Does Brand Equity of National Brand Deteriorate When Being Sold in Hard Discounters?

(1)

Master’s Thesis

Does Brand Equity of National

Brand Deteriorate When Being

Sold in Hard Discounters?

Joan Dohartha Rosabella

(2)

Does Brand Equity of National Brand

Deteriorate When Being Sold in Hard

Discounters?

Author

J. D. Rosabella j.d.rosabella@student.rug.nl Koolstraat 9A 9717KA Groningen The Netherlands +31 (0) 6 21 17 41 79

1

st

Supervisor

dr. J. E. M. van Nierop j.e.m.van.nierop@rug.nl Nettelsbosje 2 9747AE Groningen The Netherlands +31 (0) 50 36 37 065

2

nd

Supervisor

prof. dr. J. E. Wieringa j.e.wieringa@rug.nl Nettelsbosje 2 9747AE Groningen The Netherlands +31 (0) 50 36 36 37 093

Master’s Thesis

Completion Date

15.01.2018

University of Groningen

Faculty of Economics and Business

(3)

Preface

Although the concept of hard discounters is not yet popular in my home country, Indonesia, it is a very interesting topic as the rise of hard discounter has shaken up the global retail industry. That is why I was motivated to choose this topic for my master thesis. Although I find this subject to be fascinating, completing a thesis while doing 2 compulsory courses was obviously not the best idea. I still could not believe in myself that I can finally finish this, after a long struggle of not having sufficient Marketing Intelligence courses, to collecting survey from Netherlands-living people.

Therefore, I would like to use this opportunity to express my gratitude to several people without whom this paper would not have existed. First of all, I would like to thank dr. J. E. M. van Nierop, my thesis supervisor, for being patient, open-minded, and supportive. Thank you for the valuable input! Special thanks to my dear family in Indonesia for their continuous love and encouragement. Also, I’d like to thank my Netherlands-based Indonesian gang, thanks for the support during my lowest point of my life!

(4)

Management Summary

The vast growth of discounters has changed the game of retail industry. Discounters who were once viewed as “low price, low quality” are now becoming a serious threat for full-service supermarkets as well as national brands manufacturers. This tight competition then leads manufacturers to eventually create a co-opetition with hard discounters by introducing their brands in hard discounters. However, as they gain incremental profit, they also encounter the incremental risk of having their brand equity deteriorated due to customer’s in-store experience and perception. Looking at the definition, brand equity is viewed as the consumer’s subjective judgment on perceived quality, brand loyalty, and brand awareness/association of a particular brand. However, brand equity might differ according to the product category, whether the particular brand is perceived as being high in hedonic or utilitarian value. When a product is perceived to be hedonic, consumers are more lenient to their emotional judgment and willing to pay higher price premium. Ultimately, the brand equity for hedonic product is higher compared to utilitarian product. Similarly, consumers also distinguish a store image according to its hedonic and utilitarian value. This judgment is made of the in-store experience. Hence, the contrasting image of product and store may affect consumer perception of the national brand’s brand equity (NBBE) when being introduced in hard discounters. However, an incongruity may occur as the drawback of the contrasting image of the brand (hedonic) and the store (utilitarian). Consequently, consumer perception of the brand equity is expected to be lower. In addition, a sales promotion is a popular marketing tool during an introductory phase of a brand. A newly introduced brand, paired with a discount level will help to boost awareness in the market. Although this could be a beneficial strategy in the short term, it could “hurt” brand equity if applied in the long run. The decision of selling national brands in hard discounter is a profitable, yet risky move. Thus, this study tries to provide insightful contribution by answering the main research question: To what extent does the national brand’s brand equity (NBBE) change when being introduced in hard discounter, and what influences this?, which is divided into five sub questions: (1) To what extent

does the NBBE change when being introduced in hard discounter?, (2) To what extent does the sales promotion affect the NBBE?, (3) To what extent does the level of hedonic image of product affect the NBBE?, (4) To what extent does the level of hedonic image of product moderate the impact of introduction in hard  discounter to NBBE?, (5) To what extent does the level of hedonic image of store moderate the impact of introduction in hard  discounter to NBBE?  

(5)

A study with 141 respondents using an online survey was conducted. In the beginning part, consumers are asked to rate two supermarket chains (Aldi and Albert Heijn), and four different product categories (chocolate bar, detergent, soda, toothpaste) according to their hedonic value. To analyse the data, a factor analysis was firstly conducted to create 3 factors from the initial 6 items of brand equity variables. Afterwards, a chow test showed that pooling was not allowed thus the study continued with unit-by-unit model. The next step, regression analysis is executed in order to identify the main and moderating effects. Everything is done in R.

(6)

1. Introduction

Who do you think is the biggest competitor for P&G in Europe? Danone, Unilever, or Nestle? The answer is none of them, it’s Aldi (Steenkamp & Kumar, 2009). This implies how serious the competition between a national brand and a budget-private-label brand is. With the extending growth of private labels over the past years, national brands are now facing tight competition, not only from the brand manufacturers but also the distributors. These retailer brands were used to be considered as low-price, low quality products decades ago; however, they now serve as clear alternatives to national brands (Kapferer, 2008). In addition, the popularity of private label is reported to be growing across European countries. As recorded by Nielsen, the 2016 market share for retailer brands has reached its peak in 9 European countries including the Netherlands, Belgium, Germany, Italy, Poland, Austria, Sweden, Norway, and Denmark. The continuing long-term trend displays how private label’s success is a reflection of a growing confidence that shoppers have in retailer brand. (PLMA, 2017).

(8)

When entering hard discounters, national brand manufacturers are unsure whether the gained benefit would be able to offset the potential risk run while expanding their distribution channels. As the attributes that are associated with a brand name will likely to be altered by customer’s in-store experience and perception, it is possible that the image of national brands becomes damaged, or further, the brand equity could be deteriorated (Burt, 2000). Thus, managing the strength of their brand has become a challenge for national brand manufacturers. In order to do so, the manufacturers should be able to determine and measure the attributes that generate brand equity (Gill & Dawra, 2010).

Previous studies have suggested different definitions and types for measuring brand equity. At a conceptual level, the basic agreement for brand equity seems to be consistent of what has been defined by Farquhar (1989), that brand equity is the value added by the brand to the product. Furthermore, the methods for measuring brand equity generally fall into two main categories: whether it is dollar (financial-based) or utility (consumer-based) value (Srinivasan, 2005; Burt, 2000). While financial-based brand equity mainly focuses on firm value; stock price and income statement (Simon & Sullivan, 1993; Birkin, 1994), the consumer-based brand equity was either defined as an incorporation of both perceptual and behavioral dimensions of customers; which includes awareness, loyalty, perceived quality, and associations (Aaker, 1991), or defined as a price premium price premium a consumer is willing to pay for that brand over the price they are willing to pay for an identical unbranded product (Ferjani et. al., 2009).

Furthermore, brand equity differs according to the product category, whether the brand is perceived as utilitarian or hedonic. A utilitarian product may provide tangible or objective features and primarily instrumental or functional (Hirschman & Holbrook, 1982) whereas a hedonic product possesses non-tangible or subjective features that create pleasurable response from customers (Chaudhuri & Holbrook, 2001), thus providing more experimental consumption, fun, pleasure, and excitement. When a product is perceived as being hedonic, customers are willing to pay more for price premium and ultimately, its brand equity also higher than when the product is perceived as being functional (Yang, 2015). However, when a “discount” image of a store is present (i.e. selling in a hard discounter), a hedonic product may record a lower brand equity compared to functional product (Harleman, 2010).

(9)

Thus, a hedonic store may have a high-quality and status orientation, higher prices, extensive services, a relaxed and affable atmosphere, many active salespeople, and a free-form store layout (Wakefield & Barnes, 1996). While in contrast, functional stores have a value and convenience orientation, limited services, a busy and indifferent atmosphere, a few passive salespeople, and a grid layout. Ultimately, the hedonic or utilitarian nature of stores may influence how customer perceive the products that are sold there, thus proving why brand equity of product sold in utilitarian store is found to be lower than hedonic store.

Furthermore, the perceived value of a product or a store may affect the customer’s judgment of a brand. When there is a congruity of product and store image and the brand image, consumers are able to process the stimuli fluently and have more favorable preference towards the brand, as high fluency is hedonically marked (Reber et. al, 2004). In contrast, when a customer fails to process the stimuli, an incongruity may occur and it translates into lower brand equity. Therefore, the image of a product or store will have an effect on how customers perceive a brand after being sold in hard discounter.

In addition, some marketing strategies are implemented heavily during the introductory stage of a brand. For example, when a national brand is newly introduced in another shopping channel (in this case, a hard discounter), a price promotion will be given in order to gain awareness. Precedent studies found that price promotion affects brand equity, through extending brand awareness not only to the brand itself but also to the entire product category (Blattberg & Neslin, 1990). However, another study found a contrasting effect of price promotion, where repeated price promotion actually offers signals of lesser quality (Bawa & Shoemaker, 1987). In short term, price promotion may lead to immediate effect in financial performance, however, long term effect shows a price sensitivity and switching behaviour that lead to losses in brand equity (Aaker, 1991). Therefore, frequency and magnitude of sales promotion play an important role in building brand equity.

(10)

To what extent does the national brand’s brand equity (NBBE) change when being introduced in hard discounter, and what influences this?

The research question will be answered by these following sub questions: (1) To what extent does the NBBE change when being introduced in hard discounter? (2) To what extent does the sales promotion affect the NBBE?

(3) To what extent does the level of hedonic image of product affect the NBBE?

(4) To what extent does the level of hedonic image of product moderate the impact of introduction in hard discounter to NBBE?

(5) To what extent does the level of hedonic image of store moderate the impact of introduction in hard discounter to NBBE?

The purpose of this study is to fulfill the inadequacy of previous literature of determining the NBBE when being sold in hard discounters by measuring the consumer preference towards the national brand quantitatively; while manifesting different schemes to the introduction, by applying different product or store category and a marketing mix strategy (sales promotion) that can influence the change in brand equity. An earlier study by Harleman (2010) revealed that the change in brand equity of a national brand depends on how much the image of the retailer transferred to a national brand.

Furthermore, the result of this study could be used to gain insight for managers to attentively monitor the national brand “health” after being sold in hard discounters. In addition, managers could revise and customize the marketing policies of the brand over time, therefore generate some more proper strategies to target customers.

(11)

2. Literature Review

In this chapter, we will explore the theoretical framework based on precedent literature. The first part will discuss the dependent variable brand equity and its attributes, followed by the description of independent and moderator variables, including its hypothesis. Eventually, the conceptual framework will be exposed in the last part.

2.1 Brand Equity

Strong brands which posses high brand equity can help managers to achieve not only higher margins and customer loyalty but also better customer response to communications, as well as more cooperation from trade and other intermediaries (Gill & Dawra, 2010). In previous research by Buil et. al. (2013), it is proven that customers are willing to pay price premium for brand with higher brand equity, and both brand preference and purchase intentions increase with brand equity. However, measuring brand equity is none other than a challenge for companies as it should be able to capture the change and reflect its underlying dimensions, and at the same time being applicable to different market and products.

Many types of research have different ways to define brand equity, as seen in Table 1. Aaker (1991) defined brand equity from an organizational perspective, where he defined it as “a set of assets (and liabilities) linked to a brand’s name and symbol that adds (subtracts) to the value provided by a product or service to the firm and/or firm’s customers”, implying that management of brand assets are the responsibilities of companies because it is important to increase the product value for both customers and company. Therefore, a brand is a collection of its tangible aspect (product attributes) and intangible aspects (organization’s value, symbolic representatives) that cumulatively form an image or identity in consumer’s mindset.

(12)

Furthermore, the financial perspective viewed brand equity as the difference in incremental cash flow between a branded product and an unbranded product, i.e. private label (Simon & Sullivan, 1993). It is also defined as the sales and profit a company generates from heaps of marketing strategies they did in the past (Brodsky, 1991). Thus, this perspective translates brand equity into a financial aspect such as company’s balance sheet and eventually their market capitalization. In conclusion, the true value of a brand name is measured by comparing the liquidity value of a branded product versus an unbranded competitor (Fathian et. al, 2015).

Table 1. Definition of Brand Equity

All in all, the advent of a consumer based approach to brand equity established two distinct ways of conceptualizing brand equity, namely a financial (calculated in dollar metric), market based method and a consumer (calculated in utility metric) based method, which is explained in Table 2 below. In general, high CBBE in a national brand will lead to high purchase intention and consequently, will be reflected in a better margin in company’s financial report. Thus, in this study, we will focus on the change in customer perspective (CBBE method) of national brand when it is being sold at hard discounters.

(13)

The dimensions of CBBE that we are going to use is based on conceptual framework developed by Yoo et. al. (2000). Build upon a similar model by Aaker (1991), Yoo determined three dimensions to construct a brand equity: (1) perceived quality, (2) brand loyalty, and (3) brand awareness & brand association. In summary, high brand equity implies that customers have a lot of positive and strong associations related to the brand, perceive the brand is of high quality, and are loyal to the brand. These dimensions have been used by many precedent types of research in brand equity (see Table 3). Hence, the three dimensions will be discussed in detail below.

Table 3. Dimensions of Brand Equity

2.1.1 Perceived Quality

Perceived quality is the consumer judgment on the added value of a product. Zeithaml (1988) defines perceived quality as "the consumer's (subjective) judgment about a product's overall excellence or superiority". Consumer’s subjective judgment of quality can be influenced by personal product experiences, unique needs, and consumption circumstances. As perceived quality is an element to describe a brand’s value, a brand with high perceived quality would lead customers to choose the brand over other brands in the market. High perceived quality implies that customers are not only able to recognize and differentiate the brand, but also acknowledge the superiority of the brand compared to other competing brands.

(14)

However, keeping the brand to be constantly perceived as high quality could be a challenge. A study by Kirmani and Rao (2000) examined the uncertainty in consumer’s mindset when perceiving the product’s quality that sellers may offer. This uncertainty is caused by information asymmetry, where the sellers have more information than the buyers. Thus leading the buyers to make assumption based on the “incomplete” information provided by the sellers. Therefore, to solve this concern, sellers could communicate the unobservable quality of the brand through brand name, product design, packaging, advertisements and other brand identities. These elements collectively are able to create a favorable perceived quality from customers.

Perceived quality is none other than consumer’s opinion on how much a product or brand is able to accommodate their expectations. Accordingly, perceived quality has nothing to do with the actual performance of the product, but rather the consumer’s subjective assessment of that product. (Gill & Dawra, 2010). Eventually, perceived quality can be an important component to define a brand’s equity: the higher the perceived quality, the greater its brand equity.

2.1.2 Brand Loyalty

Brand loyalty is the attitudes of consumer toward a brand preference of a product. Aaker (1991) defines a brand loyalty as “a measure of the attachment that a customer has to a brand”. Moreover, Oliver (1999) defined brand loyalty as “an attained state of enduring preference to the point of determined defense”. Enduring preference means that a person purchases the product again and again, and ‘the point of determined defense’ means that the person defends himself from the attacks of competitors.

Another research by Sheth (1968) developed a factor analytic model where it defines an individual-level brand loyalty as the frequency and the pattern of purchases made by the consumer. This model was able to segregate the buyer’s manifested behavior into two components – the behavior owing to environmental effects and the behavior owing to the individual buyer himself. The importance of this measure of brand loyalty is that it is able to provide individual-level brand loyalty.

A closer look at these definitions can help to conclude that loyalty consists of both the behavioral and attitudinal dimensions. Behavioral loyalty means that there is repeated purchase by the consumers, whereas attitudinal loyalty means the attitude, beliefs, and intentions of the consumer towards the brand.

(15)

(Alvarez et al., 2000). Some researchers have considered that loyal customers are very important to a company because they reduce marketing costs (Rundle-Thiele and Bennett, 2001). Previous researches have shown that store brands have improved tremendously in terms of packaging and quality and have shed the image that they are cheap alternatives to national brands (Verhoef et al., 2009). This is proved by Rondan-Cataluna et. al. (2006) study which found that brand loyalty as a significant influencing factor for both national brands and store brands. Loyalty does not only influence the decision of store choice, but also the brand choice and the quantity of products purchased. Therefore, high loyalty of customer may translate into higher margin accumulated to a company, which also means higher brand equity.

2.1.3 Brand Awareness/Brand Association

Brand awareness enables consumers to recognize a brand from different product category and assist consumers in making decision to purchase (Heding et. al., 2009). According to Keller (1993), brand awareness involves brand recognition and brand recall. Brand recognition is the extent to which a person is able to recognize a particular brand given a set of brands, while brand recall is the extent to which a person is able to remember a brand, given a product category or need. As per Aaker (1991), brand awareness consists of many levels. These levels are brand recognition, brand recall, top of mind, brand dominance, brand knowledge and brand opinion. As one moves from brand recognition to brand opinion, the brand awareness increases.

In conclusion, brand awareness represents the strength of the brand’s presence in the mind of the target audience. The more customers can recognize, retain and remember a brand the likely fact that they will shop the particular brand. This is an advantage against competitors as customers overlook the price and other quality when the trust and know a brand, it became a part of the customer's lifestyle to shop where they are used to and have confidence in the brand (Fathian et. al, 2015).

(16)

brand’, thus represents a basis for purchase decisions and for brand loyalty. It is the strength, favorability, and uniqueness of the brand associations that are responsible for the differential effect of the consumers towards the brand (Fathian et. al., 2015).

Brand awareness with strong associations forms a specific brand image. Strong brand associations, which result in high brand awareness, are positively related to brand equity because they can be a signal of quality and commitment and they help a buyer consider the brand at the point of purchase, which leads to a favorable behavior for the brand (Yoo et. al, 2000). Furthermore, according to Allaway et. al (2011), consumers tend to purchase the most familiar, well-established brands. Especially in low involvement decision settings (i.e. the supermarket), the extent to which a brand is recognized or recalled can be decisive (Kim, Kim & An, 2003). As the national brands are far well-established than store brands, it is understandable that national brands are maximising this competitive advantage to win the market.

2.2 Introduction of National Brand at Hard Discounter

The persistent success of established hard discounter chain such as Aldi and Lidl triggers the increasing number of discount players in the market. Consequently, large hard discounter players acknowledge the need for flexibility as a necessary response to international growth ambition. They gradually shift their concept to a slightly more "moderate" middle position, in which they can be called as "soft discounter". This means more branded product, more shelves, larger stores, longer opening hours, in-store bakery, and multiple payment options (The Export Network, 2017).

From the national brand manufacturer perspective, presence on the hard discounter’s assortment is a way to alleviate their dependency on mainstream retailers. The rapid increase in the number of discounter stores will result in fewer opportunities to reach customers for national brands which are only available in mainstream retailers. On the other hand, mainstream retailers also have extensively developed their own private label lines (Ailawadi, Pauwels, & Steenkamp, 2008), thus put pressure on national brand market share.

(17)

“low price, low quality” products (Wu et al., 2011). If these concerns are not being treated accordingly by manufacturers managers, it will have a great impact on the brand equity.

Moreover, previous study found that consumer perception differs according to the store categories. While the full-service retailers are perceived as expensive and luxurious, the hard discounters are perceived as cheap (Wu et. al., 2011). As the perceived image of store types has shown a strong influence on consumer choice of store and spending (Lourenco & Gijsbrecht, 2013), it is expected that selling national brands in hard discounter will result in lower CBBE. Therefore, although the impact of introducing national brand in hard discounter from the manufacturer perspective has been previously analysed, the impact of the introduction to the consumer perception of NBBE in hard discounters remained investigated. Thus, through this study, we will examine the impact of placing national brand in hard discounter on the consumer’s perceived NBBE. Therefore, the first hypothesis is constructed:

H1: Introduction of national brand (NB) in hard discounter (HD) negatively affects national brand’s brand

equity (NBBE)

2.3 Sales Promotion

Sales promotion, in particular, price promotions (i.e., short-term price reductions such as special sales, media-distributed coupons, package coupons, cents-off deals, rebates, and refunds), are believed to erode brand equity over time despite an immediate short-term financial gain (Yoo et. al., 2000). However, different types of promotional tools (e.g., monetary and non-monetary promotions) may have different effects on sales, profitability or brand equity (Srinivasan and Anderson, 1998).

(18)

due to the gap between expected and observed prices negatively affect brand choice decisions as well as perceived quality, which leads to a decrease in brand equity. Also, price promotion campaigns do not last long enough to establish long-term brand associations, which can be achieved by other efforts such as advertising and sales management (Shimp, 1997). Relying on sales promotion and sacrificing advertising would reduce brand associations, which leads to decreasing brand equity.

Sales promotion is one of the most popular marketing strategies that is implemented during the introductory phase of a product due to the strong short effect in boosting sales. This strategy is also likely to be applied when a national brand is being introduced in hard discounters, in order to gain awareness among the shoppers. However, the continual use of sales promotions has a negative impact on perceived quality and brand association dimensions because this tool leads consumers to think primarily about price, and not about the brand (Yoo et al., 2000). Hence the first hypothesis is constructed as:

H2: Sales promotion of national brand (NB) negatively affects national brand’s brand equity (NBBE) 2.4 Hedonic Value of Product

It has been established within the literature that there are two primary behavioural attitudes behind purchase decision which drive the overall perception of goods and services—utilitarian and hedonic value (Hirschman & Holbrook, 1982). The fundamental difference between the two value lies within the motivations of purchasing the product itself. Utilitarian consumption is task related and cognitively driven as well as appealing to one’s mind and logic, while hedonic consumptions focus more on emotional opinions and responses, multisensory, and appreciation of experience (Cal & Adams, 2014). However, a product can exhibit both utilitarian and hedonic values because they may appeal in different ways towards varying demographic groups. This explains why a person may value a product as equal in both utilitarian and hedonic value, or the same product can be perceived as hedonic to a person and utilitarian to another.

(19)

related with certain image of lifestyle, thus the brand association of hedonic products is more likely to be higher than utilitarian products. Thus, a hypothesis is constructed :

H3a: The more a product perceived to be hedonic, the higher national brand’s brand equity (NBBE)

2.4.1 Moderating Effect of Hedonic Value of Product

Due to the fact that consumers perceive national brands to be associated with other entities, as we discussed in the 2.1.3 section, managers often utilise this situation by linking their brands to certain types of people, places, or other brands in order to improve their brand equity (Keller, 2003).

As hedonic products usually contribute to creating a desired image or lifestyle of a consumer, it is expected that the availability of a hedonic national brand in hard discounters may be seen as incongruent. This incongruity is built upon the failure of processing stimulus that is exposed to a consumer. Precedent studies by Winkielman (2003) proved that high fluency says something about a positive state of affairs, either within cognitive system of in the world. In other words, when a consumer is able to process the stimulus fluently, it may affect positively to their preference, because high fluency is hedonically marked (Reber et. al, 2004).

Thus, when a hedonic national brand is being sold as discounter, an incongruity will arise as consumer may not process this “stimuli” fluently because of the contrasting image of the store and brand, which may result in an unfavourable attitude towards the brand. This particular behaviour could lower their brand association and eventually, hurt the perceived CBBE. Therefore, the following hypothesis is constructed :

H3b: Level of hedonic value of product negatively moderates relationship of introduction of NB in HD and

NBBE

2.5 Hedonic Image of Store

By definition, store image is described as “the way in which the store is defined in the shopper’s mind partly by its functional qualities and partly by an aura of psychological attribute” (Martineau, 1958). Successful retailers have clear and distinctive images to target the customers and at the same time, maintaining their image.

(20)

is identified by consumer perception of the retailer-specific attributes (Lee & Hyman, 2008). For example, consumers may perceive Aldi as the hard discounter supermarket that offers limited assortment of products at low price, while Albert Heijn is seen as a full-service supermarket that sells many brands at varying price. At store-category level, image is identified by consumer perception of typical attributes among stores at the same categorical level. For example, although Lidl and Aldi are distinguished as discounter category because they both offer limited assortment with low price, consumers may perceive Lidl as “soft” discounter as it offers more national brands and more services, such as in-store bakery. While Aldi is seen as “hard” discounter because it has fewer national brands and no in-store bakery.

Similarly to product category, stores could also be classified as either hedonic or functional, according to the difference is atmosphere, floor layout, and shopper’s motivation. In detail, stores might be identified differently based on these attributes as seen in Table 4. The hedonic or utilitarian image of a store meaningfully influences consumer judgment and choice (Lee & Hyman, 2008). Therefore, managers could understand the preference of consumer better by analysing the hedonic or utilitarian image of the store.

Table 4. Difference in Hedonic and Utilitarian Store

2.5.1 Moderating Effect of Hedonic Image of Store

A theory of cue utilization suggests that consumers use an array of cues (i.e. price, color, brand name, brand image, store image, etc.) to assess a product’s quality (Olson and Jacoby, 1972), therefore consumers construct an evaluation from these cues to form an attitude towards the store and its brand (Smeijn et al., 2004). Store image has shown to strongly influence consumer choice of store and spending (Lourenço & Gijsbrechts, 2013). Furthermore, a study by Vahie and Paswan (2006) revealed that store image dimensions, such as store atmosphere and store quality, proved to positively influence the perception of the store’s brand. The influence came from an image

(21)

“transfer” from the store itself to its own brand.

However, it becomes an important task for managers to fit their brand with the appropriate store image. Similar to the incongruity theory that has been discussed on section 2.4.1, when there is a misfit between the product image and the store image, consumers will behave unfavourably to the brand equity due to lower brand association.

For example, a full-service supermarket, i.e. Albert Heijn, is evaluated as expensive and luxurious, while the hard discounters, i.e. Aldi, may be seen as cheap (Wu et al., 2011). Therefore, when a national product is being introduced in hard discounter, consumers find difficulties in processing the stimuli because of the contrasting image of the product and brand. Hence, the next hypothesis is built as following :

H4: Level of hedonic image of store negatively moderates relationship of introduction of NB in HD and

NBBE

2.6 Conceptual Model

(22)

(23)

3. Research Design

3.1 Research Design and Data Collection

The objective of this study is to examine the change of customer based brand equity (CBBE) of national brand when being introduced in hard discounter. Therefore, main effects of introduction in HD and sales promotion as well as moderating effect of hedonic value of product and store. Data is collected using an online questionnaire designed by Qualtrics, which is distributed through social media networks and forums as well as promoted through Facebook Ad. Participant anonymity will be ensured after filling the survey. Since this study is focused on introduction of national brand in hard discounter, the participant is expected to be familiar with the context of hard discounter, especially Aldi stores. Furthermore, all of the questions in questionnaire will be presented in one language, English, to reduce the distortion in translating which can create different interpretation (Chang et. al., 1999).

3.2 Measures

The objective of this study is to examine the change of customer based brand equity (CBBE) of national brand when being introduced in hard discounter. As discussed in the conceptual framework, the CBBE of a particular product will be measured through the following three dimensions: perceived quality, brand loyalty, and brand awareness/brand association. For each of the dimensions, 3 questions with 7-point likert scale from strongly disagree to strongly agree will be asked, which are based on the study of Yoo et al. (2001). The questions of each dimension are derived from the highest standardized loading and t-test value, therefore ensuring that each question is reliable and not correlated across the same construct. Table 5 represents the summary of questions asked in the questionnaire.

Table 5. Summary of Questionnaire

Dimension Item Scale

Age "How old are you?" Any numeric value Gender "What is your gender?" Male

Female Gross monthly

income "How much is your monthly income?"

< € 1.000 € 1.000 - € 2.000 € 2.001 - € 3.000 > € 3.000

(24)

HAVO VWO MBO

HBO/Hogeschool WO/University Family size "How many members in your family?"

1 person 2 persons

3 persons or more

Frequency shopping per

week "How many times do you go grocery shopping in a week?" 1 2 3 4 5 6 or more Favorite grocery

store "What is your favorite store for grocery shopping?"

Albert Heijn Jumbo PLUS Spar Spending

distribution "How would you distribute your money?"

Coop Aldi Lidl Other Perceived Quality

"The likely quality of this product is extremely high" 1 = Strongly disagree "The likelihood that this product is reliable is very high" 2 = Disagree

Brand loyalty "I consider myself to be loyal to this product" 3 = Somewhat disagree "This product would be my first choice" 4 = Neither agree or disagree Brand awarenss/

association

"I am aware of this product" 5 = Somewhat agree "I can recognize this product among other competitor

brands"

6 = Agree

7 = Strongly disagree An advanced factorial experiment will be conducted to analyse the main effect of each explanatory variables such as introduction at HD and sales promotion, followed by an analysis of moderating effect of hedonic value of product and store. Firstly, a factorial design will be built according to the level explanatory variables, which consist of :

• 2 levels of store : Albert Heijn and Aldi. Albert Heijn is a representation of full-service supermarket, which has contrasting image with Aldi which represents a hard discounter. • 3 levels of sales promotion : discount of 0%, 20%, and 40%. According to study by Li

(25)

• 4 levels of product categories : chocolate bar, soft drink, detergent, and toothpaste.

Thus, using the R software, the 2x3x4 factorial design resulted in 24 profiles which are balanced and orthogonal as shown in Table 6. In the experiment, participants are asked to value 6 (six) randomly assigned product profile, by answering 6 questions that represent the brand equity. In the beginning of the questionnaire, socio-demographic background of participants is asked.

Table 6. Variants of Experimental Design

3.3 Validation of Hedonic Product and Hedonic Store Image

As this study is analysing the effect of hedonic value to perceive a product and store image, respondents are asked to value the image of products and stores, because it is expected that brand equity will differ among product categories. For example, products such as chocolate, ice cream,

Variant Store Discount Brand

(26)

beer and chips will be more engaging to hedonic needs (Sloot et al., 2005), and other products such as detergent and toothpaste are more perceived as utilitarian products (Lee & Hyman, 2008), thus resulted in contrasting effect of the perceived hedonic value.

To validate the distinctive hedonic image of each product category, respondents are asked to value 4 attributes of hedonic, using the 7-point scale that was developed by Lee & Hyman (2008). The attributes are shown in Table 7. Before answering the questions, participants are provided with some word meaning to avoid the misunderstanding which could lead to biased judgment. Furthermore, the same questions will be asked to judge the hedonic value of a store. These questions are asked prior to the brand equity questions of product variants.

Table 7. Hedonic Attributes Attribute Interested Uninterested Exciting Unexciting Interesting Boring Fascinating Mundane 3.4 Questionnaire Design

(27)

3.5 Analytical Procedure

(28)

4. Result

In this chapter, the result of the study will be disclosed. This chapter will be started by revealing the descriptive of both dependent and explanatory variables, followed by the factor analysis and main effect as well as the moderation effect. The hypothesis testing also included in each main effect and moderation effect section.

4.1 Descriptive

This section will outline the result of the overall study. A total of 164 respondents participated in this research. However, using the listwise deletion method, 21 respondents were excluded due to missing values (incomplete survey). In addition, 2 respondents’ age is younger than 16 years old, and they are also excluded from the study. Thus, the analysis will continue with 141 respondents with characteristics as seen in Table 8. In conclusion, most respondents are female, ranging from 16 – 24 years old, with monthly income of less than € 1.000, university student, have 3 or more people in the household, and do grocery shopping twice a week.

Table 8. Descriptive of Demographic Background Gender Male 42% Female 58% Age 16-24 56% 25-29 30% 30-49 9% 50 and older 5%

Gross monthly income

< € 1.000 65%

€ 1.000 - € 2.000 21%

> € 2.001 - € 3.000 14%

Education

Basisonderwijs / Elementary 4%

Lager / Voorbereidend Beroepsonderwijs (LBO/VMBO) 2% Hoger Algemeen Voortgezet Onderwijs (HAVO) 2% Voorbereidend Wetenschappelijk Onderwijs (VWO) 10%

Middelbar Beroepsonderwijs (MBO) 4%

Hoger Beroepsonderwijs (HBO/Hogeschool) 16% Wetenschappelijk Onderwijs (WO/University) 63% Household Size

(29)

2 persons 11%

3 persons or more 66%

Frequency Shopping per Week

1 29%

2 30%

3 25%

4 and more 16%

Figure 2. Favorite Grocery Store

Figure 3. Spending Distribution

110 99 12 12 15 30 53 32 A L B E R T H E I J N J U M B O P L U S S P A R C O O P A L D I L I D L O T H E R 34% 28% 2% 2% 2% 9% 15% 8%

(30)

Figure 2 and 3 above exhibit the most popular grocery store and distribution of spending in each store. We can conclude that Albert Heijn and Jumbo are the most popular full-service supermarket. In terms of hard discounter, Lidl is quite more popular than Aldi. This finding also aligns with the distribution spending, where Albert Heijn comes first, followed by Jumbo, Lidl, Aldi, and the rest consecutively.

Furthermore, looking at the distribution of brand equity variables’ score, there are no outlier values found. This is expected as the data is based on 7-point likert scale. Hence, to test it more reliably, we will run an ANOVA test, to check whether the rating of each factor differs across store, product category, or discount level. The test will be revealed in 4.3.4 section.

Figure 4. Hedonic Perception of Product

In this study, respondents are asked to evaluate four different product categories (soft drink, chocolate bar, detergent, and toothpaste) in terms of their hedonic value. The same questions are also given to evaluate the store image between Aldi and Albert Heijn. Refer to Figure 4, it can be concluded that soft drink and chocolate bars both have high value in hedonic image (soft drink = 5.955, chocolate bar = 6.909). In terms of store image, referring to Figure 5, Albert Heijn scores high on hedonic value while Aldi scores moderately lower. This result reveals that respondents have contrasting image of the products and stores in terms of hedonic attributes.

(31)

Figure 5. Hedonic Perception of Store Image

4.2 Reliability of Brand Equity Items

To check whether the scale used in this study is internally consistent, a reliability analysis was conducted. Thus, a Cronbach’s measure reliability coefficient is calculated for the items in each construct of Perceived Quality, Brand Loyalty, and Brand Awareness/Association. Items with Cronbach’s Alpha higher than 0.7 is considered as having high internal consistency which means the scale is recommended for theory testing research (Malhotra, 2007). As seen in Table 9 below, each of the construct’s alpha is higher than 0.7, meaning that the items (questions) within the same construct are measuring the same relevant characteristics.

Table 9. Mean and Cronbach's Alpha of Items

4.3 Factor Analysis of Brand Equity Items

An explanatory factor analysis is then conducted to examine whether the items (questions) are able to form the proposed factors and also to determine whether each of the items is properly loaded

3.18

5.31

0.00 1.00 2.00 3.00 4.00 5.00 6.00

Aldi Albert Heijn

Dimension Mean Cronbach's

(32)

into the intended factors. Therefore, the items that explain same phenomenon are being combined in order to retrieve parsimoniousness and avoid multicollinearity. Thus, this section will discuss the findings from factor analysis of the questions used in the study. The factor analysis is conducted for six items, including 2 items for perceived quality (PQ_1 and PQ_2), 2 items for brand loyalty (BL_1 and BL_2), and 2 items for brand awareness (BA_1 and BA_2). We will go through the correlation analysis, Bartlett test, Kaiser Meyer Olkin test, communalities, principal component analysis (PCA), and lastly, the factor loading.

4.3.1 Correlation Analysis

Figure 6. Scatterplot Matrix

The first analysis is to examine the correlation between items. Figure 6 represents the correlation analysis using spearman method, due to the nature of ordered-categorical in the dataset (likert scale). After running the test, all of the items that belong to the same construct are found to have strongly monotonic relationship with each other, as reflected on the value for PQ_1 with PQ_2, BL_1 with BL_2, and BA_1 with BA_2 which are higher than 0,5. This means, more than 50% of variance in one variable is explained by another variable within the same construct. The result affirms the probability of pursing the items through factor analysis.

(33)

The next step, the Bartlett’s test of sphericity is being conducted in order to check the redundancy between variables that can be summarized with some factors (Snedecor et. al., 1989). The result of this test reveals the p value of <2.22e-16 and the H0 is rejected, meaning that the variables are

related and therefore suitable for structure detection. Furthermore, a Kaiser-Meyer-Olkin test is implemented to measure the sampling adequacy, which is the proportion of variance among variables that might be caused by underlying factors (Cerny & Kaiser, 1977). The result shows that the KMO value is 0.70, which Kaiser identified as “middling”, meaning that the data are likely to factor well, based on the correlation and partial correlation.

4.3.3. Communalities

Table 10. Communalities of Items

Items Initial Communalities

The likely quality of this product is extremely high 1,000 0,723 The likelihood that this product is reliable is very high 1,000 0,846

I consider myself to be loyal this product 1,000 0,647

This product would be my first choice 1,000 0,872

I am aware of this product 1,000 0,995

I can recognize this product among other competitor brands 1,000 0,652

Subsequently, a communalities check is applied to measure how much the variance in the variables has been accounted for by the extracted factors. As shown in Table 10, the communalities for each of the items are above 0,6 which means that all of these items can be considered for future factor analysis. The numbers indicate that more than 60% of each items’ variance is explained by the factor model.

4.3.4. Principal Component Analysis (PCA)

Table 11. Principal Component Analysis Table

Comp.1 Comp.2 Comp.3 Comp.4 Comp.5 Comp.6

Standard Deviation 2,942 1,840 1,397 0,875 0,767 0,663 Proportion of Variance 0,548 0,214 0,124 0,049 0,037 0,028 Cumulative Proportion 0,548 0,763 0,886 0,935 0,972 1,000

(34)

missing information from data, we will see the scree plot to see how many factors (eigenvalues) are best suited for the model.

Figure 7. Scree Plot

Figure 7 shows that three components are the best in factor analysis, and from Table 11, it is stated that those 3 factors can explain 88,6% of the total variance. Thus, we will continue using three factors instead of six items for the analysis.

4.3.5 Determining Three Factors as Dependent Variables

Eventually, the new three factors are determined by calculating the mean of two items (for example, Perceived Quality (PQ) will be defined as the average of PQ_1 and PQ_2, and same applied for Brand Loyalty and Brand Awareness/Association). Therefore, we have three independent variables which are the three factors as suggested by the factor analysis. Table 12 below explains the mean of each factor based on the overall model, by store, by product, and by discount level.

Table 12. Mean of Factors

(35)

Toothpaste 4,572 3,575 4,725

By Discount Level

0% 4,877 3,632 5,222

20% 4,685 3,783 5,250

40% 4,913 3,973 5,452

Looking at the numbers in Table 12, the mean of factors is higher in Albert Heijn, compared to Aldi. Furthermore, in terms of product, soft drink (Coca Cola) has the highest brand equity among other product types, followed by chocolate, detergent, and toothpaste. Lastly, it is revealed that the biggest discount level (40%) also recorded the highest brand equity among other levels; which is surprising as the hypothesis of this study expects otherwise. Thus, this issue will be later discussed in section 4.4.

Furthermore, an ANOVA test is conducted to test the significant difference for all the factor’s means. The result can be seen in Table 13 below.

Table 13. ANOVA Test on Mean of Factors

Perceived Quality Df Sum Sq Mean Sq F value Pr(>F) Product 3 31.17 103.900 52.326 0.001419 ** Store 1 1.23 12.256 0.6172 0.432357 Discount 2 5.98 29.889 15.053 0.222730 Brand Loyalty Df Sum Sq Mean Sq F value Pr(>F) Product 3 28.32 94.392 32.402 0.02173 * Store 1 3.63 36.265 12.449 0.26495 Discount 2 19.01 95.058 32.631 0.03890 Brand

Awareness/

Association Df Sum Sq Mean Sq F value Pr(>F) Product 3 132.56 44.187 173.096 7,97E-08 ***

Store 1 7.89 7.887 30.898 0.07925 .

Discount 2 4.14 2.070 0.8110 0.44486 Significant codes:

0,000 = “***” 0,001 = “**” 0,05 = “*” 0,1= “.”

(36)

analyse which level of attribute differs significantly. From the result which shows in Table 14, we can conclude that level of product indeed has significant difference in all of the variables, although the difference in level may vary.

Table 14. Tukey HSD Test on Mean of Factor

Variable Level Difference Lower Upper P-value

PQ Product 2-4 0,5007 0,1196 0,8819 0,0042 BL _Product _2-4 _0,4150 _-0,0477 _0,8778 _0,0967 BA Product 1-4 0,9337 0,4839 1,3835 0,0000 2-4 1,0617 0,6037 1,5197 0,0000 1-3 0,6959 0,2475 1,1443 0,0004 2-3 0,8239 0,3673 1,2805 0,0000 Store 1-0 0,2187 -0,0256 0,4629 0,0793 4.4 Pooling Test

Due to the product category is revealed to be different significantly in all variables, a chow test will be done to test whether it is feasible to pool the model across product category. Pooling the observations can be statistically justified when the null hypothesis of is not rejected (Leeflang et. al., 2015). Thus, the chow test is implemented as an F-test (Chow, 1960) with formula as follows:

Equation 1. Chow Test

𝐹 =(𝑆𝑆𝑅&''()*− 𝑆𝑆𝑅,-&''()*)/(𝑑𝑓&''()*− 𝑑𝑓,-&''()*) 𝑆𝑆𝑅_,-&''()*/𝑑𝑓_,-&''()*

where:

SSRpooled = residual sum of squares of the pooled regression

SSRunpooled = the sum of the residual sum of squares from the unit-by-unit regression

dfpooled = degrees of freedom of pooled regression

dfunpooled = total degrees of freedom unused from the unit-by-unit regression

(37)

Table 15. Residual Sum of Square (SSR)

Model Residual sum of square (SSR)

PQ BL BA Pooled 1,072.93 1,454.58 1,430.96 Unit by Unit Chocolate 250.13 301.01 213.13 Detergent 202.78 263.33 345.28 Soda 203.87 305.14 143.31 Toothpaste 209.77 327.46 438.62

Total Unit by Unit 866.56 1,196.93 1,140.34

OLSDV 1,054.64 1,441.62 1,302.63

Subsequently, using the formula in Equation 1, result for F-test and p-value are determined, as summarised in Table 16 in the following.

Table 16. F Test and p-value

Variables

PQ BL BA

Pooled & Unit by Unit

F 1.5258 1.3791 3.3793

p-value* 0.0030 0.0190 0.0007

OLSDV & Unit by Unit

F 1.4409 1.3571 0.9448

p-value* 0.0099 0.0261 0.6169

*significant at 5% level

(38)

4.5 Main Effect and Moderating Effect Analysis

In this step, main effects are analysed using linear regression. As the prior analysis suggested to continue the analysis using the unit-by-unit models, the regression is executed separately for each of product category, thus 12 models are created. Table 17 below exhibits the summary of hypothesis testing for all dimensions, separately for four product categories.

Table 17. Summary of Hypothesis Testing for Four Categories and Three Dimensions

Hypothesis

Dimensions

Perceived quality Brand loyalty awareness/ Brand association C S D T C S D T C S D T H1: Introduction of national brand (NB) in

hard discounter (HD) negatively affects national brand’s brand equity (NBBE)

** **

H2: Sales promotion of national brand

(NB) negatively affects national brand’s brand equity (NBBE)

* * .

H3a: The more a product perceived to be

hedonic, the higher national brand’s brand equity (NBBE)

*** ** ** *** .

H3b: Level of hedonic value of product

negatively moderates relationship of introduction of NB in HD and NBBE

H4: Level of hedonic image of store

negatively moderates relationship of introduction of NB in HD and NBBE

** * ** **

Significant codes:

0,000 = “***” 0,001 = “**” 0,05 = “*” 0,1= “.”

C = Chocolate bar = Accepted (significant and expected effect) S = Soft drink = Rejected (significant but opposite effect) D = Detergent = Rejected (not significant)

T = Toothpaste

(39)

participants regarding a particular store in terms of its hedonic value. Moreover, the variable “Store”, “Discount”, “Gender”, “Income”, and “Education” are categorical, therefore the parameter estimate and significance level are relative to the base attribute level. The base levels are as follow :

1. Store: Albert Heijn 2. Discount: Discount 0% 3. Gender: Male

4. Income: < € 1,000 5. Education: Elementary

4.5.1 Product Category: Chocolate Bar

Table 18. Parameter Estimate for Chocolate Bar

Perceived Quality Brand Loyalty Brand Awareness/ Association Intercept 2.83064 * 1.73324 2.339332 * Main Effect Store Aldi -0.91319 -1.503008 1.43999 Discount 20% -0.52299 * -0.117091 0.212763 Discount 40% -0.30375 -0.036652 0.21399 Hedonic Product 0.1852 0.337064 ** 0.154101 Hedonic Store 0.18798 0.065283 0.390117 *** Moderation Effect Store Aldi*Hedonic Product -0.02082 -0.173264 0.161832

Store Aldi*Hedonic Store -0.19819 -0.089425 -0.42909 **

(40)

VWO 1.14092 0.655291 0.267168 MBO 1.47296 1.681427 0.58884 Hogeschool 0.68106 0.502418 -0.635619 University 0.6553 0.057165 0.098155 Family Size 0.15042 0.258481 0.095588 Frequency shopping -0.20125 * -0.236853 * 0.018364 R square 0.1958 0.2278 0.256 p-value 0.02119 0.003418 0.0005463 Significant codes: 0,000 = “***” 0,001 = “**” 0,05 = “*” 0,1= “.”

Refer to Table 18, for main effect in chocolate category, no significant effect found on introduction in HD, thus H1 could not be supported. However, in terms of discount level, 20% discount level

is revealed to have significant negative impact on perceived quality. This results in the acceptance of H2, which states that price promotion of NB has negative effect on brand equity. Moreover,

hedonic level of product shows a positive relationship in NBBE and H3a is confirmed. Additionally, level of hedonic value of store is found to carry a positive effect on brand awareness/association.

In terms of moderating effect, there is no moderation effect found on product image on the introduction at HD, hence H3b is also not supported. However, the moderation effect of store

image on introduction at HD is found to be significantly negative on perceived quality and brand awareness/association, which is as expected in the H4, where level of hedonic value of store image

will bear a negative impact on brand equity.

For covariates, whereas in this analysis are presented in demographic background and shopping behavior; females are revealed to imply negative attitude towards perceived quality compared to males. With regard to income level, it is displayed that compared to those with monthly income less than € 1.000, people with € 2.001 – 3.000 monthly income shows negative effect on brand loyalty and brand awareness/association, while people with > € 3.000 monthly income also exhibit negative effect on perceived quality and brand awareness/association. It is also revealed that the more people shop in a week, the lower their perception towards perceived quality and brand loyalty of national brands.

(41)

Table 19. Parameter Estimate for Soft Drink

Perceived Quality Brand Loyalty Brand Awareness/ Association Intercept 0.729657 *** -1.75037 4.03896 *** Main Effect Store Aldi -2.480671 ** 0.42681 -2.08026 * Discount 20% 0.100276 0.23099 0.36683 Discount 40% 0.323669 0.50346 0.307977 Hedonic Product 0.436201 *** 0.48697 ** -0.041267 Hedonic Store 0.155371 0.24156 0.382421 *** Moderation Effect Store Aldi*Hedonic Product -0.198332 0.12487 0.097572

Store Aldi*Hedonic Store -0.460104 ** -0.18142 -0.497668 ** Covariates Gender Female -0.493155 * 0.40189 -0.232562 Age 0.003513 0.02207 0.003605 Income € 1.000 - 2.000 0.154372 0.22773 -0.053332 € 2.001 - 3.000 0.074182 -0.61836 0.081743 > € 3.000 -0.265496 -0.22219 -0.94902 * Education VMBO 0.988738 1.89245 -1.257705 HAVO -1.11906 0.36924 0.190661 VWO 1.473023 * 1.12157 0.328262 MBO -1.425526 -1.18969 -0.695501 Hogeschool 0.624278 0.4822 -0.762201 University 1.046922 . 0.60556 0.185517 Family Size 0.333197 * 0.32586 -0.068688 Frequency shopping 0.057287 0.07612 0.001643 R square 0.3392 0.2595 0.272 p-value 4.741e-06 0.001344 0.0006118 Significant codes: 0,000 = “***” 0,001 = “**” 0,05 = “*” 0,1= “.”

Table 19 represents the summary of effect analysis for soft drink category. As hypnotized in H1,

(42)

the level of hedonic value in a product also exhibits positive relationship towards perceived quality and brand loyalty, therefore the assumption of H3b, which states that the more a product is

perceived to be hedonic, the higher the brand equity, is statistically accepted. Lastly, hedonic value of a store also shows a significant positive effect on brand awareness/association, identical to the effect in chocolate.

Furthermore, hedonic value of a store is revealed to negatively moderates the effect of introduction in HD in terms of perceived quality and brand awareness/association. Thus, H4 is accepted. On

the other hand, no moderating effect is found in hedonic image of product. Looking at the covariates, females are found to show negative attitude towards perceived quality compared to males. Moreover, people with > € 3.000 monthly income also imply negative behavior towards brand awareness/association compared to those with income less than € 1.000. It is also revealed VMO and university graduates tend to perceive quality of a brand highly rather than people who are elementary diploma. Lastly, the more member in a family results in higher perception of quality.

4.5.3 Product Category: Detergent

Table 20. Parameter Estimate for Detergent

Perceived Quality Brand Loyalty Brand Awareness/ Association Intercept 5.23545 *** 1.664134 5.07117 *** Main Effect Store Aldi -0.19143 1.245599 0.33 Discount 20% -0.25445 0.024046 -0.03522 Discount 40% 0.18098 0.639207 * 0.4146 Hedonic Product 0.06226 0.555538 *** 0.16663 Hedonic Store 0.07519 -0.047263 0.07648 Moderation Effect Store Aldi*Hedonic Product 0.10742 -0.301098 0.1683

Store Aldi*Hedonic Store -0.02314 -0.001553 -0.17781

Covariates

Gender Female 0.06225 0.201842 -0.04784

Age -0.02909 * -0.027454 . -0.02667

(43)

€ 1.000 - 2.000 -0.1144 0.240243 -0.033 € 2.001 - 3.000 -0.18693 -0.065168 1.03304 * > € 3.000 0.29628 -0.098337 0.35101 Education VMBO -2.91698 ** -2.605702 * -1.98529 HAVO 1.10654 1.196948 0.28605 VWO 0.02632 0.991249 -0.72857 MBO -0.393 0.845491 -2.1326 * Hogeschool 0.28201 1.458511 * -0.67351 University -0.31645 0.756084 -0.29955 Family Size -0.09575 -0.070992 -0.41085 * Frequency shopping 0.05369 -0.071948 0.32939 ** R square 0.2275 0.2697 0.274 p-value 0.005089 0.0003489 0.0002587 Significant codes: 0,000 = “***” 0,001 = “**” 0,05 = “*” 0,1= “.”

In Table 20 above, a surprising result of positive relationship between discount level (40%) and brand loyalty is revealed. This result violates the assumption of H₂where it is hypothesized that sales promotion negatively affect the brand equity. By definition, brand loyalty does not exclusively focus on repeated purchases, but also on the internal dispositions or attitudes towards the brand (Oliver, 1999). Similar to the congruity theory which has been discussed in Chapter 2, as this phenomenon happens to a perceived utilitarian product (in this case detergent), people might expect the utilitarian benefit from this product, for example for monetary savings (Blattberg, 1990). Thus, when a utilitarian product goes on sale, people react positively because of the congruent stimuli. This also explains the contrasting result from the perceived hedonic product (in this case chocolate bare), whereas the same level of discount brought a negative effect on the brand loyalty. As chocolate is perceived to have hedonic value (refer to Figure 4) a utilitarian benefit of price promotion might produce difficulties in processing the stimuli, hence results in unfavorable attitude towards the brand. Moreover, H_3ais accepted due to strong positive relationship from hedonic product image to brand loyalty.

(44)

in family tend to show a negative effect on brand awareness/brand association, although more shopping frequency implies a positive result.

4.5.4 Product Category: Toothpaste

Table 21. Parameter Estimate for Toothpaste

Perceived Quality Brand Loyalty Brand Awareness/ Association Intercept 4.20681 *** 2.10674 4.676579 ** Main Effect Store Aldi 0.9938 1.13682 1.396371 Discount 20% -0.10518 0.33404 -0.491861 Discount 40% 0.12625 0.53374 . 0.122514 Hedonic Product -0.03292 0.2955 . 0.022823 Hedonic Store 0.21423 . 0.09664 0.293004 Moderation Effect Store Aldi*Hedonic Product 0.0602 -0.03221 0.165496

(45)

Significant codes:

0,000 = “***” 0,001 = “**” 0,05 = “*” 0,1= “.”

As the p-value of brand awareness/brand association is revealed to be not significant, the estimates from this model will not be further interpreted. Thus, we will only interpret the model for perceived quality and brand loyalty (note that the model is significant at 10%). The same phenomenon in the variable discount level also appears for the brand loyalty model, where the model estimates positive relationship of 40% discount to brand loyalty. Thus, H2 is rejected to due

to significant opposite effect. Aligned with the explanation in previous section for detergent product category, the congruity occurred for the toothpaste product category, where both happen to be within the same type of product: utilitarian. This result strengthens the congruity condition that product image and sales promotion. Moreover, the hedonic value of product and store appear to have a positive relationship towards brand loyalty and perceived quality consequently, which makes the acceptance of H3a. There is also a negative moderating effect of store image found in

terms of hedonic store image and perceived quality, hence H4 is also accepted.

(46)

5. Discussion

The extending competition in retail market push the hard discounters and manufacturers to elaborate through a co-opetition (Brandenburger & Nalebuff, 1996), where hard discounter introduces national brands in their assortment. Although both parties seem to enjoy the increasing profit from more profitable buyers (Corstjens and Lal, 2000; Dhar and Hoch, 1997), manufacturers also in a risk of having their brand’s image deteriorated due to customer’s in-store experience and perception (Burt, 2000). Thus, the purpose of this study is to examine whether the brand equity of national brand (NBBE) is indeed lower when being sold in hard discounters. Hence, the hypothesis of this study is the introduction of NB in HD has negative effect on NBBE. In addition, when a product is being newly introduced, one of the most common marketing strategies is selling them at promotional price. Therefore, this study reveals another hypothesis which is negative relationship between price promotion and NBBE. Furthermore, considering that brand equity might vary depending on the category of the product, a moderating effect of hedonic level in product is added. Lastly, as the NBBE might be affected by the in-store experience and perception, we included a hypothesis of moderating effect of hedonic level in stores. In order to conduct the research, a quantitative study with Qualtrics survey is used for the 141 participants. It is revealed that the brand equity is the highest for the product soft drink (Coca Cola).

Firstly, a factor analysis is conducted to identify possible pruning of brand equity items (as there are six items), and it suggested to shrink them into three factors. Furthermore, due to the significant difference in mean value of each product category, a chow test is executed in order to test whether pooling the data is justified. Turned out, pooling is not allowed hence continuing with unit by unit model. The analysis then leads to the creation of 12 models, where each of the four product categories has three different factors as dependent variables. Moreover, five hypotheses are tested using the twelve model.

Does Brand Equity of National Brand Deteriorate When Being Sold in Hard Discounters?

Master’s Thesis

Does Brand Equity of National

Brand Deteriorate When Being

Sold in Hard Discounters?

Joan Dohartha Rosabella

Does Brand Equity of National Brand

Deteriorate When Being Sold in Hard

Discounters?

Author

1

Supervisor

2

Supervisor

Master’s Thesis

Completion Date

15.01.2018

University of Groningen

Faculty of Economics and Business

Preface

Management Summary

Table of Contents

1. Introduction

2. Literature Review

3. Research Design

4. Result

5. Discussion