Research

Individuals are frequently tempted to delay tasks to the detriment of their well-being – even when they know they shouldn’t. I propose that costly procrastination increases when excuses, or justifications, are present. Using a survey, I document an emotional cost of procrastination that excuses attenuate. I explore the behavioral impact of this psychic cost using a within-subject longitudinal laboratory experiment. I show that participants exploit uncertainty (that can be costlessly resolved) over future work amounts as an excuse to select options with less immediate work that risk increasing total work. While 91 percent of participants minimize overall work when given full information, 37 percent avoid the information needed to do so when given the chance. The presence of an optional excuse increases procrastination fourfold and overall workload by 10 percent. Thus, the presence of an excuse has important implications for inference of individual time preferences. These results also help explain long-standing puzzles, such as why individuals do not learn from their past myopic mistakes.

This paper explores streaks as a novel technique to counteract impatient behavior in economics decision-making when actions have current costs but delayed benefits. Streaks are a powerful psychological motivator triggered by tracking the number of consecutive periods an action is performed, increasing the motivation for current effort. However, myopic reactions to broken streaks can lead to a tradeoff that causes streaks to backfire. We use a longitudinal real-effort experiment to study this tradeoff. We find that, conditional on working the day before, being randomly assigned to having consecutive days worked tracked, as opposed to total days, significantly increases the probability of working. However, conditional on \textit{not} working the day prior, those in the streaks treatment work significantly less. Allowing for flexibility, such as cheat days when effort costs are high, can act as an insurance against breaking streaks, thus mitigating both the motivating and demotivating aspects. We find that the Streaks treatment performs better in terms of payment when individuals do not experience an exogenous cost shock, while the most flexible treatment performs better when there is one. The Cheat Day treatment falls in between, with no difference in payment depending on cost shock. 

Draft forthcoming

Claiming credit for collaborative contributions is an essential part of many hiring and promotion processes. We experimentally test whether people accurately claim how much they contributed in real-effort group tasks and find evidence of systematic overclaiming. We explore factors that could exacerbate or reduce overclaiming such as imperfect memory, the degree of ambiguity , and social pressure. Overclaiming decreases when participants have to discuss respective contributions with teammates prior to making claims.  

Experimenter demand can threaten the validity of experimental results. To understand the extent of this threat, we apply the quantitative framework from de Quidt, Haushofer and Roth (2018) to explore if experimenter demand generates mistakes in core qualitative inferences drawn from experimental studies. We study this using four classic behavioral findings with clear directional predictions. We induce the potential for false negatives in settings that are well-powered replications, and false positives by using an intentionally under-powered sample. In all settings, we also examine whether we can generate comparative static reversals, a stronger test of the potential distortions resulting from experimenter demand. Using a novel laboratory population, we demonstrate that even in a stark environment with deliberate researcher attempts to manipulate participant behavior, quantitative effects are small and experimenter demand does not impact key inferences made from experimental studies.

Using observational credit report data, we show that a substantial subset of financially distressed borrowers who hold a varied debt portfolio avoid defaulting on revolving credit, resulting in an eventual default on their student loans. The severity of student loan default compared to credit card default makes this a suboptimal financial decision which has negative consequences at both the individual and aggregate level. We explore how the different time horizon of default and therefore penalties of these debts influence default decisions using a survey and find that participants rank the timing of default as the most important factor in influencing default decisions. Additionally, we find a higher level of financial literacy for credit cards than student loans.

Excuse-seeking behavior that facilitates replacing altruistic choices with self-interested ones has been documented in several domains. In a laboratory study, we replicate three leading papers on this topic: Dana et al. (2007), and the use of information avoidance; Exley (2015), and the use of differential risk preferences; and Di Tella et al. (2015),  and the use of motivated beliefs. The replications were conducted as part of a graduate course, attempting to embed one answer to the growing call for experimental replications within the pedagogic process. We fully replicate the simpler Dana et al. paper, and broadly replicate the core findings for the other two projects, though with reduced effect sizes and a failure to replicate on some secondary measures. Finally, we attempt to connect behaviors to facilitate the understanding of how each fit within the broader literature. However, we find no connections across domains.

This guide provides a detailed account of procedures for conducting traditional in-person laboratory experiments in a “virtual setting.” The main objective of these procedures is to maintain the control of traditional in-person lab studies when conducting studies over the internet. Using the participant pool of the in-person lab the key procedural steps include participants having their webcams on throughout the experiment, technical screenings and attention pledges, playing pre-recorded instructions out loud, upholding clear experimenter roles and communication protocols when interacting with participants, and finally detailed and scripted procedures for managing participants throughout the session. The described procedures have been used for more than 100 sessions and have secured results that are indistinguishable from those from the in-person lab.