Hyperbolic Discounting in Energy Consumption
Last registered on November 01, 2018


Trial Information
General Information
Hyperbolic Discounting in Energy Consumption
Initial registration date
October 29, 2018
Last updated
November 01, 2018 8:51 PM EDT
Primary Investigator
University of Münster
Other Primary Investigator(s)
Additional Trial Information
In development
Start date
End date
Secondary IDs
In a laboratory environment, we analyze the role of payment structures in energy consumption. Energy use is billed after consumption has taken place, such that a dynamic trade-off between consumption and payment of the good emerges. Such “bill-me-later”-systems give rise to uncertainty, inattention and (hyperbolic) discounting of future costs. This research focusses on the discounting of costs, by, holding information on and saliency of costs constant, randomly changing only the timing of the energy bill.
Building on recent laboratory and field experiments on present bias, we adapt the common setting with immediate effort costs and delayed benefits to a setting with immediate benefits and delayed costs. The benefits and costs are framed as consuming light and paying an energy bill. The control scenario is billing one week after consumption has taken place, the treatment scenario is billing immediately after consumption. This change in the payment structure allows us to estimate the influence of (hyperbolic) discounting on energy consumption.
External Link(s)
Registration Citation
Werthschulte, Madeline. 2018. "Hyperbolic Discounting in Energy Consumption." AEA RCT Registry. November 01. https://www.socialscienceregistry.org/trials/3503/history/36623
Experimental Details
In the control scenario, energy costs are subtracted from the second payment, i.e. one week after consumption. In the treatment scenario, energy costs are subtracted from the first payment, i.e. immediately after consumption. We only vary the timing of paying the energy costs.
Intervention Start Date
Intervention End Date
Primary Outcomes
Primary Outcomes (end points)
Absolute amount of "light" consumed (in seconds), relative share of "light" consumed (in percentage, i.e. light seconds divided by total seconds needed for all tasks)
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
We employ a within and between design in a longitudinal study over four dates. On dates one and three, subjects do real effort tasks in the laboratory and receive a first payment. On dates two and four, they just collect a second payment. There is always a week distance between each date. We randomize the timing structure, meaning that a group A experiences the control scenario on dates one and two and the treatment scenario on dates three and four, and a group B experiences the treatment scenario on dates one and two and the control scenario on dates three and four. This allows us to compare decisions between subjects on dates one and two and within subjects across all four dates.
The real effort task is designed to reproduce the energy consumption decision. The task is to find a certain letter in a table full of letters. The letters are shown with weak contrast (i.e. black letters on a grey background). For each task, subjects can decide the amount of “light” they want to consume. Switching on the light increases the contrast and therefore eases the task. We count the seconds light is switched on. The number of tasks is fixed. A price for each second of light is charged and subtracted from either the first or the second payment.
Experimental Design Details
Randomization Method
Within-session randomization done by a computer.
Randomization Unit
Was the treatment clustered?
Experiment Characteristics
Sample size: planned number of clusters
200 individuals (see below).
Sample size: planned number of observations
In a non-incentivized pre-test we measured a standard deviation for light consumed (in seconds) of 112. Assuming a treatment effect of 10%, as measured in comparable studies, the required sample size for a between-subjects analysis is 200 subjects (power: 0.8, alpha: 0.05).
Sample size (or number of clusters) by treatment arms
100 individuals in group A and 100 individuals in group B.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB Name
IRB Approval Date
IRB Approval Number
Post Trial Information
Study Withdrawal
Is the intervention completed?
Is data collection complete?
Data Publication
Data Publication
Is public data available?
Program Files
Program Files
Reports and Papers
Preliminary Reports
Relevant Papers