The theoretical framework provides mechanisms that have been put forward to explain the impact of shelf position, the number of unique private label/national brand stock keeping units, the number of private label tiers, and the price gap on the second grocery choice (i.e. private label or national brand). Given that the first choice (i.e. private label or national brand) is out of stock.
§2.1 Choice Grocery Product
Consumers have two options to choose a grocery product. They can choose a national brand or a private label (Kumar & Steenkamp, 2007). The definition of national brands (hereinafter referred to as NB), as it is given in for example Hoch (1996), will be maintained in this study. This means that NBs are brands marketed in a national market. NBs are owned, produced, and promoted by (large) manufacturers. Consistent with the private label literature, in this study, private labels (hereinafter referred to as PL) refer to brands owned, sold, and distributed by retailers (e.g. Lincoln & Thomassen, 2008). Other commonly used terms include store brand, own label, and retailer brand.
2.2.1 Growing Trend Private Labels
PLs are growing faster than NBs (e.g. Kumar & Steenkamp, 2007). Kumar & Steenkamp (2007) explored that PL share of total consumer packaged goods will increase with 50 percent in Western Europe (i.e. from 20 PL share to 30 PL share). This expected PL share growth from 2000 to 2010 is in line with the data which are established by Nielsen in November 2014 . Strictly speaking, PL is the biggest hurdle that manufacturers should overcome to gain market share and power. PL is not necessarily a threat, but it is a challenge and an opportunity for manufacturers (Kumar & Steenkamp 2007).
The power of PL is immense nowadays (e.g. Kumar & Steenkamp, 2007; Lincoln & Thomassen, 2008). According to Thomassen et al. (2006), the age of retailer power changed from 1970 to today in their advantage. Every consumer buys PLs in all key European countries, which is in line with data from ACNielsen . In addition, according to the same database from ACNielsen, PL is no longer a niche market. Although it is true that PLs are more popular among lower-income households, higher-income households are not far behind in terms of market share of PL across revenue levels (respectively 32% and 28%, according to Lincoln & Thomassen, 2008). Furthermore, according to the PL / NB comparison research of ACNielsen , people increasingly do like PLs. One of the PL success drivers of that development is the increased perceived
quality of PLs (e.g. Kumar & Steenkamp, 2007; Thomassen et al., 2006). PL has become therefore a real brand in the perception of the consumer (e.g. Lincoln et al., 2008). According to Kumar & Steenkamp (2007), PLs are increasingly imbued with emotion and imagery nowadays, which met for example the emotional satisfaction where consumers look for when choosing a brand. In other words, PLs virtually offer the same product features as NBs do, and hence serve the needs of the consumers (e.g. Kumar & Steenkamp, 2007).
Manufacturers can use several strategies to arrest PL success, including innovation, advertising, and price promotions (e.g. Steenkamp & Geyskens, 2013). On the other hand, retailers have several strategies to increase PL success. In addition to those strategies, retailers have full control of different in-store factors (Thomassen et al., 2006). Retailers could directly influence the shelf position of the products, the number of unique PL/NB stock keeping units, the number of PL tiers, and the price gap between PL and NB. This refers to the unbalanced power between manufacturers and retailers.
2.2.2 Impact of Out of Stock
Many studies investigated consumers’ responses to “out of stock”, which is a regular phenomenon in grocery stores. Those studies (e.g. Avlijas et al., 2015; Campo et al., 2000; Peckham, 1963; Verbeke et al., 1998) all approximately reached the same conclusion with regard to the different main out of stock consumer reactions. This study will consider two behavioral responses. The first is brand switch, which means that the consumers switch from NB to PL or the other way around. This study calls such consumers “switchers”. The second behavioral response is item switch, which suggest that the consumers choose from the same brand (i.e. PL or NB) another item the second time when the first choice is out of stock. This study calls such consumers “stayers”. Both behavioral responses indicate that the consumer purchased a substitute (Verhoef & Sloot, 2006). In addition, if the first choice is out of stock, then the consumer will not restart from the first stage of the consumer buying decision process when the consumer is considering a second choice (Emmelhainz et al., 1991). The consumers will start again from evaluating the alternatives based on the product first preferred (e.g. Emmelhainz et al., 1991; Verhoef & Sloot, 2006). It is therefore likely that the consumers’ second choice will be based on the same extrinsic cues as the first choice (e.g. Sloot et al., 2005; Richardson et al., 1994).
The fact that every consumer (low and high income households) buys PLs frequently (no niche market anymore) has changed the PL perceptions (Kumar & Steenkamp). These changed perceptions towards PL and the increased PL perceived quality perception have resulted in an increased likelihood to buy a PL (e.g. Kumar & Steenkamp, 2007; Thomassen et al., 2006; Lincoln & Thomassen, 2008). It is therefore likely there will be more NB switchers (i.e. NB to PL), due to the fact that the consumers know that PL is a good alternative of the first choice (which is out of stock) after re-evaluating the alternatives. On the other hand, it is likely that there will be more PL stayers, due to the fact that those stayers already have a positive attitude towards PLs (Kumar & Steenkamp, 2007).
So, if the first choice (NB or PL) is out of stock it is more likely that the second choice will be PL as well. From this the next hypotheses can be draw up:
H1: The PL choice probability the second time will be higher than that of NB, given that the first PL choice is out of stock.
H1a: The PL choice probability the second time will be higher than that of NB, given that the first NB choice is out of stock.
§2.2 Shelf Position
Determining the shelf position of products is part of shelf space management. Shelf space management is the process of allocating products on the shelf in the best possible arrangement (Frank & Massy, 1970). It is a limited resource that has a high impact on retailer’s performance (e.g. Amrouche & Zaccour, 2006). According to Amrouche & Zaccour (2006): “shelf space is one of the retailer’s most important assets. Moreover, manufacturers are rewarded with better shelf positions of their brands when they offer better services to the retailer (Mangold & Faulds, 1993). As discussed before, consumers are more inclined to buy PL (e.g. Kumar & Steenkamp, 2007; Thomassen et al., 2006). This results in improved shelf positions of store brands (e.g. Chandon et al., 2002; Nogales & Suarez, 2005).
2.3.1 Impact Shelf Position
For retailers, the key of placing the products on the shelf strategically is to increase product visibility (e.g. Chandon et al., 2002; Amrouche & Zaccour, 2006; Nogales & Suarez, 2005). The strategy of the retailer to increase visibility is to locate PLs near NBs. The main reason for that consideration is that PLs can take advantage of NBs, given the important impact of NB advertising on consumers’ perceptions (e.g. Thomassen et al., 2006; Nenycz‐Thiel & Romaniuk, 2014). This importance is demonstrated by the fact that non-users of PLs are less likely to know anything about a store brand than non-users of national brands about a NB (Nenycz‐Thiel & Romaniuk, 2014). So, advertising creates a sense of preference, but it is not that easy to convert that preference into sale(s) (e.g. Thomassen et al., 2006). It is therefore that retailers take advantage of the out-of-store advertising campaigns of manufacturers. Those campaigns create knowledge about a specific national brand in the memory of non-users (Nenycz‐Thiel & Romaniuk, 2014). In addition, it creates top of mind awareness of the brand (Thomassen et al., 2006). Both effects of the out-of-store communication stimulate the fact that consumers will search for that specific NB in the store (Steenkamp et al., 2010). The probability that the consumer will compare the advertised NB with PL will increase when PL is positioned
close to the advertised NB (e.g. Kumar & Steenkamp, 2007; Nenycz‐Thiel & Romaniuk, 2014). Moreover, the purchase decision is based, in particular, on extrinsic cues (Richardson, Dick, & Jain, 1994). The PL choice probability will increase if consumers compare the extrinsic cues of NB with PL (Kumar & Steenkamp, 2007). The comparison of NB with PL based on extrinsic cues will increase if PL has a prominent position on shelf (e.g. Nenycz‐Thiel & Romaniuk, 2014; Chandon et al., 2002). That means that PL should be positioned close to NB (Kumar & Steenkamp, 2007).
The quality of the PL determines which PL and how close it is positioned to NBs (Amrouche & Zaccour, 2006). Low quality PLs are placed far away from NBs. The opposite applies to high quality PLs, which are located close to NBs (e.g. Amrouche & Zaccour, 2006). In addition, PLs are located to the right of the (leading) NB more often, since 90% of the people are right-handed (Hoch, 1996) and the left side is a low traffic region on the shelf (Chandon et al., 2002).
Altogether, retailers allocate PL near NB to increase PL visibility. This increased visibility stimulates the comparison between NB and PL, which will increase the probability that PL is chosen. It does not depend on the consumers’ first choice (i.e. PL or NB). So, regardless of the first choice, the probability of choosing PL the second time (given that the first choice is out of stock) will increase if PL is positioned close to the first choice. As a result, comparison between NB and PL will increase. From this point of view, the second hypotheses can be formulated as:
H2: The PL choice probability the second time will increase when PL is positioned close (in the same meter) to the first choice, given that the first PL choice is out of stock.
H2a: The PL choice probability the second time will increase when PL is positioned close (in the same meter) to the first choice, given that the first NB choice is out of stock.
§2.3 Product Assortment
The concept of product assortment refers to the product assortment planning by the retailer. This study will consider one aspect of product assortment planning in particular (hereinafter, this study refers to this as PAP). This study will investigate the aspect of depth. In this context, depth means the number of different products within a specific grocery category (e.g. Mantrala et al., 2009). Depth includes therefore the number of unique PL / NB stock keeping units (hereinafter, this study refers to this as SKU), and the number of PL tiers (e.g. Mantrala et al., 2009; Geyskens et al., 2010).
2.4.1 Impact number of Unique Stock Keeping Units
When conducting PAP, retailers should become aware of the fact that consumers have different perceptions and preferences (Mantrala et al., 2009; Green and Krieger 1985). There should therefore be enough assortment to ensure that consumers’ first choice is available. In saying this, the retailers must take account of the fact that, according to Mantrala et al. (2009): “Large assortments can sometimes frustrate or overwhelm the consumer”. In addition, in many situations, the preferences of consumers are constructed when they are faced with a specific purchase decision (e.g. Bettman, Luce, and Payne, 1998). This occurs most commonly when consumers have entered the retail store (Simonson, 1999). It is therefore that consumer perceptions and preferences may change over time (McAlister & Pessemier, 1982).
Given that the consumers’ preference for NB or PL is constructed in the store, the visibility of the brand is an important factor. It is therefore that enhanced brand visibility will increase the likelihood of choosing that specific brand and hence of choosing that brand the second time, given that the first choice is out of stock (e.g. Simonson, 1999; Kumar & Steenkamp, 2007). This is due to the fact that consumers will start again from evaluating the alternatives based on the first preferred product (e.g. Emmelhainz et al., 1991; Verhoef & Sloot, 2006). The visibility may improve by increasing the unique SKUs of that specific brand (e.g. Mantrala et al., 2009).
Despite the fact that consumers could have the feeling that their choice is being pushed towards the PLs of the retailer (Kumar & Steenkamp, 2007), the PL visibility will increase with the number of unique PL SKUs (just like with NB). However, that feeling may influence the likelihood of choosing PL the second time (Kumar & Steenkamp, 2007). It is therefore that the first preferred choice (i.e. PL or NB) will strengthen or weaken the impact of the number of unique SKUs on the PL choice probability (e.g. Emmelhainz et al., 1991). From this can be argued that the number of unique PL SKUs will increase the PL choice probability the second time, but this effect will be stronger when the first PL choice is out of stock than when the first NB choice is out of stock. This is due to the fact that consumers who have chosen the 1st time PL do like PL and therefore the higher visibility on the shelf (e.g. Emmelhainz et al., 1991). The opposite is true for the number of unique NB SKUs. That means that the number of unique NB SKUs will decrease the PL choice probability the second time, but this effect will be weaker when the first PL choice is out of stock than when the first NB choice is out of stock. This is due to the fact that that consumers who have chosen the 1st time NB do like NB and therefore the higher visibility on the shelf (e.g. Emmelhainz et al., 1991).
Altogether, the effect of product assortment on preferences can be increased by improving the visibility. This could be achieved by increasing the unique SKUs of the specific brand. It is therefore that the number of unique SKUs will increase the probability that a specific brand is chosen (given that the first choice is out of stock). Hence, the choice probability of choosing a specific brand will decrease if the unique SKUs of
the other brand increase. In addition, the first choice will strengthen or weaken the impact of the number of unique SKUs. This leads to the following hypotheses which can be tested:
H3: The negative impact of the number of unique NB SKUs on the PL choice probability the second time will be weaker when the first PL choice is out of stock than when the first NB choice is out of stock.
H3a: The positive impact of the number of unique PL SKUs on the PL choice probability the second time will be stronger when the first PL choice is out of stock than when the first NB choice is out of stock.
2.4.2 Private Label Tiers
According to Geyskens et al. (2010), there is a three-tiered PL program. This so-called program follows a good, better, best approach, like for example the PL strategy of Ahold; economy (i.e. AH basic), (AH) standard, and premium (i.e. AH excellent). In other words, retailers have expanded their PL lines in order to better serve the consumer needs (e.g. Ailawadi and Keller, 2004; Nogales & Suarez, 2005). However, retailers should be aware of the effect that consumers may be confused by the different quality levels of one brand (Geyskens et al., 2010). This could therefore decrease the PL choice probability the second time.
2.4.3 Impact Private label tiers
There are two main effects of introducing different PL tiers (e.g. Geyskens et al., 2010; Simonson & Tversky, 1992). The theoretical arguments of these effects will be based on the compromise effect and similarity effect (Simonson, 1989; Huber et al., 1982). Within the context of the compromise effect, the introduction of an economy PL will increase the choice probability of choosing the standard PL, because it will become a compromise option (Simonson, 1989). This is also related to the fact that, according to Simonson (1999): “consumers tend to avoid the cheapest option in the set of options presented to them”. Consumers do not wish to be seen by others as a person who buys cheap products (Simonson, 1989). From this can be argued that consumers’ choices depend on how they will be evaluated by others (Simonson, 1989). In addition, the compromise effect is also related to the fact that the perceived difference between the standard and premium PL decreases if an economy private label is introduced (Parducci, 1974). As a result, the choice probability of choosing the standard PL will increase.
On the other hand, within the context of the similarity effect, the introduction of a new product will decrease the probability choice of the product similar to it (Simonson, 1989). It is therefore that the introduction of a premium PL will decrease the probability choice of NB. Because the premium PL will become a similar product to NB (Simonson, 1989). From this can be argued that the probability choice of the new introduced premium PL will increase. In addition, related to the similarity effect, the choice probability of choosing the standard PL will also increase, due to the fact that it will become much more unique than the
similar PL premium and NB products (Simonson, 1989). The standard PL alternative stands out from the similar premium PL and NB. In saying this, the retailers must take account of the fact that too many different PL quality levels increase the number of unique PL SKUs. This could lead to the consumers’ feeling that their choices are being pushed towards the PLs of the retailer (Kumar & Steenkamp, 2007).
In all, regardless of introducing an economy or premium PL, the likelihood of choosing PL (in total) will increase with the numbers of PL tiers (e.g. Geyskens et al., 2010). This also related to the fact that the visibility of PL will increase with the number of PL lines (e.g. Simonson, 1999; Kumar & Steenkamp, 2007). It does not depend on the first choice of the consumer (i.e. PL or NB). Hence, the probability of choosing PL the second time will increase with the number of PL tiers (given that the first choice is out of stock). Based on these arguments the following hypothesis is proposed:
H4: The PL choice probability the second time will increase with the number of PL tiers, given that the first PL choice is out of stock.
H4a: The PL choice probability the second time will increase with the number of PL tiers, given that the first NB choice is out of stock.
§2.4 Price Gap
The optimal price gap between store and national brands is hard to define. This is due to the fact that there is no optimal price gap, since it depends on the role of the PLs which is defined by the retailer (e.g. Richardson et al., 1994). Both a high and low price gap between store and national brands can be part of a specific retail strategy (e.g. Nogales & Suarez, 2005; Hoch et al., 1998). This refers to the power of the retailer (Thomassen et al., 2006).
2.5.1 Impact Price Gap between Store and National Brands
According to Richardson et al. (1994): “consumers rely on extrinsic cues when assessing product quality.” Given that price is an extrinsic cue, it can be argued that price will be an important factor in the buying decision of a grocery product (Nogales & Suarez, 2005). The factor price is also an important factor to compare PLs with NBs (Kumar & Steenkamp, 2007).
The price gap between premium PL and NBs is low (Geyskens et al., 2010). If there is a small price gap, then it is more likely that the retailer uses PL as a key differentiator (Hoch et al., 1998). The aim of premium PLs is therefore to compete directly with the leading national brands (Kumar & Steenkamp, 2007).
On the other hand, there is a large price gap between economy/standard PLs and NBs (Geyskens et al., 2010). If there is a high price gap, then it is likely that consumers will perceive lower priced PLs as “lower
quality for lower prices” (Richardson et al., 1994). From this perspective, the retailer may use a “value for money approach” (e.g. Richardson et al., 1994). From this can be argued that economy / standard PLs will be perceived as products which are the counterpart of NB. It is therefore that NB may perceived as expensive (Steenkamp et al., 2010). This is in line within the context of the prospect theory, because the losses of choosing NB over PL is higher than gains (Tversky & Kahneman, 1981). Consumers may have the feeling that they do not get value for money by choosing NB over PL (e.g. Steenkamp et al., 2010; Sethuraman & Cole, 1997). The consumers may therefore choose PL over NB. In addition, there is room for improvement of increasing willingness to pay NB over PL in the Netherlands. This is related to the fact that the Dutch people have the lowest score on willingness to pay NB over PL (Steenkamp et al., 2010). From this can be argued that Dutch consumers are more price sensitive. The effect of price gap will therefore be weaker when the first NB choice is out of stock than when the first PL choice is out of stock. This is due to the fact that consumers are less price sensitive when they are familiar with NB (e.g. Kumar & Steenkamp, 2007; Steenkamp et al., 2010). In this context, the consumers who have chosen NB the first time are familiar with the brand (e.g. Emmelhainz et al., 1991; Verhoef & Sloot, 2006, see chapter 2.2.2).
In view of the fact that price is used as an effective tool to increase the PL choice probability, manufacturers can decide to decrease the price gap between NB and PL (Kumar & Steenkamp, 2007). However, price reduction is not always the optimal solution, since price reduction is connected to the perceived value of the NB (Kumar & Steenkamp, 2007). The effectiveness of the price drop is determined by the fact that the price drop should put the NB within the price-value preference range of PL consumers (Kumar & Steenkamp, 2007, figure 1). From this can be argued that a relatively small price cut may already have a significant effect (D to D’, see figure 1). It can also be argued that in specific situations (regardless of the size of the reduction in the price) there may be no significant effect at all (A to A’, see figure 1).
Figure 1, Price Gap Management. Y axis: perceived value, X axis: price. A, B, C, and D represent manufacturer brands; PL is for private label. Source: Kumar and Steenkamp, 2007)
Altogether, the price gap plays an important role on the perceived product perceptions of consumers (e.g. Richardson et al., 1994). These perceptions may influence the preference of choosing PL or NB (e.g Geyskens et al., 2010). It is consistent with the context of prospect theory that the likelihood of choosing PL the second time will increase with the price gap between store and national brands (e.g. Tversky & Kahneman, 1981). So, the probability of choosing PL the second time will decrease with the price gap between the 1st and 2nd choice (given that the first choice is out of stock), given that NBs are more expensive than PLs (
Kumar & Steenkamp, 2007). In other words, if the price gap increase between the 1st and 2nd choice it is more likely that a NB is chosen the second time (second choice is more expensive than the first choice). The effect of price gap will be stronger for consumers familiar with NB. Hence, consumers who have chosen NB the first time. This is due to the fact those consumer are already aware of the more expensive NBs and therefore it does not matter for them if the second choice is more expensive than the first choice. From this point of view, the second hypothesis can be formulated as: