Diverse Groups Agreeing on a System of Justice in Distribution: Evidence from the Philippines

Abstract

Can unanimous (or nearly so) agreement be reached by members of a diverse community on a system of justice in distribution used to guide the re-distribution of endowments within this community? Arrow’s impossibility result suggests that this question will receive a negative answer if certain conditions are imposed on the procedures that a community employs in the attempt to make important community decisions. These conditions are reconsidered by allowing for various types of under-insured risk and uncertainty that face members of communities—especially communities in developing countries. Next, two types of uncertainty are allowed for in the designs of various game-theoretic experiments to determine if groups of individuals can come to agree unanimously on a single system of justice in distribution. (These experiments are based on those devised by Frohlich, Oppenheimer and Eavey, and Frohlich and Oppenheimer.) The participants in these experiments are asked to choose one out of four reasonable alternative systems of justice. As for the participants, they are drawn from various sections of Filipino society. The experimental results obtained indicate that groups of individuals, who possess certain characteristics (but nevertheless who hold diverse views on matters of social importance) do agree unanimously to choose a single system of justice in distribution under relevant conditions of uncertainty. The system of justice agreed upon is a version of Rawls’s first principle of justice and the priority rule—a principle that is related to Popper’s principle of minimising avoidable suffering. The participants also explain their decisions in language that reflects that they have an understanding of the uncertain economic and social realities that members of their community face. In the light of these experimental results it is indicated what alternative conditions should be used, in place of some of the Arrovian conditions, in order to generate possibility results. It is also emphasised that this possibility result need not be satisfied in all communities. This is so since the moral intuition of members of a community probably will be influence by the cultural and other relevant circumstances to be found in this community, yet these cultural and other relevant circumstances vary across communities.

The Role of Information in Choices Over Income Distributions

“…Hypothesis 1 and hypothesis 2 are also in line with the evidence from the earlier experiments (e.g., Frohlich, Oppenheimer, and Eavey 1987a). Hypothesis 3, in turn, contradicts the findings of the previous experiments, which show clear evidence on behalf of the floor constraint distribution as the most popular choice behind the veil of ignorance (e.g., Frohlich, Oppenheimer, and Eavey 1987a;Frohlich and Oppenheimer 1990;Lissowski, Tyszka, and Okrasa 1991;de la Cruz-Doña and Martina 2000;Oleson 2001). An alternative hypothesis, hypothesis 3′, is therefore formulated on the basis of the earlier experimental results.…”

“…Of the participants, 82% knew no one in her or his group beforehand, and only 8% had good friends in her or his group. constraint distribution seems to be independent of the nationality of the participants, the specific framing of the choice task, specific stakes, and the presence or absence of production in the experiment (Frohlich, Oppenheimer, and Eavey 1987a;Frohlich and Oppenheimer 1990;Lissowski, Tyszka, and Okrasa 1991;Jackson and Hill 1995;de la Cruz-Doña and Martina 2000). Moreover, de la Cruz-Doña and Martina (2000) observed 100% support for the floor constraint principle with both student and government employee participants.…”

“…18. In the earlier experiments, the floor constraint distribution was the most popular choice, both in the no-production and production conditions (Frohlich and Oppenheimer 1990;de la Cruz-Doña and Martina 2000).…”

“…This study focuses on Rawls's (1971) descriptive hypothesis, which states that people in a hypothetical original position will choose certain principles of justice. This hypothesis has been tested in laboratory experiments Eavey 1987a, 1987b;Frohlich and Oppenheimer 1990;Lissowski, Tyszka, and Okrasa 1991;Bond and Park 1991;Frohlich and Oppenheimer 1992;Jackson and Hill 1995;de la Cruz-Doña and Martina 2000;Oleson 2001).…”

“…This distribution is called the utilitarian distribution here. Rawls's (1971) argument has been tested against Harsanyi's (1976) argument in laboratory experiments Eavey 1987a, 1987b;Frohlich and Oppenheimer 1990;Lissowski, Tyszka, and Okrasa 1991;Bond and Park 1991;Frohlich and Oppenheimer 1992;Jackson and Hill 1995;de la Cruz-Doña and Martina 2000;Oleson 2001). The main results of these experiments can be summarized as follows: (1) experimental groups are able to reach unanimous agreements, (2) most of the groups choose the same principle of income distribution, and (3) the most often chosen principle is not the Rawlsian principle of distributive justice.…”