Loosing vs. Gaining Control: Enhancing Feelings of Control Reduces Present Bias

Consumers regularly make decisions about the timing of their consumption. For instance, they decide how fast to ship an online purchase, when to take vacations, or when to make a large purchase. Extant research on such decisions has shown a variety of consumer phenomena, most notably present bias (a.k.a. hyperbolic discounting). Preferences are considered present-biased when a consumer is willing to pay higher premiums to avoid a 3-day delay in shipment (e.g., $5, or $1.8 per day) than a 10-day delay (e.g., $10, or $1 per day; Malkoc and Zauberman 2006). A substantial amount of research has investigated the underlying mechanisms for present-bias. The last decade has seen a surge in research that theorized, and empirically tested, a host of cognitive explanations, based on changes in mindsets (Malkoc, Zauberman and Bettman 2010), representation of the outcomes (Malkoc and Zauberman 2006), resources (Zauberman and Lynch 2005) or time itself (Zauberman et al. 2009). In this research, we propose a more affective mechanism to account for present-biased preferences. Prior research has theorized that forgoing a current outcome is painful and leads to a feeling of deprivation (Hoch and Lowenstein 1991), resulting in impulsiveness and impatience; however, this account has not been empirically tested. More importantly, the theory is not precise as to why the act of deferral is so painful. Based on this notion, we propose that the mere act of consumption deferral leads consumers to feel that they have less control of their consumption. In a typical present bias elicitation task, participants are asked to imagine having an outcome in the present (e.g., a DVD available today) and then asked to delay this outcome for monetary gain. Such a proposition elicits a feeling of lack of control over the consumption decision. Therefore, we propose that when asked to forgo a presently available outcome, consumers feel they have less control over their consumption. As a result, consumers, now wanting to regain control, represent the subsequent intertemporal choice as an opportunity to gain control, requiring higher premiums in their immediate responses and resulting in present bias. We investigate this proposition in four studies. Experiment 1 shows that giving people control over their consumption timing decreases present bias. Experiments 2 and 3 demonstrate that a more generalized sense of power and control leads to similar reductions in present bias. Finally, experiment 4 shows that decreasing the level of control can increase present bias. In experiment 1, using a typical present bias paradigm, we asked participants to imagine having ordered a DVD online where their default delivery was scheduled for the same day. They then imagined delaying their delivery by 3 days and 10 days (within-subjects) and asked to indicate the premium they required to delay. Participants in the “choice control” condition were first asked whether they would like to delay the delivery before they indicated their delay premium. We expected that merely having a chance to express their willingness to delay to make participants feel more in control of their consumption decision, decreasing present bias. Results supported this prediction, with significantly lower level of present bias in the choice control condition (M3days-10days = .45), compared to the standard present bias paradigm (M3days-10days = .66; t(147) = 2.04, p .05). In experiment 2, we examined whether a generalized and previously evoked sense of control could influence the extent of present bias in subsequent tasks. To evoke control, we induced participants with power, which has been found to directly affect one’s sense of control (Fast et al. 2009). Since high power individuals feel they have more control over their outcomes, we predicted that high power would diminish present bias. To test this idea, half of the participants recalled an incident where they had power over others, a task that has been successfully used to induce power (Galinsky, Gruenfeld, and Magee 2003). The other half simply recalled events of the previous days. Next, all participants imagined having a cash refund ($45) that was available that day and asked to delay it by 4 and 10 weeks. As predicted, participants who were induced with power showed less present bias (M4weeks-10weeks = 1.29), compared to the control condition (M4weeks-10weeks = 2.62; t(97) = 1.78, p = .07),. In experiment 3 we replicated these findings using an alternative method to induce power (via an employee/employer task, Dubois, Rucker, and Galinsky 2010). In the final experiment we examined whether taking control away from participants would have the opposite effect and increase present bias. To that end, some participants completed a deferral task where they imagined ordering a DVD online, scheduled to arrive within 24 hours. They were then asked to indicate their premium to delay it by 3 days and 10 days. The remaining participants (low control) were instructed to imagine that they chose to pick up their DVD from a local store. However, when they arrived at the store, their DVD was not available, but the clerk offered to have it delivered to their home within 24 hours. They were then asked to indicate their delay premium to defer the receipt of this DVD by 3 days and 10 days. We predicted and found that having already lost control over their consumption, these low control participants showed greater present bias (M3days-10days = .66), compared to those in the baseline condition (M3days-10days = .38; t(171) = 2.26, p .05). Moreover, we found that feelings of control over consumption mediated the observed effect. In sum, our studies show that the feeling of control in related (experiments 1 and 4) and unrelated (experiments 2 and 3) contexts influences the extent of present bias. Further, we consistently find that decreasing (increasing) feelings of control increases (decreases) present bias. Taken together, this research suggests that the feeling of control plays an important role in present bias, and demonstrate a more affective explanation for present bias.