evidence from solar energy applications in Guinea-Bissau

This thesis aims to explore how some of the findings from behavioural economics and the social capital literature can apply in the case of electricity access in developing countries with a focus on solar off-grid electrification. And specifically on solar home systems and solar hybrid mini-grid electrification in rural Guinea-Bissau. Specifically, I am drawing from studies looking at the role of discounting anomalies on technology adoption and recurring payments, the role of trust on technology adoption and the role of computational limitations and the use of simplification strategies on the accuracy of frequency and expenditure reporting in surveys. This exercise aims to inform electrification policy in developing countries, demonstrate instances where insights from behavioural economics and social capital can enrich our understanding of the underlying barriers and drivers of electrification access, but also demonstrate how some selected case studies can help to strengthen empirical findings from other contexts. Chapter 1 provides an introduction on the issues surrounding electrification access in developing countries and introduces the research motivation and the research objectives of this thesis. This chapter also discusses the relevant gaps in the literature, how this thesis attempts to address them and the contribution to knowledge. Finally, the research location is introduced. Chapter 2 presents the results of a stated preference study that uses a choice experiment to estimate willingness to pay for a solar home system, and the trade-off between different repayment schemes and maintenance responsibilities, in the region of Bafatá in Guinea-Bissau. Results suggest that preferences are driven both by income constraints as well as self-control problems, excessive discounting and self-reported trust for a number of actors. Chapter 3 explores the main determinants in the decision to connect to a solar hybrid mini-grid, in the semi-urban community of Bambadinca in Guinea-Bissau, with a focus on social capital as expressed in trust. Connections are driven largely by the socio-economic background of the households and prior energy use patterns. However, there is evidence that social capital as expressed in self-reported trust for one’s neighbours, also has a positive effect on connections through facilitating the informal expansion of the grid, whereby households use their neighbours’ infrastructure to connect to the service. Chapter 4 explores how the technology of prepaid meters can help researchers acquire more insight regarding the accuracy of survey responses and the response strategies used. More specifically, this chapter tests the accuracy of reported energy expenditure in surveys, when using differently defined recall periods, namely a ‘usual’ week versus a ‘specific’ (i.e. last) week. We compare real expenditure data for prepaid meters for electricity, from a solar hybrid mini-grid operating in the semi-urban community of Bambadinca in Guinea-Bissau, with answers from a survey where respondents are asked to state their expenditures, randomly, in different recall periods. Overall, our results show that respondents tend to over-report the level and frequency of their energy expenditures, but reporting is more accurate when the ‘specific’ period rather than when the ‘usual’ period is used. Chapter 5 investigates the role of self-control problems on prepayment patterns for electricity provided by a solar hybrid mini-grid installed in the semi-urban community of Bambadinca in Guinea-Bissau. Prepayment patterns are found to be mostly driven by income constraints and equipment in use however there is evidence that individuals with self-control problems as well as individuals being charged with an additional time-varying tariff (a higher tariff between 7pm to 12am) resort to smaller refill levels possibly as a strategy to consume less electricity at home. Chapter 6 provides concluding remarks.