Willingness to Pay for Hermetic Grain Storage Bags

Willingness to Pay for Hermetic Grain Storage Bags in Malawi Gloria Guevara Alvarez* and William A. Masters Friedman School of Nutrition Science and Policy, Tufts University, Boston MA • Our study measures farmers’ demand for this innovation relative to traditional sisal bags, in a region of Malawi subject to very high insect loss and aflatoxin contamination. • We contrast willingness-to-pay (WTP) in June-July 2018, after widespread free distribution and demonstration events introduced hermetic bags in this area, with WTP a year earlier when hermetic bags were largely unknown. Figure 1. Demand curves implied by farmers’ willingness-to-pay for hermetic storage bags at random-price auctions in Malawi, by attendance at demonstration events, 2017 and 2018 Malawi kwacha per bag • Hermetic storage bags can protect grain against damage by animals, insects, and mold, limiting post-harvest loss and contamination by stopping the flow of oxygen, limiting respiration and growth of organisms inside the bag. Implications Results Motivation • Many respondents have WTP well below market cost. Self-selection into our 2018 market price ≈ 800 kwacha random-price auction was facilitated by having compensated respondents for survey responses, alleviating liquidity constraints. Over 90% of survey June-July 2018 market price ≈ 750 kwacha Solid line: attended event (n= 121) respondents chose to make bids, but over half were below market cost which June-July 2017 Dashed line: did not attend (n=151) could be due to either low valuation or incomplete information. Solid line: attended event (n= 29) Dashed line: did not attend (n=87) • Demonstrations are associated with demand primarily among respondents with lower WTP, confirming that promotion efforts could continue to target non-adopters and extend the reach of hermetic bags in Malawi. These results reinforce and extend findings on WTP for this innovation in other countries (e. g. Goentzel et al. 2016, Mwaijande 2017, Skula and Baylis 2017). • Results reveal the fraction of farmers likely to buy a bag at each price, and the value to them of free distribution. We can also test for links between WTP and farmer characteristics, including attendance at demonstration events. Methods • We drew stratified random samples of 293 farmers in 2018 and 266 in 2017, from the population is 67, 970 farmers enrolled in “marketing clubs” organized by the UBALE project in three very poor districts of Malawi. • We offered respondents the opportunity to participate in a market experiment using the random-price auction method of Becker, De. Groot and Marshak (1964). In BDM auctions, subjects state the highest price they are willing to pay and the interviewer draws a random price at which sale occurs if it is at or below the respondent’s offer. Bids provide the best available estimate of respondent WTP (Miller et al. 2011). • A survey prior to the auction elicited farmer characteristics, including household wealth (in quintiles of an asset index), amount of grain stored (in kg), knowledge about aflatoxins and how bags work (0 -12 questions answered correctly), and whether they had attended an UBALE project demonstration event about hermetic bags (1 or 0). • Farmers were compensated for time spent on the survey, providing liquidity for purchase in the auction, where we sold both traditional sisal bags and the new hermetic bags. Our analysis plan aims to estimate: • The distribution of WTP for hermetic bags in 2018 and 2017, and • The association between premia offered for hermetic over sisal bags and respondent characteristics after learning about the bags in 2018. • We use quantile regression to test for heterogeneity in response, hypothesizing that demand is linked to attendance at product demonstrations only at the lower end of the WTP distribution. Notes: Data shown are mean WTP at each percentile of respondents with a 95 percent confidence interval, estimated using Stata 14 lpolyci command at a bandwidth of 10. Figure 2. Association between WTP premium offered for hermetic over sisal grain bags and respondent characteristics, by quantile of premium (2018 auctions only) Attended demonstration Wealth quintile (not significant) Quantile estimate (with 95% CI) OLS est. (with 95% CI) Zero (no effect) Aflatoxin knowledge score (not significant) Zero (no effect) Amount stored (not significant) Zero (no effect) Results • Participation in the BDM auction rose sharply, from 44% (n=116) of the randomly selected interview subjects in 2017 to 92% (n=272) in 2018. Auction participants were also much more likely to have attended a demonstration event, up from 25% (n=29) to 44% (n=121) in 2018. These data are reflected in results (Figures 1 and 2): • Participants’ willingness to pay (WTP) rose sharply from 2017 to 2018. In 2017 at the start of product introduction, mean WTP less than half of the market cost; in 2018 after widespread free distribution, about half of respondents had WTP above market cost (Figure 1). Source: https: //commons. wikimedia. org/w/index. php? curid=1446503 Conclusions and next steps • Using hermetic storage to reduce post-harvest loss and contamination has required initial distribution of donated or highly subsidized bags, until farmers’ willingness and ability to pay for them rises above their market cost. • A third round of data collection in June-July 2019 will test for further shifts in demand, with repeat surveys of respondents from 2018. One specific aim is to test whether those whose random price draw led to a BDM purchase then have different WTP than similar subjects who left the auction without a bag. Zero (no effect) Notes: Data shown are estimated difference in WTP associated with the variable shown, at each quantile of WTP, estimated using Stata 14 sqreg and grqreg commands at quartiles of the distribution (0. 2, 0. 4 and 0. 6 and 0. 8). Malawi • Demand grew after product introduction. Participants’ WTP rose from well below market cost in 2017, to about half having WTP at or above market cost in 2018. Product promotion through free distribution and demonstration events could lead to market sales and welfare gains in the future. • Participants’ characteristics (attendance at demonstration events, wealth, aflatoxin knowledge and amount stored) had no significant association with WTP in 2017, but after product introduction in 2018 event attendance was significantly associated with the premium offered for hermetic over sisal bags, especially at lower levels of that premium (Figure 2). References cited: Becker, G. M. , De. Groot, M. H. , Marschak, J. (1964). Measuring utility by a single-response sequential method. Behav Sci. 9 (3): 226– 32. Goetzel, J et al. (2016), Summary report: Scaling adoption of hermetic post -harvest storage technologies in Uganda. Cambridge, MA: MIT Comprehensive Initiative on Technology Evaluation. Miller, K. M. , Hofstetter, R. Krohmer, Z. H. , Zhang, J. (2011). How should consumers' willingness to pay be measured? J. of Marketing Rsch 48(1): 172 -84. Mwaijande, F. (2017). Farmers’ adoption and willingness to pay for postharvest technologies in Tanzania. J. of Postharvest Tech, 5(1), 1 -6. Shukla, P. , & Baylis, K. (2017). Safe and secure: Food safety, food security and technology adoption in India. Urbana-Champaign, IL: University of Illinois. * Corresponding author: guevara. alvarez@tufts. edu Acknowledgements: This work was funded by the U. S. Agency for International Development under cooperative agreement AID-OAA-A-15 -00019, through the UBALE project led by Catholic Relief Services. We thank them and especially Angela Tavares and Juma Masumba for support of this study. Ethical approval was obtained from the Tufts University Institutional Review Board as IRB Study #1703012, and from the Malawi National Commission for Science and Technology as Protocol P. 06/17/181.