The effect of framing on Willingness to Pay for sustainable products  

The effect of price framing on Willingness to Pay for sustainable

products

1 Voorwoord

Aangezien ik de Master Sustainable Business and Innovation ga doen na mijn bachelor, wilde ik erg graag mijn scriptie schrijven over een onderwerp gerelateerd aan duurzaamheid. Het onderwerp Green Consumption was dus perfect, aangezien het ook gerelateerd was aan mijn tweede favoriete onderwerp, zijnde consumentengedrag. Ik heb ontzettend mijn best gedaan om er een mooi verslag van te schrijven, en heb me niet laten tegenhouden door de chaos veroorzaakt door een wereldwijde pandemie en sluiting der universiteiten. Ik ben er erg trots op en vind dit een mooie afsluiting van mijn bachelor. Maar ik heb het natuurlijk niet volledig alleen gedaan, dus bij deze een bedankje aan degenen die mij dit (af en toe toch wel zware) process hebben door gesleept.

Allereerst dank voor mijn begeleider Samuel Franssens. Zijn feedback was altijd van

geweldige kwaliteit en nog eens snel ook, en zonder zijn enthousiasme voor mijn onderwerp was het nooit zo’n mooi verslag geworden. Dan verder, in onbepaalde volgorde: dank aan mijn familie voor de opvang tijdens de coronachaos; dank aan Judith voor de gezellige middagen bakken; dank aan Jet voor de spoedcursussen statistiek; dank aan Nikki voor de werksessies op skype; and last but definitely not least, dank aan Nicole voor de super gezellige colleges de afgelopen paar jaar!

Ik hoop jullie allemaal gauw weer te zien om dit te vieren! Groetjes,

2 Statement of Originality

This document is written by Student Alexander van Wees who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

3 Abstract

Consumer’s willingness to buy sustainable products is a very important factor in the fight against climate change. To find a way to change consumer habits, this study investigated

whether the framing of the price of regular and sustainable products influenced price perceptions and Willingness to Pay. This was done through an online, between-subjects experiment. The prices were framed either as sustainable products being expensive, or regular

products being too cheap. The WTP for these conditions were compared to the WTP in a control and control+information condition. The moderating role of general environmental concern and mediating role of product-level environmental concern were also tested. It was

found that the effect of framing sustainable products as too expensive, lowers the WTP for these products in people with low environmental concern. Other than that, there were no significant framing effects, or moderation or mediation effects. However, it was also found that the participants were willing to pay significantly more for the sustainable coffee than the

price of the regular coffee. This can be explained by the fact that people the respondents indicated that they felt like regular coffee was too cheap, while they saw the price of sustainable coffee as fair. These findings imply that it would be prudent to not use the framing

that sustainable products are expensive, but also find people with low environmental concern and invest in raising their concern. Furthermore, the difference in price perceptions could open the door for a tax penalizing brands that are not sustainable. Since information was once

again confirmed as an ineffective way to influence behavior, future research should not focus on this topic. Instead, it should focus on making one proper measure for environmental

4

Table of content

Main effect of framing... 12

Moderation analysis ... 14

Mediation analysis ... 16

Moderated mediation analysis ... 17

Conclusion ... 17

General discussion ... 18

References ... 21

Appendix A: Questionnaires ... 26

5

Introduction

More and more people are starting to realize that climate change is a threat (e.g. Flohn, 1980; McCarthy, 2001). In the time since then, research has focused on the role of policymakers (e.g. Lutsey & Sperling, 2008; Ingold & Fisher, 2014), companies (e.g. Boiral, 2006; Finke, Gilchrist, & Mouzas, 2016), and consumers (see Peattie, 2010 for a full review) in causing and mitigating this crisis. This thesis will look at Green Consumption, which means it will focus on the role of consumers in the context of climate change. According to the UN Environmental Program, the goal of Green Consumption is improving quality of life for everyone on the planet, by (among other things) using fewer resources during production, and distributing the money made from production fairly along the full supply chain (UNEP, 2001). The current research will look at Green Consumption from a marketing perspective. This means it will look at the attitudes and behaviors of consumers, but not the ecological outcomes of these behaviors (Jackson, 2005). This topic was chosen because the willingness of consumers to change their behavior is one of the most important factors in making global consumption more sustainable (Peattie, 2010).

Previous research into Green Consumption has yielded positive results. Over a decade ago it was already found that the presence of eco-labels leads to a higher WTP (Loureiro & Lotade, 2005; Pelsmacker, Driesen, & Rayp, 2005). It was also found that people prefer local products and products where animals were treated humanely over those that do not meet these criteria (Howard & Allen, 2010). Part of these attitudes were explained later, when it was found that information can change product preferences. Adding information made a product’s sustainable qualities more important to consumers than price, where first the reverse was true (Goucher-Lambert & Cagan, 2015).

These positive research outcomes paint a happy picture for the future of Green

consumption, but unfortunately, the reality is different. Even in the (Western) countries where Fairtrade products are most popular, their market share is only around 8%. The country with the highest per capita expenditure on Fairtrade products in 2015 was Switzerland, with €57,7 (Lernoud & Willer, 2016), which is only 0,017% of their yearly consumption expenditure ("National economy", 2020). These numbers fly directly in the face of the positive research outcomes that are consistently reported on the topic of Green Consumption. Combine this with the fact consumers’ biggest problem with sustainable products is perceived high price (Röös and Tjärnemo, 2011; Grunert, 2011), and you may begin to see how the perception of this price could influence consumer preferences.

In this thesis, an experiment will be conducted where the price of regular products will be framed as too cheap, and the price of sustainable products as too expensive. The

differences in WTP between these conditions will be compared to two control conditions. The experiment conducted in this thesis will give insight into how framing affects Willingness to Pay, specifically for a pack of sustainable coffee. It will also tell us if these hypothetical changes are influenced by changing consumer attitudes.

Literature review

As mentioned in the introduction, one of the main reasons people state for not buying sustainable products is the perceived high price (Röös and Tjärnemo, 2011; Grunert, 2011). There are explanations to this perception. One is the reality that sustainable products are on average 10-50% more expensive than their regular counterparts (‘Pricing of sustainable products’, 2010). However, by only focusing at the difference in price, consumers miss valuable information that can influence their choice. There are three types of information consumers miss when focusing purely on the price of a product: the fact that a higher price

6 does not necessarily mean higher profits; the fact that sustainable products can often be seen as an investment; and the missing externalities from the price of regular products. Each type is discussed in this section, after which it is argued why consumers miss this information.

Firstly, consumers tend to underestimate the true cost of a product, and attribute price differences between products more to differences in profits for the producer than to

differences in costs. This erroneous attribution then leads them to believe that a product that is priced higher than its counterpart, is priced unfairly (Bolton, Warlop, & Alba, 2003). This implies that people may find sustainable products unfairly priced, but this has never been properly researched. However, in reality, the higher price of sustainable products is caused by the higher cost of sustainability. For example, sustainable production of different metals is magnitudes more expensive than the current methods (Steen & Borg, 2002). The same thing goes for sustainable farming practices (Wichelns, Oster, & Plant, 2006), and implementing active design elements like solar panels (Zhang, Platten, & Shan, 2011).

Secondly, although the price tag of sustainable products is higher, consumers may see them as an investment. For example, an eco-friendly washing machine may cost more than a regular one, but can save you money in the long run by conserving energy and water (La Faille, 2020). The same thing applies to products like solar panels, that have a high cost up front, but can save almost all yearly energy costs (Hayat et al., 2017). Consumers miss these benefits when focusing solely on purchase price, and arrive at inaccurate conclusions

regarding the value of a product.

Thirdly, consumers usually do not consider the externalities of their purchases.

Externalities are costs and benefits incurred in the production and use of products, that are not calculated into the purchase price (Bator, 1958). Economically speaking these costs should be included when calculating the true price of a product, but consumers do not do this. For a time, even economists considered externalities as an exception to the rule (Ayres & Kneese, 1969). The externalities relevant here are those related to climate change. Some examples would be that agricultural losses and deaths due to heatwaves have already cost billions of euros, and these will become more commonplace as time goes on (Stern, 2007). The costs of mitigating climate change are astronomical as well; for example the cost of heightening dykes to combat rising sea levels will also be billions of euros (Kok & Hoekstra, 2008).

One of the reasons consumers do not consider all these factors when judging product prices is framing. Sustainable products are consistently framed as more expensive than regular products, in both retail and research contexts (Peattie, 2010). News and magazine articles explaining how organic food works title the section about costs ‘Why is organic food so expensive?’, and use words like ‘more/higher costs’ to describe the prices of sustainable products (Parnes, 2020). Even outlets like Forbes use similar titles and phrases when discussing sustainable industries (Banjeree, 2019). This framing neglects two facts. On the one hand, it misses that the price for the sustainable product is the true price, including all costs necessary to make a product without exploiting land or person, and including all externalities. The other fact is that the regular product is actually too cheap, which causes environmental damage (Peattie, 2010). Since framing is a powerful tool that can influence people’s perceptions without them realizing it (Tversky & Kahneman, 1981), the framing of the price of sustainable products as ‘too expensive’ leads consumers to see the price for regular products as the default, and the price for sustainable products as one with a premium (Peattie, 2010). Because people overwhelmingly favor defaults over other presented options (Thaler & Sunstein, 2009; Johnson & Goldstein, 2003), this makes consumers favor the regular product over the sustainable version. It also anchors consumers to the regular price (for anchoring: Strack, Martin, & Schwarz, 1988), leading them further down the path of the normalization of this price.

7 It is clear that consumers have the wrong idea about how much a product costs to make, and what the environmental costs are of making products too cheap. They also attribute a higher price to profit differences, instead of cost differences. These erroneous perceptions are then strengthened by the framing of the price of sustainable products as too expensive. All of these factors can help explain the attitude-behavior gap seen with Green Consumption: if consumers do not know why a sustainable product costs more, and think these products are too expensive, they will not be interested in buying them.

In this experiment, the framing of the price of regular and sustainable products will be manipulated. Because it has been found that sustainable products are framed as more

expensive than regular products, the effect of this frame on Willingness to Pay for sustainable coffee will be measured. To see if changing the frame changes WTP, regular products will be framed as too cheap in another condition. The WTP in these conditions will be contrasted to the WTP in two control conditions; one measuring the effect of information but no frame, and one condition where neither a frame nor control are present.

Some research has been done proving the effectiveness of framing in the context of Green Consumption, but this was related to abstract vs concrete mindsets and loss and gain frames (White, Macdonnal, & Dahl, 2011; Reczek, Trudel, & White, 2018; Tu, Kao, & Tu, 2013). There has also been research into the effects of price framing on price perceptions. In Germany it has been found that stating the price differences between health plans in an absolute euro value made people switch plans more often, compared to stating it in a percentage point difference in relation to their payroll tax (Schmitz & Ziebart, 2017). The study found that this effect was persistent even when prices of these plans had not changed compared to the previous year. The effect was attributed to the fact that the price differences were more salient when stated in euro values than in percentages. In the context of consumer research, similar findings were done. For example, stating a price difference in dollars made it seem more significant to consumers than stating it in a percentage (Chen, Monroe, & Lou, 1998). Research was also done into the effects of price framing on the judgment of dynamic pricing. When the price that a disadvantaged consumer paid was framed as an exorbitant price minus a discount, the consumer felt this price was more fair compared to when they simply paid the higher price without this discount frame (Wesstein, Monroe, & Kukar-Kiney, 2013). This built on earlier research that found that the mere presence of a promotional signal increases sales, even when the price had not been cut (Inman & Hoyer, 1990). In the context of Green Consumption, a study closely related to this one was done a while ago. In the study by Okada & Mais (2010) they change the price framing by changing which product (regular or sustainable) the respondents get as a reference price, and then ask the WTP for the other product. In that study it was found that setting the sustainable product as the reference price, led to a higher WTP for this product than when the regular one was set as the reference. This led them to conclude that framing the price of a regular product as a discount leads to a higher WTP for sustainable products, compared to framing the price of sustainable products as premium (Okada & Mais, 2010). In this study, the regular product will be the reference price in all conditions. The framing will be done by placing different messages comparing the price of regular and sustainable products.

In this study, we will conduct an experiment in which we’ll manipulate the price of sustainable and regular products. In each condition, the WTP for sustainable coffee will be measured. In one condition, the price of sustainable products will be framed as more expensive than regular products. It is expected that this lowers WTP for sustainable coffee compared to the control condition. In the condition where regular products are framed as too cheap, a higher WTP for sustainable products compared to the control condition is expected. Specifically, the highest WTP is expected in the ‘regular too cheap’ condition, and the lowest in the ‘sustainable is expensive’ condition, with the control condition in between. This is H1.

8 A lot of research has been done into the antecedents of sustainable purchase behavior. Environmental concern is one interindividual characteristic that positively affects sustainable consumption. For example, it was found that people with high environmental concern are more susceptible to sustainable ads featuring nature than those with low concern (Schmuck, Matthes, Naderer, & Beaufort, 2018). It has also been found that environmental concern is a significant predictor of sustainable behavior for high-effort recycling programs (Schultz & Oskamp, 1996). In general, environmental concern is a good predictor of pro-environmental behavior (Czap & Czap, 2010). A meta-analysis confirmed that environmental concern is a major determinant of pro-environmental behavior (Joshi & Rahman, 2015). Environmental concern also increases both Willingness to Use and Willingness to Pay for biofuels (Pagiaslis & Krontalis, 2014). Those with high environmental concern were also more likely to use electronic invoicing than those with low concern (Albayrak, 2013). They were also more likely to use green energy (Bang, Ellinger, Hadjimarcou,& Traichal, 2000), and request environmental information (Bamberg, 2003) than those with low concern. So, environmental concern has direct and mediating effects on behavior.

It has also been found that environmental concern moderates the relationship between certain antecedents and behavior. For example, environmental concern influences purchase behavior in both retail and tourism contexts. In retail contexts, those with higher vs lower environmental concern are more open to green marketing (Davari & Strutton, 2014). In the tourism context, environmental concern moderates the relationship between travel intention and souvenir purchases. For those who travel with a nature immersion goal, higher

environmental concern makes people less willing to buy souvenirs that hurt the natural environment, like corral pieces, than those with low concern. It also makes them spend more on the souvenirs that they do buy (Chang, Kong, & Chen, 2017). Even in managerial contexts environmental concern matters. It has been found that managerial environmental concern moderates the relationship between sustainable innovation and profit (Ar, 2012).

Environmental concern also moderates the relationship between epistemic stances and the purchase of organic food (Kuswah, Dhir,& Sagar, 2019). It has also been found that people have been concerned about many different environmental issues for decades now (Zimmer, Stafford, & Stafford, 1992), which implies that a lot of purchasing behavior is guided by environmental concern. So, environmental concern also moderates behavioral effects.

However, while most people have concern for the environment on a general level, this concern disappears at the product-level (Grunert, Hieke, & Wills, 2014). In the

aforementioned study, people indicated they had concerns related to sustainability issues, like equal payment and sober resource usage, in the production of food. However, when asked how concerned they were about coffee production, their answers changed. Now, they were concerned about issues related to packaging, or about no issues at all (Grunert et al., 2014). The disappearance of concern at the product-level could help explain the attitude-behavior gap discussed earlier. After all, if you are not concerned about the negative impact of a certain product, you can buy it without this purchase clashing with your environmentally friendly attitude. This notion is at least partially confirmed by the fact that recently it was discovered that general environmental concern is not a strong determinant of purchase behavior (Cerri, Testa, & Rizzi, 2018). However, no other research has either confirmed or refuted the disappearance of concern at the product-level, or measured if this influences purchasing behaviors. The current research will do both, by measuring concern at a general and product level, and relating this to WTP for a sustainable product.

The studies mentioned here do not only show that environmental concern has a

significant influence on behavior, it also shows that it influences behavior in different ways. It can either directly explain it, mediate a relationship between two variables, or moderate the severity of an effect. It should also be noted that people with different levels of environmental

9 concern respond differently to price framing. Those with high levels of concern respond better to price promotions framed as gains, while those with low levels of concern preferred

messaged emphasizing the reduction of loss. Those with medium levels of concern responded equally to both frames (Wesstein, Asgari, & Siew, 2014). In this study, general

environmental concern will be seen as a moderator, and product-level concern will be seen as a mediator. It is expected in this experiment that for those with high general-level

environmental concern, the effect of the ‘regular too cheap’ frame will be higher than for people with low environmental concern. The reverse is expected for the ‘sustainable is expensive’ frame; for people with high environmental concern, the lowering of WTP is expected to be less severe, while for those with low concern it is expected to be more. This is H2. H3 is that it is expected that product-level concern will mediate the relationship between framing and WTP. In the ‘regular too cheap’ condition, this frame is expected to raise

concern, which will in turn raise WTP. In the ‘sustainable is expensive’ condition, this frame is expected to lower concern, which in turn lowers WTP. H4 holds that general-level concern is moderates this relationship, in such a way that it strengthens the rise of product-level cocnern for those with high general-level concern in the ‘regular too cheap’ condition, and weakens the lowering of product-level concern for those with high general-level concern in the ‘sustainable is expensive’ condition.

It should also be noted that only the highest levels of concern have a big impact on behavior, while the impact on lower levels is mediocre to non-existent. This is because the relationship between environmental concern and behavior is non-linear (de Langhe, Puntoni, & Larrick, 2017). This may help explain some of the contradicting results found by earlier studies, where some studies find significant effects of concern on behavior, but others do not. This can also be seen in the studies mentioned earlier. In a lot of them, the results to not simply indicate a higher vs lower effect for higher environmental concern. Instead, they indicate a moderate to low effect at moderate levels of concern, a high effect at high levels of concern, and no effect at low levels of concern.

The current research will test the relationships between framing, WTP, and different levels of environmental concern. With this experiment, the following research question will be answered:

How does the framing of the price of sustainable products influence Willingness to Pay, through the activation of product-level concern for the environment?

A conceptual model has been drawn below to illustrate the expected relationships and effects:

Framing

Willingness to Pay

Product-level

environmental concern

General environmental

concern

10

Figure 1: Conceptual model

Research design

The research question will be answered using an online experiment. The experiment will start with the measure for the moderator, general concern for the environment. This will be

measured with a questionnaire consisting of 14 questions about different environmental issues concerning food production, adapted from Grunert et al. (2014). Issues range from the use of child labor in food production, to the overuse of plastic in packaging. Respondents can indicate on a scale of 1 (not concerned at all) to 7 (extremely concerned) how concerned they are about each issue. The full questionnaire can be found in Appendix A.

The experiment will have four conditions: the control condition, the control + info condition, the ‘sustainable is expensive’ condition, and the ‘regular is too cheap’ condition. The dependent variable is the Willingness to Pay for a pack of sustainable coffee. This

variable will be measured by first giving the price of a regular pack of coffee, and then asking how much, on a scale of €0 to €15, people would be willing to pay for the sustainable version.

This will be done with the following text: “A 500 gram pack of Douwe Egberts Aroma Rood

filtercoffee costs €5,99 at Albert Heijn. How much would you be willing to pay for a 500 gram pack of sustainable coffee?”. In the control condition, this text is all there will be present; there will be no additional information or framing. This condition reflects the basic state most people are in, with no salient frame or sustainability information, and will serve as a baseline.

The effect of sustainability information on consumers has been researched thoroughly, but the results have been mixed. It has been found that information can change product preferences. Information makes the sustainability function of products more important in the eyes of consumers, while lowering the importance of price (Goucher-Lambert & Cagan, 2015). Information also raises WTP (Loureiro & Lotade, 2005; Rousseau, 2015), even when it is mentioned that there are no direct private benefits (like better taste) to buying sustainable products (Bougherara & Combris, 2009). However, since people associate sustainability with gentleness, information can backfire if consumers prefer strength-related attributes for that particular product. In these cases, information will lead consumers to prefer the regular product over the sustainable version (Luchs, Naylor, Irwin, & Raghunatan, 2010). Also, both information overload and lack of understanding of sustainability labels can negate any positive effects they might have (Horne, 2009; Grunert, 2011). These labels are also less effective on people with high environmental knowledge, because their environmental concern is already heightened (Cerri et al., 2018). What can be concluded from this is that information can be used to affect behavior in certain cases, and if it is used correctly. To disentangle the effect from the information, the control + info condition was added. In this condition, some information will be given along with the WTP question. This will be done with the following text: “During the production of sustainable products, more money is spent on labour (e.g. higher wages), resource management (e.g. diversifying crops) , and reduction of carbon emissions (e.g. changing to renewable energy). In the ‘sustainable is expensive’ condition, the price of sustainable products will be framed as more expensive than regular products. This will be done by adding the following text to the information: “This causes sustainable

products to be more expensive”. In the ‘regular is too cheap’ condition, the price of

sustainable products will be framed as reflecting the true cost of making a particular product. This will be done by adding the following text to the information: “This causes the price of a sustainable product to reflect the true cost of making this product”. It is expected that the WTP for the ‘regular is too cheap’ condition is the highest, while the ‘sustainable is

11 expensive’ condition is the lowest. The control + info and control conditions will be second and third, respectively.

The mediator, product-level environmental concern, will be measured with the same questionnaire adapted from Grunert et al. (2014) as the moderator. The scale will be the same as for the moderator, but the items will now refer to ‘coffee products’ instead of ‘food

products’. This was done so the questionnaire now measures concern at the product level. However, since the general-level and product-level questionnaires are virtually the same, the item orders for both questionnaires will be randomized. This is to prevent respondents from thoughtlessly filling in the same answers twice. It is expected that framing the price of sustainable products as fair vs too expensive will lead to a higher product-level concern. The presence of sustainability information is also expected to positively influence this concern.

The effects from environmental concern on behavior are expected to be high when concern is high, moderate to low when concern is at moderate levels, and non-existent when concern is low. So, in other words: the effect from concern on behavior will be significantly lower at low levels of environmental concern, compared to higher levels.

The final block will be a manipulation check. This was added to see if the price perceptions of regular and sustainable products change because of the framing. The check consists of two questions, where respondents can indicate how they would classify the price of regular and sustainable products. The possible answers are ‘Too expensive’, ‘Fairly priced’, and ‘Too cheap’. It is expected that respondents will indicate that sustainable products are too expensive and regular products too cheap in the control, control + info, and ‘sustainable is expensive conditions’. These answers should reverse in the ‘regular is too cheap’ condition.

The experiment, made in Qualtrics, will be conducted among people who drink coffee. This is because they have a better understanding of the value of coffee than those who do not drink it. A convenience sample will be used, as the survey will be spread to contacts on Whatsapp, and shared in Facebook groups. Because of this, most respondents will be university students aged 18-25. There will be no reward for participating in this study.

12

Results

Data collection was started on May 8th _{2020, and ended on May 12}th _{2020. In this}

time, 58 people completed the experiment. The control group has 12 people, the control+info has 15, the ‘regular too cheap’ group has 16, and the ‘sustainable is expensive’ has 15. There were also 12 people who started but did not finish the experiment. All data is available in a dropbox file, the link to which, together with the R-script used to analyze it is placed in Appendix B.

To calculate a total score for general environmental concern, the individual scores of the general concern were averaged (Cronbach’s alpha = .93). The average general level concern across groups is 4.40 on a scale from 1 to 7 (SD = 1.05), indicating a moderate level of concern.

To create a total score for product-level concern, the same method was used as for general concern (Cronbach’s Alpha = .95). The average product-level concern across groups is 4.01(SD = 1.33) on a scale from 1 to 7, indicating a moderate level of concern.

Next, a pairwise t-test indicated that across conditions, participants’ general environmental concern (M = 4.40, SD = 1.05) was higher than their product-level

environmental concern (M = 4.01, SD = 1.33; t(57) = 4.85, p < 0.01). This result is in line with the finding of Grunert et al. (2014), where concern lowered at the product-level as compared to the general level. This finding indicates that it is valid to treat these two concern-levels as different in the coming analyses.

Main effect of framing

Manipulation check

The goal of the manipulation was to change participants’ attitudes concerning the price for sustainable products. The manipulation check measured the price classification for both regular and sustainable coffee, from 1 (far too cheap) to 7 (far too expensive). The midpoint, 4, was ‘neither too cheap nor too expensive’. There were no significant differences between conditions for regular coffee. The mean classification was 3.38 (SD = 1.35), indicating that participants felt like regular products were slightly too cheap. There were also no significant differences between conditions in the classification for sustainable coffee. With a mean of 5.02 (SD = 0.96), people felt like sustainable products are slightly too expensive. However, there was a significant difference in classification between the price of regular coffee and sustainable coffee (t(57) = -9.00, p < 0.01). This finding indicates that participants classified the price of regular and sustainable coffee differently, consistently across conditions.

Specially, they felt that regular coffee was moderately too cheap, while seeing the price of sustainable coffee as slightly too expensive. The attitude that was planned to awaken in the ‘regular too cheap’ condition, that regular products are too cheap and sustainable products are fairly priced, seems to be partially present in the participants already. The levels of moderate environmental concern present in the participants (as mentioned further in this chapter) may help explain this fact.

WTP

H1 held that framing sustainable products as too expensive, would lower WTP, and framing regular products as too cheap, would raise WTP for sustainable products. In the bar graph below it can be seen that the WTP was lowest in the control condition, slightly higher in the control+info condition, and even higher for both framing conditions, with it being highest in the ‘sustainable is expensive’ condition. The results are visualized in figure 2 below. The red line is the price of the Douwe Egberts coffee mentioned in the experiment, €5.99.

13

Figure 2: The mean WTP for sustainable coffee per condition

A t-test confirmed that the average WTP for sustainable coffee was €7.44, (SD = 1.99) which is significantly higher than the DE coffee price (t(57) = 5.56, p < 0.01). This finding shows that, on average, people are willing to pay more for a sustainable product than the price of the regular version. This finding is line with earlier research, that found that consumers are willing to pay a premium for sustainable products.

To test if there was a significant difference between any of the conditions, six t-test were carried out. The null hypothesis of equal variances was not rejected (F(3) = 0.28, p = 0.84), so the t-tests could be carried out while assuming equal variances. First, the differences between the two framing conditions was tested first. There was no significant difference in WTP between the ‘regular too cheap’ (M = 7.56, SD = 1.68) and ‘sustainable expensive’ (M = 7.84, SD = 2.34) condition (t(54) = 0.384, p = 0.703).

There was no significant difference in WTP between the ‘regular too cheap’ condition and the control (M = 7.11 SD = 1.91) condition (t(54) = 0.581, p = 0.564), or ‘regular too cheap’ and the control + info (M = 7.20 SD = 2.11) condition (t(54) = 0.492, p = 0.625). It can be concluded that framing the price of regular products as too cheap, did not raise WTP for sustainable products.

There was no difference between the control and ‘sustainable expensive’ condition (t(54) = 0.929, p = 0.357). There was also no significant difference between this framing condition and the control+info condition (t(54) = 0.862, p = 0.392). Framing sustainable products as expensive did not lower the WTP for these products. It can be argued that because this framing is already abundantly present in many different contexts (Peattie, 2010), the attitude we wanted to engender in this framing condition was already present in both control conditions. This caused the WTP to not be lower (or higher) than the control condition.

Finally, there was no significant difference in WTP between the control and control+info conditions (t(54) = 0.116, p = 0.908). This contradicts earlier research (eg Loureiro & Lotade, 2005; Rousseau, 2015) that found that sustainability information raises WTP. However, the lack of effect found can be explained by the fact that functional green information, as used in this study, only affects attitudes in consumers with high environmental involvement. This is caused by the fact that only those with high involvement get value out of these messages (Matthes & Wonneberger, 2014). This is tested in the next section.

14 These tests indicate that the H0 hypothesis that framing does not affect WTP could not be disproven. There are, however, some limitations to the current research. The manipulation check indicates that the planned difference in attitude towards product prices was not present in this study. One can wonder if the frames used to engender this attitude were picked

correctly, or if they were too short and therefore only seemed to confirm the given information. It has been argued that in order for framing to be effective, they need to

drastically change the information given to respondents (Boyle, 1989), and the problem itself need to involve a high amount of risk (Cummings, Brookshire, & Schulz, 1986). For example, the ‘Asian disease’ problem by Tversky & Kahneman (1980) and the conservation problem by Samples, Dixon and Gowen (1986) involve life and death (either humans or animals). Other framing studies involve choices related to one’s health, like vaccines (e.g.

Krishnamurthy, Carter, & Blaire, 2001;Bigman, Capella, & Hornik, 2010). It can be argued that since the frames used in this study did not meet these requirements, the manipulation was not effective.

For price framing specifically, frames concerned the usage of dollar amounts. Framing effects on WTP were found when prices were either specifically stated as differences in dollars (Chen et al., 1998; Wesstein et al., 2013), or when a reference price was changed (Okada & Mais, 2013). Because the goal was to measure WTP for sustainable coffee, no price was given for this product version, and the reference price was always the same. This may help explain why this study found no significant framing effects.

The respondents could also have been anchored by the price that was given for the regular coffee (Strack et al., 1998). The €5.99 price anchor for the regular coffee may have led respondents to believe that this was the fair price for coffee, while the attitude that was supposed to awaken in consumers was the opposite. The WTP for sustainable coffee might have been higher without this price anchor present. There was no control condition to measure this effect. The selection of the maximum possible price also affects the resulting WTP. If this was set below the absolute maximum price respondents would be willing to pay for coffee, this could have negatively affected WTP (Mørkbak, Christensen, & Gyrd-Hansen, 2010).

Another limitation to this study that could help explain the lack of significance between groups is the low amount of participants. There were 58 people who finished the experiment, leading to only 12 to 16 people per condition. This is a relatively low number, which makes finding a statistically significant effect more difficult.

Moderation analysis

Next, we move to the moderating effect of general environmental concern on the relationship between framing and WTP. The general concern score did not differ across conditions (F(3,54) = 0.0799, p = 0.97).

Figure 3 below, visualizes the effect of framing on WTP depending on participants’ score on the general environmental concern questionnaire. In this graph, we can see that for the control condition, general environmental concern raises WTP. In the control+ info

condition, this effect did not occur. However, this slope of this line did not significantly differ from the control (t(50) = -0.567, p = 0.573). For both the ‘sustainable expensive’ and ‘regular too cheap’, more environmental concern led to a lower WTP. This effect seems strongest for the ‘regular too cheap’ condition. However, neither slopes differ significantly from the control group (t(50) = -1.793, p = 0.0791; t(50) = -1.334, p = 0.188). This means that the effect of general concern on WTP is the same for all four conditions. Finally, the effect of general concern on WTP in the control group was nonsignificant (t(50) = 0.831, p = 0.4097).

15 Finally, it was tested if there was an interaction effect between levels of general

concern and the condition. This interaction effect would indicate that general environmental concern had a different effect between high and low levels, depending on the condition. These interaction effects were first tested for both framing conditions, compared to the control condition. For the ‘regular too cheap’ condition, there was no difference in effect between levels of general concern (t(50) = -1.79, p = 0.079). The same thing applies for the

‘sustainable is expensive’ condition (t(50) = -1.33, p = 0.19). No significant differences were found between other conditions either.

From these tests, a couple of conclusions can be drawn. The first is that the control+info and both framing conditions did not differ significantly from the control condition. The second is that general concern did not moderate the relationship between framing and WTP on an overall level. The third is that this moderation effect was also not present at either high or low levels of concern. This means that the H0 hypothesis that general environmental concern does not moderate behavior could not be rejected.

Figure 3: Moderation analysis

16

Mediation analysis

Next, it was tested if product-level concern mediates the relationship between framing and WTP. This was done by calculating the Average Causal Mediation Effect (ACME), the Average Direct Effect (ADE), and the total effect. The total effect is the effect of the complete model, so both the mediated path as the non-mediated path. The ACME is the mediated path, or indirect effect. The ADE is the non-mediated path, or direct effect.

Between the control and ‘regular too cheap’ conditions, product-level environmental concern had no effect on WTP (ACME = 0.0370, p = 0.82), with no significant effect of framing (ADE = 0.4029, p = 0.60). The total effect was also not significant (p = 0.56). Between the ‘sustainable is expensive’ and control conditions, there was also no significant effect of environmental concern on WTP (ACME = 0.0144, p = 0.94), nor a significant effect of framing (ADE = 0.74793, p = 0.32). There was also no significant total effect (p = 0.32). The same thing applies for the control and control+info conditions; no effect from concern on WTP (ACME = -0.00788, p = 0.91), and no significant framing effect (ADE = 0.08081, p = 0.92). The total effect was also not significant (p = 0.91).

Something that may help explain the lack of mediation is that the relationship between environmental concern and behavior is not linear. Only the highest level of concern have a big impact on behavior, while the impact on lower levels is mediocre to non-existent (de Langhe, Puntoni, & Larrick, 2017). A type 3 ANOVA was done to test this assumption. However, between both the ‘regular too cheap’ and control, and ‘sustainable is expensive’ and control, a significant interaction effect was not found (t(50) = -1.36, p = 0.18; t(50) = -1.53, p = 0.13). No significant interaction effects were found between other conditions either. This means that the effect of product-level concern on WTP did not differ between high and low levels of concern.

There was no significant difference in levels of product-level concern between

conditions, with the average concern being 4.02 (SD = 1.33) on a scale of 1 from 7 (F(3,54) = 0.106, p = 0.96). This indicates moderate levels of concern. The lack of difference in product-level concern between conditions can be explained by the fact that the intended manipulation did not take place.

From this analysis it can be concluded that overall, product-level environmental concern did not mediate the relationship between framing and WTP. There was also no mediation on only high or low levels of concern. This means that the H0 hypothesis that product-level environmental concern does not mediate the relationship between framing and WTP, could not be rejected.

The lack of mediation and moderation from environmental concern can be explained by the fact that the way this attitude-behavior relationship is measured has been called into question. Firstly, every researcher tends to come up with their own measure for environmental concern. These measures, while sometimes overlapping, all measure different environmental attitudes (Weigel, 1983; Stern & Oskamp, 1987). So, if one study finds that environmental concern is positively related to behavior, another measuring a similar but different

environmental attitude may find a different result (Manieri et al., 1997). Secondly, it has been found that you only find a strong relationship between an attitude and a behavior if these two are strongly related (Hines, Hungerford & Tomera,1986-1987). This means that a general concern measure is ineffective at finding these attitude-behavior relationships, since environmental concern is split up into many different categories, which are all linked to different behaviors (Zimmer et al., 1992). These factors bring into question the validity and generalizability of the method used to measure concern in this study.

17

Moderated mediation analysis

The final step of this study was to test if general environmental concern moderated the mediated relationship discussed earlier. This was done by creating a high general concern value one SD above the mean, and low general concern value one SD below the mean.

At low levels of general concern, there was no significant effect of product-level concern on WTP between the control and ‘regular too cheap’ conditions (ACME = 0.001 p = 0.956), with no significant effect of framing (ADE = 1.66, p = 0.063). The total effect was also not significant (p = 0.063). Between these same conditions, there was also no significant effect of product-level concern on WTP at high levels of general concern (ACME = 0.0014, p = 0.90). There was also no significant effect of framing (ADE = -0.8774, p = 0.42), or a significant total effect (p = 0.38). Between these two levels of general concern, there was no significant difference in mediation (ACME = -0.002, p = 0.88; ADE = 2.5341, p = 0.1), so there was no moderated mediation. The results for most of the other conditions were the same, with similar ACME, ADE and p-values. This means that for most conditions, there was no moderated mediation. This contradicts the hypothesis, because it was expected that a higher level of general concern would have a positive effect on the relationship between framing and product-level concern for the ‘regular too cheap condition’, and a negative effect for the ‘sustainable is expensive’ condition. However, one significant effect was found. For the control and ‘sustainable is expensive’ conditions, there was no significant effect of product-level concern on WTP (ACME = 0.001, p = 0.982), nor was there a significant effect of framing (ADE = 1.175, p = 0.057). There was, however, a significant total effect (p = 0.049). This means that, for low general concern, there is in fact a significant effect of framing on WTP. However, this effect was negative, meaning that if general concern is low, framing the price of sustainable products as more expensive leads to a lower WTP than in the control condition. Conceptually, this makes sense; people who have a low amount of environmental concern are most likely less knowledgeable on environmental issues, making them more susceptible to this framing. This result is also partially in line with one of the hypotheses; however, the expected effect only takes place for those with low general environmental concern, and not for people with medium or high concern. This finding is also matches earlier research confirming the difference in behavior between people with low and high

environmental concern. As mentioned in the literature review, there were a lot of behavioral differences between these groups.

This analysis shows that general environmental concern does not moderate the relationship between framing, product-level environmental concern, and WTP. This means that the H0 hypothesis that this is the case could not be rejected. However, it was found that for low levels of environmental concern, WTP was lower than for high levels. This partially confirms the expectation set in the literature review, but cannot fully reject the H0 hypothesis.

Given the insignificant results found across this analysis, it needs to be considered again if the concepts of environmental concern were measured properly. The problem of the low number of participants also needs to be kept in mind.

Conclusion

The research question this study attempted to answer was ‘How does the framing of the price of sustainable products influence Willingness to Pay, through the activation of product-level concern for the environment?’. The answer to this question is that framing does not affect willingness to pay for sustainable products, directly or through the activation of product-level environmental concern. It was also found that there is no moderated mediation from general environmental concern. However, it was found that if you frame the price of sustainable products as being expensive, then people with low general environmental concern will be less

18 willing to pay for these products. If you frame regular products as too cheap, or give only information, this effect is not present. The effect is also not present for high levels of general environmental concern. So: the ‘sustainable is expensive’ frame has a negative effect on WTP for consumers with low general concern for the environment.

General discussion

There are a few significant discoveries made in this research. In this section, these findings and its implications will be discussed. Limitations of the current research and future research possibilities will also be mentioned and explained.

First, it was found that there is a significant difference between people’s

environmental concern at general and product level. At product-level this concern is lower than at the general level. This is fully in line with the findings of Grunert et al. (2014) and the hypothesis of the current research. This is a very significant finding because all research focuses on raising concern at the general level, while concern at the product-level may actually guide behavior more. However, no evidence was found for this in the current research. Still, future research should look into the concept of product-level environmental concern, to see why this differs from general level concern, and if and how this concept influences attitudes and behavior. If it is found that this concept indeed influences behavior, it could be very important knowledge for sustainable brands, because they could use it to convince more people to buy their products. It should be noted however, that even at the general level, most people are only moderately concerned about climate change, while scientists have been saying it will be catastrophic. Future research should try to find out why the concern for this topic is so low, and what can be done to raise it.

Secondly, partially in line with the hypothesis explained in the literature review, it was found that framing sustainable products as more expensive than regular products, leads to a lower WTP for people with low general environmental concern. However, the framing was only effective for this select group of participants. For higher levels of general concern, and different framing, there was no effect. This finding implies that people with low

environmental concern are more susceptible to this framing than those with high concern, and are thus not willing to spend as much money on sustainable products. This would imply that educating those with low concern and convincing them of the threat of climate change could help raise their WTP. Sustainable brands could use this to reach the customer segment that might be unavailable to them right now. However, they should take into account the

drawbacks and contingencies related to giving sustainability information. Otherwise, it could backfire.

This finding should also be looked at in a slightly broader context. Past research has found positive effects of environmental concern on WTP. The current research discovered there is a negative effect of low concern on behavior, where low concern actually leads to a lower WTP for sustainable products. Future research should find out what the different effects of different levels of environmental concern are on attitudes on behavior. These findings will be important, because the difference in reaction implies that certain interventions will only be effective on people with a specific level of concern. Sustainable brands could use these findings to tailor their messages better to specific customer groups, and hopefully increasing their market share. However, even with the limitations of the current research in mind, it can still be said that some frames might not be as effective at convincing people to buy

sustainable products as others. Future research should look into what frames and messages are better at influencing attitudes and behaviors, so sustainable brands can use these to get more customers. One possible avenue is the effect of social norms on purchasing behavior, in relationship with environmental concern. If prices were framed as ‘many people see the price

19 of sustainable products as fair, and regular products as too cheap’, it might have a more significant effect than the frame used in this study.

Thirdly, it was found that people see the price of sustainable products as slightly too expensive, while they see the regular version as slightly too cheap. This is good news for sustainability advocates and brands, because it implies that people are starting to see the environmental damage production of certain products causes, and that consumer price perceptions are starting to change. However, it does beg the question why people still find sustainable products too expensive, and what price they would find fair. Future research should look into this, and find out if 1) this perception is truly universal and 2) what influence it has on behavior.

Some possible theoretical explanations of the results should be considered as well. For example, there are many attitudes and beliefs that influence the relationship between

environmental concern and behavior. Capitalistic beliefs like freedom of the commons, or that technological innovation is the only route towards mitigating climate change, could stop people from acting sustainably (Hardin, 1986; Gifford, 2011). Religious beliefs can also negatively impact pro-environmental behaviors (Morrisson, Duncan, & Parton, 2015), for example when people believe God will not flood the earth a second time (Mortreux & Barnett, 2008). Also, people living comfortably under a status quo tend to resist any sort of change, in fear that this will negatively impact them (Feygina, Jost, & Goldsmith, 2010). Raising self-efficacy, the belief that one is able to carry out a task (Landry et al., 2018), has been linked to more pro-environmental behaviors (Geiger, Swim, & Frasier, 2017; Jugert et al., 2016; Lauren, Fielding, Smith, & Louis, 2016), while low self-efficacy has been identified as a barrier (Gifford, 2011). Motivated reasoning, when people reject information that does not match their world view, has also been linked to a decrease in pro-environmental behavior (Druckman & Mcgrath, 2019), and probably strengthens these effects. And finally, effort of the desired action and the person’s income influence the attitude-behavior relationship.

People do not only base their behavior on their own attitudes, but also on the behavior of so-called reference groups. It has been found that one way of conforming to these groups is by changing purchase behavior (Bearden & Etzel, 1982; Bearden, Netemeyer, & Teel, 1989). Conforming to norms like this is normal human behavior (Cialdini, 2007), and is also

effective in environmental contexts. For example, using norms has made people reuse towels in hotels (Goldstein, Cialdini, & Griskevicius, 2008), and lower their energy usage (Alcott, 2011). However, as explained in the literature review, not a lot of people buy sustainable products. So, the current social norm in place is buying regular products. People wanting to conform to this norm can help explain why their concern does not translate to behavior.

Finally, measuring the relationship between environmental concern and behavior has been the way to go in research for a long time. However, a newer stream of studies suggests that looking at environmentally concerned vs non-concerned consumers misses other categories of people that require other interventions to make them buy sustainable products. There are consumers for whom status motives are important motivators for changing

behaviors. Griskevicius, Tybur, and Bergh (2010) found that activating status motives made people buy sustainable food products, and a similar effect was found for the purchase of more sustainable cars (Sexton & Sexton, 2011). These authors suggest that the signaling of status is important to people, and that this should be harnessed by sustainable brands (Elliot, 2013). Since this study had no way to signal their willingness to pay to other people, these findings may help explain the lack of significant results. Future research should look into when these status motives can activate behavior. Because if people buy sustainable products to signal that they can afford to make a monetary sacrifice for the greater good, it makes sense that only those who can actually afford it will do so. This line of thinking matches the findings that

20 people with higher incomes spend significantly more in sustainable products than those with lower incomes (Raavensway & Blend, 1999; Vecchio & Annunziata, 2015).

This short analysis is not meant to be exhaustive. It is merely meant to show that the relationship between environmental concern and behavior is extremely complex, and that all these factors could have influenced the final results of this study. This does not mean, however, that these are all limitations or things that were missed. Measuring and accounting for all of these factors is out the scope of this study, and would have required a substantial budget and outside help. Since none of these things were available, it was decided to focus on a few factors.

In terms of policy implications, some lessons can be taken from this study. Firstly, the fact that people see the price of regular products as too cheap, and the price of sustainable products as the fair price, could be very useful to lawmakers. This view of product prices signals that there may be support for a tax on non-sustainable products. This tax would achieve three things; it would raise the price of regular products, making them less attractive than they are now relative to sustainable products. This will hopefully raise the amount of people that buy the sustainable version. The revenues from this can subsequently be used to curb the environmental damage done by the production of the regular products. The money can for example be invested in government-run projects researching sustainable energy, or into companies working to make their supply-chain more sustainable. These factors combined would then motivate the non-sustainable companies to change their production processes to get rid of this tax, which is the best possible outcome.

Secondly, with this study once again adding to the evidence that information may not be effective in convincing people to buy sustainable products, the sheer amount of labels signaling sustainability need to be revisited. Since the lack of influence on behavior could be caused by an overload of information, the development and use of one rating system signaling sustainability would be useful. It could even be a single system per product type, so one for clothing, one for food etc. A more concise display of sustainability information could be more effective than the current cluttered information. This would also allow future research to focus on the effect of one singular logo, and the changes in behavior brought about by its presence. This is easier than having to focus on many different labels and their possible interactions.

The third implication is that the way sustainable products are discussed does influence the behavior of people who have a low amount of environmental concern. The way to tackle this problem is two-fold. Firstly, even at lower levels at education, there should be lessons about climate change and the problems this will cause for our planet. This will hopefully raise environmental concern to a level where this framing does not have a negative impact on behavior anymore. Secondly, policymakers themselves should be mindful of how they discuss the prices of sustainable products. They should not brand them as ‘too expensive’, because this will negatively impact the WTP for some people, which is the exact opposite of what they wish to achieve.

21

References

Albayrak, T. (2013). The effect of environmental concern and scepticism on green purchase behaviour. Green Purchase Behavior, 31(1), 27–39.

https://doi.org/10.1108/02634501311292902

Allcott, H. (2011). Social norms and energy conservation. Journal of Public Economics, 95(9–10), 1082–1095. https://doi.org/10.1016/j.jpubeco.2011.03.003

Ar, I. M. (2012). The impact of green product innovation on firm performance and competitive capability : the moderating role of managerial environmental concern. Procedia - Social and Behavioral Sciences, 62, 854–864.

https://doi.org/10.1016/j.sbspro.2012.09.144

Ayres, R. U., & Kneese, A. V. (1969). Production , Consumption , and Externalities, 59(3), 282–297.

Banerjee, J. (2019). Why Is Sustainable Fashion So Expensive?. Retrieved 6 May 2020, from https://www.forbesindia.com/article/best-things-money-can-buy/why-is-sustainable-fashion-so-expensive/55823/1

Bamberg, S. (2003). How does environmental concern influence specific environmentally related behaviors ? A new answer to an old question. Journal of Environmental Psychology, 23, 21–32.

Bang, H. K., Ellinger, A. E., Hadjimarcou, J., & Traichal, P. A. (2000). Consumer concern, knowledge, belief, and attitude toward renewable energy: An application of the reasoned action theory. Psychology and Marketing, 17(6), 449–468.

https://doi.org/10.1002/(SICI)1520-6793(200006)17:6<449::AID-MAR2>3.0.CO;2-8 Bator, F. M. (1958). The anatomy of market failure. Quarterly Journal of Economics, 72(3). Bearden, W., & Etzel, M. J. (1982). Reference Group Influence on Product and Brand

Purchase Decisions. Journal of Consumer Research, 9(2). https://doi.org/10.1086/208911

Bearden, W. O., Netemeyer, R. G., & Teel, J. E. (1989). Measurement of Consumer

Susceptibility to Interpersonal Influence. Journal of Consumer Research, 15(4), 473– 481.

Bigman, C. A., Cappella, J. N., & Hornik, R. C. (2010). Effective or ineffective: Attribute framing and the human papillomavirus (HPV) vaccine. Patient Education and Counseling, 81(SUPPL. 1), S70–S76. https://doi.org/10.1016/j.pec.2010.08.014 Boiral, O. (2006). Global Warming: Should Companies Adopt a Proactive Strategy? Long

Range Planning, 39(3), 315–330. https://doi.org/10.1016/j.lrp.2006.07.002

Bolton, L. E., Warlop, L., & Alba, J. W. (2003). Consumer Perceptions of Price (Un)Fairness.

Journal of Consumer Research, 29(4), 474–491. https://doi.org/10.1086/346244

Bougherara, D., & Combris, P. (2009). Eco-labelled food products: What are consumers paying for? European Review of Agricultural Economics, 36(3), 321–341.

https://doi.org/10.1093/erae/jbp023

Boyle, K. J. (1989). Commodity Specification and the Framing of Contingent-Valuation Questions. University of Wisconsin Press, 65(1), 57–63.

Cerri, J., Testa, F., & Rizzi, F. (2018). The more I care , the less I will listen to you : How information , environmental concern and ethical production influence consumers ’ attitudes and the purchasing of sustainable products. Journal of Cleaner Production,

175, 343–353. https://doi.org/10.1016/j.jclepro.2017.12.054

Chang, T. D., Kong, W. H., & Chen, S. (2017). The Moderating Role of Environmental Concerns in Travel Behavior : An Exploratory Study. Journal of China Tourism

Research, 13(2), 193–210. https://doi.org/10.1080/19388160.2017.1353471

Chen, S. S., Monroe, K. B., & Lou, Y. (1998). The Effects of Framing Price Promotion Messages on Consumers ’ Perceptions and Purchase Intentions. Journal of Retailing,

22 74(3), 353–372.

Cummings, Ronald G., Brookshire, D.S., & Schulze, W. (1986). Valuing Environ-

mental Goods: An Assessment of the Contingent Valuation Method. Totowa, NJ: Row- man and Allenheld Publishers.

Czap, N. V., & Czap, H. J. (2010). An experimental investigation of revealed environmental concern. Ecological Economics, 69(10), 2033–2041.

https://doi.org/10.1016/j.ecolecon.2010.06.002

Davari, A., & Strutton, D. (2014). Marketing mix strategies for closing the gap between green consumers ’ pro-environmental beliefs and behaviors. Journal of Strategic Marketing, 22(7). https://doi.org/10.1080/0965254X.2014.914059

Druckman, J. N., & Mcgrath, M. C. (2019). The evidence for motivated reasoning in climate change preference formation. Nature Climate Change, 9(February), 111–119.

https://doi.org/10.1038/s41558-018-0360-1\

Elliott, R. (2013). The taste for green : The possibilities and dynamics of status differentiation through ‘“ green ”’ consumption. Poetics, 41(3), 294–322.

https://doi.org/10.1016/j.poetic.2013.03.003

Feygina, I., Jost, J. T., & Goldsmith, R. E. (2010). System justification, the denial of global warming, and the possibility of “system-sanctioned change.” Personality and Social Psychology Bulletin, 36(3), 326–338. https://doi.org/10.1177/0146167209351435 Finke, T., Gilchrist, A., & Mouzas, S. (2016). Why companies fail to respond to climate

change: Collective inaction as an outcome of barriers to interaction. Industrial

Marketing Management, 58, 94–101.https://doi.org/10.1016/j.indmarman.2016.05.018 Flohn, H. (1980). POSSIBLE CLIMATIC CONSEQUENCES OF A MAN-MADE GLOBAL

WARMING.

Geiger, N., Swim, J. K., & Fraser, J. (2017). Creating a climate for change: Interventions, efficacy and public discussion about climate change. Journal of Environmental

Psychology, 51, 104–116. https://doi.org/10.1016/j.jenvp.2017.03.010

Gifford, R. (2011). The Dragons of Inaction: Psychological Barriers That Limit Climate Change Mitigation and Adaptation. American Psychologist, 66(4), 290–302. https://doi.org/10.1037/a0023566

Goldstein, N. J., Cialdini, R. B., & Griskevicius, V. (2008). A Room with a Viewpoint : Using Social Norms to Motivate Environmental Conservation in Hotels. Journal of

Consumer Research, 35(3), 472–482. https://doi.org/10.1086/586910

Goucher-lambert, K., & Cagan, J. (2015). The Impact of Sustainability on Consumer Preference Judgments of Product Attributes. Journal of Mechanical Design, (March

2017). https://doi.org/10.1115/1.4030271

Griskevicius, V., Tybur, J. M., & Van den Bergh, B. (2010). Going green to be seen: Status, reputation, and conspicuous conservation. Journal of Personality and Social

Psychology, 98(3), 392–404. https://doi.org/10.1037/a0017346

Grunert, K. G. (2011). Sustainability in the Food Sector A Consumer Behaviour Perspective. International Journal on Food System Dynamics, 2(3), 207–218.

https://doi.org/10.18461/ijfsd.v2i3.232

Grunert, K. G., Hieke, S., & Wills, J. (2014). Sustainability labels on food products : Consumer motivation , understanding and use. Food Policy, 44, 177–189. https://doi.org/10.1016/j.foodpol.2013.12.001

Hardin, G. (1968). The Tragedy of The Commons. Science, 162(3859).

Hayat, M., Shahnia, F., Arefi, A., Iu, H., & Fernando, T. (2017). Comparison of the Economic Benefits and the Payback Periods of Rooftop Solar Panels in Australia.

https://doi.org/10.1109/PGSRET.2017.8251811

23 research on responsible environmental behavior: A meta-analysis. The Journal of Environmental Education, 18(2), 1–8

Hoelang roepen we elkaar al op om toch vooral gewoon te doen?. (2014). Retrieved 1 June 2020, from https://www.vandale.nl/node/1375

Horne, R. E. (2009). Limits to labels: The role of eco-labels in the assessment of product sustainability and routes to sustainable consumption. International Journal of

Consumer Studies, 33(2), 175–182. https://doi.org/10.1111/j.1470-6431.2009.00752.x Howard, P. H., & Allen, P. (2010). Beyond Organic and Fair Trade ? An Analysis of Ecolabel

Preferences in the United States. Rural Sociology, 75(2), 244–269. https://doi.org/10.1111/j.1549-0831.2009.00009.x

Ingold, K., & Fischer, M. (2014). Drivers of collaboration to mitigate climate change: An illustration of Swiss climate policy over 15 years. Global Environmental Change,

24(1), 88–98. https://doi.org/10.1016/j.gloenvcha.2013.11.021

Jackson, T. (2005). Motivating Sustainable Consumption : A Review of Evidence on Consumer a review of evidence on consumer.

Joshi, Y., & Rahman, Z. (2015). Factors Affecting Green Purchase Behaviour and Future Research Directions. International Strategic Management Review (Vol. 3). Holy Spirit University of Kaslik. https://doi.org/10.1016/j.ism.2015.04.001

Jugert, P., Greenaway, K. H., Barth, M., Büchner, R., Eisentraut, S., & Fritsche, I. (2016). Collective efficacy increases pro-environmental intentions through increasing self- efficacy. Journal of Environmental Psychology, 48, 12–23.

https://doi.org/10.1016/j.jenvp.2016.08.003

Kok, J. De, & Hoekstra, A. (2008). Aanpassen aan klimaat- verandering : twee strategieën voor dijkverhoging, 29–32.

Kushwah, S., Dhir, A., & Sagar, M. (2019). Ethical consumption intentions and choice behavior towards organic food. Moderation role of buying and environmental concerns. Journal of Cleaner Production, 236, 117519.

https://doi.org/10.1016/j.jclepro.2019.06.350

Krishnamurthy, P., Carter, P., & Blair, E. (2001). Attribute Framing and Goal Framing Effects in Health Decisions. Organizational Behavior and Human Decision Processes, 85(2), 382–399. https://doi.org/10.1006/obhd.2001.2962

La Faille, I. (2020). Zuinige wasmachines. Retrieved 27 April 2020, from https://www.consumentenbond.nl/wasmachine/zuinige-wasmachines

Landry, N., Gifford, R., Milfont, T. L., Weeks, A., & Arnocky, S. (2018). Learned

helplessness moderates the relationship between environmental concern and behavior. Journal of Environmental Psychology, 55, 18–22.

https://doi.org/10.1016/j.jenvp.2017.12.003

Lauren, N., Fielding, K. S., Smith, L., & Louis, W. R. (2016). You did, so you can and you will: Self-efficacy as a mediator of spillover from easy to more difficult pro-

environmental behaviour. Journal of Environmental Psychology, 48, 191–199. https://doi.org/10.1016/j.jenvp.2016.10.004

Lernoud, J., & Willer, H. (2016). The Organic and Fairtrade Market 2015.

Loureiro, M. L., & Lotade, J. (2005). Do fair trade and eco-labels in coffee wake up the consumer conscience? Ecological Economics, 53(1), 129–138.

https://doi.org/10.1016/j.ecolecon.2004.11.002

Lorenzoni, I., Nicholson-Cole, S., & Whitmarsh, L. (2007). Barriers perceived to engaging with climate change among the UK public and their policy implications. Global Environmental Change, 17(3–4), 445–459.

https://doi.org/10.1016/j.gloenvcha.2007.01.004