Robert Letzler, Ryan Sandler, Ania Jaroszewicz, Isaac Knowles, and Luke M. Olson, Knowing When to Quit: Default Choices, Demographics and Fraud, Oct. 8, 2014
A long literature in psychology and economics has shown that default options influence consumer choices, but it is often unclear whether individual consumers are nudged to choose optimally or simply nudged to a different choice. We study the effects of default options in a novel setting where the optimal choice is clear: the decision to escape from fraud. We employ data from one of the largest telemarketing fraud cases ever brought by the Federal Trade Commission (FTC). The telemarketer enrolled consumers into costly membership programs, which the vast majority of consumers never used. A court order issued during the FTC lawsuit created a natural experiment whereby some consumers were sent “opt-in” letters informing them they had to take action to remain enrolled while similarly situated consumers received “opt-out” letters that merely reminded them how to quit. We find that the “opt-in” letters increased cancellations by 63.4 percentage points, to essentially 100%. We then examine heterogeneity in the responses to the “opt-out” letters . We find that consumers residing in poorer, less educated Census blocks and those more likely to be minorities were more likely to cancel their subscriptions prior to the FTC lawsuit, but were relatively less likely to respond to an opt-out letter.
Conclusion, indicating that providing information isn’t enough when the default option is wrong, and that the information-only situation disproportionately harms poorer and minority consumers:
A large literature on the effects of default choice structures shows that agents are more likely to choose the default option, compared to other options. In this paper, we show that this is true even when the optimal decision is clear. Our results further indicate that informational interventions are not always an effective way of encouraging consumers to make those optimal decisions. Conversely, our results suggest that changing defaults is not a panacea when optimal choices are less clear. A standard model for a “nudge” policy involves enrolling consumers into a supposedly beneficial program and requiring them to actively opt out if they do not want to remain enrolled. It would not be surprising in a study of such a program to find only 30% or so of the target population opting out. However, in the case we study, it was likely optimal for every consumer to opt out, and relatively few did.
We also find evidence that the information intervention of the opt-out letter had heterogeneous effects across demographic characteristics, with consumers in low SES neighborhoods and those more likely to be minorities less likely to respond to the letters. Thus, the information provision policy disproportionately benefited consumers likely to be more well-off financially. Although the differences across demographic characteristics were smaller than the overall effect of the opt-out letter on cancellation, setting the correct default had bigger benefits for subscribers from lower SES neighborhoods than for subscribers from higher SES neighborhoods.