S URVIVAL OF THE P RETTIEST

(1)

Master thesis, Msc. BA Marketing Management & Research

S

URVIVAL OF THE

P

RETTIEST

P

RIVATE LABEL VS

.

N

ATIONAL BRAND

IN THE MAKE

-

UP CATEGORY DURING ECONOMIC RECESSION

University of Groningen

Faculty Economics and Business Department of Marketing Nettelbosje 2 9747 AE Groningen Author: Yvonne Flierman (1661620) Kerkstraat 278 1017 HA Amsterdam +31 623387904 y.flierman@gmail.com Supervisors:

1st Supervisor: Dr. J.E.M. van Nierop Co-assessor: Daniela Naydenova

(2)

2

I. Management Summary

This research is performed to investigate the decision process between private labels and national brands in the make-up category. The main research question of this research is: Which factors

influence consumer choice between national brands and private labels which will be bought in the mass market make-up category during harsh economic times?

Based on an extensive review of several research papers and other academic literature, nine factors which could influence the decision process between national brands and private labels were taken into account in this research for the mass make-up market. A differentiation is made between category characteristics and consumer characteristics. Perceived risk, perceived quality variation and perceived image of the retailer belong in this research to the category characteristics. Consumer characteristics which are taken into account are price consciousness, consumer innovativeness, consumer knowledge and personal involvement. Besides these two groups of variables, the three remaining variables which are expected to have an influence are rebuy intention, experience and economic recession. Four of these variables are assumed to have a mediating influence between an independent variable and the dependent variable private label purchase intention. In order to check whether all these variables have an influence on this intention, hypotheses were formulated. The data is gained through an online survey. The respondents were asked if they were familiar with the private labels of three well-known retailers. Only if they answered yes to this familiarity with one, two or three of these retailers, they had to answer the questions which belong to this specific retailer. Subsequently, multiple regressions, latent cluster analyses and latent class regression are performed in order to analyse the data. The results of the several analyses suggest that not all the nine factors that were mentioned above have an influence on the decision process of private labels and national brands in the mass make-up market. The research did suggest that the risk which is perceived by the consumer has an influence on the private label purchase intentions. The same can be said for the variable experience, perceived image of the retailer and for economic recession. Where perceived risk has a negative relation with private label purchase intention, all the other variables result in a positive relation with private label purchase intention.

Furthermore, a latent cluster analysis is performed in order to check for relevant subgroups of make-up consumers regarding their purchase behavior in the mass make-make-up market. Latent cluster analysis resulted in the following outcome, two clusters, respectively the youngsters and the high professionals cluster. The youngsters cluster has a lower income, lower education level, lower make-up expenditures, more influence of economic recession and at last a higher private label purchase intention in comparison with cluster 2. To give a better insight in the difference in effects of the independent variables between the clusters, a latent class regression is performed. The most remarkable difference in effect is concerning economic recession. This effect is positive for youngsters and negative for high professionals. A possible reason for this is the lipstick effect (Hair et all, 2012).

(3)

(4)

4

II. Preface

I am performing this research in order to understand which factors play a role in the decision process of the consumer. My interest for this topic existed especially during my internship at L’Oréal, where I was part of the L’Oreal Paris make-up team. Every month we received the market shares of the several segments and it was remarkable that the share of the private labels within the make-up category kept on growing for years now. This growing share over the last few years is a risk for national players, like L’Oréal. Based on these facts, I started writing my thesis in January of 2013, right after my internships of six months.

It was my goal to finish my thesis at the same time as I finished my internship. Quite a challenge, unfortunately this combination appeared infeasible. Right after my great internship, I had the chance to apply for a job at L’Oréal. I already started half February working for two days a week and per 1 May I started as Merchandising Manager of L’Oréal Paris Make-up. I am very happy with this great opportunity to work in a dynamic market as the make-up market and to work in such a successful FMCG company as L’Oréal. Besides this, this hard deadline of 1 May helped me to finish my thesis in a very quickly way.

I want to thank my supervisor Erjen van Nierop for his flexibility, his empathy in the make-up market  and all the help during the writing of this thesis. Next to him, I want to thank my co-assessor, Daniela Naydenova. Besides them, I want to thank my parents for all their support during my study. They always stimulated me to do whatever felt good and this pays off now in having the job which I wanted so badly. I want to thank Miss Lipgloss blogster Cynthia for sharing my research link to her big network on twitter, which helped me to have the right number of respondents. Last, but not least I want to thank the make-up team at L’Oréal, Anouk Lokhorst, Lies Uljee and my supervisor Nanja Ummels for all the insights they gave me.

(5)

5

III. 1 Introduction and Background

In November 2010, Leonard Lauder, the chairman of cosmetics firm Estée Lauder, coined the widely discussed concept the “Lipstick Effect”. In times of economic recession typically consumer spending declines, however research showed that this typical behaviour does not appears in the beauty category. Contrary women increase spending on beauty products, such as lipsticks (Hill et. al, 2012). Whereas the Lipstick Effect was the effect of the year 2010, the year 2011 brings the so-called Nail Polish Effect. With a global retail growth in nail polish of 11% in 2011 and a sales increase of 43% between 2008 and 2011, a period characterized by global financial meltdown, nail polish definitely can be seen as the new Lipstick Effect (Euromonitor International, 2012). In figure 1 (Mass Market Nail Polish Effect) the growth of the nail market in total either the growth of private labels in this category can be seen during the years 2010, 2011 and 2012.

Figure 1: Mass Market Nail Polish Effect 2010-2012 (Nielsen, P13 2012)

Besides this tremendous high growth of nail polishes, the beauty and personal care industry in general showed a global growth of 5% in 2010 as well. Irina Barbalova, head of beauty and personal care with Euromonitor International attributes this to the strong performance of mass market products in this category. Moreover future indications show even stronger presence of mass cosmetics in 2015. Euromonitor International found two main reasons for this. First reason is the dominance of Latin America in the global beauty industry. According to Euromonitor International (2010) the beauty industry in Latin America will have an expected growth of 22%, whereas the global beauty industry will have an expected growth of 12%. Secondly, the rising penetration of private labels in this category is driving the growth behind the mass cosmetics. The market is growing more in volume than in value, because of the generally lower prices of private label brands in comparison with national brands. Especially this second driver is interesting for this thesis which will focus on the Dutch cosmetic market.

(9)

9 When looking deeper into the rising penetration of private labels in general, it can be said that the success of private labels is not of the same level in every category. In 2005 there was a big gap between the private label share of refrigerated food and cosmetics, respectively a market share of 32% versus a market share of only 2%. Many researchers performed research to clarify the differences in market share between categories. For example Kumar and Steenkamp (2007) concluded that the degree of innovation activity is determinative for the private label share within a specific category. A category with low innovation activity has to deal with a 56% higher private label market share in comparison with categories with high innovation activity (Kumar and Steenkamp, 2007).

Carrie Bonner, a senior research analyst with Little Falls, N.J. – based consulting firm Kline & Co. has the following explanation for the difference between categories: “Cosmetics are a little different. Sales are not as strong as they are in other private label categori es, such as mouthwash. Name brands have a strong following and consumers are sceptical about whether private label cosmetics will meet their expectations. Consumers are loyal to cosmetic brands, because brands are fashion-forward, and branded manufacturers are regarded as authorities on beauty. Consumers might give it a try, if the price points are low enough.” The promotion pressure within the make-up category is 50% (Nielsen, YTD P1 2013). This means that one out of two make-up products which are sold in the mass-market are sold in promotion. Based on the quote of Carrie Bonner and research of Hoch and Banerji (1993) it can be said that in earlier years it was tough for private labels to gain market share in the beauty category. Hoch and Banerji (1993) dedicate this to the quality difference between private labels and national brands in that time.

Like Bonner said, it looks like consumers did give it a try. When looking at the private label growth (Figure 2: Private label share and growth), in 2005 the cosmetics category with the lowest private label share at that moment has the highest private label growth, respectively a growth of 23%. Why is this growth suddenly this high?

(10)

10 Steenkamp et al. (2010) concluded that in the Netherlands the private label share is moderate to high in comparison with the global private label share level. The research concludes that one important reason for this is that consumers are not willing to pay a price premium for a national brand over a private label. The figure below shows the results of the private label share and the willingness to pay worldwide. Our country has worldwide a number 4 position concerning private label share and a number one position concerning the lowest willingness to pay, which is visualized in figure 3 (Steenkamp et al. ,2010).

Figure 3:Willingness to pay vs. private label share worldwide, Steenkamp et al. (2010)

(11)

11 biggest threat in the make-up category is the rise of the market shares of these discount private labels, which are sold at remarkable lower prices in comparison with the prices of the national brands in this category. An overview of the price ranges of national brands and private labels can be found below in table 1 and 2. There is an obvious difference between the lowest and highest price borders of national brands and private labels.

Price Range L’Oréal Paris Maybelline Max Factor Rimmel Bourjois Eye 11.99 – 17.99 6.99 – 15.49 9.99 – 17.99 4.99 – 12.49 7.62 – 14.99

Lip 11.99 – 15.99 11.79 – 14.29 7.99 – 16.99 8.99 – 10.99 11.17 – 12.99

Teint 12.99 – 17.49 9.99 – 15.29 9.99 – 15.99 6.99 – 12.99 7.99 – 15.49

Nail 5.99 – 8.49 3.99 – 8.79 4.99 4.99 – 7.99 8.49

Table 1: Price range national brands, mass market make-up (Retail Amsterdam, January 2013) Price

Range

Etos DA HEMA

Miss Helen

Essence Catrice NYC Gosh

Eye 1.82 – 5.07 2.99 – 6.09 2.00 – 9.00 1.59 – 2.99 2.59 – 4.59 1.99 – 4.99 7.79 –11.49

Lip 2.94 – 6.09 2.02 – 6.60 2.00 – 4.00 0.99 – 1.99 2.39 – 3.99 1.99 – 3.99 8.49 – 9.49

Teint 3.04 – 7.11 3.55 – 7.11 4.00 –11.00 2.59 – 3.89 2.99 – 6.59 2.79 – 4.99 7.99 –16.99

Nail 2.53 2.99 2.25 – 5.95 1.69 – 2.59 2.69 2.29 – 2.99 7.29

Table 2: Price range private labels, mass market make-up (Retail Amsterdam, January 2013)

In the last few years private labels have reached a number one position in every segment of the make-up category (Nielsen, YTD P6 2012). The make-up category can be divided into four different segments, respectively; eye segment, lip segment, teint segment and the nail segment. In all these segments private labels are on top, followed by the big five make up A -labels (Nielsen, 2012).The table below gives an overview of the market shares of the big five mass market make-up labels and the private labels. As can be retrieved from this table, private label has the largest market share in every segment, respectively lip (32,4%); eye (31,9%); teint (34,3%); nail (50,1%) (Nielsen, YTD P6 2012).

TOTAL MAYBELLINE RIMMEL

L'ORÉAL

PARIS MAX FACTOR BOURJOIS

PRIVATE LABEL MS. Evol. MS. Evol. MS. Evol. MS. Evol. MS. Evol. MS. Evol. MS. Evol. DECO MASS 100.0% 3% 12.5% -10% 9.7% 2% 16.4% 12% 13.9% -10% 4.6% -8% 35.2% 14% LIP 17.6% -6% 10.3% -25% 8.3% -9% 19.5% 8% 20.2% -23% 3.5% -22% 32.4% 12% - Lipstick 12.6% -5% 12.1% -28% 8.5% -3% 21.4% 18% 24.6% -21% 2.4% -18% 27.7% 15% - Lipgloss 4.2% -13% 5.3% -6% 6.9% -22% 15.6% -20% 9.1% -34% 7.4% -24% 41.9% -1% - Lipcontour 0.6% -6% 12.5% -11% 15.0% -17% 14.3% 11% 13.9% -14% 1.7% -24% 36.3% 2% EYE 45.2% 1% 15.5% -11% 9.2% 10% 17.7% 3% 15.0% -5% 6.3% -7% 31.9% 15% - Mascara 28.0% 1% 21.2% -15% 9.6% 17% 22.3% 2% 21.3% -2% 4.1% -4% 19.3% 32% - Eyeshadow 7.4% -7% 6.4% 2% 5.8% -11% 8.7% -27% 6.3% -21% 9.6% -10% 54.9% 3% - Eyecontour 9.8% 8% 6.1% 34% 10.7% 4% 11.5% 46% 3.9% -25% 10.1% -9% 50.5% 10% TEINT 22.0% 10% 13.5% 11% 9.3% -2% 19.0% 20% 15.2% -1% 3.9% 2% 34.3% 18% - Foundation 12.6% 12% 19.3% 14% 5.4% 2% 25.3% 23% 18.1% 1% 3.1% 10% 25.6% 13% - Powder 5.3% 4% 4.1% -26% 16.6% -1% 10.7% 7% 14.3% -1% 3.8% 13% 43.4% 15% - Blush 2.0% 3% 4.9% -19% 8.3% -13% 12.4% 0% 9.2% -10% 9.2% -18% 49.2% 23% - Camouflage 2.1% 22% 10.3% 86% 15.0% -5% 8.4% 63% 6.1% -16% 3.4% -3% 49.9% 35% NAIL 14.8% 13% 4.4% -23% 13.5% -2% 5.5% 164% 1.3% -32% 2.1% -11% 50.1% 14%

(12)

12 As can be retrieved from the table above and many past research, private label share is high at this moment and keeps on rising. An overview of the development of the market share of private labels within the mass market make-up category is given below (figure 4: Growing mass make-up market). From 2003 onwards today there is yearly significant growth of the market share of private labels. Where the top five A-brands alternately are losing and gaining market share every year, private labels are steadily gaining market shares from these top five A-brands and a growing market.

Figure 4: Growing mass make-up market (Nielsen,YTD P6,2012)

(13)

13

Figure 5: Fluctuation MS private labels period 2003-2012 (Nielsen, YTD P6, 2012)

One of the reasons of the rise of private labels in general is the low pricing of these private labels (Batra and Sinha, 2000). Price can be seen as an important element in the decision process whether buying a national brand or a private label. However Hoch and Banerji (1993) found that in determining the success of private labels, lower price seems to be less important than high quality. Batra and Sinha (1999) describe that price consciousness of consumers is one of the reasons of the big growth in PLB sales. This price consciousness goes together with large quality improvements on PLBs.

Besides price and quality Batra and Sinha (2000) describe several consumer perceptions that influence the choice of buying a private label vs. a national brand. These perceptions, which are degree of perceived quality variation, level of perceived risk and perceived value for money all have an influence on the choice of the consumer. Concerning the perceived risk they state that well-respected brands (A-brands in the make-up category) will have a larger purchase probability therefore awareness will reduce perceived risk. If you translate this into implications for national brands which are in a fight with private labels, the suggestion of Batra and Sinha (2000) is to focus on raising the consumer’s perceived consequences of making the wrong brand choice.

(14)

14 effect on the decision to buy, which will go the other way around. Here a greater knowledge will lead to the perception that there are differences between the national brand and the private label. When people experience differences between these two, this will be in favour of the national brand purchases.

Innovativeness of the consumer can play a role in the decision process between national brands and private labels as well. Consumer innovativeness is defined by Steenkamp et al. (1999) by: the predisposition to buy new and different products and brands rather than remain with previous choices and consumption patterns”. In other words, consumer innovativeness is the degree of tendency to adopt a new product. In research of Jin and Suh (2005) the strongest factor for private label attitude was consumer innovativeness, so people with a higher degree of innovativeness will buy private labels in favour of national brands.

During economic recession market shares of private labels are increasing (Lamey and colleagues, 2007). This can also be retrieved from figure 4 and 5. Besides private labels the whole make-up market is growing in value and in units, respectively 2,6% in value and 3,7% in units (Nielsen, YTD P6, 2012). Lamey and colleagues (2007) found that private label shrinks when the economy is flourishing. The main point and a very interesting point of their research as well is the fact that during an economic recession consumers switch more extensively to private labels than they switch back to national brands in times of economic recovery. Steenkamp (2012) supports this by his statement: private label share behave counter-cyclical. This statement has consumer-related and manufacturer related causes. Consumer-related causes for this cyclical sensitivity are reduction of income, increase of price consciousness and economizing in price instead of in quantity. A manufacturer related cause is cutting down on innovations (Steenkamp, 2012). Figure 6 (Majority consumers purchase private labels) shows that the majority of consumers indeed purchased more private label brands during economic downturn (Nielsen, Q3 2010).

Figure 6: Majority consumers purchase private labels (Nielsen, Q3, 2010)

(15)

15 The main research question of this thesis is:

Which factors influence consumer choice between national brands and private labels which will be bought in the mass market make-up category during harsh economic times?

Subquestions:

1. Which category characteristics play a role in the decision process betwee n a private make-up label and a national make-up label?

1.1 Perceived risk

1.2 Perceived quality variation

2. Which consumer characteristics play a role in the decision process between a private make-up label and or a national make-up label?

2.1 Personal involvement 2.2 Consumer innovativeness 2.3 Price consciousness

3. Do consumers which currently use a private label stick to this private label use after the recession?

(16)

16

2 Theoretical Framework

This chapter will discuss relevant academic articles in order to answer the main research question and the research questions as discussed in the introduction. After answering each research question, one or more hypotheses are formulated. An overview of these hypotheses founded in academic literature is given by the conceptual framework, which will function as a guideline for the empirical research. In the hypotheses which are mentioned below, two different terms are used, both terms have the same meaning in this thesis. The first one is private label purchase intentions. The second term is national brand purchase intention. In this thesis both terms are used together, both with the same definition. A lower national brand purchase intention indicates a higher privatelabel purchase intention and vice versa.

2.1 Category characteristics

During the buying process several category characteristics play a role in the final decision of the consumer (Inman et al, 2009). Two of these category characteristics, respectively perceived risk and perceived quality variation will be discussed in this thesis.

2.1.1 Perceived risk

(17)

17

Figure 7: FCB model – feel type and think type of products (Choi et al., 2012)

(18)

18 has a considerably lower perceived risk compared with an eye-, lip- or teint-segment. Eye, lips and skin are much more sensitive to beauty products than nails are, for this reason a higher risk is expected. Besides this performance risk, there is also a social risk related to the make -up category. Ladies often share the content of their beauty cases to each other. Batra and Sinha (2000) conclude that purchases which are socially exposed to a reference group can also be of high risk.

H1: Consumers have a higher intention to buy private labels in segments of the make-up category where they perceive a low risk

2.1.2 Perceived quality difference

Many consumers traditionally believe that national brands have a higher quality in comparison with private labels (Sethuraman, 2000). This traditional belief is according to other research papers outdated. In 1992, Fittzel stated that private labels mostly perform as well as national brands. Moreover a Gallup study (1999) showed that 75% of consumers think the product quality of national brands and private labels is of the same level. Perceived quality difference is the difference which consumers perceive between the quality of national brands and private labels. When a consumer prefers a private label over a national brand, the decision might be based on the fact that the quality difference between this private label and national brands is perceived as minimal or acceptable in comparison with the price (Jin and Suh, 2005). Jin and Suh (2005) stated: less quality variation between private labels and national brands will result in higher private label purchases. To test whether this statement also works for the make-up category, the following hypothesis is formulated:

H2: A higher perceived quality difference between private labels and national brands leads to higher purchase intention of national brands

2.1.3 Perceived image retailer

The perceived image of the retailer, in other words the attractiveness of the retailer is the impression of consumers that visit this specific retailer regarding variety of offering and in -store atmosphere (Teller and Reutterer, 2008). This in-store atmosphere is formed by variables as the design of the store, merchandising activities in the store and physical characteristics (Groppel, 1993). Based on all these variables consumers create a perceived image of a retailer, which will influence the two variables perceived risk and perceived quality variation.

H3a: A positive perceived image of a retailer will lead to a lower perceived risk and perceived quality variation

(19)

19

2.2 Consumer characteristics

When a consumer buys a product, several characteristics of this consumer play a role in the decision process. There are four main characteristics, respectively psychological, personal, social and cultural characteristics. In the pyramid of Kotler (1999) figure 8, these characteristics are explained by some concepts.

Figure 8: Factors affecting consumer behavior (Kotler et al., 1999)

The pyramid of Kotler et al. (1999) will be used in the segmentation analysis which will be performed in this thesis. The focus will be on the top of this pyramid. This thesis will only look at the psychological and personal factors which influences the decision process of consumers. Kotler et al. (1999) separate the following psychological factors which influence the behaviour of a consumer: motivation, perception, learning, beliefs and attitudes. Especially perception will be widely covered in this thesis; the focus will be on the price/ quality perception and the personal involvement and knowledge within the beauty category. Personal factors which influence the consumer behaviour of a buyer are the following: age and lifecycle, occupation, economic position, lifestyle and personality.

2.2.1. Personal involvement and consumer knowledge

Assael (1998) believes that involvement is a key moderator of the strength of the relationship between attitude and behavioral intentions. Besides this moderating effect, involvement has a direct effect on the outcome of product use (Bloch and RIchins, 1983). They found that consumers who have a high involvement with a product will expect highly valued outcomes with product use. This means for the make-up category, when people are highly involved with the several or one of the make-up products, because of the trends, the atmosphere of the brand which they feel related to, or involvement in another form will associate highly valued outcomes with product use. Lichtenstein, Bloch and Black (1988) support this statement. They say that consumers who are highly involved in a product have important functional, social and physiolo gical associations with the product, which results in a high care about product quality.

(20)

20 is formed by the feeling a consumer has for a specific brand. This is also related to the self -image congruence concept. Congruence between brand -image and self--image is positively related to consumers’ product evaluations (Graeff, 1998). On the other hand research of Miquel et al. (2002) found that higher personal involvement leads to greater consumer knowledge within a specific category. Their research found two relations between consumer knowledge and private label buy. Firstly they found that consumer knowledge of a category showed consumers that there were no differences based on quality between national brands and private labels. The only difference that appeared was the price, based on this consumers did chose the private label brand. The second relation between consumer knowledge and private label buy is an inverse effect. In this case the consumer knowledge leads to a realization that there are differences between private labels and national brands, this will be in favour of the national brand purchases.

H4: High personal involvement with the make-up category leads to a higher purchase intention of national brands

H5: High personal involvement in the make-up category leads to a greater consumer knowledge within the category

H6a: Greater consumer knowledge leads to a bigger purchase intention of private

H6b: Greater consumer knowledge leads to a bigger purchase intention o f national brands

2.2.2. Consumer innovativeness

Consumer innovativeness is defined by Steenkamp et al. (1999) as: “the predisposition to buy new and different products and brands rather than remain with previous choices and consumption patterns”. A general finding concerning this predisposition of consumer innovativeness is that it leads to an early adoption of products (Goldsmith et al., 1995; Jin and Suh, 2005). Roerich (2004) performed research on this predisposition to buy innovative products, he describes four main explanations why consumer innovativeness exists: stimulation need; novelty seeking; independence toward others’ communicated experience and need for uniqueness.

(21)

21

Figure 9 : A-brand innovation activity in the category vs. private label share (Steenkamp, 2012 )

Brand innovation activity leads to big difference of private label market shares. Low brand innovation activity leads to an index of private label share of 134%, where a high brand innovation activity leads to an index of private label share of 86%. Kumar and Steenkamp (2007) confirm this statement in their research by stating that the higher the number of product launches in a particular industry, the lower the share of private labels in that particular industry will be. Several authors are confirming the statement of private labels being less successful in categories where the innovation activity is high (Wuelch and Harding, 1996; Steenkamp and Dekimpe, 1997). National brands have more budgets for research and development in their company, that’s why national brands are more innovative than private labels. Consumers who have a high degree of consumer innovativeness are expected to buy more national brands, because of their higher brand innovation activity compared to private labels.

H7: Consumers who have a greater degree of consumer innovativeness have higher purchase intention of national brands in the make-up category

2.2.3. Price consciousness

Price consciousness can be defined as a consumer degree of exclusively focusing on paying low prices for products (Jin and Suh, 2005; Lichtenstein et al. 1993). The general finding in past research is that price can be seen as the most important reason for buying a private label ( Batra and Sinha, 1999; Burton et al., 1998). However the degree of price consciousness is depending on the product category. The cosmetics category, as a component of the health and beauty aids category is such a category with a high risk of making an incorrect deci sion of purchase (Dunn, Murphy and Skully, 1986). Moreover, the economic downturn, forces consumers to trade down. There are three reasons for this, firstly the reduction of income, secondly consumers are becoming more price conscious and lastly consumers economize on expenditures through price instead of decreasing the quantity they buy (Steenkamp et al. 2012). In other words Steenkamp (2012) shows that consumers are trading down to cheaper products, which are most likely private labels. In section 2.3 the influence of economic recession on private label will be discussed. This section indicates that economic recession also has an influence on price consciousness of consumers.

134

86

0 20 40 60 80 100 120 140 160 In d e x o f Pr iv ate Lab e l S h ar e

Brand innovation activity

Low High

(22)

22 H8: In the make-up category price conscious consumers buy more private labels than non-price conscious consumers

2.3 Private label buy intention after economic recession

Much research has been performed on this topic. Lamey et al. (2007) found that private label success has counter cyclical behaviour. Which means that during economic harsh times private label share is increasing, but during flourishing economic times the private label share is shrinking. Besides this main point, the research did have two other important findings.

Firstly, they found that business cycle fluctuations cause inequality in the extent and the speed of the movements in private label share. Secondly the gained share of private labels during harsh economic times is partly permanent (Lamey et al., 2007). It is for a reason that Steenkamp (2012) recommends manufacturers to avoid private label trial of their consumers. He states exactly the same as Lamey et al. (2007); private label share grows across the business cycle, but the rate of growth is much larger than the decline in contractions, and much of these gains are permanent.” Besides this he found that it is often impossible for national brands to reach their pre-contractions market share again once the economy is in expansion.

On the other side it is stated by Ward et al (2002) that private labels are ‘discarded’ from the category once the economy is growing again. But the general finding concerning this topic is that once a consumer tried a private label during an economic downturn, most of the consumers keep on buying private labels even when the economic is flourishing (Lamey, Deleersnyder, Dekimpe & Steenkamp, 2007). Moreover looking at figure 10 (Purchase intention private labels after economic recession), 92% of European consumers will continue to purchase private label brands after economic recession, only 8% will stop private label purchase (Nielsen, Q3 2010). In addition, it is expected that economic recession leads to more price consciousness consumers. Price consciousness was discussed in section 2.2.

(23)

23 H9a: Economic harsh times lead to higher purchase intention of private labels

H9b: Economic flourishing times lead to a higher purchase intention of national brands

H9c: Consumers switch more easily to private labels in economic harsh times than they switch back to national brands during economic flourishing times

H10: Economic recession lead to more price consciousness consumers

2.4 Rebuy process and experience

Past research shows that consumers consider private labels as a group of similar brands without any differentiation. Private labels are seen as homogeneous brands (Ailawadi, Neslin, and Gedenk 2001;Bonfrer and Chintagunta 2004). Richardson (1997) confirms this with his experiment and concludes that consumers do not perceive differentiation between store brands. Szymanowski and Gijsbrechts (2012) found that because of this private labels benefit from reputation spillovers from other private label brands. This can be seen as cross-brand learning for private labels. This cross-brand learning results in a reduced uncertainty of private labels in the mind of consumers, which again leads to a higher purchase intention of other private label brands within other retailers. When looking at national brands it can be said that the process is different for national brands. National brands build their reputations purely individual, that’s why they don’t receive cross-brand benefits from other national labels (Szymanowski and Gijsbrechts, 2012).

H11a: Positive experience with a private label buy leads to a greater intention of trying other private labels

H11b: Negative experience with a private label leads to a greater intention of buying national brands

H11c: Negative experience with a private label leads to a greater intention of buying other private labels

(24)

24

2.5 Conceptual model

(25)

25

3 Research design

The following chapter gives an overview of the research methods that will be used, in order to test the several hypotheses to eventually provide an answer to the problem statement. First, the research design will be described, followed by the data collection. Lastly the plan of analysis will be outlined in this chapter.

3.1 Research Methods

The research that is performed in this thesis is a descriptive research in order to perform statistical quantitative analysis on the gathered secondary data. This descriptive research is characterized by a preceding formulation of specific hypotheses. For this reason clearly defined research problems are considered as highly important (Malhotra, 2007). Furthermore this descriptive research is characterized by a preplanned and structured research design with the final goal to describe market characteristics or functions (Malhotra, 2007). Malhotra makes a distinction between exploratory research and conclusive research. The descriptive research which is used in this thesis is a type of conclusive research. This type of research is characterized by a formal and structured research process, a large and representative sample and a conclusive result of the research. (Malhotra, 2007). In summary, the research which is conducted in this thesis is conclusive descriptive research.

In order to give an answer to the specific formulated hypothesis, an online questionnaire is performed. This online questionnaire (Appendix I) is a combination of statements related to the constructs which are measured. Female respondents are requested to give their opinion about these statements. The two main reasons for the choice to collect data with an online questionnaire are the high speed and the low costs of this method. Besides this, Malhotra (2007) cites a list of advantages for online questionnaires. Some important advantages are moderate to high diversity of questions and flexibility, high perceived anonymity of the respondent and a low social desirability.

A disadvantage mentioned by Malhotra is the low response rate of online questionnaires. The online questionnaire is spread amongst respondents that are using make-up, so mostly female respondents are expected. Besides relatives and co-students, the questionnaire also reached a large independent group of respondents. Miss Lipgloss, Hollands’ biggest beauty blogger with more than 15.000 followers on Twitter, distributed the online questionnaire among her widespread network.

3.2 Operationalizations

(26)

26 The majority of the questions have a seven-point Likert scale, this scale is ranging from strongly disagree to strongly agree. Besides this Likert scale, there is also a Likert scale which is ranging from very unlikely to very likely. In this question people are requested to say something about their future purchase intention, which means the intention to purchase a private label after the economic recession. Furthermore some nominal measurement scales (preferred brand/preferred retailer) and some interval measurement scales (age/make-up spending) are used.

3.2.1 Personal involvement

The four items that are used to measure the construct personal involvement are from Zaichkowsky (1985): 1) I would be interested in reading information about how the product is made, 2) I have a most preferred brand of this product, 3) I would be interested in reading the Consumer Reports article about this product category, 4) I have compared product characteristics among brands of this product.

3.2.2 Consumer knowledge

Consumer knowledge is measured according to Delvecchio (2001) with three items: 1) If someone asked me for help in selecting a category, I could provide reliable advice, 2) Compared to my friends, I am knowledgeable about the category, 3) In general, I am knowledgeable about the category.

3.2.3 Economic recession

To measure the influence of the construct economic recession two items from Nielsen (2010) are used: 1) During economic downturn I buy private labels, 2) When the economy improves I will continue buying private labels. Two other items are retrieved from the research paper of Strien and Wierenga (2009): 1) During economic downturn I spend less money buying less national brands 2) Economic recession has no influence on expenditures.

3.2.4 Price consciousness

Price consciousness, discussed in chapter 2 can be defined as a consumer degree of exclusively focusing on paying low prices for products (Jin and Suh, 2005; Lichtenstein et al. 1993). This construct is measured by 2 items. These two items are from Batra and Sinha (2000): 1) When buying a brand of (category), I look for the cheapest brand available, 2) Price is the most important factor when I am choosing a brand of (category). Another item is: Seeking for the best price is worth the effort (Grewal and Mamorstein, 1994).

3.2.5 Consumer innovativeness

The items for consumer innovativeness are gathered from two papers. The following three items are from Byoungho Jin and Yong Gu Suh (2005) and Roehrich (2004): 1) I often seek out information about new products and brands, 2) When I see a new brand on the shelf, I am not afraid of giving it a try, 3) I am more interested in buying new than known products/ I like to buy new and different.

3.2.6 Perceived image retailer

(27)

27 would be economical. From Keller (2008) the following item is used: I would go out of my way to shop at this retailer brand. Quelch and Harding (2012) came up with the following item: Quality in comparison with national brands is high and improving.

3.2.7 Perceived risk

According to the business dictionary, perceived risk is the consumer’s level of uncertainty regarding the outcome of a purchase decision. Perceived risk will be measured with a scale adjusted from Paul C.S. Wua, Gary Yeong-Yuh Yeh A, Chieh-Ru Hsiao (2010). The following three items are used in their research: 1) If I were to purchase private label brand for use, I become concerned that the product will not provide the level of benefits that I would be expecting , 2) As I consider to purchase a private label brand for using it at home, I think the body probably will be harmed, due to an overuse of the product, 3) As I consider the purchase of private label brand for use, I worry about whether the product will really perform as well as it is supposed to.

3.2.8 Perceived quality variation

The perceived quality variation is measured by a combination of items from two papers, respectively a paper of Batra and Sinha (2000) and a paper of Dick et al. (1995). The following four items will measure perceived quality variation: 1) The national brand and the store brand of the store I commonly purchase from are practically of the same quality/ I don’t think that there are any significant differences among different brands of (category) in terms of quality., 2) I do not think that the store brand of the store I commonly purchase from is of lower quality than a national brand, 3) Most brands of category offer similar levels of quality

3.2.9 Private label purchase intention

The intention of consumers to buy a private label is measured by five different items from Burton et al. (1998) and Byoungho Jin and Yong Gu Suh (2005): 1) Buying private label brands makes me feel good (content with private labels), 2) I love it when private label brands are available for the product categories I purchase, 3) Do not hesitate to buy private label 4) Considering value for money, I prefer private label brands to national brands, 5) When I buy a private label product, I always feel that I am getting a good deal.

3.2.10 Rebuy intention

The item which is used for the measurement of rebuy intention comes from Sinha and Batra (1999) and ZIelke and Dobbelstein (2007). Consumers intention to buy private label brands can be answered on a seven points scale with the following end points: Yes, definitely going to buy it and No, definitely not going to buy it.

3.2.11 Experience

The experience in a specific category is measured by the following items: 1) I am very familiar with this category product, 2) I know the different available brands well. These items are from a paper of Bailey (1999) and a paper of Machleit et al. (1993).

Construct Reference Question

Personal involvement Zaichkowsky (1985)  I would be interested in reading information about how the product is made

 I have a most preferred brand of this product  I would be interested in reading the Consumer

(28)

28

 I have compared product characteristics among brands of this product

Consumer knowledge Delvecchio (2001)  If someone asked me for help in selecting a category, I could provide reliable advice  Compared to my friends, I am knowledgeable

about the category

 In general, I am knowledgeable about the category

Economic recession Nielsen (2010)

A. Van Strien and B. Wierenga (2009)

 During economic downturn I buy private labels  When the economy improves I will continue

buying private labels

 During economic downturn I spend less money buying less national brands

 Economic recession has no influence on my expenditures

Price consciousness Batra and Sinha (2000) Grewel and Marmorstein (1994)

 When buying a brand of (category), I look for the cheapest brand available

 Price is the most important factor when I am choosing a brand of (category)

 Seeking for the best price is worth the effort Consumer innovativeness Byoungho Jin and Yong Gu Suh

(2005) Roehrich (2004)

 I often seek out information about new products and brands

 When I see a new brand on the shelf, I’m not afraid of giving it a try

 I am more interested in buying new than known products/ I like to buy new and different Perceived image retailer Vahie and Paswan (2006)

Knight and Kim, 2007 Brakuset al. (2009) Keller (2008)

Sweeney and Soutar (2001) Quelch and Harding (2012)

 I like the private label brand of this store very much

 This store does not care enough about the quality of its private label brand

 I would go out of my way to shop at this retailer brand

 Has an acceptable standard of quality  Has consistent quality

 Quality in comparison with national brands is high and improving

 Would be economical Perceived risk Paul C.S. Wua, Gary Yeong-Yuh

Yeh a, Chieh-Ru Hsiao (2010)

 If I were to purchase private label brand for use, I become concerned that product will not provide the level of benefits that I would be expecting  As I consider the purchase of private label brand

for use at home is that body probably will be harm, due to overuse of product

 As I consider the purchase of private label brand for use, I worry about whether the product will really perform as well as it is supposed to Perceived quality

variation

Batra and Sinha (2000) Dick et al. (1995)

 The national brand and the store brand of the store I commonly purchase from are practically the same quality / I don’t think that there are any significant differences among different brands of (category) in terms of quality

 I do not think that the store brand of the store I commonly purchase from is of lower quality than a national brand

(29)

29

Private label purchase intention

Burton et al. (1998)

Byoungho Jin and Yong Gu Suh (2005)

 Buying private label brands makes me feel good (content with private labels)

 I love it when private label brands are available, recommend it to friends

 Do not hesitate to buy private label

 Considering value for money, I prefer private label brands to national brands

 When I buy a private label brand, I always feel that I am getting a good deal

Rebuy intention Sinha and Batra (1999) Zielke and Dobbelstein (2007)

 Consumers intention to buy private label brands, used one item on a five-point scale with the end points ‘yes, defenitiely going to buy it’ and no, definitely not going to buy it’ translated into the following items:

 - After a positive experience with a private label, I am open to try other private labels

 - After a negative experience with a private label, I decide to never use a private label again

 - After a negative experience with a national brand, I decide to use another A brand  -After a negative experience with a national

brand, I decide to use private label Expierence Machleit et al. (1993)

Bailey (1999)

 I am very familiar with this category product  I know the different available brands well Table 4: Operationalization of the constructs

3.3 Pre-test

Malhotra (2007) strongly recommends performing a pre-test on the questionnaire. This pre-test can identify possible problems in the routing in the questionnaire, a lack of clarity of instructions or just glitches in wording or questions. Therefore a pre-test was conducted among eight respondents. After feedback of all eight respondents, the questionnaire was adjusted to a more understandable questionnaire. Besides this the routing in the questionnaire is optimized.

3.4 Population and sample

According to Malhotra (2007), the sampling design process consists of four different stages, respectively: 1) defining the target population, 2) determining the sampling frame, 3) selecting a sampling technique, 4) determining the sample size.

3.4.1 Target population

The target population is the larger population to whom the results of the research are to be generalized (Johnson and Christensen, 2010). In this research, the target population consists of consumers who are using make-up. In addition, the consumers should be familiar with make-up of private labels as well.

3.4.2 Sampling frame

(30)

30

3.4.3 Sampling technique

Selecting a sampling technique is the third stage of the sampling design process (Malhotra, 2007). For this research a combination of two sampling techniques are used: convenience sampling and snowball sampling, both are nonprobability methods, which means that not every member of the target population have the same chance to being a part of this research, some have no chance being included. Convenience sampling technique was chosen, because of the quick and inexpensive way to perform this research. Snowball sampling existed during the research, because respondents forwarded this research to their social network and shared in on social networks like Facebook and Twitter. This snowball effect does increase the sampling size of the research.

3.4.4 Sample size

The final step of the sampling design process is determining the sample size of the research. According to Malhotra (2007), the target for a research like performed in this thesis is a sampling size of 200 respondents. This research consists of a specific part of questions about three retailers and their private labels. The target sample size for each retailer was a minimum of 100 respondents; therefore the total sample size was targeted on 300 respondents.

This targeted sample size of 300 respondents is largely achieved. Besides this a minimum of 100 respondents for Kruidvat, Hema or Etos users is also achieved. The dataset consists of 334 respondents after deletion of 108 respondents with too many missing values. Because of the subject of the questionnaire, which is make-up, all respondents in the survey are female and all 334 respondents use make-up. It is possible that the other 108 woman didn’t use make-up and therefore a lot of missing values existed. The average age of the 334 respondents is 27 years old; the range of ages is from 13 to 59 years old. Other descriptives needed for the segmentation analysis will be discussed in chapter 4.

3.5 Reliability

There are several different aspects to check for reliability. One of these aspects concerns the internal consistency of the scales which are used in the questionnaire. This internal consistency concerns the degree to which the items that cover the scale “hang together” (Pallant, 2007). The most commonly used and well-known indicator of this internal consistency is the Cronbach’s Alpha coefficient. This Cronbach Alpha is determined for every construct. Malhotra (2007) describes the Cronbach’s Alpha as the average of all possible split-half coefficients resulting from different ways of splitting the scale items. The coeffient can have a value which varies from 0 to 1. In an ideal situation the Cronbach’s Alpha should have a value above 0.7 (DeVellis 2003). However this is an ideal situation, Malhotra (2007) uses a value of 0.6, if the value is below this number than the construct can be seen as unreliable and will not be used in the analysis. Before checking the Cronbach’s Alpha’s of each construct, some negatively worded items need to be reversed.

(31)

31 meer een huis merk make up te kopen”; “Als ik een negatieve ervaring heb met een A-merk, besluit ik om een ander A-merk te proberen.”In the perceived image retailer construct one item is worded negatively. This item is reversed: “De retailer geeft niet genoeg om de kwaliteit van het huismerk.”

Constructs Cronbach’s Alpha’s Number of items

Personal involvement 0,719 4

Consumer knowledge 0,925 3

Economic recession 0,601 4

Price consciousness 0,702 3

Consumer innovativeness 0,875 3

Perceived image retailer Etos 0,695 7

Perceived image retailer Hema 0,689 7

Perceived image retailer Kruidvat 0,736 7

Total perceived image retailer 0,762 21

Perceived risk 0,819 6

Perceived quality variation 0,583* 3

Private label purchase intention 0,885 5

Rebuy intention 0,328 4

Experience 0,761 2

* = Significant Cronbach’s Alpha’s were reached after deletion of 1 item.

α <.6

Table 5: Reliability analysis – Cronbach’s Alpha’s

The Cronbach’s Alpha for perceived quality variation is insignificant 0,449, α <.6. Therefore the third item is deleted from the construct, which resulted in a new reliable Cronbach’s Alpha of 0,583. This is still not a value of 0.6 or above, but since the value is this close to 0.6 we will take the variable Perceived Quality Variation within the estimated model. Another insignificant Cronbach’s Alpha (α <.6) is for the rebuy intention construct with a value of 0,328. After deletion of several items, the Cronbach’s Alpha did not become significant. Therefore the rebuy intention construct will not be a part of any further estimation. Hypothesis 11b has to be removed from this research. H11d: Positive experience with national brands leads to a greater intention of rebuy these national brands again.

All the other constructs which can be seen in the table 5 are reliable (α <.6) with the original set of items, thus without deletion of any item.

(32)

32 deeper into the results of the factor analysis, both the eigenvalues and the screeplot have a result of only one factor. For this reason it is decided to keep the constructs: perceived image retailer Etos, Hema and Kruidvat measured by 7 items.

When looking at the total perceived image retailer, the three above called perceived images of Kruidvat, Etos and Hema are taken together. The Cronbach’s Alpha resulted in a value of .762 which was expected with a high item number of 21.

3.6 Plan of Analysis

In the following chapter, chapter 4, an extensive description of the analyses and the results of these analyses are presented. These analyses are performed with the data obtained from the online questionnaire which was online in week 11 2013 (10 March – 17 March). Before analyzing the data in chapter 4, an overview of the descriptives of the data is given. After this process, the analyses are conducted. Firstly, regression analysis will be performed. The additional analysis in this thesis is the cluster analysis in order to perform segmentation. Clustering procedures are used to place customers in different groups based on the information on which segmentation is desired. Finally, in chapter 5, the conclusions of both analyses mentioned above will be presented in order to give an answer on the marketing research problem which is formulated in the introduction of this thesis.

3.6.1 Regression Analysis

Regression analysis is performed for two reasons. Firstly, a determination whether the independent variables explain a significant variation in the dependent variable in order to search out whether a relationship exist (Malhotra, 2007). Secondly, regression analysis is used to find out the strength of the relationship. This will be done by a determination of how much of the variation in the dependent variable can be explained by the independent variables. In order to provide an answer on all hypotheses formulated in the conceptual model a multiple regression analysis including all the constructs will be performed. Results of this multiple regression can be found in chapter 4: results. Since there are four mediators in the conceptual model, the causal step strategy of Preacher and Hayes (2008) is used to determine if these multiple mediators are really mediating the several relationships between the independent variables and the dependent variable.

The causal step strategy includes four different steps (Preacher and Hayer, 2008): In these steps the following legend will be used:

X = independent variable Y = dependent variable M = mediator

c*= the extent to which X is related to after controlling the mediatiors (Preacher and Hayer, 2008)

1) Perform a regression analysis of X on Y

2) Perform a regression analysis of M1 on X and M2 & M2 on X and M1 3) Perform a regression analysis of Y on X, M1 and M2

(33)

33

3.6.2 Latent Class Analysis

In order to create relevant subgroups of consumers regarding their buying behavior in the make-up market, an accurate way to divide consumers in homogeneous clusters is needed. There are several analyses methods which can provide in this need.

This thesis performs a latent cluster analysis by completing the following steps:

Figure 12: Latent class analysis steps

The first two steps of the latent class analysis will be described in this chapter. The other three steps will be discussed in chapter 4, the results. Latent class analysis is a type of clustering with a statistical approach calculating the probability of class memberships. To determine the ideal number of clusters, several optimal values of several cluster criteria are taken into account. These several cluster criteria are respectively CAIC, BIC, AWE and the Classification Error. Most of the time these cluster criteria yield different cluster numbers. Therefore good judgment and interpretation of these cluster criteria is needed.

In conclusion, the determination of the number of clusters is based on the outcome of the several criteria values. It is important that all the criteria values have the lowest value possible. Besides this fact it is known that the Bayesian Information Criteria (BIC) and the Consistent Akaike’s Information Criteria (CAIC) will decline as long as the number of clusters is growing. A high number of clusters is hard to use during interpretation, therefore this has to be taken into account when deciding on the number of clusters.

3.6.2.1 Problem formulation

Nowadays, during harsh economic times product managers in the make-up market are confronted with increasing levels of private labels. To understand the behavior of consumers in the mass make-up market and to target these consumers in the right way, it is important to know which factors influence consumer choice between national brands and private labels. The problem formulation of this cluster analysis is how to cluster this group of make-up consumers and how to target them in suitable way.

The cluster analysis performed in this thesis will therefore give answer to the following three sub questions:

- Are there relevant subgroups of make-up consumers regarding their purchase behavior in the mass make-up market?

- How can these segments be described?

3.6.2.2 Active variables and passive covariates

In the cluster analysis two types of variables will be taken into account. First, the active variables, these variables will be used in order to divide the group of respondents in several clusters. The active variables that will be used in the latent class analysis are six variables which are also used in the multiple regression analyses.

(34)

34 - Private label purchase intention

- Perceived image of retailer - Economic Recession - Perceived risk

- Perceived quality variation - Experience

Besides these active variables, there are several covariate variables which are described in chapter 4.1. These so called passive covariates are variables which will be used to define the clusters. This research has two types of passive covariates, respectively: economical and demographical covariates and make-up related covariates in order to describe the several clusters with as much details as possible.

3.6.3 Latent class analysis settings

The statistical program which is used to cluster the respondents is LatentGold™. In this program many settings can be used in order to find the most optimal number of clusters. Settings concerning the EM algorithm and the starting points are adjusted in order to retrieve the best cluster solution from the dataset. The EM value is doubled to 500 and the settings of random starting points are increased to 20 instead of the standard value of 10. Besides these settings, an assumption is made about the errors of the covariance. It is assumed that these are cluster independent. This will pays off in estimation with less parameter, since this will result in better models. Clustering solutions for one to five clusters are estimated in the Latent Class analysis. By a comparison on the several cluster criteria, the best cluster solution will be determined.

3.6.4 Latent class regression

Next to the latent cluster analysis, a latent class regression is performed. Where the latent cluster analysis results into the ideal number of clusters, the latent class regression shows the effect of the independent variables within these several clusters. For the latent class regression the dependent variable is private label purchase intention. The same independent variables are used for this latent class regression as used for the multiple regressions, namely: perceived risk, perceived image of the retailer, perceived quality variation, personal involvement, consumer knowledge, consumer

innovativeness, prices consciousness, economic recession and experience.

3.7 Development of the research

(35)

35

4 Results

This chapter will give an overview of the results of the several analyses. First, some general findings like economic and demographical covariate descriptives are described. After this the key results for the multiple regression and the two cluster analyses: hierarchical cluster analysis and latent class analysis are presented.

4.1 General findings

In table 6 below, the economic and demographical covariate descriptives are described for the 334 participating respondents. Table 7 gives an overview of the make-up related covariate descriptives. This table helps to get a good insight into the make-up market, respectively the several mass market make-up brands, the amount of monthly spend euros and the several points of sale.

Economic and demographical covariate descriptives

Gender % Age % Female 100 ≤ 15 years 16- 25 years 1.2 62.9 26 – 35 years 24.6 36 – 45 years 5.1 46 – 55 years 4.8 > 55 years 1.5

Household size % Highest educational level %

1 person 39.8 Junior high school 9.6

2 persons 3 persons 4 persons 27.8 11.4 12.6 MBO HBO WO 19.8 34.4 36.2 5 persons 4.5

More than 5 persons 3.9

Monthly net income of household %

< 500,- Euro 29.6 501 – 1.000,- Euro 21 1.001,- – 1.500,- Euro 15.9 1.501,- – 2.000,- Euro 18 2.001,- – 2.500,- Euro 9 2.501,- – 3.000,- Euro 2.7 3.001,- – 3.500,- Euro 1.2 3.501,- – 4.000,- Euro 4.001,- – 4.500,- Euro 4.501,- – 5.000,- Euro 0 0.3 0 > 5.000,- Euro 1.5

Table 6: Economic and demographical covariate descriptives

(36)

36

Table 7: Make-up related covariates descriptives Make-up related covariate descriptives

Usage eye make-up % Usage lip make-up %

Every day Few times a week Few times a month Few times a year

81.9 12.3 3 2.7

18.8 30.4 27.7 23.1

Usage teint make-up % Usage nail make-up %

52.9 22.2 10.5 14.5

15.2 21.6 36.9 26.2

Monthly make-up expenditures %

< 10,- Euro 43.5 11,- – 20,- Euro 31.8 21,- – 30,- Euro 14.7 31,- – 40,- Euro 1.8 41,- – 50,- Euro 5.1 51,- – 60,- Euro 0.3 61,- – 70,- Euro 0.3 > 70,- Euro 2.4

Make-up brand usage % out of 334 respondents

% Most used make-up brand %

L’Oréal Paris Maybelline Max Factor Miss Helen Rimmel Essence Etos Catrice Other brands Sephora Bourjois Douglas Gosh DA NYC Miss Sporty ICI Paris 59.3 49.1 45.5 45.5 41.9 41 37.4 36.2 36.2 22.5 22.5 17.4 17.4 12.9 12.3 11.7 10.5 L’Oréal Paris Max Factor Maybelline Catrice Rimmel Essence Other brands Etos Miss Helen Bourjois Douglas DA Sephora Miss Sporty Gosh ICI Paris NYC 17.4 12.6 11.1 8.7 7.8 7.8 7 5.4 4.8 3 3 2.7 2.4 2.1 1.8 1.5 0.9

Visited retailers to purchase make-up

% out of 334 respondents

% Most visited retailer to purchase make–up % Kruidvat Other retailers Etos Douglas DA Sephora ICI Paris Dirx 70.7 69.5 67.1 29.3 25.7 17.7 15.6 2.1 Kruidvat Etos Other retailers DA Douglas ICI Paris Sephora Dirx 43.4 24 11.4 7.8 7.8 4.2 0.9 0.6

Purchased private label % Degree private label purchase %

Yes No

100 0

Few times a week Few times a month Few times a year < Few times a year