| dc.description.abstract | This thesis consists of three essays on behavioral and financial economics. Each chapter provides insights into how consumers interact with various financial products, and offers policy recommendations related to consumer protection, behavior change, and financial inclusion. The first two essays (Chapters 2 and 3) employ randomized controlled trials to test whether nudges and incentives induce behavior change among consumers in the retail financial market. The third essay (Chapter 4) uses survey data to examine the determinants of the financial performance of savings groups. Chapter 2 investigates the impact of consumer disclosure formats on digital credit choice, motivated by pervasive evidence of inaccessible loan terms and price obfuscation in the Philippine digital credit market. Using an online discrete choice experiment with 4,000 digital credit users, we find that standardizing of complete loan attribute information leads consumers to choose digital loans with lower effective interest rates by 2.70 percentage points and a higher probability of approval at the expense of longer disbursement times and more documentary requirements, compared to information displayed through traditional marketing campaigns. Under standardization, presenting interest rates in effective (compounded) or nominal terms makes no difference. A regulator may wish to go further to allow for ranking of product attributes. Ranking by effective interest rate leads to the choice of lower cost loans, reducing the effective interest rate by 1.54 percentage points compared with standardization. Similar effects are found for some non-price attributes. Finally, while on average, consumers are responsive to disclosures about late payment fees, overconfident consumers are not. Our findings feed into the long-standing discussion of how far regulators should go in shaping the information space for consumers. Chapter 3 evaluates the relative effectiveness of cash and in-kind incentives and informational SMS interventions designed to encourage debit card use. This chapter addresses an important gap in the behavior change literature as there is relatively little research that directly compares the effectiveness of these different approaches. In a large-scale field experiment involving 1.5 million bank customers, conducted in partnership with a commercial bank in Mexico, we test various interventions aimed at shifting consumer spending from cash to debit cards. The average treatment effects of our interventions are modest but typically greater than zero. Greater behavior change is found with increased monetary value of the incentive whether it is cash or in-kind, and the longer the incentive remains active. However, incentives with higher values do not necessarily come with better cost-benefit ratios. The best performing informational treatment, which highlights the security benefits of using a debit card instead of cash, performs equally as well as the largest cash incentive in terms of increasing the amount of debit card transactions. The exact wording of the informational message matters as we find no effects in other informational treatments. Furthermore, the modest average treatment effects mask significant heterogeneity in treatment effects, suggesting targeted marketing campaigns could be more cost-effective. Using causal forests, we find that the top quartile across all treatment arms exhibits higher positive treatment effects, while the other quartiles have treatment effects that are either not statistically different from zero or negative. Finally, for customers in the top quartile of predicted treatment effects, we find no evidence of either an increase or a decrease in total spending. Chapter 4 examines the relationship between trust, formal rules, and financial performance of savings groups (SGs). SGs have become an integral component of anti-poverty and financial inclusion strategies in low-income countries, yet empirical evidence surrounding what determines their financial performance is limited. Using survey data from 420 SGs across Western Kenya and applying instrumental variable analysis to address the endogeneity of trust, we provide evidence of a causal relationship between trust and the return on savings in SGs. We also find evidence that the relationship between trust and formal rules is substitutive rather than complementary. Heterogeneity analysis suggests that formal rules become less important in high-trust SGs. These findings may temper efforts to scale and link SGs with financial institutions as trust between members remains an alternative driver of SG performance. | en |
