What to Read Next
- In 1988, Sunkist received royalties worth $10.3 million by licensing its name for use on such diverse products as soda, candy, and vitamins.
- Goodyear claims that their tires are the recommended component of Audi and Mercedes automobiles.
- Konica advertising has emphasized that corporations such as US Air and Kemper Securities use Konica copiers.
What do these recent developments in the brand management arena have in common? First, they suggest that brand names — such as Sunkist — are valuable monetary assets that can be traded. For instance, when Kraft was purchased by Philip Morris at a price of over $13 billion (more than 600 percent of its book value), industry observers noted that the monetary value of the brand name (not captured in a balance sheet) was probably worth a considerable sum.1 Second, perhaps because brand names are valuable assets, they may be combined with other brand names to form a synergistic alliance in which the sum is greater than the parts. Thus the joint promotion of Goodyear and Audi, or Konica and US Air, represents attempts by one or both brands in the alliance to secure corporate endorsements that will improve their market positions. Such activities may involve physical product integration, in which one product cannot be used or consumed without the other (in the case of IBM and Intel), or may simply involve the promotion of complementary use, in which one product can be used or consumed independently of the other (in the case of Bacardi Rum and Coca-Cola). Regardless of the nature of the association, the perception that the two brands are linked, as a consequence of their joint promotion, results in the phenomenon that we are investigating.
What led Goodyear and Konica to decide that promoting their use with another brand was an appropriate strategic option? What led to their selection of the specific brands they are associated with? These two key questions are central to understanding this ubiquitous joint-branding phenomenon. Our interest in joint branding is driven by our observation that current academic research and the popular business press focus on the value of the individual brand, which has a distinct identity independent of other brands. In reality, however, brands often coexist with other brands in the same product. Currently, Diet Coke and NutraSweet are virtually inseparable, as are IBM and Intel. This phenomenon is far more pervasive than the current level of research interest suggests, and the monetary consequences of the joint-branding decision are perhaps more significant than currently recognized.
Joint branding (such as the promotion of Frito-Lay chips with KC Masterpiece barbecue sauce) represents an alternative to in-house development of a brand name; additionally, joint branding may be an efficient alternative to traditional brand-extension strategies. The circumstances in which joint branding is an appropriate strategy and the characteristics of a good ally (which may vary depending on product characteristics) are not well understood. Therefore, from managerial as well as academic perspectives, the question of joint branding is an issue that deserves serious scrutiny.
In this article, we attempt to develop an understanding of the phenomenon of joint branding. To remain consistent with the literature on strategic alliances, from which we draw some conceptual ideas, we use the term brand alliance to describe joint-branding situations and explore the reasons for such alliances. We invoke one popular perspective in economics that is particularly relevant to brand management, information economics.
First, we briefly summarize the economic rationale for the existence of brand names. This description allows us to develop the subsequent rationale for forming brand alliances. Next we provide a managerial decision template and discuss some important elements in the decision to ally or not.2 We conclude with a discussion of how our arguments may be enriched by adding other perspectives, which, once empirically validated, would benefit brand managers and corporations contemplating brand alliances.
The Anatomy of a Brand Alliance
A brand name is generally assumed to reduce a buyer’s shopping effort by providing information about the product’s expected quality.3 This quality could be objective (e.g., the NutraSweet swirl indicates that Aspartame, not saccharin, is an ingredient) or subjective (e.g., the Chanel label has many psychological associations that supersede the objective chemical composition of the product). Consequently, consumers searching for a particular level of quality can simply search for a given brand name with which they have direct or vicarious experience. In economies where brand names do not exist, buyers may attempt to use surrogates to identify where the product was manufactured.4 To provide a foundation for our subsequent discussion of brand alliances, we next briefly summarize some of the current relevant thinking on branding.
The Purpose of Brand Names
One important function of brand names is to give consumers information about product quality. Consequently, the absence of brand names often results in the absence of information about quality. The lack of information is undesirable for both buyers and sellers; buyers are not able to determine the quality levels associated with different units, and sellers have difficulty communicating information about the quality of their products. For instance, an apocryphal story tells of consumers in the Soviet Union, unable to assess the quality of an unbranded refrigerator, who tried to identify the factory in which it was manufactured. Apparently, different factories had developed reputations for different levels of quality.5
In a classic exposition of how the absence of information about product quality can have deleterious consequences for the marketplace, George Akerlof analyzed the used car market.6 He argued that there is “information asymmetry” in the used car market: individual sellers know more about a particular car’s condition than do buyers. If buyers cannot evaluate the car’s quality before purchase, they will worry that the seller is selling a “lemon.” Therefore, buyers will offer low prices to protect themselves from being cheated.7 Consequently, honest sellers of high-quality used cars will not receive a fair (high) price for their products, and will therefore drop out of the market, leaving only low-quality sellers in the market. Thus, the market for used cars will comprise sellers of low-quality cars, and, in extreme cases, there will be no trade in such markets because any willing seller is probably selling a lemon.
One element in the Akerlof story is particularly relevant to our focus: the buyer apparently cannot evaluate quality prior to purchase. In other words, information asymmetry is a problem for experience products, whose quality is unobservable prior to purchase but is observable after purchase and use, and not for search products, whose quality is observable prior to purchase. In the information economics literature, the information asymmetry phenomenon and its consequent market degeneration problem is termed the adverse selection problem, or, perhaps more appropriately, the hidden information problem.8 One way to resolve this problem is to provide marketplace signals.
- Signaling under Hidden Information. When true quality is hidden, and the buyer wants a high-quality product, sellers often convey their quality levels through marketplace signals. A signal is an informational device that, for sellers of different levels of quality, is differentially costly to provide. This means that a used car seller that uses a signal to convey high quality must indeed be selling a high-quality car; if it turns out that the product is of poor quality, then the seller will pay dearly. In the arena of product markets, a company can sometimes signal that it is a seller of high-quality products by offering good warranties. If its product is of poor quality, offering a good warranty would be foolish, since warranty fulfillment costs would presumably be higher for poor-quality products because they are likely to have higher failure rates. Conversely, sellers of high-quality products can afford to offer good warranties since the likelihood that they will have to honor those warranties is relatively low.9 Therefore, the key to an effective signal is that those attempting to use it dishonestly would suffer harmful monetary consequences and thus would not use signals dishonestly.
- Market Signals of Product Quality. Much like warranties, brand names can signal product quality. If the claim associated with a brand is of high quality and the product turns out to be of poor quality, the product can suffer potentially grave monetary consequences. For instance, Schlitz beer toppled from 17.8 million barrels sold in 1974 and a 16.1 percent market share in 1976, to less than 5 million barrels sold in 1982, and less than 1 million barrels in 1986. Three reasons for these losses were:
- Schlitz introduced a new yeast-centered brewing process to reduce the length of the fermentation process by a factor of three, reduce cost, and reduce product variability. But shelf life was also reduced.
- In another cost-reduction effort, the company substituted corn syrup for barley malt, yielding a lighter tasting beer.
- Finally, Schlitz replaced the foam stabilizer used to ensure shelf life to avoid having to list it on the label as required by a new labeling law; the new stabilizer reacted with other chemicals, and a cloudy beer resulted.
In effect, Schlitz systematically reduced the intrinsic quality of its product — each of its actions had an impact on the taste and perceived quality of the product. Consequently, not only did Schlitz lose sales volume and market share, but profits tumbled from $48 million in 1974 to losses of $50 million in 1979, and brand equity is estimated to have been eroded by more than $900 million (over 93 percent) in six years.10 Schlitz was perceived as dishonest and paid the consequences; a brand’s claim of high quality must generally be accompanied by high quality, or else the marketplace can punish the offending brand into oblivion. In other words, a brand (and its associated claim of high quality) can be an effective signal. In this case, Schlitz may have been better served by not making a high-quality claim; apparently consumers punished Schlitz disproportionately for having betrayed their expectations of quality.11
This argument suggests that, among other functions, brand names serve as hostages in the marketplace. A brand name generally tells consumers who the manufacturer of a product is, and whom to punish should the product not perform as expected. Since such punishment means the seller will lose money, the brand name serves as a quality assurance device. Branded products are therefore less likely to debase quality than unbranded products, for fear of retribution.
The Purpose of Brand Alliances
If one brand name on a product gives a certain signal of quality, then the presence of a second brand name on the product should result in a signal that is at least as powerful, if not more powerful than, the signal in the case of the single brand name.12 (If the second brand name is unknown or hidden, the level of quality assurance is unaffected by the presence of the second brand name.) As we argue next, brand alliances can serve as quality signals when an individual brand is unable to successfully signal quality by itself.
· Signals of “Unobservable” Product Quality.
When NutraSweet first appeared on the market, it faced two hurdles. It needed to assure consumers that (1) it did not have an aftertaste, and (2) it was safe. It accomplished the first objective through the now legendary gumball promotion — large quantities of chewing gum containing NutraSweet were mailed to consumers for risk-free trial to demonstrate that the product did not have an aftertaste. The second hurdle was more difficult to overcome. NutraSweet’s claim that it was noncarcinogenic and otherwise harmless if properly used was a quality-related claim that could be confirmed only after long use. Consumers were justifiably skeptical of the claim, since, if it turned out to be false, little could be done to correct the problem or punish the vendor. (In light of the cancer-related risks associated with saccharin, and the general health-related concerns about cyclamates, any consumer skepticism was probably well entrenched.)
The claim of being a “safe” sugar substitute gained considerable credibility once Coca-Cola and Pepsi (and others) implicitly backed the claim by offering products that contained NutraSweet. In effect, these other firms were putting their reputations on the line and communicating to the marketplace that, should the product turn out to be injurious to consumers’ health, then Coca-Cola (and Minute Maid and other Coca-Cola products), Pepsi (and KFC, Taco Bell, and other PepsiCo products), and other firms allied with NutraSweet would all suffer from adverse publicity or potential consumer boycott.13 Given the monetary value of one market share point in these markets, any potential consumer boycott was a threat with teeth. Certainly, consumers do not generally penalize companies involved in a brand alliance every time a jointly branded product does not perform, yet the fear of adverse publicity and the potential monetary consequences if a competitor leverages any embarrassment are generally enough to make companies very careful when trading their brand names.
Our argument thus far has been driven by the notion of information asymmetry and the premise that brand names provide some assurance of product quality for experience products. In principle, therefore, if the sole purpose of a brand alliance is to provide a signal, then brand alliances should not be observed when there is no information asymmetry. In other words, if buyers can accurately predict the quality of a product, then the addition of a brand name does not add much value, since it does not reduce the consumer’s search cost. However, there are several alliances between brands where the quality of particular attributes is relatively easy to observe (or is known) prior to purchase. We now turn to that phenomenon.
· Information about “Observable” Product Attributes.
For products whose true quality is not a mystery, what does the addition of a brand name accomplish? Here, the additional brand name provides information about the presence of attributes that may make the jointly branded product more attractive. Unlike products in which quality is unobservable, where (at least in theory) any reputable ally could serve to signal the quality of the jointly branded product, in this case, the ally should add a functional benefit to the joint brand. Thus, an alliance between Pillsbury and M&Ms gives information about the presence of an additional attribute (that of crunchy candy) in the cookie dough. This information is in addition to any implicit quality guarantee that the M&M name may provide to the Pillsbury cookie dough (or vice versa). In such alliances, if the ally provides only a guarantee of quality, not much is offered to the consumer since, in principle, the consumer can evaluate quality through observation, and does not benefit from the additional quality guarantee provided by the joint brand. For instance, an alliance between Kemper Securities and Pillsbury cookie dough mix would not be particularly appropriate for Pillsbury, since Kemper does not add any meaningful functional benefit to Pillsbury’s product offering, and Pillsbury cookie dough (given the relatively low costs to the consumer of establishing its quality) does not particularly need a reputable ally to guarantee its quality. A case could, however, be made for Kemper Securities (a relatively more experiential product) benefiting from an alliance with Pillsbury.
The experience of Fisher-Price and McDonald’s vividly illustrates the validity of this claim. Fisher-Price and McDonald’s entered into an alliance to produce a line of play food and appliances for children, and the brand alliance failed. For this product (toys), the addition of attributes that enhanced the product’s performance quality was relatively more important than a guarantee that the product would not fail. Fisher-Price did not lack a reputation for quality; there was no concern that the toys manufactured by Fisher-Price did not work well or were unsafe. Therefore, what Fisher-Price needed, in its attempt to provide higher quality products, were products with additional attributes, or products with higher levels of existing attributes. Thus, an alliance with Nintendo, offering a new line of video-driven toys, would perhaps have been more appropriate. In any event, allying with McDonald’s did not provide any incremental benefit despite the fact that both brands address a common segment (children).14
To summarize, brand alliances can serve two purposes for a brand that needs a quality-perception boost. When the unobservable quality of a product is suspect (i.e., the product is an experience product), a brand alliance can provide reassurance about the true quality of the product. Brand alliances for such products are more likely to occur between a brand needing quality-perception enhancement (e.g., Konica copiers) and a brand possessing a reputation (e.g., Dutch Boy paints). Alternatively, a brand alliance can convey information about the enhancement of the attributes available in a product, even when product quality is observable (i.e., the product is a search product). Brand alliances for such products are more likely to occur between a brand that needs access to particular attributes that cannot be easily manufactured or purchased elsewhere, and a brand that has those attributes. For example, Sony, due to its alliance with Dolby, can provide better sound quality in its tape cassette decks; this alliance with a brand that has attributes that can profitably be sold to one segment (i.e., music aficionados) is wise, as long as a larger profit reduction does not occur because of the defection of some customers who prefer a product without the additional attribute.
Evaluating the Brand Alliance Option
The necessary conditions for forming a brand alliance are predicated on the notion that the alliance will provide a profitable quality (or quality perception) boost. In other words, from the brand’s standpoint, a brand alliance is desirable if consumers demand it and are willing to pay for it. Thus, if attributes are added to an existing product, they should enhance the overall “objective” evaluation of the product. In this section, we offer a simple approach for assessing prospective alliances.
A Decision Template
The schematic diagram in Figure 1 shows how to assess the incremental value of one (or more) brand names relative to a generic version of a product. Consider the brand alliance represented by NutraSweet (a brand seeking an ally that will endorse its quality, and offering a valuable and unique attribute, which we call the primary brand) and Diet Coke (a brand with a reputation to offer, and in need of the available attribute, the secondary brand).15
In Figure 1, a generic (unbranded) version of the product (Cell a) is compared with physically identical versions with either NutraSweet identified (Cell b), Diet Coke identified (Cell c), or both identified (Cell d). Customer’s preferences for the three branded situations (Cells b, c, and d) relative to the unbranded situation (Cell a) provides a measure of the demand for the alternative branded products.
For our purposes, we define the following four marginal comparisons to assess the incremental value of the various branded conditions. Several other comparisons may be made for other purposes, but these four comparisons allow us to begin to identify several managerial issues to consider while contemplating a brand alliance.
- The primary brand’s single-handed contribution (α1): the incremental preference associated with Cell b over Cell a reflects the degree to which a version of the product with NutraSweet identified is preferred to a generic (or unbranded) version. This quantity provides a measure of the preference for the NutraSweet name.
- The primary brand’s contribution to its partner (α2): the incremental preference associated with Cell d over Cell c reflects the degree to which a version of the product with both Diet Coke and NutraSweet identified is preferred to a version with only Diet Coke identified. This quantity provides a measure of the incremental preference associated with the NutraSweet name when the partner has been identified.
- The secondary brand’s single-handed contribution (ß1): the incremental preference associated with Cell c over Cell a reflects the degree to which a version of the product with Diet Coke identified is preferred to a generic (or unbranded) version. This quantity provides a measure of the preference associated with the name Diet Coke.
- The secondary brand’s contribution to its partner (ß2): the incremental preference associated with Cell d over Cell b reflects the degree to which a version of the product with both Diet Coke and NutraSweet identified is preferred to a version with only NutraSweet identified. This quantity provides a measure of the incremental preference associated with the name Diet Coke when the partner has been identified.
Each of the preference measures that we define above is useful for answering several different questions, but the answer to one important question is paramount: what are the benefits of a brand alliance relative to going it alone? Simply put, if α 1 and α2 are large, it implies that NutraSweet has a strong brand name. If, ß2 is large, it suggests that NutraSweet would be benefit from a brand alliance with Diet Coke. Finally, if α2 is larger than ß2, it suggests that NutraSweet brings more to the party than does Diet Coke, and so Diet Coke should be eager to form an alliance with NutraSweet.
Dynamism can easily be introduced into the analysis by considering changes to the marginal values over time. For instance, in the early 1980s, α1 and α2 were probably close to zero for NutraSweet, but are likely to be much higher now. Consequently, NutraSweet may now act as an endorser of products wanting to purchase access to the diet-conscious segment. Thus NutraSweet may signal the quality of products such as Diet Shasta and Sugar-Free Yoplait.
Elements of the Framework
What factors should managers consider when using the framework? In its simplest form, the framework can be used to fill in elements of a cost-benefit analysis. Illustrative indicators of cost and benefit elements are provided in Table 1.
· Benefits of the Alliance.
If brand alliances are formed to signal incremental quality, then such a strategy is valuable only if there is a sufficiently large number of current or potential buyers who are concerned about product quality, so that a quality boost through forming a brand alliance will actually generate significant incremental sales for the jointly branded product. The demand for quality is driven largely by two important factors: the buyer’s inherent preference (or taste) for quality (which we term quality sensitivity), and the buyer’s ability to successfully evaluate quality.
Buyer’s quality sensitivity. Buyers differ in their preference for quality. Students purchasing automobiles probably plan to replace them in four years and are perhaps more concerned about price than quality. Conversely, a college graduate purchasing a new car that reflects new-found status and wealth will probably buy a car that is more than purely functional; increasingly wealthy individuals will perhaps emphasize quality more than price in several purchase decision arenas.
The degree to which buyers are quality sensitive, or apprehensive about quality debasement, is probably influenced by several factors:
- The degree of quality variation in the marketplace —if quality does not vary at all, then buyers are probably used to receiving a particular level of quality and are satisfied with that level (or else a high-quality seller would have emerged in response to demand for higher quality). Thus, in the absence of a demand for quality, brands probably should not attempt to convey high quality through brand alliances.
- The risks associated with making a poor choice — if the product in question is publicly consumed, there are social consequences to making a poor choice; if the product goes into an important manufacturing process, a poor choice has economic consequences. Generally, when prices are relatively high, and/or the consequences of poor performance are significant, risk will be high and buyers will be quality sensitive. In such circumstances, brand alliances are one appropriate means of conveying product quality.
- The time lag between purchase and true-quality revelation — often, for unobservable quality, true quality is not revealed for a long time (e.g., durability, by definition, will not be revealed for a while). Under such circumstances, the benefits to the seller of providing low quality while claiming high quality are greater, because it will be a while before the seller is caught. (In other words, there is an opportunity cost to not cheat.) Consequently, under such circumstances, the buyer will be more suspicious and will be reassured if a reputable ally endorses the primary brand’s claim of high quality; hence, a brand alliance is one appropriate way to convey product quality.
Buyer’s ability to evaluate product quality. Two important product factors have an impact on the degree to which the buyer can readily evaluate quality. First, the greater the number of opportunities for product use, the more likely the buyer can evaluate the product prior to purchase, because of the learning that occurs during past purchase and use. In other words, the greater the frequency of purchase, the more opportunities exist to observe product quality and predict future quality. Consequently, brand alliance strategies are likely to provide enhancements to quality perceptions about experience attributes to a greater degree for infrequently purchased products (e.g., durable products such as copiers).
The second factor is the relative newness of the product. Newer products tend to be typified by sparse information about performance, and in such circumstances, brand alliance strategies are appropriate for conveying information about unobservable quality. In other words, the likelihood that there is a concern about the quality of a new product about which little quality-related information exists (e.g., NutraSweet when it was first introduced) is greater than for products that have been available for some time. Once again, brand alliance strategies are likely to provide enhancements to quality perceptions (e.g., for experience attributes such as long-term health concerns) in the case of new products such as NutraSweet.
· Costs of the Alliance.
Two important cost elements need to be kept in mind while contemplating brand alliance strategies.
Royalty fees (or the dowry). The first, direct cost is often a royalty fee like Sunkist’s for the use of its name to signal quality. The purchase price of such signaling power will vary depending on the ally being courted. The size of the royalty fee is influenced by a number of factors. First, the greater the risk of product failure, the greater will be the fee paid to the secondary brand to enter into the alliance.16 Second, the more vulnerable a brand is, the more it risks if the jointly branded product should prove to be of poor quality (and therefore the more credible is its claim of high quality). Hence, from a signaling standpoint, the reputation of the brand being courted (as measured by its vulnerability) should increase the magnitude of the royalty fee. Brands that have more to lose in their core market, in their allied markets, or from allied buyers (in the current time period or in the future), are more at risk than brands that are not exposed in multiple markets. In other words, the greater a brand’s “sales at risk,” perhaps because buyers are vindictive (or quality sensitive), the greater should be the royalty fee it commands. Third, the greater the benefits of a particular partner to the primary brand (ß2 in our framework), the greater will be the alliance’s value to the primary brand, and therefore the higher the fee it should be willing to pay for it. Conversely, the benefits that may accrue to the secondary brand, such as access to hitherto untapped markets, profits from sales to more price insensitive buyers, and the like (α2 in our framework), will reduce the fee to be paid to the secondary brand for entering the alliance.
Opportunity cost (or the cost of monogamy). A second cost component is the opportunity cost associated with a particular relationship. To the extent that a primary brand can ally with only one secondary brand (perhaps because of contractual restrictions or marketplace conditions), the long-run costs of allying with a particular brand need to be accurately assessed relative to the costs and benefits of allying with other potential allies. So, if “divorce” is not an option because of the agreement terms, or because the two brands may come to be strongly associated with each other in the marketplace, the two partners will need to consider the inherent uncertainty of the other brand’s future performance very carefully. For instance, NutraSweet’s continued association with several soft drinks despite the availability of less expensive, generic aspartame is probably driven by the secondary brand’s recognition that its brand and NutraSweet are now inextricably linked in the consumer’s mind. Therefore, if Diet Coke were to drop NutraSweet, both NutraSweet and Diet Pepsi would lose no time in informing the marketplace of this development. The short-run loss in share as some consumers temporarily stop purchasing Diet Coke is a risk that is probably unacceptable to the Coca-Cola executive responsible for this decision, despite the possibility that it may very well be optimal in the long run. Thus, particularly for relatively more permanent brand alliances, the long-run costs of being stuck with one partner when a better partner may come along in the future need to be factored into a decision to form an alliance.
We began this paper with three examples of noteworthy developments in brand management. Clearly, Sunkist, Audi, and Mercedes are benefiting from the “trading” of their brand names, either through royalty payments or through added exposure to untapped markets. Further, Konica is using US Air and Kemper (a national carrier and a conservative financial corporation, respectively) to endorse the quality of its copiers. This alliance is probably temporary, and there are no physical attribute-related synergies that benefit quality-sensitive consumers searching for a better copier; however, for quality-sensitive consumers concerned about experience attributes in a copier (such as durability and service), the presence of US Air and Kemper as brand allies of Konica successfully signals unobservable quality. Notice that these two reputable corporations are endorsing a smaller player in the copier market, thus making themselves vulnerable in their respective markets should the primary brand not have the experience attributes that are being (implicitly or explicitly) endorsed.
Our framework for analyzing brand alliances provides a simple yet powerful way to compare the relative benefits and costs to the parties in an alliance. Brand alliances may yield increased sales for a variety of reasons: access to hitherto untapped markets, access to proprietary technology or expertise, image enhancement, and so on. As we have emphasized throughout, depending on the nature of the product, different allies will be appropriate; for products requiring the endorsement of unobservable quality, a reputable ally is appropriate, while for products requiring an enhancement in their attribute set, an ally with the desired functional ability is appropriate.
The theory behind our arguments is one of several perspectives for understanding this phenomenon. Here we outline some of the other important perspectives that need to be considered to develop a better understanding of brand alliances.
· Psychological Factors.
Our approach has focused largely on objective and perceived quality-related issues. However, there are several other important elements of a brand name that have an impact on buyer behavior; for instance, that different brand names connote different meanings to consumers is obvious. The image associated with a brand often reflects its personality or core concept.17 Thus Ivory is considered a mild soap, while Lava is considered a strong soap; the Lincoln Continental is seen as a conservative and sedate automobile, while the Buick Reatta is viewed as racy and youthful. While these brand-related associations are reflective of the brand’s personality, they do not necessarily have an evaluative dimension associated with them. Nevertheless, the image or identity of a brand often has important strategic consequences for the corporation. For instance, Eastman Kodak found that its brand name, by virtue of its technical image, extended beyond photography, and it successfully entered the battery market, while Disney found that it was associated primarily with children’s entertainment and therefore had to launch a new brand name (Touchstone Pictures) to address the more mature cinema market.18
It is not our intent to ignore other important psychological aspects of brand management, but we have limited our focus to the quality-signaling property of a brand name. Future research should consider issues such as the extent to which consumers change purchasing behavior because of image alterations and perceptions unrelated to quality. The fit between the primary brand and the reputable brand is probably an important consideration when contemplating a brand alliance; the personalities of the brand, just as the personalities of people contemplating matrimony, must not clash.
· Operational Issues.
Thus far, the empirical evidence on brand alliances is largely anecdotal. Few formal consumer surveys or experiments have been conducted to examine the impact of adding different types of secondary brands (either reputable or reputationless) to search- or experience-dominated products that vary in their newness and their frequency of purchase. In our ongoing experimental work, we are manipulating several of these variables to test our propositions.
The dichotomy between search and experience is somewhat complex; it depends on (1) the degree to which product features reveal their true quality; (2) the consumer’s ability to evaluate those features; and (3) the relative importance of those features in the consumer’s decision. Thus, if an automobile’s durability is critical to a particular consumer, and the consumer, who is not an expert on cars, is therefore unable to evaluate durability simply by inspection, then an automobile is an experience product for this consumer. Notice that (1) durability is a product attribute that does not readily reveal its level; (2) most consumers will not be able to tell whether an automobile is durable without having driven it for some time; and (3) durability tends to be a relatively important attribute when purchasing an automobile. However, these difficulties are not insurmountable — several survey and experimental techniques are available to assist in the assessment of product type.19
We view the search-experience dichotomy as a continuum. Every product has some search elements (i.e., some quality-related information can be ascertained prior to purchase, at some cost), and some experience elements (i.e., some quality-related information can be ascertained only after purchase). Most products tend to fall between the end points of the continuum, and it is the extent to which their quality-related attributes can be evaluated (by the segment of interest) prior to purchase that determines the degree to which information asymmetry exists, and the type of strategic alliance that may be appropriate if a quality boost is necessary.
· Alternative Perspectives.
There are several other useful frameworks that may be applied to this problem. For instance, the alliance between two equally reputable brands can be seen as one form of alliance, and an alliance between an established brand and a new brand as another type of alliance. In this second scenario, an interesting tension prevails; it is in the primary brand’s interest to ally with the secondary brand, but it is not in the interest of the secondary brand to allow the primary brand to develop an identity through its association with the secondary brand. However, this particular alternative framework, as well as some others, could very well be analyzed using the principles we develop here.
Transaction cost economics provides a framework for analyzing the decision to internalize an operation (e.g., vertical integration of a manufacturing process).20 In this approach, two extreme options available to firms are “markets” (i.e., firms can “buy” products, services, and so on in the vendor marketplace) or “hierarchies” (i.e., firms can “make” products and services internally). It is normally preferable to buy rather than make, since the control exercised by the invisible hand of a highly competitive market will generally ensure that buying is cheaper than making.21 However, whenever the efficacy of the invisible hand is reduced because of a marketplace inefficiency (such as the presence of cartels, the absence of free flow of information, and so on), it may be preferable to make rather than buy. In other words, it is generally preferable to enter into a brand alliance, unless some characteristic of the marketplace reduces the efficacy of the invisible hand. When that situation prevails, it might be preferable not to enter into an alliance and develop the enhanced product offering (for search products) or an enhanced reputation (for experience products) internally, if circumstances will allow it.
The particular transaction cost economics-based predictions and their logic with regard to the formation of brand alliances is subject matter for another paper. Here we simply point out that there are several circumstances under which a primary brand may not want to form a relationship with a secondary brand (and vice versa) because the partner may behave opportunistically (i.e., act self-interested and deceitful) after the relationship has been formed.
Brand managers are faced with several alternatives when attempting to breathe life into a new or old brand. Typical strategies involve providing benefit-related information to segments that are initially price insensitive for new brands (a skimming strategy), and finding new uses, users, usage occasions, and the like for old brands.
One problem that a new or old brand may face is that its quality perceptions are inconsistent with, or lower than, its “true quality.” Under such circumstances, one fairly powerful means of providing assurance about product quality is to secure the endorsement of a reputable brand. A second problem that we examine is when an existing brand may benefit from adding an attribute that it could either make (by investing in R&D and modifying its current product) or buy (if it is easily available in the vendor marketplace) through the formation of a brand alliance. In several circumstances, it may be appropriate for a primary brand to buy this additional attribute through an alliance with the current manufacturer.
In general, brand alliances are an appropriate strategy when quality-sensitive buyers are apprehensive that the seller is untrustworthy, and the quality of the product or service under consideration is unobservable before purchase. Thus, in the sale and purchase of marketing research services, new products, and products where the consequences of performance failure are significant, sellers would do well to signal their true quality through a mechanism such as a brand alliance. Additionally, brand alliances are an appropriate strategy when quality is observable, if there is a demand for attribute enhancement that is costly to make in-house, relative to buying from an external vendor.
Clearly, the managerial questions involved in the brand alliance decision are not trivial. Such decisions often tend to have long-term consequences and should not be entered into lightly, without adequate analysis or thought. For the primary brand, brand managers have to decide on the types of allies, their relative reputations, and expectations for their future performance. Additionally, managers need to determine the costs and benefits of the alliance as well as the potential threat of opportunism. Conversely, a brand manager for a secondary brand needs to first establish that the quality of the primary brand is good, and likely to remain so, and then assess the potential profits from the association, given its likely costs. The particular elements that will enter into these judgments will vary from one scenario to the next; nevertheless, the principle of comparing the costs and benefits of an alliance is one that will serve brand managers well.
As we indicated earlier, empirical validation, either through assessing consumer responses to brand alliances or through an examination of brand alliance formation among corporate partners, is an appropriate next step. If empirically validated, our arguments could make the transition from being theoretically provocative and potentially relevant to immediately applicable. Further, our theoretical arguments are drawn largely from economic perspectives. For a fuller understanding of the phenomenon, further research should also incorporate psychological perspectives that consider the consumer behavior implications of brand alliances.22
1. Some have termed this monetary value “brand equity,” and the concept has begun to receive considerable scrutiny. NutraSweet is concerned about its brand equity since the patent on aspartame has expired, as is ITC (formerly the Imperial Tobacco Company, headquartered in Calcutta, India) in the wake of increasing brand proliferation and price-based competition. For an informative exposition, see:
D.A. Aaker, Managing Brand Equity (New York: Free Press, 1991).
2. Our prescriptions are based on an extensive survey of the academic literature on brand management and strategic alliances, as well as a number of discussions with executives from numerous corporations including the NutraSweet Company, Pillsbury (Grand Metropolitan), 3M, Leo Burnett, and Montgomery Ward.
3. Of course, brands do much more than provide information about intrinsic product quality. A significant literature base rooted in psychology persuasively argues that brand names develop associations, images, and personalities. For the original discussion of the symbolic value of brand names, see:
B.B. Gardner and S.J. Levy, “The Product and the Brand,” Harvard Business Review, March–April 1955, pp. 33–39.
4. That brand name has a significant impact on perceptions of product quality has often been demonstrated. See:
A.R. Rao and K.B. Monroe, “The Effect of Price, Brand Name, and Store Name on Buyers’ Perceptions of Product Quality: An Integrative Review,” Journal of Marketing Research 26 (1989): 351–357.
5. The humorous story goes on to describe how buyers also relied on the date of manufacture for a quality cue; units manufactured just before or after weekends were likely to be of suspect quality because, as a consequence of widespread alcoholism, workers were either preparing to drink heavily or were recovering from drinking heavily.
6. G.A. Akerlof, “The Market for ‘Lemons’: Quality under Uncertainty and the Market Mechanism,” Quarterly Journal of Economics 84 (1970): 448–500.
7. Our use of terms such as “cheat” and “fraud” is not intended to be pejorative. Rather, it is consistent with the vernacular in information economics and is intended to be descriptive.
8. Information economics deals with two broad classes of problems that occur when transacting parties are differentially informed about some key element of the transaction. In this paper, we address the first problem, that of adverse selection. In general, adverse selection problems are concerned with precontractual issues of determining the type of a vendor (high- or low-quality, for instance) or employee (high or low intelligence), when the true level of the attribute (quality or intelligence) is unobservable and immutable. The second problem, that of moral hazard (sometimes termed the hidden action problem), is concerned with ensuring that transacting parties act in a manner consistent with their original commitments. In general, moral hazard problems are concerned with postcontractual issues of ensuring that a vendor or employee continue to deliver high quality and continue not to shirk, respectively, when their actions are not easy to monitor. For descriptive, analytical, and empirical treatments, respectively, see:
M.E. Bergen, S. Dutta, and O.C. Walker, Jr., “Agency Relationships in Marketing: A Review of the Implications and Applications of Agency-Related Theories,” Journal of Marketing 56 (1991): 1–24;
D.M. Kreps, A Course in Microeconomic Theory (Princeton, New Jersey: Princeton University Press, 1991); and
A.R. Rao and M.E. Bergen, “Price Premium Variations as a Consequence of Buyers’ Lack of Information,” Journal of Consumer Research 18 (1992): 412–423.
9. Clearly, this strategy will only work if the provision of a good warranty by a poor-quality seller raises its costs (and price) to a level higher than that of the high-quality seller. This cost structure allows for high- and low-quality sellers to be separated from each other. If the poor-quality seller can offer a good warranty and successfully absorb the higher costs of warranty fulfillment through charging a higher price (which is still lower than that of the high-quality seller), then warranties will not be a successful signal of quality. Price, advertising, scale of operations, and many other devices have been shown to be good signals of quality under certain conditions. See:
S.J. Grossman, “The Informational Role of Warranties and Private Disclosure about Product Quality,” Journal of Law and Economics 24 (1981): 461–483;
R. Cooper and T.W. Ross, “Product Warranties and Double Moral Hazard,” Rand Journal of Economics 16 (1985): 103–113; and
T.M. Devinney, “Price, Advertising, and Scale as Information-Revelation Mechanisms in Product Markets,” in Issues in Pricing, ed. T.M. Devinney (New York: Lexington Books, 1988), pp. 63–82.
10. Aaker (1991).
11. In a mathematically elegant paper on the topic, Birger Wernerfelt makes the same point by showing how “sending a false signal puts a player in a worse situation from then on than if he had sent no signal.” See:
B. Wernerfelt, “Umbrella Branding as a Signal of New Product Quality: An Example of Signaling by Posting a Bond,” Rand Journal of Economics 19 (1988): 458–466.
12. We are restricting ourselves to alliances that are an appropriate fit in terms of consistency of meaning and context. Therefore, while Godiva and Vlasic are strong names in their respective product categories, they probably do not fit well. In other words, adding anchovies to peanut butter, regardless of how well one likes anchovies and/or peanut butter, is probably not a good idea.
13. The term “reputation” is used in a very specific sense here. A reputable firm is one that will suffer a monetarily adverse consequence in the event of poor quality. The greater a firm’s reputation, the greater the monetary damage that can be inflicted on that firm. An important corollary to this thinking is that the greater the vulnerability of a brand name, the more credible its claim of quality. Consequently, the more damage that can be done to a brand name if it is discovered to have debased quality, the more likely that it can successfully offer itself as a hostage when jointly branded with other brands. This corollary has important implications for the selection of allies.
14. D. Kiley, “Big Brands Unite on New Products,” Adweek’s Marketing Week, 1 July 1991, p. 10.
15. This nomenclature is intended to be descriptive and is not intended to convey a rank order of brands. For the significance of the Diet Coke decision to Coca-Cola, see:
J. Guyon and J. Long, “Coke to Put Much-Cherished Trademark on Diet Cola, in a $100 Million Campaign,” Wall Street Journal, 9 July 1982, p. 5.
16. Several other potential costs are more psychological in nature. For instance, what is the impact of multiple alliances on the current image of the secondary brand? Do such multiple “marriages” lessen the sharpness of the definition of this brand? For an expanded presentation, see:
17. Managing brand concepts and linking them with the appropriate image are among managers’ more important tasks. Building psychological barriers to entry by developing an invisible bond between brands and consumers through the proper management of a brand’s concept over time is expected to enhance the brand’s market performance. See:
C.W. Park, B.J. Jaworski, and D.J. MacInnis, “Strategic Brand Concept-Image Management,” Journal of Marketing 50 (1986): 135–145.
18. B.G. Yovovich, “What Is Your Brand Really Worth?” Adweek’s Marketing Week, 8 August 1988, pp. 18–24.
19. For one example, see:
Rao and Bergen (1992).
20. For an application of this approach to whether a firm should employ a salesperson or rely on independent agents/distributors, see: E. Anderson and B.A. Weitz, “Make-or-Buy Decision: Vertical Integration and Marketing Productivity,” Sloan Management Review, Spring 1986, pp. 3–18.
21. For an application of transaction cost economics to salesforce compensation plans, and for a lucid description of transaction cost economics, see:
G. John and B.A. Weitz, “An Empirical Investigation of Factors Related to the Use of Salary Versus Incentive Compensation,” Journal of Marketing Research 26 (1989): 1–14.
For an original discussion of transaction cost economics, see:
O. Williamson, “The Economics of Organization: The Transaction Cost Approach,” American Journal of Sociology 87 (1981): 548–577.
22. For an example of the application of psychological theories to branding issues, see:
B. Loken and D.R. John, “Diluting Brand Beliefs: When Do Brand Extensions Have a Negative Impact?” Journal of Marketing 57 (1993): 71–84.