STUDY 1: COMPETITIVE TARGETS AND BRAND EXTENSIONS

Should a firm position a brand with established equity directly against the prototypical competitive brand in the new category, against a nonprototypical competitor, or not use a comparative approach? Prior research on comparative (vs. noncomparative) advertising is useful for addressing this question. As the name implies, a comparative advertisement draws an explicit comparison between the advertised brand and one or more competing brands, whereas a noncomparative one does not include a competitor in the ad (Barry and Tremblay 1975).

Until recently, research on comparative vs. noncomparative advertising has been inconclusive with respect to the question of which type of advertising is most effective: on the one hand, compared to noncomparative advertising, the comparative approach has been perceived as being more offensive (Wilson 1976), more impersonal (Goodwin and Etgar 1980), less friendly and pleasant (Wilson and Muderrisoglu 1979), more aggressive and intense (Mazis 1976; Wilson and Muderrisoglu 1979), and less believable (Levine 1976; Murphy and Amundsen 1981; Prasad 1976; Shimp and Dyer 1978; Swinyard 1981; Wilson 1976). On the other hand, comparative advertising has been shown to reduce the psychological distance between the advertised brand (usually a new or a small-share brand) with the comparison brand (usually a prototypical brand) for both superiority and parity claims (Dröge and Darmon 1987; Gorn and Weinberg 1984; Pechmann and Ratneshwar 1991; Pechmann and Stewart 1990; Shimp and Dyer 1978; Walker, Swasy, and Rethans 1985). A recent meta-analysis by Grewal et al. (1997) offers several moderators (i.e., relative market positions of the sponsor and target brands, message content) as possible reasons underlying the conflicting findings--which will be elaborated in the subsequent discussion.

To determine the effect of the type of advertising in the case of brand extensions, a crucial factor to consider seems to be the perceived fit between the brand and the extension. As mentioned earlier, previous research has provided ample evidence that the higher the perceptual fit between the original brand and the extension, the greater is the affect transfer from the original brand to the extension (Aaker and Keller 1990; Park, Milberg, and Lawson 1991; Boush et al. 1987). As a result, consumers may prefer a noncomparative approach and judge an extension to be of higher quality when the fit between the brand and the extension is high because the extension is already closely related to the brand and there seems to be no need to reduce the distance between the brand and the competition. This prediction is consistent with Grewal et al. (1997)'s finding: when the sponsor brand is informative in itself, noncomparative (vs. comparative) format is deemed more effective. In other words, the extension has little to gain by associating itself with a competitive brand but it may be hurt by the negative associations that consumers have about comparative advertising. We therefore predict that, if the fit is high, the extension will be evaluated more positively if a noncomparative format than a comparative format is used.

This argument should particularly apply when a nonprototypical competitor is the target of comparison. Walker, Swasy, and Rethans (1985) showed that comparing (vs. not comparing) a new brand to nonprototypical brands results in a lower perceived similarity between the two. For example, when a new entry to the beer market was compared to nonprototypical brands in the beer category such as Heineken, Michelob, and Löwenbräu, the result was an increase in perceptual disparity between the former and the latter. The beneficial effect of perceived fit should therefore be reduced significantly in cases in which a comparison with a nonprototypical target is made.

Under conditions of low fit, on the other hand, a comparative format should be beneficial for a brand extension because the comparative format reduces the distance resulting from the poor fit between the brand and the product category. This reasoning, however, should apply only to a comparison with a prototypical competitor because the prototypical competitor is the best representative exemplar of the new product category. A nonprototypical competitor, on the other hand, is not a good example of the new product category and therefore should not be able to decrease the distance resulting from the poor perceived fit between the brand and the new product category. This prediction is consistent with Grewal et al. finding on the moderating role of relative market positions: "the comparison brand's market position significantly moderates the relationship between ad format and intention to purchase the sponsored brand." That is, when the sponsor brand is new and the target brand is the market leader, comparative format is more effective than the noncomparative ad; however, no such effect is observed when the target brand is not a market leader but an established small-share brand. In sum, we hypothesize for the brand extension context that:

H1: Effectiveness of comparative advertising will be moderated by the level of perceived fit between the parent brand and the extension. In the context of high (low) fit, the perception of the brand extension will be more favorable for a noncomparative (comparative) format vs. comparative (noncomparative) format.
H2: In a comparative format, targeting the brand extension against a prototypical (vs. nonprototypical) competitor will result in a more positive perception of the extension regardless of the level of perceived fit between the parent brand and the extension.