit ley on consumer bills.
1he inal decision we expect will depend on the role o the arm`s length organisation,s, unctions
is-a-is suppliers and their contractual positions with generators. \here the arm`s length
organisation is counter-party to the lI1 contracts, a mechanism similar to the Public Serice
Obligation ,PSO, ley is more likely.
In terms o rbo unds the support mechanism, irre.pectire of it. forv, we note that Northern Ireland is
part o a UK wide renewable energy obligation. It may thereore make sense or Northern Ireland
to be part o a UK-wide scheme in terms o unding the subsidy, rather than Northern Ireland
operating its own scheme. Ioweer, it is also not necessarily the case that Northern Ireland`s
participation in a UK-wide scheme would lead to oerall lower Northern Ireland consumer costs ,an
issue we return to below as part o the assessment o consumer costs,.
L.J0 Lstimated costs of a standalone Northern Ireland scheme
In order to support some o the qualitatie conclusions made aboe, we hae sought to model the
potential impacts on costs to consumers in Northern Ireland, under seeral dierent potential
support mechanisms reorm options ,although our model does not hae the degree o unctionality
to dierentiate between the costs o the proposed DLCC approach and a RLlI1 structure,. \e
begin by assuming that the costs o each approach are borne entirely by the Northern Ireland
customer. Note, howeer, that seeral assumptions hae to be made in modelling uture scenarios,
so a degree o caution is required in interpreting the results. Moreoer, the estimated costs do not
take into account the costs o any additional transmission or other inrastructure, which could be
aected by the choice o technology portolio
16
.
Lach reorm is set against a Baseline or the period 2012 to 2020. 1his Baseline assumes the
continuation o the RO as the principal support mechanism or large-scale renewable generation in
the UK and small-scale generation in Northern Ireland. \e then go on to assess the implications ,in
terms o consumer cost and wholesale prices, o the introduction o PlI1 and CD lI1 support
payment structures.
1able L3 shows the considerable increases in Northern Ireland consumer costs rom Northern
Ireland adopting a standalone PlI1 and lI1 CD policy relatie to the Baseline - that is, without
support costs being socialised across the UK. 1he table shows the costs o unding both large and
small-scale renewable generation.

16
1hese would be likely to ary between renewables portolios which diered rom that o the baseline.

x

1abe : Cbavge iv ôrtberv reava cov.vver co.t. vvaer atervatire poicie. iv 2020
Metric Unit NI Baseline NI PII1 NI CfD II1
Co.t. of .vb.ia,
Consumer cost in 2020 _m 0.5 353.6 268.0
Cbavge frov ba.eive _m n,a 283.1 19.5
Consumer cost
,per unit o electricity consumed,
_,M\h 6.9 34.5 26.1
Cbavge frov ba.eive _,M\h n,a 2.6 19.2
Co.t. of .vb.ia, ava tota ^ rboe.ae eectricit,
Consumer cost in 2020 _m 98.6 1,081.8 996.1
Cbavge frov ba.eive _m n,a 283.2 19.5
Consumer cost
_,M\h .9 105.5 9.1
Cbavge frov ba.eive _,M\h n,a 2.6 19.2
ivcrea.e frov ba.eive n,a 35 25
ovrce: CP.
1he analysis suggests ,under the modelled designs o PlI1 and CD lI1 policies, that a standalone
Northern Ireland lI1 scheme would be comparatiely expensie relatie to the NIRO ,in its current
orm,. 1his is because the total cost o unding inestment to meet Northern Ireland`s renewable
electricity target is assumed to be recoered solely rom Northern Ireland consumers. 1his is in
contrast to the NIRO where the design o the RO and Northern Ireland`s current lower obligation
leel mean deployment o renewable generation in Northern Ireland is in part supported by demand
or Northern Ireland ROCs by GB suppliers.
1he modelling also suggests that o the standalone Northern Ireland policies, a PlI1 policy would
be likely to result in the highest costs or Northern Ireland consumers. 1his is because in the
modelling support payment leels are set at closer to projected renewable generation costs across
dierent technology groups in the CD lI1 scenario.
1he modelled impact on Northern Ireland domestic consumer bills under the dierent policy
scenarios is illustrated in 1able L4. 1he analysis also shows projected household bills in GB sourced
rom DLCC`s Llectricity Market Reorm Impact Assessment.

xi

1abe 1: vpact of atervatire .vpport optiov. ov ^ cov.vver bi.
NI Baseline NI PII1 NI CfD II1 GB bills
J7

Dove.tic ;vo ^ efficievc, .arivg.)
2010 _496 _496 _496 _490
2011 - 2015 _508 _509 _509 _490
2015 - 2020 _523 _56 _55 _504
Dove.tic ;ritb 1 ,earov,ear ^ eectricit, cov.vvptiov efficievc, .arivg.)
2010 _496 _496 _496 _490
2011 - 2015 _498 _499 _499 _490
2015 - 2020 _48 _536 _519 _504
ovrce: CP., DCC
1able L3 illustrates that electricity bills are likely to rise relatie to today een under the Baseline.
1his is due to rising wholesale electricity prices and support payments. 1he standalone lI1 policies
,where the total cost o unding inestment to meet Northern Ireland`s renewable electricity target is
assumed to be recoered solely rom Northern Ireland consumers, clearly show the issues Northern
Ireland`s policy makers ace between meeting aspirations or renewable generation in the region and
uture impacts on consumer electricity bill costs.
In the absence o achieing year on year electricity consumption eiciency targets, the modelling
suggests that aerage consumer bills could rise by oer 20 by 2020 in Northern Ireland. 1his
would add oer _100 to the aerage consumer bill.
L.JJ Consumer costs (UK-wide scheme)
\e hae gone on to complete an assessment o the impact on Northern Ireland consumer costs i a
single UK wide low-carbon generation payment cost recoery scheme were adopted - i.e. a scheme
where UK low carbon generation support payments would orm a single pot o money recoered
across annual UK electricity consumption.
In order to assess ivaicatire Northern Ireland consumer costs in the period to 2020 we hae sourced
aerage annual GB net` payments to low carbon generators or 2020 rom DLCC`s Updated Impact
Assessment analysis or the Contracts or Dierence Llectricity Market Reorm policy modelling
scenario ,c. _2.9bn,.
18
ligure L2 illustrates wholesale electricity prices and support payment costs in
Northern Ireland ,per unit o electricity consumed in the region, in 2020 or the Baseline, a
standalone Northern Ireland PlI1, a standalone Northern Ireland CD scheme and a single UK low
carbon generation payment cost recoery scheme. 1he UK lI1 scenario also includes NI CD lI1
payments rom CLPA`s modelling.

1
18
1his includes only projected Contract or Dierence support costs.

xii

0.0
20.0
40.0
60.0
80.0
100.0
120.0
Baseline NI PFIT NI CfD UK FIT
/
M
W
h
CfD support payments
Grandfathered NIRO
Wholesale price
igvre 2: ôrtberv reava rboe.ae price. ava .vpport pa,vevt co.t. vvaer poic, .cevario. iv 2020

ovrce: CP.
ligure L2 illustrates a number o important points. lirst, a Northern Ireland PlI1 is likely to be the
more expensie o the Northern Ireland standalone lI1 policy options. As importantly, een with a
UK wide cost recoery arrangement ,i.e. a socialised cost, Northern Ireland consumers may still be
worse o ,relatie to the Baseline scenario, under a CD lI1 than they are at present in the absence
o an arrangement which might replicate the beneits o the existing lower obligation leel under the
NIRO. Len so, the modelling suggests Northern Ireland consumers might still beneit signiicantly
by participation in a UK wide cost recoery scheme ,the modelling suggests support payment costs
might be reduced by around _14 - _15 ,M\h by 2020, were costs een recoered proportiova,
across annual UK electricity consumption.
\hile the modelling results are inormatie, there are reasons to be cautious about the conclusions.
Under the RO headroom mechanism calculation, suppliers` obligations are set in proportion to UK
regional shares o electricity sales adjusted or the size o the UK obligation and the relatie regional
obligation leels. 1his ineitably leads to demand or Northern Ireland ROCs rom GB suppliers
,our modelling suggests as much as 40-50 per cent in some years, and, thereore, support payments
to Northern Ireland renewables paid by GB consumers. In other words, the consumer cost o
support payments under the RO ,per unit o total electricity consumed in the region, is lower or the
Northern Ireland consumer than or the GB consumer ,as GB suppliers hae a higher obligation
leel relatie to oerall obligation size,.
Unless some new mechanism is put in place to replicate the beneit o this lower obligation leel or
Northern Ireland, remoing the NIRO would increase the proportion o the total cost o UK low-
carbon generation paid by Northern Ireland consumers. In that situation, Northern Ireland
consumers would be worse o vve.. the decrease in total UK consumer costs rom introducing a

xiii

UK CD lI1 ,relatie to continuation o the RO, compensated or this increase. 1he modelling
suggests that it would compensate to some extent but not ully.
ligure L3 shows DLCC`s modelled GB CD support payments in the \hite Paper Impact
Assessment, both relatie to DLCC`s Baseline scenario ,continuation o the RO, and under high
and low ossil uel price projection scenarios ,net CD support payments shown in Part B are used
to derie the indicatie UK lI1 cost or Northern Ireland consumers presented in ligure L2,.
igvre : Co.t. of .vpport for or carbov vecbavi.v. ;Creat ritaiv vvaer MR reforv optiov.)

Part A: Costs o support or support mechanisms Part B: Cost o support or CD lI1 low,high uel prices

ovrce: DCC
GB`s proposed introduction o a CD lI1 may lead to quite signiicant reductions in oerall UK
support payment costs relatie to the RO both beore and ater 2020. 1hereore, een in the
absence o an arrangement similar to the lower obligation under the NIRO, Northern Ireland
consumers may not be signiicantly worse o under the proposed new arrangements. Ioweer, this
depends on a number o key assumptions, particularly uture deelopments in ossil uel prices
,which drie uture wholesale prices, and low carbon generation costs ,which drie CD strike price
leels, amongst many other market uncertainties. In the eent rboe.ae price. in the SLM were also
higher than those projected in the modelling then there would still be the potential or much higher
total costs to consumers in Northern Ireland oer the period to 2020 and beyond, een where
support payments are constrained under the CD design in the UK.
Our modelling o Northern Ireland support payments and prices in the SLM ,together with
DLCC`s modelling o GB support payments and wholesale prices in GB, suggests there are
scenarios where Northern Ireland consumers might not be made comparatiely worse o by the
proposed new arrangements. Ioweer, the assumptions that lead to this conclusion need to be
inestigated urther as this is an area o both uncertainty and importance to the oerall impacts o
any new arrangements on Northern Ireland consumers. \e would recommend, thereore, that
DL1I and NIAUR complete urther analysis as part o any uture design o reorms to the UK`s
low-carbon generation support mechanisms.

xi

0.0
20.0
40.0
60.0
80.0
100.0
120.0
NI CfD (low marine /
offshore wind)
NI CfD (projected
capacity expansion)
/
M
W
h
Support payments (/MWh)
Wholesale price (/MWh)
Marine / offshore consumer premium
L.J2 Paying for the support mechanism
1he preious section suggests Northern Ireland could beneit rom participation in a UK-wide lI1
cost recoery scheme although the impact on oerall consumer costs relatie to the Baseline
scenario is subject to uncertainty. 1hat is, there may be afforaabiit, beneits rom Northern Ireland`s
participation in a UK-wide rather than standalone scheme.
Ioweer, equally there are scenarios where consumer costs might be lower than those projected in
the Baseline modelling, in which a stand-alone scheme may be more attractie. Ioweer, to
signiicant degree, the attractieness o a stand-alone ersus a socialised cost recoery, depends upon
the composition o the renewables generation portolio being supported. By way o example, ligure
L4 compares consumer costs or the NI CD scenario under baseline renewable generation
deployment and a scenario where there is lower marine , oshore wind deployment displaced by
higher onshore wind deployment in the region. As with the preious sections, this excludes any
analysis o associated changes to inrastructure costs.
igvre 1: Cov.vver co.t. vvaer or varive , off.bore riva aepo,vevt ;CfD .cevario)

ligure L4 shows the higher resource, and thereore consumer cost, o a modelling scenario with
more marine , oshore generation deployment ,an oshore consumer premium`,. In comparison,
under a UK-wide cost recoery scheme, the cost to ôrtberv reava cov.vver. o low-carbon
generation lI1 payments is much less sensitie to the generation mix deployed in ôrtberv reava, as
a UK-wide cost recoery scheme ,similar to the NIRO currently, means consumer costs are drien
more by deployment across the UK as a whole.
A UK-wide CD lI1 would spread consumer costs, and might proide Northern Ireland with
greater lexibility to balance enironmental objecties and aordability constraints under dierent
scenarios o renewables deployment in the region. Against this, a stand-alone cost recoery scheme

x

could lower consumer costs where the portolio comprised technologies - such as on-shore wind -
where support costs were lower.
But it is diicult to see how any new arrangement could be as attractie as the existing one. Under
the current RO and NIRO arrangements, Northern Ireland has a relatiely lower obligation than
GB, or reasons including the dierence in electricity prices between the two regions. 1his lower
obligation translates into a relatiely lower cost or Northern Ireland consumers compared to those
in GB. 1his arrangement could not be replicated exactly under a lI1, because there is no equialent
o the obligation leel` in a lI1. Ioweer, other arrangements or distributing the cost are possible,
which might help in retaining some o the beneits to Northern Ireland`s consumers o the RO
arrangements. 1hese would, howeer, need to be designed with reerence to the inal structure and
institutional arrangements or a UK-wide CD lI1.
L.J3 Small-scale generation
lor both PlI1 and CD lI1 policy scenarios, we hae also modelled a Northern Ireland small-scale
generation top-up` lI1 tari oer and aboe the current NIL export tari oered to dierent
classes o small-scale generator. 1he GB small-scale generation lI1 scheme operates according to a
similar ramework whereby small-scale generators receie an export tari ,or electricity exported
o the generation site, plus a ixed top-up` ,generation, payment or total metered generation
rom the small-scale unit.
1able L5 illustrates the cost to Northern Ireland consumers o a small-scale generation lI1 ,note
these costs are captured in the consumer cost analysis in the preious section as the same lI1
structure is assumed in both the PlI1 and CD lI1 scenarios,.
1abe :: Cov.vver co.t. of .va.cae geveratiov
Metric Unit Small-scale II1
Cov.vver .vb.ia, co.t. ;iv 2020)
Consumer cost _m 52.
Consumer cost ,_,M\h renewables generated, _,M\h 125.
Consumer cost ,_,M\h demand, _,M\h 5.1
ovrce: CP.
An obious alternatie to a standalone Northern Ireland small-scale generation lI1 scheme would
be Northern Ireland participation in the current GB small-scale lI1 scheme. 1he clear adantage
with this approach is it would reduce implementation risks and administration costs. 1he main
disadantage is it may lead to reduced Northern Ireland policy discretion oer the rates set or the
region`s small-scale generators.
It should be noted, howeer, that this scheme is currently under reiew, as such, it is probably better
to wait until the results o the reiew are known, beore taking inal decisions.

xi

L.J4 Additional issues / options to consider
DL1I and NIAUR hae also requested that we explore in the modelling an additional scenario
where the NIRO would continue beyond 201. Out o all the policy options this is the most diicult
to assess ,quantitatiely, gien uture consumer costs depend on a number o policy assumptions.
lor example, at what leel would be Northern Ireland`s obligation be set going orward, and how
would the scheme be operated and administered without ongoing GB participation
1o explore this option, we modelled a scenario where the NIRO remains open operating essentially
as a PlI1 scheme with the costs recoered solely rom NI consumers. 1he tari leels oered
under the scheme approximate the banding leels currently oered to NI generators in the NIRO.
\hilst not a direct comparison to the operation o the NIRO, the modelling results do proide an
illustration o what it might cost Northern Ireland consumes to operate a standalone scheme based
on similar tari rates as under the NIRO.
1he analysis suggests that the continuation o the NIRO on a standalone basis - like the other
support payment reorm options - could be a relatiely high Northern Ireland consumer cost
scenario. 1he NIRO in its current orm proides relatiely generous support payment leels to
Northern Ireland generators. As the majority o Northern Ireland renewable deployment
,particularly oshore, is expected to take place after 201, Northern Ireland consumers might be
expected to und the costs o the scheme on a standalone basis were the scheme operated
independently o GB, resulting in high consumer costs.
1he diiculty is the inherent uncertainty o what Northern Ireland consumers might need to und
under a reormed NIRO. \hat the modelling illustrates is that were Northern Ireland to pursue the
option o retaining the NIRO, the existing concessionary obligation leel ,and thereore eicient
Northern Ireland consumer cost o the NIRO, should not be taken as a guide to suggest the option
o retaining the NIRO would necessarily lead to a reduction in projected consumer costs compared,
or example, to alternatie policy options such as a CD lI1. Indeed, there are scenarios where it
could potentially result in higher Northern Ireland consumer costs.
1here is an additional issue with the NIRO, which is that it closes in 2033 rather than 203 as
elsewhere in the UK. Unless there is new legislation to extend this, applications accepted ater April
2013 would receie less than 20 years o support. 1his may well not be material, but needs to be
considered. It may be appropriate to look into extending the lie o the NIRO to 203 to maintain
inestor conidence, irrespectie o what choices are made about new support mechanisms in GB or
Northern Ireland.
L.JS Conclusions and recommendations
Iaing considered careully the qualitatie and quantitatie analysis in this report, o the aailable
options, we beliee that neither a continuation o the NIRO nor the introduction o a PlI1, oer
workable or desirable support mechanisms or Northern Ireland. 1he choice is one o applying
DLCC`s CD lI1 proposals to the SLM, or introducing a RLlI1 type - lixed lI1 - model to
Northern Ireland.

xii

1he main dierences between the two approaches are the lows o subsidy ,in the ormer it is direct
to the generator, rather than through a supplier, and the institutional requirements necessary to
support each.
In the highly illiquid GB BL11A market, the proposed CD lI1 will need to be struck against an
index, giing rise to concerns about basis risks ,i.e. not being able to achiee the index,. 1ogether
with the potential balancing costs or non-dispatchable renewables and the need or physical o-
take to receie payments, this could mean that generators continue to seek to access markets
through a PPA. \hilst not haing been designed or the SLM, the particular eatures o a CD lI1
mean that many o the problems identiied in the GB market are unlikely to bite within the SLM.
1hus, in many ways, it might be expected that the CD lI1 would work better in the context o
Northern Ireland and the SLM than in GB.
Against this, the RLlI1 is a proen approach that carries less risk o applying something that was
designed or another market and which will need to be adapted or a new one. 1he associated costs
o introducing a RLlI1 approach would likely be lower - as there are already existing contracts
rom which to work - than or CD lI1s, or which contracts would need to be speciically written
or the SLM.
A RLlI1 type approach is also less administratiely burdensome and costly rom an institutional
perspectie, in that the public sector is not a contracting counterparty - which is likely to be the case
or the CD lI1s. Ioweer, it requires generators to contract directly with suppliers rather than
operating directly within the market.
1here are pros and cons or both options, and they need to be examined in much greater detail
beore a inal iew is arried at.
As Northern Ireland is part o a UK-wide renewable energy obligation we recommend that DL1I
seeks to socialise the costs o the support regime across the UK, unless there is a desire to ocus on
lower cost renewables technologies, which might reduce the costs to Northern Ireland customers
under a stand-alone scheme.
A UK-wide approach would spread consumer costs, and may proide Northern Ireland with greater
lexibility to balance enironmental objecties and the region`s aordability constraints under
dierent scenarios o renewables deployment.
\e recommend DL1I and NIAUR make a clear statement to the Northern Ireland market o the
uture direction electricity policy in the region so as to aoid the risk o an inestment hiatus during
a key inestment period, potentially according to the ollowing:
x Like the RO in GB, the NIRO will remain open to new accreditations until 201 ater which
the scheme will be grandathered.
x In terms o timing or joining a new scheme, Northern Ireland would plan or a similar
implementation date as the rest o the UK ,i.e. 2014,.

xiii

x Ioweer, the inal implementation date will need to be conirmed ollowing a clearer
decision on the adopted option ,s, or market integration in early 2012.
In early 2012, were signiicant reisions required or evri.agea to it with uture Luropean market
requirements, the timeline or implementing a new policy would need to be reisited. In the interim,
there is detailed analysis to examine the impacts on consumer costs urther, and how a lI1
approach might be designed or the SLM beore and ater integration with neighbouring markets.
1o support the transition to the new arrangements, the option o extending the lie o the NIRO to
203 should also be considered.

1

J. IN1RODUC1ION
CLPA has been commissioned by DL1I and NIAUR to look at the implications or Northern
Ireland ,NI,o DLCC`s Llectricity Market Reorm ,LMR, package, ocusing on the implications o
implementing a Contract or Dierence leed in 1ari ,CD lI1, or Premium leed-in 1ari ,PlI1,
in the Single Llectricity Market ,SLM,.
J.J. Context
1he United Kingdom ,UK, has signed up to binding targets or the proportion o energy sourced
rom renewable generation technologies and is committed to reducing carbon emissions. NI has the
potential to make a signiicant contribution to the achieement o the UK`s renewable electricity
targets and indeed, has set an ambitious target to source 40 per cent o its electricity consumption
rom renewable sources by the year 2020.
1he primary tool or supporting the deelopment o renewable electricity generation in NI is
currently the Northern Ireland Renewables Obligation ,NIRO,. 1he NIRO operates in parallel with
the Renewables Obligation ,RO, in Lngland and \ales and the Renewables Obligation Scotland
,ROS, to incentiise the deelopment o renewable electricity generation in the UK. 1he RO places
a legal requirement on electricity suppliers to source a set and increasing proportion o their energy
requirements rom renewable sources in the orm o Renewable Obligation Certiicates ,ROCs, or
to pay a buy-out ee or any shortall.
J.2. 1erms of reference
1he Department o Lnergy and Climate Change ,DLCC, issued a consultation on the LMR package
in December 2010. 1he two primary policy objecties o the LMR are to reorm the electricity
market arrangements so that the Goernment`s decarbonisation objecties can be met cost
eectiely and to ensure security o supply or Great Britain`s ,GB, electricity system towards the
end o the decade and beyond. Among the policy options considered by DLCC in the LMR
consultation paper and associated impact assessment were eed-in tari ,lI1, support mechanisms
or renewable generation, including a CD lI1 and PlI1.
In July the Goernment published its Lnergy \hite Paper.
19

1he \hite Paper sets out a package o reorms to GB electricity policy, including a new system o
long-term contacts in the orm o CD lI1s which will replace the RO.
20
1he \hite paper also
highlights that it is likely an organisation ,or organisations, at arm`s length rom Goernment will
administer the CD lI1s or GB.

19
20
1he RO will close to new accreditations on 31 March 201 and the Goernment will grandather RO support or all
technologies at the rate applicable on 31 March 201.

2

CLPA has been commissioned by DL1I and NIAUR to look at the implications o DLCC`s LMR
package or NI, ocusing on the implications o implementing a CD lI1 or PlI1 in the SLM. NI
has a series o unique characteristics which mean that it may be inappropriate to simply duplicate the
support mechanism adopted in GB under the LMR package. lor example, the nature o land use
means that the potential or the deelopment o certain generation technologies is greater than may
be the case in GB, and the leel o deelopment o networks mean that renewable generation creates
dierent challenges.
Lqually the single market structure ,and uture changes to the market structure resulting rom policy
at a Luropean leel or the close integration o electricity markets, means the interaction between
the support mechanism and market operation needs careul consideration, and economic actors,
such as the relatiely higher leels o uel poerty in NI than GB, mean that costs to customers are
also an important consideration.
Our study terms o reerence require us to:
x identiy and assess the main implementation risks or NI o the introduction o a CD lI1
as the region`s renewable generation support mechanism,
x assess or NI the impact and associated costs o introducing a CD based eed-in tari ,lI1
with CD, or a Premium lI1,
x identiy and assess options or recoering the costs o an NI lI1 with CD or Premium lI1
either as a standalone scheme or as a socialised cost across the UK,
x assess the implications or the NIRO rom 201 when the RO stops taking new
accreditations,
x assess the potential or the NIRO to remain open to accreditations ater the RO ceases
accreditations in 201,
x assess the implications o NI operating up to three schemes: a small-scale lI1, a lI1 with
CD and the NIRO, and
x proide a high leel assessment o the proposal to introduce a carbon price loor and the
impact on power generation in NI as part o the SLM.
\e address the terms o reerence in two parts.
In Part A o our report, we proide a more qualitatie assessment o the implications and risks o
dierent LMR options or NI. \e begin by outlining the context or renewable electricity in,
respectiely, NI, GB and the Republic o Ireland ,RoI, and the economic characteristics o
alternatie design eatures or generation support mechanisms.
\e then go on to consider how an alternatie support mechanism to the NIRO might be structured
in the SLM ,which operates as a gross mandatory pool,. \e conclude Part A with a series o issues
or more detailed quantitatie analysis.

3

In Part B we apply a modelling ramework to assess the costs and beneits o alternatie generation
support mechanism policy options. 1his is largely based on a generation inestment and CBA
modelling ramework deeloped by CLPA or NI.
1he modelling is necessarily dependent on a series o assumptions, including those on capacity build
rates, generation costs, uel prices and subsidy mechanism design, which drie the outputs the model
produces. \hile those assumptions are reasonable, the uture may turn out dierently. In addition,
models can neer capture all the actors that aect decisions and deelopments in the real world. As
such, whilst the modelling is aluable in enabling a number o hypotheses to be tested, its results
must be treated as one actor, alongside a much wider qualitatie assessment, in coming to
conclusions.
J.3. Structure of report
1he rest o this document is structured as ollows:
x Section 2 considers the context and key issues or the study,
x Section 3 sets out the options considered in the study,
x Section 4 proides our qualitatie ealuation o the options,
x Section 5 sets out our modelling ramework and assumptions,
x Section 6 presents the results rom the Baseline scenario,
x Section explores issues around Carbon Price Support in the SLM,
x Section 8 analyses alternatie lI1 options,
x Section 9 considers small-scale generation issues,
x Section 10 presents modelling results rom sensitiity analysis,
x Section 11 proides conclusions and recommendations rom the study.
1he main report is accompanied by a series o appendices:
x Appendix A proides inormation on the operation o RLlI1,
x Appendix B proides inormation on existing Premium lI1 schemes in other markets,
x Appendix C sets out the report wholesale market modelling assumptions,
x Appendix D sets out the report generation capacity expansion plan assumptions,
x Appendix L sets out the report generation cost assumptions.

4

PAR1 A: CON1LX1, ISSULS AND OP1IONS
2. CON1LX1 AND ISSULS
\e set out below the current context or renewable electricity in, respectiely, NI, Great Britain
,GB, and the Republic o Ireland ,RoI,. \e also consider the economic characteristics o
alternatie support mechanism structures. Because o its importance, we also include a separate
section on the operation o the Single Llectricity Market ,SLM, and the impacts o uture market
coupling under LU Directies.
2.J. 1he Northern Ireland context
1he oerall context is set out in the Strategic Lnergy lramework ,SLl,
21
. 1his lists the our key
energy goals:
x building competitie markets,
x ensuring security o supply,
x enhancing sustainability, and
x deeloping our energy inrastructure.
1he SLl makes it clear that Northern Ireland`s policy covtivve. to be arirev b, vrope`. \e
thereore turn to considering the Luropean context.
2.J.J. 1he Luropean context
1he LU ision is or a single Luropean energy market, with connected, low-carbon electricity
networks deliering secure and cost-eectie supplies to consumers. LU Member States hae a
target to reduce greenhouse gases by 20 by 2020, and to source 20 o energy consumption
rom renewables by the same date. \ithin that oerall 20 target, the UK has a target o 15 o
energy rom renewables by 2020.
1he LU has also taken steps to improe the unctioning o Luropean energy markets, including
the competitieness and degree o interconnection o those markets. 1his was set out in the
Directie
22
implementing the electricity aspects o the so-called 3
rd
package`. As well as
proisions on the need or unbundling o dierent aspects o the electricity supply chain ,e.g.
transmission and generation,, the package aimed to increase cross-border access or electricity
suppliers and producers. 1his o course already occurs in the SLM.

21
DL1I, 2010, verg,: . .trategic fraveror for ôrtberv reava,
http:,,www.detini.go.uk,strategic_energy_ramework__se_2010_.pd
22
OJ LU, Directire 200,2,C, covcervivg covvov rve. for tbe ivterva varet iv eectricit,

5

2.J.2. Northern Ireland's potential for renewable electricity
A preious study
23
or DL1I estimated that Northern Ireland`s practical renewable electricity
resource was around 1,00 M\ ,see table below,. 1his is comparable to Northern Ireland`s
current peak demand o around 1,900
24
M\. 1he ast majority ,1,500 M\, o the resource is
onshore wind
25
.
1abe 2.1: Reverabe eectricit, potevtia iv ôrtberv reava iv 2020, b, tecbvoog, ;figvre. iv M!)
1echnology 1heoretical maximum Practical resource
Onshore wind 3,203 1,500
Oshore wind 500 100
1idal 300 50-
\ae 0 0
Biomass 13 13
Lnergy rom waste, including landill gas 58 58
Iydro 11- 11
1O1AL 4,09S+ J,732+
ovrce: Ore .rvp
LirGrid , SONi`s Generation Capacity Statement proides a more recent assessment o the
contribution dierent renewable generation technology groups might make to NI`s renewable
electricity target by 2020. 1able 2.2 shows the renewable generation capacity mix assumed in the
LirGrid , SONi analysis.
1abe 2.2: ôrtberv reava capacit, epav.iov pav
1echnology Capacity (MW) in 2020
Onshore wind 1030
Oshore wind 600
Small-scale hydro 3
Solid biouels 200
Landill gas 2
CIP 50
1idal , \ae 300
1O1AL 22J0
ovrce: irCria , Oî
1able 2.2 shows a ar greater contribution rom emerging renewable generation technologies
such as tidal and oshore wind compared to the Oe Arup study.

23
Oe Arup, 2009, .tabi.bvevt of ôrtberv reava Reverabe ectricit, 1arget. to 2020
24
Oe Arup, 2009, based on SONI igure or 200,08 ,temperature adjusted,
25
Details o currently installed wind can be ound at http:,,maps.seai.ie,wind,

6

Northern Ireland`s electricity generation is currently based on gas ,two o the region`s three
power stations are gas-ired,. 1he SLl takes the iew that ...it i. ie, tbat ôrtberv reava ri
revaiv arge, aepevaevt ov ga. firea pavt for ba.eoaa geveratiov vvti at ea.t 200`.
2.J.3. Infrastructure
Achieing the ision o a single Luropean market will require increased physical connections
between the networks in NI, the RoI and GB. 1hese will also be useul or managing the impact
o renewables, particularly wind. Key deelopments here include the second North-South
interconnector, which under current proposals will include ...tro ver 1001 1:00M1. .ivge
circvit ive.,`
26
. An Last-\est interconnector, with a 500 M\ capacity
2
, is also under
construction between the RoI and \ales, it is due to be completed by 2012. 1his will increase
integration with the markets in those countries. \e discuss the increased integration o the
markets, and the implications or NI, in more detail in section 2.6.2.
1here will also need to be inestment in other aspects o the electricity grid. NIL estimates that
around _1 billion o grid inestment is likely to be required to support a target o 40 renewable
electricity. 1his inrastructure will hae a noticeable cost. 1he SLl notes that while the alue
depends on a number o actors, ...tbe covbivea co.t of reverabe eectricit, iv.taatiov., togetber ritb tbe
co.t of tbe gria ivre.tvevt vece..ar, to veet tbe 10 target, cova be betreev _1 ava _ per bov.eboa ov av
avvva ba.i. at cvrrevt price.`.
2.J.4. Current policy
Renewable electricity in NI is currently incentiised through the Renewables Obligation ,RO,.
1his is a UK-wide renewable generation support scheme, whereby renewable generators receie
Renewable Obligation Certiicates ,ROCs, or each M\h o electricity they generate.
1hese are then sold to UK electricity suppliers, who are obliged to purchase a speciied leel o
Renewable Obligation Certiicates ,ROCs,, or pay a buyout price. 1he number o ROCs they
need to present is calculated based on each supplier`s electricity sales in a gien obligation period.
1he oerall ROC requirement across all suppliers is set at a leel to ensure that the size o the
obligation will always be at least 10 aboe expected generation in any gien obligation period.
1his is known as headroom`. \hile ROCs issued to NI generators are known as NIROCs`,
they hae exactly the same alue as ROCs issued to generators in GB, and can be used by GB
suppliers in the same way as ROCs rom GB. 1he incentie to inest in renewable generation is
created through the ROC price realised by the renewable generator.

26
PB Power, 2009, Carav1,rove ava MeatbCarav 100r trav.vi..iov circvit.: covpari.ov of bigb rotage trav.vi..iov optiov.,
http:,,www.eirgrid.com,media,09.02.12.Comparison20o20high20oltage20transmission20options.pd
2
Source: Lirgrid,
http:,,www.eirgridprojects.com,media,Press20release20or20L\IC20Planning20Approal2015200
9202009.pd

One important dierence between the RO in NI and the rest o the UK is that the NIRO ends
in April 2033, rather than in April 203 as elsewhere. RO support is assumed to be or 20 years,
so applications made ater April 2013 will receie less support than earlier ones. 1he argument
might be that RO support leels are set on the basis o them being paid or 20 years, and so this
does not proide suicient support to renewables projects post-April 2013.
1he main counter-argument is that the dierence in practice between support oer, say, 18 years
and support oer 20 years is relatiely small. At a 10 discount rate, and assuming lat support
oer the period, the dierence in the total lietime support is around 3.5
28
. Len or projects
commissioned in April 201, which would only receie 16 years o support, the dierence is
around 8. In addition, because o the alling technology costs o wind, there may be a point at
which onshore wind in particular is economic without subsidy.
1hat said, inestor conidence may be dented i they can see support leels alling away. It may
thereore be appropriate to look into extending the lie o the NIRO to 203, in line with the
rest o the RO. \e return to this issue in Part B.
2.2. 1he Great Britain context
DLCC issued its consultation on the LMR package in December 2010. 1he LMR contains our
main energy policies or the GB electricity market:
x eeav1ariff.: these would be long term contracts that would increase clarity o long term
reenue or all orms o low carbon generation.
x Carbov price .vpport: this would proide a higher tax on ossil uels used or electricity
generation to strengthen carbon price signals to inestors.
x Capacit, pa,vevt.: these are retainer payments or lexible plant and demand side resources
that will proide a guaranteed leel o spare generating capacity.
x vi..iov. Perforvavce tavaara: this would proide a back-stop regulation that will ensure
no new inestment in unabated coal plant.
1his study ocuses on the irst o these, although we briely consider carbon price support in
Part B o our report. lrom a DL1I,NIAUR perspectie, the issue is whether a system designed
or the ery dierent BL11A market ,and seemingly with a desire or new nuclear power, is
suitable or the NI electricity system.
Since the publication o the LMR consultation, the Goernment`s \hite Paper has set out a
inalised strategy and decision on these energy policies. At the heart o the Goernment`s
strategy is a system o long-term contracts in the orm o CD lI1s.
1he key aspects o the CD lI1 designs proposed by DLCC are summarised in 1able 2.3 below
,see Annex B o the \hite Paper or urther details,.

28
In present alue terms.

8

1abe 2.: orcarbov 1 ritb Covtract. for Differevce propo.a.
Intermittent Baseload
Contract orm 1wo-way lI1 CD 1wo-way lI1 CD
Strike price Annual inlation index Annual inlation index
Market reerence price Day-ahead price
Choice o baseload or hourly price
Not aeraged oer a longer period
\ear-ahead baseload price
Choice o price sources
Contract olume Metered output 1o be conirmed, metered output or
irm olume
ovrce: DCC
In addition the Goernment has also decided to introduce a Carbon Price lloor and an
Lmissions Perormance Standard ,set as an annual limit equialent to 450g C0
2
, k\h at
baseload, to proide a clear regulatory signal on the amount o carbon new ossil-uel power
stations can emit.
1he \hite paper also highlights the Goernment`s intention to put in place an organisation ,or
organisations, at arm`s length rom Goernment to administer the CD lI1s. As illustrated by
ligure 2.1, under the GB deliery model, the arm`s length organisation is likely to be responsible
or negotiating, managing and monitoring the CD lI1s.
igvre 2.1: DCC ivaicatire aeirer, voae for eeaiv 1ariff ritb Covtract. for Differevce

Renewable electricity generation under 5M\ in GB
29
is already eligible or subsidy rom leed-In
1aris in GB. 1he subsidy is a lat rate per unit o electricity generated, with the rate per unit
arying depending on the technology and the size o the installation. Rates range rom 4.p,
k\h or wind oer 1.5M\ capacity to 43.3p,k\h or PV under 4k\ capacity which is
retroitted to an existing building
30
. 1he scheme is administered by Ogem although DLCC is
responsible or setting policy objecties and approach.

29
In Northern Ireland, support is proided through the Renewables Obligation.
30
Source: Ogem.
http:,,www.ogem.go.uk,sustainability,enironment,its,Documents1,leed-
in201ari201able20120August202011.pd

9

2.3. 1he Republic of Ireland context
1he RoI has a target o 40 o electricity rom renewables by 2020
31
. 1his is within its oerall
renewable energy target or 2020 o 16
32
.
1he Republic`s energy strategy was set out in its 200 \hite Paper
33
. 1he oerall goals are similar
to those o GB and NI: energy security, sustainability and competitieness ,including cost to
consumers,. In terms o electricity generation, signiicant growth is expected in renewables and
gas to 2020, while the quantity o electricity generated rom oil is expected to reduce.
igvre 2.1: ectricit, geveratiov iv tbe Repvbic of reava 102020

ovrce: DC^R
1he Republic`s main policy or deliering its renewable electricity target is the Renewable Lnergy
leed in 1ari ,RLlI1,
34
. 1his is described in section 2.5, and in more detail in Appendix A. As
context to the discussion o RLlI1, we irst consider the economic characteristics and context
o alternatie support mechanism designs, including leed-in 1ari ,lI1, structures that were
proposed by DLCC as part o the LMR proposals.

31
Source: DCLNR
32
OJ LU, Directie 2009,28,LC, ov tbe provotiov of tbe v.e of everg, frov reverabe .ovrce.
http:,,eur-lex.europa.eu,LexUriSer,LexUriSer.douri~OJ:L:2009:140:0016:0062:en:PDl
33
DCLNR, 200, Deirerivg . v.taivabe verg, vtvre or reava: 1be verg, Poic, raveror 2002020,
http:,,www.dcenr.go.ie,NR,rdonlyres,54C8A1L-4L96-4L28-AA-
3226220Dl2lC,2356,Lnergy\hitePaper12March200.pd
34

http:,,www.dcenr.go.ie,Lnergy,Sustainable-and-Renewable-Lnergy-Diision,Llectricity-rom-Renewables
-inc-RLlI1-and-ALR.htm

10

2.4. Subsidy design context
Support mechanisms are eectiely subsidy arrangements. Subsidies are payments which alter
costs, prices and ,or returns rom what they otherwise would be, i determined by an actual, or
hypothetical, market benchmark.
1here are two main reasons or such support mechanisms:
x irst, to increase consumption o something that would otherwise be unaordable, and
x second, to alter or distort consumption patterns rom what they otherwise might be, or
encourage the supply and consumption o one type o good or serice oer another
substitute.
Policy may desire such increases or changes to consumption in order to meet a number o
objecties. 1hese can include employment creation, protection o economic sectors, better
health outcomes and o course, enironmental beneits such as reduced carbon emissions.
2.4.J. Who pays and who benefits?
In analysing subsidy and support arrangements a starting point is always to determine who pays
and who beneits rom the subsidy. 1his analysis is not the same as who receies the subsidy
,discussed below, and is not always straightorward. \ho beneits is not always determined by
who receies the subsidy.
\ho pays, howeer, is relatiely clear. I goernment pays, it is today`s tax-payers or tomorrow`s
,when subsidies are inanced through borrowing,. In the case o cross subsidies, the entity that
pays is the entity which either pays more or a serice than a pure market based solution would
proide or, i selling ,producer to consumer subsidies, receies less than the market rate.
In terms o who beneits rom the subsidy:
x A cov.vver beneits rom a subsidy, where as a result o the subsidy, a gien leel o
product or serice is proided at a price below what would be determined by a
hypothetical market.
x A proavcer beneits where a higher inancial return is receied through the proision o a
gien product or serice than would be the case in the market.
In general, when goernment pays a subsidy consumers and producers can both beneit,
depending upon the precise incidence o the subsidy. In comparison, cross subsidies are a zero
sum game in which transers between dierent market participants take place.
In the speciic context o supporting low-carbon electricity generation, there is a dierentiation
to be made between:
x who receies the low-carbon electricity subsidy, and
x who beneits rom that subsidy.
1he RO is currently unded by UK electricity consumers with objectie to address dierences in
renewable generation technology cost and risk relatie to conentional ,carbon emitting,
electricity generation. Ioweer, because o the way the scheme is structured ,i.e. ROCs are sold

11

by generators to UK electricity suppliers, who are obliged to purchase a speciied leel o ROCs
or pay the scheme`s buyout price, it is not always straight orward to identiy who in practice
beneits rom the subsidy, electricity suppliers or generators.
lor example, in the GB market, Power Purchase Agreements ,PPAs, or renewable generation
oten contain a discount on the reenue stream to relect the risks being taken by ,and perhaps
bargaining power o, the otaker, in most cases one o the large domestic GB suppliers. lor
example, renewable wind generators typically pay the suppliers who purchase their electricity
output a ee or balancing serices.
1he leel o discount in PPAs can be applied to the power` or green` ,ROC, element o the
contract. \hile it is unclear to us whether similar discount arrangements take place in the NI
market context, the implication ,at least in a GB context, is that while generators receie the
subsidy ,in the orm o ROCs, it is the generator ava the supplier who may beneit rom the
support mechanism depending on the risks being taken by dierent parties, and the underlying
costs o alternatie generation inestment opportunities.
2.4.2. Structure of support mechanism instrument design
1his section proides an oeriew o the key design components o renewable generation
support mechanisms, with a particular ocus on lI1 structures. It is important to recognise that
there are many interdependencies between these components which must be taken into account
in the oerall instrument design.
Renewable electricity support schemes can in general be categorized along two main approaches:
price and quantity based schemes. lI1 contracts are examples o price based support schemes.
Some o the characteristics that need to be deined to structure a lI1 contract include:
x Covtract forv: is the lI1 contract a premium paid on top o the wholesale price or some
orm o Contract or Dierence ,CD, settled against a market reerence price
35

x Covtract price: or example, is the contract price ixed or ariable and what is the process
or determining the contract strike` price
x Covtract rovve: is the contract settled against metered output rom the generation plant or
is a ixed olume,number o M\s contracted
x vb.ia, price: premium lI1s structures oten ix the subsidy price
36
while CD design
structures oten ix the covtract price but allow the .vb.ia, price to ary.
x vb.ia, rovve: similar to contract olume, is the contract settled against metered output
or a ixed number o M\s
x aavcivg ri.: or some technologies it is the case that reenue realisation is subject to
Basis Risk.
3
Balancing risk can be addressed dierently according to contract orms.

35
CDs can be one-way` ,in which case the lI1 contract only requires payment in one direction, or two-way` ,in
which case the generator receies the dierence between the market reerence price in the contract and the contract
strike price when the reerence price is below the contract strike price, and when the market reerence price is aboe
the strike price, the generator pays back the dierence,.
36
lor example, the buy-out price under the RO.

12

As discussed in the preious section, there is also the issue o who receies the subsidy and who
beneits rom it to consider and duration o the contract. 1here are also interdependencies
between contract design and the wholesale electricity market structure. Dierent lI1 designs can
aim to keep renewable generation either:
x in-the-market ,i.e. the generator`s power continues to be sold into the market,, or
x out-o-the-market.
1able 2.4 ,oerlea, compares dierent lI1 structures according to these design components,
including the existing support mechanism in NI ,the NIRO,, CD lI1 designs and Premium
lI1 ,PlI1, approaches. 1he analysis is not deinitie and is intended to proide a laour o the
dierent options and sub-options that exist or lI1 structures.

3
1his is the risk that the price the renewable generation is sold at is lower than price on which the contract strike`
price is based ,the Market Reerence Price,.

1
3

1
a
b
e

2
.
1
:

1

c
o
v
t
r
a
c
t

a
e
.
i
g
v

c
o
v
p
o
v
e
v
t
.

C
o
m
p
o
n
e
n
t
s

I
i
x
e
d

I
I
1

C
l
a
s
s
i
c

C
f
D

M
e
t
e
r
e
d

C
f
D

P
I
I
1

P
I
I
1

(
c
a
p

&

c
o
l
l
a
r
)

R
O

l
o
r
m

M
a
r
k
e
t

i
n
d
e
p
e
n
d
e
n
t

G
e
n
e
r
a
t
o
r

m
u
s
t

d
i
s
p
a
t
c
h

a

i
x
e
d

o
l
u
m
e
.

S
e
t
t
l
e
d

a
g
a
i
n
s
t

a

w
h
o
l
e
s
a
l
e

m
a
r
k
e
t

r
e
e
r
e
n
c
e

p
r
i
c
e

A

p
r
e
m
i
u
m

p
a
y
m
e
n
t

o
n

t
o
p

o

w
h
o
l
e
s
a
l
e

p
r
i
c
e

P
l
I
1

b
u
t

c
a
p

a
n
d

c
o
l
l
a
r

i

w
h
o
l
e
s
a
l
e

p
r
i
c
e
s

c
h
a
n
g
e

Q
u
a
s
i

p
r
i
c
e
,
o
l
u
m
e

b
a
s
e
d

s
c
h
e
m
e

C
o
n
t
r
a
c
t

p
r
i
c
e

l
i
x
e
d

l
i
x
e
d

l
i
x
e
d

V
a
r
i
a
b
l
e

V
a
r
i
a
b
l
e

V
a
r
i
a
b
l
e

C
o
n
t
r
a
c
t

o
l
u
m
e

V
a
r
i
a
b
l
e

l
i
x
e
d

V
a
r
i
a
b
l
e

-

n
o
m
i
n
a
t
e
d

c
l
o
s
e

t
o

d
i
s
p
a
t
c
h

V
a
r
i
a
b
l
e

-

n
o
m
i
n
a
t
e
d

c
l
o
s
e

t
o

d
i
s
p
a
t
c
h

V
a
r
i
a
b
l
e

-

n
o
m
i
n
a
t
e
d

c
l
o
s
e

t
o

d
i
s
p
a
t
c
h

V
a
r
i
a
b
l
e

-

n
o
m
i
n
a
t
e
d

c
l
o
s
e

t
o

d
i
s
p
a
t
c
h

S
u
b
s
i
d
y

p
r
i
c
e

1

V
a
r
i
a
b
l
e

2

V
a
r
i
a
b
l
e

2

V
a
r
i
a
b
l
e

2

l
i
x
e
d

Q
u
a
s
i

l
i
x
e
d

3

Q
u
a
s
i

l
i
x
e
d

4

S
u
b
s
i
d
y

o
l
u
m
e

V
a
r
i
a
b
l
e

l
i
x
e
d

V
a
r
i
a
b
l
e

V
a
r
i
a
b
l
e

V
a
r
i
a
b
l
e

V
a
r
i
a
b
l
e

B
a
l
a
n
c
i
n
g

r
i
s
k

N
o

\
e
s

-

b
u
t

d
e
p
e
n
d
s

o
n

m
a
r
k
e
t

r
u
l
e
s
.

\
e
s

-

b
u
t

d
e
p
e
n
d
s

o
n

m
a
r
k
e
t

r
u
l
e
s
.

\
e
s

-

b
u
t

d
e
p
e
n
d
s

o
n

m
a
r
k
e
t

r
u
l
e
s
.

\
e
s

-

b
u
t

d
e
p
e
n
d
s

o
n

m
a
r
k
e
t

r
u
l
e
s
.

\
e
s

-

b
u
t

d
e
p
e
n
d
s

o
n

m
a
r
k
e
t

r
u
l
e
s
.

R
o
l
e

i
n

t
h
e

m
a
r
k
e
t

O
u
t

o

t
h
e

m
a
r
k
e
t

I
n

t
h
e

m
a
r
k
e
t

I
n

t
h
e

m
a
r
k
e
t

I
n

t
h
e

m
a
r
k
e
t

I
n

t
h
e

m
a
r
k
e
t

I
n

t
h
e

m
a
r
k
e
t

o
v
r
c
e
:

C
P
.

^
o
t
e

1
:

v
b
.
i
a
,

v
e
t

o
f

t
b
e

r
b
o
e
.
a
e

e
e
c
t
r
i
c
i
t
,

p
r
i
c
e

^
o
t
e

2
:

v
b
.
i
a
,

;
v
e
t

o
f

r
b
o
e
.
a
e

p
r
i
c
e
)

r
a
r
i
e
.

a
.

r
b
o
e
.
a
e

p
r
i
c
e

c
b
a
v
g
e
.

^
o
t
e

:

a

c
a
p

a
v
a

c
o
a
r

P
1

a
o
r
.

.
o
v
e

r
a
r
i
a
t
i
o
v

i
v

1

t
o
p
v
p

p
a
,
v
e
v
t

b
a
.
e
a

o
v

v
a
r
e
t

p
r
i
c
e

^
o
t
e

1
:

b
e
c
a
v
.
e

t
b
e

b
v
,
o
v
t

p
r
i
c
e

i
.

f
i
e
a

14

2.S. Operation of the RLII1
Iaing considered the design components o lI1 contract structures, in this section we reiew the
operation o RLlI1 in the RoI. 1he operation o RLlI1 is particularly releant to NI gien it
operates through the SLM mechanism. \e note that while RLlI1 operates like a CD it is closer in
design to a lixed lI1 than the CD lI1 structures currently being considered by DLCC as part o
the LMR proposals ,see Appendix A or urther details,.
1he RLlI1 was launched in 2005 to replace the Alternatie Lnergy Requirement ,ALR,
mechanism. In summary, the scheme gies a ixed price or renewable electricity to inestors. It does
this by proiding .vppier. with a guaranteed reerence price or renewable electricity through what is,
eectiely, a one-way CD plus a small ixed premium payment. 1he supplier then enters into a PPA
with the generator, typically or 15 years, which speciies the amount that the supplier will pay the
generator or each unit o electricity produced.
Payments under RLlI1 hae three parts:
1. 1he irst part is a small ixed payment, equal to 15 o the oerall RLlI1 reerence price
or onshore wind.
2. 1he second is a market equalisation` payment, which coers any dierence between the
price that the supplier can sell the renewable electricity or, and the RLlI1 reerence price.
1his is a one-way contract, so i the market price is higher than the RLlI1 reerence price,
the supplier simply receies no market equalisation` payment, rather than haing to repay
anything.
3. 1he third is a technology` payment, paid or technologies that are more expensie than
onshore wind. 1his coers the additional cost to the supplier o buying electricity rom those
technologies, and so is set by reerence to the price in the PPA between the supplier and the
generator.
\e consider the option o a lI1 system similar to RLlI1 in general in Section 3 below.
2.6. Operation of the Single Llectricity Market (SLM)
2.6.J. Current market design
lor our purposes, the most important aspect o the SLM is that it operates as a gross mandatory
pool. All generators are required to bid into the SLM pool at their short-run marginal cost ,SRMC,,
and all power
38
must be bought and sold through it. All generators then receie the marginal SLM
price
39
. Some generators, including wind, are so-called price takers`
40
and are treated as i their bid

38
Aboe a de minimis leel o 10 M\
39
Rather than their bid, as in the GB market.
40
\e assume or the purposes o this report that wind power continues to be a price taker. As outlined in our preious
report, howeer, we understand that the RAs are considering options which would result in what are currently classiied

15

into the SLM is zero. Like other generators in the SLM, wind also receies a capacity payment or
being aailable to generate. \e assume that this would not change with any change in the renewable
support mechanism.
In some cases, a wind generator may produce electricity but not be dispatched ,i.e. the electricity is
not used,. 1his is known as being constrained o` ,i due to transmission limits, or being
curtailed` ,i there is simply too much wind generation or demand,. \hen this happens, wind
receies compensation.
1he wholesale market arrangements in the SLM are dierent to those in BL11A. 1his is important
to consider when assessing alternatie options or the renewable generation support mechanism in
NI and GB. 1able 2.5 compares wholesale market design in the SLM and BL11A ,with reerence
to renewable generators,. More details on the operation o the SLM can be ound in Appendix C.
1abe 2.:: Covpari.ov of M ava 11. rboe.ae varet ae.igv
Issue SLM BL11A
Market structure Gross mandatory pool Bilateral contracts, net-market
Payment structure
Separate energy and capacity
market
Lnergy only market
Dispatch Central ,priority, dispatch Sel-dispatch
Balancing Central Balancing market
Principles o dispatch and balancing in wholesale electricity markets, and their interactions with
support mechanism design, are important or policy makers and inestors in low-carbon and other
orms o generation. Dispatch and balancing inluence how electricity markets operate, consumer
costs and also inestors` perception o market risk.
2.6.2. Iuture market coupling
One o the key aims o the LU is to create a single, competitie, market in goods and serices across
its Member States. 1his extends to electricity markets, through preious initiaties such as the so-
called 1hird Lnergy Package` on opening up Luropean electricity markets to competition.
\ork on integrating Luropean markets is also underway, drien by Luropean legislation regulating
the cross-border low o electricity. 1his aims to acilitate and increase the low ,or trading, o
electricity between national networks. It aims to moe towards a system where prices are consistent
across a region that is deined by transmission constraints, rather than prices simply being consistent
within a country, or part o a country. Box 2.1 below sets out at a high leel the so-called Luropean
1arget Llectricity Model`.

as ariable price taking generators becoming price making generators ,i.e. being required to submit bids into the market,.
1his change could hae a number o impacts on the SLM.

16

Box 2.J: Luropean 1hird Lnergy Package and 1arget Llectricity Model
1he aim o the 1hird Lnergy Package is to create a single LU energy market. 1he package consists o
two Directies, one concerning common rules or the internal market in gas ,2009,3,LC, and one
concerning common rules or the internal market in electricity ,2009,2,LC,. Under the third package,
the 1arget Llectricity Model is or cross-border market coupling with price zones deined by congestion
,i.e. network issues, rather than national boundaries. Network codes will be deeloped and apply to
network capacity allocation and congestion management between dierent the dierent price zones in
the LU electricity market. 1he Luropean network codes will deine a price zone as a bidding area, i.e. a
network area, within which market participants shall submit their bids day-ahead, in intraday and in the
longer term timerames.
Day-ahead market pricing and capacity allocation
Price-based market coupling is the mandated capacity allocation method in the day-ahead ramework.
lor day ahead capacity allocation, Luropean 1ransmission System Operators ,1SOs, will be required to
implement capacity allocation on the basis o implicit auctions. 1his will be achieed ia a single price
coupling algorithm which simultaneously determines olumes and prices in all releant zones, based on
marginal pricing principles. Calculated zonal prices will only dier in the case o congestion between the
concerned price zones.
lorward capacity allocation
1here will be long-term transmission rights, physical and inancial, to proide market participants with
long-term hedging solutions against congestion costs and day-ahead congestion pricing compatible with
pricing zone delimitation. 1he network codes shall proide or a single platorm or the allocation and
nomination o long term-term transmission rights.
Intra-day capacity allocation
1he network codes will implement an intra-day market that enables market participants to trade energy as
close to real time as possible in order to balance their position. 1he network codes will implement a pan-
Luropean intraday trading platorm supporting trading and pricing o intraday transmission capacity.
Conclusion
1he Llectricity 1arget Model will create proound changes in the LU electricity market. 1his includes
electricity markets in the UK and the Republic o Ireland. 1he Llectricity 1arget Model may require
uture ,potentially structural, change to the SLM.
ovrce: .CR ,CP.
1he practical implications or the all-island market are that there will need to be changes made to
allow the SLM and the GB electricity market ,BL11A, to link. In particular, there will need to be
irm day-ahead prices or both markets, to allow the lows on the interconnectors between the SLM
and GB to be determined based on these prices.
1his should hae beneits, in that it will allow each market to import rom the other at times when
meeting demand locally would be more expensie. In eect, it will moe towards a situation where
plant in GB and on the island o Ireland is dispatched in order o cost across both islands.

1

1his situation should be more eicient, and will in eect be a moe towards a single electricity price
or both islands. Since prices in the SLM are generally higher, a moe towards a single price or both
islands should reduce the SLM price, all other things being equal. In practice, the limited
connections currently in place between the two islands mean that a single electricity price at all times
is unlikely.
1here could also be beneits or renewables, in that at times when the SLM has excess wind
generation, it could export it to GB rather than curtailing it. Ioweer, gien that wind patterns
across the two islands are strongly linked, GB may hae excess wind at the same time. 1hereore,
the ability to export cannot be relied upon.
Realising these beneits will though require some changes or additions to the SLM. 1he price in the
SLM is only determined ex post, and so - unlike the GB market - the SLM does not hae a irm
day-ahead price. 1he implications or the design o the renewable electricity subsidy mechanism are
that one that is linked to the electricity market structure ,such as a CD lI1, is less robust to uture
changes than one that is not ,such as the RO or a Premium lI1,.
More speciically, since a CD lI1 is paid as the dierence between the actual leel o a price and
some set alue, it needs to be clearly speciied what that price is. lor example, it could be the
outturn SMP in the SLM or a particular hal hour, or the aerage price or GB across an entire 24
hour period. I the market structure changes, that price may no longer be aailable, and so the CD
would need to be re-written. Since the RO and the PlI1 make no assumptions about market prices,
they do not need to change i the market structure changes.
1he scale o this issue will be proportional to the scale o the changes in the SLM, this should
become clearer by early 2012. 1he important points or this project are:
x a renewable support scheme designed or the current SLM may or may not need to be
signiicantly reised in a ew years` time,
x it may be diicult to justiy haing a dierent renewable support scheme in Northern
Ireland compared to the rest o the UK and Ireland, particularly i the SLM and BL11A
conerge, and
x the changes to the markets are due in the next ew years, beore the RO is expected to close
to new projects in 201.
1he SLM Committee recently published its initiation o the SLM Market Integration Project.
41

Amongst the options it will consider are day ahead market coupling using inancial contracts market
and more ull-scale market redesign.
A SLM Committee decision in lebruary 2012 will outline whether to proceed with the eolutionary
approach and transitional arrangements.

41
SLM Committee ,2011, Market Integration Project Plan
http:,,www.allislandproject.org,GetAttachment.aspxid~3a36dab-3652-440-a4d-028c938993

18

2.7. Issues
DLCC`s LMR package, as discussed aboe, was drien by DLCC`s oerall objecties or the GB
electricity market ,BL11A,. \hile at a high leel the objecties are the same as or NI there are
crucial dierences, as shown in the table below.
1abe 2.: Creat ritaiv, ôrtberv reava ava Repvbic of reava eectricit, varet. ava ob;ectire.
1opic DLCC/ GB DL1I/ NI RoI Notes
Reverabe
eectricit, target
Around
42

30 by 2020
40 by 2020 Iigher target in NI relects greater
potential. Legally binding target is
or UK as a whole.
^vcear Sees a role or
new nuclear
No new nuclear oreseen Proposed structure o GB`s CD
lI1 drien by the objectie to keep
low carbon generation in the market
and address inestment in nuclear as
well as other orms o low-carbon
electricity generation ,or example,
oshore wind,.
Maret
.trvctvre
BL11A -
bilateral
contracts, net
market
SLM - gross pool
Reverabe
.vpport
Renewables Obligation RLlI1 \hile RLlI1 operates like a CD it
is closer in design to a lixed lI1
than the CD lI1s being considered
in the GB context.
ovrce: CP.
2.7.J. Investor perspective
Section 2.1 aboe showed that the renewables target will require well oer _1 billion o additional
inestment in Northern Ireland electricity generation by 2020. Attracting this inestment to
Northern Ireland will thereore be a key consideration in the design o any NI renewables subsidy
scheme.
\e consider the perspectie o inestors in more detail later in the paper when we come to consider
speciic options. lor now, we simply note that inestors are attracted by high, predictable and
relatiely low risk returns, within stable and predictable regulatory enironments. Inestors also
alue simplicity and comprehensibility in policy, and dislike new and untested policy designs.
2.7.2. Small-scale and micro-generation
1he GB electricity market includes a separate support mechanism
43
or small-scale ,under 5M\,
generation, which is structured as a lI1. 1his does not apply to Northern Ireland, which supports

42
GB has not set a speciic target or renewable electricity. 1he igure o around 30 is based on the lead scenario in the
UK`s 2009 Renewable Lnergy Strategy.

19

small-scale generation through the RO. I the RO will no longer be aailable to new small-scale
generation rom 201, there is a question o how it will be subsidised rom that date. 1here are three
broad options:
1. 1hrough the same scheme as large-scale generation. \hile this would hae the adantage o
consistency, it might not be optimal or the dierent characteristics o micro-generation
,such as the act that it is usually operated by those who are not power market experts,.
2. 1hrough a scheme tailor-made or small-scale generation. 1his could be targeted more
closely at the needs o small-scale generation, but introduces additional complexity and
possible pererse incenties.
3. Not to subsidise small-scale generation at all. Small-scale generation is generally expensie
compared to large-scale generation, as illustrated in ligure 2.2 below. On the other hand, it
may be diicult to justiy giing small-scale generation less support than large-scale.
See Section 3.9 or more on small-scale renewables.
igvre 2.2: Reatire co.t. of reverabe tecbvoogie., b, .ie
44

ovrce: CP.

43
http:,,www.decc.go.uk,en,content,cms,what_we_do,uk_supply,energy_mix,renewable,eedin_tari,eedin_tari
.aspx
44
Source: CLPA, PB, 2010, Detervivatiov of tbe appropriate forv of .vpport for ivcevtiri.ivg tbe aereopvevt of reverabe eectricit,
geveratiov iv ôrtberv reava
C W 8 A u
L W
S v
D
t
S M L

20

2.7.3. State Aid
Any subsidy mechanism could be required to seek State Aid approal rom the Luropean
Commission. 1he Renewables Obligation, or example, receied State Aid approal in Noember
2001
45
. Lnsuring that the aims o the subsidy were clear, and that it was well-targeted, would help to
secure this approal. A similarity with existing schemes, which already hae State Aid approal,
might also be helpul.
2.7.4. 1ransition issues
In any scenario the uture support mechanism or renewables in NI will be dierent. Len i the
NIRO were retained, the orthcoming changes to the RO in GB will hae a major eect on the
support that the NIRO would delier. So the question is not how can transition costs be aoided
but how can they be kept to a manageable leel. 1he likely need or urther changes to any support
scheme to deal with SLM, BL11A coupling also needs to be actored in here.
2.8. Summary
1his section has set out the current context or renewable electricity in, respectiely, NI, GB and the
RoI. \e hae also considered the economic characteristics o alternatie support mechanism
structures ,including Premium and Contract or Dierence lI1 structures,. 1he section which
ollows considers the options or NI`s renewable generation support mechanism package,
speciically how the scheme might be structured and the subsidy unded.

45
http:,,ec.europa.eu,eu_law,state_aids,comp-2000,n504-00.pd

21

3. OP1IONS
As noted aboe, NI has an ambitious target or renewable electricity generation by 2020. Because
renewables are more expensie than conentional generation, some orm o subsidy will be needed
to achiee the leels o deployment targeted. 1his leads to two questions:
x Iow should the subsidy or renewables in NI be structured, bearing in mind the context in
the UK and the RoI
x Iow should the subsidy be unded
\hile there are links between the two questions, in our iew they are suiciently discrete to be
considered separately. \e consider them in turn below.
3.J. Options for structuring subsidy
Once the decision is made to subsidise renewables, there are seeral ways this can be done.
lollowing the discussion in 2.4.2, the options can be categorised by answers to the ollowing
questions:
1. \hat does the generator hae to do or someone to receie the subsidy Possible answers
include:
a. Build generation
b. Connect generation to the grid
c. Be aailable to generate
d. Generate
e. Be dispatched
2. Is the subsidy ixed, or ariable I it aries, how does it do so
3. I the subsidy is or production or sale o electricity, is it in addition to ,some, electricity
price, or is the electricity price netted o in some way
4. \ho is the subsidy paid to
5. \ho sets the leel o the subsidy
6. Iow does the leel o the subsidy ary by technology ,i at all,
1he subsidy design components discussed in Section 2.4.2 also clearly apply.

22

In this discussion, we ocus on options which are paid or generating or being dispatched, rather
than ,say, capital grant schemes
46
. \ithin that, we consider the ollowing options summarised in
1able 3.1 below.
1abe .1: Reverabe vpport optiov.
Option Is subsidy fixed? Is subsidy net of electricity price?
lixed lI1 \es \es
RLlI1 No \es
CD lI1 \es i 2-way, no i 1-way \es
Premium lI1 \es No
Renewables Obligation No No
ovrce: CP.
Payments can be made to generators directly or to generators ia suppliers ,as in the case o both the
RO and the RLlI1,. 1he leel o subsidy can also be set either by the goernment or through some
orm o auction or market-based process ,as or the RO,. \e consider these questions in the
sections below on each option. Lxamples o the options are set out in detail in the appendices.
3.2. Iixed II1
A classic ixed lI1 is a ixed payment, in _, M\h, paid to generators or each unit o electricity
generated or dispatched. 1he leel can be set either by an administrator, or through some orm o
auction or tender process. One possible ixed lI1 approach is illustrated in ligure 3.1 below ,arrows
show lows o money,.

46
1hese were considered in CLPA`s preious report or DL1I. \e concluded that they were unsuitable because o their
administratie costs, lack o ocus on output and the perception that they could be relatiely short-term leading to
boom and bust` behaiour in the market.

igvre .1: iea 1 approacb - pa,vevt. to geverator.

1he main diiculty with a ixed lI1 in the SLM is t
through the SLM Pool and receie the SLM price ,the
a mechanism or the generator to pay this SMP to Go
or there has to be an intermediary, which receies
ixed lI1. Both options are possible,
complex ,or Goernment, since there are likely to
other hand, suppliers may demand some orm o compe
to and rom generators.
3.3. RLII1 arrangement
An alternatie to the lixed lI1 approach illustrated in ligure 3.1 is a s
similar to RLlI1 in the Republic. In this scenario,
letter o oer` rom goernment which conirms to an electricity supp
entering into a PPA with the generator they will be
reerence prices set or the scheme.
In this case, rather than Goernment needing t
enters into a PPA with the generator and recoers t
through the PSO mechanism. 1he 1ransmission System Operator ,1SO, or Distribut
Operator ,DSO, would apply these charges to suppliers who would then recoer
consumers through the PSO ley.

4
Lxcept rom generators below 10M\
pa,vevt. to geverator.

he main diiculty with a ixed lI1 in the SLM is that all
4
electricity generated has to be sold
through the SLM Pool and receie the SLM price ,the SMP,. 1his means that either there has to be
a mechanism or the generator to pay this SMP to Goernment ,and receie the ixed lI1 in return,
or there has to be an intermediary, which receies the SMP on the generator`s behal and pays it the
ixed lI1. Both options are possible, but haing generators make payments directly would
complex ,or Goernment, since there are likely to be many more generators than suppliers. On the
other hand, suppliers may demand some orm o compensation or the cost o managing payments
lixed lI1 approach illustrated in ligure 3.1 is a support mechanism structure
similar to RLlI1 in the Republic. In this scenario, renewable generators would need to apply or a
om goernment which conirms to an electricity supplier that in return or
entering into a PPA with the generator they will be receie balancing payments against the
reerence prices set or the scheme.
rather than Goernment needing to be a counterparty to the lI1 payment the supplier
enters into a PPA with the generator and recoers the balancing payments under the scheme
1he 1ransmission System Operator ,1SO, or Distribut
y these charges to suppliers who would then recoer
consumers through the PSO ley.

23
electricity generated has to be sold
her there has to be
ernment ,and receie the ixed lI1 in return,
the SMP on the generator`s behal and pays it the
but haing generators make payments directly would be more
be many more generators than suppliers. On the
nsation or the cost o managing payments
upport mechanism structure
would need to apply or a
lier that in return or
against the agreed
o be a counterparty to the lI1 payment the supplier
ments under the scheme
1he 1ransmission System Operator ,1SO, or Distribution System
y these charges to suppliers who would then recoer the costs rom

24

^
^D
S
M
S 8Lll1
SC
l
1SC uSC SC 8Lll1

REFIT payments
igvre .2: R1 approacb

Under a RLlI1 approach, subsidy lows through suppliers to the generators, with the generator
receiing a ixed price or their power through a PPA arrangement. Suppliers would handle all sale
and dispatch administration with the market operator and 1SO. In Appendix A, we consider the
RLlI1 and look at some o the possible lessons or NI i it chose to implement a similar design o
lI1. 1able 3.2 below sets out how a RLlI1 approach might work in NI, and compares it to the
operation o the RLlI1 in the Republic.
1abe .2: R1 ae.igv i..ve.
Llement RLII1 NI RLII1 RoI
Contract orm Generator PPA with suppliers Generator PPA with suppliers
Contract price
Agreed between the supplier and the
generator
Agreed between the supplier and the
generator
Cost recoery
Suppliers receie payments against
administered reerence prices.
Suppliers receie payments against
administered reerence prices.
Administration
DL1I proides letters o oer`
NIAUR administers PSO payments
DCLNR proides letters o oer`
CLR administers PSO payments

3.4. CfD II1
3.4.J. NI CfD
A lI1 implemented through a Contract or Dierence
rom consumers ,ia Goernment
generating renewable electricity or sourcing it to
or clawbacks`, where i the price o electricity i
money back. 1his is known as a two
consumers and generators, suppliers, to distinguish it rom a one
lows rom consumers.
1he dierences in practice in the short
almost always more expensie than electricity rom
ineitably more complex to calculate. On the other
beneits o a two-way CD should increase. 1hey also hae the beneit
consumers and preenting windall
unlikely to be reached requently in practice and so this beneit is more
ligure 3.3 illustrates a CD lI1 with a similar or
Ioweer, this case the CD would settled against th
igvre .: CfD 1 approacb - C voae rbere pa,vevt i. tbrovgb

SMP paid to generators and CfD II1 paid to generators

A lI1 implemented through a Contract or Dierence ,CD, is eectiely a top
Goernment, to generators, to compensate them or the increas
generating renewable electricity or sourcing it to supply consumers. CDs can also inole reunds`
or clawbacks`, where i the price o electricity is aboe the leel o the lI1, consumers receie
money back. 1his is known as a two-way` CD ,since money can low in both directions b
suppliers, to distinguish it rom a one-way` CD where money only
1he dierences in practice in the short-term are likely to be small, since renewable electr
almost always more expensie than electricity rom conentional sources. 1wo-way CDs are also
ineitably more complex to calculate. On the other hand, as the costs o wind power start to all, the
way CD should increase. 1hey also hae the beneit o limiting costs to
consumers and preenting windalls or renewable generators, suppliers, een i thos
requently in practice and so this beneit is more perceied than real.
ligure 3.3 illustrates a CD lI1 with a similar orm and structure as has been
Ioweer, this case the CD would settled against the SLM SMP.
C voae rbere pa,vevt i. tbrovgb geverator.
and CfD II1 paid to generators

25
,CD, is eectiely a top-up` payment
, to generators, to compensate them or the increased cost o
ers. CDs can also inole reunds`
s aboe the leel o the lI1, consumers receie
way` CD ,since money can low in both directions between
way` CD where money only
term are likely to be small, since renewable electricity is
way CDs are also
hand, as the costs o wind power start to all, the
o limiting costs to
s or renewable generators, suppliers, een i those limits are
perceied than real.
and structure as has been proposed in GB.

26

As ligure 3.3 illustrates. the GB proposal or CD lI1s is currently that lI1 payments are likely to
be made directly to the geverator ,through a orm o arm`s length Goernment agency, rather than to
suppliers. As illustrated in the matrix below, generators will still be responsible or dispatch and
balancing in the GB market. A NI CD lI1 would operate dierently within the SLM arrangements
and central dispatch.
igvre .1: CfD pa,vevt vatri
II1 payment to:
Ceverator vppier.
Cevtra
NI CD lI1 RLlI1 , NIRO
Dispatch
Proposed GB CD lI1 RO ,Lngland, \ales and Scotland,
ef

One o the proposed beneits ,in the GB context, o CD lI1 payments made directly to generators
rather than suppliers is the potential to reduce the need or the scale o discounts under PPAs
oered by otakers ,or example, in relation to balancing costs as discussed in the preious section,.
Ioweer, these beneits depend on a number o prerequisites, including whether generators will be
able to borrow without a PPA being in place, or without some other means o ensuring that physical
output is purchased.
1he payment route or a CD lI1 payment needs to be considered separately in NI gien the SLM`s
gross mandatory pool structure. In the absence o an NI standalone institution, NI generators would
need to contract with the GB body separately rom the SLM.
3.4.2. Worked example payments under a possible CfD II1
A 100M\ wind generator is connected to the grid. It receies a two-way CD lI1 contract rom
DL1I, NIAUR or 20 years, with a contract price o _60 per M\h.
At the end o each month, the generator reports to DL1I, NIAUR its output in each SLM period
in that month. DL1I, NIAUR looks up the SMP or each period when the generator was
dispatched. Suppose or example in period 5 on the 10
th
day o the month, the SMP was _50 per
M\h, and the generator was running at 50 load. SLM periods are hal an hour, so in that time the
generator produced 25M\h ,100M\ times 50 load times ' hour,. 1he dierence between the

2

SMP and the contract price was _10 per M\h, so DL1I, NIAUR pays the generator _250
,25M\h times _10, or its output in that period.
1he issue o how the CD lI1 would work i the generator were constrained o would need to be
addressed.
3.4.3. What are the design questions for a CfD II1?
As is highlighted by the discussion in section 2.4.2, a CD lI1 would raise complex design issues or
NI`s policy makers:
x \hat should be the contract orm
x \hat is the contract duration
x \ho would be the seller and who would be the buyer
x \hat should the CD market reerence price be and how would it interact with the SLM`s
Capacity Payment Mechanism ,CPM,.
x \hat should the CD strike price be Should it be lat within the year ,i.e. no time o use
signal, Should it be subject to indexation ,e.g. RPI,
x Should CD payments be based on:
R market scheduled quantities,
R dispatched quantities, or
R just aailability ,capacity,
x Is the CD one-way or two-way
x Iow is the price leel set
R Administratiely
R Competitiely ,through auction or tender,
1he answer to these questions depends on the type o low-carbon generation technology which the
CD contracts are deeloped or and the wholesale market structure. A crucial dierence between
GB and NI is whether nuclear is included and the dierent bilateral contracting and gross
mandatory pool structures.
1able 3.2 below sets out how a CD lI1 might work in NI, and compares it to the current GB
proposal ,please see Annex B o the \hite Paper or urther details on how the CD is expected to
be structured or dierent renewable technologies.

2
8

1
a
b
.
2
:

C
f
D

1

a
e
.
i
g
v

i
.
.
v
e
.

C
f
D

I
I
1

d
e
s
i
g
n

H
o
w

m
i
g
h
t

i
t

w
o
r
k

i
n

N
I
?

W
h
a
t

h
a
s

b
e
e
n

p
r
o
p
o
s
e
d

f
o
r

G
B
?

C
o
n
t
r
a
c
t

o
r
m

a
,

D
i
e
r
e
n
t

o
r
m
s

o

c
o
n
t
r
a
c
t

l
i
k
e
l
y

o
r

b
a
s
e
l
o
a
d
,

i
n
t
e
r
m
i
t
t
e
n
t

a
n
d

l
e
x
i
b
l
e

l
o
w
-
c
a
r
b
o
n

g
e
n
e
r
a
t
i
o
n
.

D
i
e
r
e
n
t

o
r
m
s

o

c
o
n
t
r
a
c
t

l
i
k
e
l
y

o
r

b
a
s
e
l
o
a
d
,

i
n
t
e
r
m
i
t
t
e
n
t

a
n
d

l
e
x
i
b
l
e

l
o
w
-
c
a
r
b
o
n

g
e
n
e
r
a
t
i
o
n
.

C
o
n
t
r
a
c
t

o
r
m

b
,

1
w
o
-
w
a
y

C
D

1
w
o
-
w
a
y

C
D

M
a
r
k
e
t

r
e
e
r
e
n
c
e

p
r
i
c
e

C
l
e
a
r

o
u
t
t
u
r
n

r
e
e
r
e
n
c
e

p
r
i
c
e

o
r

t
h
e

c
o
n
t
r
a
c
t

-

t
h
e

S
L
M

S
y
s
t
e
m

M
a
r
g
i
n
a
l

P
r
i
c
e
.

N
o

s
i
n
g
l
e

r
e
e
r
e
n
c
e

p
r
i
c
e
.

D
i
e
r
e
n
t

o
p
t
i
o
n
s

b
e
i
n
g

c
o
n
s
i
d
e
r
e
d

o
r

d
i
e
r
e
n
t

g
r
o
u
p
s

o

t
e
c
h
n
o
l
o
g
i
e
s
.

N
o

e
x
-
p
o
s
t

a
d
j
u
s
t
m
e
n
t

i
s

p
r
o
p
o
s
e
d
.

C
o
n
t
r
a
c
t

o
l
u
m
e

L
i
k
e
l
y

t
o

b
e

m
e
t
e
r
e
d

o
u
t
p
u
t

g
i
e
n

N
I

r
e
n
e
w
a
b
l
e
s

p
r
e
d
o
m
i
n
a
n
t
l
y

i
n
t
e
r
m
i
t
t
e
n
t
.

1
o

b
e

c
o
n
i
r
m
e
d
,

m
e
t
e
r
e
d

o
u
t
p
u
t

o
r

i
r
m

o
l
u
m
e
.

C
o
n
t
r
a
c
t

a
l
l
o
c
a
t
i
o
n

A
d
m
i
n
i
s
t
e
r
e
d

o
r

a
l
l
o
c
a
t
e
d

t
h
r
o
u
g
h

a
n

a
u
c
t
i
o
n

m
e
c
h
a
n
i
s
m
.

G
o
e
r
n
m
e
n
t

a
g
e
n
t

i
s

c
o
u
n
t
e
r
-
p
a
r
t
y
.

A
d
m
i
n
i
s
t
e
r
e
d

o
r

a
l
l
o
c
a
t
e
d

t
h
r
o
u
g
h

a
n

a
u
c
t
i
o
n

m
e
c
h
a
n
i
s
m
.

G
o
e
r
n
m
e
n
t

a
g
e
n
t

i
s

c
o
u
n
t
e
r
-
p
a
r
t
y
.

C
o
s
t

r
e
c
o
e
r
y

S
t
i
l
l

t
o

b
e

r
e
s
o
l
e
d
.

S
t
i
l
l

t
o

b
e

r
e
s
o
l
e
d
.

A
d
m
i
n
i
s
t
r
a
t
i
o
n

C
o
u
l
d

b
e

D
L
1
I

,

N
I
A
U
R

o
r

a

a
r
m
`
s

l
e
n
g
t
h

b
o
d
y

r
e
s
p
o
n
s
i
b
l
e

o
r

a
d
m
i
n
i
s
t
e
r
i
n
g

a
n
d

n
e
g
o
t
i
a
t
i
n
g

C
D
s

D
i
e
r
e
n
t

p
r
o
p
o
s
e
d

r
o
l
e
s

o
r

D
L
C
C

a
n
d

t
h
e

a
r
m
`
s

l
e
n
g
t
h

b
o
d
y

s
e
t
t
i
n
g

p
o
l
i
c
y

a
n
d

a
d
m
i
n
i
s
t
e
r
i
n
g

C
D
s

B
a
l
a
n
c
i
n
g

r
i
s
k

\
o
u
l
d

d
e
p
e
n
d

o
n

h
o
w

t
h
e

s
c
h
e
m
e

o
p
e
r
a
t
e
d

i
s
-
a
-
i
s

t
h
e

S
L
M

b
u
t

n
o
t

l
i
k
e
l
y

t
o

b
e

m
a
t
e
r
i
a
l

L
i
k
e
l
y

t
o

b
e

s
h
o
r
t
-
t
e
r
m

b
a
s
i
s

r
i
s
k

r
o
m

b
a
l
a
n
c
i
n
g

m
a
r
k
e
t
.

29

One issue with any CD lI1 is that it would ineitably be compared to the RLlI1 in the Republic.
1his would also apply to the CD lI1 or GB.
Iaing a similar support mechanism to neighbouring countries would be beneicial, all other things
being equal, because o the ease o explaining the mechanism to inestors and the reduced potential
or pererse interactions. I the expectation is that the GB market and the SLM will be more
strongly coupled in uture - in many ways, orming a single market - haing a broadly similar
support mechanism across the whole o the British Isles is likely to make sense.
1here is though the issue o subsidy competition` - inestors comparing the subsidy they might
get in neighbouring regions and ocusing their capital where the return is highest. 1here will
ineitably be questions about any dierences in support leels and degrees o risk between NI and
the RoI. \hile there could be an attempt to set the same or similar leels, NI support will be in _,
while that in the Republic will be in t, so dierences will persist, and will ary oer time as the
exchange rate aries.
It would, in theory, be possible to implement a CD lI1 across the whole UK. Ioweer, the
dierences in market structure would create complications and administration costs or policy
makers and consumers. 1here may also be inestor impacts i they beliee there are dierent
risks,return balances across market areas in the UK.
One key issue would be that a single market reerence price and contract structure would not be
sensible or the UK as a whole - it would likely lead to signiicant undesirable consequences to hae
payments to NI generators based on the GB electricity price.
On the other hand, there is nothing preenting the leel o the CD lI1 ,the .trie price) being the
same across the UK. Suppose the strike price was _60 per M\h, the SLM price was _50 and the
BL11A price was _45. Generators in NI could receie a _10 top-up to the SLM market price, while
those in GB could receie a _15 top-up to the BL11A price. Both would receie _60 in total. As
highlighted in this example, the leel o subsidy payments could dier between GB and NI, but it
would be possible to make the total reenue the same.
O course, while the leel o the CD lI1 could be the same across the UK, this is not the same as
saying that it should be. 1here may be good reasons ,as or current RO banding leels, why the leel
should be dierent in NI.
A CD lI1 that applied across the whole UK might hae the adantage o it being clearer that the
costs should be recoered rom consumers across the UK, rather than the costs o deployment in
one region being recoered rom consumers in that region.
3.S. Premium II1
A Premium lI1 pays a certain _,M\h irrespectie o the electricity price. As or the ixed lI1, this
can be set either by an administrator or through some orm o auction or tender process.
Since the Premium lI1 is paid whateer the electricity price, the interactions with the electricity
market are reduced and the arrangements can be simpler. 1here are though still interactions, such as

a tendency or generators to oer their power into the market at ery low or e
that they can receie the Premium lI1.
1he operation o the PlI1 paid through an arm`s length body ,similar to the p
model, is illustrated below. Note that there is
making this a relatiely simpler approach than the
that the lI1 is ixed makes it simpler to calculate
igvre .:: Previvv 1 approacb

Ioweer, because the premium
dependent on the uture electricity price. 1his mak
although less risky than the RO, since the leel o
3.S.J. Worked example how a Premium II1 could work
A 100M\ wind generator is connected to the grid. It
NIAUR or 20 years, with a contract price o _30 pe
At the end o each month, the generator reports to DL1I, NIAUR its output in each
in that month. Suppose or example in period 5 on t
running at 50 load. SLM periods are hal an hour,
,100M\ times 50 load times ' hour,. DL1I, NIAUR pays
_30, or its output in that period. 1he generator a

oer their power into the market at ery low or een negatie prices so
that they can receie the Premium lI1.
paid through an arm`s length body ,similar to the proposed GB CD lI1
is illustrated below. Note that there is no payment rom generators to DL1I, NIAUR,
making this a relatiely simpler approach than the lixed lI1 or lI1 with CD. In addition, the act
that the lI1 is ixed makes it simpler to calculate the payments.

premium lI1 payment is ixed, the deployment o renewables is still
dependent on the uture electricity price. 1his makes it a more risky proposition than a CD lI1,
, since the leel o support at least is ixed.
how a Premium II1 could work
A 100M\ wind generator is connected to the grid. It receies a premium lI1 contract rom DL1I,
NIAUR or 20 years, with a contract price o _30 per M\h.
generator reports to DL1I, NIAUR its output in each
in that month. Suppose or example in period 5 on the 10
th
day o the month, the generator was
running at 50 load. SLM periods are hal an hour, so in that time the generator produced 25M\h
00M\ times 50 load times ' hour,. DL1I, NIAUR pays the generator _50 ,25M\h times
_30, or its output in that period. 1he generator also receies whateer the SMP was or that period.

30
en negatie prices so
roposed GB CD lI1
no payment rom generators to DL1I, NIAUR,
lixed lI1 or lI1 with CD. In addition, the act
is ixed, the deployment o renewables is still
es it a more risky proposition than a CD lI1,
receies a premium lI1 contract rom DL1I,
generator reports to DL1I, NIAUR its output in each SLM period
day o the month, the generator was
so in that time the generator produced 25M\h
the generator _50 ,25M\h times
lso receies whateer the SMP was or that period.

31

3.S.2. What are the design questions for a Premium II1?
1he implementation questions or a PlI1 are a subset o those or the CD lI1. In particular, there
is no need to consider the relationship between the leel o the lI1 and the price o electricity. 1he
high-leel remaining questions are:
x Iow is the lI1 price leel set
R Administratiely
R Competitiely ,through auction or tender,
x \hat happens i a generator is scheduled to run, but is constrained o
More details on how PlI1s can work, and how they hae worked in practice in Spain, can be ound
in Appendix B.
\e expect that in general, a PlI1 ,as described in DLCC`s LMR package and the design options
aboe, would operate largely as does the current ROC system, except with a ixed payment rather
than one that aries with the ROC price.
1here is one main exception to this. 1he RO is ulnerable to changes in the olume o ROC-eligible
generation, although the goernment has tried to reassure inestors that the mechanism will be
presered through adjustments to headroom. In contrast, a PlI1 would be contractually enorced
and may thereore be considered more aourably by inestors. Under a Premium lI1, there may
also need to be a counterparty to lI1 contracts oered in the SLM.
A PlI1 would also operate in a similar way as the NIRO within the current trading and settlement
rules or the SLM:
x \here a renewable generator elected to be price taking, they would continue to hae a
market scheduling bid o zero but receie the System Marginal Price as their energy payment.
x \here they elect to be ariable price makers, proided the alue o P
lLOOR
is negatie,
generators could continue to bid their subsidy into the market.
As with a CD lI1, there would be contract allocation issues to consider. lor example, would the
premium taris be set administratiely ,like under the NIRO where Northern Ireland`s policy
makers set the banding leels or renewable technologies, or through an auctioning ,competitie`,
mechanism as has been proposed in the LMR package
A stand-alone NI PlI1 scheme would require the region`s own public sector organisations to
manage and administer the scheme ,unlike under the RO where administration is currently handled
by Ogem through a serice agreement with NIAUR,. 1he implication is a Premium lI1 approach
,assuming GB were to adopt a CD lI1 approach, could lead to higher administration costs than at
present, although as discussed in the section on CDs, the dierent structure o the SLM means that
there will be some need or DL1I, NIAUR administration whateer option is chosen.

32

\e note a key principle o LMR ,as well as encouraging new generators with access to new orms o
inance, is cost eectieness and aordability ,the support mechanism design should delier its
objecties eiciently to minimise cost increases or consumers,. Gien the challenges o setting
PlI1 leels ,as illustrated by the banding reiews under the RO, policy makers would also need to
consider whether there is a need to adopt caps and loors on PlI1 payments to address risks o
windall gains under such a regime.
3.S.3. How a Premium II1 might work in NI
1able 3.3 below sets out how a PlI1 might work in NI, and compares it to the scheme in operation
in Spain and the operation o the NIRO.

3
3

1
a
b
:

P
r
e
v
i
v
v

1

a
e
.
i
g
v

I
I
1

d
e
s
i
g
n

i
s
s
u
e

H
o
w

m
i
g
h
t

i
t

w
o
r
k

i
n

N
I
?

H
o
w

h
a
s

i
t

w
o
r
k
e
d

i
n

S
p
a
i
n
?

N
I
R
O

C
o
n
t
r
a
c
t

o
r
m

N
o
r
t
h
e
r
n

I
r
e
l
a
n
d

c
o
u
l
d

a
d
o
p
t

a

c
o
n
s
t
a
n
t

P
r
e
m
i
u
m

l
I
1

c
o
n
t
r
a
c
t

,
c
o
n
s
t
a
n
t

t
o
p
-
u
p

a
b
o
e

t
h
e

w
h
o
l
e
s
a
l
e

p
r
i
c
e
,

o
r

a

m
o
r
e

c
o
m
p
l
e
x

c
o
n
t
r
a
c
t

d
e
s
i
g
n

,
e
.
g
.

w
i
t
h

c
a
p

a
n
d

l
o
o
r

o
n

p
a
y
m
e
n
t
s
,
.

S
l
i
d
i
n
g

P
r
e
m
i
u
m

l
I
1

,
w
i
t
h

c
a
p

a
n
d

l
o
o
r

o
n

t
a
r
i

p
a
y
m
e
n
t
s
,
.

D
i
e
r
e
n
t

c
o
n
t
r
a
c
t

d
u
r
a
t
i
o
n
s

o
r

d
i
e
r
e
n
t

t
e
c
h
n
o
l
o
g
i
e
s
.

O
b
l
i
g
a
t
i
o
n

p
l
a
c
e
d

o
n

s
u
p
p
l
i
e
r
s
.

C
o
n
t
r
a
c
t

p
r
i
c
e
s

a
n
d

o
l
u
m
e

A
d
m
i
n
i
s
t
e
r
e
d

o
r

a
l
l
o
c
a
t
e
d

t
h
r
o
u
g
h

a
n

a
u
c
t
i
o
n

m
e
c
h
a
n
i
s
m
.

G
o
e
r
n
m
e
n
t

h
a
s

s
e
t

r
e
e
r
e
n
c
e

m
a
r
k
e
t

p
r
i
c
e
s
,

a
l
o
n
g

w
i
t
h

c
a
p
s

a
n
d

l
o
o
r
s
.

R
O
C

b
u
y
-
o
u
t

p
r
i
c
e

s
e
t

b
y

g
o
e
r
n
m
e
n
t
.

O
b
l
i
g
a
t
i
o
n

s
i
z
e

a
n
d

t
e
c
h
n
o
l
o
g
y

b
a
n
d
i
n
g

a
l
s
o

s
e
t

b
y

g
o
e
r
n
m
e
n
t
.

C
o
n
t
r
a
c
t

a
l
l
o
c
a
t
i
o
n

1
h
e

o
b
l
i
g
a
t
i
o
n

s
i
z
e

,
w
i
t
h

h
e
a
d
r
o
o
m

m
e
c
h
a
n
i
s
m
,

i
s

s
e
t

t
o

a
c
c
o
m
m
o
d
a
t
e

e
x
i
s
t
i
n
g

a
n
d

u
t
u
r
e

r
e
n
e
w
a
b
l
e

g
e
n
e
r
a
t
i
o
n
.

1
h
e

o
b
l
i
g
a
t
i
o
n

o
n

s
u
p
p
l
i
e
r
s

c
r
e
a
t
e
s

a

m
a
r
k
e
t
`

o
r

g
e
n
e
r
a
t
o
r
s

R
O
C
s
.

C
o
s
t

r
e
c
o
e
r
y

L
e
y

o
n

c
o
n
s
u
m
e
r

b
i
l
l
s
.

L
e
y

o
n

c
o
n
s
u
m
e
r

b
i
l
l
s
.

S
p
a
i
n

h
a
s

c
o
n
s
i
d
e
r
e
d

a
d
o
p
t
i
n
g

e
x
e
m
p
t
i
o
n
s

o
r

l
o
w
-
i
n
c
o
m
e

h
o
u
s
e
h
o
l
d
s
.

S
u
p
p
l
i
e
r
s

p
a
s
s

t
h
e

c
o
s
t

o

p
u
r
c
h
a
s
i
n
g

R
O
C
s

t
h
r
o
u
g
h

t
o

c
o
n
s
u
m
e
r

b
i
l
l
s

,
i
.
e
.

i
m
p
l
i
c
i
t

l
e
y

i
n

i
n
a
l

e
l
e
c
t
r
i
c
i
t
y

b
i
l
l
,
.

A
d
m
i
n
i
s
t
r
a
t
i
o
n

\
e
r
e

G
B

t
o

a
d
o
p
t

a

C
D

l
I
1

t
h
e
n

t
h
e

s
c
h
e
m
e

m
a
y

n
e
e
d

t
o

b
e

a
d
m
i
n
i
s
t
e
r
e
d

b
y

N
o
r
t
h
e
r
n

I
r
e
l
a
n
d
`
s

p
o
l
i
c
y

m
a
k
e
r
s

,
D
L
1
I
,
N
I
A
U
R
,
.

1
a
r
i
s

s
e
t

b
y

g
o
e
r
n
m
e
n
t
.

R
O
C
s

a
r
e

t
r
a
d
e
d

b
e
t
w
e
e
n

U
K

s
u
p
p
l
i
e
r
s

a
n
d

g
e
n
e
r
a
t
o
r
s
.

O
g
e
m

i
s

r
e
s
p
o
n
s
i
b
l
e

o
r

a
d
m
i
n
i
s
t
e
r
i
n
g

t
h
e

s
c
h
e
m
e
.

34

3.6. RO (or other quantity based instrument)
1his is a similar arrangement to a PlI1, except that the additional payment aries with the alue o a
Renewable Obligation Certiicate. 1his is thereore inherently more risky or inestors than the
Premium lI1, and slightly more diicult to understand.
Because the RO is a UK-wide scheme, any use o it needs to consider what is happening in GB. 1he
DLCC LMR package includes a proposal to close the RO in Lngland & \ales to new projects rom
April 201. Lxisting projects would continue to receie ROCs until 203, unless moed oer to a
new lI1 scheme. New projects would be subsidised through a CD lI1 or Premium lI1.
1he most important question or the uture o the NIRO beyond 201 is what the implications o
this closure might be. An initial analysis o the issues suggests that while it might be made to work,
the costs and uncertainties are likely to make it unattractie or both NI and or inestors in
renewables. \e set out some o the key concerns below.
1oatiit, of tbe ROC price
Projects accredited under the RO receie support or a maximum o 20 years
48
. 1he RO has been in
existence since 2002
49
and so by the 2020s, some o the older projects unded under the RO will no
longer be supported ,i they hae not already closed,. 1his means that there will be ewer ROCs
produced, and so ,unless the RO target is reduced, the price o a ROC will increase, increasing costs
or NI consumers. 1hese closures, and the regular re-setting o the RO target, are likely to make the
RO price highly olatile.
1he \hite Paper includes an Annex on maintaining support or renewable technologies in this
transition period.
Reqvirevevt for ^ to aavivi.ter tbe .cbeve aove
I the Lngland and \ales RO closes to new projects rom April 201, it can be closed by April
203. A NIRO that continued to take on new applications ater April 201 would hae to guarantee
continuing payments beyond April 203, which would imply that NI would coer all
50
administratie
costs or the RO beyond that date.
In any case, GB is likely to argue that NI should take on an increasing proportion o the costs o
administering the RO as the Lngland and \ales RO winds down.

48
Ogem, 2011, Reverabe. Obigatiov: Cviaavce for geverator.
http:,,www.ogem.go.uk,Sustainability,Lnironment,RenewablObl,Documents1,RO20Generator20Guidance
20May20201120inal.pd paragraph 6.21
49
Renewables Obligation Order, 2002, http:,,www.legislation.go.uk,uksi,2002,914,pds,uksi_20020914_en.pd
50
Unless Scotland also continued with the RO

35

.iv. ava ob;ectire.
1he aims and objecties o a GB-based administrator eectiely winding down the GB RO are
likely to be ery dierent to those o a NI administrator looking to encourage new renewables
through the NIRO.
Inestors might worry that the RO would be drien by the desire to wind up the larger part o it,
which is unlikely to make it an attractie long-term scheme.
3.7. Combinations of options
1he options discussed aboe are not mutually exclusie, and there is no reason in principle why
DL1I, NIAUR could not implement two or more o the options. 1he balance o arguments is
though probably in aour o a single option. 1o start with, implementing multiple options is o
course more expensie than implementing only one. 1here is also the question o how the options
would interact - including whether there would be pererse incenties.
One argument oten adanced in aour o haing more than one option is that small-scale
renewables need to be incentiised through a simpler ,and possibly more generous, scheme than
large-scale, because their owners are not large sophisticated power market participants. Dealing with,
or example, a two-way contract or dierence with quarterly or annual settlement might be too
complex or the smallest generators.
1he argument about complexity is less strong i the option is a RLlI1 approach implemented
through suppliers ,since generators would receie a ixed payment through a PPA, or a Premium
lI1 ,since this is the simplest possible subsidy rom the generator`s point o iew,. 1he issue about
the leel o support needed can be dealt with through separate subsidy bands or small scale
renewables within an oerall subsidy approach that is likely to hae separate subsidy bands or
dierent large-scale renewables in any case.
1he other main argument adanced or haing multiple schemes is the need to proide
grandathered support or existing renewables. It is true that grandathering ,and the promise o
grandathering, is important or inestor conidence, but it does not ollow that inestments need to
be coered by the same support mechanism or their whole lie. A transer to equialent support in
a new scheme can also be acceptable. Ioweer, lenders ,including to renewable generators, oten
hae clauses in their contracts that allows the loans to be treated as in deault ,and so allowing the
lender to change the terms, i there is a material change in the law or in regulation. Care would be
needed in making the transition to aoid these clauses being triggered. 1he judgement here is about
the balance between the costs o administering multiple schemes and the cost o calculating
equialent subsidies in a new scheme and migrating existing renewables to it.
Some countries do hae more than one support scheme operating. lor example, in Spain and the
Czech Republic, generators hae a choice o either a lixed or a Premium lI1. I we subsequently
choose a Premium lI1 or large-scale generation, it is hard to see why any other option would be
used or small-scale. I a CD lI1 is chosen or large-scale generation, there may be an argument or

36

a Premium lI1 or small-scale. 1he question or the Renewables Obligation then becomes whether
to close it immediately and migrate existing projects to the new lI1, or to continue it or existing
projects only ,which is the GB approach,.
3.8. Options for funding subsidy
1here are two questions to be asked about the source o unding: should consumers or taxpayers be
the source, and rom which country, ies 1his gies us, in theory, six possible options:
1. NI consumers
2. NI taxpayers
3. UK consumers
4. UK taxpayers
5. NI and RoI consumers
6. NI and RoI taxpayers
\e assess these options or easibility, equity and cost to NI.

3

1abe .1: ...e..vevt of .vb.ia, fvvaivg optiov.
Option Ieasibility Lquity Cost implications Consider?
NI consumers
Possible through
PSO ley
Arguable, since NI
consumers paying
or deliery o their
own target ,but or
more than their
share o the UK
target,
See Part B o report,
but likely higher than
now.
Impact on uel
poerty could be
signiicant
\es
NI taxpayers
Unclear. DLCC
Renewable Ieat
Incentie unded
rom general
taxation, but in
general unding
rom tax likely to be
politically diicult.
Could be argued that
cost should be borne
in proportion to
electricity used,
which supports
unding by
consumers not
taxpayers
As or NI consumers No
UK
consumers
1hrough UK-wide
ley
Positie. Cost o
achieing UK
renewables target
borne by UK
consumers as a
whole
Drien by GB
targets - NI
population roughly
1,40 o GB so
impact is negligible
\es
UK taxpayers
As or NI taxpayers
option. In addition,
would require
agreement o IM
1reasury
As or NI taxpayers
- breaks link
between cost and
electricity
consumption
Unclear No
NI and RoI
consumers
Not clear although
there may be
precedents in the
SLM
No clear case or NI
consumers unding
RoI renewables
targets and ice ersa
Unclear No
NI and RoI
taxpayers
Likely to be
extremely diicult
rom tax
soereignty
perspectie
No clear case. Unclear No
Based on this initial assessment, we conclude that we should ocus our analysis on the UK and NI
consumer-unding options, or reasons o equity and easibility.
1his brings us to the question o deliery. \hile more analysis is needed, we expect that the NI
consumer-unding option could be managed through the existing PSO ley. It is not clear what
DLCC has in mind or unding its CD lI1 proposal. \e see two basic options.
1he irst is to use the same approach as the RO, which is unded by imposing an obigatiov on
suppliers, who then recoer the cost o this rom their customers. Many other policies operate in
this way ,e.g. the Supplier Obligation,. It could be made to work or a lI1, with suppliers being

38

obliged to oer PPAs ,there would need to be some agreed terms, to renewable generation. 1here
would hae to be some orm o reenue sharing mechanism agreed between suppliers, to preent
one or two suppliers being orced to coer all the costs o renewables.
1he alternatie is to und the lI1 through an explicit ley on consumer bills. 1here may need to be
legislation to implement this ,as there was in the 2010 Lnergy Act or the - neer used - CCS ley,.
1here does not seem to be any clear reason why such a ley could not apply UK-wide. 1he cost
would then be set out explicitly on consumer bills.
3.9. 1reatment of small-scale renewables
1he support mechanism or small-scale
51
generation also needs to be considered. \e see three basic
options:
1. Support small-scale renewables through the same support mechanism as or large-scale.
2. A separate Northern Ireland-speciic mechanism or small-scale
3. Using the GB scheme or small-scale generation
A basic question is how material small-scale support would be. 1able 2.1 and 2.2 suggests that the
potential or small-scale generation in NI is limited. \e thereore need to question whether the
administration and other costs o a separate scheme would be justiied. 1his suggests that option ,2,
is probably unattractie, unless there is a strong desire to retain lexibility or NI to set its own rates,
which could not be guaranteed under ,3,. 1he choice between options ,1, and ,3, will depend on
what large-scale option is chosen.
I it is an NI CD lI1, or a Premium lI1, the scheme it will need to be designed as simply as
possible or small-scale generators. lor example, under either approach, the lI1 could be proided
as top-up` payment aboe small-scale generation export taris oered by suppliers ,as currently
under the NIRO, or alternately the wholesale price in the SLM.
lor consistency, and to allow NI ull control oer the leel o taris, this may be preerred to ,3,.
Ioweer, i the chosen option is a CD lI1 deliered through generators, option ,3, may be worth
considering. 1he administration in that case would be done by Ogem, which would presumably
look to DL1I or unding to support this.

51
lor Northern Ireland we assume small-scale is anything under the 10M\ ae vivivi. threshold in the SLM. In GB,
small-scale is deined as under 5M\, with micro-generation` as anything under 50k\.

39

4. LVALUA1ION OI OP1IONS
Iaing set out options in the preious section, in this section we proide a short qualitatie
ealuation o those options. 1his is ollowed by an assessment o the issues , questions that are
considered in more detailed in the remainder o the report ,Part B,.
4.J. Lvaluation of options
4.J.J. Criteria
1he ideal option or our purposes would be ery attractie to inestors in both large and small-scale
renewable electricity generation, hae no negatie implications or security o supply, and be low
cost and low risk. Based on this and on other DL1I, NIAUR objecties, we propose the criteria in
1able 4.1.
1abe 1.1: Criteria for a..e..ivg optiov.
Aim Criteria Notes
Attractie to
inestors in
renewable
generation
Leel o return 1he higher the better
Riskiness o return 1he lower the better
Complexity or inestors
A new or complex scheme requires inestors to spend
time learning about it, and is inherently more risky
since untried
Similarity to existing
schemes
Security o
supply
Impact on security o NI
electricity supply
n,a
Cost 1otal cost
52

Oxera
53
argues that cost o lI1 and RO is similar, or at
least can be made to be similar i you choose the bands
and taris appropriately. Cost dierences likely to be
drien by relatie risk o options.
Other
Suitability or small
generators ,under 10M\
54
,
n,a
Cost and riskiness o
implementation
Administratie cost and
complexity
ovrce: CP.

52
Iow the cost is distributed is considered in the preious section.
53
Oxera, 2009, .tabi.bvevt of tbe ôrtberv reava Reverabe. Obigatiov to 2020
54
lor consistency with the SLM ae vivivi. threshold

40

4.J.2. Qualitative evaluation
1he table below shows our high-leel assessment o the options against the criteria aboe. \e score
the options on a scale o 1-5, where 5 is best.
1abe 1.1: vitia bigbere a..e..vevt of optiov. agaiv.t criteria
Criteria RLII1 CfD II1 Premium II1 NIRO
Leel o return 3 4 4 5
Riskiness o return 5 5 2 2
Complexity or
inestors
4 3 4 2
Similarity to existing
schemes
4 2 3 5
Impact on security o
NI electricity supply
3
55

1otal cost
56
4 5 2 1
Suitability or small
generators
3 3 4 4
Cost and riskiness o
implementation
3 3 3 1
5

Administratie cost
and complexity
4 3 4 3
OVLRALL 33 3J 29 26
ovrce: CP.
1he oerall score needs to be treated with caution, since it assumes that all the criteria hae equal
weight, but does suggest a ocus on lI1s rather than the NIRO is justiied.
Maintaining the NIRO long-term is not really practicable as it is a UK wide approach, realistically
requiring it to operate more widely than just NI. Continuation o the scheme beyond 201 would be
administratiely complex and potentially uncertain or inestors.
A starting point or understanding the dierences between the lI1 support mechanisms is to
understand the speciic eatures o their design, the risks this gies rise to and how these risks might
be mitigated ,as discussed in the preious section,.

55
No clear reason why any option should be better or worse or security o supply
56
See Part B or urther details
5
\hile the RO is already implemented, implementing the changes needed to keep it open to new installations post
201 would be extremely complex

41

1he appropriateness o a particular mechanisms can be drien by both varet reqvirevevt. as well as
the requirements o the t,pe. of tecbvoog, being supported.
1he RLlI1 - which rom the point o iew o the generator can be characterised as a lixed lI1 -
allocates market and balancing risks to suppliers, through a standardised PPA, with the supplier
receiing a contracted allowance to manage them.
58
In this case, suppliers are required to manage
market reenues receied across trading periods and the olatile nature o renewable output. 1he
lixed lI1 would proide a ixed predicable reenue stream or renewable generators.
O the new options, it is the PlI1 which most closely resembles the existing NIRO. It could
operate in any uture market structure by being delinked rom the wholesale electricity price.
Inestors are comortable with a ixed reenue stream which they know they can receie alongside
wholesale market reenues ,which are more risky,.
Ioweer, an eectie CD lI1 design structure should ,in theory, be perceied more aourably by
inestors as it proides a more predictable reenue stream or generators. In comparison, under a
PlI1 policy generators bear wholesale electricity price risk, as currently under the NIRO. By
proiding a more predicable reenue stream, a CD lI1 could help to reduce risk and long term
consumer costs through a slightly lower cost o capital.
59

Maret arravgevevt.
1he proposals being deeloped by DLCC are to deal with the speciic requirements o the GB
market. 1hus, whilst haing the same support mechanism structure as GB appears attractie, the
practical implications o applying it within the SLM need to be taken into account. In ealuating the
practicability o dierent support mechanism or NI it is also essential that both cvrrevt and fvtvre
market arrangements are taken into account.
As noted in the preious two sections, at the moment, the SLM is quite dierent to BL11A
operating in GB, which is a much less liquid, so called net market`, with most electricity being sold
through bi-lateral contracts ,PPAs, between generators and suppliers. 1he SLM in contrast is a gross
mandatory market with a separate energy component and a capacity payment, central dispatch and
intermittent generation ,such as wind, treated with priorit, ai.patcb.
1he market dierences between GB and the SLM are more important in deeloping a lixed lI1 or
the CD lI1 approach than in the case o the NIRO or a PlI1, as a lI1 CD approach, in
particular, is much more embedded within the electricity market, rather than being essentially an
additional income stream that is added to that o the wholesale reenues.

58
Suppliers receie a balancing payment - 15 o the large wind category tari - to coer the cost o managing the
short term ariable production o wind energy.
59
See CLPA ,2011,: Note on the impacts o the CD lI1 support package on costs and aailability o capital`

42

Under the SLM 1rading and Settlement Code a renewable generator can sell electricity either
directly into the pool or enter into a PPA with a supplier or intermediary to sell the energy or its
generator`s behal. \hile a generator ,participating, in the SLM does not ace balancing risks in the
same way as a GB generator, it will still need to manage its output and the requirements rom
participating in the market. Market risks and requirements rom participating in the market may
dier by renewable generation technology.
1o begin with, this means that the reerence price or a CD lI1 policy will need to be determined
dierently within the SLM to what it is in GB. Currently, the reerence price proposed or GB ,in
the case o intermittent generation, is a day-ahead price sourced rom the best representation o day-
ahead prices at the time the CD lI1 is allocated. 1he aim o the proposed reerence price in GB is
to try and keep generators within the already illiquid GB wholesale market. In contrast, in the
current market a CD lI1 in the SLM would need to reerence ,in some orm, the single System
Marginal Price ,SMP, or the SLM.
In itsel, such a dierent reerence price would not appear to be an insurmountable hurdle. \hereas,
a concern with introducing a CD lI1 within the GB market is the basis risk that this gies rise to,
generators within NI might not ace such a basis risk with the SLM. 1he central SMP and the
balancing arrangements or the market, mean that in the SLM, there may not be the additional
uncertainty through the introduction o basis risk which, or example, might be introduced through
a CD lI1 policy based on an aerage` wholesale traded price index as has proposed in GB.
60

lrom this perspectie, a CD lI1 policy may actually it better within the SLM than it does in the
BL11A, at least rom those considering inesting in generation in NI. Ioweer, at a more detailed
leel there may still be risks that we hae not currently identiied. Indeed, these might only be
identiied when new CD lI1 contracts are deeloped or generators in NI, which cannot be based
entirely on those deeloped or GB, due to the dierent market arrangements that need to be taken
account o.
61

62

vb.ia, pa,vevt for
1here are certain attractions o a structure similar to RLlI1 where the subsidy lows through the
suppliers to the generators, and remuneration o arious balancing and market participation costs
incurred by the generators are ixed rather than negotiated. Such an arrangement would proide

60
Generators in the SLM would receie wholesale energy payments while an ex-post outturn SMP would proide a clear
reerence or settlement o the CD lI1.
61
lor example, the limited number o market participants who contacted during the course o our project highlighted
uture issues around compensation or constraints and curtailment o wind generation and the appropriate contract
olume ,should payment be based on metered olume or irm olume ,i.e. aailability,, \e hae not been able to
explore the materiality o these issues urther within the scope o our study.
62
See LirGrid , SONi All Island 1SO lacilitation o Renewables Studies` or challenges o managing the technical and
operational implications associated with high shares o wind power in the All Island power balance.

43

renewable generators a ixed payment through PPAs, and has the beneits o simplicity, being
enorced contractually, and being applicable to all sizes o generation including small-scale.
63

Under this approach, NI`s policy makers might be responsible or setting oerall policy and
administering the scheme ,as currently under the NIRO, while suppliers` would be responsible or
managing renewable output in the market and the settlement ,cash-out`, o wholesale and subsidy
payments across trading periods. Institutionally the approach is also simpler although the scheme
would still require administration.
1he alternatie would be to ollow a similar route as proposed in GB with goernment responsible
or negotiating, managing and monitoring the CD lI1s and the subsidy paid directly to NI`s
generators. 1his model would allow renewable generators to sell directly into the pool and receie
subsidy payments directly. 1his could proide more commercial lexibility than generators haing to
contract with registered electricity suppliers in NI who then act as intermediaries or the generators
in the market.
64

\hile a renewable generator in GB generally require a PPA to dispatch, renewable generators in the
SLM ,through priority dispatch rules, hae a guaranteed route to market. 1he issue or NI`s
generators ,although compensation or constraints and curtailment may in uture mean olumes
become more o an issue depending on the design o support mechanism, would be the price
realised or their power. A payment route airect, to tbe geverator would thereore seem to address price
risk and allow renewable generators to continue to operate within the market.
Ioweer, compared to a RLlI1 approach, a direct payment route requires a more complicated
contractual structure. 1his might deter inestors. In some ways it is also an aaravtage rather than
disadantage that a subsidy payment low tbrovgb .vppier. means intermediary agents manage
renewable output in the market, rather than eery generator in NI haing to interace with the
market and manage these actiities directly.
\ith substantial numbers o new, predominantly wind, generators granted access to the SLM
transmission operation is expected to become increasingly constrained, and compensation or
constraints and curtailment may make market reenues more uncertain and costly or renewable
generators. Part o the rationale or retaining RLlI1`s 15 per cent balancing payment within the
SLM has also been to remunerate suppliers or managing ex-post correction o payments under the
R-actor correction process which can take oer a year.
1he implication is a direct payment route to generators could lead to high leels o administration
and complexity. Generators would need to manage cash payments on metered output across the
year`s trading period rather than receiing a ixed payment stream.

63
1he model is also consistent with how RLlI1 operates in the Republic and would be consistent with supplier
64
A risk with the RLlI1 approach ,payment to suppliers, is generators ability to achiee a air price or their power is
diminished with only the small number o suppliers to contract with in Northern Ireland.

44

Clearly there are arguments in aour o either approach. NI`s policy makers need to consider the
uture role they enisage or renewable generators in the SLM, and how 1rading and Settlement
rules acilitate generators contracting inancially through intermediary structures, rather than directly
interacing with the market.
Maret covpivg
1here are urther complications, because o changes to the SLM to meet Luropean electricity
market integration goals, speciically in terms o the need or market coupling and it would seem, a
market design more along the lines o the BL11A than the SLM.
I the expectation is that the GB market and the SLM will be more strongly coupled in uture - in
many ways, orming a single market - haing a broadly similar .trvctvre or a support mechanism
across the whole o the British Isles is likely to make sense. Indeed, the SLM Committee is
currently considering a number o options or integrating the SLM with its neighbouring electricity
markets.
65

lrom market structure perspectie, this might support delaying a moe to a new scheme until uture
market arrangements become clearer. Against this, there is also an argument that with GB adopting
a CD lI1 rom 2014 ,and RLlI1 in place in the RoI, inestment could be undermined in NI,
where there is delay in implementing a new ,and potentially improed, support scheme.
1here are pros and cons either way, clearly there may be alue ,rom both an inestor and policy-
maker perspectie, in delaying a inal decision on rbev to vore to a new scheme until the requirement
or SLM redesign ,i any, is clearer ,expected to be early 2012,. In the interim, more detailed analysis
is required to examine how a CD lI1 might be designed or the SLM beore and ater integration
with neighbouring markets.
.avivi.tratiov arravgevevt.
\hilst as regards markets, there may be arguments on both sides as to whether to pursue the
proposed GB approach, or that o the RLlI1, we beliee that these are less ambiguous when it
comes to the institutional administration requirements o adopting the GB arrangements in
Northern Ireland.
Under the likely GB deliery model, an arm`s length organisation is likely to be responsible or
negotiating, managing and monitoring the CD lI1s. 1he arm`s length body may also be contract
counterparty to CD lI1s. I this were to be applied in NI, either a standalone parallel entity would
need to be established or the SLM, or the GB entity would need to be the counterparty to contracts
operating in the SLM.
1he irst would be likely to be costly, whereas the latter would require an understanding o the SLM
as well as the BL11A, so could also be administratiely more complex than the operation o the

65
Including day ahead coupling using the inancial contracts market and the potential long term redesign o the SLM to
meet Luropean single market requirements.

45

NIRO and potentially expensie too. In comparison, under a RLlI1 arrangement, the suppliers
would be responsible or managing both wholesale and subsidy lows through the PPA - as in the
Republic o Ireland - which would seem administratiely simpler.
vvaivg
1he \hite Paper makes no clear decision on how CD lI1 costs will ultimately be unded accept
that unding will be rom electricity consumers. One option would be to use the same approach as
the RO, which is unded by imposing an obigatiov on suppliers, who then recoer the cost o this
rom their customers. 1he alternatie is to und the lI1 through an explicit ley on consumer bills.
1he inal decision we expect will depend on the role o the arm`s length organisation ,s, unctions
is-a-is suppliers and their contractual positions with generators. \here the arm`s length
organisation is counter-party to the lI1 contracts, a mechanism similar to the Public Serice
Obligation ,PSO, ley is more likely.
In terms rbo unds support mechanism, irre.pectire of it. forv, we note that Northern Ireland is part o
a UK wide renewable energy obligation. It may thereore make sense or Northern Ireland to be
part o a UK-wide scheme in terms o unding the subsidy, rather than Northern Ireland operating
its own scheme. Ioweer, it is also not necessarily the case that Northern Ireland`s participation in
a UK-wide scheme would lead to oerall lower Northern Ireland consumer costs ,an issue we return
to below as part o the assessment o consumer costs below,.
4.2. Questions and issues to be considered in Part B
Iaing considered the options on the basis o a relatiely high-leel qualitatie assessment, in this
section we discuss the issues which need to be considered in more detail in the remainder o this
report ,through modelling and quantitatie analysis, to conirm these conclusions.
4.2.J. Identifying the costs of the NIRO
As the existing support mechanism in place in NI, the NIRO proides the counteractual against
which other approaches to supporting renewable generation in NI need to be ealuated rom a cost-
beneit analysis perspectie.
\e hae set out a Baseline or NI renewable inestment in the SLM or the period 2012 to 2020
based on current NI renewable energy policy. \e then consider the costs o the NIRO ,in its
current orm, including its projected impact on NI domestic consumer bills and the oerall
wholesale cost o electricity in the region.
4.2.2. Assessing alternative options
Iaing considered the cost to NI customers and ability to incentiise renewable generation
deployment under the NIRO we go on to compare modelling outcomes to those under alternatie

46

support mechanism options discussed in the subsections aboe, in particular an NI PlI1 or CD
lI1 scheme implemented in the SLM.
1he leel o renewable inestment that may be encouraged through a system o lI1s will depend on
the structure o how the lI1 contract is set ,as discussed in the preious section, as well as the leel
o support oered relatie to projected renewable generation costs. 1he questions we seek to
address through modelling are as ollows:
x Iow might NI consumer costs be expected to change under alternatie design parameters o
NI lI1 policy including a CD lI1 or PlI1 policy in the SLM
x Iow do projected NI consumer costs in the modelling compare to projected consumer
costs in GB under LMR package options
x linally, what might be the impact on NI consumer costs were a socialised lI1 policy
adopted or the whole o the UK
x \e also consider the implications o Carbon Price Support ,CPS, or the SLM and NI`s
renewable energy policy ,drawing on quantitatie analysis proided by SLMO ,the SLM
market operator,,.
As part o the modelling we also consider the resource ,inestment, costs o renewable generation
projected in the modelling alongside a series o additional cost and beneit metrics which support
the oerall ealuation and assessment o options.
4.2.3. Assessing the costs of a II1 policy for small-scale generation
1he deployment o small-scale renewable generation is a strategic objectie or NI. \e proide
indicatie consumer cost analysis were NI small-scale generators to receie a top-up` lI1 payment
oer and aboe current generation export taris oered to small-scale generators in NI. 1he
questions we speciically consider are:
x \hat are the costs o small-scale and micro generation and what leel o subsidy may be
needed to stimulate deelopment
x \hat contribution could small-scale generation be expected to make to the oerall
consumer costs o renewable generation in the region
\e note that consumer costs o small-scale generation are particularly uncertain gien uncertainty
oer resource and deployment o dierent technologies in the region ,beyond simply the inancial ,
economic actors addressed in the modelling,. \e note the leel o consumer costs projected in the
modelling would be lower in any year i the projected take-up o small-scale generation in the
modelling were not realised.

4

4.2.4. Sensitivities under alternative scenarios
\e conclude Part B by considering the sensitiity o the modelling results to key input assumptions,
such as uel prices and the projected capacity expansion plan o renewable generation in NI and the
SLM. \e also consider an alternatie option that DL1I and NIAUR hae requested that we model
which is continuation o the NIRO beyond 201.
4.2.S. Conclusions
1his allows us to deelop, alongside the more qualitatie analysis presented in Part A,
recommendations and inal conclusions in respect o the orm o support mechanism or renewable
generation in NI.

48

PAR1 B: ANALYSIS OI POLICY OP1IONS
In the preious section, we set out a range o policy options or reorming NI`s renewable electricity
generation support mechanism. In this section we proide a more quantitatie assessment o the
dierent options, ollowed by an assessment o the risks and implementation issues associated with
the dierent options.
As a starting point or the quantitatie analysis we hae set out a Baseline or the period 2012 to
2020. 1his is based on:
x current policy ,e.g. continuation o the Northern Ireland Renewables Obligation ,NIRO,,,
x expected deelopments in the NI and Great Britain ,GB, electricity sector,
x SLM 1SO projections o renewable generation resource and build rates, and
x DLCC`s assumptions o uel prices, demand and generation capital costs.
1he only change we hae made to policy in the Baseline is to assume that the NIRO end-date will be
extended to 203 to be consistent with the RO.
1he Baseline is used as a comparison ,counteractual`, or the assessment o dierent options. \e
go on to assess the implications o the introduction o:
x a Carbon Price Support mechanism in the SLM,
x PlI1 options in the SLM, and
x CD lI1 options.
Unlike the rest o the UK ,where a small-scale generation lI1 scheme is in orce, the current
support regime or large and small-scale renewable generation in NI is the NIRO. Alternatie LMR
options in NI thereore need to consider sub-options and packages o support regime that would
address NI`s large and small generation sectors. As the issues or small-scale renewable generation
are somewhat dierent to larger, more commercial scale renewables projects, we consider these
issues in a separate section.

49

S. MODLLLING IRAMLWORK AND ASSUMP1IONS
In this section we describe the modelling ramework which has been deeloped to assess the
impacts o LMR options. \e begin by proiding a high leel structural oeriew o the model,
beore going on to discuss the modelling approach or dierent support options and the outputs
produced by the model. \e then discuss the key assumptions in the model, including generation
costs and wholesale electricity prices.
S.J. Model structure
1he objectie o the model is to assess how alternatie approaches to supporting renewable
generation in NI compare to the costs, beneits and leels o deployment deliered by other support
mechanisms. 1hese scenarios are based around the interaction o a number o policy support
options - including the NIRO ,in its current orm,, Contract or Dierence ,CD, lI1s and a
Premium lI1 ,PlI1, - and the SLM.
1he structure o the modelling ramework is summarised at a high-leel in ligure 5.1.
igvre :.1: CP. voaeivg approacb

ovrce: CP.
S.J.J. Modelling approach
1he irst stage o the modelling assumes NI renewables build is an exogenous ariable. A capacity
expansion plan or dispatchable and non-dispatchable plant or the all-island market is deeloped
consistent with the SONi,LirGrid Generation Adequacy statement.
66

66
See Appendix 2 o that document or urther details.
Assumptions / inputs Calculations CBA
Supply curves
Build assumptions
Capital costs
Financing costs
Operating costs
Learning curves
Administrative costs
Benefit inputs
Environmental assumptions
Renewable support
policy option
Generation mix &
build assumptions
SEM Revenues
Generation costs
Required revenues
Required level
of subsidy
Monetary costs
Monetary benefits
Investor
cost of capital
Scenario NPV
Model Outputs
Assumptions / inputs Calculations CBA
Supply curves
Build assumptions
Capital costs
Financing costs
Operating costs
Learning curves
Administrative costs
Benefit inputs
Environmental assumptions
Renewable support
policy option
Generation mix &
build assumptions
SEM Revenues
Generation costs
Required revenues
Required level
of subsidy
Monetary costs
Monetary benefits
Investor
cost of capital
Scenario NPV
Model Outputs

50

\holesale electricity prices in the SLM are then deried rom a simple dispatch model that uses a
load duration cure and the SLM generation capacity expansion plan merit order stack. ligure 5.2
illustrates the key phases in the market modelling ,a more detailed description o the assumptions
and approach adopted is proided in Appendix A,.
igvre :.2: !boe.ae price voaeivg

ovrce: CP.
1he second phase o the modelling uses the outturn SLM price projections rom Phase 1 to assess -
or a gien renewable generation resource potential - the proportion o projects that would be
iable, based on the expected costs o dierent technology types,scales ,which ary oer time, and
the reenues those projects would expect to receie ia the SLM, other sources o reenues and the
selected RLS support mechanism. 1his deries a projected NI renewable generation supply cure
or each year, which is then assessed against the renewable generation capacity expansion plan
assumed in the original SLM market modelling.
1he third phase o the modelling iterates the SLM dispatch model to update the SLM wholesale
price projections to be consistent with the projected NI renewable generation supply cure. Note
that or lI1 support options - where the leel o payments are set to delier the capacity expansion
plan enisaged in the 1SO generation adequacy statement - the projected RLS generation expansion
plan is consistent with the irst phase o the modelling.
1he ourth phase o modelling ramework calculates the costs and beneits o the selected support
mechanism scenario and presents a range o model outputs demonstrating expected costs and
beneits in the modelling scenario.

SEM load duration curve
SEMgeneration capacity
development plan
Plant SRMC merit order
Plant dispatch to set SMP
Time-weighted wholesale
price
Generation plant data
Time-weighted capacity
payment
Time-weighted SEMprice
R
e
n
e
w
a
b
l
e

g
e
n
e
r
a
t
i
o
n

i
n
v
e
s
t
m
e
n
t

m
o
d
e
l
l
i
n
g
Calculations
Fuel / carbon price and plant
parameter assumptions
0
1000
2000
3000
4000
5000
6000
7000
M
W
h
Hours

51

S.J.2. Model outputs
1he outputs generated by the model are grouped as ollows:
x Deployment - An assessment o the leel o RLS-generation deployment, speciically
whether NI`s 2020 target is met and the contribution o dierent generation technologies to
the deployment.
x Costs and benefits - A comparison o the costs and beneits o the Baseline and alternatie
lI1 options, including the resource cost, any costs o oer subsidy, emission saings and
,where applicable, other economic beneits.
x Cost to consumers - \e also proide an analysis o the likely costs to consumers o the
support mechanism, including the cost per unit o electricity consumption and the cost per
unit o renewable generated under a scheme.
Speciic cost and beneit outputs presented rom the modelling analysis include:
x Resource cost equals total expenditure comprising capital and operating costs on
generation plant under a gien policy scenario.
x Production subsidy cost equal to the total cost o the subsidy proided to renewable
generators.
x Deadweight loss windfall gains - deined as the leel o subsidy cost oer and aboe
that which would strictly be required to make inestment iable.
x Value of carbon savings - which shows, based on an assumed counteractual uel mix
within the SLM, the beneits associated with displacing carbon emitting generation.
S.2. Modelling LMR options
In this section, we describe each LMR policy option and the way that they hae been implemented
in the modelling. In DLCC`s LMR Impact Assessment ,IA, an important consideration was the
impact on inestor risk including cost o capital impacts o dierent support packages. 1his section
thereore also sets out how we hae sought to capture inestment impacts rom dierent options in
our modelling ramework. 1his includes:
x 1he potential reduction in the co.t. of capita or inestors under a PlI1 or CD lI1 rom
costs o capital under the existing NIRO.
x 1he potential impact o proposed support packages on setting the ere o subsidy renewable
generators beneit rom in NI.
\e proide a summary o the economic characteristics o each support regime we hae modelled
and how they aect inestor decision making.

52

S.2.J. II1 options
In this section we describe how dierent lI1 options hae been implemented in the modelling.
As we noted in our preious report to DL1I,NIAUR
6
, policy makers ace a diicult task when
administering a long term lI1 scheme. 1hey are only presented with current inormation and
assumptions on generation technology costs and how these are expected to deelop in the uture
when designing a scheme that will inluence inestment decisions oer a signiicant time period.
1hey ace inormation asymmetries relatie to project deelopers and uncertainty with respect to the
long term costs o emerging renewable technologies.
In contrast, the goals and objecties or a lI1 scheme ,or example achieement o NI 2020 targets
or renewable electricity generation, are oten apparent, and the costs o not achieing objecties
high. Policy makers also need to set a long term inestment signal to the market and inestors o
expected returns and the structure o the support scheme.
\hen setting lI1 leels, policy makers thereore ace a dilemma, that i the leel o support
proided to RLS generation under a lI1 is set too low, while the risk o windall gains to inestors
is reduced, the risk o ailing to incentiise suicient deployment in renewable generation increases,
and thereore the risk o ailing to achiee public policy objecties and renewable energy targets also
increases. In contrast, i the leel o support proided under a lI1 is set to high, inestment and
deployment o renewable generation may result, but with potential windall gains to inestors.
lor example, under a CD lI1 approach the issue is the leel at which the CD strike price` is set
while under a PlI1 the issue is the required lI1 proided on top o the wholesale price ,which is
inherently uncertain oer long term planning periods,.
Policy makers hae sought to address these issues through use o competitie allocation mechanisms
,such as auctions, which aim to reeal deeloper inestment costs rather than the tari leels being
set administratiely by Goernment or its agent.
ligure 5.4 illustrates the policy dilemma diagrammatically.

6
CLPA , PB ,2010,: Determination o the appropriate orm o support or incentiising the deelopment o
renewable electricity generation in Northern Ireland - Volume B`

53

igvre :.1: v.tratire ai.tribvtiov of geveratiov co.t.

ovrce: CP.
1hese implementation issues can be explored in the modelling ramework ,under alternatie lI1
policy designs, by assessing the sensitiity o the results to key assumptions, or example changes in
renewable generation capital costs and required inestor hurdle rates. Ioweer, the uncertainty and
the risks associated with these assumptions mean a range o costs, beneits and support mechanism
design structures need to be considered or a ull impact assessment o the dierent policy options
in NI. \e discuss the speciic PlI1 and CD lI1 design options CLPA has modelled in the
subsections which ollow.
Previvv 1
As discussed aboe, a PlI1 pays a certain _,M\h or renewable electricity whateer the electricity
price. Because the PlI1 payment is ixed, the deployment o renewables is still dependent on
perceptions o uture electricity prices in the SLM.
lor the purposes o the modelling, we assume the PlI1 is paid based on output, as is the case under
the NIRO, gien the price taker arrangements or non-dispatchable plant bidding in the SLM. \e
assume the taris are administered by goernment.
1he PlI1 leels are set to bridge the enironmental premium` or unding gap` or dierent
renewable technologies in NI. An assessment o the required enironmental premium is thereore
necessarily dependent on assumptions o generation costs o dierent technologies.
1he modelling adopts the ollowing approach or the setting PlI1 leels:
Mid-point
(average) RES
generation costs
RES-
generation
costs in
Northern
Ireland
Low RES
generation costs
High RES
generation costs
Renewables
supply curve
costs /MWh

54

x PlI1 leels or dierent technologies are set by subtracting their leelised cost rom the
leelised cost o a CCG1 and rounding to the nearest _5,M\h,
x the lI1 premia are then iterated until NI generation targets ,the capacity expansion plan in
the modelling, is met.
68

Consistent with the DLCC IA, we assume that the PlI1 policy is implemented in 2014. Renewable
generation plant commissioned on or ater 2014 would all under the PlI1 scheme. Plant that are
commissioned beore 2014 all under the NIRO. lrom 2014 onwards, plant alling under the NIRO
are proided grand athered payments set equal to the prealent buy-out price plus ten per cent
headroom.
69

lrom our work or DLCC on cost o capital issues, we understand that lenders and inestors are
reasonably comortable with existing ROCs as there is a track record. 1his is also most likely the
main reason why some respondents to DLCC`s consultations appear to aour a PlI1 - with
amiliarity and being comortable with existing arrangements being the arguments adanced.
It is clear rom this that inestors across the UK are reasonably comortable with a portion o
reenue being exposed to wholesale price risk. 1hus, a PlI1 is assumed to operate largely as per a
NI ROC in the modelling and as such is assumed to hae no impact on cost o capital relatie to
current support arrangements ,i.e. the NIRO,.
CfD 1
A lI1 implemented through a CD is eectiely a top-up` payment rom NI consumers to
suppliers or generators, to compensate them or the increased cost o generating renewable
electricity or sourcing it to supply consumers. 1he principle is similar to a lixed lI1 in that a stable
earnings stream independent o the SLM wholesale price is proided to NI renewable generators ,or
to suppliers, through an agreed CD strike price` ,with the CD top-up payment deried rom the
CD wholesale market reerence price,. \e hae made the ollowing assumptions or modelling
CD lI1 payments:
x the contract is paid on metered generation ,i.e. it is not a classic CD whereby a ixed olume
o M\h are procured,,
x CD contract strike prices are set to coer the leelised ,LRMC, cost o renewable
generation plant in the SLM,
x the CDs operate two-way` - i.e. i the wholesale price rises aboe the contract strike price
,set at leelised cost, the generator pays the dierence to the contract counterparty,
x as with a PlI1 scheme, the policy is assumed to be implemented in 2014 with the same
arrangements or grandathering the NIRO.

68
\e note a similar assumption was adopted in the original LMR Impact Assessment modelling.
69
1his was the assumption adopted also adopted in the DLCC IA.

55

Gien the structure o the model ramework, a CD lI1 is assumed to operate in a similar way to a
lixed lI1 ,i.e. we assume generators bear no basis risk in the inestment analysis,. 1his is consistent
with how a CD payment is likely to be structured within the current SLM gross pool structure and
the current trading and settlement rules in the SLM.
Under a CD lI1 regime we hae modelled a ie percentage point increase in gearing or oshore
wind and other emerging renewable generation technologies ,e.g. tidal,. \e assume a maximum
increase in gearing o 2.5 or onshore wind. 1his causes a reduction in inestor cost o capital or
the CD lI1 scenario ,i.e. generation costs relatie to the baseline,.
va.cae geveratiov
lor both PlI1 and CD lI1 policy scenarios, the modelling assumes NI small-scale generators
receie a top-up` lI1 payment oer and aboe the current NIL small-scale generation export tari
or classes o generator who currently qualiy or the scheme. Small-scale generator classes who
currently do not qualiy or the NIL export tari receie a top-up` payment aboe the wholesale
price in the SLM. 1he GB small-scale generation lI1 scheme operates according to a similar
ramework whereby small-scale generators receie an export tari ,or electricity exported o the
generation site, plus a ixed top-up` ,generation, payment or total metered generation rom the
small-scale unit.
eeaiv tariff co.t recorer,
In terms o how lI1 costs are recoered, the modelling adopts two approaches:
x a mechanism ,such as the PSO ley which recoers the cost o the Directed Contract costs
in the SLM, is used to ley the total cost o lI1 payments on NI electricity consumers only.
x a mechanism is used to socialise the costs o lI1s and a grandathered NIRO across UK
electricity consumers as a whole.
\e inestigate the impacts o NI orming part o a UK wide ,rather than standalone, lI1 cost
recoery mechanism based on the indicatie CD lI1 costing analysis proided in DLCC`s LMR
Impact Assessment. 1he lI1 costs o NI RLS generation are added to the aerage wholesale energy
price and net support costs projected or GB.
S.2.2. Summary
1able 5.1 summarises the economic characteristics assumed in the modelling or each o the LMR
support regime options.

56

1abe :.1: vvvar, of .vpport vecbavi.v ae.igv iv tbe voaeivg
Llement PII1 CfD II1
\holesale price Variable lixed
Volume
llexible ,applies to metered
output,
llexible ,applies to metered
output,
lorm o subsidy Premium payment to RLS
Agreed strike price paid to
RLS
Payment leel LRMC
0
LRMC o RLS
Implementation 2014 2014
Impact on NIRO Grandathered Grandathered
Small-scale 1op-up lI1 1op-up lI1
ovrce: CP.
S.3. Model inputs
Iaing set out the outputs produced rom the modelling ramework, in this section we introduce
the main model inputs and assumptions.
1he model has our main sets o exogenous inputs that drie the modelling results:
x Rgeveratiov co.t a..vvptiov. - including operating, capital and inancing costs o new-build
RLS-generation.
x ectricit, aevava a..vvptiov. - that underpin the wholesale market modelling scenarios and
the subsidy costing analysis.
x Rgeveratiov vi ava bvia a..vvptiov. - scenarios o the contribution o RLS technologies
haing regard to the aailable resource in NI and build-constraints to 2020.
x ve price a..vvptiov. - scenarios o uture o projections or uel costs, such as the cost o
natural gas, distillate oil and coal.
x Carbov price a..vvptiov. - scenarios o uture carbon prices including the UK carbon price
support mechanism
\e discuss each in turn below.

0
Long run marginal cost ,LRMC, o renewable generation minus LRMC o CCG1.

5

S.3.J. Generation cost projections
Capital costs assumptions or new build generation ,conentional ,or the wholesale market
modelling, and renewable plant, hae been taken rom a range o sources including the Mott
McDonald UK Llectricity Generation Costs Update report June 2010 or DLCC and the cost
analysis proided by PB in CLPA`s preious report to DL1I,NIAUR.
1
\e also hae adjusted
technology generation cost assumptions where NI speciic data has been aailable. 1hese input
assumptions are used to estimate the leelised , bus-bar `, costs per M\h or a range o renewable
generation technologies and sizes.
ligure 5.5 illustrates current leelised cost estimates or a range o larger scale renewable
technologies. Capital cost scalers are applied across the modelling period to relect projected alls in
renewable generation costs ,learning rates`,.
igvre :.:: erei.ea co.t e.tivate. ;cvrrevt) ;_,M!b)

ovrce: CP. ava rariov.
ligure 5.5 illustrates that generator costs, on a leelised basis, are much higher or emerging
renewable generation technologies, such as 1idal and Oshore \ind, compared to more established
and mature technologies such as onshore wind.

1
See: CLPA,PB ,2010,: Determination o the appropriate orm o support or incentiising the deelopment o
renewable electricity generation in Northern Ireland - Volume B`

0
50
100
150
200
250
/
M
W
h

58

S.3.2. Llectricity demand
Llectricity demand assumptions assumed in the modelling are illustrated in ligure 5.6. NI demand
orecasts are sourced rom SONi,LirGrid`s 2011-2020 Generation Capacity Statement.
2
1he annual
demand orecasts or GB correspond to the ULP 2010 DLCC Central scenario or total electricity
supply. In the context o GB, electricity supply is deined as gross generation less the amount o
electricity used on station sites ,own use,. It thereore corresponds to the term Supplied ,gross,`
used in the Digest o United Kingdom Lnergy Statistics ,DUKLS,.
igvre :.: ectricit, aevava pro;ectiov. ;1!b)

ovrce: DCC c Oî , irCria
S.3.3. Renewable energy resource potential
Seeral preious studies hae sought to quantiy the increase in generation capacity that is easible
and that will be required to meet targets. ligure 5. shows projected NI generation capacity in the
central modelling scenario. It can be seen that the bulk o deelopments in NI generating capacity
under this scenario are expected to come rom onshore and oshore wind. Inormation o current
wind applications ,sourced rom the NI planning serice, has been used to inorm expected
inestment sizes o onshore wind installations. DL1I`s Strategic Lnergy Assessment and the SONi
, LirGrid 2011 - 2020 Generation Adequacy statement hae been used to inorm expected capacity
resource o small-scale generation and marine energy.

2
SONi,LirGrid ,2010,: All-island Generation Capacity Statement.
8.4
8.6
8.8
9.0
9.2
9.4
9.6
9.8
10.0
10.2
10.4
358
360
362
364
366
368
370
372
374
376
378
2012 2013 2014 2015 2016 2017 2018 2019 2020
N
I

D
e
m
a
n
d
G
B

D
e
m
a
n
d
GB NI

59

igvre :.: Reverabe geveratiov capacit, ;cevtra .cevario) ;2020)

ovrce: D1 , Ore .rvp , irCria , Oî
ligure 5. projects a signiicant quantity o NI marine energy to enter the SLM by 2020. \hile
realisation o marine energy resource potential is a strategic objectie or NI energy policy, we
understand it is unclear at this stage which marine technologies will be adopted and the extent which
total resource can in act be realised gien the construction risks and constraints around connections
to the transmission system.
S.3.4. Iuel price assumptions
ligure 5.8 shows the uel price assumptions assumed in the Baseline modelling. 1he central
projections are drawn rom a ariety o sources including DLCC`s Updated Lnergy Projections
,ULP, June 2010 Central price case.
0 200 400 600 800 1000 1200
Wind Onshore
Wind Offshore
Biomass
Landfill gas / AD / EfW etc.
Tidal
Hydro
PV
MWs

60

igvre :.: ve price a..vvptiov. ;200 price.)

ovrce: DCC , CP.
S.3.S. Carbon price assumptions
ligure 5.9 illustrates DLCC`s Central LUA carbon price assumptions and the Carbon Price Support
leel used in the modelling.
igvre :.: Carbov price .vpport a..vvptiov. ;_,tovve COe)

ovrce: DCC; M 1rea.vr,
0
100
200
300
400
500
600
700
2012 2014 2016 2018 2020 2022 2024 2026 2028 2030
Coal ( / tonne) Gas (p /therm)
Gas Oil ( / tonne) HFO ( / tonne)
uLCC C C C S

61

1he carbon price policy assumptions or the UK that were used in the modelling were current at the
time the modelling was completed.
S.4. Summary
1his section has set out the analytical ramework and assumptions used in the modelling. In the
section which ollows we proide results or the Baseline scenario ,i.e. a counteractual where the
RO continues to be the support regime or large scale renewable generation in the UK and small-
scale generation in NI,. \e then go on to assess the impact o alternatie LMR options and support
packages in the sections that ollow.

62

6. BASLLINL
6.J. Overview
1his section inestigates the expected consequences o continuing to support RLS in NI ia the
NIRO in its current orm. 1his Baseline` scenario assumes no changes are made to the current
structure o the RO other than those included in the RO ,Amendment, Order 2010 and the NIRO
,Amendment, Order 2010, and, as such, is the benchmark ,counteractual, against which other
support options are compared and ealuated. \e begin by setting out Base Case speciic
assumptions beore turning to the outputs generated by the model.
6.2. Baseline assumptions
lor the purposes o the analysis, the NIRO is deined to include the ollowing components and
assumptions:
x the existing list o NIRO-eligible technologies and banding assumptions,including recent
reisions to banding or small-scale generation,,
x NI`s obligation size increases to 6.30 by 2012 and remains at this leel until the NIRO end
date in 2033 ,we inestigate a sensitiity where the obligation is increased to the GB leel,,
x the headroom mechanism or the UK RO increases to 10 in 2011,12 as set out in the RO
,Amendment, Order,
x the buyout price is held constant in real terms or the duration o the modelling period, set
equal to the actual buyout price o _36.99 per ROC or 2010.
In addition to the aboe assumptions, the Baseline uses the ollowing assumption sets ,scenarios,
or uel and carbon prices and plant inestment:
x DLCC`s ULP central gas price projections,
x DLCC uel price projections,
x Carbon Price Support ,CPS, takes eect in NI rom 2013,
x DLCC`s central assumptions or LU Allowance ,LUA, carbon prices,
x SONi,LirGrid Generation Capacity Assessment or 2020.
As described in Section 2, in the initial phase o the modelling, SLM wholesale prices relect an
exogenous dispatchable and non-dispatchable all-island market capacity expansion plan. 1he output
o the RLS-generation iability modelling ,based on projected wholesale prices and NIRO reenue
streams, is used to update the NI RLS generation capacity expansion plan and correspondingly SLM
wholesale prices. 1he analysis is repeated or a number o iterations.

63

6.3. Results investment and plant mix
Cumulatie SLM generation capacity between 2012 and 2020 projected in the modelling under the
Baseline assumptions is illustrated in ligure 6.1.
igvre .1: a.eive capacit, iv tbe M

ovrce: irCria , Oî c CP.
ligure 6.2 shows NI renewable capacit, broken down by plant type ,the igure includes installed and
new-build capacity oer the modelling period,. As is clear rom the proile o inestment, the NI
capacity expansion plan assumed in the SONi,LirGrid Generation Capacity Statement or the all-
island market is relected in the modelling outputs. 1he rate o incremental annual inestment in all
NI RLS-generation technologies increases towards the end o the 2020 deployment period ,see
discussion below, in line with the SONi,LirGrid expansion plan.

0
2,000
4,000
6,000
8,000
10,000
12,000
14,000
16,000
18,000
2012 2013 2014 2015 2016 2017 2018 2019 2020
M
W
Gas DO Coal Peat HFO Renewables

64

D
t
v
1 W nl
CP
L C
8
S
C
C
igvre .2: ^ reverabe capacit, vi ba.eive

ligure 6.3 shows NI renewable geveratiov broken down by plant type ,the igure includes installed
and new-build capacity oer the modelling period,. Biomass makes a greater contribution to NI`s
RLS mix in generation terms ,compared to capacity, because o its higher load actor.
igvre .: ^ reverabe geveratiov vi ba.eive

D
t
v
P
1
L W
A u
8
C W
C W

65

1he capacity expansion plan assumes a large amount o onshore wind generation is added to the NI
system oer the next ew years. It is this increasing leel o onshore wind that is expected to be the
main contributor to achieing NI`s renewable electricity target. Alternatie renewable generation
projects to onshore wind are also assumed to be commissioned by 2020, including oshore wind
,600 M\,, tidal ,300 M\, and a large scale biomass plant ,200 M\,.
3
A small amount o
distributed small-scale biomass, anaerobic digestion and landill gas ired generation plant is assumed
to come on to the system by 2020.
6.4. Results wholesale prices
ligure 6.4 shows the time-weighted aerage wholesale price o electricity in the Baseline. 1he
wholesale price rises towards 2030 on the back o rising uel and carbon price assumptions.
igvre .1: 1ivereigbtea arerage M rboe.ae eectricit, price. ;rea) ;2010 price.)

ovrce: CP.
6.S. Results renewable generation support payments
ligure 6.5 shows the leel o support payments made to NI renewable generator technologies under
the Baseline scenario and support mechanism ,i.e. the NIRO,. 1he assumed payment leels are
shown or large scale plant ,i.e. greater than 5 M\, and like a PlI1 are proided in addition to
generator wholesale reenue streams rom the SLM.

3
As the Generation Capacity Statement notes it is unclear as to which tidal technology will be used to achiee this. lor
the purposes o the analysis tidal power is assumed to be generated by tidal stream using turbine technology rather than
tidal lagoons and barrages.
D
t

66

igvre .:: ^RO .vpport pa,vevt. ;rea) ;2010 price.)

ovrce: CP. , DCC, D1
6.6. Results cost benefit analysis
6.6.J. Consumer costs
1he cost to consumers o the NIRO can be measured in a number o ways, which capture costs
within a particular year or the sum oer a numbers o years. 1able 6.1 presents the cvvvatire cost o
the NIRO to NI consumers using a range o metrics to 2020, some including the cost o electricity
while others show only the cost o the NIRO to consumers.
1abe .1: Cvvvatire co.t. ^ cov.vver. vvaer tbe a.eive to 2020 ;2010 price.)
Metric Unit Cost to NI consumers
Co.t. of .vb.ia, ov,
Cumulatie consumer cost to 2020 _m 425.4
Cumulatie consumer cost ,_,M\h consumed, _,M\h 4.9
Co.t. of .vb.ia, ava epevaitvre ov rboe.ae eectricit,
Cumulatie consumer cost to 2020 _m 6,458.3
Cumulatie consumer cost ,_,M\h consumed, _,M\h 4.2
ovrce: CP.
1able 6.2 shows the avvva cost o the NIRO to NI consumers using a range o consumer metrics
or 2020. As with 1able 3.3 some metrics include the total cost o wholesale electricity in consumer
costs while others show only the cost o the NIRO to the NI consumer.

-
10
20
30
40
50
60
70
80
90
2012 2013 2014 2015 2016 2017 2018 2019 2020
/
M
W
h
Onshore Wind Offshore Wind Biomass Tidal

6

1abe .2: .vvva co.t. ^ cov.vver. vvaer tbe a.eive iv 2020 ;2010 price.)
Metric Unit Cost to NI consumers
Co.t. of .vb.ia, ov,
Consumer cost in 2020 _m 0.5
Consumer cost in 2020 ,_,M\h renewables generated, _,M\h 10.1
Consumer cost ,_,M\h consumed, _,M\h 6.9
Co.t. of .vb.ia, ava epevaitvre ov eectricit,
Consumer cost in 2020 _m 98.6
Consumer cost in 2020 ,_,M\h renewables generated, _,M\h 114.1
Consumer cost ,_,M\h consumed, _,M\h .9
ovrce: CP.
1he key points to note rom 1able 6.1 and 1able 6.2 are as ollows:
x 1he NIRO ,renewable generation support costs only, is projected to cost NI electricity
consumers around _0m annually by 2020.
x 1his is a cost o _6.90 per unit o total electricity consumed in the region in 2020 and a cost
o _10.10 per unit o renewable electricity generated.
x Cumulatiely, the cost o the NIRO or NI consumers is projected to cost around _425m to
2020 ,a cost o _4.90 per unit o electricity demand,.
x 1otal wholesale costs ,per unit o electricity consumed in 2020 including the cost o the
NIRO and SLM prices, are around _8 M\h.
x 1his compares to an aerage NI wholesale electricity cost or the period 2012-2020 o
around _4 per M\h.
1he result that oer 2G\ o renewable generation capacity can be supported or around _0m o
NI consumer subsidy on irst appearance is surprising. 1he eicacy o the NIRO ,in terms o NI
consumer cost, is drien by the structure o the scheme assumed in the modelling whereby NI
operates at a lower obligation leel than GB ,as illustrated in ligure 6.6 below,. Under the design o
the RO ,and the NI obligation assumption in the modelling, deployment o renewable generation in
NI is in part relected by demand or NI ROCs by GB suppliers. 1his is because the demand and
supply o ROCs operates on a UK-wide basis.

68

igvre .: ^ reverabe. obigatiov reatire to C vvaer a.eive a..vvptiov

ovrce: CP.
1he way the UK RO is currently designed implies the cost to NI consumers o the NIRO ,and
thereore supporting renewable generation, is capped - equal to the RO buyout price multiplied by
the size o the NIRO in ROCs.
But in the eent the headroom mechanism is triggered ,which gien projected renewable build rates
in the UK is almost certain ,were the scheme retained,, then:
x headroom increases the |K obigatiov ere to ensure there is at least 10 headroom` aboe
projected |K geveratiov in the obligation period, but
x the increase in the NI obligation is proportional to the region`s share o electricity supply in
the UK and the relatie obligation leels in NI and GB.
1his means i) the cost o the NIRO or NI consumers is somewhat diorced rom actual RLS-
generation in NI and ii) deployment o NI RLS-generation is linked to the demand or NI ROCs by
GB electricity suppliers.
ligure 6. shows the projected number o NI ROCs produced in the Baseline renewable generation
capacity expansion plan compared to the NI obligation size ,also in ROCs, with and without
headroom. It illustrates the projected low o ROC payments rom GB suppliers to NI generators
under the current design o the UK RO scheme.
1his result is important to note when comparing the change in costs arising rom support
mechanism support options. As well as changing the .trvctvre o the support mechanism or
renewable generation in NI, adoption o an alternatie scheme ,such as a CD lI1 o Premium lI1,
in the absence o counteracting measures, may also change the low o payments to NI generators,
L W
S
8
C
8 C
n l
8
C

69

including who pays or the enironmental premium` element o renewable generation inestment
in NI ,aboe the wholesale price o electricity set in the SLM,.
igvre .: Cbavge. iv ^ obigatiov ave to beaaroov

ovrce: CP.
6.6.2. Additional cost-benefit analysis metrics
1able 6.3 below compares a number o cost-beneit metrics or the Baseline scenario, including the
resource cost o NI renewable generation ,per unit o renewable electricity generated, and the dead-
weight loss ,oer-subsidy`, under current NIRO banding leels.
1abe .: Co.tbevefit vetric.
Metric Baseline

Resource cost o new plant built in 2020 ,_m, 594.6
Resource cost ,_,M\h renewables generated, in 2020 85.0

Dead-weight loss ,oer-subsidy`,,_m, in 2020 - 245.1
Dead-weight loss ,_,M\h renewables generated, in 2020 - 45.5
ovrce: CP.
1he analysis shows the resource cost ,generation, operating and capital costs, o new build NI
renewable generation under the Baseline expansion plan is projected to be around _0.6bn in 2020,
this is a resource cost, per M\h o renewable generation, o oer _ 85 per M\h in 2020.
1he deadweight costs ,oer-subsidy`, o the Baseline support regime ,the NIRO, is projected to be
around _245m by 2020, the dead-weight cost ,oer-subsidy, per unit o renewable electricity
generated in 2020 is _45 per M\h in the modelling.
0
2
4
6
8
10
12
2012 2013 2014 2015 2016 2017 2018 2019 2020
R
O
C
s

(
m
i
l
l
i
o
n
s
)
NI obligation NI obligation (with headroom) NI ROCs

0

1able 6.4 shows the impact on domestics consumer bills under the Baseline, including the aerage
impact on bills or period 2010 - 2020. 1he impact on the consumer bill is estimated as the change
in wholesale prices under the Baseline. 1he analysis assumes other price elements o the domestic
consumer bill ,or example, network charges, remain constant. 1he analysis also assumes aerage
annual domestic consumption remains constant.
1abe .1: vpact of a.eive ov ^ cov.vver bi.
Baseline change from 20J0
Dove.tic
2010 _496 -
2011 - 2015 _508 2.5
2015 - 2020 _523 5.4
2020 _533 .4
Aerage 2011 - 2020 _515 3.9
ovrce: CP.
1able 6.4 suggests NI consumer bills will rise towards the end o the decade on the basis o rising
wholesale energy costs in the SLM and rising costs o the NIRO. By way o context, the modelling
suggests the cost o support payments to renewable generators would comprise approximately _36
o the aerage annual NI domestic electricity bill by the year 2020.
In practice, while the unit cost o electricity is expected to rise towards 2020 this might be oset by
energy eiciency saings oer the period thus reducing household electricity consumption and, cetri.
pariba., NI household electricity bills. 1o illustrate this aect, 1able 6.5 shows the impacts on NI
consumer bills were a 1 year on year reduction in household electricity consumption obsered
oer the inestment period to 2020.
1abe .:: vpact of a.eive ov ^ cov.vver bi.
Baseline change from 20J0
Dove.tic
2010 _496 -
2011 - 2015 _498 0.4
2015 - 2020 _48 -1.8
2020 _48 -1.9
Aerage 2011 - 2020 _493 -0.
ovrce: CP.
1he analysis suggests that were a year on year NI electricity consumption eiciency rate o 1
obsered or the inestment period to 2020 this might mitigate the rise in the NI domestic
consumer bill caused by the rise in wholesale costs obsered in the modelling.

1

v
P
W
1
L W
8
C W
C W
6.6.3. 1echnology resource costs
ligure 6.8 illustrates the resources costs o the NI renewable capacity mix in 2020 in the Baseline.
1he analysis illustrates how more emerging RLS-generation technology groups, such as tidal and
oshore wind, orm a relatiely expensie component o NI`s renewable capacity mix compared to
more established mature` generation technologies such as onshore wind.
igvre .: 1ecbvoog, covtribvtiov co.t. vvaer tbe a.eive

ovrce: CP.
6.7. Summary
1he modelled capacity expansion plan under the NIRO deliers suicient inestment in renewable
generation to delier the region`s 40 per cent renewable electricity consumption target. 1his comes
at a cost to the NI electricity consumer.
\ithin the current design o the UK RO, the .vb.ia, cost o NI RLS-generation is not borne in ull
by NI consumers as deployment in the region is in part relected by demand or NI ROCs by GB
electricity suppliers. 1his proides or a relatiely high eicacy scheme, in terms o NI consumer
cost, because the costs o NI renewable generation are in part spread across UK consumers.
1he sections which ollow analyse alternatie support regimes and reorm options compared to the
Baseline scenario results.

2

7. CARBON PRICL SUPPOR1

| As agreed with DL1I and NIAUR, we hae not proided an analysis o the Carbon Price lloor
and, instead, the ocus o this report is on policy options or low carbon support [

3

8. ANALYSIS OI II1 OP1IONS
8.J. Overview
1his section inestigates the expected impacts and costs o Premium and CD lI1 reorm options,
relatie to the Baseline scenario.
8.2. Premium II1 policy
8.2.J. Assumptions
As described in Section 5, CLPA has deeloped a hypothetical PlI1 policy in order to project
deployment and costs to NI consumers under a PlI1 scheme designed to meet NI renewable
electricity targets ,and thereore the capacity expansion plan assumed in the wholesale market
dispatch modelling,.
1he approach adopted or setting the PlI1 leels is as ollows:
x Initial PlI1 leels or dierent technologies are deried by subtracting their leelised cost
rom the leelised cost o a CCG1 and rounding to the nearest _5M\.
x 1he model applies a degression rate to the PlI1 leels consistent with the projected all in
technology capital cost in the model.
x 1he PlI1 leles are ,iteratiely, increased upwards until the projected NI capacity expansion
plan is achieed.
4

luel price, generation capacity expansion plans and generator cost assumptions remain the same as
in the Baseline scenario. \e assume the PlI1 policy is implemented in 2014. Renewable generation
plant commissioned on or ater 2014 all under the PlI1 scheme. Plant that are commissioned
beore 2014 all under the NIRO.
ligure 8.1 shows the leel o PlI1 payments assumed in the modelling.

4
A similar assumption was adopted in the original LMR Impact Assessment modelling.

4

igvre .1 : P1 Poic, 2 .vpport pa,vevt ere.

8.2.2. Results investment and plant mix
By adjusting the renewable support leels, the PlI1 leels hae been designed to achiee the
capacity expansion plan assumed in the wholesale market dispatch modelling. ligure 8.2 shows the
trajectory o NI renewable generation. It illustrates the same proile o generation build as in the
Baseline ,see ligure 6.3,. lrom 2012 to 2013 renewable generation in NI is built under the NIRO,
the cost o which is then grandathered and added to the consumer cost o PlI1s paid to new build
generation built rom 2014 onwards.
igvre .2: Reverabe geveratiov ovtpvt vvaer a P1 poic,
0
20
40
60
80
100
120
2012 2013 2014 2015 2016 2017 2018 2019 2020
/
M
W
h
Onshore wind Offshore wind
Biomass (large dedicated) Biomas CHP
Landfill gas Tidal

D
t
v
P
1
L W
A u
8
C W
C W

5

8.2.3. Results consumer costs
1able 8.1 shows the change in NI consumer costs rom NI adopting a standalone PlI1 policy
relatie to the Baseline. 1he analysis suggests that under the modelled design o PlI1 policy that a
standalone NI PlI1 scheme would be comparatiely expensie relatie to the NIRO ,in its current
orm,. 1his is because the total cost o unding the inestment to meet NI`s renewable electricity
target would need to be recoered solely rom NI consumers through a ley on NI consumer bills
or through NI electricity suppliers.
1abe .1: Cbavge iv ^ cov.vver co.t. P1 poicie. iv 2020
Metric Unit Baseline PII1
Co.t. of .vb.ia,
Consumer cost in 2020 _m 0.5 353.6
Cbavge frov ba.eive _m n,a 283.1
Consumer cost
_,M\h 6.9 34.5
Cbavge frov ba.eive _,M\h n,a 2.6
Consumer cost in 2020 _m 98.6 1,081.8
Consumer cost
_,M\h .9 105.5
ovrce: CP.
1he high consumer cost o a PlI1 policy ,relatie to the Baseline, is not surprising, gien the
structure o the UK RO, the PlI1 modelling approach and the RLS-generation capacity expansion
plan projected or NI. NI is required to und the enironmental premium` o NI RLS-generation
wholesale costs on a standalone basis ,rather than the .vb.ia, co.t o renewable generation in the UK
being in part unded by GB supplier demand or NI ROCs,.
Points to note with regards the results presented in 1able 5.1 include:

6

x 1he PlI1 leels are set independently o the wholesale price. \hile they delier the
projected NI RLS generation expansion plan they are not set necessarily set at the minimum
eicient leel strictly required to acilitate build.
x Assuming PlI1s were ixed or dierent RLS generation technologies, the model results are
sensitie to the wholesale power price in the RLS-generation iability modelling and
projected generation costs.
1he impact on consumer bills under the PlI1 policy scenarios is inestigated as part o the
comparatie consumer cost analysis o lI1 options in Section 8.5.
8.3. CfD II1 policy
1his section inestigates the expected impacts and costs o a CD lI1 reorm option, relatie to the
Baseline scenario. 1he assumptions adopted in the modelling are summarised in the next sub-section
ollowed by a summary and description o the results. As with the Baseline, the analysis assumes
CPS takes eect in NI rom 2013.
8.3.J. Assumptions
\e hae made the ollowing assumptions or modelling CD lI1 payments:
x the contract is paid on metered generation ,i.e. it is not a classic CD whereby a ixed olume
o M\s is procured,,
x CD contract strike prices are set to coer the leelised ,LRMC, cost o renewable
generation plant in the SLM,
x as CD strike prices are set to coer leelised cost the modelling assumes a lower leel o
subsidy than under the NIRO and PlI1 policies,
x the CDs operate two-way` - i.e. i the wholesale price rises aboe the contract strike price
,set at leelised cost, the generator pays the dierence to the contract counterparty,
x as with a PlI1 scheme, the policy is assumed to be implemented in 2014 with the same
arrangements or grandathering the NIRO.
Gien the structure o the model ramework, a CD lI1 is assumed to operate in a similar way to a
lixed lI1 ,i.e. we assume generators bear no basis risk in the inestment analysis,. 1his is consistent
with how a CD payment is likely to be structured within the current SLM structure ,gross pool
system, and the current trading and settlement rules.
ligure 8.3 shows the strike` price payments made to large-scale renewable generators under the
cost and CD lI1 structure assumptions in the modelling.

igvre .: CfD 1 pa,vevt ere.

8.4. Results investment and plant mix
1he CD lI1 policy sets ixed tari leels ,CD strike prices`, to delier the capacity expansion
plan assumed in the SLM market dispatch modelling. 1he proportion o renewable technologies in
trajectory o deployment to 2020 ,see ligure 6.1, is thereore the same as the Baseline and PlI1
policy options relecting the projected build constraints , easible capacity expansion plan projected
or NI and the SLM.
8.4.J. Results consumer costs
1able 8.2 shows the change in NI consumer costs rom NI adopting a standalone CD lI1 policy
relatie to the Baseline.
Like an NI standalone PlI1 policy, the results suggest ,under the modelled design o CD lI1
policy, that a standalone NI scheme would be more expensie than the NIRO ,in its current orm,.
1he total cost o unding the generation inestment to meet NI`s renewable electricity target would
need to be recoered solely rom NI consumers, either through NI electricity suppliers or a ley on
NI electricity consumer bills.

0
50
100
150
200
250
300
2012 2013 2014 2015 2016 2017 2018 2019 2020
/
M
W
h
Onshore wind Offshore wind Biomass (dedicated)
Landfill gas Tidal

8

1abe .2: Cbavge iv ^ cov.vver co.t. CfD 1 poicie. iv 2020
Metric Unit Baseline CfD II1
Co.t. of .vb.ia,
Consumer cost
_,M\h 6.9 26.1
Consumer cost
_,M\h .9 9.1
ovrce: CP.
1he modelling suggests the total cost o deliering the RLS capacity plan in NI could be
comparatiely less under a CD lI1 policy than a PlI1 policy. 1his is drien by:
x CD lI1 payments ,strike` prices, being set at a more eicient leel to acilitate inestment
under the capacity expansion plan in the modelling, and
x assumed lower cost o capital under a CD lI1 policy relatie to the NIRO , PlI1 policies
where inestors continue to bear wholesale price risk.
1he leel o support payment costs can be considered relatiely eicient gien the certainty oer
current and uture generation costs in the modelling. In practice, lI1 strike prices might need to be
set much higher to support the inestment in capacity expansion plan adopted. 1his would lead to
higher NI consumer costs.
In the modelling, NI consumers are still projected to pay _19,M\h more or their wholesale
electricity in 2020 under a standalone CD lI1 policy relatie to the Baseline. \ere NI to adopt a
standalone CD lI1 support scheme, the modelling suggests NI`s policy makers would need to
consider careully the eicacy o awarding lI1 contracts to all the assumed projects in the capacity
expansion plan gien the signiicant consumer cost.

9

8.S. Consumer cost comparison
1he preious sections set out projected consumer costs o a standalone NI PlI1 policy and CD
lI1 policy under a projected NI generation capacity expansion plan. In this scenario, NI consumers
are assumed to bear the ull cost o renewable electricity generation in NI unlike the NIRO where
costs are in part unded by GB supplier demand or NI ROCs.
In this section we compare projected consumer costs in GB with those projected under the Baseline,
NI PlI1 and NI CD lI1 policy scenarios. \e then go on to assess NI consumer costs were an
lI1 policy part o a UK wide low-carbon generation payment cost recoery scheme. In this scenario
the costs o NI lI1 payments would be recoered rom UK electricity consumers.
8.S.J. Comparison of domestic consumer bills
1able 8.3 compares annual NI domestic consumer electricity bills rom our modelling and GB
domestic consumer bills in DLCC`s LMR IA ,again, we show the results or the Contracts or
Dierence policy scenario,. 1he table shows consumer bill impacts where no electricity eiciency
saings are assumed and where 1 year on year eiciency saings are assumed ,consistent with the
bill impact assessment methodology used in Section 6,.
1abe .: vpact of atervatire .vpport optiov. ov ^ cov.vver bi.
Sample NI Baseline NI PII1 NI CfD II1 GB bill
Dove.tic ;vo ^ efficievc, .arivg.)
2010 _496 _496 _496 _490
2011 - 2015 _508 _509 _509 _490
2015 - 2020 _523 _56 _55 _504
Dove.tic ;1 ,earov,ear ^ efficievc, .arivg.)
2010 _496 _496 _496 _490
2011 - 2015 _498 _499 _499 _490
2015 - 2020 _48 _536 _519 _504
ovrce: CP., DCC
1he modelling suggests that domestic electricity bills or the period to 2020 could on aerage be
higher or NI compared to GB ,this result is o course dependent on the generation capacity
expansion plan and generation costs assumed in the modelling and the respectie year on year
eiciency saings in electricity consumption in the two regions o the UK,.
1he standalone lI1 policies ,where the total cost o unding inestment to meet NI`s renewable
electricity target is assumed to be recoered solely rom NI consumers, highlight clearly the balance
NI`s policy makers ace between meeting aspirations or renewable generation in the region and
uture impacts on consumer electricity bill costs.

80

0.0
20.0
40.0
60.0
80.0
100.0
120.0
Baseline NI PFIT NI CfD UK FIT
/
M
W
h
CfD support payments
Grandfathered NIRO
Wholesale price
8.S.2. NI consumer costs under a UK support mechanism
In order to derie an estimate o NI consumer costs were a UK wide low-carbon generation
payment cost recoery scheme adopted ,whereby the costs o NI lI1s are recoered rom electricity
consumers across the whole o the UK, we hae assumed:
x Grandathered ,legacy, NIRO costs are recoered rom the NI consumer ,across the
electricity demand base in the region,.
x NI lI1 costs in 2020 are spread proportiova, across projected UK electricity demand
,sourced rom DLCC`s Updated Lnergy Projections,.
x 1hese are added to net` low carbon payments to GB generators under the Contract or
Dierence LMR policy ,also in 2020,.
5

x 1he enironmental premium` consumer cost per unit o electricity demand ,_,M\h, is
then added to the projected wholesale electricity price in NI.
x lor the purposes o modelling a socialised lI1 scheme, we assume NI CD lI1 costs are
added to GB net support payments.
1he results are summarised in ligure 8.4 below.
igvre .1: ôrtberv reava rboe.ae price. ava .vpport pa,vevt co.t. vvaer poic, .cevario. iv 2020

ovrce: DCC
ligure 8.4 illustrates a number o important points. lirst, an NI PlI1 is likely to be the more
expensie o the NI standalone lI1 policy options. As importantly, een with a UK wide cost
recoery arrangement ,i.e. a socialised cost, NI consumers may still be worse o ,relatie to the
Baseline scenario, under a CD lI1 than they are at present in the absence o an arrangement which

5
Annual GB net` payments to low carbon generators or 2020 are sourced rom DLCC`s Updated Impact Assessment
analysis or the Contracts or Dierence Llectricity Market Reorm policy modelling scenario ,c. _2.9bn,.

81

might replicate the beneits o the existing lower obligation leel under the NIRO. Len so, the
modelling suggests NI consumers might still beneit signiicantly by participation in a UK wide cost
recoery scheme ,the modelling suggests support payment costs might be reduced by around _14 -
_15 ,M\h by 2020, were costs recoered een proportiova, across annual UK electricity
consumption.
\hile the modelling results are inormatie, there are reasons to be cautious about the conclusions.
Under the RO headroom mechanism calculation, suppliers` obligations are set in proportion to UK
regional shares o electricity sales adjusted or the size o the UK obligation and the relatie regional
obligation leels. 1his ineitably leads to demand or NI ROCs rom GB suppliers ,our modelling
suggests as much as 40-50 per cent in some years, and, thereore, support payments to NI
renewables paid by GB consumers. In other words, the consumer cost o support payments under
the RO ,per unit o total electricity consumed in the region, is lower or the NI consumer than or
the GB consumer ,as GB suppliers hae a higher obligation leel relatie to oerall obligation size,.
Unless some new mechanism is put in place to replicate the beneit o this lower obligation leel or
NI, remoing the NIRO would increase the proportion o the total cost o UK low-carbon
generation paid by NI consumers. In that situation, NI consumers would be worse o vve.. the
decrease in total UK consumer costs rom introducing a UK CD lI1 ,relatie to continuation o
the RO, compensated or this increase. 1he modelling suggests that it would compensate to some
extent but not ully.
ligure 8.5 shows DLCC`s modelled GB CD support payments in the \hite Paper Impact
Assessment, both relatie to DLCC`s Baseline scenario ,continuation o the RO, and under high
and low ossil uel price projection scenarios ,net CD support payments shown in Part B are used
to derie the indicatie UK lI1 cost or NI consumers presented in ligure 8.5,.
igvre .:: Co.t. of .vpport for or carbov vecbavi.v. ;Creat ritaiv vvaer MR reforv optiov.)

Part A: Costs o support or support mechanisms Part B: Cost o support or CD lI1 low,high uel prices

ovrce: DCC

82

GB`s proposed introduction o a CD lI1 may lead to quite signiicant reductions in oerall UK
support payment costs relatie to the RO both beore and ater 2020. 1hereore, een in the
absence o an arrangement similar to the lower obligation under the NIRO, NI consumers may not
be signiicantly worse o under the proposed new arrangements. Ioweer, this depends on a
number o key assumptions, particularly uture deelopments in ossil uel prices ,which drie uture
wholesale prices, and low carbon generation costs ,which drie CD strike price leels, amongst
many other market uncertainties. In the eent rboe.ae price. in the SLM were also higher than those
projected in the modelling then there would still be the potential or much higher total costs to
consumers in NI oer the period to 2020 and beyond, een where support payments are constrained
under the CD design in the UK.
6

Our modelling o NI support payments and prices in the SLM ,together with DLCC`s modelling o
GB support payments and wholesale prices in GB, suggests there are scenarios where NI consumers
might not be made comparatiely worse o by the proposed new arrangements. Ioweer, the
assumptions that lead to this conclusion need to be inestigated urther as this is an area o both
uncertainty and importance to the oerall impacts o any new arrangements on NI consumers. \e
would recommend, thereore, that DL1I and NIAUR complete urther analysis as part o any uture
design o reorms to the UK`s low-carbon generation support mechanisms.

8.S.3. Additional cost-benefit analysis metrics
In this section we illustrate the results rom additional cost-beneit analysis metrics, including the
resource costs o renewable generation under the NI lI1 policy scenarios and the deadweight loss
,oer-subsidy`, obsered in the modelling.
Re.ovrce co.t.
ligure 8.6 shows annual resource costs oer time under the dierent NI policy scenarios, including
the Baseline, PlI1 and CD lI1 scenarios.

6
One reason projected wholesale ,and inal retail, prices might by higher in NI ,compared to GB, is higher
transmission costs ,per customer, in the SLM system, or example, rom constraint payments.

83

0
100
200
300
400
500
600
700
2012 2013 2014 2015 2016 2017 2018 2019 2020
m
NIRO
PFIT
CfD
v
P
W
1
L W
8
C W
C W
igvre .: Re.ovrce Co.t. vvaer aifferevt poic, .cevario. ;_v)

1he key points to note are as ollows:
x resource costs are the same between the policy scenarios. 1his is because leelised costs do
not change between scenarios, and
x the capacity expansion plan ,i.e. the NI renewable generation mix, also remains constant
between the policy scenarios.
ligure 8. shows, or the CD scenario, the breakdown in the annual resource costs between the
dierent renewable generation technologies as compared to the capacity expansion plan oer the
modelling period ,i.e. to 2020,.
igvre .: reaaorv of re.ovrce co.t. ava tbe capacit, epav.iov pav vvaer ^ CfD .cevario
.vvva re.ovrce co.t.

84

D
t
v
1 W nl
CP
L C
8
S
C
C
.vvva capacit, epav.iov

1he key points to note rom ligure 8. are:
x up to 201, the increase in resource costs is drien by inestment in onshore wind
generation projects, which has a relatiely low leelised cost in comparison to other less
established renewable technologies, but
x during the period 2018-2020 there is a signiicant increase in oshore wind, biomass and
tidal generation, all o which hae relatiely high resources costs, which dries the increase in
resource costs towards the end o 2020,
x the contribution o emerging generation technologies ,or example, oshore wind and tidal,
to total annual resource costs is relatiely greater than their contribution to the deployment
o renewable generation in capacity terms.
Deaareigbt o.. , profit for geverator.
ligure 8.8 below shows deadweight loss under the dierent policy scenarios. Deadweight loss
measures the additional subsidy that is proided to the generators oer and aboe what is required
or new-build iability, and so can be described as oer-subsidy`. lor example, i the SLM price is
_60,M\h and the subsidy proides an additional _40M\h, then total reenue is _100,M\h,
howeer, i the leelised cost or that technology is _80,M\h, then the generators receies
_20,M\h more than is strictly required in order to build.
1hereore deadweight loss is equialent to supernormal proits or generators ,aboe the normal`
amount o proit that would be required to meet the inestor`s cost o capital,.

85

0
50
100
150
200
250
2012 2013 2014 2015 2016 2017 2018 2019 2020
m
NIRO PFIT CfD
igvre .: Deaareigbt o.. vvaer poic, .cevario. ;_v)

Key points to note rom ligure 8.8 are as ollows:
x 1he illustration shows that deadweight loss ,oer-subsidy`, aries considerably between the
dierent scenarios ,and so is a drier o consumer cost,.
x 1he NIRO experiences the highest deadweight loss ,at oer _200m,, ollowed by modelled
NI PlI1 ,c. _125m,.
x 1he CD scenario, by the nature o its design in the modelling, has an almost zero
deadweight loss.
x 1his is because it gies the generator the dierence between the SLM price and the strike
price` ,set at the generator leelised cost,.

It should be noted that although the CD scenario shows irtually ery little deadweight loss, this is a
theoretical outcome as it assumes that the policymaker can set the strike price` at the exact leel o
the leelised cost or each technology, both now and in uture years. 1hereore, practically, it is
possible that the CD ,i implemented, might hae a greater deadweight loss than that shown by the
modelling in ligure 8.8.
8.6. Summary
1his section has summarised modelling results or an NI CD and PlI1 lI1 policy. 1he analysis
shows an NI standalone PlI1 and CD lI1 scheme is likely to be relatiely expensie compared to

1he CD scenario does not hae exactly zero deadweight loss because o RO grandather payments which are
assumed to occur or generators built in 2012 and 2013 and the contract payments hae been rounded to the nearest
_10 ,as described in Section 4,.

86

the NIRO in its current orm. 1he model results suggest that een with a UK wide support cost
recoery arrangement ,i.e. a socialised cost,, NI consumers are still likely to be worse o ,relatie to
the NIRO, than they are at present in the absence o an arrangement which would replicate the
beneits o the existing lower obligation leel under the NIRO ,although NI consumers may still
beneit signiicantly rom such an arrangement,. 1he modelling also suggests that the total cost o
deliering the RLS capacity expansion plan in NI could be comparatiely less under a CD lI1
policy than a PlI1 policy. 1his is drien by lower net support payments in the modelling and an
assumption o lower cost o capital or certain renewable generation technologies.
1he modelling helps conirm the conclusion made in Section 4 that an NI CD lI1 policy is likely
to hae adantages oer an NI PlI1 policy.

8

9. SMALL-SCALL GLNLRA1ION
1his section ocuses on the costs o NI small-scale generation, which we hae deined as generation
with a capacity up to 5M\. 1he consumer costs presented are already included in the analysis
presented in the preious sections. 1his section, thereore, simply presents consumer costs o small-
scale generation on a standalone basis.
9.J. Small-scale II1 design
As described in Section 5, or both the PlI1 and CD lI1 policy scenarios, the modelling assumes
NI small-scale generators receie a top-up` payment oer and aboe the current NIL export tari
or classes o small-scale generator who currently qualiy or the scheme. Small-scale generator
classes who currently do not qualiy or the NIL export tari instead receie a top-up` payment
aboe the wholesale price in the SLM.
1able 9.1 shows the export taris assumed in the modelling.
1abe .1: va.cae geverator eport tariff. ;!b)
Period Night Peak Day
Oct - Mar Apr - Sep Apr - Sep Oct - Mar Oct Mar
100k\ 5.49p
100 k\ - 1M\ 3.98p 3.85 .36 5.63
ovrce: ^
Small-scale lI1s are paid on metered generation and are unded on an NI standalone basis ,unlike
under the NIRO where NI ROCs ,een or small-scale generators can be sold to GB suppliers,. 1he
range o lI1 payment leels ,and tari degression rates, assumed in both o the lI1 policy
scenarios are illustrated in 1able 9.2.

88

igvre .2: va.cae geveratiov 1 ere.
Sample
PII1 scenario (1op-up J)
1ariff (/MWh) Degression
Onshore wind 5k\ 240 5
Onshore wind 5M\ 40 3
Biomass 500k\ 80 -
Biomass 5M\ 80 -
AD 500k\ 80 1
Land ill gas 5M\ 25 1
Iydro 50k\ 120 -
Iydro 500k\ 120 -
Iydro 5M\ 40 -
PV 5k\ 200 6
PV 50k\ 200 6
PV 500k\ 200 6
PV 5M\ 200 6
ovrce: CP.
9.2. Generation mix
ligure 9.1 shows the total NI small-scale generation mix in the modelled capacity expansion plan.
Consistent with the LirGrid , SONi Generation Capacity Statement, by 2020 the illustration shows
an NI small-scale generation mix which includes biomass, anaerobic digestion, small-scale wind and
land-ill gas generation.

89

-
50,000
100,000
150,000
200,000
250,000
300,000
350,000
400,000
450,000
2012 2013 2014 2015 2016 2017 2018 2019 2020
M
W
h
Onshore Wind Biomass Energy from Waste & Anaerobic Digestion Hydro PV
igvre .1: va.cae geveratiov vi

9.3. Consumer costs
1able 9.3 proides an estimate o the consumer subsidy cost o supporting small-scale generation
under a PlI1 and CD policy scenario.
1abe .: Cov.vver co.t. of .va.cae geveratiov
Metric Unit Small-scale II1
Consumer cost _m 52.
Consumer cost ,_,M\h renewables generated, _,M\h 125.
Consumer cost ,_,M\h demand, _,M\h 5.1
ovrce: CP.
1he modelling suggests that a standalone NI small-scale generation lI1 scheme might cost around
_50m by 2020. In the modelling, renewable electricity rom small-scale generation ,sub-5M\, is
projected to be around 0.41\h by 2020 - around 4 o electricity demand in the region. 1he leel
o consumer costs would be lower in any year i the projected take-up o small-scale generation in
the modelled capacity expansion plan were not realised.
1he contribution o small-scale generation to total consumer costs under the PlI1 and CD lI1
policy scenarios is illustrated in 1able 9.4. 1he analysis suggests the take-up o small-scale generation
could contribute around 14-18 o the oerall consumer subsidy cost by around 2020 depending on
the design o NI lI1 scheme adopted.

90

1abe .1: Cov.vver co.t. of .va.cae geveratiov
Metric Unit PII1 CfD II1
Consumer cost ,total, _m 353.6 26.
Consumer cost ,small-scale generation, _m 52.3 52.3
Small-scale generation costs o total consumer costs 15 19
ovrce: CP.
9.4. Implementation issues
Under more complex lI1 design structures or large-scale generation, or example a CD lI1, there
is also a policy implementation issue,risk to consider. Support or small-scale generation may
require the operation o two alternatie lI1 design structures in Northern Ireland ,one or large-
scale and one or small-scale generation, like currently in GB where the RO operates alongside the
small-scale generation lI1 scheme.
A CD lI1 in particular, is unlikely to be appropriate or small-scale generators who operate outside
o the SLM. In this case ,like in the modelling, some orm o ixed payment would need to operate
alongside the CD payments or large scale generation. 1his could create additional administration
costs and operational requirements or Northern Ireland policy makers and goernment.
An obious alternatie to a standalone NI small-scale generation lI1 scheme would be NI
participation in the current GB small-scale lI1 scheme. 1he clear adantage with this approach is it
would reduce implementation risks and administration costs. 1he main disadantage is it may lead to
reduced NI policy discretion o the rates set or the region`s small-scale generators.
9.S. Conclusions on small-scale generation
\here a UK-wide CD lI1 is adopted or large-scale generation, it would seem sensible to adopt a
UK-wide scheme or small-scale generation. It should be noted, howeer, that this scheme is
currently under reiew, as such, it is probably better to wait until the results o the reiew are known,
beore taking inal decision.
NI participation in the GB scheme would also need the ollowing questions to be considered:
x \hat should NI contribute to the spending enelope o the scheme
x Should the spending enelope be expanded with NI participation
x \hat discretion would NI`s policy makers hae to set tari rates

91

J0. SLNSI1IVI1ILS
1his section tests the sensitiity o the modelling outputs to key uncertainties and assumptions. It
also presents results rom an additional scenario DL1I and NIAUR hae requested we explore in
the modelling - continuation o the NIRO beyond 201.
J0.J. Continuation of the NIRO beyond 20J7
DL1I and NIAUR hae also requested that we explore in the modelling an additional scenario
where the NIRO would continue beyond 201. Out o all the policy options this is the most diicult
to assess ,quantitatiely, gien uture consumer costs depend on a number o policy assumptions.
lor example, at what leel would be NI`s obligation be set going orward, and how would the
scheme be operated and administered without ongoing GB participation
1o explore this option, we hae modelled a scenario ,NIRO PlI1`, where the NIRO remains
open operating essentially as Premium lI1 scheme with the costs recoered solely rom NI
consumers. 1he tari leels oered under the scheme approximate the banding leels currently
oered to NI generators in the NIRO.
\hile not a direct comparison to the operation o the NIRO the modelling results proide an
illustration o what it might cost Northern Ireland consumes to operate a standalone scheme based
on similar banding leels as under the NIRO.
1he results are presented in 1able 10.1.
1abe 10.1: Cov.vver co.t. ;covtivvatiov of tbe ^RO)
Metric Unit NIRO PII1 NI PII1 NI CfD II1
Consumer cost _m 392. 353.6 268.0
Consumer cost ,electricity consumer, _,M\h 38.3 34.5 26.1
Cov.vver co.t. ivcvaivg .vpport pa,vevt. ava tota co.t of eectricit, ;iv 2020)
Consumer cost _m 1,120.9 1,081.8 996.1
ovrce: CP.
1he analysis suggests that the continuation o the NIRO on a standalone basis - like the other
support payment reorm options - could be a relatiely high NI consumer cost scenario. 1he NIRO
in its current orm proides relatiely generous support payment leels to NI generators. As the
majority o NI renewable deployment ,particularly oshore, is expected to take place after 201, NI
consumers might be expected to und the costs o the scheme on a standalone basis were the
scheme operated independently o GB, resulting in high consumer costs.

92

D
t
C L
1he diiculty is the inherent uncertainty o what Northern Ireland consumers might need to und
under a reormed NIRO. \hat the modelling illustrates is that were Northern Ireland to pursue the
option o retaining the NIRO, the existing concessionary obligation leel ,and thereore eicient
consumer cost o the NIRO, should not necessarily be taken as a guide to suggest the option o
retaining the NIRO would necessarily lead to a reduction in projected consumer costs relatie to
alternatie options such as lI1 policies. Indeed, there are scenarios where it could potentially result
in higher consumer costs.
1here is an additional issue with the NIRO, which is that it closes in 2033 rather than 203 as
elsewhere in the UK. Unless there is new legislation to extend this, applications accepted ater April
2013 would receie less than 20 years o support. 1his may well not be material, but needs to be
considered. It may be appropriate to look into extending the lie o the NIRO to 203 to maintain
inestor conidence, irrespectie o what choices are made about new support mechanisms in GB or
NI.
J0.2. Low-gas price sensitivity
1his section tests the sensitiity o consumer costs and the uptake o renewable generation under
low gas price projections.
Gas prices are a strong drier o electricity prices in the SLM and thereore reenues and returns
receied by renewable generators. lor the gas price sensitiity we hae kept lI1 leels the same as
or the central gas price scenario. 1he result is the electricity price is lower but ixed lI1 leels ,or
example, PlI1 leels, remain unchanged.
ligure 10.1 shows the time-weighted wholesale power price under the low gas price sensitiity
relatie to the central gas price projection used in the modelling in the preious sections. 1he
illustration shows that lower prices lead to lower wholesale prices in the SLM.
igvre 10.1: 1ivereigbtea rboe.ae eectricit, price.

93

ligure 10.1 shows modelled wholesale power prices ,under both a central and low gas price
projection, where renewable generation uptake in NI ollows the projected renewable capacity
expansion plan in the SONi,LirGrid Generation Capacity Statement.
ligure 10.2 shows the projected uptake o renewable generation under an NI PlI1, an NI CD lI1
and the NIRO with projected wholesale prices under the low gas price projection. 1his is compared
to the projected expansion plan in the baseline modelling ,Sections 6 to 9,.
igvre 10.2: Pro;ectea vptae vvaer or pro;ectea ga. price.

lor a number o scenarios, ligure 10.2 shows projected uptake is lower than the projected
expansion plan in the baseline modelling. In the case o the NI PlI1, this is due to non-iability o
tidal projects at the ixed PlI1 leels and projected wholesale power prices under low projected gas
prices. Similarly in the case o the NIRO, this is also due to non-iability o tidal projects in 2018
and small-scale commercial onshore wind projects ,2012-2014, that receie the SLM price rather
than a ixed export tari.
1he analysis suggests a CD lI1 o all the modelled options is most robust to long-term
uncertainties ,such as low gas prices, as CD payments rise with lower SLM wholesale electricity
prices ,and so deliers the capacity expansion plan in the baseline modelling,, similarly support
payments would all i wholesale prices were higher than projected ,or example, i wholesale gas
prices were higher than the central price projection,.
lor each scenario, we hae updated the wholesale power price projections in the modelling to relect
indiidual expansion plans illustrated in ligure 10.2. 1able 10.2 shows outturn consumer costs or
each scenario - the CD lI1 in this case has the highest consumer costs because the scenario has
the highest projected build ,including tidal,.

0
1
2
3
4
5
6
7
8
2012 2013 2014 2015 2016 2017 2018 2019 2020
T
W
h
Projected expansion plan PFIT CfD NIRO

94

D
t
v
1 W nl
CP
L C
8
S
C
C
1abe 10.2: Cov.vver co.t. - or ga. price .ev.itirit,
Metric Unit NIRO NI PII1 NI CfD II1
Consumer cost _m 0.5 319.9 33.1
Cov.vver co.t. ivcvaivg .vpport pa,vevt. ava tota co.t of eectricit, ;iv 2020)
Consumer cost _m 583.2 832. 885.8
ovrce: CP.
J0.3. Low marine / offshore wind deployment
1his section tests the sensitiity o consumer costs under a scenario where there is low marine
,oshore wind generation deployment scenario replaced by higher onshore wind deployment.
ligure 10.3 shows the NI renewable capacit, broken down by plant type ,the igure includes installed
and new-build capacity oer the modelling period, under this alternatie scenario.
igvre 10.: Capacit, epav.iov pav vvaer or

ligure 10.4 compares consumer costs or the NI CD scenario under baseline renewable generation
deployment and a scenario where there is lower marine , oshore wind deployment displaced by
higher onshore wind deployment.

95

0.0
20.0
40.0
60.0
80.0
100.0
120.0
NI CfD (low marine /
offshore wind)
NI CfD (projected
capacity expansion)
/
M
W
h
Support payments (/MWh)
Wholesale price (/MWh)
Marine / offshore consumer premium
igvre 10.1: Cov.vver co.t. vvaer or varive , off.bore riva aepo,vevt ;CfD .cevario)

ligure 10.4 shows the higher resource, and thereore consumer cost, o a modelling scenario with
more marine , oshore generation deployment ,an oshore consumer premium`,. In comparison,
under a UK-wide cost recoery scheme, the cost to ôrtberv reava cov.vver. o low-carbon
generation lI1 payments is much less sensitie to the generation mix deployed in Northern Ireland,
as a UK-wide cost recoery scheme ,similar to the NIRO currently, means consumer costs are
drien more by deployment across the UK as a whole.
J0.4. Summary
A CD lI1 o all the modelled options is most robust to long-term uncertainties ,such as low gas
prices, as CD payments rise with lower SLM wholesale electricity prices, similarly they all i
wholesale prices are higher than projected.
A scenario where there is low marine ,oshore wind generation deployment scenario replaced by
higher onshore wind deployment illustrates that the modelling results rom preious sections
contain a marine and oshore consumer premium`.
In the eent that oshore wind and tidal deployment was lower than projected, an NI standalone
scheme might result in lower consumer costs compared to NI`s participation in a UK-wide scheme
as GB low-carbon payments are also expected to rise with the deployment o large olumes o low-
carbon generation such as oshore wind.
Ioweer, a UK-wide CD lI1 spreads consumer costs, and might proide Northern Ireland with
greater lexibility to balance enironmental objecties and aordability constraints under dierent
scenarios o renewables deployment in the region.

96

JJ. CONCLUSIONS
Iaing considered careully the qualitatie and quantitatie analysis in this report, o the aailable
options, we beliee that neither a continuation o the NIRO nor the introduction o a PlI1, oer
workable or desirable support mechanisms or NI. 1he choice is one o applying DLCC`s CD lI1
proposals to the SLM, or introducing a RLlI1 type - lixed lI1 - model to NI.
1he main dierences between the two approaches are the lows o subsidy ,in the ormer it is direct
to the generator, rather than through a supplier, and the institutional requirements necessary to
support each.
In the highly illiquid GB BL11A market, the proposed CD lI1 will need to be struck against an
index, giing rise to concerns about basis risks ,i.e. not being able to achiee the index,. 1ogether
with the potential balancing costs or non-dispatchable renewables and the need or physical o-
take to receie payments, this could mean that generators continue to seek to access markets
through a PPA. \hilst not haing been designed or the SLM, the particular eatures o a CD lI1
mean that many o the problems identiied in the GB market are unlikely to bite within the SLM.
1hus, in many ways, it might be expected that the CD lI1 would work better in the context o NI
and the SLM than in GB.
Against this, the RLlI1 is a proen approach that carries less risk o applying something that was
designed or another market and which will need to be adapted or a new one. 1he associated costs
o introducing a RLlI1 approach would likely be lower - as there are already existing contracts
rom which to work - than or CD lI1s, or which contracts would need to be speciically written
or the SLM.
A RLlI1 type approach is also less administratiely burdensome and costly rom an institutional
perspectie, in that the public sector is not a contracting counterparty - which is likely to be the case
or the CD lI1s. Ioweer, it requires generators to contract directly with suppliers rather than
operating directly within the market.
1here are pros and cons or both options, and they need to be examined in much greater detail
beore a inal iew is arried at.
As NI is part o a UK-wide renewable energy obligation we recommend that DL1I seeks to
socialise the costs o the support regime across the UK, unless there is a desire to ocus on lower
cost renewables technologies, which might reduce the costs to NI customers under a stand-alone
scheme.
A UK-wide approach would spread consumer costs, and may proide NI with greater lexibility to
balance enironmental objecties and the region`s aordability constraints under dierent scenarios
o renewables deployment.

9

\e recommend DL1I and NIAUR make a clear statement to the NI market o the uture direction
electricity policy in the region so as to aoid the risk o an inestment hiatus during a key inestment
period, potentially according to the ollowing:
x Like the RO in GB, the NIRO will remain open to new accreditations until 201 ater which
the scheme will be grandathered.
x In terms o timing or joining a new scheme, NI would plan or a similar implementation
date as the rest o the UK ,i.e. 2014,.
x Ioweer, the inal implementation date will need to be conirmed ollowing a clearer
decision on the adopted option ,s, or market integration in early 2012.
In early 2012, were signiicant reisions required or evri.agea to it with uture Luropean market
requirements, the timeline or implementing a new policy would need to be reisited. In the interim,
there is detailed analysis to examine the impacts on consumer costs urther, and how a lI1
approach might be designed or the SLM beore and ater integration with neighbouring markets.
1o support the transition to the new arrangements, the option o extending the lie o the NIRO to
203 should also be considered.

98

APPLNDIX A: OPLRA1ION OI RLNLWABLL LNLRGY ILLD-IN 1ARIII (RLII1)
Introduction
1he Republic o Ireland`s existing renewable generation support mechanism, the RLlI1, is a orm
o Contract or Dierence ,CD, leed-in 1ari ,lI1,. 1his annex looks at how the RLlI1 works,
and at some o the lessons or any possible introduction o a Northern Ireland CD lI1 in the
Single Llectricity Market ,SLM,.
RLII1 contract design and interaction with SLM
acgrovva
1he RLlI1 was launched in 2005 with the aim o stimulating suicient capacity to meet Ireland`s
2010 renewable energy targets. 1he RLlI1 applies to new projects only. In addition, in order to
ensure that projects qualiying or support were in a position to build within a reasonable timerame,
there is a requirement to hae a alid grid connection oer in place.
1he scheme guarantees a minimum price or renewable electricity to inestors. lor historic reasons,
the money does not low directly to renewable generators, but instead to supply companies. 1hey
enter into a PPA with the generator, typically or 15 years, which speciies the amount that the
supplier will pay the generator or each unit o electricity produced.
Proce..
1he irst stage in the process is that the generator applies or a letter o oer` rom the
Department o Communications, Lnergy and Natural Resources ,DCLNR,. As noted aboe, to get
the letter they must hae must hae planning permission and a grid connection oer or their
project. 1he letter o oer conirms to any electricity supplier ,licensed to supply in the Republic o
Ireland, that in return or entering into a PPA with the generator or 15 years they will receie a
balancing payment` in accordance with the terms o the RLlI1 scheme. Suppliers recoer the
balancing payments through the PSO mechanism.
8

Covtract ae.igv ava .vppier revvveratiov
1he irst phase o the RLlI1 ,RLlI1 I, was announced in 2006 and proides support or onshore
wind generation, hydro and biomass or up to 15 years. 1he leel o support increases with
inlation
9
i inlation is positie. 1he second phase, RLlI1 II, supports renewable electricity rom
Anaerobic Digestion, CIP, wae , tidal energy and oshore wind. 1he tables below show the
current reerence tari leels.

8
IPA Lnergy and \ater Consulting ,2008,: Innoatie Llectricity Markets to Incorporate Variable Production`
9
CPI

99

1abe ..1: Referevce price. for R1
1echnology band 20J0 20JJ
R1
Large \ind t66.353,M\h t66.353,M\h
Small \ind t68.681,M\h t68.681,M\h
Iydro t83.814,M\h t83.814,M\h
Landill t81.486,M\h t81.486,M\h
Biomass t83.814,M\h t83.814,M\h
R1
ôt ivea to CP
Ocean energy t220,M\h
Oshore wind t140,M\h
ivea to CP
AD CIP 500 k\ t150,M\h
AD CIP 500 k\ t130,M\h
AD ,non CIP, 500k\ t110,M\h
AD ,non CIP, 500k\ t100,M\h
Biomass CIP 1500k\ t140,M\h
Biomass CIP 1500k\ t120,M\h
ovrce: DC^R
Under the RLlI1 there are three streams o compensation, accruing to the supplier:
x lixed payment ,l, - independent o the market price o electricity and set equal to 15 o
the support per M\h gien to large onshore wind installations. 1his is to coer balancing
costs associated with contracting with undispatchable generators.
x Market equalisation ,ML, payment - 1his is an opportunity cost payment to relect any extra
cost o contracting with RLlI1 generators relatie to the price o energy rom the SLM. It is
equal to the dierence between the RLlI1 tari leel or onshore wind and the wholesale
price, or zero i the wholesale price is higher.
x 1echnology dierence ,1D, payment - paid in addition to the reerence price or all
renewables other than large scale wind. 1his is to promote diersity in renewables.
As illustrated below, the market equalisation ,opportunity cost`, payment to the supplier operates
like a one-way CD.

100

igvre ..1: Overa, CfD varet eqvai.atiov pa,vevt

1he total yearly RLlI1 payment paid to a supplier who has entered into a generator using
technology r that generates e
r
M\h in that year is:

1he RLlI1 balancing and technology dierence payments are on metered generation, or botb in-
market and out-o-market generation, and are gien in ex-ante estimates and reconciled ex-post with
known alues through the Public Serice Obligation ,PSO, ley ,known as the R-actor`,.
1he sections which ollow:
x set out in detail how the technology dierence payment and opportunity cost payment
interact with the wholesale electricity price in the SLM,
x proide a couple o worked examples ,with reerence to onshore and oshore wind, o how
the RLlI1 payments are calculated or suppliers.
Maret eqvai.atiov ;opportvvit, co.t) pa,vevt.
lor ivvaret RLlI1,
80
the opportunity cost payment is calculated as the dierence between the total
reenues receied rom the market, ersus the total cost o purchasing metered energy rom the
generator, based on the onshore wind RLlI1 reerence price or the releant PSO period. \here
costs are greater than reenues, the supplier is entitled to compensation. \here costs are less than or
equal to reenues, there is no opportunity cost payment to the supplier. 1his calculation is estimated
ex-ante, and then reconciled ex-post with known alues ,also or inclusion in the R-actor,.

80
1hat is, generators who participate in the SLM.
P1
Wholesale power
price
C/MWh
REFIT (Onshore Wind)
Reference price
Wholesale market
(SEM) price
REFIT pays ME
payment = 0
REFIT revenue paid at
difference between
REFIT reference price
and wholesale price
P2 P3 P4

101

1abe ..2: vvaret opportvvit, co.t pa,vevt.
Revenues (received from the SLM) Costs (incurred under the PPA)
Lnergy ~ SMP x MSQLl
81
x 1PD
82

RLlI1 Benchmark Price x MG x 1PD Constraints ,where releant,
Capacity payments
ovrce: CR
lor ivvaret opportunity cost payments, reenues are calculated on the basis o Market Scheduled
Quantity Loss lactored ,MSQLl, rather than Metered Generation ,MG,. At its simplest, the in-
market opportunity cost payment ,illustrated in 1able A2, thereore keeps the supplier whole or any
dierence in costs incurred under the PPA with the renewable generator, and the actual reenues it
is able to obtain rom the wholesale market.
1he supplier is compensated or dierences in the ai.patcbea everg, and veterea geveratiov under the
PPA through the market constraint payment.
lor the purposes o the out-o-market opportunity cost payment, the dierence between the cost to
suppliers at the RLlI1 reerence price and what it would hae cost them to buy the equialent
olumes rom the market is used to determine i any payment is due. I the market cost is greater
than or equal to the RLlI1 reerence price, there is no payment to the supplier. I market cost is
less than the RLlI1 reerence price, the supplier will be entitled to compensation under the
opportunity cost payment support. 1able A.3 shows the calculations or out-o-market RLlI1
opportunity cost payments.
1abe ..: Ovtofvaret opportvvit, co.t pa,vevt.
Market cost Contract cost
Pool Purchase Price ,SMP - Capacity Charge -
Imperections Charge - Market Operator Charge, x
,MG or MGLl, x 1PD
RLlI1 Reerence Price x ,MG or MGLl, x 1PD
ovrce: CR
1ecbvoog, aifferevce pa,vevt.
1he technology dierence payment or RLlI1 I is paid independently o the market price. As a
result, i the wholesale power price rises aboe the reerence price or the renewable generation
technology the .vppier continues to be paid the top-up between the RLlI1 reerence price or

81
Under the 1SC the MSQ o a ariable price taker is set to its actual metered output ,Metered Generation ,MG, ~
Actual Output x 1PD, when its dispatch is in accordance with its MSQ. \hen such a generator is constrained down in
dispatch, its MSQ is set to the greater o its Actual Output and the time weighted aerage o its outturn aailability.
1hus, or the purposes o the reenue calculation, the ormula MSQLl x 1PD ensures that the correct market reenue
,MG unless constrained down, accruing to the supplier is calculated.
82
1PD is the 1rading Period Duration in the 1SC and is 0.5 and deines the ' hour trading period in the SLM. 1he
energy payment is thus calculated in M\h.

102

onshore wind and the alternatie renewable technology ,or example, biomass,
83
. 1he geverator
continues to receie the PPA price. Ioweer, like the market equalisation payment, the RLlI1 II
technology dierence payment operates like a one-way CD.
1he calculation o the technology dierence payment is the most complex aspect o the RLlI1, and
depends on the PPA price, the RLlI1 reerence price and the SLM price. 1he diagram aboe
assumes, or simplicity, that the PPA price and the RLlI1 price are the same. Ioweer, i the PPA
price is greater than the RLlI1 price ava the RLlI1 price is greater than the wholesale price, then
the .vppier receies the dierence between the RLlI1 price or that technology and the RLlI1
reerence price or large onshore wind. I the PPA price is less than the releant technology RLlI1
reerence price bvt greater than the wholesale price, then the .vppier receies the dierence between
the PPA price and the wholesale power price.
I the PPA price is less than the wholesale price, then the .vppier receies no technology dierence
payment.
!orea eavpe
1able A.4 proides a worked example o RLlI1 payments or oshore wind and onshore wind.
1he worked example illustrates a scenario where the wholesale power price is assumed to be t50 per
M\h ,so e.. than the reerence RLlI1 price,. 1he PPA price is assumed to be equal to the RLlI1
reerence price or the technology.
1abe ..1: v.tratire R1 cacvatiov.
Llement Onshore wind Offshore wind
RLII1 I RLII1 II RLII1 I RLII1 II
lixed payment 10
Market equalisation
payment
16
1echnology dierence
payment
0 0 0 90
1O1AL payment 26 26 26 JJ6
R1 aavivi.tratiov ava co.t recorer,
1he cost o the scheme is passed on to Republic o Ireland consumers in the PSO payment. 1he
PSO ley is charged to all electricity consumers as an additional cost in the electricity bill. 1he PSO
R-actor ,which coers other mechanisms than the RLlI1, is used to reconcile the dierence
between the ex-post amounts due to suppliers ,based on submitted, audited statements that are
consistent with goerning legislation and CLR reporting requirements, and the ex-ante estimates
paid to them under the PSO ley.

83
1he size o top-up depends on whether the PPA price is greater or less than the technology reerence price.

103

Other points to note on administration,cost recoery elements o RLlI1 include:
x CPI igures are applied to the reerence price on an annual basis using Ireland`s LSRI
quarterly economic orecasts.
84

x Interest
85
is applied ex-post to R-actor payment alues owed to,rom the PSO und. It is
intended to relect that monies accruing under the R-actor are paid two PSO periods ater
the period to which the payment reers.
x 1he RLlI1 ex-ante benchmark price is set at parity with other support proided under the
PSO to proide transparency and consistency with the State Aid Notiication to the LU
regarding the RLlI1.
x PSO payments to suppliers rom the PSO ley are made on a monthly proportionate basis.
1he 1SO,DSO acts as an independent party who makes the inal monthly payments to the
releant suppliers.
ligure A.2 illustrates payment and remuneration processes o RLlI1 and its interactions with the
SLM mechanism.
igvre ..2: R1 revvveratiov proce..

84
Although not all taris under RLlI1 II are linked to CRI.
85
At LURIBOR, the Luro Interbank Oered Rate. 1he LURIBOR rates are based on the aerage interest rates at
which a panel o more than 50 Luropean banks borrow unds rom one another and are considered be the most
important reerence rates in the Luropean money market.
L
S

S
S
A
W
WW
S

S

8
^
^DW

8Lll1
1

l
8Lll1
C
8Lll1

1

P
kll1

104

1he preceding discussion shows that ,although the RLlI1 calculation process appears complex, the
RLlI1 is by and large a ixed lI1 or .vppier. ,and indeed because payments to generators are
through a PPA, it is precisely a ixed lI1 or them,. 1he complexity arises because o the need to
take account o the SLM market price, and o the dierences in cost between dierent renewable
technologies.
Note that the RLlI1 does not obligate suppliers to purchase RL under the scheme, but priority o
dispatch principles in the SLM
86
,as required under LU Renewables Directie, means there is a clear
route to market or suppliers who enter into PPAs under the scheme.
Lessons for a Northern Ireland stand-alone scheme
Covtract forv ava ae.igv
x 1he SLM has a clear market reerence price or determining payments under a CD contract
or RLlI1 policy. 1he gross pool in the SLM makes the design o a CD instrument simpler
compared to a physical contracting market such as BL11A.
x Ioweer, or an eectie hedge against pool price olatility to be achieed, an ex-post
reconciliation actor ,as is the case or RLlI1, is likely to need to apply to reconcile
payments to actual outturn prices in the SLM.
x As RLlI1 has an ex-post adjustment ,and so eectiely is a ixed lI1 that remoes basis
risk or the supplier, a Northern Ireland scheme not settled ex-post would look relatiely
unattractie to inestors in the market.
8

x RLlI1 operates like a one-way CD. A two-way CD design ,settled ex-post, would remoe
price risk or generators. Ioweer, this would also take away upside or inestors.
Generators would need to repay reenues where the wholesale price than the strike price.
x Under the SLM`s gross pool system ariable renewable generation ,such as wind, are price
takers. 1here is not the issue ,as is in GB, o keeping renewable generation in the market`
as all generation ,aboe a ae vivivi. threshold, is already sold through the pool and centrally
dispatched by System Operator.
x A CD lI1 designed to delier a ixed reenue stream could operate consistently with this
process and the SLM. Ioweer, as with RLlI1 or the proposed GB CD lI1 scheme, the
issue is whether the subsidy payment low should operate through suppliers or be paid
directly to the generator.
88

86
See Principles o Dispatch and the Design o the Market Schedule in the 1rading & Settlement Code - A SLM
Committee decision paper`, September 2010
8
1he RO does not remoe price risk or inestors. Ioweer, the ere o subsidy proided under the scheme ,in its
current orm, is relatiely generous and so inestors are willing to manage price risk.
88
As illustrated by RLlI1.

105

Covtract aocatiov
x 1he contract allocation process or CD lI1s is related to the counterparty to those
contracts. Under RLlI1, generators source a PPA rom a .vppier - RLlI1 then guarantees
the price the .vppier is paid or that power in the SLM or a veterea olume o power.
89

x In GB, there are ongoing discussions as part o the LMR process as to who could be the
counterparty to new CD lI1 contracts. 1he proposal is or Goernment to establish a
central body to administer and be the counterparty to the contracts.
x Under the LMR proposals ,as currently, there will be no obligation on suppliers to source a
ixed olume o renewable electricity.
90
1hat is, the support mechanism would work purely
as a price support mechanism.
ôrtberv reava cova aavivi.ter .et referevce price. ;ie vvaer R1) or aereop av avctiovivg vecbavi.v for covtract
aocatiov tbat rova .et tbe tariff ere..
x 1able A.5 compares the pros and cons o an administered approach s. an auctioning
approach ,drawing on the LMR consultation and our own analysis,.
x It is clear that both approaches hae adantages and disadantages.

1abe ..:: Pro. ava cov. of aavivi.terea ava avctiovivg proce..e. for .ettivg 1.
Scheme type Pros Cons
Administered approach
\ell understood process as is how
support leels are set under the RO.
Likely to hae lower upront
administratie costs.
Can be complex to implement and
relies on accurate data. Does not hae
the same price reelation properties as
an auction. RO illustrates that
approach can lead to oer subsidy i
olume targets are binding.
Auctioning
Price discoery characteristics o an
auction should enable inancial
support to be set at a leel just high
enough to lead to deployment ,i.e.
reduce excessie economic rents ,oer
subsidy,,.
May be considered risky and
unattractie by inestors and policy
makers. \ould require Goernment
to determine what share o the
electricity mix should be low-carbon
and a iew o the technology mix.
Co.t recorer,
x 1he obious implication o a .tavaaove .cbeve or Northern Ireland is the cost will be leied
solely on Northern Ireland`s electricity consumer. In contrast the RO supports Northern

89
Note that suppliers hae no obligation to buy renewable electricity.
90
1his is also the case under the RO. \hile an obligation or a ixed olume o renewable electricity is placed on
suppliers they hae the option o paying the buy-out price rather than sourcing the power.

106

Ireland generators through a UK wide obligation on electricity suppliers to source renewable
electricity.
x As Northern Ireland has a lower obligation under the RO than the rest o the UK, a
standalone Northern Ireland CD lI1 scheme ,to delier Northern Ireland`s 40 renewable
electricity consumption target, would almost certainly increase household electricity bills in
the region.
91

x Ioweer, there is a simple option or recoering the cost ,were it deemed aordable, rom
consumers. 1he PSO ley ,which recoers the cost o the Directed Contract costs in the
SLM, could be used to ley lI1 costs on Northern Ireland consumers just as the RLlI1
uses the PSO ley to recoer costs in the Republic o Ireland.
.avivi.tratiov
x A stand-alone scheme or Northern Ireland would require the region`s own public sector
organisations to manage and administer the contract allocation and payment processes
,unlike under the RO where administration is currently handled by Ogem through a serice
agreement with NIAUR,.
x Under a RLlI1 model, DL1I would issue letters o oer`. Under a GB CD policy
contract allocation and signature would presumably also be handled by DL1I and the
Department o linance and Personnel ,unless Northern Ireland participated in the GB
Goernment arm`s length body,.
x In a similar role to CLR with RLlI1, it is likely that NIAUR ,with its knowledge and ires
in the SLM, would need to play an important administratie role during management and
administration o the RLlI1 payments through the PSO. Similar to the role played by CLR
with RLlI1 in the Republic.

91
1he increase in bills is likely to be both in absolute terms ,as the rovve o renewable generation supported under av,
support mechanism increases, and relatie to the NIRO in its current orm.

10

APPLNDIX B: PRLMIUM II1 IN 1HL SLM
Introduction
One o the options DLCC are considering or the Llectricity Market Reorm ,LMR, project is a
Premium leed-in 1ari ,lI1,. A Premium lI1 has been adopted in other Luropean countries ,or
example, Spain, as a mechanism or incentiising renewable generation.
1his annex explores:
x the design eatures o Premium lI1s adopted in other Luropean countries ,with particular
reerence to Premium lI1 design in Spain,, and
x how a stand-alone Northern Irish Premium lI1 could be made to work in the SLM and the
design characteristics that might be adopted.
A Premium lI1 has similar characteristics to Northern Ireland`s existing renewable generation
support mechanism, the NIRO. \e highlight these similarities and any potential dierences
throughout the annex discussion.
Premium II1 design options
A Premium lI1 design oers a top-up on market electricity prices ,see ligure B.1 below,. 1he
premium payment can be designed to achiee two objecties: 1, to represent the enironmental
attributes o renewable generation, or 2, to approximate renewable electricity costs.
igvre .1: Previvv 1 voae

ovrce: ^R ;2010): . poic,vaer`. gviae to eeaiv 1ariff Poic, ae.igv

108

Orerrier of optiov.
1here are seeral types o Premium lI1 model including:
x Cov.tavt Previvv 1 poicie. - a predetermined adder on top o the spot market, and
x iaivg Previvv 1 poicie. - which allow payments to ary based on market price.
x Percevtage Previvv 1 poicie. - the market price plus a set percentage o market price.
1able B.1 summarises some o the pros and cons with a constant and sliding Premium lI1, each o
which hae been implemented in other Luropean countries. A key adantage highlighted with
Premium lI1s is their compatibility with liberalised electricity markets. Unlike a lixed lI1 they
maintain commercial incenties on renewable generators who continue to sell their power through
the releant jurisdiction`s market arrangements.
1abe .1: Previvv 1 ae.igv i..ve.
Llement Constant Premium II1 Sliding Premium II1
Description Constant pre-determined added on top
o the wholesale price.
1he subsidy continues to be oered een
i wholesale prices increase.
Can be dierentiated by technology type,
uel type and project sizes.
Allows payments to ary based on
market prices.
Lxamples include:
- Caps and loors on total premium.
- Percentage based approaches.
92

Laluation Relatiely simple to implement and
administer.
Risk or project deelopers that prices
decrease along with project returns.
Risk o windall gains i wholesale prices
increase relatie to when Premium lI1
set.
As wholesale market prices increase ,or
decrease, the premium amount can be
designed to decline ,and ice ersa, to
minimize windall proits.
But it is challenging to establish what the
caps and loors should be and to adjust
them oer time i technology costs
change signiicantly.
93

Lxamples
94
Czech Republic
Sloenia
Netherlands,
Denmark ,or onshore wind energy,
Spain
Denmark ,or onshore wind energy,

92
\here the Premium lI1 is deined as a percentage o the spot market price.
93
Similar to the challenges o a banding under the RO.
94
Note the examples are not all the current support mechanism in the releant jurisdictions.

109

A constant Premium lI1 is the simplest o the dierent options. Ioweer, because o the design
challenges with setting the leel o a Premium lI1 ,and the risks perceied by inestors who
continue to manage wholesale price risk, the remuneration o the premium option is generally set
higher than that o a ixed tari option in order to compensate renewable electricity inestors or
the higher risks associated ,i the same inestment in new installations is to be achieed,.
95
1his in
turn implies higher costs or the electricity consumer.
In response to the potential problems created by constant Premium lI1 approaches, certain
jurisdictions hae allowed the lI1 payments to ary based on market price. lor example, Spain has
adopted caps and loors on the premium paid ,a sliding premium option,. 1he subsection which
ollows reiews how that scheme has worked.
paiv .iaivg .cae Previvv 1

Spain introduced a sliding premium option that includes both a payment cap and a payment loor on
the previvv avovvt ,in t,M\h, paid to renewable generators.
9
I aerage wholesale market prices
increase, the premium payment also begins to decline. A loor price is also proided, below which
the combined reenues o the premium price and the market price cannot drop. ligure B.2
illustrates Spain`s lI1 premium payment leels or onshore wind. 1he loor price is set at
t3.66,M\h. As the electricity pool price increases, the premium amount declines until the aerage
market price rises aboe t8.9,M\h. At this point the premium lI1 oered alls to zero.
igvre .2: pavi.b .iaivg 1 for ov.bore riva ;200)


95
See Poyry, 2009, Qvaitatire Cov.iaeratiov. for C 1 ae.igv
96
1his section draws heaily on ^R, 2010, . poic,vaer`. gviae to eeaiv 1ariff Poic, ae.igv
9
Note a sliding Premium lI1 can also be designed to be a cap,loor on the total payment receied by the renewable
generator ,wholesale market price - premium payment,.

110

Some o the other key eatures o Spain`s lI1 ramework include:
x Renewable generators hae the option o a guaranteed ,ixed, lI1 or the guaranteed bonus
,premium, lI1. 1he goernment publishes reerence tari leels ,or dierent technologies,
or both orms o contract ,see 1able B2 below, .
x Generators choose at the beginning o each year whether they want to sell the electricity as
part o the bundled ixed-price lI1 tari or on the spot market ,where they would all under
the premium lI1 option,.
x \hile the ixed-tari design guarantees a purchase obligation, there is none or generators
who choose the premium option. Under the latter, electricity is sold on the spot market
instead, or ia bilateral contracts.
x Under its ixed-price option, Spain requires all RL projects larger than 10 M\ to orecast
their anticipated supply 30 hours beore a day starts by contacting the regional grid operator
,OMLL,. It is still possible to correct orecasted amounts up to one hour beore the hourly
deadline.
x Ioweer, i the inal hour-ahead electricity orecast diers rom the actual electricity
supplied by more than 20 or solar and wind, and more than 5 or other technologies,
the owners are charged a ee o 10 o the aerage electricity price on eery kilowatt-hour
o deiation.
x Renewable energy deelopers with multiple projects at dierent sites can oset some o
these deiations with surplus production elsewhere on the grid.
x Under the Premium lI1 option, Spain employs the market rules, which requires all
renewable electricity generators to orecast their anticipated output, regardless o the
generation plant size.

111

1abe 2: 1 ere. iv paiv

ovrce: CP.
trevgtb. ava reave..e. of a Previvv 1 approacb iv ôrtberv reava
One o the key strengths o a constant Premium lI1 is its simplicity.

112

Another strength is its similarities to the NIRO which appears well understood by inestors and has
already stimulated considerable leels o renewable generation inestment in Northern Ireland. It
may thereore hae the least disruption costs during any transition rom the NIRO.
1he less complex characteristics o a Premium lI1 mean it may also be more lexible to uture
changes in the wholesale market structure in Northern Ireland ,as the subsidy sits on top o the
wholesale price, howeer this may be determined,.
98

But a CD lI1 is currently DLCC`s aoured option as the low-carbon generation support
mechanism or the GB market:
x I Northern Ireland were to adopt a dierent approach, it would be a standalone scheme
next to two larger generation zones in the Republic o Ireland and GB both operating orms
o CD lI1s ,although both are likely to operate ery dierently,.
99

x Proided a CD lI1 can be designed in the GB market, and gien the design o RLlI1
,eectiely a ixed lI1,, Northern Ireland could be considered relatiely risky by inestors as
a location or inestment.
x 1his is because in Northern Ireland ,under a Premium lI1 regime, inestors would continue
to bear wholesale price risk ,as under the RO, unlike under a CD type regime that would
operate in the Republic o Ireland and BL11A ,although under the current GB proposals
generators would still bear some short term risk under a CD approach,.

98
Although we note that Premium lI1s with caps and loors ,such as Spain, hae similar policy design challenges in
competitie electricity markets as a CD lI1 design.
99
Although we note the detailed designs may in practice be dierent.

113

APPLNDIX C: WHOLLSALL LLLC1RICI1Y PRICL PROJLC1IONS
In this appendix we outline our approach or orecasting wholesale electricity prices in the SLM
including the assumptions used in the modelling. As context we reiew historical eidence o SLM
wholesale prices and other research studies o the uture changes in the SLM.
CJ. Market structure and historical evidence of SLM wholesale prices
1he SLM is a separate energy and capacity market. In the energy market, the Bidding Code o
Practice ,BCP, requires generators to only relect SRMC in bids. As well as an energy payment, an
additional capacity payment is paid to generators to meet total customer demand ,plus a margin or
security, based on the ixed costs o a Best New Lntrant Peaking Plant` ,BNL Peaker,. 1his
ensures that generators can recoer their operating costs plus the ixed costs o the BNL peaker.
I energy payments to market participants were designed to purely compensate the Short-Run
Marginal Cost ,SRMC, ,aoidable cost, relected in their bids, there would be no incentie or
generators to inest in plant which had higher ixed costs but lower aoidable costs than the BNL
peaker, as such generators would be paid only the lower ixed costs o the BNL peaker and the
lower bid or their actual generation. 1his problem is aoided in the SLM by paying generators, not
the aoidable cost o production, but at a System Marginal Price ,SMP,.
1he uel price o the marginal generation plant will set the shadow price ,SRMC, that is used to
determine the SMP and, with capacity payments, the wholesale price receied by generators. ligure
A1 shows the weekly and 28-day aerages or the shadow price, the uplit, their sum, the SMP, and
the capacity payments that, when added to the SMP, determine the total payments to dispatched
generators ,those aailable but not dispatched just receie the capacity payment,. 1he data is or the
period No 200 to April 2010.
ligure C1 shows:
x the two components o generator reenue - capacity payments and SMP,
x SMPs in the SLM hae been olatile across the trading year,
x prices were on aerage lower in 2009 than 200,2008.

114

igvre C1: Covpovevt. of rboe.ae price, ava 2aa, vorivg arerage

SMP in the SLM is calculated on a hal-hourly basis and ,as illustrated in ligure C1, is measured in
Luros per Megawatt-hour ,t,M\h,. As context to the price data in ligure C1, as contracts are
dierentiated by base, mid-merit 1 ,-23 hrs,, mid-merit 2 ,-19 hrs, and peak ,1-21 hrs,, the
eolution o SRMCs or mid-merit and peaking plant and one o the releant uel costs are plotted
in ligure C2. 1his shows a strong correlation between the moement in SMP and gas prices, which
is not surprising gien the majority o SLM demand is sered by gas ired power plant.
100

100
SLM MMU ,2009,: Market Monitoring Unit: Market Update`
SMP, SRMC and capacity payments in Irish SEM Nov 2007-ApriI 2010
0
10
20
30
40
50
60
70
80
90
100
110
1
-
N
o
v
-
0
7
1
-
D
e
c
-
0
7
1
-
J
a
n
-
0
8
1
-
F
e
b
-
0
8
1
-
M
a
r
-
0
8
1
-
A
p
r
-
0
8
1
-
M
a
y
-
0
8
1
-
J
u
n
-
0
8
1
-
J
u
l
-
0
8
1
-
A
u
g
-
0
8
1
-
S
e
p
-
0
8
1
-
O
c
t
-
0
8
1
-
N
o
v
-
0
8
1
-
D
e
c
-
0
8
1
-
J
a
n
-
0
9
1
-
F
e
b
-
0
9
1
-
M
a
r
-
0
9
1
-
A
p
r
-
0
9
1
-
M
a
y
-
0
9
1
-
J
u
n
-
0
9
1
-
J
u
l
-
0
9
1
-
A
u
g
-
0
9
1
-
S
e
p
-
0
9
1
-
O
c
t
-
0
9
1
-
N
o
v
-
0
9
1
-
D
e
c
-
0
9
1
-
J
a
n
-
1
0
1
-
F
e
b
-
1
0
1
-
M
a
r
-
1
0
1
-
A
p
r
-
1
0
E
u
r
o
s
/
M
W
h
SMP+capacity payment
weekly average SMP
28-day average SMP
28-day shadow price
capacity payments
28-day uplift

115

igvre C2: rovtiov of .egvevt. of tbe rboe.ae varet, 2aa, arerage.
101

ovrce: CP.
1able C.1 below shows SMP statistics or 1 No. 200 through to 31 Dec. 2008 as presented in the
SLM Market Monitoring Unit`s ,MMU, public report on the market. Oer the irst three quarters o
2009 the aerage SMP was t44.43,M\h.
102
In the study period, the MMU also note that the CPM
added approximately t15 or each M\h consumed.
1abe C.1: MP .tati.tic.
SMP Summary t/MWh 2007 QJ 2008 Q2 2008 Q3 2008 Q4 2008
Aerage SMP ,time weighted, 62.34 4.95 88.40 86.51 2.65
Aerage SMP ,demand weighted, 6.21 8.81 92.49 91.6 .0
Standard Deiation ,All, 45.0 41.68 34.92 31.0 3.34
Standard Deiation ,Daily Aerage, 10.3 10.40 14.15 12.56 12.2
Standard Deiation ,Intraday Aerage, 32.62 23.85 20.82 22.01 24.26
Minimum 29.31 30.19 2.45 30.23 3.39
Maximum 524.65 468.32 525.0 551.46 696.85
1
st
Quartile 43.33 54.96 62.62 63.46 52.45
3
rd
Quartile 6.25 80.98 108.34 106.34 83.6
ovrce: M MM|

101
M\he is a megawatt hour o electricity. 1he conention adopted here is that M\h reer to electricity, and M\ht is
the thermal energy content o uel. At 50 eiciency 2 M\ht o gas can generate 1 M\he o electricity
102
Ibid.
Mid merit and peaking short-run marginaI costs and mid-merit fueI costs
0
20
40
60
80
100
120
1-Nov-07 1-Feb-08 1-May-08 1-Aug-08 1-Nov-08 1-Feb-09 1-May-09 1-Aug-09 1-Nov-09 1-Feb-10
E
u
r
o
s
/
M
W
h
e
peak
mid-merit 2
mid-merit 1
gas at 36% efficiency

116

As the MMU note in their 2009 Public Report, aerage prices were at their highest during Quarter 2
and 3 o 2008, when uel prices were also at their highest. Aerage prices were at their lowest during
the irst two months o the market inluenced by the low cost o carbon in this period ,note prices
hae subsequently allen urther since the MMU analysis,.
1he MMU has also inestigated price-setting plant` in the SLM. 1he percentage o periods in which
selected Generator Units were detected by the MMU as setting the Shadow Price ,or the study
period 1 No. 200 through to 31 Dec. 2008, olumetrically grouped into the hal-hours o the
1rading Day is illustrated in ligure A3.
1he analysis in ligure C3 shows that the majority o price setting plant in the SLM are either gas, oil
or coal ired plant. 1his supports the obsered strong correlation o SMP with trends in wholesale
commodity prices, such as the price o natural gas.
igvre C: Price .ettivg geverator. b, tive of aa,

ovrce: M MM|
C2. Studies on future generation portfolios and prices in the SLM
In 2009, the Regulatory Authorities ,NIAUR and CLR ,hereater the RAs,, conducted an impact
assessment o high leels o wind penetration in 2020 on the SLM.
103
1he study ealuated ie uture
generation portolios ,based on the All Island Grid Study
104
,.

103
http:,,www.allislandproject.org,GetAttachment.aspxid~20c228-2b30-48a-a0-539a3c65523c
104
DCLNR, DL1I ,2008,: All Island Grid Study`

11

1able C.2 shows the existing thermal and new generation plant assumed in the RAs modelling under
the alternatie All Island Grid Study portolios.
1abe C.2: 2020 M geveratiov portfoio. ;M!)
Plant type Portfolio J Portfolio 2 Portfolio 3 Portfolio 4 Portfolio S
Coal 1,163
CCG1 1,294 1,200 1,200 1,200
OCG1 1,450 829 1,968 311 829
ADG1 89 535 535 111
Baseload renewables 183 183 183 183 361
Variable renewables 2 2 2 2 2
\ind 2,000 4,000 4,000 4,000 6,000
Interconnector ,Last,\est, 500 500 500 500 500
Subtotal new generation S,S87 7,3J8 7,2S7 7,427 9,200
Lxisting Base Generation 6,70J 6,70J 6,70J 6,70J 6,70J
1otal installed capacity J2,288 J4,0J9 J3,9S8 J4,J28 JS,90J
As well as assessing the costs and beneits o increased wind on the system, the study also looked at
the eect o the ie generation portolios on SMP and capacity payments to judge whether
wholesale prices in the SLM would be higher or lower with increased wind generation in 2020.
ligure C4 illustrates the annual aerage SMP or Portolios 1 to 5, in time-weighted terms. 1he
ranges around the central SMP estimates relect high and low uel scenarios where uel and carbon
prices are assumed to be 50 o the central scenario.
igvre C1: Pro;ectiov. for 2020 arerage MP

0
50
100
150
200
250
Portfolio 1 Portfolio 2 Portfolio 3 Portfolio 4 Portfolio 5
C
/
M
W
h
Central Time weighted SMP

118

1he RAs analysis showed that while SMPs are generally lower the more price-taker generation ,i.e.,
wind, there is on the all-island system, this is not uniersally the case. 1he type o conentional plant
that makes up the rest o the portolio was also illustrated to be critical.
C3. Poyry low carbon generation scenarios in the SLM (203S)
Poyry has also examined the main generator technologies that could be deployed on the island o
Ireland in the next twenty years and the potential impact on wholesale prices. 1hey modelled six
possible generation portolios, each with a particular technology theme or a study year o 2035. All
scenarios built on the 2020 renewables policy targets while also recognising that in this time rame
the existing coal, oil and peat ired stations will hae reached end o lie. ligure A5 shows the SLM
generation-weighted aerage annual wholesale prices.
105

igvre .:: Moaeea M rboe.ae varet price. iv 20:

ovrce: Po,r,
ligure A5 shows the portolios with a higher proportion o renewables ,with their lower ariable
generation costs, typically set lower market prices.
ligure A6 shows the wholesale prices with Pory`s central gas price scenario, and the shaded bars
show the higher wholesale prices in our high gas price scenario. All scenarios show similar ariation,
which is basically a result o gas-ired plant setting the market price on the island or a large
proportion o the time.

105
I.e. it coers all generator SMP reenues and capacity payments weighted by SLM electricity generation.

119

igvre .: vpact of bigb fve price. ov voaeea M rboe.ae varet price. iv 20:

ovrce: Po,r,
As with the RA`s analysis o uture wholesale prices in the SLM in 2020, the Poyry analysis
illustrates the uture uncertainty o wholesale prices in the SLM and thereore the reenue streams
or generators in the market.
C4. Wholesale price modelling assumptions
Generation plant assumptions are based on the public data set proided in:
http:,,www.allislandproject.org,en,market_decision_documents.aspxpage~2&article~cb4ee33b-
a83a-4ce-956a-6c30900495
CLPA has also made a number o simpliying assumptions ,or example, on outage rates and
eiciency, where plant data was not publicly aailable.

120

APPLNDIX D: GLNLRA1ION CAPACI1Y ASSUMP1IONS
;i) D1
1abe D1: D1 Capacit, .tivate. ;M!)
1echnology Description
Capacity Lstimates (2020) in MW
Low Medium High
Onshore \ind 1 M\ 800 1000 1200
1 M\ 20 40 60
Oshore \ind 200 600 800
Biomass 1 M\ 50 150 300
Using uel 5 9 13
Lnergy rom \aste 10 30 40
1idal 100 300 400
\ae - - -
Iydro 5 6 10
PV 5 25 50
1otal J,J9S MW 2,J60 MW 2,873 MW
Off.bore !iva, 1iaa ava !are accovvt for tbe foorivg: 00 M!; 00 M!; 1200 M!. rov D1. off.bore
reverabe everg, geveratiov cov.vtatiov ;p.), tbi. .eev. to be tro tbira. off.bore riva, ove tbira tiaa ava ero rare porer.
(ii) LirGrid/SONi - All Island Generation Capacity Statement (20JJ - 2020) (p.66)
1abe D2: irCria,Oî Capacit, .tivate. ;M!), ba.ea ov D1. Meaivv cevario
Sample Capacity (2020) Comments (comparison to DL1I medium)
Onshore \ind 1030 Similar
Oshore \ind 600 Same
Small Scale Iydro 3 Similar
Solid Biouels 200 Area o potential dierence |see below[
Landill Gas 2 Similar
CIP 50 Area o potential dierence |see below[
Industrial , DSU 3 n,a
1idal,\ae 300 Same
1otal 2,2J3 MW

121

(iii) CLPA - model inputs
1abe D: CP. fiva voae ivpvt. for geveratiov capacit, iv 2020 ;M!)
Sample Capacity (2020) in MW
Onshore \ind 1030
Oshore \ind 600
Biomass 209
Lnergy rom \aste 30
1idal 300
\ae 0
Iydro 6
PV 5
1otal 2,J80 MW
(iv) Breakdown
In order to obtain accurate model inputs, we hae aimed to reconcile published igures. \here this
was diicult or not possible, we aimed to make sensible assumptions. 1hese are detailed below:
x Oerall, the Lirgrid igures are reasonably similar to the DL1I igures. 1he total is only
50M\ oer ,close, and the wind igures are irtually the same.
x 1here are two main dierences in the igures:
R lirstly, the biomass igures: 1here is 159 M\ in the DL1I igures, whilst in the
Lirgrid igures there is 200 M\ rom biouels and a urther 50 M\ rom CIP. \e
hae chosen to use the Lirgrid igure o 200 M\ or biomass, whilst also including
the 9 M\ o 'biomass using uels' rom the DL1I igures.
R Secondly, PV igures: Lirgrid igures do not show any expected increase in capacity
rom PV technology. 1hereore we hae decided to utilise the 'low scenario' DL1I
igures o 5 M\ ,as opposed to the 'medium scenario' igures o 25 M\,.
1he project model has six size categories or each technology, ranging rom 'Domestic' ,5k\, to
'Very Large' ,10M\,.
106
1hereore, we hae made some high-leel assumptions to split these
capacity estimates by technology size, based on the inormation aailable rom the DL1I and
LirGrid,SONi igures, as well as our own understanding o the NI energy context. 1he table below
presents the assumptions or the medium scenario, except or PV where we hae chosen to take
DL1I's low case scenario igure.

106
1he ull list is: Domestic ,5k\,, Small ,50k\,, Medium 1 ,500k\,, Medium 2 ,5M\,, Large ,10M\,, and
Very Large ,10M\,.

122

1abe D1: CP. capacit, breaaorv ava covvevt.
1echnology Scale
Generation
Mix
Capacity
(MW)
Comments
Onshore
\ind
Domestic 0.2 2.4
Based on the DL1I igs ,medium scenario,,
c.4 o capacity is 1M\. So assume this is
distributed between Domestic, Small and Med
1, with proportions based on the Ogem ROC
register o existing installed capacity.
Small 1. 1.3
Med 1 1.9 19.9
Med 2 3.4 34.9
Distribution o Med 2 to V. large scaled
according to the total o the existing and
proposed applications or onshore wind
generators ,NI Direct Goernment \ebsite,. Large 11.1 114.3
V. Large 81. 841.2
Oshore
\ind
Domestic - -
CLPA Assumption: 100 o oshore wind is
ery large , 10 M\,.
Small - -
Med 1 - -
Med 2 - -
Large - -
V. Large 100 600
Biomass
Domestic - -
x 1ake the 200M\ o 'Solid Biouels' rom
Lirgrid's igures. Assume this capacity is all
ery large ,i.e. 10 M\,.
x Assume that the 9 M\ o 'Biomass using
uel' ,DL1I's igures, is part o the 50 M\
o CIP ,Lirgrid's igures,. Assume that the
other 41 M\ is rom gas CIP and
thereore non-renewable.
x O these 9 M\, assume it is split between
the medium size categories, with 25 in
Med 1 and 5 in Med 2.
Small - -
Med 1 1.1 2.25
Med 2 3.2 6.5
Large - -
V. Large 95. 200

123

1echnology Scale
Generation
Mix
Capacity
(MW)
Comments
Lnergy rom
\aste
Domestic - - x Assume that the 30 M\ ,DL1I igures,
mostly corresponds to the 2 M\ o
landill gas in Lirgrid's igures. Assume all
o this alls under Med 2 , 500 k\, but
5 M\,.
x O the remaining 3 M\, assume that this
alls under Anaerobic Digestion, and
categorise this under Med 1 , 50 k\, but
500 k\,.
Small - -
Med 1 10 3
Med 2 90 2
Large - -
V. Large - -
1idal Domestic - - Assume the majority o tidal stream
deployment is large scale plant.
Small - -
Med 1 - -
Med 2 - -
Large - -
V. Large 100.0 60
\ae Power Domestic - - Not applicable. Based on the DL1I igures,
there neither any currently installed capacity in
Northern Ireland, not is there any expected
generating capacity or wae power in 2020.
Small - -
Med 1 - -
Med 2 - -
Large - -
V. Large - -
Iydro Domestic - - Assume 6 M\ in total, taken rom DL1I's
igures ,medium scenario,. Assume it is
distributed according to the proportions o
existing installed capacity according to the
current Ogem ROC register.
Small .1 0.2
Med 1 51.8 1.6
Med 2 41.1 1.2
Large - -
V. Large - -
PV Domestic 53.0 2. x Lirgrid igures do not include any expected
PV generation capacity in 2020. 1hereore
take DL1I's 'low' scenario o 5 M\.
x Assume this 5 M\ is distributed according
to the proportions o existing installed
capacity according to the current Ogem
ROC register.
Small 2.0 1.3
Med 1 20.0 1.0
Med 2 - -
Large - -
V. Large - -
1otal 2,J77 MW

124

APPLNDIX L: GLNLRA1ION COS1 ASSUMP1IONS
L.J. Overview of approach
Renewable energy technologies selected were those that could reasonably be expected to be
deployed in Northern Ireland by 2020. Note that tidal energy in the context o this study reers to
tidal stream technology rather barrages or other tidal range technologies.
1he size bands assessed were chosen to:
x Align with the size thresholds or the support mechanisms in GB and participation in the
SLM, and
x Proide reasonable ranges within which costs could be speciied, while using a manageable
number o bands or modelling.
lor each technology, a typical plant size within each band was selected or modelling purposes. In
analysing capital and operating costs or each technology, there will be project speciic ariability, in
addition to market uncertainty on actual cost at a particular point in time.
\e hae thereore sought to establish reasonable cost ranges based on our experience and a broad
range o sources.
L.2. Key public cost and performance sources
x UK Llectricity Generation Costs Update`, Mott McDonald, June 2010
http:,,www.decc.go.uk,assets,decc,statistics,projections,1-uk-electricity-generation-
costs-update-.pd
x Costs o low-carbon generation technologies`, Mott McDonald, May 2011
http:,,hmccc.s3.amazonaws.com,Renewables20Reiew,MML20inal20report20or
20CCC20920may202011.pd
x Note on the impacts o the CD lI1 support package on costs and aailability o capital`,
CLPA, July 2011
http:,,www.decc.go.uk,media,iewile.ashxiletype~4&ilepath~11,policy-
legislation,LMR,214-cepa-paper.pd&minwidth~true
x Discount rates or low-carbon and renewable generation technologies`, Oxera, May 2011,
http:,,hmccc.s3.amazonaws.com,Renewables20Reiew,Oxera20low20carbon20di
scount20rates20180411.pd
x Design o leed-in 1aris or sub-5M\ Llectricity in Great Britain`, Llement
Lnergy,Poyry, July 2009
x Powering the Nation update 2010`, Parsons Brinckerho, April 2010,
http:,,www.pbworld.co.uk,index.phpdoc~528

125

x luture Marine Lnergy`, Carbon 1rust, 2006,
http:,,www.carbontrust.co.uk,Publications,pages,PublicationDetail.aspxid~C1C601
x Lnergy Saings 1rust case studies, http:,,www.energysaingtrust.org.uk,Generate-your-
own-energy,Real-lie-case-studies
x Projected Costs o Generating Llectricity`, ILA,NLA, 2010,
http:,,www.iea.org,w,bookshop,add.aspxid~403
x Digest o United Kingdom Lnergy Statistics`, DLCC, 2009,
http:,,www.decc.go.uk,en,content,cms,statistics,publications,dukes,dukes.aspx
x 1he Oshore Valuation`, Oshore Valuation Group, May 2010,
http:,,www.oshorealuation.org
x Agri-lood & Biosciences Institute, http:,,www.abini.go.uk,index,serices,serices-
specialist-adice,renewable-energy,re-anaerobic-digestion,re-anaerobic-digestion-intro.htm
x ILA Lnergy 1echnology documents, http:,,www.iea.org,techno,index.asp

L.3. Outturn levelised costs
ligure L1 illustrates current leelised cost estimates or a range o larger scale renewable
technologies. Capital cost scalers are applied across the modelling period to relect projected alls in
renewable generation costs ,learning rates`,. 1able L1 ,oerlea, proides leelised costs by year
and technology type.
igvre 1: erei.ea co.t e.tivate. ;cvrrevt) ;_,M!b)

0
50
100
150
200
250
/
M
W
h

126

1abe 1: erei.ea co.t. v.ea iv tbe voaeivg

ovrce: 1ariov.

20J2 20JS 2020
Onshore \ind Domestic _23.58 _223.9 _10.23
Onshore \ind Small _23.58 _236.3 _204.68
Onshore \ind Medium 1 _158.65 _144.4 _125.06
Onshore \ind Medium 2 _99.2 _91.6 _9.23
Onshore \ind Large _80.06 _.92 _4.3
Onshore \ind Very Large _80.06 _.92 _4.3
Oshore \ind Very Large _14.6 _140.46 _123.43
Biomass Medium _114.82 _114.82 _114.82
Biomass Very Large _101.4 _101.21 _100.05
Anaerobic Digestion Small _11.80 _115.6 _111.01
Anaerobic Digestion Medium , Large _60.03 _59.5 _58.52
1idal _242.53 _19.4 _95.40
Iydro Small _199.29 _199.29 _199.29
Iydro Small,Medium _156.60 _156.60 _156.60
Iydro Medium _115.92 _115.92 _115.92
Iydro Large _103.01 _103.01 _103.01
PV Small _352.80 _24.22 _189.08
PV Small , Medium _323.03 _251.59 _14.20
PV Medium _366.3 _284.81 _196.05
PV Large _206.29 _13.98 _120.4

Cepa Report - Analysis of Emr Policy Options For Northern Ireland 2

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Cepa Report - Analysis of Emr Policy Options For Northern Ireland 2

Încărcat de

Drepturi de autor:

Formate disponibile

1

LLLC1RICI1Y MARKL1 RLIORM

In this case, suppliers are required to manage

1SC uSC SC 8Lll1

S-ar putea să vă placă și