Uni e si y o he Basque Coun y
Euskal He iko Unibe si a ea
Impac o he phase ou o F ench nuclea plan s on he Spanish
elec ici y ma ke .
Mas e Thesis submi ed o he deg ee o
Mas e in Economics: Empi ical Applica ion and Policies
Au ho : Cyn hia Gab iela Ve a Ve a
Supe iso s: C is ina Piza o-I iza and Ai o Cia e a
Sep embe 2020
Con en s
1 In oduc ion 7
2 Li e a u e Re iew 9
3 Desc ip i e analysis o he elec ici y ma ke in Spain and F ance 11
3.1 Nuclea p oduc ion in F ance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.2 The Spanish elec ici y in e connec ion wi h F ance . . . . . . . . . . . . . . . . 11
3.3 Analysis o he p ice o ma ion in he Spanish elec ici y ma ke . . . . . . . . . 12
4 Empi ical S a egy and Da a 18
4.1 Da a.......................................... 18
4.2 Me hodology ..................................... 19
4.2.1 Nuclea ene gy in F ance . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.2.2 Spain in e connec ion capaci y . . . . . . . . . . . . . . . . . . . . . . . . 20
4.2.3 P oduc ion o coal and enewable ene gy . . . . . . . . . . . . . . . . . . 21
4.3 Scena ios........................................ 22
5 Resul s and Discussion 25
5.1 Scena ios wi h syn he ic bids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
5.2 Cos -Bene i Analysis ................................ 29
5.3 Robus ness es .................................... 32
6 Conclusions and Policy Implica ions 34
7 Appendix 36
2
Lis o Figu es
1 Impo and Expo Capaci y du ing 2019 (MW). . . . . . . . . . . . . . . . . . 12
2 Agg ega e Demand and Agg ega e Supply on Feb ua y 13 h, a 3am. . . . . . . 13
3 Agg ega e Demand and Agg ega e Supply on Feb ua y 13 h, a 9pm. . . . . . . 14
4 Agg ega e Demand and Agg ega e Supply on Sep embe 11 h, a 3am. . . . . . . 15
5 Agg ega e Demand and Agg ega e Supply on Sep embe 11 h, a 9pm. . . . . . . 15
6 Ma ginal Sys em P ice du ing 2019, a 3am and 9pm (eu o pe MWh). . . . . . 16
7 Sliding window co ela ion coe icien s be ween F ance and Spain elec ici y p ices. 17
8 Ke nel Densi y Es ima ion o Benchma k and Scena io 1. . . . . . . . . . . . . . 36
9 Ke nel Densi y Es ima ion o Benchma k and Scena io 2. . . . . . . . . . . . . . 37
10 Ke nel Densi y Es ima ion o Benchma k and Scena io 3. . . . . . . . . . . . . . 37
11 Ke nel Densi y Es ima ion o Benchma k and Scena io 4. . . . . . . . . . . . . . 38
12 Ke nel Densi y Es ima ion o Benchma k and Scena io 5. . . . . . . . . . . . . . 38
13 Ke nel Densi y Es ima ion o Benchma k and Scena io 6. . . . . . . . . . . . . . 39
14 Ke nel Densi y Es ima ion o Benchma k and Scena io 7. . . . . . . . . . . . . . 39
15 Ke nel Densi y Es ima ion o Benchma k and Scena io 8. . . . . . . . . . . . . . 40
16 Ke nel Densi y Es ima ion o Benchma k and Scena io 9. . . . . . . . . . . . . . 40
17 Ke nel Densi y Es ima ion o Benchma k and Scena io 10. . . . . . . . . . . . . 41
18 Ke nel Densi y Es ima ion o Benchma k and Scena io 11. . . . . . . . . . . . . 41
19 Ke nel Densi y Es ima ion o Benchma k and Scena io 12. . . . . . . . . . . . . 42
20 Ke nel Densi y Es ima ion o Benchma k and Scena io 13. . . . . . . . . . . . . 42
21 Ke nel Densi y Es ima ion o Benchma k and Scena io 14. . . . . . . . . . . . . 43
22 Ke nel Densi y Es ima ion o Benchma k and Scena io 15. . . . . . . . . . . . . 43
3
Lis o Tables
1 Desc ip i e s a is ics o elec ici y p ices in Spain and F ance in 2019. . . . . . . 19
2 E olu ion o he ins alled powe o elec ic ene gy (MW). . . . . . . . . . . . . . 21
3 Benchma k and scena ios p epa ed wi h syn he ic bids. . . . . . . . . . . . . . . 23
4 Weigh ed A e age P ice (eu os) and Mon hly Sum (TWh) o Benchma k and
scena ios wi h changes in nuclea ene gy. . . . . . . . . . . . . . . . . . . . . . . 26
5 Weigh ed A e age P ice (eu os) and Mon hly Sum (TWh) o Benchma k and
scena ios wi h changes in in e connec ion and nuclea ene gy. . . . . . . . . . . . 27
6 Weigh ed A e age P ice (eu os) and Mon hly Sum (TWh) o Benchma k and
scena ios wi h changes in coal, enewable ene gy and nuclea ene gy. . . . . . . . 28
7 Weigh ed A e age P ice (eu os) and Mon hly Sum (TWh) o Benchma k and
scena ios wi h changes in coal, enewable ene gy, in e connec ions and nuclea
ene gy.......................................... 29
8 Mon hly and annual di e ences o he p oduc o p ice and quan i y wi h espec
o he Benchma k (in millions o eu os). . . . . . . . . . . . . . . . . . . . . . . 31
9 K uskal-Wallis equali y-o -popula ions ank es and cumula i e dis ibu ion unc-
ion esul s. ...................................... 33
4
Lis o Abb e ia ions
- AD: Agg ega e Demand.
- AS: Agg ega e Supply.
- ENTSO-E: Eu opean Ne wo k o T ansmission Sys em Ope a o s o Elec ici y.
- GW: Gigawa .
- MIBEL: Ibe ian Elec ici y Ma ke .
- MSP: Ma ginal Sys em P ice.
- MTV: Maximum T adable Volume.
- MW: Megawa .
- MWh: Megawa - hou .
- OMIE: Ibe ian Ene gy Ma ke Ope a o .
- PNIEC: Na ional In eg a ed Ene gy and Clima e Plan.
- REE: Elec ici y Ne wo k in Spain (Red Eléc ica de España).
- RES-E: Elec ici y om Renewable Ene gy Sou ces.
- RES: Renewable Ene gy Sou ces.
- TW: Te awa .
- TWh: Te awa - hou .
5
Abs ac
One o he key elemen s in he ansi ion o low ca bon economies is he phase-ou o ossil- uel
based echnologies. Nuclea powe , despi e no being a high emi ing sou ce, is one o he
echnologies a he hea o he deba e, mainly due o secu i y issues. Howe e , nuclea elec-
ici y gene a ion is s ill one o he baseload echnologies in many coun ies and i s p og essi e
phase-ou will hus ha e impo an economic implica ions. In pa icula , F ance, which is one
o he coun ies wi h he highes nuclea pa icipa ion wo ldwide (i.e. 70.6% o o al gene a-
ion), decided o lowe nuclea p oduc ion o 50% o o al gene a ion by 2035. Since F ance is
a e y well in e connec ed coun y in e ms o ene gy, and i is also a ne elec ici y expo e ,
he p ices o all he neighbou elec ici y ma ke s will also be a ec ed by his poli ical decision.
Bianco and Sca pa (2018) analyzed he e ec ha he educ ion o he impo o elec ici y
om F ance o I aly due o he phase-ou o F ench nuclea plan s had on he I alian elec ici y
ma ke when he in e connec ion size was 2,650 MW. In his mas e hesis we p opose o explo e
he e ec ha his phase-ou would ha e on he Spanish elec ici y ma ke , whe e he cu en
elec ici y low om one coun y o ano he is 2,800 MW. In ac , F ance al eady expe ienced
a majo supply c isis in Janua y 2017 due o he s oppage o a la ge pa o i s nuclea plan s.
In ha momen , F ance had o impo Spanish elec ici y o co e hei demand and Spanish
elec ici y p ices inc eased by a 28% compa ed o he same mon h o he p e ious yea . Ou
esul s would ha e in e es ing policy implica ions o he in eg a ion o he single Eu opean
elec ici y ma ke , which is in he Eu opean 2030 agenda. The esul s show ha he educ ion
o nuclea ene gy in F ance, inc ease he p ices and dec ease he quan i ies, bu educing a
100% o coal and inc easing he in e connec ion capaci y o he maximum and he p oduc ion
o enewable ene gies, he p ices ends o dec ease and quan i ies inc ease again. Hence, his
pape ecommends suppo ing he educ ion o coal and an inc ease in in e connec ion capaci y
and enewable ene gies as a way o coun e ac he inc ease in he p ices due o he phase-ou
o F ench nuclea ene gy.
Key wo ds: Ene gy ma ke , nuclea ene gy, in e connec ion.
6
1 In oduc ion
Spain is p ac ically an ene gy island and i s only in e connec ions wi h Eu ope a e h ough
F ance, acco ding o he Elec ici y Ne wo k in Spain (REE, o i s Spanish ac onym), his
in e connec ion capaci y is cu en ly 2,800 MW. In ac , one o he goals o he Eu opean
Union in he nex yea s is o c ea e he "Ene gy Union", o which inc easing he size o
he in e connec ions all o e Eu ope is a p io i y. In his ega d, he Na ional In eg a ed
Ene gy and Clima e Plan (PNIEC, o i s Spanish ac onym) o 2021- 2030 aims o inc ease
he in e connec ion capaci y be ween Spain and F ance o 8,000 MW, ha inc ease ep esen s
mo e han 200% o e exis ing le els.
On he o he hand, ano he key elemen o he Eu opean ene gy policy o he ollowing
decades is he ansi ion o a low ca bon economy. The use o Renewable Ene gy Sou ces
(RES) will be os e ed and ossil uels will be g adually phased-ou , which also includes nuclea
powe . In his sense, he F ench nuclea lee is cu en ly he second la ges in he wo ld in
e ms o ins alled capaci y, only a e USA, i is composed o 56 nuclea eac o s, dis ibu ed
among 18 powe plan s. In 2019, p oduc ion was 379.5 TWh, o 70.6% o elec ici y p oduc ion
in F ance (CDE 2019).
The F ench Law o "Ene gy T ansi ion and G een G ow h" is a law wi h mul iple objec i es
whe e one is men ioned o "P ese ing human heal h and he en i onmen , mi iga ing he
emission o g eenhouse gases, limi ing indus ial isks, educing he exposu e o ci izens o ai
pollu ion and ensu ing nuclea sa e y". In No embe 2019, he "Ene gy and clima e" law o
F ance published plans o educe his pa icipa ion o nuclea ene gy o 50% by 2035, ollowing
he Eu opean di ec i es.
Since F ance is he main in e connec ion be ween Spain and Eu ope, any change in F ench
elec ici y p ices would ha e an impac on Spanish elec ici y p ices (and ice e sa). In ac ,
he p o i om he in e connec ion be ween Spain and F ance in 2019 was 168 million o eu os
(24.1% less compa ed o he p e ious yea ). In his sense, educing nuclea p oduc ion in F ance
would a ec he elec ici y p ice in Spain. Acco ding o his esea ch, wi h he inc ease o he
in e connec ions capaci y o he maximum, he p o i will be 1,276.96 million o eu os.
A i s analyzes o he case o a Ge man nuclea phase-ou in he con ex o he in e nal elec-
ici y ma ke in Eu ope by Hos e (1998) ound ha in he long e m he phase-ou would only
lead o a mode a e inc ease in he a e age cos s o elec ici y gene a ion bu in a compe i i e
in eg a ed elec ici y ma ke , cos s a e signi ican ly lowe han in closed elec ici y ma ke s.
Mo e ecen ly, Bianco and Sca pa (2018) ca ied ou a esea ch abou he impac o he phase-
ou o F ench nuclea plan s on he I alian powe sec o , showing he ele ance o his si ua ion
o I alian ene gy secu i y, because a la ge pa o he elec ici y ha I aly uses is impo ed
om F ance, bu a u u e elimina ion o nuclea powe plan s will signi ican ly educe expo
7
capaci ies wi h subs an ial consequences on he I alian ene gy sys em.
Gi en his si ua ion, i is ex emely impo an o ca y ou an analysis o he e ec o he
phase-ou o F ench nuclea plan s on he Spanish elec ici y ma ke , aking in o accoun he
expec ed changes in he in e connec ion. This esea ch aims ills he gap o he Ibe ian ma ke
and answe s he ollowing ques ion: "Wha will happen o he Spanish elec ici y p ices when
he phase-ou o F ench nuclea plan s?". In o de o answe he esea ch ques ion he objec i e
o his wo k is o quan i y he impac o he phase-ou o F ench nuclea plan s on he Spanish
elec ici y ma ke .
We ake 2019 as he s udy yea since i is he mos ecen yea wi h comple e hou ly da a
and hey a e accessible, because hey a e downloaded om he websi e o he elec ici y ma ke
ope a o designa ed o he managemen o he daily elec ici y ma ke in he Ibe ian Peninsula
and om he Eu opean Ne wo k o T ansmission Sys em Ope a o s o Elec ici y (ENTSO-E)
websi e.
Taking he idea o Bianco and Sca pa (2018), in his pape we ca y ou a simila analysis
o he e ec o he phase-ou o F ench nuclea plan s on he Spanish elec ici y ma ke . To
de elop his esea ch, we combine wo di e en me hodologies ha ha e al eady been used
by o he au ho s. On he one hand, we use he algo i hm elabo a ed by Cia e a, Espinosa
and Piza o-I iza (2014) o compu e he ou come o he hou ly auc ion o he elec ici y
whole sale ma ke . On he o he hand, we use he idea o "syn he ic bidding" employed by
Cia e a, Espinosa and Piza o-I iza (2017), in o de o build "syn he ic" supply cu es o
di e en scena ios o in e connec ions, enewable capaci y, nuclea phase-ou and changes in
in e connec ions.
The con ibu ion o his wo k consis s o measu ing he e ec o he phase-ou o F ench
nuclea plan s on he Spanish elec ici y ma ke , and will also se e as he basis o subsequen
esea ches who could ake mo e yea s in o accoun , ocus on he echnologies o con inue he
analysis o he e ec as he F ench a ge o 2035 is me . Ou esul s will con ibu e o he
ene gy policy analysis ha is being conduc ed in Eu ope o achie e a ansi ion o a low ca bon
economy a he lowes cos .
The es o he a icle is o ganized as ollows: Sec ion 2 p esen s a li e a u e e iew including
in e connec ions and nuclea phase-ou in Eu ope. Sec ion 3 exposes heo e ical amewo k
wi h he cu en si ua ion o he nuclea phase-ou in F ance and elec ici y ma ke in Spain,
whe e ma ke p ices and he impo ance o doing his esea ch a e analyzed. Sec ion 4, de ails
he empi ical s a egy ca ied ou . Sec ion 5 p esen s he main esul s. Sec ion 6 ends he
esea ch wi h conclusion and some policy implica ions.
8
2 Li e a u e Re iew
Re iewing p e ious publica ions abou in e connec ions, Child a all (2019) analyzes he impo -
ance o lexible elec ici y gene a ion, in e connec ions, and s o age o ob ain ully enewable
elec ici y sys ems h ough wo scena ios, he i s one, by independen ly modeled egions and
he second one, aking in o accoun he in e connec ion be ween egions. They men ion ha
when making he ansi ion o enewable ene gy he i s egion p esen ed a educ ion in p ice,
bu in he second scena io, in which he egions a e in e connec ed he elec ici y p ice is e en
lowe . The esul s o hese scena ios e i y he inc eased cos sa ings. The second scena io,
which conside s he in e connec ions be ween egions, p esen s sa ings o 26 billion eu os pe
yea compa ed o he scena io modeled independen ly, hey say ha "mo e apid de ossilisa ion
and g ea e cos sa ings can be achie ed h ough he es ablishmen o inc eased in e connec-
ions be ween he egions o Eu ope".
The au ho s men ion F ance as he coun y wi h he highes pa icipa ion o nuclea ene gy
o e he o al ene gy p oduc ion in he Eu opean Union, so i s ene gy policies a ec he coun-
ies wi h which i is connec ed ene ge ically and hei s udy shows ha by inco po a ing low
cos s o gene a ion and s o age o enewable ene gy and choosing he igh suppo ins umen s
and in line wi h he objec i es o he Eu opean Union, a ansi ion owa ds ene gy sus ainabili y
in Eu ope can be eached.
Con inuing wi h he in e connec ions, Ries, Gauda d, and Rome io (2016) analyze he case
o Mal a and i s in e connec ion wi h he Eu opean ma ke by building me i o de cu es.
They conclude ha he p ice le el o elec ici y pe consume does no dec ease wi h he newly
ins alled in e connec ion bu also depends on ins alled gene a ing capaci y, oil p ice, and ma ke
design.
Thei s udy men ions he secu i y o ene gy supply and he nega i e e ec s o in as uc u e
p ojec s as wo impo an poin s ha should no be igno ed by decision make s when imple-
men ing policies. Among hei ecommenda ions, hey men ion he impo ance ha should be
gi en o inc easing in e connec ions as well as eplacing old gene a o s.
Lea ing aside he in e connec ions, bu aking in o accoun F ance’s phase-ou policies, we
ind Malischek and T üby, (2016), hey analyze he phase-ou o nuclea ene gy in F ance in
h ee aspec s, i s , he cos s o phasing ou nuclea powe in F ance, second, how much o he
cos s will be passed on o he es o he Eu opean powe sys em and hi d, wha e ec does
he unce ain y ega ding u u e nuclea policy in F ance ha e on sys em cos s. Among hei
esul s, hey men ion ha he addi ional cos o phasing ou nuclea powe in F ance will be
76 billion eu os and ha his cos is highe i he phase-ou o he nuclea powe plan s occu s
be o e he end o he echnical li e ime. The cos s o he Eu opean elec ici y sys em depend
on how quickly neighbo ing coun ies implemen policies o coun e ac he phase-ou o ench
9
Figu e 6: Ma ginal Sys em P ice du ing 2019, a 3am and 9pm (eu o pe MWh).
Sou ce: Own elabo a ion wi h da a om OMIE.
The Figu e 6 shows he beha io o he MSP du ing 2019, a o -peak (3am) and on-peak
(9pm). The blue line, ha co esponds o 3am o o -peak, is highe han he ed line, ha
co esponds o 9pm o on-peak. The g ea es a ia ions a e obse ed du ing he mon hs o
Ap il and May, while he smalles a ia ions a e obse ed du ing he mon hs o June, July and
Augus . The mon hs wi h he smalles gap be ween 3am MSP and 9pm MSP a e Oc obe ,
No embe , Decembe and Janua y.
Con inuing wi h he analysis o p ices, has been made a calcula ion o he sliding o un-
ning window co ela ion be ween F ance elec ici y p ices and Spanish elec ici y p ices wi h a
window o 168 (7 days and 24 hou s).
Sliding-window c oss-co ela ion is a common me hod o es ima e ime a ying co ela ions
be ween signals. I p oduces a co ela ion alue be ween wo signals (posi i e o nega i e) o
e e y ( ime,lag) pai o alues. In p inciple, he expec ed alue o he co ela ion o any pai
o ( ime,lag) alues mus be compu ed by a e aging x( ) y( +lag) o e many ealiza ions o he
s ochas ic p ocess (Bäcke & Cassenae , 2002).
Figu e 7 shows he sliding window co ela ion coe icien s be ween F ance elec ici y ma ke
p ices and Spain elec ici y ma ke p ices, in which i is obse ed ha he esul s g ea e han
ze o (posi i e) and s able, in some cases e y high wi h alues abo e 0.8.
The posi i e esul s ob ained indica e ha he ela ionship be ween he p ice o elec ici y
16
ma ke in F ance and he p ice o elec ici y ma ke in Spain is posi i e, ha is, gi en an
inc ease in p ices in F ance (Spain), he p ices in Spain (F ance) end o inc ease, and gi en a
dec ease in p ices in F ance (Spain), p ices in Spain (F ance) end o dec ease. I can be seen
ha i is ime- a ying.
Figu e 7: Sliding window co ela ion coe icien s be ween F ance and Spain elec ici y p ices.
Sou ce: Own elabo a ion wi h da a om OMIE.
17
4 Empi ical S a egy and Da a
In his sec ion we desc ibe he da abases (4.1), he me hodology used in ou analysis, including
he algo i hm ha we model (4.2) and he scena ios ha we simula e (4.3).
4.1 Da a
We use wo di e en da a sou ces o he p epa a ion o his wo k. On he one side, we use
his o ical da a (ex-pos ) o he Spanish elec ici y hou ly bids du ing 2019, which we e ex ac ed
om he websi e o OMIE. The a iables o his da abase a e he ollowing:
- Da e: Day, mon h and yea o he obse a ion (Since Janua y, 1s ,2019 o Decembe 31s ,
2019).
- Hou : Time he p ice was cap u ed (1 am o 12 pm).
- quan i y: ene gy bid in MWh by one p oduc ion uni a a ce ain p ice.
- p ice: p ice bid in EUR/MWh by one p oduc ion uni o a ce ain quan i y o ene gy.
On he o he hand, he da abase used o wholesale F ench and Spanish p ices has been aken
om he ENTSO-E websi e. The a iables o his da abase a e he ollowing:
- Da e: Day, mon h and yea o he obse a ion (Since Janua y, 1s ,2019 o Decembe 31s ,
2019).
- Hou : Time he p ice was cap u ed (1 am o 12 pm).
- p ice_sp: P ice o he elec ici y in Spain.
- p ice_ : P ice o he elec ici y in F ance.
Once he p ice da a o Spain and F ance ha e been ex ac ed, he mean, s anda d de ia ion,
maximum, minimum, skewness and ku osis can be seen in he Table 1.
Compa ing he esul s o he p ices o Spain and F ance, i can be seen ha he mean is
highe o Spain wi h 47.8682, howe e , he s anda d de ia ion is highe in F ance wi h 14.0206.
Fo Spain he maximum is 74.74 and he minimum 0.01, and o F ance he maximum is 121.46
and he minimum is -24.92, because in F ance nega i e p ices a e allowed.
The skewness nega i e coe icien indica es ha p ices in Spain p esen a nega i e o le -
handed asymme ic dis ibu ion, while he posi i e coe icien o p ices in F ance indica es ha
i p esen s a posi i e o igh -handed asymme ic dis ibu ion. The ku osis esul s indica e ha
18
p ices in Spain, ha ing he coe icien highe han p ices in F ance, show a highe concen a ion
o alues a ound i s mean.
Table 1: Desc ip i e s a is ics o elec ici y p ices in Spain and F ance in 2019.
Desc ip i e S a is ics p ice_sp p ice_
Mean 47.8682 39.4517
S anda d De ia ion 10.8147 14.0206
Maximum 74.74 121.46
Minimum 0.01 -24.92
Skewness -0.955627 0.279215
Ku osis 5.744057 4.388169
Sou ce: Own elabo a ion wi h da a om ENTSO-E.
I is impo an o men ion ha all he igu es and ables p esen ed a e o own elabo a ion
wi h da a om he OMIE and ENTSO-E websi e wi h he s a is ical package S a a.
4.2 Me hodology
Fo his esea ch, he me hodology o Cia e a, Espinosa and Piza o-I iza (2014) has been
aken in o accoun . We use da a o he day-ahead ma ke and measu e he Ma ginal Sys em
P ice (MSP) and he Ma ginal T adable Volume (MTV) o he spo ma ke . We simula e
di e en scena ios and we measu e he p ice di e ences be ween hem, d i en by he me i
o de e ec .
To build he algo i hm we ake in o accoun h ee equa ions:
qmin(pi) = min{qask(pi), qbid(pi)}(1)
q aded =max
pi
{qmin(pi)}(2)
p aded =q−1
bid(q aded)(3)
Whe e:
-qask(pi): Agg ega e olume o ask o de s a p ices (pi).
-qbid(pi): Agg ega e olume o o e s a p ices (pi).
Equa ion (1) exp esses he ac ha o each p ice he quan i y aded would be he sho
side o he ma ke . Equa ion (2) compu es he quan i y aded (q aded)as he maximum o he
quan i ies ob ained in Equa ion (1). Fo i s pa , Equa ion (3) ind he ma ke clea ing p ice
o ma ke p ice (MSP) acco ding o he ma ke ules. Quan i ies a e exp essed in MWh and
p ices in eu os pe MWh.
19
Fu he mo e, Cia e a, Espinosa and Piza o-I iza ’s (2017) c ea e a syn he ic supply cu e
s uc u e based on a e e ence yea including changes in some o he echnologies. Hence, we
p oceed based on he algo i hm o Cia e a, Espinosa and Piza o-I iza ’s (2014) and Cia e a,
Espinosa and Piza o-I iza ’s (2017) syn he ic bidding s uc u e.
Fo he elabo a ion o he scena ios and he analysis o he e ec o hese on he elec ici y
p ices in Spain, he da a o 2019 ha e been modi ied in h ee ins ances, i s one, modi ying
he p oduc ion o nuclea ene gy in F ance (4.2.1), second one, modi ying he in e connec ion
capaci y be ween F ance and Spain (4.2.2) and inally, modi ying he p oduc ion o coal and
enewable ene gies (4.2.3).
4.2.1 Nuclea ene gy in F ance
In he i s ins ance, o he changes in he p oduc ion o nuclea ene gy in F ance, ha e been
aken in o accoun he equi emen s o he "Ene gy and clima e" law o F ance, which plans
o educe he pa icipa ion o nuclea ene gy o 50% by 2035. Fo his pu pose, a "Baseline"
scena io has been p epa ed, which educes he p oduc ion o F ench nuclea ene gy o 50% in
line wi h i s plan. Howe e , assuming ha F ance canno achie e his goal, a "pessimis ic"
scena io is p epa ed in which F ance only educes i s nuclea ene gy p oduc ion by 25%. On
he o he hand, assuming ha F ance will no only mee i s a ge o 2035, bu will also educe
100% o nuclea ene gy in F ance, he "op imis ic" scena io is c ea ed.
As he p oduc ion o nuclea ene gy in F ance co esponds o 70.6% o elec ici y p oduc ion,
he pa icipa ion o nuclea ene gy o e he o al elec ici y expo s o Spain is conside ed o be
p opo ional. Once he nuclea ene gy educ ion is applied, he MSP and MTV a e ecalcula ed
and compa ed o he cu en scena io o Benchma k o measu e he e ec o he educ ion o
nuclea ene gy.
4.2.2 Spain in e connec ion capaci y
In he second ins ance, aking in o accoun ha PNIEC 2021-2030 aims o inc ease he in e -
connec ion capaci y be ween Spain and F ance o 8,000 MW, he o iginal bases a e aken, he
in e connec ion capaci y is inc eased om 2,800 MW o 8,000 MW and he MSP and MTV a e
ecalcula ed o compa e wi h he Benchma k.
In acco dance wi h he abo e, h ee o he scena ios a e p oposed esul ing om modi ying
in he o iginal da abase he maximum in e connec ion capaci y and he educ ion o nuclea
ene gy in F ance as indica ed in Sec ion 4.2.1.
20
4.2.3 P oduc ion o coal and enewable ene gy
In he hi d ins ance, he p oduc ion o coal has been educed and some enewable echnologies
ha e been inc eased acco ding o he PNIEC 2021-2030, as i is shown in he Table 2. Ha e
been chosen o be modi ied wind ene gy, pho o ol aic sola ene gy and he moelec ic sola
ene gy because hey a e conside ed o be he only in e mi en enewable ene gies and hose
ha mos a ec p ice ola ili y. Also, hese echnologies a e he ones ha cu en ly ha e he
la ges pa icipa ion and hose ha acco ding o he PNIEC 2021-2030 will ha e he mos
changes.
Also a he ime o making changes in echnology has been aken in o accoun he capaci y
ac o , o wind ene gy is usually 10-40%, o pho o ol aic sola ene gy is 10-30% and he
he moelec ic sola ene gy can ope a e wi h a e y high capaci y ac o . Once i has inc eased
enewable ene gy, aking in o accoun i s capaci y ac o and he educ ion o coal o ze o, has
been ecalcula ed MSP and MTV o compa e wi h he Benchma k.
Table 2: E olu ion o he ins alled powe o elec ic ene gy (MW).
Technologies 2020 2030
Wind Ene gy 28,033 50,333
Pho o ol aic Sola Ene gy 9,071 39,181
The moelec ic Sola Ene gy 2,303 7,303
Coal 7,897 0
Sou ce: Own elabo a ion wi h da a om PNIEC 2021-2030.
Con inuing wi h he analysis o he e ec o he educ ion o coal and he inc ease o enewable
ene gies, his scena io is combined wi h he scena ios in which nuclea ene gy is modi ied 4.2.1,
he MSP and MTV a e ecalcula ed o be compa ed wi h he Benchma k o base scena io.
Con inuing he same analysis, he changes made o coal and enewable ene gy, aking in o
accoun hei capaci y ac o , is combined wi h he inc ease o he in e connec ion capaci y o
he maximum as men ioned in 4.2.2, he MSP and MTV a e ecalcula ed o be compa ed wi h
he Benchma k o base scena io.
Finally, he p e ious scena io which conside s changes in coal, changes in enewable ene gy
and in e connec ion capaci y a e combined wi h scena ios ha modi y nuclea ene gy in F ance
(4.2.1) has also been ecalcula ed he MSP and he MTV o be compa ed wi h he Benchma k,
and hese las h ee can be conside ed as he mos ealis ic scena ios, since hey ake in o
accoun bo h wha is es ablished by he laws in F ance, and wha is es ablished in he PNIEC
2021-2030 in Spain.
21
4.3 Scena ios
Fi s o all, we p esen a cu en scena io o Benchma k, which does no conside he objec i es
es ablished ei he in he PNIEC 2021- 2030 o he o he laws abou he inc ease o in e con-
nec ion and enewable ene gy, and he educ ion o coal and nuclea ene gy. I e lec s he
ac ual bidding s uc u e (no changes a all). We will conside his scena io o compa e he
o he coun e ac ual scena ios, whe e we upda e he in e connec ion size be ween Spain and
F ance, he nuclea capaci y in F ance o he capaci y o he di e en Spanish echnologies.
To analyze he e ec ha he phase-ou o F ench nuclea ene gy will ha e on he Spanish
elec ici y p ices, he ollowing scena ios wi h syn he ic bids ha e been elabo a ed, hey can
also be ound in Table 3.
- Scena io 1: Op imis ic scena io wi h a 100% educ ion in nuclea ene gy, keeping he
o he a iables cons an .
- Scena io 2: Baseline scena io wi h a 50% educ ion in nuclea ene gy, conside ing ha
F ance plans o educe he pa icipa ion o nuclea ene gy o 50% by 2035, keeping he
o he a iables cons an .
- Scena io 3: Pessimis ic scena io wi h jus a 25% educ ion in nuclea ene gy, keeping he
o he a iables cons an .
- Scena io 4: Scena io o maximum in e connec ion capaci y, aking in o accoun ha he
PNIEC 2021-2030 aims o inc ease he in e connec ion capaci y be ween Spain and F ance
o 8,000 MW, keeping he o he a iables cons an .
- Scena io 5: Op imis ic scena io wi h a 100% educ ion in nuclea ene gy and maximum
in e connec ion capaci y.
- Scena io 6: Baseline scena io wi h a 50% educ ion in nuclea ene gy and maximum
in e connec ion capaci y.
- Scena io 7: Pessimis ic scena io wi h jus a 25% educ ion in nuclea ene gy and maximum
in e connec ion capaci y.
- Scena io 8: Scena io wi h a educ ion o coal o 100% and an inc ease in enewable
ene gies, aking in o accoun he objec i es es ablished in he PNIEC 2021-2030, keeping
he o he a iables cons an .
- Scena io 9: Op imis ic scena io wi h a 100% educ ion in nuclea ene gy, a educ ion o
coal o 100% and an inc ease in enewable ene gies.
22
- Scena io 10: Baseline scena io wi h a 50% educ ion in nuclea ene gy, a educ ion o coal
o 100% and an inc ease in enewable ene gies.
- Scena io 11: Pessimis ic scena io wi h jus a 25% educ ion in nuclea ene gy, a educ ion
o coal o 100% and an inc ease in enewable ene gies.
- Scena io 12: Scena io o maximum in e connec ion capaci y and a educ ion o coal o
100% and an inc ease in enewable ene gies, keeping he o he a iables cons an .
- Scena io 13: Op imis ic scena io wi h a 100% educ ion in nuclea ene gy, wi h maxi-
mum in e connec ion capaci y, a educ ion o coal o 100% and an inc ease in enewable
ene gies.
- Scena io 14: Baseline scena io wi h a 50% educ ion in nuclea ene gy, wi h maximum
in e connec ion capaci y, a educ ion o coal o 100% and an inc ease in enewable ene -
gies.
- Scena io 15: Pessimis ic scena io wi h jus a 25% educ ion in nuclea ene gy, wi h max-
imum in e connec ion capaci y, a educ ion o coal o 100% and an inc ease in enewable
ene gies.
Table 3: Benchma k and scena ios p epa ed wi h syn he ic bids.
Scena ios
Changes in nuclea ene gy Maximum
In e connec ion
Capaci y
Changes in Coal
and
Renewable Ene gy
Op imis ic Baseline Pessimis ic
100% 50% 25%
Benchma k
Scena io 1 X
Scena io 2 X
Scena io 3 X
Scena io 4 X
Scena io 5 X X
Scena io 6 X X
Scena io 7 X X
Scena io 8 X
Scena io 9 X X
Scena io 10 X X
Scena io 11 X X
Scena io 12 X X
Scena io 13 X X X
Scena io 14 X X X
Scena io 15 X X X
Sou ce: Own elabo a ion.
23
The esul s ex-an e ob ained wi h hese syn he ic bids, ha conside di e en ma ke si ua-
ions, a e explained and analyzed in he nex sec ion.
24
5 Resul s and Discussion
In his sec ion we desc ibe he esul s ob ained in each scena io (5.1), a b ie we a e analysis
is pe o med (5.2) and h ough he K uskal-Wallis equali y-o -popula ions ank es we seek o
s eng hen he esul s. (5.3).
5.1 Scena ios wi h syn he ic bids
In his sec ion di e en scena ios a e p esen ed wi h ex-an e esul s assuming changes in nuclea
ene gy and he combina ion o hese wi h changes in in e connec ion capaci y, changes in coal
and changes in enewable ene gy, since he PNIEC 2021-2030 aims o o inc ease he in e con-
nec ion capaci y be ween Spain and F ance and he "Ene gy and Clima e" law o F ance plans
o educe he pa icipa ion o nuclea ene gy o e he o al ene gy p oduc ion.
Changes in nuclea ene gies ha e always aken in o accoun h ee possible scena ios, an
op imis ic scena io wi h he educ ion o 100% in nuclea ene gies, a baseline scena io wi h he
educ ion o 50% in nuclea ene gies and a pessimis ic scena io wi h he educ ion o only 25%
o nuclea ene gy.
To ca y ou he analysis o he i s scena ios modi ying he nuclea ene gy p oduced by
F ance, Table 4 shows he esul s o he weigh ed a e age p ice (in eu os) and he mon hly
sum o he Benchma k (in TWh) o cu en scena io, compa ed o he i s h ee scena ios,
op imis ic scena io, baseline scena io and pessimis ic scena io. I can be seen also he annual
weigh ed a e age p ice and he annual sum o each scena io.
The esul s show ha he educ ion o nuclea ene gy in F ance, inc ease he p ices and
dec ease he quan i ies. La ge educ ion in nuclea ene gy, inc ease he p ice, in o he wo ds,
he op imis ic scena io is he one wi h he highes weigh ed a e age p ice. Because, as expec ed,
a a lowe p oduc ion o ene gy, p ices would end o inc ease and as p oduc ion dec eases and
his dec ease is no ollowed by ano he policy o coun e ac he e ec , p ices will be highe
and highe .
On he o he hand conside ing only he quan i y and no he p ice, la ge educ ion in nuclea
ene gy, lowe he quan i ies, in o he wo ds, he op imis ic scena io is he one wi h he smalles
quan i ies. This e ec is expec ed because a lowe p oduc ion o nuclea ene gy is no ollowed
by ano he policy ha incen i izes he main enance o he same nego ia ed quan i ies.
To he p e ious scena ios in which a policy o nuclea ene gy educ ion is ca ied ou in
F ance, a policy o inc easing he in e connec ion capaci y in Spain o he maximum is now
applied and Table 5 shows he weigh ed a e age p ice (in eu os) and he mon hly sum (in TWh)
o Benchma k and scena ios wi h changes in in e connec ion and nuclea ene gy (op imis ic,
baseline and pessimis ic). I can be seen also he annual weigh ed a e age p ice and he annual
sum o each scena io.
25
5.3 Robus ness es
Finalizing he p ice analysis, o he obus ness o he esul s ob ained in each scena io, has
been pe o med a K uskal-Wallis equali y-o -popula ions ank es . I es ed whe he se e al
independen samples come o no om he same popula ion. I can be conside ed as a gene -
aliza ion o he Wilcoxon Rank Sum es .
The null hypo hesis es ablishes ha he dis ibu ion o he p ices o he Benchma k and he
dis ibu ion o he p ices o each scena io a e equal, while he al e na i e hypo hesis es ablishes
ha he dis ibu ion o he p ices o he Benchma k and he dis ibu ion o he p ices o each
scena io a e no equal. The esul s ob ained can be seen a Table 9 and he Ke nel Densi y
Es ima ion o Benchma k and each scena io can be seen in he Appendix.
In he case o he Benchma k wi h Scena io 5, he p- alue is s ic ly g ea e han 0.01. Hence,
we can’ ejec he null hypo hesis and we conclude ha he dis ibu ion o he p ices o he
Benchma k and he dis ibu ion o he p ices o Scena io 5 a e equal.
Fo he o he scena ios, he p- alue is s ic ly less han 0.01 so he null hypo hesis is ejec ed
and i can be concluded ha he dis ibu ion o he p ices o he Benchma k and he dis ibu ion
o he p ices o each scena io (excep scena io 5) a e no equal.
To iden i y he sign, o he Benchma k and Scena io 1, whe e he p ice dis ibu ion is no
equal, he hi d column indica es ha he cumula i e dis ibu ion unc ion o he Benchma k
is below he cumula i e dis ibu ion unc ion o Scena io 1, and o his case he Benchma k is
be e .
On he o he hand, o he Benchma k and Scena io 4, whe e he dis ibu ion o p ices is also
no equal, he hi d column indica es ha he cumula i e dis ibu ion unc ion o he Bench-
ma k is abo e he cumula i e dis ibu ion unc ion o Scena io 4, and o his case Scena io 4
is be e .
F om he esul s o his analysis, i can be concluded ha scena io 5, co esponding o an
op imis ic scena io in which 100% o nuclea ene gy in F ance is educed and in e connec ion
capaci y is inc eased o he maximum, is no s a is ically signi ican a he 1% signi icance le el.
In o he wo ds, aking nuclea ene gy educ ion and capaci y inc ease policies o he ex eme
would no be con enien scena ios.
32
Table 9: K uskal-Wallis equali y-o -popula ions ank es and cumula i e dis ibu ion unc ion
esul s. Benchma k wi h P obabili y Cumula i e dis ibu ion unc ion
Scena io 1 0.0001 F0(x)< F1(x)
Scena io 2 0.0001 F0(x)< F2(x)
Scena io 3 0.0055 F0(x)< F3(x)
Scena io 4 0.0001 F0(x)> F4(x)
Scena io 5 0.0224 F0(x) = F5(x)
Scena io 6 0.0001 F0(x)> F6(x)
Scena io 7 0.0001 F0(x)> F7(x)
Scena io 8 0.0001 F0(x)> F8(x)
Scena io 9 0.0001 F0(x)> F9(x)
Scena io 10 0.0001 F0(x)> F10(x)
Scena io 11 0.0001 F0(x)> F11(x)
Scena io 12 0.0001 F0(x)> F12(x)
Scena io 13 0.0001 F0(x)> F13(x)
Scena io 14 0.0001 F0(x)> F14(x)
Scena io 15 0.0001 F0(x)> F15(x)
Sou ce: Own elabo a ion wi h da a om OMIE.
33
6 Conclusions and Policy Implica ions
In his esea ch, ha e been ca ied ou i een possible scena ios, aking in o accoun he "Ene gy
and clima e" law o F ance ha plans o educe 50% o he pa icipa ion o nuclea ene gy on
he o al ene gy p oduc ion and he objec i es p oposed by he PNIEC (2021 -2030) ha aims
o inc ease in e connec ion capaci y, educe coal p oduc ion and inc ease enewable ene gy
p oduc ion.
The esul s in Sec ion 5 show ha he educ ion o nuclea ene gy in F ance, inc ease he
p ices and dec ease he quan i ies. The op imis ic scena io is he one wi h he highes weigh ed
a e age p ice and he smalles quan i ies.
Wi h an inc ease in he in e connec ion capaci y o he maximum, p ices dec ease d as ically,
bu i his scena io is combined wi h he educ ion o nuclea ene gy, p ices inc ease. In he
op imis ic scena io he p ice is highe han he Benchma k. When inc easing he in e connec-
ion capaci y o he maximum, he quan i y inc eases, bu as he nuclea ene gy dec eases, he
quan i ies also dec ease, in he case o he op imis ic scena io, he quan i y eaches o be lowe
han he Benchma k.
Wi h a educ ion o coal in a 100% and an inc ease in he p oduc ion o enewable ene gy, he
weigh ed a e age p ice dec eases, howe e , as nuclea ene gies a e educed, he p ice inc eases
again, bu does no exceed he weigh ed a e age p ice o he cu en scena io o Benchma k.
On he quan i y side, wi h he educ ion o coal and he inc ease in enewable ene gies, he
quan i y inc eases. Bu , as he nuclea ene gies a e educed, he quan i ies dec ease again, bu
hey a e no lowe han he Benchma k.
Wi h a educ ion o 100% o coal, he p oduc ion o enewable ene gy is inc eased and he
in e connec ion capaci y is inc eased by maximum, he weigh ed a e age p ice is much lowe .
Howe e , as nuclea powe is dec eased, he p ice inc eases again. I he quan i ies a e aken in o
accoun , he quan i y d as ically inc eases, bu as he nuclea ene gy is educed he quan i ies
end o dec ease, howe e in he op imis ic scena io, whe e 100% o he nuclea ene gies a e
educed, he amoun is no less han he Benchma k.
Due o he a o emen ioned and he esul s ob ained, as a policy implica ion i can be men-
ioned ha i is impo an o suppo he educ ion o coal and he inc ease in in e connec ion
capaci y and p oduc ion o enewable ene gies as a way o coun e ac he inc ease in p ices due
o he phase-ou o nuclea ene gy in F ance.
Taking in o accoun he di e ence o he p oduc o p ice and quan i y wi h espec o he
benchma k, he scena io wi h he bes esul s is he one in which coal, enewable ene gies and
in e connec ion capaci y a e modi ied (Scena io 12). And, he scena io wi h he wo s esul s
is he one ha educes 100% o nuclea ene gy in F ance (Scena io 1). F om he bes and he
wo s scena io, when compa ing he p ice a ia ion wi h he cu en o Benchma k scena io,
34
we can conclude ha he maximum p ice educ ion is equal o 9.7 eu os pe MWh and he
maximum p ice inc emen is 2.27 eu os pe MWh.
Thanks o K uskal-Wallis equali y-o -popula ions ank es esul s i has been possible o
conclude ha scena io 5, in which maximum in e connec ion capaci y is combined wi h he
elimina ion o a 100% o nuclea ene gy in F ance, is no s a is ically signi ican a he 1%
signi icance le el, so his pape does no ecommend he implemen a ion o policies ha would
ake he elimina ion o nuclea ene gy in F ance o he ex eme.
I can be men ioned ha demand has no been modeled, his documen has only concen a ed
on he supply side, bu scena ios whe e demand is modi ied and he s a egic beha io o he
ma ke pa icipan s (p ice endogenei y) could be included as u he esea ches.
35
7 Appendix
Figu e 8: Ke nel Densi y Es ima ion o Benchma k and Scena io 1.
Sou ce: Own elabo a ion wi h da a om OMIE.
36
Figu e 9: Ke nel Densi y Es ima ion o Benchma k and Scena io 2.
Sou ce: Own elabo a ion wi h da a om OMIE.
Figu e 10: Ke nel Densi y Es ima ion o Benchma k and Scena io 3.
Sou ce: Own elabo a ion wi h da a om OMIE.
37
Figu e 11: Ke nel Densi y Es ima ion o Benchma k and Scena io 4.
Sou ce: Own elabo a ion wi h da a om OMIE.
Figu e 12: Ke nel Densi y Es ima ion o Benchma k and Scena io 5.
Sou ce: Own elabo a ion wi h da a om OMIE.
38
Figu e 13: Ke nel Densi y Es ima ion o Benchma k and Scena io 6.
Sou ce: Own elabo a ion wi h da a om OMIE.
Figu e 14: Ke nel Densi y Es ima ion o Benchma k and Scena io 7.
Sou ce: Own elabo a ion wi h da a om OMIE.
39
Figu e 15: Ke nel Densi y Es ima ion o Benchma k and Scena io 8.
Sou ce: Own elabo a ion wi h da a om OMIE.
Figu e 16: Ke nel Densi y Es ima ion o Benchma k and Scena io 9.
Sou ce: Own elabo a ion wi h da a om OMIE.
40
Figu e 17: Ke nel Densi y Es ima ion o Benchma k and Scena io 10.
Sou ce: Own elabo a ion wi h da a om OMIE.
Figu e 18: Ke nel Densi y Es ima ion o Benchma k and Scena io 11.
Sou ce: Own elabo a ion wi h da a om OMIE.
41
