Schedule - Parallel Session 3 - Intertemporal Choices 1

WMG IDL03 (IDL Auditorium) - 15:40 - 17:10

Attention Shapes Valuation in Intertemporal Choice: A Test of Procedure Invariance

Lisheng He; Kenneth Lim; Marc Scholten; Adam Sanborn; Daniel Read


Background Violations to procedure invariance provide an avenue to understanding the cognitive processes underpinning choice and behavior. One approach to the relationship between cognitive processes and procedure variance is through the role of attention. Empirical studies taking this approach usually link visual attention to choices in a correlational fashion or manipulate attention by forcing options’ exposure time respectively. In this paper we propose (and test) a new method: Using the variation in an option’s context to manipulate the direction of attention. Methods We conducted two experiments to test how attention shapes intertemporal choice and thus induces violations to procedure invariance. In addition, we assessed whether the effects of attention are moderated by cognitive style, as measured by the Cognitive Reflection Test. Both experiments involved between-subject designs. In Experiment 1, participants made 81 intertemporal choices. In all items, the smaller-sooner (SS) option was held constant [$110 in 2 months], but the larger-later (LL) option varied in both the outcome [which could take nine values ranging from $120 to $200] and delay [ranging from 4 to 20 months]. The 81 choices were divided into nine blocks, either LL-outcome-varying (LOV) blocks or LL-delay-varying (LDV) blocks contingent on the experimental condition. In each LOV block only one value of LL delays appeared within the block, but all nine outcomes appeared. In each LDV block, the situation was reversed with the outcomes constant and the LL delays varying. In Experiment 2, the method was very similar except that the LL option was held constant, and there were nine values for the SS outcome and the SS delay respectively. In each SS-outcome-varying (SOV) block only one value of SS delays appeared within the block and, in each SS-delay-varying (SDV) block, only one value of SS outcomes appeared. Results Results from both experiments exhibited attention’s bidirectional effects. First, respondents in the LOV condition were more likely to choose the larger-later option than those in the LDV condition (Experiment 1); By contrast, those in the SOV condition were more likely to choose the smaller-sooner option than those in the LDV condition (Experiment 2). Second, whichever the varying attribute is, respondents were more sensitive to a change to this attribute than when it is kept constant in a block (both experiments). Cognitive reflection moderated attention’s effects too. Experiment 1 showed that high-reflection respondents were less susceptible to the attention framing than low-reflection respondents, while Experiment 2 showed a less significant moderation effect, however, in an opposite direction. It is also implied that preference is always uncertain and the preference uncertainty may differ across people with different cognitive styles, highlighting the importance of the measurement of preference uncertainty.

Lisheng He

Student, Warwick Business School

The Value of Nothing: Asymmetric Attention to Opportunity Costs Drives Intertemporal Decision Making

Daniel Read; Christopher Olivola; David Hardisty


We propose a novel account of intertemporal choice based on asymmetric attention given to opportunity costs. We argue that the costs of delayed consumption are more salient than the costs of earlier consumption. This produces a bias in favor of smaller, sooner rewards over larger, later ones. Our account implies that highlighting the opportunity costs of choosing smaller, sooner rewards should lead people to become more patient, whereas highlighting the opportunity costs associated with delayed gratification should have little or no effect (since people are naturally aware of these latter costs). Patience is measured through choices between smaller, sooner (SS) and larger, later (LL) payments, such as “$100 today OR $150 in one year.” The opportunity cost of choosing LL is that you will receive $0 instead of $100 today, while the opportunity cost of choosing SS is that you will receive $0 instead of $150 in one year. In several studies we show that reminding people of the SS opportunity cost increases patience, while reminding them of the LL opportunity cost has no effect. We do this primarily by extending a design introduced by Magen, Dweck and Gross (2008) who found that eliciting choices in the Explicit zero frame below produced greater patience: Hidden zero: $100 today OR $150 in one year Explicit zero: $100 today and $0 in one year OR $0 today and $150 in one year We show that this effect is due to adding the zero to the SS item. We will report on eight experiments. The first five reproduce the SS zero effect, and show that it is robust to outcome sign (it works for losses), for outcomes of varying magnitude, for delayed outcomes, and even when the wording of the options deviates from the “Zero” frame just described. We next show that the asymmetric opportunity cost effect holds when people are reminded of these costs in a different way – by being given a strong reminder that (for example) if they choose the later of two options, they will not get the earlier option. Finally we show the effect occurs for non monetary goods such as lives saved, air quality, and chocolates.

Daniel Read

Professor, Warwick Business School

Measuring Time Preferences and Anticipation: A Lab Experiment

Lea Bousquet


Several biases related to time preferences, in particular present and future biases, have important consequences on economic decisions. Present bias concerns individuals’ self-control problem. Future bias can be defined as the anticipatory emotions individuals feel in the period preceding the consequences of their decision. Individual’s acumen for anticipating also has economics consequences. The individual aware of his bias seeks to constrain his action to overcome the consequences of his bias. Whereas a naive individual does not anticipate his bias and might choose the wrong commitment or the wrong action plan. This paper proposes a methodology for eliciting anticipation of time preferences using a lab experiment. Though extensive literature exists on the measurement of time preferences and related biases, scant attention has been given to the elicitation of the anticipation of these biases. This paper reports on a lab experiment in two rounds with the same subjects. On the one hand, the second round elicits their time preferences regarding the temporal allocation of monetary rewards between two dates (CTB method of Andreoni and Sprenger (2012)). The comparison between allocation decisions for which the sooner allocation date is « Today » — the immediate present is involved — with the allocation decisions for which the two allocation dates are in the future allows me to elicit if they are present-biased, future-biased or if they exhibit no bias. On the other hand, the first round elicits their anticipation of their allocation choices. During this round, two weeks before the second round, the participants were asked to anticipate what they think their allocation decisions will be. By comparing their anticipated allocations at the first round and the ones chosen at the second round, I am able to determine whether participants are naive, i.e., if they underestimate their bias. It is assumed that participants have (β, δ) preferences (Laibson (1997)). Whereas δ represents the traditional long-term discount rate, β represents the short-term discount rate used only between the immediate present and the future. This latter parameter captures the relative weight participants attribute to immediate utility compared to future utility. Given, the estimated value of this parameter, it is possible to determine whether the participant is present-biased, future-biased or exhibits no bias. Moreover, I assume that the anticipation accuracy of their allocation decisions depends on their belief on their bias, i.e., their belief on their short-term discount rate. Estimating this belief and comparing it with the estimated value of β indicates whether biased participants are also naive. I find that even though a majority of the participants can not be considered as biased and accurately anticipate their time preferences, when they are biased, both present- or future-biased participants tend to be naive about their bias, i.e., they underestimate their bias.

Discounting Future Gains and Losses of Time

Cedric Gutierrez; Mohammed Abdellaoui; Emmanuel Kemel


In many decisions, saving or losing time, now or in the future, is as relevant as saving or losing money now or in the future. Benefits of projects such as automation solutions for high-speed production lines or enlarging expressways are economically justified by the value of time saved. It is however usual to assess the economic value of such saved/lost time by discounting the corresponding monetary values over time rather than discounting the very values of time over time. While intertemporal choice of money has been studied extensively, very few studies have analyzed the way people discount time, despite the fact that it is a scarce and valuable resource. We investigate this issue in a laboratory experiment where consequences are measured in units of time and real incentives are implemented in gains and losses. We created a concrete scenario in order to facilitate subjects’ understanding of the concepts of gains and losses of time, and to give them a common reference point from which gains and losses of time were defined. We used this scenario for the implementation of real incentives. We also measured intertemporal decisions with monetary consequences, as a benchmark. We estimated all the components of the discounted utility model using maximum likelihood. We took a descriptive perspective and allowed for discount rates to vary over time. We also measured utility of time and money in order to disentangle attitudes towards delays from attitudes towards outcomes. Finally, we used mixed modeling in order to take into account heterogeneity in intertemporal preferences. We find that subjects discount more gains of time than gains of money. This difference between time and money is reversed for losses: people discount less losses of time than losses of money. Moreover, we observe that subjects are more likely to exhibit non-constant discount rates for time than for money in both gain and loss domains. Analyses also show that there is a significant correlation between impatience towards money and impatience towards time: subjects who are impatient for money also tend to be impatient for time. Finally, mixed modeling analyses show that there is more heterogeneity in discounting behavior for time consequences than for monetary consequences. As a conclusion, we find that people do not discount time as they discount money. The paper also reveals the complexity and diversity of intertemporal preferences.

Cedric Gutierrez

PhD Student, HEC Paris