The Role of Information Source and Effect Statement for Valuation of Sustainable Foods

Overall there will be one control group without an information treatment and four five different information treatment arms with varying information source and content. The treatments are exclusive.

2020-11-12

2020-11-30

The primary outcome of interest is the respondents’ intention to buy sustainable foods, measured as stated WTP. In the survey, we measure WTP premium for a chocolate bar of 100 g of unknown brand and flavor with a sustainable label (Fairtrade) compared to a similar chocolate without a label. Thereby, we measure the WTP for Fairtrade certification. We choose this outcome because we want to know how different information treatments influence consumers’ valuation. We elicit the WTP because we are interested in a quantifiable amount of this stated preference. The outcome of the WTP premium is measured in Euro.

Behavior in terms of WTP will be measured using the contingent valuation strategy, which is a direct measurement that captures hypothetical WTP. This strategy is particularly useful to capture the price premium that a consumer is willing to pay for one additional attribute (here the label) of the product (Liebe, 2007). It is most suitable when prices are non-existing in the market place, such as the case for a particular attribute (Petrakis, 2015). According to Katt and Meixner (2020) the contingent valuation method is one of the most frequent stated preference methods for eliciting WTP in the context of organic food consumption.

We will elicit the WTP of the participants by using the introduced iterative multiple price list method (iMPL) by Andersen et al. (2007) to overcome some of the disadvantage of interval measures using multiple price lists, multiple switching points. We ask the respondents iteratively to choose one of the presented intervals of premiums they would be maximal willing to pay in addition to the current price of a similar chocolate without the Fairtrade label, i.e. what the highest € amount is before they would rather not like to buy the Fairtrade chocolate but prefer to take the conventional chocolate without a premium. In the first round participants have to choose whether to pay a premium at all or one of the three different intervals: 0.01 € to 1,00€, 1,01 € to 2,00 €, or 2,01 € to 3,00 €. In the second round, participants who chose an interval, are now asked to choose one of ten more refined intervals where i indicates the smallest full Euro amount in the chosen interval and j the largest full Euro amount: i.01 € to i.10 € , i.11 € to i.20 €, i.21 € to i.30 €, i.31 € to i.40 €, i.41 € to i.50 €, i.51 € to i.60 €, i.61 € to i.70 €, i.71 € to i.80 €, i.81 € to i.90 €, or i.91 € to j.00 €.
The selected range is supported by the literature. Didier and Lucie (2008) find an average bit in their study for fair-trade chocolate of 1.18 € with the smallest WTP premium within subjects in different stages is 0.25 €. For a fair-trade 100g chocolate bar compared to a conventional one Rousseau (2015) finds a price premium of 2.03 € and Teyssier et al. (2015) 0.458 € in a private WTP elicitation and 1.044 € in a public WTP elicitation. Poelmans and Rousseau (2016) find a WTP premium of 10.86 € per 250g for fair-trade chocolate compared to the same product without a label (This equals a WTP premium of €4.344 per 100g), however, their sample consists of chocolate lovers and not average consumers. We choose intervals of 10 cents as the finest interval option; we are confident that the 10-cent interval ensures enough precision for eliciting the WTP premium for Fairtrade chocolate, as the smallest detected WTP premiums for fair-trade chocolate compared to conventional chocolates are larger than €0.20 (Didier and Lucie, 2008; Poelmans and Rousseau, 2016; Rousseau, 2015; Teyssier et al., 2015).
The iMPL method identifiers intervals clearly and so ensures that there is only one switching point (Andersen et al., 2006).
Even though a price anchor, i.e. stating how much the conventional chocolate bar costs, would increase precision by reducing the spread of responses, we do not provide an anchor price because the price of a bar of chocolates is well-known among consumers. This also implies that we cannot use the measured WTP premium as an absolute value, however, this is no obstacle for our study as we are only interested in identifying the relative WTP depending on the treatment.

As simplified outcomes of purchasing intentions we measure the self-stated likelihood or willingness of purchasing Fairtrade chocolate in the future as categorical variables based on self-report. We choose these outcomes because they can provide additional evidence for changes of the purchase intention and so support the findings using the WTP measure.

To examine the channel of how information treatments influence purchasing intention, we include supplementary outcomes. These measure the perception of the information source and the information content which are both crucial to information processing. They could add to explaining differences in valuation or intention to buy for the different treatments.

For the other outcomes of interest we adapt the measures of self-stated likelihood or willingness to purchase Fairtrade chocolate in the future from Hansen et al. (2018). We also adjust the measures to a seven-point Likert-Scale. We choose to have an option of indifference despite a possible framing effect to ensure that we allow for all possible preferences of the participants.
The supplementary outcomes are adapted (translated into German) from Untilov and Ganassali (2020) and measure the participants’ perception of information source and content with each three questions. We also adjust the measures to a seven-point Likert-Scale. We choose to have an option of indifference despite a possible framing effect to ensure that we allow for all possible preferences of the participants. Apart from the categorical outcomes, we will perform principal component analysis in order to narrow down the statement batteries to the core of each concept.

Overall there will be one control group without an information treatment and five different information treatments. The treatment and control status is randomly assigned. Before the treatment covariates are asked for. After the treatment the outcome measures are elicited. The experiment is set up for a duration of 20 to 25 minutes.

Randomization down by the software UniPark in which we programmed the survey and the participants will respond in.

We randomize on the individual level.

No

No clusters are planned.

2500 German consumers

The groups are randomly assigned.We do not fix the participants per treatment or control group.