National Brands vs. Private Labels The exceptions to the rule

National Brands vs. Private Labels

The exceptions to the rule

Master Thesis, MSc Marketing Management University of Groningen

Faculty of Economics and Business Department of Marketing

11 July, 2012

Tim Buiting

E

XECUTIVE SUMMARY

It is a more or less accepted notion that once the economy turns sour, consumers tend to switch to the usually cheaper private label (PL) causing the market share of national brands (NBs) in

supermarkets to plummet. Remarkably though, all research done on this topic grouped the entire market of fast moving consumer goods (FMCG) into two categories: PLs and NBs. It is very well possible that within one of these groups there are brands that show the opposite behaviour, but they would not show due the size of the aggregate group. So it is possible that there are brands that display a pro-cyclical behaviour, meaning they move alongside with the business cycle instead of counter-cyclical behaviour. Identifying which brands would be these ‘exceptions to the rule’ is essential as it may have a profound effect on how marketing budgets are (re-)allocated when being slashed or reviewed due to economic contraction periods.

This paper’s literature review identified two brand characteristics that are essential if a brand is to grow during times of economic contractions. The first and most important characteristic is that of hedonic value. A consumer might seek to substitute a high level expense with a lower level expense, e.g. dining out for dining in. Consumers would subsequently seek brands that offer a hedonic experience that now will be missed due to this substitution behaviour. In addition consumers will, more than during economic high-tides, seek products that offer them security and simplicity in choice-making. Consumers seek to limit the (financial-) risk and therefore look for products that are worth the money. When there is a high perceived quality differential, a brand is assumed to be superior in quality than its counterparts which would limit the performance risk involved.

12 brands from three product categories were examined by using the baseline sales (corrected for promotion-effect) from these brands over a period of five years. Subsequently it was examined whether these are somehow affected by fluctuations in the business cycle which is represented by the consumer confidence index (CCI). After attempting to reaffirm the general accepted notion of a pro-cyclical behaving NB sales and counter-cyclical behaving PL sales in hypothesis 1, the moderating effects of hedonic value and perceived quality differential are tested in hypothesis 2. To objectively determine the value for hedonic value and perceived quality differential per brand a questionnaire was designed. A high Cronbach Alphs led to the combining of these two values into an average combined value for each brand.

If a significant relation can be found in future research, one major impact would be that of the allocation of the marketing budget. As marketing budgets are usually under pressure during

economic contractions, these findings could have an impact of the (re-) allocation of funds. First, the brands that grow during an economic contraction need less support and can therefore be labelled as cash cows, allowing for marketing budgets to be re-allocated to brands that really need it. Secondly, extra support may be given to brands that have not yet attained the hedonic value and/or perceived quality difference that would allow it to grow during economic contractions.

PREFACE

The origins of this paper may be found at H.J. Heinz the Netherlands where I held an internship position from August 2011 to February 2012. Within the marketing department I was part of the drinks team which coordinated the brands Roosvicee and Karvan Cévitam. Being part of such a team is a great and educational experience. This internship was also meant as part of the finalization of my Master degree in Business Administration. It laid the foundation for my thesis which will be the last challenge before the actual finalization of my degree.

It goes without saying that I owe much gratitude to H.J. Heinz and my former colleagues not only because of a great internship, but also for making available the necessary data for my thesis. I would like to express special thanks to Dorien Dijkstra, brand manager Roosvicee, who supported me during and after my internship. Without Dorien, this thesis would not have been possible.

Nathalie van Herk is the brand manager for the brands Heinz (sandwich spread), Brinta and De Ruijter. Her desk was located near mine and it is because of this that I overheard a conversation in which she mentioned how surprising it was that a premium brand as De Ruijter doesn’t seem to drop in sales due to a recession. Without knowing what the context of this remark was it turned out to be the inspiration for my thesis. Therefore I thank Mrs. Van Herk for being, be it unintentionally, a source of inspiration.

My time at H. J. Heinz showed me how much knowledge is present in such a large organization and how it is used to better serve the consumer and the market as a whole. However it also showed me how much marketing-driven firms like H.J. Heinz crave for more, new knowledge and insights. Marketing research is therefore a never ending knowledge generating process of which I am glad to have been a part of.

I thank all who have supported and guided me in the past months, especially Dr. Karel-Jan Alsem for his excellent guidance and support.

Tim Buiting

CONTENT

2.1 National Brands during economic contractions ... 8

2.1.1 The PL phenomenon... 8

2.1.2 The perceived quality differential ... 9

2.2 The business cycle and its effect on NBs and PLs ... 10

2.2.1 General effects on business and consumption ... 10

2.2.2 NB share and the business cycle ... 11

2.2.3 Business cycle effect on traditional consumer segments ... 12

2.3 The exception to the rule: countercyclical movement of NBs ... 12

2.3.1 Hedonic values and substitution behaviour ... 12

2.3.2 Consumer Purchasing Process: Simplification and risk-averse behaviour ... 14

2.4 Problem statement... 15 2.5 Hypotheses ... 16 3 Methodology ... 17 3.1 Variables ... 17 3.1.1 Dependent variable ... 17 3.1.2 Independent variable ... 18

3.1.3 Moderator 1: Hedonic value ... 19

3.1.4 Moderator 2: perceived quality differential ... 19

3.2 Preparing data ... 20

3.3 Data Collection ... 21

3.3.1 Dependent Variable... 21

3.3.2 Independent Variable - CCI... 22

3.4 Research plan ... 23

4 Results ... 23

4.1 Descriptive data ... 23

4.1.1 Dependent variables visualized ... 23

4.1.2 Seasonality adjustments ... 25

4.1.3 Qualitative evaluation focal brands dependent variable ... 25

4.1.4 Independent variable - Consumer Confidence index ... 28

4.2 Hypothesis 1 ... 28

4.3 Hypothesis 2 ... 29

5 Conclusion & Discussion ... 30

5.1 General results ... 30

5.2 Discussion ... 31

5.3 Factors influencing results... 31

5.4 Implications ... 33

5.5 Suggestions for further research & limitations ... 33

6 References ... 35

7 Appendix 1 ... 40

1 Introduction

1.1 Research background

The brand manager of Koninklijke De Ruijter once mentioned: “…some consumers still seek

indulgence when their disposable income shrinks, but they seek it in different forms. Many of them seek it in the form of the luxury brand such as De Ruijter.” What she meant was that when consumers can’t afford, or have the feeling they can’t afford, large luxury expenditures anymore they seem to seek it in relatively smaller purchases such a chocolate sprinkles. In this case, De Ruijter may be considered to be the top premium brand in the category. Where consumers switch to the cheaper private label for most of their purchases, they reward themselves by doing a small purchase of a more premium brand. This could explain why for De Ruijter market shares actually don’t look all that bad even though we are in one of the most severe economic crises of our time.

The observation made by the brand manager of De Ruijter may be at first perceived as rather peculiar as common sense directs us to think that during an economic slowdown premium brand such as De Ruijter all but increase in sales. One would expect cheaper brands and store brands, or private labels (PLs), to gain significant market share at the cost of more expensive national brands. However the observation also makes one wonder whether this ‘conventional wisdom’ of declining national brand (NBs) sales as compared to cheaper private label is really an absolute truth, or if maybe some exceptions apply. This paper will attempt to answer this question.

1.2 Relevance

The relevance of such an investigation is found in the implications of the above suggested

conventional wisdom of declining NB share and increasing PL share. Much literature is available on how to combat PL (e.g. Quelch & Harding, 1996; Kumar, 2006) and on NBs and how to make them survive an economic slowdown (Ang et al, 2000). These authors all build their investigation upon the notion that during economic slowdowns PL share goes up and NB share goes down. For example, Ang et al. (2000) provide a list of possible strategies that business could employ to survive an economic crisis; e.g. marketing mix strategies or pricing strategies. In addition, Lamey et al. (2012) argue that major innovations are needed to combat PL growth and that the sharp decline in promotional activities during economic contractions manufacturers of NBs inflict themselves wounds they may never recover from.

It is therefore relevant for both manufacturers of NBs and retailers to consider those brands and adapt their strategies accordingly.

1.3 Paper build-up

In this paper the above notion will be examined by first examining the literature on the topic of PLs and its relation to NBs. After this, theory on business cycles in relation to the topic of NBs and/or PL will be explored. This will lay the foundation for the problem statement and the subsequent research which will lead to a conclusion which will be formulated in the final chapter. The conclusions will have implications for (marketing-) management which will be discussed in the respective chapter.

2 Theoretical framework

In this section the different concepts are discussed after which the relevant theories will be outlaid. First the relevant literature on NBs and PLs will be addressed after which literature will shed some light on the business cycle and its effect on NBs. Subsequently it will be discussed what

characteristics a brand should have in order to have the potential to grow when the economy contracts. The chapter will end by stating the problem statement and a conceptual framework.

2.1 National Brands during economic contractions

As described in chapter 1.2 there is a general consensus concerning the effect of an economic contraction on the market shares of NBs and PLs. Lamey et al. (2007) confirm the earlier researched notion of flourishing PL market shares during economic contractions. Their research built upon earlier work from Quelch and Harding (1996), Ang et al.(2000), Hoch and Banerji (1993) and Baxter and King (1999) who all investigated the same phenomenon. So as Lamey et al. (2007) have confirmed, the market share over PL tends to increase during economic contractions and decrease again during economic expansion periods; for NBs an opposite trend is noticeable. In addition to confirming the earlier research, Lamey et al. (2007) also concluded that not all market share that was lost to PLs during contraction periods is returned to NBs during expansion periods suggesting an increase in PL popularity over time.

2.1.1 The PL phenomenon

It should be mentioned that Lamey et al (2007) define PLs as being generally low priced items despite the fact that some PLs have a sub-range actually aiming for the more expensive segments such as Albert Heijn’s AH Excellent or Coop’s Coop Superieur. Hoch (1996) argued that the traditional view of PLs as being inferior quality products is no longer accurate. Quality differences decrease and PLs are present in all sorts of market segments. However Hoch (1996), as well as the before mentioned literature, confirm that a consumer’s financial situation is still of significant influence on PL sales. The significant increase in PL sales when disposable income shrinks suggests that the consumer abides by the more traditional view of PLs; that of a low-cost option. Even though Deleersnyder et al (2007) confirm that discounters increasingly try to find other ways than price to differentiate themselves, this investigation will maintain the view of PLs as the low-cost option as this still seems the

predominant view used my many authors (Lamey et al., 2007; 2012).

2.1.2 The perceived quality differential

Lamey et al. (2007) observed that as consumers tend to be reluctant to switch back to NBs when the economy is expanding after they had switched to PL during an economic contraction which creates a permanent increase in market share for PLs. A possible cause for this constant increase could be the increased effort retailers put into their PL programs (Hoch, 1996) and due to the increase in quality of PL brands (Steenkamp and Dekimpe, 1997).

The above would suggest that certain macro-economic factors negatively influence the perceptual quality differential during an economic contraction as it was mentioned earlier that after consumers have switched to PL due to economic down-turn, they are reluctant to switch back (Lamey et al., 2007).

Quelch and Harding (1996) argue that NB are capable of prevailing over PLs, however NBs need dedicated management and attention to succeed. The authors state that one should not forget that brands allow consumers to simplify the purchasing process as brands have the potential to assure a consumer of its quality and hereby limiting the perceived risk.

But from a consumer perspective there are also some factors that moderate the extent to which consumers prefer PL over NB. One of the most extensively researched factors is income level. Akbay (2005) observed that in the context of healthy versus less healthy products, higher income, and higher educated consumers, consumers tend to buy not only healthier products but also more premium products. More general, Sethuraman and Cole (1999) found that consumers with a higher education and/or income are less sensitive to prices and more willing to pay the price premium of NBs. This could suggest that more wealthy consumers are more likely to stay loyal to their favourite NB even when the economy contracts.

2.2 The business cycle and its effect on NBs and PLs

The business cycle is a phenomenon usually referred to when people speak of the economic trend and cycle of a country or region. Mascarenhas and Aaker (1989) refer to this phenomenon as “…economic activity fluctuations, whose periodicity and amplitude may not be predictable, controllable, uniform, or unique, that often involve changing prices and changing relative prices among inputs and outputs”. Currently the world is experiencing a rather deep economic crisis which has its effect on business and every-day life.

In this paper, the movement with or against the business cycle is referred to as the co-movement of a variable. This term suggests a relationship between the variable in question and the business cycle variable chosen. If a variable would move in the opposite direction from the business cycle one would speak of counter-cyclical behaviour. For the opposite behaviour, a variable moving along with the business cycle, the term ‘pro-cyclical behaviour’ is used.

2.2.1 General effects on business and consumption

co-move with the business cycle are innovation and prices. Innovation activities behave pro-cyclical just like advertisement spending (Axarloglou, 2003) where prices clearly behave counter-cyclical meaning that prices increase during difficult times and drop when the economy is flourishing (Chevalier and Scharfstein, 1996). Rising prices of NBs during a period where disposable income shrinks might significantly increase the perceived risk of NBs (Sethuraman and Cole, 1999). Consumers may demand higher quality as the perceived quality differential is shrinking, which in other words would mean consumers need the quality differential to grow in order to accept the price premium. Otherwise the risk perception will grow beyond acceptable levels (Grewel et al., 1994).

One important effect of a negative business cycle is that consumers tend to become more critical. Estelami et al. (2001) argue that consumers become more price sensitive which can have significant effects as the price gap between PL and NB is expected to grow in economic contraction periods. And together with a decreasing quality gap between PL and NB it can have a significant effect on the Willingness-To-Pay (WTP) for the price premium of NB (Steenkamp et al., 2010).

2.2.2 NB share and the business cycle

When purchasing power decreases during economic downturn, the consumers’ willingness and ability to purchase also decreases and consumers tend to postpone purchases and reduce the quantity bought. However, Shama (1981) argues that the situation for Consumer Packaged Goods (CPG) is slightly different because of their more necessary nature and therefore people are more prone to capitalize on their spending; they appear to focus more on lower prices. This, and other incentives, laid the foundation for Lamey et al.’s (2007) research on the link between PL market share and economic state; or, as Lamey et al. (2007) describe it, the counter-cyclical movement of the PL share with the aggregate business cycle.

2.2.3 Business cycle effect on traditional consumer segments

The first challenge of a marketer of NBs during economic contraction periods is to understand that the market is more volatile than ever before and that new customer segments develop themselves (Quelch and Jocz, 2009). Where previously demographics or lifestyle might have been a most relevant segmentation variable, the emotional response of consumers to the recession might be more relevant now (see table 1 for the new segments according to Quelch and Jocz).

Besides having to reassess one’s view on market segments, recession usually also has a negative effect on maintaining and supporting brands (Lamey et al., 2012; Quelch and Jocz, 2009). Both articles strongly urge marketers to reconsider slashing the marketing and communication budget during economic contractions. Lamey et al.(2012) argue that it is sound brand-supporting

advertisement that can limit the market share erosion and the permanent loss of market share to PLs. Quelch and Jocz (2009) go as far as by saying that “On average, increases in marketing spending during a recession have boosted financial performance throughout the year following a recession”.

2.3 The exception to the rule: countercyclical movement of NBs

No research has yet been done on the counter-cyclical movement behaviour of certain types of CPG within the PL or NB categories. It is therefore that many signs that would suggest the possibility of a counter-cyclical moving phenomenon within the NBs category come from either different areas of business or non-academic or practitioner’s journals. However, a broader search perspective allows for more academic evidence to present itself.

2.3.1 Hedonic values and substitution behaviour

In-depth research on possible exceptions to the general notion of PLs prevailing over NBs during an economic contraction is scarce. However there are signs that suggest that there are some exceptions

Table 1: New consumer segments during a recession as defined by Quelch and Jocz (2009)

Slam-on-the-brakes Pained-but-patient Comfortably-Well-Off Live-for-Today

This segment feels vulnerable and hardest hit by the recession. This group reduces all types of spending by eliminating, postponing, and substituting purchases

This segment tends to be resilient and optimistic about the long-term but less confident about the short-term. They economize in all areas, though less aggressively

This segment feels secure about their ability to ride the recession out. They consumer near-prerecession levels, but they have become a bit more selective about their purchases.

to the rule. Besides the remarks made by the brand manager of De Ruijter there are other sources from other areas of business that observe the same trend. Taylor (2003) mentioned as an example that during the 1990-1991 recession in the United States ice cream sales dropped, however the premium brand Haagen-Dazs saw its sales increase rapidly. Although the actual reasons for such a trend in this particular case are beyond the scope of this paper, Taylor suggests that consumers seek enjoyment in experiencing the same as those who are more fortunate. Another explanation is provided by White-Sax (2009) who commented in an article on skyrocketing luxury chocolate-brand sales and argued that consumers simply seek those small sweet indulgences that they can still afford, and by Reyneke et al. (2012) who reported a high demand in fine wines making this sector seem nearly recession proof. These reports from the field all have in common that they involve premium brands all of which provide a certain emotional benefit rather than just a utilitarian benefit.

Quelch and Jocz (2009) provide academic evidence by stating that consumers may substitute a purchase in a high category (treats; indulgences whose immediate purchase is justifiable) with a comparable purchase in a lower category (essentials: survival, central to well-being). In this framework the higher in the ranking, the more expendable the purchase becomes (Postponables: purchases that can be put off; Expendables: purchases seen as unnecessary and unjustifiable). As an example, one can think of going out to dinner (treat) to be substituted by cooking at home

(essential). And as dining out moves up the ranking it is very likely that buying premium brands for dinner becomes the treat and that dining out is not an option anymore. In addition one could argue that when large expenditures which used to provide the consumer with sense of luxury or treating oneself are being postponed or put off all together, these hedonic experiences are being sought in smaller purchases such as premium ice cream (Taylor, 2003).

The above would suggest that there is a possibility for a growing demand for brands that would allow oneself to treat oneself occasionally. This is supported by non-academic sources such as White-Sax (2009) as stated earlier.

To support the above notion, one can view the example of the dining out being substituted by cooking at home from a consumer psychology point of view. Once cooking becomes a treat, one might view the purchase of PLs as too risky as one is not sure whether this PL will be able to deliver the hedonic value sought after in a treat (Richins, 1994). Richins continues by saying that

desired consumption pleasure and performance. In practise this means that PLs are less likely than NBs to offer the desired acceptable performance risk (Sethuraman & Cole, 1999). Important to mention is that the term used in this context is ‘value’, and not ‘quality’. Value is obviously similar to quality, however it is higher-level concept which differs from quality in two ways: 1) it is more individualistic and personal than quality; 2) it is a ‘give-and-get’ concept where there is a trade-off involved (Zeithaml, 1988). This latter point-of-difference is also defined by some as the ‘emotional pay-off’ (Young & Feigen, 1975) which relates well to the concept of hedonic value. In line with the above stated theory one could argue that brands that particularly focus on hedonic values have the potential to grow during times when consumers seek simplicity and security in their purchase behaviour (Flatters and Willmott, 2009; Qualch and Jocz, 2009).

2.3.2 Consumer Purchasing Process: Simplification and risk-averse behaviour

One of the reasons for the positive approach of literature towards advertisement spending in relation to sales during an economic contraction period could be that consumers seek simplicity (Flatters and Willmott, 2009; Qualch and Jocz, 2009). Consumers seek insurance that the choices made are the right ones, and strong brands are able to offer that. Take the iPod for instance, this product was introduced in during an economic contraction period in 2001 in the high-end segment, yet it was the embodiment of simplicity at the time. This product was just one of many extremely successful introductions during economic low-tides (Matheson and Staba, 2010). Simplicity may be defined in several ways, but both the articles define simplicity as a degree of choice-making; consumers seek to simplify the choice-making process. This theory would suggest that consumers seeking simplicity would seek purchases that would minimize (financial-) risk (Dunn, 1986). One important way of doing that is by creating a sufficient perceived quality differential (Sethuraman and Cole, 1999). A national brand with a significant perceived quality differential is being perceived as superior in quality to other brands (usually PL) in the particular category. In addition the same authors mention that it is an accepted and proven notion that if this perceived quality is superior to the opposing brand (or PL) consumers are willing to pay a higher premium for this brand. This is different from the performance risk discussed earlier as the motive behind the risk is not hedonic but more utilitarian, “Is the product’s quality worth the price premium?” (Red.). In short this would mean that brands able to provide a significant better perceived quality than its counterparts should be able to ensure the consumer of its performance and thereby limiting the risk involved; hence simplifying the purchasing process.

This line of reasoning would suggest that strong brands who manage to generate a significant perceived quality differential would be capable of simplifying the consumer’s purchasing process by limiting risk. Hence the potential for strong brands to maintain or gain market share during an economic contraction

2.4 Problem statement

As the literature review demonstrates there is non-academic as well as academic evidence that suggests that a counter-cyclical movement of NBs is possible. There is quite some literature available on how to combat PLs (e.g.: Flatters and Willmott, 2009; Lamey et al., 2007; Quelch and Jocz, 2009; Kumar, 2006) which would suggest that in theory it is possible for NBs to be successful in periods of economic contraction.

To summarize the literature review one can clearly state that there are definitely possibilities for exceptions to the rule meaning that NBs actually grow during times of economic contractions. The literature suggests that the main two underlying causes for this could be the hedonic value that brands may have and the perceived quality differential that is present within a category. It is therefore that the main research questions reads as follows:

To what extend do can NBs high in hedonic value and perceived quality differential move countercyclical (in the opposite direction) to the aggregate business cycle?

Figure 1: Conceptual Framework

2.5 Hypotheses

The purpose of this chapter is to formulate clear hypotheses that would allow for this investigation to commence in an orderly fashion. The literature review and the results from the commenced

questionnaire which objectively determined the degree of hedonic value and perceived quality differential produced the below stated hypotheses. The hypotheses are visualized in the conceptual model in figure 1.

Based upon a vast amount of literature (Lamey et al., 2007; Quelch and Harding, 1996; Ang et al., 2000; Hoch and Banerji, 1993; Baxter and King, 1999) one can formulate a clear expectation of how PL sales would behave. The literature clearly states that PL sales tend to behave counter-cyclical and NB sales pro-cyclical. To test whether the dataset used in this investigation abides by this notion, the following hypothesis may be formulated:

H1: The base sales are affected significantly different for NBs as opposed to PLs by the CCI. As mentioned before, it is essential for this investigation that it encompasses categories of products and brands that provide the consumer with a clear and salient choice. One of the most important differentiating factors, price, will be discussed further in chapter 3.1.1, but obviously a brand positioning is more than just price. There is a vast positioning difference between brands in the categories chosen for this investigation all attempting to create a perceived quality difference as opposed to PLs or to generate a significant hedonic value. Some category brands build upon a strong supra-categorical brand, such as Heinz, which is active in many categories like ´warm wet sauces´; ´cold wet sauces´ and ´canned food´. Heinz is an example of a brand that derives its brand

associations from their core business (ketchup) and extended their brand from there into other categories. Another often used positioning strategy is that of ‘country of origin’ (Keller, 2008) such as used by Bertolli and Grand’Italia. If these attempts to differentiate have the desired effect and a brand has managed to have one of the highest scores on hedonic value and perceived quality differential, the question becomes whether this would allow for a brand to grow during times of declining consumer confidence.

3 Methodology

As little research is available on a brand-level, it remains speculation as to what brands would qualify to be the exception to the rule by growing during times of economic contraction. It is therefore that a questionnaire is designed to objectively determine the degree of hedonic value and perceived quality differential. This chapter will go into more detail about this questionnaire and all subsequent methods used to eventually come to a sound conclusion.

3.1 Variables

Both the independent variable (IV) and dependent variable (DP) are gathered from available data and do not require primary research. These variables will be discussed next.

3.1.1 Dependent variable

There are several options that were considered as dependent variable. One option was to take market share, which was what Lamey et al. (2007; 2012) did in their investigation of the relationship between PL & NB market share and the business cycle. However, the research done on this topic grouped all supermarket sales together in a PL-group and a NB-group. This allows for a good comparison of the variation in market share of PLs and NBs in general. In this way, the mutation of total supermarket sales is not of influence. It only matters whether this growth in total supermarket spending is spent on PLs or NBs.

However, as this investigation will go into more detail and will not group NBs together one cannot simply take market shares for several reasons. First, an overall increase in total supermarket sales does not have to show in market shares. Theoretically the increase in total supermarket sales can be evenly spread over all brands which would leave market shares unchanged despite a sales growth. Using sales data instead of market shares allows for much smaller mutations to be noticed. It is therefore that it was chosen to use sales data from brands in selected categories in Dutch quality supermarkets as a dependent variable. The latter meaning that discounter such as Lidl and Aldi will not be included. This is due to the fact that the data collector (Information Resource or IRI) does not keep track of these figures and therefore is not able to provide this particular data. To control for the fact that the category in which the brand is situated might have a moderating effect, these are incorporated in the regression analysis as dummy variables.

Purifying sales data

The Fast Moving Consumer Goods (FMCG) sector is characterised by its swift and tangible marketing efforts. Most new product introductions are supported by extensive (TV-) campaigns and brand teams are constantly thinking of new introductions or brand extensions. However, marketing is only part of the engine that drives the FMCG companies. The other part is the sales department who are responsible for the actual sale of the brands to the supermarkets. Where a marketing department has a budget at their disposal to design and run campaigns and such, the sales department has a budget available to fund promotional activities. These activities may be defined as activities conducted to increase sales, also called merchandising (Little, 1998). Merchandising activities very often include price-cuts (e.g. 3 for 2, 50% off). These activities would temporarily increase sales (drastically) and do not necessarily have a long-lasting effect. As this can seriously distort the raw data, the raw data must be corrected for this irregular impurity. To do this, IRI separates sales that would have occurred without these merchandising activities and sales that occurred due to these merchandising activities, baseline sales and incremental sales respectively. Data collection companies such as IRI or A.C. Nielson Company take data from weeks without merchandising and generate a smooth line through sales over time, correcting for several market effects (little, 1998). Baseline sales are widely used in evaluations and assessments as they provide a more realistic perspective on sales growth or decline.

As the investigation is mainly interested in purchase behaviour related to business cycles, incidental fluctuations due to merchandising is not desirable. Therefore it is chosen to use baseline sales numbers as a dependent variable.

3.1.2 Independent variable

The independent variable is a measure that somehow represents the business cycle in The Netherlands. There are plenty macro-economic measurements to choose from, such as GDP and unemployment rates. However, as this paper examines consumer spending in supermarkets it is important to stay close to the consumer and to select a measurement that closely reflects the consumer’s experience of the business cycle. As common macro-economic measurements tend to treat the consumer as a rational being (Becker, 1962) it could be more useful to select a

It must be noted though that there are some doubts about its predictive capabilities, but that is beyond the scope of this paper. So, as the consumer confidence index is a good representation of the state of the economy and its cycle, and because it is an index that remains close to the consumer and their sentiment, this measurement is chosen.

3.1.3 Moderator 1: Hedonic value

As not just quality, but also a brand’s ability to provide the consumer with a degree of pleasure it is important that a brand possesses a certain degree of hedonic value. As mentioned before, a high level of hedonic value would limit the performance risk of the product.

To measure the performance risk the 9-point hedonic value scale as developed by David R. Peryam (1957) is used in a questionnaire. Even though this scale was developed several decades ago, it is still widely used and holds well in tests with comparable tests (Hein et al, 2008). Respondents were asked in a questionnaire to indicate their opinion on a brand by selection a value between the extremes of ‘extremely like’ or ‘extremely dislike’.

3.1.4 Moderator 2: perceived quality differential

The perceived quality differential allows for brands to be identified that possess the added qualitative value to limit performance risks. Looking at the price-quality inference would say something about how the brand’s price holds against its perceived quality, but not how it would hold against PLs. As comparing NBs to PLs is central in this investigation it is chosen to use the perceived quality

differential as a moderating factor. The usability of the perceived quality differential is strengthened by the fact that consumers often perceive a higher priced product as being of high quality (Rao & Monroe, 1989). So looking at the perceived quality differential also includes some degree of price-quality inference.

To measure the perceived quality differential respondents were asked to rate a NBs quality as

opposed to its PL counterpart. Per product category respondents were shown a visual representation of PLs and products of the NB and were asked to indicate how they believed the quality of the NB holds against that of the PL by using a 9-point likert scale ranging from ‘much worse quality’ to ‘equal quality’ and eventually to ‘much better quality’.

3.2 Preparing data

To effectively draw conclusions based upon this investigation, it is necessary to take a closer look at the two brand characteristics that are expected to act as moderators in the relationship between the CCI and a brand’s sales. As mentioned earlier, it is hypothesised that brands with a significant level of hedonic value and a significant perceived quality differential should have the potential to display a counter-cyclical behaviour as opposed to a pro-cyclical behaviour. After these brands have been identified a short qualitative evaluation of these brands will follow.

To investigate the moderating effect of these two brand characteristics it is necessary that one identifies those brands within each category which possess a high level of one or both of these characteristics. As mentioned in chapter three this is done by conducting a questionnaire (appendix 2). The results are summarized in table 2.

Table 2: Results survey

Category Brand Average score

‘hedonic value’ Average score ‘Perceived quality differential’ Average score combined Warm wet sauces Heinz 5,8 5,7 5,7

Grand’Italia 6,6 6,3 6,5

Bertolli 6,6 6,4 6,5

Cold wet sauces Heinz 6,6 6,8 6,7

Calvé 6,4 6,2 6,3

Remia 6,4 6,0 6,2

Gouda’s Glorie 5,7 5,7 5,7

Spreadable toppings Venz 6,5 6,3 6,4

De Ruijter 7,2 7,0 7,1

Choca 5,0 5,1 5,0

Fred & Ed 4,6 4,8 4,8

Bolletje 5,8 5,9 5,9

N: 42

3.3 Data Collection

As mentioned earlier, the data collection method used is mining pre-existing databases for the relevant information. Below the different variables will be further elaborated on.

3.3.1 Dependent Variable

As mentioned earlier the source for data for the dependent variable is IRI (IRI, 2011) and is made available by H.J. Heinz. This means that the

supermarket product categories available for investigation are limited to the categories H. J. Heinz is active in, in The

Netherlands. However this does not mean that any category will do. As research shows that the main reason consumers switch to PLs is price, consumers seek low added value in these

products (Guerrero et al., 2000). So if one is to investigate a possible counter-cyclical movement of NBs there has to be a price premium involved which indicates added value to a consumer

(Sethuraman and Cole, 1999). Categories where the price is not only extremely low but also where the price difference between NBs is hardly significant are not suited to be used for this investigation. This is supported by Zeithaml (1988) who stated that in product categories with a low price variation, price is not likely to be seen as a cue for quality or value perception. In appendix 1, several brands within 4 categories are stated with the price per unit and the price per standardized measure (e.g. per litre). A summary can be found in table 4.

Table 3: Categories and brand in Dutch quality supermarkets under investigation in this paper

Category Explanation Brands investigated

Warm wet sauces Pasta sauces from a jar or can, not in powdered form.

 Heinz

 Grand’Italia

 Bertolli

 PL Cold wet sauces Sauces eaten cold,

such as ketchup and mayonnaise  Heinz  Calvé  Remia  Gouda’s Glorie  PL Spreadable topping (bread-) toppings

This overview clearly shows that the ‘fruit drink unchilled’ category is unsuitable for this investigation as the price variation is minimal and the prices of NB lies very close to that of the PL (Albert Heijn store brand). The other 3 categories depicted in table 4 show a clear price differentiating factor between brands which automatically means that there is a significant price difference with the PLs in that particular category.

As this investigation is limited not only to the categories that are relevant and suitable, but also to that which is made available by H.J. Heinz, it will be limited to the three categories depicted in table 3. These categories contain brands which significantly differ in price and seem to have a very different positioning strategy (section 4.1.3) providing the consumer with a true choice.

As mentioned earlier, the data that is provided consists of sales numbers in baseline units sold. The time span for the data is 2007 to 2011 with quarterly data providing the research with 20

observations which will be considered sufficient.

The data will be corrected for seasonal influences as well. If the magnitude of the seasonal

fluctuations seemed to vary little over the time period whether the trend line goes up or down, an additive model is used to correct for seasonality. If the seasonal component is more irregular or, for example, increases in magnitude when the trend line rises a multiplicative model is more

appropriate.

3.3.2 Independent Variable - CCI

The Independent variable will consist of consumer confidence figures from The Netherlands with a corresponding time frame to that of the dependent variable, 2007 to 2011. The data will be

extracted from the Dutch Central Bureau of Statistics (CBS) and will be available in quarterly terms.

The consumer confidence index (CCI) In the Netherlands is compiled, as mentioned before, by the CBS. The CBS surveys households about their own financial situation and the general economic climate which is then translated into a score. The score is calculated by categorizing the interviewees

Table 4: summary of the findings in appendix 1

Category Top Brand Bottom brand Price difference

in %

Fruit based drinks unchilled

Roosvicee Original €0,99 p/l AH Multivitamine drank € 0,79 p/l

25,3%

Warm wet sauces Grand’Italia Toscana pastasaus € 6,77 p/l

Spaghetini Tradizional €2,31 p/l

193%

Cold wet sauces Calvé YoFresh Top Down € 4,16 p/l

Gouda’s Glorie Mayonaise € 2,10 p/l

98%

Spreadable toppings De Ruijter Specials intens

puur €8,08 p/kg

Euroshopper Hagelmix € 2,20 p/kg

into positivists, pessimists and indifferent and consequently subtracting the percentage of pessimists from the optimists (CBS, 2012). This provides the CCI score.

3.4 Research plan

After the brand-specific values for Hedonic value and Perceived quality differential are determined, the first hypothesis will be tested using a linear regression analysis. To control for a possible

moderating effect of categories, dummy variables will be generated that represent the three possible categories. The moderating effect will subsequently be evaluated by using a hierarchal regression analysis.

4 Results

Here one can find the results from the analyses performed. First descriptive results from the dependant variable and independent variable will be stated after which the results from the hypotheses tests will be stated.

4.1 Descriptive data

To illustrate the raw data, first the dependent variables will be displayed in graphs before as well as after seasonal adjustment. After this the independent variable will be visualized in a graph.

4.1.1 Dependent variables visualized

Below one can see the base sales visualized in graphs. The X-axis represents a time period of 5 years divided into 20 quarters.

0 500 1000 1500 2000 2500 3000 3500 4000 4500 Sal es (Tho u sa n ts)

Graph 1: Warm Wet Sauces

HEINZ BERTOLLI

GRAND ITALIA PRIVATE LABEL

0 500 1000 1500 2000 2500 3000 3500 4000 4500 Sal e s (Th o u san ts)

Graph 2: Warm Wet Sauces Seasonally adjusted

SHeinz SBertolli

In the Graphs 1, 3 and 5 one can see the raw data of the baseline sales in all three categories. The data’s time-span is from the first quarter of 2007 to the fourth quarter of 2011. From these descriptive graphs one can see some interesting movements of the base sales over time.

In the first graph on the warm wet sauces, one can clearly see an upward trend in baseline sales of PL and Grand’Italia. These two seem to steadily grow during the observation time.

The descriptive results of cold wet sauces category in graph 3 show a remarkable trend in baseline sales. One can see a clear up-and-down movement of the baseline sales of PL and Calvé.

To a lesser extent, the same trend is visible for Remia and Heinz as well. This graph would even suggest that the higher the sales level, the stronger the up-and-down fluctuation over time.

Interestingly, the sales hit their lowest point every first quarter annually and their highest point in the second quarter. This would suggest that for some reason the consumer tends to buy cold wet sauces

0 2000 4000 6000 8000 10000 Sal e s (Th o u san ts)

Graph 3: Cold wet sauces

CALVE GOUDA S GLORIE

HEINZ REMIA PRIVATE LABEL 0 2000 4000 6000 8000 10000 Sal e s (Th o u san ts)

Graph 4: Cold wet sauces seasonally adjusted SCALVE SGG SHeinz SRemia SPL 0 1000 2000 3000 4000 5000 6000 7000 Sal e s (Th o u san ts)

Graph 5: Spreadable Toppings

DE RUIJTER CHOCA

FRED&ED PRIVATE LABEL

VENZ BOLLETJE 0 1000 2000 3000 4000 5000 6000 7000 Sal e s (Th o u san ts)

Graph 6: Spreadable Toppings seasonally adjusted

SDe Ruijter SChoca SF&E

especially during the months May through July. A possible explanation could be the barbeque season which runs from late spring throughout the summer. The data would have to be corrected for such a seasonal disturbance.

The baseline sales of spreadable toppings as shown in graph 5 don’t seem to show a considerable trend. However, it is very interesting to notice that starting in 2008 the PLs and De Ruijter move in a similar fashion. This is remarkable as PL is the cheapest and De Ruijter the most premium priced brand in the market, yet this seems to make no difference over time.

4.1.2 Seasonality adjustments

Data of a certain time series tend to consist of several components: a trend-, seasonal- and error component. To effectively make use of the data, it is important that the fluctuations caused by the seasonal influence on the data are removed. Therefore, seasonal decomposition is applied to the dataset. Looking at graphs 1, 3 and 5 one can observe that nearly all lines display some seasonal fluctuation. As mentioned in the previous sub-chapter, seasonal influences such as the summer and barbeque periods need to be filtered out to purify the data. Therefore all baseline sales will be seasonally decomposed using the decomposition tool in SPSS.

The seasonally adjusted data can be found in graphs 2, 4 and 6 displayed next to its raw-data constituent. The effects for the ‘warm wet sauces’ category and ‘spreadable toppings’ category are minimal; however the ‘cold wet sauces’ category showed a clear seasonal component in graph 3 which was nicely filtered out (graph 4) leaving a more useful base-sales line.

4.1.3 Qualitative evaluation focal brands dependent variable

Warm Wet Sauces

The category of ‘warm wet sauces’ is dominated by tomato-based sauced predominantly for pasta dishes. This particular category is based for the largest part on the Italian cuisine. It is therefore not

the ‘warm wet sauces’ category. Heinz did not ignore the ‘Italian sauce’ -approach and did

incorporate Italian words on their packaging, but it is not their primary source of credibility as it is for Bertolli and Grand’Italia. Looking at table 2 one could say that in terms of perceived quality

differential and hedonic value; the approach of Grand’Italia and Bertolli seems more effective. Both brands could be considered as a premium and strong brand that cleverly uses the country-of-origin effect (Keller, 2008). Bertolli has extended itself within and outside of the investigated categories positioning themselves as the experts on true Italian cuisine. The strength of the brand is reaffirmed by the nomination of Bertolli for the annual Nima marketing award (Adformatie, 2004) and the study of unequivocal brands in The Netherlands done by the European Institute for Brand Management (EURIB) in 2010 who placed Bertolli on the 26th place of strong food brands. So if it may be assumed that the brand has been able to saliently communicate its values, it could be suggested that this is a brand that is able to offer the consumer a hedonic experience that potentially substitutes dining out.

Cold Wet Sauces

The ‘cold wet sauces’ category is diverse in the sense that there are a high variation in flavours and types of sauces. One thing that catches the eye is the vastly different packaging used by different brands. There is the iconic glass bottle by Heinz, but also metal tubes, plastic or glass jars, top-down plastic bottles and plastic containers (image

2). A brand like Calvé seems to focus on health by offering low-fat alternatives whereas Heinz uses their credibility as a ‘tomato expert’. An observation worth mentioning is that the more geared towards a niche and premium sub-brands such as Calvé YoFresh come in relatively small packaging and the more mainstream and cheaper brands or PL come in large packaging (e.g. Albert Heijn Frites Saus and Gouda’s Glorie).

So the brand that is expected to grow during economic contraction is the brand which offers the consumer not just a high quality product, but also an experience rather than just ketchup or curry. Looking at table 2, one can say that the brand that offers this experience more than others is Heinz. In their communication they take the consumer on a journey from ‘seed to bottle’ and position themselves as the one and the only brand for ketchup which they support with a rich heritage. They use this credibility for extensions throughout the category with different flavours and sauces. This

apparently translates is a significantly higher score on the ‘hedonic value’ and ‘perceived quality’ test.

Spreadable Toppings

The ´spreadable toppings’ category is rather diverse. There are the more general sub-brands, those geared towards a younger audience, and those aimed at chocolate loving adults. This is also a category with a very high price difference (table 4) between the premium brand and the lowest priced brand.

Looking at image 3 one can see a clear difference in positioning. If one would leave price out of consideration, there is still plenty of choice for the different brands and products. The kids’ brands employ ‘fun’ as a positioning strategy by means of their packaging and product itself. Others use their heritage to back-up their credibility as a qualitative brand. Choca is an example of a very functional brand which communicates exactly what it is, and nothing more.

De Ruijter in general communicates more hedonic values than utilitarian values. De Ruijter Specials however, is a sub-brand that deserves extra attention as it is a clear example of a sub-brand which communicates indulgence, pleasure and quality. It comes in a smaller amount than any other (sub-) brand, contains more cacao than its counterparts and its communication campaign aims clearly at positioning it as an indulgence product. Quite possibly the Utz certification displayed front-of-pack serves as a justification tool; “it is Utz certified so it is OK to buy” (red.). However, this remains speculation at this point.

The above observations are supported by the results from the questionnaire. One can see a rather high average score for both the ‘hedonic value’ and ‘perceived quality differential’ test (table 2).

4.1.4 Independent variable - Consumer Confidence index

In graph 7 the consumer confidence index is displayed over the time-span corresponding with the dependent variables; quarterly data from 2007 to 2011.

As one can see in the graph there are clear peaks visible suggesting both positive and negative attitudes of consumers towards the economy.

4.2 Hypothesis 1

To test whether the sales of NBs behave differently from PLs in relation to the CCI a linear regression analysis was performed. The results of this analysis can be found in table 5. The variables used in this regression analysis were:

 Sales (dependant variable)  CCI (independent variable)  NB/PL dummy (NB=1, PL=0)  Interaction term NB/PL (IAVNB/PL)  Category 2 Dummy

 Category 3 Dummy

Table 5: Results linear regression H1

B Sig. CCI -0.116 0.248 NB/PL Dummy -0.406 0.000 IAVNB/PL 0.105 0.336 Category 2 Dummy -0.485 0,000 Category 3 Dummy -0.421 0,000

As one can see in table 5 the NB/PL dummy, category 2 dummy and category 3 dummy variable are statistically significant. This would mean that the average sales level is significantly lower for NBs as it is for PLs. However this is not entirely true as the PL component of this investigation is an aggregate

number over brands meaning that all PLs have been grouped into one. Therefore no conclusions may be drawn from this number.

To control for the effect of the different categories, a hierarchal regression was done by comparing the R2 with and without these moderating variables. The results can be seen in table 6.

Adding the category dummies to the regression analysis (model 2) shows that these do have a moderating effect of the CCI on the base sales as can be seen in table 6 when comparing them to the same model without these moderating variables (model 1).

Table 6: Model summary

Model R-squared R-Squared Change F-change Sig.

1 0.196 0.196 24.010 0.000

2 0.407 0.211 52.382 0.000

The most important variable on this investigation is the interaction term IAVNB/PL as this variable would indicate a difference between NBs and PLs in sensitivity of sales in relation to economic fluctuations. However as not only the interaction term but also the main relationship are not significant, no conclusions can be drawn from this.

4.3 Hypothesis 2

Hypothesis 2 is tested in a similar fashion using a linear regression analysis.

The variables used in the below regression are:

 Sales (dependant variable)  CCI (independent variable)

 Average score hedonic value & Perceived quality differential (HQ average)

 Interaction term Average score hedonic value & Perceived quality differential (IAVHQAv)  Category 2 Dummy

4.3.1 Hypothesis testing

Table 7 presents the results of the regression analyses performed on the three categories.

As one can see in table 7 only the dummy variables for the different categories show a statistically significant result.

A hierarchical regression allows for a check on how the category dummies moderate the effect of CCI on base sales. As one can see in table 8, the significance of the first model without the category dummies is not statistically

significant. Interestingly though, the second model is statistically significant with the consideration of controlling category fluctuation.

5 Conclusion & Discussion

In this chapter the conclusion of the analysis performed will be discussed.

5.1 General results

This paper made a first step into a more in-depth view of the supermarket environment beyond the division in NBs and PLs. However as most results are not statistically significant one can’t draw conclusions from the presented data.

The results of this investigation lead to the rejection of both H1 and H2 due to mostly statistically insignificant data, however that does not mean that the suggested effects of macro-economic figures on the sales of national brands are not present.

Table 7: Results linear regression

Variable B Sig. CCI -0.237 0.581 HQAverage 0,095 0,190 IAVHQAv 0.210 0,759 Category 2 Dummy -0.457 0.000 Category 3 Dummy -0.426 0.000

Table 8: Model summary

Model R-squared R-Squared Change F-change Sig.

1 0.015 0.015 1.537 0205

5.2 Discussion

This paper is based upon a large body of work that nearly all suggests that during economic contractions PL sales increase and NB sales decrease. The results of this investigation should therefore reflect the co-moving behaviour of the brands with the economic tide. However as one can see in table 7 no significant evidence was found that supports this notion as well as the hypothesised exceptions to this notion. There are several possible reasons which will be discussed here. After discussing the results, the possible forces of influence on the results will be discussed.

5.3 Factors influencing results

There are several factors that could have affected the results as displayed in table 5 and 7 which will be discussed below .

The literature review builds upon the notion that consumers always seek a certain hedonic value in products. Going out to dine at a restaurant usually is not driven by utilitarian motives (consuming a nutritious meal) but by hedonic motives (an enjoyable night out). The same could be expected from products bought in the supermarket. Combining this with the knowledge that consumers might seek this hedonic experience more in supermarket products during a recession due to substitution behaviour (dining out for dining in) one can say that in theory

there is potential for certain NB to actually grow during economic contraction periods. A prerequisite would be that this NB is able to offer the hedonic value the consumer is expecting from it.

Taking the above into consideration the research structure was designed in a way that PLs and NBs were both investigated by comparing baseline sales and the CCI. The factor, however, that could affect the research outcome is the fact that the dependent variable was a brand as a whole (per category). It is without a doubt that one can argue that a NB such as De Ruijter is more likely to be able to deliver a hedonic experience than a PL brand might be. But in many cases, a NB such as De Ruijter has several sub-brands all of which can have a different positioning. To better illustrate this De Ruijter is used as an example. Take the De Ruijter Vruchtenhagel and De Ruijter Specials Intens Puur sub-brands. Only by looking at the vast difference in price (€ 3.60 p/kg and € 8.08 p/kg respectively) one can already see a significant difference in positioning. A stricter application of the theory as presented in the literature review could argue that it is possible that the NB as a whole would not grow during economic contraction periods, but top sub-brands would. The sub-brands that are specifically positioned to serve the top segment of the market by explicitly stating the

hedonic benefits of the product are therefore more likely to grow during a contraction period than its ‘regular’ counterparts.

To illustrate the difference between the positioning of sub-brands one can look at image 5. The positioning of the product on the right, De Ruijter Specials Intens Puur, is obviously different from the ‘regular’ De Ruijter Puur on the left. Also taking into account the advertising campaign launched after the introduction of the De Ruijter Specials (De Ruijter, 2012) one can observe a positioning as a product of indulgement and luxury. The sub-brand name is ‘Specials intens puur’ emphasizing the intense and special characteristics of the product, such as the 60% cacao. The narrator of the commercial emphasizes this by stating things she loves, such as jewels, velvet and art. These observations are supported by the statement of the De Ruijter brand team after winning the ‘Distrifood Wheel of Retail 2012’ award for best new introduction in the breakfast category. The brand team stated that with De Ruijter Specials they attempted to really differentiate themselves from PLs by following the trend for more pure and intense cacao and indulgence (Distrifood, 2012).

Obviously the above made observation is not conclusive and qualitative of nature, but it does illustrate the vast differences between sub-brands within a category. The same differences are likely to exist in the other categories investigated. According to the theory suggesting a growth of NBs during economic contraction periods could support the growth of these specific sub-brands rather than the NB as a whole. Hence the possible influencing factor of having investigated the NB as a whole rather than the sub-brands.

It was mentioned before that the time span of the observations may have affected the results. But it could potentially have an effect on the significance levels as well. Even though the time span is 5 years (2007-2011), only quarterly data is available adding-up to 20 observations. This limited amount of observations may have a dampening effect on the statistical significance of the research results.

5.4 Implications

As mentioned earlier, this paper derives its relevance from the fact that marketing departments are constantly being cut in their budget during economic contractions. Advertisement spending

decreases, investments in innovation comes to a halt and budgets are being reallocated. As none of the results were significant there are no statistically supported implications, however if one is to assume that the expectations derived from the literature review have some ground one could draw the following implications from it:

 Cash-Cow position- NBs that would possess the brand characteristics necessary to allow for a counter cyclical behaviour hardly need extra support during economic contraction periods allowing for marketing departments to cut-back on support-budgets for these brands. These brands would be places in a ‘cash cow’ position (In terms of the growth-share matrix) where little will be invested and cash generated from these brands may be used to support other brands (Aaker & McLaughlin, 2007).

 Invest to become the exception to the rule- Besides the fact that there is academic evidence supporting an increase in advertisement spending during economic contractions, the

literature review in this investigation support the notion that marketing- and/or brand managers should invest more in premium brands that have the potential of offering a salient and strong positioning, invest in a perceived quality differential and a true hedonic

experience to the consumer. This goes against the common practise of slashing prices to regain market share in a market where consumers increasingly favour PLs over NBs. Invest significantly in the establishment of a difference in perceived quality within a product category and perhaps the brand in question might become the exception to the rule.

5.5 Suggestions for further research & limitations

It is suggested that this investigation would yield better results if the amount of observations would increase. Not just because of the increase in statistical confidence in the results, but also because it would then encompass multiple business cycles which could more accurately display the strength of the relationship.

The problem with investigating certain sub-brands is that many sub-brands tend to have a rather short life cycle. The sub-brand discussed earlier, De Ruijter Specials Intens Puur, was only introduced in 2011 and would therefore not qualify for a longitudinal investigation. Besides this, sub-brands tend to come and go making it difficult to look at a sub-brand level in general when wanting to do a similar longitudinal investigation.

Once certain NBs have been identified that have the potential to grow during economic contractions according to the theory as presented in this paper, one should investigate the exact implications on management and marketing. This is in particular important because it could present marketing managers with opportunities they did not have before. From personal experience, and from literature, it is clear that economic contractions result is massive cut backs internally and price slashing externally. This opportunity would shake-up this paradigm and presents a marketer with more options to strengthen their brands.

The most significant constraint of this research is that of the IRI data, or base-sales numbers. The research was completely dependent on H.J. Heinz to make the data available.

6 References

Aaker, D. A., McLaughlin D. (2007). Strategic Market Management, European Edition. John Wiley & Sons, Ltd: West Sussex

Adformatie, (2004). Nima, Bertolli, Heineken, Hema, Interpolis of Rituals. [online]. Available at: < http://www.adformatie.nl/nieuws/bericht/nima-bertolli-heineken-hema-interpolis-of-rituals> [Accessed 21 May 2012]

Akbay, C., Jones, E. (2000). Food Consumption Behaviour of Socioeconomic Groups for Private Labels and National Brands. Food Quality and Preference, Vol. 16 (7), pp. 621-631

Albert Heijn (2012). Albert Heijn Webshop. [online] Available at: <http://webwinkel.ah.nl> [Accessed 25 April 2012].

Ang, S. H., Siew, M. L., Kotler, P. (2000). The Asian Apocalypse: Crisis Marketing for Consumers and Business. Long Range Planning, vol. 33 (February), pp. 97–119.

Axarloglou, K. (2003), The Cyclicality of New Product Introductions. Journal of Business, Vol. 76 (January), pp. 29-48

Becker, G. S. (1962). Irrational Behaviour and Economic Theory. Journal of Political Economy. Vol. 70 (1), pp. 1-13

Batra, R., Sinha, I. (2000). Consumer-level factors moderating the success of PL brands. Journal of Retailing. Vol. 76 (2), pp. 175-191

Baxter, M., King, R. G. (1999). Measuring Business Cycles: Approximate Band-Pass Filters For Economic Time Series. Review of Economics and Statistics. Vol. 81 (November), pp. 575-593

CBS (2012). Consumentenvertrouwen index. [online] Centraal Bureau voor Statistiek. Available at:

<http://www.cbs.nl/nl-NL/menu/themas/dossiers/conjunctuur/publicaties/conjunctuurbericht/inhoud/conjunctuurklok/toe lichtingen/ck-03.htm> [Accessed 25 April 2012]

Deleersnyder, B., DeKimpe, M. G., Steenkamp, J.B.E.M, Koll, O. (2007). Win-Win Strategies at Discount Stores. Journal of Retailing and Consumer Services, Vol. 14, pp. 309-318

Deleersnyder, B., DeKimpe, M. G., Steenkamp, J.B.E.M., Leeflang, P. S. H. (2009). The Role Of National Culture In Advertising’s Sensitivity To Business Cycles: An Investigation Across Continents. Journal of Marketing Research, 46 (October)m pp. 623-636.

De Ruijter, 2012. De Ruijter Specials [video online]. Available at

<http://www.youtube.com/watch?v=ygJu_KSdz5Q> [Accessed 20 May 2012]

Distrifood (2012). Distrifood Winnaar Wheel of Retail 2012: Ontbijtproducten. [online] Available at: <http://www.distrifood.nl/web/Evenementen/Wheel-of-Retail-2012/Ontbijtproducten-De-Ruijter-Specials.htm> [Accessed 9 June 2012]

Dunn, M.G., Murphy, P.E., Skelly, G.U. (1986). Research note: the influence of perceived risk on brand preference for supermarket products. Journal of Retailing, Vol. 62 (Summer), pp. 204-17.

Estalami, H., Lehman, D. R., Holden, A. C. (2001). Macro-Economic Determinants Of Consumer Price Knowledge: a Meta-Analysis of Four Decades of Research. International Journal of Research in Marketing. Vol. 18 (December), pp. 341-355.

European Institute for Brand Management (2010). EURIB Top 100 Onmiskenbare Merken. EURIB, Rotterdam, The Netherlands

Familieshopper.nl (2012). Familie shopper webshop. [online] Available at <http://www.familyshopper.nl/funkyvlokkies-melk> [Accessed 25 April 2012]

Flatters, P., Willmott, M. (2009). Understanding the Post-Recession Consumer. Harvard Business Review, Vol. 87 (7/8), pp. 106-112.

Garretson. J. A., Fischer, D., Burton, S. (2002). Antecedents of PL attitude and NBs promotion attitude: Similarities and Differences. Journal of Retailing. Vol. 78, pp. 91-99

Grewel, D., Gotlieb, J., Marmorstein, H. (1994). Moderating Effects of Message Framing and Source Credibility on the Price-Perceived Risk Relationship. Journal of Consumer Research, Vol. 21 (June) pp. 145-153