Introducing national brands at hard discounters in the Netherlands : the extent to which National Brand manufacturers in the FMCG grocery industry can introduce National Brands at Hard Discounters to increase sales, whi

(1)

1

Master Thesis

Introducing National Brands at

Hard Discounters in the Netherlands

The extent to which National Brand manufacturers in the FMCG grocery

industry can introduce National Brands at Hard Discounters to increase sales,

while minimizing channel conflict.

Name | Isabelle van der Waals Student number | 11143886

Supervisor | Dr. Soulimane Yajjou Submission date | 27-01-2017

Version | Final

Master | MSc. Business Administration - Marketing track Institution | University of Amsterdam

(2)

ii

Statement of originality

This document is written by Isabelle van der Waals who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of

(3)

iii

Acknowledgement

First of all, I would like to express my gratitude to my thesis supervisor Dr. Soulimane Yajjou. Throughout the thesis process he was always willing to lend his guidance when running into a

challenge or when I had a question regarding my research. He always allowed this paper to be my own, yet pointed me in the right direction whenever we thought necessary. Moreover, his expertise within the area of research methodology as well as the retail industry made him a great source of knowledge.

Furthermore, I would like to thank Unilever and especially Arie Punt and Dick van Creij for granting and providing me with access to the retail scanner data in order to conduct this thesis.

(4)

iv

Abstract

National brand (NB) manufacturers and traditional supermarkets (TSs) have employed various retaliation maneuvers including price promotions and the introduction of budget private labels in order to retaliate against the growing market share of hard discounters (HDs). Despite these retaliation maneuvers, HD market share is still increasing in Western economies and the academic research available is limited, especially concerning ways in which NB manufacturers can respond. Recently, NB manufacturers have started to introduce NBs at HDs; however, there is a dearth in research concerning the impact of introducing NBs at HDs for NB manufacturers, HDs and how this may impact the sales at TSs. Through the use of Nielsen scanner data, this study explores the effect that NB introductions have on the sales of the respective manufacturer, the hard discounter as well as other market incumbents including traditional supermarkets in the Netherlands.

This study illustrates that the introductions of NBs at HDs increase the sales for manufacturers in the total market without compromising sales at TSs as it causes HD

shoppers to trade up to premium NB alternatives as opposed to drawing new consumers who traditionally purchase their NBs at TSs. Simultaneously, when introducing NBs in order to maximize effectiveness of the brands introduced, NBs result in higher sales in differentiated product categories. However, it is important that NBs result in overall category expansion at HDs as opposed to category cannibalization of the HDs PLs, which are usually considered to have higher margins and hence result in higher value sales overall to ensure that all market incumbents remain satisfied.

Key words: National brands, hard discounters, market share, market incumbents, brand introduction

(5)

v

Introduction

Hard discounters, from here on referred to as ‘HDs’, have become dominant players in the grocery retailing industry over the past two decades (Vroegrijk, Gijsbrechts & Campo, 2016; Cleeren, Verboven, Dekimpe, & Gielens, 2010; Steenkamp & Kumar, 2009), and have only been fuelled forward by the economic crisis (Lamey, 2014). HDs are a type of retailer differentiated from traditional supermarkets, from here on referred to as ‘TSs’, such as Albert Heijn and Tesco in three major ways; they hold limited assortments of approximately 1,000 lines compared to 100,000 lines at TSs, they offer few or no manufacturer national brands, from here referred to as ‘NBs’, and have an assortment led by private labels, from here referred to as ‘PLs’ (Vroegrijk, Gijsbrechts & Campo, 2013; Dekimpe, Gielens, Raju, & Thomas, 2011; Cleeren et al., 2010). With comparable or even superior quality offerings to NB manufacturers, HDs are a cause of concern for both TSs and NB manufacturers, as they have altered business models, show continuous growth and have gained substantial market share in Western Europe (Vroegrijk, Gijsbrechts & Campo, 2013). With a focus on

operational efficiency, low assortment sizes, and lower service levels, they are able to offer quality products at lower prices (Steenkamp & Kumar, 2009; Lamey, 2014; Weiswig, 2015; Vroegrijk, Gijsbrechts & Campo, 2016).

HDs, like Aldi and Lidl, have secured substantial market share over the years in Western countries, where in “Germany and Norway, their market share already exceeds 30%” (Dekimpe et al., 2011, p. S22) in their respective countries and “in many other European countries (e.g. Austria, Belgium, Denmark) shares are in the 10-15% range” (Dekimpe et al., 2011, p. S22). This thesis focuses on the grocery retail environments in the Netherlands, where HDs have registered a growth of 5% in 2015 and have a market share of approximately 10% (Euromonitor, 2016a). As manufacturers are reliant on TSs to sell the majority of their

(8)

2

consumer products, the efficacy of HDs has forced NB manufacturers to re-evaluate their strategies in order to remain competitive and maintain market share in the future (Dekimpe et al., 2011; Cleeren et al., 2010).

While both Aldi and Lidl are prime HDs, Aldi has lost market share in the Netherlands since 2004 despite its aggressive increase in number of stores, whereas Lidl has managed to maintain its growth by capitalizing on market developments and repositioning itself by emphasizing on service and store attributes like store atmosphere and shopper experience (Punt, 2016; Nielsen, 2016). The assortment at Lidl, in the Netherlands, holds no NBs whereas the retailer does include NBs in their assortment abroad (Steenkamp & Kumar, 2009). Aldi, on the other hand, has recently introduced NBs to their assortment in the Netherlands in various product categories including Unilever’s; Dove, Unox and Bertolli. Aldi’s strategy of including NBs in its assortment has been an attempt to increase its market share by capitalizing on the attractiveness and traffic-driving capabilities of NBs (Punt, 2016; Nielsen, 2016), while this strategic decision affects both NB manufacturers and retailers it is important to explore the effects on market incumbents (Lourenço & Gijsbrechts, 2013).

This study further builds upon two previous studies that have had distinguished impact on the knowledge in this particular field. Firstly, the study of Deleersnyder et al., (2007) examine to what extent both NB manufacturers and HDs can benefit from listing NBs at HD formats in Germany, Italy, and Spain. The authors show how NB introductions can lead to a mutually beneficial situation for both HDs and NB manufacturers. Moreover, this study takes the perspective of Dekimpe et al., (2011) into consideration, who suggest NB manufacturers should consider placing their NBs at HDs in order to increase their revenue streams. Dekimpe et al., (2011) illustrate that trends in the environment have led to changes in consumer

preferences and purchasing habits and highlight that further research is necessary regarding the way in which TSs may react when NBs are introduced at HDs and what effect this could

(9)

3

have on NB manufacturers. Both papers consider the perspectives of NB manufacturer and HDs yet fail to consider the possible channel conflicts and side effects on other market incumbents in the industry. Therefore, this thesis answers Deleerysnyder et al.’s (2007) call for further research to determine “the broader impact of NB additions across different retailers” (p316) as well as Dekimpe et al.’s (2011) call for further research to determine possible effects introducing NBs at a HD may have on market incumbents. While a

considerate amount of research has revolved around TSs retaliating against HDs, there is a dearth in research concerning the way in which NB manufacturers can most effectively deal with the increasing market share of HDs in the fast moving consumer goods (FMCG) grocery industry.

TSs have retaliated against HDs by introducing PLs which has resulted in limited empirical support and “the defensive power of EPLs [economy private labels]” (Vroegrijk, Gijsbrechts & Campo, 2016, p309) is questionable as this has not yet been a successful manner to regain market share to TSs. NB manufacturers, on the other hand, have focused on product innovations and promotions in order to maintain and increase market share, despite these retaliation maneuvers, HDs are still gaining momentum. NB manufacturers seek new ways to deal with the rise of HDs yet are still unsure how to effectively do so without compromising relationships with TSs (Dekimpe, et al., 2011). TSs and NBs have been co-dependent, therefore as NBs introduce their products at HDs, TSs, which were known to be the only formats to sell NBs may now lose (part of) their value proposition as NBs are now also available at HDs. To conclude, the previous research on this subject area is limited, fragmented and as illustrated by Lourenço and Gijsbrechts (2013), fails to produce a clear picture of the true effect on all incumbents of when NBs are introduced at HDs. Therefore, the objective of this research is to determine whether NB manufacturers can increase their overall

(10)

4

sales by introducing their NBs at HDs, without compromising the relationships with TSs. This has led to the following main research question:

To what extent can national brand manufacturers in the fast moving consumer goods grocery industry introduce national brands at hard discounters to increase their sales, while minimizing channel conflicts with traditional supermarkets?

All in all, the following research questions will be tackled in order to answer the main research question:

• Research questions 1: To what extent do national brand introductions at a hard discounter result in a growth of sales at the hard discounter for the national brand manufacturer?

• Research question 2: To what extent do the sales of the hard discounter in which a national brand has been introduced increase in the product categories in which national brands have been introduced?

• Research question 3: To what extent are the sales of traditional supermarkets affected in the product categories in which national brands have been introduced at the hard discounter?

• Research question 4: To what extent do product category characteristics moderate the sales of particular introduced national brands?

This paper closes this research gap by exploring the effect of introducing NBs at an HD on all market incumbents in the Netherlands. Hence from an academic perspective, this research will provide insights into between-store competition. As mentioned by Geyskens et al. (2010), preceding research has focused on competition within store, compromising of the effects of high and low tiered product introductions as well as the performance of HDs and NB manufacturers. Additionally, research has been conducted predominantly in Western markets concerning the introductions of NBs at HDs yet fail to mention the Netherlands

(11)

5

(Deleersnyder, et al., 2007), this is interesting seeing as in comparison to Germany and Spain for example, goods sold in on promotion in the Netherlands are considerably higher than its European neighbors (IRi, 2016b). Additionally, the majority of previous papers disregard the effect on the overall market and its incumbents (Lourenço & Gijsbrechts, 2013).

Besides the academic relevance, this paper also has managerial relevance for both NB manufacturers and HDs. For NB manufacturers, this study will help determine two major managerial issues; firstly, whether NB manufacturers can increase and maintain sales without compromising their relationship with TSs by introducing NBs at HDs and hence avoid

channel conflict. This can be achieved by evaluating whether TSs experience a decrease in sales after the introduction of NBs at HDs. Secondly, this research can help determine whether it is beneficial for NB manufacturers to place their products at HDs in terms of total market value sales, to help determine to what extent channel conflict is minimized.

Furthermore, from the HD perspective; this research will provide additional insights on whether the products introduced at HDs result in an increase in product categories sales overall as researched by Deleersnyder et al. (2007) and whether this offsets possible

cannibalization of HD PLs. Therefore, this also study responds to both Geyskens et al. (2010) and Lourenço & Gijsbrechts’ (2013) call for future research regarding the effects of

introducing NBs at HDs by also determining to what extent NBs can lead to a mutually beneficial situation for both HDs and NB manufacturers in the Netherlands (Deleersnyder, et al., 2007).

The conceptual framework is tested by means of a natural experiment, by analyzing 21 NB introductions at a HD, in the Netherlands. The data consists of historical scanner data before and after the introduction of NBs and will be analyzed by means of regression analysis. The structure of this paper is as follows; the subsequent chapter features the background literature of this thesis followed by the research design and methodology. Furthermore, the

(12)

6

results will be elaborated on as well as a discussion chapter of the results followed by the managerial implications, limitations as well as further research opportunities.

Literature review

Background Literature

Traditional supermarkets generally offer “national brands, store brands, and generic or unbranded products” (Juhl, Esbjerg, Grunert, Bech-Larsen & Brunsø, 2006, p.332). Where NBs represent products of “leading national brands of manufacturers” (Steiner, 2004, p.105) and store brands represent brands of TSs (Steiner, 2004, p.105), which are also referred to as PLs. According to Ailawadi & Keller (2004), the overall store image of a retailer is the main influence of shopper choice in terms of grocery store shopping. The store image concept has several dimensions including price, quality, and variety of assortment (Mazursky & Jacoby, 1986; Ailawadi & Keller, 2004). These dimensions are correlated with the four key

dimensions of retail competition including price, variety, assortment and store location (Fox & Sethuram, 2006), hence the retailer strategy influences consumer store choice and spending significantly. The price dimension includes varying prices across categories as well as price promotions and is emphasized by the type of retailing strategy a retailer has utilized, which will be elaborated on in the following chapters. The variety (or breadth) entails the number of categories carried, while the assortment includes the number of brands (or depth) within a category held by the retailer. Lastly, the store location of a retailer includes where retailers are located in relation to its consumers and hence convenience of the store’s location. (Fox & Sethuram, 2006)

Furthermore, it is important to note that a retailer’s image can adapt with time and is updated continuously (Louenço & Gijsbrechts, 2013). To illustrate; the store image of HDs have altered dramatically over the past decade leading to a significant increase in market

(13)

7

share (Dekimpe, et al., 2011). While a decade ago HDs were associated with low-quality products and extremely low prices – this perception of quality has improved and hence have resulted in numerous concerns for all market incumbents (Steenkamp & Kumar, 2009). NB's signal high-quality products, therefore NB presence in a retailer’s assortment play an

important role for the overall store image of a retailer, acting as a perceptual cue for

consumers (Ailawadi & Keller, 2004; Dawar & Parker, 1994). While NB manufacturers have preferred supplying TSs in the past, given their mutual interests of substantial advertising, as well as large assortments (Deleersnyder, et al., 2007), the rise of HDs have become

undeniable. Therefore, NB manufacturers are gradually encouraged to consider supplying HDs as this may be of strategic importance in the long run; however, the strategic

implications of introducing NBs at HDs for all market incumbents is in its infancy (Lourenço & Gijsbrechts, 2013; Deleersnyder, et al., 2007). The following subchapters address the possible challenges as well as benefits of introducing NBs at a HD for all market incumbents. First, however, past retaliation maneuvers on behalf of TSs and manufacturers are discussed followed by channel conflict consequences among market incumbents. Subsequently the effect introducing NBs at HD will be elaborated on in the ‘NB success drivers’ section, followed by the effects on the HD in the ‘Halo effect at the hard discounter’ chapter and the possible effects for TSs are evaluated in the ‘Retailer pricing formats’ section. Lastly, product category moderation is tackled to determine whether this influences NB proliferation. The literature review allows for a progressive multidimensional argumentation about the implication of NBs at HDs and the effects this may have on market incumbents.

Retaliation maneuvers

Given the increasing market share of HDs in the past decade, NB manufacturers and TSs have employed retaliation maneuvers in order to gain back market share. Berman (2015) introduces four strategic options in retaliating against low-cost competition including; “(1)

(14)

8

waiting and watching, (2) deciding not to match new competitors’ price levels, (3) matching or coming close to low-cost competitor price levels and (4) developing new fighter brand or PL brand to be sold along with a company’s traditional brands” (Berman, 2015, p87). These retaliation options are virtually identical to those of Hoch (1996) who provided NB

manufacturers four strategic alternatives in dealing with PL proliferation. However, one key difference includes that as opposed to ‘not matching competitors’ price levels’ (Berman, 2015, p87), Hoch (1996) recommended focusing on product innovations as a fourth strategic option, which is one of the main driving factors highlighting the quality gap between PLs and NBs (Deleersnyder et al. 2007). The following subchapters discuss the predominant

retaliation maneuvers NB manufacturers and TSs have undertaken in attempt to regain market share.

Innovations

Increasing investments in research and development to ensure frequent product innovations has been a dominant way in which NB manufacturers have attempted to regain consumer sales and hence retaliate against HDs (Ailawadi & Keller, 2004). Consumers’ need for variety can be satisfied, by introducing frequent product innovations (Ter Braak, et al., 2014) and hence encourage less price focus among consumers yet increase quality focus. Additionally, HD’s emphasis on keeping costs as low as possible decrease their ability to become prime innovators in the industry and instead attempt to copy NB product innovations (Steenkamp, Van Heerde, Geyskens, 2010; Steiner, 2004). Furthermore, innovations hence stand out in product categories where little innovations take place (Deleersnyder et al., 2007).

Decreasing prices

Another dominant retaliation maneuver on behalf of NB manufacturers include frequent price promotions, however, increasing number of promotions can be detrimental for

(15)

9

a brand and decrease the overall willingness-to-pay for NBs in the long run (Ter Braak et al., 2014). Frequent promotions for NBs makes consumers more price-oriented and hence predominantly purchase NBs while they are in promotion (Guyt & Gijsbrechts, 2014), simultaneously, this also makes standard PLs obsolete seeing as consumers tend to wait for NB promotions which occur frequently, commonly referred to as a pre-promotional dip (van Heerde, Leeflang and Wittink, 2000). While gaining market share from PLs is a favorable outcome for NBs, this also risks a decrease in base sales after promotions becoming enduring and “may erode their brands perceived quality levels” (Geyskens et al., 2010, p.805) resulting in long-term lower margins. Furthermore, NB manufacturers have also resorted to decreasing costs and by decreasing prices of older, less prominent brands in their brand portfolios in attempt to level with the offerings of PLs. However, this is not always successful seeing as price decreases are not consequently implemented by TSs (Keller, 2013). Additionally, Procter & Gamble resorted to decreasing prices of their more prominent brands Always sanitary napkins and Pampers in order to endure the HD competitiveness in Germany (Steenkamp & Kumar, 2009). Furthermore, while increasing operational efficiency and the decreasing prices remain a common retaliation maneuver (Steenkamp & Kumar 2009) this is not a long-term solution for manufacturers or TSs. A business cannot continue decreasing prices as this will not only lead to a deterioration of percentage margin (Berman, 2015). In addition, NB manufacturers need higher margins in order to fund their marketing which allows them to ask a premium price in the first place (Steenkamp, van Heerde & Geyskens, 2010; Juhl et al., 2006).

Introducing fighter brands

Another strategic options highlighted by both Hoch (1996) and Berman (2015) entails NB manufacturers introducing ‘fighter brands’ where the main goal is to compete with HDs and TS PLs as well as defend their premium quality brands (Keller, 2013). However,

(16)

10

introducing ‘fighter brands’ comes at a high price as they are difficult to implement and risk of cannibalization of higher tiered brands is high (Berman, 2015). For example, in 2009 Procter & Gamble introduced ‘Tide Basic’, a less expensive version of their premium washing detergent ‘Tide’ in attempt to defend against the growing market share of HDs (Dekimpe et al., 2011). A year later, Tide basic was delisted as consumers were having difficulty distinguishing between the budget version and the regular version of the detergent, which then led to cannibalization of its premium brand (Ziobro, 2013). As a result, the main issue with NBs creating budget NBs would be the risk of permanent damage to the brand's long-term equity and permanent cannibalization (Dekimpe et al., 2011).

Furthermore, this strategic option has also been employed by TSs by introducing budget PLs in order to fight low-cost competition. Geyskens et al. (2010) investigate the introduction of economy and premium PLs and demonstrate how this can influence the brand choices made by consumers regarding manufacturer NBs. Multiple-tiered portfolios are becoming prominent in the retailing industry. According to Ailawadi, Pauwels, and Steenkamp (2008), there are various reasons for the increase in PL penetration of the past years including; higher margins, better negotiating leverage with NB manufacturers and increased customer loyalty. While standard PLs were typically marketed as being similar quality to NBs, now, TSs are adding lower quality PL products to their portfolios in order to compete against the products and prices of HDs (Geyskens et al., 2010; Kumar & Steenkamp, 2007; Vroegrijk, Gijsbrechts, & Campo, 2016) and hence drive market share back to the TSs. However, this tactic has been called to question; according to Ter Braak, Geyskens, and Dekimpe (2014), this has merely resulted in lower margins for the retailer, low product satisfaction as well as cannibalization of the retailer’s higher tiered products. In addition, Vroegrijk, Gijsbrechts, & Campo (2016) conclude that introducing budget PLs to their assortment in order to drive back market share to TSs has thus far deemed unsuccessful and

(17)

11

leads to category losses as opposed to gains. Lower quality PLs emphasize a focus on lower prices and lower quality in comparison to standard PLs and hence as opposed to increasing store loyalty among consumers, has increased the focus on price comparisons and can reflect poorly on the overall perception of the store image (Vroegrijk, Gijsbrechts, & Campo, 2016).

Channel conflict

Channel conflict is important to consider seeing as NBs are dependent on their

distribution channels to supply their product to the final user, namely the consumer, and hence the relationship market incumbents have with one another can have a significant impact on the way they conduct business. Distribution channel conflicts can occur when partners in the distribution channels do not see eye-to-eye and can be enhanced when there is a

disproportionate distribution of power between two partners (Skinner, Gassenheimer & Kelley, 1992). TSs hold a powerful position in terms of competition between brands in the grocery industry, as they determine shelf space and overall visibility of the product including promotions, displays, and product facings within store solutions (Ailawadi et al., 2008). Hence manufacturers invest substantially in promoting their products and are dependent on successful negotiations with TSs and the brand equity of their NBs to determine how well the product is positioned in store (Martenson, 2007; Juhl et al., 2006). Simultaneously, NBs create traffic for TSs and TSs increase their own brand image by holding familiar and reputable NBs (van Heede & Geyskens, 2010), which in turn “generate consumer interest, patronage and store loyalty” (Aliawadi & Keller, 2004, p332). While TSs often prefer to promote own PLs due to higher profit margins (Ailawadi & Harlam, 2004; Juhl et al., 2006), if they replace NBs with own PLs it will affect their competitiveness in the market

(Martenson, 2007) as this could make TSs less appealing to consumers.

On the other hand, despite the co-dependent relationship between TSs and NB manufacturers and their co-dependence, the introductions of multiple quality tiered PLs by TSs have not

(18)

12

been in the best interest of NBs. While one of the main motivations of implementing store PLs was to increase store loyalty (Ailawadi, Pauwels & Steenkamp, 2008) and hence drive more market share back to TSs as opposed to HDs, the introductions of PLs compete directly with NBs in the same product categories (Ter Braak, et al., 2014). The rise in HD dominance has therefore led to decreasing trade conditions for NB manufacturers for two main reasons. Firstly, TSs have increased PL productions which result in more competition for NBs.

Secondly, TSs set barriers for NBs who wish to place new product innovations on the shelves and delist products with higher frequency (Deleersnyder, et al., 2007).

The declining market conditions for NBs encourage NB manufacturers to seek alternative solutions and guidelines on how to improve their current market situation and hence find new markets and/or channels for their products. While previous literature has focused on the retaliation of TSs, Ailawadi & Keller (2004) and Dekimpe et al. (2011) encourage NB manufacturers to consider supplying HDs which could restore the power balance between TSs and NB manufacturers; however, permitting NBs to be sold at HDs at lower prices could induce channel conflict in addition to deciding to supply HDs at all. Therefore, when supplying NBs to HDs it is not only vital to ensure mutually beneficial trade conditions between HDs and NB manufacturers; however, considering TSs is just as

important in order to avoid channel conflicts and jeopardizing existing distribution channels. A few ways in which channel conflicts can be minimized as highlighted by Dekimpe et al. (2011) can be to produce different sizes and products for the HD channels. Furthermore, as highlighted by Deleeysnyder et al., (2007), if the introduction of NBs is not mutually

beneficial for both HDs and NB manufacturers the HD can choose to discontinue the line of NBs. As previously mentioned, HDs have limited assortments, as well as assortments led by PLs, therefore, it is vital NBs contribute to the overall profitability of the categories in which they have been introduced and minimize category cannibalization.

(19)

13

National Brand success

Drivers of NB success

NB manufacturers have built brand equity over the past decades and have a reputation of providing consistent quality. According to Steenkamp, van Heerde and Geyskens (2010), there are four main driving factors highlighting the quality gap between PLs and NBs including new product innovation, packaging, advertising and price promotions.

In addition to being a retaliation maneuver, product innovation is an important aspect driving the quality gap between PLs and NBs. Conventionally, NBs are known for following a differentiating strategy as opposed to low-cost strategy, which is followed by mainly PLs (Verhoef, Nijssen, & Sloot, 2002), therefore for NBs, product innovation is vital to remain competitive, relevant and differentiating from PLs (Porter, 1980). Simultaneously, according to Deleersnyder et al. (2007) the addition of innovative NBs is more likely to result in

increasing the overall attractiveness of the HDs product category. Distinct packaging is used by NB manufacturers to increase the quality perception of their NBs in comparison to PLs (Steenkamp, van Heerde and Geyskens, 2010). Advertising is a key tactic used by NB manufacturers in order increase the brand awareness, quality perception of the NBs and long term brand equity of their products. According to Kirmani and Wright (1989), the perceived advertising expenditure can be perceived as an indication of the quality of the product. NBs are more likely to have higher advertising budgets in comparison to TSs, as a result,

consumers perceive the quality of NBs higher in which they are prominently advertised (Keller, 1998; Steenkamp, van Heerde & Geyskens, 2010). Price Promotions are mainly used by NBs in order to increase sales in the short run as well as for market penetration purposes and increasing market share (Juhl, et al., 2006). There are contesting views on the efficacy of price promotions as highlighted by Guyt & Gijsbrechts (2014) as well as ter Braak, Geyskens & Dekimpe (2014) while on the one hand price promotions are able to increase manufacturer

(20)

14

revenue in the short run in comparison to competitors yet in the long run the strategic value is limited (van Heerde, Leeflang & Wittink, 2000; Srinivasan, Pauwels, Hanssens, & Dekimpe, 2016).

According to the study conducted by Juhl et al. (2006) PLs are generally products that are not top of mind compared to NBs, which can be explained by the level of advertising, product innovation, distinctive packaging and price promotions (Steenkamp, van Heerde, Geyskens, 2010). This in turn has had a positive effect on consumers’ preferences for NBs due to “consumers reliance on extrinsic cues in quality assessment” (Martenson, 2007, p548). Hence, NBs are a quality and risk assurance to consumers and simplify the purchase decision-making process by serving as a cue, decision-making it easier for consumers to make a choice in product categories (Quelch & harding, 1996) resulting in an overall preference for NBs in TS outlets.

NBs as a risk reduction strategy

Keller (2003) refers to brand equity as the more favorable consumer response to marketing actions in comparison to competitors. According to Fischer, Völckner and Sattler (2010), the summary of preceding literature highlights two main functions of a brand, namely; risk reduction and social demonstrance. Brands are a means by which consumers can identify the source and producer of a product and hence acts as a way of decreasing the chance of making an error in their purchases. Determining the quality of a product can be difficult before consumption of the product, therefore the marketing activities and recognizability of a brand act as a risk-reduction approach used by consumers. (Fischer, Völckner and Sattler, 2010)

Secondly, the social demonstrance function refers to the symbolic function of a brand and emphasizes that brands are a way in which consumers can demonstrate their self-image and lets consumers identify with the ideals of a brand (Laurent & Kapferer, 1985; Fischer,

(21)

15

Völckner and Sattler, 2010; Steenkamp, Van Heerde, & Geyskens, 2010). While PLs are introducing higher quality tiered products and attempt to add brand meaning to their products, NBs still hold an advantage in creating meaning for consumers (Ailawadi & Keller, 2004). As one of the issues HDs and PLs struggle with are the perceived risks of PLs, NBs, on the other hand, excel in the brand imagery and are perceived to have superior quality products and hence have less risk associated with their brands (Ter Braak, et al., 2014). Therefore, given the fact that NBs act as a function of simplifying the purchase decision-making process by serving as a cue as well as providing products with a higher perception of quality and less perceived risk this would, therefore, lead to a preference for NBs.

All in all, given the success drivers of NBs as well as their main function as social demonstrance and risk reduction it is expected that NB introductions at the HD will lead to additional sales for NB manufacturers. This is because shoppers will now be tempted to combine their lower-priced purchases with high-priced NB purchases as well as purchase products in categories in which NB were previously not available. Hence the following hypothesis is proposed:

H1: There is a positive relationship between the introduction of NBs at HDs and the growth in sales of the respective manufacturer.

Halo Effect at the Hard Discounter

Traditional supermarkets are known to have broader and deeper assortments in comparison to HDs, hence keeping supply chain costs to a minimum and HDs proliferating their own PLs (Steenkamp & Kumar, 2009; Vroegrijk, Gijsbrechts & Campo, 2016). Given the low-cost nature of HDs, it is essential that NB introductions increase total category

attractiveness in order for the introductions to be perceived as successful at the HD. While NB manufacturers are predominantly focused on increasing market share of their brand in a given product category, HDs, on the other hand, determine brand success based on the attractiveness

(22)

16

of the entire product category in store. Therefore, for HDs it is vital that the introduction of NBs results in an overall increase in product category attractiveness, which results in trading up to premium products and new shoppers for the HD leading to category expansion

(Deleerysnyder et al., 2007).

Shoppers value variety in assortments because preferences are for stores that offer what they need and the larger the assortments the greater the chance of finding what they are looking for (Hoch, Bradlow & Wansink, 1999). Therefore the breadth and depth of the retailer's assortment is an important element of the overall store image of the retailer because for variety-seeking consumers this will provide greater utility (Ailawadi & Keller, 2004). Given the drivers of NB success previously mentioned, adding NBs which are priced higher than PLs, to an HD assortment will most likely increase the overall product category

perceptions of quality and variety increasing the general category appeal (Dhar, Hoch & Kumar, 2001) especially in an assortment which is dominated by PLs. Furthermore, the increased diversification of the assortments will cater to the heterogeneous preferences of consumers, leading to an increase in the category sales by attracting additional shoppers to the store looking for NB offerings (Ailawadi & Harlam, 2004;Ailawadi, Neslin, & Gedenk, 2001; Deleeysnyder et al., 2007). Hence this could lead to less competition between PLs and NBs in the HD because it will attract different consumer segments (Deleeysnyder et al., 2007). In addition, by adding NB offerings at the HD this could potentially help the HD increase share of wallet of its loyal customers by providing NBs in product categories for which they would have otherwise visited TSs, hence tapping into new markets (Lourenço & Gijsbrechts, 2013). On the other hand, if introductions of NBs are highly comparable to existing product offerings of PLs they are less likely to increase the overall appeal of the category seeing as consumers then see no differences between the NB and PL offerings. Therefore, one could argue the bigger the quality and price differences between the product

(23)

17

introductions the higher the likelihood that the overall attractiveness of the category will increase leading to higher overall sales in the product categories.

According to Jacoby and Mazursky (1984), strong NBs are able to improve the overall consumer store image. In turn, this increase in store perception may have positive spillover effects on the HDs PLs and hence lead to overall higher ratings and sales for the HD’s PLs (Simmons, Bickart and Buchanan, 2000). Similarly, Ailawadi, Harlam, César & Trounce (2006) illustrate that through the use of promotions a store can realize a positive net impact on the overall sales in a given store. In other words, promotions draw consumers to store who then purchase other products that are not on promotion leading to an overall increase in sales, referred to as a “halo” effect. Ailawadi et al. (2006) find a positive influence of promotions on the overall net sales of the store and in this case, it is proposed that NBs can draw consumers to the store who then purchase other products in addition to merely purchasing the NBs. Provided with these arguments, it is likely that the incremental sales from expanding consumer share of wallet, attracting new shoppers as well as the “Halo effect” of NBs will offset the cost of cannibalization. Hence the following proposed hypotheses will help determine whether the total sales of the HD have increased and secondly whether the NBs have increased the total sales in which the product categories have been introduced. These insights will help answer the second research question; focusing on the extent to which the sales of the HD in which the NB has been introduced, change.

H2a: There is a positive relationship between the introduction of NBs at the HD and the total sales of the HD after the NB introduction.

H2b: There is a positive relationship between the introduction of NBs and the sales of the respective product categories of the HD.

(24)

18

Retailer pricing formats

According to Inman, Shankar & Ferraro (2004) and Bell, Chiang and Padmanabhan (1999) consumers relate particular product categories to distinctive retail formats like TSs and HDs as well as to different quality tiers. The pricing strategy of a retailer is one of the key components of how consumers choose which retailer to visit. According to Gijsbrechts, Campo and Nisol (2008) shoppers purchase at different stores for two main reasons; product category differences and overall pricing strategy of the retailer. Despite the growing market share of HDs, they are not likely to be interchangeable with TSs entirely seeing as HDs predominantly sell their own brands and overall have smaller or no NB offerings (Vroegrijk, Gijsbrechts & Campo, 2013; Deleersnyder, et al., 2007). Additionally, to date, NBs are a core traffic driver for consumers for TSs, combined with their frequent price promotions (Cleeren, et al., 2010) and deal-prone consumers this is unlikely to decrease patronage completely at TSs. Also, according to Martenson (2007), there is a positive influence on store loyalty as a result of customer satisfaction resulted from a high choice variety of NBs. Multiple-store purchasing by consumers has branded the grocery retailing industry the past decade especially as a result of deal-prone consumers (Gijsbrechts, Campo & Nisol, 2008). Purchasing at both HDs as well as TSs allows consumers to save costs in some product categories while

‘splurging’ in higher quality product categories dominated by NBs where “choice variety and brand equity play more important roles” (Vroegrijk, Gijsbrechts & Campo, 2013, p609). While Deleeyrsnyder et al. (2007) indicate that the effort of visiting more stores is

compensated by the acquisition utility, Fox & Sethurman (2006) on the other hand, suggest consumers prefer to shop at one store at a time in order to be efficient. Hence, if NBs are introduced in HDs this could increase the acquisition utility of the HD in question and

decrease the likelihood of needing to visit multiple stores, seeing as consumers would be able to splurge in the categories in which NBs are available. This would indicate that consumers

(25)

19

prefer to visit one store that provides a broad enough assortment to cater to their needs as opposed to having to visit numerous stores.

There are two principal pricing strategies in the retailing industry, including EveryDay Low Price (EDLP) and High-Low promotional pricing (HiLo) where most retailers are

generally not one or the other yet employ a strategy either leaning towards EDLP or HiLo (Ailawadi & Keller, 2004). EDLP retailing strategies include retailers that charge an overall constant lower price and have little to no price promotions, while HiLo retailing strategies have higher prices but implement frequent promotions priced below EDLP constant prices (Ailawadi & Keller, 2004; Hoch, Dreze & Purk, 1994). According to Ho, Tang and Bell (1998), the reason that EDLP and HiLo retailing strategies are both dominant is because both attract different types of shoppers. HiLo outlets are generally visited by consumers who visit more frequently but have smaller purchasing baskets whereas EDLP outlets generally have less frequent visits but bigger basket sizes, this is as a result of the overall pricing strategies (Ailawadi & Keller, 2004). Seeing as HDs are more price focused in comparison to HiLo TSs, consumers will anticipate lower constant NB prices at HDs (Ailawadi & Keller, 2004).

Hence, HDs employ a retailing strategy comparable to that of EDLP as opposed to a HiLo strategy seeing as they use little to no price promotions and have constant lower prices (Steenkamp, Van Heerde, Geyskens, 2010). As a result, this can act as a motivating factor to switch stores to the HDs as they now offer lower prices for similar products offered at TSs (Deleerysnyder, 2007).

According to Dhar and Hoch (1997), the market share of PLs depends on the retailing strategy of a particular retailing outlet, meaning that PL overall quality perception can differ according to what type of retailing strategy the retailer employs. HiLo retailing strategies are associated with high-quality given their overall higher priced products, in turn, this acts as a quality cue to consumers signaling an overall high-quality perception of the retailer’s PLs. On

(26)

20

the other hand, EDLP retailers are associated with lower quality given their overall lower prices. This is important to note seeing as by adding NBs to HDs, it is necessary to determine what effects this could have on other market incumbents. When NBs are introduced at HDs they become more comparable to both HiLo and EDLP TSs meaning that consumers who previously did not visit HDs because of their lacking NBs in their assortments might do so now, given the traffic driving capabilities of NBs (van Heede & Geyskens, 2010).

Furthermore, the retailing strategy of the TS may have an impact on the extent to which they are affected by the introduction of NBs at the HD. This is because TSs that have a HiLo strategy, which is associated to overall higher quality and have broader assortments are able to cater to a wider consumer base in comparison to both EDLP retailers and HDs. As a result, it is expected that the introduction of NBs at the HD will have little effect on HiLo retailer outlets seeing as their PL line is associated to higher quality products in comparison to EDPL outlets and offer broad and deep assortments. As a result, the following hypothesis is

proposed:

H3a: There is a small negative relationship between the introduction of NBs at HDs and the sales at traditional supermarkets that employ a HiLo retailing strategy.

On the other hand, retailers operating under an EDLP retailing strategy operate with lower average prices and hence have an overall lower perception of their PLs (Dhar and Hoch, 1997). As a result, it is expected that TSs that operate with an EDLP strategy will be more negatively influenced by the introduction of NBs at HDs seeing as they are more comparable to HDs yet employ slightly higher prices and have a lower quality perception of their PLs, hence driving consumers to HDs.

H3b: There is a strong negative relationship between the introduction of NBs at HDs and the sales at traditional supermarkets that employ an EDLP retailing strategy.

(27)

21

Moderating factor of product category characteristics

As previously mentioned, it is possible that different NBs have differing effects on the sales for both the NB manufacturer and the HD as a result of introducing NBs in product categories in which consumers see little to no differences with existing product offerings of PLs. In addition, NBs function as a means of risk reduction and social demonstrance for consumers; however, one can argue that while overall NBs may be preferable, it can depend on the given product category in question depending on the level of risk or social

demonstance that is linked to the product category overall. Therefore this chapter highlights that while NBs may be preferable overall, the product category characteristics of a given product category may act as a moderating factor in the extent to which NBs are preferred.

Previous research highlights the way in which consumer factors have had an effect on the success of PL penetration in varying product categories. According to Vroegrijk,

Gijsbrechts, & Campo (2016), the effect of introducing a budget brand differs not only per consumer segment but also per product category. As illustrated by Steenkamp & Kumar (2009), one of the misconceptions of HDs is that they are merely patronized by consumers with a low income yet, in reality, higher income segments also frequently visit HDs, because HDs offer high-quality products for a low price. Therefore, the main goal is to determine whether NBs are more or less successful in different product categories as a result of the level of differentiation. Koschate-Fischer Cramer & Hoyer (2014) illustrate how the degree of differentiation can influence the link between PL brand share and store loyalty. While one could argue that the majority of FMCG purchases are based on habit, consumers’ need for variety may offset habitual purchase behavior. Furthermore, Vroegrijk, Gijsbrechts & Campo (2016) and Ailawadi & Keller (2004) illustrate that the impact of budget lines will differ per product category demonstrated by noteworthy differences in PL shares across different product categories. In this section, the extent to which product category characteristics can

(28)

22

influence NB preference is elaborated on, which may contribute to the findings of both Vroegrijk, Gijsbrechts & Campo (2016) and Ailawadi & Keller (2004) in illustrating that the impact of NBs may also differ depending on the product category.

Level of differentiation of product categories

According to Steiner (1993), non-differentiated products are those where consumers can or do not differentiate greatly between different brands and the extent to which consumers see products in the product category as equal. Research indicates consumers have a higher willingness to pay for NBs when the perceived quality and degree of differentiation is higher for NBs than for PLs in the given product category (Koschate-Fischer, Cramer & Hoyer, 2014). This also indicates, however, that the less differentiated a product is, the less perceived risk is associated with the brand, in which case relatively more expensive NBs are not always preferred (Steenkamp, Van Heerde, and Geyskens 2010). In addition, Vroegrijk, Gijsbrechts & Campo (2016) also illustrate that PLs will be most successful in less differentiated product categories when the difference in quality and risks are minimal. Seeing as NB manufacturers mainly excel in differentiated product categories, one could argue whether it would be wise for a NB manufacturer to introduce brands into product categories where the overall product differentiation is low. The main function of a brand is to reduce functional and social risks; however, if there are none of such risks, to begin with, an NB manufacturer should

contemplate whether it is worth introducing branded products into those categories. According to Sethuraman & Gielens (2014), when there are little to no quality differences between products, PLs will have a higher market share compared to NBs; consequently, consumers will purchase PLs when the perceived product has better value than NBs.

Furthermore, perceived risk and social risk are the predominant barriers of PL proliferation, when the perceived quality of a product is questionable this then leads to greater uncertainty of whether lower priced PLs are of acceptable quality (Sethuraman & Gielens, 2014). NBs

(29)

23

have higher perceived quality of their products by consumers (Ter Braak, Geyskens, &

Dekimpe, 2014), therefore, when the perceived quality in a given category is undifferentiated, consumers are more motivated to find cheaper alternatives. Hence to answer the fourth

research question and determine whether product category characteristics, and more

specifically level of differentiation of a product category can moderate the sales of NBs the following hypothesis is proposed:

H4: Differentiated product categories have a greater positive moderating effect on the respective category sales after the introduction of an NB in the respective category in comparison to undifferentiated product categories.

Conceptual Framework

The conceptual framework shown below illustrate the hypotheses formulated in the previous sections of the literature review. The first model illustrates the introduction of NB at the HD (independent variable) influence on the sales for the respective manufacturer, HD and TSs (dependent variables). The second model illustrates how the relationship between the introduction of NB at the HD (independent variable) and the HD category sales of the product category in which an NB has been introduced (dependent variable) is moderated by the level of differentiation of product categories (moderator).

Main effect of introducing NBs at HD and effect on total sales for NB manufacturer, HDs and TSs

(30)

24

Moderating effect of differentiation in product categories on the category sales at the HD

Research Design

This section includes the methodological approach in order to address the research questions and hypotheses to determine how the introduction of an NB affects NB

manufacturer sales, HD sales as well as the sales of TSs. The total sales of grocery retailers in the Netherlands are considered in order to answer the main research question. The following section includes a description and reasoning behind the methodological approaches taken, description of the variables as well as the research model.

Methodological approach

This research is conducted using a secondary data source consisting of historical sales data exported from A.C. Nielsen. Retail scanner data is often used in research concerning market share and sales and is essentially historical checkout scanner data, providing information on product sales (Rossi, DeLurgio, & Kantor, 2000). The nielson scanner data will be acquired and analyzed in order to attain information regarding the sales of NBs before and after the introduction of NBs at the HD in different retail outlets in the Netherlands. The dataset entails a natural experiment of the introduction of 21 Unilever products at the HD; Aldi, in the Netherlands in order to examine the hypotheses. Aldi and Lidl are the two HDs in the Netherlands and while Lidl has gained more market share than Aldi since 2013, Aldi accounts for 57% of all HD outlets in the Netherlands with 519 stores at the end of 2015

(31)

25

(Euromonitor, 2016c).

The data set includes 40 weeks of data, spanning January 2016 to November 2016 and includes sales data from Aldi, Albert Heijn, Jumbo and Superunie. While Aldi represents the HD in the data set, Albert Heijn, Jumbo and Superunie represent the TSs. Superunie;

however, includes data from all smaller retailers aggregated in the Netherlands. This is because individual level data for all retailers in Superunie has not been made available; however, aggregated data including all retailers within Superunie is. The data set includes all product category sales in which Unilever introduced brands at Aldi as well as brand-level data on a weekly basis for all retailers. This provides a comprehensive overview into the changes in market share over a 40-week period. The specific product brand introductions were chosen because all were introduced within approximately five weeks of each other as indicated in table 3 and all products are from the same manufacturer and differ across different product categories.

Retail scanner data analysis

Variables

The purpose of this analysis is to determine whether an introduction of an NB at an HD (independent variable) will have an effect on the sales (dependent variable),

communicated in terms of total euro revenue of the brand, of the retailer, the manufacturer as well as the remaining TSs. The independent variable includes the introductions of the NBs before and after the introduction at the HD. Therefore the sales before and after the

introduction of the 21 NBs are measured. One product was introduced in week 25 of 2016, whereas the majority (13 out of 21 introductions) were introduced in week 26 of 2016. The remaining seven brand introductions at Aldi vary between week 27 and week 31. The dependent variable is the sales of the manufacturer for the introduced brands as well as sales

(32)

26

for the respective 21 brand introductions which are measured for all market incumbents as well as the product categories to which they belong. The effect of introducing NBs at an HD can be measured in several ways including for example measuring total volume sold;

however, this does not control for price differences between the retailers. To clarify, when comparing two retailers; retailer A may have sold 100 products for 1 euro, whereas retailer B may have sold 80 products for 1,50 euro. While at first glance retailer A may have had better sales, retailer B attained higher value for the product. Therefore the dependent variable will be measured in euro sales, which is price multiplied by volume, as opposed to volume. The moderating variable is categorical in this natural experiment and includes the level of differentiation of category in which the 21 brand introductions were introduced. Table 2 provides an overview of all variables used in the analysis as well as their notations and operationalization.

Validation of classification into differentiated and undifferentiated product categories In order to determine whether the brand introductions were in differentiated or undifferentiated product categories, a pretest survey was conducted of 44 Dutch consumers. The pretest survey was composed of 2 constructs adopted from Koschate-Fischer Cramer & Hoyer (2014) representing level of differentiation. The overview of constructs can be seen in table 2 and a preview of the survey can be seen in appendix 2. The survey asked participants to answer the questions based on a 7-point Likert scale ranging from “strongly disagree” to “strongly agree”. Furthermore, a means test further validated the results presented in table 1. A reliability test was conducted to determine the internal consistency of the constructs, which resulted in a Cronbach’s alpha of 0,86, indicating that the two constructs demonstrate high internal consistency.

(33)

27

Table 1 Classification into differentiated and undifferentiated product categories Validation according to

pretest survey

Average rating of perceived level of differentiation of product category *

Take home Differentiated 2,23

Deodorant Differentiated 1,82

Fabric Cleaning Undifferentiated 5,09

Fabric Conditioning Undifferentiated 5,09

Household cleaning Undifferentiated 5,45

Mayo Undifferentiated 4,64

Meals & Dishes Undifferentiated 4,51

Other miscellaneous Differentiated 2,77

RTD Tea Differentiated 2,00

Sauces Undifferentiated 4,57

Yellow Fats Undifferentiated 5,11

Impulse Differentiated 2,23

Skin Cleansing Differentiated 1,80

Soups Undifferentiated 3,84

Wash & Care Differentiated 2,34

* Respondents were asked a series of questions (Overall, I see no major difference between brand name products and store’s own brands in category x” & “Overall, brand name products and store’s own brands are equivalent in category x”) as seen in table 2 to determine their level of differentiation for the given product categories. The scale ranges from “7 = strongly agree to 1 = strongly disagree”. The average rating of perceived level of differentiation has therefore been translated to represent the following rating: 1 = high differentiation – 7 = low differentiation

Control variables

A number of control variables are included in the analysis, as the sales of a product can be influenced by a number of different factors. The subsequent control variables have been included in the analysis as influencing the sales.

• The average price of a brand: this continuous variable includes the average prices of the brands across retailers per week. The price is important as it influences consumers purchasing decisions.

• Week: this continuous variable represents the weeks in which the sales were recorded for all brands in the data set and controls for sales growth per week.

(34)

28

Table 2: Variable operationalization

VARIABLE NOTATION OPERATIONALIZATION SOURCE

Dependent variable

Total sales LnTotalUnileverSales t, b

LnAldiSalest, b

LnlTSSales t, b

Total logged sales for Unilever expressed in euros, purchased at weekt in brands, b.

Total logged sales for Aldi expressed in euros, purchased at weekt in brands, b.

Total logged sales for all TSs expressed in euros, purchased at weekt in brands, b.

Nielsen database

Independent variable

NB brand introductions Dummy_Introt, b,

Dummy_takehome t

Dummy_deodorant t

Dummy_impulse t

Dummy_fabriccleaning t

Dummy_fabricconditioning t

Dummy_meals & dishes t

Dummy_other t

Dummy_householdcleaning t

Dummy_skin cleansing t

Dummy_wash & care t

Dummy_yellowfats t Dummy_soups t Dummy_sauces t Dummy_rtd t Dummy_brand X t Dummy_AHt, b Dummy_Superuniet, b Dummy_Jumbot,b

Dummy indicator, equals 1 when Aldi carries an NB in week t for brand b, otherwise 0.

Dummy indicator, equals 1 when Aldi carries take home category in week t, otherwise 0.

Dummy indicator, equals 1 when Aldi carries deodorant category in week t, otherwise 0.

Dummy indicator, equals 1 when Aldi carries impulse category in week t, otherwise 0.

Dummy indicator, equals 1 when Aldi carries fabriccleaning category in week t, otherwise 0. Dummy indicator, equals 1 when Aldi carries fabricconditioning category in week t, otherwise 0. Dummy indicator, equals 1 when Aldi carries meals&dishes category in week t, otherwise 0. Dummy indicator, equals 1 when Aldi carries other miscellaneous foods category in week t, otherwise 0. Dummy indicator, equals 1 when Aldi carries householdcleaning category in week t, otherwise 0. Dummy indicator, equals 1 when Aldi carries skin cleansing category in week t, otherwise 0.

Dummy indicator, equals 1 when Aldi carries wash & care category in week t, otherwise 0.

Dummy indicator, equals 1 when Aldi carries yellow fats category in week t, otherwise 0.

Dummy indicator, equals 1 when Aldi carries soups category in week t, otherwise 0.

Dummy indicator, equals 1 when Aldi carries sauces category in week t, otherwise 0.

Dummy indicator, equals 1 when Aldi carries RTD category in week t, otherwise 0.

Dummy indicator, equals 1 when Aldi carries brand X in week t, otherwise 0.

Dummy indicator, equals 1 for all sales made for Albert Heijn in week t in brand b, otherwise 0.

Dummy indicator, equals 1 for all sales made by Superunie in week t in brand b, otherwise 0.

Dummy indicator, equals 1 for all sales made by Jumbo in week t in brand b, otherwise 0.

Moderating variable

Degree of

differentiationa Dummy_Differentiated t, b Dummy indicator, equals 1 when the product category _{is differentiated in week t in brand b, otherwise 0.}

“Overall, I see no major difference between brand name products and store’s own brands in category x” & “Overall, brand name products and store’s own brands are equivalent in category x” (Variable mean-centered across categories, STD. ! = 0,86) Koschate-Fischer Cramer & Hoyer (2014) Control variables Price Weekly trend LnUnileverPrice t, b LnAldiPrice t, b LnTSPrice t, b Week t, b

Average logged price for all Unilever brands sold in entire market expressed in Euros, purchased in week t in per brand b.

Average logged price for all available brands at Aldi expressed in Euros, purchased in week t in per brand b. Average logged price for all brands at traditional supermarkets expressed in Euros, purchased in week t in per brand b.

Week indicator

Nielsen database

NOTE: a_{Measured on 7 point Likert scale (1= “strongly disagree” and 7 = “strongly agree” )}

(35)

29

Research Model

For this research application, the base model applied for testing the hypothesis is an ordinary least squared (OLS) regression model where multiple predictor variables are entered into the model. OLS regression allows one to estimate the dependent variable, in the research model: value sales, by reducing the difference between observed responses and predicted responses leading to a model better fitting to the data (Hayes and Cai, 2007). The additive model shown below includes logarithms for the sales values as well as price indicators. The subsequent model is the base model used for testing the hypotheses; however, these are adjusted according to the hypotheses and elaborated on in the respective subsections. As previously mentioned, given the not normally distributed pattern of the data the data is

transformed, Ln indicates the logarithm. In order to determine the specific effects on the sales for all market incumbents, the sales data alters to reflect the hypothesis and the sales are considered, in a given week (t) for a specific brand (b).

LnSales !,! = α + LnPrice!,! ∗ β!+ week !,! ∗ β!+ dummy_intro!,! ∗ β!

+ dummy_productcategory_!,∗ β_!+ ε

Data and Operationalization

Descriptive statistics

The Nielsen scanner data includes sales data from 21 different brand introductions at a HD, and contains data of two TSs including Albert Heijn and Jumbo and the purchasing organization Superunie. This research takes 40 weeks in consideration in total, with

approximately 25 – 30 weeks prior to NB introduction at the HD and 10 – 15 weeks after the introductions as indicated in table 3.

Table 4 and 5 illustrate the brand and product category market share change for the manufacturer and the HD before and after the introduction of the NB at the HD. Two periods

(36)

30

are measured after the introduction seeing as this takes into consideration possible time that consumers take to become aware of the introduction of NBs at the HD. In order to take into consideration possible weekly variances in market share as a result of marketing activities on behalf of the NB manufacturer including promotions for example, the average market share was taken for the period before and after the introduction. Meaning the market share indicated includes the average of 5 weeks before, 5 weeks after and the 5 weeks preceding that. Table 5 illustrates the market shares by brand level of the manufacturer in the total market before and after the introduction of the NB at the HD while table 4 illustrates the market shares of the HD by product category level. For the manufacturer, the brand share in the total market increased after the introduction at the HD in 17 of the 21 brand introductions. Furthermore, table 5 illustrates the effect the NB introductions have had on the entire product category at the HD. All product categories experienced an overall increase in total market share except the product category other miscellaneous, skin cleansing and soup, illustrating the initial positive effects of the implementation of NBs at the HD. Figure 1 illustrates an example timeline of a NB introduction in the product category RTD, where after the introduction at the HD, there were overall 1,2% higher

sales in comparison to before including all four retailers. On the other hand,

excluding the HD sales the total sales are 0,66% lower than before the introduction. It is important to note while considering the raw data

Figure 1: Timeline of NB introduction at HD

(37)

31

that the manufacturer registers a positive increase in market share, it is important to discuss this data with restraint seeing as they have not been corrected for endogeneity. In order to assess the true effect of the introduction of the NBs, it is necessary to apply a statistical analysis followed in the subsequent chapter.

Table 3: Product category details

*Please refer to appendix 1 for a full overview of product category definitions.

Table 4: Brand descriptives: market share (%) in total market before and after NB introduction for manufacturer

Product category Brand 1 month before intro % 5 weeks after intro % 10 weeks after intro % Change ∆ in market share % Take home Magnum 16,7 14,8 16,9+ _0,2+

Deodorant Rexona 11,1 11,5 12,5+ _1,4+

Fabric cleaning Robijn 23,3 22,1 25,9+ _2,6+

Fabric conditioning Robijn 50,1 44,9 49,6- _0,5-

Household cleaning Cif 3,3 2,6 3,7+ _0,4+

Household cleaning Andy 2,9 2,5 3,8+ _0,9+

Household cleaning Glorix 12,0 14,3 14,4+ _2,4+

Mayonnaise Calve 19,0 20,3 20,7+ _1,7+

Meals & dishes Knorr 61,4 62,2 66,1+ _4,7+

Other miscellaneous Unox 24,5 29,7 28,7+ _4,2+

Ready-to-drink Lipton 48,6 52,3 49,9+ _1,3+

Sauces Bertolli 6,9 8,9 9,4+ _2,5+

Take home Cornetto 4,5 5,6 6,0+ _1,5+

Yellow fat Blueband 16,7 16,8 17,0+ _0,3+

Take home Viennetta 1,0 1,3 1,6+ _0,6+

Yellow fat Becel 18,8 19,3 19,4+ _0,6+

Impulse Hertog 52,4 52,7 45,8- _6,6-

Skin cleansing Axe 4,3 4,5 4,7+ _0,4+

Skin cleansing Dove 8,9 8,8 8,1- _0,8-

Soup Unox 43,6 45,4 44,7+ _0,9+

Introducing national brands at hard discounters in the Netherlands : the extent to which National Brand manufacturers in the FMCG grocery industry can introduce National Brands at Hard Discounters to increase sales, whi

Master Thesis