What can we learn from experience? An impact analysis of experience on households' preferences for microfinance

Purpose - This article examines the impact of experience on rural households' preferences for microfinance attributes, using household data from Sichuan province in China. Design/methodology/approach - We use the Bayesian updating method to account for the learning process involved in acquiring experience on microfinance. We then use the generalized multinomial logit model that accounts for both preference and scale heterogeneity to estimate the choice probabilities and impact of experience on preferences and willingness-to-pay (WTP) for microfinance. Findings - The empirical findings show that experience with microfinance products or lending institutions helps households in their selections of microfinance institutions. In particular, experience with financial institutions increase the scale parameter and help respondents to feel assured about their choices, while experience with individual lenders have no such effects. The results also indicate that the willingness-to-pay estimates vary across experiences, with WTP for installment credit displaying the biggest change and the Bayesian updating making the changes even much larger. Originality/value - This study contributes to the research gap by taking experience as a latent variable that influences personal specific tastes and as an integral part of total utility of rural households in their decision-making processes. We develop a conceptual framework that is based on McFadden's random utility theory and Bayesian inference.