Investigation of the effects of disclosure statements in rental car advertisements.
Richards, Jef I. and Ivan L. Preston (1987), "Quantitative Research: A Dispute Resolution Process for FTC Advertising Regulation," Oklahoma Law Review, 40(Winter): 593-619.
Roth v. United States, 354 U.S. 476 (1957).
Russo, J. Edward (1974), "More Information is Better: A Reevaluation of Jacoby, Speller and Kohn," Journal of Consumer Tremendous controversy has surrounded increased activity by the attorneys general of several states in their attempts to regulate national advertising (Bishop 1988; Mattox and Stratton 1988). Representing their states, they were responding to a perceived lack of federal interest in defending consumer rights (Richards 1991). By taking action as a group, through the National Association of Attorneys General (NAAG), states have begun establishing their own national advertising standards, thereby entering an arena formerly the near exclusive domain of the Federal Trade Commission (FTC). The first national "guidelines," adopted by more than 40 states in 1987, dealt with airline advertising. Encouraged by airline compliance, NAAG members formed a task force to suggest possible guidelines for rental car advertising (Lawrence 1988). This article evaluates the effectiveness of these car rental guidelines.
In 1988, NAAG released a preliminary report from its task force on rental car advertising (Pepper, Hamilton & Scheetz 1988). This report alleged industry-wide abuses regarding advertised rental prices, potentially resulting in consumer deception. It noted that actual rates charged were frequently twice the rate advertised, because the stated charge excluded fuel charges, airport access fees, and other frequently encountered costs. The report also charged those ads often failed to disclose other material restrictions, such as mileage costs and geographic locations where the offer applies.
Later that year draft guidelines were completed, with emphasis on abuses in broadcast commercials. To rectify the alleged nondisclosures, certain mandatory disclosures were recommended. Where the broad range of typical qualifications applied to a rental offer, these guidelines provided that an appropriate disclosure would include substantially the following:
Offer available through June 15 at participating locations, 30 cents mileage fee per mile after first 100 miles, collision damage waiver 12 dollars per day, other important restrictions apply. (National Association of Attorneys General 1988, section 2.2)
The American Advertising Federation (AAF) filed comments with the NAAG, arguing that lengthy disclosures would be counterproductive (American Advertising Federation 1988b). While it did not dispute the benefits of providing consumers with more information, the AAF suggested that ads cannot serve as a replacement for the information available from a reservations agent and that lengthy disclosures discouraged price advertising and rendered broadcast advertising virtually impossible, because of the time required for the disclosure. The AAF proposed that a better approach would be to include a simple statement in ads referring consumers to the sales or reservations agent for restrictions and other information (American Advertising Federation 1988a). In particular, the AAF argued that "|t~oo much information in advertising can confuse consumers and render their retainment of the information less likely" (American Advertising Federation 1988c, 2).
NAAG allegations of deceptiveness, their proposed solution, as well as the AAF alternative solution apparently were based upon experience, fears, and biases of their respective members rather than empirical behavioral tests of the ads and the recommended disclosure. While this is the conventional approach to advertising regulation, empirical research offers the potential of more equitable results (Richards and Preston 1987). Rather than focus on information provision, information disclosure programs should direct attention to the effects such information has on its receivers (Jacoby, Chestnut, and Silberman 1977). While these effects may be political, economic, social, or psychological, this article concentrates on the impact of such disclosures on consumer "belief" structures: the essence of deception (Richards 1990).(1)
Set against this background, this study was designed to compare the validity of the conflicting NAAG and AAF arguments. Three important qualifications, however, should be noted. First, no attempt is made to judge the wisdom of a policy aimed at providing consumers with more information. The goal of this study is merely to test which of these two means of providing such information is the better method of eradicating deceptiveness. Second, the AAF position makes the rather large assumption that reservations agents voluntarily disclose information of the type envisioned by the NAAG recommendation, or that consumers know enough to ask for that information. Because deceptiveness legally is dependent only upon effects caused by the challenged advertisement and not by information from other sources (Jacoby and Hoyer 1987), that assumption is neither challenged nor supported. Third, the focus of the controversy is on broadcast commercials, so the separate issues raised by print ads are not considered. Also, because NAAG has expressed a preference for oral disclosures the issue of visual disclosures on television is not addressed.
To put this study in perspective, a brief review of the role of mandatory disclosures in the regulatory process is presented in the next section. In following sections the study is described, and the findings are presented. Finally, conclusions and implications of those findings are discussed.
ROLE OF AFFIRMATIVE DISCLOSURE OF INFORMATION
In his 1962 address to Congress, President Kennedy hailed the rights of consumers (1) to safety, (2) to be informed, (3) to choose, and (4) to be heard, along with the obligation of government to facilitate these rights. Under the right to be informed, Kennedy stressed that consumers have the right "to be protected against fraudulent, deceitful, or grossly misleading information, advertising, labeling, or other practices, and to be given the facts he needs to make an informed choice" (Kennedy 1962, 4140). In fact, a major rationale for government intervention in a free market/free speech system is to facilitate the flow of accurate information to individuals, thereby enhancing consumer sovereignty (Central Hudson 1980; Emerson 1981; Virginia Pharmacy Board 1976).
To ensure relatively accurate information is received, the U.S. legal system repeatedly acknowledges a preference for more speech, rather than enforced silence, as a means of regulation (Central Hudson 1980; Linmark v. Willingboro 1977; Whitney v. California 1927). This suggests mandatory information disclosure is normally preferable to total prohibition of an advertising claim; the more information provided, the better. Such an ideal, of course, is not always considered workable, and the history of the United States is replete with instances where enforced silence was the regulation of choice (Chaplinsky v. New Hampshire 1942; Posadas de Puerto Rico 1986; Roth v. United States 1957; Whitney v. California 1927). Mandatory disclosure, however, remains a constitutionally favored "less restrictive alternative" to banning speech (Bates v. State Bar of Arizona 1977; Central Hudson 1980; Richards 1987).
While never acknowledging such a preference, the FTC has frequently invoked affirmative disclosure as a remedy for omissions of material information (Wilkie 1982), and the Supreme Court has declared these legally mandated disclosures constitutionally permissible (Zauderer 1985, 651). Whether demanded by law, required by media, or inserted as a precaution, disclaimers and other disclosures have become a familiar adjunct to commercial messages (Elliott 1988; Foxman, Muehling, and Moore 1988). Of course, the justification for such disclosures is that a significant number of consumers benefit by receiving the added information.
When fashioning a remedy for deceptive advertising, one must consider not only potential benefits to consumers but also the costs to be incurred by sellers (Craswell 1985; Mazis et al. 1981). Although deceptive speech can claim no sanctuary under the constitution (Central Hudson 1980), any remedy inevitably infringes somewhat upon nondeceptive speech.
Both buyer and seller have rights under the law. Generally, restrictions on ad claims must not unduly impair nondeceptive speech and therefore must be narrowly tailored to reasonably fit the public's interest in regulating those claims. Consequently, where a significant public interest is directly advanced by the restrictions and the regulation is sufficiently narrow to avoid unreasonable (corporeal or incorporeal) costs to the seller, the regulation is constitutionally permissible (Board of Trustees of SUNY v. Fox 1989). On the other hand, where the consumer accrues little or no benefit, the public's interest is not served and significant cost to the seller cannot be justified (Central Hudson 1980; Craswell 1985; Mazis et al. 1981).
Benefits of Disclosure
Applying this cost-benefit principle to the rental car disclosure issue, the first question that must be addressed is whether consumers obtain a measurable benefit from the NAAG Disclosure. Clearly, the intended benefit is to provide previously omitted information to consumers, thereby dispelling inherent deceptiveness created by these omissions. It is well established in law (Donaldson v. Read Magazine 1948; P. Lorillard Co. v. FTC 1950) and behavioral science (Harris 1977; Preston 1967; Shimp 1978) that omissions can be as misleading as explicit falsities.
On the other hand, the AAF suggests such a disclosure may cause consumer confusion and inhibit recall of the information (1988c). There is some theoretical and empirical support for this position. Information must be not only available to consumers; it must be processable (Bettman and Kakkar 1977). Broadcast advertising, by its nature, presents a continuous stream of rapid-fire information requiring consumers to attempt processing some information while being bombarded with more. While earlier data are being transferred from short-term to long-term memory, later information must be squeezed into short-term memory (Bettman, Payne, and Staelin 1986). Miller (1956) found, however, that short-term memory capacity is severely limited. People are unable to process and retain all information available, so some is necessarily filtered out (Bettman, Payne, and Staelin 1986).
Research also indicates that if the volume of information surpasses a consumer's cognitive capacity, processing accuracy may suffer (Henry 1980) and can affect the quality of consumer decisions (Jacoby and Kohn 1974). Following a series of studies investigating information load effects, Jacoby concluded
Information overload refers to the fact that there are finite limits to the ability of human beings to assimilate and process information during any given unit of time. Once these limits are surpassed, the system is said to be "overloaded" and human performance (including decisionmaking) becomes confused, less accurate, and less effective. (Jacoby 1977, 569)
For example, Malhotra (1982) found that information overload can contribute to dysfunctional performance in a purchase decision situation when either the number of alternatives in a consumer's choice set or the number of product attributes used for choice comparison becomes too great. A more recent study by Keller and Staelin (1987) provided additional evidence that where there is too much available information consumers experience a decrease in decision efficiency.
However, research concerning the application of this overload threshold to information disclosure programs is limited in scope, and many of the studies have been subjected to severe criticism (see Russo 1974; Summers 1974; Wilkie 1974). The controversy over whether or not consumers can experience an information "overload" and what consequences it might cause continues to be debated (Keller and Staelin 1989; Meyer and Johnson 1989). In addition, it is difficult to apply previous findings to the present dispute because most studies have used some form of print media, some have used numerical data as the "information" provided (e.g., involving disclosure of nutritional information) and both dependent and independent variables have varied widely. Further, Jacoby (1977) noted the complexities of the overload phenomenon make it difficult even to define operationally what constitutes an appropriate unit of "information" (in terms of time or content).
Given these limitations in and questions about information load theory, we are left with little guidance as to how much information from a 15-, 30-, or 60-second commercial is comprehended and encoded into long-term memory. If, in the context of a typical rental car commercial, the 29-word NAAG Disclosure exceeds this capacity it may fail to serve the public interest in regulating these ads. In this event a shorter disclosure like that recommended by the AAF might be superior. On the other hand, if the NAAG Disclosure does not unduly tax consumers' information processing abilities it could reasonably be expected to eliminate more deceptiveness than the simple referral suggested by the AAF, thereby better serving the public's interest within the confines of a brief broadcast commercial.(2)
Costs of Disclosure
On the cost side of this equation, some proponents of information disclosure programs suggest that such disclosures require no direct out-of-pocket expense to an advertiser and therefore provide benefits at no cost (Elliott 1988). However, the AAF (1988c) argues current broadcast advertising figures reveal that less than five percent of all television commercials are 60 seconds, almost 25 percent are 15 seconds, and the vast majority (roughly 70 percent) are 30 seconds. With nearly all commercials being 30 seconds or less, a disclosure requiring five to ten seconds eliminates a large portion of the time available to the seller.(3) In practical terms, this means either that less information intended by the seller can be conveyed to consumers or that less time is available to emphasize and reinforce the seller's message. Arguably, the disclosure may be a cost to sellers' rights of free speech and potentially to their profits.(4)
Mazis et al. concluded "In general, disclosures increase the cost of information.... This is especially true for broadcasting since the time for the total message is fixed" (1981, 19). It follows, therefore, that unless some significant benefit to consumers can be substantiated to effectively outweigh these costs, such regulation might not be justified. This, in turn, raises questions about whether a broadcast commercial, constrained by realities of time and cost limitations, can adequately fulfill consumers' right "to be informed" as set forth by President Kennedy.
The present study was designed to evaluate the relative benefits of the proposed NAAG guidelines for rental car ads in terms of reducing deceptiveness of those messages. This research is aimed primarily at helping to answer the public policy questions raised by the present controversy. Because significant effort was made to remain a neutral party in the controversy, the authors felt it more appropriate to state the issues relevant to this study in the form of research questions than hypotheses. Four questions are addressed.
To determine whether the proposed disclosure will inure any benefit to consumers, vis-a-vis deceptiveness, the first question is
Q1. Are rental car price claims deceptive, where typical qualifications are not disclosed?
If the answer is no, the regulation has no merit. To answer this question, however, what constitutes legally regulable deceptiveness must be examined.
This is a complex question, particularly in the present situation, because the parties to the controversy have not agreed upon a single definition of deceptiveness. The definition of deceptiveness to which the FTC adhered for nearly 70 years--permitting regulation of claims possessing a "capacity or tendency" to mislead even the "foolish or feebleminded"--was supplanted with a new policy in 1984. This change has stimulated significant conflict within the FTC, because it is perceived as diminishing the amount of protection afforded consumers (Ford and Calfee 1986; Preston and Richards 1986). It is also clearly one reason for the recent activity of NAAG in the arena of national advertising regulation (Calvani 1989; Mattox and Stratton 1988).
The current FTC definition was chosen for the present study because no single definition encompasses all viewpoints, and the FTC approach is the default national standard.(5) The Commission, in the Cliffdale case, specified this standard:
|W~ill find an act or practice deceptive if, first, there is a representation, omission, or practice that, second, is likely to mislead consumers acting reasonably under the circumstances, and third, the representation, omission, or practice is material. (Cliffdale 1984, 164-165)
In the auto rental controversy we are clearly dealing with an omission. The questions remaining to be answered, therefore, are whether these omissions (1) are likely to mislead consumers acting reasonably and (2) are material. Material omissions are further defined by the FTC as ones involving "information that is important to consumers and, hence, likely to affect their choice of, or conduct regarding, a product" (Cliffdale 1984, 165).
To meet the Commission's standard, therefore, one must also ask
Q2. Are rental car price claims material?
Again, the regulation has no merit if the answer to this question is no. If yes, however, the following must be asked
Q3. Will the proposed NAAG Disclosure remove a significant degree of deceptiveness?
If not, there is no benefit to the regulation as proposed. If yes, the next question is
Q4. Is a proposed alternative shorter disclosure an adequate substitute for the longer disclosure?
A negative answer suggests the NAAG Disclosure is probably a justifiable remedy, but a positive answer indicates that the shorter disclosure might be more appropriate.(6) To investigate these questions, an experimental study was conducted. The design of this study is described.
Data were collected from 188 undergraduate students at a large southwestern university, using a carefully pretested questionnaire. Undergraduate students were deemed appropriate for two reasons. First, use of student samples for studies presented as evidence in deceptiveness cases is well-established (e.g., Sun Oil 1974). Second, both the FTC and the courts have indicated that sample data need not be statistically generalizable to the population-at-large to be probative of the facts (Preston 1987). Though it would be advantageous to replicate this study with a more generalizable sample, the submission of data from a student sample constitutes research acceptable as one piece of evidence in a legal proceeding involving deceptiveness.
To examine the issue of information disclosure in broadcast advertising, a rental car radio commercial was written and produced.(7) This commercial (Basic message) contained only a 24-second selling message and was supplemented by one of three disclosure messages, which are presented in Figure 1. The Basic message with the NAAG Disclosure constituted approximately a 30-second commercial.
The Basic message spot for Avis rental cars was derived from an Avis print ad which appeared in a January 1989 issue of Time magazine. For ecological validity, the commercial was produced at a local radio station using music, sound effects, and professional talent. Also, to add realism and disguise the nature of the investigation, the Avis commercial was sandwiched between a 30-second "distractor" commercial for Bumper-to-Bumper auto parts stores and one for Jello gelatin.
The three different disclosure messages developed and added to the Basic message were
(1) The "NAAG Disclosure" of nine seconds, which used the exact wording suggested in the NAAG guidelines, except that a date closer to the time of the experiment was inserted to enhance realism.
FIGURE 1 Script of Avis Radio Commercials |Basic Message~ Music: Fanfare drum roll with cymbals crash! Announcer: Avis presents the Supreme Weekend Deal. Now you can drive away to your weekend getaway in a sporty Oldsmobile Cutlass Supreme for about the same price as a compact. Music: Fanfare drum roll with cymbals crash! Announcer: 25 dollars a day! And with our fast service and convenient locations, Avis makes it easier than ever to get away in style! To take advantage of this incredible offer and get your Cutlas Supreme for 25 dollars a day, call Avis now! Avis. We're trying harder than ever. |NAAG Disclosure~ Offer available through March 15 at participating locations, 30 cent mileage fee per mile after first 100 miles, collision damage waiver 12 dollars per day, other important restrictions apply. |Reversed Disclosure~ Offer available through from now on at all Avis locations, free unlimited mileage is included, a collision damage waiver is also included at no charge, there are no other restrictions which apply. |Alternate Disclosure~ Contact on Avis reservartions agent for details.
(2) The "Reversed Disclosure" of nine seconds designed to effectively reverse the claims made in the NAAG Disclosure.
(3) The "Alternate Disclosure" of two seconds derived from the AAF recommendation.(8)
Assuming that the NAAG Disclosure is true, the Reversed Disclosure presents totally false claims. As explained below, the Reversed Disclosure was included to provide a means of measuring an effective response range for the advertising messages within the context of an individual study.
Subjects were randomly assigned to one of five treatment groups. Group A was exposed to the Avis Basic message only, without being sandwiched between the Bumper-to-Bumper and Jello commercials. The remaining four groups (labeled B-E) were asked to listen to three radio commercials, played in the following sequence: Bumper-to-Bumper, Avis, and Jello. All subjects were given the following description of their task as participants:
We appreciate your help with this research project. As a participant, you will be listening to three radio commercials, and after you have heard the commercials we would like you to complete a short questionnaire. Please listen carefully to the commercials, and then answer the questions as honestly as possible.
The task description was modified for group A to reflect one rather than three commercials.
After listening to the commercial, subjects in group A were asked questions only about what type of restrictions or qualifications were likely to apply to the offer described in the rental car commercial. They were first asked to indicate on a nine-point scale how likely they felt it was that "there are restrictions or qualifications regarding the terms of the offer described in the commercial." Next, they were asked to respond to an open-ended question: "What type(s) of restrictions or qualifications, if any, regarding the terms of the Avis offer do you believe would be likely to apply?"
Subjects in groups B-E were exposed to different versions of the Avis commercial sandwiched between the Bumper-to-Bumper and Jello commercials. All began with the same 24-second Basic message followed by a different type of disclosure message as identified in Table 1.
TABLE 1 Group, Treatment Presented, Length of Message, and Number in Each Group Group Treatment Time n A Basic Message ONLY, without the Bumper-to-Bumper and Jello commercials 24 sec. 60 B Basic Message ONLY 24 sec. 31 C Basic Message + NAAG Disclosure 33 sec. 32 D Basic Message + Reversed Disclosure 33 sec. 34 E Basic Message + Alternate Disclosure 26 sec. 31
TABULAR DATA OMITTED
Following exposure to the commercials, subjects in groups B-E were asked a series of questions about the relative performance (e.g., memory and attitude) of all three commercials--Bumper-to-Bumper, Avis, and Jello. These initial questions were used primarily to mask the purpose of the study.
Next, subjects were told they would be asked detailed questions about only one of the three commercials, but in fact all subjects were assigned to respond to the rental car commercial. They responded to two different series of questions.
First, to measure subjects' specific beliefs about claims potentially implied in the Avis offer, a series of nine-point semantic differential type scaled questions was used. These involved the five "beliefs" statements listed in the left-hand column of Table 2. Each scale provided a "true" description of the offer at one end and a "false" description at the other.(9) For example:
What do you believe to be true about the offer described in the commercial? There ARE There are NO restrictions which restrictions which apply 1 2 3 4 5 6 7 8 9 apply
Polarity of the scales, i.e., which end had the true description and which had the false description, was randomized to avoid bias. Each of these four groups responded to the same questionnaire.
Second, to measure the materiality of each claim another nine-point scale was used immediately after each "belief" scale. This question took the following form.
Would this feature be a factor in your decision whether or not to take advantage of this offer? Would NOT be a Would be a factor 1 2 3 4 5 6 7 8 9 PRIMARY factor
This approach reflects the longstanding legal standard for materiality used by the FTC (Richards 1990).
The measure of deceptiveness used is reasonably sophisticated and requires some explanation. The FTC, when using empirical research to assess deceptiveness of an ad claim, typically relies upon some measure of consumer comprehension of the claim as an indicator of the claim's "likelihood to mislead" (Preston and Richards 1986). The fact that a claim is understood to make certain promises, however, does not ensure the consumer believes these promises. The difficulty created by using comprehension instead of belief is illustrated by the popular use of spoof claims.(10) Actual deception of the consumer does not occur without such belief. A better measure than message comprehension, therefore, is consumer belief about the product/service attributes after exposure to the commercial. When using an experimental rather than a survey method, it is possible to test these beliefs while fully complying with FTC legal standards for testing deceptiveness (Richards 1990).
Although a few experimental methods have been developed to test deceptiveness (Armstrong and Russ 1975; Gardner 1975; Russo, Metcalf, and Stephens 1981), Richards (1990) identified problems with each and proposed an alternative method: the Relative Proximity Index (RPI). This technique is based on an examination of the responses of two groups of subjects. One group is asked to respond to a scaled question after exposure to an advertising message that includes a true assertion regarding the attribute of interest; a second group responds to the same scaled question after exposure to an ad message that includes a false assertion regarding the same attribute. Hence, the RPI approach empirically frames the range of response within which a specific advertising message is capable of moving consumer perceptions, in the context of the individual study.
The thinking behind the RPI method is realistic in that it recognizes that consumers react to an advertising message with preconceived beliefs. That is, they bring to bear their experience gleaned from evaluations of similar ad messages, experiences in the marketplace, and what they found to be true regarding other similar offers by marketers. As O'Toole (1981) pointed out, consumers in U.S. culture are not naive regarding what sort of information they receive from marketers. A healthy skepticism exists regarding the believability of ad claims. In addition, research has shown that a certain amount of miscomprehension is inherent in the communication process (Jacoby and Hoyer 1987; Jacoby, Hoyer, and Sheluga 1980) and cannot be eradicated (Preston and Richards 1988).
To address these problems, Richards proposed measures be taken on three stimuli: the true claim, the false claim, and the original ad claim. Measures of the true and false claims are then used to compute a midpoint (grand mean) of the two. For example, as shown in Figure 2, where the group mean of a true message is 2.8 and that of a false message is 8.6, the midpoint value is (2.8 + 8.6)/2 = 5.7 |for another example, see note to Table 3~. If the mean of the ad claim differs significantly from the midpoint, in the direction of the false claim, it is concluded to be deceptive. Because of these experimental manipulations, this method automatically controls for subjects' ineradicable miscomprehension and prior beliefs about rental car advertising.
Questions of scale construction and validity of the RPI method were addressed by Richards (1990). However, note that the RPI method avoids the dilemma in which the group mean of those exposed to the original ad claim is significantly different from the "true" group's mean in the direction of deceptiveness, but is also significantly different from the "false" group's mean in the direction of truth (see discussion in Russo, Metcalf, and Stephens 1981). The RPI approach, by relying on an empirically determined midpoint, unambiguously addresses the issue of deceptiveness within the range established by the two extremes.
The findings are organized in three major sections: (1) Perceptions of Likely Restrictions, (2) Effects of Different Disclosures on Beliefs TABULAR DATA OMITTED About the Offer, and (3) Materiality of the Claims. This is followed by a discussion of the conclusions and implications of these findings.
Perceptions of Likely Restrictions
To investigate consumers' beliefs about what types of restrictions most likely apply to the offer made in the Basic message, one treatment group was asked to listen only to the basic Avis commercial (group A). After listening to the commercial, subjects were asked the likelihood restrictions or qualifications apply to the offer. Eighty-eight percent of subjects indicated they felt it was likely. On a nine-point scale, with 1 = "Extremely Likely" and 9 = "Extremely Unlikely," the group mean was 2.68. The frequency distribution is shown in Figure 3.
Next, this group was asked to identify the types of restrictions or qualifications, if any, which likely apply. Responses to this open-ended question were mileage charge (60 percent), time limitation (53 percent), age (52 percent), insurance fee |e.g., collision damage waiver fee~ (28 percent), limited locations (22 percent), major credit card (20 percent), limited number of cars (20 percent), driver's license (18 percent), and miscellaneous others (30 percent).
Finally, subjects in this group were asked "How likely do you believe it would be for this offer to include the following qualifications?" The qualifications listed were the four specifically addressed in the proposed NAAG Disclosure. The mean response for each, on a nine-point scale (1 = "Extremely Likely" and 9 = "Extremely Unlikely"), was "NOT available at all Avis locations" = 4.56; "Available ONLY for a limited time" = 1.98; "Mileage charge for mileage over 100 miles per day" = 2.75; and "Additional charge for a collision damage waiver" = 1.97.
Effects of Different Disclosures on Beliefs About the Offer
This section presents the responses of the four treatment groups (B-E) used to assess deception using the Relative Proximity Index (RPI) discussed earlier. Note that although polarity of the scales was randomized in the questionnaire, the results presented have been rescaled to facilitate comparisons. All data presented in this section were collected using a nine-point scale and the means discussed should be interpreted as falling between the "true" description = 1 (e.g., "There are restrictions which apply") and the "false" description = 9 (e.g., "There are no restrictions which apply").
The mean responses of each of the four treatment groups on each of the five scales used to explore deception are presented in Table 2. Note that the five scales covered perceptions of restrictions, generally, and of the four specific restrictions covered in the NAAG Disclosure. For example, group B was exposed to the basic message only sandwiched between the two distractor commercials. When asked whether they believed the Avis offer was subject to restrictions, their responses averaged 3.32. Only 9.7 percent of subjects were on the "no restrictions" side of the scale.
Next, the findings with respect to legal deceptiveness were evaluated using the RPI. Table 3 presents the findings when the midpoints, calculated between the NAAG Disclosure and Reversed Disclosure groups, were compared with the means of the treatment groups which were exposed to the Basic message only (group B) and the Alternate Disclosure commercials (group E). A one-group ANOVA was used to determine if a significant difference exists between the midpoint and the average response of the Basic message only and Alternate Disclosure groups.
The Basic message only treatment group's beliefs about the likelihood restrictions would apply were examined. This revealed a significant difference on only one of the five scales ("restrictions-no restrictions") in the direction of deceptiveness. Beliefs about a collision damage waiver were also significantly different from the midpoint, but in this instance significance was in the direction of nondeceptiveness.
Table 3 also presents the findings for the group that listened to the Alternate Disclosure message. On all five of the claims the RPI analysis revealed no deceptiveness. On the one scale where there was a significant difference between the midpoint and the Alternate Disclosure mean, the Alternate Disclosure mean was closer to the NAAG Disclosure ("true") end of the scale.
Materiality of the Claims
As indicated earlier, for a deceptive claim to be regulable it must also be material or important. While the question of materiality is normally determined in FTC hearings without the benefit of empirical evidence, such evidence has the potential of affecting the outcome of a legal action (Richards 1990).
To determine the materiality of the five claims for which beliefs were measured, subjects were asked to indicate if each claim would be a factor in their decision whether or not to take advantage of the offer. In every case the mean of the group which was exposed to the Basic message sandwiched between the other two commercials (group B) fell on the "material" side of the scale.(11) The responses (1 = Would NOT Be a Factor and 9 = WOULD Be a Factor) were as follows: "Restrictions apply" = 6.58; "NOT available at all Avis locations" = 6.84; "Expires on a specific date" = 6.42; "Does NOT include unlimited free mileage" = 6.74; and "Collision damage waiver NOT included in price" = 5.90.
Effectiveness of the NAAG Disclosure
In all cases but one ("Collision damage waiver included"), the mean of the NAAG Disclosure condition (group C) fell closer to the "true" end of the scale than did the mean of the Basic message alone (group B). A one-way ANOVA comparing each found that addition of the NAAG Disclosure did not significantly affect evaluation of the Basic message on the four specific claims: "Available at all locations" (F = 0.34, p = .56); "Expires on a specific date" (F = 3.58, p = .06); "Includes unlimited mileage" (F = 2.54, p = .12); and "Collision damage waiver included" (F = 0.26, p = .61). Only the overall "restrictions apply" evaluation was significantly affected by addition of the NAAG Disclosure (F = 14.26, p = .001).
CONCLUSIONS AND IMPLICATIONS
The present study was designed to evaluate the relative benefits of the proposed NAAG Disclosure for rental car advertising versus a shorter disclosure. In order to accomplish this, four research questions were investigated. The conclusions drawn from the findings regarding each of these questions are discussed.
Q1. Are rental car price claims deceptive, where typical qualifications are not disclosed?
After listening to a radio commercial which described a price special, roughly nine out of ten subjects indicated that they believe restrictions or qualifications apply to the offer. Even under the previous FTC standard it was unusual for the agency to regulate a claim where as little as ten percent of consumers were found to be deceived (Preston and Richards 1986). Current FTC policy seems to require that deceptiveness involve at least 20 to 25 percent of consumers before an ad is regulated (Preston and Richards 1986). As less than 12 percent in group A and ten percent in group B fall on the "False" end of the scale, the FTC probably would not find this deceptive.
An analysis of the data which focused on the issue of legal deceptiveness using the RPI method reveals that in only one of five claims is there deceptiveness among the treatment group exposed to the Basic message commercial. On each of the specific claims covered by the NAAG Disclosure, no deception is indicated. At most, it appears that the commercial with no disclosure is only marginally deceptive.
Q2. Are rental car price claims material?
On each of the five scaled questions regarding possible restrictions, subjects indicated that the restriction would be a factor in their decision whether or not to take advantage of the offer. Therefore, all restrictions are material.
Q3. Will the proposed NAAG Disclosure remove a significant degree of deceptiveness?
On the one scale where there is a significant difference between the treatment group exposed to the Basic message (group B) versus the group exposed to the Basic message with the NAAG Disclosure added (group C), the NAAG Disclosure causes the mean of the commercial to fall significantly closer to the "true" end of the scale ("There are restrictions which apply"). Therefore, the NAAG Disclosure removes deceptiveness.
Q4. Is a proposed alternative shorter disclosure an adequate substitute for the longer disclosure?
As indicated by the RPI analysis, the longer NAAG Disclosure provides no significant advantage over the shorter disclosure on any of the scales examined. Although the Basic message is found slightly deceptive on the overall ("restrictions-no restrictions") question, addition of the short Alternate Disclosure removes this deceptiveness. Therefore, the shorter disclosure appears to be an adequate substitute for the NAAG Disclosure.
The findings of any single investigation must be qualified, for example, in light of the sample and whether or not it may be projected to other populations, the type of data collection instrument used, the data analysis procedures applied, and so on. The use of forced exposure conditions in the experiment certainly affect the results of the study, though the authors feel this an acceptable limitation, offsetting the lack of exposure frequency found in a natural setting. As with any exploratory study, the study should be replicated, perhaps with a broader sample of targeted rental car prospects, prior to attempting to draw definitive conclusions.
Further, the present study examines the impact of only one short disclosure statement. There are a myriad of variations of a short disclaimer which could be examined and which might lead to different conclusions. For example, the short disclaimer could include suggestions of the types of information about which potential renters should inquire.
Along these lines, it is worth noting that the use of a disclosure statement which suggests that consumers contact a reservations agent holds some potential for abuse. The reservations agent could, for example, misrepresent the offer and/or withhold information valuable to the consumer in making an informed choice. While this is an important public policy issue, it falls outside the context of deception in advertising; the focus of the present study.
With such qualifications in mind, the findings of the present investigation seem clear. While the findings indicate that the proposed NAAG Disclosure of qualifications on rental car price offers is an effective remedy for deceptiveness, a shorter disclosure statement is equally potent in affecting consumers' beliefs about such an offer.
Furthermore, subjects were well aware of the types of restrictions which are likely to apply to such offers. On the one scale where a statistically significant difference occurred in analyzing the responses of group B exposed to the Basic message commercial, the mean (3.32) is well toward the "restrictions apply" end of the scale. On all of the claims examined, there is no deceptiveness when the effects of the short disclosure are analyzed.
This study was specifically designed to evaluate the merits of two competing regulatory approaches. Consequently, its results are of greatest import to legal rather than behavioral theory. However, failure of the longer and more specific NAAG Disclosure to provide greater informational benefit over the much shorter Alternate Disclosure demands some behavioral explanation.
One potential explanation can be found in subjects' expectations of restrictions on this offer as revealed by the Basic message only treatments (groups A and B). The short Alternate Disclosure may have been sufficient to reinforce the confidence with which subjects held these preexisting beliefs, thereby affecting their evaluations enough to impede a finding of deceptiveness for group E. If so, the efficacy of this short disclosure may be unique to the prior held beliefs attributable to this particular product/service.
An equally possible explanation is the one provided by information load theory (Keller and Staelin 1987; Malhotra 1982). The Basic message coupled with the NAAG Disclosure squeezes 107 words into approximately a 30-second broadcast commercial. It seems reasonable that the addition of the six copypoints in this longer disclosure might push the limits of consumer information processing, thereby resulting in some miscomprehension (or noncomprehension) of these copypoints.
This suggests research regarding disclosures is needed in two areas: (1) to discover the extent to which the strength and direction of pre-existing beliefs can affect the relative effectiveness of short versus long disclosures, and (2) to explore the possibility that disclaimer length can be subject to a ceiling effect resulting from information overload. Such information would be of critical assistance to regulators who need to know how much information can be provided before it becomes dysfunctional. The application of such insights could apply across a broad range of products and services.
Finally, the present study demonstrates that empirical research has potentially important applications in examining public policy issues. Consumers' responses to advertisers' messages should be investigated empirically prior to making policy decisions. Consumer information can be provided in many forms and through several mechanisms. Although any efforts by regulators to facilitate informed decision-making may be laudable, failure to ensure that the chosen method of presentation is appropriate for consumer use can make those regulations worthless or even detrimental to consumer interests. If consumers are unable to understand or recall the information in the legally mandated form another disclosure technique (e.g., through the rental agent) may be more efficacious. In addition after adopting this requirement, regulatory officials may presume they have done their duty and move on to other matters, without realizing it is ineffective. In such instances a better approach might never be considered, leaving the consumer no better off than before adoption of that regulation. This certainly was not the goal of President Kennedy.
Empirical studies like the one described in this article, which are tailored to fit a specific product/service category, offer the public policymaker a means of examining the impact of proposed guidelines. The NAAG, the FTC, FDA, and other governmental agencies should not make important decisions in a vacuum of information. As others have stressed (Gardner 1975; Preston and Richards 1986; Wilkie 1982), consumer research assessing the necessity and impact of proposed changes in advertising content should be an integral part of public policymakers' decision-making processes.
John H. Murphy is Professor and Jef I. Richards is Assistant Professor, Department of Advertising, The University of Texas at Austin.
1 "Belief" as used here concerns consumer beliefs about the product or service being advertised. Jacoby and Hoyer (1987) describe such beliefs as "referent beliefs."
2 While outside the context of this article, however, note that consumers might accrue other benefits from a longer disclosure, such as reduced costs of information search. Further, if the advertiser's message represents a diseconomy to consumers, inclusion of a disclosure might serve to increase the total value of the message by reducing the dysfunctional information.
3 That most TV spots are 30 seconds or less is not, by itself, a good argument for failing to require full disclosure of important information. If in some situations, relatively short commercials do not permit inclusion of necessary information, commercials of this length may not be appropriate.
4 The benefits are characterized as accruing to consumers and the costs being borne by advertisers. Note, however, that disclosure statements may result in some costs to consumers as well as some benefits to advertisers.
5 Although some states may prefer the Commission's previous definition, several states are bound by their own laws to follow the FTC approach. For example, one of the most active consumer protection advocates associated with NAAG, Texas, is so bound (TEX. CODE ANN. 1987).
6 Note that only a single alternative message is considered here, and some other "shorter disclosure" might have quite different effects.
7 Radio, rather than television, was used in the experiment for five reasons. First, the NAAG guidelines directly apply to "broadcast commercials," which clearly includes radio. Second, rental car companies invest heavily in radio. For example, according to Leading National Advertisers Avis invested approximately $4,500,000 in network radio during 1988 and approximately $1,900,000 in 1989. In 1988 network radio represented 17 percent of Avis's total advertising expenditures, and in 1989 it represented 11 percent. Third, the cost of producing a set of television commercials for use in this experiment was prohibitive. Fourth, radio commercials could be produced in a more professional manner than television given budgetary considerations. Fifth, the combination of both sound and video in television introduced additional confounds not in radio.
8 Howard Bell, President of the AAF, stated, "Instead of these mandated disclosures, the AAF believes a better approach would be a simple statement in ads referring consumers to the sales or reservations agent for restrictions and other information" (AAF 1988a, 1). However, the AAF has adopted no specific alternative wording. The wording developed for use in this study reflects the gist of the AAF proposal and was chosen as the shortest possible alternative. The word "details" was selected because it connotes a broader range of information than "restrictions," but it may also bias the results against a finding of relative efficacy for the alternative disclosure.
9 Note that "true" and "false" refer to objective reality, i.e., what is actually being offered by Avis. Consequently, the "true" description remains the same no matter what claims are made in or omitted from the message.
10 Spoofs have gained wide attention recently. For example, the advertising of Isuzu automobiles features "Joe Isuzu." When Joe Isuzu claims that his new trucks cost only $1.00, consumers comprehend the false claim but do not believe it. If comprehension is accepted as the dependent variable, the claim would be found deceptive. However, if belief is used, there is no deceptiveness.
11 Materiality was determined solely from this treatment group, to avoid the potential confounding influence of the experimental manipulations used in the other treatments.
American Advertising Federation (1988a), AAF News (August 1): 1.
American Advertising Federation (1988b), AAF Washington Report (August/September): 5.
American Advertising Federation (1988c), letter from Howard H. Bell and Wallace S. Snyder to Art Weiss (October 19).
Armstrong, Gary M. and Frederick A. Russ (1975), "Detecting Deception in Advertising," MSU Business Topics, 23: 21-31.
Bates v. State Bar of Arizona, 433 U.S. 350 (1977).
Bettman, James R. and Pradeep Kakkar (1977), "Effects of Information Presentation Format on Consumer Information Acquisition Strategies," Journal of Consumer Research, 3: 233-240.
Bettman, James R., John W. Payne, and Richard Staelin (1986), "Cognitive Considerations in Designing Effective Labels for Presenting Risk Information," Journal of Public Policy and Marketing, 5: 1-28.
Bishop, Nancy (1988), "Ad Groups, AGs Come to Table: Industry Airs Concerns Over Greater Regulation," Adweek, SW Edition (October 3): 1.
Board of Trustees of SUNY v. Fox, 109 S.Ct. 3028 (1989).
Calvani, Terry (1989), "Advertising Regulation: The States v. FTC," Antitrust Law Journal, 58: 253-266.
Central Hudson Gas v. Public Service Commission of New York, 447 U.S. 557 (1980).
Chaplinsky v. New Hampshire, 315 U.S. 568 (1942).
Cliffdale, 103 F.T.C. 110 (1984).
Craswell, Richard (1985), "Interpreting Deceptive Advertising," Boston University Law Review, 65: 657-732.
Donaldson v. Read Magazine, 333 U.S. 178 (1948).
Elliott, Stuart (1988), "More Fine Print Clutters Commercials," USA Today (March 31): 8B.
Emerson, Thomas I. (1981), "The Affirmative Side of the First Amendment," Georgia Law Review, 15: 795-849.
Ford, Gary T. and John E. Calfee (1986), "Recent Developments in FTC Policy on Deception," Journal of Marketing, 50: 82-103.
Foxman, Ellen R., Darrel D. Muehling, and Patrick A. Moore (1988), "Disclaimer Footnotes in Ads: Discrepancies Between Purpose and Performance," Journal of Public Policy and Marketing, 7: 127-137.
Gardner, David M. (1975), "Deception in Advertising: A Conceptual Approach," Journal of Marketing, 39(January): 40-46.
Harris, Richard J. (1977), "Comprehension of Pragmatic Implications in Advertising," Journal of Applied Psychology, 62: 603-608.
Henry, Walter A. (1980), "The Effect of Information-Processing Ability on Processing Accuracy," Journal of Consumer Research, 7(June): 42-48.
Jacoby, Jacob (1977), "Information Load and Decision Quality: Some Contested Issues," Journal of Marketing Research, 14(November): 569-573.
Jacoby, Jacob, Robert W. Chestnut, and William Silberman (1977), "Consumer Use and Comprehension of Nutrition Information," Journal of Consumer Research, 4: 119-128.
Jacoby, Jacob and Wayne D. Hoyer (1987), The Comprehension and Miscomprehension of Print Communications: An Investigation of Mass Media Magazines, Hillsdale, NJ: Lawrence Erlbaum Associates.
Jacoby, Jacob, Wayne D. Hoyer, and David A. Sheluga (1980), Miscomprehension of Televised Communications, Hillsdale, NJ: Lawrence Erlbaum Associates.
Jacoby, Jacob and Carol A. Kohn (1974), "Brand Choice Behavior as a Function of Information Load," Journal of Marketing Research, 11(February): 63-69.
Keller, Kevin Lane and Richard Staelin (1987), "Effects of Quality and Quantity of Information on Decision Effectiveness," Journal of Consumer Research, 14: 200-213.
Keller, Kevin Lane and Richard Staelin (1989), "Assessing Biases in Measuring Decision Effectiveness and Information Overload," Journal of Consumer Research, 15: 504-508.
Kennedy, John F. (1962), "Consumers' Protection and Interest Program," U.S. Congressional Code and Administrative News: 4139-4147.
Lawrence, Jennifer (1988), "State Ad Rules Face Showdown: Victory Could Bolster National Role," Advertising Age (November 28): 4.
Linmark v. Willingboro, 431 U.S. 85 (1977).
Malhotra, Naresh K. (1982), "Information Load and Consumer Decision Making," Journal of Consumer Research, 8: 419-430.
Mattox, James and Hal Stratton (1988), "Should States Regulate Ads?" Advertising Age (August 8): 18.
Mazis, Michael B., Richard Staelin, Howard Beales, and Steven Salop (1981), "A Framework for Evaluating Consumer Information Regulation," Journal of Marketing, 45(Winter): 11-21.
Meyer, Robert J. and Eric J. Johnson (1989), "Information Overload and the Nonrobustness of Linear Models: A Comment on Keller and Staelin," Journal of Consumer Research, 15: 498-503.
Miller, George A. (1956), "The Magical Number Seven, Plus or Minus Two: Some Limits on Our Capacity for Processing Information," Psychological Review, 63(January): 81-97.
National Association of Attorneys General (1988), Task Force on the Car Rental Industry: Draft Guidelines (September 26).
O'Toole, John (1981), The Trouble With Advertising, New York: Chelsea House.
Pepper, Hamilton & Scheetz (law firm) (1988), PH&S Advertising Law Update, 1(September): 5.
P. Lorillard Co. v. FTC, 186 F.2d 52 (4th Cir. 1950).
Posadas de Puerto Rico Associates v. Tourism Company, 106 S.Ct. 2968 (1986).
Preston, Ivan L. (1967), "Logic and Illogic in the Advertising Process," Journalism Quarterly, 44: 231-239.
Preston, Ivan L. (1987), "Extrinsic Evidence in Federal Trade Commission Deceptiveness Cases," Columbia Business Law Review, 3: 633-694.
Preston, Ivan L. and Jef I. Richards (1986), "Consumer Miscomprehension as a Challenge to FTC Prosecutions of Deceptive Advertising," John Marshall Law Review, 19(3): 605-635.
Preston, Ivan L. and Jef I. Richards (1988), "Consumer Miscomprehension and Deceptive Advertising: A Response to Professor Craswell," Boston University Law Review, 68(2): 431-438.
Richards, Jef I. (1987), "Clearing the Air About Cigarettes: Will Advertisers' Rights Go Up in Smoke?" Pacific Law Journal, 19: 1-70.
Richards, Jef I. (1990), Deceptive Advertising: Behavioral Study of a Legal Concept, Hillsdale, NJ: Lawrence Erlbaum Associates.
|Printer friendly Cite/link Email Feedback|
|Author:||Murphy, John H.; Richards, Jef I.|
|Publication:||Journal of Consumer Affairs|
|Date:||Dec 22, 1992|
|Previous Article:||Financing of college education by minority and white families.|
|Next Article:||Hedonic analysis of reliability and safety for new automobiles.|