Conference Agenda

Promoting household-level adaptation measures is an important part of climate change adaptation policies that aim to reduce the vulnerability of (agro-)pastoral communities to droughts in sub-Saharan Africa. To develop effective policies, it is important to get a better understanding of the needs of the communities. In this study, we therefore present the results of a discrete choice experiment in which we identify preferences for four different types of drought adaptation support in (agro-)pastoral communities in Kenya. A novelty of our study is that we link the results from our discrete choice experiment to behavioural factors of three established decision-making theories from economics and psychology: expected utility theory, theory of planned behaviour, and protection motivation theory. Including these theories in our analysis to get a better understanding of the causal relationship between adaptation behaviour and preferences for adaptation support. We observe clear heterogeneity in preferences for support related to behavioural factors from the different theories. We discuss the implication of our results for drought risk adaptation policy.

Discrete Choice Experiments (DCE) represent a methodology with a rich tradition of usage in various fields of knowledge, such as Economics, Health, or Consumer Preference. There exists a substantial body of literature on optimizing their design, focusing on elements such as the optimal number of alternatives, choice cards, or attributes. However, little attention has been given to a crucial aspect – the optimal visual orientation of alternatives on choice cards presented to respondents. In this study, we scrutinize the implications of the visual orientation of a discrete choice experiment on the answers given by respondents and the characteristics as well as the goodness fit of the resulting final model. To achieve this, we designed two versions of the same Discrete Choice Experiment, one with a horizontal arrangement and another with a vertical arrangement. The obtained results suggest that the visual presentation of Discrete Choice Experiments carries significant implications, thereby warranting further investigation in this regard in future research.

Stated preference methods are widely used for valuing public goods such as tap water quality improvements. Nevertheless, whether respondents reveal their true preferences in surveys is still the subject of academic debate. Existing literature suggests that truthful disclosure of preferences in valuation surveys may pivot on the number of alternatives presented in a single choice task. Although, on a theoretical basis, the use of a single two-alternative choice format (one of which is a “no change” option) has long been recommended because of its potential incentive-compatible properties, empirical studies provide nascent evidence that using more than two choice alternatives may be a more natural setting, because it makes it easier for respondents to find an option that matches their preferences. The lack of consensus about the impact of the number of alternatives on respondents’ stated preferences motivates this study. Using data from a discrete choice experiment, we examine whether willingness-to-pay (WTP) estimates depend on the number of alternatives provided in a choice task. We employ a split-sample design that uses two- and three-alternative formats in a stated preference survey of proposed public policies for the improvement of tap water quality (iron and chlorine contents, hardness) in Milanówek, a town in the Warsaw agglomeration in Poland. Drawing on the mixed logit model with scale heterogeneity, we find no significant differences in the mean WTP values elicited with two- and three-alternative tasks, while the WTP estimates based on three-alternative tasks appear to have lower standard errors compared with two-alternative tasks. This finding indicates that the use of three or more alternatives per choice task may offer a way to increase efficiency without significantly biasing the results.

This paper estimates the willingness to pay for enhanced electricity service quality in Sierra Leone, a West African country where millions of households experience highly unreliable power supply. We investigate the economic cost of outages in a nationally representative sample of 1,047 grid or minigrid-connected households, via two different valuation methods, the discrete choice experiment (a stated preferences approach), and the consumer damage function (a revealed preferences method that has not previously been applied in a lower- or middle-income country context). For analysis of the stated preference, we estimate a generalized mixed logit model that allows for preference and scale heterogeneity to obtain household weights for a set of specific attributes related to the nature of service interruptions. The choice model results show that households differ in their relative preferences, but that, on average, they prefer shorter outages, fewer daytime or evening outages compared to nighttime outages, and prior notification. For the revealed preferences analysis, we then use a two-part model to estimate the costs incurred by households during two recent outage experiences. We observe that outage damages incurred by households are higher for longer duration interruptions and notified outages, while daytime or evening outage costs are lower than nighttime outages. We then discuss why some of these patterns may be inconsistent across methods. Our study provides insights that can help inform investment decisions on whether and how to make service reliability improvements, to best respond to households' preferences and reduce the damage they incur.