Defaults: a double-edged sword in governing common resources

TL;DR

The paper investigates how simple default extraction values in a CPRD influence collective and individual outcomes. By comparing a Self-serving default, a Pro-social default, and a no-default control, the study shows that selfish defaults increase extraction during the nudged rounds while pro-social defaults reduce extraction only briefly, with effects dissipating after the default is removed. Personal preferences, particularly SVO and risk attitudes, modulate these effects, revealing heterogeneous responses and reinforcing the idea that nudges can be more effective when aligned with target groups. The findings highlight the potential for low-cost, ethically deployed defaults to promote sustainability, while cautioning against unintended backfire and emphasizing the need to consider user heterogeneity and context. Overall, defaults act as non-persistent levers that can shape resource use in the short term but require careful design to avoid adverse outcomes.

Abstract

Extracting from shared resources requires making choices to balance personal profit and sustainability. We present the results of a behavioural experiment wherein we manipulate the default extraction from a finite resource. Participants were exposed to two treatments -- pro-social or self-serving extraction defaults -- and a control without defaults. We examined the persistence of these nudges by removing the default after five rounds. Results reveal that a self-serving default increased the average extraction while present, whereas a pro-social default only decreased extraction for the first two rounds. Notably, the influence of defaults depended on individual inclinations, with cooperative individuals extracting more under a self-serving default, and selfish individuals less under a pro-social default. After the removal of the default, we observed no significant differences with the control treatment. Our research highlights the potential of defaults as cost-effective tools for promoting sustainability, while also advocating for a careful use to avoid adverse effects.

Figures (11)

• Figure 1: Visual representation of the experiment designed for this work. We use the Common Pool Resource dilemma (CPRD) walker_rent_1990 to understand collective resource management. In this game, individuals extract resources (as tokens) and receive payoffs (in experimental units, or ECUs) proportionate to their extraction levels. Panel A shows four individuals who request 45 tokens in total. Panel B transforms these requests into a collective payoff of 52.87 ECUs, determined by a payoff function curve shown in panel C. (more details in the Methods section). D: Experiment flow. the experiment consists of three tasks: In task 1 (brown box), participants complete an incentivised Social Value Orientation task (SVO) murphy_social_2013, allowing us to identify participants' resource allocation preferences. Task 2 (green box), assessed participants' risk aversion (also incentivised) dave_eliciting_2010eckel_sex_2002. Finally, in task 3, participants engage in the CPRD as part of one of three treatments, which cover 2 different types of defaults and one treatment without a default for comparison (panel E): Pro-social (default = 11 tokens, optimal group extraction yielding 52.8 ECUs), Self-serving (default = 23, excessive extraction yielding no returns), and a Control with no preset extraction. The green dots represent the rounds, where the darker green represent the first five rounds with default values, and the lighter green the rounds with no default. More details on these treatments are provided in the Methods section.
• Figure 2: Mean extraction per treatment. A) Mean extraction by round and by treatment: Pro-social $n = 156$, Self-serving $n = 156$ and Control $n = 100$. The blue dotted line represents the Nash Equilibrium of the game, while the dotted lines on both extremes represent the default values presented to participants. Vertical lines represent $95\%$ confidence interval. (B, C and D) Estimated difference of mean extraction between treatments the red is the estimated difference, as given by the mixed-effect model, while the vertical lines represent the round where this significant difference can be found. The colour shaded areas represent the $95\%$ confidence interval from the model.
• Figure 3: Difference in extraction depending on participants' SVO over time. A and B: Estimated difference given by the mixed-effects model, where non-white parts show a significant difference between two treatments. The horizontal lines show the SVO categories for reference and the colours the difference in extraction. Negative (or positive) differences show a larger (or smaller) extraction than the results in the Control treatment. C and D: Mean extraction over time given by the experimental data, where the most pro-social categories of SVO (cooperative and altruist) are grouped in panel C and the most selfish categories of SVO (individualistic and competitive) are grouped in panel D. The highlighted points in both panels are the ones given by the model in panels A and B. Vertical bars show the 95% confidence interval.
• Figure 4: Screenshot of the third task. In this task, participants had to choose their desired extraction for ten rounds. At the top, the platform showed the round number and the time left to make their decision. The participants had their previous extraction and the extraction of the other members of the group, and their payoff, as shown on the left side of the figure. At the bottom, participants had a sandbox at their disposal to calculate their potential earnings depending on theirs and others' extraction.
• Figure 5: Group payoff as a result of a given group extraction. As shown in the green line, the maximum return the CPRD can provide is when participants extract 46 tokens as a group (average 11.5 tokens per player), which is shown with the yellow dashed line. If participants extract more or less than that quantity as a group, they will be under or over extracting the resource, respectively. The pink line shows that if participants extract 92 tokens as a group (average 23 tokens per player) they will all receive zero ECUs.