Eliciting Willingness‐to‐Pay through Multiple Experimental Procedures: Evidence from Lab‐in‐the‐Field in Rural Ghana

Abstract

This paper has the objectives of (a) comparing estimated willingness‐to‐pay (WTP) across three elicitation mechanisms (a Becker‐DeGroot‐Marschak [BDM] auction, a kth price auction, and a choice experiment [CE]) and (b) examining how these vary by participation fee. The product under consideration is kenkey made with nutritious maize, biofortified with vitamin A, which gives it a distinct orange color, in contrast to the white and yellow varieties that are traditionally consumed. We use an experiment consisting of 14 treatment arms, conducted in rural Ghana. Our estimation strategy explicitly accounts for the censored (typically at the market price) nature of the bids in the auctions, and the apparently lexicographic choices of several individuals in the CE. We find no evidence of economically meaningful (defined by the minimum currency unit of five pesewas) differences in WTP (although they may be statistically significant) across elicitation mechanisms, or by participation fee, a result that is in contrast to that found in much of the literature. A secondary finding is that the provision of nutrition information positively and significantly affects the marginal WTP for the new maize.

Information and consumer willingness to pay for biofortified yellow cassava: evidence from experimental auctions in Nigeria

“…BDM is a widely applied auction‐like mechanism in consumer acceptance analyses in rural Africa (e.g., Banerji et al., ; De Groote et al., ; Hoffmann et al., ; Kiria et al., ; Morawetz et al., ). In a BDM mechanism, the individual places a bid, b , for the product on sale.…”

“…The prevailing market prices for half a Kongo (500 g), or 3 Nescafé cups (300 g), of gari in Oyo and Imo states, respectively, varied between ₦20 to ₦50 (US$1≈ ₦150 when the study was implemented). Based on recent studies on vitamin A biofortified orange sweet potato in Uganda (Chowdhury et al., ) and orange maize in Zambia (Meenakshi et al., ) and Ghana (Banerji et al., ), we anticipated effects of 15% or 20%, corresponding to ₦5 or ₦7 as well as a standard deviation of ₦30–₦35. Power calculations indicated that for the pooled data over the two states and for one‐tailed test, 450 (150*3) participants were required for the three treatments, resulting in 225 participants per state.…”

“…Banerji et al. () note that participants who state bids equal to the market price may have higher WTP due to perceived transaction costs of obtaining the same product outside the experiment but bid the price at which they could buy the product in the market.…”

“…In order to meet these objectives, we designed a consumer acceptance study using the Becker–DeGroot–Marschak (BDM) mechanism (Becker et al., ) adapted from experimental economics literature, and hedonic tests adapted from food science literature (Corrigan et al., ; List, ; Tomlins et al., ). The BDM mechanism was chosen among various preference elicitation techniques because it is easier to implement on an individual basis in rural settings (Banerji et al., ). Unlike other field experiments on consumer acceptance of biofortified foods, this study eliminated participation fees and participants were told that they were expected to make out of pocket payments.…”

“…Unlike other field experiments on consumer acceptance of biofortified foods, this study eliminated participation fees and participants were told that they were expected to make out of pocket payments. Standard theory suggests that initial endowments can distort optimal bidding behavior (Corrigan and Rousu, ), though the empirical evidence in this regard has been mixed (Banerji et al., ; Loureiro et al., ; Morawetz et al., ). Thus, a fourth objective of this article is to compare the bids of participants who paid out of pocket against those who were unable to pay due to liquidity constraints.…”