Sales

Pricing in the UK retail energy market,

Description
Pricing in he UK reail energy marke, Caherine Waddams Price Cenre for Comeiion Policy, Universiy of Eas Anglia Minyan Zhu Cenre for Comeiion Policy, Universiy of Eas Anglia CCP Working Paer 13-12
Categories
Published
of 24
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
Pricing in he UK reail energy marke, Caherine Waddams Price Cenre for Comeiion Policy, Universiy of Eas Anglia Minyan Zhu Cenre for Comeiion Policy, Universiy of Eas Anglia CCP Working Paer Absrac UK governmens and he energy regulaor have shown increasing concern abou he healh of comeiion in he residenial energy marke, following heir ioneering deregulaion a he end of he las cenury. We idenify he effecs of inroducing he non discriminaion clauses in 2009, a major regulaory inervenion and he firs since deregulaion. We exlore he effec of his inervenion on he rice movemens of he six major layers, and find ha he naure of comeiion in he indusry has changed, wih less effecive rivalry beween he regional incumbens and large regional comeiors following he inervenion; comanies seem o have rereaed o heir home regions, leaving a marke where ricing behaviour resembles more closely a duooly beween Briish Gas and he regional incumben. ISSN Pricing in he UK Reail Energy Marke, 2005 o Caherine Waddams Price 2 and Minyan Zhu 3 ESRC Cenre for Comeiion Policy and Norwich Business School Universiy of Eas Anglia Absrac UK governmens and he energy regulaor have shown increasing concern abou he healh of comeiion in he residenial energy marke, following heir ioneering deregulaion a he end of he las cenury. We idenify he effecs of inroducing he non discriminaion clauses in 2009, a major regulaory inervenion and he firs since deregulaion. We exlore he effec of his inervenion on he rice movemens of he six major layers, and find ha he naure of comeiion in he indusry has changed, wih less effecive rivalry beween he regional incumbens and large regional comeiors following he inervenion; comanies seem o have rereaed o heir home regions, leaving a marke where ricing behaviour resembles more closely a duooly beween Briish Gas and he regional incumben. December We graefully acknowledge funding from he ESRC, and many helful discussions wih and suggesions from Sehen Davies, Maxine Frerk, Moren Hviid, Sehen Lilechild, and Ben Smihers, while we reain sole resonsibiliy for he analysis and is inerreaion. We are graeful for commens from colleagues a he Cenre for Comeiion Policy, and aricians in discussions a he Ausralian Comeiion and Consumer Council 1. Inroducion and hisory 2008 marked a change in he UK energy regulaor s aiude o he residenial reail marke. While i had reviously ioneered and chamioned reail comeiion in his secor, i became increasingly concerned wih he fairness of he comeiive rocess, seeing he comeiion glass much more as half emy han half full. We exlore he effec of his change in olicy on he indusry by idenifying changes in he way ha he major suliers o he reail marke have se heir rices. The debae coninues o be driven by oliical and consumer concerns, and o be fuelled by oliical inervenion, ranging from ex cahedra saemens by he rime miniser ha he would ensure ha everyone was on he cheaes ariff in he marke (Cameron, 2012) o he romise of a seveneen monh rice freeze if he Labour ary is eleced in 2015 (Miliband, 2013), and couner moves from he governmen o remove axes which fund energy efficiency measures from rices (Osborne, 2013). Unil 1996, each residenial consumer in Grea Briain was served by wo monooliss a naional gas sulier, Briish Gas 4 (known as Scoish Gas in Scoland), and one of foureen regional elecriciy suliers. The energy markes were oened beween 1996 and 1999, and each of hese monooly suliers enered each ohers markes. A rocess of consolidaion hrough akeover and exi led by 2002 o he emergence of five major successors o he elecriciy incumbens, each reviously he monooly sulier in wo or hree regions, and Briish Gas. These firms dominaed suly, wih oher enrans gaining less han 1% of he marke over he nex decade, and wih no long erm survivors amongs hese enrans, who were aken over or exied he marke. The regulaor reduced barriers o enry afer some years, and by he end of 2013 here were several new enrans, whose join share of he marke had grown o 3%, he larges for many years. Neverheless suly coninues o be dominaed by he Big 6. Analysis of energy rices from hese firms a he ime when he las rice cas were removed from he reail secor 5 in 2002 showed ha while ars of he marke were comeiive, incumbency mark-us remained, suggesing considerable consumer ineria; and rice variaions did no reflec differences in coss for consumers using reaymen (ay as you go) meers, indicaing considerably less well develoed comeiion in his marke (Salies and Waddams Price, 2004). The removal of rice cas on incumben suliers, coincided wih he end of a eriod of consolidaion in he indusry which culminaed in he esablishmen of he Big 6. In he hree years following ha consolidaion he surviving comanies chose rice srucures which effecively searaed he marke, wih some offering ariffs aricularly aracive for users of large quaniies, and ohers offering ariffs which were beer for users of small amouns of elecriciy (Davies e al., 2014). These 4 Pied gas is no available in many rural areas, so around 17% of households do no use gas. 5 Disribuion and ransmission rices remained regulaed. 2 seem o have evolved from he reeaed ineracion of he comanies in he regional elecriciy and naional gas markes raher han from any exlici collusion; hey are consisen boh wih innovaion in he marke and as an effecive way of sofening head-on comeiion beween suliers. The very raid increase in wholesale energy coss afer 2005 seems o have desabilised his ariff aern. However one aern which coninued in hese laer years, and which resuled in subsanial regulaory inervenion, was he ersisence of an incumben mark-u of around 10%, similar o he level idenified when he rice cas were removed in 2002 (Salies and Waddams Price, 2004). In is 2008 Energy Suly Probe, he regulaor idenified such rice differences beween regions as a symom of comeiion concern and a major roblem for fairness; hey inroduced a new license condiion (25A, which we refer o as he nondiscriminaion clause, or NDC), o reven comanies charging higher mark-us o consumers in heir home regions han in ohers, i.e. he same incumben mark-us which had been idenified a he ime of deregulaion 6. The regulaor ursued his olicy on he grounds of fairness, and a concern ha vulnerable consumers were more likely no o have swiched, and so be aying higher rices (Ofgem, 2008), desie acknowledging oenial damage o comeiion (Ofgem 2009). The regulaor was also moivaed by comlains from oenial enrans from ouside he indusry ha he heavy discouns offered by major layers ou of heir home markes aced as a barrier o enry o smaller layers wihou a home base where hey could charge higher rices o recou heir coss. Following reresenaions from he Big 6, hey were allowed o comee hrough secial offers o arac new consumers, so long as hese were emorary; as rediced 7, such secial offers resuled in a roliferaion of ariffs, and concerns ha consumers, aricularly vulnerable grous, migh no fully undersand he emorary naure of hose offers, since when hey exired consumers were generally reurned o higher defaul ariffs. The 2008 suly robe also inroduced oher measures o imrove comeiion and remove barriers o swiching, including an annual saemen o rom consumer awareness, igher rules on mis-selling of energy and resricions on how far comanies could reven swiching by consumers who owed hem money. However afer accumulaed evidence ha he marke had been damaged by he NDCs (Hviid and Waddams Price, 2012; Lilechild, 2012) he regulaor reversed is decision o renew he clauses in 2012, bu announced coninued vigilance in his regard and inroduced a number of oher consrains on ariffs o simlify choices for consumers. In his aer we examine he evoluion of elecriciy rice movemens of one major ariff since 2005, in aricular he ineracion beween differen firms in he marke. Prices doubled beween 2005 and 2013 (from abou 250 o 500 a year) and have recenly 6 The regulaor simulaneously imlemened a Euroean direcive, requiring ha differences in he erms and condiions offered in resec of differen aymen mehods is cos reflecive, in license condiion SLC 27.2A. While his is no he focus of our analysis, we commen briefly on is oenial effec in he conclusions. 7 See for examle he evidence from Waddams (2009) o he Ofgem consulaion. 3 become he focus of considerable oliical aenion (Miliband, 2013; Cameron, 2012). We find ha he aern by which firms se heir rices changed a he ime of he NDC, confirming concerns ha his inervenion has adversely affeced he naure of comeiion in he indusry. This suly side change has been mirrored by falling consumer engagemen in he marke, wih swiching raes halving beween heir eak in 2008 and 2013 (DECC, 2013). The nex secion resens descriive saisics on rice changes in he residenial marke and exlains he daa, and heir limiaions. Secion 3 uses causaliy ess o idenify rice leadershi, and secion 4 discusses he olicy imlicaions and concludes. 2. Tariffs and daa Since 2005, he level of reail energy rices has risen in real erms, bu wih some decreases as well as increases. This rising rend in rices is shown in figure 1 for each of he main suliers, using an unweighed average across all regions of he annual bill of a direc debi elecriciy consumer on he sandard ariff, using a medium quaniy. Figure 1: Annual elecriciy charges by Big 6 for consumer aying by direc debi and using 3300kWh er annum (deflaed using CPI, average across regions) Daa Source for all grahs and analysis: Consumer Focus Price Comarison Facshees and auhors calculaions 4 Figure 2 shows hese daa groued according o ye of sulier in each region, namely incumben (which varies beween regions), cheaes and median offer from among he res of he Big 5 (i.e. oher han Briish Gas) and Briish Gas. To disinguish beween hese comanies and smaller firms who have enered he indusry from ouside, we label hese large comanies wih regional elecriciy incumbency regions as majorsaway when oeraing ouside heir home areas. The ga beween he average incumben and median majoraway bills before 2008 illusraes he background o he regulaor s inroducion of he NDC which revened suliers from charging higher margins in areas where i was incumben han where i was ouside is home region. The regulaor had found ha in he eriod leading u o 2008 suliers had charged around en er cen more in incumben areas (where consumers sayed wih i as defaul rovider unless hey swiched rovider) han in oher areas (where he majoraway had o em consumers away from he incumben rovider in ha region). Of course he ideniy of he cheaes (and median) majoraway varies boh beween regions and across ime eriods, since we are ineresed in he bes challenge o he incumben a any one ime. Figure 2a: Annual elecriciy charges for a consumer aying by direc debi and using 3300kWh er annum (deflaed using CPI, average across regions) 5 Figure 2b: Annual elecriciy charges for consumer aying by direc debi and using 3300kWh er annum (deflaed using CPI) Figure 2a shows he convergence of rices following he imosiion of he NDC in 2009, and hese reducions in differenials are shown in more deail in figures 3 o 5. We see his aern even more clearly in some individual regions, for examle he Easern region, shown in Figure 2b. Price differences beween incumben and he bes majoraway sared o fall in early 2009 and reached heir lowes oin, where hey remained for he following wo years, around Aril 2011 (figure 3). The difference beween he charges of he incumben and Briish Gas changed considerably over he eriod. Afer a eriod of relaively high rices, Briish Gas made significan rice reducions in 2007 o close he ga wih incumbens, and by early 2008 is rices became more comeiive han hose of incumbens. The average incumben-briish Gas rice sread has sayed above zero and relaively fla since hen (figure 4). The Briish Gas sraegy is refleced in rice differences beween Briish Gas and he bes majoraway (figure 5), wih a decreased ga afer he rice cus, and a furher decrease in rice differences afer he imosiion of he NDC (figure 5). 6 Figure 3 Difference in annual elecriciy charges beween incumben and bes riced majoraway (consumer using direc debi aymen and 3,300 kwh er year, deflaed using CPI), average across regions Figure 4 Difference in annual elecriciy charges beween incumben and Briish Gas (consumer using direc debi aymen and 3,300 kwh er year, deflaed using CPI), average across regions 7 Figure 5 Difference in annual elecriciy charges beween Briish Gas and bes riced majoraway (consumer using direc debi aymen and 3,300 kwh er year, deflaed using CPI), average across regions We exlore he imlicaions of hese differences for comeiion in he reail marke. Hviid and Waddams Price (2012) showed ha because he srong marke of each of he major layers (where hey were radiional incumbens) coincided wih he weak markes of comeiors (where hey had no incumbency base), and Briish Gas ook a naional aroach, he NDC was likely o lead each comany o concenrae on is home markes and comee less aggressively in ohers, resuling in higher rices. Iniial high level evidence suggess ha his is recisely wha has haened. Price differenials beween sandard ariffs have indeed fallen; since comanies were allowed o offer secial offers, hese increased dramaically, enabling he comanies o segmen he marke, and focus heir comeiive acions away from he sandard ariffs. Swiching raes have fallen, confirming widesread evidence (Ofgem, 2008,.152, Giuliei e al., 2005, Flores and Waddams Price, 2013, Waddams Price e al., 2013) ha oenial rice gains are he main drivers of consumer aciviy in he residenial energy marke. Published figures a an aggregae level, and monhly figures obained from he regulaor a regional level, show a dramaic fall in swiching raes following he inroducion of he non-discriminaion clauses. If he oher olices which he regulaor inroduced a he same ime o remove barriers o swiching were effecive, hese figures may undersae he fall in swiching due 8 o he NDC alone. On he oher hand, comanies gradually wihdrew from direc markeing from 2011 onwards, which will iself resumably have reduced swiching raes. Because of he volailiy of usream coss, aricularly wholesale energy rices, i is difficul o idenify direcly wheher rice differenials beween incumbens and majorsaway narrowed because incumben rices fell (as he regulaor inended) or because majoraway rices rose as comeiive consrains weakened. However figures from he regulaor on he rofiabiliy of he Big 6 seem o indicae he laer, as he rofi margins rose from levels near zero when he clauses were inroduced, o margins of jus over 100 er consumer in 2013 (Ofgem, 2013). In his aer we exlore he changes in comanies ricing behaviour which followed he inroducion of he ackage of reforms arising from he 2008 Energy Suly Probe and he NDC. We do his by analysing he aern of rice behaviour amongs he big six comanies for heir sandard elecriciy cusomers (off line, of average quaniies, and aying by direc debi, he aymen mehod used by mos consumers swiching energy sulier). These daa are aken from rice shees ublished by Consumer Focus which show he main ariffs for he Big 6; while laer informaion includes online ariffs, he earlier ublicaions do no, so we resric our analysis o offline ariffs. Since he non-discriminaion ariffs rovided incenives for comanies o use secial offers o comee (because hese were no ied o rices charged in home markes), comeiion shifed somewha o hese non-sandard ariffs, which are no ublished. To he exen ha comanies no longer exec o recrui new consumers hrough sandard ariffs, our analysis will herefore oversae any damening effec on comeiion; we reurn o his in our conclusions. 3. Paerns of rice changes and causaliy ess This secion analyses differences in suliers ricing behaviour; o exlore he effec of he NDC, we divide he daa ino wo subsamles, before and afer he inroducion of he clauses. We omi he eriod from July 2008 o Augus 2009 as a ransiion eriod, since discussions abou he new condiions sared around June 2008 and were already affecing differenials before he ublicaion of he Probe in Ocober 2008, bu were no formally imosed unil Seember Model Secificaion Figures 1-5 above show he variaion in rice differences beween suliers. To exlore how suliers rice reail elecriciy relaive each oher, we fi a Vecor Auoregressive Model (VAR) model o esimae he relaionshi beween rice changes (including changes in 9 revious eriods) of he incumben, Briish Gas and he bes riced majoraway 8 in each region. An advanage of using he VAR model is ha i reas all variables as being endogenous and does no require rior resricions on he srucure of he model. Effecively, we allow he daa o seak, and he ricing relaionshis beween he suliers idenified by he VAR model should be a good saisical reresenaion of he underlying ricing model. A sandard VAR, wrien as, is a model in which variables are secified as linear funcions of of heir own lags, lags of he oher variables, and addiional exogenous variables We secify a -order VAR model of rice changes inroduced by he incumben, Briish Gas and he bes riced majoraway, wih exogenous variables including seasonal dummies and a ime rend as follows: y v Ay 1... A y B 1 0 x u (1) M B I M B I Where y (,, )' is a 3 1 vecor, and,, are he rice changes of he bes majoraway, Briish Gas and incumben, resecively, in eriod ; A1 o A are 3 3marices of he arameers o be esimaed; x is a 4 1vecor of exogenous variables, namely quarerly dummy variables and a ime ' rend; B is a 3 4 marix of coefficiens; 0 v is a 3 1vecor of arameers (consans); and u is assumed o be whie noise. If he u are disribued wih a zero mean, i. i. d. vecor rocess, and y and x are covariance saionary and are no correlaed wih he, consisen and efficien esimaes of he, he and are obained via seemingly unrelaed regression, yielding esimaors ha are asymoically normally disribued. Since all he equaions for he variables y have he same se of regressors on he righ hand side, equaion-by-equaion OLS esimaes are equivalen o he condiional maximum likelihood esimaes. In he above VAR model, effecively we have 3 equaions esimaed, namely of he curren rice change of he bes majoraway, incumben and Briish Gas u 8 Since our analysis concerns he behaviour of he Big 6, who conrolled 98% of he marke during his eriod (100% of i for mos of i) we omi oher enrans from his analysis. The ideniy of he bes majoraway is likely o change from ime o ime, so his does no necessarily reresen changes in rices offered by a single comany, unlike rice changes for Briish Gas and he incumben. 10 resecively, wih he same regressors, i.e. he rice changes of he same layers and he same exogenous variables in each equaion on he righ hand side. To deermine he selecion of lag lengh (he value of ), we use a sequence of log likelihood raio (LR) ess using a maximum of four lags 9. We also imlemen a Lagrange-mulilier (LM) es for auocorrelaion in he residuals of he VAR models (as resened by Johansen, 1995) o ensure ha he disurbance is no auo-correlaed. Tes saisics are reored in he aendix. LM ess wih a maximum five lags are imlemened ha show no hin of model missecificaion for all regions. Saisics for wo lags are reored in he aendix. Noe ha before we esimae he VAR model, Dicky-Fuller uni roo ess confirm ha all rice changes are saionary in each region. Prior o he VAR model secificaion, we also conduced coinegraion ess and do no find robus evidence of coinegraion of he rice series of he suliers 10 in he regions in each eriod. Afer fiing he VAR model, we conduc Granger causaliy ess 11 o es he saisical significance of causaliy from one sulier s rice changes o anoher. For insance, I M in columns 2 and 3 of ables 1 and 2 reors wheher as rice changes of he incumben influence he curren rice changes of he cheaes majoraway. 3.2 Resuls Table 1 shows he resuls of he VAR model and Granger causaliy ess in Period 1, before June The coefficiens reored measure he effec of one se of rice changes on he rice changes being exlored, while conrolling for all oher variables (including revious rice changes by boh he sulier iself and oher acors in he mar
Search
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks