9AM GMT (6AM Rio de Janeiro, 9AM London, 10AM Paris, 12PM Istanbul, 2:30PM New Delhi, 5PM Shanghai, 6PM Tokyo/Seoul, 8PM Sydney, 10PM Auckland)
Huaxia Zeng (School of Economics, SUFE) "Equity in Strategic Exchange"
Host: Marcus Pivato
Abstract. New fairness notions aligned with the merit principle are proposed for designing exchange rules. We show that for an obviously strategy-proof, efficient and individually rational rule, (i) an agent receives her favorite object when others unanimously perceive her object the best, if and only if preferences are single-peaked, and (ii) an upper bound on fairness attainable is that, if two agents' objects are considered the best by all agents partitioned evenly into two groups, it is guaranteed that one, not both, gets her favorite object. This indicates an unambiguous trade-off between incentives and fairness in the design of exchange rules.
(Joint work with Peng Liu)
2PM GMT (9AM Montréal/Toronto, 11AM Rio de Janeiro, 2PM London, 3PM Graz, 5PM Istanbul, 7:30PM New Delhi, 11PM Tokyo/Seoul)
Steven Kivinen (University of Graz) "Robust Median Voter Rules"
Host: Marcus Pivato
Abstract. Generalized median voter (GMV) rules on the single-peaked preference domain are group strategy-proof. We show that if incomplete information coexists with the ability to commit to coalitional agreements, then GMV rules can be susceptible to insincere voting by groups with heterogeneous beliefs. We identify strategic compromise as a novel source of insincere voting in this environment. Our two main results characterize the set of fair, efficient, and robust voting rules: those that ensure sincere voting under asymmetric information and coalition formation. Each result uses a different notion of robustness, and both give (at most) two alternatives special treatment, with the remaining alternatives chosen according to a type of consensus.
(Joint work with Norovsambuu Tumennasan)
8PM GMT (12PM Vancouver, 2PM Austin, 3PM Toronto/Montréal, 5PM Rio de Janeiro, 8PM London, 9PM Paris, 11PM Istanbul, 9AM Wednesday in Auckland)
Harvey Lederman (University of Texas, Austin) "Maximal Social Welfare Relations on Infinite Populations Satisfying Permutation Invariance"
Host: Marcus Pivato
Abstract. We study social welfare relations (SWRs) on an infinite population. Our main result is a characterization of the common core shared by prominent utilitarian SWRs over distributions which realize finitely many welfare levels on this population. We characterize them as the largest SWR (with respect to set-inclusion when the weak relation is viewed as a set of pairs) which satisfies Strong Pareto, Permutation Invariance (elsewhere called ``Relative Anonymity'' and ``Isomorphism Invariance''), and a further ``Pointwise Independence'' axiom.
(Joint work with Jeremy Goodman)
2PM GMT (9AM College Park, 11AM Rio de Janeiro, 2PM London, 3PM Paris, 5PM Istanbul, 7:30PM New Delhi, 11PM Tokyo/Seoul)
Eric Pacuit (University of Maryland) "Characterizations of voting rules based on majority margins"
Host: Marcus Pivato
Abstract. In the context of voting with ranked ballots, an important class of voting rules is the class of margin-based rules (also called pairwise rules). A voting rule is margin-based if whenever two elections generate the same head-to-head margins of victory or loss between candidates, then the voting rule yields the same outcome in both elections. Although this is a mathematically natural invariance property to consider, whether it should be regarded as a normative axiom on voting rules is less clear. In this paper, we address this question for voting rules with any kind of output, whether a set of candidates, a ranking, a probability distribution, etc. We prove that a voting rule is margin-based if and only if it satisfies some axioms with clearer normative content. A key axiom is what we call Preferential Equality, stating that if two voters both rank a candidate x immediately above a candidate y, then either voter switching to rank y immediately above x will have the same effect on the election outcome as if the other voter made the switch, so each voter's preference for y over x is treated equally.
This is joint work with Yifeng Ding and Wes Holliday
5PM GMT (9AM Berkeley, 12PM Montréal/Toronto, 2PM Rio de Janeiro, 5PM London, 6PM Paris, 8PM Istanbul, 10:30PM New Delhi)
Snow Zhang (University of California, Berkeley) "Coherent combinations of experts' opinions"
Host: Marcus Pivato
Abstract. One popular rule for aggregating multiple experts' point estimates is linear averaging. However, this rule is incompatible with Bayesian conditionalization except in trivial cases of expert consensus (Dawid et al., 1995; Ranjan & Gneiting, 2010; Bradley, 2018; Gallow, 2018). This paper proves a generalization of this impossibility result. If time permits, we'll also discuss how to extend the analysis to the imprecise setting.
9AM GMT (6AM Rio de Janeiro, 9AM London, 10AM Paris, 12PM Istanbul, 2:30PM New Delhi, 6PM Tokyo/Seoul, 8PM Sydney, 10PM Auckland)
Vassili Vergopoulos (Université Paris II Panthéon-Assas) "Egalitarianism in Preference Aggregation under Uncertainty"
Host: Marcus Pivato
Abstract. This paper puts forward a theory of preference aggregation under uncertainty that incorporates egalitarian concerns both ex ante and ex post and accommodates heterogeneity in individual utilities and beliefs. Existing approaches often appeal to a compromise between an ex ante perspective committing to the Pareto Condition (PC) and an ex post perspective committing to Subjective Expected Utility (SEU). In contrast, we maintain each of PC and SEU in their full force on adequately restricted domains. Key to our approach is Choquet integration and its Fubini-like properties.
(Joint work with Federica Ceron)
5PM GMT (10AM Vancouver, 1PM Toronto/Montréal, 2PM Rio de Janeiro, 6PM London, 7PM Saint Etienne, 8PM Istanbul, 10:30PM New Delhi)
Antoinette Baujard (Université Jean Monnet Saint-Etienne) "How people understand novel voting rules: approval voting, evaluative voting and majority judgment"
Host: Marcus Pivato
Abstract. A primary condition of individuals' empowerment in their selection or use of a voting rule is that they understand it. Two papers analyze people's understanding of three voting rules: approval voting, evaluative voting and majority judgment. We draw on results from two data sets, first a lab experiment on incentivized voting where participants are exogenously assigned single-peaked preferences (first paper) and, second, a survey based on two representative samples of 1000 French voters, organized during the first round of the French presidential election (second paper). We distinguish three components of understanding of voting rules: how to fill in the ballot; how votes are aggregated; and how to vote strategically. We scrutinize each component by observing the respondents' voting behaviors and their answers to comprehension questions on the rules. First, from both data sets, we find that most participants understand how to fill in the ballot with the three novel voting rules. Second, in the lab experiment, participants' understanding of vote aggregation under majority judgment is lower and, crucially, more heterogeneous than with evaluative voting and approval voting. The fact that majority judgment is poorly understood is particularly striking in the representative survey. Third, participants' voting behavior is oddly similar between evaluative voting and majority judgment. Data from the lab experiment confirm the theoretical prediction that under evaluative voting there will be a high incidence of strategic voting through the use of extreme grades, but contradict the prediction that under majority judgment voters will vote less strategically. We also find that with majority judgment, the better voters understand how votes are aggregated, the more they use extreme grades. Fourth, the responses from the representative survey revealed genuine confusion among participants testing majority judgment, who conflated its vote aggregation process with the one used in evaluative voting. These results complement social choice theory by enabling an analysis of the properties of voting rules in practice.
(Joint work with Roberto Brunetti and Isabelle Lebon)
2PM GMT (10AM Toronto/Montréal, 11AM San Luis, 3PM London, 4PM Saint Etienne, 5PM Istanbul, 7:30PM New Delhi, 11PM Tokyo/Seoul)
Agustín Bonifacio (GATE Saint-Étienne and Instituto de Matemática Aplicada San Luis) "Obvious Manipulations by Groups"
Host: Marcus Pivato
Abstract. We introduce the notion of obvious manipulation by groups, extending obvious manipulability from individual to coalitional deviations. While existing work focuses on individual incentives, coordinated deviations are often natural, calling for a notion of obviousness at the group level.
We propose a definition based on simple best- and worst-case reasoning and study its implications for voting rules. In tops-only domains, ruling out obvious group manipulations imposes strong restrictions: it implies efficiency, monotonicity, and an almost-unanimity property.
Our main result shows that any tops-only, monotonic voting rule that is not obviously manipulable by groups must be dictatorial. This is somewhat surprising in light of the positive results for individual non-obvious manipulability, where rich classes of rules can be sustained. By contrast, robustness to obvious group deviations sharply limits the design of voting rules.
We also show that many standard rules satisfying the majority criterion are vulnerable to obvious group manipulations. Overall, our results provide a first step toward a theory of group obviousness and uncover new tensions between collective incentives and classical voting principles.