Does it pay-off to go green?
A study to the effect on brand credibility and willingness to pay of adding
sustainable assortment at slow and fast fashion retailers
Lisa Stapel
June, 2016
Does it pay-off to go green?
A study to the effect on brand credibility and willingness to pay of adding
sustainable assortment at slow and fast fashion retailers
Lisa Stapel
Master Thesis
Faculty of Economics and Business
University of Groningen
June, 2016
Dr. C. Hofstede de Grootkade 11-53
9718 KA Groningen
tel. 06 46 10 88 41
lisastapel@hotmail.com
S2784904
Management Summary
Increasingly, major fashion brands and retailers are subjected to close scrutiny due to unsustainable practices such as unfair working conditions and environmental pollution. The negative media attention these brands received might have harmed their brand image. In response to these unsustainable production practices, fashion designers as Stella McCartney, but also high street fashion brands like H&M have developed sustainable apparel (Lundblad, 2016). For brands that offer sustainable alternative or want to do so in the future, it is interesting to know if it really pays off to go green.
Previous research focused on the motivations to engage in sustainable consumption and identified the characteristics of consumers who purchase sustainable apparel. Limited research has investigated if it is beneficial for brands to offer sustainable items. Existing studies point out there is an attitude-behavior gap; even though the attitude of consumers towards sustainable consumption is often positive, this often does not lead to actual purchase of sustainable apparel. This study has build upon previous research by investigating whether consumers would be willing to pay more for sustainable apparel and if sustainable assortments will contribute to a higher brand credibility.
In a controlled experiment, it has been tested whether the amount of sustainable jeans within an assortment, the fashion type and a premium price for sustainable items had an effect on brand credibility and willingness to pay. Participants were exposed to an assortment of a web shop of either a fast or slow fashion retailer. This assortment consisted of six jeans attached with a label indicating whether the jeans were sustainable or not. These assortments were either not, partly or fully sustainable. The jeans had price indication of 19.99 euros for a jeans from the fast fashion brand and 79.99 for a jeans from the slow fashion brand. In some conditions, the sustainable jeans were priced 25 per cent higher to see if respondents would be willing to pay a premium for sustainable items. After a short introduction about one the retailers, respondents were asked to evaluate the brand and give an indication of how much they would be willing to pay for a jeans from the assortment.
are partly sustainable, but only if the price level of sustainable items are 25 per cent higher. Furthermore, willingness to pay is also higher for fully sustainable assortments with a higher price level compared to when assortments are not sustainable and have regular prices. Next, this research found that perceived credibility positively mediates the relationship between level of sustainability and willingness to pay. Last, both fashion type and price level have an effect on willingness to pay, but solely fashion type influences perceived credibility.
Preface
During my Bachelor Communication at the Hanze University in Groningen several Marketing related subjects were covered in the course material. I really enjoyed participating in these courses and after three years I decided that I want to do a Master in Marketing once I would have obtained my Bachelor degree. I kept my promise and three years later, after I successfully completed the pre-Master, I started with the Master programme. I participated in many interesting courses, and especially the course ‘Retail Marketing’ broadened my horizons. The time flew and at the end of the year I already had to subscribe for a Master Thesis topic. When I found out that one topic was about Retail Marketing, I directly knew it would be the number one in my motivation to apply for a Thesis group. Luckily I got chosen and now, half a year later, my Master Thesis is finished. I would like to thank my first supervisor prof. dr. Laurens Sloot and my friends and family for their help and support.
Table of Content
Introduction 7
1 Literature Review 9
1.1 Sustainable Fashion Consumption 9
1.2 Brand Rankings 9
1.3 Consumer Profiles 11
1.4 Consumers’ Associations with Sustainability 11 1.5 Barriers to Sustainable Apparel Consumption 11
1.6 Consumer Characteristics and Motives 12
1.7 Consumers’ Evaluations about Sustainability 14
1.8 Labeling 14
2 Conceptual Model and Hypotheses 16
2.1 Level of Sustainability 16
2.2 Willingness to Pay 17
2.3 Perceived Credibility 17
2.4 Price Premium Sustainable Items 18
2.5 Type of Fashion: Fast versus Slow Fashion 19
2.6 Conceptual Model 20
3 Methodology 21
3.1 Research Design 21
3.2 Sampling Frame 22
3.3 Measurements 23
3.3.1 Direct Relationship of Level of Sustainability on Willingness to Pay 23 3.3.2 Measuring the Mediator Perceived Credibility 24
3.3.3 Price Premium Sustainable Items 25
3.3.4 Type of Fashion 25
4 Results 27
4.1 Sample Descriptives 27
4.2 Assortment Evaluation and Likelihood of Purchasing 28
4.2.1 Assortment Evaluation 28
4.2.2 Likelihood of Purchasing 28
4.3 Willingness to Pay 29
4.3.1 Level of Sustainability on Willingness to Pay 29 4.3.2 Price Premium Sustainable Items and Fashion Type on Willingness to Pay 31 4.3.3 All Independent and Control Variables on Willingness to Pay 31
4.4 Perceived Credibility 33
4.4.1 Correlation Analysis and Factor Analysis 33 4.4.2 Level of Sustainability, Price Premium Sustainable Items and Fashion Type on
Perceived Credibility 34
4.4.3 All Independent Variables on Perceived Credibility 34
4.5 Perceived Credibility as a Mediator 35
4.5.1 Correlation Analysis and Regression Analysis 36
4.6 Preferences and Willingness to Pay for Sustainable Apparel 38
4.6.1 Consumer Preferences 38
4.6.2 Willingness to Pay 39
5 Conclusions and Recommendations 41
5.1 Conclusions 41
5.2 Recommendations 42
5.3 Limitations 43
5.4 Suggestions for Future Research 44
References 45
Introduction
The documentary ‘The True Cost’ sheds light on the dramatically increasing human environmental costs of purchasing clothes in the fast fashion industry. The collapse of the textile factory Rana Plaza Bangladesh in 2013 was a major disaster with implications for the fashion industry. Even when taken into account these circumstances, fast fashion can be seen as a double-edged sword. Some argue that the clothes consumers buy at big chains as H&M, Zara, and Primark are at cost of factory workers in developing countries. However, on the other hand bringing back the production from those countries will also have a huge impact on production workers and in more general, society.
apparel, however in actual shopping situations environmental concerns are most of the time not taken into account (Butler and Francis, 1997).
Chan and Wong (2012) note that it is important to promote eco-fashion consumption and demonstrate that store-related attributes such as store design and environment and shop convenience positively influence consumers eco-fashion purchase decision making. However, this relationship can be weakened by the price premium level of the apparel. Joergens (2006) also shows that the ‘Made in’ sign in clothing labels is not always an indication of the conditions in which the apparel has been produced. It might be helpful to include a label with ethical information, but Joergens (2006) questions if the implementation would be effective since only a minimal number of consumers read a label when shopping for apparel.
1!Literature Review
1.1!Sustainable Fashion Consumption
Apparel is used to fulfill basic needs as coverage and modesty. However, fashion also expresses symbolic meanings and shares mutual social understanding (Bly, Gwozdz and Reisch, 2015). Solomon and Rabolt (p. 6, 2004) define fashion as “a style of dress that is accepted by a large group of people at any given time.” In the past several years, taking care of the environment has become of growing interest among fashion brands and retailers. Fashion designers as Stella McCartney, but also high street fashion brands like H&M increasingly develop sustainable apparel (Lundblad, 2016). Joergens (p. 361, 2006) defines ethical fashion as “fashionable clothes that incorporate fair trade principles with sweatshop free labor conditions while not harming the environment or workers by using biodegradable and organic cotton.”
The attitude of consumers towards sustainable consumption is often positive, however this often does not lead to actual purchase of sustainable apparel. Due to the challenges clothing retailers and manufacturers, but also other parties in the textile industry face with regard to sustainability, it is important to gather insights about consumers regarding (un)sustainable fashion consumption. Cohen and Van der Wijst (2015) point out that youth are important actors in (un)sustainable fashion consumption, since they frequently buy clothes, purchase more, have more spare time, and continue buying apparel during their whole life cycle.
1.2!Brand Rankings
investigated brands communicate in some way about sustainability. However, most of this information is unsubstantial and does not focus on relevant issues (Konietzko, Van Woersem, and Simpson, 2014). These results are of main concern regarding environmental and social impacts on the fashion industry especially when taking into account the water scarcity, the severe labor conditions and when looking at the leather production in particular, the widely used chemicals but also animal cruelty.
Konietzko, Van Woersem, and Simpson (2014) make a distinction between eleven subsectors in the fashion industry: Sport & Outdoor Clothing, Fashion Retailers, Surf, Beach & Swimwear, Sport & Outdoor Clothing & Shoes, Premium Brands, Shoes & Footwear, Lingerie & Underwear, Casual Clothing, Jeans & Denim, Luxury Brands, and Baby & Children Clothing. Brands are listed on the ‘Greenwashing Alert’ when they receive an ‘E’ label when there is no information, or no substantial and relevant information available about their sustainability practices. Brands in the Shoes & Footwear and Lingerie & Underwear are most often reported on the Greenwashing Alert list, 52 and 50 per cent respectively. Notably, amongst others premium brands as Louis Vuitton and Hugo Boss, but also other well-known brands like New Yorker, Hollister, and Carhartt are all listed on the Greenwashing Alert. When compared to the Rank a Brand sustainability criteria, performance is quite well developed in the Fashion Retailer and Sport & Outdoor Clothing subsectors, whereas it is underdeveloped in the Luxury Brands and Baby & Children Clothing categories.
1.3!Consumer Profiles
The GfK Roper Consumer Styles model (2016) makes a distinction between different consumer profiles in order to better understand behavior of consumer groups. The groups are positioned according to two factors; whether the consumer is materialistic or rather post-materialistic and if the consumer is more risk and pleasure seeking or risk-averse. The different profiles consist of Dreamers, Homebodies, Settled, Adventurers, Rational-realists, Open-minded, Organics, and Demanding. The segments that focus most on post-materialism, quality, and authenticity are the Open-minded, Organics, and Demanding. These three groups form 38 per cent of the total population in The Netherlands and pay more attention to sustainability than the other profiles (Cohen and Van der Wijst, 2015). Appendix III illustrates the distribution of consumer profiles among the Dutch population. Cohen and Van der Wijst (2015) explain sustainability as the extent to which companies deal with both labor conditions, society, and environment as well as the environmental impact caused by the products itself. Based on sales of a couple of different retailers, Appendix IV gives an overview in which consumer profiles customers of those retailers are located. These retailers are both Dutch and international retailers, both offline and online.
1.4!Consumers’ Associations with Sustainability
Results of a research about sustainable consumption about youth in The Netherlands indicates that among the age group 16 to 35 years old, 57 up to 76 per cent does not know to what extent fashion brands are sustainable (Cohen and Van der Wijst, 2015). In general, retailers with lower prices are often seen as less sustainable whereas retailers with hardly any private labels are more often associated with sustainability. Another finding suggests that Dutch retailers are considered more sustainable than brands from other countries, especially concerning department stores. The fact that retailers with lower prices are seen as less sustainable has do to with consumers’ decision making based on heuristics; ‘if a product is relatively cheap it will probably not be sustainable’ whereas more expensive products are in general associated with being more sustainable (Cohen and Van der Wijst, 2015).
1.5!Barriers to Sustainable Apparel Consumption
(Cohen and Van der Wijst, 2015). Other important barriers are finding the right information and price; more than half of the respondents argue that sustainable apparel is relatively more expensive (Cohen and Van der Wijst, 2015). The research also shows that higher knowledge about certain quality labels lead to more sustainable consumption. Limited knowledge as a barrier is confirmed in a study by Hiller Connell (2010) who found that a majority of consumers have limited awareness of the relation between environmental issues and fashion production and consumption.
Furthermore, there is a strong relationship between caring about the environment and sustainable consumption, since 96 per cent of consumers who care about the environment often take into account environmental and social factors when making a purchase decision. The other way around holds for people who barely care about the environment; they show very little sustainable fashion consumption. Cohen and Van der Wijst (2015) also point out that consumers who consider themselves as ‘fashionable’ more often engage in sustainable consumption behaviors.
Another barrier for consumers to engage in sustainable fashion consumption is the availability of environmental friendly clothes (Hiller Connell, 2010). Consumers indicate that there are relatively few choices for sustainable clothes as compared to un-conscious apparel of what the majority of retailers offer. Furthermore, sustainable produced apparel is perceived as unattractive and therefore less stylish and unfashionable which makes it less likely to be purchased (Hiller Connell, 2010). However, the finding of Cohen and Van der Wijst (2015) is contradicting to this statement since they suggested that fashion-oriented consumers engage more often in sustainable consumption.
1.6!Consumer Characteristics and Motives
shopping as they also buy clothing for men especially older women, buying for their husband or partner for example. Youth between 16 and 25 engage less in sustainable fashion behavior, in particularly until 20 years old. This might have to do with the fact that they have less budget and therefore price plays an important role.
Consumers from 16 to 20 years old buy clothing most frequently. For female youngsters, shopping is an important social activity. Furthermore, consumers with a higher education level indicate to care more about the environment than people with a lower education level. 42 per cent of high-educated people feel it as a moral duty to purchase pro-environmental products in contrast to 18 per cent of low-educated consumers (Cohen and Van der Wijst, 2015). In contrast, Butler and Francis (1997) suggest that there is only a limited influence of age and education on both attitudes and behavior.
In general, women from 16 to 35 years old indicate that the most important factors (in descending order) when purchasing apparel are price, fit, comfort, ways to combine, size, style, shape, fabric, colors, and lifetime (Van der Wijst, 2014). When making purchasing decisions, consumers might consider environmental issues. However, other criteria such as price, style, and fit may be more dominant in making a purchase decision (Butler and Francis, 1997). Consumers take into account many different criteria when shopping for clothing. This suggests that when purchasing apparel, it is more difficult for consumers to consider environmental attitudes next to all other criteria to apply than for products in other categories.
and see both the shortness of time and limited availability of clothing as a benefit in purchasing fast fashion since it will contribute to create unique, hard-to-copy looks.
Consumers of fast fashion apparel differ in their purchase and consumption, post-consumption evaluation, and divestment of stages from slow fashion consumers (Watson and Yan, 2013). In the purchase and consumption stage, fast fashion consumers enjoy buying large quantities of clothes for low prices, whereas slow fashion consumers want to buy versatile apparel to build a long-lasting wardrobe. At the post-consumption evaluation, fast fashion consumers indicated to be satisfied after purchasing but this changed after the consumption stage, mainly because the apparel would be damaged after a few times wearing it. However, they indicated to be satisfied because of the fact that they expect that the relatively cheap clothes will damage within a couple wears. In the divestment stage, both groups of consumers would donate, re-sell, pass along or reuse their clothes, but the fast fashion consumers indicated that they were likely to throw clothing away where slow fashion consumers will not.
1.7!Consumers’ Evaluations about Sustainability
Giving consumers information about a company’s sustainability practices influences their evaluation of the company and also purchase intentions. When described as not being sustainable, consumers’ evaluation about the company and their intentions to purchase are negative. More specific, consumers are more sensitive to companies with poor sustainability practices than companies with a strong focus on sustainability (Choi and Ng, 2011). Consumers’ reactions towards a company’s poor environmental sustainability are more negative than to a company’s poor commitment to economic sustainability. When having information about a company’s poor environmental sustainability, consumers respond unfavorable to low prices (Choi and Ng, 2011).
1.8!Labeling
Yan, Ogle, and Lee, 2012, p. 51). In contrast to Joergens suggestion that relatively few consumers read product labels, Hyllegard, Yan, Ogle, and Lee (2012) found that the frequency of reading clothing labels among consumers has increased in recent years.
When a tag is attached to a product containing highly explicit messages such that the definitions of terms used are clear and the presented information is factual, evaluations of the clothing labels will be more favorable than when the information provided is less explicit. Evaluations of labels on apparel will also be more positive when a label contains a social responsibility logo of a third-party (Hyllegard, Yan, Ogle, and Lee, 2012). The favorability of labels including highly explicit messages and social responsibility logos of third-parties leads to positive attitudes toward the clothing brand, which subsequently affects purchase intentions (Hyllegard, Yan, Ogle, and Lee, 2012). Consumers might not read product labels containing social responsibility information in the same way. Consumers that are more interested in fashion and shopping use social responsibility labels more often. They are also more familiar with these specific labels and are more likely to purchase apparel that is social responsible (Gam, Ma, and Banning, 2014).
2!Conceptual Model and Hypotheses
Major fashion brands and retailers received negative publicity about their unethical production practices (Shaw, Hogg, Wilson, Shui and Hassan, 2006). In respond to these unaccepted behaviors, apparel retailers increasingly offer sustainable alternatives. These ethical retailers are often relatively small companies having low economies of scale which makes it hard for them to compete with major fast fashion retailers (Arnold, 2009).
Even though the demand for fast fashion characterized by its relatively low prices and low quality is not diminishing, people increasingly value quality and more sustainable alternatives. In recent years, big fashion chains have introduced sustainable products and product lines such as H&M with its Conscious collection. Goworek, Hiller, Fisher, Cooper and Woodward (2012) argue that cheap, disposable apparel and sustainable fashion can co-exist and at the same time increase demand, even when being offered within the same fashion chain.
However, the choice of conscious, stylish and fashionable apparel seems limited (Hiller Connell, 2010). Shaw, Hogg, Wilson, Shui and Hassan (p.439, 2006) point out that “consumers are calling for ethical products to be available on the High Street so they are convenient and easy to access, while also demanding that they conform to fashion.” This makes it interesting for fashion retailers to introduce sustainable clothing lines, also since they have the position to influence and encourage consumers to purchase sustainable fashion (Goworek, Hiller, Fisher, Cooper and Woodward, 2012).
2.1!Level of Sustainability
2.2!Willingness to Pay
Attitudes and perceptions influence behavior, but consumer attitudes alone are a poor predictor of actual purchase behavior (De Pelsmacker, Driesen, and Rayp, 2005). An explanation for the attitude-behavior gap could be that people tend to give socially desirable answers on questions about attitudes. These attitudes of consumers might therefore be more positive than their actual behavior (King and Bruner, 2000). Instead of using attitudes to predict behavior, willingness to pay is a more valid measurement of buying behavior since it is closely related to actual behavior and therefore a realistic indicator of actually purchasing sustainable or ethical products (De Pelsmacker, Driesen, and Rayp, 2005).
The question this research addresses is whether consumers are willing to pay a premium for brands that offer sustainable apparel. Previous research found that a significant amount of consumers would be willing to pay more for sustainable or ethical alternatives. In their study, Ha-Brookshire and Norum (2011) found that half of the respondents would be willing to pay more for organic, sustainable and locally or nationally produced cotton. For a shirt that costs 30 dollars, consumers were willing to pay 5 dollars more if it was either organic, sustainable or US-grown. Results also showed that age and gender influenced willingness to pay for organic and sustainable shirts. Willingness to pay for organic cotton shirts was higher for younger than older consumers and women were more likely to pay a premium for sustainable shirts (Ha-Brookshire and Norum 2011).
Due to the fact that existing research found that consumers in general are willing to pay more for sustainable items, it expected that when the level of sustainability increases, the willingness to pay for the assortment also will increase.
H1: Level of sustainability positively influences willingness to pay.
2.3!Perceived Credibility
the validity of those claims. The meaning of a product with a claim like ‘organic’, ‘fair trade’ or ‘eco-friendly’ might be unclear and consumers are skeptical towards its credibility (Keller, 2012).
If a fashion chain would consider introducing apparel that has been produced ethically, would people believe that these products are sustainable? “Brand credibility describes the extent to which customers see the brand as credible in terms of three dimensions: perceived expertise, trustworthiness, and likeability” (Keller, 2012, p. 117). Brand credibility is part of brand judgments, in which consumers form opinions about a brand and evaluate that particular brand based on different associations they have in mind. Credibility measures whether consumers; evaluate the brand as good at what it does, think the brand is concerned about its customers, and simply like the brand or not.
Previous research points out that corporate credibility of a brand or organization could increase when organizations implement a corporate image strategy that either positions the company as being innovative, being concerned about the environment, or taking care of the community (Keller and Aaker, 1990). “Corporate societal marketing could affect all three dimensions of credibility, because consumers may think of a firm willing to invest in CSM as caring more about customers and being more dependable than other firms, at least in a broad sense, as well as being likable for ‘doing the right things.’” (Keller, 2012, p. 424).
Based on the suggestions from previous research, it is expected that the perceived credibility of fashion brands mediates the relationship between level of sustainability and willingness to pay. When consumers evaluate a brand as more credible, the willingness to pay for sustainable fashion items will go up.
H2: Perceived brand credibility of the retailer positively mediates the relationship between level of sustainability and willingness to pay.
2.4!Price Premium Sustainable Items
more expensive than regular products. This suggestion is confirmed by De Pelsmacker, Driesen, and Rayp (2005) who state that the prices of fair trade coffee are higher than non-fair trade coffee. They also found that 10 per cent of the respondents in their research was willing to pay the price premium of 27 per cent for coffee when a fair-trade label is attached to the product.
As a fashion retailer, a way to account for the accumulated costs of sustainable production is to pass on these costs or part of these costs to the customer by asking a price premium for sustainable items. One might then question whether consumers would still purchase these items if the prices will rise with a certain percentage or amount. This research will therefore investigate if consumers are willing to pay for sustainable items that are priced higher than their regular alternatives offered by fashion retailers. Also, since consumers might be aware of the fact that the costs of the production of sustainable items are higher compared to apparel that is not sustainable, credibility might play a role as well. When sustainable items have the same prices as clothes that are not, consumers might be skeptical and wonder if these items are indeed sustainable. It is therefore expected that the perceived credibility will be higher when sustainable alternatives have a higher price compared to regular items. Since previous research (Ha-Brookshire and Norum, 2011; Pelsmacker, Driesen, and Rayp, 2005) found that some consumers are willing to pay a premium for sustainable items, it is also expected that charging a price premium for sustainable fashion will have a positive influence on willingness to pay.
H3a: Compared sustainable and unsustainable fashion with similar prices, charging a price premium for sustainable items will have a higher influence on perceived brand credibility of the retailer.
H3b: Compared sustainable and unsustainable fashion with similar prices, charging a price premium for sustainable items will have a positive influence on willingness to pay.
2.5!Type of Fashion: Fast versus Slow Fashion
H4a: Compared to fast fashion, slow fashion is positively related to the perceived brand credibility of the retailer.
H4b: Compared to fast fashion, willingness to pay will be higher for slow fashion brands.
3!Methodology
3.1!Research Design
This quantitative research is conducted with a controlled experiment. An experiment is used to manipulate the independent variables in order to measure their effect on the dependent variable and to control for extraneous variables. After conducting the experiment, valid conclusions can be drawn and valid generalizations can be made to a larger population (Malhotra, 2009). This experiment has a fractional factorial statistical design. In statistical designs, different basic experiments are conducted at the same time. A factorial design measures the effect of the independent variables at multiple levels and allows for interactions between variables (Malhotra, 2009). The experiment consists of a 3 x 2 x 2 between-subjects design, with level of sustainability as independent variable with three levels, fashion type as second independent variable with two levels, and premium price sustainable items as a third independent variable with two levels.
The two levels of the third variable are regular prices and premium prices for sustainable items. Since premium prices are only charged for sustainable fashion items due to higher production costs, the two conditions in which the level of sustainability is not sustainable and where a price premium is asked are irrelevant and therefore not taken into account in this research. This implies that only a portion of the full factorial design will be measured and therefore a fractional factorial design is used consisting of 10 conditions as illustrated in table 1 and 2. Data is gathered by inviting respondents to fill in a survey with pen and paper. These respondents are randomly selected and randomly assigned to one of the ten groups.
REGULAR PRICE Type of fashion
Level of sustainability Fast fashion Slow fashion Not sustainable Fast fashion not
sustainable
Slow fashion not sustainable
Adding a conscious line Fast fashion with
a conscious line
Slow fashion with a conscious line
Fully sustainable Fast fashion fully sustainable
Slow fashion fully sustainable
PREMIUM PRICE SUSTAINABLE ITEMS Type of fashion
Level of sustainability Fast fashion Slow fashion
Not sustainable x x
Adding a conscious line Fast fashion with
a conscious line
Slow fashion with a conscious line
Fully sustainable Fast fashion fully
sustainable Slow fashion fully sustainable
Table 2: 3 x 2 x 2 between-subjects design with premium prices for sustainable items
As already indicated, group 7 and 10 are irrelevant and will not be used in this research. Therefore, the experiment consists of the next ten groups:
Group 1: Fast fashion – not sustainable – regular price Group 2: Fast fashion – with a conscious line – regular price Group 3: Fast fashion – fully sustainable – regular price Group 4: Slow fashion – not sustainable – regular price Group 5: Slow fashion – with a conscious line – regular price Group 6: Slow fashion – fully sustainable – regular price
Group 8: Fast fashion – with a conscious line – price premium sustainable items Group 9: Fast fashion – fully sustainable – price premium sustainable items Group 11: Slow fashion – with a conscious line – price premium sustainable items Group 12: Slow fashion – fully sustainable – price premium sustainable items
For each group, a webpage of a web shop is designed which all have a subtle difference in order to manipulate the independent variables. In the next subsections, an explanation is given of how the three independent are manipulated to measure their effects on the dependent variable. In Appendix V, all ten different websites are illustrated. Control variables used in this research are age, education level and income. Since the web shop only includes jeans for women, only female respondents participated in this study.
3.2!Sampling Frame
2012), this group might be of greatest interest in this research. Before distributing the survey, a small pretest will be done with a group of participants to exclude difficulties and
inconsistencies (Malhotra, 2009). The surveys will be distributed in universities among students but also in offices where young professionals are working since both groups form a part of the population of interest. For each condition, 27 participants will be invited which will lead to a total sample size of 270 respondents.
3.3!Measurements
3.3.1! Direct Relationship of Level of Sustainability on Willingness to Pay
Participants are shown a webpage of a web shop from a fictitious fashion brand with six jeans. All jeans are attached with a label which indicate how ‘sustainable’ the particular jeans are. Labels containing information about social responsibility of a brand might be effective means to encourage ethical clothing (Gam, Ma, and Banning, 2014). With these labels, companies can communicate about their sustainability practices and convey this information to consumers. Therefore, this experiment used the sustainability labels initiated by Rank a Brand (2016) in which brands are ranked from A to E, with A as most sustainable and E as least sustainable brand. These jeans included a label with either an ‘E’ indicating that the apparel is relatively unethical and unsustainable or an ‘A’ indicating that the clothing is relatively sustainable and a conscious choice. Respondents were exposed to one of the following options:
1)! Jeans with only ‘E’ labels indicating that the assortment is unsustainable.
2)! Four jeans with an ‘E’ label and two jeans with an ‘A’ label indicating that the assortment is partly sustainable; four jeans are unsustainable and the other two jeans are part of a conscious line.
would be willing to pay for the jeans of their choice. By measuring the difference in outcomes between all groups, it can be concluded whether level of sustainability influences willingness to pay.
Image 1: Example jeans with E-label Image 2: Example jeans with A-label
3.3.2! Measuring the Mediator Perceived Credibility
An existing scale from Keller (2012) is used to measure perceived brand credibility, that consist of three sub-constructs: expertise, trustworthiness, and likability (Keller, 2012). Since the brands that the respondents have to evaluate are fictitious brands and therefore they are not familiar with them, participants are asked to rate one of the brands at a first glance. Participants will indicate how they perceive the brand in terms of expertise, trustworthiness and likability to measure the perceived credibility of each brand. The three sub-constructs consist of the following items that will be used in the experiment:
Expertise
!! How knowledgeable are the makers of this brand? !! How innovative are the makers of this brand? !! How much do you trust the makers of this brand? Trustworthiness
Likability
!! How much do you like this brand? !! How much do you admire this brand? !! How much do you respect this brand?
These questions are measured using a 7-point Likert scale. In order to test the hypothesis, first a factor analysis is performed to see whether the nine questions could be combined into one variable that measures perceived credibility. After performing the factor analysis, mediation analysis is conducted to investigate whether perceived credibility indeed mediates the relationship between level of sustainability and willingness to pay.
3.3.3! Price Premium Sustainable Items
In order to measure whether charging a price premium for sustainable items will have a higher influence on perceived brand credibility and willingness to pay, under the pictures of the jeans in the web shop a price indication of the jeans is given. In groups one till six, the prices of all jeans, either not sustainable with an E-label or sustainable with an A-label, are the same and thus regular prices within an assortment. For fast fashion brands the regular price is 19.99 euros whereas the jeans from the slow fashion brand have a regular price of 79.99 euros. In groups eight till twelve, a price premium is asked for the jeans that are sustainable. For these groups, prices of sustainable, ‘green’ jeans are 25 per cent higher than the regular jeans within the assortment. This implies that the jeans with an E-label have a regular price, and the jeans with an A-label are priced 25 per cent higher than the regular prices. Fast fashion sustainable jeans will then be 24.99 euros and slow fashion sustainable jeans 99.99 euros. By comparing the groups that all have a regular priced assortment to those that have sustainable jeans with a price premium, it can be investigated whether participants perceive a brand that offers sustainable items for higher prices as more credible and if they are willing to pay a higher price for sustainable items.
3.3.4! Type of Fashion
this experiment: ‘Sarah Vie’ as slow fashion brand and ‘Propenza’ as fast fashion brand. With an introduction text in the survey about either the slow or fast fashion brand, the brand and its characteristics are shortly described. Also, each brand has its own logo shown in the web shop. Both logos are illustrated in image 3 and 4. Half of the groups will be exposed to the web shop of ‘Sarah Vie’ and the other half to ‘Propenza’. By comparing both groups, it can be investigated whether perceived credibility and willingness to pay will be higher for the slow fashion brand compared to the fast fashion brand.
Image 1: Logo Propenza Image 2: Logo Sarah Vie
The introduction text of Sarah Vie:
“Sarah Vie is a fashion brand with its own web shop and offers clothing in the relatively higher price segment. Every season the brand introduces a new collection that is highly accepted by the international fashion press. Sarah Vie is a brand that is known by its timeless fashion items which lasts for several seasons or even years. A notable competitor of Sarah Vie is Abercrombie & Fitch.”
The introduction text of Propenza:
4!Results
4.1!Sample Descriptives
In total, 270 participants filled out a survey, each condition was filled in 27 times. When cleaning the data, no out of range, logically inconsistent or extreme values are identified, probably because the surveys are filled in with pen and paper. However, in some cases missing responses were found and these cases were removed from the dataset. After data purification, 257 cases were left. In this study, 257 female respondents participated with an average age of 23. Approximately 78 per cent of the participants has a relatively low income of 0 to 1000 euros a month. Almost half of the respondents completed high school (44.7%) and 45.1 per cent even has a Bachelor or Masters’ degree from either a University or University of Applied Sciences, which indicates that the sample is relatively highly educated.
Since only females who are on average relatively young participated in this study, results might not be generalizable among the whole Dutch population. However, previous research found that especially young women buy often more clothing and pay more attention to quality labels than men (Cohen and Van der Wijst, 2015; Hyllegard, Yan, Ogle, and Lee, 2012). Therefore, this particular group is of great interest in this research and results could be generalized among these young, Dutch females.
Demographical data like age, education level and income should be equally distributed among the ten different conditions. To test whether there are significant differences in age, a One-way ANOVA with age on condition is performed. The One-way ANOVA shows that there are significant differences in age between all groups F(9, 247) = 5.097, p = .000*. This implies that
age is not equally distributed among the conditions. Another One-way ANOVA with education level on condition is performed. This One-way ANOVA was also significant, F(9, 247) = 2.095, p = .044**, meaning that education level is also not equally distributed among all conditions. A
third One-way ANOVA with income on condition shows that income is also not equally distributed among the conditions F(9, 247) = 6.077, p = .000*. Table 3 shows how age,
Condition Frequency Mean Age Completed high school Obtained Bachelor or Master degree Monthly income between 0 and 1,000 euros a month Monthly income over 1,000 euros a month Condition 1 26 24 26.9% 61.5% 73.1% 26.9% Condition 2 25 24 44% 48% 64% 36% Condition 3 25 27 32% 64% 60% 40% Condition 4 27 21 59.3% 29.6% 92.6% 7.4% Condition 5 26 21 64% 34.6% 69.2% 30.8% Condition 6 26 21 57.6% 26.9% 88.5% 11.5% Condition 8 26 24 38.5% 57.7% 80.8% 19.2% Condition 9 25 30 20% 64% 52% 48% Condition 11 26 21 50% 30.8% 100% 0% Condition 12 25 21 56% 36% 100% 0% Total 257 23 44.7% 45.1% 78.2% 21.7%
Table 3: Distribution among groups
4.2!Assortment Evaluation and Likelihood of Purchasing
4.2.1! Assortment Evaluation
In the experiment, respondents were asked to rate the assortment of the brand they were exposed to on a 7-point Likert scale. Descriptives show that on average, respondents rate the assortment as slightly positive (4.75 out of 7 points). Before moving on to the hypotheses testing, it might be interesting to see if assortment evaluation differs per level of sustainability. A One-way ANOVA of level of sustainability on assortment evaluation is performed. This One-way ANOVA was not significant, F(2, 254) = .894, p = .410. This finding indicates that level of sustainability does not influence assortment evaluation. Assortments with a ‘green’ line or assortments that are fully sustainable are not evaluated as more positive than assortments that are not sustainable.
4.2.2! Likelihood of Purchasing
sustainable assortment, rather than buying a jeans from a brand that only offers jeans that are not sustainable.
4.3!Willingness to Pay
4.3.1! Level of Sustainability on Willingness to Pay
Next, the effects of level of sustainability, fashion type and price level on willingness to pay will be measured. In order to analyze whether or not the willingness to pay differs per level of sustainability, a One-way ANOVA of level of sustainability on willingness to pay is performed. Results indicate that level of sustainability influences willingness to pay, however these findings are not significant F(2, 254) = 1.238, p = .292, as shown in table 4, model 1a. Additional analysis illustrate that only for a slow fashion brand that charges a premium price for sustainable clothes, willingness to pay significantly differs per level of sustainability (F1, 49) = 3.007, p = .089***. People are willing to pay around 8 euros more when the assortment is fully sustainable as compared to when the same assortment is only partly sustainable, shown in table 4, model 1b. Model 1a Model 1b Willingness to pay Not sustainable €42.73 - Partly sustainable €42.96 €55.96 Fully sustainable €46.97 €64.06 Model fit df 2, 254 1, 49 F 1.238 3.007 Sig. .292n.s. .089***
Table 4: ANOVAs with level of sustainability on willingness to pay Model 1a: level of sustainability on willingness to pay
Model 1b: level of sustainability on willingness to pay with slow fashion and price premium for sustainable items
***significance level of 10%
β = 12.144, t = 1.734, p = .089*** as illustrated in table 5, model 2b. This implies respondents are willing to pay more for sustainable jeans with a higher price than regular jeans that are not sustainable, only when these jeans are offered by slow fashion brand. Since the effect is marginally significant, results have to be interpreted with caution.
Model 2a Model 2b Intercept β 41.957 .000* β 51.915 .000* Level of Sustainability β 4.812 .130n.s. β 12.144 .089*** Model fit R2 .009 .058 df 1, 256 1, 50 F 2.308 3.007 Sig. .130n.s. .089***
Table 5: Regressions with level of sustainability on willingness to pay Model 2a: level of sustainability on willingness to pay
Model 2b: level of sustainability on willingness to pay with slow fashion and price premium for sustainable items
***significance level of 10%
Additional analyses show that when unsustainable assortments including jeans with regular prices are compared with fully sustainable assortments where sustainable items have a premium price and therefore are priced 25 per cent higher than regular jeans. Results of the regression reveal a marginally significant effect R2 = .034, F(1, 102) = 3.510, p = .064*** and willingness to pay is higher for jeans in a fully sustainable assortment, β = 7.507, t =1.873, p = .064*** (marginally significant).
4.3.2! Price Premium Sustainable Items and Fashion Type on Willingness to Pay
To investigate whether price level and fashion type also have an effect on willingness to pay, first a regression analysis is performed with price level regressed on willingness to pay. The regression analysis was marginally significant, R2 = .011, F(1, 256) = 2.753, p = .098***. The price level influences willingness to pay, β = 4.296, t = 1.659, p = .098***. Another regression analysis with fashion type regressed on willingness to pay showed that fashion type significantly influences willingness to pay, R2 = .347, F(1, 256) = 135.666, p = .000* with β = 23.971, t = 11.648 and p = .000*.
4.3.3 All Independent and Control Variables on Willingness to Pay
Model 3
Intercept β 4.877 .383n.s.
Level of Sustainability β .966 .702n.s.
Price Premium Sustainable Items β 4.023 .049**
Fashion Type β 29.087 .000* Age β .448 .004* Education Level β 1.037 .178n.s. Income β 5.186 .002* Model fit R2 .451 df 6, 256 F 34.187 Sig. .000*
Table 6: Effects on willingness to pay
*significance level of 1% **significance level of 5%
Model 4 Model 5
Intercept β 7.791 .207n.s. β 44.483 .007*
Level of Sustainability β -1.920 .548n.s. β 3.687 .345n.s.
Price Premium Sustainable Items β 2.076 .408n.s. β 6.382 .056***
Age β .493 .001* β -.577 .534n.s. Education Level β .668 .462n.s. β 2.159 .152n.s. Income β 5.359 .001* β 8.123 .099*** Model fit R2 .301 .074 df 5, 126 5, 129 F 10.417 1.970 Sig. .000* .088***
Table 6: Effects on willingness to pay Model 4: Fast fashion
Model 5: Slow fashion
*significance level of 1% **significance level of 5%
4.4!Perceived Credibility
4.4.1! Correlation Analysis and Factor Analysis
To test the hypotheses whether level of sustainability, price level, and type of fashion influence on perceived credibility, first one variable measuring perceived credibility should be computed. A correlation analysis showed that all 9 items that measure perceived credibility significantly correlate (for all items p = .000, except for ‘Expertise: trust’ and ‘Expertise: innovate’ where p = .001 and ‘Expertise: innovate’ and ‘Likability: respect’ where p= .016). This indicates that when consumers consider the brand as an expert, they also trust and like the brand, and vice versa.
Next, a factor analysis is performed with the items that should measure perceived credibility. The KMO measure is .856, which higher than .5 and Bartlett’s Test for Sphericity is significant (p = .000). However, when looking at the communalities all items score higher than .4, except ‘Expertise: Innovative’. Another factor analysis is performed with the 8 remained items. Here, the KMO measure is .852, which is higher than .5 and also Bartlett’s Test is significant (p = .00). The communalities are all higher than .4, so therefore it is decided to continue the factor analysis with 8 items.
one factor. A reliability analyses with the 8 items showed that the Cronbach’s alpha is .870, which is higher than .6. Based on the fact that it is already an existing and developed scale (Keller, 2012) and one factor has a relatively high eigenvalue, the eight items are taken together and the mean of perceived credibility is calculated in order to continue with the analyses.
4.4.2! Level of Sustainability, Price Premium Sustainable Items and Fashion Type on Perceived Credibility
In order to analyze whether or not level of sustainability influences perceived credibility, a regression analysis is performed with level of sustainability regressed on perceived credibility. The regression analysis was marginally significant, R2 = .015, F(2, 256) = 3.805, p = .052***. The level of sustainability influences perceived credibility, β = .268, t = 1.951, p = .052***.
Next, it is investigated whether price level influences perceived credibility. A regression analysis indicated that price level does not influence perceived credibility, R2 = .000, F(1, 256)
= .076, p = .783 with β = .031 and t = .276. Furthermore, to measure whether fashion type influences perceived credibility, a regression analysis is performed with fashion type regressed on perceived credibility. The regression analysis was significant, R2 = .034, F(1, 256) = 9.065,
p = .003*. This finding suggests that the type of fashion influences perceived credibility, β = .328, t = 3.011, p = .003*.
4.4.3! All Independent Variables on Perceived Credibility
Model 6 in table 8 shows the effect of level of sustainability, a price premium for sustainable items and fashion type on perceived credibility. The overall model is significant, R2 = .049, F(3, 256) = 4.378, p = .005*, and level of sustainability significantly influences perceived credibility β = .278, t = 1.975, p = .049**; when the level of sustainability increases, the perceived credibility will be higher. Charging a price premium for sustainable items does not have a significant effect on perceived credibility β = -.027, t = -.235, p = .814. This finding indicates that when fashion brands charge higher prices for sustainable apparel compared to regular clothes that are not sustainable, participants do not evaluate the brand as more credible. On the contrary, fashion type does have a significant influence on perceived credibility β = .328, t = 3.025, p = .003*. Slow fashion brands are being evaluated as more credible than fast
Model 6
Intercept β 3.853 .000*
Level of Sustainability β .278 .049**
Price Premium Sustainable Items β -.027 .814n.s.
Fashion Type β .328 .003* Model fit R2 .049 df 3, 256 F 4.378 Sig. .005*
Table 8: Effects on perceived credibility
*significance level of 1% **significance level of 5%
4.5!Perceived Credibility as a Mediator
It is expected that perceived brand credibility mediates the relationship between level of sustainability and willingness to pay. Mediation implies that the effect of the independent variable on the dependent variable is explained by another variable; the mediator (Baron and Kenny, 1986). This will be the case when the influence of level of sustainability (X) on willingness to pay (Y) is dependent on perceived credibility (Z) as depicted in figure 1. Before conducting a mediation analysis, first a correlation and regression are performed.
4.5.1! Correlation Analysis and Regression Analysis
First, it is investigated whether both perceived credibility and willingness to pay correlate. A correlation analysis showed that perceived credibility and willingness to pay significantly correlate (r = .198, p = .001). When people evaluate the brand at first glance as more credible, they are willing to pay more for an item of its product assortment. Next, a regression analysis with perceived credibility on willingness to pay is performed to measure whether perceived credibility has an influence on willingness to pay. The regression analysis was significant, R2 = .039, F(1, 256) = 10.436, p = .001*. This finding indicates that perceived credibility has a significant influence on willingness to pay, β = 4.554, t = 3.230, p = .001*.
4.5.2! Mediation Analysis
To test for mediation, the following equation should be used:
Total effect (C) = direct effect (C’) + indirect effect (A*B) (Hayes, 2013).
When mediation exist, the mediator is caused by the independent variable level of sustainability while at the same time it is a cause for the dependent variable willingness to pay. To test for mediation, the mediate macro by Hayes (2013) is used. The following four steps of Baron and Kenny (1986) must hold in order to conclude whether perceived credibility (Z) indeed mediates the relationship between level of sustainability (X) and willingness to pay (Y).
1)! The effect of X on Y (C) has to be significant
The first condition for mediation is that the effect of level of sustainability on willingness to pay has to be significant. The total effects model in the output of the mediation analysis shows that the overall model is not significant F(1, 256) = 2.3085, p = .1299. Therefore, it can be concluded that the first condition does not hold.
2)! The effect of X on Z (A) has to be significant
The second requirement is that the effect of level of sustainability on perceived credibility has to be significant. The overall model is significant, F(1, 256) = 3.8049, p = .0522*** which implies that the second condition holds.
3)! The effect of Z on Y has to be significant
perceived credibility is also significant, p = .0024*. It can be concluded that the third condition also holds.
4)! Including Z must make the effect of X on Y (C’) non-significant for full mediation, or lower its effect substantially for partial mediation
The last condition requires that when including perceived credibility, the effect of level of sustainability on willingness to pay should be non-significant for full mediation or lower its effect substantially for partial mediation. There is indeed an indirect effect, since the effect of sustainability on willingness to pay is not significant, p = .2468.
Even though the first condition does not hold, all other requirements are met. The total effect of level of sustainability on willingness to pay is not significant, however it is relatively close to a significance level of 90 per cent. Besides, the p-value in the indirect effect is higher than in the total effects model. Kenny and Judd (2014) recently suggested that the power of the direct effect path C is often less valuable than the power of the indirect effect path A and B. Therefore, it could be assumed that in this case mediation exists.
4.5.3! Mediation Analysis with Covariates
The mediation analysis could be performed when also accounting for covariates. These covariates include the other independent variables price level and fashion type, but also the control variables age, education level, and income.
1)! The effect of X on Y (C) has to be significant
The first condition for mediation is that the effect of level of sustainability on willingness to pay has to be significant. The total effects model in the output of the mediation analysis shows that the overall model is significant F(6, 256) = 34.1873, p = .000*. However, the effect of sustainability is not significant, p = .7024 and therefore the first condition does not hold. 2)! The effect of X on Z (A) has to be significant
3)! The effect of Z on Y has to be significant
In the third condition, the effect of perceived credibility on willingness to pay has to be significant. The overall model is significant F(7, 256) = 31.3001, p = .000*. Also, the effect of perceived credibility on willingness to pay is also significant, p = .0047*. It can be concluded that the third condition is also met.
4)! Including Z must make the effect of X on Y (C’) non-significant for full mediation, or lower its effect substantially for partial mediation
The last condition requires that when including perceived credibility, the effect of level of sustainability on willingness to pay should be non-significant for full mediation or lower its effect substantially for partial mediation. There is indeed an indirect effect, since the effect of sustainability on willingness to pay is not significant, p = .9492.
Results of the mediation analysis including covariates is comparable to the mediation analysis where the covariates are left out. Even though the p-values in the total effects model and indirect effect are much higher when adding covariates in the analysis, the only condition that is not met is that the effect level of sustainability on willingness to pay is not significant in the total effects model. Also here, the p-value in the indirect effect is higher than in the total effects model. As already mentioned in the previous mediation analyses, Kenny and Judd (2014) found that the power of the direct effect path C is often less valuable than the power of the indirect effect path A and B. Therefore, it could be assumed that also in this case mediation exists.
4.6!Preferences and Willingness to Pay for Sustainable Apparel
Previous results are based on analyses on a chain or brand level, in which the effects on perceived credibility and willingness to pay are measured based on a whole assortment of a brand. It might also be interesting to see if there are differences in for example preferences and willingness to pay within an assortment, measured on a product level.
4.6.1! Consumer Preferences
was partly sustainable, two out of six jeans were part of a “green” line and thus sustainable. In the two other conditions, all six jeans were either “green” and thus sustainable or not sustainable. When comparing the amount of respondents that preferred the “green” jeans in the partly sustainable condition to the respondents who chose the same jeans in the other two control conditions, it could be measured if consumer preferences for sustainable alternatives are higher. A cross table with Chi-square indicated that preferences for sustainable jeans are significantly higher (59,6%) compared to the control conditions (30.8%), Chi-square (1) = 9.762, p = .002*. Even when the conditions where the price level of sustainable alternatives is 25 per cent higher are added in the analysis, preferences for sustainable jeans are still significantly higher (49.5%) compared to the control conditions (33.8%), Chi-square (1) = 6.374, p = .012**.
4.6.2! Willingness to Pay
Even if preferences for sustainable fashion items are higher than for the same items that are not sustainable, this does not imply consumers are also willing to pay more for sustainable apparel compared to when the same clothes are not sustainable. Therefore, it is interesting to measure whether willingness to pay differs for a jeans from a “green” line, where the other part of the assortment is not sustainable, compared to when the same jeans are offered in an assortment where level of sustainability does not differ (either in an assortment where all jeans are sustainable or in an assortment where all jeans are not sustainable). The conditions in which the assortment is partly sustainable were compared to the control conditions in which the assortment was either not or fully sustainable.
Next, it is also investigated if there are differences in willingness to pay for sustainable jeans and jeans that are not sustainable within the same assortment. An independent samples t-test with choice of jeans and willingness to pay is performed. Only the conditions in which the assortment is partly sustainable and where the prices of sustainable jeans and regular jeans are the same were selected. The independent samples t-test was not significant, t(49) = .44, p = .662. The average willingness to pay for unsustainable jeans (M = 43.18, SD = 19.19) does not differ from the average willingness to pay for sustainable jeans (M = 40.86, SD = 18.28). From both analyses, on a product level, it can be concluded that the willingness to pay for sustainable jeans is not higher than for jeans that are not sustainable.
4.7!Overview of Supported and Rejected Hypotheses
Table 8 gives an overview whether the hypotheses are either supported or rejected.
Hypotheses Results
H1 Level of sustainability positively influences willingness to pay. Partly supported H2 Perceived brand credibility of the retailer positively mediates the
relationship between level of sustainability and willingness to pay.
Supported H3a Compared sustainable and unsustainable fashion with similar
prices, charging a price premium for sustainable items will have a higher influence on perceived brand credibility of the retailer.
Rejected H3b Compared sustainable and unsustainable fashion with similar
prices, charging a price premium for sustainable items will have a positive influence on willingness to pay.
Supported H4a Compared to fast fashion, slow fashion is positively related to
the perceived brand credibility of the retailer. Supported H4b Compared to fast fashion, willingness to pay will be higher for
slow fashion brands. Supported
5!Conclusions and Recommendations
5.1!Conclusions
This study wanted to investigate whether it pays off for fashion brands to offer sustainable alternatives, either in a ‘green’ line or in a fully sustainable assortment. On average, participants rated the assortment as slightly positive, however the findings indicate that it was slightly unlikely that they would buy a jeans from one of the assortments. Results show that in general, respondents did not evaluate a brand with a partly or fully sustainable assortment as more positive than brands that offer only items that are not sustainable. Also, the likelihood of purchasing a jeans was not higher in partly or fully sustainable assortments compared to assortments with only unsustainable options.
These findings could partly explain the result that level of sustainability does not have a significant effect on willingness to pay. Likelihood to purchase was not higher for partly or fully sustainable assortments as compared to when assortments were unsustainable. Hence, it could be assumed that willingness to pay would not differ between unsustainable or partly and fully sustainable brands either. Even though in general level of sustainability did not have an effect on willingness to pay, for the slow fashion brand where the price level of sustainable alternatives was 25 per cent higher, level of sustainability did have an influence on willingness to pay. This finding shows that participants were willing to pay around 8 euros more for a jeans when the assortment was fully sustainable as compared to when the same assortment was only partly sustainable, only when the price of sustainable jeans was higher.
The type of fashion also has a significant effect on willingness to pay. Participants were willing to pay more for jeans from a slow fashion brand than from a fast fashion brand. Last, age and income influence willingness to pay as well. Older respondents and participants with a higher income were willing to pay more for a jeans than younger respondents and participants with a lower income.
which implies that perceived credibility is higher for slow than for fast fashion brands. Furthermore, it is found that even though there is no direct effect of level of sustainability on willingness to pay, perceived credibility positively mediates this relationship.
Previous conclusions were based on analysis on chain or brand level, in which the effects on perceived credibility and willingness to pay were measured based on a whole assortment of a brand. Additional analyses tried to investigate if there are also differences in for example preferences and willingness to pay within an assortment, measured on a product level. Results show a significantly higher amount of participants preferred a sustainable jeans over a non-sustainable jeans. Even if the price of the non-sustainable jeans were 25 per cent higher, still the sustainable options were preferred. Yet, willingness to pay for sustainable jeans was not significantly higher than for jeans that are not sustainable. In addition, willingness to pay was also not significantly higher for sustainable jeans than for jeans that are not sustainable within the same assortment.
In conclusion, level of sustainability in general has no significant direct effect on willingness to pay, however under certain conditions level of sustainability does have an influence on the willingness to pay a premium. Willingness to pay is higher for assortments from slow fashion brands that are fully sustainable than when those assortments are partly sustainable, but only if the price level of sustainable items are 25 per cent higher. Furthermore, willingness to pay is also higher for fully sustainable assortments with a higher price level compared to when assortments are not sustainable and have regular prices. Next, this research found that perceived credibility positively mediates the relationship between level of sustainability and willingness to pay. Last, both fashion type and price level have an effect on willingness to pay, but solely fashion type influences perceived credibility.
5.2!Recommendations
conditions for their factory workers. The negative media attention could damage the image of these fashion brands which in turn may lead to turnover losses. The most obvious solution to prevent this issue is to transform their business format towards sustainable production, but this also implies higher production costs (Pollin, Burns and Heintz, 2004; Stenzel, 2012). This research found that it might be beneficial for these brands to improve their business and offer apparel that is produced and manufactured with care and consideration for people and environment.
For both slow and fast fashion brands that currently only offer fashion items that are not sustainable it would pay off to transform their business to 100 per cent sustainable production, only when they also increase the price level of their apparel. For these brands, it should be examined whether the benefits of changing the business format of the brand to go ‘fully green’ will outweigh the costs, as sustainable production is relatively more expensive. However, brands having the image of caring about the environment and community receive greater acceptance and are seen as more credible (Keller and Aaker, 1990). This study found that perceived credibility mediates the relationship between level of sustainability and willingness to pay, which implies that the more sustainable alternatives fashion brands offer, the more credible the brand will be, which will in turn lead to a higher willingness to pay for their assortments.
Adding just several sustainable items in a ‘green’ line such as H&M does with its conscious line will not have any impact on willingness to pay compared to assortments with only items that are not sustainable. Fashion brands that already have a ‘green’ or ‘conscious’ line in their assortment, could consider making their whole assortment sustainable since consumers will only pay more for brands that are fully sustainable. When brands consider going fully sustainable, the question which consumers to target might arise. Depending on the positioning of the brand, targeting older consumers with a higher income would generate the highest profit, since they are willing to pay the most for sustainable apparel.
