Publications
Geoff Allen and Matthew Jenkins. (2025). Too Small to Fail? Persistence and protest among small party supporters in Germany and New Zealand. 21st Century Political Science Review , 35(2), 93-108.
Abstract
Over the last two decades, party systems across advanced industrial democracies have experienced a general fragmentation. A consequence of this in many countries is an increasing vote share going to very small parties that fail to win any seats. We know little about these parties, or their supporters, given their relative scarcity. In this paper, we argue that understanding their motivations and behaviors is increasingly important given the continued fragmentation of party systems. We utilize a combination of survey data and
electoral data from New Zealand and Germany to show that supporters of very small parties can be distinguished from supporters of other political parties that win seats. These countries, which utilize mixed-member electoral systems, allow us to both observe small
party supporter behavior, and observe how these voters behave when they are not given the choice to vote for their preferred party in single-member contests. We argue that the differences in observable and self-reported behavior are largely a reflection of protest-motivated voting: because these voters feel a certain sense of dissatisfaction with the current system and its party choices, they are more likely to cast their votes for any protest party rather than make a more strategic choice and vote for a less-preferred systemic alternative. This, we believe, has important implications for our understanding of how voters behave when existing party systems lose legitimacy.
Allen, Geoff and Heather Stoll. "A number most convenient? The representational consequences of legislative size." Electoral Studies 82 (2023): 102594.
Abstract
This paper explores the political representational consequences of the size of democratic legislatures, which it makes the case for viewing as a feature of the electoral system. Focusing on descriptive representation, we argue that larger legislatures facilitate the descriptive representation of traditionally underrepresented groups, and that this effect should be stronger in majoritarian electoral systems. We find support for the hypothesis about the greater impact of legislative (assembly) size in majoritarian systems using a cross-national analysis of women's representation. We then employ a within-case empirical analysis of the descriptive representation of several racial and ethnic minorities and women in (majoritarian) U.S. state legislatures. Support for the hypothesis about the effects of relative legislative size (the ratio of seats to persons) is found for most minorities and women. More mixed but still somewhat supportive findings emerge about absolute legislative size (the raw number of seats).
Allen, Geoff. "The resilience of ethnic minority parties in the face of strategic voting incentives: A research note." Party Politics 23.5 (2017): 560-565.
Abstract
Recent research has indicated that social heterogeneity impacts party system size, even in restrictive settings. This research as yet has not established whether it is minority or majority voters who are behaving outside Duvergerian expectations. This study argues that it is ethnic voters that seem to defect from their parties at lower rates, which explains why small parties proliferate and persist in heterogeneous states. This hypothesis is tested on party-in-district level election returns in the German land Schleswig-Holstein. The results show that small ethnic parties suffer notably less defection than small non-ethnic parties. The study proposes a number of potential causal mechanisms that could be driving ethnic voters, as a group, to defect at lower rates than non-ethnic voters.
Abstract
Over the last two decades, party systems across advanced industrial democracies have experienced a general fragmentation. A consequence of this in many countries is an increasing vote share going to very small parties that fail to win any seats. We know little about these parties, or their supporters, given their relative scarcity. In this paper, we argue that understanding their motivations and behaviors is increasingly important given the continued fragmentation of party systems. We utilize a combination of survey data and
electoral data from New Zealand and Germany to show that supporters of very small parties can be distinguished from supporters of other political parties that win seats. These countries, which utilize mixed-member electoral systems, allow us to both observe small
party supporter behavior, and observe how these voters behave when they are not given the choice to vote for their preferred party in single-member contests. We argue that the differences in observable and self-reported behavior are largely a reflection of protest-motivated voting: because these voters feel a certain sense of dissatisfaction with the current system and its party choices, they are more likely to cast their votes for any protest party rather than make a more strategic choice and vote for a less-preferred systemic alternative. This, we believe, has important implications for our understanding of how voters behave when existing party systems lose legitimacy.
Allen, Geoff and Heather Stoll. "A number most convenient? The representational consequences of legislative size." Electoral Studies 82 (2023): 102594.
Abstract
This paper explores the political representational consequences of the size of democratic legislatures, which it makes the case for viewing as a feature of the electoral system. Focusing on descriptive representation, we argue that larger legislatures facilitate the descriptive representation of traditionally underrepresented groups, and that this effect should be stronger in majoritarian electoral systems. We find support for the hypothesis about the greater impact of legislative (assembly) size in majoritarian systems using a cross-national analysis of women's representation. We then employ a within-case empirical analysis of the descriptive representation of several racial and ethnic minorities and women in (majoritarian) U.S. state legislatures. Support for the hypothesis about the effects of relative legislative size (the ratio of seats to persons) is found for most minorities and women. More mixed but still somewhat supportive findings emerge about absolute legislative size (the raw number of seats).
Allen, Geoff. "The resilience of ethnic minority parties in the face of strategic voting incentives: A research note." Party Politics 23.5 (2017): 560-565.
Abstract
Recent research has indicated that social heterogeneity impacts party system size, even in restrictive settings. This research as yet has not established whether it is minority or majority voters who are behaving outside Duvergerian expectations. This study argues that it is ethnic voters that seem to defect from their parties at lower rates, which explains why small parties proliferate and persist in heterogeneous states. This hypothesis is tested on party-in-district level election returns in the German land Schleswig-Holstein. The results show that small ethnic parties suffer notably less defection than small non-ethnic parties. The study proposes a number of potential causal mechanisms that could be driving ethnic voters, as a group, to defect at lower rates than non-ethnic voters.
| The resilience of ethnic minority parties |
| A number most convenient? |
| Too small to fail? |
Papers in Progress and Working Papers
(Please do not cite these works without first consulting the authors)
Delivering Women's Representation: The comparative effectiveness of political institutions (with Heather Stoll)
Abstract
Scholars of political representation, women’s and minority politics, and comparative political institutions have all argued that a number of different political institutions should facilitate the descriptive representation of minorities and traditionally underrepresented groups such as women. Recent work on women’s representation has narrowed the focus of study onto the impacts of electoral systems and quotas, finding both substantively important to the level of descriptive representation for women. While these institutions are undeniably important, the narrowed focus has come at the expense of two institutions long tied to representational outcomes: legislature size and executive regime type. We argue that researchers cannot entirely ignore the impact of legislature size or executive regime type, and should consider re-evaluating the role of these institutions in women’s representation. We conduct a quantitative, cross-national, and hence broadly generalizable empirical study that compares the relative effects of these various institutional mechanisms that are believed to facilitate a group’s descriptive representation. We do so by studying women’s descriptive representation in recent elections in all minimally democratic countries. We find that quota systems (especially mandatory quotas) have the largest substantive effect by far, dwarfing the effect of the variable that has received the most attention to date: that of the electoral system. We also find evidence that legislative size likely impacts women’s descriptive representation, particularly in states with restrictive electoral systems. Presidentialism, meanwhile, plays a substantively important role in limiting women’s representation.
Abstract
Scholars of political representation, women’s and minority politics, and comparative political institutions have all argued that a number of different political institutions should facilitate the descriptive representation of minorities and traditionally underrepresented groups such as women. Recent work on women’s representation has narrowed the focus of study onto the impacts of electoral systems and quotas, finding both substantively important to the level of descriptive representation for women. While these institutions are undeniably important, the narrowed focus has come at the expense of two institutions long tied to representational outcomes: legislature size and executive regime type. We argue that researchers cannot entirely ignore the impact of legislature size or executive regime type, and should consider re-evaluating the role of these institutions in women’s representation. We conduct a quantitative, cross-national, and hence broadly generalizable empirical study that compares the relative effects of these various institutional mechanisms that are believed to facilitate a group’s descriptive representation. We do so by studying women’s descriptive representation in recent elections in all minimally democratic countries. We find that quota systems (especially mandatory quotas) have the largest substantive effect by far, dwarfing the effect of the variable that has received the most attention to date: that of the electoral system. We also find evidence that legislative size likely impacts women’s descriptive representation, particularly in states with restrictive electoral systems. Presidentialism, meanwhile, plays a substantively important role in limiting women’s representation.
| Delivering Women's Representation |
Representation, Electoral Systems and Stranded Minority Parties
Ethnic minority parties, by virtue demographics, are limited in their electoral pursuits. In the standard policy, office or votes framework, they face structural constraints in pursuing both votes or political office. These structural constraints necessarily focus ethnic minority parties on policy influence. This paper focuses on how electoral institutions shape the ability of ethnic minority parties to influence the policy debates within a country. It compares the effectiveness in policy pursuits of ethnic minority parties representing three similarly sized groups in three different countries: Hungarians in Slovakia, Swedes in Finland and Serbs in Croatia. These groups are comparable not only because of their similar size, but because of their status as what I call stranded minorities, groups with a kin-state that can serve as a credible backdrop for policy demands. I argue that electoral systems that force ethnic minority parties to compete against non-minority parties create a stronger incentive for non-minority parties to consider ethnic minority issues as part of the broader national policy debate. By contrast, legislative reservations seem to create incentives for mainstream parties to effectively sideline ethnic minority policy concerns, even when ethnic minority parties are members of the governing coalition.
Ethnic minority parties, by virtue demographics, are limited in their electoral pursuits. In the standard policy, office or votes framework, they face structural constraints in pursuing both votes or political office. These structural constraints necessarily focus ethnic minority parties on policy influence. This paper focuses on how electoral institutions shape the ability of ethnic minority parties to influence the policy debates within a country. It compares the effectiveness in policy pursuits of ethnic minority parties representing three similarly sized groups in three different countries: Hungarians in Slovakia, Swedes in Finland and Serbs in Croatia. These groups are comparable not only because of their similar size, but because of their status as what I call stranded minorities, groups with a kin-state that can serve as a credible backdrop for policy demands. I argue that electoral systems that force ethnic minority parties to compete against non-minority parties create a stronger incentive for non-minority parties to consider ethnic minority issues as part of the broader national policy debate. By contrast, legislative reservations seem to create incentives for mainstream parties to effectively sideline ethnic minority policy concerns, even when ethnic minority parties are members of the governing coalition.
| Representation, Electoral Systems and Stranded Minority Parties |
Recent and Upcoming Conference Presentations
The Impact of Communally Reserved Legislative Seats: A new theory with evidence from Croatia
(presented at the American Political Science Association Annual Meeting 2016, Philadelphia)
Abstract
States, particularly in Eastern Europe, have increasingly seen reserved legislative seats for national minority groups as an effective strategy for creating descriptive representation at the national level. This institutional arrangement undeniably creates descriptive representation, but questions remain about the substantive policy influence minority communities actually exert in such a system. To date, there is no comprehensive, cross-national theory of the impact of communally reserved legislative seats. Progress in the field has largely focused on case studies, from states as diverse as Bolivia, Niger and Romania. These studies, along with studies conducted at the regional and local level in India, consistently find little to no policy impact for minority communities granted reserved legislative seats. In this paper, I will argue that this outcome can no longer be seen as contextually driven, but instead the consequence of the electoral conditions and policy bargaining conditions generated by this institutional arrangement. I use bargaining theoretic and principal agent arguments to show how this institutional arrangement creates strong pressures for status-quo maintenance in regards to issues of minority interest. To investigate the plausibility of these hypotheses, I conducted a qualitative plausibility probe in the state of Croatia, which makes wide use of communally reserved legislative seats, to understand the impacts of the institution on minority representation. Semi-structured interviews with minority community leaders, elected legislators and government officials point to widespread disagreement on the effectiveness of minority legislators in pushing for policy, but general consensus that the system, as currently structured, has major flaws. Community leaders, particularly those not directly involved with national governance, largely discount the value of their elected representatives, while minority legislators and government officials argue that community leaders are not able to see the difficulties of national level governance. The sum of the evidence from the case study indicates the plausibility of the hypotheses generated at the beginning of the paper. To conclude, I discuss ways in which these hypotheses can be tested robustly in a cross-national study, with particular emphasis on ways of drawing out the complex nature of how to measure policy influence and policy bargaining efficacy given the divergent contextual settings found in any cross-national work.
(presented at the American Political Science Association Annual Meeting 2016, Philadelphia)
Abstract
States, particularly in Eastern Europe, have increasingly seen reserved legislative seats for national minority groups as an effective strategy for creating descriptive representation at the national level. This institutional arrangement undeniably creates descriptive representation, but questions remain about the substantive policy influence minority communities actually exert in such a system. To date, there is no comprehensive, cross-national theory of the impact of communally reserved legislative seats. Progress in the field has largely focused on case studies, from states as diverse as Bolivia, Niger and Romania. These studies, along with studies conducted at the regional and local level in India, consistently find little to no policy impact for minority communities granted reserved legislative seats. In this paper, I will argue that this outcome can no longer be seen as contextually driven, but instead the consequence of the electoral conditions and policy bargaining conditions generated by this institutional arrangement. I use bargaining theoretic and principal agent arguments to show how this institutional arrangement creates strong pressures for status-quo maintenance in regards to issues of minority interest. To investigate the plausibility of these hypotheses, I conducted a qualitative plausibility probe in the state of Croatia, which makes wide use of communally reserved legislative seats, to understand the impacts of the institution on minority representation. Semi-structured interviews with minority community leaders, elected legislators and government officials point to widespread disagreement on the effectiveness of minority legislators in pushing for policy, but general consensus that the system, as currently structured, has major flaws. Community leaders, particularly those not directly involved with national governance, largely discount the value of their elected representatives, while minority legislators and government officials argue that community leaders are not able to see the difficulties of national level governance. The sum of the evidence from the case study indicates the plausibility of the hypotheses generated at the beginning of the paper. To conclude, I discuss ways in which these hypotheses can be tested robustly in a cross-national study, with particular emphasis on ways of drawing out the complex nature of how to measure policy influence and policy bargaining efficacy given the divergent contextual settings found in any cross-national work.
| The Impact of Communally Reserved Legislative Seats.pdf |
Too Small to Win, Too Important to Fail? The Paradox of Small Party Support in Mixed Member Electoral Systems
(with Matthew David Jenkins)
(Presented at the American Political Science Association Annual Meeting 2017, San Francisco)
Strategic voting theories largely predict that restrictive electoral rules will result in the winnowing of political competition through mechanical and psychological effects. Recent research suggests that these theories may not be empirically robust in all situations. Scholars of mixed member electoral arrangements have been at the forefront of this research. In this piece, we demonstrate that under mixed-member electoral arrangements in South Korea and New Zealand, small parties not only survive in the more restrictive single-member electoral districts, but in fact gain electoral support in these settings relative to their performance in the proportional representation tier. In contrast, more competitive parties, those that we label ‘coalition-makers’, either suffer electoral losses between tiers (New Zealand) or gain voters at a lower rate than their smaller competitors as competition diminishes between the tiers (Korea). After presenting these findings, we conclude with a discussion of potential organizational, group and individual level explanations of the observations, as well as thoughts on implications for our general understanding of electoral politics.
(This project is part of a broader project, Compensatory vs Parallel: The Strategic Impacts of System Choice)
(with Matthew David Jenkins)
(Presented at the American Political Science Association Annual Meeting 2017, San Francisco)
Strategic voting theories largely predict that restrictive electoral rules will result in the winnowing of political competition through mechanical and psychological effects. Recent research suggests that these theories may not be empirically robust in all situations. Scholars of mixed member electoral arrangements have been at the forefront of this research. In this piece, we demonstrate that under mixed-member electoral arrangements in South Korea and New Zealand, small parties not only survive in the more restrictive single-member electoral districts, but in fact gain electoral support in these settings relative to their performance in the proportional representation tier. In contrast, more competitive parties, those that we label ‘coalition-makers’, either suffer electoral losses between tiers (New Zealand) or gain voters at a lower rate than their smaller competitors as competition diminishes between the tiers (Korea). After presenting these findings, we conclude with a discussion of potential organizational, group and individual level explanations of the observations, as well as thoughts on implications for our general understanding of electoral politics.
(This project is part of a broader project, Compensatory vs Parallel: The Strategic Impacts of System Choice)
| Too Small to Win, Too Big to Fail? |
Book Project
Presence and Impotence: The perils of guaranteed descriptive representation
| full_draft.pdf |
