The Regulatory Policy Institute Research Group

The challenge of removing a mistaken price cap

This is a pre-proof of a forthcoming publication in Economic Affairs. Expected publication: October.


The UK Competition and Markets Authority in 2016 calculated a detriment of £1.4 billion–£2 billion in Great Britain’s retail energy market, attributed to weak customer response. The government in 2019 imposed a tariff cap until competition is effective. I argue that the cap was a mistake: there was no such detriment and there are valid reasons for customers not changing supplier. The market was not previously uncompetitive and inefficient as suggested. The cap has rendered the sector loss-making and led to supplier exit. The assessments of effective competition by the Office of Gas and Electricity Markets have been arbitrary and implausible. Some alternative ways ahead are noted, but latest government policy invokes behavioural economics to propose even greater intervention. A postscript discusses dramatic latest developments.



The UK Competition and Markets Authority (CMA, 2016) found that many British residential energy customers were not switching from higher- to lower-priced suppliers, and attributed this to “weak customer response”. It estimated an annual customer detriment of £1.4 billion–2 billion in the form of excess profits and inefficient costs. Following a political outcry, in 2019 the UK government, under the Tariff Cap Act 2018, introduced a temporary tariff cap on residential energy prices, renewable annually until 2023 at the latest if the Office of Gas and Electricity Markets (Ofgem)1 deemed competition still not effective. The cap was claimed to reduce individual annual bills by some £75–100. Surprisingly, the feared adverse impacts on competition (reduced price differentials leading to reduced customer switching) seemed not to materialise. What was not to like?

This article argues that the tariff cap was a mistake and seeks to correct the narrative on evolving competition, costs and efficiency, the miscalculation of customer detriment, and alleged weak customer response. Customer response was not ‘weak’: other factors including product and supplier differentials and regulatory restrictions could explain customer reluctance to change suppliers. There was no customer detriment of £1.4 billion–2 billion: the CMA used a mistaken calculation, inconsistent with economic theory, with previous competition authority practice, and with its own Guidelines. The adverse impact of the cap on price differentials and customer switching was initially masked by falling wholesale prices, but is now seen to be much greater than realised. It has also rendered the sector loss-making and led to supplier exit rather than entry. This seems likely to jeopardise the investment, innovation and government–customer–supplier partnership that government and Ofgem seek.

Ofgem’s annual assessment of whether competition is effective has been misdirected, including by focusing on ‘fair tariffs’ that may be convenient for regulators but have no economic substance. There is no recognition that the tariff cap itself is the most serious restriction on competition.

If there were any regulatory and political appetite to remove the cap before 2023, then various measures could be considered to facilitate this. These include correcting the narrative as explained; shifting the focus from more switching to better switching and encouraging customer loyalty where deserved; assessing and highlighting the downsides of the tariff cap; gradually phasing it out so there is no longer a credible threat of significant tariff increases once the cap is removed; providing information about evolving levels of costs to inform assessment of evolving prices; and limiting the tariff cap to specified vulnerable customers at a reasonable economic level.

However, the regulatory and political appetite is apparently for the opposite. In July 2021, much influenced by behavioural and ‘nudge’ considerations, the government proposed to extend the cap indefinitely and to promote extensive opt-in and opt-out switching of customers between suppliers – not only from higher-priced to lower-priced tariffs and suppliers, but also to “take consumers on the journey to a net-zero energy system” (BEIS, 2021b, p. 4). Competition is a process for discovering and seeking to provide what customers prefer. But what is now envisaged in the retail energy market is a process for securing the products and suppliers that government and regulator deem appropriate.

Although this is a particular issue in Britain at present, retail energy price caps or related controls were also introduced by the Australian Competition and Consumer Commission at about the same time, amid concerns about price discrimination (ACCC, 2018; Simshauser & Whish-Wilson, 2017; Simshauser, 2018; 2021; Mountain & Burns, 2021). There are still some intendedly transitional price caps or default tariffs in various European countries and US states (Littlechild, 2021b). Of course, ‘temporary’ price controls have a long history, as for example the 1947 New York City rent controls that survive to this day. The appropriate policy for removing (or continuing) them will depend upon the circumstances of each case, but some of the ideas in this article may be of relevance elsewhere.



In introducing its Strategic Narrative concerning its Forward Work Plan, Ofgem (2019a, p. 14) claimed that: “the legacy of many years of a concentrated, uncompetitive retail market is widespread cost inefficiency”. This is incomplete and misleading: it reflects an imagined explanation for the incorrect CMA finding of inefficiency and customer detriment. The market was not particularly concentrated and was never uncompetitive, although there were regulatory restrictions on competition in the period before the CMA review. And although there were significant cost differences between suppliers in 2015, when the CMA looked at the market, this did not constitute ‘widespread cost inefficiency’ as properly understood. Rather, cost differentials and the gradual substitution of lower-cost for higher-cost suppliers are indications of competition as a rivalrous discovery process taking place over time.

To explain briefly, retail competition at the domestic (residential) level was introduced for both natural gas and electricity in the late 1990s. The government’s golden shares in the Regional Electricity Companies (RECs), which prevented takeover, had lapsed in 1995. From then until about 2002 there were many mergers and takeovers reshaping the retail market, including the entry of large generating companies and several overseas companies. Each new owner brought ideas for innovation and greater efficiency. In addition, the RECs and British Gas extended into each other’s markets, and increasingly customers became ‘dual-fuel customers’ taking both electricity and natural gas from the same supplier. The generally held view was that retail suppliers needed to be large to compete effectively, partly to reap economies of scale in retail and partly to sustain their own generation fleets since the wholesale markets were not yet fully competitive. The evidence supports this judgement: several small and independent new suppliers entered the market before 2010 but only half a dozen survived, with aggregate market share never exceeding about 1 per cent.

Nonetheless, this was not a particularly concentrated market – for example, it was about the third or fourth least-concentrated retail energy market in Europe. And it was very competitive. By 2002 Ofgem had phased out its initial transitional price caps for domestic (residential) customers, and until 2007 repeatedly praised the competitiveness of the market, also noting innovations in tariff structure and payment methods.  All players were making significant incursions into the territories of historic incumbent retailers. The supplier SSE went from fourth to second largest, increasing its electricity market share from 14 per cent in Q1 2004 to 20 per cent in Q1 2009, and its gas market share from 9 per cent to over 15 per cent (Ofgem Data Portal, n.d.).  There was a steady increase in the customer switching rate between suppliers, from around 16 per cent per year in 2005 to 20 per cent per year by 2008, as high as anywhere in the world at that time.

Retail prices declined steadily until 2008 and then began to rise sharply, in both cases mainly reflecting movements in wholesale prices. There was then a change in Ofgem personnel and policy. Under pressure to explain the price increases or ‘do something’, Ofgem diverted attention to relative prices (Yarrow, 2014, pp. 4–5). In October 2008 it introduced a non-discrimination condition. Over the next few years, it severely restricted doorstep selling. In 2011, Ofgem indicated its intention to limit suppliers to a maximum of four ‘simple tariffs’. There was pressure to go further: in September 2013 Ed Miliband, leader of the Labour Party in opposition, suggested a temporary (20-month) freeze on energy prices, “to reset the market” (BBC News, 2013).

Although the market did not become ‘uncompetitive’, the regulatory actions nonetheless restricted competition, and by 2013 the annual switching rate had halved to 10 per cent. There was no change in the rank ordering of suppliers by market share during the five years 2008–12. British Gas (which being a national supplier had not previously been able to differentiate its prices between in-area and out-of-area customers) benefited particularly from the new restriction on its competitors, and increased its market share. The aggregate average earnings before interest and tax (EBIT) profit rate of the large legacy suppliers increased from under 1 per cent in 2009 (artificially low because of lagged response to the wholesale cost increases – which does not suggest market power until then) to over 4 per cent in 2012. This increased market profitability was one of the factors (though not the only one, see below) that led to significant new entry from 2013 onwards.

Economists documented the adverse effects of regulatory policy (Pollitt & Haney, 2014; Littlechild, 2014; Waddams Price & Zhu, 2016). In 2014 Ofgem, with a new chairman, referred the energy sector for investigation by the CMA, because of “concerns that competition was not delivering the desired outcomes” (CMA, 2014, p. 1), but not least to assess Ofgem’s own policy. The CMA agreed with Ofgem’s concerns about the market but soon found that Ofgem’s non-discrimination condition had likely contributed to a softening of competition, and that its ‘simple tariffs’ policy had had an adverse effect on competition. In November 2014 Ofgem confirmed that its non-discrimination condition had expired, and in August 2015 indicated its intention to remove the ‘simple tariffs’ restrictions. Switching steadily increased again.

There were not, then, ‘many years of a concentrated, uncompetitive market’. The market was never particularly concentrated by international standards, and it has been very competitive for two decades. Regulatory policy did reduce the effectiveness of competition during 2008–14 but, as shown below, new entry had an increasing impact from the early 2010s.



What about the alleged legacy of inefficient costs? There is no reason why the evolving nature of retail competition since the market opened, or Ofgem’s later restrictions on competition, should have reduced the incentives of the large suppliers to be efficient. On the contrary, an important aim of the mergers in the 2010s was to reduce costs, and in the early 2010s the six large suppliers were actively seeking to increase their efficiency and effectiveness by unifying and modernising the paper-based legacy IT and billing systems of their various component companies. Four of these six suppliers (Scottish Power, nPower, EDF, and British Gas, in roughly that order) adopted the German SAP system, a fifth supplier (E.On) did so in part. At that time SAP was considered the best system available. In the event, however, the customisation and integration of these SAP systems proved significantly more problematic, time-consuming and expensive than expected, and in some cases led to temporary failures in customer service as well as higher costs.

The large suppliers had an additional incentive to increase efficiency because for the first time, also during the early 2010s, new entry on a smaller scale became economic. Improved and lower cost IT and billing and collection systems were being developed, geared to quick and small-scale entry. These various off-the-shelf and ‘supplier in a box’ models provided potential entrants with companies that had been taken through Ofgem and other entry processes, and they could be rented rather than purchased outright. Increasingly they also offered outsourced customer management services and related support infrastructure.

In addition, competition in the wholesale market was increasing, and large wholesale energy providers were now willing to provide low-collateral trading and management of small suppliers’ trading positions. The growth of price comparison websites (PCWs) served to inform, assist and encourage potential switchers, and by the same token new entrants could appear in these comparisons and benefit from their massive marketing. Government and Ofgem took steps to facilitate new entry, including by exempting new entrants from significant social and environmental costs. The number of entrants increased fourfold, from six at end-2010 to 23 in early 2015.

Thus, when the CMA (2016) took its snapshot of the industry, it found most of the large suppliers still hit by the unexpectedly high costs of the SAP systems but still accounting for 90 per cent of the market, and two dozen small and medium suppliers with new and low-cost systems accounting for only 10 per cent of the market. Unfortunately, the CMA interpreted this as a static, concentrated, inefficient and uncompetitive market instead of realising that it was a very competitive market at the beginning of a second phase of radical transformation. The growth of the small and medium suppliers at the expense of the original large suppliers had only just begun and had not yet had time to play out.

Increasingly, this aspect of the competitive process has indeed played out. The aggregate market share of new entrants remained at under 1 per cent from 2004 to 2012 in both gas and electricity, and then steadily increased to 30 per cent by 2019 (Ofgem, 2019b, figures 5 & 6; Ofgem Data Portal, n.d.).

Also, the external and internal restructuring has continued. Thus, a former medium supplier (Ovo) has taken over the one former large incumbent supplier (SSE) that had not attempted to modernise its systems via SAP and has transferred both businesses to its new platform, Kaluza. The market share of the five other former large suppliers, at about 80 per cent in early 2013 and about 60 per cent in early 2019, was down to about 55 per cent in late 2020. E.On, the large supplier that only partially adopted the SAP system, has taken over another large supplier (nPower) and invested in a new customer platform, Kraken. This was developed by Octopus Energy, a supplier that entered the market in late 2015 and had only 1,500 customers in April 2016, as the CMA finalised its report. British Gas Centrica is transferring its customers to a new platform with ENSEK. The remaining small and former medium suppliers increased their numbers from 23 in 2015 to a peak of 64 suppliers in 2018 and tripled their market share from about 10 per cent to about 30 per cent (Ofgem Data Portal, n.d.). Two suppliers (Bulb Energy and Octopus Energy) that had barely entered the market in 2015 are now classed as large suppliers with over a million customers each (and each has ambitions to have 100 million customers worldwide within another ten years).

Thus, despite cost differences at any point in time, the retail energy market of Great Britain has always been, and continues to be, an extremely competitive retail market. In 2015 some new entrants had lower costs than incumbents, but it was not feasible for incumbents to access those lower costs immediately. What has been observed since 2011 is the realistic rate of structural transformation and implementation of cost efficiencies, as driven primarily by competition, rather than the unrealistic idealised benchmark used by the CMA, as I now explain.



The CMA (2016) took the view that energy was a homogenous product – it was the same whoever supplied it. The CMA noted that there were significant differences in the prices charged by different suppliers, and indeed in the levels of different tariffs offered by the large suppliers, and expressed surprise: Why didn’t customers move to a lower-cost supplier, or to lower-price tariffs? This was particularly puzzling because, for lower-income customers, the reduction of their annual energy bill by choosing cheaper tariffs would be quite significant.

The CMA’s explanation was ‘weak customer response’, which could be interpreted to mean that many customers didn’t know what they were doing, or were doing the wrong thing, or not doing anything. The CMA therefore recommended that Ofgem take steps to promote more active customer engagement (in other words, to explain to customers what they should be doing and persuade them to do so).

‘Weak customer response’ was not the invention of that particular CMA panel, or something specific to the retail energy sector. Rather, it was a concept actively developed and promoted in concert by government, competition authorities and Ofgem (Littlechild, 2015). Briefly, in the late 2000s behavioural economics was attracting growing regulatory attention. The Office of Fair Trading (2010, pp. 9–10) referred to the “virtuous circle between consumers and competition” in which “well informed, confident, rational and effective consumers can play a key role in activating vigorous competition between firms” who in turn deliver what customers want. But behavioural biases “can impact on the extent to which consumers play their active, effective, and rational part in this virtuous circle”. Shortly afterwards, the new Coalition Government set up its Behavioural Insights Team (the ‘Nudge Unit’). In introducing its simple tariffs policy, Ofgem (2011, pp. i, 9) explained that it had “implicitly used insights from behavioural economics for many years” to deal with “weak customer engagement”.

In 2011 the government (quoted in Littlechild, 2015, pp. 43–5) proposed “accelerating the shift towards more informed consumers” and creating the Competition and Markets Authority to “play a leading role in achieving the over-arching objectives and delivering the desired outcomes”, including by “joint working between the CMA and sector regulators”, guided by a “strategic Steer”. This government Steer said that “these customer behavioural issues should be central to the CMA’s analysis of whether markets are working well, and where relevant, should inform the remedies it puts in place”. The revised competition authority Guidelines for Market Investigations (Competition Commission, 2013) (subsequently just Guidelines) had a 20-paragraph discussion of weak customer response and mentioned behavioural economics concepts some 50 times.

In 2013 government, CMA and Ofgem worked to develop a joint Assessment Framework for the energy sector. This was followed by a joint CMA–Ofgem State of the Market Assessment in 2014 (Ofgem, 2014), which duly identified weak customer response as a concern. This was followed in turn by an agreed reference to the CMA in which Ofgem duly asked the CMA particularly to investigate weak customer response.

Of course, it was up to the CMA (2016) panel members to make up their own minds, but the agreed view and policy of government, CMA and Ofgem was that weak customer response was a problem. Hence, the main tasks were to quantify the detriment and to recommend behavioural solutions. There is no evidence that the government, Ofgem or CMA (2016) took any notice of, or were even aware of, the strong cautionary advice of OFT (2010, pp. 30–5) against undue regulatory intervention, which concluded that these points “all caution us against being too paternalistic even when behavioural biases point to problems within the market”.2



As noted, the CMA (2016) duly found ‘weak customer response’. This might not have been a problem were it not for the CMA’s quantification of the implied detriment. How then to estimate what would have been the price in a fully competitive market?

The Competition Commission (2014), the CMA’s predecessor, faced precisely that question in its examination of the cement market. It ranked the existing companies in order of increasing unit cost, noted their maximum capacities, and calculated the competitive price as the unit cost of the marginal plant – that is, the cost of the highest-cost plant needed to meet demand. It measured customer detriment by excess profit – roughly, the extent to which market price exceeded that competitive price. This was in accordance with standard economic analysis.

In contrast, the CMA (2016) effectively assumed that the competitive price in the retail energy market was the unit cost of the least-cost plant available in the market. More specifically, it defined the competitive price as the price that would be charged by “a hypothetical construct, a ‘supplier’ that is a combination of the suppliers that we have identified as being the most competitive in the markets” (2016, p. 602). This seems to refer to two new entrant companies with lower costs than most of the other companies. Using this approach, the CMA estimated a customer detriment averaging £1.4 billion a year, reaching almost £2 billion in 2015.

However, this ‘hypothetical construct’ is essentially what the CMA Guidelines call “an idealized perfectly competitive market” (Competition Commission, 2013, p. 11). The Guidelines say explicitly that such a benchmark will not be used. Hence the CMA’s calculation was inconsistent with its Guidelines, with previous practice, and with conventional economic analysis.

As indicated above (also Littlechild, 2020a), the CMA’s assumptions were unrealistic at that time. The larger companies could not have achieved the lower costs of some medium-sized ones, nor could the small and medium new entrant companies, then accounting in aggregate for about 5 per cent of the market, have expanded instantaneously to meet the market demand at the cost levels of the most efficient ones. Moreover, as explained shortly, many customers were not persuaded that other suppliers were as satisfactory as their own supplier. If the CMA had used a more realistic and economically sound approach, understood the history of the market, and acknowledged companies’ actual costs and capacities – and customers’ preferences for staying with their existing supplier – it would not have found such an enormous customer detriment.

In assessing detriment, normal competition authority practice, including at the Competition Commission and the CMA, was and is to give considerable weight to excess profits (but not including alleged inefficiency). Using as a benchmark the profit rate in the acknowledged very competitive market for large industrial customers, excess profits in the domestic market might have averaged about £170 million per year over 2007–14 (Littlechild, 2020a). This was around £6 per household per year, an order of magnitude lower than the £75 per household cited in the CMA report. A finding of about £170 million detriment would have been consistent with the evidence, noted above, that Ofgem’s simple tariffs policy had restricted competition and led to increased profits.

Such a detriment would have been addressed by the CMA’s recommendation that Ofgem withdraw those regulatory restrictions. When Ofgem did so in 2015, the average switching rate did indeed recover and continued to rise until early 2019.

However, the CMA had identified a behavioural problem, and that necessitated a behavioural remedy. The CMA recommended that Ofgem experiment with various methods to encourage more active customer engagement. Which Ofgem duly did, particularly in the form of five Collective Switch trials that invited customers to ‘opt-in’ to a change of supplier (Ofgem, 2019b). What Ofgem might then have done in the absence of a tariff cap is a matter of conjecture, but at the time Ofgem appeared content not to pursue an active opt-in or opt-out switching policy.

A correct estimate of customer detriment certainly did not justify a tariff cap. Indeed, the CMA itself, even having found a £2 billion detriment, opined that a tariff cap was not appropriate. Unfortunately, one dissenting member favoured such a cap precisely because of the magnitude of the calculated detriment. And customer detriment of £2 billion per year, caused by supplier exploitation of weak customer response, was the only message that the media and politicians heard in the run-up to the 2017 general election. The inevitable result was the tariff cap.



Although the CMA (2016) and previous policy held that weak customer response was the most plausible explanation for customer behaviour in the retail energy sector, is that still a tenable argument? Brennan (2007) suggested early on that some customers may prefer not to choose, but in recent years various studies have sought to better understand how customers actually do choose (e.g. He & Reiner, 2018; Flores & Waddams Price, 2018). Customers generally act consistently with conventional economic principles, and searching, evaluating and switching to alternative suppliers is costly (e.g. Giulietti, Waterson, & Wildenbeest, 2014). Ros (2020) finds that US evidence is consistent with customers making rational economic decisions.

In a careful study of The Big Switch (TBS) run by the Consumers’ Association’s Which? magazine, Deller et al. (2021, p. 109) find that

A range of non-price factors – various sources of uncertainty, the non-monetary characteristics of different offers, concerns about the switching process and time pressures when TBS occurred – are all associated with the switching decision. … most of the factors we identify are consistent with consumers making a largely ‘rational’ decision when declining to switch, even if this results in substantial monetary savings apparently being left on the table.

These non-price preferences confirm that consumers do not regard energy as a homogeneous product, despite the view of many analysts. So our second policy conclusion is that actions which automatically move consumers to a cheaper supplier may reduce utility for at least some consumers, since they do not regard suppliers as completely interchangeable.

The following five more informal pieces of recent UK evidence provide further indication that energy products and suppliers are not homogeneous, and that suppliers, comparison sites and customers do not regard them as such.

First, trade association EnergyUK (2020) has introduced a Vulnerability Commitment by which suppliers can pledge additional support to customers in vulnerable circumstances. As of 24 June 2021, 14 suppliers plus seven white labels (such as supermarkets rebranding as their own products energy supplied to them by other suppliers) had signed up, including six of the present seven largest suppliers. In contrast, of the 12 suppliers offering the 20 cheapest tariffs as of 19 June 2021, only five had signed up, and only two out of the six suppliers offering the cheapest ten tariffs had done so (Cornwall Insight, 2021).

Second, leading PCW Uswitch has begun to offer Green accreditation of energy tariffs.3 Again as of 19 June 2021, four of the seven largest suppliers offered standard variable tariffs (SVTs) that met Bronze standard. One did not do so but offered seven other tariffs in the Silver category. (The other two large suppliers’ SVTs were not listed on the Uswitch site.) In contrast, of the 20 cheapest tariffs, seven met Bronze standard, as did only two of the ten cheapest.

Third, an examination of the many retail energy offers listed on Energy Helpline,4 a PCW used by Ofgem, suggested that the savings available to three hypothetical but plausible customers on 4–5 October 2019 were significantly less than first appeared (Littlechild, 2021a, s. 8). The three customers were assumed to have average annual consumption and to be on an SVT with an incumbent large supplier, priced at the tariff cap. Customers paying by direct debit, on receipt of bill and by prepayment meter (PPM) would initially appear to have, respectively, over 150, over 60 and 10 tariffs to choose from that offer significant annual savings of up to £328, £259 and £159, respectively. But suppose they have various different but plausible preferences (e.g. for variable rather than fixed tariff, for paper bills rather than online-only tariff, for a supplier well rated by Which? magazine, for availability of Warm Home Discount). Then the number of tariffs offering annual savings of more than £5 falls to just two for each customer, and their maximum savings reduce to, respectively, £37, £89 and £27 (plus a £50 voucher). Such savings are an order of magnitude lower than the available offers first suggest, and these customers might quite reasonably consider that switching supplier is not worth the effort. Put another way, there are significantly lower prices available, but only for different products than the ones that they actually want, and their lack of switching is not attributable to disengagement or weak customer response.

Fourth, some lower-priced suppliers are riskier or less well regarded, and not without reason. From 2015 to 2017, on average, four out of the ten cheapest suppliers at any time subsequently exited the market (Cornwall Insight, 2020). Similarly, of the cheapest 20 suppliers in the market on 30 January 2021, as listed by Cornwall Insight, nearly half (9/20) were not well enough established to be rated by Which? and Citizens Advice, and none of the cheapest five suppliers were so rated. Moreover, of the 11 suppliers that were rated by these two organisations, over one third (4/11) were on average ranked in the lower half of an index of these ratings (the Overall Customer Satisfaction score described below).

Fifth, customers prefer familiar suppliers as well as lower prices. Uswitch processes about 1.5 million switches per year, about 20 per cent of the UK total. It confirms that price is an important determinant of switching: the switching rate is nearly double for a saving of £300 compared with £150 (Neudegg, 2020). And the proportion of customers that switch to a particular tariff is about twice as high if it is the lowest price tariff rather than the second lowest, and similarly for the third- and fourth-lowest price tariff. This is all consistent with economic rationality. Also, importantly for present purposes, ‘brand matters’. For a tariff in any position in the ranking (i.e. lowest price, second-lowest price, and so on), the switching rate to a large former incumbent supplier is greater than that to a medium supplier, which in turn is greater than that to a small supplier. The proportions seem to be roughly 5:4:2.

In other words, for large suppliers, and to a lesser but increasing extent medium suppliers, ‘brand’ is an intangible asset because of the greater value that customers attach to their product. The large suppliers are better known and more trusted, because they have been around the longest, are known to the most people, have provided continuous and acceptable service to most customers, and none of them has gone bust in the history of the industry. Not unreasonably, this reputation is significant for customers, especially when other lesser-known suppliers come and go, often causing concern and disruption. Customers are prepared to pay more for a well-known and reputable brand, quite significantly more in some cases. In a competitive market, suppliers earn a return on the value of this intangible reputational asset – a return which a tariff cap expropriates.

The implication of all this recent research and evidence is that customers are more economically rational than a simplistic interpretation of the behavioural approach would imply. Customers that do not switch to an apparently lower-cost supplier are not necessarily disengaged or acting irrationally. A low switching rate does not mean that competition is not effective.

Furthermore, promoting switching to correct alleged ‘weak customer response’ may encourage some customers to move to a product or supplier that is lower-cost but less suited to their needs. It also reduces the incentive on suppliers to establish a good reputation. A higher switching rate should thus not be a prerequisite for removing the tariff cap.



The CMA (2016) recommended that Ofgem impose a temporary cap on PPM tariffs, which it did in 2017. In February 2018 Ofgem extended this cap to about one million customers on the Warm Home Discount scheme, and was contemplating extending it to a further two million vulnerable customers. However, the Tariff Cap Act 2018 required Ofgem to impose a cap on all standard variable tariffs (those not fixed for a defined period of time) and default tariffs (that apply if the customer does not explicitly choose a tariff). Ofgem did so from January 2019. Ofgem may adjust the cap from time to time: it has done so each six months.

In setting the tariff cap, Ofgem must act to protect customers, and in doing so have regard to creating incentives for suppliers to improve their efficiency, enabling them to compete effectively, maintaining the incentives for customers to switch suppliers, and ensuring that efficient suppliers can finance their authorised activities.

Each year, Ofgem must carry out a review into whether conditions are in place for effective competition. This review must, among other things, consider the extent to which progress has been made in installing smart meters. The smart metering installation programme has been controversial, and suppliers, Ofgem, government, and the Data Communications Company, have all been subject to criticism. Since smart metering was mentioned by the CMA, it is not surprising to find it mentioned in the Act. However, and rightly, the Act does not say that progress in smart metering is a precondition for effective competition.

Ofgem must then report on the outcome of its review and recommend whether the tariff cap should be extended for a further year. After considering this report, the Secretary of State for Business, Energy and Industrial Strategy then has to say whether he or she considers that the conditions are in place for effective competition. The cap falls away unless the Secretary of State considers that such conditions are not in place.

Importantly, the Act does not require that Ofgem’s recommendation whether to extend the tariff cap be determined by, or even be dependent on, its findings about effective competition. That is, it is open to Ofgem to find that conditions for effective competition are in place but nonetheless the tariff cap should be extended for a further year (perhaps to provide additional benefits to particular sets of customers). Alternatively, Ofgem might find that the conditions for effective competition are not in place, but nonetheless the tariff cap should not be extended for a further year (perhaps because its various disadvantages outweigh its advantages).

Unfortunately, Ofgem does not seem to have appreciated this distinction (as noted by former energy regulators; Littlechild, McCarthy, Marshall, Smith, & Spottiswoode, 2019). As a result, its first (August 2020) appraisal looks only at effective competition, and provides no assessment of the various beneficial or adverse consequences of the tariff cap, which could and should be relevant in informing its recommendation whether to extend the cap. Unfortunately, too, the Act does not draw the same distinction for the Secretary of State as for Ofgem, between the conditions for effective competition and the decision whether to continue or remove the tariff cap. It is apparently not open to the Secretary of State to consider the wider considerations that Ofgem may deem relevant. This seems to be a mistake and an inconsistency in the drafting of the Act.



How plausible was the first review of whether competition was effective? Ofgem decided that three conditions would need to be in place.

Ofgem’s Condition 1 was “Structural changes from the government, Ofgem and the wider market are facilitating competition” (Ofgem, 2020a, p. 5). Ofgem noted improvements in licensing conditions and the development of price comparison websites and automatic switching services. However, installation of smart meters had been slower than expected, and future improvements to the switching process were not yet in place.

These statements are true but delays in smart metering and faster switching have not prevented effective competition for the last two decades, or in other jurisdictions. It was surely untenable to cite them as a justification for maintaining a tariff cap.

Ofgem’s Condition 2 was “The competitive process should be expected to work well in the absence of the price cap” (2020a, p. 6). Ofgem had three concerns here.  First, “it is not clear that engagement levels across default tariff customers would be sufficient to constrain the price setting of default tariffs, in the absence of the cap” (2020a, p. 6). However, the evidence presented earlier suggests that it was primarily Ofgem’s regulatory restrictions that reduced the switching rate after 2008, and that removing those restrictions had enabled the switching rate to return to its previous level by the end of 2018. That level was higher in the UK than in almost any other energy market in the world, and higher for energy suppliers than for almost any other domestic product. It also seems that the tariff cap itself may have restricted engagement levels, insofar as the switching rate ceased to continue growing after the cap was imposed in early 2019, and switching is now falling.

Second, “The larger suppliers are pursuing efficiency programmes to bring costs closer in line with the efficient benchmark of the default tariff cap. However, these programmes will take time to bear fruit and deliver lasting productivity gains” (2020a, p. 7). Well, of course they will. Competition is a process that takes place over time. The unrealistic assumption that effective competition requires all market participants to be equally ‘efficient’, with the market in some sort of long-run equilibrium, was the key mistake that the CMA made. If a tariff cap had to be imposed on a market where some of the higher-cost participants were pursuing but not yet completing efficiency programmes, then every competitive market would need a tariff cap. Moreover, even Professor Cave’s dissent to the CMA report, recommending a tariff cap, noted that “it puts cost pressure on the larger suppliers to become more efficient” (CMA, 2016, p. 1417) but did not suggest that such increased efficiency would or should be achieved within his proposed two-year duration of the cap.

Strangely, Ofgem’s summary of Condition 2 had nothing to say about alleged excess profits of the large suppliers, which were the other reason for the alleged customer detriment and hence for the tariff cap being imposed. Its text on this issue said blandly, “The aggregate EBIT margin has fallen annually since 2016” (2020a, p. 38). Yes, but that hardly summarises the situation adequately. The CMA is cited as saying that a 2 per cent EBIT margin is an appropriate benchmark (2020a, p. 38n), but the tariff cap reduced the aggregate EBIT margin to minus 1.48 per cent in 2019. Only one of these large suppliers made a positive margin (of 2 per cent). Surely Ofgem should have acknowledged that the tariff cap had rendered the whole sector loss-making (see below), and in that sense was itself now preventing effective competition?

Third, “The Covid-19 pandemic has placed financial strain on both consumers and suppliers” (2020a, p. 7). Maybe so, although Ofgem and suppliers seem to have handled this well, most suppliers seem to be coping with some quite costly obligations, and the tariff cap has probably imposed greater financial strain. Surely the pandemic provided no basis for arguing that competition is not effective.

Ofgem’s Condition 3 was “The competitive process should deliver fair outcomes for consumers” (2020a, p. 7). When regulators propose ‘a fair deal for consumers’, or say that a network price control is ‘tough but fair’, this is generally rhetorical. (The author prohibited use of the word ‘fair’ while head of the Office of Electricity Regulation, because it too often seemed a substitute for careful thought.) No economist believes that the competitive process delivers prices that meet some arbitrary ethical definition of fairness. The adjective ‘fair’ is too vague to serve as a sensible criterion for evaluating the effectiveness of competition.

Ofgem clarifies that “fair outcomes” means that consumers should not be overcharged through excessive profits or inefficient costs, should receive a good quality of service, and should have access to a range of different tariffs (2020a, p. 40).  Yet Ofgem has nothing further to say in this section about excessive profits and inefficient costs. Instead, it agonises about differentials between fixed and variable tariffs, which it seems to feel might be too high, but it gives no indication of why, or what ‘competitive’ or ‘fair’ differentials might be. It concludes that these price differentials “have not evolved as we expected under the default tariff cap but may still be lower than if the cap were not in place. There is therefore uncertainty as to how prices and price differentials would evolve post-cap …” (2020a, p. 40). That Ofgem cannot say what fair tariff differentials are, and failed to predict the impact of the tariff cap on differentials, and is uncertain what they will be in future, are hardly reasons for concluding that competition is not effective. Moreover, an examination of how tariff differentials have actually evolved (see below) suggests that it is unduly simplistic to expect that more competition means lower tariff differentials, or that this is necessarily a good thing.

On quality of service, Ofgem says “For competition to be effective, we would expect that consumers have the confidence that switching to potentially cheaper deals will not result in a reduction in the quality of service they receive” (2020a, p. 51). Surely the opposite? In a competitive market one would expect that higher quality of service would command a higher price and that lower quality of service would have to accept a lower price. Effective competition means variations in quality of service to serve customers with different incomes and tastes. That is what has been observed in the market, as the evidence mentioned earlier suggests.

Ofgem also says “Overall, it is not clear that customer service is improving” (2020a, p. 7). But if higher-quality service is more costly to provide, why should it keep improving? Surely effective competition does not mean that customers should keep paying ever more for ever higher-quality service?

Ofgem’s three arguments as to why conditions for effective competition are not in place are arbitrary, vague and implausible. In part, they are influenced by the mistaken and unrealistic CMA benchmark (as indeed are Ofgem’s calculations of efficient cost in setting the tariff cap). Regrettably, as noted earlier, Ofgem made no attempt to assess the case for or against extending the tariff cap, which in the Act is distinct from the question whether the conditions for effective competition are in place. As a result, Ofgem gave no consideration to possible adverse effects of the tariff cap, which are surely relevant in considering whether it should be extended.

Ofgem’s second appraisal in August 2021 (Ofgem, 2021) yielded similar answers: positive developments in certain respects, not in others, not clear if competition would be sufficient to protect customers if the tariff cap were removed. There was no indication of overall progress or otherwise, no suggestion that competition ever could become effective. There was some useful further analysis (see section 9) but still little or no recognition that the tariff cap could have adverse effects.



 9.1 | Effects on spread of prices and switching

The CMA was concerned at the size of the spread of prices (between the large suppliers’ high standard variable tariffs (SVTs) and the various lower tariffs available in the market), which it thought reflected weak customer response and ineffective competition. But, later, Ofgem was concerned that a tariff cap, by constraining SVTs, would reduce the spread of prices in the market, and thereby reduce the extent of customer switching. It estimated that switching rates could fall by between 33 and 50 per cent (Ofgem, 2018, para 2.83), which it considered would reduce the effectiveness of competition.

In the event, savings of up to several hundred pounds against the tariff cap continued to be available, even higher than in earlier times, and the switching rate rose to an all-time high level. What was going on? Did a high spread indicate competition or a lack of it? If the tariff cap brought benefits to default tariff customers without adversely affecting competition, what’s not to like?

Ofgem (2021, figure 11) has thrown useful light on the situation. Apparently surprising movements in price differentials can be explained by movements in wholesale prices, partly because movements in wholesale prices feed through into fixed tariffs more quickly than into variable tariffs restricted by the tariff cap.

Thus, during 2017 and early 2018 the average large legacy SVT was around 15 per cent above the average fixed tariff of other suppliers, fell nearly to par over the last half of 2018, then, surprisingly given Ofgem’s prediction, rose again to around 15 per cent above in the first half of 2019 after the tariff cap was imposed, and indeed rose to around 20 per cent above in the first half of 2020. However, Ofgem’s figure 11 also shows that the high price differentials occurred primarily during periods of falling wholesale prices, which offset the impact of the tariff cap from early 2019 to mid-2020, and the low price differentials occurred primarily during periods of rising wholesale prices, notably since mid-2020.

So, after allowing for movements in wholesale prices, experience does not suggest that Ofgem was wrong in predicting that the tariff cap would have a dampening effect on price differentials. Indeed, price differentials were lower in early 2021 than at any time during the previous five years. Ofgem acknowledges that “differentials may be lower under the cap than if it were not in place” and that “narrower price differentials between default and fixed tariffs have meant lower potential savings from switching” (2021, p. 29).

The six-monthly resetting of the tariff cap further distorts competition by restricting the ability of the suppliers to respond more quickly and flexibly to changing market conditions, at least with respect to their variable tariffs. The cap artificially increases the spread of prices at times of falling wholesale prices (including by encouraging suppliers to price up to the cap, higher than they otherwise would have done) and constraining it at times of rising wholesale prices. As of 16 August 2021, wholesale costs were rising so quickly that some two-thirds of suppliers were pricing their cheapest fixed-price tariff above their variable tariff. And only nine companies (two of which went bust a couple of weeks later) were offering any tariffs more than £12 below the tariff cap. This seems a severe distortion of the market.

Insofar as the six-monthly setting of the cap delays but accentuates the impact of changing underlying wholesale costs, it also means that SVT customers at cap level are faced with sudden and significant tariff changes, most recently increases of £96 on an annual bill of just over £1,000 effective from April 2021, followed by an increase of £139 [to be?] effective October 2021. Moreover, not only are prices artificially low because of the mistaken basis on which Ofgem sets the control, they are also at present a further £139 below that because of the inflexibility in resetting the control. And a further increase of £278 is predicted next time (Byers Hussain, & Gosden, 2021). Unconstrained competitive markets respond much more quickly and smoothly than this artificially and severely constrained regulatory process.

There was also an early increase of £117 effective from April 2019, but a subsequent High Court decision indicated that Ofgem had set the initial January 2019 tariff cap artificially (and illegally) low. The general understanding is that Ofgem’s initial cap level was to deliver the politically promised level of initial tariff reduction. This suggests that the cap level is influenced by political as well as economic considerations, which is not unexpected but neither is it reassuring.

What has been the impact on switching? Ofgem (2021, figure 4) helpfully shows the rolling average annual switching rate (although only since January 2016). As noted earlier, there was a steady increase following the removal of Ofgem’s various simple tariff restrictions, to just over 20 per cent in March 2019, just after the tariff cap was imposed. The switching rate then remained around 20 per cent for electricity, fractionally higher at 21 per cent in February 2020, but has since remained at around 20 per cent. It has apparently fallen below 18 per cent for gas.

No doubt COVID restrictions have had some impact since lockdown took effect in spring 2020, but it seems plausible that the cap may have contributed to the levelling off in switching after the cap was introduced. A more systemic analysis looking also at price differentials would seem appropriate.

More significantly, there are now signs (see below) that the failure of the cap to keep up with wholesale price increases has rendered switching pointless because hardly any tariffs are available below the cap price. A period of near-zero switching may be in prospect.

9.2 | Effects on suppliers’ own tariff differentials

How, if at all, would the tariff cap impact on the decisions by suppliers to differentiate the prices of their own products? Ofgem and more recently the CMA have been concerned about the alleged ‘loyalty premium’, the difference between the prices that suppliers offer to new customers as opposed to their existing (assumed loyal) customers. The Ofgem/CMA view is that this reflects exploitation of disengaged customers. An alternative view, proffered by Baumol (2006) and accepted (I conjecture) by most economists, is that in a competitive market most suppliers have no alternative but to offer low tariffs at about marginal cost, typically but not always for a short fixed period, in order to attract new customers, either to grow or even to maintain existing customer numbers in order to survive in the market. It is simply not viable for most suppliers to offer such low marginal-cost prices to all their customers. So suppliers do not choose to differentiate their prices in order to exploit market power; they are forced to do so by the strength of competition.

Some medium-large suppliers have offered only one tariff (which might be fixed or variable) or have offered fixed and variable tariffs closely aligned to each other, on the basis that this is ‘fair’ to customers and that significant tariff differentials are ‘unfair’ because they exploit loyal or less active customers. And some have supported a tariff cap, too, as being in the interests of customers. It must be acknowledged, however, that this is an argument for government and/or regulatory policy that supports their own commercial policies. Moreover, the cap is also likely to be in their own longer-run commercial interests, as hitting their largest and smallest competitors harder than themselves.

Although uniform or near-uniform pricing policies sound ‘fair’, will they prove commercially viable over the longer term? Or will all or most suppliers need to offer differential tariffs in order to survive? Or is a coexistence of uniform and differential pricing policies possible and indeed economic because there are differences among suppliers and among customers? Competition as a rivalrous discovery process should reveal this, and also reveal what, if any, impact the tariff cap has had on supplier policies to date.

Consider the SVTs and the lowest fixed tariffs offered by individual suppliers at two sample dates: 31 January 2018, a year before the introduction of the tariff cap, and 2 November 2020, nearly two years after its introduction. On the first date, the spread between Ofgem’s average large legacy supplier SVT and cheapest tariff basket (a proxy for wholesale price) was £294; on the second it was £188. According to Ofgem (2021, figure 11), on the first date average large legacy SVT was around 15 per cent above average fixed tariff of other suppliers, on the second date it was between 5 per cent and 10 per cent above. On the first date, average wholesale cost was around £0.04 and relatively stable, on the second date it was around £0.03 and rising.

So several factors were changing. One would expect that, on the second date, increasing wholesale cost plus the tariff cap would render less feasible very large differentials between a supplier’s own tariffs, but what if any effect would the cap have on whether suppliers differentiated their tariff prices at all?

Table 1 summarises the findings. The four suppliers that offered a very large tariff differential (over £200) in 2018 no longer did so, but nearly the same number of suppliers (nine as opposed to ten) still offered at least a large differential (over £100). The number of suppliers offering a medium differential (£31–99) increased from four to nine, and the number offering a small differential increased from one to six. The most striking change was that the number of suppliers offering a single tariff for all customers fell from 12 to two.

TABLE 1   Summary of changes in tariff differentials

Size of differential

No. of suppliers Jan 2018

No. of suppliers Nov 2020

Very large > £200



Large £100–199



Medium £31–99



Small £2–30



Single tariff






This is by no means definitive because it provides just a snapshot of suppliers active at two points in time, under different cost conditions. Nonetheless, it appears that the picture on tariff policies is very mixed: different suppliers have very different policies. A movement towards a single tariff per supplier, or even to smaller tariff differentials, does not seem under way.

On the contrary, by November 2020 almost all the large and medium suppliers had continued or adopted a policy of large (over £100) tariff differentials. In Q4 2020 these suppliers accounted for about three-quarters of total customers. Eight smaller suppliers that had previously had zero differentials had now introduced medium differentials (£31–99). Admittedly, Bulb Energy maintained a policy of a single variable tariff, and Octopus Energy had a set of tariffs with small (under £30) differentials. Their combined market share in Q4 2020 was 11.8 per cent. Together with another half-dozen smaller suppliers each with a single tariff, they perhaps accounted for about 12.5 per cent or one eighth of all customers. This leaves another ten or so suppliers with medium (£35–99) differentials supplying the remaining one eighth of the market.

The variety continued. As of 19 June 2021, for example, 38 suppliers were offering tariffs, of which five had large differentials (two positive, three negative), 17 had medium differentials (five positive, 12 negative), eight had small differentials (three positive, five negative), seven offered variable tariff only, and one offered fixed tariff only.

The very existence of negative differentials – indeed, two-thirds of these differentials were negative – is an indication of how far variable tariff prices have been held down below fixed tariff prices – in effect, the opposite of a loyalty premium. Since June, the situation has worsened.  As of mid-September, the major suppliers are no longer offering their variable tariffs to new customers online (Gosden, 2021). When the author checked Uswitch it listed only seven suppliers that would save over £20 on a typical dual fuel bill, to none of which it could switch the customer, and those six suppliers to which it could switch the customer would increase the annual bill by about £360 to £650. Compare the Market says “We’re really sorry but we can’t get you a quote at the moment. We’re temporarily pausing our energy comparison as energy suppliers are currently restricting the number of tariffs available.”5

All this suggests that it is wrong to characterise price differentiation as an exploitative pricing policy practised by those six former incumbents that have long-standing loyal but disengaged customers. Rather, it is a type of policy that is necessary for most suppliers to practise to some degree, simply to survive in a very competitive market like this one. This is – or was – what prices look like in a competitive market where marginal costs are below average costs and where customers have different elasticities of demand.  But now, the disappearance of products and services is the predictable consequence of a misconceived price cap set below cost.

9.3 | Effects on profits and the number of suppliers

As noted earlier, the tariff reductions as a result of the cap have rendered the retail sector loss-making. From 2012 to 2017 the aggregate average pre-tax profit rate (EBIT) of the six large legacy suppliers (per Ofgem data portal, n.d.) was more or less constant at just over 4 per cent. In 2018 it fell below 3 per cent, in 2019 to minus 1.48 per cent and in 2020 it was minus 1.02 per cent. Only one of those large companies made a profit in the last two years, but a profit that is presently below the level that the CMA deemed competitive.

To attract and keep customers, other suppliers must maintain prices below those of the large incumbent suppliers, so they too are hit by the tariff cap – in fact more severely. For 24 such non-legacy and non-renewable suppliers the range of net profit margins posted for financial year 2019 or later was from minus 1 per cent to minus 29.9 per cent, with mean of minus 10.9 per cent and median of minus 8.1 per cent.6 The obvious question is: how is this consistent with Ofgem’s duty, under the Tariff Cap Act s. 1(6)(d), “to ensure that holders of supply licences who operate efficiently are able to finance activities authorised by the licence”?

For all these companies, the cap has cut the ground from under them. They entered and invested in the retail market in the expectation that it would be competitive and without price controls; then suddenly a tariff cap was imposed at an unrealistic level, and they were pushed into unprofitability. Not surprisingly, an increasing proportion of suppliers has been unable to survive.

As in all competitive markets, some of the exiting suppliers were not sufficiently experienced, organised or well-funded. It is now apparent that some of the smaller suppliers were not sufficiently hedged against wholesale cost increases, or even hedged at all. However, other exiting suppliers were experienced professional companies (like Engie, formerly Gaz de France, iSupply owned by Vattenfall, and Green Network Energy, part of the Green Network Group in Italy since 2003), and/or backed by serious investors (like Tonik backed by Mitsui).

From 2004 to 2010 the number of active suppliers was relatively constant at about a dozen. From 2010 to the first two quarters of 2018 the number rose steadily to 70. But since the Tariff Cap Act passed in July 2018, it has been downhill almost all the way. In the 2.5 years from Q1 2016 to Q2 2018, just five suppliers exited the market. In contrast, in the 2.75 years from Q3 2018 to Q1 2021, 24 suppliers exited, nearly five times as many. The exit rate is still accelerating: another exit in August 2021, four more exits in the first half of September, and a further four suppliers understood to be seeking exit. On 21 September Ofgem issued provisional orders against five suppliers (three domestic) for failure to make payment to the feed-in tariff renewables scheme, which is often a precursor to exit. “Baringa, the analysts, projects that another 39 suppliers could fail in the next 12 months leaving only ten remaining in the market” (Byers et al. 2021, p. 4). “The Government is in talks with the industry over how to deal with the predicted collapse of dozens of small suppliers… Taxpayers face a multibillion-pound bill to help energy companies cope…” (Courea, Gosden, Martin, & Ellery, 2021).

The tariff cap experience, including the recent intention to extend it indefinitely (see below), is sending a clear message to investors: the UK retail energy market is now politicised and subject to price controls set at artificially below-cost levels. It is a market that prudent investors are now avoiding or exiting rather than entering.

Government and Ofgem are not responsible for the present wholesale cost increases, nor for the policies adopted by suppliers. But they have imposed and continued this tariff cap in the mistaken belief (as argued here) that it reflects the price that would characterise a competitive market. In fact, it was nearly £100 below that level, which forced new entrant suppliers to cut their prices and costs by about £100 more than otherwise would have been necessary, in order to attract and retain customers. Government and Ofgem thus bear substantial responsibility, not only for the short-term distress caused to customers of those suppliers forced out of the market, and the costs of tens or hundreds of millions, or even billions, of pounds imposed on other customers and/or taxpayers, but also for the longer-term adverse consequences of discouraging future new entry, investment and innovation.

Ofgem (2020b) has said that “The management of the energy sector, in particular the transition to net zero, requires a partnership between the regulator, governments, energy companies, and energy consumers.”  If it is government and regulatory policy artificially to hold down retail prices and profits, indeed to force losses on all suppliers, and to continue to urge customers to leave their present suppliers, this will jeopardise any such partnership.  



In 2019 the Department for Business, Energy and Industrial Strategy (BEIS) and Ofgem consulted on further retail policy, including with respect to the tariff cap. Nearly 50 organisations responded.7 The focus here is more specific. If there were a political and regulatory interest in removing or ameliorating the tariff cap, so that government and regulatory attention could be focused more productively, what steps might facilitate this? Here are five main suggestions.

First, move on from the incorrect narrative of a long-standing uncompetitive and inefficient retail market with significant customer detriment. Instead, develop and communicate more broadly a better understanding of how this competitive market works, why observed price differentials are not necessarily exploitation of loyal customers, but reflect differences in product and quality or reliability of supplier. They also provide opportunities for customers. Explain, too, that although the tariff cap may have advantages for some customers, it has disadvantages for others and for competition and investment.

The earlier part of this article attempts to provide the basis for this first task. It means abandoning much of the CMA analysis, and acknowledging various competitive aspects of the market, and various limitations of regulatory intervention, that have not been fully acknowledged in recent years.

Second, shift the focus from increased switching – trying to persuade more customers to leave their present supplier – to better switching. Explain that customer loyalty is in general a good thing where it is deserved, not an indication of weak response or an undermining of competition. Acknowledge that although price is important it is not the only consideration. Assist customers to make more informed choices of tariff and product, considering supplier reputation, so that customers do not need repeatedly to switch supplier.

Although there may still be some interest in trying to increase switching, on which I have made suggestions elsewhere (Littlechild, 2019b), the case for doing so (weak customer response) seems increasingly unpersuasive in the face of other explanations for customer behaviour, and the case against (the danger of persuading customers to move to lower-priced but less satisfactory or more risky suppliers) seems increasingly strong.

I have suggested elsewhere the calculation of an Overall Customer Satisfaction (OCS) score (Littlechild, forthcoming). This is the average of four individual scores related to data or ranking published by Ofgem (on customer complaints), Citizens Advice, Which? magazine, and customers’ own views as expressed on Trustpilot. Just over two dozen suppliers are typically in the OCS league at any time – nearly half the total number of suppliers. These are the suppliers that have proved their ability to attract or retain a significant number of customers. The simple aim is to develop and make available information that could be helpful in establishing more quickly the reputations of suppliers, so that customers (and regulators) can be more confident that customers are making informed choices whether to be loyal to suppliers, or which suppliers to turn to if they are considering switching.

Third, gradually relax the present tariff cap, thereby minimising the concern that there will be a significant increase in prices when the cap is removed. The tariff cap has already been set for October 2021 – March 2022 and would need to be reset four times if it remained in place until end-2023. If, as often asserted, the initial imposition of the cap reduced average SVT prices by about £100, it would be sufficient to increase the cap by about £25, cumulatively, beyond the present method of calculation, at each of these four resettings. To achieve this, the calculation of ‘efficient cost’ might gradually transition from the CMA’s unrealistic perfect competition concept to a more realistic assessment.

Then, by October 2023 the cap would be at a level such that no further tariff increases could reasonably be expected (other than to reflect legitimate cost increases) if the tariff cap is not renewed. It is likely that large suppliers would gradually find it profitable to price below the cap before it is increased by the full £100, as their efficiency programmes increasingly take effect. Ofgem and BEIS could then point to the successful reintroduction of competition.

Fourth, replace the tariff cap by an ongoing cost calculation, which might be made public, to inform and reassure Ofgem and others whether increases in energy prices (or lack of reduction) are justified. In 2008, Ofgem was unable to decide, and was led to various interventions that were subsequently found to have been counterproductive. In 2021, by contrast, Ofgem itself announced two very significant increases in the tariff cap, and although it was subject to severe criticism, particularly from fuel poverty organisations,8 it was able to justify the increases because it had done the calculations itself.

Fifth, continue a tariff cap for a specified and limited set of vulnerable customers, at a reasonable level that reflects actual costs rather than hypothetical ‘efficient’ costs. This would discharge Ofgem’s obligation to have regard to the interests of particular sets of vulnerable customers and would provide a marker against which other tariffs could be assessed.

In principle, this could be consistent with one supplier’s suggestion that “The price cap behaves like the National Minimum Wage – it is a minimally interventionist ‘decency’ limit under which competition thrives and companies are incentivised to keep costs down” (Octopus Energy, 2019, p. 6). However, whether such policies are viable depends critically upon the coverage and levels set. Around 7 per cent of UK workers are paid at or below the Minimum Wage, whereas the tariff cap constrains tariffs to over 70 per cent of SVT customers – an order of magnitude difference. A Minimum Wage that constrained over 70 per cent of workers’ wages would be so high and distorting as to be unviable. Thus, it is the seriously mistaken level and coverage of the present tariff cap, rather than the principle of it, that makes it untenable. In addition, the inability of a tariff cap to respond to rapidly changing conditions in a particular market sector is another factor differentiating it from an overall minimum wage.

Of course, judgement must be exercised in calculating costs, and there is always the possibility that Ofgem or government would try to influence the course of prices, and suppliers’ pricing decisions, by modifying (or failing to modify) the cost calculations underlying any cap. But if a cost index or vulnerable customer tariff cap were sufficiently limited in scope, and sufficiently moderate and flexible to enable justified competitive price movements that would otherwise be politically unacceptable and lead to more severe intervention, that could be a price worth paying.



As this article was being revised in summer 2021, BEIS (2021a,b,c,d) published a retail energy market strategy paper, a consultation on two switching schemes (opt-in and opt-out), and an announcement outlining its plan to enable the default tariff cap to be potentially extended beyond 2023. Ofgem hastened to indicate its support.

These proposals have been somewhat eclipsed by recent events, so only two brief points here. First, retail energy market policy has now jumped on board the ‘loyalty penalty’ bandwagon launched by the CMA (2018) responding to a Citizens Advice complaint. The claim is that competition is not fully effective in numerous sectors because new customers are offered a competitive price (down to marginal cost) whereas loyal but disengaged existing customers are exploited by a price above that. The contrary argument of Baumol (2006) and others (see above) is that, in markets like these, it is the strength of competition, not its weakness, that forces firms to differentiate prices in this way.

The CMA policy is informed by a review of Economic Research on Loyalty Price Determination (E.CA Economics, 2020) which at over 100 pages and around 150 literature references appears exhaustive, not to say exhausting. Yet it does not reference Baumol (2006) or any of the 31 references therein (except for brief mention of Joan Robinson’s classic 1933 volume and a more recent survey paper). Nor for that matter do BEIS (2021a–d). The CMA/BEIS policy on the retail energy market is apparently not informed by the economic literature explaining how observed pricing policies can reflect effective competition rather than the lack of it. Prohibiting or restricting such pricing can have an adverse effect on competition, as the CMA found with previous Ofgem policies aimed at improving competition, and as tariff cap experience is now demonstrating.

Second, BEIS envisages significant government and regulatory intervention to implement net zero policies, and opt-in and opt-out switching are the chosen means to influence customer choice. Thus, from 2024 onwards the strategy is to “prompt consumers to consider alternative tariffs/services, improving competition and engagement, through opt-in switching”. By the late 2020s the strategy envisages “Potentially more action (e.g. through opt-out switches) to nudge consumers and suppliers towards more competitive and/or low carbon products” (BEIS, 2021b, p. 5).

If BEIS and Ofgem were to find that competition is now effective in the retail market, opt-in and opt-out switching would no longer be needed to stimulate more customer engagement. BEIS and Ofgem would then have to make the case for such proactive switching on the grounds of changing inappropriate customer preferences. More convenient, then, to continue to find that competition is not yet effective?

The BEIS proposal is to “work with the market to take consumers on the journey to a net-zero energy system” (2021b, p. 4). “Where they are not engaged, [consumers should] still be taken on a journey to products and services that contribute to decarbonisation (potentially in the absence of active choice)” (2021b, p. 8). Six duties are proposed for energy companies, and another four for the market framework.

Can the envisaged retail energy market still be called competitive? Whereas competition may be described as a process for discovering and providing what customers want, the BEIS proposal is that customers should be led to consume the products chosen by government. Customers can have any colour electricity they like, as long as it’s green. The role of suppliers will be to discover and provide what government wants customers to have. What, if anything, might constitute effective competition in such a market remains unclear, but likely unachievable. The prospect of removing the mistaken price cap recedes beyond the horizon.



As of proofreading in late September 2021, events are moving fast. At current gas price levels, the tariff cap would need to rise by a further £550, in addition to the increase of £139 planned for October 2021, if it is to cover the cost of supplying a customer for the next year (Financial Times, 2021).

Two more suppliers went bust on 22 September. Of the 20 cheapest tariffs in the market as at 7 August, 12 were from suppliers that had gone bust by 22 September, and another was from a supplier that is working with restructuring consultants. If up to 40 small and medium-sized suppliers fail in the next few months, the larger better-financed suppliers will have neither the capacity nor the economic ability to take their millions of customers, given that each customer will be heavily loss-making at the tariff cap price.

Various alternative solutions are reportedly under consideration, including moving these customers to a temporary ‘bad bank’ supplier; appointing a special administrator, in effect nationalising suppliers temporarily; and loan guarantees to big suppliers to foot these costs, subsequently recouping them through domestic energy bills. The bill for subsidising these customers is estimated at several billions of pounds (Courea et al., 2021).

On 20 September the Business and Energy Secretary and the Ofgem CEO issued a joint statement saying that “Central to any next steps is our clear and agreed position that the Energy Price Cap will remain in place”.9 No thought seems to have been given to how the nature of the price cap might usefully evolve. It might appear that the challenge of removing a mistaken price cap is more difficult than ever.

However, the mainstream media is recognising the disadvantages of the price cap, and turning against it. Thus, the lead editorial in The Times (22 September) uses phrases like “cravenness as well as complacency” and “a system in which competition is evaporating and risks will have to be borne by the taxpayer”; it argues that failure to remove the cap will mean “a sector that in future will have to be far more tightly regulated, leading to less choice, less investment and less innovation”. Warner (2021), referring to the price cap as “a blatant piece of vote buying populism”, says “There has been no good precedent internationally for price caps, yet the Government now trumpets the policy as the best way of protecting consumers from the spike in energy prices. That’s rich, given that it may partially have caused the problem in the first place.”

Now that the serious limitations of the tariff cap are acknowledged, is it not worth exploring a way through here? That is, remove the cap for most SVT customers, but keep it at a higher level, consistent with allowing competition, for a specified subset of vulnerable customers (such as the ones Ofgem previously had in mind)? Perhaps also increase support for vulnerable customers, for example by increasing the Warm Home Discount. And invite suppliers themselves to offer social tariffs for vulnerable customers, as indeed they used to do, before regulation discouraged this.

Stephen Littlechild

Emeritus Professor, University of Birmingham, and Associate, Cambridge Judge Business School, UK



[1] Ofgem regulates energy markets in Great Britain (its remit does not run in Northern Ireland).

[2] The OFT advice included, “In most circumstances, the pricing, marketing, and advertising practices of firms can still be viewed as benign with no need for action”; “markets can be self-correcting and interventions can potentially do more harm than good”; “In many instances, the problems arising from behavioural biases will be solved by the actions of market participants themselves”; and “it may be that authorities simply do not have the level of expertise required to make delicate interventions. In such situations an authority would be wise to be conscious of its own limitations”.

[3] https://www.uswitch.com/gas-electricity/green-energy/green-accreditation/ (accessed 15 September 2021).

[4] https://www.energyhelpline.com/ (accessed 15 September 2021).

[5] https://cdn2.comparethemarket.com/market/cms/energy-holding/index.html (accessed 18 September 2021)

[6] mikewhiskeytango.com, as of 12 February 2021.

[7]https://www.ofgem.gov.uk/system/files/docs/2020/01/stakeholder_consultation_responses_-_future_energy_retail_market_review_3.zip (accessed 22 September 2021).

[8] For example, the End Fuel Poverty Coalition. http://www.endfuelpoverty.org.uk/category/news/ (accessed 15 September 2021).

[9] https://www.gov.uk/government/news/gas-market-and-prices-joint-statement-from-government-and-ofgem (accessed 22 September 2021).



ACCC (Australian Competition and Consumer Commission) (2018). Restoring Electricity Affordability and Australia’s Competitive Advantage. Canberra: ACCC.

Baumol, W. (2006). Regulation Misled by Misread Theory: Perfect Competition and Competition-Imposed Price Discrimination. Washington, DC: AEI Press.

BBC News (2013). Ed Miliband: Energy market needs to be reset. 25 September. https://www.bbc.co.uk/news/av/uk-politics-24237800 (accessed 14 September 2021).

BEIS (Department for Business, Energy and Industrial Strategy) (2021a). Update on the Energy Price Cap and Energy Retail Strategy, Open Letter, 23 July. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1005308/open-letter-price-cap-announcement.pdf (accessed 15 September 2021).

BEIS (2021b). Energy Retail Market Strategy for the 2020s: Helping Consumers on the Net Zero Journey. July. London: BEIS.

BEIS (2021c). Flexible and Responsive Energy Retail Market, Summary of Consultation Responses, July. London: BEIS.

BEIS (2021d). Tackling the Loyalty Penalty in the Retail Energy Market: Analytical Annex. July. London: BEIS.

Brennan, T. (2007). Consumer preference not to choose: Methodological and policy implications. Energy Policy, 35(3), 1616–1627.

Byers, D., Hussain, A., & Gosden, E. (2021). Energy bills set to rise as price of gas soars and small firms go bust. The Times, 18 September, p. 4.

CMA (Competition and Markets Authority) (2014). Energy Market Investigation: Summary of hearing with Ofgem on 10 November 2014. https://assets.publishing.service.gov.uk/media/54bf95f540f0b6158a000010/Summary_of_hearing_with_Ofgem.pdf (accessed 14 September 2021).

CMA (2016). Energy Market Investigation: Final Report. 24 June. London: CMA. https://assets.publishing.service.gov.uk/media/5773de34e5274a0da3000113/final-report-energy-market-investigation.pdf (accessed 15 September 2021).

CMA (2018). Tackling the Loyalty Penalty, Response to a Super-complaint made by Citizens Advice on 28 September 2018. 19 December. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/892058/Loyalty_Penalty.pdf (accessed 6 September 2021).

Competition Commission (2013). Guidelines for Market Investigations: Their Role, Procedures, Assessment and Remedies (CC3).  1 April. (Adopted by CMA, updated 5 July 2017). London: Competition Commission. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/284390/cc3_revised.pdf (accessed 15 September 2021).

Competition Commission (2014). Aggregates, Cement and Ready-Mix Concrete Investigation: Final Report. 14 January. London: Competition Commission. https://assets.publishing.service.gov.uk/media/552ce1d5ed915d15db000001/Aggregates_final_report.pdf (accessed 15 September 2021).

Cornwall Insight (2020). Domestic Tariff Report, January.

Cornwall Insight (2021). Domestic Tariff Report, 21 June.

Courea, E., Gosden, E., Martin, B., & Ellery, B. (2021). Taxpayers face bill for propping up energy firms: Suppliers on brink of collapse as prices soar.  The Times, 20 September.

Deller, D, Giulietti, M., Loomes, G., Waddams Price, C., Moniche, A., & Jeonet, J.  (2021). Switching energy suppliers: It’s not all about the money. The Energy Journal, 42(3), 95–120.

EnergyUK (2020). Energy UK launches Vulnerability Commitment. 1 December. https://www.energy-uk.org.uk/media-and-campaigns/press-releases/463-2020/7715-energy-uk-launches-vulnerability-commitment.html (accessed 15 September 2021).

E.CA Economics (2020). Economic Research on Loyalty Price Determination:  Report Prepared for the Competition and Markets Authority. London: E.CA Economics.

Financial Times (2021). Industry experts warn of steep rises in UK energy bills.  22 September.

Flores, M. & Waddams Price, C. (2018). The role of attitudes and marketing in consumer behaviours in the British retail electricity market. The Energy Journal, 39(4), 153–179.

Giulietti, M., Waterson, M., & Wildenbeest. M. (2014). Estimation of search frictions in the British Electricity Market. Journal of Industrial Economics, 62(4), 555–590.

Gosden, E. (2021). Energy firms sell fixed deals far above price cap. The Times, 4 September.

He, X. & Reiner, D. (2018). Consumer Engagement in Energy Markets: The Role of Information and Knowledge. EPRG Working Paper 1835. Energy Policy Research Group, University of Cambridge.

Littlechild, S. (2014). Promoting or Restricting Competition? Regulation of the UK Retail Residential Energy Market since 2008. EPRG Working Paper 1415. Energy Policy Research Group, University of Cambridge. https://www.eprg.group.cam.ac.uk/wp-content/uploads/2014/09/1415-PDF1.pdf (accessed 15 September 2021).

Littlechild, S. (2015). The CMA energy market investigation, the well-functioning market, Ofgem, Government and behavioural economics. European Competition Journal, 11, 574–636.  (Online 11 March 2016.)

Littlechild, S. (2019a). Providing for a transition back to a competitive retail energy market: A response to the CMA’s Invitation to Comment on its proposed review of the Prepayment Charges Restriction Order 2016. 17 January. Energy Policy Research Group, University of Cambridge. https://www.eprg.group.cam.ac.uk/wp-content/uploads/2019/01/S.-Littlechild_CMA-with-PS-Jan19.pdf (accessed 15 September 2021).

Littlechild, S. (2019b). Replacing the tariff caps and protecting vulnerable customers by harnessing the competitive process. Energy Policy Research Group, University of Cambridge. 18 July. https://www.eprg.group.cam.ac.uk/wp-content/uploads/2019/07/S.-Littlechild-Replacing-the-tariff-cap_17Jul19.pdf (accessed 15 September 2021).

Littlechild, S. (2020). The CMA’s assessment of customer detriment in the UK retail energy market. Journal of Regulatory Economics, 57(3), 203–230.

Littlechild, S. (2021a). Time to stop digging: A submission to Ofgem’s Review of its Forward Work Plan. 19 February.  https://www.eprg.group.cam.ac.uk/wp-content/uploads/2021/02/S.-Littlechild_Ofgem_19Feb2021.pdf (accessed 15 September 2021).

Littlechild, S. (2021b). The evolution of competitive retail electricity markets. Chapter 4 in J.-M. Glachant, P. Joskow, & M. Pollitt (Eds), Handbook on Electricity Markets. Cheltenham: Edward Elgar Publishing, November (forthcoming).

Littlechild, S. (forthcoming). Exploring customer satisfaction in Great Britain’s retail energy sector, Part III: A proposed overall Customer Satisfaction Score. Utilities Policy. An earlier version, An Overall Customer Satisfaction Score for GB Energy Suppliers, is available as EPRG Working Paper 2027. Energy Policy Research Group (September 2020), University of Cambridge. https://www.eprg.group.cam.ac.uk/wp-content/uploads/2020/09/2027-Text.pdf (accessed 15 September 2021).

Littlechild, S., McCarthy, Sir C., Marshall, M., Smith, S., & Spottiswoode, C. (2019). Response to Ofgem on effective competition. 19 July. https://www.ofgem.gov.uk/system/files/docs/2019/07/stephen_littlechild.pdf (accessed 20 September 2021).

Mountain, B. & Burns, K. (2021). Loyalty taxes in retail electricity markets: Not as they seem? Journal of Regulatory Economics, 59(1), 1–24.

Neudegg, R. (2020). Drivers of Switching Beyond Price. Presentation at Utility Week Energy Customer Conference, 22 January, London.

Octopus Energy (2019).  Flexible and Responsive Energy Retail Markets: A Consultation Response [to BEIS & Ofgem]. September.

 https://www.ofgem.gov.uk/system/files/docs/2020/01/stakeholder_consultation_responses_-_future_energy_retail_market_review_3.zip (accessed 22 September 2021).

Ofgem (Office of Gas and Electricity Markets) (2011). What Can Behavioural Economics Say about GB Energy Consumers? 21 March. London: Ofgem.

Ofgem (2014). State of the Market Assessment. 27 March. London: Ofgem.

Ofgem (2018). Decision – Default Tariff Cap – Overview Document. 6 November. London: Ofgem.

Ofgem (2019a). Our Strategic Narrative for 2019–23. 11 July. London: Ofgem.

Ofgem (2019b). Ofgem’s Collective Switch Trials. 27 September. London: Ofgem.

Ofgem (2019c). Decision – Framework for Assessing Whether Conditions are in Place for Effective Competition in Domestic Supply Contracts. 3 October.  London: Ofgem.

Ofgem (2020a). Report: Outcome of Review into Whether Conditions Are in Place for Effective Competition in Domestic Supply Contracts. 7 August. London: Ofgem.

Ofgem (2020b). Forward Work Programme 2021/22 Consultation. 15 December.  London: Ofgem. https://www.ofgem.gov.uk/publications/forward-work-programme-202122-consultation (accessed 16 September 2021).

Ofgem (2021). Report: Outcome of 2021 Review into Whether Conditions Are in Place for Effective Competition in Domestic Supply Contracts. 6 August. London: Ofgem.

Ofgem Data Portal (n.d.). https://www.ofgem.gov.uk/energy-data-and-research/data-portal (accessed 15 September 2021).

OFT (Office of Fair Trading) (2010). What does Behavioural Economics Mean for Competition Policy? (OFT1224). March. London: OFT. https://webarchive.nationalarchives.gov.uk/ukgwa/20140402182927/http://www.oft.gov.uk/shared_oft/economic_research/oft1224.pdf (accessed 15 September 2021).

Pollitt, M. & Haney, A. (2014). Dismantling a competitive retail electricity market: Residential market reforms in Great Britain. The Electricity Journal, 27(1), 66–73.

Ros, A. J. (2020). Does electricity competition work for residential consumers? Evidence from demand models for default service and competitive residential electricity services, Journal of Regulatory Economics, 58(1), 1–32.

Simshauser, P. (2018). Price discrimination and modes of failure in deregulated energy markets. Energy Economics, 75, 54–70.

Simshauser, P. (2021). Lessons from Australia’s National Electricity Market 1998–2018: Strengths and weaknesses of the reform experience. Chapter 8 in J.-M. Glachant et al. (Eds), Handbook on Electricity Markets. Cheltenham: Edward Elgar Publishing.

Simshauser, P. & Whish-Wilson, P. (2017). Price discrimination in Australia’s retail electricity markets: An analysis of Victoria and Southeast Queensland. Energy Economics, 62(17), 92–103.

Waddams Price, C. & M. Zhu (2016). Non-discrimination clauses: Their effect on GB retail energy prices. The Energy Journal, 37(2), 111–132.

Warner, J. (2021). Energy crisis exposes deep flaws in Britain’s ‘seat of the pants’ economy. Telegraph, 22 September.

Yarrow, G. (2014). Energy Market Investigation: Response to the CMA’s Statement of Issues. 14 August. https://assets.publishing.service.gov.uk/media/53f1c32fe5274a48c400000f/George_Yarrow_response_to_IS.pdf (accessed 17 September 2021).



Domestic Gas and Electricity (Tariff Cap) Act 2018. https://www.legislation.gov.uk/ukpga/2018/21/notes/division/6/index.htm (accessed 16 September 2021).


Share on Twitter
Share on LinkedIn