Abstract
We consider seat-posted (or designated-seat) committee elections, where disjoint sets of candidates compete for each seat. We assume that each voter has a collection of seat-wise strict rankings of candidates, which are extended to a strict ranking of committees by means of a preference extension. We investigate conditions upon preference extensions for which seat-wise Condorcet candidates, whenever all exist, form the Condorcet winner among committees. We characterize the domain of neutral preference extensions for which the committee of seat-wise winners is the Condorcet winning committee, first assuming the latter exists (Theorem 1) and then relaxing this assumption (Theorem 2). Neutrality means that preference extensions are not sensitive to the names of candidates. Moreover, we show that these two characterizations can be stated regardless of which preference level is considered as a premise.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Change history
12 October 2021
A Correction to this paper has been published: https://doi.org/10.1007/s10058-021-00268-1
Notes
The reader may refer to Lang and Xia (2016) for a review of methods that partially overcome the trade-off between expressivity and costs.
A preference over committees is separable if there exists a set of seat-wise rankings such that for any two committees disagreeing on a single seat, the preferred one is comprising the more preferred candidate for that seat.
This result extends previous ones obtained for multiple referenda. In particular, (Özkal-Sanver and Sanver 2006) show that if preferences over committees are separable, every anonymous seat-based procedure is Pareto inefficient. For the same setting, Cuhadaroglu and Lainé (2012) show that the committee formed by seat-wise plurality winners is Pareto-optimal at any profile of (separable) preferences over committees if and only if these preferences emanate from the Hamming distance criterion.
If a voter’s ideal committee is A, the Hamming distance between A and another committee B is the number of seats with different appointed candidates.
A detailed argument is provided in Sect. 6.
A CCG is exposed to several paradoxes (see Ratliff 2003; Diss and Mahajne 2020). Studies of voting methods selecting a CCG can be found in Barberà and Coelho (2008), Kamwa and Merlin (2015), Diss and Doghmi (2016), Aziz et al. (2017), Kamwa (2017a, 2017b), Diss et al. (2020) and Bubboloni et al. (2020).
In this context, a preference extension maps every ranking of candidates to a ranking of sets of candidates. Kamwa and Merlin (2018) consider the Leximin and the Leximax preference extensions.
A preliminary step towards the answer is provided in Kamwa and Merlin (2018), who prove that if preferences over committees are separable, a CCG coincides with the CCF if the latter exists.
The reader may refer to Ratliff and Saari (2014), who analyze at-large committee selection methods that address diversity constraints such as gender balance.
Another example of a lexicographic preference extension is given by the choice problem defined above Definition 3. In this example, \(\delta ^{i}\) is q-lexicographic with 1 q 2.
Under neutrality, a preference extension generates a unique committee preference up to a reshuffling of candidates’ names.
By Lemma 3 a responsive preference extension induces a separable committee preference. The fact that a profile of separable preference generates a separable majority preference and its consequences are emphasized in particular by Kadane (1972), Koehler (1975), Miller (1975), Hollard and Le Breton (1996) and Vidu (2002).
As this relabelling does not interact with the one made above, it does not imply any loss of generality.
We write the projection of \(C_{k}\) to \(\{k^{*},\ldots ,k\}\) as \( C_{k}\mid _{\{k^{*},\dots ,k\}}\)instead of \(C_{k\{k^{*},\dots ,k\}}\) in order to avoid a too heavy notation.
Intuitively, \(q_{\delta }\) defines a priority ordering over seats. Observe that the definition of \(\delta |_{\Omega }\) induces the inverse of this priority order. This expresses the fact that if seat s is given lower priority than seat \(s^{\prime }\), and when starting from the best rank vector \(\overrightarrow{1}\), losing one rank in s should be less harmful than losing one rank in \(s^{\prime }\).
Observe that one cannot have \(s^{*}=S\), since \(\delta \) is responsive.
Note that if \(s^{*}=1\), the designated columns for \(1\le s<s^{*}-1\) in preferences are redundant. And also straightforward changes are required in the notation, e.g., \((\overset{s^{*}-1}{\overbrace{1,...,1}},\overset{ i}{\overbrace{2,...,2}},1,\overset{S-s^{*}-i}{\overbrace{2,...,2}})\) becomes \((\overset{i}{\overbrace{2,...,2}},1,\overset{S-s^{*}-i}{ \overbrace{2,...,2}})\).
Observe that one may have \(\forall {\mathbf {P}}\in {\mathcal {L}}_{{\mathcal {C}}}\), \(\varepsilon ({\mathbf {P}})\in {\mathbb {L}}\). For instance, denote by \(A^{*}( {\mathbf {P}})\) the first-best committee according to \({\mathbf {P}}\), and define \( P_{s}^{i}\) by \(\forall s\in {\mathcal {S}}, \forall a_{s},b_{s}\in {\mathcal {C}} _{s}\), \(a_{s} \varepsilon _{s}({\mathbf {P}}) b_{s} \Leftrightarrow [a_{s},A^{*}({\mathbf {P}})_{-s}] {\mathbf {P}}^{i} [b_{s},A^{*}({\mathbf {P}})_{-s}]\). With words, for any seat s, we elicit voter i ’s preferences for two different candidates \(a_{s}\) and \(b_{s}\) by considering all committees which overlap the first-best committee for all but seat s. If \({\mathbf {P}}\) ranks the committee comprising \(a_{s}\) above the one comprising \(b_{s}\), we say that i prefers \(a_{s}\) to \(b_{s}\). This elicitation method generalizes the one above by allowing to pick different reference committees for different seats. However, for the elicitation maps characterized in our results, some committee preferences are mapped to \( {\triangle }\).
The reader may refer to Gehrlein and Lepelley (2017) for an overview of the main probabilistic approaches.
References
Aslan F, Dindar H, Lainé J (2019) Choosing a committee under majority voting. In: Morais D, Carreras A, de Almeida A, Vetschera R (eds) Group decision and negotiation: behavior, models, and support. GDN 2019. Lecture notes in business information processing, vol 351. Springer, Berlin
Aziz H, Elkind E, Faliszewski, P, Lackner M, Skowron, P (2017) The Condorcet principle for multiwinner election: from short listing to proportionality. In: Proceedings of the twenty-sixth international joint conference on artificial intelligence (IJCAI-17), pp 84–90
Barberà S, Coelho D (2008) How to choose a non-controversial list with k names. Soc Choice Welf 31:79–96
Benoît J-P, Kornhauser LA (1991) Voting simply in the election of assemblies. Starr Center for Applied Economics. Working paper #91-32, New York University, CV
Benoît J-P, Kornhauser LA (2010) Only a dictatorship is efficient. Games Econom Behav 70:261–270
Bernholz P (1973) Logrolling, arrow-paradox and cyclical majorities. Public Choice 15:87–95
Bezembinder T, van Acker P (1985) The ostrogorski paradox and its relation to nontransitive choice. J Math Sociol 11:131–58
Black D (1948) On the rationale of group decision-making. J Polit Econ 56:23–34
Bubboloni D, Diss M, Gori M (2020) Extensions of the Simpson voting rule to the committee selection setting. Public Choice 183:151–185
Cuhadaroglu T, Lainé J (2012) Pareto efficiency in multiple referendum. Theor Decis 72:525–536
Daudt H, Rae D (1976) The Ostrogorski paradox: a peculiarity of compound majority decision. Eur J Polit Res 4:391–398
Deb R, Kelsey D (1987) On constructing a generalized Ostrogorski paradox: necessary and sufficient conditions. Math Soc Sci 14:161–174
Diss M, Doghmi A (2016) Multi-winner scoring election methods: Condorcet consistency and paradoxes. Public Choice 169:97–116
Diss M, Mahajne M (2020) Social acceptability of Condorcet committees. Math Soc Sci 105:14–27
Diss M, Kamwa E, Tlidi A (2020) On some k-scoring rules for committee elections: agreement and Condorcet principle. Revue d’Economie Politique 130(5):699–725
Elkind E, Faliszewski P, Skowron P, Slinko A (2017) Properties of multiwinner scoring rules. Soc Choice Welf 48:599–632
Fishburn PC (1981) An analysis of simple voting systems for electing committees. SIAM J Appl Math 41:499–502
Gehrlein WV (1985) The Condorcet criterion and committee selection. Math Soc Sci 10:199–209
Gehrlein WV, Lepelley D (2017) Elections, voting rules and paradoxical outcomes. Studies in social choice and welfare. Springer, Berlin
Hodge JK, Schwallier P (2006) How does separability affect the desirability of referendum election outcomes? Theor Decis 61:251–276
Hollard G, Le Breton M (1996) Logrolling and a McGarvey theorem for separable tournaments. Soc Choice Welf 13:451–455
Kadane JB (1972) On division of the question. Public Choice 13:47–54
Kamwa E (2017) On stable rules for selecting committees. J Math Econ 70:36–44
Kamwa E (2017) Stable rules for electing committees and divergence on outcomes. Group Decis Negot 26(3):547–564
Kamwa E, Merlin V (2015) Scoring rules over subsets of alternatives: consistency and paradoxes. J Math Econ 61:130–138
Kamwa E, Merlin V (2018) Coincidence of Condorcet committees. Soc Choice Welf 50:171–189
Kaymak B, Sanver MR (2003) Sets of alternatives as Condorcet winners. Soc Choice Welf 20:477–494
Koehler DH (1975) Vote trading and the voting paradox: a proof of logical equivalence. Am Polit Sci Rev 69:954–960
Laffond G, Lainé J (2006) Single-switch preferences and the Ostrogorski paradox. Math Soc Sci 52:49–66
Laffond G, Lainé J (2009) Condorcet choice and the Ostrogorski paradox. Soc Choice Welf 32:317–333
Laffond G, Lainé J (2012) Searching for a compromise in multiple referendum. Group Decis Negot 21:551–569
Lang J, Xia L (2016) Voting on combinatorial domains. In: Brandt F, Conitzer V, Endriss U, Lang J, Procaccia AD (eds) Handbook of computational social choice. Cambridge University Press, Cambridge, pp 197–221
Miller NR (1975) Logrolling and the arrow paradox: a note. Public Choice 21:107–110
Moulin H (1988) Axioms for cooperative decision-making. Cambridge University Press, Cambridge
Özkal-Sanver I, Sanver MR (2006) Ensuring Pareto optimality by referendum voting. Soc Choice Welf 27:211–219
Ratliff TC (2003) Some startling inconsistencies when electing committees. Soc Choice Welf 21:433–454
Ratliff TC (2006) Selecting committees. Public Choice 126:343–355
Ratliff TC, Saari DG (2014) Complexities of electing diverse committees. Soc Choice Welf 43:55–71
Vidu L (1999) An extension of a theorem on the aggregation of separable preferences. Soc Choice Welf 16:159–167
Vidu L (2002) Majority cycles in a multi-dimensional setting. Econ Theor 20:373–386
Acknowledgements
Funding was provided by Istanbul Bilgi University (Grant No. BILGI Research Development Innovation Programme, POlarization viewed from SOcial choice Perspective (POSOP)) National Research Development and Innovation Office (Grant No. TKP2020, National Challenges Program, BME NC TKP2020).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Authors are grateful to two reviewers for their valuable comments and suggestions. This research has been partially funded by the BILGI Research Development Innovation Programme, POlarization viewed from SOcial choice Perspective (POSOP), and the “TKP2020, National Challenges Program” of the National Research Development and Innovation Office (BME NC TKP2020).
Rights and permissions
About this article
Cite this article
Aslan, F., Dindar, H. & Lainé, J. When are committees of Condorcet winners Condorcet winning committees?. Rev Econ Design 26, 417–446 (2022). https://doi.org/10.1007/s10058-021-00260-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10058-021-00260-9