A FIRM’S STRATEGY TO CONSUMER BEHAVIORAL DECISIONS: PERCEIVED OBSOLESCENCE AND THE TIMING OF REPLACEMENT

다국어 초록 (Multilingual Abstract)

Adding new attributes is the main strategy firms use to attract consumers in many industries, but the impact of new attributes may be ambiguous, as indicated by the results of several studies (Bertini, Ofek and Ariely, 2009; Griffin and Broniarczyk, 2010; Nam, Wang and Lee, 2012; Sun, Keh and Lee, 2012; Zhang and Fitzimons, 1999; Zhang, Kardes and Cronley, 2002). The demand for innovative, upgraded, and integrated products is higher than ever before, and firms’ ability to meet this demand is tenuous in many cases. Take the example of the iPhone, where the touch pad option is the new feature that overcomes the disadvantages of a regular keypad, and where other options such as SIRI and other convenient software technologies provide good reasons to purchase or upgrade a smart phone. Although the transition from keypad to touch screen is occurring rapidly in the technology market, many consumers still opt to utilize their old phones rather than upgrading. The alignability of the attributes of the smart phone may affect their decision-making. Self-regulatory factors moderate decision-making based on alignable and non-alignable attributes. Promotion and prevention self-regulation systems managed across separate mental accounts are involved consumer information flow process in the sense that alignable attributes engender risk aversion, whereas non-alignable attributes engender risk seeking (Zhou and Pham, 2004). More cognitive effort may be required to evaluate non-alignable attributes due to the absence of inferential references inherent in the newness of features (Nam et al., 2012; Sun et al., 2012). Accordingly, if there is no significant difference in alignable attributes between products, consumers will put more weight on non-alignable attributes (Brown and Carpenter, 2000). For high-technology products, if alignable attributes satisfy consumers’ perceived utility as weighed against the cost incurred, consumers may compare non-alignable attributes in order to justify their purchases. Even if non-alignable attributes are trivial, they may be important to the final decision. Based on evaluations of the latter, consumers can justify the purchase (Brown and Carpenter, 2000). Although non-alignable attributes may seem peripheral and complementary, they may contribute to the success of a product due to their novelty, uniqueness and role in simplifying the purchase choice (Carpenter, Glazer and Nakamoto, 1994). However, if consumers perceive that non-alignable attributes fundamentally affect product quality, they may infer that its alignable attributes are relatively inferior. This thinking process may result in a negative evaluation of the product (Simonson, Nowlis and Simonson, 1993). As mentioned earlier, consumers may associate non-alignable attributes with unnecessary costs (Brown and Carpenter, 2000). In the experiment of Bertini et al. (2009), consumers preferred the MP3 player with the BOSS earphone over the one with no other option, which led to a decrease in the willingness to upgrade the product. They do not evaluate non-alignable attributes in isolation, but may use those instrumentally. In the case of high technology products, when information about non-alignable attributes becomes available, ambiguity arises from the integration of distinctive features and functionalities, which are non-alignable attributes, and moreover from the integration of non-alignable attributes with alignable attributes. Consumer beliefs and attitude resulted from the mediating role of this ambiguity are formed holistically by non-selective attention for integral stimuli (alignable and non-alignable attributes) (Nosofsky, 1986, 1987). When consumers ambiguous towards a bundle of attributes, which is nothing but the product, they are inclined to interpret it positively (Bar-Hill and Budescu, 1995; Goldsmith and Amir, 2010). Consumers are hypothesized to exhibit innate optimism, being biased towards unforeseen benefits. They are even sensitive to the specific types of non-alignable attributes: central vs. peripheral attributes. Thus, in strategizing for product marketing purposes, non-alignable attributes should be positioned as peripheral factors rather than as central factors, especially in the early stages of product introduction. Hypothesis Development Consumers make judgments about the obsolescence of the products they currently use. Various factors influence the decision to upgrade. Obsolescence is a loss in value since the launch of a product, not because the product has become less useful, but because a new product with upgraded features and designs has become available to consumers (Bayus, 1991; Chung, Han and Sohn, 2012; Levinthal and Purohit, 1989). Consumers’ expectations about improvement in the new version are based upon two types of obsolescence that influence their willingness to upgrade (pay) for the new product: technological obsolescence and psychological obsolescence. Much research has been done (Bayus, 1991; Chung et al., 2012; Levinthal and Purohit, 1989); however, consumers’ strategic decision-making has not always been clearly understood. They may not know in advance what product to choose, or they may lack information to utilize during the buying process. In addition, technological progress is so rapid and uncertain that consumers may not be able to keep up with the novelty of new technologies. They may have difficulty selecting among alternatives that will be standard later in the market. Accordingly, although they may perceive the technological obsolescence of the product, they may not replace it until they feel psychologically justified in doing so. They may not intend to replace the product only for reasons of technological obsolescence due to loss of the sunk cost incurred by its replacement (Okada, 2006). While alignable attributes are more likely to be deterministic, functional and important, non-alignable attributes are more likely to be marginal, hedonic and trivial. Accordingly, consumers may form cognitive attitudes toward alignable attributes and expectations as to how these should change to become more useful. In other words, they may develop their own ideal level (point) for each alignable attribute as they become familiar with them. Thus, consumers experience uncertainty within a range unique to each individual. They judge the feasibility of a purchase insofar as it corresponds to their own ideal points. For example, when Microsoft upgraded from MS Office 2002 to MS Office 2007, the text menu-driven UI (user interface) was changed to become ribbon menu-driven (graphic). Many users of MS Office 2002 had difficulty using the upgraded functions of MS Office 2007. Frequently used features and options in the previous version, even such as cut and paste, were suddenly not easy to use, and consumers who upgraded to the next version were puzzled at the fact that they were unable to properly use the function they got used to. Complexity may be detrimental to ease-of-use judgment with inferred learning costs. Although consumers make inferences based upon their ideal points, the uncertainty involved in this process relates to cognitive referral. Since alignable attributes are representative of prevention, these engender risk aversion for consumers. Consumers who are primed with prevention-focus may prefer not to exert effort to find out how to use features which were easy to use in the previous version just for the sake of upgrading to the new product. Accordingly, uncertainty about ease of use of alignable attributes has a negative effect on consumers’ willingness to replace a product As more information on a product’s non-alignable attributes becomes available, consumers’ ambiguity about the updated product will decrease and their likelihood of replacement will decrease (Smith and Amir, 2010; Norton, Frost and Ariely, 2007: Bar-Hillel and Budescu, 1995; Krizan and Windschitl, 2007). In addition to technological obsolescence, curiosity and self-regulation in terms of the newness and novelty of a product may motivate consumers to replace a product. Existing consumers may be very curious about new information and non-alignable attributes, making every effort to process available information (Alba and Hutchinson, 1987; Maheswaran and Sternthal, 1990; Nam et al., 2012; Sujan, 1985). When non-alignable attributes become available, consumers do not look at them separately from the alignable ones with which the upgraded product is already equipped; instead, they evaluate the product holistically (Bertini et al., 2009). In an identification–classification task, subjects that were able to attend selectively to the relevant dimension and filter the irrelevant one for separable stimuli were unable to do so for integral stimuli. Non-alignable attributes directly affect consumers’ willingness (likelihood) to replace a product. Therefore, a mediating role of ambiguity is evident between non-alignable attributes and consumers’ willingness to replace the product. Thus, we present the following hypotheses: Hypothesis 1-1: When information on alignable attributes becomes available, consumers will be primed with prevention-focus. Hypothesis 1-2: When information on non-alignable attributes is available, consumers will be primed with promotion-focus. Hypothesis 2-1: If consumers are uncertain about the alignable attributes of a product, they will less likely purchase the updated version of the product. Hypothesis 2-2: If consumers are uncertain (ambiguous) about the non-alignable attributes of a product, they will more likely purchase the updated version of the product. Each individual tries to minimize the distance from a current state to a desired state by means of self-regulation (Higgins, Kruglanski and Pierro, 2003). The mode of self-regulation may be either locomotive or assessable. Assessment is a self-regulatory mode in which some objects and states are critically and analytically judged relative to desired means and goals in order to select the best among many alternatives (Kluglanski et al., 2000). Assessors intend to minimize the distance from the current state to the desired state by means of comparative processes to measure, construe and evaluate among alternatives (Higgins et al., 2003). Accordingly, they make cognitive efforts to compare and analyze the alternatives prior to the choice behavior. The other self-regulatory mode, locomotion, involves judging of objects and states with reference to a particular goal in the process of selecting alternatives (Kruglanski et al., 2000; Pierro et al., 2006). Because locomotors emphasize behavior and progress, they are quick to analyze the alternatives and direct in their selection. Locomotors are different from assessors, who select the best alternative via comparative processes. Locomotors try to select the best alternative realistically and heuristically to achieve their goals. They may not commit to analytical assessment of all possible problems in the process. In the process of evaluating alternatives, assessors focus on all attributes rather than on some particular attributes. They continuously try to assess both alignable and non-alignable attributes. By contrast, locomotors want to achieve a particular goal rapidly. When they perceive technological obsolescence, they may rely on information about alignable attributes because they are eager to benefit from improved features and functions. In the same vein, when they perceive psychological obsolescence, they may rely on information about non-alignable attributes because of the differences in value of these attributes from those of the existing product. Therefore, we offer the following hypotheses: Hypothesis 3: When assessors evaluate an updated product, they will put more weight on non-alignable attributes than on alignable attributes. Hypothesis 4: When locomotors evaluate an updated product, they will depend upon attribute information corresponding to the two types of obsolescence. EXPERIMENT 1 Method Experiment 1 explores consumers’ regulatory focus which would be differently primed either with alignable attributes or with nonalignable attributes (H1 and H2). And it explores consumers’ purchase intention via the amount of information (low-information vs. high-information) and regulatory modes (locomotion vs. assessment. A pretest was conducted with a separate sample (n=64) to ensure that the product profile used in the experiment matched the situation where participants perceive alignable attributes to be deterministic, functional and salient, and where they perceive nonalignable attributes to be peripheral, hedonic and novel. We picked the external HDD that participants associated with salience and newness. In the main experiment, two hundred seventy one participants read information about the external HDD and then answers on the alignable and nonalignable attributes of this product. The between-subject experimental design consists of information type (alignable vs. nonalignable attributes) x information amount (low-information vs. high-information). Results We examined that consumers will be primed with different regulatory focus (H1 and H2) upon the type of information via a gap analysis and a mediated regression analysis. The test results were that respondents exposed to the information type of alignable attributes would be primed with prevention-focus (4.60) than with promotion-focus (4.20) (t = -3.180, p < .001), and that respondents exposed to the information type of nonalignable attributes would be primed with promotion-focus (4.44) than with prevention-focus (3.99) (t = -2.746, p < .05). Next, the type of information was treated as a dummy variable, where it takes values of “0” for nonalignable attributes and of “1” for alignable attributes. Respondents’ dominant regulatory focus was set to be the value of prevention which was subtracted from the value of promotion. The primed regulatory focus fully mediates between the type of information and purchase intention (Sobel Z-score = -3.827, p < .01). That is, if consumers are exposed to the information type of alignable attributes, they will be primed with prevention-focus, which will negatively affect consumers’ purchase intention. Accordingly, Hypothesis 1-1 and 1-2 were supported. Furthermore, we examined the differential effect on purchase intention via a 2 (type of information) × 2(amount of information) between-subjects ANOVA (H1-2 and H2-2) as in Figure 2. In the case of respondents exposed to alignable attributes, the purchase intention was higher for certain information (3.75) than for uncertain information (3.20) (t = 2.077, p < .05), whereas in the case of respondents exposed to nonalignable attributes, the purchase intention was higher for uncertain information (4.48) than for certain information (3.81) (t = 2.478, p < .05). Prior knowledge and involvement on the stimulus were controlled as covariates, and respondents were divided into groups by means of the information type and the information amount. The interaction between the variables was investigated (F(1, 271) = 13.413, p < .01). Accordingly, H2-1 and 2-2 were supported. Finally, we examined the differential effect on purchase via 2 (regulatory mode) × 2 (type of information) between-subjects ANOVA (H4 and H5) as in Figure 3. Regulatory mode was dichotomized into locomotion and assessment by the median of values of assessment subtracted from those of locomotion. In the assessors’ case, a difference between the purchase intentions by means of the type of information – alignable attributes (3.60) and nonalignable attributes (3.73), was not found (t=-.533, p > .10). Hypothesis 3 was rejected. On the other hand, in the locomotors’ case, the purchase intention of respondents exposed to the information type of nonalignable attributes (4.53) would be higher than that of those exposed to the information type of alignable attributes (3.38) (t = 3.826, p < .01). Accordingly, hypothesis 4 was supported. Prior knowledge and involvement on the stimulus were controlled as covariates, respondents were divided into groups by means of regulatory mode and the information type. The interaction among purchase intention, regulartory mode and the information type (F(1,271) = 7.647, p < 0.01).