Consumer Perceptions of Product Warranties: an Exploratory Study

Donald R. Lehmann, Columbia University
Lyman E. Ostlund, Columbia University
[ to cite ]:
Donald R. Lehmann and Lyman E. Ostlund (1974) ,"Consumer Perceptions of Product Warranties: an Exploratory Study", in NA - Advances in Consumer Research Volume 01, eds. Scott Ward and Peter Wright, Ann Abor, MI : Association for Consumer Research, Pages: 51-65.

Advances in Consumer Research Volume 1, 1974    Pages 51-65

CONSUMER PERCEPTIONS OF PRODUCT WARRANTIES: AN EXPLORATORY STUDY

Donald R. Lehmann, Columbia University

Lyman E. Ostlund, Columbia University

[Support for this research from the Columbia Faculty Research Program is gratefully acknowledged.] [Donald R. Lehmann is Associate Professor, Graduate School of Business, Columbia University.] [Lyman E. Ostlund is Assistant Professor, Graduate School of Business, Columbia University.]

Consumer product warranties have been criticized by consumer activists as confusing, inadequate and often deceptive. In fact, Walter Sandbach, Executive Director of Consumers' Union stated recently that warranty complaints are among the most numerous and bitter of those received by C.U., amounting to 5,000 letters annually (Fisk, 1970). It must be remembered, however, that the product warranty serves two distinct purposes: to limit the manufacturer's liability that would exist if no written warranty were attached to the product and to make explicit for the consumer his remaining coverage, and how to obtain it. For certain products, warranties are customarily ignored by consumers, hence, the warranty serves only the manufacturer's purpose. For certain other products, particularly those purchased infrequently, those perceived by consumers as complex or those from relatively unknown (low market share) manufacturers, the warranty may be deemed far more significant in the eyes of the consumer, as Udell and Anderson (1968) concluded. Thus, many consumer warranty complaints may originate in those product categories, or for those brands, where the manufacturer uses the warranty only to limit his liability, not to attract customers. That is, he does not intend the warranty to be of any particular value, even though a portion of his customers will behave as if it does.

With the increased complexity of products, both consumers and manufacturers face great uncertainty as to what one can expect from a product under extended use. Although apparently without documentation, Fisk (1970) claims that "... perhaps as many as 80 percent of all consumer claims for warranty service may not be necessary. It is not that consumers are stupid, but they seldom read the operating instructions that accompany their purchases or obey these instructions faithfully even when they know them." Fisk suggests the pretesting of warranties to assure that consumers understand both the warranty's coverage and the buyer's obligations under it.

Few manufacturers bother with such warranty testing for a variety of reasons. For industry as a whole it is perhaps unfortunate that more attention and resources are not devoted to providing straightforward, valuable warranties in those product categories where consumers show significant interest in warranty coverage. Obviously legislation may result from this neglect if it persists. Already there are signs of deepening distrust of households, Barksdale and Darden (1972) exposed several rather strong, negative consumer attitudes toward U.S. manufacturers. From their questionnaire, five of the forty statements that appear relevant to warranties are reproduced in Table 1 with the responses that were obtained. In answer to statements 1 and 2 over one-half of their respondents appear to doubt the trustworthiness of manufacturers. Answers to statements 3 and 5 suggest that most consumer find it difficult to obtain prompt help with product problems due to the procedures of manufacturers. Only in statement 4 did the respondents seem to acknowledge that at least manufacturers are trying to handle complaints fairly, although they evidently are not succeeding too well. In total, consumers seem to have become skeptical of manufacturer motives and critical of their awkward procedures, but are perhaps still hopeful of improvements to come.

Despite the widespread concern over product warranties, little research has been conducted on how consumers behave toward warranties. Specifically, for what product categories are warranties considered important? How important are they relative to price, brand name, style or dealer reputation? Do consumers read warranties before they buy? Do they expect to understand the typical warranty? If a given warranted product failure arises, how much effort will they make to obtain a remedy? This paper reports the results of an exploratory study of these questions.

METHODOLOGY

A sample of 78 Columbia MBA students completed a questionnaire which dealt with attitudes toward warranties in general plus three specific warranties. In light of the findings by Udell and Anderson (1968) three warranties were selected which were intended to represent low, medium and high complexity and uncertainty. The low complexity warranty concerned a clock radio by a nationally known manufacturer. This warranty was relatively direct in the statement of its coverage and straightforward in its language (Appendix A). The second warranty for a refrigerator was a bit qualified in its coverage, consequently a bit complex and longer although quite readable (Appendix B). The third warranty for a water heater was very restrictive and specific as to its coverage, complex in its language and would require considerable study in order to comprehend its worth (Appendix C). Each subject answered first the general questions on warranties, then specific questions concerning each of the three selected warranties. Each one-third of the subjects received a questionnaire with the three warranties in a different sequence to eliminate any order bias.

Given the nature of the sample, the results would not be considered representative in any-sense of American consumers. Instead the study was intended to illuminate any methodological difficulties in conducting such warranty research and thereby aid the design of a national study. One benefit of the sample is that, because of their training and inclination toward business and in spite of their limited experience due to their age (most were in their 20's), members of the sample should be especially adept at understanding the provisions and the value of warranties. Hence inability on the part of this sample to interpret warranties would suggest that, even with considerable consumer education, the general public will have considerable difficulty in understanding warranties as they are currently written.

FINDINGS

General Questions on Warranties

So as to learn the importance attached to product warranties generally, the subjects were first asked to rank the importance of several rather basic dimensions in buying a television set. Forty-eight percent ranked price as the most important factor, and thirty-seven percent considered brand name most important (Table 2). Style and product warranty were each most important for 8 percent of subjects and dealer reputation was first for the remaining 3 percent. The product warranty was most commonly ranked the third or fourth most important factor. Thus among these subjects the product warranty itself is not considered very important in buying this product of considerable performance risk and complexity.

TABLE 1

ATTITUDES OF AMERICAN CONSUMERS TOWARD MANUFACTURERS AND THEIR PRACTICES ON COMPLAINTS

TABLE 2

RANKING OF FACTORS AS TO IMPORTANCE WHEN BUYING A TV SET

TABLE 3

CORRELATIONS AMONG GENERAL QUESTIONS ON WARRANTIES

Despite these reactions, 48 percent of the subjects claimed they definitely read the warranty in buying most products selling for over $10. Only 14 percent said they definitely do not read such warranties. Interestingly, those who attached relatively little importance to warranties were nearly as likely to read them as those subjects attaching greatest importance.

These rather curious results are not clarified by response to the questions which followed. All but 19 percent considered different warranties within a product category to be essentially alike, yet they still read them. This unexpected pattern of responses seems to suggest that the subjects read warranties only as a matter of habit, hoping against belief to learn something useful about which brand to buy. While these results may be idiosyncratic to the sample, any future study should investigate this area in detail.

Notwithstanding this curious attitudinal set, the subjects are apparently well aware that "good" warranties cost money. Moreover, they would be willing to pay in order to get comprehensive coverage. In an open-ended question, subjects were asked to indicate how much they would be willing to pay for a product warranty of 5 years which provided for free replacement of a $200 appliance if it should fail to operate. The question was deliberately worded simply so as to avoid any confusion over qualifications to the coverage intended. Almost one third of the subjects named the single figure $20, which is 10 percent of the hypothetical purchase price. Amounts under $20 were given by 28 percent of the sample. Twenty-four percent give responses between $21 and $30 with fifteen percent quoting figures even higher.

It should be noted that the above data showed no significant differences between students according to their area concentration (marketing - 20 percent, finance - 52 percent, or other - 28 percent). Nor were there significant differences between foreign versus domestic students or single versus married students.

Taken together the questions on general attitudes do not exhibit any startling relationships (Table 3). Importance of the warranty in buying a television set correlated positively with the likelihood of reading warranties (p < .01 level) but not with any other general question. That is, warranty importance does not relate to confidence in understanding warranties, perceived differences between them, or the amount subjects said they would pay for a comprehensive warranty on a $200 appliance. Confidence in understanding a warranty and likelihood of reading a warranty display a significant correlation (p < .05 level). One could reasonably have expected the question on perceived differences to relate positively with likelihood of reading, but such is not the case. Moreover, those subjects confessing to some lack of confidence in understanding warranties might have been willing to pay more money for the full coverage 5 year warranty on the $200 hypothetical Appliance, but again nothing of that sort emerges. In total, a satisfactory theory of how our subjects perceive and behave toward warranties is not evident. To repeat the earlier conclusion, the subjects apparently read warranties while still not expecting to discover differences that would sway a buying decision.

Specific Questions on the Three Test Warranties

For each of the warranties, the subjects were given a statement describing a particular product failure and asked what action they would take. The product failures were:

1. "The alarm (on the clock radio) stopped working 80 days after purchase."

2. "After 1 1/2 years, the freezer section (of the refrigerator) is no longer operating."

3. "After 2 years, the water heater develops a leak."

The alternative courses of action were:

1. Ignore it.

2. Throw it out.

3. Try to fix it yourself.

4. Have it repaired and pay for it yourself.

5. Complain to the dealer.

6. Complain to the manufacturer.

Given that the question asked the subjects their hypothetical course of action, there is no right or wrong answer per se. However, each warranty did advise the consumer to contact the manufacturer or his service agent in the event of failure. Table 4 presents the data in answer to these question on what action the subjects said they would take. For analysis, the response categories have been collapsed to three: suffer silently (first 4 categories); complain to dealer; and complain to manufacturer. For the clock radio 44 percent said they would complain to the dealer instead of the manufacturer. One cannot judge if those subjects confused the expression, "Service center or Authorized Service Station" with simply dealer or whether their true first action would be to contact the dealer regardless of what the warranty directs. Given the large percentages of subjects that chose the same action for the refrigerator and water heater warranties, it appears that the dealer is considered the natural warranty wailing wall, regardless of warranty terms. It is also of interest to note that 17 percent of subjects are willing to take no action (suffer silently) if the alarm on the clock radio should stop working in 80 days and similarly 19 percent of subjects will do nothing if the water heater develops a leak after 2 years of service. These results contrast sharply with only 5 percent of suffer silently responses for the refrigerator, should the freezer section fail. Perhaps the clock radio is not considered sufficiently costly to justify complaint effort while the refrigerator is, since the warranties for both products are rather easy to act upon. However, for the water heater the warranty is obviously worded to minimize manufacturer warranty losses. To obtain a replacement heater in the event of a leak the consumer must pay shipping charges from Philadelphia or Kentucky on the replacement model, plus shipping charges to return the leaking heater. That is not a simple or inexpensive requirement. Thus 19 percent of the subjects may have decided the remedy isn't worth the effort.

TABLE 4

ACTION THAT WOULD BE TAKEN UPON PRODUCT FAILURE (PERCENTS)

TABLE 5

WHAT ACTION COULD BE REQUIRED OF THE MANUFACTURER (PERCENT)

The next question asked for each warranty had a "correct" answer in each case. The response alternatives were:

1. free replacement or repair to original condition;

2. free parts only;

3. free parts and labor, but not shipping;

4. free parts, labor and shipping;

5. nothing;

6. don't know.

For purposes of analysis, these categories were collapsed to:

1. absolutely free replacement or repair;

2. free replacement or repair except for shipping;

3. no action by manufacturer;

4. don't know.

For all three products and their given product failures, the warranties allow replacement or repair with the buyer paying the shipping charges. Table 5 presents the responses to this question for each warranty. Concerning the clock radio only 33 percent of subjects recognized that the failure was covered except for shipping charges. Sixty-three percent expected no charges and a handful expected no action (4 percent) or were somehow confused over the warranty coverage (1 percent). For the refrigerator 45 percent selected the correct response, namely free replacement or repair except for shipping. Surprisingly, 17 percent were uncertain as to what remedy they could require, and responded, don't know. For the water heater warranty a bare majority, 56 percent, recognized that free replacement without shipping could be required. As with the refrigerator about one-third would expect absolutely free replacement of the leaking heater and 8 percent would expect no action from the manufacturer. Five percent were unsure of what to expect. Contrary to what was expected, however, the proportion of correct answers rose from clock radio through to water heater.

Overall, the responses to the above question suggest that these rather sophisticated subjects are not overly successful at relating the facts of product failures to warranty terms. In a sense the results remind one of the confusion college-educated housewives exhibited in a shopping experiment by Friedman (1966) when instructed to merely buy a list of items which had the highest value (volume or weight per dollar) for the money within their respective product categories. This experiment, conducted before the days of unit pricing in supermarkets, demonstrated that even well educated subjects could not cope with the proliferation of container sizes and prices in attempting to attain the single objective of obtaining the most product for the money. Likewise in this warranty study, most MBA students, given all the time desired, failed to match product failure characteristics to warranty items. One would expect consumers generally to have even more difficulty.

In addition to providing information about their probable and legally possible courses of action, subjects were asked to rate the effort that it would take to obtain the remedy they selected from the above question. The results are given in Table 6. In agreement with experimental design, subjects would expect the required effort to be the least under the clock radio warranty and most under the water heater warranty. The same progression applied in answer to the question, "How understandable is this warranty?" Table 7 indicates that even for this sophisticated group, the clock radio warranty was not very understandable to 10 percent of subjects. The percentage grew to 24 percent for the refrigerator warranty and 33 percent for the water heater warranty. Clearly the idea presented by Fisk (1970) that warranties should be consumer tested is a very reasonable one, given such results.

TABLE 6

EFFORT JUDGED TO BE REQUIRED IN OBTAINING SUITABLE REMEDY TO THE PRODUCT FAILURE

TABLE 7

"HOW UNDERSTANDABLE IS THIS WARRANTY?"

Finally, the subjects were asked to rate the value of each warranty. From Table 8. it is clear that the refrigerator warranty was perceived as most valuable. While its coverage is more complicated, respondents recognized that in dollar terms its worth would obviously be greater than that for the clock radio. Had the water heater warranty not been so one-sided in its terms for obtaining remedy, it may well have been rated higher in value. Despite the much higher price of the water heater and its warranty promising replacement parts for several years, the buyer must truly exert himself to collect on the coverage. Consequently the subjects concluded that it had limited value and rated it lower than the value of the clock radio warranty.

TABLE 8

PERCEIVED VALUE OF EACH WARRANTY

Generally speaking, to the degree a warranty was thought to require much effort in order to obtain a suitable remedy, it too was considered difficult to understand and have relatively little value (Table 9). All such correlations for the clock radio warranty are statistically significant. For the refrigerator warranty, effort required and perceived value of the warranty related negatively but feel short of statistical significance. Thus, the high appraisal of that warranty's value was not diluted by complexity in its language. The remaining two relationships for the refrigerator warranty were significant, however. For the water heater warranty, the degree to which subjects understood the warranty did relate positively to the value they perceived it to have but did not attain statistical significance. This fact may be due to the high proportion (48 percent) of subjects that considered the warranty of little value. Effort required for remedy did relate negatively with how understandable the warranty was and its perceived value.

TABLE 9

CORRELATIONS OF SPECIFIC QUESTIONS CONCERNING EACH WARRANTY

In order to see how the respondents' general attitudes toward warranties related to specific attitudes about the three sample warranties, the correlations between the general question concerning the importance of a warranty in buying a television set and responses to specific questions about the three warranties were calculated. The results, shown in Table 10, show no significant correlations. This surprising finding suggests one of two interesting conclusions:

1. The importance of warranties is not related to the perceived value, understandability, or required effort to obtain a remedy inherent in particular warranties.

TABLE 10

CORRELATIONS BETWEEN THE IMPORTANCE ATTACHED TO WARRANTIES GENERALLY AND PERCEPTIONS OF THE THREE SPECIFIC WARRANTIES

2. The importance of a warranty varies across products for a given person. (In other words, there is no segment for whom warranties are in general important.)

The second explanation seems more plausible in this case, although further research is clearly needed to test both of these explanations.

CONCLUSION

This study has investigated the attitudes of a particular set of respondents to warranties. Three basic findings emerged:

1. The respondents do not consider warranties too valuable either in absolute terms or in terms of comparisons with other determinants of purchase such as price and brand name.

2. A substantial fraction of the sample did not understand what their rights were in the case of three particular warranties, and a substantial fraction of those who recognized them would not take the trouble to obtain their legal remedy.

3. The respondents did perceive differences in the three warranties, especially in terms of understandability.

It must be pointed out that these conclusions are tentative due to the small sample size (n=78), the peculiar characteristics of the sample (all were students studying for their MBA degrees), and the fact that the respondents were estimating what action they would take in a hypothetical situation described by a scenario rather than in a situation which affected them personally. Nonetheless, some implications from the results regarding warranties seem obvious:

1. There is substantial support, at least among a highly educated segment of the market, for the service contract type of warranty.

2. Warranties as they are presently written are confusing even to many MBA's, and hence obviously to a substantial segment of the population at large.

These conclusions suggest that a market opportunity exists for manufacturers willing co provide a useful warranty designed for the layman rather than the lawyer. Unfortunately, some problems exist in providing such warranties. First, cooperation of dealers must be obtained in servicing warranty problems since customers tend to turn first to the dealer in the event of a problem. Given the trend to discount store operations, this is a serious but not insurmountable problem. Second, the quality of construction, usage instructions, and ease of use must be sufficiently high, clear, and simple that the majority of users will not find it necessary to make warranty complaints. In the past many customers have been restrained from exercising warranty options because of difficulty in understanding them and/or anticipated or real difficulty in obtaining the remedy. Since a useful warranty would by definition be easy to use, a side benefit of improved warranties might well be improved product quality and utility. Third, pretesting of warranties is needed to determine which forms are the most understandable, and even then some consumer education will undoubtedly be needed, as experience with the Truth-in-Lending Bill would indicate. Finally, some means must be found of protecting the companies from the small minority of customers whose complaints, which would generally be considered unfounded, could significantly increase companies' cost of offering the warranties and hence the price of consumer in general.

Thus there are some obvious problems involved in introducing the "ideal" warranty. Nonetheless the opportunity may well be worth the trouble involved for the companies and any additional cost to the consumers, a determination which can be made only after research and testing. It does appear clear that, given the disenchantment with present warranties, that unless companies act the government may act in their absence.

APPENDIX A

Assume you just purchased a clock radio whose warranty is reproduced below:

Warranty

Your new radio is warranted against factory defects in material and workmanship for a period of 90 days from the date of purchase or receipt as a gift. We will elect to either repair the radio or replace it, with a new or factory serviced unit at no cost to you for labor, materials or return transportation, if it is delivered, prepaid to any Servicenter or Authorized Service Station in the United States.

This warranty does not cover failures beyond the control of the Company and is the only one applicable to this product.

APPENDIX B

Now assume you just purchased a new refrigerator. Please read the guarantee below and answer the questions which follow.

Guarantee

* One (1) Year Replacement Guarantee on All Components

* Five (5) Year Replacement Guarantee on Sealed System

* One (1) Year Service Guarantee on All Components - No Extra Charge for Labor

* Five (5) Year Service Guarantee on Sealed System - No Extra Charge for Labor

COMPONENTS AND SERVICE LABOR - ALL COMPONENTS

For one (1) year after delivery of your refrigerator-freezer, will replace any defective component and perform any required product repair without charge to you for the replacement component or service labor.

SEALED SYSTEM AND SERVICE LABOR

For five (5) years after delivery of your refrigerator-freezer, will replace any defective component and perform any required repair of the sealed system of your refrigerator-freezer without charge to you for the replacement components or service labor. This five (5) year guarantee covers only the compressor, evaporator, condenser and interconnecting tubing of the sealed refrigeration system.

QUALIFIED SERVICE

If any component in your Product should become defective within the guarantee periods, contact your nearest Qualified Service Center. You can obtain their locations through your dealer or calling (as you normally dial long distance) special number free of charge.

This Guarantee of Product Performance is offered to you as the original United States purchaser of this Product for use in your home. You may exercise this Guarantee anywhere you live or may live in the United States.

THIS GUARANTEE IS SOLE GUARANTEE AND IS IN LIEU OF ALL OTHER GUARANTEES, EXPRESS OR IMPLIED, INCLUDING THE IMPLIED GUARANTEES OF MERCHANTABILITY AND OF FITNESS FOR A PARTICULAR PURPOSE, AND OF ANY OTHER OBLIGATION. IT DOES NOT COVER DAMAGE CAUSED BY MISUSE OR ACCIDENT, FIRE, FLOOD OR OTHER CAUSES NOT WITHIN THE CONTROL OF THE COMPANY.

APPENDIX C

Now assume you just purchased a new hot-water heater for your home. Please read the warranty below and answer the questions which follow.

THE YEAR REPLACEMENT PLAN

1. The undersigned Company will, during the first year the heater is installed, under normal domestic (single family unit) use furnish replacement parts to the original owner, at original installation address, if proven to our satisfaction that these parts are defective in material or workmanship. (See paragraph 4 below.)

2. We further agree, if it be proven to our satisfaction that a leak has developed in the original tank, that we will furnish to such original owner, at the original installation address, during the period of the first to fifth year inclusive from date of original installation, a new heater of the prevailing comparable model, and that if such leak should occur during the period of the sixth to tenth year, inclusive, from the date of original installation, we will furnish a new heater of the prevailing comparable model at a percentage of the Company's suggested retail price at time of replacement on the following schedule:

Within 6th year, 50%

Within 7th year, 60%

Within 8th year, 70%

Within 9th year, 80%

Within 10th year, 90%

A replacement heater provided under the terms of this paragraph will carry only the unexpired portion of the original heater's warranty. (See paragraph 4 below.)

3. If the heater has been used for other than normal domestic (single family unit) use, the company agrees only to furnish to the original installation address, a new heater of the prevailing comparable model, if proven to-our satisfaction that a leak has developed in the original installation. (See paragraph 4 below.)

4. Heater or replacement parts will be shipped f.o.b. plant, Philadelphia, Pennsylvania, or Louisville, Kentucky, and the customer will be liable for shipping, delivery and labor charges for removal and installation. Any heater or parts being replaced will become the property of the Company and must be returned to the Company's nearest plant with transportation paid by the customer.

5. This contract is VOID:

a) If the heater is not equipped with approved temperature and pressure relief valve: or

b) If it is installed in violation of any local plumbing code, statute, ordinance or regulation; or

c) If it has been subjected to misuse, alteration, neglect or external damage, regardless of fault.

6. This contract is in lieu of all warranties expressed or implied and of all other obligations on our part. THERE IS NO WARRANTY OF MERCHANTABILITY, or any other warranty not expressed herein. We neither assume nor authorize any person to assume for us any other obligation or liability in connection with the sale or use of this heater. Rusty or discolored water in itself does not constitute tank failure within the scope of the contract. We assume no liability for consequential damages, resulting from defective product.

THE GUARANTEE CARD ATTACHED BELOW MUST BE RETURNED TO THE MANUFACTURER WITHIN 30 DAYS OF INSTALLATION IN ORDER THAT THIS GUARANTEE MAY BE PUT INTO EFFECT. FAILURE TO DO THIS WILL MAKE THIS GUARANTEE NULL AND VOID.

SELECTED REFERENCES

Barksdale, Hiram C., and Darden, William R., "Consumer Attitudes Toward Marketing and Consumerism," Journal of Marketing, Vol. 36, No. 4 (October, 1972), pp. 28-35.

Fisk, George, "Guidelines for Warranty Service After Sale," Journal of Marketing Vol. 34, No. 1 (January, 1970), pp. 63-67.

Friedman, Monroe Peter, "Consumer Confusion in the Selection of Supermarket Products," Journal of Applied Psychology, Vol. 50, No. 6, 1966, pp. 529-34.

Udell, Jon G., and Anderson, Evan E., "The Product Warranty as an Element of Competitive Strategy," Journal of Marketing, Vol. 32, No. 1 (October, 1968), pp. 1-8.

----------------------------------------