Back to front?

Back to front?


Market prices are powerful economic signals, and interpreting them is a big part of the business of microeconomics. Certain kinds of pricing behaviour are almost bound to attract the attention of authority – both the sector regulators and the competition watchdogs, as well as consumer lobby groups and government policy-makers. One of the areas that has attracted increasing regulatory attention over the past 20 years is “back-book” pricing (or so-called “loyalty penalty”). In particular, regulators face the complaint that long-term customers get trapped by ignorance or inertia in the economic equivalent of an abusive relationship – being persistently charged more, or rewarded less, than new customers. It is an issue that has raised particular concern among consumer groups with respect to financial services. This article explores the phenomenon and the regulatory response.

It is rare to find a product or service for which the price is always the same for all
customers all the time. Obvious, commonplace examples include variations for:

  • different groups of people – e.g. £5 cinema tickets for those over 60, or a 10% student discount at a fashion retailer
  • different times of the day – e.g. Liverpool to London standard single, ranging from £36 to £162.50
  • different moments of booking – e.g. £219 ticket for a flight from London to Madrid on the day, when the customer is desperate to fly, £33 for a flight to be taken in a month’s time; or £4,000 for a cabin booked three months in advance, £900 for an identical one booked at the last minute, when the cruise company is desperate to fill the ship
  • different kinds of customers – e.g. discount vouchers offered through print media or third-party websites to target customers with particular characteristics
  • different quantities – 3 for 2, or a discount for booking the fifth visit to the spa in a month – a “loyalty bonus”
  • different periods – £12 for a first quarterly subscription to The Economist and then £65 per quarter – a “loyalty penalty”.

Of all these variations, it is the second last – rewarding customers’ loyalty – that feels
instinctively right. And in terms of good business, and good business practice, it makes sense in many markets, as it is usually much easier to get repeat business than acquire a >whole new set of customers. So it is not surprising that explicit loyalty schemes are commonplace in travel, leisure and throughout the retail sector.

But there are markets where exactly the opposite happens – new customers get a better deal than old ones. Introductory offers (such as the subscription example above) are commonplace on everything from grocery deliveries to plant catalogues. But the issues become more complicated where an “essential” service is involved – energy, broadband or banking products – if “back-book” customers are persistently charged more or paid less than new ones.

This is an example of “price discrimination”, a term that describes situations in which firms charge different prices to different customers for the same thing, depending on their perceived sensitivity to price. Online markets are providing rapidly increasing opportunities for sophisticated discrimination, but the practice is common in “old economy” markets too. Regulators do not find price discrimination intrinsically objectionable, if it is transparent and easily countered by the customer (e.g. by switching supplier). But questions arise if it is hidden or hard to avoid – being explicitly or implicitly based on personal characteristics, or if switching is difficult.

So what is it about the economics of these markets that turns pricing back to front, and when should regulators be concerned? In addition to the continuing service relationship, the distinctive feature of these markets is that customers typically face some switching cost in moving to a new product and cancelling an existing contract. These costs may be explicit (such as early repayment charges on loans) or implicit (such as the time required to search for a new provider and fill in a new bunch of forms). It is these switching costs that mean providers can charge prices higher than the competitive market rate to their long-term customers.

However, it is the very fact that providers can charge them higher prices that makes loyal customers so valuable. This, in turn, means that providers will compete hard for new customers that may turn into old. These pressures tend to lead to the opposite of a loyalty bonus – a pattern of charges that rise with the “life” of the customer relationship, as Box 1 explains.

This is an example of price discrimination







Draw a caricature of an economist, and she (or mostly he) must be playing with a “simplified model”. And, all right, these models don’t provide direct answers to any of the complex problems our clients face; but they can help us to zoom in on the nub of the particular problem under the microscope.

So here’s a simplified model to illustrate the problem of pricing ongoing service relationships: Let’s suppose that:
— there are plenty of providers, offering an identical service
— there are only two periods in the life of their relationship with the customer, i.e. the customer is either “new” or “old”
– just acquired, or on the “back book” — the competitive market price for the service (i.e. the total cost of providing the service, including the provider’s profit margin) is £100 per period
— the cost to the customer of switching providers is £10 (i.e. customers will only switch between providers if they save more than £10 by doing so)
— customers choose providers solely on price.

This simplified model allows us to explore the pricing problem. First, and most obviously, the provider needs to earn revenue of at least £200 across the two periods to maintain profit margins. (We ignore the time
value of money.)

Suppose one provider decides to increase profits by charging £110 in the second period, knowing that’s not quite enough to make “old” customers switch to another provider charging £100. The problem is, because all providers know they can do this, they may decide to spend some of the extra profit earned by reducing prices in the first period to attract extra custom.

If they cut the charge to “new” customers to £90, they will both attract new custom and persuade “old” customers paying £110 elsewhere to switch to them. But to maintain their profit margins, they too will then need to charge £110 for the second period, and so they too will lose  custom. To cut a long story short, the process of competition will drive prices to £95 in the first period and £105 in the second.

Any other pricing outcome is impossible in this model: charge more than £105 in the second period, and you lose all your customers. Charge less than £95 in the first period, and you won’t be able to recover your investment from profit on the “back book” in the second period. And if you set your introductory price above £95, you won’t acquire any customers to begin with.



Another piece of anti-competitive behaviour of which those who tilt the price curve against long-staying customers are often accused is, of course, cross-subsidy. Competition regulators are on the lookout for the use of cross-subsidies to attempt to foreclose the market. Set too low a price in one market, and others cannot compete with you. This then leaves you free to increase the prices in that market over time – or protect the profits in the linked market.

There are many cases where firms have been found to do just this, particularly since a seminal European judgement in 2001. The EC found that Deutsche Post had been foreclosing the parcel delivery market by cross-subsidising its business there with (legal) monopoly profits from its domestic letter business.

At first glance, that looks quite like what is going on in markets with a loyalty penalty – from which providers extract enough profit to price initial offers below cost. But the regulatory test imposes a higher standard of proof of (unacceptable) behaviour. In a regulatory sense, cross-subsidies exist only when three conditions hold:

  • the cross-subsidised product or customer group is loss-making;
  • such losses are intentional; and
  • there are linked sales making excess returns that are used to fund the loss made on the cross-subsidised product or group.

It is not clear that the existence of a loyalty penalty meets that test. Firstly, the initial discount may not be large enough to render the price in the introductory period loss-making, on an incremental basis. For many services, the cost of adding another customer may be small, relative to the fixed and common costs of the business. The cost of delivering one more copy of a magazine (printing and postage) is modest in comparison to all the costs that are shared by all the subscribers (editorial staff and print works, for example); the cost of supplying an online subscription smaller still.

Secondly, the framework for the assessment of whether cross-subsidies exist in the supply of a long-term service is not a cross-section of the customer base at a single point in time, with “newbie” prices compared with those paid by “loyals”. Rather, the assessment should be longitudinal, covering the (expected) lifetime revenues from each customer.

It is unlikely that many businesses offer  (cheap) introductory offers without expecting  that they will be able to earn compensating  profits over the customer lifetime, while  customers paying higher prices probably  benefited from introductory offers when they  first joined. So, thirdly, the long-term services  on which we focus in this article don’t usually  have a linked product deliberately chosen to  fund the loss made on the cross-subsidised one.

Nonetheless, complaints about loyalty  penalties feature strongly in consumer debates,  with repeated calls for regulators to ensure a  “fairer” outcome. Over our first 20 years,  Frontier has been involved in numerous  regulatory investigations into markets that  involve long-term service relationships.  From this vantage point we have been able  to observe how the regulators’ responses  have evolved over time. 

Complaints about loyalty penalties feature strongly in consumer debates, with repeated calls for regulators to ensue a "fairer outcome"






In their attempts to encourage the least  active customers to explore the market more  thoroughly, regulators began by putting a lot  of emphasis on improving disclosure. Examples  include the prominent warnings that have to be  given on store card statements (“cheaper credit  may be available elsewhere”), imposed by the  Competition Commission in 2007. There have  been frequent interventions by the Financial  Conduct Authority (FCA) to ensure that, for  example, general insurers have to show last  year’s premium in the annual renewal letter.

More recently, however, confidence in  these types of remedies has been waning.  The evidence suggests that they have not been  particularly effective in changing market  outcomes. And, in particular, they have not  energised the most inert customers. A sense of  frustration, combined with a shift in political  attitudes, has changed the weather.

In its 2018 green paper, Modernising  Consumer Markets, the UK Department for  Business, Energy and Industrial Strategy  (BEIS) announced:

…a new approach by government and  regulators to safeguard consumers who, for  whatever reason, remain loyal to their existing  supplier so that they are not materially  disadvantaged… Government and regulators  should be prepared to act when this is the case,  and to be adaptable and flexible in responding  to this new world, using both competition and  intervention where appropriate to get the best  possible outcomes for consumers. 

In the response to its consultation,  we may learn something more about what  it believes to be the threshold for “material  disadvantage”. But BEIS has already made it  clear in the green paper that it is focused  on those services which make up significant  proportions of household spending,  particularly among the less well off, and where  the vulnerable make up a disproportionate  share of those suffering a loyalty penalty:

…For example in the energy market, the  consumers least likely to have switched in the past  three years are consumers with one or more of the  following characteristics: household incomes under  £18,000 a year; living in rented social housing;  without qualifications; aged 65 and over; with a  disability; or on the Priority Services Register. 

In September 2018, Citizens’ Advice  launched a “super-complaint” (a right confined  to certain recognised organisations) to the  CMA. This highlighted concerns that not  enough had been done to tackle loyalty  penalty issues in five other markets: mobile,  broadband, cash savings, home insurance  and mortgages. Using Citizens’ Advice  research, the CMA put the total of the  “loyalty penalty” in these markets at over  £4 billion, and quickly responded that more  had to be done to tackle this. 

The CMA identified three types  of loyalty penalty, or occasions when  long-standing customers pay more for the  same service unless they actively intervene:

  • In some markets there is a sharp  increase after the introductory price  (a “price jump”), as in energy.
  • In others there are successive price  rises (“price walking’), as in insurance.
  • Elsewhere, customers on older tariffs  sometimes pay higher prices for  similar services (“legacy pricing”),  as in broadband. 

And it considered this most  “problematic” where:

  • suppliers make it more difficult than it  need be for customers to exercise choice
  • the price gap is large, with some paying  very high prices, or affects many people
  • it particularly harms the vulnerable
  • it happens in “essential markets”. 

Both government and the CMA have  clung to the presumption that competition  can do the trick, if it can be suitably enhanced. But one can sense the CMA’s frustration with consumers for not making competition work  better in its work on the energy market:

...our view is that the overarching feature  of weak customer response gives suppliers a  position of unilateral market power concerning  their inactive customer base and that suppliers  have the ability to exploit such a position through  their pricing policies.

So what can regulators do to wake  the dozy consumer up? There are broadly four  types of “competition assistance” they can take   or stimulate:

1. Smarter information. Regulators have  not altogether lost faith in transparency: they still debate whether the failure of past information assaults stemmed   from a poor understanding of how to communicate well with consumers – when/how/where to present the information – or a complete failure of   the information tool. Much work still continues into “sunlight remedies”. A better understanding of consume  behaviour has led to the introduction  of more powerful messages and “nudges”, which in some markets have had modest success.

2. Provider good practice rules. In its response to the super-complaint, the   CMA has listed a number of provider practices that it considers essential:

  • Exit/entry equivalence: people must be able to exit a contract at least   as easily as they can enter it.
  • Auto-renewal should generally be on an opt-in basis upfront, and include a   clear and prominent option without auto-renewal in most markets.
  • Exit fees should not be used after any initial minimum/fixed term.
  • Auto-renewal on to a fresh fixed term should not generally be used.
  • Customers must be sufficiently informed about the renewal and any   price changes (through sufficient notifications) in good time.
  • Switching should generally be managed by the gaining supplier so that customers do not have to   contact their existing supplier if   they want to move.  

Whether the CMA has the capacity or capability to supervise all markets to   ensure these “rules” are being followed is questionable, but the list will both act as a warning and influence the approach taken by sector regulators.  

3. Data portability. The government’s Smart Data Review is intended to accelerate the development and use of new data-driven technologies and services to “improve the consumer experience” in regulated markets (in other words, get the inert customer out of bed). Making switching easier is clearly key to improving the market dynamics, and that requires a mixture of technological and regulatory change.

4. Simplification of choice. Over time, regulators have come to understand   that choice – the driving force of competition – can seize up if it becomes too complicated. However, interventions requiring regulated firms to simplify   their tariffs have not always had great outcomes, and regulators have begun to lean more on.

5. Default options. One way of responding to the problem of customer inertia is to require providers to move them on to a “default” account, laying on the provider the obligation to ensure that this is in line with their interests. Such an obligation, for example, is laid on the trustees of defined contribution pension schemes, if members do not make an   active selection of an investment strategy for themselves. And the proposal on which the FCA has been consulting for a basic savings rate falls in somewhat the same category. 

6. Price caps. Common in the UK right through to the 1970s, these became   anathema to governments of the right and even the centre-left from the 1980s   onwards. Since the financial crash undermined faith in free markets, however, even a Conservative government has encouraged or imposed price caps on regulated industries. Ofgem, the sector regulator, introduced a price cap for customers with pre-payment energy meters in April 2017.   And, disregarding the CMA’s doubts, the   government required Ofgem to introduce a wider cap on standard variable rates and default tariffs in January 2019. Meanwhile the FCA has won plaudits   for imposing a cap on interest rates charged for pay day loans. However, apart from the anti-competitive effect, such interventions can prove hard to implement effectively. Humiliatingly, Ofgem had to raise its price cap on standard variable tariffs substantially, a mere two months after it was introduced.  




Loyalty penalties present regulators with a real dilemma, because, as we have seen, they do not necessarily mean that competition is ineffective; indeed perhaps the reverse. And as an aside, loyalty benefits might actually dampen competition: they make it harder for customers to leave their current provider, and consequently limit the potential for new providers to expand.

If competition is working effectively in a long-term service contract market, it is likely that the higher profits earned from longer-term customers will be invested back into acquiring those customers in the first place. Yet in major regulated markets these higher profits have caused so much anger that regulators have had to promise that they will mount a serious attack. The difficulty is to do so effectively, without further damaging competition.

A regulatory intervention that seeks to limit price increases for loyal customers will clearly alter the distribution of customer outcomes. The “waterbed” effect (push down here, it will bump up there) will create a balance of winners and losers. Which makes the first question for regulators: is such a change sufficiently necessary, or at least desirable, to merit intervention in the market?

In the extreme, a regulatory intervention would force the equalisation of prices to long-standing and new customers. Less active (switching) customers would benefit, at the expense of more active ones. Whether enforcing that change is justifiable depends on your view of the service, as well as your views of customer inertia. How essential is the service to customers (compare magazine subscriptions with energy supply)? How much do you want to encourage its persistent use (magazines versus savings accounts)? What kinds of customers are “inactive” (lazy or vulnerable ones)?

The fundamental problem may sometimes be that customers do not value choice as much as the regulators would like them to do. Of course, where that is the case, they may not object to regulatory interventions to fix prices, since that will save them the bother of having periodically to search for good new deals. But for regulators, the next question is: what will be the knock-on effects of intervention on the way competition works? With no (or less) benefit from switching, there would be less of it going on. Firms would compete less for other people’s customers, and face less of a threat of losing their own. This would reduce the incentive to improve and innovate – in short, it would blunt the positive forces of competition, and be likely to lead to higher prices in the future.

Of course, there are many different ways of intervening, some of which may constrain back-book prices without completely destroying front-book competition. Regulators may then be able to convince themselves that the net effect of capping back-book prices will be to drive prices down overall rather than push them up. But whatever the modelling shows in advance, it will take a lot of post-implementation analysis to ensure that the intervention continues to be worth the candle.




The financial sector presents regulators with a wider range of challenges than any other. Following the financial crisis, appreciation of systemic risk led to the role being split in the UK – the major economy most dependent on financial services – with prudential regulation being gradually re-absorbed into the Bank of England.

That did not, however, leave the new Financial Conduct Authority, created in 2013, with any shortage of tasks in fulfilling its three operational objectives of protecting consumers, ensuring market integrity, and promoting effective competition. The pricing practices discussed in this article have presented the FCA with particular challenges, given the specific requirements placed on regulated financial services businesses to “treat customers fairly”. And savings, home insurance and mortgages were three of the five markets highlighted in the CMA’s response to the Citizens’ Advice “super-complaint” about loyalty penalties.

Bank savings accounts had already attracted the attention of the FCA. In 2015, its Cash Savings Market Study raised concerns about back-book pricing, given the substantial proportion of accounts which were more than five years old. And the FCA introduced several “nudge” techniques to penetrate customer inertia, as Figure 7.1 (from the FCA discussion paper) illustrates:

However, the randomised controlled trials carried out by the FCA indicated that these demand-side measures would not have sufficient effect on customer behaviour. So in 2018 the regulator decided it needed to think further about supply-side intervention, and began consultation on the introduction of a Basic Savings Rate (BSR). Providers would be obliged to set a single BSR (a variable interest rate), which would apply to all easy-access cash savings accounts and easy-access cash ISAs after they had been open for a set period of time, such as 12 months.

The FCA argued that a BSR would work by pooling more active “mid-book” customers (accounts between 12 months and 2.5 years) with the inert back-book ones; its modelling work suggested this pooling would drive up rates without destroying front-book competition, and so the waterbed effect would be complemented by modest increases in interest rates overall.

A BSR would be the least restrictive of four supply-side options on which the FCA canvassed opinion: the others, in ascending degree of invention, were a superseded accounts rule, ratio-based price regulation and a total ban on price discrimination.

Nonetheless, it would clearly be a significant intervention, and many market operators have argued that it would be disproportionate to the scale of the “harm” the regulator had identified, and to the effect such an intrusive regulation was likely to have. Given the low level of interest rates on cash savings currently, the likely increase, even for the most inert of long-standing customers, would be pretty modest.

The CMA, meanwhile, has strongly supported the idea of a BSR, arguing that if its effect was modest the FCA should go further and impose an interest rate floor, while also urging the FCA to go further on the demand side by extending “open banking” (intended to be fully in place by September 2019 for current accounts) to savings accounts.

Conceptually, “open banking” means allowing one provider access to customers’ accounts held with another provider in order to offer services such as personal finance tools, and is being introduced both in implementation of a European directive and as part of a general push towards what BEIS calls “Smarter Data”. There are, however, obstacles in the way of its extension to savings accounts; and questions as to whether such sophisticated demand-side measures would really attract the kind of long-standing, unsophisticated savers the FCA is concerned about.

However, an emphasis on such measures is at least consistent with the objective of promoting competition. It is a little hard to see the CMA’s enthusiasm for an interest rate floor in the same light, along with the contortions into which it sometimes goes to argue that “weak customer response” amounts to a failure to ensure competition.

For the FCA, with a broader mixture of objectives, the sequence of questions is a little more straightforward, if still not easy to answer. If long-standing customers value long-standing relationships and dislike change, what degree of price discrimination in response to these characteristics is acceptable, and what amounts to a failure to “treat customers fairly”? What kind of intervention is justified by the degree of harm, and the degree of vulnerability of the people affected by it? And what consequences for other objectives should be weighed against the answer?