Lecturer in Law, Institute for Commercial Law Studies, University of Sheffield

E-mail: christian.twigg-flesner@sheffield.ac.uk

Copyright © Christian Twigg-Flesner 2002

First Published in Web Journal of Current Legal Issues

In early July 2002, the Office of Fair Trading (OFT) published its latest report on extended warranties. On previous occasions (OFT 1994; 1996), it had concluded that there was a lack of transparency in the pricing of extended warranties, and that retailers had often adopted pressure selling techniques to persuade consumers to buy them. In response, the British Retail Consortium (BRC) adopted a Code of Practice on the sale of extended warranties. However, the OFT had become concerned about the monitoring of this code and conducted a ’mystery shopping’ exercise. The results suggested that there were a significant number of instances of non-compliance with the BRC code, and a full investigation was launched by the OFT’s new Market Investigations Division. This considered not only problems for consumers, but also the lack of competition in the sale of extended warranties. Its findings suggest a significant lack of competition. In order to get a more detailed picture, the Director-General of Fair Trading (DGFT) has asked the Competition Commission (CC) to carry out a monopoly investigation under the Fair Trading Act 1973 (FTA)⁽¹⁾.

The purpose of this article is to examine the issues raised in the OFT’s report and to examine the legal and other issues of extended warranties (particularly those given on domestic electrical appliances). It will also consider how some of the problems could be remedied if the CC decides that there is a public interest problem in this sphere. There will first be an examination of extended warranties and why these may be attractive for consumers. This will be followed by an analysis of the relationship between extended warranties, manufacturers’ guarantees and the statutory rights of consumers under the Sale of Goods Act 1979. There will then be a consideration of the competition and consumer protection problems identified by the OFT. The final part will speculate as to the potential responses to the CC’s report into extended warranties.

---

 

An extended warranty is a contract which is purchased by a consumer and which offers protection in case the product(s) covered by the warranty break down. Usually, coverage extends beyond defects that existed at the time the product was supplied to the consumer to include defects that may arise during the lifetime of the product, such as accidental damage or break-down due to wear and tear. These are not manufacturing defects and therefore not normally covered by the manufacturers’ guarantee, nor by the buyer’s statutory rights⁽²⁾. Between one-fifth and one-third of all purchasers of electrical goods take out an extended warranty (OFT, 1994). Extended warranties are more often purchased when the product is expensive and an irregular purchase (Kelley, Conant and Brown, 1988). An important aspect of such warranties is that they are generally deferred, in that they only take effect after the free manufacturer’s guarantee has expired. Many warranties are therefore misleadingly described as “three-year’ or ‘five-year’ extended warranties, because they do not take effect immediately after purchase and therefore only cover two or four years.

There are four classes of providers of extended warranties: retailers, manufacturers, specialist extended warranty providers and credit card companies (for products purchased with their cards). This multiplicity of providers may confuse consumers. This is especially so in circumstances where they pay for a product by credit card and are persuaded to purchase an extended warranty from a retailer. The credit card company may already offer an extended warranty⁽³⁾. Furthermore, the OFT (2002) distinguishes between breakdown insurance cover and service contracts, both of which are described as extended warranties. Most extended warranties are underwritten by an insurance company, but not all. The recent failure of the Tempo chain of stores illustrates the problems that may arise in the case of extended warranties which are not backed by insurance. In this instance, customers were left without protection after Tempo had gone into liquidation and, in effect, had to pay twice for repairs (see Which?, January 2002, p.5). This issue will be considered in more detail below.

A retailer who offers an extended warranty is sending mixed messages to a consumer. On the one hand, the provision of extended warranties may suggest that a particular business is committed to after-sales service (Kelley, Conant and Brown, 1988). On the other, some consumers may lose confidence in a product they have chosen if an extended warranty is offered to them just before completing their purchase (Voss and Ahmed, 1992). Offering an extended warranty may, in fact, create a bad product image in that it might suggests a greater chance of the product breaking down soon after purchase (Kelley, Conant and Brown, 1988). Although retailers generally seem to play on the risk of product breakdown, the OFT noted that no mention is made of the difference in reliability between different brands (OFT 2002, para 4.10).

The latest OFT report indicates that 4 per cent of consumers bought extended warranties because they thought it was a ‘good idea’, and a further 37 per cent did so for ‘peace of mind’. It seems that consumers buy extended warranties in order to minimise three types of risk. First, there is the performance risk - will the product work? It has been suggested that some consumers are unduly pessimistic about the standard of reliability they can expect from a new product (OFT, 1994). In such circumstances, an extended warranty might provide a form of reassurance, to boost consumer confidence. This might explain the high volume of sales recorded in 1994 of 5 million extended warranties with a market value of £650 million (OFT, 1994). More recent figures suggest that the market could now be worth more than £1 billion⁽⁴⁾, although the OFT (2002) puts the figure more cautiously at ‘well over £500 million per year’ (para 3.3).The difficulty for consumers is that they do not have ready access to data on product reliability or repair costs (para 4.15). Such information can be obtained e.g., through reports published by the Consumers’ Association, but these are not accessible to the vast majority of consumers, either. Consumers may therefore be looking to the retailer to provide such information, but retailers may be unwilling to provide all the information they have because being aware of the small likelihood of a product breakdown may dissuade consumers from purchasing additional protection.

Second, there is the financial risk - how much will it cost to repair the product if it does not work? A consumer may believe that the price of an extended warranty will be lower than the cost of potential repairs and will therefore opt to buy such a warranty. He will have to balance the potential expenditure on repair over the period covered by the extended warranty with the cost of the extended warranty. Consumers who are risk averse are likely to buy an extended warranty (OFT, 1994). Some consumers may previously have encountered a product failure and may buy an extended warranty to minimise the potential cost of a further breakdown (Voss and Ahmed, 1992). Others might purchase an extended warranty because they anticipate that they will subject the product to heavy use which increases the likelihood of it breaking down (Padmanabhan, 1995). Moreover, if a product is used frequently by a number of different users, it might make more sense from the consumer’s perspective to acquire an extended warranty. In contrast, if only the consumer himself will use the product, he may decide that by taking adequate care, he can avoid product breakdowns and thus not need an extended warranty. Those consumers who believe they will need to rely on an extended warranty have the greatest incentive to buy it. This is known as ‘adverse selection’ (cf. Holdych and Mann, 1996, p.797). Consumers who expect a product to break down because of the way they use it are more likely to take advantage of the extended warranty if this will protect them against the financial consequences of a product breakdown.

The third factor is the potential effort that may be required to effect repairs when the product fails. Extended warranties may help because the warranty provider will have put in place a mechanism for ensuring the speedy repair of the product. Consumers may prefer this to having to arrange for repairs themselves⁽⁵⁾.

These seem to be the most significant factors that influence consumers in deciding whether to take out an extended warranty. However, consumers are already protected against many (although not all) types of product failure under the Sale of Goods Act 1979 and under any voluntarily given manufacturers’ and retailers’ guarantees. Furthermore, many household insurance policies provide some protection against the cost of repairing faulty items. The desire of consumers to buy extended warranties may therefore be based on a limited understanding of their existing protection. The OFT can be criticised for not considering this point in any detail in its report. The following part will consider the relationship between consumers’ legal rights and the additional protection offered by extended warranties.

The OFT accepts that there are consumers who want to buy extended warranties, but it does not consider whether consumers should need to buy extended warranties in the first place:

The OFT’s assumption therefore is that there is demand for extended warranties. However, it may be dangerous to make such an assumption without investigating the underlying causes of consumer demand for extended warranties. The OFT accepts that consumers buy extended warranties to get peace of mind, but it does not consider if consumers really need to buy additional protection. Consumers already enjoy a significant degree of ‘free’ protection, both under the Sale of Goods Act 1979 (SoGA) and under free guarantees offered by manufacturers. This section will examine the relationship between the rights under the SoGA, free guarantees and extended warranties. It will become apparent that extended warranties can be useful, but probably only for particular categories of consumers. However, these consumers do not seem to be the main purchasers of extended warranties in practice. The OFT has failed to identify this mismatch.

The SoGA requires that goods sold by a seller to a consumer should be of satisfactory quality (the term implied by s.14(2)), and makes available a range of remedies if goods do not meet this standard (see Twigg-Flesner and Bradgate, 2000). Compliance with this term is ascertained at the time that risk passes from seller to consumer, which is usually at the time of sale⁽⁶⁾. In order to assess whether a particular product is of satisfactory quality, it is necessary to consider the description of the product, its price (if this is relevant) and all the other relevant circumstances (s 14(2A)). S.14(2B) provides a list of factors which may be ‘relevant circumstances’ in some cases. One of these factors is ‘durability’. However, the Act does not define ‘durability’ further. A useful explanation of what is meant by ‘durability’ in a non-legal sense is provided by Jarmin:

This can be contrasted with ‘reliability’:

On this basis, a product would not be ‘reliable’ if it failed to perform the function it was designed to perform, either on a permanent or on a temporary basis. The product would not be ‘durable’ if it required maintenance or repair on a very frequent basis. For example, a washing machine which does not spin on some occasions may be regarded as unreliable, whereas a washing machine which fails to work after a few washes would not be durable. The distinction between these concepts should not be overstated, although it is nevertheless important to appreciate that they are not identical.

Although these explanations do not interpret the meaning of ‘durability’ for the purposes of s.14, they may nevertheless be of assistance. In the context of the ‘satisfactory quality’ test, an important factor in considering the durability criterion could be whether a particular product has ‘endured’ sufficiently. The SoGA does not require that products have to be durable for a specific period of time. It would be impossible to specify a suitable period in a statutory standard. Rather, the durability of a particular product in relation to the durability to be expected of the type of product generally is considered. For example, it is common for washing machines to work without problems for five or six years (or even longer). If a particular washing machine broke down after seven months of normal use, it would seem probable that there was something wrong with it at the time of sale. The consumer would therefore have a good chance of claiming successfully that the washing machine is not of satisfactory quality⁽⁷⁾.

It is difficult to say with certainty for how long a particular product should perform without breaking down before durability is no longer a relevant factor in applying the satisfactory quality test. For example, in Thain v Anniesland Trade Centre ((1997) SLT 102, Sh Ct.), the court suggested obiter that a new car could be expected to work trouble-free at least for the duration of the manufacturer’s guarantee. Although this comment was not made directly with reference to the durability criterion in s.14(2B), it could be followed in a subsequent case. There are two points to be made against this. First, referring to a manufacturer’s guarantee in applying the satisfactory quality test could result in a reduction in consumers’ rights. This is because the guarantee period is determined by the manufacturer, and taking this into account would in effect operate as a limitation of the scope of the satisfactory quality test. This contradicts the general policy that consumers’ rights should not be limited or excluded (see e.g., s.6 of the Unfair Contract Terms Act 1977). At best, it may be possible to argue that there is a presumption that goods which fail before the guarantee period has expired were not of satisfactory quality. Secondly, the court’s suggestion seems to conflict with certain comments of Mustill LJ in Rogers v Parish (Scarborough) Ltd ([1987] QB 933, CA). Here, the claimants had bought a Range Rover which cost £14,000. This was defective but the claimants agreed to a replacement vehicle. However, there were numerous problems with the replacement car, and the claimants eventually sought to reject the car and obtain a refund. One of the questions to be considered by the Court of Appeal was whether the existence of a manufacturer’s guarantee was relevant in considering whether there had been a breach of the term implied by s.14(2). The crucial passage is this:

It would therefore be wrong to suggest that a product is of satisfactory quality if it has not failed during the guarantee period, but soon after, because this would have the effect of allowing the guarantee to operate as a limitation clause. It would also conflict with Court of Appeal authority. In any event, Thain does not rule out that ‘durability’ would also be a relevant factor if a product failed outside the guarantee period. Although there is no further guidance available from court judgments (the scope of ‘durability’ in the context of the satisfactory quality test has yet to be considered in detail by a higher court), this must surely be the case. It was suggested above that the durability factor is only then relevant where a particular product has broken down significantly sooner than would ordinarily be expected of a product of the same type⁽⁸⁾. It may be possible to produce empirical evidence of the average time a product will give trouble-free use before it is likely to develop a fault. Products which fail at a point significantly before that average time would then not be of satisfactory quality.

It is therefore apparent that consumers already enjoy some protection against latent defects which existed at the time of sale and caused a product to break down much sooner than could reasonably be expected. There is, however, a practical difficulty. If a product breaks down soon after purchase, it will be relatively easy to show that there was something wrong at the time of delivery. However, the longer the period since delivery, the greater the burden on the consumer to prove that the product was faulty when he got it. A consumer may have to go to court and produce detailed evidence to convince a judge that the product was not of satisfactory quality, and one may suspect that few consumers will be able, or willing, to do so. An extended warranty would obviate the need for having to prove this because as long as the product breaks down during the warranty period, the provider will make available a remedy in accordance with the terms of the warranty. In effect, an extended warranty could provide an uncomplicated means of obtaining redress if a product has broken down.

If a product is not of satisfactory quality, a consumer is currently entitled to reject it and terminate the contract if he has not yet ‘accepted’ the product (s.11(4) SoGA). Additionally, he can claim damages for non-delivery (s.51 SoGA). If he is deemed to have accepted the product, he will only be entitled to bring a claim for damages. In consumer cases, ‘acceptance’ is most likely to occur after a reasonable period of time has lapsed. The length of this period will depend on the nature and complexity of the goods (Bernstein v Pamsons Motors (Golders Green) Ltd. ([1987] 2 All ER 220), although it is often likely to last for no more than four to six weeks after sale. Damages are assessed on the basis of the difference between the value of the goods at the time of delivery, and the value the goods would have had, had they been in compliance with the implied terms (s.53(3) SoGA). Usually, this will cover at least the cost of repairing the product.

In contrast, extended warranties will often promise a specific remedy of repair if a product covered by the warranty has broken down. A replacement may be provided if the product cannot be repaired. In this respect, extended warranties offer an additional benefit to consumers, because they are given a contractual entitlement to demand repair. This is not something currently available under the Sale of Goods Act. However, Directive 99/44/EC on Consumer Sales⁽⁹⁾ will be implemented into UK law by the end of 2002. This requires, inter alia, the introduction of repair and replacement as remedies under domestic sale of goods legislation. The current advantage offered by extended warranties in this respect will therefore become less significant.

The preceding discussion has also mentioned that there may be a manufacturer’s guarantee available. Such guarantees are given free of charge and promise that a product is free from defects in workmanship and materials. Should a problem materialise within a specified period of time after purchase, the guarantor usually undertakes to repair or replace the product free of charge. An obvious advantage of such guarantees is that they give consumers an alternative target if something has gone wrong. Under the SoGA, only the final seller is liable⁽¹⁰⁾.

As with the satisfactory quality test, the guarantee only covers faults which existed at the time of sale. Most guarantees expressly exclude faults which are the result of misuse by the consumer, accidental damage or normal wear and tear. Moreover, guarantees are usually only given for a period which is relatively short if compared to the potential service life of the corresponding product. Although there can be little doubt that product quality continues to improve, the most common guarantee period is still twelve months. There are some products which come with longer guarantees, but these tend to be products which are less complex and easier to maintain (a good example being steel bedframes).

It has been suggested in the past that manufacturer’s guarantees could be an indicator of product quality. In giving a guarantee, a manufacturer may be viewed as ‘signalling’ to a consumer that a product meets a certain level of quality (cf. Priest, 1981; Willett, 1992; Agrawal et.al., 1996). One view is that the decision to offer a guarantee signals general confidence in the product (Willett, 1992), but since the giving of a manufacturers’ guarantee has now become common practice, it would probably be an unwise commercial decision not to offer a guarantee. A more interesting suggestion is that by offering a guarantee on terms which are better than those of guarantees given on competing products (such as a longer duration), a manufacturer may signal that his product is of higher quality than competing products. The court’s view in Thain v Anniesland Trade Centre (1997) that goods should perform without trouble at least during the guarantee period seems to reflect the notion that a guarantee can be a signal of quality. If this theory is correct, it may be expected that a product which will give at least four or five years of trouble-free service is accompanied by a four-year guarantee. Such longer, free guarantees would reduce the need for additional, purchased protection in the form of an extended warranty.

However, this is usually not the case, and guarantees tend to have a much shorter duration than the average life-span of the corresponding product. A significant body of research has since identified numerous factors that affect the signalling function of guarantees (see in particular, Agrawal et.al., 1996; Wein, 2001). Moreover, if the guarantee is to signal product quality to the consumer, it assumes that a consumer will (a) take the guarantee into account in deciding whether to make a purchase and (b) is able to examine the guarantee before purchase. As most consumers do not do so, it seems unlikely that guarantees have a strong signalling function. Moreover, there are a number of factors at play which further limit the degree to which a guarantee can be an adequate signal of product quality. These are primarily relevant in setting the duration of a free guarantee. First, when a faulty product is returned to the guarantor, the initial difficulty is to establish whether the fault is a result of poor workmanship or due to consumer misuse. The longer a consumer has used the goods, the greater the likelihood that a fault will have been caused by the way the consumer has used the product (Eddy, 1977, p.844). By restricting the guarantee period to a relatively short term after purchase, the manufacturer minimises the likelihood of providing a remedy for faults which were, in fact, caused by the consumer. Secondly, consumers do not use products in a uniform manner – some take a lot of care whereas others do not. If high-intensity use will increase the likelihood of a product breakdown, then a guarantor will set the guarantee period such as to minimise the number of claims made as a result of such use (Emons, 1989). Thirdly, and linked to the previous points, there is moral hazard (Lutz, 1988). This is a concept which, in essence, refers to the fact that the availability of insurance is likely to reduce the efforts made by the consumer to take care of the product in order to prevent the occurrence of the event insured against. A guarantee can be characterised as a form of insurance against latent manufacturing faults⁽¹¹⁾. If a guarantee is available, it is less likely that a consumer will take as much care of a particular product compared as he would if the product were not covered by a guarantee. A guarantor will take into account the likelihood of claims which are the result of inadequate care and seek to minimise such instances by reducing the guarantee period.

These important factors offer at least a partial explanation for the comparatively short duration of manufacturers’ guarantees. They do not, of course, explain why many guarantees are still restricted to one year. There are only some manufacturer’s guarantees which last for a longer period (for example, a Japanese car manufacturer has recently advertised that all their cars are covered by a five year guarantee). Perhaps this can be explained by the fact that some manufacturers themselves offer extended warranties to consumers. Such manufacturers target consumers who have registered their free guarantee with the manufacturer. Such extended warranties are an additional source of income for manufacturers and may also be a significant factor in the decision to offer a free guarantee limited to a relatively short period.

In light of these points, it is now possible to suggest that extended warranties have an important role to play in providing additional protection to consumers. For this, it is useful to refer to the theory of the consumer guarantee put forward by George Priest (1981) in what is still the most influential theoretical examination of manufacturers’ guarantees. Priest argued that the substance of guarantees, i.e., the terms and conditions on which guarantees are provided, is best explained on the basis of his ‘Investment Theory’ (or ‘Comparative Advantage Theory’ – see Schwartz and Wilde, 1983, p.1398). In essence, a guarantee divides responsibility for allocative investments, i.e., preventive actions to minimise the risk of a product breakdown, between guarantor and consumer. The party who is in the best position to assume responsibility for a particular problem should do so. The basis of this theory is therefore the ‘least-cost avoider’ (Oughton and Lowry, 2000, p.205). Thus, the manufacturer can improve the design of the product to minimise the probability of the product needing repairs, but only in so far as this would be economically viable. The consumer can take sufficient care when using the product to reduce the likelihood of a product breakdown, and may also ‘invest’ in prolonging the life of the product by carrying out some minor repairs himself. Priest’s example is that of a fridge – the consumer can prevent his children from swinging from the fridge door, and he can replace a shelve inside the fridge that may have fallen out.

Market insurance would cover those aspects which cannot be subject to allocative investments. Guarantees offered by manufacturers are one form of market insurance. Although the consumer does not pay separately for the guarantee, the price of the product includes a sum for the guarantee. However, as demonstrated above, the amount and intensity of use of a particular product vary between consumers, and guarantees are generally limited to reflect this (Priest, 1981, p.1317). The longer the consumer uses the product, the greater is the likelihood that a product failure is caused by the consumer’s way of using the product. The guarantee is therefore limited to a period of time after purchase during which product breakdowns are more likely to be the result of a manufacturing defect. The consumer is expected to take sufficient care to ensure that the product does not break down after that period⁽¹²⁾. If the consumer wants further protection, he will have to obtain this through additional market insurance. Extended warranties provide such additional insurance.

Moreover, extended warranties offer protection for problems which are not covered by the SoGA or free guarantees, such as accidental damage and breakdown due to normal wear and tear. The consumer can be expected to invest in minimising the risk of accidents, but will not be able to eliminate it altogether. Thus, for accidental damage and other product breakdowns, insurance is provided in the form of extended warranties. One aspect of this function of extended warranties is illustrated by the decision in Thain v Anniesland Trade Centre (1997). Here, a second-hand car which was about five years old and had travelled about 80,000 miles had a latent defect which resulted in a fault in the gear-box within a fortnight of purchase and the car was put off the road. The court held that the car was still of satisfactory quality. The defect did not exist at the time of sale but was likely to develop at any time with a second-hand car. It is significant to note that the court put some emphasis on the fact that the seller had offered an extended warranty which would have provided some protection against the risk of expensive repairs. In declining this offer, the consumer accepted the risk of expensive repair immediately. This aspect of Thain is interesting. The car met the level of quality a reasonable person would expect. There was a risk that a defect might develop, but this did not affect the satisfactory quality of the car and there was therefore no breach of the implied term. Had the consumer wanted additional protection against this risk, she would have had to buy additional insurance in the form of the extended warranty offered to her. Thain therefore provides an example of how extended warranties and statutory rights might interact. However, it is debateable whether the decision in this case is correct. It seems at least arguable that there was a breach of the implied term because the car failed very soon after purchase. The fact that an extended warranty might have offered protection seems to have been a significant factor in the court’s decision. It is not clear whether a similar decision would have been reached had no extended warranty been offered. The worrying aspect of this case is the extent to which the availabilty of additional protection influenced the application of the satisfactory quality test.

The preceding discussion suggests that there are clear rationales for the existence of extended warranties. The OFT has accepted the desire to obtain peace of mind as the reason why most consumers purchase extended warranties. This discussion suggests that this is too simplistic an analysis and that there are more complex factors at play. The OFT’s failure to analyse this point is a major omission. It states that its objective has been to examine whether the market works well for consumers (para 2.6). It must be asked how it can do so without considering the relevant legal framework and come to useful conclusions. A fundamental conclusion arising from the preceding paragraphs is that extended warranties should generally be purchased by higher risk consumers. Yet, the OFT found that it is generally low-income consumers (Class C2DE) who purchase extended warranties (para 4.48). It seems rather unlikely that all higher-risk consumers are also on low incomes. It is regrettable that the OFT has not examined this aspect in depth, and consequently, some doubts may be raised about the value of the report’s conclusions.

Consumers may value the reassurance that extended warranties can bring, but one objective for any system of consumer protection is to ensure that consumers are better informed about the choices they make. The OFT has identified a number of problems with such warranties, some of which are classed as ‘competition’ problems, whereas others are regarded as ‘consumer protection’ problems. These will be examined in the following sections.

Prior to considering the competition and consumer protection problems, it is useful to set out the alternatives to extended warranties. If a consumer does not buy an extended warranty and a product breaks down outside the manufacturers’ guarantee period, he might be able to ask the retailer to provide a remedy. However, as explained above, this is only possible if the cause of the breakdown was inherent in the product at the time of sale, and only for a limited time after sale. For other faults, consumers will have to pay to have the product repaired. There are both independent repairers and repair networks set up by manufacturers, as well as some retailers, who provide such a service. Although consumers will have to cover the cost of repairs themselves, the OFT’s report suggests (at para 4.47) that repairs are generally likely to cost less than an extended warranty, although the report does not come to a definite conclusion on this point.

The OFT examined both competition and consumer protection problems caused by extended warranties. This section will consider the competition problems. Although these are dealt with separately, there is a considerable overlap with consumer protection concerns, and the competition problems identified by the OFT have an impact on consumers in that they limit consumers’ choice, which in turn has an impact on the price they pay.

There is considerable variation in the scope of extended warranties, as noted by the OFT (1994) and the Consumers’ Association (1994). For example, some domestic appliance warranties offer a replacement of the faulty product on a ‘new-for-old’ basis. In contrast, others take account of the depreciation in product value that will have happened over time. There is further divergence in the way that payment for repairs to a faulty product are handled. In some cases, the warranty provider will pay for the service directly whereas in others the consumer will have to pay for the repair first and then claim back the money from the warranty provider. Finally, there can be considerable variation in the various terms and conditions, such as the exclusion of certain items or parts of appliances, or terms dealing with related service, such as the fixing of appointments.

One would, therefore, expect the market for extended warranties to be a healthy, competitive one. Yet, there has been a much lamented absence of any obvious competition (OFT, 1994). Most retailers seem to offer only one type of extended warranty for the products they sell. Furthermore, these are offered for products sold by that retailer. Only a few retailers offer extended warranties on products purchased elsewhere, but this is not widely advertised (OFT, 2002).

The OFT concluded that retailers have such a strong position in the sale of extended warranties that extended warranties offered by manufacturers present only a small challenge to the retailer’s ‘point of sale advantage’ (para 4.26). One explanation for this is that manufacturers do not make a significant effort to compete with retailers on the provisions of extended warranties. Retailers are, of course, the main customer base for manufacturers – few manufacturers sell direct to customers. Many retailers are of equal or stronger bargaining power with manufacturers and could decide to cease stocking a particular brand if the manufacturer also offered an extended warranty to consumers at the point of sale. Manufacturers therefore often make a conscious decision not to compete at point of sale and limit their business to consumers who have registered their free guarantee.

This is an issue which the Competition Commission will undoubtedly have to investigate further. It must be noted that an understanding between a retailer and manufacturer that the manufacturer will not offer an extended warranty that competes with the retailer’s warranty, may breach the Chapter I prohibition of the Competition Act 1998 (CA). Section 2(1) CA prohibits agreements or concerted practices which have as their object or effect the prevention, restriction or distortion of competition within the United Kingdom. In agreeing not to offer an extended warranty that would compete with a retailer’s warranty, a manufacturer and retailer are in effect reducing competition by limiting the choice that is available to consumers⁽¹³⁾.

A further factor is the variation in the scope of protection which, whilst positive for competition, may also make it more difficult for consumers to shop around for the ‘best’ extended warranty. This is because consumers often find it difficult to understand exactly what is provided by the various warranties on offer (OFT, 2002, para 4.23).

The lack of competition is also blamed for the high cost of extended warranties in relation to the likely cost of repairs. A fundamental tenet of competition policy is that greater competition will lead to a reduction in prices. If competition on extended warranties were to intensify, consumers should have to pay less. However, increased competition may not be the solution to this problem. It is necessary to examine extended warranty pricing more closely.

An important question is whether the price of an extended warranty bears any relation to the likely cost of repairing or replacing a product under the warranty. It seems that this is often not the case. The OFT’s initial examination of extended warranties on electronic goods (OFT, 1994) noted that the cost of repairing a product is usually lower than the cost of taking out an extended warranty. The estimated repair costs for selected electrical goods were all lower than the prices of extended warranties offered for these products. In some instances, the warranty was twice as expensive. Consequently, consumers pay over the odds to obtain peace-of-mind protection.

At the same time, the profit margins on extended warranties appear to be wide - businesses seem to profit substantially from consumers purchasing extended warranties. Extended warranties are a significant source of income, particularly for retailers. Indeed, research indicates that many businesses in the electronics sector rely on the sale of extended warranties to make a profit⁽¹⁴⁾. In some cases, a retailer can have a profit margin of as much as 80 per cent of the total price of the extended warranty. Furthermore, it seems that the ratio of claims under extended warranties is considerably below the average claims ratio for insurance products generally (para 4.31).

A partial explanation for the high cost of extended warranties may be that this takes account of the expenditure incurred in providing a remedy under a free guarantee (Chen and Ross, 1994). By offering extended warranties with prices exceeding expected costs, manufacturers and retailers could recover the cost of additional service under normal guarantees. This may be preferable to including the cost of free guarantees in the original product price to be more competitive. However, this would probably only be of relevance for warranties provided by manufacturers, and the proportion of such warranties is relatively small. This cannot therefore be a sufficient reason.

Extended warranties are a type of insurance, and the objective of insurance is to spread the cost of likely losses among a large number of people to lower the actual cost to any particular individual suffering loss. Thus, one might reasonably expect the cost of an extended warranty to be lower than the actual cost of repairing a defective product. The fact that this is generally not the case warrants a consideration of the factors that may be relevant in determining a suitable premium for an extended warranty. The following paragraphs seek to explore this question, but cannot provide a definitive answer.

There are a number of factors which need to be taken into account by a warranty provider in setting the price of, or premium for⁽¹⁵⁾, an extended warranty. The basic starting point is to estimate the risk of occurrence of the event insured against. Thus, an extended warranty provider needs to consider how many units of a particular product are likely to break down during the period covered by the warranty. Furthermore, he needs to be aware of the cost of repairing the product once it has broken down. For example, it is assumed that around 35 per cent of all Brand X washing machines will develop a fault during the first five years of use, and the average cost of repair is £50. In order to cover the cost of providing insurance against product breakdown, the insurer has to collect sufficient premiums to cover the anticipated costs. In this example, the basic premium can be established by multiplying the risk by the cost of repair, and the insurer needs to collect £17.50 from every insured person (cf. Shavell, 1997).

The insurer needs to be sure that there are sufficient pooled assets (i.e., the total of all premiums received) to meet all the claims that may be made. There are a number of factors which affect the calculation of the insurance premium. As noted previously, with any insurance policy, there is a moral hazard problem (Lutz, 1989). This is the problem that an insured who might take greater care to prevent the occurrence of the event insured against if no insurance were available will be less careful and thereby increase the likelihood of having to draw on the insurance policy. This may lead to a greater number of claims than anticipated. Additionally, those who anticipate that they will have to make a claim are more likely to take out insurance than those who think that they will not need the insurance (adverse selection - see e.g. Akerlof, 1970). Both moral hazard and adverse selection are further factors that may result in a higher premium. Moreover, those consumers who buy extended warranties are likely to be higher-risk consumers (see also Padmanabhan, 1995). Priest (1981) suggests that premiums for extended warranties are likely to be higher because the consumers who take out an extended warranty have a greater than average risk of needing the insurance (p.1317). It must be remembered that this is the theoretical position rather than practice – as noted previously, those who actually buy extended warranties are on low incomes and often not high risk consumers.

All these factors indicate that the number of claims made under the extended warranty policies may be greater than the average. In order to reduce the likelihood of not being able to meet all claims, the insurer has to increase the ‘risk bearing capacity’ of the pool, i.e., the amount by which the value of the pool exceeds the average. The simplest way of doing this is by increasing the premium paid by the insureds to a sum which exceeds the average expected loss (Purves, 2001, p.634). Moreover, the insurer will also need to take into account his overheads, such as the cost of administering the extended warranties, staff costs and maintaining business premises. At least some of these costs need to be worked into the premium. The premium will need to be based on more than the risk of loss and the likely amount of loss. It is not possible within the confines of this paper to provide specific figures, but it would seem strange if the premium exceeded the actual loss that may be suffered by an individual insured. Yet, this exactly seems to be the case with most extended warranties. In respect of this, it is significant to note that the OFT found that mot providers do not calculate the premium for an extended warranty on the basis of the actual risk of failure of a particular model or brand, but instead charge ‘catch-all’ standard fees. The only variations that were identified were based on the price bands for the relevant products (OFT 2002, para 4.13). The perception that extended warranties generally cost too much is reinforced by this fact.

There is therefore reasonable evidence to suggest that the price of extended warranties is excessive, even when one takes into account the fact that those consumers who take out extended warranties are likely to be higher-risk consumers. The Consumers’ Association’s report (1994) included a table of figures which set out the basic likelihood of product breakdown and identified a average repair cost for various products. However, these calculations were not sufficiently detailed to provide an accurate picture. The Competition Commission should undertake a detailed analysis of this point and identify a realistic level of premium based on the various factors that may be taken into account by extended warranty providers. This might make it possible to consider whether increased competition could have a noticeable impact on the premiums paid by consumers for extended warranties. Prima facie, the high profit margin and low number of claims suggests that premiums are too high, but it is not clear by how much. This needs to be investigated further.

The lack of competition and the high premiums are often regarded as inter-linked issues. Obviously, this has a detrimental impact on consumers in that they have to pay more for extended warranties compared to a competitive market. In addition, there are specific consumer protection problems, which are considered in the next section.

The OFT report identified three aspects of concern to consumers, in addition to the high cost of extended warranties. First, it seems that some consumers feel pressured into taking out an extended warranty. Secondly, some warranties contain terms which may be unfair or could operate unfairly against consumers. Finally, the fact that not all providers of extended warranties appear to be subject to regulatory controls means that some consumers may be left without protection if the warranty provider goes out of business. These issues will be examined in turn.

Although it was seen above that there are good reasons why consumer might want to purchase extended warranties, it seems that most consumers do not set out to buy an extended warranty when they shop for a particular product. Most extended warranties are purchased as an afterthought and generally only after the retailer has drawn their availability to the attention of the consumer (OFT, 2002, para 4.1). Providers of extended warranties target potential customers at two stages - at the point of sale of the original product and towards the end of the manufacturers’ guarantee period by direct mail (OFT, 1994). However, retailers seem to be in a significantly stronger position, with 71 per cent of consumers stating that extended warranties are available from retailers, and only 23 per cent mentioning manufacturers and 14 per cent insurance companies (OFT, 2002, para 4.21).

Particular concern has been expressed with regard to pressure-selling techniques for extended warranties adopted by retailers at the point of sale. Some retailers market extended warranties quite aggressively at point of sale, whilst others sell these in a passive manner (Day and Fox, 1985). The second OFT report (OFT, 1996) discovered that many retailers did not volunteer information about manufacturers’ extended warranties, or claimed that these were not available because the retailer himself offered an extended warranty. This practice obviously prevented consumers from being able to make an informed decision when they contemplate buying an extended warranty. Salespeople are often encouraged to market extended warranties because they are paid a commission on warranty sales. In response to the OFT reports, the BRC code of practice for extended warranties requires retailers to provide information about extended warranties and not to push the sale of their extended warranties. However, a recent mystery shopping exercise (Taylor Nelson Sofres, 2001) still showed significant problems, which resulted in the latest OFT investigation into the provision of extended warranties (OFT, 2002). The BRC code appears to have been unsuccessful, and it is interesting to see that there is no intention to apply for recognition of the code under the OFT’s new criteria (OFT, 2002, para 2.3). Consequently, the limited degree of control offered through the current self-regulatory regime would be lost and there would be no regulatory framework for extended warranties at all.

The code was aimed at promoting transparency in the sale of extended warranties and at allowing consumers to make a choice, rather than to be pressured into buying a warranty. The OFT’s research suggests that around 9 per cent of all consumers felt pressured to buy an extended warranty, and a further 33 per cent were encouraged to purchase one (para 4.5). These figures are based on all consumers who bought domestic electrical goods. Extended warranties were not discussed as part of all of these transactions. If only those transactions where an extended warranty was discussed are taken as a basis, 15 per cent of consumers felt pressured and almost 60 per cent were encouraged to buy an extended warranty (OFT 2002, para 4.5). The OFT does not explain where it has drawn the line between ‘pressure’ and ‘encouragement’, but it does seem that the retailers still take the lead in offering extended warranties, rather than to respond to consumer enquiries about such warranty. Often, this is the result of the sales techniques adopted by retailers, who paint a bleak picture of the reliability of new products. As consumers are generally not able to establish the likelihood of a product breakdown before purchase, the retailer’s comments may be very influential. Moreover, consumers from low-income groups are more likely to buy an extended warranty (para 4.48). Such consumers may find the prospect of having the unexpected and unquantifiable cost of repairs or replacement particularly daunting. This is of concern because consumers who are already financially stretched are committing themselves to further expenditure which, on balance, may be unnecessary. This is an aspect which the OFT has failed to give sufficient attention.

It may be that part of the problem is a lack of awareness of the existing level of protection under the Sale of Goods Act, both on the part of consumers and retailers. Greater awareness of their statutory rights may encourage more consumers to query the need for additional protection, particularly when faced with claims by a retailer that the product may break down unexpectedly soon after purchase.

The OFT (2002) noted that most terms and conditions do not seem to be problematic, although it did raise two features which cause some concern. First, some extended warranties include a provision by which depreciation in the value of goods is taken into account. This may, in some circumstances, result in a product being deemed a write-off and no repair or replacement being provided under the warranty (para 4.54). Such terms may be regarded as unfair under the Unfair Terms in Consumer Contracts Regulations 1999. These Regulations implement Directive 93/13/EEC on Unfair Terms in Consumer Contracts. They apply to contractual terms which have not been ‘individually negotiated’ (Regulation 5(1)). These are terms which have been drafted in advance and the substance of which the consumer has not been able to influence (Regulation 5(2)). A pre-drafted term will be regarded as unfair ‘if, contrary to the requirement of good faith, it causes a significant imbalance in the parties’ rights and obligations under the contract, to the detriment of the consumer’ (Regulation 5). In Director-General of Fair Trading v First National Bank plc [2001] UKHL 52, [2001] 1 All ER 97, the House of Lords noted that ‘good faith’ has procedural and substantive aspects (per Lord Bingham, paragraph 17). With regard to the procedural aspects, good faith requires that a consumer should be made aware of all the relevant terms, and particular attention should be drawn to important or onerous terms. As far as the substantive element is concerned, the House of Lords held that some terms always create a significant imbalance to the detriment of a consumer (per Lord Steyn, paragraph 37). Such terms are inherently contrary to the requirement of good faith and therefore unfair. In assessing whether other terms are unfair, it is necessary to take into account the nature of the goods or services covered by the contract, to consider all the circumstances attending the conclusion of the contract and to the other terms of the contract (Regulation 6(1)). If a term is regarded as unfair, then it will not be binding on the consumer (Regulation 8(1)). No assessment will be made of the fairness of a term which defines the main subject matter of the contract, nor of the adequacy of the consideration provided by the consumer (Regulation 6(2)). It is still possible to refer to these when assessing the fairness of other terms in the contract (see Bradgate, 2000, p.92). The Regulations contain a set of rules for their enforcement by public authorities and other organisations with a legitimate interest in protecting consumers (see Twigg-Flesner, 2000). It is possible for such organisations to challenge terms. This enforcement mechanism is co-ordinated by the OFT, which also publishes an Unfair Contract Terms Bulletin that provides details of terms which have been challenged successfully.

The concern in respect of depreciation clauses is that consumers may discover that they have paid for an extended warranty, but will receive little or nothing in return if the warranty provider declares that the product is a write-off. Such a provision appears to be contrary to the purpose of an extended warranty, because a consumer would effectively have to pay twice if the product breaks down – once for the extended warranty which then does not provide a great deal by way of remedy, and then again to obtain a replacement product. The 1999 Regulations could be used to control such terms to ensure that consumers do not suffer unnecessary detriment as a result of such terms, but to the extent that such terms are regarded as fair, the question remains whether these are always applied correctly.

A second feature is the ‘cash back’ provision in some extended warranties. These provide that a consumer can obtain a refund of the premiums paid if no claim is made under the extended warranty. The procedures for claiming a refund are described as ‘rigid and complex’ (para 4.55) and require a claim to be made within strict time limits. The Financial Ombudsman and the OFT have both challenged such terms under the 1999 Regulations (see para 4.56). If such terms are insufficiently clear and precise, they may be interpreted in favour of the consumer, but in individual cases, it may be difficult for a consumer to succeed with his claim for a refund of the premium.

It may therefore be concluded that there are no obvious concerns regarding the terms and conditions on which extended warranties are provided. Those terms which have the potential to be unfair can be controlled under the 1999 Regulations. All that may be required in this regard is to step-up the policing of extended warranties to ensure that terms which may be unfair are no longer used. However, what is not clear is what actually happens when a consumer claims under an extended warranty. The OFT does not provide any evidence on the administration of extended warranties, and what looks acceptable in print may not work in the consumer’s favour in practice. A problem of much greater concern is how the premiums are used by the extended warranty provider and what might happen if the provider becomes insolvent.

On a number of occasions, providers of extended warranties have become insolvent and consumers been left without any protection. If an extended warranty provider becomes insolvent, the very least a consumer will want is to be able to recover his premiums. However, consumers who have paid premiums to a provider are likely to be treated as unsecured creditors in the event of the trader’s insolvency. This means that their claim to any money still held by the provider will be considered only after those of the secured creditors⁽¹⁶⁾ have been dealt with. All unsecured creditors are treated equally and they will have to share any remaining funds rateably between them. Consequently, they will receive back less than they are owed, if anything at all.

If a consumer is to be able to reclaim the premium, it is necessary to identify the premium in the hands of the insolvent warranty provider. In effect, a consumer would have to argue that the premium is held on trust for the duration of the extended warranty policy⁽¹⁷⁾. This type of trust is know as a Quistclose trust after the leading case Barclays Bank Ltd v Quistclose Investments Ltd ([1970] AC 567). Such a trust was established in the context of pre-payments made by consumers for goods ordered by mail-order in Re Kayford Ltd ([1975] 1 WLR 279).

There are a number of requirements that need to be fulfilled in order to succeed with such a claim. First, the consumer would have to prove that there was an intention to create a trust over the premium payments. It seems that in the normal course of dealings, the presumption is that a provider should be free to use any payments made by a consumer for his normal business activities. The provider will not usually keep these payments separate from other funds needed for his business. It is unlikely that the presumption would be displaced if the provider kept the premium payments in a separate account without further evidence of an intention to hold these on trust. This is well illustrated by Re Multi Guarantee Co Ltd ([1987] B.C.L.C. 257, and see Campbell, 1987). Multi Guarantee (MG) was a specialist provider of extended warranties. These were provided inter alia to customers of an electrical appliances retailer. The retailer would collect the premiums and pay these to MG in return for extended warranty cover. Once the premiums had been paid to MG, they became part of MG’s assets. The retailer became concerned about the insurance arrangements made by MG⁽¹⁸⁾ and steps were taken to improve security of the premiums paid by the customers by setting up a joint bank account to which the premium payments were transferred. When it had been confirmed that MG’s insurance position was inadequate, MG petitioned for winding-up. The question arose whether it had constituted itself as trustee over the funds in the joint account. The Court of Appeal held that because both MG and the retailer might have a claim to he money⁽¹⁹⁾, it was not clear who had a right to make withdrawals from the fund. There was insufficient certainty for the creation of a trust. The liquidator was entitled to claim the funds in the joint account on behalf of MG’s creditors.

If no separate account for extended warranty premiums is set up, it may be more difficult to identify the sums in the hands of the warranty provider. Although it is not impossible for equity to trace the payments made by a consumer into a mixed bank account⁽²⁰⁾, this will not be of assistance unless the account has continuously been in credit (Bradgate and White, forthcoming). As an insolvent provider is unlikely to have an account which is in credit, and an attempt by a consumer to trace his premium will fail. The creation of a separate bank account for premium payments would therefore be a necessary pre-requisite for a Quistclose trust to be established. This is unlikely to occur in practice because of the provider’s need to use the payment for his business activities.

Even if these hurdles can be overcome successfully, there may a further problem. As in Quistclose itself, the provider’s bank needs to have notice of the trust. It may not be necessary to notify the bank explicitly, and setting up a separate account with a name such as ‘active extended warranty policy premium account’ could be sufficient to give notice to the provider’s bank (cf. Re Kayford)⁽²¹⁾.

From the point of view of a consumer, the separation of premium payments and the imposition of a trust would have the advantage of preventing the use of these payments for general business purposes until the relevant extended warranty policies have expired. This would ensure that all consumers could obtain a refund of their premiums if the retailer or manufacturer became insolvent. However, there would be a number of drawbacks for the provider. The premiums would be ring-fenced and could not be used for normal business purposes. For most providers (particularly retailers), the income from the sale of extended warranties is crucial and it could lead to financial difficulties for some of them. Moreover, it may be difficult to prove that a trust has been created.

It is unlikely that a Quistclose trust would be of assistance in the context of extended warranty premium payments. However, there is an alternative. It has already been explained that extended warranties function as a form of insurance against product defects. If they fall within the legal definition of insurance, protection may be offered to consumers in two ways. Insurance law requires that insurers maintain a solvency margin to ensure that all claims can be met. Moreover, insurance companies are regulated by the Financial Services Authority (FSA). The FSA operates the Financial Services Compensation Scheme which compensates consumers if an insurance company becomes insolvent and therefore cannot honour any claims (cf. OFT, 2002, para. 4.59). Consumers who have obtained their warranties from providers who are authorised insurers themselves or whose policies are underwritten by an authorised insurer would therefore be protected.

The question therefore arises whether extended warranties should be regarded as contracts of insurers and providers of them as insurers who should comply with the regulatory regime. If this is the case, future problems could be avoided through improved enforcement of the regulatory system. However, if this is not the case, then it may be necessary to introduce specific rules to protect purchasers of extended warranties. Interestingly, this is a problem which has arisen in other jurisdictions (both in Australia and the United States). It has been a question of particular concern in Ohio where a number of cases have tried to discern a boundary between warranties and contracts of insurance. Interestingly, such cases arose when consumers were left out of pocket, following the insolvency of a warranty provider (Samini, 1993).

The first point is whether extended warranties should be treated as contracts of insurance. It is, of course, well-known that there is no formal definition of ‘contract of insurance’, but it is possible to identify some key aspects of such contracts (Birds and Hird, 2001, pp 11-17). These may be summarised as follows. First there has to be a binding contract which provides that one party will do something for the other party if a particular event occurs. This event may or may not happen – whether it will has to be uncertain⁽²²⁾. Furthermore, the person in whose favour the contract will operate (i.e., the insured) must have an insurable interest in the item(s) insured (see Prudential Insurance Company v Inland Revenue Commissioners [1904] 2 KB 658). It is not necessary that the insurer pays a sum of money to the insured, should the event happen, provided that the insurer provides a service (such as repair) which is clearly worth money (Re Sentinel Securities [1996] 1 W.L.R. 316). Finally, it seems that the event insured against cannot be within the control of the party assuming the risk of the uncertain event occurring (Birds and Hird, 2001). Although there is no case-law authority for the requirement of control, it seems that this is a factor which the FSA will take into consideration (Purves, 2001, p.633).

Does the standard extended warranty satisfy these requirements? There will be a contract between consumer and warranty provider. The uncertain event is the failure of the product which the consumer has bought. However, the crucial question is whether the extended warranty provider can be deemed to be ‘in control’ of the event insured against, i.e., the breakdown of the product.

This point needs to be examined more closely. A two-fold distinction can be drawn. First, there are the different parties who might provide an extended warranty, i.e., the manufacturer, the retailer or a third-party provider. Secondly, there are the different events which the extended warranty might cover: defective workmanship and materials, accidental damage and breakdown due to wear and tear. It seems that only those extended warranties which cover an event which cannot be said to be within the control of the warranty provider would be considered contracts of insurance. Out of the various permutations that can be derived from the distinction set out in this paragraph, only one combination could be construed as involving an event which is within the control of the insurer: an extended warranty offered by a manufacturer which only covers defects in workmanship and materials. It is arguable that in this case, the manufacturer has ‘constructive control’ (Twigg-Flesner, 1999) over the defect, because the manufacturer makes the product and could prevent the occurrence of such defects. In all the other situations, the person providing the insurance has no control over the event insured against. Thus, a retailer who provides an extended warranty against defective workmanship would be providing insurance, as would a manufacturer who offers a warranty against accidental damage.

What about the types of ‘extended warranty’ which are more like service contracts? A broad distinction may be drawn between two types of service contract (Twigg-Flesner and Bohling, 1999, p. 195): those which involve a regular (usually annual) inspection of the relevant product and those which do not. It seems arguable that the regularity aspect of the first type of service contract would remove this from the scope of insurance, whereas the second type is very much like a contract of insurance. The fact that a regular visit to service the product in question is carried out regardless of whether there is a fault introduces an element of certainty into the contract which is unlikely to be compatible with the suggestion that such a contract is akin to a contract of insurance⁽²³⁾. This view is in line with the decision in Department of Trade and Industry v St Christopher Motorists Association [1974] 1 All ER 395. In this case, the Association undertook to provide its members with a chauffeur service if they had been disqualified for drink driving. This constituted insurance. It appears to have been significant that the cost of providing a chauffeur service was an additional cost for the association, rather than a part of its general running costs. If the provision of this service had added nothing to the expenses of the association, it would probably not have been regarded as providing insurance (per Megarry VC in Medical Defence Union Ltd v Department of Trade [1979] 2 All ER 421 at 430). In the context of service contracts, it is arguable that those contracts where no regular visit is included would be insurance, but those where there is such a visit would not. There could therefore be a thin dividing line between those service contracts which fall within the understanding of ‘insurance’ and those which do not. Service contracts which do not include a regular service of the product and the type of extended warranty discussed above should be treated in the same way, and therefore as contracts of insurance. This would be in line with the court’s view in Re Sentinel Securities, which states that the crucial factor whether a contract is a contract of insurance is not what it is called but whether the substance of the obligation undertaken under the contract is to assume the risk of an uncertain event. There is also no significant qualitative difference between service contracts with and without regular service of the product, and therefore, they should also be treated in the same way. It may be concluded from this discussion that extended warranties and service contracts, no matter what they are called, should fall within the meaning of ‘contracts of insurance’.

This leads on to the second aspect of this analysis, i.e., whether extended warranty providers should be subject to the regulatory regime. Generally, those involved in carrying of the business of insurance have to comply with specific regulatory requirements: first, they require authorisation under the Financial Services and Markets Act 2000 to carry on the regulated activity. The Financial Service and Markets Act (Regulated Activities) Order 2001 confirms that insurance is a regulated activity. Secondly, the Act requires insurers to maintain a minimum amount by which their assets exceed their liabilities. This should protect insureds who have taken out an insurance policy, and therefore also protect buyers of extended warranties.

One possible way of analysing the provision of extended warranties would be to say that all providers, including retailers and manufacturers, are carrying on the business of providing insurance, and would therefore have to comply with the regulatory requirements. It is unlikely that such an analysis would be correct: retailers and manufacturers are primarily interested in selling their goods, and the provision of extended warranties is an ancillary part of their business. If extended warranties are contracts as insurance, and retailers and manufacturers offered these without the backing of an insurance company, then they would be offering a ‘mixed grill’ of contracts (per Megarry VC in Medical Defence Union Ltd v Department of Trade [1979] 2 All ER 421 at 431). In such circumstances, it seems likely that the provider would not be regarded as carrying on an insurance business. In the Medical Defence case, Megarry VC did not have to decide this point, but his obiter comments (at 431-432) suggest that the regulatory regime should only apply to companies who predominantly enter into contracts of insurance. If the opposite conclusion were reached, then it would be impossible for the provider to carry on any other business. This would lead to the paradoxical result that retailers or manufacturers could not carry on their main activity of making or selling products⁽²⁴⁾.

The concern for consumers is that there have been spectacular failures of extended warranty providers which have left many consumers out of pocket and without any protection against product failure. With the exception of some particular cases, extended warranties fall within the legal understanding of insurance, but it seems unlikely that manufacturers and retailers who provide such warranties would be deemed to be carrying on an insurance business. As a result, they do not have to comply with the strict solvency requirements imposed by the regulatory regime. However, in order to prevent the consequences of future failures of non-insurance providers, it may be desirable to consider two solutions: first, it may be possible to require all providers of extended warranties to obtain insurance cover to ensure that consumers will be able to recover their premiums if the retailer or manufacturer goes out of business. In effect, all retailers and manufacturers wishing to provide extended warranties would have to enter into an arrangement with an authorised insurer to underwrite their extended warranty policies. Retailers and manufacturers will, of course, have to make sure that the insurer has the appropriate authorisation. In Re Cavalier Insurance ([1989] 2 Lloyd’s Rep. 430), an insurance company which provided the underwriting of extended warranties offered by a specialist business (Multi Guarantee Co Ltd) did not have the appropriate authorisation, and all underwriting contracts were unenforceable⁽²⁵⁾.

In practice, not all providers are authorised insurance companies or use an insurance company to underwrite their policies. It seems that a change in Insurance Premium Tax led to a number of retailers to discontinue their links with insurance companies. Such providers do not have to comply with the solvency requirement, nor would their customers be covered by the compensation scheme. Consequently, some consumers would not be protected in case the provider becomes insolvent. This and the difficulties of establishing a Quistclose trust may leave some consumers in a precarious position.

An alternative solution may be to introduce specific provisions into insolvency legislation to give extended warranty purchasers a preferential status, should a provider of non-insurance backed warranties become insolvent. This would be so regardless of whether the Quistclose requirements have been satisfied. However, it unlikely that such changes would be made if there are alternative means of dealing with this problem. As suggested above, there is no real difference between extended warranties and other contracts of insurance. One possible solution for the avoidance of doubt would be to introduce a general requirement that all extended warranties be contracts of insurance which must be underwritten by an authorised insurance company. It is therefore submitted that a rule should be adopted which brings all types of extended warranty clearly within the scope of insurance regulation.

The DGFT has asked the Competition Commission to investigate the provision of extended warranties under the Fair Trading Act⁽²⁶⁾. There are two types of monopoly that can be investigated by the CC under the FTA – scale monopolies and complex monopolies (see Whish, 2001). A scale monopoly exists where 25% of goods/services of a particular description are supplied by or to one and the same person. A complex monopoly exists where 25% of the goods or services are supplied by or to a group of people who, whether voluntarily or not, and whether by agreement or not, conduct their affairs in any way to prevent, restrict or distort competition in connection with those goods or services. This looks at the behaviour of the companies concerned.

The CC has been asked to carry out what is known as a ‘full public interest reference’ to establish whether a complex monopoly exists and if so, whether there are adverse implications on the public interest. The public interest criteria are set out in s.84 of the FTA and include all matters which appear to the CC to be relevant in the particular circumstances, and in particular: maintaining and promoting effective competition, the balanced distribution of employment and industry in the UK, and competitive activity in markets outside the UK by UK companies, as well as promoting the interests of consumers, purchasers and other users of goods and services, and the reduction of costs and development of new techniques. If the CC concludes that a monopoly situation exists and that it operates against the public interest, the Secretary of State (SoS) for Trade and Industry has the power to make orders to remedy the problem. The SoS is under no obligation to act.

The present investigation into extended warranties may identify a number of aspects which operate against the public interest, but it may equally be the case that the CC concludes that there are no adverse implications for the public interest. If the CC decides that there are adverse consequences, then a number of options may be available to the SoS for Trade and Industry. If the CC finds that there is a problem in respect of competition and that this is the result for the high cost of extended warranties, the SoS may decide that legislation to promote competition might be the best way forward. This would need to ensure that consumers are aware not only of the existence of other warranty providers, but also of the price of such other warranties. In fact, this was suggested by the OFT in its 1994 enquiry into the provision of extended warranties. It considered utilising s.26 of the Consumer Protection Act, 1987 to make an order on pricing of goods or services.

This section⁽²⁷⁾ enables the Secretary of State for Trade and Industry to make an order (by Regulations) for the purpose of regulating the circumstances and manner in which any person gives an indication as to the price at which goods or services will be available, or indicates any other matter in respect of which such an indication may be misleading. Under s.26(3), such Regulations may, in particular, require an indication as to a price or other matter is accompanied by additional information and that such information is accurate. In the context of extended warranties, the Secretary of State could adopt Regulations in respect of the way prices for extended warranties are given to consumers. If there is sufficient evidence that consumers do not receive an adequate amount of information, Regulations may specify that certain details must be given every time a person is offering an extended warranty to a consumer. This could be a statement advising the consumer to consider whether an extended warranty is needed, bearing in mind the protection under the SoGA and under free guarantees. It may also include the price and availability of a manufacturer’s extended warranty, if this is offered. Other relevant details may be the scope of the warranty, emphasising any important differences between retailer’s and manufacturer’s extended warranties, as well as the price and availability of third-party extended warranties. Other matters could also be deemed relevant. Thus, where reliability data is available, it may be possible to require that consumers should be given this information before deciding whether to buy an extended warranty.

A s.26 order may not be appropriate to deal with this problem, and may not be enough to resolve the competition concerns that exist. The SoS has the power to introduce Regulations that would address the public interest concerns raised by the CC’s report, and he does not have to make use of existing statutory provisions. It may, for example, be necessary to deal with the pressure selling techniques adopted by some retailers. In this context, it may be noted that the improved enforcement mechanism in Part 8 of the Enterprise Bill, which is expected to become law in late 2002, only permits action by the relevant enforcement authorities if a trader has failed to comply with relevant civil and criminal law obligations. This does not extend to behaviour which, although unfair, is not illegal. Although pressure selling techniques may be regarded as unfair, they are not in themselves contrary to existing consumer protection rules. Legislation in response to the CC’s report could therefore seek to deal with this. There would be a number of possibilities. For example, the SoS could decide to introduce a cooling-off period for all extended warranties which would allow consumers to cancel their extended warranties within a specific period. This could be set at seven or 14 days (which seem to be common periods) after purchase, although it may also be possible to introduce a longer-term cooling off period. Such a longer period would reflect the fact that most extended warranties do not take effect until the manufacturer’s guarantee period has expired. More drastically, the SoS could decide to ban the sale of extended warranties at the time of purchase of the relevant product. This would not only eradicate the pressure selling problem, but also allow consumers to compare prices and obtain a better deal. However, it seems probable that such a response would be treated as disproportionate. Whatever the outcome of the CC’s investigation (other than a finding that there are no adverse implications for the public interest), the SoS will have the opportunity to address the provision of extended warranties and to introduce rules that could ensure that extended warranties provide an realistic additional level of protection.

This article has sought to explore the legal issues surrounding the provision of extended warranties in the light of the OFT’s report and the ongoing Competition Commission investigation. Although the OFT’s report is helpful in putting extended warranties back on the agenda, it is also disappointing for its lack of rigorous analysis and a failure to examine the broader context in which extended warranties are provided. The key problems are understanding the relationship between statutory rights and extended warranties; the price of extended warranties and the financial protection of consumers if providers become insolvent; certain terms and conditions, and sales techniques.

There are some consumers for whom an extended warranty is clearly attractive because it may reduce the risks associated with the financial consequences of a product breakdown. Extended warranties offer protection that goes beyond the level of protection provided under the Sale of Goods Act 1979 and any voluntarily given manufacturers’ guarantees. However, there are concerns over the pricing of such warranties and whether consumers are given adequate choice. It seems that consumers could be better informed with regard to the durability of particular products, as well as the likely cost of repairs if a product breaks down. This would enable them to make a better informed decision about the need for an extended warranty.

The Competition Commission will need to investigate if there is a real lack of competition and if competition would result in a significant lowering of the price of such warranties. It is easy to suggest that a lack of competition is to blame for high prices, but without a proper understanding of how extended warranties should be priced, it is difficult to say what a more realistic pricing level of warranties would be, nor can it be concluded with confidence that increased competition would inevitably bring about a lower price. The OFT’s report creates the impression that improved competition will solve many of the problems with extended warranties, but this may be doubted. An issue which has not received sufficient consideration is the fact that extended warranties are purchased disproportionately by lower income consumers. Surely a lack of competition cannot explain this pattern. Moreover, there are consumer protection concerns, notably with respect to the selling techniques adopted by some providers and the financial consequences for consumers if a provider becomes insolvent. These will need to be addressed and the CC’s report will hopefully provide the much-needed incentive to address this issue. The CC has to report by July 2003 and its findings are awaited with interest.

Agrawal, J , Richardson, P S and Grimm, P E (1996). ‘The Relationship Between Warranty And Product Reliability’, 30, Journal of Consumer Affairs, 421-443.

Akerlof, G (1970). ‘The market for ‘Lemons’: Quality Uncertainty and the Market Mechanism’, 84, Journal of Law and Economics, 488-500.

Birds, J and Hird, N J (2001). Birds’ Modern Insurance Law, 5^th edition (London: Sweet & Maxwell).

Bradgate, R (2000). Commercial Law, 3rd edition (London: Butterworths).

Bradgate, R and Twigg-Flesner, C (2002). ‘Expanding the Boundaries of Liability for Quality Defects’, 25, Journal of Consumer Policy (forthcoming).

Bradgate, R and White, F (forthcoming). ‘Insolvent Sellers and Pre-Paying Buyers’

Campbell, C (1987). ‘Trusts outside liquidation’, 8, The Company Lawyer, 280-281.

Chen Z and Ross, T W (1994). ‘Why are extended warranties so expensive?’, 45, Economics Letters, 253-257.

Consumers’ Association (1994). Extended Warranties: A consumer appraisal (London: CA)

Eddy, J A (1977). ‘Effects Of The Magnuson-Moss Act Upon Consumer Product Warranties’, 55, North Carolina Law Review, 835-877.

Emons, W (1989). ‘On the Limitation of Warranty Duration’, 37, Journal of Industrial Economics 287-301.

Holdych, T J and Mann, B D (1996). ‘The Basis Of The Bargain Requirement: A Market and Economic Based Analysis Of Express Warranties - Getting What You Pay For and Paying For What You Get’, 45, DePaul Law Review, 781-857.

Jarmin, A (1994). ‘Consumer Product Testing’ in John, R (ed.) The Consumer Revolution - Redressing the balance (London: Hodder & Stoughton).

Kelley, C A and Conant, J S (1991). ‘Extended Warranties: Consumer and Manufacturer Perceptions’, 25, Journal of Consumer Affairs, 68-83.

Kelley, C A, Conant, J S and Brown, J J (1988). ‘Extended Warranties: Retail Management and Public Policy Implications’ in G.Frazier et.al., American Marketing Association Educator’s Proceedings (American Marketing Association: Chicago, 1988).

Lutz, N A (1989). ‘Warranties as Signals under Consumer Moral Hazard’, 20, RAND Journal of Economics, 239-255.

Office of Fair Trading (OFT, 1994). Extended Warranties on Electrical Goods (London: OFT).

Office of Fair Trading (OFT, 1996). Price Transparency in the Sale of Extended Warranties by Electrical Retailers (London: OFT).

Office of Fair Trading (OFT, 2002). OFT 387 - Extended Warranties on Domestic Electrical Goods – A report on an OFT investigation (London: OFT).

Padmanabhan, V (1995). ‘Usage Heterogeneity and Extended Warranties’, 4, Journal of Economics and Marketing Strategy, 33-53.

Priest, G L (1981). ‘A theory of the consumer product warranty’, 90, Yale Law Journal, 1297-1352.

Purves, R L (2001). ‘The Expanding (or Shrinking) Scope of ‘Insurance’ in FAS Regulation’ [2001] Journal of Business Law 623-645.

Samini, K (1993). ‘Third Party Extended Warranties and Service Contracts: Drawing the Line Between Insurance and Warranty Agreements’ (1993) Ohio State Law Journal 537 – 556.

Schwartz, A and Wilde, LL (1983). ‘Imperfect Information in Markets For Contract Terms: The Examples of Warranties and Security Interests’, 69, Virginia Law Review, 1387-1485.

Shavell, S (1987). ‘The Allocation of Risk and the Theory of Insurance’ in Economic Analysis of Accident Law (Cambridge, Harvard University Press), reprinted in Katz, Foundations of the Economic Approach to Law (Oxford, Oxford University Press, 1998).

Taylor Nelson Sofres (2001). Report into Extended Warranties (Gerrards Cross: Taylor Nelson Sofres).

Twigg-Flesner, C (2000). ‘The Unfair Terms In Consumer Contracts Regulations 1999 - New Powers For The Regulators and the Consumers’ Association’, 11, Utilities Law Review 41.

Twigg-Flesner, C (1999). ‘Consumer Product Guarantees and Insurance Law’, Journal of Business Law, 279-285.

Twigg-Flesner, C and Bohling, H (1999). ‘Insurance Regulation: When A Warranty Is Not A Guarantee’, 9, Consumer Policy Review, 193.

Twigg-Flesner, C and Bradgate, R (2000). ‘The E.C. Directive On Certain Aspects of the Sale of Consumer Goods and Associated Guarantees - All Talk and No Do?’, Web Journal of Current Legal Issues No.2 <http://webjcli.ncl.ac.uk/2000/issue2/flesner2.html>.

Voss, G B and Ahmed I (1992). ‘Extended Warranties: A behavioral perspective’, 19, Advances in Consumer Research, 879-886.

Wein, T (2001). ‘Consumer Information Problems – Causes and Consequences’ in Grundmann, S, Kerber, W and Weatherill, S (eds.) Party Autonomy and the Role of Information in the Internal Market (Berlin: de Gruyter).

Whish, R (2001). Competition Law, 4^th edition. (London: Butterworths).

Willett, C (1992). ‘Consumer Guarantees, Citizen’s Charter’s and Access to Justice’, 11, Civil Justice Quarterly, 99-105.

^(*) I am grateful to Robert Bradgate and Richard Bragg for their helpful comments and suggestions.

⁽¹⁾ Once the Enterprise Bill becomes law, the provisions in the Fair Trading Act 1973 will be repealed and an improved framework for market investigations will be put in place. However, the current investigation will be carried out within the framework of the FTA.

⁽²⁾ The shorthand “statutory rights” is used in this article to refer to the terms implied by s.13, s.14(2) and s.14(3) of the Sale of Goods Act 1979 and the remedies for breach of those terms made available under the Act.

⁽³⁾ Also, credit card providers are jointly liable for contractual claims, including those arising from a breach of the terms implied by the Sale of Goods Act 1979, if the goods in question are worth more than £100 (s.75 Consumer Credit Act 1974).

⁽⁴⁾ OFT Press Release PN40/01, 3^rd October 2001.

⁽⁵⁾ Although some extended warranties require a consumer to arrange for repairs themselves and then claim back the cost subsequently.

⁽⁶⁾ In Lexmead (Basingstoke) Ltd v Lewis [1982] AC 225, Lord Diplock (at p.276) suggested that the implied term as to merchantable quality (now satisfactory quality) was a “continuing term”. The better view seems to be the one expressed in the main text, i.e., that compliance with the term is assessed at the time risk passes. Defects which manifest themselves at a later date are covered if these were latent at the time risk passed.

⁽⁷⁾ Although “reliability” is not a factor expressly mentioned in s.14(2B), it could nevertheless be a “relevant circumstance” in an “appropriate case”.

⁽⁸⁾ This comparison may also be made in relation to other factors. A particular brand may be considerably more durable and its products may generally last for longer than products made by competing brands.

⁽⁹⁾ Directive 99/44/EC on Certain Aspects of the Sale of Consumer Goods and Associated Guarantees (1999) O.J. L177/12; 7 July 1999.

⁽¹⁰⁾ A clear case can be made for extending liability to the manufacturer: see Bradgate and Twigg-Flesner (2002).

⁽¹¹⁾ Whether it is strictly a contract of insurance is considered below.

⁽¹²⁾ Although this still leaves unresolved the question why most guarantees are limited to one year. It seems that there is at least some degree of arbitrariness involved in this.

⁽¹³⁾ It must be noted that such an agreement would be horizontal in nature, rather than vertical, because manufacturer and retailer would be operating at the same level of the supply chain and would therefore be actual or potential competitors in the provision of extended warranties. The distribution agreement between retailer and manufacturer will be vertical and therefore be excluded from the Chapter I prohibition by virtue of the Competition Act 1998 (Land and Vertical Agreements Exclusion) Order 2000, S.I. 2000/310. This Order is currently under review.

⁽¹⁴⁾ The Consumers’ Association Report on Extended Warranties (Consumers’ Association, 1994) includes extracts from various annual reports which suggest that the revenue from the sale of extended warranties makes up a major part of overall profits.

⁽¹⁵⁾ The words “price” and “premium” are used interchangeably to refer to the sum paid by the consumer in return for the extended warranty.

⁽¹⁶⁾ “Secured creditors” are those creditors who have obtained a charge over some or all of the assets of the provider. This entitles them to realise these assets and keep as much of the proceeds as is owed by the provider.

⁽¹⁷⁾ The author is grateful to Robert Bradgate for drawing this point to his attention. The following paragraphs draw heavily on Bradgate and White (forthcoming).

⁽¹⁸⁾ In Re Cavalier Insurance, mentioned below, it was held that the insurance company which acted as underwriter for MG’s extended warranties did not have the requisite authorisation.

⁽¹⁹⁾ MG may have needed some of the funds to pay for repairs and the retailer may have needed the funds to obtain alternative cover for their customers.

⁽²⁰⁾ It is likely that the premiums will be paid into the provider’s business bank account. In that case, the premium payment changes in character and become a debt due to the provider from the provider’s bank. A consumer would therefore have to trace his premium into this new asset.

⁽²¹⁾ Even a successful claim under the Quistclose line of cases may fail if the premium account is only set up as the provider is approaching insolvency. The protection given to consumers by the trust device might be regarded as a preference which could be set aside under s.239 of the Insolvency Act 1986.

⁽²²⁾ In the context of life assurance, the uncertainty is obviously not whether the person insured will die, but rather when that person will die.

⁽²³⁾ In the U.S., it seems that this distinction has become fundamental and is known as the “service-indemnity distinction”: see Samini (1993).

⁽²⁴⁾ The Court of Appeals of New York reached a different conclusion in Ollendorf Watch Co. v Pink (Superintendent of Insurance) (1938) 17 N.E.2d 676. In this case, a manufacturer had given a “warranty” which offered a replacement if the original watch was stolen. This was held to be a violation of the relevant statute prohibiting a “transaction of insurance” without authorisation. The UK’s authorisation system seems to have set a higher standard than that which applied in Ollendorf.

⁽²⁵⁾ Thus proving correct the concerns felt by the electrical retailers in the earlier case involving Multi Guarantee (Re Multiguarantee Co Ltd [1987] B.C.L.C. 257.).

⁽²⁶⁾ It should be noted that the replacement system under the Enterprise Bill will have significant differences. This will not affect the present enquiry, which is carried out under the existing scheme.

^{(27) Section 4 of the Prices Act 1974 also empowers the Secretary of State to make an order on the way prices are indicated to consumers. Under this section, orders can be made regarding the indication of the price itself as well as the manner in which prices are given. However, this is limited to indicating prices and the most recent measure under this section is the Price Marking Order 1999, requiring an indication of prices per unit. The primary objective of this measure would be to facilitate price comparison.}