Price discrimination in the British retail market for gas and
electricity
Claude Crampes1 and Mathias Laffont2
Incomplete draft
June 17, 2015
1
2
Toulouse School of Economics
Toulouse School of Economics
Abstract
We revisit the 2009-2012 episode in which Ofgem imposed a non discrimination clause
to large retailers in the UK energy market. Standard Licence Condition 25A was introduced to prevent suppliers from charging higher prices to their incumbent customers,
compared to their non-incumbent customers. SLC 25A included a “sunset clause”, to
allow the condition to lapse three years after its implementation. Several IO economists
protested that the prohibition of spatial price discrimination would eventually lead to
competition weakening. At the end of the three-year period, Ofgem decided not to reinsert SLC 25A for any period. To have a better understanding of Ofgem’s motivations
and the arguments of its opponents, we build a model where two local monopolists
compete on a third market. We determine conditions where the obligation to fix the
same price in and out of the local market can result in a better or a worse situation for
consumers.
JEL codes: L11, L41, L94, Q41
Keywords: Price discriminarion, Energy retail, Competition policy
1
Introduction
Started during the Thatcherian era, the liberalisation of the UK energy industry is
supposed to be achieved since the full opening of retail to competition, including for
residential consumers in 1999. Nevertheless, fifteen years later, the industry regulator
(Office of Gas and Electricity Markets, Ofgem), seems strongly dissatisfied with the
way retailers behave, in particular with the pricing schemes they apply. In June 2014,
Ofgem referred the retail energy market for full investigation to the Competition and
Market Authority (CMA)1 to make the market simpler, clearer and fairer, to encourage
more new entrants and increase competitive pressure on prices. This action follows a
series of attempts by Ofgem to restrict retailers’ freedom.2 Our paper focusses on one of
these regulations, the non-discrimination condition between regions (Standard Licence
Conditions SLC 25A). It has been implemented in September 2009 for a three years test
and abandonned in 2012, in particular because of a campaign by economists claiming
its anticompetitive effects.3
TBC: relevant references on price discrimination
In the first section, we present the UK energy retail market and the tentative by
Ofgem to constrain the pricing strategy of sellers. The second section provides a basic
model of price competition in a framework that mimics the main features of the UK
market. We show that it is very difficult to know ex ante whether price discrimination
is good or bad for consumers. Then, even though it appeared ex post that Ofgem got it
wrong with its non-discrimination condition, it was right in checking the consequences
of a ban accompanied by a sunset clause.
2
Pricing on the British retail market for energy
To understand the offensive of Ofgem against the pricing policy of the big energy retailers, we must first present an overview of the present state of the industry and its history.
We then explain the arguments used by Ofgem to identify anti-competitive practices
and apply price control remedies. Finally, we introduce the counter-arguments of the
opponents to the price regulation that was implemented.
2.1
The Six Large Energy Firms
The retail market for gas and electricity is currently dominated by six firms, often
referred to as the Big 6, with over 90% share of domestic customers: British Gas, EDF
Energy, npower, E.ON UK, Scottish Power, and SSE. These companies are the heirs
1
From 1 April 2014, CMA took over most of the functions of the Competition Commission (CC) and
the Office of Fair Trading (OFT).
2
Littlechild (2014) provides a summary of the Ofgem interventions to promote competition in energy
retail.
3
One of the most virulent opponents to Ofgem interventions is Stephen Littlechild who was Director
General of Electricity Supply and Head of the Office of Electricity Regulation (Offer) during the period
1989-1998. After merging with Office of Gas Supply (Ofgas), Offer became Ofgem in 2000.
1
Figure 1: Domestic electricity supply market shares.
Source: Ofgem (2014), ”State of the Market Assessment”.
of the 12 Regional Electricity Companies (RECs) created in 1988 by the reorganisation
of the nation-wide Central Electricity Generating Board and the Area Boards that
purchased electricity (almost all of it from the CEGB) and distributed and sold it to
customers within their designated areas.
The market being deregulated, firms considered profitable to vertically integrate
production and supply activities through a series of mergers. The partial or total acquisition of RECs by power generators gave them supply capability. The result of a
long series of take-overs and mergers (some by French, German and Spanish utilities)
is an oligoply made of six producers/retailers, each with medium market power at the
national level. Figure 1 shows that in the nation-wide electricity retail activity, each big
firm has between 10 and 25% market share, whereas small suppliers hardly reach 5%
altogether.
Electricity being a perfectly homogeneous product, and knowing that ”two is enough”,
one would expect strong competition, resulting in low margins with prices more or less
uniform across the UK. Actually, competition does not work as it should. This is because each of the Big 6 remains dominant in its historical selling zone, as shown in
Figure 2 (except for British gas).4 Ofgem “observed differences in prices between dif4
The map in Figure 2 is from the website of ecotricity, one of the small suppliers that compete
against the six large energy firms. (http://www.ecotricity.co.uk/for-your-home/check-our-prices/your-
2
Figure 2: Dominant regional suppliers.
Source: www.ecotricity.co.uk/for-your-home/check-our-prices/your-regional-supplier
ferent domestic customer groups” (”State of the Market Assessment” p. 10, 27 March
2014). In particular, “Customers on single fuel tariffs with their legacy supplier are
likely to be paying higher prices than those who have switched to another supplier, either
to single or dual fuel tariffs.” (ibid.).
If a significant minority of consumers are stuck to their legacy supplier, competion
is only for the non-stuck consumers, and differential pricing is a natural outcome. But
Ofgem does not like this outcome and has tried repeatedly to fix it. The next section
presents one of these attempts.
2.2
2.2.1
Ofgem and the economists
Ofgem attacks
In ”Energy Supply Probe - Initial Findings Report” (6 October 2008), pp. 8-9, Ofgem
expressed concern as regards differential pricing. In particular, “Until very recently,
the five former incumbent electricity suppliers charged electricity customers in their
former monopoly areas an average of over 10 per cent higher prices than comparable
“out-of-area” customers. (...) Based on data provided to us by the companies, we
regional-supplier)
3
can find no cost basis for this premium, nor are similar premiums found in gas.5 ...
Overall, these price differentials mean that companies charge more to existing (“sticky”)
customers whilst maintaining competitiveness in more price sensitive segments of the
market. The ability to price differentially in this way means that pressure on prices
in the most competitive segments of the market does not always constrain prices for all
other consumers. There is evidence in the companies’ business plans and from interviews
with the Big 6 that they are aware of these dynamics and take them into account in their
pricing decisions.”
As a consequence, Ofgem considered placing a new condition in the licences of the Big
6 suppliers that would either impose a prohibition on price discrimination or introduce a
form of relative price control. To address ”undue discrimination”, it eventually decided
to impose energy suppliers a licence condition that would limit the differences in the
tariffs that they can charge to their ‘in-area’ customers compared with their ‘out-ofarea’ customers. This is standard licence condition SLC 25A, starting September 1,
2009. The condition included a three-year sunset clause, to allow it to lapse three years
after its implementation.6 Coming from an entity in charge of citizens’welfare, we can
imagine that the expected outcome of this regulation was at least a price decrease for
‘in-area’ customers at no cost for the others.
2.2.2
The economists’ campaign
Condition SLC 25A has been strongly criticized, particulary by S. Littlechild.7
“All these proposals are based on an elementary economic fallacy: the assumption
that the range of tariffs available in a competitive market would remain unchanged if
new obligations of the kinds proposed were introduced. Patently they would not. If a
supplier were required to put all its customers on its best tariff, it would withdraw that
’best’ tariff quicker than a politician could say ’philosopher’s stone’.
To illustrate, [a] Times article suggested that the average annual dual fuel bill is
about £1300 and the cheapest dual fuel tariff is about £1165 , a difference of about
£135 . It said that roughly 75 per cent of customers are on standard tariffs and 25 per
5
Ofgem also explains that a number of the price differentials between payment types do not appear
to have a cost justification - particularly for those customers who pay by standard credit. Another
complain is that suppliers compete vigorously in the online market with heavily discounted offers, the
cheapest of which may be below cost.
In our paper, we focus on the spatial pricing dimension, even though the two other problems (payment
mode and internet contact) can be interpreted using the same type of IO model.
6
See the text in the Appendix.
Also see ”Guidelines on Cost Reflectivity between Payment Methods and the Prohibition of Undue Discrimination in Domestic Gas and Electricity Supply Contracts”, 7 August 2009, https://www.ofgem.gov.uk/ofgem-publications/85032/2-19guidelinesoncostreflectivityandunduediscriminationinsupply.pdf
7
The Littlechild′ s opinion is now detailed in Littlechild (2014). The main arguments he presented during the period 2009-2012 to criticize the non-discrimination clause are summarized in
http://www.iea.org.uk/blog/ofgem-and-the-philosophers-stone.
Other economists were among the cons, in particular Catherine Waddams Price, George Yarrow (who
resigned from the Gas and Electricity Markets Authority because of the Ofgem decision) and John
Vickers (a former director of the Office of Fair Trading).
4
cent on cheaper ones. Consider a major supplier with 3 m customers on standard tariffs
and 1 m customers on cheaper ones. Which would be more profitable: to cut prices by
£135 to 3 m customers, at a loss of revenue of £405 m per year, or to withdraw the
cheap tariffs and lose 1 m customers that were of only borderline profitability? This is
a no-brainer.”
Ofgem’s policy actually made the retail market less competitive than it was formerly.
Waddams Price and Zhu (2013) “find that the nature of competition in the industry has
changed, with less effective rivalry between the regional incumbents and large regional
competitors following the intervention; companies seem to have ‘retreated’ to their home
regions, leaving a market where pricing behaviour resembles more closely a duopoly
between British Gas and the regional incumbent.”
2.2.3
Ofgem surrenders
This no-discrimination provision expired on 31 July 2012. In advance of the expiry
of SLC 25A, Ofgem issued a consultation on whether it was appropriate to retain the
condition for a further two year period. Following the consideration of responses to the
consultation, the Gas and Electricity Markets Authority (the governing authority of
Ofgem) decided not to re-insert SLC 25A for a further two-year period or for any other
period.8 Apparently, this last decision was not clear enough since on 18 December 2014,
Ofgem felt necessary to send a letter to ’Domestic suppliers, consumer representatives
and other interested parties’, reiterating that “Standard Condition 25A is no longer in
effect. The provision expired on 31 July 2012 and suppliers are not bound by it in any
way. We are not considering reintroducing such a licence condition, nor would this be
appropriate during the CMA’s investigation.”9
3
A model of price competition with spatial differentiation
Whether price discrimination is good or bad for competition and welfare is a difficult
question with no clearcut answer (see Armstrong, 2006). Using the tools of IO, we
build a model of price competition with spatial differentiation to highlight the procompetitive and anti-competitive consequences of a prohibition of price discrimination
in a framework similar to the UK energy retail market but simple enough to give clearcut
answers. We assume there are three markets and two firms: each firm is a monopolist
on its local market and it competes against the other on the third market. We then
successively determine the price equilibrium in the case where pricing is free and in the
case where each firm must fix the same price in its local market and in the competitive
segment. We finally show that, even in this over-simplified framework, one cannot
predict whether banning price discrimination is good or bad for consumers.
8
Decision on Standard Condition 25A in the Gas and Electricity Supply Licences, 26 October 2012, https://www.ofgem.gov.uk/ofgem-publications/74952/decision-standard-condition-25a-gasand-electricity-supply-licences.pdf
9
www.ofgem.gov.uk/ofgem-publications/92094/standardcondition25ainthegasandelectricitysupplylicences.pdf
5
3.1
Hypotheses and notations
Two firms sell a homogenous product in three markets:
• market a is the historical ground of firm a. Demand is given by the linear function
qa = a − p2a .
• market b is the historical ground of firm b. Demand is given by the linear function
qb = b − p2b where a ≥ b.
• market c is a versatile market where firms a and b can compete. All consumers
have the same willingness to pay c, where c < b. They are uniformly distributed
along a segment of length 1. Firm a (resp. b) is installed at the left (resp. right)
end of the segment. The firms fix FOB prices and consumers have to pay the
extra cost of ’travelling’ towards their preferred supplier at a cost of t per unit of
distance, with t ≤ c. A consumer who does not buy has a zero net utility.
Because of large switching costs, it is too costly for a (resp. b) to enter the local
market served by b (resp. a). Competition, if any, is limited to market c.
Firms have zero production and selling costs.
3.2
3.2.1
The unconstrained equilibrium
Local markets
On market i (i = a, b), firm i is a monopolist. It solves maxpi pi qi (pi ) = pi i −
can easily compute
pi
2
. We
• the monopoly price
pm
i =i
(1)
i
2
(2)
• the monopoly quantity
qim =
• the monopolist’s profit
m
pm
i qi =
i2
2
(3)
• the global surplus
m
um
i−
i − cost = 2qi
qim
2
3
− 0 = i2
4
(4)
• the consumer’s surplus
3 2 i2
i2
m m
um
−
p
q
=
i
−
=
i
i i
4
2
4
6
(5)
3.2.2
The competitive market
The consumer located at distance x from the left end of the segment (where firm a is
installed) enjoys the net utility c−pa −tx if she buys one unit at a’s and c−pb −t (1 − x)
if she buys at b’s. We assume that c is large enough so that all consumers can buy one
unit either at a or at b, even though only one firm is active.
* To indentify the constraint created by this requirement, suppose that only firm a is
active in market c. Then the marginal consumer is located at x such that c−pa −tx = 0,
and since consumers are distributed uniformly with density 1, x also is the quantity
a
demanded to firm a. Then, as long as x = c−p
≤ 1 the best choice of a is the solution
t
to
c − pa
(6)
max pa
pa
t
c− c
The resulting monopoly price is 2c , so that the marginal consumer is located at t 2 = 2tc .
Since the consumer most remote from 0 (where firm a is installed) is located at 1, to be
sure he will be served we assume that
c
≥1
2t
(7)
Under this assumption, if let alone, the best choice of firm i is to fix
pcm
i =c−t
(8)
which gives it the whole market of size 1.
* When both firms are active, the marginal consumer is located at distance x such
that c − pa − tx = c − pb − t (1 − x) . Because of the uniform distribution hypothesis, x
a
also is the quantity demanded to firm a. Let us denote it by qa (pa , pb ) = 12 + pb −p
and,
2t
p
−p
1
a
b 10
similarly, qb (pa , pb ) = 2 + 2t .
Solving for
max pi qi (pi , pj ) = pi
pi
1 pj − pi
+
2
2t
i = a, b j = a, b i = j
we obtain the best response function of firm i, Ri (pj ) =
N
N
N we can deduce
equilibrium pN
a = Ra pb , pb = Rb pa
t+pj
2 .
(9)
From the Nash
• the equilibrium duopoly prices
N
pN
a = pb = t
(10)
1
2
(11)
• the duopoly quantities
qaN = qbN =
10
Actually, qi (pi , pj ) = min
1
2
+
pj −pi
2t , 1
. We see that qi (pi , pj ) = 1 if pj ≥ pi + t.
7
• the dupolists’ profit
t
N
pN
i qi = , i = a, b
2
(12)
• the global surplus
cN
u
1
2
=2
(c − tx) dx = 2
0
c
t
−
2 8
=c−
t
4
(13)
• the consumer’s surplus
1
2
2
0
5
(c − tx) dx − 2piN q iN = c − t
4
(14)
Clearly, all profits in this market come from the differentiation index (cost of transportation) t. If t = 0, we have Bertrand competition (price = zero marginal cost).
Keeping the hypothesis (7) , all consumers in market c are served and the higher t,
the lower the consumers’ surplus and global surplus, the higher the profits and the
deadweight loss.
3.3
3.3.1
No price discrimination
Constrained duopoly prices
N
The regulator considers that price discrimination pm
i = i > pi = t is harmful for
consumers. She imposes firm i to fix the same price in its historical market where
t+pj −pi
.
demand is qi = i− p2i and in the competitive market where demand is qi (pi , pj ) = 2t
Then, if the two firms are active in market c, firm i solves
max i −
pi
pi
t + pj − pi
pi +
pi
2
2t
i = a, b j = a, b i = j
t+p −2p
From the FOC (i − pi )+ j2t i = 0, we derive the best response function Ri (pj ) =
2ti+t+pj
N
N
N
N
2(1+t) . The Nash equilibrium is defined by pa = Ra pb , pb = Rb pa . Solving for
the two best-response functions, we obtain the equilibrium non discriminatory prices
pN
i =
3.3.2
(3 + 2t) t + 4t (1 + t) i + 2tj
4 (1 + t)2 − 1
i = a, b j = a, b i = j
(15)
Sharp competition in market c
We can obtain a simple global view of the effects of the non discrimination clause in the
case where t = 0. With a zero transaction cost, market c is a battlefield for Bertrand
competition.
a) Absent the regulatory constraint on prices, from (10), we obtain pN
i = 0. It results
i2
m
that the overall profit of firm i is exclusively coming from its local market: pm
i qi = 2
8
b) Under the obligation to fix the same price on markets i and c, firm i has two pure
strategies:
i2
m
b1) To quit market c, which guarantees pm
i qi = 2 from its home market i;
b2) To stay in market c, with a profit that depends on the decision of its competitor.
If the competitor stays in, the price is given by (15) with t = 0, that is pN
i = 0, which
means a nil global profit. If the competitor leaves market c, from (8) firm i can at best
cm
cm i − pi
fix pcm
= c. The resulting profit is (pcm
= c 1 + i − 2c . Then,
i
i × 1) + pi
2
if αj is the probability of j leaving market c, the expected profit of i when staying in
market c is Ei = αj c 1 + i − 2c .
We see that firm i is indifferent between serving only market i at the local monopoly
cm = c if
price pm
i = i or serving the two markets at the non-discriminatory price pi
2
i
c
2 = αj c 1 + i − 2 , that is if firm j abandons market c with probability
αj =
i2
c (2 + 2i − c)
(16)
Therefore, we can have an equilibrium in mixed strategies where each firm i, j = a, b,
i = j randomizes its decision to leave market c with probability (16) . The equilibrium
in mixed strategies remains to be analyzed.
3.4
The effects of the no-discrimination rule
To keep things simple, in the following we keep on assuming that t = 0 and we add the
following assumption:
a2
c
b2
− ac > c 1 −
>
− bc
(17)
2
2
2
i.e., firm a (resp. b) is better off (resp. worse off) when serving only its local
market. When these two inequalities are satisfied, the no-discrimination constraint has
the following effects
* to push firm a out of market c where firm b becomes a de facto monopolist
facing a (endogenous) price cap constraint;
* to leave the situation on market a unchanged
cm
* to decrease the price on market b from pm
b = b to pb = c
* to increase the price on market c from pbN = 0 to pcm
b =c
The two last effects are not a surprise. What about the overall effect?
We can use essentially two gauges to assess the global effect of the no-discrimination
rule: a normative one by comparing the levels of global surplus and consumer’s surplus,
and a practical one by comparing the average prices.
3.4.1
Surplus index
Before the constraint on price is imposed, total welfare is
3
3
m
cN
W D = um
= a2 + b2 + c
a + ub + u
4
4
9
(18)
Under the no-discrimination rule, total welfare becomes
3 2
c2
cm
cm
2
W R = um
a + ub + uc = a + b −
4
4
+c
(19)
We clearly have W R > W D since the regulation leaves the quantities sold unchanged
in markets a and c whereas trade increases in market b. Then from the global point of
view, in our oversimplified model, the obligation to sell at the same price in and out of
area is a good intiative.
However, the rule was implemented in the interest of consumers. What can we say
about their surplus?
Absent the constraint on price
m m
m
m m
cN
N
− 2pN
CS D = (um
=
a − pa qa ) + (ub − pb qb ) + u
i qi
a2 b2
+ +c
4
4
(20)
Under the no-discrimination rule, the consumers’ surplus becomes
m m
cm
cm cm
cm
cm cm
CS R = (um
a − pa qa ) + (ub − pb qb ) + (uc − pc qc ) =
c
a2
+ b−
4
2
2
+ 0 (21)
We see that the consumers’ surplus is unchanged by regulation in market a, it decreases
in market c and it increases in market b. Therefore, there is no guarantee that the
2
2
regulation has positive effects for consumers. We see that CS D −CS R = b4 +c− b − 2c
that we can rewrite as
CS D − CS R = c 1 −
c
−
2
b2
− bc
2
−
b2 − c2
4
By (17), the term into brackets is positive but after substracting
(22)
b2 −c2
4
> 0, the
sign of CS D − CS R cannot be certified. For example if b = 1, c = 12 , we obtain
CS D − CS R =
3
16
(23)
1
9
(24)
whereas if b = 1, c = 13 , we obtain
CS D − CS R = −
In a nutshell, when c = 12 consumers are globally better off without the price regulation and the opposite is true when c = 13 . This is because the consumers’ surplus
in market c is fully confiscated by firm b when the latter is left alone because of the
regulation. Therefore, we have that for a ”high” (resp. ”low”) value of c, the nodiscrimination clause is harmful (beneficial) to consumers.11
11
It is easy to check that when b = 1 the value of c that changes the sign of CS D − CS R is c ≃ .4.
10
3.4.2
Average price
The surplus index cannot be calculated without a good knowledge of the willingness-topay of consumers. Regulators may rather use an observable index, such as the average
i
m
m
price p = pq
q . When prices are freely fixed, we have pi = i, qi = 2 on the local
1
N
markets and pN
i = 0, qi = 2 on the shared market (i = a, b) . The average price is then
a2
2
D
p =
a
2
+
+
b2
2
b
2
+1
(25)
a
m
cm
Under the no-discrimination regulation, we have pm
a = a, qa = 2 on market a, pb =
c
m
cm
cm
c, qb = b − 2 on market b, and pb = c, qb = 1 on market c. Then the average price
R
p =
a2
2
a
2
+ b − 2c c + c
+ b − 2c + 1
(26)
Observe first that more consumers are served under the no-discrimination requirement. Indeed, markets a and c have the same quantity sold whereas consumption
increases in market b where the price has decreased. Then, the denominator is larger in
pR than in pD .
Now consider the expenditures, that is the numerator in the two average prices. In
2
market a, the regulation leaves consumers’ expenditures unchanged at a2 . In market
2
b, expenditures decrease (from b2 to b − 2c c), while they increase in market c (from
0 to c). Nevertheless, taking together markets b and c, expenditures increase since this
higher revenue is the reason why firm b remain active in market c despite the price
constraint (see the second inequality in (17)). It results that the numerator is larger in
pR than in pD .
As a consequence, pD can be larger or smaller than pR depending on the values of
the parameters a, b and c.
For example, assume that a = 2 and b = 1.
* If c = 12 ,
1
pD − pR = −
(27)
22
* If c = 13 ,
4
(28)
51
which confirms the result we have obtained in terms of consumer’s surplus: when
c = 12 , consumers are better off absent the price constraint and it is the opposite when
c = 13 .
Nevertheless, the way we have measured the average price is disputable since it
integrates consumers in market a whereas their situation does not change when the
regulation is implemented. Observe that consumers of market a are canceled out in the
pD − pR =
11
index in terms of surplus variation. If we restrict the computation of the average price
to markets b and c, we obtain
b2
1
D
p = b 2 =
(29)
3
2 +1
when b = 1, and
pR = c
(30)
Then, we can keep the former assertion that for a ”high” (resp. ”low”) value of c, the
no-discrimination clause is harmful (beneficial) to consumers.
However,the threshold values of the parameters differ so that different indexes can
give opposite answers. For example, with b = 1, for 13 < c < .4, we have that pD −pR < 0
meaning that the constraint on prices is bad for consumers and CS D − CS R < 0 which
means the opposite.
4
Conclusion
Apparently, Ofgem was not fully certain that prohibiting spatial discrimination would
benefit consumers, since it had added a sunset clause to its regulation. By contrast,
the British economists who wrote against the Ofgem’s decision were convinced that it
was anti-competitive. Actually, as we have shown in a oversimplified framework, it is
difficult to assess ex ante whether price discrimination is good or bad both globally
and individually for each category of players: incumbents, rivals and consumers. As
Armstrong (2006) states, “Price discrimination can lead to more intense competition
which benefits consumers. When firms have difficulty committing to prices, they often
are forced to charge low prices. In such situations, a policy which forbids discrimination
endows a firm with commitment power and prevents the firm competing with itself, to
the detriment of consumers and welfare.” This is precisely what occured in the British
case. Nevertheless, Armstrong clearly writes that price discrimination can lead to
more intense competition. It may be that competition is not intense enough under price
discrimination. Finally, Ofgem was not totally wrong since the sunset clause allowed
to stop a sort of market experiment showing that price discrimination was better for
consumers than uniform pricing.
12
References
[1] Armstrong M. (2006), ‘Price Discrimination’, chapter 12
cirossi (Ed) (2008), Handbook of Antitrust Economics,
else.econ.ucl.ac.uk/papers/uploaded/222.pdf
in P. BucMIT Press
[2] Cooper J.C., L. Froeb, D.P. O’Brien and S. Tschantz (2005) ”Does price discrimination intensify competition? Implications for antitrust”, Antitrust Law Journal,
Vol. 72, No. 2, pp. 327-373
[3] Littlechild,
S
(2012),
”Ofgem
and
the
www.iea.org.uk/blog/ofgem-and-the-philosophers-stone.
Philosopher’s
Stone”,
[4] Littlechild, S (2014), ”Promoting or restricting competition?: Regulation of
the UK retail residential energy market since 2008”, EPRG Working Paper
1415, Cambridge Working Paper in Economics, www.eprg.group.cam.ac.uk/wpcontent/uploads/2014/09/1415-PDF.pdf
[5] Pollitt, M G and A B Haney (2014), ”Dismantling a Competitive Retail Electricity
Market: Residential Market Reforms in Great Britain”, Electricity Journal, 27 (1),
January: 66-73.
[6] Thisse J.F and X. Vives (1988), ”On The Strategic Choice of Spatial Price Policy”,
The American Economic Review, Vol. 78, No. 1., March: 122-137.
[7] Waddams Price C. and M. Zhu (2013), Pricing in the UK retail energy market,
2005 — 2013 , CCP Working Paper 13-12, competitionpolicy.ac.uk
13
5
Appendix: Condition 25A. Prohibition of undue discrimination in supply
source: https://epr.ofgem.gov.uk/Content/Documents/
Electricity supply standard licence conditions consolidated%20- Current%20Version.pdf
Prohibition of undue discrimination
25A.1 This condition applies in relation to the supply of electricity by the licensee
under a Domestic Supply Contract or a Deemed Contract for the supply of electricity
to Domestic Premises.
25A.2 Subject to paragraph 25A.3, the licensee must ensure that in supplying or
offering to supply electricity, the Principal Terms on which it does so do not discriminate
without objective justification between one group of Domestic Customers and any other
such group. For the purposes of this condition it shall be for the Authority to decide
whether there is any such objective justification.
25A.3 The licensee shall only be in breach of this condition if and to the extent that
the nature of the discriminatory terms offered and/or their impact on any Domestic
Customers is material in any respect.
Compliance with this condition
25A.4 This licence condition shall be interpreted and enforced in accordance with
guidance issued by the Authority and until such guidance is published this licence
condition shall not be enforceable.
25A.5 The Authority may from time to time revise the guidance referred to in
paragraph 25A.4 with a view to:
(a) removing or reducing inconsistencies between Electricity Suppliers in their interpretation and application of its provisions; and
(b) clarifying how the licensees’ compliance with the obligations imposed by this
condition will be monitored and enforced
25A.6 Before revising guidance under paragraph 25A.5, the Authority shall give
Notice that it proposes to do so to:
(a) Electricity Suppliers in whose licences Section B of the standard conditions is
effective;
(b) the National Consumer Council; and
(c) such other persons as the Authority considers it appropriate to consult in relation
to the proposal.
25A.7 A Notice given by the Authority under paragraph 25A.6 must:
(a) state that the Authority proposes to issue the revised guidance and specify the
date on which it intends that this should take effect;
(b) set out the text of the guidance and the Authority’s reasons for proposing to
revise it; and
(c) specify the time (which must not be less than a period of 28 days from the date
of the Notice) within which representations or objections with respect to the proposal
may be made, and the Authority must consider any representations or objections which
are duly made and not withdrawn.
14
25A.8 The licensee is not required to comply with this condition if it supplies electricity to fewer than 50,000 Domestic Customers or such other number as may from
time to time be directed by the Authority.
Termination of this Provision
25A.9 This condition will cease to have effect on 31 July 2012.
15