From More than Red and Blue: Political Parties and American Democracy

Abstract

The old case for a two-party system did not: (a) fully grapple with ‘presidential’ democracy, (b) foresee the dangers of polarization, or (c) give sufficient weight to demands for representation. We therefore sketch a vision of multiparty presidential democracy, introducing relevant literature along the way. This vision rests on reforms to make multiple parties viable, push that system toward pre-election coalition, and improve descriptive representation regardless of how many parties there are. Key features are proportional representation (PR) for U.S. House, then reforms of singleseat offices (like President and Senator) to let multiple parties compete as parties. We give some ‘pros and cons’ of three PR forms: mixed-member proportional, single transferable vote, and open-list proportional. We also explain why PR might not lead to the sort of fragmentation that some critics and proponents alike expect.

A Different Kind of Party Government

A Different Kind of Party Government

The notion of a system of responsible political parties as articulated in the 1950 APSA report was one in which the parties would set out a “choice between the alternatives of action.”¹APSA Committee on Political Parties. 1950. Towards a More Responsible Two-Party System. American Political Science Association. The report specifically, in its very title, understood that to mean two possible courses of action — a responsible two- party system. The core notion of a “responsible party” is one that not only sets out a course of action, but is able to follow through on it if elected to power.

Our contribution to this discussion nearly three quarters of a century later will not be to elaborate on whether such a model is or ever was desirable. Independent of its desirability, a responsible party is almost impossible under a presidential form of government. The separation of executive and legislative electoral processes and the separate survival of the elected branches (via fixed terms) provide too many countervailing incentives for party-policy responsibility to be feasible. Parties under such conditions are much more likely to be “presidentialized” than responsible. What this means is that presidential candidates set out their own priorities to win their separate contests, while legislators cater to localities or interest groups that may have different priorities from those of the presidential candidate.²Samuels, David J., and Matthew S. Shugart. 2010. Presidents, Parties, and Prime Ministers: How the Separation of Powers Affects Party Organization and Behavior. Cambridge University Press. Once in office, the absence of the parliamentary confidence mechanism means there’s little to hold the executive to a collectively agreed policy platform.³Azari, Julia. 2017. Review of Relic by William G. Howell and Terry M. Moe. The Journal of Politics 79 (4): e79-e80.

Thus we do not attempt to reiterate the case for responsible parties, understood as collective teams offering competing governing options. Rather, we propose a different kind of “responsible” party government, consistent with the themes of this wider task force, that could institutionalize a process of reflecting the diversity of both opinion and socio-demographic constituencies currently either subsumed within or left outside of the two-party system.

As a starting point, we posit that it would be inaccurate to claim that the U.S. party system finally arrived at the “responsible two-party system” in the form of today’s polarized parties.⁴Mounk, Yascha. 2018. “The Rise of McPolitics.” The New Yorker. June 25. https://www.newyorker.com/magazine/2018/07/02/the-rise-of-mcpolitics. What we have in the American system today is not a variant of responsible parties. If anything, parties have become irresponsible. Their nomination and policy-setting processes allow highly organized groups to pull each party away from the median voter.⁵We mean “median voter” in the multidimensional sense (Huber and Powell 1994, 293; McGann and Latner 2012, 831). In the U.S., this person might be ‘operationally’ liberal and ‘symbolically’ conservative (Cayton and Dawkins 2020; Ellis and Stimson 2012; Grossmann and Hopkins 2015). Yet the polarization and unwillingness to compromise seen most especially in the GOP⁶Hacker, Jacob S., and Paul Pierson. 2015. “Confronting Asymmetric Polarization.” Chap. 3 in Solutions to Political Polarization in America, ed. Nathaniel Persily. Cambridge University Press. is a far cry from the model of responsible parties setting out competing programs of government.

It is not that a wide range of opinions about policy and ideological options are not already represented in the U.S. two parties. However, most voters have little opportunity to cast an effective vote to express their preferred paths, due to winner- take-all contests, including at the primary nomination stage.

We sketch (and introduce relevant literature on) a different way of representing the diversity of ideological and policy preferences of American voters. This alternative relies on a form of proportional representation (PR) for the U.S. House, combined with other reforms for bodies where PR is not practical due to the election of single offices, such as the presidency and Senate (absent major constitutional amendment).

What we have in the American system today is not a variant of responsible parties. If anything, parties have become irresponsible.

Contra some critiques of PR (see below), we see it as a potential contribution to responsible multiparty politics. We mean “responsible” in a similar way as the 1950 report: voters signaling policy direction via their party choices. However, the “responsibility” for implementing policy would rest with coalitions of parties. Those coalitions could emerge either efore or after elections. Future elections would offer opportunities for voters to shift to different parties if they were unhappy with the records of their previous parties. Crucially, that would not mean shifting to the single party on the other side of the political divide.

It is likely that the presence of the Senate and the presidency would encourage parties to form electoral alliances (pre-election coalitions). Therefore, in most elections, there would continue to be two major such alliances. With PR for the House (and other rules for other bodies that we shall address briefly later) some parties would agree to cooperate in elections with other parties with whom they share basic affinity. For instance, “progressive” and “center left” could be distinct parties within a broad left alliance, and social conservatives and economic conservatives could be distinct parties but would cooperate in a broad right. Or, when circumstances called for something different (like a ‘pro-democracy alliance’), the institutions we describe might help bolster that. Such a model can offer voters more voice in the setting of policy direction, without sacrificing the building of electoral majorities. In fact, alliances of this sort are common in existing PR systems⁷Carroll, Royce, and Gary W. Cox. 2007. “The Logic of Gamson’s Law: Pre-Election Coalitions and Portfolio Allocations.”American Journal of Political Science 51 (2): 300-313.⁸Ganghof, Steffen. 2015. “Four Visions of Democracy: Powell’s Elections as Instruments of Democracy and beyond.”Political Studies Review 13 (1): 69-79.⁹Golder, Sona N. 2006. The Logic of Pre-Electoral Coalition Formation. Columbus, OH: Ohio State University Press., and PR generally is more favorable to ensuring legislative majorities that reflect a majority of the electorate.¹⁰Lijphart, Arend. 1997. “Back to Democratic Basics: Who Really Practices Majority Rule?” Chap. 8 in Democracy’s Victory and Crisis, ed. Axel Hadenius. Oxford: Oxford University Press.¹¹McDonald, Michael D., and Ian Budge. 2005. Elections, Parties, Democracy: Conferring the Median Mandate. Oxford: Oxford University Press.¹²Powell, G. Bingham. 2000. Elections as Instruments of Democracy Majoritarian and Proportional Visions. Princeton: Princeton University Press.

In a sense, this notion of alliance-facilitating PR could harness the best of the current U.S. party system, yet marginalize that system’s worst features. The current system sees progressive and center-left actors in the Democratic Party, while the Republican Party hosts both economic conservatives and authoritarians. Consider the observation of Henry Droop¹³Droop, Henry R. 2012 [1869]. “On the Political and Social Effects of Different Methods of Electing Representatives,” in Electoral Systems, Vol. III, ed. By David M. Farrell and Matthew S. Shugart. London: Sage. on two-party politics and “majority voting” (i.e., winner-take-all):

As every representative is elected to represent one of these two parties, the nation, as represented in the assembly, appears to consist only of these two parties, each bent on carrying out its own programme. But, in fact, a large proportion of the electors who vote for the candidates of the one party or the other really care much more about the country being honestly and wisely governed than about the particular points at issue between the two parties; and if this moderate non-partisan section of the electors had their separate representatives in the assembly, they would be able to mediate between the opposing parties and prevent the one party from pushing their advantage too far, and the other from prolonging a factious opposition. With majority voting they can only intervene at general elections, and even then cannot punish one party for excessive partisanship, without giving a lease of uncontrolled power to their rivals.

We do not have to believe a strict interpretation of Droop’s words — that most voters are “moderate” — to understand the value to current conditions of the prescription he offered more than a century and a half ago: a proportional system. The problem of one party “pushing their advantage too far” is an even greater problem in an era of two-party polarization, as is the inability of voters who are less aligned with the mainstream of their preferred party to rein it in other than by voting for the opposing party that they likely find unacceptable. The larger point is that the country consists of more options than any given voter has placed before her by the candidates of the two dominant parties.

A PR system might expand the menu, allowing different parties to reflect different “alternatives of action.” It also, as we shall discuss, allows for a given party to have multiple candidates, whose personal attributes reflect different socio-demographic groups, placed before the voters. At present, the only opportunities voters have to select from among policy options and candidate attributes within these broad left and right camps is at presidential primaries—and even then, often only for voters who happen to be in early states on the primary calendar when a plurality of voices are still competing for delegates.¹⁴And it is well worth remembering that this competition for delegates takes place within a form of proportional representation in the Democratic Party, and also does so in several states (especially those early in the calendar) in the Republican Party (Jones, McCune, and Wilson 2020). A PR system for the House would permit this sort of competition to take place in forging the main majority-seeking caucus options inside the House of Representatives instead of just at the quadrennial party conventions (and only for some primary voters). It therefore gives voters a chance to weigh in at general elections for Representatives and not only at the candidate-selection stage. It also does not present them with the stark choice at present, which is either to swallow their disagreements with the dominant tendencies in their preferred parties or cross party lines (an untenable option for many voters, at least at present).

PR and Fragmentation

PR and Fragmentation

What many readers think of when PR comes up is party fragmentation, difficulty building governing majorities, and amplification of fringe voices.¹⁵Hermens, Ferdinand A. 1941. Democracy or Anarchy? A Study of Proportional Representation. The Review of Politics, University of Notre Dame: Notre Dame, IN. However, these concerns are mostly caricatures of real-world PR systems. To the extent that these concerns contain grains of truth, they are largely irrelevant to the U.S. context. Or they are mostly problems of elite will to compromise.¹⁶Lijphart, Arend. 2013. “Steps in My Research and Thinking about Power Sharing and Democratic Institutions.” Taiwan Journal of Democracy (special issue): 1-7.¹⁷Rosenfeld, Sam. 2020. “Can More Political Parties Fix American Democracy?” Boston Review, April 20. https://www.bostonreview.net/articles/sam-rosenfeld-it-takes-three-or-more.¹⁸Santucci, Jack. 2020. “Multiparty America?” The Journal of Politics 82 (4): e34-39.¹⁹Ziblatt, Daniel. 2017. Conservative Parties and the Birth of Democracy. Cambridge University Press.

From a strictly technical perspective, there are at least two reasons why any realistic version of PR for From a strictly technical perspective, there are at least two reasons why any realistic version of PR for the U.S. House would be unlikely to foster ‘excessive’ fragmentation. First, the sort of PR system that could be reasonably adopted in the U.S. is a moderate version. With 435 House seats — or even a plausible increase to say 600 —and 50 states, an extremely proportional system is not in the cards.²⁰See Hermens (1936, 412-3) for an example of “extreme” proportionality. The interwar German electoral law set up several ‘layers’ of nested districts, so that very few votes would be ‘wasted.’

We assume the multi-seat districts required for any PR system could not span state lines.²¹That is, nationwide proportionality is out of the question, including systems of “compensation” where there might be local districts, but proportionality would be determined by pooling votes across districts (Elklit and Roberts 1996) and thus across all states or groups of states. We assume such designs are non-starters because the Constitution states that House seats are apportioned to states, which implies they can’t be effectively shared between states. That is, a state could serve as a multi-seat district (electing members “at large” by a proportional method). Quite likely, the larger states would be subdivided into multiple districts, in order to avoid excessively large numbers elected per district. Most advocates of PR in the U.S. indicate a preference for a range of district magnitude (the number of seats per district) of no more than 5 or 7.²²For states with as few as three Representatives, PR is still feasible. For states with two, PR means that most of the time each of the top two parties would earn one seat from the state. For states with just one, PR is impossible. This need not doom a PR system as a whole; some existing PR systems have a few districts that elect just one member. Solutions such as expanding the House can minimize the number of such states, or a minimum of three per state could be set (tolerating some degree of malapportionment of states in exchange for proportional representation of voters). These are complicated questions that we shall not attempt to resolve here, but which need not detain us from evaluating the potential effects of the sort of moderate PR system sketched here. Such a PR design prevents the extreme fragmentation associated with PR in countries such as the Netherlands (with its single nationwide district of 150 seats) or Israel (with its single district of 120). Moreover, the seat product — the assembly size times mean district magnitude, which is a strong predictor of the shape of a party system ²³Shugart, Matthew S. and Rein Taagepera. 2017. Votes from Seats: Logical Models of Electoral Systems. Cambridge: Cambridge University Press.²⁴Shugart, Matthew S. and Rein Taagepera. 2020. “Predicting Party Systems from Electoral Systems.” In Oxford Research Encyclopedia of Politics. Oxford University Press.²⁵Taagepera, Rein. 2007. Predicting Party Sizes: The Logic of Simple Electoral Systems. Oxford: Oxford University Press. — would be modest.²⁶The current seat product of the US House is 435 (the number of seats in the House times district magnitude which is 1). If a PR system had a mean magnitude of 4, for example, the seat product would be 1740— a bit smaller than the modest PR systems of Spain or Sweden and a far cry from the values of really extreme PR systems like Israel (14,400) or the Netherlands (22,500).

The second reason, aside from a relatively low district magnitude and modest seat product, why a U.S. PR system might not be fragmenting is the presence of the Senate and presidency, for which plurality and majority systems remain the only realistic options (again absent major reform for the Senate²⁷One “major reform” might give each state three Senators, elected at the same time.). Thus parties competing with distinct party labels in proportional contests (for U.S. House seats) could have incentives to present joint (or “fusion”) candidacies for these other offices. In turn, the incentive to present joint candidacies might hold down fragmenting tendencies in the proportional contests.²⁸Again, the incentive to coalesce depends on pre-existing will to compromise; as noted earlier, such pre-election coalitions are common in many existing PR systems.

Descriptive Representation and Institutional Design

Descriptive Representation and Institutional Design

The political representation of racial, ethnic and linguistic minorities is a perennial issue in American politics. Resolutions have taken several forms over the years: disenfranchisement²⁹Valelly, Richard M. 2016. “How Suffrage Politics Made—and Makes—America.” Chap.. 22 in The Oxford Handbook of American Political Development, eds. Richard Valelly, Suzanne Mettler, and Robert Lieberman. Oxford University Press., incorporation on dominant-group terms³⁰Shefter, Martin. 1986. “Political Incorporation and the Extrusion of the Left: Party Politics and Social Forces in New York City.” Studies in American Political Development 1: 50-90., the post-Voting Rights Act³¹Davidson, Chandler, and Bernard Grofman. 1994. Quiet Revolution in the South: The Impact of the Voting Rights Act, 1965-1990. Princeton University Press. settlement on single-seat districts (SSD), and recent attempts to use that law to reduce descriptive representation.³²Latner, Michael, Charles Smith, Alex Keena, and Anthony McGann. 2021. “Electoral Engineering and the Freedom to Vote.”Scientific American, October 21. https://www.scientificamerican.com/article/electoral-engineering-and-the-freedom-to-vote.

Single-seat districts (SSDs) have proven capable of representing some groups more-or-less in proportion to their numbers.³³Collingwood, Loren, and Sean Long. “Can States Promote Minority Representation? Assessing the Effects of the California Voting Rights Act.” Urban Affairs Review, December 31, 2019, 1078087419896854. doi: 10.1177/1078087419896854.³⁴Davidson, Chandler, and Bernard Grofman. 1994. Quiet Revolution in the South: The Impact of the Voting Rights Act, 1965-1990. Princeton University Press.³⁵Lublin, David, Thomas L. Brunell, Bernard Grofman, and Lisa Handley. 2009. “Has the Voting Rights Act Outlived Its Usefulness? In a Word, ‘No.’” Legislative Studies Quarterly 34 (4): 525-53. Substantively, representation via SSDs has reduced economic inequality, at least in jurisdictions that the VRA used to cover.³⁶Aneja, Abhay P., and Carlos F. Avenancio-León. 2019. “Disenfranchisement and Economic Inequality: Downstream Effects of Shelby County v. Holder.” AEA Papers and Proceedings 109: 161-65. doi.org/10.1257/pandp.20191085.³⁷Cascio, Elizabeth U., and Ebonya Washington. 2013. “Valuing the Vote: The Redistribution of Voting Rights and State Funds Following the Voting Rights Act of 1965.” The Quarterly Journal of Economics 129 (1): 379-433.³⁸Eubank, Nicholas, and Adriane Fresh. 2022. “Enfranchisement and Incarceration after the 1965 Voting Rights Act.”American Political Science Review 116 (3):791-806.³⁹The Supreme Court invalidated the VRA ‘coverage formula’ (Section 4) in 2014.

Yet limits to the SSD remedy are well known: it works where groups are geographically concentrated and where there are relatively few ‘communities of interest’ to represent.⁴⁰Abott, Carolyn and Ayse Magazinnik. 2020. “At-Large Elections and Minority Representation in Local Government.”American Journal of Political Science 64 (3): 717-733.⁴¹Gimpel, James G., and Laurel Harbridge-Yong. 2020. “Conflicting Goals of Redistricting: Do Districts That Maximize Competition Reckon with Communities of Interest?” Election Law Journal: Rules, Politics, and Policy 19 (4): 451-71.⁴²Leib, Jonathan I. 1998. “Communities of Interest and Minority Districting after Miller v. Johnson.”Political Geography 17 (6): 683-99. Protecting more geographically dispersed or internally heterogeneous groups (e.g., Latino and Asian voters) has proven more difficult under SSDs.⁴³Kogan, Vladimir, and Eric McGhee. 2012. “Redistricting California: An Evaluation of the Citizens Commission Final Plans.” California Journal of Politics and Policy 4 (1). Moreover, ascribing protected classes to a single constituent interest can foster “tokenism” and weaken coalition policymaking capacity.⁴⁴Guinier, Lani. 1992. “The Representation of Minority Interests: The Question of Single-Member Districts.” Cardozo Law Review 14 (5): 1135-1174.⁴⁵Lublin, David, and D. Stephen Voss. 2000. “Racial Redistricting and Realignment in Southern State Legislatures.” American Journal of Political Science 44 (4): 792-810.

Partisanship also increasingly matters. The electoral “capture” of voters of color by the Democratic Party makes it possible to take majority-minority districts for granted and focus resources onto “swing” voters.⁴⁶Frymer, Paul. 2010. Uneasy Alliances: Race and Party Competition in America. Princeton University Press. It also gives Republicans in state legislatures an incentive to undermine the VRA by packing targeted groups into uncompetitive Democratic districts.⁴⁷Keena, Alex, Michael Latner, Anthony J. McGann McGann, and Charles Anthony Smith. 2021. Gerrymandering the States: Partisanship, Race, and the Transformation of American Federalism. Cambridge University Press.⁴⁸Levitt, Justin. 2013. “Section 5 as Simulacrum.” Yale Law Journal 123. Meanwhile, the U.S. Supreme Court is increasingly calling into question any use of race-conscious districting, referring to the case law as “notoriously unclear and confusing.”⁴⁹Merrill v Milligan, 95 U. S. Nos. 21A375 and 21A376 (2022). If the Court further insulates state legislatures from federal voting rights protections, alternatives to the SSD regime will likely be in higher demand.

Proportional representation is one possible response to these challenges. The argument that PR improves minority representation typically rests on the ability of racial and ethnic parties to win seats through lower ‘thresholds of exclusion.’⁵⁰Lijphart, Arend. 2004. “Constitutional Design for Divided Societies.” Journal of Democracy 15 (2): 96-109.⁵¹Norris, Pippa. 2004. Electoral Engineering: Voting Rules and Political Behavior. Cambridge University Press.⁵²Reynolds, Andrew. 1995. “Debate: PR and Southern Africa: The Case for Proportionality.”Journal of Democracy 6 (4), 117-124. Yet it also rests on evidence that larger parties run more inclusive slates under PR.⁵³Latner, Michael and Anthony McGann. 2005. “Geographical representation under proportional representation: The cases of Israel and the Netherlands.” Electoral Studies 24 (4): 709-734.⁵⁴Shugart, Matthew S. and Rein Taagepera. 2017. Votes from Seats: Logical Models of Electoral Systems. Cambridge University Press. PR permits dispersed groups to aggregate their votes over larger geographic areas. It also might permit groups less attached to established parties to form new ones if they so desired.⁵⁵See Dyck and Johnson (2022) on how party identification varies over time for some groups in the U.S.

Yet PR is not a panacea. Even in proportional systems, candidate-centered ballots permit voters to withhold support from minority candidates.⁵⁶Negri, Margherita. 2017. “Preferential Votes and Minority Representation in Open List Proportional Representation Systems.” Social Choice and Welfare 50 (2): 281-303.⁵⁷Protsyk, Oleh, and Konstantin Sachariew. 2012. “Recruitment and Representation of Ethnic Minorities under Proportional Representation: Evidence from Bulgaria.” East European Politics and Societies 26 (2): 313-39.⁵⁸Sipinen, J., and Peter Söderlund. 2022. “Explaining the Migrant-Native Vote Gap Under Open-List Proportional Representation.” Party Politics. doi.org/10.1177/13540688221113665 Some forms (like MMP below) may not be viable without constitutional amendment. Others require extensive voter education and elite coordination.⁵⁹Pildes, Richard H. and Kristen A. Donoghue. 1995. “Cumulative Voting in the United States.”University of Chicago Legal Forum 1: 241-313. More generally, those who have fought for generations to secure representation under current rules have good reason to look skeptically at changing those rules.

What type of PR?

What type of PR?

This section gives some ‘pros and cons’ of three common PR forms. One of them, mixed-member proportional (MMP), combines two kinds of seats: one ‘tier’ elected in single-seat districts, then a second tier from party lists from which seats are allocated so that parties’ final seat shares are proportional to their vote shares.⁶⁰For details on MMP, see Shugart and Wattenberg (2001) or the explainer offered by the New Zealand Electoral Commission: https://elections.nz/democracy-in-nz/what-is-new-zealands-system-of-government/what-is-mmp. Another is single transferable vote (STV), recently dubbed ‘proportional ranked-choice voting’.⁶¹For an introduction to ‘pure’ STV, see C.G.P. Grey’s video explainer: https://www.cgpgrey.com/blog/politics-in-the-animal-kingdom-single-transferable-vote. ‘Pure’ means without mechanisms like ticket voting. A third is open-list proportional representation (OLPR), which permits voters to set party-list order by choosing among candidates.⁶²See Kosar and Santucci (2021) for an explainer. A vote then helps elect a candidate and their party. For space considerations, we do not cover closed-list PR (in which voters choose among parties only).⁶³Moreover, many see closed-list PR as a non-starter for U.S. conditions due to its lack of candidate choice. Eberhard (2018) gives focus-group results to this effect.

Mixed-Member PR (MMP)

Political scientists who specialize in electoral systems typically rate MMP as among the very best options.⁶⁴Bowler, Shaun and David M. Farrell. 2006. “We Know Which One We Prefer but We Don’t Really Know Why: The Curious Case of Mixed Member Electoral Systems.” The British Journal of Politics and International Relations 8 (3): 445-560.⁶⁵Carey, John, Simon Hix, Shaheen Mozaffar, and Andrew Reynolds. 2013. “Report from the Field: Two Surveys of Political Scientists.” Chap. 7 in Political Science, Electoral Rules, and Democratic Governance: Report of the Task Force on Electoral Rules and Democratic Governance, eds. Mala Htun and G. Bingham Powell. Washington, DC: American Political Science Association. Recent research confirms its ability to balance national expertise-based policymaking with local responsiveness.⁶⁶Shugart, Matthew S., Matthew E. Bergman, Cory L. Struthers, Ellis S. Krauss, and Robert Pekkanen. 2021. Party Personnel Strategies: Electoral Systems and Parliamentary Committee Assignments. Oxford: Oxford University Press. MMP also might not disrupt race-conscious (single-seat) districting (although we have noted other problems with that). Two issues nonetheless raise questions about viability in the U.S. context: how to construct the ‘compensation tier,’ and the potential for ‘decoy lists’ in that tier.

Achieving proportionality under MMP requires compensation via the party-list tier. It would work best with a much larger U.S. House. MMP often has 50% of seats in single-seat districts and 50% from compensation lists. It can deliver a high degree of proportionality with a lower share (perhaps as low as 25%) but only on condition that the compensation regions be relatively large — ideally nationwide. We assume that nationwide or multi-state regional compensation is a nonstarter in the U.S. It likely is unconstitutional on its face. The Constitution stipulates that seats are apportioned among states, not shared among them. Thus only state-level compensation is doable, leaving even a significantly larger House insufficient for proportionality.⁶⁷In current House apportionment, thirteen states have 1 or 2 members, and MMP arguably requires four (perhaps three) in a state as an absolute minimum for compensation to work. Even with a House of 600 seats, several states would have fewer than three members. MMP nonetheless may be viable for state legislatures (Nagel 2014). On the possibility of expanding the House, see Drutman, et al. (2021).

The second problem is the possibility of large parties defeating the compensation mechanism via decoy lists. These are possible under ‘two-vote’ MMP (which would be necessary if reformers sought to induce multiparty politics). Party X directs voters to vote for Party X candidates in districts, but to cast their list vote for its decoy. When this happens, instead of Party X getting its district seats and only whatever number of list seats it needs to compensate for disproportionality arising from the district tier, it gets those district seats plus a fully proportional share of the list seats for its decoy. This practice can be avoided only by having a nationwide electoral administrative agency overseeing list registration (as in Germany and New Zealand) or by having only a single vote for both tiers (which however vitiates key advantages of MMP).⁶⁸Regulation might be done on a state-by-state level, but this would empower one or a few states to undermine a national election’s integrity by turning a blind eye to decoy lists.

Single Transferable Vote

STV is theoretically compatible with nonpartisan elections and permits electoral coalitions that defy party grouping.⁶⁹Lakeman, Enid and James Douglas Lambert. 1970. How Democracies Vote: A Study of Majority and Proportional Systems. London: Faber.⁷⁰Richie, Rob. 2022. “Whither and Whether Proportional Voting in the USA.” DemocracySOS, May 20. https://democracysos.substack.com/p/whither-and-whether-proportional-0b4. These properties make it popular but raise questions about longevity, given parties’ likely responses. For a sense of administrative issues, which include voter education, see Anthony et al.⁷¹Anthony, Joseph, Amy Fried, Robert Glover, and David C. Kimball. 2021. “Ranked Choice Voting in Maine from the Perspective of Local Election Officials.” Election Law Journal: Rules, Politics, and Policy 20 (3):254-271.

Where STV has been stable, various mechanisms exist for tempering its nonpartisanship.⁷²Bowler, Shaun and Bernard Grofman, eds. 2000. Elections in Australia, Ireland, and Malta under the Single Transferable Vote: Reflections on an Embedded Institution. Ann Arbor: University of Michigan Press.⁷³Clark, Alistair. 2020. “The Effects of Electoral Reform on Party Campaigns, Voters and Party Systems at the Local Level: from Single Member Plurality to the Single Transferable Vote in Scotland.” Local Government Studies 47 (1): 79-99.⁷⁴Farrell, David M. and Richard S. Katz. 2014. “Assessing the Proportionality of the Single Transferable Vote.”Representation 50 (1): 13-26. These include disciplined multi-party politics (Ireland), rules to give first-choice votes outsized importance (Malta), and a ticket-voting option so widely used that the system works effectively as closed-list PR (Australian Senate).⁷⁵Since 2016, Australian voters have had the option of ranking parties rather than either voting for one party ticket or else having to rank all candidates on their Senate ballot (McAllister & Muller 2018). This new option is still more akin to closed-list PR than to any other system, except that it allows the vote to transfer from one party to another.

In the U.S., by contrast, STV historically has not been stable. It was adopted in 22 cities from 1915-47, then repealed in all but one by 1962.⁷⁶This figure does not include two federally administered towns with advisory councils or two more former single-tax colonies. Reasons for this trajectory include: frustration with vote counts and election results⁷⁷Harris, Joseph P. 1930. “The Practical Workings of Proportional Representation in the United States and Canada.”National Municipal Review 19 (5): 3-50., bipartisan opposition from party elites⁷⁸Amy, Douglas J. 1996. “The Forgotten History of the Single Transferable Vote in the United States.”Representation 34 (1): 13-20.⁷⁹Weaver, Leon. 1986. “The Rise, Decline, and Resurrection of Proportional Representation in Local Governments in the United States.” Chap. 8 in Electoral Laws and their Political Consequences, eds Bernard Grofman and Arend Lijphart. New York: Agathon Press., and weak party discipline due to party change.⁸⁰Santucci, Jack. 2022. More Parties or No Parties: The Politics of Electoral Reform in America. New York: Oxford University Press.

Open-List PR

We have arrived at OLPR by working through what it might take to implement MMP, taking seriously some challenges associated with STV, and looking for a reasonable alternative. We are not the first to have followed such a path.⁸¹Lien, Arnold J. 1925. “Eight Years of Proportional Representation in Boulder, Colo.” Washington University Studies 13: 247-266.⁸²Gosnell, Harold F. 1939. “A List System with Single Candidate Preference.” The American Political Science Review 33 (4): 645-650.⁸³MMP did not ‘exist’ when those studies were published. However, the broad conclusion in each was that simple ballot formats might not have provoked adverse reactions by voters and election officials.

OLPR systems come in many ‘flavors,’⁸⁴For instance, “flexible” lists are not truly “open” but are sometimes conflated with them: voters have votes for candidates but these votes alter party-set list order only when a candidate’s votes cross some threshold. “Free” lists permit voters to cast multiple votes for candidates on more than one list. and a ‘one- vote’ version may be easiest to implement. It would not make new demands on voters or election officials. It just means each voter’s vote is for both a candidate and the list as a whole. In this way, the system might offer the advantages of PR, while remaining relatively familiar to stakeholders.

Descriptive Representation in STV and OLPR

Space constraints prevent an extended discussion of how these systems might shape racial and gender representation. This is an active research area. Key issues include: the extent to which voters ‘shun’ candidates from target groups⁸⁵Crowder-Meyer, Melody, Shana Kushner Gadarian, and Jessica Trounstine. 2023. “Ranking Candidates in Local Elections: Neither Panacea nor Catastrophe.” Journal of Experimental Political Science: 1-18. doi:10.1017/XPS.2023.6.⁸⁶Protsyk, Oleh, and Konstantin Sachariew. 2012. “Recruitment and Representation of Ethnic Minorities under Proportional Representation: Evidence from Bulgaria.” East European Politics and Societies 26 (2): 313-39.⁸⁷Sipinen, J., and Peter Söderlund. 2022. “Explaining the Migrant-Native Vote Gap Under Open-List Proportional Representation.” Party Politics. doi.org/10.1177/13540688221113665, whether parties field such candidates in the first place⁸⁸McGing, Claire. 2013. “The Single Transferable Vote and Women’s Representation in Ireland.” Irish Political Studies 28 (3): 322-340., and whether the need to maximize party vote share leads party leaders to nominate fewer such candidates.⁸⁹Valdini, Melody Ellis. 2012. “A Deterrent to Diversity: The Conditional Effect of Electoral Rules on the Nomination of Women Candidates.” Electoral Studies 31: 740-749. How do different forms of PR compare to one another in terms of delivering descriptive representation? Comparisonof closed lists with STV and OLPR suggests closed lists outperform both.⁹⁰Dhima, Kostanca, Sona N. Golder, Laura B. Stephenson, and Karine Van der Straeten. 2021. “Permissive Electoral Systems and Descriptive Representation.” Electoral Studies 73: 102381. A tentative conclusion might be that nominations matter — racist/sexist parties mean racist/sexist outcomes. Cultural attitudes also matter⁹¹Valdini, Melody Ellis. 2012. “A Deterrent to Diversity: The Conditional Effect of Electoral Rules on the Nomination of Women Candidates.” Electoral Studies 31: 740-749., but again these may shape nomination practices⁹²Hirczy, Wolfgang. 1995. “STV and the Representation of Women.” PS: Political Science and Politics 28 (4): 711-713.. Quotas also seem to shape party behavior in the long run.⁹³Barnes, Tiffany D., and Mirya R. Holman. 2020. “Gender Quotas, Women’s Representation, and Legislative Diversity.”The Journal of Politics 82 (4): 1271-1286.

Reforms for Single-Seat Offices

Reforms for Single-Seat Offices

How might results of U.S. Senate and Presidential elections be aligned with those to the House? One possible reform is cross-endorsement ballot fusion, which permits multiple parties to endorse the same candidate.⁹⁴Cross-endorsement fusion stands in contrast to cross-filing, whereby one candidate may declare multiple party designations. See Masket (2009) on cross-filing as an anti-party reform. This would be compatible with allowing OLPR (for the House) to feature joint lists.⁹⁵Shugart, Matthew S. and Rein Taagepera. 2017. Votes from Seats: Logical Models of Electoral Systems. Cambridge University Press. Another possibility for these offices is single-seat STV, also known as ‘instant runoff’ or the Alternative Vote.⁹⁶Getting inter-party coordination under instant runoff might mean requiring voters to rank all candidates (Reilly and Maley 2000; Reilly 2021).

Conclusion: Potentially More Than Two Parties

Conclusion: Potentially More Than Two Parties

PR adoption in the United States far from guarantees a multiparty system. It is important to recognize that the U.S. already has a smaller number of parties than it ‘should have’ given the fundamentals of its current House electoral system. Even electoral systems consisting only of winner-take-all SSDs should be expected to have more than two parties if they have hundreds of districts.⁹⁷Shugart, Matthew S. and Rein Taagepera. 2017. Votes from Seats: Logical Models of Electoral Systems. Cambridge University Press.⁹⁸Shugart, Matthew S. and Rein Taagepera. 2020. “Predicting Party Systems from Electoral Systems.” In Oxford Research Encyclopedia of Politics. Oxford University Press.⁹⁹Taagepera, Rein. 2007. Predicting Party Sizes: The Logic of Simple Electoral Systems. Oxford: Oxford University Press. The experiences of Canada and even the United Kingdom show that SSDs with plurality rule are compatible with multiparty politics. Thus SSDs are not the only factor constraining the number of parties. Other constraining factors would persist even if the House were elected by PR.

One such constraint is the Senate (with its coequal powers). A body for which, at any given election, only 33 or 34 seats are being filled in such a high- population country suppresses the emergence of additional parties.

Another constraint is ‘unit-rule’ allocation of presidential electors, which may lead voters to ‘desert’ minor-party candidates.¹⁰⁰Presidentialism per se is not the reason for a lower than expected number of parties in the USA. Shugart and Taagepera (2017) show that the seat product model of how party systems are connected to the assembly electoral system is just as reliable in a large dataset of elections around the world when a system is presidential as when it is parliamentary. The reforms we proposed for single-seat offices might alleviate voters’ incentive to desert. Or they might lead to fewer ‘spoiler’ candidacies overall.¹⁰¹Fusion, instant runoff, and other single-seat reforms do not fully obviate ‘spoiler’ candidacy. The key issue is whether the putative ‘spoiler’ otherwise would be part of some larger coalition (Riker 1982, 765).

Finally, we should not discount the role of primaries. When we look at the range of countries with first-past- the-post (FPTP) elections (given no primaries), none with an assembly larger than Jamaica’s (63) has a strict two-party system. These countries include the United Kingdom and Canada (where multiparty competition is in fact nationwide). Whether the U.S. should be called ‘FPTP’ itself is dubious, and not only because some states (e.g. Georgia) hold runoffs or use the Alternative Vote (e.g. Maine). Rather, the U.S. has an unusual two-round system in which the first round winnows the field. This usually is at the intraparty level, although sometimes it is without regard to party (e.g. in Alaska and California). Some of that winnowing- round competition might become interparty at the general election if PR were in place. On the other hand, it is perhaps difficult to imagine total abolition of primaries, and if they were to remain, new-party entry might remain more limited than otherwise expected.

In sum, adopting PR for the House of Representatives, particularly a model in which district magnitude is typically not much higher than five, might not proliferate parties as much as its critics fear and some proponents desire. We nonetheless hope to have sketched (and introduced scholarship on) how a responsible multiparty system might work.

The analysis, views, and conclusions contained herein reflect those of the author(s) and do not reflect the views of the American Political Science Association or Protect Democracy.