The Downside of Defaults

The use of default options to influence behavior in the retirement savings arena has expanded significantly in recent years. While the use of defaults is sometimes portrayed as a Pareto improvement because it guides behavior without constraining individual choice, the welfare implications of defaults depends critically on why people default and whether the default is well-suited to those affected. In this paper, we use survey and administrative data on participants in a large public retirement system to explore who defaults on an important economic decision, why they default, and whether they subsequently regret their decision (or lack of decision). We find that information problems are an important contributor to the likelihood of default. We show that the likelihood of default increases with information problems, and that this holds true even after controlling for general and decision-specific knowledge, preferences and beliefs, and a variety of socioeconomic characteristics. We also find significant heterogeneity in the self-reported reasons for default, with sizable numbers of participants attributing their default to each of the commonly hypothesized reasons (e.g., endorsement effects, complexity, and procrastination), as well as to "deliberate defaulting" by those who believe the default option represented the best choice, and to beliefs that the decision was not important. We also find that individuals who are passively defaulted are substantially more likely to regret their decision (as measured by whether they would make the same choice today) than individuals who make an active choice. In total, our results suggest that there are potentially important negative welfare consequences when default options are used for complex, high-stakes welfare decisions.