ABSTRACT
Modern and gig economy businesses collect voluntary contributions (i.e., tips) from consumers via screen-based payment systems (i.e., $1, $2, $3; 10%, 15%, 20%). The use of these systems has been criticized by the popular media for forcing consumers to leave large tips in contexts where they previously would have left small tips or where tips were not required. The authors employ a multi-method approach, including an analysis of secondary data (N = 51,825), a field experiment (N = 1,810), and laboratory experiments (N = 2,321) to show that an absolute dollar frame (vs. percentages) leads to higher tip payments especially for low bill amounts. These effects are attenuated when (1) absolute options are presented in cents (e.g., $0.50), leading consumers to infer that small tip amounts are acceptable and, (2) absolute options start at high levels. Countering conventional wisdom, the authors further show that open-ended formats can lead to higher tip payments compared to closed-ended response formats in specific conditions. Theoretically, these results add to the behavioral pricing, prosocial behavior, and labor economics literatures. Managerially, the results are relevant for decision makers in the multi-billion-dollar digital service industry.
ABSTRACT
This research examines how the choice architecture of tip options used in screen-based payment collection systems affects consumers’ tipping behavior. Eight lab and field experiments show that consumers choose to avoid a numerical zero tip option (i.e., 0). Replacing the dominant opt-out default “No Tip” with 0% in a choice set, nudges people to opt-in to tipping. This effect is robust to bill size, ranges of alternatives, service level, order presentation of defaults and is mediated by self-image concerns. Furthermore, replacing a non-zero option (i.e., 1%, 5%, 10%) with 0% when a “No Tip” option is also present as an additional means of opting out, counterintuitively leads consumers to tip higher amounts. This work further builds on the survey method literature to show that the number 0 may not be used as a source of information to make tip judgments as this option is ignored. These results have theoretical implications for choice architecture, numerical cognition, prosocial behavior, and behavioral pricing. Importantly, these findings provide practical implications for consumer and labor welfare, and to businesses within the new age of the digital service economy.