Prominent numbers in DCE cost vectors: A review and an application to wind energy

Nearly all discrete choice experiments in environmental valuation have a cost attribute to estimate willingness to pay for the goods and services in question. The composition of the cost vector, however, has received little attention in the literature so far. This paper focuses on whether or not the cost vector should include "prominent amounts" (e.g., €1, €2, €5, €10, €20, €50). On the one hand, these amounts are familiar to participants as they are used in everyday life (e.g., coins or banknotes). On the other hand, respondents may not consider the amounts to reflect the cost of environmental programs, affecting credibility. In a review of published discrete choice experiment articles on wind energy, we find little information on how the levels of the cost attribute vectors were chosen and why prominent amounts are (not) included, reflecting the lack of guidance in the literature. In a split-sample survey on renewable energy, we find that the composition of the cost vector does not affect choice frequencies or the perceived difficulty of the task. However, we observe that the cost vector impacts credibility, suggesting that it may impact the validity of the welfare estimates. • The social science literature suggests that prominent numbers influence decision-making. • Cost vectors in discrete choice experiments often include prominent numbers (1, 2, 5, 10, 20, 50, etc.). • Respondents reveal no tendency to choose alternatives with prominent amounts more often. • We observe that the composition of the cost vector can enhance the validity of the results.