Back to History Current Version

Excuse-driven present bias

Last registered on September 10, 2019


Trial Information

General Information

Excuse-driven present bias
Initial registration date
June 07, 2019

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
June 21, 2019, 11:39 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Last updated
September 10, 2019, 1:47 PM EDT

Last updated is the most recent time when changes to the trial's registration were published.



Primary Investigator

Central European University

Other Primary Investigator(s)

PI Affiliation
Central European University

Additional Trial Information

In development
Start date
End date
Secondary IDs
In this study we measure whether and by how much people behave in a more present biased way when they can use excuses to rationalize their present-biased behavior.
External Link(s)

Registration Citation

Drucker, Luca Flóra and Marc Kaufmann. 2019. "Excuse-driven present bias." AEA RCT Registry. September 10.
Former Citation
Drucker, Luca Flóra and Marc Kaufmann. 2019. "Excuse-driven present bias." AEA RCT Registry. September 10.
Experimental Details


In our experiment, we measure how subjects change trading off current vs future effort when they have excuses available to them (in the form of risk) and when they have no excuses available to them. The main intervention consists in randomizing both whether an excuse is available and whether an excuse is desirable. By desirable we mean whether there is anything that the person might want to excuse, which means whether it would justify working less right now -- as opposed to justifying working less later.
Intervention Start Date
Intervention End Date

Primary Outcomes

Primary Outcomes (end points)
We measure how much subjects value a given option, which consists in a small probability of not having to do work in the future. Our first primary outcome is the difference in this option's value when the option can serve as an excuse compared to when it cannot.

Our second primary outcome (secondary design) is whether subjects use risk on money as an excuse to distort their trade-offs in effort allocation always in favor of working less in the present.
Primary Outcomes (explanation)
See detailed description in the Experimental Design section.

Secondary Outcomes

Secondary Outcomes (end points)
Our secondary outcome is the change in our measure of present bias under the assumption that all of the change comes from a distortion of present bias -- or equivalently, if one estimated a structural model as in Augenblick and Rabin (2018) without accounting for the role of excuse-making.
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Our design allows to estimate quantitatively how much present-biased behavior increases due to having excuses available, but it requires us to difference out the impact of risk.
Experimental Design Details
Not available
Randomization Method
Done by computer
Randomization Unit
Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters
In our initial pilots, we will recruit subjects from MTurk.

80 individuals for the pilot in which we test sliders vs price lists; and to calibrate the number of tasks and payment levels for the main study.

We are aiming for 150-300 subjects for the main study. This will depend on:

(i) How much we need to pay to ensure low level of attrition (main concern)
(ii) How large difference in utility there is between doing the work today compared to tomorrow for different levels of effort

The idea for point (ii) is that if subjects don't care much about the difference between working today and tomorrow (say, because it is only 5 tasks), then there may be little reason in the first place to make an excuse. If however they do care, then the incentive for excuse-making is higher. (We have no proof that this is the case, but is more plausible than assuming that subjects want to excuse the small but not the important.) Given that we will incur large fixed costs to ensure low attrition, we will almost surely benefit from cranking up our treatment effect, which we think grows with the size of what can be excused.

For example: if we need to pay $15 to ensure that everyone comes back, and if we get a difference in stated utility of $0.50 vs $3 when subjects do merely 5 tasks vs 20, it may be better to have 20 tasks rather than 5 tasks, since we get a noticeable difference in utility (and thus hopefully something worth excusing), especially since the fixed costs per person are $15 + variable costs.
Sample size: planned number of observations
Based on the pilot we will determine the size of the primary and secondary designs (see power calculations).
Sample size (or number of clusters) by treatment arms
Based on the pilot we will determine the size of the primary and secondary designs (see power calculations).
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
We will compute the power after our pilot, based on the standard errors and precision of the methods we try out there. The constraints will be determined by (a) the outcome of the pilot and the SE of our measurement methods (b) our fixed research budget (in the short term at least) and (c) the required amount per participant that gives a fair payment and avoids attrition.
Supporting Documents and Materials

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Institutional Review Boards (IRBs)

IRB Name
CEU Ethical Research Committee
IRB Approval Date
IRB Approval Number