Experiment: The effect of unconditional basic income and human capital accumulation on lifetime consumption and leisure pathways

Last registered on November 01, 2023

Pre-Trial

Trial Information

General Information

Title
Experiment: The effect of unconditional basic income and human capital accumulation on lifetime consumption and leisure pathways
RCT ID
AEARCTR-0012358
Initial registration date
October 25, 2023

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
November 01, 2023, 9:14 PM EDT

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

Locations

Region

Primary Investigator

Affiliation
Radboud University

Other Primary Investigator(s)

PI Affiliation
Radboud University

Additional Trial Information

Status
In development
Start date
2023-11-13
End date
2024-11-16
Secondary IDs
-
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
The study considers the effect of unconditional basic income on consumption, leisure and human capital accumulation decisions in a Life-Cycle Consumption Theory experiment following Duffy & Li (2019).
External Link(s)

Registration Citation

Citation
Füllbrunn, Sascha and Daniel Waters. 2023. "Experiment: The effect of unconditional basic income and human capital accumulation on lifetime consumption and leisure pathways." AEA RCT Registry. November 01. https://doi.org/10.1257/rct.12358-1.0
Sponsors & Partners

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

Request Information
Experimental Details

Interventions

Intervention(s)

Intervention (Hidden)
The experimental design is explained below. We consider the baseline treatment (B), a treatment where unconditional basic income is added to the baseline (U), a treatment where participants can accumulate human capital (H) and a treatment combining the two latter treatments (UH). Hence, we have a classical 2x2 design with the dimensions UBI (UBI and NoUBI) and human capital accumulation (HCA and NoHCA).
Intervention Start Date
2023-11-13
Intervention End Date
2023-11-16

Primary Outcomes

Primary Outcomes (end points)
The experiment yields key outcome variables in tokens (experimental currency units) and euro cents. At period level, we consider leisure in terms of utility (=eurocent), effort in terms of fraction of capacity, consumption in tokens and utility (eurocents), and whether participants decided to invest in human capital (Yes/No). At the lifetime level, we consider lifetime income (in tokens), lifetime utility (leisure + consumption), and human capital stock (level of human capital achieved). We contrast these empirical variables with the conditional and unconditional optimal decisions.
Primary Outcomes (explanation)
Next to absolute measures as taken from the experiments, we will consider measures to be different from theoretical conditional and unconditional consumption benchmarks following Duffy and Li (2019) in the first case but invent a new benchmark in the second case.

Secondary Outcomes

Secondary Outcomes (end points)
To control misunderstandings, we ask for self-reported abilities to speak English (we have a subject pool with mostly Dutch students), do mathematical calculations, and understand the instructions. In addition, we ask about the general willingness to take risks, as part of the tasks includes risk-taking. We make use of a 10-point Likert scale.
Secondary Outcomes (explanation)
see above

Experimental Design

Experimental Design
During a 25-period lifetime consisting of a working phase and a retirement phase, the participants decide how much to work - to earn an income; the residual leisure earns utility, whether to invest in human capital - to improve productivity, yielding a higher income, and to consume income - yielding consumption utility – or save – earning interest. The project enriches the literature on life-cycle models (Duffy and Li, 2019), unconditional basic income (Füllbrunn et al., 2019), but stated-effort experiments (e.g. Charness et al., 2018).

- Charness, G., Gneezy, U., & Henderson, A. (2018). Experimental methods: Measuring effort in economics experiments. Journal of Economic Behavior & Organization, 149, 74-87.
- Duffy, J., & Li, Y. (2019). Life-cycle consumption under different income profiles: Evidence and theory. Journal of Economic Dynamics and Control, 104, 74-94.
- Füllbrunn, S., Delsen, L., & Vyrastekova, J. (2019). Experimental Economics: A Test-Bed for the Unconditional Basic Income?. Empirical Research on an Unconditional Basic Income in Europe, 171-199.
Experimental Design Details
The experiment reflects a ‘lifetime’ of 25 periods, 17 ‘working’ periods (WPs) and 8 ‘retirement’ periods (RPs). In each period, the participants decide on how much to consume, c_t, or save, s_t, given a token inventory of w_t=(1+r)s_(t-1)+y_t, the sum of last period’s savings next to interest and this period’s income (during WPs only). During WPs, we enforce a compulsory consumption of c ̅ . The consumed tokens convert to eurocent via the ‘consumption utility’ function u_c (c_t )=C_t=a_c×ln⁡(b_c (c_t-c ̅)+d_c ), i.e. every additional token consumed has a lower eurocent value. The parameters a, b, and c shape the utility function in a meaningful way. The compulsory consumption yields no euro cents. During the RPs, the token inventory reduces to w_t=(1+r)s_(t-1). In the very last RPs, c_T=w_T holds. The income y_t equals the excerpted effort e_t∈{0,E} times a wage factor w(h_t) depending on a human capital level, h_t, such that y_t=w(h_t )e. We see E as the period capacity that can be used for working (e_t) or leisure l_t=E-e_t. Hence, the period capacity not used for work yields euro cents via the ‘leisure utility’ u_l (l_t )=L_t=a_l×ln⁡(b_l (E-e_t)+d_l ), i.e. every additional leisure unit has a lower eurocent value. Within a working period, the participant first decides on e_t, then learns about w_t, and finally decides on c_t. The environment described so far reflects the baseline treatment (B) without unconditional basic income (UBI) and human capital accumulation (HCA). The following treatment adds the additional features.

Treatment U (the UBI treatment)): We add the unconditional basic income to treatment B in that we add a fixed payment ubi_t=ubi to the period token inventory such that w_t= y_t+(1+r) s_t+ubi. The RPs earn no UBI.

Treatment H (the HCA treatment): We add human capital accumulation as follows. Participants can opt to undertake ‘training’ to improve human capital level h_t∈\{1,2,…\}, and thus their wage level w(h_t )=w×h_t. A training period needs the entire capacity such that neither income nor leisure utility can be generated, i.e. y_t=0 and L_t=0. Moreover, not everyone will make it to the next level; the training is a memoryless stochastic process, with a 2/3 chance of improving to the next level, during the RPs, y_t=0 and L_t=0.

Treatment UH combines the treatments U and H, i.e. participants earn ubi and can train to increase their human capital level.
Setup: In a between-subjects design, the participants go through two lifetimes. The first one is practice; the second one translates to their Euro payment. The participant's payment equals the show-up X plus the earnings from the second lifetime, i.e. Total=X+ [∑ _(t=1)^25 C_t ] + [ ∑ _(t=1)^17 L_t .

We will run the experiment in an in-person lab setting, recruiting students via ORSEE (Greiner, 2015) who will sit at a computer cube submitting decisions using zTree software (Fischbacher, 2007).

Based on simulations with our model adjustment of the Duffy & Li (2019) setup, we set the value in line with the optimal decision path {c_ip^(**),e_ip^(**),h_ip^(**) \}_(p=1)^P such that optimal players will earn the average student per-hour salary.
Randomization Method
Participants subscribed to Orsee (Greiner, 2015) and will be invited to participate considering a list of sessions. Participants can only take part in one session.
Randomization Unit
Individual students take part in the sessions, i.e. there is no clustering.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
We invite students from one faculty.
Sample size: planned number of observations
We plan to have 200 student participants.
Sample size (or number of clusters) by treatment arms
We plan to have 50 students per treatment; the Duffy & Li (2019) experiment had about 30 student participants.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Simple treatment comparisons allow us to detect an effect size of 0.565 or higher, assuming α=0.05 and 1-β=0.8 (Faul et al. 2008). Based on our new environment and the different dependent variables, we cannot meaningfully provide information at the unit, standard deviation, or percentage level. Faul, F., Erdfelder, E., Lang, A. G., & Buchner, A. (2007). G* Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior research methods, 39(2), 175-191.
IRB

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number

Post-Trial

Post Trial Information

Study Withdrawal

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

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials