Experimental Design Details
To allow us to test our hypotheses, respondents will first be asked to participate in 4 experimental tasks to investigate their contributions to public goods in different scenarios. The order of these decisions is randomised to reduce any order effects, and the order of presentation will also be controlled for in the regression analysis. The tasks include two public goods game tasks, where these tasks follow the typical public goods game set-up: respondents are given a monetary endowment, and are asked to stipulate an amount (which can be zero) from this endowment that they would like to put into a group investment. Participants are informed that contributions from the four members of their group (including themselves) are doubled, and that the resulting amount will be divided equally among the four group members. Any amount that participants do not contribute to the group investment are kept. Since we use a multiplier of 2, this gives a marginal per capita return on investment of 0.5. That is, our public goods game follows the common approach where the dominant strategy equilibrium outcome for participants concerned only with their own payment would be to contribute nothing; while the social optimum is for all participants to contribute their full endowment. One of the public goods game tasks has all four participants receive the same endowment: R50. The other has unequal endowments, where half of the participants are given an endowment of R30 and the other half of the respondents are given an endowment of R60. In this task, all participants know that their group includes 2 participants with R30 and 2 participants with R60.
The other two tasks use a modified dictator game, where participants again receive an endowment, and must now choose how much of the endowment to keep and how much to contribute to a public good, in this case the Solidarity Fund. We vary the impact of contributions in these tasks by having two scenarios. The first allows for a direct comparison with the equal endowment public goods game: participants receive an endowment of R50, and any contributions are multiplied by 2. The second scenario allows us to directly compare contributions with higher and lower impacts: here the endowment remains R50, but contributions are muliplied by 4, doubling the impact of each rand contributed.
In all cases, examples are used to explain the payments resulting from possible decisions. To reduce possible experimenter demand effects, examples include a range of possible contributions.
To allow us to investigate the impact of beliefs about others' contributions on contribution decisions, we ask participants to report what they believe to be the average contribution of other participants for each of the games. This task is incentivised by paying the participant guessing closest to the true average for each game an incentive of R50.
In addition to these experiment questions, we ask participants to complete a short survey on their household income, education level and demographics (age, gender, race). We also ask respondents whether they own a small business, and how the small business has fared following Covid-19 and the associated lockdowns. This will allow us to see whether people who stand to benefit directly from the Solidarity Fund are more or less likely to be willing to contribute to this fund.